Alcoholismo y Cirrosis

MD Consult - Case Of The Week
April 21, 2000
Conde Petra
April 14, 2000
A 47-Year-Old Women With Alcoholic Cirrhosis and

SOB
A 47-year-old woman is admitted to the hospital secondary to increasing shortness
of breath. The patient has a history of alcoholic cirrhosis (diagnosed by liver biopsy
3 months ago) with ascites. She states that her abdomen has increased in size and
that her breathing has become increasingly difficult. She has had mild abdominal
discomfort for the last week but denies fever, nausea, vomiting, hematemesis,
melena, or mental status changes. She denies excess fluid or salt intake and has
been compliant with her medications, which include Lasix and Aldactone. Her
social history is significant for heavy alcohol intake for 15 years. She denies current
drug or tobacco use.
You are the resident on the floor admitting the patient. The patient carries a
Question 1 diagnosis of cirrhosis; however, you realize it is important to review the
relationship between alcohol and liver function before examining the patient.
How does ethanol damage the liver?
On physical examination her vital signs reveal a low-grade temperature and a

Question 2 respiratory rate of 28 breaths/minute. Her sclerae are icteric, and she has several
spider nevi on the chest. She has bibasilar crackles and bilateral E-A changes.
Her abdominal examination reveals tense distension and shifting dullness. Liver
size is difficult to evaluate due to the ascites. There is mild right upper quadrant
tenderness but no peritoneal signs. There is no peripheral edema, and rectal
examination reveals brown stool, heme negative.
What are the causes and clinical features of cirrhosis?
http://home.mdconsult.com/das/stat/view/cow?nid=39981 (1 of 4) [21/04/2000 23:04:22]

The laboratory data for your patient reveal a mild transaminase elevation,
Question 3 prothrombin time 16.2 sec, INR 1.9, albumin 1.8 g/dl, total protein 4.9 g/dl, WBC
13.3 K, plts 78 K, Hgb 10.8 g/dl, electrolytes are normal, viral hepatitis profile is
negative, ammonia 32, total bilirubin 2.2 g/dl, and alkaline phosphatase 167.
Arterial blood gas is consistent with respiratory alkalosis and a PaO2 of 58. Chest
x-ray shows small bilateral pleural effusions with compressive atelectasis but no
infiltrates. Abdominal ultrasound shows a large amount of ascites, irregular liver
surface, and splenomegaly. Upon receiving the above information, you place the
patient on 2 L O2 by nasal cannula and perform a therapeutic abdominal
paracentesis. While waiting for the infusion you try to remember the complications
of cirrhosis.
How much do you know about the complications of cirrhosis? Test your
knowledge.
The initial physical examination revealed a protuberant abdomen with a fluid

Question 4 wave, shifting dullness, and prominent flanks.
With what entity are these physical findings consistent and what are
the associated complications?
The next day the medical student on the case comes to you concerned about the
Question 5 patient. She states that the patient thought that she was in Mexico and living on a
tobacco farm and that the student was actually a mule used to haul the tobacco.
What is the likely diagnosis leading to such a mental status change?

What is another complication that may lead to renal failure?
The large-volume paracentesis was completed without complications, and the

Question 6 patient tolerated the procedure well. You were able to remove 5 L of fluid that you
sent for analysis. The cell count was normal, and the gram stain was negative.
The patient is feeling and breathing much better and remains very stable for the
next 24 hours. You are preparing to discharge the patient when the medical
students ask you about liver transplantation for this patient.

What are the indications and eligibility criteria for liver transplantation?
The patient was discharged 2 days after paracentesis with resolution of dyspnea.
Case Follow-Up The patient was referred to a regional transplant center for evaluation and is
currently on the waiting list for orthotopic liver transplantation.
Black M, Friedman AC: Ultrasound examination in the patient with ascites, Ann Intern Med
Bibliography
110(4):253-255, 1989(editorial)

Braunwald E, et al: Harrison's principles of internal medicine, ed 12, New York, 1993, McGraw-Hill

Cotran RS, Kumar V, Robbins SL: Robbins pathologic basis of disease, ed 4, Philadelphia, 1989, WB
Saunders

Gines P, et al: Norfloxacin prevents spontaneous bacterial peritonitis recurrence in cirrhosis:

results of a double-blind, placebo-controlled trial, Hepatology 12(4):717-723, 1990

Gines P, et al: Paracentesis with intravenous infusion of albumin as compared with

peritoneovenous shunting in cirrhosis with refractory ascites, N Engl J Med325(12):830-842, 1991

Gines P, et al: Randomized comparative study of therapeutic Paracentesis with and without
intravenous albumin in cirrhosis,Gastroenterology94:1493-1502, 1988

Hoefs JC: Diagnostic paracentesis, a potent clinical tool,Gastroenterology 98:230-236, 1990

Jensen DM: Portal-systemic encephalopathy and hepatic coma, Med Clin North Am
70(5):1081-1091, 1986

Jensen DM, Payne JA: Patient selection for liver transplantation. In Williams JW, ed: Hepatic
transplantation, Philadelphia, 1990, WB Saunders

Kandel G, Diamant NE: A clinical view of recent advances in ascites,J Clin Gastroenterol
8(1):85-99, 1986

Munoz SJ: Keeping current with the indications for liver transplantation, Intern Med, March 1994; 38

Rikkers LF: Variceal hemorrhage, Gastroenterol Clin North Am 17(2):289-301, 1988

Rossle M, et al: The transjugular intrahepatic portosystemic stent-shunt procedure for variceal
bleeding, N Engl J Med 330(3): 165-171, 1994

Runyon BA, Antillon MR, Montano AA: Effect of diuresis versus therapeutic paracentesis on
ascitic fluid opsonic activity and serum complement, Gastroenterology 97:158-162, 1989

Wilcox CM, Dismukes WE: Spontaneous bacterial peritonitis, a review of pathogenesis,

diagnosis and treatment, Medicine 66(6):447-455, 1987

Wyngaarden JB, Smith L, Bennett JC, eds: Cecil, textbook of medicine, ed 19, Philadelphia, 1992,

WB Saunders

Copyright 2000 MD Consult L.L.C.

http://www.mdconsult.com
Bookmark URL: /das/stat/view/cow?nid=39981&sid=0

April 21, 2000
Conde Petra
April 14, 2000
Question 1 - Alcohol Damage of the Liver

Alcohol abuse is the most common cause of liver disease in the Western world.
Alcohol exerts its damaging effects by several different cellular pathways. Ethanol
has a direct effect on membrane fluidity, leading to loss of membrane regulatory
control and cell swelling. Another pathway involves the production of excess
acetaldehyde from the metabolism of alcohol. Acetaldehyde exerts toxic effects on
the hepatocyte and is responsible for a more severe lesion. During alcohol
metabolism NADPH is formed, leading to an elevated redox potential favoring
accumulation of hepatic triglycerides.
There are three patterns of alcoholic liver disease. Steatosis (fatty liver) involves
the accumulation of triglycerides in hepatocytes. This process begins within the
first few days of alcohol intake and progresses with continued intake. This
condition is completely reversible if alcohol intake is discontinued. Clinically it
presents as moderate to massive hepatomegaly, right upper quadrant pain, and
mild transaminase elevation. Diagnosis is confirmed by liver biopsy showing large
droplet fat occupying most of the hepatocyte that resolves upon cessation of
drinking.
Alcoholic hepatitis is a more serious pattern and implies acute liver cell necrosis
with an inflammatory reaction. The histologic triad characteristic of this pattern
includes (1) alcoholic hyaline (eosinophilic aggregates) also known as Mallory
bodies, (2) infiltration by inflammatory cells, and (3) development of a network of
intralobular connective tissue. The clinical presentation of this disease is highly
variable, spanning the spectrum from asymptomatic to hepatic failure. Symptoms
may include fever, jaundice, anorexia, nausea, vomiting, and weight loss. Physical
examination reveals stigmata of liver disease. Laboratory evaluation reveals only
mild elevation of transaminases SGOT:SGPT >2, leucocytosis, prolonged
prothrombin time, and hypoalbuminemia. Diagnosis is confirmed by liver biopsy.
The outcome is unpredictable but return to normal is possible with discontinuation
of alcohol intake. Most commonly, alcoholic hepatitis leads to cirrhosis and may
precipitate hepatic failure and death.
Here are some citations about alcohol damage of the liver in

chronic alcoholics:
Proceed With Case

Alcoholism And Alcohol Abuse

Goldman: Cecil Textbook of Medicine, 21st Ed.
Copyright 2000 W. B. Saunders Company
Alcohols and Glycols

Ellenhorn's Medical Toxicology, 2nd ed.
Copyright 1997 Williams & Wilkins Alcohols and Glycols
ALCOHOL AND LIVER REGENERATION

Diehl A - Clinics in Liver Disease - 1998 Nov; 2(4); 723-738
Effects of prostaglandins on ethanol damage in primary cultured rat hepatocytes.

Yang JM - Korean J Intern Med - 1998 Feb; 13(1): 1-9
From NIH/NLM MEDLINE
Ethanol-induced alterations of the microtubule cytoskeleton in hepatocytes.

Yoon Y - Am J Physiol - 1998 Apr; 274(4 Pt 1): G757-66
Ethanol-induced retention of nascent proteins in rat hepatocytes is accompanied

by altered distribution of the small GTP-binding protein rab2.
Larkin JM - J Clin Invest - 1996 Nov 1; 98(9): 2146-57
Screening for Problem Drinking

- U.S. Preventive Services Task Force Guideline, Guide to Clinical Preventive Services, 2/e - 1996; ;
Alcoholism
Copyright Nidus Information Services 2000
Well-Connected


MD Consult - Reference Books
April 21, 2000
Conde Petra
Goldman: Cecil Textbook of Medicine, 21st Ed.,

About the Publication
Part IV - PREVENTIVE HEALTH CARE
16 - ALCOHOLISM AND ALCOHOL Goldman: Cecil Textbook of Medicine, 21st Ed., Copyright 2000 W.
B. Saunders Company
ABUSE
DEFINITIONS
ETIOLOGY
Chapter 16 - ALCOHOLISM
PREVALENCE
EPIDEMIOLOGY
AND ALCOHOL ABUSE
PATHOGENESIS
Ivan Diamond
ETHANOL ABSORPTION,
Cheryl A. Jay
DISTRIBUTION, AND
ELIMINATION.
ETHANOL METABOLISM. DEFINITIONS

CLINICAL MANIFESTATIONS
Alcoholism is characterized by addiction to ethanol. In contrast
ALCOHOL INTOXICATION. to behavioral and socioeconomic definitions of alcoholism, in a
medical setting alcoholism is a chronic disease in which the
TOLERANCE TO ALCOHOL. alcoholic craves and consumes ethanol uncontrollably, becomes
NATURAL HISTORY OF tolerant to its intoxicating effects with repetitive drinking, and
ALCOHOLISM. has symptoms and signs of alcohol withdrawal ( physical
dependence) when drinking is stopped. Individuals who drink
ALCOHOL WITHDRAWAL excessively without evidence of dependence have an alcohol
SYNDROME. abuse disorder. Binge drinking refers to bouts of excessive
TREMULOUSNESS. drinking for several days at a time. The distinction between
alcoholism and alcohol abuse has practical implications since
DISORDERED alcoholics require more intensive medical intervention.
PERCEPTIONS.
ALCOHOL WITHDRAWAL ETIOLOGY

CONVULSIONS.
Genetic susceptibility and environmental factors interact to
DELIRIUM TREMENS. produce alcoholism, often in families. Twin studies show that a
DIAGNOSIS
monozygotic twin of an alcoholic is more likely to be alcoholic
than is a dizygotic twin. Such studies yield broad heritability
TREATMENT estimates of 50%, which suggests that about half the variance
http://home.mdconsult.com/das/book/view/882/21.h...stty=cow&nid=39982&tlt=Z&pos=&tag=relinfo&anc=bo (1 of 17) [21/04/2000 23:05:55]

ACUTE ALCOHOL
for the development of alcoholism may be attributed to genetic
INTOXICATION.
factors. Even stronger support for genetic vulnerability to
ALCOHOL WITHDRAWAL alcoholism comes from adoption studies. Children of alcoholic
SYNDROME. parents who were adopted early in life by non-alcoholic parents
are over three times more likely to become alcoholics than are
ALCOHOLISM AND ALCOHOL
control adoptees. This pattern is particularly evident for
ABUSE.
"male-limited" alcoholism in fathers and sons with antisocial,
Bibliography impulsive, novelty-seeking behavior, who often begin drinking
as children or early adolescents. Adoption studies suggest that
this type of alcoholism in the biologic father is a much greater
predictor for alcoholism in the son than is the environment in
which the boy is raised.
50
PREVALENCE
A recent U.S. survey found the 12-month prevalence of alcohol
abuse and alcoholism to be 2.5% and 7.2%, respectively; the
lifetime prevalence was 9.4% and 14.1%. At least twice as
many men are alcoholic as women. Coexisting psychiatric
abnormalities include antisocial personality disorder,
schizophrenia, depression, anxiety disorders, and drug abuse.
Alcoholism and alcohol abuse affect 20% or more of
ambulatory and hospitalized patients. Physicians should be
aware that steady employment and social stability do not
exclude the diagnosis. Alcoholism develops in individuals of all
races and socioeconomic classes; only 5% of alcoholics fit the
"skid row" stereotype.
EPIDEMIOLOGY
Nearly two thirds of Americans older than 14 years drink
alcoholic beverages. Their per capita consumption is the
equivalent of 9.7 gallons of whiskey, 89 gallons of beer, or 31
gallons of wine per year. Heavy drinkers account for half of the
alcohol consumed and nearly all of the socioeconomic and
medical complications of alcoholism and alcohol abuse. The
annual cost of these problems to American society is about
$100 billion, a figure that includes costs to treat alcoholism and
related medical complications and lost productivity. Excessive
alcohol consumption ranks as the third leading preventable
cause of death, behind cigarette smoking and obesity, and
accounts for 5% of the total U.S. mortality, or about 100,000
deaths annually.

PATHOGENESIS
ETHANOL ABSORPTION, DISTRIBUTION, AND
ELIMINATION.
Ethanol is absorbed completely from the gastrointestinal tract

and is detected in the blood within minutes of ingestion. About
25% enters the bloodstream from the stomach and 75% from
the intestine, but gastrointestinal absorption is also affected by
food; the rate of drinking; the concentration, amount, and type
of alcoholic beverage; variations in gastrointestinal motility;
and gender. Most foods in the stomach delay gastric absorption,
and high concentrations of alcohol in the stomach can cause
pylorospasm, which slows gastric emptying and retards
intestinal absorption. Rapid gastric emptying or gastrectomy
increases rates of alcohol absorption from the small intestine.
Alcohol vapor can also be absorbed through the lungs. Women
have lower gastric alcohol dehydrogenase activity and hence
have higher blood alcohol concentrations than men do after
consuming similar amounts of ethanol per kilogram of body
weight.
Ethanol readily crosses biologic membranes, particularly in the
brain, and equilibrates rapidly into total body water. Ninety to
98% is removed in the liver, and the remainder is excreted by
the kidneys, lungs, and skin. Elimination proceeds at a constant
rate, independent of the blood alcohol concentration (zero-order
kinetics); a 70-kg man can metabolize 5 to 10 g ethanol per
hour. Since the average drink contains 12 to 15 g ethanol, blood
alcohol levels continue to rise when an individual drinks at a
rate greater than metabolism; however, when drinking is
discontinued, blood levels fall by about 10 to 25 mg/dL/hour.
ETHANOL METABOLISM.
Ethanol oxidation to acetaldehyde by alcohol dehydrogenase in

the liver is the rate-limiting step and accounts for more than
90% of ethanol metabolism in vivo. Alcohol dehydrogenase has
a high affinity for ethanol and accounts for essentially all
ethanol oxidation at low to moderate doses. When the blood
alcohol concentration is high, however, a microsomal
ethanol-oxidizing system with a lower affinity for ethanol can
also generate acetaldehyde (Fig. 16-1) . This oxidizing system
can be induced by ethanol to accelerate drug metabolism in the
liver (see Chapter 148) . Barbiturates have a similar effect,
which accounts for the metabolic cross-tolerance between these
agents.
Acetaldehyde is converted to acetate by aldehyde

dehydrogenase, a metabolic event with important clinical
ramifications. For example, in 50% of Japanese and other Asian
people, a genetic variation in an aldehyde dehydrogenase
isoenzyme results in reduced enzyme activity in vivo. Shortly
after drinking alcohol, affected individuals have increased blood
acetaldehyde levels and experience an alcohol-flush reaction
characterized by vasodilatation with facial flushing, hot
sensations, tachycardia, and hypotension. These unpleasant
experiences appear to deter drinking; Japanese and Chinese
people with this isoenzyme have a lower rate of alcoholism.
Pharmacologic inhibition of aldehyde dehydrogenase can cause
severe aversive symptoms after drinking alcohol and is the
reason why disulfiram (Antabuse) has been used to discourage
drinking. Disulfiram inhibits aldehyde dehydrogenase (and
other sulfhydryl-containing enzymes), but it is not ordinarily
toxic when taken therapeutically without ethanol. After drinking
alcohol, however, patients on prophylactic disulfiram therapy
have significant increases in blood acetaldehyde levels, and a
more severe acetaldehyde syndrome develops. They can
experience dysphoria, intense palpitations, sweating, thirst,
throbbing headache, dyspnea, nausea and vomiting, weakness,
vertigo, and syncope. Moreover, other drugs that likewise
inhibit aldehyde dehydrogenase, such as metronidazole
(Flagyl), may also make patients ill if they drink ethanol.
Disulfiram does not cure alcoholism and is not widely used.
In peripheral tissues, acetate derived from acetaldehyde is
converted to acetyl coenzyme A and subsequently to CO2 and
water. Complete oxidation of ethanol yields 7.1 kcal/g, and
ethanol may account for 5 to 10% of the total caloric intake in
the United States. Alcoholics often obtain 50% of their calories
from ethanol, and serious nutritional deficiencies, particularly
protein, thiamine, folate, and pyridoxine deficiency, develop in
many (Table 16-1) (see Chapter 231) . Moreover, as a
consequence of ethanol metabolism, alcoholics are prone to
hypoglycemia (Chapter 243) , lactic acidosis (Chapter 102) ,
hyperuricemia (Chapter 299) , and hypertriglyceridemia
(Chapter 206) . Binge drinking, inadequate diet, and severe
vomiting on a background of chronic alcohol consumption can
lead to alcoholic ketoacidosis (see Chapter 102) .

ALCOHOL INTOXICATION.
The blood-brain barrier to ethanol is virtually non-existent, and

shortly after drinking, the concentration
Figure 16-1 Ethanol metabolism. Alcohol

dehydrogenase (ADH) predominates at low to
moderate ethanol doses. The microsomal
ethanol-oxidizing system (MEOS) is induced at high
ethanol levels or chronic exposure and by certain drugs. Aldehyde
dehydrogenase (ALDH) inhibition (genetic or drug induced) leads to
acetaldehyde accumulation.
51
TABLE 16-1 -- ALCOHOL-RELATED MEDICAL

DISORDERS
AFFECTED ORGAN OR
SYSTEM DISORDERS
Nutrition Deficiencies of
Vitamins: Folate, thiamine,
pyridoxine, niacin, riboflavin
Minerals: Magnesium, zinc,
calcium
Protein
Metabolites and electrolytes Hypoglycemia, ketoacidosis,
hyperlipidemia,
hyperuricemia,
hypomagnesemia,
hypophosphatemia
GI tract Liver: Fatty liver, hepatitis,
cirrhosis
Gut: Esophagitis, gastritis
Pancreatitis

Nervous system Brain: Hepatic

encephalopathy,
Wernicke-Korsakoff
syndrome, cerebellar
degeneration, central pontine
myelinolysis,
Marchiafava-Bignami disease,
dementia
Neuromuscular: Neuropathy,
myopathy
Amblyopia
Cardiovascular Heart: Arrhythmia,
cardiomyopathy
Hypertension
Bone marrow Macrocytosis, anemia,
thrombocytopenia, leukopenia
Endocrine Pseudo-Cushing's syndrome,
testicular atrophy, amenorrhea
Other Traumatic injury
Aerodigestive neoplasms
Osteopenia
Fetal alcohol syndrome
of alcohol in the brain is nearly the same as in the blood. In a

non-alcoholic, intoxication occurs at blood alcohol levels of 50
to 150 mg/dL (Table 16-2) , and legal intoxication ranges from
80 to 100 mg/dL in most states. After two to three average
drinks (12 oz beer, 5 oz wine, or 1.5 oz 80-proof spirits), the
blood alcohol concentration approaches the legal limit,
depending on body weight and gender; women may achieve
higher blood alcohol concentrations with fewer drinks. A
general "rule of thumb" is that ethanol metabolism removes one
drink per hour. Symptoms vary directly with the rate of
drinking and are more severe when blood alcohol
concentrations are rising than falling. Most individuals feel
euphoric, lose social inhibitions, and manifest expansive,
sometimes garrulous behavior; others may become gloomy,
belligerent, or even explosively combative. Some people do not
experience euphoria but become sleepy after moderate drinking;
they rarely abuse alcohol. Neurologic signs of intoxication
include impaired cognition, slurred speech, incoordination, mild
truncal ataxia, and slow or irregular eye movements. Signs of
increased sympathetic activity include mydriasis, tachycardia,

and skin flushing. Cerebellar and vestibular function

deteriorates at higher blood alcohol levels, and drunkenness is
characterized by dysarthria, more severe ataxia, nystagmus, and
diplopia. Patients may become lethargic with bradycardia,
reduced blood pressure, and diminished respirations, sometimes
complicated by vomiting and pulmonary aspiration. In
non-tolerant individuals, stupor and coma may supervene at 400
mg/dL, and fatalities occur above this level, usually because of
respiratory
TABLE 16-2 -- BLOOD ETHANOL LEVELS AND
SYMPTOMS
BLOOD
ETHANOL SYMPTOMS
LEVELS
(mg/dL) Sporadic Drinkers Chronic Drinkers
50-100 Euphoria, Minimal or no effect
gregariousness,
incoordination
100-200 Slurred speech, Sobriety or
ataxia, labile mood, incoordination
drowsiness, nausea Euphoria
200-300 Lethargy, Mild emotional and
combativeness motor changes
Stupor, incoherent
speech
Vomiting
300-400 Coma Drowsiness
>500 Respiratory Lethargy, stupor,
depression, death coma
depression with respiratory acidosis and hypotension. The

median lethal dose for ethanol is approximately 450 mg/dL.
Other central nervous system depressants such as narcotics and
sedative-hypnotics act synergistically with alcohol.
Alcoholic blackouts sometimes complicate acute alcohol
intoxication during the consumption of large amounts of
ethanol. These episodes, which can occur in alcoholics or
sporadic drinkers, are characterized by amnesia for several
hours without impaired consciousness. The patient reports an
inability to remember new events but has no difficulty with
long-term memory or immediate recall. These symptoms

resemble the syndrome of transient global amnesia (see

Chapters 449 and 470) . Ethanol can depress myocardial
function at moderate doses, and binge drinking can cause
arrhythmias, or the holiday heart syndrome (see Chapter 64) .
Relaxation of vascular smooth muscle causes vasodilation,
which can lead to hypothermia, particularly in cold
environments.
TOLERANCE TO ALCOHOL.
A reduced response to ethanol, or tolerance, develops both

acutely and chronically during drinking and is due to adaptive
changes in the central nervous system, not ethanol metabolism.
Acute tolerance occurs during a single episode of drinking and
is characterized by greater intoxication at a given blood alcohol
concentration when the level is rising than when falling
(Mellanby effect). Chronic tolerance occurs in alcoholics and is
characterized by greater resistance to the intoxicating effects of
ethanol; they may appear to be sober at levels of 400 to 500
mg/dL, concentrations known to produce stupor, coma, or death
in naive individuals. The highest blood alcohol level reported is
1510 mg/dL in an ambulatory chronic alcoholic who had
stopped drinking 3 days earlier. Tolerance appears to be due to
ethanol-induced changes in gene expression and intracellular
signaling cascades involving neurotransmitter receptors, ion
channels, and protein kinases.
NATURAL HISTORY OF ALCOHOLISM.
Most people who consume alcohol begin drinking in

adolescence or early adulthood. Up to half of male drinkers
have alcohol-related problems such as blackouts, fighting, or a
single alcohol-related arrest during their late teens or early
twenties. Most learn to moderate their alcohol consumption by
their late twenties. Those who continue to accumulate
alcohol-related problems often become alcoholic. Craving and
uncontrolled drinking accompanied by tolerance and symptoms
of withdrawal signal the development of alcoholism. Episodes
of abstention and failed efforts to control drinking are common
and highlight the relapsing and remitting course of the disease.
Some alcoholics achieve long-term sobriety on their own; those
who do not face increased mortality from trauma and medical
complications.
ALCOHOL WITHDRAWAL SYNDROME.
Ethanol is a central nervous system depressant. In alcoholics,

the nervous system appears to adapt to chronic exposure to

ethanol by increasing the activity of neural mechanisms that
counteract alcohol's depressant effects. When drinking is
abruptly reduced or discontinued, these adaptive neural
mechanisms are left unrestrained by ethanol, and physical
dependence is manifested by a hyperexcitable alcohol
withdrawal syndrome. The alcohol withdrawal syndrome
typically evolves in a recognizable temporal sequence (Fig.
16-2) and consists of tremulousness, disordered perceptions,
seizures, and delirium tremens of varying severity.
TREMULOUSNESS.
Tremor, the earliest, most common, and most apparent

symptom, begins about 6 to 8 hours after the last drink, usually
the morning after an overnight abstinence ("morning shakes").
Tremor is generalized, coarse, and rapid and often accompanied
by irritability, nausea, and vomiting. The patient usually senses
an inner tremulousness even when the tremor is not severe.
Self-treatment is usually a morning drink to "quiet the nerves,"
followed by drinking for the rest of the day. If the alcoholic
does not resume drinking, tremor intensifies by 24 to 36 hours
and is exacerbated by motor activity or stress. The tremor can
be so severe that it interferes with walking, eating, or speech.
Accompanying symptoms and signs of sympathetic
hyperactivity are also apparent. The patient is increasingly
anxious and easily startled by minor stimuli and complains of
insomnia and anorexia. Increased sweating, facial flushing,
mydriasis, tachycardia, and mild hypertension are noted. Most
abnormalities subside in a few days, but increased arousal and
anxiety may persist for 2 weeks.
52
Figure 16-2 Time course of alcohol withdrawal.
DISORDERED PERCEPTIONS.
Disordered perceptions accompany tremor and sympathetic

hyperactivity in approximately 25% of patients, become most
pronounced at 24 to 36 hours, and clear in a few days. Often,
vivid nightmares interfere with sleep; while awake, ordinary
visual, auditory, and tactile experiences become distorted and
misinterpreted. Isolated and prolonged auditory hallucinations

may develop in alcoholics undergoing withdrawal ( alcoholic
hallucinosis) despite being alert, oriented, and without memory
loss. Hallucinations may persist for weeks even though other
signs of ethanol withdrawal have improved and the patient is
less agitated and tremulous. In the absence of sympathetic
hyperactivity, persistent auditory hallucinations may be
confused with acute schizophrenia (see Chapter 450) .
However, alcoholic hallucinosis is closely associated with
ethanol withdrawal and usually subsides in weeks to months.
ALCOHOL WITHDRAWAL CONVULSIONS.
Generalized tonic-clonic seizures develop in about one third of

alcoholics, most often within 12 to 24 hours after reducing or
stopping drinking. Some propose that the first seizure in
alcoholics may be a consequence of ethanol toxicity. However,
ethanol dependence is followed by withdrawal seizures in
animals, particularly in mice bred to have convulsions during
withdrawal, thus suggesting a role for genetic vulnerability in
humans. Ethanol withdrawal seizures usually follow chronic
daily drinking but can also occur after 5 to 7 days of binge
drinking. Alcoholics who have seizures during one episode of
withdrawal are likely to have them again when alcohol
withdrawal is repeated. One isolated convulsion or several
seizures may occur, usually within a 6-hour period. Focal
seizures are less common and indicate a cerebral lesion, either
old or new. Status epilepticus occurs in about 3% of cases, and
ethanol withdrawal accounts for about 15% of all patients
evaluated for status epilepticus. Status epilepticus is a medical
emergency and requires immediate treatment with
anticonvulsants (see Chapter 484) .
DELIRIUM TREMENS.
Delirium tremens, the most alarming manifestation of the

ethanol withdrawal syndrome, occurs in about 5% of alcoholics.
It consists of agitated arousal, global confusion and
disorientation, insomnia, and vivid, often threatening
hallucinations and delusions. Signs of sympathetic hyperactivity
include tremor, mydriasis, tachycardia, fever, and intense
diaphoresis. In contrast to tremulousness, disordered
perceptions, and seizures, which appear earlier after withdrawal,
delirium tremens begins abruptly within 2 to 4 days of
abstinence, sometimes as a surprising development in an
unrecognized alcoholic admitted to the hospital for other

reasons. Patients are terrified by their hallucinations and can be

combative, destructive, and very dangerous. Episodes of
delirium tremens last from 1 to 3 days and end as abruptly as
they begin. However, relapses occur, and the disorder may
continue for days to weeks with intervening periods of lucidity.
When signs of sympathetic hyperactivity are absent, it may be
difficult to distinguish delirium tremens from an acute
psychosis. However, the diagnosis is usually suggested by the
evolution of symptoms in a chronic alcoholic undergoing
withdrawal. The differential diagnosis includes alcoholic
hypoglycemia, overdose with anticholinergic agents,
intoxication with amphetamines, cocaine, and phencyclidine,
and withdrawal from other sedating drugs. Metabolic
disturbances, cerebral infection, encephalitis, meningitis, sepsis,
or thyrotoxicosis should also be considered.
DIAGNOSIS
Alcohol-related problems are common in medical practice, and
physicians should be alert to their diverse clinical
manifestations. End-organ complications such as ulcers,
hepatitis, cirrhosis, pancreatitis, cardiomyopathy, or peripheral
neuropathy should prompt consideration of alcoholism as an
underlying cause. Recurrent trauma, particularly skeletal
fractures, sleep disorders, fatigue, depression, sexual
dysfunction, and labile hypertension, should also arouse
suspicion of alcoholism. Hospitalization may precipitate the
unexpected appearance of an alcohol withdrawal syndrome
within several days of admission.
Physicians should identify individuals who do not meet the
criteria for alcohol dependence but who drink alcohol at levels
that pose potential health risks. Laboratory-based screening
such as elevated mean corpuscular volume (MCV) or
gamma-glutamyl transpeptidase (GGTP) may help the
physician in confronting patient denial, but elevated MCV or
GGTP values are neither specific for alcoholism nor sufficiently
sensitive to serve as effective screens. Simple screening
questionnaires such as the CAGE (Fig. 16-3) outperform
laboratory measures in detecting excessive alcohol consumption
in a variety of clinical settings. Because advice and discussion
by a concerned physician are simple, cost-effective ways to
reduce heavy drinking, routine screening for alcohol disorders
is recommended, particularly in primary care.
Complex ethanol-drug interactions must also be considered
when prescribing medications to patients who use alcohol.
Ethanol potentiates the central nervous system depressant

effects of narcotic, sedative, and psychoactive drugs (see
Chapter 26) , and patients for whom these agents are prescribed
should be counseled explicitly about this interaction. In
contrast, the induction of microsomal enzymes in the liver by
ethanol accelerates the elimination of drugs metabolized by
these enzymes (see Chapter 26) . As a result, ethanol can
decrease drug efficacy by lowering the amount of active drug
available or increase toxicity and promote unusual side effects
caused by drug metabolites.
TREATMENT
ACUTE ALCOHOL INTOXICATION.
Mild to moderate ethanol intoxication requires no specific

therapy. Severe acute alcohol intoxication, defined by a
depressed level of consciousness, can be fatal and is a medical
emergency. Administration of sedatives to intoxicated patients
who are agitated and combative can lead to stupor, coma, and
respiratory arrest from synergistic depressant effects and should
be avoided. The immediate history should include information
about the quantity of alcohol consumed, the rate of drinking,
use of other drugs including methanol and ethylene glycol,
complicating medical and psychiatric disorders, and prior
alcohol abuse or alcoholism. If the patient is stuporous and
unable to walk, the airway must be evaluated immediately.
Indications for endotracheal intubation and assisted ventilation
include marked hypoventilation, accumulating secretions, or
coma. Complications such as hypoglycemia, meningitis,
subdural hematoma, and hepatic encephalopathy must be
considered. Evidence of head trauma or focal cerebral signs
suggests urgent intracranial pathology, and a computed
tomography scan should be performed immediately. Otherwise,
routine scans for alcohol intoxication are not indicated. Gastric
lavage may be performed if the obtundation is due to recent and
massive alcohol consumption, but it must be preceded by
endotracheal intubation. Hemodialysis should be considered if
the blood alcohol concentration exceeds 500 mg/dL or when
methanol or ethylene glycol has been ingested concurrently.
After a history and physical examination, patients with adequate
vital signs and acceptable mental status but without evidence of
other disorders can be kept calm under observation until
sobriety returns. However, medical information is usually
incomplete, and it is often necessary to anticipate complications

commonly associated with severe alcohol intoxication or

alcoholism. Routine blood counts and laboratory studies may
uncover anemia (see Chapter 159) , hypokalemia,
hypophosphatemia, and hypomagnesemia.
53
Figure 16-3 Screening and brief intervention for alcohol

problems in clinical practice.
Alcoholic hypoglycemia (see Chapter 243) can be evaluated

rapidly by a bedside blood glucose determination. If laboratory
results are delayed, 12.5 to 25 g glucose should be given
intravenously but must be preceded by or accompanied by 100
mg intravenous thiamine to avoid precipitating Wernicke's
encephalopathy (see Chapter 489) . Alcoholic ketoacidosis (see
Chapter 102) will be improved by infusion of 5% dextrose in
half-normal saline, also with thiamine. If the blood alcohol level
is too low to account for obtundation or if improvement does
not occur as expected, it is necessary to search for other causes
of stupor and coma (see Chapter 444) , including other sedating
agents.
Alcoholics stop drinking for many reasons, including serious

alcohol-related medical, surgical, or psychiatric conditions.
Hence symptoms or signs of trauma, infection, liver disease,
gastritis, pancreatitis, arrhythmia, or electrolyte disturbance
should be sought. One hundred milligrams of thiamine should
be given intravenously to all patients undergoing ethanol
withdrawal to prevent or treat Wernicke's encephalopathy (see
Chapter 489) and should be followed by daily multivitamins.
The alarming symptoms of ethanol withdrawal are best

managed by substituting another central nervous system
depressant. However, alcoholics undergoing withdrawal are
very resistant to sedatives (cross-tolerance), so large doses are
often required to calm their agitation. Benzodiazepines are
widely used to manage tremulousness and disordered
perceptions during ethanol withdrawal. The goal is to suppress

symptoms and produce mild sedation, and the drug dosage is

adjusted to the severity of the withdrawal reaction. Treatment
includes managing delirium and autonomic stability and
preventing seizures. A sedative-hypnotic agent, typically a
benzodiazepine, is prescribed as a substitute for alcohol, and the
dose is tapered over several days. Patients with mild
tremulousness and few associated symptoms usually respond to
oral diazepam, 5 to 10 mg every 4 to 6 hours. The dosage is
then reduced by 20 to 25% on successive days or increased if
symptoms of ethanol withdrawal return. beta-Blockers are
useful ancillary therapy; they attenuate the symptoms of
autonomic hypersensitivity but are not anticonvulsants and do
not appear to reduce delirium. Detoxification can be carried out
with close monitoring in an outpatient setting in socially stable
patients with mild withdrawal. If withdrawal is more severe or
accompanied by significant medical, surgical, or psychiatric
illness or the patient is in an unstable social setting, inpatient
detoxification may be needed. In such instances,
benzodiazepines such as diazepam (Valium), chlordiazepoxide
(Librium), oxazepam (Serax), or lorazepam (Ativan) are
administered orally or parenterally in doses sufficient to keep
the patient calm. Benzodiazepines should not be given
intramuscularly because of inconsistent absorption. Patients
may require hourly medication at doses that would be fatal in
non-tolerant individuals. The first several days of severe alcohol
withdrawal may require intravenous administration of total
daily diazepam doses exceeding 400 mg (or the equivalent of
other benzodiazepines) to achieve mild sedation. Multivitamin
and thiamine supplementation should be continued, as should
meticulous attention to electrolyte status. The benzodiazepine
dosage can then be tapered by approximately 20 to 25% on
successive days, with an increase in dosage if withdrawal
symptoms recur. Once the symptoms of ethanol withdrawal are
suppressed, it is necessary to avoid oversedation and the danger
of respiratory depression by carefully titrating the dose of
diazepam to just keep the patient calm.
Alcohol withdrawal seizures can often be managed with
intravenous benzodiazepines such as diazepam or lorazepam.
Phenytoin does not prevent seizures during withdrawal.
Management of status epilepticus is the same as in other
situations (see Chapter 484) . Alcoholics are at increased risk
for head trauma and central nervous system infection; studies to
exclude these more serious diagnostic possibilities should be
performed when seizures occurring in the setting of withdrawal
display focal features or are accompanied by a prolonged

post-ictal state or when status epilepticus intervenes. Long-term

anticonvulsant therapy is not indicated for typical alcohol
withdrawal seizures.
Delirium tremens requires hospitalization and vigorous
management in an intensive care setting. Mortality has reached
15% in the past, primarily because of injuries or associated
medical disorders complicated by hyperthermia and
dehydration. Volume depletion accompanying delirium tremens
may cause circulatory collapse, and fluid losses can require
replacement of 4 to 10 L in the first day. The goal of treatment
is to control behavior and suppress symptoms without danger to
the patient. Five to 10 mg or more of diazepam is given
intravenously every 5 to 15 minutes until the patient is calm,
and maintenance therapy is continued every 1 to 4 hours, as
needed. Initially, as much as 200 mg of diazepam may be
required before the agitation subsides. Seizures are unusual in
patients with delirium tremens and should be evaluated
promptly because of the possibility of meningitis or other
disorders. Coexisting hepatic and cardiac disease may
complicate fluid management, and the possibility that sedative
agents may precipitate hepatic encephalopathy should be kept in
mind.
54
ALCOHOLISM AND ALCOHOL ABUSE.
Alcoholics and alcohol abusers come to medical attention

because of alcohol-related medical or psychiatric conditions, by
referral from social service or criminal justice agencies, or
through screening in clinical practice. Family members provide
valuable collateral history. Physicians should confront
alcoholics in a firm but non-judgmental fashion, educate them
about health risks, and assess their motivation to stop drinking
(see Fig. 16-3) . Heavy drinkers should be counseled to reduce
consumption. It is valuable to establish a contract with the
patient to decrease drinking and return for follow-up
assessments. Alcoholics should be referred to a rehabilitation
program but may first require inpatient detoxification. Diverse
psychosocial interventions have been tested; all are equally
effective, and a successful outcome is related more to interested
personal intervention than to psychotherapy matched to the
patient's condition. Medication may be a useful adjunct in some
instances. Disulfiram can be helpful in highly selected patients.
The opiate antagonist naltrexone (ReVia), the only other agent

currently approved by the Food and Drug Administration

(FDA) for the treatment of alcoholism, appears to decrease the
relapse rate in abstinent alcoholics. Acamprosate, a drug not yet
approved by the FDA, has also shown promising results in
clinical trials.
Intervention is more effective earlier in the course of the illness,
before the onset of associated medical disorders. Alcoholics
who continue to drink shorten their lifespans by at least 15
years. Many alcohol-related medical complications such as
ulcer disease, acute pancreatitis, hepatitis, myopathy, and
neuropathy stabilize or regress with continued abstinence.
Others such as cirrhosis with portal hypertension,
Wernicke-Korsakoff syndrome, or dilated cardiomyopathy
frequently cause permanent disability or death. About half of
socially stable, middle-class alcoholics remain sober for at least
a year after rehabilitation. Alcoholics Anonymous and Al-Anon
provide low-cost support for alcoholics and their families in
virtually all communities in the United States.
Litten RZ, Allen JP: Medications for alcohol, illicit drug, and
tobacco dependence: An update of research findings. J Subst
Abuse Treat 16:105, 1999. New approaches to alcohol
dependence.
O'Connor PG, Schottenfeld RS: Patients with alcohol problems.
N Engl J Med 338:592-602, 1998. A useful review of the
diagnosis and treatment of alcohol-related problems.
Saitz R, O'Malley SS: Pharmacotherapies for alcohol abuse:
Withdrawal and treatment. Med Clin North Am 81:881, 1997. A
helpful review of the pathophysiology, treatment goals, and
medication options for alcohol withdrawal and dependence.
Schorling JB, Buchsbaum DG: Screening for alcohol and drug
abuse. Med Clin North Am 81:845, 1997. An excellent summary
of the rationale for alcohol screening in primary care, use of
screening questionnaires such as the CAGE, and practical
guidelines for brief physician intervention.
US Department of Health and Human Services: Ninth Special
Report to the US Congress on Alcohol and Health. Rockville,
MD, National Institute on Alcohol Abuse and Alcoholism,
1996. A comprehensive discussion of the major biomedical and
socioeconomic problems of alcoholism and alcohol abuse.
MD Consult L.L.C. http://www.mdconsult.com

Bookmark URL: /das/book/view/882/21.html/top

MD Consult - Book Text
Goldman: Cecil Textbook of Medicine, 21st Ed., Copyright 2000 W. B. Saunders Company
Chapter 16 - ALCOHOLISM AND ALCOHOL ABUSE
Ivan Diamond
Cheryl A. Jay
DEFINITIONS
Alcoholism is characterized by addiction to ethanol. In contrast to behavioral and socioeconomic
definitions of alcoholism, in a medical setting alcoholism is a chronic disease in which the alcoholic
craves and consumes ethanol uncontrollably, becomes tolerant to its intoxicating effects with repetitive
drinking, and has symptoms and signs of alcohol withdrawal ( physical dependence) when drinking is
stopped. Individuals who drink excessively without evidence of dependence have an alcohol abuse
disorder. Binge drinking refers to bouts of excessive drinking for several days at a time. The distinction
between alcoholism and alcohol abuse has practical implications since alcoholics require more intensive
medical intervention.
ETIOLOGY
Genetic susceptibility and environmental factors interact to produce alcoholism, often in families. Twin
studies show that a monozygotic twin of an alcoholic is more likely to be alcoholic than is a dizygotic
twin. Such studies yield broad heritability estimates of 50%, which suggests that about half the variance
for the development of alcoholism may be attributed to genetic factors. Even stronger support for genetic
vulnerability to alcoholism comes from adoption studies. Children of alcoholic parents who were adopted
early in life by non-alcoholic parents are over three times more likely to become alcoholics than are
control adoptees. This pattern is particularly evident for "male-limited" alcoholism in fathers and sons
with antisocial, impulsive, novelty-seeking behavior, who often begin drinking as children or early
adolescents. Adoption studies suggest that this type of alcoholism in the biologic father is a much greater
predictor for alcoholism in the son than is the environment in which the boy is raised.
50
PREVALENCE
A recent U.S. survey found the 12-month prevalence of alcohol abuse and alcoholism to be 2.5% and
7.2%, respectively; the lifetime prevalence was 9.4% and 14.1%. At least twice as many men are
alcoholic as women. Coexisting psychiatric abnormalities include antisocial personality disorder,
schizophrenia, depression, anxiety disorders, and drug abuse. Alcoholism and alcohol abuse affect 20%
or more of ambulatory and hospitalized patients. Physicians should be aware that steady employment and
social stability do not exclude the diagnosis. Alcoholism develops in individuals of all races and
http://home.mdconsult.com/das/book/body/0/882/21.html (1 of 11) [21/04/2000 23:06:19]

socioeconomic classes; only 5% of alcoholics fit the "skid row" stereotype.
EPIDEMIOLOGY
Nearly two thirds of Americans older than 14 years drink alcoholic beverages. Their per capita
consumption is the equivalent of 9.7 gallons of whiskey, 89 gallons of beer, or 31 gallons of wine per
year. Heavy drinkers account for half of the alcohol consumed and nearly all of the socioeconomic and
medical complications of alcoholism and alcohol abuse. The annual cost of these problems to American
society is about $100 billion, a figure that includes costs to treat alcoholism and related medical
complications and lost productivity. Excessive alcohol consumption ranks as the third leading
preventable cause of death, behind cigarette smoking and obesity, and accounts for 5% of the total U.S.
mortality, or about 100,000 deaths annually.
PATHOGENESIS
ETHANOL ABSORPTION, DISTRIBUTION, AND ELIMINATION.
Ethanol is absorbed completely from the gastrointestinal tract and is detected in the blood within minutes
of ingestion. About 25% enters the bloodstream from the stomach and 75% from the intestine, but
gastrointestinal absorption is also affected by food; the rate of drinking; the concentration, amount, and
type of alcoholic beverage; variations in gastrointestinal motility; and gender. Most foods in the stomach
delay gastric absorption, and high concentrations of alcohol in the stomach can cause pylorospasm,
which slows gastric emptying and retards intestinal absorption. Rapid gastric emptying or gastrectomy
increases rates of alcohol absorption from the small intestine. Alcohol vapor can also be absorbed
through the lungs. Women have lower gastric alcohol dehydrogenase activity and hence have higher
blood alcohol concentrations than men do after consuming similar amounts of ethanol per kilogram of
body weight.
Ethanol readily crosses biologic membranes, particularly in the brain, and equilibrates rapidly into total
body water. Ninety to 98% is removed in the liver, and the remainder is excreted by the kidneys, lungs,
and skin. Elimination proceeds at a constant rate, independent of the blood alcohol concentration
(zero-order kinetics); a 70-kg man can metabolize 5 to 10 g ethanol per hour. Since the average drink
contains 12 to 15 g ethanol, blood alcohol levels continue to rise when an individual drinks at a rate
greater than metabolism; however, when drinking is discontinued, blood levels fall by about 10 to 25
mg/dL/hour.
ETHANOL METABOLISM.
Ethanol oxidation to acetaldehyde by alcohol dehydrogenase in the liver is the rate-limiting step and
accounts for more than 90% of ethanol metabolism in vivo. Alcohol dehydrogenase has a high affinity
for ethanol and accounts for essentially all ethanol oxidation at low to moderate doses. When the blood
alcohol concentration is high, however, a microsomal ethanol-oxidizing system with a lower affinity for
ethanol can also generate acetaldehyde (Fig. 16-1) . This oxidizing system can be induced by ethanol to
accelerate drug metabolism in the liver (see Chapter 148) . Barbiturates have a similar effect, which
accounts for the metabolic cross-tolerance between these agents.

Acetaldehyde is converted to acetate by aldehyde dehydrogenase, a metabolic event with important

clinical ramifications. For example, in 50% of Japanese and other Asian people, a genetic variation in an
aldehyde dehydrogenase isoenzyme results in reduced enzyme activity in vivo. Shortly after drinking
alcohol, affected individuals have increased blood acetaldehyde levels and experience an alcohol-flush
reaction characterized by vasodilatation with facial flushing, hot sensations, tachycardia, and
hypotension. These unpleasant experiences appear to deter drinking; Japanese and Chinese people with
this isoenzyme have a lower rate of alcoholism. Pharmacologic inhibition of aldehyde dehydrogenase can
cause severe aversive symptoms after drinking alcohol and is the reason why disulfiram (Antabuse) has
been used to discourage drinking. Disulfiram inhibits aldehyde dehydrogenase (and other
sulfhydryl-containing enzymes), but it is not ordinarily toxic when taken therapeutically without ethanol.
After drinking alcohol, however, patients on prophylactic disulfiram therapy have significant increases in
blood acetaldehyde levels, and a more severe acetaldehyde syndrome develops. They can experience
dysphoria, intense palpitations, sweating, thirst, throbbing headache, dyspnea, nausea and vomiting,
weakness, vertigo, and syncope. Moreover, other drugs that likewise inhibit aldehyde dehydrogenase,
such as metronidazole (Flagyl), may also make patients ill if they drink ethanol. Disulfiram does not cure
alcoholism and is not widely used.
In peripheral tissues, acetate derived from acetaldehyde is converted to acetyl coenzyme A and
subsequently to CO2 and water. Complete oxidation of ethanol yields 7.1 kcal/g, and ethanol may
account for 5 to 10% of the total caloric intake in the United States. Alcoholics often obtain 50% of their
calories from ethanol, and serious nutritional deficiencies, particularly protein, thiamine, folate, and
pyridoxine deficiency, develop in many (Table 16-1) (see Chapter 231) . Moreover, as a consequence of
ethanol metabolism, alcoholics are prone to hypoglycemia (Chapter 243) , lactic acidosis (Chapter 102) ,
hyperuricemia (Chapter 299) , and hypertriglyceridemia (Chapter 206) . Binge drinking, inadequate diet,
and severe vomiting on a background of chronic alcohol consumption can lead to alcoholic ketoacidosis
(see Chapter 102) .
ALCOHOL INTOXICATION.
The blood-brain barrier to ethanol is virtually non-existent, and shortly after drinking, the concentration
Figure 16-1 Ethanol metabolism. Alcohol dehydrogenase (ADH) predominates at low to moderate
ethanol doses. The microsomal ethanol-oxidizing system (MEOS) is induced at high ethanol levels or
chronic exposure and by certain drugs. Aldehyde dehydrogenase (ALDH) inhibition (genetic or drug
induced) leads to acetaldehyde accumulation.
51
TABLE 16-1 -- ALCOHOL-RELATED MEDICAL DISORDERS

AFFECTED ORGAN OR SYSTEM DISORDERS

Nutrition Deficiencies of
Vitamins: Folate, thiamine, pyridoxine, niacin,
riboflavin
Minerals: Magnesium, zinc, calcium
Protein
Metabolites and electrolytes Hypoglycemia, ketoacidosis, hyperlipidemia,
hyperuricemia, hypomagnesemia,
hypophosphatemia
GI tract Liver: Fatty liver, hepatitis, cirrhosis
Gut: Esophagitis, gastritis
Pancreatitis
Nervous system Brain: Hepatic encephalopathy,
Wernicke-Korsakoff syndrome, cerebellar
degeneration, central pontine myelinolysis,
Marchiafava-Bignami disease, dementia
Neuromuscular: Neuropathy, myopathy
Amblyopia
Cardiovascular Heart: Arrhythmia, cardiomyopathy
Hypertension
Bone marrow Macrocytosis, anemia, thrombocytopenia,
leukopenia
Endocrine Pseudo-Cushing's syndrome, testicular atrophy,
amenorrhea
Other Traumatic injury
Aerodigestive neoplasms
Osteopenia
Fetal alcohol syndrome
of alcohol in the brain is nearly the same as in the blood. In a non-alcoholic, intoxication occurs at blood
alcohol levels of 50 to 150 mg/dL (Table 16-2) , and legal intoxication ranges from 80 to 100 mg/dL in
most states. After two to three average drinks (12 oz beer, 5 oz wine, or 1.5 oz 80-proof spirits), the
blood alcohol concentration approaches the legal limit, depending on body weight and gender; women
may achieve higher blood alcohol concentrations with fewer drinks. A general "rule of thumb" is that
ethanol metabolism removes one drink per hour. Symptoms vary directly with the rate of drinking and
are more severe when blood alcohol concentrations are rising than falling. Most individuals feel
euphoric, lose social inhibitions, and manifest expansive, sometimes garrulous behavior; others may
become gloomy, belligerent, or even explosively combative. Some people do not experience euphoria but
become sleepy after moderate drinking; they rarely abuse alcohol. Neurologic signs of intoxication
include impaired cognition, slurred speech, incoordination, mild truncal ataxia, and slow or irregular eye
movements. Signs of increased sympathetic activity include mydriasis, tachycardia, and skin flushing.
Cerebellar and vestibular function deteriorates at higher blood alcohol levels, and drunkenness is

characterized by dysarthria, more severe ataxia, nystagmus, and diplopia. Patients may become lethargic
with bradycardia, reduced blood pressure, and diminished respirations, sometimes complicated by
vomiting and pulmonary aspiration. In non-tolerant individuals, stupor and coma may supervene at 400
mg/dL, and fatalities occur above this level, usually because of respiratory
TABLE 16-2 -- BLOOD ETHANOL LEVELS AND SYMPTOMS
BLOOD ETHANOL SYMPTOMS
LEVELS (mg/dL) Sporadic Drinkers Chronic Drinkers
50-100 Euphoria, gregariousness, Minimal or no effect
incoordination
100-200 Slurred speech, ataxia, labile mood, Sobriety or incoordination
drowsiness, nausea Euphoria
200-300 Lethargy, combativeness Mild emotional and motor changes
Stupor, incoherent speech
Vomiting
300-400 Coma Drowsiness
>500 Respiratory depression, death Lethargy, stupor, coma
depression with respiratory acidosis and hypotension. The median lethal dose for ethanol is
approximately 450 mg/dL. Other central nervous system depressants such as narcotics and
sedative-hypnotics act synergistically with alcohol.
Alcoholic blackouts sometimes complicate acute alcohol intoxication during the consumption of large
amounts of ethanol. These episodes, which can occur in alcoholics or sporadic drinkers, are characterized
by amnesia for several hours without impaired consciousness. The patient reports an inability to
remember new events but has no difficulty with long-term memory or immediate recall. These symptoms
resemble the syndrome of transient global amnesia (see Chapters 449 and 470) . Ethanol can depress
myocardial function at moderate doses, and binge drinking can cause arrhythmias, or the holiday heart
syndrome (see Chapter 64) . Relaxation of vascular smooth muscle causes vasodilation, which can lead
to hypothermia, particularly in cold environments.
TOLERANCE TO ALCOHOL.
A reduced response to ethanol, or tolerance, develops both acutely and chronically during drinking and is
due to adaptive changes in the central nervous system, not ethanol metabolism. Acute tolerance occurs
during a single episode of drinking and is characterized by greater intoxication at a given blood alcohol
concentration when the level is rising than when falling (Mellanby effect). Chronic tolerance occurs in
alcoholics and is characterized by greater resistance to the intoxicating effects of ethanol; they may
appear to be sober at levels of 400 to 500 mg/dL, concentrations known to produce stupor, coma, or
death in naive individuals. The highest blood alcohol level reported is 1510 mg/dL in an ambulatory
chronic alcoholic who had stopped drinking 3 days earlier. Tolerance appears to be due to
ethanol-induced changes in gene expression and intracellular signaling cascades involving

neurotransmitter receptors, ion channels, and protein kinases.
NATURAL HISTORY OF ALCOHOLISM.
Most people who consume alcohol begin drinking in adolescence or early adulthood. Up to half of male
drinkers have alcohol-related problems such as blackouts, fighting, or a single alcohol-related arrest
during their late teens or early twenties. Most learn to moderate their alcohol consumption by their late
twenties. Those who continue to accumulate alcohol-related problems often become alcoholic. Craving
and uncontrolled drinking accompanied by tolerance and symptoms of withdrawal signal the
development of alcoholism. Episodes of abstention and failed efforts to control drinking are common and
highlight the relapsing and remitting course of the disease. Some alcoholics achieve long-term sobriety
on their own; those who do not face increased mortality from trauma and medical complications.
Ethanol is a central nervous system depressant. In alcoholics, the nervous system appears to adapt to
chronic exposure to ethanol by increasing the activity of neural mechanisms that counteract alcohol's
depressant effects. When drinking is abruptly reduced or discontinued, these adaptive neural mechanisms
are left unrestrained by ethanol, and physical dependence is manifested by a hyperexcitable alcohol
withdrawal syndrome. The alcohol withdrawal syndrome typically evolves in a recognizable temporal
sequence (Fig. 16-2) and consists of tremulousness, disordered perceptions, seizures, and delirium
tremens of varying severity.
TREMULOUSNESS.
Tremor, the earliest, most common, and most apparent symptom, begins about 6 to 8 hours after the last
drink, usually the morning after an overnight abstinence ("morning shakes"). Tremor is generalized,
coarse, and rapid and often accompanied by irritability, nausea, and vomiting. The patient usually senses
an inner tremulousness even when the tremor is not severe. Self-treatment is usually a morning drink to
"quiet the nerves," followed by drinking for the rest of the day. If the alcoholic does not resume drinking,
tremor intensifies by 24 to 36 hours and is exacerbated by motor activity or stress. The tremor can be so
severe that it interferes with walking, eating, or speech. Accompanying symptoms and signs of
sympathetic hyperactivity are also apparent. The patient is increasingly anxious and easily startled by
minor stimuli and complains of insomnia and anorexia. Increased sweating, facial flushing, mydriasis,
tachycardia, and mild hypertension are noted. Most abnormalities subside in a few days, but increased
arousal and anxiety may persist for 2 weeks.
52
DISORDERED PERCEPTIONS.

Disordered perceptions accompany tremor and sympathetic hyperactivity in approximately 25% of

patients, become most pronounced at 24 to 36 hours, and clear in a few days. Often, vivid nightmares
interfere with sleep; while awake, ordinary visual, auditory, and tactile experiences become distorted and
misinterpreted. Isolated and prolonged auditory hallucinations may develop in alcoholics undergoing
withdrawal ( alcoholic hallucinosis) despite being alert, oriented, and without memory loss.
Hallucinations may persist for weeks even though other signs of ethanol withdrawal have improved and
the patient is less agitated and tremulous. In the absence of sympathetic hyperactivity, persistent auditory
hallucinations may be confused with acute schizophrenia (see Chapter 450) . However, alcoholic
hallucinosis is closely associated with ethanol withdrawal and usually subsides in weeks to months.
ALCOHOL WITHDRAWAL CONVULSIONS.
Generalized tonic-clonic seizures develop in about one third of alcoholics, most often within 12 to 24
hours after reducing or stopping drinking. Some propose that the first seizure in alcoholics may be a
consequence of ethanol toxicity. However, ethanol dependence is followed by withdrawal seizures in
animals, particularly in mice bred to have convulsions during withdrawal, thus suggesting a role for
genetic vulnerability in humans. Ethanol withdrawal seizures usually follow chronic daily drinking but
can also occur after 5 to 7 days of binge drinking. Alcoholics who have seizures during one episode of
withdrawal are likely to have them again when alcohol withdrawal is repeated. One isolated convulsion
or several seizures may occur, usually within a 6-hour period. Focal seizures are less common and
indicate a cerebral lesion, either old or new. Status epilepticus occurs in about 3% of cases, and ethanol
withdrawal accounts for about 15% of all patients evaluated for status epilepticus. Status epilepticus is a
medical emergency and requires immediate treatment with anticonvulsants (see Chapter 484) .
DELIRIUM TREMENS.
Delirium tremens, the most alarming manifestation of the ethanol withdrawal syndrome, occurs in about
5% of alcoholics. It consists of agitated arousal, global confusion and disorientation, insomnia, and vivid,
often threatening hallucinations and delusions. Signs of sympathetic hyperactivity include tremor,
mydriasis, tachycardia, fever, and intense diaphoresis. In contrast to tremulousness, disordered
perceptions, and seizures, which appear earlier after withdrawal, delirium tremens begins abruptly within
2 to 4 days of abstinence, sometimes as a surprising development in an unrecognized alcoholic admitted
to the hospital for other reasons. Patients are terrified by their hallucinations and can be combative,
destructive, and very dangerous. Episodes of delirium tremens last from 1 to 3 days and end as abruptly
as they begin. However, relapses occur, and the disorder may continue for days to weeks with
intervening periods of lucidity. When signs of sympathetic hyperactivity are absent, it may be difficult to
distinguish delirium tremens from an acute psychosis. However, the diagnosis is usually suggested by the
evolution of symptoms in a chronic alcoholic undergoing withdrawal. The differential diagnosis includes
alcoholic hypoglycemia, overdose with anticholinergic agents, intoxication with amphetamines, cocaine,
and phencyclidine, and withdrawal from other sedating drugs. Metabolic disturbances, cerebral infection,
encephalitis, meningitis, sepsis, or thyrotoxicosis should also be considered.

DIAGNOSIS
Alcohol-related problems are common in medical practice, and physicians should be alert to their diverse
clinical manifestations. End-organ complications such as ulcers, hepatitis, cirrhosis, pancreatitis,
cardiomyopathy, or peripheral neuropathy should prompt consideration of alcoholism as an underlying
cause. Recurrent trauma, particularly skeletal fractures, sleep disorders, fatigue, depression, sexual
dysfunction, and labile hypertension, should also arouse suspicion of alcoholism. Hospitalization may
precipitate the unexpected appearance of an alcohol withdrawal syndrome within several days of
admission.
Physicians should identify individuals who do not meet the criteria for alcohol dependence but who drink
alcohol at levels that pose potential health risks. Laboratory-based screening such as elevated mean
corpuscular volume (MCV) or gamma-glutamyl transpeptidase (GGTP) may help the physician in
confronting patient denial, but elevated MCV or GGTP values are neither specific for alcoholism nor
sufficiently sensitive to serve as effective screens. Simple screening questionnaires such as the CAGE
(Fig. 16-3) outperform laboratory measures in detecting excessive alcohol consumption in a variety of
clinical settings. Because advice and discussion by a concerned physician are simple, cost-effective ways
to reduce heavy drinking, routine screening for alcohol disorders is recommended, particularly in
primary care.
Complex ethanol-drug interactions must also be considered when prescribing medications to patients
who use alcohol. Ethanol potentiates the central nervous system depressant effects of narcotic, sedative,
and psychoactive drugs (see Chapter 26) , and patients for whom these agents are prescribed should be
counseled explicitly about this interaction. In contrast, the induction of microsomal enzymes in the liver
by ethanol accelerates the elimination of drugs metabolized by these enzymes (see Chapter 26) . As a
result, ethanol can decrease drug efficacy by lowering the amount of active drug available or increase
toxicity and promote unusual side effects caused by drug metabolites.
TREATMENT
ACUTE ALCOHOL INTOXICATION.
Mild to moderate ethanol intoxication requires no specific therapy. Severe acute alcohol intoxication,
defined by a depressed level of consciousness, can be fatal and is a medical emergency. Administration
of sedatives to intoxicated patients who are agitated and combative can lead to stupor, coma, and
respiratory arrest from synergistic depressant effects and should be avoided. The immediate history
should include information about the quantity of alcohol consumed, the rate of drinking, use of other
drugs including methanol and ethylene glycol, complicating medical and psychiatric disorders, and prior
alcohol abuse or alcoholism. If the patient is stuporous and unable to walk, the airway must be evaluated
immediately. Indications for endotracheal intubation and assisted ventilation include marked
hypoventilation, accumulating secretions, or coma. Complications such as hypoglycemia, meningitis,
subdural hematoma, and hepatic encephalopathy must be considered. Evidence of head trauma or focal
cerebral signs suggests urgent intracranial pathology, and a computed tomography scan should be
performed immediately. Otherwise, routine scans for alcohol intoxication are not indicated. Gastric
lavage may be performed if the obtundation is due to recent and massive alcohol consumption, but it
must be preceded by endotracheal intubation. Hemodialysis should be considered if the blood alcohol

concentration exceeds 500 mg/dL or when methanol or ethylene glycol has been ingested concurrently.
After a history and physical examination, patients with adequate vital signs and acceptable mental status
but without evidence of other disorders can be kept calm under observation until sobriety returns.
However, medical information is usually incomplete, and it is often necessary to anticipate complications
commonly associated with severe alcohol intoxication or alcoholism. Routine blood counts and
laboratory studies may uncover anemia (see Chapter 159) , hypokalemia, hypophosphatemia, and
hypomagnesemia.
53
Figure 16-3 Screening and brief intervention for alcohol problems in clinical practice.
Alcoholic hypoglycemia (see Chapter 243) can be evaluated rapidly by a bedside blood glucose
determination. If laboratory results are delayed, 12.5 to 25 g glucose should be given intravenously but
must be preceded by or accompanied by 100 mg intravenous thiamine to avoid precipitating Wernicke's
encephalopathy (see Chapter 489) . Alcoholic ketoacidosis (see Chapter 102) will be improved by
infusion of 5% dextrose in half-normal saline, also with thiamine. If the blood alcohol level is too low to
account for obtundation or if improvement does not occur as expected, it is necessary to search for other
causes of stupor and coma (see Chapter 444) , including other sedating agents.
Alcoholics stop drinking for many reasons, including serious alcohol-related medical, surgical, or
psychiatric conditions. Hence symptoms or signs of trauma, infection, liver disease, gastritis, pancreatitis,
arrhythmia, or electrolyte disturbance should be sought. One hundred milligrams of thiamine should be
given intravenously to all patients undergoing ethanol withdrawal to prevent or treat Wernicke's
encephalopathy (see Chapter 489) and should be followed by daily multivitamins.
The alarming symptoms of ethanol withdrawal are best managed by substituting another central nervous
system depressant. However, alcoholics undergoing withdrawal are very resistant to sedatives
(cross-tolerance), so large doses are often required to calm their agitation. Benzodiazepines are widely
used to manage tremulousness and disordered perceptions during ethanol withdrawal. The goal is to
suppress symptoms and produce mild sedation, and the drug dosage is adjusted to the severity of the
withdrawal reaction. Treatment includes managing delirium and autonomic stability and preventing
seizures. A sedative-hypnotic agent, typically a benzodiazepine, is prescribed as a substitute for alcohol,
and the dose is tapered over several days. Patients with mild tremulousness and few associated symptoms
usually respond to oral diazepam, 5 to 10 mg every 4 to 6 hours. The dosage is then reduced by 20 to
25% on successive days or increased if symptoms of ethanol withdrawal return. beta-Blockers are useful
ancillary therapy; they attenuate the symptoms of autonomic hypersensitivity but are not anticonvulsants

and do not appear to reduce delirium. Detoxification can be carried out with close monitoring in an
outpatient setting in socially stable patients with mild withdrawal. If withdrawal is more severe or
accompanied by significant medical, surgical, or psychiatric illness or the patient is in an unstable social
setting, inpatient detoxification may be needed. In such instances, benzodiazepines such as diazepam
(Valium), chlordiazepoxide (Librium), oxazepam (Serax), or lorazepam (Ativan) are administered orally
or parenterally in doses sufficient to keep the patient calm. Benzodiazepines should not be given
intramuscularly because of inconsistent absorption. Patients may require hourly medication at doses that
would be fatal in non-tolerant individuals. The first several days of severe alcohol withdrawal may
require intravenous administration of total daily diazepam doses exceeding 400 mg (or the equivalent of
other benzodiazepines) to achieve mild sedation. Multivitamin and thiamine supplementation should be
continued, as should meticulous attention to electrolyte status. The benzodiazepine dosage can then be
tapered by approximately 20 to 25% on successive days, with an increase in dosage if withdrawal
symptoms recur. Once the symptoms of ethanol withdrawal are suppressed, it is necessary to avoid
oversedation and the danger of respiratory depression by carefully titrating the dose of diazepam to just
keep the patient calm.
Alcohol withdrawal seizures can often be managed with intravenous benzodiazepines such as diazepam
or lorazepam. Phenytoin does not prevent seizures during withdrawal. Management of status epilepticus
is the same as in other situations (see Chapter 484) . Alcoholics are at increased risk for head trauma and
central nervous system infection; studies to exclude these more serious diagnostic possibilities should be
performed when seizures occurring in the setting of withdrawal display focal features or are accompanied
by a prolonged post-ictal state or when status epilepticus intervenes. Long-term anticonvulsant therapy is
not indicated for typical alcohol withdrawal seizures.
Delirium tremens requires hospitalization and vigorous management in an intensive care setting.
Mortality has reached 15% in the past, primarily because of injuries or associated medical disorders
complicated by hyperthermia and dehydration. Volume depletion accompanying delirium tremens may
cause circulatory collapse, and fluid losses can require replacement of 4 to 10 L in the first day. The goal
of treatment is to control behavior and suppress symptoms without danger to the patient. Five to 10 mg
or more of diazepam is given intravenously every 5 to 15 minutes until the patient is calm, and
maintenance therapy is continued every 1 to 4 hours, as needed. Initially, as much as 200 mg of
diazepam may be required before the agitation subsides. Seizures are unusual in patients with delirium
tremens and should be evaluated promptly because of the possibility of meningitis or other disorders.
Coexisting hepatic and cardiac disease may complicate fluid management, and the possibility that
sedative agents may precipitate hepatic encephalopathy should be kept in mind.
54
ALCOHOLISM AND ALCOHOL ABUSE.
Alcoholics and alcohol abusers come to medical attention because of alcohol-related medical or
psychiatric conditions, by referral from social service or criminal justice agencies, or through screening
in clinical practice. Family members provide valuable collateral history. Physicians should confront
alcoholics in a firm but non-judgmental fashion, educate them about health risks, and assess their
motivation to stop drinking (see Fig. 16-3) . Heavy drinkers should be counseled to reduce consumption.
It is valuable to establish a contract with the patient to decrease drinking and return for follow-up

assessments. Alcoholics should be referred to a rehabilitation program but may first require inpatient
detoxification. Diverse psychosocial interventions have been tested; all are equally effective, and a
successful outcome is related more to interested personal intervention than to psychotherapy matched to
the patient's condition. Medication may be a useful adjunct in some instances. Disulfiram can be helpful
in highly selected patients. The opiate antagonist naltrexone (ReVia), the only other agent currently
approved by the Food and Drug Administration (FDA) for the treatment of alcoholism, appears to
decrease the relapse rate in abstinent alcoholics. Acamprosate, a drug not yet approved by the FDA, has
also shown promising results in clinical trials.
Intervention is more effective earlier in the course of the illness, before the onset of associated medical
disorders. Alcoholics who continue to drink shorten their lifespans by at least 15 years. Many
alcohol-related medical complications such as ulcer disease, acute pancreatitis, hepatitis, myopathy, and
neuropathy stabilize or regress with continued abstinence. Others such as cirrhosis with portal
hypertension, Wernicke-Korsakoff syndrome, or dilated cardiomyopathy frequently cause permanent
disability or death. About half of socially stable, middle-class alcoholics remain sober for at least a year
after rehabilitation. Alcoholics Anonymous and Al-Anon provide low-cost support for alcoholics and
their families in virtually all communities in the United States.
Litten RZ, Allen JP: Medications for alcohol, illicit drug, and tobacco dependence: An update of research
findings. J Subst Abuse Treat 16:105, 1999. New approaches to alcohol dependence.
O'Connor PG, Schottenfeld RS: Patients with alcohol problems. N Engl J Med 338:592-602, 1998. A
useful review of the diagnosis and treatment of alcohol-related problems.
Saitz R, O'Malley SS: Pharmacotherapies for alcohol abuse: Withdrawal and treatment. Med Clin North
Am 81:881, 1997. A helpful review of the pathophysiology, treatment goals, and medication options for
alcohol withdrawal and dependence.
Schorling JB, Buchsbaum DG: Screening for alcohol and drug abuse. Med Clin North Am 81:845, 1997.
An excellent summary of the rationale for alcohol screening in primary care, use of screening
questionnaires such as the CAGE, and practical guidelines for brief physician intervention.
US Department of Health and Human Services: Ninth Special Report to the US Congress on Alcohol and
Health. Rockville, MD, National Institute on Alcohol Abuse and Alcoholism, 1996. A comprehensive
discussion of the major biomedical and socioeconomic problems of alcoholism and alcohol abuse.


Figure 16-3 Screening and brief intervention for alcohol problems in clinical practice.
http://home.mdconsult.com/das/book/body/0/882/I7.fig (1 of 2) [21/04/2000 23:07:42]


Bookmark URL: /das/book/view/882/I7.fig/top
http://home.mdconsult.com/das/book/body/0/882/I7.fig (2 of 2) [21/04/2000 23:07:42]


http://home.mdconsult.com/das/book/body/0/882/I6.fig [21/04/2000 23:08:13]

Chapter 15 - IMMUNIZATION
Walter A. Orenstein
Immunization is one of the most cost-effective means of preventing morbidity and mortality from infectious diseases. Routine
immunization, particularly of children, has resulted in decreases of 90% or more in reported cases of measles, mumps, rubella,
congenital rubella syndrome, polio, tetanus, diphtheria, and pertussis. In many circumstances, immunization not only prevents morbidity
and mortality but also, in the long run, reduces health care costs.
GENERAL CHARACTERISTICS OF IMMUNIZATIONS

Immunization protects against disease or the sequelae of disease through the administration of an immunobiologic: vaccines, toxoids,
immune globulin preparations, and antitoxins. Protection induced by immunization can be active or passive.
ACTIVE IMMUNIZATION.
Administering a vaccine or toxoid causes the body to produce an immune response against the infectious agent or its toxins. Vaccines
consist of suspensions of live (usually attenuated) or inactivated microorganisms or fractions thereof. Toxoids are modified bacterial
toxins that retain immunogenic properties but lack toxicity. Active immunization generally results in long-term immunity, although the
onset of protection may be delayed because it takes time for the body to respond. With live attenuated vaccines, small quantities of living
organisms multiply within the recipient until an immune response cuts off replication. In contrast, inactivated vaccines and toxoids
contain large quantities of antigen. In the majority of recipients, a single dose of a live vaccine generally induces an immune response
that closely parallels natural infection and induces long-term immunity. Killed vaccines, in contrast, often require multiple doses.
PASSIVE IMMUNIZATION.
Passive immunization using immune globulins or antitoxins delivers pre-formed antibodies to provide temporary immunity. Immune
globulins obtained from human blood may contain antibodies to a variety of agents, depending on the pool of human plasma from which
they are prepared. Specific immune globulins are made from the plasma of donors with high levels of antibodies to specific antigens,
such as tetanus immune globulin. Most immune globulins must be injected intramuscularly. Antitoxins are solutions of antibodies
derived from animals immunized
41
with specific antigens (e.g., diphtheria antitoxin). Passive immunization is usually indicated to protect individuals immediately before
anticipated exposure or shortly after known or suspected exposure to an infectious agent (Table 15-1) , when active immunization either
is not possible or has not been adequate.
ROUTE AND TIMING OF VACCINATION.
Each immunobiologic has a preferred site and route of administration. In adults, vaccines containing adjuvants should be injected
intramuscularly, preferably in the deltoid muscle. For men, a 1-inch needle is adequate, whereas for women, recommended needle
lengths vary from 5/8 inch for women weighing less than 60 kg to 1 inch for those weighing 60 to 90 kg and 1.5 inches for women
heavier than 90 kg. Use of the buttocks is discouraged except when large volumes are required both because of the potential for damage
to the sciatic nerve and because of diminished immune response to some vaccines, such as hepatitis B. Subcutaneous vaccines are also
usually administered in the deltoid area, and intradermal vaccines are usually given on the volar surface of the forearm. In general,
inactivated vaccines and toxoids can be given simultaneously at different sites. With vaccines that frequently cause side effects, such as
cholera and inactivated typhoid vaccines, it may be best to separate administration by at least a week. With the exception of cholera and
yellow fever vaccines, which should ideally be administered at least 3 weeks apart, live and inactivated vaccines can be administered at
the same time. For example, measles, mumps, and rubella (MMR) vaccine can be administered with oral polio vaccine (OPV). However,
with the exception of MMR and OPV, which can be administered at any interval, live vaccines not delivered on the same day should be
separated by at least 1 month because immune globulin may also interfere with the take of live vaccines other than OPV; ideally, such
vaccines should be administered at least 2 weeks prior or 3 to 11 months after immune globulin.

ADVERSE REACTIONS.
Hypersensitivity to vaccine components such as animal proteins, antibiotics, preservatives, and stabilizers can lead to local and systemic
reactions ranging from mild to severe. The egg protein contained in vaccines grown in chicken eggs (influenza and yellow fever
vaccines) may cause reactions in persons allergic to eggs. In general, persons without anaphylactic-type allergies to eggs can be given
these vaccines safely, but persons with anaphylactic reactions to eggs should not generally receive these vaccines except when absolutely
necessary and then only under established protocols by physicians who are expert in such situations. Even though measles and mumps
vaccines are grown in chick embryo tissue culture, the risk of anaphylaxis even in those with severe hypersensitivity to eggs is very low,
so such persons can be vaccinated without prior testing but should be observed for at least 20 minutes and preferably 90 minutes after
immunization.
No vaccine is completely safe or completely effective. Two major groups make comprehensive, detailed recommendations regarding
immunization of adults: (1) the Task Force on Adult Immunization of the American College of Physicians (ACP) and The Infectious
Diseases Society of America (IDSA), which publishes the Guide for Adult Immunization, and (2) the Advisory Committee on
Immunization Practices of the U.S. Public Health Service. The latter group publishes its information in Morbidity and Mortality Weekly
Report. Suspected adverse events temporally related to vaccinations should be reported to the Vaccine Adverse Events Reporting System
(1-800-822-7967).
GENERAL CONSIDERATIONS.
Immunizations for adults depend on age, lifestyle, occupation, and medical conditions (Table 15-2) . All adults should have a primary
series of tetanus and diphtheria toxoids with boosters of combined toxoids (Td) every 10 years. Persons born in or after 1957 should
have evidence of immunity to measles, mumps, and rubella. Vaccination of susceptible adolescents and adults against varicella is
desirable. Pneumococcal vaccine
TABLE 15-1 -- PASSIVE IMMUNIZATIONS FOR ADULTS
DISEASE NAME OF MATERIAL COMMENTS AND USE
Tetanus Tetanus immune globulin, human Management of tetanus-prone
wounds in persons without
adequate prior active immunization
and treatment of tetanus
Cytomegalovirus Cytomegalovirus immune globulin, Prophylaxis for bone marrow and
intravenous kidney transplant recipients
Diphtheria Diphtheria antitoxin, equine Treatment of established disease,
high frequency of reactions to
serum of non-human origin
Rabies Rabies immunoglobulin, human Post-exposure prophylaxis of
animal bites
Measles Immune globulin, human Prevention or modification of
disease in contacts of cases, not for
control of epidemics
Hepatitis A Immune globulin, human Protection of household contacts,
pre-exposure prophylaxis for
travelers who need protection
before immunity can be achieved
with hepatitis A vaccine
Hepatitis B Hepatitis B immune globulin, Prophylaxis for needlestick or
human mucous membrane contact with
HBsAg-positive persons, for sexual
partners with acute hepatitis B or
hepatitis B carriers, for infants born
to mothers who are carriers of
HBsAg, for infants whose mother
or primary caregiver has acute
hepatitis B

Varicella Varicella-zoster immune globulin Persons with underlying disease

and at risk of complications from
chickenpox who have not had
varicella or varicella vaccine and
who are exposed to varicella. May
be given post-exposure to known
susceptible adults, particularly if
antibody-negative
Vaccinia Vaccinia immune globulin Treatment of eczema vaccinatum,
vaccinia necrosum, and ocular
vaccinia following vaccinia
(smallpox) vaccination
Erythroblastosis fetalis Rh immune globulin Rh-negative women who give birth
to Rh-positive infants or who abort
Hypogammaglobulinemia Immune globulin, intravenous Maintenance therapy
Idiopathic Immune globulin, intravenous Therapy for acute episodes
thrombocytopenic purpura
Botulism Trivalent A, B, and E antitoxin, Treatment of botulism
equine
Snakebite Antivenin, equine (North American Specific for North American coral
coral snake antivenin) snake, Micrurus fulvius
Crotalidae, polyvalent Effective for viper and pit viper
bites, including rattlesnakes,
copperheads, moccasins
Spider bite Antivenin, equine Specific for black widow spider,
Latrodectus mactans, and other
members of the genus
HBsAg = hepatitis B surface antigen.
42
TABLE 15-2 -- SELECTED IMMUNIZING AGENTS INDICATED FOR ADULTS *

MAJOR
IMMUNIZING CONTRAINDICATIONS
DISEASE AGENT INDICATIONS SCHEDULE AND PRECAUTIONS COMMENTS
Cholera Inactivated vaccine Meeting Two 0.5-mL Of limited
international travel doses SC or IM effectiveness; cholera
requirements or two 0.2-mL and yellow fever
doses ID 1 wk vaccine should be
to 1 mo apart; administered at least 3
booster dose wk apart
every 6 mo

Diphtheria Tetanus and diphtheria All adults Two doses IM 4 History of neurologic or
toxoids combined wk apart, 3rd severe hypersensitivity
dose 6-12 mo reaction following a
after 2nd dose previous dose
for primary
series, booster
every 10 yr; no
need to repeat if
schedule is
interrupted
Hepatitis A Inactivated hepatitis A Travelers to highly 2 doses at least Hypersensitivity to vaccine Should be considered
vaccine or intermediately 6 mo apart for components for outbreak control for
endemic countries, children and
persons 2
men who have sex adolescents in
years of age
with men; illegal communities with
drug users (injectors intermediate to high
and non-injectors), underlying
persons who work transmission
with virus-infected
primates or do
research with the
virus, persons with
chronic liver
disease, recipients
of clotting factors
Hepatitis B Inactivated virus Adolescents, health IM; three doses Pregnancy should not
vaccine care and public at 0, 1, and 6 mo be considered a
safety workers contraindication if the
potentially exposed woman is otherwise
to blood, clients and eligible. Health care
staff of institutions workers who have
for the contact with patients or
developmentally blood should be tested
disabled, 1-2 mo after
hemodialysis vaccination to
patients, sexually determine serologic
active homosexual response
men; users of illicit
injectable drugs,
recipients of clotting
factors, household
and sexual contacts
of HBV carriers,
inmates of
long-term
correctional
facilities,
heterosexuals
treated for sexually
transmitted diseases
or with multiple
sexual partners,
travelers with close
contact for 6 mo
with populations
with high
prevalence of HBV
carriage

Influenza Inactivated influenza Annual Anaphylactic

All adults 65 yr;
virus vaccine vaccination; see hypersensitivity to eggs
other adults with
annual ACIP
high-risk conditions;
recommendation
adults caring for
persons with
high-risk conditions,
including medical
personnel (see text);
women who will be
in 2nd or 3rd
trimester of
pregnancy during
influenza season
Japanese Inactivated virus Travelers to Asia Three 1-mL Persons with histories of No data exist on
encephalitis vaccine spending at least 1 doses SC on urticaria at greater risk of concurrent
mo in endemic areas days 0, 7, 30; adverse reactions to administration with
during transmission shortened vaccine; pregnancy vaccines other than
season schedule of 0, 7, DTP, drugs (e.g.,
14 days may be chloroquine,
used when mefloquine), or other
necessary. biologics
Booster doses
may be given
after 2 yr
Lyme disease Inactivated outer Primarily persons Three doses IM Duration of immunity
surface protein A who live, work, or at 0, 1 month, and need for boosters
(OspA) visit areas at high or and 12 months are unknown.
moderate risk for
Lyme disease and
who will have
frequent or
prolonged exposure
to a tick-infested
habitat during
transmission season.
Measles Live-virus vaccine All adults born after One dose Altered immunity (e.g., Persons with
1956 without sufficient for leukemia, lymphoma, anaphylactic allergies
history of live most adults; 2 generalized malignancy, to eggs may be
vaccine on or after doses at least 1 congenital vaccinated (see text).
1st birthday, mo apart immunodeficiency, Vaccine should be
physician-diagnosed indicated for immunosuppressive administered to
measles, or persons entering therapy), immune globulin persons with
detectable measles college or or other blood products asymptomatic HIV
antibody; persons medical facility within prior 3-11 mo infection and should be
born before 1957 employment, depending upon dose of considered for
can generally be traveling immune globulin or blood symptomatic HIV
considered immune abroad, or at product received, untreated patients, except those
risk of measles tuberculosis, anaphylactic with severe
during hypersensitivity to immunocompromise
outbreaks neomycin or gelatin,
pregnancy,
thrombocytopenia

Meningococcal Polysaccharide vaccine Terminal One dose Consider revaccination

disease containing tetravalent complement after 3-5 yr for adults
A, C, W135, and Y component at increased risk of
deficiencies, disease
anatomic or
functional asplenia,
travelers who will
live in areas with
hyperendemic or
epidemic diseases;
may be useful
during localized
outbreaks
Mumps Live-virus vaccine All adults born after One dose Altered immunity (e.g., Although persons born
1956 without leukemia, lymphoma, before 1957 are
history of live generalized malignancy, generally immune,
vaccine on or after congenital vaccine can be given to
1st birthday, immunodeficiency, adults of all ages and
physician-diagnosed immunosuppressive may be particularly
mumps, or therapy), immune globulin indicated for
detectable mumps or other blood products post-pubertal males
antibody; persons within prior 3-11 mo, who are thought to be
born before 1957 anaphylactic susceptible. Persons
can generally be hypersensitivity to with anaphylactic
considered immune neomycin or gelatin, allergies to eggs may
pregnancy, be vaccinated.
thrombocytopenia if
administered with measles
vaccine
Pneumococcal 23-valent Adults with One dose IM or
disease polysaccharide vaccine cardiovascular SC. A 2nd dose
disease, pulmonary should be
disease, diabetes considered 5 or
mellitus, more yr later for
alcoholism, adults at high
cirrhosis, risk of disease
cerebrospinal fluid (e.g., asplenic
leaks, splenic patients), as
dysfunction or well as those
anatomic asplenia, who lose
Hodgkin's disease, antibody rapidly
lymphoma, multiple (e.g., nephrotic
myeloma, chronic syndrome, renal
renal failure, failure,
nephrotic syndrome, transplant
immunosuppression, recipients).
HIV infection; Revaccinate
high-risk adults who
populations such as received a first
certain Native dose when <65
Americans and all yr who are now
adults 65 yr 65 years and
who received
their vaccine at
least 5 years
earlier.

Poliomyelitis IPV (inactivated), OPV Certain adults who For OPV: immune deficiency Adults who have not
(live attenuated) are at greater risk of unvaccinated diseases, patients with been adequately
exposure to wild adults, IPV is altered immune status (e.g., immunized against
poliovirus than the preferred: two leukemia), household polio are at a very
general population, doses SC 4 wk contacts of small risk of polio
including travelers apart and a 3rd immunodeficient patients, when their children are
to countries where dose 6-12 mo household contacts with a vaccinated with OPV.
polio is epidemic or after the 2nd; if family history of The child can be
endemic, members less than 4 wk immunodeficiency until the vaccinated with OPV
of community or available before immune status of regardless of the
specific populations protection is individuals is established. immune status of the
groups with disease needed, a single On theoretic grounds, parents. An acceptable
caused by wild dose of OPV or pregnant women should alternative, provided
poliovirus, IPV. For not receive IPV or OPV. that the full
laboratory workers incompletely However, if immediate immunization of the
handling specimens immunized protection is needed, IPV child is not
that may contain adults, complete or OPV can be used compromised, is to
poliovirus, health primary series vaccinate the parents
care workers in with either first with IPV or use an
close contact with vaccine; all-IPV schedule for
patients who may be primary series the child
excreting wild consists of three
poliovirus, doses of IPV or
unvaccinated adults OPV; no need to
whose children will restart
receive OPV interrupted
series. A single
dose of OPV or
IPV can be
given to adults
who previously
completed a
primary series
Rabies Inactivated vaccine, High-risk persons, Pre-exposure History of severe Further doses needed
HDCV, PCEC, or RVA including animal prophylaxis: hypersensitivity reaction following exposure. If
handlers, selected three doses of to be given
laboratory and field 1.0 mL IM for concurrently with
workers, and HDCV, PCEC, chloroquine, only IM
persons traveling for or RVA on days route should be used
0, 7, and 21 or
1 mo to areas
28. For HDCV
with high risk of
only, three
rabies
doses of 0.1 mL
ID on days 0, 7,
and 21 or 28
Rubella Live-virus vaccine Adults, particularly 1 dose SC Pregnancy, altered Women should be
women of immunity (e.g., leukemia, counseled to avoid
childbearing age, lymphoma, generalized pregnancy for 3 mo
who lack history of malignancy, congenital
rubella vaccine and immunodeficiency,
detectable immunosuppressive
rubella-specific therapy), immune globulin
antibodies in serum; or other blood products
both males and within the 3-11 mo prior to
females in vaccination, anaphylactic
institutions where hypersensitivity to
rubella outbreaks neomycin. Administration
may occur, such as of blood products should

hospitals, the not contraindicate
military, and post-partum vaccination;
colleges. Persons thrombocytopenia if
born prior to 1957, administered with measles
except women who vaccine
can become
pregnant, can
generally be
considered immune
Tetanus Tetanus and diphtheria All adults Three doses IM History of neurologic or Special
toxoids combined needed for severe hypersensitivity recommendations for
primary series: reaction following a wound treatment (see
two doses 4 wk previous dose text). Persons with
apart, 3rd dose GBS within the 6 wk
6-12 mo after after immunization,
2nd dose, particularly adults who
booster every 10 received a prior
yr; no need to primary series, should
repeat if probably not be
schedule is revaccinated in most
interrupted circumstances
Typhoid fever Heat-phenol-inactivated Travelers to areas Inactivated Severe local or systemic Efficacy only 50-77%.
vaccine, Vi capsular where the risk of vaccine: two reaction to a prior dose. Food and water
polysaccharide vaccine, prolonged exposure 0.5-mL doses Ty21A vaccine should not precautions essential.
live attenuated Ty21a to contaminated SC 4 or more be administered to persons Ty21a and Vi
oral vaccine food and water is wk apart, with altered immunity or polysaccharide
high; may be boosters of 0.5 those receiving vaccines preferred over
considered for mL SC or 0.1 antimicrobial agents heat-phenol-inactivated
family and intimate mL ID every 3 vaccine
contacts of carriers yr; Vi
and laboratory polysaccharide
workers who work vaccine: one
with Salmonella dose IM 0.5 mL,
typhi boosters every 2
yr; Oral
vaccine: 4 doses
on alternate
days, boosters
every 5 yr
Varicella Attentuated varicella Persons, including 2 0.5-mL SC Immunocompromise, Adults with a history
vaccine, OKA strain health care workers, doses 4-8 wk pregnancy, allergy to of prior varicella
who have contact apart for persons vaccine components; avoid should be considered
with patients, at salicylate use for 6 wk after immune. Vaccine virus
13 yr of age
high risk of vaccination has rarely been
complications from transmitted to contacts
varicella; persons from healthy vaccinees
who work with in whom rash
children (e.g., developed. Women
teachers), persons in who receive vaccine
institutions that may should not become
have outbreaks (e.g., pregnant for 1 mo
colleges),
non-pregnant
women of
childbearing age,
international
travelers; desirable
for other susceptible

adolescents and
adults
Yellow fever Live attenuated virus Persons living or One dose; Immunocompromised
(17D strain) traveling in areas booster every 10 persons; history of
where yellow fever yr anaphylactic allergies to
exists eggs; pregnancy on
theoretic grounds, although
may be given if risk is high
SC = subcutaneously; IM = intramuscularly; ID = intradermally; HBV = hepatitis B virus; ACIP = Advisory Committee on
Immunization Practices; DTP = diphtheria-tetanus-pertussis; HIV = human immunodeficiency virus; IPV = inactivated polio vaccine;
OPV = live-virus trivalent oral polio vaccine; HDCV = human diploid cell vaccine for rabies; RVA = rabies vaccine absorbed; PCEC
= purified chick embryo cell culture rabies vaccine; GBS = Guillain-Barre syndrome.
*See the text and package inserts for further details, particularly regarding indications, dosage, mode of administration, side effects, and adverse reactions and
contraindications.
and annual vaccination against influenza are indicated for all adults 65 years and older and younger adults with certain medical
conditions that place them at high risk of complications. Health care workers exposed to blood or blood products should receive hepatitis
B vaccine. Those caring for patients at high risk of complications from influenza should receive annual vaccination. Health care workers
likely to come in contact with persons transmitting measles, mumps, rubella, or varicella should be immune to those diseases.
IMMUNOCOMPROMISE.
Patients with conditions that compromise their immune systems should not receive live attenuated vaccines. Such patients include those
with immunodeficiency diseases, leukemia, lymphoma, and generalized malignancy and those who are immunosuppressed from therapy
with corticosteroids, alkylating agents, antimetabolites, and radiation. An exception is infection with human immunodeficiency virus
(HIV). Asymptomatic patients should receive MMR vaccine. MMR should be considered for symptomatic patients with HIV; however,
severely immunocompromised persons should not be vaccinated. Because of the availability of enhanced-potency inactivated polio
vaccine (IPV), all patients known to be infected with HIV should receive IPV instead of OPV. Patients with leukemia in remission who
have not been receiving any chemotherapy for at least 3 months may receive live-virus vaccines. Short-course therapy (<2 weeks) with
corticosteroids, alternate-day regimens with low to moderate doses of short-acting corticosteroids, and topical applications or tendon
injections are not ordinarily contraindications to the administration of live vaccines.
Immunocompromised patients can receive inactivated vaccines and toxoids, although the efficacy of such preparations may be
diminished. Patients with known HIV infection should receive pneumococcal vaccine and annual influenza vaccination.
PREGNANCY.
In general, live vaccines should not be given to pregnant women because of the theoretic concern that such vaccines could adversely
affect the fetus. No significant adverse events attributable to vaccination of pregnant women with MMR or varicella have been
documented; nevertheless, pregnant women should not receive MMR, and women who do receive MMR or varicella should wait 3
months or 1 month, respectively, before becoming pregnant. Polio and yellow fever vaccines should not usually be given to pregnant
women unless the risk of disease is substantial. Td vaccination is especially indicated for pregnant females who are not appropriately
vaccinated to prevent neonatal tetanus in their infants. Vaccination is best performed after the first trimester. All pregnant women should
be screened for hepatitis B surface antigen (HBsAg). Offspring of HBsAg carrier mothers should receive hepatitis B vaccine and
hepatitis B immune globulin. Women who will be in the second or third trimester of pregnancy during the influenza season should
receive influenza vaccine.
INDIVIDUAL IMMUNOBIOLOGICS
Tetanus and Diphtheria
Tetanus (see Chapter 337) toxoid is one of the most effective immunizations, with over 95% protection after a primary series. The
adsorbed is preferred over the fluid preparation because it induces protective levels of antitoxin that persist longer after fewer doses. A
primary series consists of three doses. In persons aged 7 years or older, it should always be used in combination with diphtheria (see
Chapter 333) toxoid (Td), which is more than 85% effective in preventing disease. Doses need not be repeated if the schedule is
interrupted. Boosters are recommended every 10 years. An easy way to remember is to schedule immunization at the middle of each

decade (e.g., 25 years, 35 years, etc.). The ACP/IDSA Task Force on Adult Immunization has recently suggested that a single Td booster
at age 50 may be sufficient to maintain protective antibody levels in older adults who have received a primary series with boosters as a
teenager and young adult.
After a wound, persons of unknown immunization status or those who have received fewer than three doses of tetanus toxoid should
receive a dose of Td regardless of the severity of the wound. Td is also indicated for those who have previously received three or more
doses if more than 10 years has elapsed since the last dose, in the case of clean, minor wounds, and if more than 5 years has elapsed for
all other wounds. Tetanus immune globulin should be
46
administered simultaneously at a separate site to persons who have not received at least three doses of toxoid and who have wounds that
are not clean and minor. Most reactions to Td consist of local inflammation and low-grade fever. However, Guillain-Barre syndrome and
brachial neuritis have very rarely been associated with tetanus toxoid.
Measles
Measles (see Chapter 381) immunization is recommended for all persons born in or after 1957 who lack evidence of prior
physician-diagnosed measles or laboratory evidence of immunity or appropriate vaccination. Prior to 1989, appropriate vaccination
consisted of a single dose of live vaccine administered on or after the first birthday. Now, a routine two-dose schedule is recommended:
the first dose, which is 93 to 98% effective, at 12 to 15 months of age and the second dose at entry to primary school. By 2001, all
children from kindergarten through the 12th grade should have a second dose. Most adults are considered to have been appropriately
vaccinated if they received one dose of vaccine administered on or after their first birthday. Some adults, however, who are at increased
risk of measles (health care workers with direct patient contact, students in college, international travelers) should receive a second dose
of vaccine unless they have documentation of prior physician-diagnosed measles or serologic evidence of immunity. Persons embarking
on foreign travel should ideally have received two doses or have other evidence of measles immunity. Persons born before 1957 are
usually immune as a result of natural infection and do not require vaccination, although vaccination is not contraindicated if they are
believed to be susceptible.
During outbreaks of measles in institutions, all persons at risk who have not received two doses or who lack other evidence of measles
immunity should be vaccinated. Measles vaccine is usually administered along with mumps and rubella vaccine (MMR) to ensure
immunity against all three diseases. Individuals already immune to one or more of the components, however, may receive MMR without
harm.
Measles vaccine is contraindicated for pregnant women on theoretic grounds, for persons with moderate to severe acute febrile illnesses,
and for persons with altered immunocompetence, except those with HIV infection who are not severely immunocompromised. Patients
with anaphylactic reactions to eggs can be vaccinated without prior skin testing.
In approximately 5 to 15% of susceptible recipients of measles vaccine, temperatures of 39.4 C or higher develop between 5 and 12
days after vaccination and last 1 to 2 days. Transient rashes develop in about 5%. Thrombocytopenic purpura has been reported rarely
after MMR. The overall rate of reactions after the second dose of a measles-containing vaccine is substantially lower than after the first
dose. Encephalopathy or encephalitis following measles vaccination has been reported at a rate lower than the background or expected
rate.
Rubella
Rubella (see Chapter 382) vaccine is indicated for susceptible adults born in 1957 or later and for susceptible women of any age who are
considering becoming pregnant. Persons without a prior history of vaccination on or after the first birthday or laboratory evidence of
immunity should be considered susceptible. A single dose of vaccine is 95% or more effective. Many persons receive two doses of
rubella vaccine via the two-dose schedule of MMR.
Follow-up of 305 susceptible women who received rubella vaccines within 3 months of the estimated date of conception has failed to
reveal any evidence of defects compatible with congenital rubella syndrome in their offspring. Nevertheless, vaccine is contraindicated
in pregnant women on theoretic grounds, and conception should be delayed for 3 months after rubella vaccination.
Reactions occur only in susceptible persons. Arthralgia, usually of the small peripheral joints, develops in up to 40% of susceptible
adults, and frank arthritis develops in 10 to 20%. Joint symptoms usually begin 1 to 3 weeks following vaccination and persist for 1 day
to 3 weeks. Very rarely have chronic recurrent or persistent joint symptoms developed following vaccination, but controlled studies have
shown that the incidence of these events in vaccinees is similar to that of non-vaccinees. Other infrequent adverse events include
transient peripheral neuritis and pain in the arms and legs. Thrombocytopenic purpura has been reported rarely when rubella vaccine is
administered as MMR. Rubella vaccine is contraindicated for persons with moderate to severe acute febrile illnesses and for persons
with reduced immunocompetence. When given with measles vaccine, it may be administered to those with asymptomatic HIV infection

and considered for those with symptomatic infection without severe immunocompromise. Rubella vaccine is grown in human diploid
cells and can be administered without problems to persons with allergy to eggs.
Mumps
Mumps (see Chapter 384) vaccine is indicated for all persons, especially susceptible males, without a prior history of vaccination on or
after the first birthday, physician-diagnosed mumps, or laboratory evidence of immunity. Most persons born prior to 1957 can be
considered immune as a result of natural infection, although vaccination is not contraindicated if such persons are thought to be
susceptible. In clinical trials, a single dose of vaccine has induced seroconversion in more than 90% of recipients.
Adverse events following mumps vaccine are uncommon--fever, parotitis, and allergic manifestations. Thrombocytopenic purpura has
been reported rarely in those administered MMR. Mumps vaccine is contraindicated for pregnant women on theoretic grounds, for
persons with moderate to severe acute febrile illnesses, and for persons with altered immunocompetence. When combined with measles
vaccine, it may be given to those with asymptomatic HIV infection and considered for those with symptomatic infection if they are not
severely immunocompromised. Patients with anaphylactic reactions to eggs can be vaccinated without skin testing (see Measles earlier).
Varicella
A live attenuated varicella vaccine (Oka strain) was licensed in March 1995. The vaccine protects 70 to 90% of recipients against any
disease and more than 95% of recipients against severe disease. Breakthrough infections in persons who have previously seroconverted
have been reported in 2 to 4% per year following vaccination with the licensed product. Such breakthroughs are typically mild and
average fewer than 50 lesions as compared with several hundred lesions in unvaccinated persons with varicella. Breakthrough illnesses
do not appear to increase in incidence or severity with increasing time since vaccination, a finding compatible with long-term protection
following initial vaccination. Persons 13 years or older require two doses at least 4 weeks apart to achieve seroconversion rates of
approximately 99%, a rate comparable to that in younger children after one dose.
The most common side effect is soreness at the injection site, which is reported in 25 to 35% of recipients 13 years or older.
Varicella-like rashes at the injection site (median of two lesions) have been reported in 3% of recipients in this age group after the first
dose and in 1% after the second dose. Non-localized rashes with a median of five lesions have been reported in 5.5% of recipients after
the first dose and in 0.9% after the second dose. The incidence of herpes zoster (shingles) is substantially lower than would be expected
after natural varicella (see Chapter 383) . Although more severe events occurring in temporal relation to the vaccine have been reported
very rarely, a causal relationship has not been established. Transmission of vaccine virus to a contact is extremely rare and appears to
take place only with vaccinees in whom a varicella-like rash has developed.
Varicella vaccine is indicated routinely for all children. Persons with a prior history of varicella disease can be considered immune and
do not need vaccination. Whereas a negative or unknown history of disease is predictive of susceptibility in children, many adults with
such histories are immune. Serologic screening of adults in some situations may be cost-effective, provided that identified susceptible
adults are vaccinated. The vaccine is contraindicated in the immunocompromised, those with anaphylactic allergies to vaccine
components, and pregnant women. Varicella vaccine is more temperature sensitive than other vaccines used in the United States. It must
be stored frozen at -15 C or colder to retain
47
potency. It should be discarded if not used within 30 minutes of reconstitution.
Hepatitis B
Hepatitis B (see Chapter 149) vaccine is the first vaccine that can prevent cancer (an estimated 800 persons per year in the United States
die of hepatitis B-related liver cancer; many times more do so in the developing world). It can also prevent acute and chronic
complications of hepatitis B, including an estimated 4000 deaths annually from cirrhosis and 250 deaths annually from fulminant hepatic
disease in the United States. The original hepatitis vaccine in the United States consisted of purified, inactivated, alum-adsorbed, 22-nm
HBsAg particles obtained from human plasma. Currently produced vaccines are derived from inserting the gene for HBsAg into
Saccharomyces cerevisiae. Hepatitis B vaccine, the first licensed vaccine made by using recombinant techniques, produces adequate
antibody responses in more than 90% of normal adults and more than 95% of normal infants, children, and adolescents when
administered in a three-dose series. The dosage depends on the product, the age group, and the underlying clinical condition and can be
determined by consulting the package insert. The duration of vaccine-conferred immunity is not known, although follow-up of vaccinees
for 11 years indicates persistence of protection against clinically significant infections (i.e., detectable viremia and clinical disease).
Booster doses are not currently recommended. Vaccine must be injected intramuscularly, preferably in the deltoid.
Because strategies targeting hepatitis B vaccine use only to high-risk populations have not had a significant impact on hepatitis B
incidence, universal vaccination is now recommended. Universal infant vaccination is presently recommended for all populations. All

adolescents who have not been previously vaccinated should be immunized. Universal screening for HBsAg is recommended for all
pregnant women, with administration of three doses of vaccine and one dose of hepatitis B immune globulin recommended for infants of
carrier mothers.
The major side effect is soreness at the injection site. Alopecia, which is usually reversible, has been reported very rarely. Among adults
receiving plasma-derived vaccine, the risk of Guillain-Barre syndrome is increased after the first dose, but the overall increase, if real, is
very small and is outweighed by the substantial benefits of vaccination. Recombinant vaccine, which is now the standard, does not
appear to increase the risk of Guillain-Barre syndrome. There is no risk of acquiring HIV infection from either vaccine.
Influenza
Annual influenza (see Chapter 379) vaccination is indicated for adults at high risk of complications from the disease: persons with
chronic cardiopulmonary disorders, residents of nursing homes or other chronic care facilities, persons aged 65 or older, patients with
other chronic diseases (such as diabetes mellitus, kidney dysfunction, hemoglobinopathies, and immunosuppression) who have required
regular medical follow-up or hospitalization in the prior year, and children receiving long-term aspirin therapy. Women who will be in
the second or third trimester of pregnancy during the influenza season (usually late December through mid-March) should also be
vaccinated. In addition, transmission of influenza to high-risk patients can be reduced by annually vaccinating health care workers and
household contacts of high-risk patients.
The efficacy of influenza vaccine varies with the host's condition and the degree to which antigens in the vaccine match viruses in
circulation the following season. Current vaccines contain whole or split inactivated viruses of three major antigenic types--A (H3N2), A
(H1N1), and B. Provided that the match is good, vaccine efficacy is usually 70 to 90% in normal healthy young adults. Efficacy is
substantially lower, often between 20 and 40%, in the institutionalized elderly; nevertheless, it appears to be 60 to 80% protective against
pneumonia and death. Ideally, vaccines should be administered between October and mid-November of each year, although earlier in the
autumn suffices if circumstances require.
Persons with anaphylactic allergies to eggs should not be immunized. The most common side effect is soreness at the injection site.
Fever, malaise, and myalgia may begin 6 to 12 hours after vaccination and persist for 1 to 2 days, although such reactions are most
common in children exposed to vaccine for the first time. Severe allergic reactions are rare. If current influenza vaccines cause
Guillain-Barre syndrome, it is likely to be very rare, on the order of 1 case per million doses. A live attenuated trivalent influenza
vaccine for intranasal administration may soon become available.
Pneumococcal Vaccine
Pneumococcal vaccine consists of purified polysaccharide capsular antigens from the 23 types of Streptococcus pneumoniae that are
responsible for 85 to 90% of the bacteremic disease in the United States (see Chapter 319) . Most adults, including the elderly and
patients with alcoholic cirrhosis and diabetes mellitus, have a two-fold or greater rise in type-specific antibodies within 2 to 3 weeks of
vaccination. Although the serologic response is generally acceptable, estimates of vaccine efficacy in preventing disease vary widely.
Efficacy may be lower in some patients, such as those with alcoholic cirrhosis or Hodgkin's disease. There is good evidence that
vaccination is approximately 60% effective against bacteremic pneumococcal disease, which accounts for an estimated 50,000 cases
annually. However, evidence regarding efficacy against pneumonia in high-risk populations is not clear. Regardless, the preponderance
of information supports the use of pneumococcal vaccine in high-risk populations, including all persons older than 65 years.
Immunity may decrease 5 or more years after initial vaccination; boosters should therefore be considered at that time for adults at highest
risk of disease, such as asplenic patients, as well as for those who lose antibody rapidly, such as patients with nephrotic syndrome or
renal failure. Persons older than 65 years who received a dose more than 5 years earlier when they were younger than 65 years should be
revaccinated.
Local reactions are frequent. Fewer than 1% of vaccinees experience severe local reactions or systemic illness such as fever and malaise.
Severe events such as anaphylaxis are rare. Because of the rarity of severe reactions in revaccinated patients, persons with indications for
vaccination but with unknown histories of prior vaccination should be vaccinated.
Special efforts should target hospitalized patients. Approximately two thirds of patients later admitted with pneumococcal disease had
been hospitalized for other reasons within the preceding 5 years.
Poliomyelitis
The last documented cases of indigenously acquired poliomyelitis (see Chapters 389 and 476) caused by wild polioviruses in the United
States were reported in 1979. All indigenous cases since 1981, approximately 8 per year, have been linked epidemiologically and/or via
laboratory tests to OPV exposure. Between 1980 and 1994, the overall risk of acquiring vaccine-associated polio was 1 case for every
2.4 million doses distributed. The risk is more than 3000 times higher for immunodeficient persons than normal recipients. Vaccine
polioviruses may spread from recipients to contacts, and cases among the latter account for more than one third of the total

vaccine-associated cases. A goal has been established to eradicate wild poliovirus from the world by the end of 2000. Between 1988,
when the goal was announced, and 1997, cases of polio reported world-wide have decreased by 85% and indigenous wild poliovirus
transmission has been eliminated from the Americas since late 1991. Because of the lower risk of wild virus exposure and the continued
occurrence of vaccine-associated polio with an all-OPV schedule, all major immunization advisory groups have changed
recommendations for routine immunization. A primary series of four doses of IPV, or two doses of IPV followed by two doses of OPV,
is an acceptable alternative for vaccinating children. OPV, even for just the first two doses, should be restricted to special situations such
as refusal of IPV, imminent international travel, and a child who is behind in immunizations and for whom an all-IPV regimen would be
a barrier to catching up.
The Advisory Committee on Immunization Practices prefers the sequential IPV/OPV schedule for public health reasons to take
advantage of the individual protection of IPV as well as the gut immunity induced by OPV, which can immunize contacts and enhance
community protection; this schedule should reduce vaccine-associated polio by 50 to 75%. With further progress in polio
48
eradication, an all-IPV schedule is envisioned prior to stopping vaccination, probably by 2001.

Adults are at increased risk of paralytic disease from receiving OPV; routine vaccination of adults is therefore not warranted given the
small risk of exposure to wild virus in the United States. The major indication for adult vaccination is travel to areas where wild
poliovirus is endemic or epidemic. For previously unvaccinated adults, IPV is indicated. Adult travelers who have histories of partial
vaccination should complete a primary series of three doses of either IPV or OPV. Adults who formerly received three doses should
receive a booster of OPV or IPV. Health care personnel who come in contact with wild viruses should be immune to polio. IPV is the
vaccine of choice in such persons to protect both the recipient and any immunocompromised persons with whom the health care worker
has contact from exposure to OPV. Parents of children to be vaccinated with OPV may elect to receive IPV before vaccinating their
child or alternatively elect to give their children all IPV.
A primary series of both OPV and IPV consists of three doses. A fourth dose is administered to children at school entry. The sequential
schedule requires four doses to obtain the full benefits of both vaccines. No serious side effects of IPV have been reported. OPV should
never be given to immunocompromised individuals or to a child living in a household with immunocompromised persons.
Hepatitis A
Two inactivated hepatitis A (see Chapter 149) vaccines are available in the United States. Seroconversion rates after a single dose of
either vaccine in persons older than 2 years exceed 95%. Antibody levels shown to be protective in animals develop in almost all
persons. The most common side effect has been tenderness and soreness at the injection site. Although rare and more serious adverse
events have been reported in temporal association with vaccination, a causal relationship has not been established.
The vaccine is indicated primarily for persons traveling to countries, primarily the developing world, with high or intermediate
endemicity for hepatitis A. In addition, children living in communities with high rates of endemic hepatitis A (anti-hepatitis A
prevalence of 30 to 40% by 5 years of age) should be vaccinated. Health care workers have not been shown to be at higher risk than the
general population for hepatitis A and do not need routine immunization. Although food handlers are not at increased risk of hepatitis A
when compared with the general population, the consequences of infection or suspected infection in this group, which can lead to
extensive public health investigations, may make vaccination cost-effective in some settings. Hepatitis A vaccine should be given to
children 2 years of age or older to control outbreaks in communities with high rates of prior infection and be considered for communities
with intermediate levels of prior infection (anti-hepatitis A seroprevalence of 10 to 25% by 5 years of age). In 1999, the Advisory
Committee on Immunization Practices voted to recommend universal vaccination of children who reside in states or counties with an
average annual incidence rate of hepatitis A between 1987 and 1997 of 20/100,000 population. Universal vaccination may also be
considered for areas with average annual incidence rates between 10 and 20/100,000. Doses vary by age and product. All schedules call
for a second dose at least 6 months after the first dose with a permissible range for one of the products as long as 18 months after the
initial dose. Vaccines are not indicated for children younger than 2 years because of the absence of adequate data on safety and efficacy.
Lyme Disease
On December 21, 1998, the Food and Drug Administration licensed LYMErix (Smith Kline Beecham Biologicals). The vaccine contains
lipidated recombinant outer-surface protein A (OspA) of Borrelia burgdorferi, the cause of Lyme disease in the United States. The
vaccine works by inducing antibodies against OspA, an antigen expressed on the surface of the spirochete in the tick vector. Expression
of OspA is either very limited or absent when spirochetes infect humans. When an infected tick feeds on blood with antibodies to OspA,
the blood kills the spirochetes in the tick gut prior to the time transmission can take place. The vaccine was found to be 49% effective
after two doses at 0 and 1 month and 76% effective after three doses at 0, 1, and 12 months in persons 15 to 70 years of age. Efficacy
was higher against asymptomatic infection, 83% and 100%, after two or three doses, respectively. Another unlicensed product as of
February 1999 containing purified OspA produced by Pasteur, Merrieux, Loynaught reported efficacies of 68% and 92% after two or

three doses, respectively, in persons 18 to 92 years of age.
The major side effect of LYMErix is local reactions (24.1% vs. 7.6% in placebo recipients). Myalgias, influenza-like illness, fever, and
chills were all significantly higher in vaccinees than placebo recipients, but their attributable risk was 1.4% or less. The vaccine is
indicated primarily for persons living or working in endemic tick-infested areas or visitors who will have substantial exposure during
transmission season. The duration of immunity and hence the need for boosters is unknown.
Meningococcal Polysaccharide Vaccine
A quadrivalent meningococcal polysaccharide vaccine containing serogroups A, C, Y, and W135 is now available. These groups account
for approximately 50% of meningococcal disease in the United States (see Chapter 329) . Serogroup A and C vaccines have had 85 to
100% efficacy in epidemic settings, whereas vaccines for the other groups have documented good immunogenicity in adults. The
duration of immunity is unknown, although protection in older children and adults probably persists for at least 3 years. Protection in
pre-school children may be shorter. Routine vaccination is not recommended in the United States because of the low risk of infection. A
single dose is indicated for high-risk persons. Vaccination may also be useful during localized epidemics of serogroups in the vaccine.
Meningococcal vaccine may be offered to travelers and persons who will live in areas with hyperendemic or epidemic disease, e.g., the
"meningitis belt" of sub-Saharan Africa stretching from Mauritania to Ethiopia.
Revaccination should be considered 2 to 3 years after primary immunization for children younger than 4 years at the initial vaccination.
Revaccination 3 to 5 years after the initial dose may also be considered for older adolescents and adults at continued risk. The major side
effects are local reactions lasting 1 to 2 days.
Rabies
Rabies (see Chapter 478) vaccine is indicated for pre-exposure prophylaxis of high-risk persons, including animal handlers, selected
laboratory and field workers, and persons traveling for more than 1 month to areas where rabies is a constant threat. The pre-exposure
regimen consists of either three 1.0-mL intramuscular injections on days 0, 7, and 21 or 28 for all rabies vaccines or, for the human
diploid cell vaccine only, three 0.1-mL intradermal injections on days 0, 7, and 21 or 28. Testing for serum antibody or a booster every 2
years is indicated for persons with continuing risk. Post-exposure treatment depends on prior exposure to vaccine (see Chapter 478) .
Human rabies immune globulin is indicated for previously unvaccinated persons who are exposed.
VACCINES INTENDED PRIMARILY FOR INTERNATIONAL TRAVELERS

Yellow Fever
Yellow fever (see Chapter 391) now occurs only in areas of South America and Africa. Vaccination with a single dose of the live
attenuated 17D strain of virus confers protection to almost all recipients for at least 10 years. Boosters are recommended every 10 years
for those at risk. Side effects are uncommon. Yellow fever vaccine should not be given to immunocompromised persons or those with
anaphylactic allergies to eggs. The vaccine is contraindicated in pregnant women on theoretic grounds, although if such women must
travel to a high-risk area, they may be vaccinated.
Typhoid Vaccine
Three types of vaccines, a live attenuated Ty21a oral vaccine, a parenteral heat-phenol-inactivated vaccine, and a capsular
polysaccharide vaccine (ViCPS), appear to be of comparable efficacy (50 to 77%). Typhoid (see Chapter 340) vaccine is indicated
primarily for travelers to areas where the risk of prolonged exposure to contaminated food and water is high. The vaccine is not
optimally effective; food and water precautions are still essential. The vaccine
49
may also be considered for family or other intimate contacts of typhoid carriers and laboratory workers who work with Salmonella typhi.
For adults and children 6 years and older, any of the vaccines may be used. For Ty21a, one enteric-coated capsule is taken every other
day for four doses. Alternatively, a single dose of the ViCPS vaccine or two doses of heat-phenol-inactivated vaccine separated by 4 or
more weeks may be given. The duration of protection with Ty21a is not known; repetition of the primary series is recommended every 5
years for persons at risk. Boosters are recommended every 2 years for the ViCPS vaccine and every 3 years for recipients of the
heat-phenol-inactivated vaccine if they continue to be at risk. The ViCPS vaccine can be given to children as young as 2 years. The
heat-phenol-inactivated vaccine can be used in children 6 months or older.
Adverse reactions are much less common after the Ty21a and the ViCPS vaccines than after the inactivated vaccine. Thus, whenever
feasible, the Ty21a and ViCPS vaccines are preferred.

Cholera
Cholera (see Chapter 344) vaccines offer only about 50% protection after completion of a primary series of two doses 1 week to 1 month
apart. Peak protection appears about 2 months after the last dose, and protection wanes by 3 to 6 months. Vaccination often results in
significant local reactions accompanied by fever. Neurologic reactions are rare. The major indication is to meet requirements imposed by
some countries for entry.
Japanese Encephalitis Vaccine
Japanese encephalitis (see Chapter 392) vaccine is primarily indicated for travelers to Asia who will spend a month or longer in endemic
areas during the transmission season, especially if travel will include rural areas. In all instances, travelers should be advised to take
personal precautions to reduce exposure to mosquito bites. The vaccine appears to be 80 to 91% effective in preventing clinical disease.
The primary series consists of three subcutaneous 1-mL doses given on days 0, 7, and 30 (see Table 10-2) . A shortened schedule given
on days 0, 7, and 14 may be used when necessary. Booster doses may be given after 2 years. Local reactions are common and occur in
about 20% of vaccinated persons, and systemic symptoms of fever, headache, chills, nausea, and abdominal pain have been noted in
about 10%. A delayed urticaria-angioedema syndrome may occur a median of 12 hours after the first dose of vaccine and up to 2 weeks
after the second dose. Vaccinees should be observed for at least 30 minutes after inoculation and, during the subsequent 10 days, should
remain in areas with ready access to medical care. The vaccine is contraindicated for pregnant women on theoretic grounds, but if such
women travel to an endemic, high-risk area, they may be vaccinated.
OTHER VACCINES
A number of other vaccines used in selected circumstances include smallpox (vaccinia) vaccine, which is used by the military and
laboratory workers who handle orthopoxviruses; BCG (bacille Calmette-Guerin) vaccine to prevent tuberculosis, which has very limited
use in the United States; anthrax vaccine, which is indicated in selected high-risk populations; and plague vaccine, which may be
considered for workers at risk and for some travelers. In addition, trivalent botulism antitoxin is available for the treatment of suspected
cases of botulism.
Although not available today, a number of vaccines are under development and may be licensed in the future. For example, extensive
field trials have occurred or are planned with pneumococcal and meningococcal conjugate vaccines, acellular pertussis vaccines in
adults, and human immunodeficiency virus. Because of the biotechnology revolution, it is likely that many more vaccines will become
available in the future.
ACP Task Force on Adult Immunization and Infectious Diseases Society of America: Guide for Adult Immunization, 3rd ed.
Philadelphia, American College of Physicians, 1994, pp 1-218. An excellent comprehensive guide covering all aspects of adult
immunization. A must for the physician who cares for adults, whether in primary, secondary, or tertiary care.
Centers for Disease Control: Update on Adult Immunization. Recommendations of The Advisory Committee on Immunization Practices
(ACIP). MMWR 40(RR-12):1, 1991. A compendium of ACIP statements on immunizations for adults, as well as valuable information on
other aspects of immunization. Somewhat dated at this point. More current ACIP statements on individual vaccines are published as
available in the Morbidity and Mortality Weekly Report, Recommendations and Reports Supplement.
Centers for Disease Control and Prevention: General recommendations on immunization: Recommendations of The Advisory
Committee on Immunization Practices (ACIP). MMWR 43(RR-1):1, 1994. A comprehensive review of vaccination schedules,
precautions, contraindications, and adverse events, as well as information about federal laws on injury compensation and record
keeping.
Centers for Disease Control and Prevention: Health Information for International Travel. Washington, DC, US Government Printing
Office, 1996-1997. A complete guide for the international traveler, including required and recommended vaccinations. Revised every 1
to 2 years.
Centers for Disease Control and Prevention: Update: Vaccine side effects, adverse reactions, contraindications, and
precautions--Recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR 45(RR-12):1, 1996. A
comprehensive evaluation of the Institute of Medicine reports on adverse events with modifications, where appropriate, in precautions
and contraindications.
Committee on Infectious Diseases, American Academy of Pediatrics: Report of the Committee on Infectious Diseases, 23rd ed. Elk
Grove Village, IL, American Academy of Pediatrics, 1997, pp 1-764. The "Red Book" is published every 2 to 3 years and addresses in a
comprehensive manner vaccination of children and adolescents, as well as other issues relating to prevention, control, and treatment of
infectious diseases.
National Immunization Program (NIP), Centers for Disease Control and Prevention. The NIP has established toll-free numbers for

answering questions from both the general public and physicians: 1-800-232-2522 (English), 1-800-232-0233 (Spanish). Inquiries can
be made to the NIP by e-mail: nipinfo@ cdc.gov or the NIP web site at www.cdc.gov/NIP.
Plotkin SA, Orenstein WA: Vaccines, 3rd ed. Philadelphia, WB Saunders, 1999, pp 1-1230. A thorough review of each of the available
vaccines, vaccine-preventable diseases, and etiologic agents. Covers such topics as vaccine production, handling, indications, adverse
events, and public health impact.


Chapter 14 - VIOLENCE AND INJURY
Elizabeth McLoughlin
DEFINITIONS
Violence in the United States is a public health emergency and can be caused by institutional and
personal actions. The root causes of violence include inequitable social and economic conditions.
Personal violence is the intentional use of physical or psychological force against another person or
against oneself that may result in injury or death. An injury is damage to tissue usually caused by
excessive energy transfer. That energy can be kinetic (causing fractures, lacerations, and contusions),
thermal (burns and scalds), electrical (electrocutions), or chemical (poisonings). The mechanism is
somewhat different for drowning and suffocation, which result when tissue is deprived of oxygen.
Injuries may be classified in many ways, primarily by type, by cause, and by intent. Type of injury
includes, for example, a fracture, laceration, or burn. Cause groupings distinguish among, for example,
injuries caused by a car crash, a bullet, poisons, or a fall. Intent categories address whether the injury was
unintentional, intentionally self-inflicted (the most severe outcome being suicide), or intentionally
inflicted by another (the most severe outcome being homicide). Violent injuries such as homicide and
suicide are positioned at the intersection of violence in general and all injuries.
EPIDEMIOLOGY
In 1995, 6% of all deaths in the United States were caused by an injury; 8% of all hospital discharges had
a first listed diagnosis of injury and 37% of all emergency department visits were for injuries. Figure
14-1 presents the burden of injury at four levels of severity. Suicide and homicide account for about 35%
of deaths from injury (Table 14-1) .
The leading external causes of injury death, regardless of intent, are motor vehicle traffic crashes,
firearms, poisoning (primarily by drug overdose), suffocation (which includes suicide by hanging), falls,
drownings, and fire. Although firearms slightly exceed motor vehicles as the primary mechanism for
injury death for males, the age-related profiles are remarkably alike across the age range (Fig. 14-2) .
Among firearm deaths, the peak in young men is primarily homicide, whereas the peak in older men is
primarily suicide. Motor vehicle deaths exceed firearm deaths in the young and the very elderly, age
groups that are vulnerable to pedestrian as well as vehicle occupant deaths.
The external causes of fatal and non-fatal injury differ dramatically. In states with databases where one
can compare the causes of fatal and non-fatal injury, for example, in California in 1995, falls accounted
for fewer than 10% of the deaths but over one third of the hospitalizations for injury. In comparison,
motor vehicles and firearms together accounted for more than half of the deaths but fewer than 20% of
the hospitalizations.

Violence in families is an increasingly recognized and complex problem. Children's Protective Services
determined that over 1 million children were victims of abuse or neglect in 1995. The National Family
Violence Surveys estimate that 116 per 1000 women experience a violent act and 34 per 1000 experience
severe violence at the hands of an intimate partner. No estimates of the prevalence of elder abuse have
been made, but the problem is serious and may be increasing as the population ages.
Alcohol consumption is a major risk factor for all types of injury. In 1987, the estimated percentage of
unintentional injury deaths associated with alcohol were 42% for motor vehicles, 20% for other road
vehicles, 20% for water transport, 16% for air transport, 35% for falls, 45% for fires, and 38% for
drowning. For suicide and homicide, the percentages of deaths associated with alcohol are estimated to
be 28% and 46%, respectively.
SECULAR TRENDS IN INJURIES

Injuries are preventable. Data on injury deaths from 1910 to 1995 show a significant decrease in other
(i.e., non-motor vehicle) unintentional injury death rates (Fig. 14-3) (Figure Not Available) . This
decrease was due in part to improved safety design of occupational machinery and other protective
measures, the mechanization of agriculture and
Figure 14-1 Burden of injury: United States, 1995. (Data from National Vital Statistics System,
National Hospital Discharge Survey, National Hospital Ambulatory Medical Care Survey, National
Health Interview Survey. Reprinted from Fingerhut LA, Warner M: Injury Chartbook. Health, United
States 1996- 97. Hyattsville, MD, National Center for Health Statistics, 1997.)
38
TABLE 14-1 -- NUMBERS * AND RATES OF INJURY DEATHS IN THE UNITED STATES
(1995) BY INTENTIONALITY AND MECHANISM BY GENDER
TOTAL MALES FEMALES
DEATHS No. Rate No. Rate No. Rate %

All deaths 147,891 56.3 105,645 78.4 42,246 25.4 100.0
Distribution
by
Intentionality
Unintentional 90,402 34.4 60,066 43.3 30,336 16.8 61.1
Suicide 31,284 11.9 25,369 18.6 5,915 4.1 21.2
Homicide 22,552 8.6 17,408 14.5 5,144 4.0 15.2

Other 3,653 1.3 2,802 2.0 851 0.5 2.5

Distribution
by
Mechanism
Motor 42,452 16.2 28,490 22.2 13,962 9.9 28.7
vehicle
traffic
Firearms 35,957 13.7 30,724 24.1 5,233 4.0 24.3
Poisoning 16,307 6.2 11,568 8.4 4,739 3.2 11.0
Falls 11,275 4.3 6,387 3.6 4,888 1.4 7.6
Suffocation 10,376 4.0 7,106 5.1 3,270 1.7 7.0
Drowning 5,071 1.9 4,015 3.1 1,056 0.8 3.4
Fire 4,235 1.6 2,528 1.8 1,707 1.1 2.9
Other 22,218 6.7 14,827 10.2 7,391 3.4 15.0
From National Vital Statistics System, 1995. Atlanta, National Center for Health Statistics, Centers for
Disease Control and Prevention, 1998.
*Excludes adverse event-related deaths (International Classification of Diseases, Injuries, and Causes of Death codes
E870-E879 and E930-E949)
Per 100,000 population. Crude rates are used for "total rate"; age-adjusted rates are used for "gender rate."
industry, labeling and packaging of drugs and toxic products, and improved medical care.
Between 1912 and 1995, unintentional work deaths per 100,000 population were reduced 90%, from 21
to 2. In 1912, an estimated 18,000 to 21,000 workers' lives were lost. In 1996, in a work force more than
triple in size and producing 13 times the goods and services, there were only 4800 work-related deaths.
The death rate from motor vehicle crashes increased 10-fold from 1910 to 1930 as cars became the
primary form of transportation. However, this death rate has decreased 30% in the last two decades
owing in part to improved safety features in vehicles and roads, temporary lowering of speed limits,
increased legal drinking age, and public intolerance of drinking and driving.
The homicide rate increased from 6 per 100,000 population in 1910 to 9 in 1930, decreased during World
War II and the post-war period (1940s to 1960s) to approximately 5, and then increased to 10 in the
1980s and 1990s. Recent increases are attributed to the enormous number of guns in circulation,
currently estimated to be from 150 to 200 million, one third of which are handguns. The suicide rate has
shown less variability but has been consistently higher than the homicide rate throughout this century.

WAYS TO REDUCE INJURY

An example of the effectiveness of a public policy to prevent injury is the reduction in deaths and
non-fatal head injuries in motorcyclists after enactment of the mandatory helmet laws. Hospital charges
for injuries to California motorcyclists declined markedly in the 2 years (1992 to 1993) following
enactment of the law, with charges related to head injuries falling by 58%. Certain personal and family
actions can prevent injury from known hazards (Table 14-2) , and enactment of certain public policies
has the potential to reduce injury:
MOTOR VEHICLE CRASHES.
Wear seat belts to maximize the protection offered by air bags, keep children properly restrained in the
back seat, wear helmets while riding motorcycles and bicycles, and drive sober.
POLICIES.
Enact or maintain motorcycle helmet laws; improve public transportation to reduce dependence on cars.
FIREARMS.
Remove guns from the home (or at least store unloaded, locked, and out of reach of children).
POLICIES.
Restrict the purchase and possession of handguns in the home, up to and including bans (official policy
of the American
Figure 14-2 Firearms and motor vehicle traffic crashes: male death rates by age groups, United
States, 1995. (Data drawn from 1995 Injury Mortality Data, National Center for Health Statistics,
Centers for Disease Control and Prevention, 1998.)
39
Figure 14-3 (Figure Not Available) Trends in injury death rates, United States, 1910 to 1995. (Updated from Baker SP,
O'Neill B, Ginsburg MJ, et al: The Injury Fact Book. 2nd ed. New York, Oxford University Press, 1992.)
Academy of Pediatrics and the American Public Health Association).
FIRE AND BURN INJURIES.
Install and maintain smoke detectors or residential sprinklers; reduce the temperature settings in
residential hot water heaters to 125 F.
POLICIES.
Establish mandatory flammability performance standards for cigarettes to prevent furniture ignition.

DROWNING.
Wear flotation devices while boating.
POLICIES.
Require four-sided isolation fences with self-latching gates on all residential pools.
FALLS.
Install guards on balconies and windows in high-rise buildings; improve lighting and install handgrip
devices in the home.
ALCOHOL-RELATED INJURY.
Avoid binge drinking, promote alcohol moderation, and drive sober.
POLICIES.
Permit community control over the number and location

TABLE 14-2 -- SOME KEY MEASURES TO PREVENT INJURY
INJURY FOR PUBLIC POLICY FOR PATIENTS
Motor vehicle Enact and enforce seat belt, child Use seat belts, child restraints
restraint, and motorcycle helmet Use helmets when riding
laws. bicycles and motorcycles
Firearms Ban assault weapons Remove firearms from home
Enforce waiting period for If gun in home, store locked and
firearm purchase unloaded
Fire and burns Mandate fire-safe cigarettes Maintain smoke detectors; install
sprinklers
Drowning Mandate 4-sided fencing of pools Wear flotation devices when
boating
Falls Install guards on balconies,
windows; install handrails in
elders' homes
Alcohol Advocate for local control of Drive sober
alcohol outlets Avoid binge drinking
Restrict alcohol advertising that
appeals to children

Actions for physicians Counsel patients about injury and

prevention (see above).
Identify and refer abused patients
to social/legal services.
Insist on adequate
physical/occupational
rehabilitation.
Improve injury databases and use
them to design preventive
strategies.
Become knowledgeable about
and advocate sound public
policies.
40
of alcohol outlets in neighborhoods; restrict alcohol advertising that is attractive and available to
children.
Violence and injuries are complex, pervasive problems that must be reduced through comprehensive,
multidisciplinary interventions. As is the case with preventing diseases such as smoking-associated
cancers and acquired immune deficiency syndrome, preventing violence and injuries requires that
physicians intervene both at the individual level and in the social and political processes that determine
the prevalence of these conditions.
IMPLICATIONS FOR MEDICAL PRACTICE

Traditionally, physicians have focused on treating and counseling individual patients and have
concentrated on knowledge, attitudes, and behaviors. Evidence from successful injury prevention efforts
suggests that equal attention should be given to public policies to prevent injury. To reduce violence and
injuries, physicians can do the following:
1. Counsel patients about injury risk and prevention. All physicians, regardless of specialty, have the
opportunity to advise patients and families about actions they can take to prevent injury. The
American Academy of Pediatrics has developed injury prevention messages for parents that
emphasize a child's developmental stages and how to avoid age-specific risks. Although
physician-patient counseling may not always result in behavior change, it can be a powerful
educational message delivered by a trusted authority.
2. Identify and refer abused patients. Hospitals, clinics, and doctors' offices may provide the first
opportunity for abused patients, particularly adult women, to acknowledge the abuse, receive
support, find protection, and break the cycle of violence. Medical and nursing professional
organizations have prepared guidelines for institutionalizing the health care response to family
violence through the development of model protocols, staff training materials, and proposed
modification of intake forms for hospitals and clinics (e.g., American Medical Association

guidelines). Policies and procedures should be adapted to individual hospital needs and address
state-specific regulations about reporting abuse to authorities. Health care providers can best assist
abused patients by working collaboratively with local social and legal services and by referring
patients to these resources.
3. Emphasize rehabilitation and community follow-up. Tertiary prevention involves minimizing
functional disability, a consequence of serious injury. Physicians can help their patients return to
productive lives by ensuring that patients receive appropriate physical and occupational therapy
and that they have access to community services after discharge. The independent living
movement and local centers for independent living, as well as state departments of rehabilitation,
can provide role models and resources for people with disabilities. Because community social and
mental health services are essential for prevention and rehabilitation, physicians can serve their
patients by publicly speaking out in support of these services.
4. Improve the injury database for research and prevention. Information about the mechanisms and
intentionality of injury must be gathered by coroners, medical examiners, and health care providers
through history taking and documentation in official records. The usefulness of non-fatal injury
data would be increased if all states established centralized hospital and emergency department
databases that included external cause of injury codes.
5. Advocate for public policy solutions to the violence and injury problem. Physicians have played a
leadership role in injury control in such diverse areas as traffic safety, burns from tap water and
clothing ignition, and firearms policy. Today's injury problems call for augmented medical
leadership in policy areas. Legislators and journalists turn to physicians for information about
disease and injury because physicians have daily contact with sick and injured people and can thus
speak from personal experience about the problem. Informed physicians can advocate for solutions
by testifying at legislative hearings, by granting media interviews, by making presentations at
professional meetings, and by teaching medical students and residents about injury prevention
principles and strategies. The World Wide Web sites suggested in the reference section provide the
most recent data on statistics, policies, and programs related to violence and injury.
The following agencies can direct investigators to additional sources of data, background materials,
rationale for specific policies, and updates on the current status of policy initiatives and program
interventions.
For Violence and Firearm Injury Control Policies
The Pacific Center for Violence Prevention, San Francisco General Hospital, San Francisco, CA 94100;
Web: www.pcvp.org. Federal government information about criminal justice from the Justice
Information Center at www.ncjrs.org.
For Motor Vehicle Injury Control Policies
Advocates for Highway and Auto Safety, 750 First Street, NE, Suite 901, Washington, DC 20002; Web:
www.saferoads.org. Federal government information from the National Highway Traffic Safety
Administration at www.nhtsa.dot.gov.
For Falls Control Policies (and for Injury in General)
The National Center for Injury Prevention and Control (Centers for Disease Control and Prevention),
Office of Communications Resources, Mailstop K65, 4770 Buford Highway NE, Atlanta, GA
30341-3724; Web: www.cdc.gov/ncipc.

For Flame/Burn Control Policies (and for Injury in General)

Trauma Foundation, San Francisco General Hospital, San Francisco, CA 94100; Web: www.tf.org.
Federal government information from the US Fire Administration at www.usfa.fema.gov.
For Alcohol Control Policies
The Marin Institute for the Prevention of Alcohol and Other Drug Problems, 24 Belvedere Street, San
Rafael, CA 94901; Web: www.marininstitute.org.


Figure 16-1 Ethanol metabolism. Alcohol dehydrogenase (ADH) predominates at low to moderate ethanol doses.
The microsomal ethanol-oxidizing system (MEOS) is induced at high ethanol levels or chronic exposure and by
certain drugs. Aldehyde dehydrogenase (ALDH) inhibition (genetic or drug induced) leads to acetaldehyde
accumulation.


April 21, 2000
Conde Petra
Ellenhorn's Medical Toxicology, 2nd ed., Copyright

1997 Williams & Wilkins
Part IV - CHEMICALS
55 - Alcohols and Glycols Ellenhorn's Medical Toxicology, 2nd ed., Copyright 1997 Williams &
Wilkins
ETHANOL
ISOPROPYL ALCOHOL
METHANOL
ETHYLENE GLYCOL Part IV - CHEMICALS

PROPYLENE GLYCOL
DIETHYLENE GLYCOL
2-MERCAPTOETHANOL Chapter 55 - Alcohols and

ETHYLENE GLYCOL
DERIVATIVES
Glycols
PROPYLENE GLYCOL ETHERS 1127
POLYETHYLENE GLYCOL
ETHANOL
BENZYL ALCOHOL
ACUTE TOXICITY
UNAPPROVED DRUGS (UNITED
STATES)
Clinical toxicities related to ethanol use are summarized in
Table 55-1 (Table Not Available) . A survey of clinical and
laboratory manifestations seen in the critical care setting with
the alcohols is seen in Table 55-2 (Table Not Available) . [1]
Figure 55-1 (Figure Not Available) presents a summary of
statutory limits for alcohol while driving in Europe. Similar
levels are applicable for the United States. Although lower
BACs (0.01 to 0.09 g/dL) can cause driving impairment
associated with an increased risk for fatal crash involvement,
the risk is substantially greater for high levels of alcohol (BACs
more or less than 0.10 g/dL). [2] In the UK there is some support
for changing the legal drinking-driving limit from 17.4 mmol/L
(0.8 g/L) to 10.9 mmol/L. [3] [4]
http://home.mdconsult.com/das/book/view/402/1044....&stty=cow&nid=39982&tlt=Z&pos=&tag=relinfo&anc=bo (1 of 2) [21/04/2000 23:12:58]


http://home.mdconsult.com/das/book/view/402/1044....&stty=cow&nid=39982&tlt=Z&pos=&tag=relinfo&anc=bo (2 of 2) [21/04/2000 23:12:58]

Ellenhorn's Medical Toxicology, 2nd ed., Copyright 1997 Williams & Wilkins
Part IV - CHEMICALS
Chapter 55 - Alcohols and Glycols

1127
ETHANOL
ACUTE TOXICITY
Clinical toxicities related to ethanol use are summarized in Table 55-1 (Table Not Available) . A survey
of clinical and laboratory manifestations seen in the critical care setting with the alcohols is seen in Table
55-2 (Table Not Available) . [1]
Figure 55-1 (Figure Not Available) presents a summary of statutory limits for alcohol while driving in
Europe. Similar levels are applicable for the United States. Although lower BACs (0.01 to 0.09 g/dL) can
cause driving impairment associated with an increased risk for fatal crash involvement, the risk is
substantially greater for high levels of alcohol (BACs more or less than 0.10 g/dL). [2] In the UK there is
some support for changing the legal drinking-driving limit from 17.4 mmol/L (0.8 g/L) to 10.9 mmol/L.
[3] [4]

http://home.mdconsult.com/das/book/body/0/402/1044.html [21/04/2000 23:13:25]

TRAUMA
An increase in blood alcohol has been associated with fatal aircraft accidents, [5] fatal aqauatic activity, [6]
[7] pedestrian deaths, [7] and fatal [5] motor vehicle crashes. [8] Interestingly, a prospective cohort study
suggests that chronic but not acute alcohol abuse adversely affects outcome from blunt or penetrating
trauma (increase in complications, particularly pneumonia, and longer hospital stays). Screening trauma
patients for chronic alcohol abuse may help to confirm this observation. [9]


OVER-THE-COUNTER FORMULATIONS WITH ETHANOL
Tables 55-3 and 55-4 (Table Not Available) summarize ethanol content in some common products.


URETHANE
Urethane, a carcinogen in animals, forms naturally during the fermentation process (Table 55-5) . [10] The
U.S. Food and Drug Administration does not at present have recommendations limiting urethane in
alcoholic beverages.
1128
TABLE 55-1 -- Acute Toxicity of the Alcohols a

(Not Available)
a Adapted from Ellenhorn MJ. In: Hall JB et al, eds. Principles of critical care. New York: McGraw-Hill, 1982, 2080-2093.
TABLE 55-2 -- Ethanol Toxicity in the Critical Care Setting a

(Not Available)
a Adapted from Ellenhorn MJ. In: Hall JB et al, eds. Principles of critical care. New York: McGraw-Hill, 1982, 2080-2093.


TOXICOKINETICS
Absorption
Gastric Alcohol Dehydrogenase
Alcohol dehydrogenase in the gastric mucosa contributes substantially to alcohol metabolism (gastric
first-pass metabolism). Studies of alcohol dehydrogenase activity in gastric biopsies of women suggest a
significant decrease in such gastric alcohol dehydrogenase activity in women when compared with men.
This can explain the findings in women of higher peak blood alcohol concentrations, smaller volumes of
distribution, higher bioavailability of alcohol, and the presence of liver damage after consumption of
relatively smaller quantities of alcohol when compared with men. [11] [12] This finding is reversed in men
over age 50 years. [13]
Elimination/Drug Interactions (Table 55-6) (Table Not Available)
P450IIE1 Induction
P450IIE1 is an ethanol-inducible form of cytochrome P-450 that may, after long-term consumption of
alcohol, be capable of metabolically activating some other compounds to metabolites or to compounds
known to be hepatotoxic. Susceptible compounds include carbon tetrachloride, bromobenzene, and
anesthetic agents, aflatoxin B-induced necrosis and steatosis, isoniazid, phenylbutazone, acetaminophen,
cocaine, nitrosodimethylamine, and methadone. Short-term alcohol use, on the other hand, inhibits
microsomal demethylation of methadone, thus enhancing brain and liver concentrations of the drug by
direct competition for cytochrome P-450. Ethanol can inhibit metabolism of tranquilizers and
barbiturates, enhancing their concentration in the blood. [14]
Both aspirin and H2 receptor antagonists inhibit the action of gastric alcohol dehydrogenase and thereby
increase the blood levels of alcohol. [15] [16] Cigarette smoking appears to slow gastric emptying and
thereby delays alcohol absorption with resultant reductions in peak blood alcohol concentrations after
ingestions of 0.5 g/kg of ethanol (from 13.5 mML/L [63.1 mg/dL] to 11.4 mML/L [511 mg/dL]). [17]
1129
Figure 55-1 (Figure Not Available) One for the road: holidaymakers driving in Europe this summer will find widely
differing limits on drink-driving. In some countries, such as Bulgaria, Romania and Turkey, drivers must be teetotal. Those
who like a tipple will discover that the rules are most relaxed in Republic of Ireland. (From The European. May 24, 1991.).
Cocaethylene (See Cocaine Chapter)/Cocapropylene
Cocapropylene (propylcocaine) is formed by transesterification of cocaine with n-propanol in whole

human liver homogenates in vitro. [18] This product appears to have strong topical anesthetic activity. [19]
Its importance in clinical cocaine toxicity has yet to be determined. [19]



ACETALDEHYDE
Clinical Presentation
Acetaldehyde, produced by all known oxidative pathways of ethanol metabolism, is in turn converted to
acetate by aldehyde dehydrogenase. Asians often harbor an inactive aldehyde dehydrogenase variant that
causes them to experience high blood acetaldehyde levels when they drink, with subsequent development
of ethanol intolerance and flushing. [20] Disulfiram, an inhibitor of acetaldehyde dehydrogenase, raises the
acetaldehyde levels similarly in most subjects after drinking and causes flushing and other adverse
effects. Similar elevations of plasma acetaldehyde and facial flushing follow ingestions of calcium
carbamide. [21] Other inhibitors of acetaldehyde dehydrogenase include metronidazole, sulfonylurea
antidiabetic drugs, the fungicide thiram, and the ink cap mushroom, Coprinus atramentarius.
Acetaldehyde is also a metabolite of paraldehyde. Pathologic findings in deaths from acetaldehyde
poisoning include pulmonary edema, nausea, narcosis, respiratory failure, cardiac dilatation,
cardiovascular collapse, congestive heart failure, seizures, and sudden death. [22] [23] Death of a
17-year-old boy after ethanol ingestion was associated with a blood acetaldehyde level of 1 mg/mL and
an ethanol level of 110 mg/dL. [23]
Mechanism of Action
Acetaldehyde in blood is decreased by reactions with compounds containing sulfhydryl and amino
groups (e.g.,
1130
TABLE 55-3 -- Alcohol Content of Some Liquid Cold Preparations

Product Alcohol (%)
Antihistamines and Combinations
Vicks Children's Cough Syrup 5
Cheracol D Cough Formula 4.75
Co-Tylenol Cold Medicine 7.5
Demazin Decongtestant-Antihistamine Syrup 7.5
Dimetane Decongestant Elixir 2.3
Dimetane Elixir 3
Formula 44 Cough Mixture 10
Formula 44D Decongestant Cough Mixture 10

Novahistine Elixir Cold and Hay Fever Formula 5

Nyquil 25
Triaminic Expectorant 5
Decongestants (Oral and Combinations)
Cheracol Plus 8
Children's Co-Tylenol Liquid Cold Formula 8.5
Cogespirin Liquid Cold Medicine 10
Contac Severe Cold Formula Night Strength Liquid 25
Contac Jr 10
Coricidin Cough Syrup 0.5
Formula 44M Multi-Symptom Cough Mixture 20
Naldecon-DX Pediatric Syrup 5
Robitussin 3.5
Robitussin-CF 4.75
Robitussin-DM 1.4
Robitussin-PE 1.4
Robitussin Night Relief Colds Formula 25
Mouthwashes
Cepacol Mouthwash/Gargle 14
Scope (SD Alcohol 38F) 18.5
Other
Geritol 12
Paregoric tincture 45
Terpin hydrate elixir 42
TABLE 55-4 -- Approximate Percent Ethanol Content of Some Common Products a

(Not Available)
a From Vogel C et al. Clin Toxicol 1995;33:25-33.

d-penicillamine). Cardiovascular sequelae develop as a result of sympathomimetic effects involving

catecholamine release (positive inotropic and chronotropic responses, vasoconstriction and
hypertension), which may be secondary to the reaction of acetaldehyde with tissue sulfhydryl groups.
Acetaldehyde also acts on the liver to depress mitochondrial function, decrease fatty acid oxidation,
enhance glycogenolysis, and decrease gluconeogenesis from glycerol. These effects may be produced
with concentrations of ethanol that are observed following acute ethanol intoxication. Acetaldehyde may
be involved in some of the hepatotoxic effects that are incident to chronic ethanol ingestion. Alterations
in norepinephrine turnover in the brain and changes in the metabolic pattern of other monoamines such
as dopamine and serotonin are induced by acetaldehyde. [22]


TERATOGENESIS
Data on the quantity of alcohol drunk by the mother during pregnancy and the subsequent development
of congenital malformations in the neonate suggest that one to two drinks per day taken by the mother
during the first trimester may not be associated with more teratogenicity than that associated with
nondrinkers. [24] Studies on the human placenta suggest that the placenta oxidizes ethanol to acetaldehyde
and releases acetaldehyde to the fetus. This may be a factor in production of the fetal alcohol syndrome
[25] (Table 55-7) (Table Not Available) .
Three criteria for the diagnosis of fetal alcohol syndrome have been recommended by the Research
Society on Alcoholism Fetal Alcohol Study Group. [26]
TABLE 55-5 -- Wine and Whiskey Sampling a
FDA and the U.S. Bureau of Alcohol, Tobacco and Firearms has sampled wines and whiskeys from
domestic and foreign producers to determine urethane levels.
The following are the results of two samplings measuring average urethane levels in parts per billion
(ppb). The 1987 figures represent the FDA-ATF initial survey of domestic and imported alcoholic
beverages, collected from January 1986 through August 1987. This compares with an ATF sampling
done in 1991 that shows urethane levels decreasing in most instances.
Average Urethane Level (ppb)

1987 1991
Product Domestic Imported
Brandy (grape) 40 10 45
Brandy (fruit) 1,200 5 255
Bourbon (retail) 150 70 55
Rum 20 2 5
Liqueur 100 10 25
Scotch 50 b 55
Sherry 130 10 40
Port 60 23 26
Grape wine 13 10 15
Sake 300 55 60
a Adapted from Foulke FL. FDA Consumer. 1993;27(1):23.
b Scotch is not manufactured domestically.

1131
TABLE 55-6 -- Adverse Interactions With Alcohol a

(Not Available)
aAdapted from Interactions of drugs with alcohol. Med Lett Drugs Ther 1981;23:34. Used by special permission from The
Medical Letter.
1. Prenatal growth retardation of length, weight, or head circumference ( 2 SD) or postnatal growth
retardation of length, head circumference ( 2 SD), or relative weight ( 10%) or both.
2. Dysfunction of central nervous system indicated (a) by performance below -2 SD on the Bayley
Mental Scale or in the Reynell Verbal Comprehension Test or (b) by performance better than -2
SD in the developmental tests used, but with difficulties in active speech, fine motor development,
or perception needing further follow-up.
3. Craniofacial criteria with at least two of the following: (a) head circumference 2 SD, (b)
palpebral fissure 2 SD, and (c) hypoplastic philtrum and thin upper lip.
The diagnosis of fetal alcohol effects is made when only two of the criteria are present. [27]
1132
TABLE 55-7 -- Features Observed in Fetal Alcohol Syndrome/Fetal Alcohol Effects a

(Not Available)
a From Committe on Substance Abuse. American Academt of Pediatrics. Pediatrics 1993;91:1004-1006.
Fetal alcohol syndrome (FAS) is difficult to recognize in newborns because facial stigmata of FAS are
often subtle. Some types of central nervous system deficits in infants are difficult to detect and the birth
weight of some affected infants is normal. [28]
The National Institute on Alcohol Abuse and Alcoholism (NIAAA)--which joined the National Institutes
of Health (NIH) on October 1, 1992--has the major responsibility for research on FAS and ARBD. At

NIAAA's Fetal Alcohol Research Center at Wayne State University, Detroit, Michigan, Robert J. Sokol,
M.D., and associates Susan S. Martier, MSSA, and Joel W. Ager, PhD, have developed a four-question
test (T-ACE) that takes less than 1 minute to administer. The quiz circumvents the problems of denial
and underreporting that historically make self-reporting, the only other screening technique available, of
limited value.
Known as T-ACE, the test has the further advantage of not seeming to pry into current drinking habits,
which might prompt untruthful answers. The key question concerns tolerance, one of the best predictors
of continued drinking throughout pregnancy: ``How many drinks does it take to make you feel high?''
A woman who replies ``more than two,'' Sokol's team found, is more likely to drink enough alcohol to
bear an infant with alcohol-related birth defects or fetal alcohol syndrome. That risk is amplified by
positive responses to at least one of T-ACE's other queries about whether she has been annoyed by
criticism of her drinking, has felt she should cut down, and has ever had a drink first thing in the morning
to steady her or get rid of a hangover (eye-opener).
Lactation
Following ingestion of ethanol by the lactating mother, breast milk samples will contain alcohol that can
be perceived by odor. Studies of infant behavior suggest that alcohol in breast milk will lead to a reduced
consumption of milk by the infant. [29]


PATHOPHYSIOLOGY
Abuse
A positive association between the A1 allele of the D2R dopamine receptor gene and alcoholism has
been reported, suggesting that a mutation that confers susceptibility to this clinical condition is present in
the vicinity of the restriction site of the Taq 1 enzymes, located in the DRD2 gene. The number of DRD2
sites is reduced in human alcoholics. Further data to validate the significance of these observations are
indicated. [30] [31] [32]
Alcohol Dependency
Diagnostic tests have been devised to identify problem drinkers or alcohol dependence. Alcohol
dependence represents a syndrome diagnosed by DSM-III-R (Diagnostic and Statistical Manual of
Mental Disorders, Revised, Third Edition--American Psychiatric Association) and the International
Statistical Classification of Diseases, Tenth Revision (ICD-10) (Table 55-8) (Table Not Available) .
Several questionnaires have been developed for the detection of alcohol disorders, including the CAGE
(cut down, annoyed by criticism, guilty about drinking, eye-opener drinks) questionnaire, the Michigan
Alcoholism Screening Test (MAST) (Table 55-9) (Table Not Available) , and the AUDIT (Table 55-10)
(Table Not Available) , and they are summarized by Allen and Colleagues (Table 55-11) (Table Not
Available) . [33] The most widely used are the CAGE questionnaire and the MAST. Of these the MAST
has been more thoroughly studied in terms of reliability and accuracy. However, the MAST and its
shortened versions are more complicated than the CAGE questionnaire. The CAGE questionnaire is
short, easily memorized, and reasonably accurate, making it the screening test of choice for busy house
officers and practitioners.
1133
Some authors contend that ingestion of four or more drinks per day in man and two or more drinks per
day in women constitute a ``hazardous'' consumption level that increases the risk of alcohol dependence
and medical problems. A ``drink'' is defined as equivalent volume amounts that have an ethanol content
of 0.6 oz. Twelve
TABLE 55-8 -- DSM III-R and ICD-10 Diagnostic Criteria for Substance Abuse, Harmful Use, and
Substance Dependence a
(Not Available)
a DSM-III-R indicates Diagnostic and Statistical Manual of Mental Disorders, Revised Third Edition; ICD-10,
International Statistical Classification of Diseases, 10th Revision.

ounces of beer, 5 oz of wine, and 1.5 oz of liquor all contain 0.6 oz of ethanol.
The CAGE questionnaire can be a useful tool in the diagnosis of DSM-III-R-defined abuse and
dependence and very heavy drinking (>8 drinks per day). Scores of 3 or 4 strongly support the diagnosis
of alcohol abuse. The CAGE questionnaire has not been tested as a tool for identifying persons who may
be engaged in hazardous drinking of lesser amounts of alcohol, for example, 4 drinks per day. The
AUDIT was recently developed to identify hazardous drinkers. [34]
Wernicke's Encephalopathy
Acute pancreatitis, hyperemesis gravidarum, anorexia nervosa, prolonged fasting, malnutrition in

infancy, gastric plication, prolonged feeding (including total parenteral nutrition), leukemia, lymphoma,
chronic renal failure, thyrotoxicosis, and acquired immunodeficiency syndrome, in addition to ethanol,
have been associated with the development of Wernicke's encephalopathy and Korsakoff's syndrome. [35]
[36]


CLINICAL PRESENTATION
Acute Intoxication
Children
Percutaneous alcohol intoxication has been described in young children after use of alcohol-soaked
gauze pads. [37] Chemical burns of the skin in neonates can follow applications of alcoholic skin
preparations. [38] In juvenile alcohol intoxication, metabolic acidosis and decreased blood pH may be
correlated with the blood alcohol concentration and loss of consciousness. Hypoglycemia is the most
common reported symptom in children under 5 years of age. The hypoglycemic effects of ethanol are not
dose dependent. The fasting state may predispose a child to ethanol-induced hypoglycemia. [39]
Hypokalemia is an important concomitant finding. Alcohol abuse occurs in very young children. [40] [41]
Adults
Ethanol is a selective CNS depressant in low doses and a generalized depressant in high doses.
Comparison of cognitive and psychomotor skills at blood ethanol levels of 90 and 135 mg/dL indicates
that attention, concentration, motor coordination, and reaction time are significantly more affected at the
higher level. At these same levels no difference was observed in visual and verbal memory. [42] Initially,
ethanol produces exhilaration, which progresses to loss of restraint, behavioral abnormalities,
loquaciousness, slurred speech, ataxia, gait disturbances, irritability, drowsiness, and finally stupor and
coma. A flushed face, dilated pupils, excessive sweating, and gastrointestinal distress may accompany
CNS symptoms. Rarely, alcohol-induced urticaria occurs, which is partially mediated by histamine. [43]
Ethanol can produce dysrhythmias (e.g., atrial fibrillation) in nontolerant binge drinkers, as well as in
chronic alcoholics. [44] Ethanol is a venodilator that produces decreased preload, afterload, and systemic
vascular resistance in healthy adults after acute ingestion. When these factors are corrected, acute
ingestion also has a myocardial depressant effect. [45] Tolerance
1134
TABLE 55-9 -- Life-style Risk Assessment Instrument, Northeastern Vermont Regional Hospital, St
Johnsbury a
(Not Available)
a Adapted from Graham AW. Arch Intern Med 1991;151:958-964.
1135

TABLE 55-10 -- Alcohol Use Disorders Identification Test (AUDIT) Questions a

(Not Available)
a From Kitchens JA. JAMA 1994;272:1782-1787.
TABLE 55-11 -- Characteristics of Self-report Alcoholism Screening Tests a

(Not Available)
a Adapted from Allen et al. Arch Interm Med 1995;155:1726-1730.
1136
lessens acute ethanol effects, but may exacerbate chronic metabolic effects. In all acutely inebriated
patients, search for concurrent trauma (e.g., subdural hematoma), underlying disease, and coingestion of
drugs and toxic alcohol substitutes (i.e., methanol, ethylene glycol).
Chronic Ethanolism (Alcoholism)
In 1972 the Criteria Committee of the National Council on Alcoholism proposed criteria for the
diagnosis of alcohol. [46] This has been recently modified. [47]
Alcoholism is a primary, chronic disease with genetic, psychosocial, and environmental factors
influencing its development and manifestations. The disease is often progressive and fatal. It is
characterized by impaired control over drinking, preoccupation with the drug alcohol, use of alcohol
despite adverse consequences, and distortions in thinking, most notably denial. Each of these symptoms
may be continuous or periodic.
Nervous System
Diminished fine motor skills, diminished cognition, peripheral motor/sensory neuropathy, and
Wernicke's-Korsakoff's syndrome have been observed. Movement disorders are
TABLE 55-12 -- Movement Disorders Associated With Alcoholism a
(Not Available)
a Adapted from Neiman J et al. Neurology 1990;40:741-746.

summarized in Table 55-12 (Table Not Available) . [48] Data on alcohol consumption suggests that low
levels of alcohol consumption (below 390 g weekly) may have some protective effect on the cerebral
vasculature, but heavy consumption (over 400 g weekly) appears to predispose to both hemorrhagic and
nonhemorrhagic stroke. [49]
Gastrointestinal Tract
Acute pancreatitis may be associated with a retinopathy characterized by multiple cotton-wool patches.
Vision may be impaired. [50]
Liver
A prospective multicenter study suggests that in alcoholic cirrhosis with or without alcoholic hepatitis,
progression to cirrhosis from alcoholic hepatitis occurs in about half of proven chronic alcoholics. The
presence or absence of Mallory bodies does not correlate with either the severity or mortality of patients
with alcoholic hepatitis and cirrhosis. There is still controversy relating to the reversibility of cirrhosis
and to whether abstinence from alcohol improves survival. Risk factors for survival include the patient's
age, race, prothrombin time, ALT levels, AST:ALT ratio, ascites, histologic severity score, alcohol
intake prior to admission, and clinical disease severity. [51] Professor Sheila Sherlock estimates that the
minimum alcohol intake associated with appreciable liver damage is 16 units of alcohol daily for 5 years
(1 unit, 10 g of alcohol is contained in 28 mL [1 fluid ounce] of whiskey or similar spirits, 85 mL of
wine, or 230 mL of beer). [52]
Muscle
Patients may develop proximal-muscle weakness with elevated serum creatine kinase levels and
myoglobinuria. Myopathy generally appears in middle-aged alcoholics after many years of drinking. It is
estimated that myopathy is more likely to occur in a 70 kg man who drinks more than 12 oz of 86 proof
whiskey (120 g of ethanol) a day for 20 years--a lifetime dose of 876 kg. [53] [54]
Movement Disorders
Movement disorders associated with alcoholism are summarized in Table 55-12 (Table Not Available) .
Hematologic Abnormalities
Hematologic effects of alcoholism on the platelets, red cells, and neutrophils are presented in Tables
55-13 and 55-14 (Table Not Available) .
TABLE 55-13 -- Hematologic Abnormalities

Decreased platelets
On withdrawal
Reticulocytosis
Rise in white cell count
Fall in serum iron

1137
TABLE 55-14 -- Alcohol Effects on Neutrophil Function a

(Not Available)
a Adapted from MacGregpr RR, JAMA 1986;256:1474-1479.
TABLE 55-15 -- Initial Electrocardiographic Findings in Alcoholic Cardiomyopathy a

(Not Available)
a Adapted from Moushmoush B, Abi-Mansour P. Arch Interm Med 1991;151:38.
Cardiac Dysfunction
The ``Holiday Heart Syndrome'' reflects a supraventricular arrhythmia induced by drinking binges (over
6 drinks a day). Atrial fibrillation is the most common arrhythmia, but atrial flutter, atrial tachycardia,
junction tachycardia, and multiple atrial premature beats have also been observed.
An isolated episode of atrial fibrillation is often the first complaint. About 10 to 20 years of high alcohol
use (about 200 mL--7 oz of 86 proof whiskey a day) may be required before cardiac decompensation
becomes apparent. [53] [55] Right- and left-sided heart failure then become more apparent. [56] Early
electrocardiographic changes include left ventricular hypertrophy with abnormal T-waves and
nonspecific ST-T-wave changes. There are few specific changes found on myocardial biopsy. Therapy
should include abstinence from alcohol, dietary salt restriction, diuretics, digoxin, vasodilator drugs,
possibly angiotensin-converting enzyme inhibitors, and thiamine.
Primary ventricular arrhythmias culminating in fibrillation may partially explain why alcoholics die
suddenly and unexpectedly. [56] Decreased variability of heart rate--a sign of cardiac vagal neuropathy
and a factor notorious for increase in the risk of death after myocardial infarction--is a relatively common
finding among men dependent on alcohol. Most episodes of cardiac arrhythmias terminate within 24 to
48 hours either spontaneously or after treatment with beta-blockers combined with adequate sedation,
rehydration, and treatment of any potassium and magnesium depletion. Standard alcoholism
questionnaire (Tables 55-6 (Table Not Available) and 55-7) (Table Not Available) should be
administered to all patients presenting with otherwise unexplained tachyrhythmias. [57]
Alcoholic Cardiomyopathy (Table 55-15) (Table Not Available)

Alcoholic cardiomyopathy shares certain features with the Beriberi heart failure found in malnourished,
vitamin-deficient alcoholics: cardiac chamber dilatation, tachycardia, elevated venous pressure, and
peripheral edema. However, the thiamine-deficient patient exhibits a high cardiac output state and warm
extremities, while the chronic alcoholic patient has depressed cardiac output and ventricular
hypocontractility. [56] Concomitant toxic substances such as cobalt chloride (additive used as a beer-foam
stabilizer resulting in death), [58] features of chronic arsenic intoxication in some wine drinks, [59] and lead
contamination in some moonshiners [60] may have been contributing factors affecting the development of
cardiomyopathy in particular groups of chronic alcoholics. Hypokalemia, hypophosphatemia, and
hypomagnesemia are contributory factors in some patients with alcoholic cardiomyopathy. [56] A
controlled study suggests that susceptibility to alcoholic cardiomyopathy and myopathy appears to be
more pronounced in women than in men. [61]
Cardiac Conduction
Cardiovascular death is the most important cause of mortality in alcoholics, yet alcohol may protect
against ischemic heart disease. QT-interval prolongation in some patients with alcoholic liver disease is
associated with an adverse prognosis, especially sudden cardiac death. [62]
Hypertension
Acutely, alcohol causes a modest fall in blood pressure. [63] Continued consumption of more than the
amount contained
1138
in two usual portions a day (one portion contains 10 to 12 g of ethanol) results in a dose-dependent rise in
blood pressure. [64] [65]
Bone
Hip fractures [66]
Osteopenia and fractures, especially of the spine and ribs, are associated with osteoporosis rather than
osteomalacia; circulatory levels of Vitamin D metabolites are observed. The role of parathyroid hormone
and calcitonin has yet to be established. [67] [68]
Immune Defense
Tables 55-16 (Table Not Available) and 55-17 (Table Not Available) summarize the effects of alcohol on
cell-mediated and humoral immunity. [69]
Transient Hypoparathyroidism
Short-term alcohol administration causes a decline in the secretion of parathyroid hormone and this may
account at least in part for the transient hypocalcemia, hypercalciuria, and hypermagnesemia that follow
alcohol ingestion. [70]
Magnesium Deficiency

Alcoholism is probably the most important cause of magnesium deficiency (see Table 55-18) (Table Not
Available) . Alcoholics ingest low levels of magnesium in their diet, excrete more in their urine, and have
decreased albumin (with cirrhosis of the liver) for binding magnesium. The serum magnesium level may
not reflect this deficit. [71] Hypomagnesemia may be present, but serum levels of magnesium do not
predict body deficits accurately. Magnesium deficiency interferes with thiamine action. Thiamine should
be administered with magnesium to prevent Wernicke-Korsakoff syndrome. Administer the 30% solution
slowly when given intravenously to avoid pain and sclerosis. The chronic alcoholic has a mean
magnesium deficit of about 1.2 mEq/hour. For life-threatening states (e.g., dysrhythmia) MgSO4 may be
administered in a dosage of up to 4 g over 3 to 4 minutes. [72]
Renal
Patients with chronic alcoholism have a variety of renal tubular abnormalities that are independent of
chronic liver disease, pancreatitis, and rhabdomyolysis and that occur in the presence of normal
glomerular filtration. These abnormalities are reversible, disappearing after 4 weeks of abstinence despite
many years of alcohol abuse. [73]
TABLE 55-16 -- Alcohol Effects on Cell-Mediated Immunity a
(Not Available)
TABLE 55-17 -- Alcohol Effects on Humoral Immunity a

(Not Available)
1139
TABLE 55-18 -- Symptoms and Signs of Magnesium Deficiency Syndrome a

(Not Available)
a Adapted from Miller G. Compr Ther 1985;11:58-64.

Alcoholic Pellagra Encephalopathy [1]
Recent studies indicate that some alcoholic patients who have received thiamine and pyridoxine but not
niacin have developed a secondary pellagra consisting of confusion or an altered state of consciousness,
oppositional hypertonia, and myoclonus. Such findings may develop over a period of weeks or during
hospitalization several days after admission and apparent recovery from Wernicke's encephalopathy.
Treatment with niacin may result in dramatic improvement. Stimulation of metabolic pathways by
pyridoxine and thiamine may increase the relative deficit of niacin.


ALCOHOLIC KETOACIDOSIS [1]
Alcoholic ketoacidosis (AKA) follows withdrawal from alcohol and develops in chronic alcoholics with
a recent history of heavy episodes. [1] [74] [75] Such patients have often experienced symptoms of nausea,
vomiting, abdominal pain, [76] and decreased food intake often due to gastritis, hepatitis, or pancreatitis or
related to alcohol withdrawal, fatty liver infiltration, or aspiration pneumonia. They become volume
depleted and usually have abruptly stopped or markedly decreased their alcohol intake 24 to 72 hours
before presentation. The patient becomes confused, drowsy, and occasionally comatose. Tachypnea and
tachycardia are common signs, and the patient may present with breathlessness in a Kussmaul breathing
pattern compensatory for the ketoacidosis. These patients often have no measurable blood alcohol levels
when first seen in a health care facility. The blood glucose level is usually normal to slightly elevated.
Most patients will respond to glucose-containing intravenous fluids without insulin. In some areas
alcoholic ketoacidosis is the causative factor in up to 20% of patients presenting with ketoacidosis.
Patients are usually conscious and able to give a good history.
The moderate-to-severe ketoacidosis is due to the formation of beta hydroxybutyrate (BOHB) and
acetoacetate (AcAc). The BOHB usually predominates and therefore testing with Ketostix and Acetest
(which are most sensitive to acetoacetate, less so to acetone, and not at all to beta hydroxybutyrate) may
show only a weakly positive reaction when levels of BOHB are highest. The finding of ketonuria without
glycosuria suggests the diagnosis. Serum lactate levels are only moderately elevated. Severe lactic
acidosis would suggest another serious disorder such as hypoxemia or hypoperfusion.
Metabolic effects associated with alcoholic ketoacidosis include hormonal changes (increased levels of
cortisol, growth hormone, glucagon, free fatty acids, catecholamines, decreased levels of insulin, and
ADH) and effects secondary to any increase in the NADH/NAD ratio (increase in the BOHB/AcAc ratio
and lactate production; decreased gluconeogenesis and citric acid cycle activity).
Electrolyte, glucose, and arterial blood gas measurements are essential in making a diagnosis. The basic
acid-base abnormality in alcoholic ketoacidosis is an elevated anion-gap metabolic acidosis.
Hypokalemia and hypochloremia are often seen due to the bouts of prolonged vomiting. Serum
potassium must be carefully monitored during treatment. The blood pH may vary from 6.96 to 7.61. A
significant respiratory alkalosis may be seen as a compensatory response to the metabolic acidosis or due
to alcohol withdrawal or other associated illnesses. The protracted vomiting can lead to a primary
metabolic alkalosis. The initial blood pH and bicarbonate levels are not good indicators of eventual
outcome.
Serum ketones are markedly elevated. AcAc levels over 2 mEq/L (normal:<0.05) and BOHB levels over
10 mEq/L (normal:<0.05) may be present. The BOHB/AcAc ratio (normally 1:1) rises to 4:10.1. When
alcoholics with ketoacidosis are treated, the BOHB is oxidized to AcAc and later to acetone. Thus the
nitroprusside test may worsen when the patient is actually improving. Remember that when severe
elevations of the anion gap (>30 mEq/L) are found in ketoacidosis, hyperosmolar coma, lactic acidosis,
and ingestion of ethylene glycol or methanol, an osmolal gap is present. [77] [78]

Treatment
Management of these patients requires correction of volume depletion and administration of glucose. [1]
[71] [72] [73] [74] [75] The volume depletion is usually amenable to infusion of solutions of normal saline with
dextrose. When the volume deficit is corrected (normal orthostatic blood pressure and pulse), 0.5 N
saline with dextrose may be continued. Such intravenous therapy is continued until the serum
bicarbonate level reaches 18 to 20 mEq/L, signs of orthostasis have resolved, and oral fluids are well
tolerated. Patients will respond to therapy within 12 hours. Close monitoring (every 4 to 6 hours) of
serum potassium and phosphorus levels during treatment is important because hypokalemia and
hypophosphatemia may ensue quickly. Potassium supplementation may be required. Sodium bicarbonate
administration is usually not necessary except in severe cases of acidosis
1140
(pH <7.1). Insulin therapy is not required. Thiamine (50 to 100 mg) should be given to prevent
development of the Wernicke-Korsakoff syndrome. Magnesium and multivitamins may be considered.


LABORATORY
Markers may assist in the diagnosis of alcoholism (Table 55-19) (Table Not Available) . [79]
Analytic Methods
Two methods provide rapid quantitative determination of blood alcohol concentrations (BAC). One uses
an electrochemical method and the other, saliva. [80] [81] Limitation to the saliva test may include
cross-reactivity with other congeners such as methanol. Mouthwash, phenol-containing lozenges,
tobacco products, and patient cooperation may affect these determinations. Vomitus containing alcohol
may produce a falsely elevated test result. Further data is required before clinical usefulness of these
procedures can be defined.
Blood Levels (Table 55-20) (Table Not Available)
Ethanol doses calculated to achieve and maintain blood ethanol concentrations of 100 mg/dL in a 70-kg
adult are presented in Table 55-21 (Table Not Available) . A 30-month-old 13-kg child
TABLE 55-19 -- Markers for Alcoholism a
(Not Available)
a Adapted from Mihas AA, Travassoli M. Am J Med Sci 1992;303:415-428.
became comatose after ingesting up to 16 ounces of wine containing 20% ethanol. Despite the initial
blood ethanol level of 98.78 mmol/L (455 mg/dL), the child recovered following prompt gastric
decontamination and maintenance of adequate hydration and euglycemia. [82] In spite of intensive
investigation, there is still no satisfactory useful clinical laboratory marker for surreptitious alcohol
ingestion. Ingestion
TABLE 55-20 -- Stages of Acute Alcoholic Influence/Intoxication in Nontolerant Individuals a
(Not Available)
a Adapted from Dubowski KM. Am J Clin Pathol 1980;74:747-750.
1141

TABLE 55-21 -- Ethanol Doses Calculated to Achieve and Maintain Blood Ethanol Concentrations of
100 mg/dL in a 70-kg Adult a
(Not Available)
a Adapted from McCoy HG, Cipolle RJ, Ehlers SM et al. Am J Med 1979;67:806.
TABLE 55-22 -- Effect of Some Solutes on Serum Osmolality a

(Not Available)
a Adapted from Ellenhorn MJ In: Hall JB et al, eds. principles of critical care. New York: McGraw-Hill, 1992;2080-2093.
of 80 g of ethanol over 30 minutes by heavy drinkers (720 to 2000 g/week) may produce a significant
rise in plasma glutathione-S-transferase with peak values 60 minutes after alcohol ingestion, suggestive
of mild subclinical acute liver damage. [83] Further work is required to determine clinical usefulness of
this test. Additional important prognostic abnormalities of alcoholic liver disease include serum albumin
levels less than 2.5 g/dL and serum bilirubin values over 136 mumol/L (7.5 mg/dL). [84] [85]
Abnormalities
Serum Osmolality (Table 55-22) (Table Not Available)
Geller and colleagues found that the osmolal gap (mOsm/kg) was related to the serum ethanol
concentration (nmol/L) by the formula:
Hypophosphatemia
Hypophosphatemia may be observed in alcoholics and is often found in malnourished individuals.

Reduced phosphate reabsorption capacity in alcoholics is probably the result of a proximal tubular
dysfunction related to brush border damage. Data suggest that liver-function impairment is not required
for this proximal tubular dysfunction. [87] Associated neurologic deficits have varied from anisocoria,
ballismus, paresthesia, hyporeflexia, ataxia, convulsions, coma, and even death. Coma is usually
accompanied by seizures. [88] Management includes intravenous phosphate, thiamine, and potassium and
magnesium replacement as required.
Lactic Acidosis
The most common cause of elevated blood lactate, in general, is circulatory shock. Acute or chronic
alcohol abuse predisposes to sepsis, gastrointestinal hemorrhage, pancreatitis, and other disorders that
can lead to shock. Other causes of lactic acidosis encountered in emergency departments include

seizures, liver disease, alcoholic ketoacidosis, thiamine deficiency, and poisoning with methanol,
ethylene glycol, acetaminophen, cyanide, and carbon monoxide. Lactic acidosis solely attributable to
ethanol is uncommon. Look for an underlying pathophysiologic process. [89]
Acetone
Abnormally high concentrations of acetone in the blood might occur if a person drinks 2-propanol,
undertakes a prolonged fast, has diabetes mellitus, or engages in strenuous exercise. The highest recorded
concentration of acetone (61.9 mug/mL) was observed in the blood of a drunk driver (blood alcohol
concentration 0.11 g %). This suggests that in a population of motorists an individual with an abnormally
high concentration of acetone in the blood or breath is unlikely to be observed. The risk of acetone
interfering with breath-alcohol analyzers may be exaggerated. [90]
Carbohydrate-Deficient Transferrin
A new and potentially useful diagnostic marker of alcohol abuse utilizes carbohydrate-deficient
transferrin in the serum. [91] This transferrin abnormality measures an accumulated effect of alcohol
consumption, appearing after regular intake of 50 to 80 g of ethanol/day for at least 1 week and
normalizing slowly during abstinence (half-life = about 15 days). Koppel considers this to be a specific
marker for alcohol abuse patients admitted to the intensive care unit. This test is not yet widely used. [92]
1142
Selenium
A controlled clinical study suggests that chronic alcohol abuse may be associated with a decreased serum
selenium level. The significance of this finding must be validated by further study.92a


TREATMENT
Flumazenil
Flumazenil (3 mg IV) may aid in reversing the respiratory depression associated with ethanol ingestion,
but this observation has not been clinically validated. [93] Analeptic agents should not be used.
RO 15-4513
RO 15-4513 is a synthetic chemical closely related to the benzodiazepine antagonist flumazenil. It is an

antagonist of the action of ethanol in animals [94] and specifically is antagonistic to ethanol-induced
depression of the central nervous system, where it inhibits the neurochemical and behavioral effects of
ethanol. [95] RO 15-4513 may be valuable in attenuating drunkenness, but is useless in life-threatening,
ethanol-overdosage-induced coma.
Since it can reverse the central nervous system effects of ethanol while allowing the peripheral effects of
the drug to remain, RO 15-4513 would probably lead to an increase in the incidence of alcohol-related
disease of the liver, cardiovascular system, and gastrointestinal tract. The manufacturer has decided not
to market the product. [91] [92]
Antagonists
Calcium Carbamide
This investigational drug inhibits the enzyme aldehyde dehydrogenase, which metabolizes acetaldehyde
to acetic acid. Ingestion of alcohol causes an accumulation of acetaldehyde and brings on nausea and
vomiting. Optimum dose and dosage schedule have not been established; adverse reactions and
deleterious drug interactions preclude its use at present. [96]
Carbamazepine
Studies with carbamazepine suggest that it is effective in treating alcohol withdrawal, including delirium
tremens. [97] It appears to be effective without adjunctive medication. [98] Carbamazepine may offer the
advantage of rapid return to work or early induction into an alcoholism treatment program. [99]
Chlormethiazole
(Sold in Great Britain as Heminevrin.) Chlormethiazole has hypnotic, anxiolytic, and anticonvulsant
properties. In Britain, where it is given in a rapidly reducing dosage over 6 days it is the most popular
drug used for alcohol withdrawal. However, alcoholics rapidly become dependent on this drug.
Chlormethiazole abuse may lead to serious self-poisoning with deep coma and centrally mediated
respiratory depression that may be fatal.
Preliminary reports suggest the use of clonidine [100] (60 to 180 mug/hour IV) and gamma-hydroxybutyric
acid [101] [102] (50 mg/kg orally) for the treatment of withdrawal symptoms. Confirmatory controlled

clinical studies will be required to evaluate these findings.

Disulfiram
(See Disulfiram chapter.)

Lithium Carbonate
Maintenance of therapeutic serum levels of lithium appears to assist in maintenance of sobriety. [103]
Further studies are required to substantiate the dose and efficacy of this drug in alcoholism. [103] [104] [105]
4-Methylpyrazole
Initial studies suggest that a dose of 7 mg/kg of intravenous 4-methylpyrazole appears to decrease the
rate of elimination of ethanol and to suppress typical manifestations of ethanol ingestion in humans. [106]
Naltrexone
In 1994 the Food and Drug Administration approved naltrexone (Revia) as a treatment for alcoholism.
Research indicates that the drug appears to reduce the craving for alcohol. Naltrexone should not be used
in patients receiving opioids or currently dependent on them, those in acute opioid withdrawal, those who
have a history of sensitivity to naltrexone, or those patients with acute hepatitis or liver failure. Patients
with alcohol-induced liver dysfunction may be poor candidates for naltrexone therapy. The
recommended dose of naltrexone for treatment of alcohol dependence is 50 mg once a day for 12 weeks.
Long-term trials are lacking. There is no evidence that the drug is effective without regular counseling.
[107]
Ritanserin
A potent and specific 5HT2 -receptor antagonist that may act to decrease alcohol intake in chronic
alcoholics without harmful side effects. It remains in the investigational phase at present. [108]
Tiapride
Tiapride, an atypical neuroleptic agent, is a selective dopamine D2 -receptor antagonist. It facilitates

management of alcohol withdrawal, but its use in patients at risk of severe reactions in acute withdrawal
should be accompanied by adjunct therapy for hallucinosis and seizures. The usefulness of tiapride in this
setting is likely to be limited. Tiapride ameliorates psychologic distress, improves abstinence, reduces
drinking behavior, and in the short-term facilitates reintegration within society. The potential risk of
tardive dyskinesia at the dosage employed (300 mg/day) requires evaluation and necessitates medical
supervision. Tiapride does not appear to cause physical or psychologic dependence.
1143
Figure 55-2 (Figure Not Available) Severity of signs and symptoms of alcohol withdrawal. (From Freedland ES,
McMicken DB. J Emerg Med 1993;11:605-618.).
TABLE 55-23 -- Ethanol Withdrawal Syndromes Times of Onset after Cessation of Drinking a

(Not Available)
a Adapted from Ellenhorn MJ In: Hall JB et al, eds. principles of critical care. New York: McGraw-Hill, 1992;2080-2093.
Toxicokinetics
Bioavailability of tiapride is about 75% following oral or intramuscular administration. Peak plasma
tiapride concentrations are achieved within about 0.4 to 1.5 hours. The drug is rapidly distributed and
does not bind appreciably to plasma proteins. Tiapride is mainly eliminated by renal excretion,
principally in the unchanged form. The elimination half-life is approximately 3 to 4 hours and may
increase with age and declining renal function (VD = 1.436 L/kg).
Clinical presentation
The most frequently reported adverse events (>1%) are drowsiness, extrapyramidal syndromes,
dizziness, and orthostatic hypotension.
Dosage
For the treatment of delirium or predelirium during alcohol withdrawal, intravenous or intramuscular
tiapride 400 to 1200 mg/day given every 4 to 6 hours is recommended, increased to 1800 mg/day if
required. [109]
Zimeldine
Blocks serotonin uptake, possibly leading to a decrease in daily intake of alcohol and an increase in days
of abstinence in chronic alcoholics. Zimeldine was withdrawn worldwide because of reports of hepatitis
and Guillain-Barre syndrome. [93]
TABLE 55-24 -- Equivalent, Potential Initial Doses of Benzodiazepines Frequently Used for Treatment
of Alcohol Withdrawal a
(Not Available)
a Adapted from Lohr MH. Mayo Clin Proc 1995;70:777-782.


WITHDRAWAL SYNDROMES (Fig. 55-2) (Figure Not Available) (See Table 55-23) (Table
Not Available)
Abstinence [110]
The majority of patients (over 95%) experiencing acute alcohol withdrawal probably do not require
psychotropic drug therapy. Studies with clonidine 0.2 mg orally given several times daily over a 4-day
period suggest that it is effective in reducing some of the adrenergic manifestations of alcohol
withdrawal. [111]
Additional studies with other drugs have suggested potential uses for dexamethasone, [112] phenobarbital,
[113] chlormethiazole, beta-blockers (for mild symptoms), [114] subanalgesic doses of nitrous oxide, [115]
clorazepate, [116] haloperidol, [111] and hydroxybutyric acid [101] in ameliorating some of the symptoms
associated with alcohol withdrawal. They may be useful as supplements to benzodiazepine therapy or for
patients resistant to benzodiazepines (Table 55-24) (Table Not Available) . Additional work is required
with each group of drugs before specific recommendations can be made.
1144


ALCOHOLIC SEIZURES
Seizures (Table 55-23) (Table Not Available)
Seizures may be classified in an alcoholic population as follows:

1. Solitary, convulsive seizure in alcoholics. No prior epileptic convulsions, no other epileptogenic
disease, no relation to withdrawal or massive intake of alcohol.
2. Convulsive seizure of alcoholics. Includes withdrawal seizures and seizures related to massive
alcohol intake.
3. Seizures in alcoholics with potentially epileptogenic disease such as head injury, idiopathic
epilepsy, cerebrovascular disease, and other drugs facilitating seizures.
4. Alcoholic epilepsy. Recurrent seizures in alcoholics; no history of epilepsy, potentially
epileptogenic diseases, withdrawal, or massive intake of alcohol.
Risk Factors
Factors associated with alcohol withdrawal and considered most likely to precipitate seizures are
hypoglycemia, hypomagnesemia, and respiratory alkalosis. Alcohol withdrawal also heightens photic
sensitivity and can lead to television-induced seizures.
A history of seizures before age 18 years or before the onset of heavy drinking is usually due to
idiopathic epilepsy. Alcohol withdrawal seizures appear 6 to 48 hours after either cessation or precipitous
decline of alcohol intake. The true alcohol withdrawal seizure will be manifest prior to the onset of
delirium tremens (DT). It is a generalized seizure and does not manifest an aura, a focal onset, or a
significant period (e.g., more than 30 minutes) of postictal confusion, agitation, or aggression. Persons
who differ from this pattern (i.e., have an aura, begin the seizure with a focal presentation, have onset
during DT, have an extended period of postictal confusion, suffer a second seizure) should be carefully
evaluated for other conditions. Alcohol withdrawal may exacerbate partial (focal) seizures common with
posttraumatic epilepsy. Partial seizures must be considered indicative of a mass lesion until proven
otherwise. Seizures in a setting of alcohol consumption or withdrawal will usually not require long-term
anticonvulsant therapy since the seizures are self-limited. [1]
Any patient arriving at an emergency department with seizures should be questioned about alcohol
intake. It is involved in up to 40% of adults with seizures admitted to a hospital and in about 15% of
patients with status epilepticus.
Seizures with alcohol use are dose-dependent and may be causal, independent of alcohol withdrawal.
Alcohol contributes to seizure frequency in the general epileptic population. This may be enhanced by
sleep deprivation, enhanced photic sensitivity, and accelerated metabolism of antiepileptic drugs due to
drinking alcohol. Sudden withdrawal of phenytoin may enhance the convulsive effects of alcohol
withdrawal.

Seizure treatment should include an intravenous line with 5% dextrose and saline solution; 100 mg
thiamine, 25 g dextrose, and 1.2 mg naloxone given intravenously to reverse Wernicke's syndrome.
Hypoglycemia or narcotic ingestion, metabolic disorders, toxic ingestion, infection, and structural
abnormality are ruled out by history, repeated physical examinations, laboratory data, and computed
tomography (CT) scans, if required.
Status Epilepticus
If an alcoholic develops status epilepticus (uncommon in alcohol withdrawal), an attempt is made to

terminate the seizure in <60 minutes to prevent irreversible brain damage. Oral airway or an endotracheal
tube is maintained and supplementary oxygen used. Two intravenous lines are started. After thiamine,
dextrose, and naloxone are given, a loading dose of phenytoin (13 to 18 mg/kg) is started in one line with
normal saline with a filter (0.22 or 0.45 mum) designed to remove phenytoin microcrystals that may form
when it is mixed in solution. An intravenous infusion pump is recommended. The flow rate is kept at 40
to 50 mg/minute, with caution in patients with preexisting heart disease. In a second line, diazepam is
given at 2 to 4 mg/minute up to 20 mg. Diazepam has a short duration of action (20 to 30 minutes) and is
given for immediate seizure control with caution in the elderly who are more vulnerable to respiratory
depression and hypotension; here, lorazepam may be useful. If seizures persist, a diazepam drip is begun,
with 50 to 100 mg diazepam diluted in 500 mL of 5% dextrose in water (D5 W) and run at 40 mL/hour. If
there is no response to phenytoin or diazepam, or if cardiac disease or phenytoin allergy preclude its use,
intravenous phenobarbital is given at 15 to 50 mg/minute after a loading dose of 7 to 20 mg/kg.
Phenobarbital should not be administered together with an intravenous diazepam drip because of their
potential incompatibilities in solution. By the time that phenobarbital is added, endotracheal intubation
and mechanical ventilation are invariably helpful.
For the alcoholic in withdrawal with no present seizure but a past history of withdrawal give either an
intravenous loading dose to those admitted or an oral loading dose of 19 to 20 mg/kg in two to three
divided doses over 6 hours with no more than 6000 mg/dose to outpatients. For withdrawal seizures, the
patient is maintained on phenytoin 300 mg/day for 5 days, but a prospective, randomized,
placebo-controlled study suggests that phenytoin is not effective in preventing withdrawal seizures. [117]
The alcoholic who has already had a single or short burst of seizures and is now alert may only need
treatment for alcohol withdrawal. Further seizures will be rare if adequately treated with a
benzodiazepine. The patient should be observed for at least 6 hours after the seizures before discharge.
Possible causes of a decreased level of consciousness following seizures in the obtunded patient include
postictal state, occult head trauma, unrecognized metabolic disorder, or poisoning.
Wernicke's-Korsakoff's Syndrome
Even when a Korsakoff's state is evident, use of thiamine may assist about 25 to 50% of patients in
making at least a partial recovery. [118]
1145
Treatment
Patients suspected of having Wernicke's encephalopathy should be treated immediately with 100 mg
thiamine daily, infused slowly in 500 mL fluid for at least 5 days. At the same time, deficiencies of other

vitamins, including niacin, minerals, electrolytes, and especially magnesium, should be corrected.
Intravenous glucose should be given only in conjunction with thiamine since the glucose alone can
precipitate Wernicke's encephalopathy in thiamine-deficient patients. Fortification of alcoholic beverages
with thiamine has been suggested. [1]
Transplantation
An extensive discussion of alcoholic liver disease (cirrhosis, hepatitis, encephalopathy) is outside the
scope of this book. Alcohol-associated progressive impairment of the liver may, however, develop
despite abstaining from alcohol ingestion. Hematemesis, advanced portal hypertension, hepatocellular
carcinoma, intractable ascites, or encephalopathy may provide reasons for transplantation, especially in
patients who have no serious disease of other organs, no history of alcohol dependence, excellent family
and social support, and an estimated length of survival of less than 1 year. [119] This remains a
controversial area of interest, [120] [121] but tends to favor serious consideration of employing this
procedure in selected alcoholic patients. [118] [122]
Children
Once a poisoning has occurred, the blood ethanol and glucose should be monitored and the child should
be treated with glucose if necessary. [123] A retrospective study of 102 cases suggests that children who by
history have ingested up to 105 mL of cologne, perfume, or after-shave (containing 15 to 99% ethanol)
and who remain asymptomatic can be observed at home if parents refuse to bring their asymptomatic
children to the emergency department. Home assessment must include extremely close hourly
observation for symptoms of central nervous system depression and hypoglycemia for at least 3 to 6
hours postingestion. [12] [14]
The Toxic Dose
In small children a blood ethanol concentration greater than 20 mg/dL may produce hypoglycemia. [124]
The average lethal concentration in adults is quoted as being 450 mg/dL. [125] The most common lethal
dose of ethanol reported in patients not receiving supportive therapy is 5 to 8 g/kg in adults and 3 g/kg in
children. [126]
Unlike in adults, poor nutritional status or a prolonged fast does not appear to be a prerequisite for
hypoglycemia to occur in children. Serious ethanol poisoning with hypoglycemia also results from
children ingesting mouthwash products. If a child consumes a volume of an ethanol-containing product
that can produce a blood ethanol concentration of 50 mg/dL, evaluate for immediate and delayed
hypoglycemia. [123]
The expected BEC can be calculated by the following formula:

Rearrangement of the above equation allows calculations of the amount ingested if the BEC and the
preparation's percent ethanol concentration is known. [123]


REFERENCES--ETHANOL
1. Ellenhorn MJ. The alcohols. In: Hall JB, Schmidt GA, Wood LDH eds. Principles of critical care. New York:
McGraw-Hill, 1992, 2080-2093.
2. MMWR. Update. Alcohol-related traffic fatalities--United States 1982-1993. JAMA 1994;272:1892.
3. GuppyA. At whole blood concentration should drunk-driving be illegal? Something lower than 17.4 mmol/L (0.8 g/L).
Br Med J 1994;308:1055-1056.
4. Bradley KA, Donovan DM, Larson EB. How much is too much? Advising parents about safe levels of alcohol
consumption. Arch Intern Med 1993;153:2734-2740.
5. Canfield
DV, Kupiec T, Huffine E. Postmortem alcohol production in fatal aircraft accidents. J Forensic Sci
1993;38:914-917.
6. CDC. Alcohol use and aquatic activities--United States, 1991. MMWR 1993;42:675-682.
7. CDC. Alcohol involvement in pedestrian fatalities--United States, 1982-1992. MMWR 1993;42:716-719.
8. CDC. Quarterly table reporting alcohol involvement in fatal motor vehicle crashes. MMWR 1993;42:729.
9. Jurkovich GJ, Rivara FP, Gurney JG, Fligner C, Ries R, Mueller BA, Copass M. The effect of acute alcohol intoxication
and chronic alcohol abuse on outcome for trauma. JAMA 1993;270:51-66.
10. Foulke FL. Alcohol beverages under investigation. FDA Consumer 1993;27(1):23.
11. Frezza
M, di Padova C, Pozzato G, Terpin M, Baraona E, Lieber CS. Higher blood alcohol levels in women. The role of
decreased gastric alcohol dehydrogenase activity and first-pass metabolism. N Engl J Med 1990;322:95-99.
12. Schenker
S, Speeg KV. The risk of alcohol intake in men and women. All may not be equal. N Engl J Med
1990;322:127-129.
13. Seitz HK, Egerer G, Simanowski UA. High blood alcohol levels in women. N Engl J Med 1990;323:58.
14. Lieber
CS. Biochemical and molecular basis of alcohol-induced injury to liver and other tissues. N Engl J Med
1988;319:1639-1650.
15. Holt S. Alcohol and H2 receptor antagonist: over the counter, under the table? Am J Gastroenterol 190;185:516-517.
16. Roine
R, Gentry T, Hernandez-Munoz R, Baraona E, Lieber CS. Aspirin increases blood alcohol concentrations in
humans after ingestion of ethanol. JAMA 1990;264:2406-2408.
17. Reference Deleted.

18. Bailey DN. Cocapropylene (propylcocaine) formation by human liver in vitro. J Anal Toxicol 1995;19:1-4.
19. Novy FG. Some higher analogues of cocaine. Am Chem J 1988;10:145-148.
1146
20. Harada S, Misawa S, Agarwal DP, Goedde HW. Liver alcohol and aldehyde dehydrogenase in the Japanese: isozyme
variation and its possible role in alcohol intoxications. Am J Hum Genet 1980;32:8-15.
21. JonesAW, Neiman J, Hillbom M. Concentration-time profiles of ethanol and acetaldehyde in human volunteers treated
with the alcohol-sensitizing drug calcium carbamide. Br J Clin Pharmacol 1988;25:213-221.
22. Brien JF, Loomis CW. Pharmacology of acetaldehyde. Can J Physiol Pharmacol 1983;61:1-22.
23. Schootstra
R, Bloemhof H, Bouma P, Uges DRA. An unusual case of acetaldehyde intoxication. In: Uges DRA, de
Zeeuw RA, eds. Proc Int Assoc Forensic Toxicol. 25th Meeting, Groningen, Netherlands: June 1988, pp. 85-91.
24. MillsJL, Graubard BI. Is moderate drinking during pregnancy associated with an increased risk for malformations?
Pediatrics 1987;80:309-314.
25. Karl
PI, Gordon BHJ, Lieber CS, Fisher SE. Acetaldehyde production and transfer by the perfused human placental
cotyledon. Science 1988;242:273-275.
26. Rosett HL. A clinical perspective of the fetal alcohol syndrome. Alcohol Clin Exp Res 1980;4:119-122.
27. Autti-Ramo I, Korkman M, Hilakivi-Clarke L, Lehtonen M, Halmesmaki E, Granstrom M-L. Mental development of
2-year-old children exposed to alcohol in utero. J Pediatr 1992;120:740-746.
28. CDC. Fetal alcohol syndrome--United States 1979-1992. MMWR 1993;42:339-341.
29. MennellaJA, Beauchamp GK. The transfer of alcohol to human milk. Effects on flavor and the infant's behavior. N
Engl J Med 1991;325:981-985.
30. Amadeno S, Abbou M, Fourcade ML, Waksman G, Leroux MG, Madec A et al. DZ dopamine receptor genes and
alcoholism. J Psychiatr Res 1993;27:173-179.
31. Noble EP, Blum K. Alcoholism and the D2 dopamine receptor gene. JAMA 1993;270:1547.
32. Gelernter J, Risch N, Goldman D. Alcoholism and the D2 dopamine receptor gene. JAMA 1993;270:1547-1548.
33. Allen
JP, Maisto Connors GJ. Self-report screening tests for alcohol problems in primary care. Arch Intern Med
1995;155:1726-1730.
34. Kitchens JM. Does this patient have an alcohol problem? JAMA 1994;272:1782-1787.
35. Engel
PA, Grunnet M, Jacobs B. Wernicke-Korsakoff syndrome complicating T-cell lymphoma: unusual or
unrecognized. South Med J 1991;84:253-256.
36. Fried RT, Levy M, Leibowitz AB, Bronster DJ, Iberti TJ. Wernicke's encephalopathy in the intensive care patient. Crit

Care Med 1990;18:779-780.
37. Da
Dalt L, Dall'Amico R, Lawerda AM, Chemollo C, Chiandelti L. Percutaneous ethyl alcohol intoxication in a
one-month-old infant. Pediatr Emerg Care 1991;7:343-344.
38. Watkins AMC, Keogh EJ. Alcohol burns in the neonate. J Paediatr Child Health 1992;28:306-308.
39. Hornfeldt
CS. A report of acute ethanol poisoning in a child: mouthwash versus cologne, perfume and after-shave. Clin
Toxicol 1992;30:115-121.
40. Lammingoa A, Vilska J. Acute alcohol intoxications in children treated in hospital. Acta Paediatr Scand
1990;79:847-854.
41. Kingston R, Saxona K, Sioris LJ, Lelwica T. Alcohol abuse in a 47-month-old child. Vet Hum Toxicol 1991;33:385.
42. Minocha A, Roberson DG, Herold DA et al. Impairment of cognitive and neuromuscular function by ethanol in social
drinkers. Vet Hum Toxicol 1985;28:319.
43. Elphinstone PE, Black AK, Greaves MW. Alcohol-induced urticaria. J Roy Soc Med 1985;78:340-341.
44. Thorton JR. Atrial fibrillation in healthy non-alcoholic people after an alcoholic binge. Lancet 1984;2:1013-1014.
45. Lang
RM, Borow KM, Neumann A et al. Adverse cardiac effects of acute alcohol ingestion in young adults. Ann Intern
Med 1985;102:742-743.
46. Criteria
Committee, National Council on Alcoholism. Criteria for the diagnosis of alcoholism. Am J Psychiatry
1972;129:129.
47. Morse RM, Flavin DK. Joint Committee of the National Council on Alcoholism. JAMA 1992;268:1012-1014.
48. NeimanJ, Lang AE, Fornazzari L, Carlen PL. Movement disorders in alcoholism: a review. Neurology
1990;40:741-746.
49. Gill
JS, Shipley MJ, Tsementzis SA, Hornby RS, Gill SK, Hitchcock ER et al. Alcohol consumption--a risk factor for
hemorrhagic and non-hemorrhagic stroke. Am J Med 1991;90:489-497.
50. Steel
JK, Cockcroft JR, Ritter JM. Blind drunk: alcoholic pancreatitis and loss of vision. Postgrad Med J
1993;69:151-152.
51. ChedidA, Mendenhall CL, Gartside P, French SW, Chen T, Rubin L, VA Cooperative Study Group. Am J
Gastroenterol 1991;86:210-216.
52. Sherlock S. Alcoholic liver disease. Lancet 1995;345:227-229.
53. Diamond I. Alcoholic myopathy and cardiomyopathy. N Engl J Med 1989;320:458-460.
54. Urbano-Marquez A, Estruch R, Navarro-Lopez F, Grau JM, Mont L, Rubin E. The effects of alcoholism on skeletal and
cardiac muscle. N Engl J Med 1989;220:409-415.
55. Alderman LE, Coltart DJ. Alcohol and the heart. Br Med Bull 1982;38:77-80.

56. Moushmoush B, Abi-Mansour P. Alcohol and the heart. The long-term effects of alcohol on the cardiovascular system.
Arch Intern Med 1991;151:36-42.
58. Knieriem HJ, Herbertz G. Electron-microscopic findings and photometric activation analytical results in experimental
cardiac insufficiency caused by cobaltous chloride. Virchows Arch B Cell Pathol 1969;2:32-46.
59. Munzinge W. Cardiomyopathy and arsenic intoxication. Arch Klin Med 1987;19:444.
60. Asokan SK, Witham AC. Myocardial malfunction of unknown cause. Cardiovasc Clin 1972;4:113-132.
61. Urbano-MarquezA, Estruch R, Fernandez-Sola J, Nicolas JM, Parc JC, Rubin E. The greater risk of alcoholic
cardiomyopathy and myopathy in women compared with men. JAMA 1995;274:149-153.
62. Day CP, James OFW, Butler TJ, Campbell RWF. QT prolongation and sudden cardiac death in patients with alcoholic
liver disease. Lancet 1993;341:1423-1428.
63. Abe H, Kawano Y, Jojima S et al. Biphasic effects of repeated alcohol intake on 24-hour blood pressure in hypertensive
patients. Circulation 1994;89:2626-2633.
64. Kei
N, Swales JD, Grobbee DE. Alcohol intake and its relation to hypertension. Cardiovasc Risk Factors
1993;33:189-200.
65. Kaplan NM. Alcohol and hypertension. Lancet 1995;345:1588-1589.
66. Felson
DT, Kiel DP, Anderson JJ, Kannel WB. Alcohol consumption and hip fractures. The Framingham Study. Am J
Epidemiol 1988;128:102-110.
67. Bikle
DD, Stesin A, Halloran B, Steinbach L, Recher D. Alcohol-induced bone disease: relationship to age and
parathyroid hormone levels. Alcohol Clin Exp Res 1993;17:690-695.
68. Rico H. Alcohol and bone mineral density. Br Med J 1993;307:939.
69. MacGregor RR. Alcohol and immune defense. JAMA 1986;256:1474-1479.
70. Laitinen
K, Lamberg-Allardt C, Tunninen R, Karonen S-L, Tahtela R, Ylikahri R et al. Transient hypoparathyroidism
during acute alcohol intoxication. N Engl J Med 1991;324:721-727.
71. Miller G. Magnesium deficiency syndrome. Compr Ther 1985;11:58-64.
72. Freedland
ES, McMichen DB. Alcohol-related seizures. II. Clinical presentation and management. J Emerg Med
1993;11:605-618.
73. DeMarchi S, Cecchin E, Basile A, Bertotti A, Nardini R, Bartoli E. Renal tubular dysfunction in chronic alcohol
abuse--effects of abstinence. N Engl J Med 1993;329:1927-1934.
1147

74. Adams SL, Mathews JJ, Flaherty JJ. Alcoholic ketoacidosis. Ann Emerg Med 1987;16:90-97.
75. ThompsonCJ, Johnston DG, Baylis PH, Anderson J. Alcoholic ketoacidosis: an underdiagnosed condition? Br Med J
1986;292:463-465.
76. Duffens K, Marx JA. Alcoholic ketoacidosis--a review. J Emerg Med 1987;5:399-406.
77. WrennKD, Slovis CM, Minion GE, Rutkowski R. The syndrome of alcoholic ketoacidosis. Am J Med
1991;91:119-128.
78. Schelling
JR, Howard RL, Winter SD, Linas SI. Increased osmolal gap in alcoholic ketoacidosis and lactic acidosis.
Ann Intern Med 1990;113:580-582.
79. Milas
AA, Tavassoli M. Laboratory markers of ethanol intake and abuse: a critical appraisal. Am J Med Sci
1992;303:425-428.
80. WaxPM, Hoffman RS, Goldfrank LR. Rapid quantitative determination of blood alcohol concentration in the
emergency department using an electrochemical method. Ann Emerg Med 1992;21:254-259.
81. Christopher TA, Zeccardi JA. Evaluation of the QED saliva alcohol test: a new rapid accurate device for measuring
ethanol in saliva. Ann Emerg Med 1992;21:1135-1136.
82. LopezGP, Yealy DM, Krenzelok EP. Survival of a child despite unusually high blood ethanol levels. Am J Emerg Med
1989;7:283-285.
83. Beckett
GJ, Hayes JD. Plasma glutathion-S-transferase measurements and liver disease in man. J Clin Biochem Nutr
1987;2:1-24.
84. Blake
J, Orrego H. Monitoring treatment of alcoholic liver disease: evaluation of various severity indices. Clin Chem
1991;37:5-13.
85. Chedid
A, Mendenhall CL, Gartside P, French SW, Chen T, Rabin L et al. Prognostic factors in alcoholic liver disease.
Am J Gastroenterol 1991;86:210-216.
86. Geller
RJ, Spyler PA, Herold DA, Bruns DE. Serum osmolal gap and ethanol concentration: a simple and accurate
formula. Clin Toxicol 1986;24:77-84.
87. AngeliP, Gatta A, Caregaro L, Luisetto G, Menon F, Merkel C et al. Hypophosphatemia and renal tubular dysfunction
in alcoholics. Are they related to liver function impairment? Gastroenterology 1991;100:502-512.
88. Naughton M, Grand J. Hypophosphatemia in a comatose alcoholic. Med J Aust 1991;155:723-724.
89. MacDonald
L, Kruse JA, Levy D, Marulendra S, Sweeny PJ. Lactic acidosis and acute alcohol intoxication. Am J
Emerg Med 1994;12:32-35.
90. JonesAW. Update on the concentration of acetone in blood from drinking drivers, Type-1 diabetes, out-patients and
healthy blood donors. Proc Am Acad Forensic Sci, 44th Annual Meeting, New Orleans: February 17-22, 1992; p 193.
91. Stibler
W. Carbohydrate-deficient transferrin in serum: a new marker of potentially harmful alcohol consumption
reviewed. Clin Chem 1991;37:2029-2093.
92. Koppel C, Muller C, Wrobel N. Diagnostic value of carbohydrate-deficient transferring for identifying patients with

drug overdose at risk of developing an alcohol withdrawal syndrome (Personal communication). 92a. Koppel C, Rosick U,
Bratter P. Influence of chronic alcohol abuse on selenium status in ICU patients. Ann Emerg Med 1995;26:723.
93. Donnelly
A, Paloucek F, Leikin J. Possible reversal of ethanol-induced respiratory depression of flumazenil. Vet Hum
Toxicol 1991;33:389.
94. Koch HP. The story of the anti-alcohol drug RO 15-4513. Int Pharmacol J 1988;2:85-86.
95. Littleton J. Alcohol intoxication and physical dependence. A molecular mystery tour. Br J Addict 1989;84:267-276.
96. NaranjoCA, Sellers EM, Wu PH, Lawrin MO. Moderation of ethanol drinking: role of enhanced serotonergic
neurotransmission. In : Naranjo CA, Sellers EM, eds. Research advances in new psychopharmacological treatments for
alcoholism. Amsterdam: Elsevier, 1985.
97. Cook C, Lipsedge M. Chlormethiazole and alcohol: a lethal cocktail. Br Med J 1987;294:1099.
98. AgricolaR, Mazarino M, Urani R. Treatment of acute alcohol withdrawal syndrome with carbamazepine: a
double-blind comparison with tiapride. J Int Med Res 1982;10:100-105.
99. Ballenger
JC, Post RM. Carbamazepine in alcohol withdrawal syndromes and schizophrenia psychoses.
Psychopharmacol Bull 1984;20:572-584.
100. YamPCI, Forbes A, Kox WJ. Clonidine in the treatment of alcohol withdrawal in the intensive care unit. Br J Anaesth
1992;68:106-108.
101. Gallimberti
L, Canton G, Gentile N, Ferri M, Cibin M, Ferrara SD et al. Gamma-hydroxybutyric acid for treatment of
alcohol withdrawal syndrome. Lancet 1989;2:787-789.
103. Flemenbaum A. Affective disorders and ``chemical dependence'': lithium for alcohol and drug addiction? Dis Nerv
Syst 1974;35:281-289.
104. Flemenbaum A. Lithium carbonate prophylaxis of alcoholism: its time has come. Arch Gen Psychiatry 1989;46:290.
105. FawcettJ, Aagesen CA, Tilkin JM, McGuire M, Clark DC, Pisani VD et al. Lithium carbonate prophylaxis of
alcoholism. Its time has come. Arch Gen Psychiatry 1989;46:290-291.
106. Baud FJ, Galliott M, Astier A, VuBien D, Garnier R, Likforman J, Bismuth C. Treatment of ethylene glycol poisoning
with intravenous 4-methylpyrazole. N Engl J Med 1988;319:97-100.
107. Naltrexone for alcohol dependence. Med Lett Drug Ther 1995;37(Issue 953):64-66.
108. Monti J, Alterwain P. Ritanserin decreases alcohol intake in chronic alcoholics. Lancet 1991;337:60.
109. Peters
DH, Vaulds D. Tiapride. A review of its pharmacology and therapeutic potential in the management of alcohol
dependence syndrome. Drugs 1994;47:1010-1032.
110. Lewis DC, Femind J. Management of alcohol withdrawal. Pharmacol Physicians 1982;16:1.
111. BaumgartnerGR, Rowen RC. Clonidine vs chlordiazepoxide in the management of acute alcohol withdrawal
syndrome. Arch Intern Med 1987;147:1223-1226.

112. Pol S, Nalpas B, Berthelot P. Dexamethasone for alcohol withdrawal. Ann Intern Med 1991;114:705-706.
113. Ives
TJ, Mooney AJ III, Gwyther RE. Pharmacokinetic dosing of phenobarbital in the treatment of alcohol withdrawal
syndrome. South Med J 1991;84:18-21.
114. Rosenbloom A. Emerging treatment options with alcohol withdrawal syndrome. J Clin Psychiatry 1988;49 (Suppl
12):28-31.
115. Gillman
MA, Lichtigfeld FJ. Analgesic nitrous oxide for alcohol withdrawal: a critical appraisal after 10 years' use.
Postgrad Med J 1990;66:543-546.
116. Haddox VG, Bidder TG, Waldron LE, Derby P, Achen SMW. Clorazepate use may prevent alcohol withdrawal
convulsions. West J Med 1987;146:695-696.
117. Fish
SS, Heeren T. Does phenytoin prevent alcohol withdrawal seizures: a meta analysis. Vet Hum Toxicol
1993;35:366.
118. Korsakoff's syndrome. Lancet 1990;336:912-913.
119. NeubergerJM. Transplantation for alcoholic liver disease contraindicated by alcohol dependence or extrahepatic
disease. Br Med J 1989;299:693.
120. Moss AH, Siegler M. Should alcoholics compete equally for liver transplantation? JAMA 1991;265:1295-1298.
121. CohenC, Benjamin M, and the Ethics and Social Impact Committee of the Transplant and Health Policy Center. Ann
Arbor, MI: JAMA 1991;205:1299-1301.
122. BirdGLA, O'Grady JG, Harvey FAH, Calne RY, Williams R. Liver transplantation in patients with alcoholic cirrhosis:
selection criteria and rates of survival and relapse. Br Med J 1990;301:15-17.
123. Vogel C, Caraccio T, Mofenson H, Hart S. Alcohol intoxication in young children. Clin Toxicol 1995;33:25-33.
124. Leung AK. Ethyl alcohol ingestion in children. A 15-year review. Clin Pediatr 1986;25:617-619.
125. Gibson PJ, Cant AJ, Mant TG. Ethanol poisoning. Acta Paediatr Scand 1985;74:977-978.
126. Redetzki
HM. Ethanol (Management Treatment Protocol): Poisindex Information System. Rumack BH, ed. Colorado:
Micromedex, Inc. Edition expires 8/30/92.
1148


ISOPROPYL ALCOHOL
Figure 55-3 (Figure Not Available) Pharmacokinetics of IPA and acetone (semilogarithmic scale). (From Vicas IMO,
Beck R. Clin Toxicol 1993;31:473-481.).
In hospitals isopropyl alcohol (IPA) is often colored with blue dye to distinguish it from many other clear
and colorless liquids; this has led to the designation blue heaven by abusers. [1]
TOXICOKINETICS
Absorption
Ingestion of 1 oz of 70% isopropyl alcohol (0.4 mL/kg) by volunteers led to peak serum isopropyl
alcohol concentrations of about 28 mg/dL in 30 minutes; peak serum acetone concentrations of about 34
mg/dL were not observed until approximately 4 hours after ingestion. Urine tests positive for acetone are
measurable by 3 hours postingestion. A positive urine test for acetone may still be present at 24 hours
postingestion. Serum acetone is measurable within 30 minutes postingestion (Fig. 55-3) (Figure Not
Available) . If no acetone is quantified by 30 minutes postexposure, exposure to isopropyl alcohol is
unlikely. [2] A large overdose may delay absorption. Skin absorption is probably relatively small, but
contributes to toxicity with prolonged contact.


LABORATORY
Blood Levels
A given isopropanol blood level is roughly twice as toxic as the same blood ethanol level.
Endogenous Isopropanol Formation
Isopropanol may be found in type I insulin-dependent acetonemic diabetes mellitus patients not exposed
to isopropyl alcohol who are hyperglycemic and usually acidotic, indicating that acetone may be
converted to isopropanol in physiologic conditions in which reduced nicotinamide adenine dinucleotide
is elevated. Isopropanol serum levels up to 29.7 mg/dL have been observed in such patients with acetone
levels up to 32.1 mg/dL.3 Chronic alcoholics, who also are known to produce elevated NADH
concentrations, may develop acetonemia (e.g., through starvation ketosis) and theoretically could
produce detectable concentrations of isopropyl alcohol. This remains to be confirmed.
Acetone (Fig. 55-3) (Figure Not Available)
Endogenous concentrations of acetone in the blood of healthy individuals range from about 0.1 to 0.5
mg/dL. [4] The highest concentration of acetone from 500 randomly selected blood specimens in one
series was 6 mg/dL. [5] Levels of 20 to 30 mg/dL are considered toxic. A lethal concentration of 55
mg/dL has been reported. [6] Others have reported acetone blood concentrations of over 200 mg/dL in
nonfatal cases of isopropanol ingestion. [5] [7]


CLINICAL CASE REPORT
Inhalation of 70% isopropyl alcohol by a neonate led to sedation, hypotension, cyanosis, bradycardia,
asystole, and death. [8] The isopropyl alcohol elimination half-life was 9.6 hours.


ABNORMALITIES
Osmolal Gap
Elevations of endogenous glycerol, acetone, and acetone metabolite levels may also be causes for an
increased osmolal gap in the alcoholic patient. Before alcohol therapy and/or hemodialysis is instituted in
patients with both an increased anion gap metabolic acidosis (Table 55-22) (Table Not Available) and
increased osmolar gap, alcoholic acidosis and lactic acidosis should be excluded. Any contribution of
ethyl alcohol to the increased osmolal and anion gap can be evaluated from an initial serum ethyl alcohol
level. Each 10 mg/dL of ethanol adds 2.3 mOsm/kg H2 O to the serum osmolality. Isopropyl alcohol may
increase the osmolal gap and induce ketosis since it is metabolized to acetone, but metabolic acidosis is
rare. [9]
1149


TREATMENT
Large doses of activated charcoal can absorb significant amounts of isopropanol and acetone. [10]


REFERENCES--ISOPROPYL ALCOHOL
1. Rich J, Scheife RT, Katz N, Caplan LR. Isopropyl alcohol intoxication. Arch Neurol 1970;417:322-324.
2. Lacoutre PG, Heldreth DD, Shannon M, Lovejoy FH Jr. The generation of acetonemia/acetonuria following ingestion of
a subtoxic dose of isopropyl alcohol. Am J Emerg Med 1989;7:38-40.
3. Bailey DN. Detection of isopropanol in acetonemic patients not exposed to isopropanol. Clin Toxicol 1990;28:459-466.
4. Levey S, Balchun OJ, Medrano V, Jung R. Studies of metabolic products in expired air. II. Acetone. J Lab Clin Med
1964;63:574-584.
5. Jones AW. Driving under the influence of isopropanol. Clin Toxicol 1992;30:153-155.
6. Stead
AN, Moffat AC. A collection of therapeutic, toxic and fatal blood drug concentrations in man. Hum Toxicol
1983;3:437-464.
7. Kelner M, Beuley DN. Isopropanol ingestion: interpretation of blood concentrations and clinical findings. J Toxicol Clin
Toxicol 1983;20:497-507.
8. Vicas IMO, Beck R. Fatal inhalational isopropyl alcohol poisoning in a neonate. Clin Toxicol 1993;31:473-481.
9. Braden GL, Strayhorn CH, Germain MJ, Mulhern JG, Skutcher CL. Increased osmolal gap in alcoholic acidosis. Arch
Intern Med 1993;153: 2377-2380.
10. Burkhart
KK, Martinez MA. The absorption of isopropanol and acetone by activated charcoal. Clin Toxicol
1992;30:371-375.


METHANOL
USES
An anecdotal report suggest that methanol poisoning may follow intentional ``sniffing.'' [1]


Initial Presentation
Clinical Effects
A patient survived a serum methanol level of 493 mg/dL without loss of eyesight following aggressive
therapy with an ethanol drip, bicarbonate, and hemodialysis. [2]
Gastrointestinal Tract
Methanol is a mucosal irritant and produces nausea, vomiting, and abdominal pain in over one half of
cases. Absence of gastrointestinal symptoms does not rule out serious toxicity. Pancreatitis, as defined by
elevated serum amylase, occurs commonly, [3] appearing in two-thirds of a recent series of cases. [4]
Hemorrhagic pancreatitis may appear on autopsy. Elevation of hepatic aminotransferases usually is mild
and transient.


LABORATORY
Analytic Methods
A modified headspace gas chromatographic method for analysis of formate in blood has a limit of
detection of 2.5 mg/dL. Ocular toxicity may correlate better with formate concentration than with
methanol concentration. [5]
Abnormalities
Acidosis.
Blood Levels
Serum and urine formate levels do not appear to be good biologic markers of methanol intoxication. [6]
A patient fell into a vat of furniture finish stripping solution. Three hours later his peak methanol level
was 247 mg/dL. With intravenous ethanol, bicarbonate, folate, gastric lavage, and hemodialysis the blood
methanol dropped to 4 mg/dL 43 hours later. The patient survived. [7]
A 6-week-old infant was fed Similac infant formula accidentally diluted with a methanol-containing
windshield washer fluid. The infant, who appeared normal, was treated with activated charcoal and folic
acid and developed a serum methanol level of 45.6 mg/dL (14.2 mmol/L) on the third day. No abnormal
ophthalmologic fundus optic changes or severe metabolic acidosis developed, and no ethanol therapy
was instituted. The long half-life of methanol in this case (28 hours) was probably due to the relative
inactivity of alcohol dehydrogenase in this age group. Formate levels were not measured. [8]
Urine Levels
There is a correlation between occupational exposure to methanol vapor and levels of methanol measured
in shift-end urine samples. Levels of about 42 mg methanol/liter of urine are excreted in a shift-end urine
sample following 8 hours of exposure to methanol at 200 ppm (current permissible limit). [9]
Formate Levels--Predictive Value
Criteria predictive of severe methanol poisoning possibly leading to permanent sequelae can include: (a)
an interval between ingestion and treatment exceeding 10 hours and (b) blood formate levels about 0.5
g/L (or 11.1 mmol/L) [7] [10] (Fig. 55-4) (Figure Not Available) .
Ancillary Tests
Hypomagnesemia occurs following ethyl alcohol and methanol ingestion, diuretic therapy, and
sympathomimetic use. [11] Hypokalemia may be due to the formation of potassium formate. In the
presence of a metabolic acidosis associated with hypokalemia, methanol poisoning should be considered.
[12]

Reduction in pH will not begin to occur before 6 hours after ingestion. Plasma bicarbonate levels and
percent change in plasma bicarbonate levels correlate poorly with the time after ingestion.12a
An anecdotal study suggests that magnetic resonance imaging was useful in the evaluation of a patient
with methanol-induced toxic optic neuropathy. [13]
1150
Figure 55-4 (Figure Not Available) Metabolic pathways involved in methanol metabolism and relationship with folate
metabolism. THF, tetrahydrofolate. (Adapted from Kruse JA. Intensive Care Med 1992;18:292-297.).


TREATMENT
Stabilization
If the patient is asymptomatic and methyl alcohol ingestion is suspected, perform gastric lavage with
activated charcoal. Obtain serum methanol and arterial blood gas.
Chronic Alcoholics
Chronic alcoholics on drinking bouts may often drink solutions containing both methanol and ethanol
and exhibit no signs of a formate-induced metabolic acidosis on admission in spite of high methanol and
ethanol blood levels. The methanol content of 20 commercial wines ranged from 5.0 to 32.5 mg/dL, [14]
and the level in 24 distilled liquors ranged from 1.3 to 10.6 mg/dL. Therefore consumption of large
amounts of wine or liquor theoretically can result in detectable levels of serum methanol. [15] [16]
Such patients may not require hemodialysis. The diagnosis and treatment of combined methanol and
ethanol poisoning should be based on the case history, clinical signs, and the presence of a metabolic
acidosis, not on blood methanol concentrations alone. [17]
Elimination Enhancement
Consideration should be given to the reduction of hemorrhage complicating brain necrosis by performing
hemodialysis without heparinization, using an artificial kidney with a biocompatible membrane such as
polymethylmethacrylate and an albumin coating. [18]
Forced diuresis is not effective, but hemodialysis effectively removes methanol (100 to 200 mL/min
clearance), as well as formaldehyde and formic acid. [19] Hemoperfusion removes neither methanol nor
formate well. [20] Although peritoneal dialysis increases methanol clearance, hemodialysis is about eight
times more effective. [21] Indications for dialysis procedures include the following:
1. A peak methanol level over 50 mg/dL is recommended in the medical literature, [22] but the exact
level is debatable. Dialysis does reduce the prolonged intensive care time required for ethanol
therapy at methanol levels above 50 mg/dL.
2. Metabolic acidosis is not immediately correctable with bicarbonate therapy. High formate levels
(i.e., over 20 mg/dL) suggest the need for hemodialysis.
1151
3. Any visual impairment.

4. Renal failure.
Dialysis may be stopped when the methanol level falls below 25 mg/dL. Remember that ethanol also is
dialyzed and therefore maintenance levels must be increased during dialysis.
Antidotes

Administration of ethanol blocks the formation of formaldehyde and formic acid because of the
preferential affinity of ethanol for alcohol dehydrogenase. Ethanol levels should be maintained between
100 and 150 mg/dL to completely inhibit toxic metabolite formation. Average dosages necessary to
maintain a blood ethanol concentration of 100 mg/dL in a 70-kg patient are listed in Table 55-21 (Table
Not Available) .
4-MP (4-methyl-pyrazole) exhibits nonlinear elimination kinetics and probably induces its own
metabolism. This may make it difficult to establish a safe dosage regimen with the drug. Multiple dosing
with 4-MP may cause transient hepatotoxicity. Further work will be required to determine the safe use of
4-MP. [23]
Ethanol Administration
Intravenous administration is more reliable than oral administration, but ethyl alcohol is irritating to
veins. An intravenous solution of 10% ethanol in D5 W is optimal. Note that maintenance infusion must
be increased during dialysis. Blood must be drawn frequently before, during, and after dialysis until a
steady-state ethanol level is confirmed. Continue ethanol infusion until the methanol level falls below the
range of 20 to 25 mg/dL. Ethanol prolongs the elimination half-life of methanol to 24 to 30 hours; hence
several days may be required to reduce the methanol level below 25 mg/dL when hemodialysis is not
used.
Ethanol Indications
1. Peak methanol level over 20 mg/dL.

2. Any patient with a history of ingestion of 0.4 mL/kg or any symptomatic patient should receive
ethanol pending confirmatory blood methanol levels.
3. Acidosis.
4. Any patient considered for hemodialysis.
5. Palatnick and colleagues suggest that ethanol may prolong the risk for toxicity and potential
complications. Hemodialysis should be considered for methanol poisoned patients who are treated
with ethanol infusions. [24]
Supportive Care
1. If the methyl alcohol level is below 40 mg/dL and the blood pH is normal, further alcohol
administration or hemodialysis is usually not necessary.
2. If the methyl alcohol level is about 40 to 50 mg/dL and the blood pH is normal, use either a
continuous IV alcohol infusion with frequent monitoring of both methanol and ethanol serum
levels (requires prolonged hospitalization) or use IV alcohol and hemodialysis, which will usually
remove the methanol in about 5 hours and decrease hospitalization.
3. If the methyl alcohol level is above 50 mg/dL and the blood pH is normal, use IV alcohol and
begin hemodialysis.
4. If the blood pH supports a metabolic acidosis with normal lactic acid levels, the concentration of
methanol is not a determining factor, and the patient should be treated with IV ethyl alcohol and

hemodialysis to remove formaldehyde and formic acid from the blood. [25]
5. Frequent assessment of vital signs (hourly) until stable. Watch for variations in blood pressure,
hypothermia, tachycardia, arrhythmias, cyanosis, and dyspnea.
6. Measurements of ethanol and methanol levels and calculation of anion and osmolar gaps provide
an estimate of the elimination rate. [26]
7. Folic acid probably accelerates the detoxification of the toxic metabolite, formic acid.


REFERENCES--METHANOL
1. McCormick MJ, Mogabgab E, Adams SL. Methanol poisoning as a result of inhalational solvent abuse. Ann Emerg Med
1990;19:639-642.
2. Pamies RJ, Sugar D, Rives L, Herold AH. Methanol intoxication. Case report. J Fla Med Assoc 1993;80:465-467.
3. BennettJL, Cary FH, Mitchell GL et al. Acute methyl alcohol poisoning: a review based on experiences in an outbreak
of 323 cases. Medicine 1953;32:431-463.
4. Swartz RD, Millman RP, Billi JE et al. Epidemic methanol poisoning: clinical and biochemical analysis of a recent
episode. Medicine 1981;60:373-382.
5. Fraser
AD, MacNeil W. Gas chromatographic analysis of methyl formate and application in methanol poisoning cases. J
Anal Toxicol 1989;13:73-76.
6. D'Alessandro A, Osterloh J, Chumers P, Quinlan P, Kell T, Becker C. Formate in serum acid urine following controlled
methanol exposure at the threshold limit value. Vet Hum Toxicol 1993;35:358.
7. KellerK, Pearigen PD, Olsen KR. Severe methanol poisoning and chemical burn after submersion in a furniture
stripping solution. Vet Hum Toxicol 1991;33:366.
8. Brent J, Lucas M, Kulig K, Rumack BH. Methanol poisoning in a 6-week-old infant. J Pediatr 1991;118:644-646.
9. Kawai T, Yasugi T, Mizunama K, Horiguchi S, Hirase Y, Uchida Y, Ikeda M. Methanol in urine as a biological indicator
of occupational exposure to methanol vapor. Int Arch Occup Environ Health 1991;63:311-318.
10. MahieuP, Hassoun A, Lauwerys R. Predictors of methanol intoxication with unfavourable outcome. Hum Toxicol
1989;8:135-137.
11. Harchelroad F. Hypomagnesemia during methanol intoxication. Vet Hum Toxicol 1993;35:364.
12. HassounA, Mahieu P, Lauwerys P. Hypokalemia in acute methanol poisoning. Proc Eur Assoc Pois Cont Clin Toxicol,
Birmingham, UK: May 1993.12a. McGuigan MA. Analysis of the temporal development of acidosis in uncomplicated
methanol poisoning. Ann Emerg Med 1995;26:725.
13. Bernstein
JM, McNally J, Boyer L. Magnetic resonance imaging of methanol-induced optic nerve toxicity. Vet Hum
Toxicol 1993;35:365.
14. Carroll RB. Analysis of alcoholic beverages by gas-liquid chromatography. QJ Stud Alcohol 1970; 5(Suppl)6-19.
15. Tintinalli JE. Serum methanol in the absence of methanol ingestion. Ann Emerg Med 1995;26:393.
16. Lee
CY. Acree TE, Butts RM. Determination of methyl alcohol in wine by gas chromatography. Anal Chem
1975;47:747-748.

17. Martensson E, Olofsson U, Heath A. Clinical and metabolic features of ethanol-methanol poisoning in chronic
alcoholics. Lancet 1988;1:327-328.
1152
18. Phang
PT, Passerini L, Mialke B, Berendt R, King EG. Brain hemorrhage associated with methanol poisoning. Crit
Care Med 1988;16:137-140.
19. McCoy HG, Cipolle RJ, Ehlers SM et al. Severe methanol poisoning: application of a pharmacokinetic model for
ethanol therapy and hemodialysis. Am J Med 1979;67:804-807.
20. WhalenJE, Richards CJ, Ambre J. Inadequate removal of methanol and formate using the sorgent based regeneration
hemodialysis delivery system. Clin Nephrol 1979;11:318-321.
21. Settler
JG, Singh R, Brackett NC et al. Studies on the dialysis of methanol. Trans Am Soc Artif Intern Organis
1967;13:179-182.
22. Gonda A, Gault H, Churchill D et al. Hemodialysis for methanol intoxication. Am J Med 1978;64:749-758.
23. Jacobsen
D, McMartin KE. Methanol and ethylene glycol poisoning: 4-methylpyrazole or ethanol. Proc Intern Cong
Eur Assoc Poison Control Centres, Milan, Italy: September 25-29, 1990; p. 142.
24. Palatnick
W, Redman LW, Sitar DS, Tenenbein M. Methanol half-life during ethanol administration: implications for
management of ethanol poisoning. Ann Emerg Med 1995;26:202-207.
25. McCarron MM. Methyl alcohol. September, 1993 (Personal communication).
26. King ML. Acute methanol poisoning: a case study. Heart Lung 1992;21:260-264.


ETHYLENE GLYCOL
Toxic Dosage
The approximate minimum lethal dose is 1 to 1.5 mL/kg or approximately 100 mL in an adult. Persons
who attempted suicide by ingesting 1 and 2 L and who were treated within 1 hour have survived. [1] [2]


TOXICOKINETICS
Absorption
Ethylene glycol is rapidly absorbed orally but not by lung or dermal routes. Peak levels occur 1 to 4
hours postingestion.
Distribution
Since ethylene glycol is highly water soluble, it distributes evenly throughout body tissue.
Elimination
The renal glomeruli filter and then passively reabsorb most of the absorbed ethylene glycol dose. [3]
Approximately 20% of a dose of 1 mg/kg is excreted unchanged; less than 1% of the ethylene glycol is
metabolized to oxalic acid at this dose. The liver oxidizes ethylene glycol primarily to glycoaldehyde,
glycolate, and then glyoxylate.
The metabolism of glyoxylate follows several pathways that depend on the cofactors thiamine and
pyridoxine. The oxidation of ethylene glycol to glyoxylate and subsequently to oxalate requires the
conversion of NAD to NADH. The altered NAD/NADH ratio shifts pyruvate to lactate and thereby helps
produce lactic acidosis. The acidic metabolites are more toxic than the parent compound. The order of
toxicity appears to be glyoxylate > glycoaldehyde > ethylene glycol. [4]
The plasma half-life of ethylene glycol is approximately 3 to 5 hours. At ethanol levels of 100 to 200
mg/dL, the half-life of ethylene glycol is prolonged to 17 hours because of the 100-times-greater affinity
of ethanol for alcohol dehydrogenase. [5]


PATHOPHYSIOLOGY
Ethylene glycol produces roughly the same CNS depression as ethanol, but ethylene glycol produces
toxic metabolites. The metabolic acidosis and anion gap result primarily from glycolic acid formation
and some lactic acid formation. [6] The inhibition of the citric acid cycle resulting from reduced
NAD/NADH ratios and formation of oxalic acid contribute, to a limited extent, to the metabolic acidosis.
Oxalate formation produces myocardial depression and acute tubular necrosis, although the exact
mechanism is unclear since only a small amount of oxalate is formed. Glycoaldehydes, glycolic acid, and
glyoxylic acid may contribute to CNS depression and may contribute to renal toxicity by producing renal
edema; however, McChesney and colleagues [7] and Clay and Murphy [8] have shown that little oxalic
acid, glycoxylic acid, glycoaldehyde, or formic acid is found during ethylene glycol intoxication.
Hypocalcemia may result from chelation of oxalate, [1] although there are scant data available to support
this hypothesis.


The classic three-stage presentation depends on the amount of severity of ingestion. [9] Hepatic damage
usually is minimal.
Stage 1: CNS Depression (1 to 12 Hours Postingestion)
Transient exhilaration occurs without the odor of ethanol. Gastrointestinal complaints include primarily
nausea and vomiting. Acidosis, coma, convulsions, and myoclonic jerks also may be present. The optic
fundus is usually normal, although the occasional presence of papilledema may confuse the clinical
presentation with that of methanol. Nystagmus and ophthalmoplegias may appear. Cerebral edema
secondary to cytotoxic damage and calcium oxalate deposition synergistically depress CNS activity in
severe poisoning.
Stage 2: Cardiopulmonary Symptoms (12 to 24 Hours Postingestion)
Tachycardia, tachypnea, and mild hypertension often occur. Congestive heart failure and circulatory
collapse are seen in severe ingestions.
Stage 3: Renal Stage (24 to 72 Hours Postingestion)
This stage is characterized by oliguria, flank pain, acute tubular necrosis, renal failure, and rarely bone
marrow arrest. [10] Renal damage may be permanent.
Ethylene glycol toxicity is suggested by the following: ethanol-like intoxication with no odor,
large-anion-gap acidosis and coma, osmolal gap, calcium oxalate crystals, and mental status changes.
1153


ACUTE INTOXICATION (See Table 55-1) (Table Not Available)


CHRONIC INTOXICATION
Cranial Nerve Deficit and Peripheral Nerve Deficits
Multiple cranial nerve deficits can develop 1 to 2 weeks after ethylene glycol ingestion. [11] Cranial nerve
deficits may follow within 1 to 3 weeks the ingestion of generally over 100 mL of ethylene glycol in
spite of aggressive therapy (hemodialysis with correction of fluid and acid-base abnormalities) of the
underlying acute poisoning. [11] Most patients exhibit a bilateral facial nerve (VII) paralysis, although
hearing loss, dysarthria, dysphagia, anisocoria, and blurred vision may also be present. [12] The deficit
may be permanent. [3] Several weeks after the cranial nerve deficits have been established, a
demyelinating sensorimotor peripheral neuropathy may develop with proximal muscle weakness,
stocking-glove sensory loss, and areflexia. [13]


LABORATORY
Abnormalities
Urine
Although oxalate normally is a minor metabolic product of ethylene glycol metabolism, urinary oxalate
crystals are a common, but not invariable, feature of ethylene glycol intoxication (Figs. 55-5 (Figure Not
Available) and 55-6) (Figure Not Available) . [14] There are two forms of urinary calcium oxalate
crystals: the octahedral or tent-shaped form of the dihydrate crystals, and the prism or dumbell-shaped
monohydrate form. [15] The latter form is stable under normal physiologic conditions; the dihydrate form
appears only during high urinary calcium and oxalate concentrations, as seen in ethylene glycol
poisoning. The dihydrate form can transform into the monohydrate form. [16]
Analytic Methods
2,3-Butanediol
Ethylene glycol determinations frequently are performed by precipitation of plasma proteins with
acetonitrile, formation of the cyclic phenylboronate ester derivative of ethylene glycol, and analysis of
gas chromatography with OV-17 as the stationary phase. Alcoholics may drink industrial alcohol
preparations that contain 2-butanone. This ketone may be converted in the liver to 2,3-butanediol, which
has an identical retention time to ethylene glycol and may be mistakenly reported as ethylene glycol.
This problem may be easily corrected by changing the stationary phase from OV-17 to, for example,
SE-30. [17]
Propylene Glycol
Propylene glycol (PG) has been used as an internal standard for quantifying ethylene glycol by gas
chromatography in serum. Intravenous administration of drug formulations containing PG (e.g.,
phenytoin--Dilantin, diazepam--Valium) may lead to potentially serious underestimations of ethylene
glycol concentrations in serum. [18]
Urine
Check the urine each hour for at least 5 hours after ingestion before ethylene glycol intoxication is ruled
out for consideration. If urine oxidate crystals are seen and a second 1-hour urine specimen also shows
calcium oxalate crystals, begin IV alcohol and hemodialyze (Fig. 55-6) (Figure Not Available) .
Some commercial ethylene glycol antifreeze products contain sodium fluorescein as a colorant to aid in
the detection of automobile cooling system leaks. The urine of a patient suspected of an antifreeze
ingestion may disclose visually detectable fluorescein under a Wood's lamp in the first several hours after
ingestion. This observation must be confirmed by appropriate quantitative tests. [19]
In anuric patients, irrigation of the urinary bladder with 50 to 100 mL of saline, centrifugation of the

irrigant, and examination of the sediment for calcium oxalate crystals may increase the chances of
detection of calcium oxalate crystalluria in a suspected ethylene glycol ingestion. [20]
Blood
Hypocalcemia may occur and is manifested by QT prolongation on the ECG; tetany may result.
Myalgias, elevated serum creatinine levels, and increased serum creatine phosphokinase levels may be
seen. [6] Anion-gap metabolic acidosis indicates the production of organic acids. Toxicity may occur
without significant elevation of the osmolal gap. Serious ethylene glucol toxicity (50 mg/dL) produces an
approximate rise in the osmolal gap of 10 mOsm. The osmolal and anion gaps may remain elevated in
spite of low serum ethylene glycol levels because of the accumulation of the blood glycolate. [21]
Serum Levels
Quantitative levels require gas chromatography and are not routinely included on toxicology screens.
Serum levels exceeding 50 mg/dL suggest the need for hemodialysis. Recovery has occurred with
aggressive treatment in the presence of ethylene glycol levels of 145 mg/dL (8-hour level) and 560
mg/dL (1-hour level). [2] [22] Reported serum ethylene glycol levels in survivors ranged up to 650 mg/dL,
whereas levels between 98 and 775 mg/dL were reported in fatalities. [2] [5] [23] [24] Glycolic acid and
bicarbonate levels correlated better than serum ethylene glycol levels with the clinical picture, since the
former two levels reflect the action of the toxic metabolites.


TREATMENT
Stabilization
Do not wait for symptoms to appear before treatment. Time elapsed between ingestion (fatalities reported
if longer than 12 hours) and treatment and the dose ingested are major predictive factors of morbidity
fatality. [25]
Elimination Enhancement
A primary indicator for hemodialysis (HD) is a significant metabolic acidosis with a pH less than 7.15.
Patients with very high concentrations of ethylene glycol (EG) (over 60 mmol/L or 373 mg/dL) without
acidosis may also be
1154
Figure 55-5 (Figure Not Available) Calcium monohydrate crystals. Left, 400; right, scanning electron micrograph,
1,200; white arrow indicates a point of x-ray fluorescence. (From Terlinsky AS, Grochowski J, Geoly KL et al. Am J Clin
Pathol 1981;76:224-225.).
Figure 55-6 (Figure Not Available) Calcium dihydrate crystals, 400. (From Terlinsky AS, Grochowski J, Geoly KL et
al. Am J Clin Pathol 1981;76:224-225.).
candidates for HD to shorten the course of treatment and reduce the risk of renal complications. HD
should be carried out at EG levels as low as 50 mg/dL (10 mmol/L) if the patient is anuric. During HD
there is no need for ethanol administration intravenously or in the dialysate since glycolate is effectively
eliminated. After termination of HD ethanol may be administered (IV boluses of 500 mL 5% glucose
with 10% ethanol over 20 to 30 minutes, followed by a continuous infusion of 70 to 100 mL/hour).
Monitor serum ethanol every hour until a constant level of 20 to 30 mmol/L (92 to 135 mg/dL) is
obtained. Ethanol is continued until the serum concentration of EG is less than 10 mmol/L (about 60
mg/dL). [26]
Glycolate
Glycolate has a low volume of distribution (0.5 to 0.6 L/kg) and a low molecular weight and is rapidly
cleared by hemodialysis. Careful monitoring can proceed with hourly acid-base balance without ethanol
administration since glycolate assays are often not easily available. [26] [27] Glycolate appears to be the
causative factor for acidosis in ethylene glycol poisoning. It is effectively reversed by hemodialysis. [28]
Antidotes
Ethanol
If the urine fluoresces under a Wood's lamp, begin intravenous alcohol and hemodialyze the patient.
However, if the

1155
serum concentration of ethylene glycol (EG) is high (over 30 mmol/L to 180 mg/dL) without metabolic
acidosis, there is no immediate indication for hemodialysis. Ethanol will effectively inhibit the
metabolism of EG, so that HD may be deferred until staff and equipment are available. During HD
glycolate (low volume of distribution--0.5 to 0.6 L/kg, low molecular weight) is effectively eliminated. If
the patient is admitted with metabolic acidosis, ethanol treatment and HD on an emergency basis is
indicated to remove glycolate as soon as possible. [26]
Supportive Management
Margaret McCarron has presented a useful regimen for the management of ethylene glycol poisoning:
1. Ethylene glycol serum concentrations are not performed in most hospital laboratories; a specimen
should be taken and sent to a contract lab.Results will be obtained in 3 to 5 days; therefore
ETHYLENE GLYCOL SERUM LEVELS are NOT USEFUL IN DETERMINING ACUTE
TOXICITY. They should be obtained for later confirmation of the diagnosis.
2. Gastric lavage with charcoal for any suspected case. The diagnosis of ethylene glycol poisoning
should be made by the history and examination of the urine.
. Waiting for symptoms to appear before treatment is too late, because the symptoms are due
to the metabolic abnormalities that often cannot be adequately treated.
b. Examination of the urine is used to substantiate the diagnosis. Recommend that urine be
checked every hour if negative, for at least 5 hours after the ingestion, before ethylene
glycol intoxication is ruled out.
1. Check urine for calcium oxalate crystals. If next specimen in 1 hour shows more
crystals: IV ALCOHOL AND HEMODIALYZE.
2. Check urine under Wood's lamp. If antifreeze has been taken, the urine will fluoresce
from the fluorescent dye in the product. If urine fluoresces: IV ALCOHOL AND
HEMODIALYZE.
3. If the patient enters the emergency room with metabolic acidosis and is in coma:
1. Treat with bicarbonate for pH 7.20 or less.
2. IV alcohol infusion.
3. Hemodialysis. Note: Addition of 95% ethanol to dialysate is necessary to replace the ethanol
lost during the procedure.
4. Pyridoxine 50 mg and Thiamine 100 mg IM q.i.d. 2 days.
5. Monitor fluid and electrolytes, especially calcium and magnesium.
4-Methylpyrazole
4-Methylpyrazole has been shown to reduce blood glycolate levels after ethylene glycol ingestion and
appears to be effective after both oral and intravenous administration. The product has not yet received
FDA approval in the United States. [29] [30] [31]
Continuous arteriovenous hemofiltration dialysis may be an alternative when hemodialysis and
4-methylpyrazole therapy are not available. This observation has not been validated by controlled clinical

trials. [32]
Supportive Care
1. Follow electrolyte fluid balance carefully. Renal clearance of ethylene glycol is inversely related to
water absorption; therefore maintenance of good urine volumes is necessary to enhance urinary
elimination.
2. Monitor serum calcium level and replace as indicated with 10% calcium gluconate intravenously.
3. Follow arterial pH and correct pH below 7.2 with intravenous bicarbonate.


REFERENCES--ETHYLENE GLYCOL
1. Scully R, Galdabini J, McNealy B. Case records of the Massachusetts General Hospital. Discussion of Levinsky NG.
Case 38:1979. N Engl J Med 1979;301:650-657.
2. Stokes JB, Averon F. Prevention of organ damage in massive ethylene glycol ingestion. JAMA 1980;243:2065-2066.
3. Von Oettinggen VF. Ethylene glycol. US Public Health Bull 1943;281:166-174.
4. Beasley VR, Buck WB. Acute ethylene glycol toxicoses: a review. Vet Hum Toxicol 1980;22:255-263.
5. Peterson DC, Collins AJ, Himes JM et al. Ethylene glycol poisoning: pharmacokinetics during therapy with ethanol and
hemodialysis. N Engl J Med 1981;304:21-23.
6. Gabow PA, Clay K, Sullivan JB et al. Organic acids in ethylene glycol intoxication. Ann Intern Med 1986;105:16-20.
7. McChesney EW, Goldberg L, Parekh CK et al. Re-appraisal of the toxicology of ethylene glycol. II. Metabolism studies
in laboratory animals. Food Cosmet Toxicol 1971;9:21-38.
8. Clay KL, Murphy RC. On the metabolic acidosis of ethylene glycol intoxication. Toxicol Appl Pharmacol
1977;39:39-49.
9. Brown CG, Trumbull D, Klein-Schwartz W et al. Ethylene glycol poisoning. Ann Emerg Med 1983;12:501-506.
10. Bobbit
WH, Williams RM, Freed CR. Severe ethylene glycol intoxication with multisystemic failure. West J Med
1986;144:225-228.
11. Spillane
L, Roberts JR, Meyer AE. Multiple cranial nerve deficits after ethylene glycol poisoning. Ann Emerg Med
1991;20:208-210.
12. Factor
SA, Lava NS. Ethylene glycol intoxication: a new stage in the clinical syndrome. NY State J Med
1987;87:179-180.
13. ThomasD, Claussen G, Suggs S, Oh SJ, Joy JL. Ethylene glycol-induced peripheral neuropathy. Neurology
1990;40(Suppl 1):344.
14. Turk J, Morrell L, Avioli LV. Ethylene glycol intoxication. Arch Intern Med 1986;146:1601-1603.
15. TerlinskyAS, Grochowski J, Geoly KL et al. Identification of atypical calcium oxalate crystalluria following ethylene
glycol ingestion. Am J Clin Pathol 1981;76:223-226.
16. Burns
JR, Finalyson B. Changes in calcium oxalate crystal morphology as a function of concentration. Invest Urol
1980;18:174-177.
17. Jones AW, Nilsson L, Gladh SA, Karlsson K, Beck-Friis J. 2,3-Butanediol in plasma from an alcoholic mistakenly

identified as ethylene glycol by gas-chromatographic analysis. Clin Chem 1991;37:1453-1455.
18. LeGatt
DF, Tisdell RH. Ethylene glycol quantifications. Avoid propylene glycol as an internal standard. Clin Chem
1990;36:1860-1861.
19. WinterML, Ellis MD, Snodgrass WR. Urine fluorescence using a Wood's lamp to detect the antifreeze additive sodium
fluorescein: a qualitative adjunctive test in suspected ethylene glycol ingestion. Ann Emerg Med 1990;19:663-667.
20. Goodkin DA. Ethylene glycol poisoning. Am J Med 1990;88:201.
1156
21. HewlettTP, McMartin KE. Ethylene glycol poisoning: the value of glycolic acid determinations for diagnosis and
treatment. Clin Toxicol 1986;24:389-402.
22. Underwood F, Bennett WM. Ethylene glycol intoxication. JAMA 1973;57:143-150.
23. Parry MF, Wallach R. Ethylene glycol poisoning. Am J Med 1974;57:143-150.
24. GodolphinW, Meagher EP, Sanders HD et al. Unusual calcium oxalate crystals in ethylene glycol poisoning. Clin
Toxicol 1980;16:479-486.
25. Groszek
B. Ethylene glycol poisoning: why so high mortality? Proc Eur Assoc Pois Cont Clin Toxicol, Birmingham,
UK: May 26-28, 1993.
26. Malmlund H-O, Berg A, Korlman G, Magnusson A, Lillman B. Considerations for the treatment of ethylene glycol
poisoning based on analysis of two cases. Clin Toxicol 1991;29:231-240.
27. Curtin
L, Kramer J, Wine H, Savitt D, Abuelo JG. Complete recovery after massive ethylene glycol ingestion. Arch
Intern Med 1992;152:1311-1313.
28. Jacobsen D, Orrebo S, Ostborg J, Sejerst OM. Glycolate causes the acidosis in ethylene glycol poisoning and is
effectively removed by hemodialysis. Acta Med Scand 1994;216:409-416.
29. Baud FJ, Galliott M, Astier A, VuBien D, Garnier R, Likforman J, Bismuth C. Treatment of ethylene glycol poisoning
with intravenous 4-methylpyrazole. N Engl J Med 1988;319:97-100.
30. Baud F, Bismuth C, Garnier R, Galliott M, Astier A, Maistre G et al. 4-Methylpyrazole may be an alternative to ethanol
therapy for ethylene glycol intoxication in man. J Toxicol Clin Toxicol 1986-87;24:463-483.
31. Saladino
R, Shannon M. Accidental and intentional poisoning with ethylene glycol in infancy: diagnostic clues and
management. Pediatr Emerg Care 1991;7:93-96.
32. ChristianssonKK, Kapersson KB, Kulling PEJ, Orrebo S. Treatment of severe ethylene glycol intoxication with
continuous arterio-venous hemofiltration dialysis. J Toxicol Clin Toxicol 1995;33:267-270.


PROPYLENE GLYCOL
USES
Propylene glycol has been used as a nontoxic antifreeze in dairies and breweries, as a component of
automotive brake fluids and antifreeze preparations, in the production of varnishes and synthetic resins,
as a flavoring agent in baking and candy production, and as an emulsifier and nontoxic preservative in
the foods industry. [1] Among prescription medications it can be found in several oral antibiotics, and it is
a major ingredient in the formulation of several parenteral preparations, including diazepam (Valium)
and phenytoin (Dilantin) in which its concentration may be as high as 40%. [1]
Propylene glycol is used in commercially available IV nitroglycerin solutions in quantities of 30 to 96%
or more (Table 55-25) (Table Not Available) . [2] [3] It is also present in other parenteral medications
(Table 55-26) (Table Not Available) . [3]


TOXICOKINETICS
Intravenous administration of propylene glycol (PG) in amounts from 3 to 15 g/m2 is followed by a

maximum plasma concentration of 60 to 425 mug/mL, respectively, with a half-life of 1.8 to 3.3 hours, a
volume of distribution of 0.51 to 0.88 L/kg, and a clearance rate of about 300 mL/min/1.73 m2 . [4]
Cerebrospinal fluid concentrations are as high as 85% of the serum concentrations. [5]


Propylene glycol may cause hemolysis, deafness, a high anion-gap acidosis due to elevated lactate levels,
sudden collapse, cardiac arrhythmias and asystole, hepatic damage, renal damage, hemolysis and serum
hyperosmolarity with a marked osmolar gap. [2] There appears to be a significant increase in seizures
among infants who have received a multivitamin preparation (MVI-12 with propylene glycol 3 g/day)
when compared with infants receiving MVI concentrate (propylene glycol 300 mg/d). [6]
In experimental animals, propylene glycol possesses one-third the central nervous system depressant
properties of ethanol. [7] A 15-month-old child had several episodes of hypoglycemia while ingesting 7.5
mL of propylene glycol per day. [8] Seizures developed in an 11-year-old boy with multiple endocrine
problems and systemic candidiasis who ingested a medication containing propylene glycol. [9] A patient
who presented with a propylene glycol blood level of 70 mg/dL developed stupor and lactic acidosis. [10]
Hemolysis, hemoglobinuria, skin irritation, deafness, and other neurologic disturbances may be observed
after propylene glycol administration. [11] The FDA considers propylene glycol safe in small doses for
pharmaceutical preparations.
A patient was admitted with a plasma propylene glycol (PG) concentration of 4 mg/mL derived from
drinking fruit juice (PG 0.6 mg/mL). She developed intractable epilepsy, respiratory depression, plasma
hyperosmolality, and a metabolic acidosis, and she recovered rapidly (PG plasma level fell to less than
100 mug/ml within 6 hours). [12]


LABORATORY
Propylene glycol is osmotically active and produces a concentration-dependent increase in serum

osmolality. Propylene glycol contained in enoximone was infused intravenously into an infant with heart
failure at the rate of 2.4 mg/kg/min. Serum osmolality rose from 304 mmol/kg before the start of the
infusion to 385 mmol/kg. Estimations of the serum propylene glycol concentration were 10,000 mug/mL.
[11] A rapid gas-liquid chromatographic assay has been described with a detection limit of 1 mug/mL.
Data suggests that serum PG concentrations greater than 177 mug/mL are required to increase the lactate
concentration by 6 mug/mL and to result in an elevated `ànion gap.'' [5] A serum PG concentration in
excess of 1520 mug/mL is required to yield an increase in osmolality of 20 mOsm/kg. [5] A method for
the gas chromatographic-mass spectrometric identification and quantification of ethylene glycol and
diethylene glycol in plasma is available. The detection limit is less than 10 mg/liter. [13]
A theoretical formula for the estimation of propylene glycol concentration from the osmolal gap is:
propylene glycol (mg/dL) = osmolal gap 7.6, or 47.5 + (osmolal gap 9.2). [14] [15]


TREATMENT
Infusion rates of over 1 mL/minute of products containing propylene glycol (PG) should be avoided.
PG-containing products should not be administered through the same
1157
TABLE 55-25 -- Composition of Intravenous Nitroglycerin Solutions Available in Different Countries a

(Not Available)
a Adapted from Derney HE et al. Intensive Care Med 1988;14:221-226.
intravenous line as packed red cells. When amounts of PG exceeding 300 mg per day are used, serum PG
levels and serum osmolality should be monitored. [16]


DIETHYLENE GLYCOL
HISTORY
The Elixir Sulfanilamide disaster of 1937 was one of the most consequential mass poisonings of the 20th
century. This tragedy occurred shortly after the introduction of sulfanilamide, the first sulfa antimicrobial
drug, when diethylene glycol was used as the diluent in the formulation of a liquid preparation of
sulfanilamide known as Elixir Sulfanilamide. One hundred and five patients died from its therapeutic
use. [17]


USES
Diethylene glycol is used in industrial solvents and antifreeze.
1158
TABLE 55-26 -- Parenteral Drug Products a

(Not Available)
a Adapted from Smolinske SC. Handbook of food, drug and cosmetic excipients. Boca Raton, FL: CRC Press, 1992.


A single fatal oral dose of diethylene glycol in humans is about 1.2 mL/kg (1 to 2 g/kg).
Epidemiology
Children (seven) admitted with diethylene glycol poisoning presented with a prodromal febrile illness
leading to vomiting, anuria, and diarrhea. In addition, dehydration, tachypnea, hepatomegaly, depressed
consciousness, irritability, palpable kidneys, papillitis, and meningeal signs were observed. The serum
alanine aminotransferase (SGPT) is elevated usually in all children studied. Death may occur despite
rehydration and apparent improvement. [18]
Adulteration of some wines with up to 10 to 20 g/L of diethylene glycol was observed in 1985 in the
Netherlands. Elevated serum creatinine levels were observed in some patients. Further health data are not
available. [19] A number of wines adulterated with diethylene glycol were imported from Austria, West
Germany, and Italy. [20] Polyethylene glycol solutions may contain quantities of diethylene glycol as a
contaminant. [21]
Topical application of a 1% silver sulfadiazine formulation used in Spain for application to second- or
third-degree burns (500 to 4000 g/day) contained 6.2 to 7.1 g/kg of diethylene glycol stearate and free
diethylene glycol. [22] These substances are not present in the United States' silver sulfadiazine product
(Silvadene). [23] After 3 to 6 days of topical treatment with the formulation, oliguria, metabolic
1159
and lactic acidosis, a increased anion gap, and irreversible coma developed. All patients died despite
bicarbonate replacement. No calcium oxalate crystals were observed. [18] Fourteen patients ingested
glycerine contaminated with diethylene glycol. They developed severe gastrointestinal symptoms, a
metabolic acidosis, and renal failure. All died. Autopsy revealed acute renal cortical necrosis,
centrilobular hepatic necrosis, and extensive hemorrhages in the adrenal medullae. [24] Paracetamol elixirs
with diethylene glycol as a diluent led to 51 deaths and an epidemic of acute renal failure in early 1990 in
Bangladesh. [25]
Symptoms of oral diethylene glycol poisoning occur within 24 hours of ingestion and include nausea,
anorexia, vomiting, abdominal pain, and diarrhea. Neurologic effects include headache, dizziness, and
narcosis. Renal involvement include polyuria and oliguria. Jaundice and ascites indicate liver pathology.
Near death at 2 to 22 days postingestion, the patient may exhibit seizures, coma, anuria, edema, fluid and
electrolyte imbalance, and acidosis. At autopsy renal tubular damage and centrilobular liver necrosis are
observed.


LABORATORY
The patient exhibits elevated serum osmolality, abnormal electrolytes, low blood glucose, elevated BUN
and creatinine, an abnormal urinalysis with cells and casts, and elevated liver function test.


TREATMENT
Decontamination is recommended within 1 to 2 hours of an acute oral ingestion. Ipecac is not advised
because of the potential for early seizures. Ethanol therapy has not been subjected to controlled clinical
studies. Hepatitis is treated with lactulose and a low-protein diet. Comatose patients need ventilatory
support. Seizures are treated with standard anticonvulsants. Monitor fluid status. Repeat hemodialysis
may be indicated. [26]


2-MERCAPTOETHANOL
STRUCTURE
SH-CH2 -CH2 -OH


USES
2-Mercaptoethanol is a colorless liquid with a strong unpleasant odor used in laboratories for reduction
of protein disulfide bonds and for protection of sulfhydryl enzymes. [27] It also has industrial uses. [28]


2-Mercaptoethanol is irritating to the skin, eyes, and mucous membranes. In man ingestion of about 10 to
20 mL may induce emesis and end fatally with extensive subendocardial ventricular hemorrhages. The
toxicity of 2-mercaptoethanol may be due to the thiol group in the molecule, which is known to inhibit
cytochrome oxidase. 2-Mercaptoacetate, a metabolite, is found in the urine and may be secreted into the
gastric contents. [25]


TREATMENT
Treatment is symptomatic and supportive.


ETHYLENE GLYCOL DERIVATIVES

ETHYLENE GLYCOL MONOMETHYL ETHER
Ethylene Glycol Monomethyl Ether (EGME) and Acetate (EGMEA)
Ethylene Glycol Monoethyl Ether (EGEE) and Acetate (EGEEA)
NIOSH has recommended an occupational standard: 0.2 ppm EGME or EGMEA as a TWA for up to a
10-hour day during a 40-hour work week, and 0.5 ppm for EGEE or EGEEA as a 10-hour TWA. NIOSH
also recommends that dermal contact be prohibited for all four of these glycol ethers. [29]
Synonyms
2-Ethoxyethanol Cellosolve
CG monoethyl ether
2-Butyoxyethanol EGBE (Butyl cellosolve)
2-Methoxyethanol EGME (Methyl cellosolve)


USES
These derivatives are used as solvents for resins and paints and as constituents of cellulose inks,
industrial coatings (wood stains, epoxies, varnish, paints), and cleaning compounds (Tables 55-27 (Table
Not Available) and 55-28) (Table Not Available) .


PATHOPHYSIOLOGY
The alkoxyacetic acid metabolite, or its glycine conjugate, has been found in the urine from man or
animals after the administration of 2-methoxy-, 2-ethoxy-2-isopropoxy-, and 2-butoxyethanol. It appears
that many of the toxic effects of the short-chained alkoxyethanols can be attributed to their acid
metabolites. [30]


Ethylene Glycol Monodiethyl Ether
An adult ingested 1 liter of an ethyl glycol monobutyl ether (EGBE)-containing window cleaner over 3
days and became comatose with hyperventilation. Treatment with hemodialysis led to recovery. Marked
metabolic acidosis after the ingestion of a large amount of EGBE remains the main symptom caused by
the oxidating of EGBE to butoxyacetic acid (Fig. 55-7) (Figure Not Available) . [31] [32] Less than 1 mL of
a commercial glass/window cleaner containing less than 17% EGBE does not appear to require specific
therapy. [33] Bone marrow aspirates of seven lithograph workers exposed to glycol ethers, among other
hydrocarbons, revealed myeloid hypoplasia
1160
TABLE 55-27 -- Chemical Names, Chemical Structures, Trade Names, and Synonyms for Monoalkyl
Ethers of Ethylene Glycol a
(Not Available)
a Adapted from Browning RG, Curry SC. Hum Exp Toxicol 1994;13:325-335.
TABLE 55-28 -- Common Household Products Containing Glycol Ethers a

(Not Available)
a Adapted from Browning RG, Curry SC. Hum Exp Toxicol 1994;13:325-335.
and stromal injury in three that could not be explained by known risk factors. [34] Studies in animals
indicate that the glycol ethers Cellosolve and methyl Cellosolve (but not butyl Cellosolve) cause birth
defects and testicular damage at levels near legal exposure limits (100 ppm). A consistent pattern of
embryotoxic and teratogenic effects occurs with lower-molecular-weight glycol ethers perhaps acting as
tumor-promoting agents. [35]
Ethylene Glycol Monomethyl Ether (EGME)
Blood seems to be the main target of toxicity in cases of chronic EGME exposure (macrocytic anemia,
leukopenia, increased proportion of lymphocytes). [36] Hematologic changes are reversible after exposure

is discontinued.
Following a symptom-free interval of 8 to 18 hours, two patients who had ingested 100 mL of EGME
became confused and complained of nausea and weakness. [37] The subsequent course included a
profound metabolic acidosis, tachypnea, rise in serum creatinine, and a marked oxaluria, from which they
recovered. Treatment included intravenous sodium bicarbonate and ethyl alcohol. [38] Cutaneous
occupational exposure to EGME for about 6 months led to an encephalopathy in factory workers. This
was accompanied by an anemia, leukopenia, and thrombocytopenia. [39]
Monoethyl Ether (EGEE)
About 23% of absorbed EGEE is recovered in the urine. [40] Exposure to EGME and EGEE at work has
led to some decrease in testicular size but no apparent alteration in fertility. [41] [42]
Monobutyl Ether (EGBE)
Ethylene glycol butyl ether (EGBE) is an ingredient in commercial window glass cleaners (Table 55-29)
(Table Not Available) . [43] Children aged 7 months to 9 years who have ingested 5 to 300 mL of liquid
products containing EGBE (0.5 to 9.9%) have all recovered.
EGBE ingestion may lead to coma, hypotension, metabolic acidosis, renal injury, hematuria, hemolysis,
oxaluria, and noncardiogenic pulmonary edema. [36] [44] Patients have recovered with symptomatic and
supportive therapy.
EGBE is absorbed through the skin, lungs, and gastrointestinal tract. The elimination half-time is brief
(40 minutes). It is oxidized by alcohol dehydrogenase in the liver. Its principal metabolite is butoxyacetic
acid responsible for the metabolic acidosis [49] and possibly hemolysis. [45] Ethanol therapy or
4-methylpyrazole administration may be useful because it competitively inhibits alcohol dehydrogenase.
Ingestion of 5 to 10 mL of EGBE-containing glass/window cleaners (0.5 to 9.9%) by children aged 7
months to 9 years was treated with oral fluids at home. The patients were asymptomatic. Two children
ingested amounts over 15 mL and were treated with gastric emptying and admission to a health care
facility for 24 hours. No hemolysis, CNS depression, acidosis, or renal compromise was noted. Data
suggest that ingestions of less than 10 mL of a commercial liquid glass/window cleaner containing less
than 10% EGBE can be safely treated with simple dilution in the home setting. [50] An adult ingested
about 500 mL of a window cleaning agent containing 12.7% v/v EGBE and 3.2% v/v ethanol. About 200
to 250 mL were absorbed corresponding to a total dose of 25 to 30 g of EGBE. Butoxyacetic acid, the
main metabolite of EGBE, peaked in the urine about 24 hours after ingestion. The half-life of EGBE was
210 minutes. Clinically, the patient was admitted in coma with
1161
Figure 55-7 (Figure Not Available) Suggested metabolic patterns of EGBE. (Adapted from Rambourg-Schepens MO et al.
Hum Toxicol 1988;7:187-189.)
TABLE 55-29 -- Common Commercial EGBE-Containing Glass Cleaners a

(Not Available)

a Adapted from Dean BS, Krenzeleck EP. J Toxicol Clin Toxicol 1992;30:557-563.
mydriasis, bradycardia, hypotension, and metabolic acidosis with a high anion gap. The patient
developed anemia and hematuria on the second day, concurrent with maximum levels of butoxyacetic
acid but with no oxaluria. Treatment was supportive (intubation, assisted ventilation, activated charcoal,
gastric lavage, forced diuresis, dopamine), and hemodialysis was required to control the metabolic
acidosis and to remove the toxic agent and its metabolites. [46]
An adult ingested 250 to 500 mL of a window cleaner containing 12% EGBE. The patient became
comatose, developed a metabolic acidosis, hypokalemia, a rise in serum creatinine, a marked increase in
urinary excretion of oxalate crystals, anemia, and a hemoglobinuria.
Monophenyl Ether (EGPE)
Percutaneous absorption after exposure to about 500 mL of 2-phenoxyethanol (EGPE) per day may
induce headache, lightheadedness, slurred speech, euphoria, grogginess, diminished strength and
sensation in the hands and fingers, forgetfulness and irritability. Alcohol intolerance may be observed. [47]


LABORATORY
Monitoring of the urinary excretion of the alkoxyacetic acid metabolites may be a useful indicator of
human exposure to ethylene glycol ethers. In ten male workers exposed to ethylene glycol monoethyl
ether, the maximal urinary excretion of ethoxyacetic acid occurred in 3 to 4 hours, and the urine biologic
half-life was 21 to 24 hours.
Analytic Method
A sensitive method for butoxyacetic acid (BAA) and other alkoxyacetic acids in use is based on ion-pair
extract with a capillary gas chromatograph and electron captive detection (GC-ECD). This method has a
detection limit and a practical quantification limit of about 0.05 and 7 mumol/L, respectively. A single
urine sample of 0.2 mL is sufficient to measure BAA. [35] A marked increase of EGBE in blood and urine
and a marked decrease of BAA excretion follows administration of ethanol corresponding to above
1162
0.3% in blood prior to exposure to EGBE vapor 20 ppm for 2 hours.


TREATMENT [48]
1. Diagnosis is based on the history and clinical presentation. The hallmarks of acute glycol ether
toxicity are central nervous system depression and metabolic acidosis similar to that seen after
poisoning from methanol, ethylene glycol, paraldehyde, iron, isoniazid, salicylates, cyanide,
hydrogen sulfide, and carbon monoxide and that seen with diabetic and alcoholic ketoacidosis,
renal failure, and lactic acidosis.
2. The glycol ethers are rapidly absorbed. Gastric lavage would appear to be of little value unless
performed very early after ingestion.
3. Syrup of ipecac is contraindicated due to the possibility of rapid progression to coma, with risk of
aspiration.
4. Activated charcoal probably would be inefficient at binding glycol ethers in the gut, similar to its
relative lack of efficacy in similar molecular weight alcohols and glycols.
5. Blockade of the alkoxyacetic acid metabolite may follow competitive inhibition of alcohol
dehydrogenase with alcohol or possibly 4-methylpyrazole. An intravenous loading dose of 800
mg/kg of ethanol is followed by a continuous drip of 80 to 150 mg/kg with adjustments to
maintain a plasma ethanol level of 100 to 150 mg/dL.
6. Support measures, if required, include airway control, ventilatory assistance, and cardiovascular
support with fluids and vasopressors.
7. Hemodialysis may be useful for severe, refractory acidosis or renal insufficiency. There have been
no controlled studies to validate this theory.
8. The likelihood of fatality is low. Delayed toxicity may result from toxic metabolites of EGBE after
it is metabolized by alcohol dehydrogenase. Alcohol should be given early to patients with a
significant ingestion of EGBE. This could be ceased later if no toxicity develops and anion and
osmolar gaps remain normal. The osmolar gap is mainly within the normal range with EGBE
poisoning. [49] It would take extremely high and clinically unlikely concentrations of glycol ethers
to produce a detectable increase in the osmolal gap. [50]


PROPYLENE GLYCOL ETHERS

STRUCTURE
Propylene glycol ethers commonly used include methoxypropanol, methoxypropylacetate, and

dipropylene glycol monomethyl ether. Both alpha and beta isomers are present in commercial
methoxypropanol and methoxypropyl acetate. The isomers differ in toxicity.
Synonyms (S) and conversion factors (CF)
Methoxypropanol, alpha isomer
S: Propylene glycol monomethyl ether
Propylene glycol methyl ether (PGME)
CF: ppm = 3.68 pm/m3
1 mg/mm3 = 0.272 ppm
odor threshold for PGME = 36 mg/m3
Methopyl propanol beta isomer
S: Propylene glycol monomethyl ether
Propylene glycol methyl ether: beta PCME
CF: 1 ppm = 3.368 mg/m3
1 mg/m3 = 0.272 ppm
Methoxyl propyl acetate, alpha isomer
S: Propylene glycol monomethyl ether acetate
Propylene glycol methyl ether acetate PGMEA
CF: 1 ppm = 5.40 mg/m3
1 mg/m3 = 0.185 ppm
Methoxyl propyl acetate, beta isomer
S: Propylene glycol monomethyl ether acetate
Propylene glycol methyl ether acetate (beta PGMEA)
CF: 1 ppm = 5.40 mg/m3

1 mg/m3 = 0.185 ppm

Dipropylene glycol methyl ether
S: Dipropylene glycol monomethyl ether
Dipropylene glycol methyl ether (DPGME)
CF: 1 ppm = 6.06 mg/m3
1 mg/m3 = 0.165 ppm
Propylene glycol ethers are colorless liquids at room temperature. They have a sweet, etherlike odor and
are completely miscible with water and a number of organic solvents.


USES
Propylene glycol ethers are used industrially as solvents for paints, lacquers, resins, oils, and fat.
Propylene glycol ethers have increased in use in the past 10 years. One reason for the increase is
probably the replacement of ethylene glycol ethers by propylene glycol ethers because of the
reproductive toxicity associated with the former group of solvents. [51]


TOXICOKINETICS
The major metabolic pathways of propylene glycol ethers can be summarized as follows: (a) acetate
esters are rapidly hydrolyzed to the corresponding either alcohol; (b) ether alcohols, both alpha and beta
isomers, are conjugated with sulfite and glucuronic acid; (c) beta isomers, being primary alcohols, are
oxidized to carboxylic acid; (d) alpha isomers, being secondary alcohols, are also oxidized to carbon
dioxide after cleavage of the ether bond.


PGME
Exposure at 300 ppm for 5 minutes produces mild irritation of the eyes, nose, and throat. At 750 ppm the
PGME vapor is extremely irritating. At concentrations of 47 to 1000 ppm for 1 to 7 hours no
abnormalities in liver enzymes are observed. At exposure levels of 100 ppm to 400 ppm
light-headedness, a negative Romberg test, and eye, nose, and throat irritation may be observed.
1163
DPGME
The irritating concentration of DPGME in man is 74 ppm (450 mg/m3 ) or about twice the odor
threshold. Exposure to 0.6 to 6.4 ppm have resulted in bone marrow injury. There are no reports of any
effects on the peripheral nervous system. There are no signs of subacute or chronic organ-specific
damage at levels below 1500 ppm PGME, 1000 ppm PGMEA, and 300 to 400 ppm DPGME. There are
no present indications of mutagenicity or genotoxicity.


POLYETHYLENE GLYCOL
STRUCTURE
OH-CH2 -O-(CH2 -CH2 -O)n -CH2 -CH2 OH

Table 55-30 (Table Not Available) summarizes some species of polyethylene glycol used clinically. [52]
Polyethylene glycols have been used as a component of suppositories where they serve as an `ìnert'' base
and carrier for various `àctive'' drugs. Preliminary animal data suggest that polyethylene glycol
suppositories may predispose cells in the rectal or vaginal mucosa to malignant change. Further data is
required on this subject. [53]
Polyethylene glycol (PEG) is used for preoperative bowel cleaning, as a medicinal to treat constipation,
and as a purgative in the management of poisoning. Diethylene glycol may be a contaminant of
commercial polyethylene glycol (PEG) solutions. [26]


Metabolic acidosis, renal insufficiency, and cardiorespiratory arrest have followed topical use in a
2-year-old infant. [54]


LABORATORY
Polyethylene glycol may be present in temazepam, a drug often abused orally and by the intravenous
route. Detection of polyethylene glycol in the urine may aid in detection and clinical treatment of drug
abusers. [55]
TABLE 55-30 -- Range of Molecular Weights and Polymer Lengths for Polyethylene Glycols Used in
These Studies a
(Not Available)
a Adapted from Schiller LR et al. Gastroenterology 1988;94:933-941.


TREATMENT
Treatment consists of supportive care. Hemodialysis and ethanol administration have not been protective.


BENZYL ALCOHOL
(See also Anesthetic Drug Chapter)


PARENTERAL MEDICATIONS
Parenteral medications formulated with benzyl alcohol include atropine sulfate, glycopyrrolate,
physostigmine, pyridostigmine, heparin, crystalloid solutions, succinylcholine, atracrurium,
pancuronium, tubocurarine, doxapram, and metoclopramide. [56]
With dosages of benzyl alcohol ranging from 99 to 234 mg/kg/day, hypotension, severe metabolic
acidosis with an increased anion gap secondary to increased blood concentrations of benzoic acid,
changes in consciousness, leukopenia, thrombopenia, hyperammonemia, and respiratory gasping are
observed with high mortality rates. Diazepam 1 mL contains 5 mg of diazepam, 0.5 mmol of sodium, and
15.7 mg of benzyl alcohol. When a continuous intravenous diazepam infusion is administered for status
epilepticus, no dosage of more than 1 mg/kg/hour should be used. With this dosage, benzyl alcohol 75
mg/kg/day and sodium 2.4 mmol/kg/day are administered. When using dosages between 0.5 and 1
mg/kg/hour, frequent cardiovascular and respiratory monitoring are required, and serum electrolytes,
blood gases, and anion gap must be measured every 6 to 12 hours. Serum and urine benzoic acid
concentrations must be determined at least every 24 hours. [57]


The term gasping syndrome describes the progressive neurologic deterioration of premature infants
suffering from benzyl alcohol poisoning. [58] Retrospective studies of neonates revealed a substantial
decline in both mortality and major intraventricular hemorrhage among infants weighing less than 1 kg
after the use of benzyl alcohol-containing solutions was discontinued. [59]


UNAPPROVED DRUGS (UNITED STATES)

CALCIUM 2-AMINOETHANOL PHOSPHATE
Calcium-2-aminoethanol phosphate is claimed to inhibit autoimmune processes and to be useful in the

treatment of osteoporosis. [60] [61] It is brought into the United States from Europe and may be in use by
practitioners in the United States for the treatment of multiple sclerosis.
1164


A 53-year-old female with a history of multiple sclerosis was given an intravenous infusion of calcium
2-aminoethanol phosphate and developed a cardiopulmonary arrest, followed by massive hemolysis,
renal failure, adult respiratory distress syndrome, shock liver, and disseminated intravascular
coagulation. [62]


TREATMENT
Treatment of overdose/or adverse reactions is supportive and symptomatic.


REFERENCES--OTHER GLYCOLS AND ALCOHOLS
1. Catanzaro JM, Smith JG Jr. Propylene glycol dermatitis. J Am Acad Dermatol 1991;24:90-95.
2. Demey HE, Daelemans RA, Verpooten GA, de Broe ME, Van Campenhout CM, Lakiene FV et al. Propylene
glycol-induced side effects during intravenous nitroglycerin therapy. Intensive Care Med 1988;14:221-226.
3. Smolinske SC. Handbook of food, drug and cosmetic excipients. Boca Raton, FL: CRC Press, 1992.
4. Speth PAJ, Vree TB, Neilen NFM, de Mulder PHM, Newell DR, Gore ME et al. Propylene glycol pharmacokinetics and
effects after intravenous infusion in humans. Ther Drug Monitor 1987;9:255-258.
5. Kelner MJ, Bailey DN. Propylene glycol as a cause of lactic acidosis. J Anal Toxicol 1985;9:40-42.
6. MacDonald MG, Getson PR, Glasgow AM, Miller MK, Boeckx RL, Johnson EL. Propylene glycol: increased incidence
of seizures in low birth weight infants. Pediatrics 1987;79:622-625.
7. Lehman AJ, Newman HW. Propylene glycol: rate of metabolism, absorption and excretion with a method for estimation
in body fluids. J Pharmacol Exp Ther 1937;60:312-322.
8. Martin G, Finberg L. Propylene glycol: a potentially toxic vehicle in liquid dosage. J Pediatr 1970;77:877-878.
9. Arulanantham K, Genel M. Central nervous system toxicity associated with ingestion of propylene glycol. J Pediatr
1978;93:515-516.
10. Cate JC, Hendricks R. Propylene glycol intoxication and lactic acidosis. N Engl J Med 1980;303:1237.
11. Huggon I, James I, Macrae D. Hyperosmolarity related to propylene glycol in an infant treated with enoximone
infusion. Br Med J 1990;301:19-20.
12. LolinY, Francis DA, Flanagan RJ, Little P, Lascelles PT. Cerebral depression due to propylene glycol in a patient with
chronic epilepsy--the value of the plasma osmolal gap in diagnosis. Postgrad Med J 1988;64:610-613.
13. Maurer
H, Kessler C. Identification and quantification of ethylene glycol and diethylene glycol in plasma using gas
chromatography-mass spectrometry. Arch Toxicol 1988;62:66-69.
14. Glasgow AM, Boeckx RL, Miller MK, MacDonald MG, August GP, Goodman SI. Propylene glycol plasma level.
Pediatrics 1985;76:654.
15. HallAH, Bronstein AC, Smolinske SC, Doutre WH, Kulig KW, Rauch BH, Spoerke DG. Propylene glycol plasma
levels. Pediatrics 1985;76:656.
16. Smolinske
SC, Vandenberg SA, Spoerke DG, Rumack BH. Propylene glycol content of parenteral medications. Vet
Hum Toxicol 1987;29:491.

17. WaxPM. Elixirs, diluents and the passage of the 1938 Federal Food, Drug and Cosmetic Act. Ann Intern Med
1995;122:456-461.
18. Bowie MD, McKenzie D. Diethylene glycol poisoning in children. S Afr Med J 1972;46:931.
19. Van
der Linden-Cremers PMA, Sangster B. Diethylene glycol in wine: experiences in the Netherlands. Summer 1985.
Vet Hum Toxicol 1987;29(Suppl 2):12-13.
20. U.S. finds traces of poison chemical in 10 more wines. Los Angeles Herald-Examiner, April 16, 1986.
21. WoolfA, Pearson K. Presence of diethylene glycol in commercial polyethylene glycol (PEG) solutions. Clin Toxicol
1995;33:475-486 (abstract 10).
22. Cantarell
MC, Fort J, Camps J, Sans M, Piera L, Rodamilans M. Acute intoxication due to topical application of
diethylene glycol. Ann Intern Med 1987;106:478-479.
23. McAlinney PG. Silver sulfadiazine and oliguric renal failure. Ann Intern Med 1987;107:264.
24. Pandya SK. An unmitigated tragedy. Br Med J 1988;297:117-119.
25. HanifM, Mobarak MR, Ronan A, Rahman D, Donovan JJ Jr, Bennish ML. Fatal renal failure caused by diethylene
glycol in paracetamol elixir: the Bangladesh epidemic. Br Med J 1995;33:88-91.
26. Woolf AD. Diethylene glycol. Clin Toxicol Rev 1994;17(3):1-2.
27. White FH Jr. Reduction and reoxidation of sulfide bonds. Methods Enzymol 1972;25(Part B):387-392.
28. EriksonA, Mohlin L, Nilsson L, Sorbo B. Mercaptoethanol poisoning. Report of a fatal case and analytical
determinations. J Anal Toxicol 1989;13:60-62.
29. Wess JA. Reproductive toxicity of ethylene glycol monomethylether, ethylene glycol monoethyl ether and their
acetates. Scand J Work Environ Health 1992;18(Suppl 2):43-45.
30. Johanson G. Urine butoxyacetic acid as a therapeutic guide. Clin Toxicol 1993;31:501-552.
31. Rambourg-Schepens MO, Buffet M, Bertlaut R, Jaussaud M, Journe B, Fay R et al. Severe ethylene glycol butyl ether
poisoning. Kinetics and metabolic pattern. Hum Toxicol 1988;7:187-189.
32. FelgenahuerN, Zilker T, Clarmann M. A case of severe ethylene glycol monobutyl ether poisoning. Proc Eur Assoc
Pois Cont Clin Toxicol, Birmingham, UK: May, 1993.
33. Dean BS, Krenzelok E. 10 mL of 10% solution is not ``significant.'' Clin Toxicol 1993;31:503-504.
34. CullenMR, Rado T, Waldron JA et al. Bone marrow injury in lithographers exposed to glycol ethers and organic
solvents used in multicolor offset and ultraviolet curing printing processes. Arch Environ Health 1983;38:347-354.
35. HardinBD, Lyon JP. Summary and overview: NIOSH symposium on toxic effects of glycol ethers. Environ Health
Perspect 1984;57:273-275.
36. Larese
F, Fiorito A, de Zotti R. The possible haematological effects of glycol monomethyl ether in a frame factory. Br J
Ind Med 1992;49:131-133.

37. Johanson
G, Kronborg H, Naslund PH, Nordqvist MB. Toxicokinetics of inhaled 2-butoxyethanol (ethylene glycol
monobutyl ether) in man. Scand J Work Environ Health 1986;12:594-602.
38. Nitter-Hauge S. Poisoning with ethylene glycol monomethyl ether. Acta Med Scand 1970;188:277-280.
39. Ohi
G, Wegman DH. Transcutaneous ethylene glycol monomethyl ether poisoning in the work setting. J Occup Med
1978;20:675-676.
40. GroesenekenD, Veulemans H, Maaschelein R, Van Vlem E. Ethoxyacetic acid: a metabolite as ethylene glycol
monoethyl ether acetate in man. Br J Ind Med 1987;44:488-493.
41. Welch LS, Plotkin E, Schrader S. Indirect fertility analysis in painters exposed to ethylene glycol ethers. Sensitivity and
specificity. Am J Ind Med 1991;20:229-290.
42. Cook RR, Bodner KM, Kolesar RC, Utilmanu CS, Van Peenen PFD, Dickson GS et al. A cross-sectional study of
ethylene glycol monomethyl ether process employees. Arch Environ Health 1982;37:346-351.
43. DeanBS, Krenzelok EP. Clinical evaluation of pediatric ethylene glycol monobutyl ether poisonings. Clin Toxicol
1992;30:557-563.
44. BauerP, Weber M, Mur JM, Protois JC, Bollaert PE, Condi A et al. Transient non-cardiogenic pulmonary edema
following massive ingestion of ethylene glycol butyl ether. Intensive Care Med 1992;18:250-251.
1165
45. DeanBS, Krenzelok EP. Clinical evaluation of pediatric ethylene glycol monobutyl ether poisonings. Vet Hum Toxicol
1991;33:362.
46. Gijsenbergh FP, Jenco M, Veulemans H, Groeseneken D, Verberckmoes R, Delooz HH. Acute butylglycol intoxication.
A case report. Hum Toxicol 1989;8:243-245.
47. Morton WE. Occupational phenoxyethanol neurotoxicity: a report of three cases. J Occup Med 1990;32:42-45.
48. Browning RG, Curry SC. Clinical toxicology of ethylene glycol monalkyl ethers. Hum Exp Toxicol 1994;13:325-335.
49. Buckley N, Whyte IM, Dawson AH. Letters to the Editor: EGBE. Clin Toxicol 1993;31:499-500.
50. Browning RG, Curry SC. Effect of glycol ethers on plasma osmolality. Hum Exp Toxicol 1992;11:488-490.
51. NEG and NIOSH basis for an occupational health standard: propylene glycol ethers and their acetates. Center for
Disease Control USDHHS. DHHS (NIOSH) Publication No. 91-103.
52. Schiller
LR, Emmett M, Santa Ana CA, Fordtran JS. Osmotic effects of polyethylene glycol. Gastroenterology
1988;94:933-941.
53. Greene
MH, Young TI, Eisenbarth GS. Polyethylene glycol in suppositories: cardinogenic? Ann Intern Med
1980;93:781.
54. Rodriguez ZM, Gutierrez JCL, Jimenez JMT, Mor ZR. Metabolic acidosis, intoxication caused by polyethylene glycol.

Proc Eur Assoc Pois Cent Clin Toxicol, Birmingham, UK: May 26-28, 1993.
55. Farrell M, Herrod J, Smith B, Strang J. Detection of intravenous drug use. Br Med J 1990;300:612-613.
56. Weissman DB, Jackson SH, Heicher DA, Rockoff MA. Benzyl alcohol administration in neonates. Anesth Analg
1990;70:673.
57. Lopez-Herce
J, Bonet C, Meana A, Albajara L. Benzyl alcohol poisoning following diazepam intravenous infusion.
Ann Pharmacother 1995;29:632.
58. Gershanik
J, Boecler B, Ensley H et al. The gasping syndrome and benzyl alcohol poisoning. N Engl J Med
1982;307:1384-1388.
59. Hiller
JL, Benda GI, Rahatzad M et al. Benzyl alcohol toxicity: impact on mortality and intraventricular hemorrhage
among very low birth weight infants. Pediatrics 1986;77:500-506.
60. Nieper
HA. A comparative study of the clinical effect of CA-1-dl-aspartate (Calciretard) of Ca-2-aminoethanol
phosphate (Ca--EAP) and of the cortisones. Aggressologie 1968;9:471-475.
61. Nieper HA. A clinical study of the calcium transport substances Ca 1-dl-aspartate and CA-2-amino ethanol phosphate
as potent agents against autoimmunity and other anti-cytological aggressions. 2nd communication. Aggressologie
1967;8:395-406.
62. Sauter
D, Goldfrank L, Charash BD. Cardiopulmonary arrest following an infusion of calcium-2-amino-ethanol
phosphate. J Emerg Med 1990;8:711-720.


Chapter 56 - Anesthetics
1166
This chapter will survey aspects of anesthetic drug use as they relate to problems encountered following
overdose or inadvertent misuse. Excellent texts are available that cover practical approaches to the
clinical practice of anesthesiology. [1] [2] [3] Opioid agonists and antagonists are reviewed in the Opiate
chapter.
ANESTHETIC ACCIDENTS AND DEATH

Anesthesia has been associated with a low mortality in recent years, but it is still associated with a
significant morbidity [4] [5] (Table 56-1) (Table Not Available) . [6] The risk of death due to anesthesia
decreased in the United States from 1 in 2680 [1] [7] to 1 in 10,000 in the period from 1954 to 1982. [8]
Human error remains an important factor in the causation of anesthesia faults, near accidents, and
accidents (Tables 56-2 (Table Not Available) through 56-5). [9] [10]
The University Hospital, Leiden, has established a Faults, Accidents and Near Accidents Committee
(FONA), which collects and analyzes reported faults, accidents, near accidents, and complications that
occur in the hospital. These data also include anesthetic procedures. Definitions used by the committee
include the following: [1]
Fault: A procedure that has resulted (or could have resulted) in injury or harm to a patient, while
this injury or harm (or chance thereof) could have been prevented by another procedure.
Accident: Any event, except a fault or a complication, that has resulted in harm or injury to a
patient.
Near Accident: An event that could have developed into a fault, a complication, or an accident but
that was prevented from so developing either by chance or by a previously unplanned intervention.
Complication: Harm or injury to a patient occurring as a result of a known risk of a treatment or a

diagnostic procedure.
DENTAL ANESTHESIA AND DEATHS
Prolonged anesthesia and complex techniques carry considerable risks. General anesthesia given by the
operating dentist alone increases risks of death. Local anesthetics are
1167

TABLE 56-1 -- Causative Factors Concerning Anesthetic Cardiac Arrest and Death a
(Not Available)
a From Kubota Y et al. J Clin Anesth 1994;6:227-238.
TABLE 56-2 -- Faults in Drug Administration a

(Not Available)
a Adapted from Chopra V, Borill JG, Spierdijk J. Anaesthesia 1990;45:3-6.
TABLE 56-3 -- 1000 Deaths Associated With Anesthesia--Anesthesia Implicated in 589 a

(Not Available)
a Adapted from Dinnick OP. Anaesthesia 1964;19:536-556.
TABLE 56-4 -- 600 Deaths Associated With Anesthesia--Anesthesia Implicated in 400 Approximately a
(Not Available)
a Adapted from Dinnick OP. Anaesthesia 1964;19:536-556.
TABLE 56-5 -- Causes of Death and Cerebral Damage Classified Into Those Apparently Due to
Misadventure and Those Apparently Due to Error a
(Not Available)
a Adapted from Utting JE, Gray TC, Shelley FC. Can Anaesth Soc J 1979;26:472-478.

not a serious risk factor. [11] Low blood volume, underventilation, regurgitation and vomiting, and a
collapse after intravenous block are the main factors in deaths. [12]


MD Consult - Journals
April 21, 2000
Conde Petra
Additional Article
This article is not currently cited in
MEDLINE, but was found in MD Clinics in Liver Disease
Consult's full-text literature database. Volume 2 Number 4 November 1998
Full Text
Frontmatter 723
GENERAL OVERVIEW OF LIVER

REGENERATION ALCOHOLIC LIVER DISEASE
REGULATION OF HEPATOCYTE
PROLIFERATION AFTER PH
Requirement 1: Injury-Related
Changes in the Microenvironment
that Promote Regeneration
Accumulation of the
Injury-Related Cytokine, TNF
Anna Mae Diehl MD
Regulation of TNF Biological
Activity
Johns Hopkins University School of Medicine, Baltimore, Maryland
Cytokine Production by Address reprint requests to
Extra-Hepatic Tissues Anna Mae Diehl, MD
912 Ross Building
Hepatocyte Mitogens and 720 Rutland Street
Comitogens Baltimore, MD 21205
e-mail: amdiehl@welchlink.welch.jhu.edu
Requirement 2: Proliferation of a
Differentiated Population of Cells
That Ordinarily Exist in a The liver is a major target tissue for alcohol-induced
Nonproliferative State damage. Given the prevalent use of alcohol, it is not
Initiation of the Proliferative
surprising that alcohol abuse is the most common cause of
Response in Hepatocytes (G0 clinically significant liver disease in the United States.
-to-G1 Transition)
Deaths from cirrhosis (the eventual consequence of chronic
liver injury) correlate with per capita consumption of
alcohol. Alcoholic cirrhosis, the end-stage of alcohol-induced
liver injury, currently is the fourth leading cause of death in
middle-aged men and is likely to remain a major cause of
http://home.mdconsult.com/das/journal/view/N/1088...&stty=cow&nid=39982&tlt=Z&pos=&tag=relinfo&anc=ja (1 of 26) [21/04/2000 23:21:50]
Progression from the
morbidity and mortality due to recent increases in alcohol
Prereplicative Period (G1 ) into
consumption among women and young teenagers. [36]
the Replicative Phases (S, M)
Unfortunately, there is no effective medical therapy for
of the Cell Cycle
most patients with alcoholic liver disease. Orthotopic liver
Requirement 3: Temporary transplantation improves the survival of patients with
Suspension of the Normal terminal alcoholic cirrhosis; however, transplantation
Mechanisms that Couple Cell carries a risk of perioperative mortality and necessitates
Cycle Re-entry to Cell Death lifelong immunosuppressive therapy with attendant risks of
CONCLUSION
serious infection and malignancy. Furthermore, if all
patients with advanced alcoholic liver disease received liver
References transplants, there would be no donor organs available for
patients with other forms of cirrhosis. [84] Therefore, there
still is a desperate need to develop effective medical therapy
for alcoholic liver disease.
Over the past 3 decades, considerable progress has been
made towards understanding the mechanisms by which
alcohol produces liver injury. [82] It now is clear that liver
injury is the end result of multiple events that are triggered
by the oxidation of ethanol to acetaldehyde, a highly
reactive metabolite. One important intermediary event in
this pathogenic process appears to be the release of
pro-inflammatory cytokines, including tumor necrosis
factor-alpha (TNF). Plasma concentrations of TNF and
TNF-inducible cytokines, including interleukins 6 (IL-6)
and 8 (IL-8), correlate with mortality in patients
hospitalized with alcoholic hepatitis. [10] [41] [49] [62] [63] [64] This
suggests that anticytokine agents may be useful for
interrupting the pathogenesis of alcohol-induced liver
injury. Indeed, treatments that decrease endotoxemia (a
potent stimulus for TNF production)
724
inhibit the evolution of hepatitis in rats that are being fed

alcohol. [1] [70] Unfortunately, unlike experimental animals
that receive the antidote and the toxin concomitantly, most
alcoholic patients already have sustained significant liver
damage before they seek medical attention. Therefore, in
clinical practice another major goal of treatment is to
facilitate liver repair. Data from a number of groups, [37] [38]
[43] [59] [77] [92] including the author's, [28] [29] [30] [35] indicate
that ethanol inhibits the normally immense regenerative

capacity of the liver. Thus, alcohol consumption may
produce liver disease both by injuring the liver and by
impairing the regenerative response to that injury.

Relatively little is known about the mechanisms underlying

ethanol's antiregenerative actions; however, these must be
defined and reversed to optimize speedy and complete
recovery from alcohol-induced liver injury.
GENERAL OVERVIEW OF LIVER

REGENERATION
The liver is the only vital organ in adult mammals that has the
ability to regenerate efficiently after it has been damaged. The
capacity of mature, terminally differentiated hepatocytes to
proliferate in the injured liver is surprising since, in adults, these
cells rarely proliferate in the healthy liver. Rather, hepatocytes
exist in a growth-arrested (G0 ) state, preoccupied with
accomplishing the vital synthetic and excretory functions of the
liver. After liver injury, remaining hepatocytes promptly leave
G0 and enter the proliferative stages of the cell cycle during
which they duplicate necessary components and synthesize new
proliferation-specific molecules (G1 phase), replicate their DNA
(S phase), and divide (M phase). Rounds of proliferation
continue until the premorbid mass of functional hepatocytes is
restored. During liver regeneration, nonparenchymal cell (e.g.,
endothelial, Kupffer, stellate, and bile duct epithelial cell)
proliferation and new matrix synthesis also are required to
reconstitute the organ. Despite the increased workload
necessary for organ regeneration, liver-specific, differentiated
functions must be maintained or death of the organism ensues.
This latter requirement imposes a burden on proliferating
hepatocytes that distinguishes liver regeneration from most fetal
or neoplastic growth, during which the expression of
proliferative genes generally is afforded a higher priority than
that of tissue-specific genes. The healthy liver can regenerate
quite efficiently, that is, increase proliferative activity while
maintaining differentiated functions. For example, following
70% (partial) hepatectomy (PH), liver functions generally are
preserved and the resected liver mass is restored within days to
weeks in rats and weeks to months in humans. [67] Because
facets of the normal regenerative response fail or are
dysregulated during many acute or chronic liver diseases,
workers seeking to treat these conditions are interested in
understanding liver regeneration.
PH has been a particularly useful model for defining events that
regulate the proliferation of mature ("differentiated")

hepatocytes because it provides a unique opportunity to
evaluate cell cycle progression in vivo. Because virtually all
hepatocytes are in G0 at the time of PH, and PH is a temporally
circumscribed "insult," all remaining hepatocytes enter
proliferative phases of the cell cycle synchronously. In rats,
hepatocyte DNA synthesis peaks 24 hours after PH, almost a
full day before there is appreciable DNA synthesis in
nonparencymal cells. [12] [67] Thus, the initial 24 hours after PH
offers a "window" to scrutinize the events that regulate G0 -G1
and G1 -S transition in adult hepatocytes. Indeed, it may be the
only system available to study G0 -G1 transition in normal
hepatocytes, because the overwhelming majority of freshly
isolated and cultured primary
725
or neoplastic hepatocytes already have entered G1 . [12] [32] [67]

Furthermore, since the remnant liver lobe was not directly
manipulated at the time of PH, it is a purely regenerating system
without coincident hepatic necrosis or inflammation to
confound data interpretation. The dearth of overt injury in the
remnant liver has prompted some concern that the PH model
may not reliably mimic events that regulate hepatocyte
proliferation after toxic or infectious liver injury. [32] The latter
criticism, however, may not be entirely valid given emerging
evidence that injury-associated cytokines (including TNF)
promote proliferation in surviving hepatocytes after PH. [2] [94]
Because a relatively detailed understanding of the molecular
events that regulate hepatocyte proliferation in vivo has been
developed for the PH model, this review focuses on that data.
REGULATION OF HEPATOCYTE
PROLIFERATION AFTER PH
In all species studied to date, regeneration of damaged tissues
has three general requirements: an injury-related change in the
microenvironment that favors the proliferation of a
differentiated population of cells that ordinarily exist in a
non-proliferative state while the usual mechanisms that couple
cell cycle re-entry to cell death are suspended temporarily. [67]
Information concerning each of these requisites for liver
regeneration after PH is provided in following sections. In each
section, initial comments summarize data in normal, healthy

animals and subsequent attention details the effects of acute or

chronic ethanol consumption on the normal response.
Requirement 1: Injury-Related Changes in the

Microenvironment that Promote Regeneration
Accumulation of the Injury-Related Cytokine, TNF

Several years ago, the author's research group proposed that
TNF was one of the factors released during liver injury that was
required to initiate the regenerative response in the damaged
liver. [2] At the time, several lines of evidence supported this
concept. TNF is a cytokine synthesized by monocytes,
macrophages, endothelium and some epithelium. Hence, TNF is
present in most, if not all, tissues. However, the liver is felt to
be a major source of TNF, particularly when exposed to certain
bacterial products such as endotoxin (i.e., lipopolysaccharide
[LPS]). [23] Cornell reported that restriction of gut-derived
endotoxin via antibiotics or endotoxin binding agents impaired
liver regeneration after PH. [20] Similarly, he showed that liver
regeneration after PH was significantly delayed in germ-free,
athymic, and LPS-resistant mice, strains that have a limited
ability to release cytokines in response to gut derived endotoxin.
[18] He also demonstrated that endotoxin pretreatment, a potent
stimulus for hepatic TNF release, enhanced liver regeneration

after PH. [19] Several other groups subsequently showed that
administration of recombinant human TNF increased DNA
polymerase alpha activity as well as the incorporation of H3
-thymidine into DNA in the livers of healthy adult rats. [8] [39] [40]
[65] Because serum or tissue levels of cytokines had not been
measured in any of Cornell's studies, and because relatively

large doses of recombinant TNF were delivered systemically in
the latter studies, the physiologic significance of TNF as a
regulator of hepatocyte proliferation was uncertain.
To clarify the physiologic role of TNF in the compensatory
hepatic regeneration
726
that is triggered by PH, the author's research group treated adult

rats 1 hour before PH with neutralizing polyclonal antibodies to
TNF alpha. Several parameters of liver regeneration in
anti-TNF antibody treated rats were compared with similar end
points in control rats treated with antibodies to rat IgG.
Treatment with anti-TNF antibodies pre-PH significantly
inhibited post-PH increase in H3 -thymidine incorporation into

hepatic DNA, and the expression of PCNA, an S-phase-related

antigen, by both hepatocytes and liver nonparenchymal cells. [2]
Many were skeptical about the physiologic significance of these
results as we were unable to detect TNF protein in the
circulation after PH using the bioassays that were available at
the time; the polyclonal antisera used may not have been
specific for TNF alpha. Furthermore, there was abundant
evidence that TNF played an important role in experimental
liver injury, [6] [17] [46] [47] [50] [51] [69] and recombinant TNF was
known to induce cytotoxicity when micromolar to milimolar
concentrations were added to hepatoma cell lines and primary
hepatocytes under certain culture conditions. [16]
In contrast, a few labs, including our own, began to report that
human recombinant TNF augmented DNA synthesis by primary
hepatocytes cultured in the presence of other mitogenic factors.
[33] We also showed that pretreatment of rats with soluble TNF
receptors (which bind to and inactivate TNF) inhibited

hepatocyte DNA after PH. Because soluble TNF receptors are
relatively specific for TNF alpha and have been used clinically
to neutralize the biologic actions of TNF, our finding
strengthened the argument that TNF may initiate the hepatic
regenerative process. [86] Eventually, the development of
sensitive reverse transcriptase-polymerase chain reactions
(RT-PCR) to amplify low abundance mRNAs and
commercially available ELISA's to detect picogram amounts of
rodent cytokines permitted our lab and other groups to
demonstrate that increases in hepatic TNF mRNAs and
circulating TNF protein occur within minutes to hours after PH
in otherwise healthy rats. [79] More recently, other laboratories
have completed work that also supports the theory that TNF
plays an important, positive growth-regulatory role during
normal liver regeneration. Yamada et al reported that post-PH
liver regeneration is inhibited in transgenic mice that had
targeted disruption of the TNF receptor-1 (TNFR-1) gene, [94]
and Cressman et al published evidence that liver regeneration is
inhibited in transgenic mice with inactivation of the gene that
encodes IL-6, [21] a TNF-inducible cytokine. [23] Thus, there is
now compelling evidence that injury-related cytokines help to
promote regeneration of the damaged liver.
There is some evidence that chronic ethanol consumption, at
least by rodents, does not obviate the role of TNF in hepatic
regeneration. We found that rats that had been fed
nutritionally-balanced, ethanol-containing diets for several
weeks actually were more vulnerable to effects of anti-TNF
antibodies than isocalorically-fed, control rats. Compared with

control rats that were pretreated with nonimmune IgG before

PH, post-PH hepatocyte DNA synthesis was 50% lower in
control rats that had been pretreated with anti-TNF antibodies.
In contrast, hepatocyte DNA synthesis was more than 90%
inhibited in anti-TNF, pretreated, ethanol-fed rats when
compared with a group of ethanol-fed rats that received
nonimmune IgG pre-PH. [3] Recent work indicates that hepatic
expression of TNF mRNAs and circulating levels of TNF
protein are generally increased after PH in such ethanol-fed rats
when compared with pair-fed control animals. [94A] Thus, more
TNF appears to be necessary for hepatocytes to achieve a
proliferative response in ethanol-fed animals; that is, chronic
ethanol consumption may result in a relative resistance to the
proliferative actions of TNF. If this is true, then strategies to
eliminate TNF may be particularly counterproductive once
alcoholic liver injury has occurred; however, given strong
evidence that
727
TNF is involved in the genesis of ethanol-mediated liver injury,

additional investigation is necessary to characterize the signals
that mediate the proliferative actions of TNF so that conditions
can be manipulated therapeutically to optimize TNF's
proliferative, as opposed to its cytotoxic, effects. Regulation of
TNF Biological Activity
Since TNF and TNF-inducible cytokines, such as IL-1 and IL-6,
regulate diverse functions in many target cells, there has been
intense interest in defining mechanisms that regulate TNF itself.
Over the last 5 years, work in various models indicates that
TNF production and biologic activity are regulated by other
cytokines. Some of these, including IL-12 and interferon
gamma (IFN gamma), act as TNF-inducers by promoting the
generation of TNF and increasing target cell sensitivity to TNF.
[76] [78] [89] Others, including IL-10 and transforming growth
factor beta (TBF beta), generally down-regulate the expression

and activity of TNF-inducers, TNF, and TNF-inducible
cytokines. [7] [52] [56] [72] The picture has been further complicated
by the recent discovery of other cytokines (lymphotoxin alpha
and lymphotoxin beta) that are related structurally to TNF. At
least one of these (lymphotoxin alpha, also called TNF beta)
binds to types 1 and 2 TNF receptors and appears to activate
some (but not all) of the same signaling mechanisms as TNF.
[66] [83] [85]
As stated earlier, the liver is an important producer of TNF,

particularly after exposure to gut-derived endotoxin (i.e., LPS)

[23] ; LPS increases in the portal blood after PH [87] and this is
thought to play a positive role in regulating liver regeneration

after PH. [13] [19] [20] In sepsis, LPS stimulation of TNF
expression is influenced by the relative abundance of
TNF-regulatory cytokines, including IL-12, IFN gamma, and
IL-10, which can be induced by LPS. [4] [5] [42] [71] [73] [88] [97]
Increased TNF prompts the production of several other
cytokines, including IL-1 and IL-6. TNF, therefore, is induced
always in the context of other cytokines in vivo. As a result, the
ultimate consequences of LPS (and other stimuli that increase
cytokine production) are mediated by a network of cytokines
that regulate each other's expression and activity. This partially
explains why the actions of any given cytokine are pleiotropic:
cytokine effects vary with the state of activation of the target
cell(s), which, in turn, is influenced by the presence of other
cytokines and the ability of the target cell to produce bioactive
autocrine factors. The importance of cytokine interactions in
defining physiology and pathophysiology is well-illustrated by
work in models of endotoxic shock, which demonstrates that
specific neutralization of any one of a number of different
cytokines (e.g., TNF, [86] [91] IL-12, [89] IFN gamma, [14] IL-10 [44]
) significantly influences mortality after treatment with LPS.
To date, relatively little has been learned about the nature and
timing of the cytokine response that occurs during liver
regeneration after PH. Our initial experiments with anti-TNF
antibodies demonstrated that IL-6 is induced after PH and that
TNF plays an important role in promoting IL-6 induction in this
setting. [2] These findings were confirmed by recent evidence
that post-PH increases in hepatic IL-6 mRNAs are severely
attenuated in TNFR-1-null transgenic mice. [94] Furthermore,
TNF and IL-6 appear to interact to regulate hepatocyte
proliferation after PH. This concept is supported by
observations that pre-treatment with IL-6 inhibits liver
regeneration after PH in normal mice but restores post-PH
hepatocyte proliferation in TNFR-1 null mice (which produce
little IL-6 mRNA in the liver after PH) [94] and in IL-6-null
mice, which have TNF receptors and presumably express TNF
after PH. [21] We recently noted that the induction of IL-6
mRNAs was slightly attenuated and delayed in the livers of
ethanol-fed rats after PH [94A] ; however, the physiologic
implications of this finding
728

are uncertain as ethanol feeding is associated with an

up-regulation of IL-6 expression in other tissues.
Several other TNF-related cytokines also increase, at least
transiently, after PH. For example, we have noted increases in
cytokines that promote TNF's activity (e.g., IL-12 and IFN
gamma, IL-1 beta). Sequential analysis of the expression of
various cytokine mRNAs in the liver over the initial 24 hours
after PH demonstrates a transient induction of TNF that begins
within 5 minutes after PH and persists for about 1 hour. [80] [81]
During this time period, expression of IL-1 beta also increases
transiently although induction of IFN gamma and IL-12 occur
somewhat later. [94A] Other cytokines (e.g., IL-10 and TGF beta)
that down-regulate the effects of TNF also are induced after PH.
For example, there is a rapid, transient, up-regulation of IL-10
mRNAs in the liver within the initial 30 to 60 minutes after PH.
[81] TGF beta induction occurs a bit later. [11] [81] We have
reported that chronic ethanol feeding generally enhances the

regenerative induction of all of these cytokines after PH. [94A]
Almost nothing is known about the cellular source of these
various cytokines during liver regeneration; however, emerging
evidence suggests that different cells may be predominately
responsible for producing a given cytokine at different time
points in the regenerative process. This was elegantly
demonstrated by Bissell's work, which showed that Kupffer
cells are a major source of TGF beta Tbefore PH, but that an
increased expression of this gene by hepatocytes accounts for
the induction of TGF beta in the regenerating liver remnant. [11]
Similarly, in situ hybridization analysis demonstrates that
hepatic and portal venous endothelia and biliary epithelia are
the major producers of TNF during the prereplicative period
following PH. [60] The latter finding was quite unanticipated,
because tissue macrophages (e.g., Kupffer cells) are a major
source of TNF after LPS exposure. [23] Experiments with
gadolinium chloride, a Kupffer cell depleting drug, suggest that
the in situ hybridization results are valid because hepatic levels
of TNF mRNA and circulating levels of TNF protein are
increased in the livers of gadolinium chloride pre-treated rats
after PH. [81] Those same studies also revealed that Kupffer cells
are likely to be important producers of IL-10 after PH, because
regenerative induction of IL-10 was virtually abolished in rats
that were pre-treated with gadolinium chloride. Taken together,
those findings, coupled with the known anti-TNF actions of
IL-10, [4] [7] [44] prompted us to speculate that Kupffer cells
normally act to restrict the TNF response to PH. [79] [80] [81] More
work is required to identify which cell populations produce

each of the various cytokines during normal liver regeneration.

It also is important to determine if conditions that lead to
impaired liver regeneration and chronic liver damage alter this
normal pattern of cytokine production. For example, several
groups have demonstrated increased production of TNF by
Kupffer cells during the evolution of experimentally-induced
alcoholic liver damage [1] [47] [51] but it is not clear if or how this
might influence the regenerative response.
Cytokine Production by Extra-Hepatic Tissues
Very recently, we demonstrated that extra-hepatic tissues,
particularly white adipose tissue, are important cytokine
producers after PH. In healthy adult rats, we found little
evidence of adipose TNF, IL-6, or IL-10 expression before PH;
however, within a few hours after PH, mRNAs for each of these
cytokines were easily demonstrated in white adipose tissue. In
healthy rats, the induction of cytokine gene expression in fat
generally lagged behind the induction of these same genes in
the liver but followed a similar pattern as the hepatic response.
[94A] Interestingly, all of these cytokines were relatively
overexpressed in the fat of
729
rats that had been fed ethanol chronically. Indeed, ethanol-fed

rats exhibited constitutive expression of TNF mRNA in white
adipose tissue, and PH led to a more rapid, intense and
sustained induction of TNF and the other cytokines in the
ethanol-fed rats when compared with the pair-fed controls.
Thus, in aggregate, the evidence suggests that chronic
consumption of ethanol generally increases the production of
injury-related, proinflammatory cytokines in several tissues,
including the liver and fat. Hepatocyte Mitogens and
Comitogens
Only recently has much attention been given to the
growth-regulatory roles of proinflammatory cytokines. It has
long been recognized that many other molecules are potent
growth regulators. Molecules that stimulate the proliferation of
cultured cells in the absence of serum are termed "mitogens."
Those that enhance the growth-stimulatory actions of mitogens,
but alone cannot affect DNA synthesis, are designated
"comitogens." Although there is evidence that each of these
factors is likely to have some role in hepatic regeneration, few,
if any, of the hepatocyte mitogens are specific for hepatocytes
and most are produced in many tissues, not only the liver. Thus,
it is not clear why liver cells proliferate relatively selectively in
http://home.mdconsult.com/das/journal/view/N/108...stty=cow&nid=39982&tlt=Z&pos=&tag=relinfo&anc=ja (10 of 26) [21/04/2000 23:21:50]
response to liver injury. Hepatocyte comitogens also are

somewhat curious, since most are hormones with principle roles
to regulate cellular substrate use and energy homeostasis.
Presumably, timing is the secret to the initiation of a
proliferative response in hepatocytes, which spend most of their
adult life "outside" of the cell cycle. That is, the cytokines,
mitogens, and comitogens must be presented to liver cells in the
appropriate sequence in order to elicit the sequential responses
that move hepatocytes into and through various stages of the
cell cycle. The growth response is likely to be initiated by local
factors that are most abundant within the injured liver, and then
amplified and propagated by factors that are recruited from
extra-hepatic sites.
Little is known about the mechanisms that operate within the
regenerating liver to coordinate the local production and release
of mitogenic factors. Errors in this critical process are likely to
underlie the genesis of chronic liver injury and, perhaps, even
the evolution of cirrhosis. The latter is suggested by recent
evidence that hepatic stellate cells, long acknowledged to be the
principle source of liver extracellular matrix, also are major
producers of mitogens during normal liver regeneration. For
example, several groups have noted that after PH, stellate cells
are rich sources of hepatocyte growth factor (HGF), a potent
hepatocyte mitogen. [85A] We found that TNF plays a role in the
induction of stellate cell HGF production after PH. After PH,
the degree of stellate cell activation (as indicated by loss of
vitamin A and increased expression of a smooth muscle actin)
and hepatic HGF expression correlate with levels of TNF.
Furthermore, treatment of cultured primary stellate cells with
recombinant TNF elicits a dose-related induction of IL-6
expression, activation of the IL-6 responsive transcription
factor, C/EBP beta, and increased expression of HGF, a gene
that is transcriptionally activated by C/EBP beta. [78A] Thus,
liver nonparenchymal cells also appear to be targets for injury
related cytokines and, under at least some circumstances, TNF
stimulates stellate cells to produce factors that are mitogenic for
hepatocytes. Presumably, the kinetics of the latter response are
such that stellate-derived mitogens are available maximally
when hepatocytes are optimally sensitive to mitogens. Because
liver nonparenchymal cell proliferation begins well after
hepatocyte proliferative activity has peaked post-PH, it also is
likely that hepatocytes produce factors that regulate the
proliferation of liver nonparenchymal cells. Whether or
730

not chronic alcohol consumption influences stellate cell

production of hepatocyte mitogens or hepatocyte production of
factors that regulate the phenotype of stellate cells is unknown.
A better understanding of the cell-to-cell communications that
orchestrate the normal regenerative response to liver injury is
required before we can design treatments to optimize the
response in patients with chronic liver injury caused by alcohol
or other agents.
Requirement 2: Proliferation of a Differentiated

Population of Cells That Ordinarily Exist in a
Nonproliferative State
Initiation of the Proliferative Response in Hepatocytes (G0

-to-G1 Transition)
TNF and other injury-related cytokines initiate hepatocyte entry
into the prereplicative phase of the cell cycle (i.e., G1 ). Studies
in transgenic mice, with deletions of TNF receptor genes and
neutralizing anti-TNF antibody treatment of normal rodents
subjected to PH, complement work that has been done in many
types of cultured cells and identify specific signals that mediate
this response. It is now well established that after PH, TNF
interacts with both classes of TNF receptors on hepatocytes to
activate discrete, but overlapping, pathways that regulate the
activity of various kinases which, in turn, modulate the activity
of transcription factors that regulate the transcription of genes
that encode molecules required for proliferation as well as
genes that encode hepatocyte-specific products (e.g., albumin,
clotting factors). For example, after PH, TNF-dependent
activation of Jun nuclear kinase is necessary for the
proto-oncogene product, c-Jun, to accumulate in liver cell
nuclei and participate in DNA complex formation with activator
protein (AP)-1, a transcription factor that modulates the
expression of many growth-regulatory genes. [33] TNF also
plays a major role in promoting the redistribution of other
transcription factors, including NF kB, C/EBP beta, and C/EBP
delta, from the cytosol and into the nuclei of hepatocytes during
the initial hours after PH. [34] [94] [95] The latter two
transactivators displace one of their family members, C/EBP
alpha. C/EBP alpha is the major transcriptional activator of the
albumin gene, but it also functions as a constitutively expressed,
growth-arrest gene in hepatocytes. [26] Thus, TNF/TNF receptor
interactions help to initiate the hepatocyte proliferative response
both by inducing the expression of genes (e.g., c-Jun) that
stimulate proliferation and by inhibiting the expression of other

genes (e.g., C/EBP alpha) that normally keep hepatocytes from

proliferating. Furthermore, by promoting the substitution of
closely related, growth-permissive transcriptional regulators
(e.g., C/EBP beta and C/EBP delta) for constitutive, growth
inhibitory transcription factors (e.g., C/EBP alpha), TNF helps
to assure that hepatocyte-specific, differentiated gene
expression is maintained as these cells prepare to proliferate.
Little is known about the effect of ethanol consumption on the
propagation of the signals that regulate re-entry into the cell
cycle. We have reported that acute ingestion of an intoxicating
dose of ethanol actually amplifies the activation of the
stress-related protein kinase, Jun nuclear kinase (JNK), and
speeds the nuclear accumulation of c-Jun after PH; however, we
found that chronic consumption of ethanol had the opposite
effect on these responses, inhibiting the induction of JNK and
c-Jun after PH. [96] Similarly, we noted that acute ethanol
ingestion accelerated the activation of NF kB after PH, [96] and
chronic ingestion of ethanol virtually abolished the activation of
this transcription factor after PH. [94A] These opposing effects of
acute and chronic alcohol appear to be relatively
731
specific for stress- or oxidant induced responses because we

noted that both acute and chronic exposure to ethanol inhibit the
induction of mitogen activated kinases after PH. [96] The latter is
consistent with the work of other groups, which indicates that
ethanol exposure inhibits the ability of the mitogen, epidermal
growth factor (EGF), and the comitogen, insulin, to induce
proliferation in primary cultures of hepatocytes. [15] Curiously,
although chronic ethanol consumption inhibited at least two
stress-related responses (i.e., the induction of NF kB and JNK)
in regenerating hepatocytes, it failed to block the activation of
Stat-3, [94A] an IL-6-responsive transcription factor that is
normally induced after PH. [15] Furthermore, we have had
difficulty identifying ethanol-related inhibition of other
IL-6-regulated responses, such as the induction of C/EBP beta
or C/EBP delta mRNAs after PH. [31] Thus, it does not appear
that ethanol inhibition of liver regeneration can be explained by
an interruption of IL-6-related signals, although the latter
clearly are required for the normal regenerative response. [21] [94]
Progression from the Prereplicative Period (G1 ) into the
Replicative Phases (S, M) of the Cell Cycle
Unless they have already entered the prereplicative phase of the
cell cycle (G1 ), hepatocytes respond poorly to mitogens. This is
nicely illustrated by evidence that classical hepatocyte

mitogens, including epidermal growth factor (EGF),
transforming growth factor alpha (TGF alpha), and HGF, have
only weak growth stimulatory actions when infused into the
livers of healthy adult animals. [67] These same factors are
effective mitogens for primary hepatocytes in culture. [67] This
apparent paradox has been resolved by recent evidence that, like
other types of liver injury, the hepatocyte isolation protocol acts
as a cell cycle initiator, inducing stress-activated protein kinases
and immediate early genes in hepatocytes. It remains uncertain
precisely how the activation of these kinases and genes sensitize
hepatocytes to the trophic actions of mitogens; however, the
intracellular events that follow the interaction of various
mitogens with their receptors have been extensively
categorized.
Most growth factor receptors either possess intrinsic tyrosine
kinase activity (e.g., the EGF or HGF or insulin receptors) or
they are capable of recruiting cytosolic tyrosine kinases of the
Janus kinase family to the plasma membrane (e.g., the IL-6
receptor). Receptor-associated tyrosine kinases recruit other
adaptor proteins (e.g., Grb-2, SOS) that couple to various signal
transduction pathways, including those that activate
phospholipid hydrolysis, promote the accumulation of
intracellular calcium, and induce other kinases and
phosphatases. Mitogenic factors that activate seven
transmembrane-spanning domain receptors (e.g., glucagon,
adrenergic hormones) can also interact with signaling
components of the receptor tyrosine kinase cascades via their
associated GTP-binding proteins.
The series of intracellular events that propagate
mitogen-initiated signals at the cell surface into the nucleus
have been a particular focus of investigation because activating
mutations of the components of this signal transduction
pathway often lead to malignant transformation. The
mitogen-activated kinase cascade culminates with the activation
of mitogen activated protein kinase (also called Extracellular
regulated kinase [Erk]), a serine threonine kinase, whose targets
include growth regulatory transcription factors, such as the
proto-oncogene, c-Jun. Phosphorylation of c-Jun and other
DNA-binding proteins alters their DNA binding activities or
transactivating potential, thus modulating the level of
expression of all of the genes that are regulated by these factors.
Thus, mitogen- and stress-activated kinase cascades share
downstream targets (e.g.,

732
c-Jun) that regulate a wide array of cellular genes and thus,

affect global changes in the hepatocyte phenotype.
There is no doubt that chronic consumption of ethanol prevents
mitogen-stimulated hepatocytes from entering S phase given
results published by several groups who have studied the PH
model, [27] [37] [43] [77] [92] as well as data obtained from
hepatocyte culture experiments. [15] Although chronic ethanol
exposure thwarts the propagation of some pre-replicative
signals, many others are transduced. Ultimately, the induction
of the immediate-early gene response that characterizes
successful entry into the cell cycle occurs more-or-less normally
in ethanol-treated rats after PH. [30] Recently, we discovered that
chronic exposure to ethanol prevents the induction of many
delayed-early genes, such as cyclin D1 , cyclin D3 , cyclin A,
p53, p21, that normally occurs as hepatocytes prepare to exit G1
and enter S phase. [94A] These findings identify the site of cell
cycle arrest because of ethanol (i.e., chronic ethanol ingestion
arrests initiated hepatocytes in mid-late G1 ).
Because the hepatocyte is bombarded with many mitogens and
comitogens after PH, it is improbable that ethanol would be
able to arrest hepatocytes in prereplicative stages of the cell
cycle by inhibiting a single growth factor or its receptor. Rather,
it is likely that ethanol-associated G1 growth arrest reflects a
block in some common growth-regulatory pathways that are
downstream from multiple growth factor receptors. The latter
concept is supported by evidence that chronic ethanol
consumption inhibits the induction of ornithine decarboxylase
(ODC), the rate limiting enzyme for polyamine biosynthesis. [28]
Activation of ODC is an effect of many mitogens and is
necessary for liver regeneration after PH. [61] Furthermore,
treatment with supplemental putrescine, the immediate product
of ODC, normalizes hepatic polyamine levels and improves
liver regeneration after PH in ethanol-fed rats. [28] Increases in
polyamine synthesis begin in mid-late G1, around the time
when induction of the delayed-early, cell-cycle-dependent
genes begins. Thus, it is conceivable that ethanol affects the
propagation of one or more proximal signals that activate
various targets in mid-G1 . Indeed, chronic ethanol exposure is
known to inhibit insulin or insulin-related growth
factor-dependent phosphorylation of insulin-receptor substrate
(IRS)-1. [8A] It also interferes with normal trafficking of the
EGF receptor [22A] and alters the activation of heterotrimeric

GTP-binding proteins that couple alpha and beta-adrenergic

receptors to their intracellular targets. [29] One consequence of
the latter is that chronic ethanol exposure inhibits regenerative
activation of adenylyl cyclase, blocking the accumulation of
cAMP in the liver after PH. [29] Ethanol-related inhibition of
insulin/IGF signaling and decreases in cAMP are predicted to
have global effects on hepatocyte intermediary metabolism and
energy homeostasis. Indeed, chronic consumption of ethanol is
known to reduce liver ATP stores and inhibit mitochondrial
respiration. [22] Given the energy demands of tissue repair, it is
conceivable that, ultimately, ethanol may limit liver
regeneration by creating an intracellular environment that
compromises hepatocyte energy homeostasis. If confirmed, this
hypothesis could have important pathophysiologic implications
as mitochondrial dysfunction and energy depletion also are
believed to contribute to alcohol-induced liver injury.
Requirement 3: Temporary Suspension of the Normal

Mechanisms that Couple Cell Cycle Re-entry to Cell
Death
The size of an adult organ is guarded carefully. Thus, cells

proliferate normally only to replace their senescent neighbors.
Because programmed hepatocyte
733
death (i.e., apoptosis) is a relatively rare event in the healthy

liver, hepatocyte proliferation also occurs infrequently.
However, insults that increase the rate of hepatocyte death
generally result in an increased rate of hepatocyte proliferation.
A breakdown in the mechanisms that couple hepatocyte
proliferative activity to the rate of hepatocyte death results in a
net increase in liver mass. This has been induced experimentally
in transgenic mice with dysfunctional Fas receptors (FasR). The
latter animals develop massive hepatomegaly because they have
decreased hepatocyte apoptosis, and they clear senescent
hepatocytes inefficiently. Because the low endogenous rate of
hepatocyte proliferation is not balanced by deletion of senescent
hepatocytes in FasR-mutant mice, the total number of
hepatocytes gradually increases over time and the liver
continues to grow in adulthood.
In healthy adult mammals, liver size remains constant despite
hepatocyte turnover, because the rates of hepatocye
proliferation and apoptosis are coupled. The adult liver cannot
recover from sudden, exogenous insults that abruptly decrease
the number of hepatocytes (e.g., PH or hepatotoxins) unless

these normal homeostatic mechanisms that couple proliferative
and apoptotic activities are suspended. This temporary
uncoupling of proliferation and apoptosis permits the rate of
hepatocyte proliferation to outstrip the rate of apoptosis until
the liver recovers its pre-injury mass. The latter implies that
successful liver regeneration requires relative inhibition of
hepatocyte apoptosis.
Consistent with this concept, several anti-apoptotic genes (e.g.,
bcl-2 and bcl-x) and genes that regulate the production of
cytoprotective factors (e.g., inducible nitric oxide synthases
(iNOS) are induced in the mid-to-late prereplicative period
following PH. [24] [25] [57] [74] [75] [90] Furthermore, PH triggers
increased hepatocyte apoptosis, rather than proliferation, in
transgenic mice with targeted deletion of one of these gene (i.e.,
iNOS). [78A] It is interesting that although TNF is required for
liver regeneration after PH, it can also induce hepatocyte
apoptosis. The latter only occurs in hepatocytes when RNA or
protein synthesis are inhibited. This observation suggests that
when exposed to TNF, hepatocytes protect themselves by
synthesizing one or more antiapoptotic factors. Given evidence
that TNF induces the expression of bcl-2, [58] iNOS, and heat
shock protein (Hsp)-70 [9] [48] [54] [55] [68] in cultured hepatocytes,
it is conceivable that TNF may promote the induction of these
factors in regenerating hepatocytes after PH.
It is not known if ethanol alters the induction of TNF-regulated
antiapoptotic factors after PH. Although a few groups have
suggested that chronic exposure to ethanol via inhalation [45] or
intragastric infusion [93] increases basal apoptosis in rodent
livers, and apoptotic bodies have been reported in patients with
alcoholic hepatitis, [53] increased apoptosis was not noted in
patients with simple alcohol-induced steatosis. [53] Indeed, mice
with ethanol-induced fatty livers have increased hepatocyte
expression of two antiapoptotic proteins, bcl-2 and bcl-x. After
2 to 3 weeks of consuming ethanol-containing diets, most of the
murine hepatocytes in acinar zone 3 and many of those in acinar
zone 2 express these antiapoptotic factors. [81A] Because cell
death and proliferation are fundamentally related to each other,
ethanol effects on hepatocyte apoptosis are likely to influence
its effects on hepatocyte proliferation and vice versa. Efforts to
clarify how ethanol influences apoptosis are likely to improve
our understanding of how it inhibits liver regeneration.

CONCLUSION
Chronic consumption of ethanol is one of the few insults that
can lead to liver injury without inducing a compensatory
proliferative response in surviving
734
hepatocytes. This observation suggests that ethanol selectively

blocks the growth stimulatory actions of injury-related
cytokines and results in a differential sensitivity to their toxic
actions. Thus, efforts to understand how ethanol produces these
responses may identify therapeutic targets that can be
manipulated to abort injury and coax regeneration in
ethanol-damaged livers.
References
1. Adachi Y, Moore LE, Bradford BU, et al: Antibiotics prevent liver

injury in rats following long-term exposure to ethanol. Gastroenterology
108:218-224, 1995
2. Akerman P, Cote P, Yang SQ, et al: Antibodies to tumor necrosis factor

alpha inhibit liver regeneration after partial hepatectomy. Am J Physiol
263:G579-G585, 1992
3. Akerman PA, Cote PM, Yang SQ, et al: Long-term ethanol consumption
alters the hepatic response to the regenerative effects of tumor necrosis
factor alpha. Hepatology 16:1212-1219, 1993
4. AraiT, Hiromatsu K, Kobayashi N, et al: IL-10 is involved in the

protective effect of dibutyryl cyclic adenosine monophosphate on
endotoxin-induced inflammatory liver injury. J Immunol 155:5743-5749,
1995
5. BarsigJ, Kusters S, Vogt K, et al: Lipopolysaccharide-induced

interleukin-10 in mice: Role of endogenous tumor necrosis factor-alpha.
Eur J Immunol 25:2888-2893, 1995
6. Bemelmans MHA, Gouma DJ, Greve JW, et al: Cytokines, tumor

necrosis factor and interleukin-6 in experimental biliary obstruction in
mice. Hepatology 15:1132-1136, 1992
7. Berg DJ, Kuhn R, Rajewsky K, et al: Interleukin-10 is a central regulator

of the response to LPS in murine models of endotoxic shock and the
Shwartzman reaction but not endotoxin tolerance. J Clin Invest
96:2339-2347, 1995
8. BeyerHS, Stanley M, Theologides A: Tumor necrosis factor-alpha

increases hepatic DNA and RNA and hepatocyte mitosis. Biochemistry

International 22:405-410, 1990
8A. Bhavani K, de la Monte S, Brown NV, et al: Effect of ethanol on p36

protein kinase substrate and insulin receptor substrate 1 expression and
tyrosyl phosphorylation in human hepatocellular carcinoma cells. Alcohol
Clin Exp Res 19:441-446, 1995
9. Billiar
TR, Curran RD, Harbrecht BG, et al: Modulation of nitrogen
oxide synthesis in vivo: NG-monomethyl-L-arginine inhibits
endotoxin-induced nitrite/nitrate biosynthesis while promoting hepatic
damage. J Leukoc Biol 48:565-569, 1990
10. Bird GLA, Sheron N, Goka AKJ, et al: Increased plasma tumor necrosis
factor in severe alcoholic hepatitis. Ann Intern Med 112:917-920, 1990
11. Bissell
DM, Wang SS, Jarnagin WR, et al: Cell-specific expression of
transforming growth factor-beta in rat liver. Evidence for autocrine
regulation of hepatocyte proliferation. J Clin Invest 96:447-455, 1995
12. BucherNLR: Liver regeneration: An overview. J Gastroenterol Hepatol

6:615-624, 1991
13. Campbell KA, Wu Y, Chacko VP, et al: P31-MR spectroscopic changes

during liver regeneration. J Surg Res 49:244-247, 1990
14. CarBD, Eng VM, Schnyder B, et al: Interferon gamma receptor

deficient mice are resistant to endotoxic shock. J Exp Med 179:1437-1444,
1994
15. Carter
EA, Wands JR: Ethanol inhibits hormone stimulated hepatocyte
DNA synthesis. Biochem Biophys Res Commun 128:767-774, 1985
16. Chapekar MS, Hugget AC, Thorgeirsson SS: Growth modulatory

effects of a liver-derived growth inhibitor, transforming growth factor beta
1 and recombinant tumor necrosis factor alpha, in normal and neoplastic
cells. Exp Cell Res 185:247-257, 1989
17. Coletti
LM, Remick DG, Burch GD, et al: The role of tumor necrosis
factor-alpha in the pathophysiologic alterations after hepatic
ischemia/reperfusion in the rat. J Clin Invest 85:1936-1943, 1990
18. Cornell RP, Leljequist BL, Bartizal KF: Depressed liver regeneration
after partial hepatectomy of germ-free, athymic and
lipopolysaccharide-resistant mice. Hepatology 11:915-922, 1990
735
19. Cornell RP: Gut-derived endotoxin elicits hepatotrophic factor secretion

for liver regeneration. Am J Physiol 249:R551-R562, 1985
20. Cornell RP: Restriction of gut-derived endotoxin impairs DNA

synthesis for liver regeneration. Am J Physiol 249:R563-R569, 1985
21. Cressman DE, Greenbaum LE, DeAngelis RA, et al: Liver failure and
defective hepatocyte regeneration in interleukin-6-deficient mice. Science
274:1379-1383, 1996
22. CunninghamCC, Coleman WB, Spach PI: The effects of chronic

ethanol consumption on hepatic mitochondrial energy metabolism.
Alcohol Alcohol 25:127-136, 1990
22A. Dalke DD, Sorrell MF, Casey CA, et al: Chronic ethanol
administration impairs receptor-mediated endocytosis of epidermal growth
factor by rat hepatocytes. Hepatology 12:1085-1091, 1990
23. DeckerK: Biologically active products of stimulated liver macrophages

(Kupffer cells). Eur J Biochem 192:245-261, 1990
24. Decker KF, Obolenskaya MY: Cytokines, nitric oxide synthesis and
liver regeneration. J Gastroenterol Hepatol 10:S12-S17, 1995
25. Diaz-Guerra MJM, Velasco M, Martin-Sanz P, et al: Nuclear factor kB

is required for the transcriptional control of type II NO synthase in
regenerating liver. Biochem J 326:791-797, 1997
26. Diehl AM, Johns D, Yang SQ, et al: Adenovirus-mediated transfer of

CCAAT/enhancer binding protein alpha identifies a dominant
anti-proliferative role for this isoform in hepatocytes. J Biol Chem
271:7343-7350, 1996
27. DiehlAM, Chacon MA, Wagner P: The effect of chronic ethanol

feeding on ornithine decarboxylase activity and liver regeneration.
Hepatology 8:237-242, 1988
28. DiehlAM, Wells M, Brown ND, et al: Effect of ethanol on polyamine

synthesis during liver regeneration in rats. J Clin Invest 85:385-390, 1990
29. Diehl
AM: Effects of alcohol on liver regeneration. Alcohol Health Res
World 17:279-283, 1993
30. DiehlAM, Thorgeirsson SS, Steer CJ: Ethanol inhibits liver

regeneration in rats without reducing transcripts of key protooncogenes.
Gastroenterology 99:1105-1112, 1990
31. DiehlAM, Yang SQ: Regenerative changes in C/EBP alpha and C/EBP
beta expression modulate binding to the C/EBP site in the c-fos promoter.
Hepatology 19:447-456, 1994
32. DiehlAM, Rai R: Regulation of signal transduction during liver

regeneration. FASEB J 10:215-227, 1996
33. DiehlAM, Yin M, Fleckenstein J, et al: Tumor necrosis factor alpha

induces c-Jun during the regenerative response to liver injury. Am J
Physiol 267:G552-G561, 1994

34. Diehl
AM, Yang SQ, Yin M, et al: Tumor necrosis factor alpha
modulates CCAAT/Enhancer binding proteins activities and promotes
hepatocyte-specific gene expression during liver regeneration. Hepatology
22:252-261, 1995
35. D'Souze-Gburek S, Batta KP, Nikias GA, et al: Liver transplantation for
jejunoileal bypass-associated cirrhosis: Allograft histology in the setting of
an intact bypassed limb. Liver Transpl Surg 3:23-27, 1997
36. Dufour MC, Stinson FS, Fe Cases M: Trends in cirrhosis morbidity and
mortality: United States, 1979-1988. Semin Liver Dis 13:109-125, 1993
37. Dugay L, Couter D, Hetu C, et al: Inhibition of liver regeneration by

chronic alcohol administration. Gut 23:8-13, 1982
38. FangJWS, Bird GLA, Nakamura T, et al: Hepatocyte proliferation as

an indicator of outcome in acute alcoholic hepatitis. Lancet 343:820-823,
1991
39. Feingold KR, Barker ME, Jones AL, et al: Localization of tumor
necrosis factor-stimulated DNA synthesis in the liver. Hepatology
13:337-339, 1991
40. Feingold
KR, Soued M, Grunfeld C: Tumor necrosis factor stimulates
DNA synthesis in the liver of intact rats. Biochem Biophys Res Commun
153:576-582, 1988
41. Felver
ME, Mezey E, McGuire M, et al: Plasma tumor necrosis factor
alpha predicts decreased long-term survival in severe alcoholic hepatitis.
Alcohol Clin Exp Res 14:255-259, 1990
42. Flesch
IEA, Hess JH, Huang S, et al: Early interleukin 12 production by
macrophages
736
in response to mycobacterial infection depends on interferon gamma and

tumor necrosis factor alpha. J Exp Med 181:1615-1621, 1995
43. FrankWO, Rayes AN, Washing A, et al: Effect of acute ethanol

administration upon hepatic regeneration. J Lab Clin Med 93:402-413,
1979
44. Gerad C, Bruyens C, Marchant A, et al: Interleukin 10 reduces the

release of tumor necrosis factor and prevents lethality in experimental
endotoxemia. J Exp Med 177:547-550, 1993
45. GoldinRD, Hunt NC, Clark J, et al: Apoptotic bodies in a murine

model of alcoholic liver disease: Reversibility of ethanol-induced changes.
J Pathol 171:73-76, 1993
46. Goto M, Takei Y, Kawano S, et al: Tumor necrosis factor and

endotoxin in the pathogenesis of liver and pulmonary injuries after

orthotopic liver transplantation in the rat. Hepatology 16:487-493, 1992
47. Hansen J, Cherwitz DL, Allen JI: The role of tumor necrosis factor
alpha in acute endotoxin-induced hepatotoxicity in ethanol-fed rats.
Hepatology 20:461-474, 1994
48. HarbrechtB, Stadler J, Demetris A, et al: Nitric oxide and

prostaglandins interact to prevent hepatic damage during murine
endotoxemia. Am J Physiol 266:G1004-G1010, 1994
49. HillD, Marsano L, Cohen D, et al: Increased plasma interleukin-6

activity in alcoholic hepatitis. J Lab Clin Med 119:507-512, 1992
50. Hishinuma I, Nagakawa J, Hirota K, et al: Tumor necrosis factor-alpha

in galactosamine-induced hepatitis. Hepatology 12:1187-1191, 1990
51. HonchelR, Ray RB, Marsano L, et al: Tumor necrosis factor in alcohol
enhanced endotoxin liver injury. Alcohol Clin Exp Res 16:665-669, 1992
52. IanaroA, Xu D, O'Donnell CA, et al: Expression of TGF-beta in

attenuated Salmonella typhimurium: Oral administration leads to the
reduction of inflammation, Il-12 and IFN-gamma, but enhancement of
IL-10, in carrageenin-induced oedema in mice. Immunology 84:8-15, 1995
53. Kawahara H, Matsuda Y, Takase S: Is apoptosis involved in alcoholic

hepatitis? Alcohol Alcohol 29:113-118, 1994
54. Kim YM, Lee BS, Yi KY, et al: Upstream Nf-kappa B site is required
for the maximal expression of mouse inducible nitric oxide synthase gene
in interferon-gamma plus lipopolysaccharide-induced RAW 264.7
macrophages. Biochem Biophys Res Commun 236:655-660, 1997
55. Kim YM, Talanian RV, Billiar TR: Nitric oxide inhibits apoptosis by
preventing increases in caspase-3-like activity via two distinct
mechanisms. J Biol Chem 272:31138-31148, 1997
56. KnolleP, Schlaak J, Uhrig A, et al: Human Kupffer cells secrete IL-10
in response to lipopolysaccharide (LPS) challenge. J Hepatol 22:226-229,
1995
57. KrenBT, Trembley JH, Krarewski S, et al: Modulation of

appoptosis-associated genes bcl-2, bcl-x, and bax during liver
regeneration. Cell Growth Differ 7:1633-1642, 1996
58. Kroemer G: The proto-oncogene Bcl-2 and its role in regulating

apoptosis. Natl Med J 3:614-620, 1997
59. Leevy CM: In vitro studies of hepatic DNA synthesis in percutaneous

liver biopsy specimens. J Lab Clin Med 61:761-779, 1963
60. Loffreda S, Rai R, Yang SQ, et al: Bile ducts and portal and central
veins are major producers of tumor necrosis factor alpha in regenerating
rat liver. Gastroenterology 112:2089-2098, 1997
61. Luk GD: Essential role of polyamine metabolism in hepatic

regeneration: Inhibition of protein synthesis and tissue regeneration by
difluoromethylornithine in the rat. Gastroenterology 90:1261-1267, 1986
62. McClain CJ, Hill DB, Schmidt J, et al: Cytokines and alcoholic liver
disease. Semin Liv Dis 13:170-182, 1993
63. McClainCJ, Cohen DA: Increased tumor necrosis factor production by

monocytes in alcoholic hepatitis. Hepatology 9:349-351, 1989
64. McClain CJ, Cohen DA, Dinarello CA, et al: Serum interleukin-1
(IL-1) activity in alcoholic hepatitis. Life Sci 39:1479-1485, 1986
65. MealyK, Wilmore DW: Tumor necrosis factor increases hepatic cell
mass. Br J Surg 78:331-333, 1991
737
66. Medvedev AE, Espevik T, Ranges G, et al: Distinct roles of the two
tumor necrosis factor (TNF) receptors in modulating TNF and
lymphotoxin alpha effects. J Biol Chem 271:9778-9784, 1996
67. MichalopoulosG: Liver regeneration: Molecular mechanisms of growth

control. FASEB J 4:176-187, 1990
68. Mohr S, Zech B, Lapetina EG, et al: Inhibition of caspase-3 by

S-nitrosation and oxidation caused by nitric oxide. Biochem Biophys Res
Commun 238:387-391, 1997
69. Nagakawa J, Hishinuma I, Hirota K, et al: Involvement of tumor

necrosis factor alpha in the pathogenesis of activated
macrophage-mediated hepatitis. Gastroenterology 99:758-765, 1990
70. Nanji
AA, Khettry U, Sadrzadeh SM: Lactobacillus feeding reduces
endotoxemia and severity of experimental alcoholic liver disease. Proc Soc
Exp Biol Med 205:243-247, 1994
71. Narumi
S, Wyner LM, Stoler MH: Tissue-specific expression of murine
IP-10 mRNA following systemic treatment with interferon gamma. J
Leukoc Biol 52:27-33, 1992
72. NeurathMF, Fuss I, Kelsall BL, et al: Antibodies to interleukin 12

abrogate established experimental colitis in mice. J Exp Med
182:1281-1290, 1995
73. Nill
MR, Oberyszn TM, Ross MS, et al: Temporal sequence of
pulmonary cytokine gene expression in response to endotoxin in C3H/HeN
endotoxin-sensitive and C3H/HeJ endotoxin-resistant mice. J Leukoc Biol
58:563-574, 1995

74. Obolenskaya M, Schulze-Specking A, Plaumann B, et al: Nitric oxide

production by cells isolated from regenerating rat liver. Biochem Biophys
Res Commun 204:1305-1311, 1994
75. Obolenskaya MY, Vanin AF, Mordvintcex PZ, et al: EPR evidence of
nitric oxide production by the regenerating rat liver. Biochem Biophys Res
Commun 202:571-576, 1994
76. Orange JS, Salazar-Mather TP, Opal SM, et al: Mechanism of

interleukin 12-mediated toxicities during experimental viral infections:
Role of tumor necrosis factor and glucocorticoids. J Exp Med
181:901-914, 1995
77. OrregoH, Crossley IR, Saldiva V, et al: Long-term ethanol

administration and short- and long-term liver regeneration after partial
hepatectomy. J Lab Clin Med 97:402-413, 1981
78. Pauli
U: Control of tumor necrosis factor gene expression. Crit Rev
Eukaryot Gene Exp 4:323-344, 1994
78A. Rai RM, Anania FA, Bagby G, et al: Proinflammatory cytokines

induce hepatocyte growth factor expression in the regenerating rat liver
and in cultured rat stellate cells. Am J Physiol (in press).
79. RaiR, Diehl AM: Review: Regulation of liver generation by

proinflammatory cytokines. J Gastroenterol Hepatol 11:466-470, 1996
80. Rai RM, Yang SQ, McClain C, et al: Kupffer cell depletion by
gadolinium chloride enhances liver regeneration after partial hepatectomy
in rats. Am J Physiol 270:G909-G918, 1996
81. RaiRM, Loffreda S, Karp CL, et al: Kupffer cell depletion abolishes
induction of interleukin-10 and permits sustained over-expression of tumor
necrosis factor alpha mRNA in the regenerating rat liver. Hepatology
25:889-895, 1997
81A. Rashid A, Lin HZ, Yang SQ, et al: Mitochondrial proteins that
regulate apoptosis and necrosis are induced in fatty liver. Hepatology (in
press).
82. RothschildMA, Berk PD, Lieber CS: Alcoholic liver disease. Semin
Liv Dis 13, 1993
83. Ruddle
N: Tumor necrosis factor (TNF-alpha) and lymphotoxin
(TNF-beta). Curr Opin Immunol 4:327-332, 1992
84. Schenker S, Perkins HS, Sorrell MF: Should patients with end-stage
alcoholic liver disease have a new liver? Hepatology 11:314-319, 1990
85. Schuchmann M, Hess S, Bufler P, et al: Functional discrepancies

between tumor necrosis factor and lymphotoxin alpha explained by trimer
stability and distinct receptor interactions. Eur J Immunol 25:2183-2189,
1995
85A. Tomiya T, Ogata I, Fujiwara K: Transforming growth factor alpha

levels in liver and blood correlate better than hepatocyte growth factor
with hepatocyte proliferation during liver regeneration. Am J Pathol
153:955-961, 1998
738
86. Tracey KJ, Cerami A: Tumor necrosis factor, other cytokines and
disease. Ann Rev Cell Biol 9:317-343, 1993
87. Trautwein C, Rakemann T, Niehof M, et al: Acute-phase response

factor, increased binding, and target gene transcription during liver
regeneration. Gastroenterology 110:1854-1862, 1996
88. Trinchieri
G: Interleukin-12: A cytokine produced by
antigen-presenting cells with immunoregulatory functions in the
generation of T-helper cells type 1 and cytotoxic lymphocytes. Blood
84:4008-4027, 1994
89. Trinchieri
G: Interleukin-12: A proinflammatory cytokine with
immunoregulatory functions that bridge innate resistance and
antigen-specific adaptive immunity. Ann Rev Immunol 13:251-276, 1995
90. Tzung SP, Fausto N, Hockenberry DM: Expression of Bcl-2 family

during liver regeneration and identification of Bcl-x as a delayed early
response gene. Am J Pathol 150:1985-1995, 1997
91. Vassalli
P: The pathophysiology of tumor necrosis factor. Ann Rev
Immunol 10:411-452, 1992
92. Wands JR, Carter EA, Bucher NLR, et al: Inhibition of hepatic
regeneration in rats by acute and chronic ethanol intoxication.
93. Yacoub LK, Fogt F, Griniuviene B, et al: Apoptosis and Bcl-2 protein
expression in experimental alcoholic liver disease in the rat. Alcohol Clin
Exp Res 19:854-859, 1995
94. Yamada Y, Kirillova I, Peschon JJ, et al: Initiation of liver growth by

tumor necrosis factor: Deficient liver regeneration in mice lacking type I
tumor necrosis factor receptor. Proc Natl Acad Sci USA 94:1441-1446,
1997
94A. Yang SQ, Lin HZ, Yin M, et al: Effects of chronic ethanol
consumption on cytokine regulation of liver regeneration. Am J Physiol
1998 (in press).
95. YinM, Yang SQ, Lin HZ, et al: Tumor necrosis factor alpha promotes
nuclear localization of cytokine-inducible CCAAT/enhancer binding
protein isoforms in hepatocytes. J Biol Chem 1996; 271:17974-17978

96. ZeldinG, Rai R, Yang SQ, et al: Effects of alcohol on

cytokine-inducible transcription factors. Alcohol Clin Exp Res
20:1639-1645, 1996
97. Zhang R, zur Hansen A, Hoffmann R, et al: Rat liver macrophages

express the 55 kDa tumor necrosis factor receptor: Modulation by
interferon-gamma, lipopolysaccharide and tumor necrosis factor alpha.
Biochem Chem Hoppe-Sey 375:249-254, 1994

Bookmark URL: /das/journal/view/N/10888400?ja=129184&PAGE=1.html&ANCHOR=top

MD Consult - Journal Article
Clinics in Liver Disease

Volume 2 Number 4 November 1998
723
ALCOHOLIC LIVER DISEASE
Anna Mae Diehl MD
Johns Hopkins University School of Medicine, Baltimore, Maryland

Address reprint requests to
Anna Mae Diehl, MD
912 Ross Building
720 Rutland Street
Baltimore, MD 21205
e-mail: amdiehl@welchlink.welch.jhu.edu
The liver is a major target tissue for alcohol-induced damage. Given the prevalent use of alcohol, it
is not surprising that alcohol abuse is the most common cause of clinically significant liver disease
in the United States. Deaths from cirrhosis (the eventual consequence of chronic liver injury)
correlate with per capita consumption of alcohol. Alcoholic cirrhosis, the end-stage of
alcohol-induced liver injury, currently is the fourth leading cause of death in middle-aged men and
is likely to remain a major cause of morbidity and mortality due to recent increases in alcohol
consumption among women and young teenagers. [36] Unfortunately, there is no effective medical
therapy for most patients with alcoholic liver disease. Orthotopic liver transplantation improves
the survival of patients with terminal alcoholic cirrhosis; however, transplantation carries a risk of
perioperative mortality and necessitates lifelong immunosuppressive therapy with attendant risks
of serious infection and malignancy. Furthermore, if all patients with advanced alcoholic liver
disease received liver transplants, there would be no donor organs available for patients with other
forms of cirrhosis. [84] Therefore, there still is a desperate need to develop effective medical therapy
for alcoholic liver disease.
Over the past 3 decades, considerable progress has been made towards understanding the
mechanisms by which alcohol produces liver injury. [82] It now is clear that liver injury is the end
result of multiple events that are triggered by the oxidation of ethanol to acetaldehyde, a highly
reactive metabolite. One important intermediary event in this pathogenic process appears to be the
http://home.mdconsult.com/das/article/body/jorg=journal&source=&sp=10888400&sid=0/N/129184/1.html (1 of 18) [21/04/2000 23:22:31]

release of pro-inflammatory cytokines, including tumor necrosis factor-alpha (TNF). Plasma

concentrations of TNF and TNF-inducible cytokines, including interleukins 6 (IL-6) and 8 (IL-8),
correlate with mortality in patients hospitalized with alcoholic hepatitis. [10] [41] [49] [62] [63] [64] This
suggests that anticytokine agents may be useful for interrupting the pathogenesis of
alcohol-induced liver injury. Indeed, treatments that decrease endotoxemia (a potent stimulus for
TNF production)
724
inhibit the evolution of hepatitis in rats that are being fed alcohol. [1] [70] Unfortunately, unlike
experimental animals that receive the antidote and the toxin concomitantly, most alcoholic patients
already have sustained significant liver damage before they seek medical attention. Therefore, in
clinical practice another major goal of treatment is to facilitate liver repair. Data from a number of
groups, [37] [38] [43] [59] [77] [92] including the author's, [28] [29] [30] [35] indicate that ethanol inhibits the
normally immense regenerative capacity of the liver. Thus, alcohol consumption may produce liver
disease both by injuring the liver and by impairing the regenerative response to that injury.
Relatively little is known about the mechanisms underlying ethanol's antiregenerative actions;
however, these must be defined and reversed to optimize speedy and complete recovery from
alcohol-induced liver injury.
GENERAL OVERVIEW OF LIVER REGENERATION

The liver is the only vital organ in adult mammals that has the ability to regenerate efficiently after it has
been damaged. The capacity of mature, terminally differentiated hepatocytes to proliferate in the injured
liver is surprising since, in adults, these cells rarely proliferate in the healthy liver. Rather, hepatocytes
exist in a growth-arrested (G0 ) state, preoccupied with accomplishing the vital synthetic and excretory
functions of the liver. After liver injury, remaining hepatocytes promptly leave G0 and enter the
proliferative stages of the cell cycle during which they duplicate necessary components and synthesize
new proliferation-specific molecules (G1 phase), replicate their DNA (S phase), and divide (M phase).
Rounds of proliferation continue until the premorbid mass of functional hepatocytes is restored. During
liver regeneration, nonparenchymal cell (e.g., endothelial, Kupffer, stellate, and bile duct epithelial cell)
proliferation and new matrix synthesis also are required to reconstitute the organ. Despite the increased
workload necessary for organ regeneration, liver-specific, differentiated functions must be maintained or
death of the organism ensues. This latter requirement imposes a burden on proliferating hepatocytes that
distinguishes liver regeneration from most fetal or neoplastic growth, during which the expression of
proliferative genes generally is afforded a higher priority than that of tissue-specific genes. The healthy
liver can regenerate quite efficiently, that is, increase proliferative activity while maintaining
differentiated functions. For example, following 70% (partial) hepatectomy (PH), liver functions
generally are preserved and the resected liver mass is restored within days to weeks in rats and weeks to
months in humans. [67] Because facets of the normal regenerative response fail or are dysregulated during
many acute or chronic liver diseases, workers seeking to treat these conditions are interested in
understanding liver regeneration.

PH has been a particularly useful model for defining events that regulate the proliferation of mature
("differentiated") hepatocytes because it provides a unique opportunity to evaluate cell cycle progression
in vivo. Because virtually all hepatocytes are in G0 at the time of PH, and PH is a temporally
circumscribed "insult," all remaining hepatocytes enter proliferative phases of the cell cycle
synchronously. In rats, hepatocyte DNA synthesis peaks 24 hours after PH, almost a full day before there
is appreciable DNA synthesis in nonparencymal cells. [12] [67] Thus, the initial 24 hours after PH offers a
"window" to scrutinize the events that regulate G0 -G1 and G1 -S transition in adult hepatocytes. Indeed,
it may be the only system available to study G0 -G1 transition in normal hepatocytes, because the
overwhelming majority of freshly isolated and cultured primary
725
or neoplastic hepatocytes already have entered G1 . [12] [32] [67] Furthermore, since the remnant liver lobe
was not directly manipulated at the time of PH, it is a purely regenerating system without coincident
hepatic necrosis or inflammation to confound data interpretation. The dearth of overt injury in the
remnant liver has prompted some concern that the PH model may not reliably mimic events that regulate
hepatocyte proliferation after toxic or infectious liver injury. [32] The latter criticism, however, may not be
entirely valid given emerging evidence that injury-associated cytokines (including TNF) promote
proliferation in surviving hepatocytes after PH. [2] [94] Because a relatively detailed understanding of the
molecular events that regulate hepatocyte proliferation in vivo has been developed for the PH model, this
review focuses on that data.
REGULATION OF HEPATOCYTE PROLIFERATION AFTER PH

In all species studied to date, regeneration of damaged tissues has three general requirements: an
injury-related change in the microenvironment that favors the proliferation of a differentiated population
of cells that ordinarily exist in a non-proliferative state while the usual mechanisms that couple cell cycle
re-entry to cell death are suspended temporarily. [67] Information concerning each of these requisites for
liver regeneration after PH is provided in following sections. In each section, initial comments
summarize data in normal, healthy animals and subsequent attention details the effects of acute or
chronic ethanol consumption on the normal response.
Requirement 1: Injury-Related Changes in the Microenvironment that Promote

Regeneration
Accumulation of the Injury-Related Cytokine, TNF

Several years ago, the author's research group proposed that TNF was one of the factors released during
liver injury that was required to initiate the regenerative response in the damaged liver. [2] At the time,
several lines of evidence supported this concept. TNF is a cytokine synthesized by monocytes,
macrophages, endothelium and some epithelium. Hence, TNF is present in most, if not all, tissues.
However, the liver is felt to be a major source of TNF, particularly when exposed to certain bacterial
products such as endotoxin (i.e., lipopolysaccharide [LPS]). [23] Cornell reported that restriction of
gut-derived endotoxin via antibiotics or endotoxin binding agents impaired liver regeneration after PH.

[20]Similarly, he showed that liver regeneration after PH was significantly delayed in germ-free, athymic,
and LPS-resistant mice, strains that have a limited ability to release cytokines in response to gut derived
endotoxin. [18] He also demonstrated that endotoxin pretreatment, a potent stimulus for hepatic TNF
release, enhanced liver regeneration after PH. [19] Several other groups subsequently showed that
administration of recombinant human TNF increased DNA polymerase alpha activity as well as the
incorporation of H3 -thymidine into DNA in the livers of healthy adult rats. [8] [39] [40] [65] Because serum
or tissue levels of cytokines had not been measured in any of Cornell's studies, and because relatively
large doses of recombinant TNF were delivered systemically in the latter studies, the physiologic
significance of TNF as a regulator of hepatocyte proliferation was uncertain.
To clarify the physiologic role of TNF in the compensatory hepatic regeneration
726
that is triggered by PH, the author's research group treated adult rats 1 hour before PH with neutralizing
polyclonal antibodies to TNF alpha. Several parameters of liver regeneration in anti-TNF antibody
treated rats were compared with similar end points in control rats treated with antibodies to rat IgG.
Treatment with anti-TNF antibodies pre-PH significantly inhibited post-PH increase in H3 -thymidine
incorporation into hepatic DNA, and the expression of PCNA, an S-phase-related antigen, by both
hepatocytes and liver nonparenchymal cells. [2] Many were skeptical about the physiologic significance
of these results as we were unable to detect TNF protein in the circulation after PH using the bioassays
that were available at the time; the polyclonal antisera used may not have been specific for TNF alpha.
Furthermore, there was abundant evidence that TNF played an important role in experimental liver
injury, [6] [17] [46] [47] [50] [51] [69] and recombinant TNF was known to induce cytotoxicity when micromolar
to milimolar concentrations were added to hepatoma cell lines and primary hepatocytes under certain
culture conditions. [16]
In contrast, a few labs, including our own, began to report that human recombinant TNF augmented
DNA synthesis by primary hepatocytes cultured in the presence of other mitogenic factors. [33] We also
showed that pretreatment of rats with soluble TNF receptors (which bind to and inactivate TNF)
inhibited hepatocyte DNA after PH. Because soluble TNF receptors are relatively specific for TNF alpha
and have been used clinically to neutralize the biologic actions of TNF, our finding strengthened the
argument that TNF may initiate the hepatic regenerative process. [86] Eventually, the development of
sensitive reverse transcriptase-polymerase chain reactions (RT-PCR) to amplify low abundance mRNAs
and commercially available ELISA's to detect picogram amounts of rodent cytokines permitted our lab
and other groups to demonstrate that increases in hepatic TNF mRNAs and circulating TNF protein
occur within minutes to hours after PH in otherwise healthy rats. [79] More recently, other laboratories
have completed work that also supports the theory that TNF plays an important, positive
growth-regulatory role during normal liver regeneration. Yamada et al reported that post-PH liver
regeneration is inhibited in transgenic mice that had targeted disruption of the TNF receptor-1 (TNFR-1)
gene, [94] and Cressman et al published evidence that liver regeneration is inhibited in transgenic mice
with inactivation of the gene that encodes IL-6, [21] a TNF-inducible cytokine. [23] Thus, there is now
compelling evidence that injury-related cytokines help to promote regeneration of the damaged liver.
There is some evidence that chronic ethanol consumption, at least by rodents, does not obviate the role of
TNF in hepatic regeneration. We found that rats that had been fed nutritionally-balanced,
ethanol-containing diets for several weeks actually were more vulnerable to effects of anti-TNF

antibodies than isocalorically-fed, control rats. Compared with control rats that were pretreated with
nonimmune IgG before PH, post-PH hepatocyte DNA synthesis was 50% lower in control rats that had
been pretreated with anti-TNF antibodies. In contrast, hepatocyte DNA synthesis was more than 90%
inhibited in anti-TNF, pretreated, ethanol-fed rats when compared with a group of ethanol-fed rats that
received nonimmune IgG pre-PH. [3] Recent work indicates that hepatic expression of TNF mRNAs and
circulating levels of TNF protein are generally increased after PH in such ethanol-fed rats when
compared with pair-fed control animals. [94A] Thus, more TNF appears to be necessary for hepatocytes to
achieve a proliferative response in ethanol-fed animals; that is, chronic ethanol consumption may result
in a relative resistance to the proliferative actions of TNF. If this is true, then strategies to eliminate TNF
may be particularly counterproductive once alcoholic liver injury has occurred; however, given strong
evidence that
727
TNF is involved in the genesis of ethanol-mediated liver injury, additional investigation is necessary to
characterize the signals that mediate the proliferative actions of TNF so that conditions can be
manipulated therapeutically to optimize TNF's proliferative, as opposed to its cytotoxic, effects.
Regulation of TNF Biological Activity
Since TNF and TNF-inducible cytokines, such as IL-1 and IL-6, regulate diverse functions in many
target cells, there has been intense interest in defining mechanisms that regulate TNF itself. Over the last
5 years, work in various models indicates that TNF production and biologic activity are regulated by
other cytokines. Some of these, including IL-12 and interferon gamma (IFN gamma), act as
TNF-inducers by promoting the generation of TNF and increasing target cell sensitivity to TNF. [76] [78]
[89] Others, including IL-10 and transforming growth factor beta (TBF beta), generally down-regulate the
expression and activity of TNF-inducers, TNF, and TNF-inducible cytokines. [7] [52] [56] [72] The picture
has been further complicated by the recent discovery of other cytokines (lymphotoxin alpha and
lymphotoxin beta) that are related structurally to TNF. At least one of these (lymphotoxin alpha, also
called TNF beta) binds to types 1 and 2 TNF receptors and appears to activate some (but not all) of the
same signaling mechanisms as TNF. [66] [83] [85]
As stated earlier, the liver is an important producer of TNF, particularly after exposure to gut-derived
endotoxin (i.e., LPS) [23] ; LPS increases in the portal blood after PH [87] and this is thought to play a
positive role in regulating liver regeneration after PH. [13] [19] [20] In sepsis, LPS stimulation of TNF
expression is influenced by the relative abundance of TNF-regulatory cytokines, including IL-12, IFN
gamma, and IL-10, which can be induced by LPS. [4] [5] [42] [71] [73] [88] [97] Increased TNF prompts the
production of several other cytokines, including IL-1 and IL-6. TNF, therefore, is induced always in the
context of other cytokines in vivo. As a result, the ultimate consequences of LPS (and other stimuli that
increase cytokine production) are mediated by a network of cytokines that regulate each other's
expression and activity. This partially explains why the actions of any given cytokine are pleiotropic:
cytokine effects vary with the state of activation of the target cell(s), which, in turn, is influenced by the
presence of other cytokines and the ability of the target cell to produce bioactive autocrine factors. The
importance of cytokine interactions in defining physiology and pathophysiology is well-illustrated by
work in models of endotoxic shock, which demonstrates that specific neutralization of any one of a
number of different cytokines (e.g., TNF, [86] [91] IL-12, [89] IFN gamma, [14] IL-10 [44] ) significantly
influences mortality after treatment with LPS.

To date, relatively little has been learned about the nature and timing of the cytokine response that occurs
during liver regeneration after PH. Our initial experiments with anti-TNF antibodies demonstrated that
IL-6 is induced after PH and that TNF plays an important role in promoting IL-6 induction in this setting.
[2] These findings were confirmed by recent evidence that post-PH increases in hepatic IL-6 mRNAs are
severely attenuated in TNFR-1-null transgenic mice. [94] Furthermore, TNF and IL-6 appear to interact to
regulate hepatocyte proliferation after PH. This concept is supported by observations that pre-treatment
with IL-6 inhibits liver regeneration after PH in normal mice but restores post-PH hepatocyte
proliferation in TNFR-1 null mice (which produce little IL-6 mRNA in the liver after PH) [94] and in
IL-6-null mice, which have TNF receptors and presumably express TNF after PH. [21] We recently noted
that the induction of IL-6 mRNAs was slightly attenuated and delayed in the livers of ethanol-fed rats
after PH [94A] ; however, the physiologic implications of this finding
728
are uncertain as ethanol feeding is associated with an up-regulation of IL-6 expression in other tissues.
Several other TNF-related cytokines also increase, at least transiently, after PH. For example, we have
noted increases in cytokines that promote TNF's activity (e.g., IL-12 and IFN gamma, IL-1 beta).
Sequential analysis of the expression of various cytokine mRNAs in the liver over the initial 24 hours
after PH demonstrates a transient induction of TNF that begins within 5 minutes after PH and persists for
about 1 hour. [80] [81] During this time period, expression of IL-1 beta also increases transiently although
induction of IFN gamma and IL-12 occur somewhat later. [94A] Other cytokines (e.g., IL-10 and TGF
beta) that down-regulate the effects of TNF also are induced after PH. For example, there is a rapid,
transient, up-regulation of IL-10 mRNAs in the liver within the initial 30 to 60 minutes after PH. [81] TGF
beta induction occurs a bit later. [11] [81] We have reported that chronic ethanol feeding generally enhances
the regenerative induction of all of these cytokines after PH. [94A]
Almost nothing is known about the cellular source of these various cytokines during liver regeneration;
however, emerging evidence suggests that different cells may be predominately responsible for
producing a given cytokine at different time points in the regenerative process. This was elegantly
demonstrated by Bissell's work, which showed that Kupffer cells are a major source of TGF beta Tbefore
PH, but that an increased expression of this gene by hepatocytes accounts for the induction of TGF beta
in the regenerating liver remnant. [11] Similarly, in situ hybridization analysis demonstrates that hepatic
and portal venous endothelia and biliary epithelia are the major producers of TNF during the
prereplicative period following PH. [60] The latter finding was quite unanticipated, because tissue
macrophages (e.g., Kupffer cells) are a major source of TNF after LPS exposure. [23] Experiments with
gadolinium chloride, a Kupffer cell depleting drug, suggest that the in situ hybridization results are valid
because hepatic levels of TNF mRNA and circulating levels of TNF protein are increased in the livers of
gadolinium chloride pre-treated rats after PH. [81] Those same studies also revealed that Kupffer cells are
likely to be important producers of IL-10 after PH, because regenerative induction of IL-10 was virtually
abolished in rats that were pre-treated with gadolinium chloride. Taken together, those findings, coupled
with the known anti-TNF actions of IL-10, [4] [7] [44] prompted us to speculate that Kupffer cells normally
act to restrict the TNF response to PH. [79] [80] [81] More work is required to identify which cell populations
produce each of the various cytokines during normal liver regeneration. It also is important to determine
if conditions that lead to impaired liver regeneration and chronic liver damage alter this normal pattern of
cytokine production. For example, several groups have demonstrated increased production of TNF by

Kupffer cells during the evolution of experimentally-induced alcoholic liver damage [1] [47] [51] but it is not
clear if or how this might influence the regenerative response.
Cytokine Production by Extra-Hepatic Tissues
Very recently, we demonstrated that extra-hepatic tissues, particularly white adipose tissue, are important
cytokine producers after PH. In healthy adult rats, we found little evidence of adipose TNF, IL-6, or
IL-10 expression before PH; however, within a few hours after PH, mRNAs for each of these cytokines
were easily demonstrated in white adipose tissue. In healthy rats, the induction of cytokine gene
expression in fat generally lagged behind the induction of these same genes in the liver but followed a
similar pattern as the hepatic response. [94A] Interestingly, all of these cytokines were relatively
overexpressed in the fat of
729
rats that had been fed ethanol chronically. Indeed, ethanol-fed rats exhibited constitutive expression of
TNF mRNA in white adipose tissue, and PH led to a more rapid, intense and sustained induction of TNF
and the other cytokines in the ethanol-fed rats when compared with the pair-fed controls. Thus, in
aggregate, the evidence suggests that chronic consumption of ethanol generally increases the production
of injury-related, proinflammatory cytokines in several tissues, including the liver and fat. Hepatocyte
Mitogens and Comitogens
Only recently has much attention been given to the growth-regulatory roles of proinflammatory
cytokines. It has long been recognized that many other molecules are potent growth regulators.
Molecules that stimulate the proliferation of cultured cells in the absence of serum are termed
"mitogens." Those that enhance the growth-stimulatory actions of mitogens, but alone cannot affect DNA
synthesis, are designated "comitogens." Although there is evidence that each of these factors is likely to
have some role in hepatic regeneration, few, if any, of the hepatocyte mitogens are specific for
hepatocytes and most are produced in many tissues, not only the liver. Thus, it is not clear why liver cells
proliferate relatively selectively in response to liver injury. Hepatocyte comitogens also are somewhat
curious, since most are hormones with principle roles to regulate cellular substrate use and energy
homeostasis. Presumably, timing is the secret to the initiation of a proliferative response in hepatocytes,
which spend most of their adult life "outside" of the cell cycle. That is, the cytokines, mitogens, and
comitogens must be presented to liver cells in the appropriate sequence in order to elicit the sequential
responses that move hepatocytes into and through various stages of the cell cycle. The growth response is
likely to be initiated by local factors that are most abundant within the injured liver, and then amplified
and propagated by factors that are recruited from extra-hepatic sites.
Little is known about the mechanisms that operate within the regenerating liver to coordinate the local
production and release of mitogenic factors. Errors in this critical process are likely to underlie the
genesis of chronic liver injury and, perhaps, even the evolution of cirrhosis. The latter is suggested by
recent evidence that hepatic stellate cells, long acknowledged to be the principle source of liver
extracellular matrix, also are major producers of mitogens during normal liver regeneration. For example,
several groups have noted that after PH, stellate cells are rich sources of hepatocyte growth factor (HGF),
a potent hepatocyte mitogen. [85A] We found that TNF plays a role in the induction of stellate cell HGF
production after PH. After PH, the degree of stellate cell activation (as indicated by loss of vitamin A and
increased expression of a smooth muscle actin) and hepatic HGF expression correlate with levels of

TNF. Furthermore, treatment of cultured primary stellate cells with recombinant TNF elicits a
dose-related induction of IL-6 expression, activation of the IL-6 responsive transcription factor, C/EBP
beta, and increased expression of HGF, a gene that is transcriptionally activated by C/EBP beta. [78A]
Thus, liver nonparenchymal cells also appear to be targets for injury related cytokines and, under at least
some circumstances, TNF stimulates stellate cells to produce factors that are mitogenic for hepatocytes.
Presumably, the kinetics of the latter response are such that stellate-derived mitogens are available
maximally when hepatocytes are optimally sensitive to mitogens. Because liver nonparenchymal cell
proliferation begins well after hepatocyte proliferative activity has peaked post-PH, it also is likely that
hepatocytes produce factors that regulate the proliferation of liver nonparenchymal cells. Whether or
730
not chronic alcohol consumption influences stellate cell production of hepatocyte mitogens or hepatocyte
production of factors that regulate the phenotype of stellate cells is unknown. A better understanding of
the cell-to-cell communications that orchestrate the normal regenerative response to liver injury is
required before we can design treatments to optimize the response in patients with chronic liver injury
caused by alcohol or other agents.
Requirement 2: Proliferation of a Differentiated Population of Cells That Ordinarily Exist

in a Nonproliferative State
Initiation of the Proliferative Response in Hepatocytes (G0 -to-G1 Transition)

TNF and other injury-related cytokines initiate hepatocyte entry into the prereplicative phase of the cell
cycle (i.e., G1 ). Studies in transgenic mice, with deletions of TNF receptor genes and neutralizing
anti-TNF antibody treatment of normal rodents subjected to PH, complement work that has been done in
many types of cultured cells and identify specific signals that mediate this response. It is now well
established that after PH, TNF interacts with both classes of TNF receptors on hepatocytes to activate
discrete, but overlapping, pathways that regulate the activity of various kinases which, in turn, modulate
the activity of transcription factors that regulate the transcription of genes that encode molecules required
for proliferation as well as genes that encode hepatocyte-specific products (e.g., albumin, clotting
factors). For example, after PH, TNF-dependent activation of Jun nuclear kinase is necessary for the
proto-oncogene product, c-Jun, to accumulate in liver cell nuclei and participate in DNA complex
formation with activator protein (AP)-1, a transcription factor that modulates the expression of many
growth-regulatory genes. [33] TNF also plays a major role in promoting the redistribution of other
transcription factors, including NF kB, C/EBP beta, and C/EBP delta, from the cytosol and into the
nuclei of hepatocytes during the initial hours after PH. [34] [94] [95] The latter two transactivators displace
one of their family members, C/EBP alpha. C/EBP alpha is the major transcriptional activator of the
albumin gene, but it also functions as a constitutively expressed, growth-arrest gene in hepatocytes. [26]
Thus, TNF/TNF receptor interactions help to initiate the hepatocyte proliferative response both by
inducing the expression of genes (e.g., c-Jun) that stimulate proliferation and by inhibiting the expression
of other genes (e.g., C/EBP alpha) that normally keep hepatocytes from proliferating. Furthermore, by
promoting the substitution of closely related, growth-permissive transcriptional regulators (e.g., C/EBP
beta and C/EBP delta) for constitutive, growth inhibitory transcription factors (e.g., C/EBP alpha), TNF
helps to assure that hepatocyte-specific, differentiated gene expression is maintained as these cells
prepare to proliferate.

Little is known about the effect of ethanol consumption on the propagation of the signals that regulate
re-entry into the cell cycle. We have reported that acute ingestion of an intoxicating dose of ethanol
actually amplifies the activation of the stress-related protein kinase, Jun nuclear kinase (JNK), and speeds
the nuclear accumulation of c-Jun after PH; however, we found that chronic consumption of ethanol had
the opposite effect on these responses, inhibiting the induction of JNK and c-Jun after PH. [96] Similarly,
we noted that acute ethanol ingestion accelerated the activation of NF kB after PH, [96] and chronic
ingestion of ethanol virtually abolished the activation of this transcription factor after PH. [94A] These
opposing effects of acute and chronic alcohol appear to be relatively
731
specific for stress- or oxidant induced responses because we noted that both acute and chronic exposure
to ethanol inhibit the induction of mitogen activated kinases after PH. [96] The latter is consistent with the
work of other groups, which indicates that ethanol exposure inhibits the ability of the mitogen, epidermal
growth factor (EGF), and the comitogen, insulin, to induce proliferation in primary cultures of
hepatocytes. [15] Curiously, although chronic ethanol consumption inhibited at least two stress-related
responses (i.e., the induction of NF kB and JNK) in regenerating hepatocytes, it failed to block the
activation of Stat-3, [94A] an IL-6-responsive transcription factor that is normally induced after PH. [15]
Furthermore, we have had difficulty identifying ethanol-related inhibition of other IL-6-regulated
responses, such as the induction of C/EBP beta or C/EBP delta mRNAs after PH. [31] Thus, it does not
appear that ethanol inhibition of liver regeneration can be explained by an interruption of IL-6-related
signals, although the latter clearly are required for the normal regenerative response. [21] [94] Progression
from the Prereplicative Period (G1 ) into the Replicative Phases (S, M) of the Cell Cycle
Unless they have already entered the prereplicative phase of the cell cycle (G1 ), hepatocytes respond
poorly to mitogens. This is nicely illustrated by evidence that classical hepatocyte mitogens, including
epidermal growth factor (EGF), transforming growth factor alpha (TGF alpha), and HGF, have only
weak growth stimulatory actions when infused into the livers of healthy adult animals. [67] These same
factors are effective mitogens for primary hepatocytes in culture. [67] This apparent paradox has been
resolved by recent evidence that, like other types of liver injury, the hepatocyte isolation protocol acts as
a cell cycle initiator, inducing stress-activated protein kinases and immediate early genes in hepatocytes.
It remains uncertain precisely how the activation of these kinases and genes sensitize hepatocytes to the
trophic actions of mitogens; however, the intracellular events that follow the interaction of various
mitogens with their receptors have been extensively categorized.
Most growth factor receptors either possess intrinsic tyrosine kinase activity (e.g., the EGF or HGF or
insulin receptors) or they are capable of recruiting cytosolic tyrosine kinases of the Janus kinase family to
the plasma membrane (e.g., the IL-6 receptor). Receptor-associated tyrosine kinases recruit other adaptor
proteins (e.g., Grb-2, SOS) that couple to various signal transduction pathways, including those that
activate phospholipid hydrolysis, promote the accumulation of intracellular calcium, and induce other
kinases and phosphatases. Mitogenic factors that activate seven transmembrane-spanning domain
receptors (e.g., glucagon, adrenergic hormones) can also interact with signaling components of the
receptor tyrosine kinase cascades via their associated GTP-binding proteins.
The series of intracellular events that propagate mitogen-initiated signals at the cell surface into the
nucleus have been a particular focus of investigation because activating mutations of the components of

this signal transduction pathway often lead to malignant transformation. The mitogen-activated kinase
cascade culminates with the activation of mitogen activated protein kinase (also called Extracellular
regulated kinase [Erk]), a serine threonine kinase, whose targets include growth regulatory transcription
factors, such as the proto-oncogene, c-Jun. Phosphorylation of c-Jun and other DNA-binding proteins
alters their DNA binding activities or transactivating potential, thus modulating the level of expression of
all of the genes that are regulated by these factors. Thus, mitogen- and stress-activated kinase cascades
share downstream targets (e.g.,
732
c-Jun) that regulate a wide array of cellular genes and thus, affect global changes in the hepatocyte
phenotype.
There is no doubt that chronic consumption of ethanol prevents mitogen-stimulated hepatocytes from
entering S phase given results published by several groups who have studied the PH model, [27] [37] [43] [77]
[92] as well as data obtained from hepatocyte culture experiments. [15] Although chronic ethanol exposure
thwarts the propagation of some pre-replicative signals, many others are transduced. Ultimately, the
induction of the immediate-early gene response that characterizes successful entry into the cell cycle
occurs more-or-less normally in ethanol-treated rats after PH. [30] Recently, we discovered that chronic
exposure to ethanol prevents the induction of many delayed-early genes, such as cyclin D1 , cyclin D3 ,
cyclin A, p53, p21, that normally occurs as hepatocytes prepare to exit G1 and enter S phase. [94A] These
findings identify the site of cell cycle arrest because of ethanol (i.e., chronic ethanol ingestion arrests
initiated hepatocytes in mid-late G1 ).
Because the hepatocyte is bombarded with many mitogens and comitogens after PH, it is improbable that
ethanol would be able to arrest hepatocytes in prereplicative stages of the cell cycle by inhibiting a single
growth factor or its receptor. Rather, it is likely that ethanol-associated G1 growth arrest reflects a block
in some common growth-regulatory pathways that are downstream from multiple growth factor
receptors. The latter concept is supported by evidence that chronic ethanol consumption inhibits the
induction of ornithine decarboxylase (ODC), the rate limiting enzyme for polyamine biosynthesis. [28]
Activation of ODC is an effect of many mitogens and is necessary for liver regeneration after PH. [61]
Furthermore, treatment with supplemental putrescine, the immediate product of ODC, normalizes hepatic
polyamine levels and improves liver regeneration after PH in ethanol-fed rats. [28] Increases in polyamine
synthesis begin in mid-late G1, around the time when induction of the delayed-early,
cell-cycle-dependent genes begins. Thus, it is conceivable that ethanol affects the propagation of one or
more proximal signals that activate various targets in mid-G1 . Indeed, chronic ethanol exposure is
known to inhibit insulin or insulin-related growth factor-dependent phosphorylation of insulin-receptor
substrate (IRS)-1. [8A] It also interferes with normal trafficking of the EGF receptor [22A] and alters the
activation of heterotrimeric GTP-binding proteins that couple alpha and beta-adrenergic receptors to their
intracellular targets. [29] One consequence of the latter is that chronic ethanol exposure inhibits
regenerative activation of adenylyl cyclase, blocking the accumulation of cAMP in the liver after PH. [29]
Ethanol-related inhibition of insulin/IGF signaling and decreases in cAMP are predicted to have global
effects on hepatocyte intermediary metabolism and energy homeostasis. Indeed, chronic consumption of
ethanol is known to reduce liver ATP stores and inhibit mitochondrial respiration. [22] Given the energy
demands of tissue repair, it is conceivable that, ultimately, ethanol may limit liver regeneration by
creating an intracellular environment that compromises hepatocyte energy homeostasis. If confirmed,

this hypothesis could have important pathophysiologic implications as mitochondrial dysfunction and
energy depletion also are believed to contribute to alcohol-induced liver injury.
Requirement 3: Temporary Suspension of the Normal Mechanisms that Couple Cell

Cycle Re-entry to Cell Death
The size of an adult organ is guarded carefully. Thus, cells proliferate normally only to replace their
senescent neighbors. Because programmed hepatocyte
733
death (i.e., apoptosis) is a relatively rare event in the healthy liver, hepatocyte proliferation also occurs
infrequently. However, insults that increase the rate of hepatocyte death generally result in an increased
rate of hepatocyte proliferation. A breakdown in the mechanisms that couple hepatocyte proliferative
activity to the rate of hepatocyte death results in a net increase in liver mass. This has been induced
experimentally in transgenic mice with dysfunctional Fas receptors (FasR). The latter animals develop
massive hepatomegaly because they have decreased hepatocyte apoptosis, and they clear senescent
hepatocytes inefficiently. Because the low endogenous rate of hepatocyte proliferation is not balanced by
deletion of senescent hepatocytes in FasR-mutant mice, the total number of hepatocytes gradually
increases over time and the liver continues to grow in adulthood.
In healthy adult mammals, liver size remains constant despite hepatocyte turnover, because the rates of
hepatocye proliferation and apoptosis are coupled. The adult liver cannot recover from sudden,
exogenous insults that abruptly decrease the number of hepatocytes (e.g., PH or hepatotoxins) unless
these normal homeostatic mechanisms that couple proliferative and apoptotic activities are suspended.
This temporary uncoupling of proliferation and apoptosis permits the rate of hepatocyte proliferation to
outstrip the rate of apoptosis until the liver recovers its pre-injury mass. The latter implies that successful
liver regeneration requires relative inhibition of hepatocyte apoptosis.
Consistent with this concept, several anti-apoptotic genes (e.g., bcl-2 and bcl-x) and genes that regulate
the production of cytoprotective factors (e.g., inducible nitric oxide synthases (iNOS) are induced in the
mid-to-late prereplicative period following PH. [24] [25] [57] [74] [75] [90] Furthermore, PH triggers increased
hepatocyte apoptosis, rather than proliferation, in transgenic mice with targeted deletion of one of these
gene (i.e., iNOS). [78A] It is interesting that although TNF is required for liver regeneration after PH, it can
also induce hepatocyte apoptosis. The latter only occurs in hepatocytes when RNA or protein synthesis
are inhibited. This observation suggests that when exposed to TNF, hepatocytes protect themselves by
synthesizing one or more antiapoptotic factors. Given evidence that TNF induces the expression of bcl-2,
[58] iNOS, and heat shock protein (Hsp)-70 [9] [48] [54] [55] [68] in cultured hepatocytes, it is conceivable that
TNF may promote the induction of these factors in regenerating hepatocytes after PH.
It is not known if ethanol alters the induction of TNF-regulated antiapoptotic factors after PH. Although
a few groups have suggested that chronic exposure to ethanol via inhalation [45] or intragastric infusion
[93] increases basal apoptosis in rodent livers, and apoptotic bodies have been reported in patients with
alcoholic hepatitis, [53] increased apoptosis was not noted in patients with simple alcohol-induced
steatosis. [53] Indeed, mice with ethanol-induced fatty livers have increased hepatocyte expression of two
antiapoptotic proteins, bcl-2 and bcl-x. After 2 to 3 weeks of consuming ethanol-containing diets, most of
the murine hepatocytes in acinar zone 3 and many of those in acinar zone 2 express these antiapoptotic

factors. [81A] Because cell death and proliferation are fundamentally related to each other, ethanol effects
on hepatocyte apoptosis are likely to influence its effects on hepatocyte proliferation and vice versa.
Efforts to clarify how ethanol influences apoptosis are likely to improve our understanding of how it
inhibits liver regeneration.
CONCLUSION
Chronic consumption of ethanol is one of the few insults that can lead to liver injury without inducing a
compensatory proliferative response in surviving
734
hepatocytes. This observation suggests that ethanol selectively blocks the growth stimulatory actions of
injury-related cytokines and results in a differential sensitivity to their toxic actions. Thus, efforts to
understand how ethanol produces these responses may identify therapeutic targets that can be
manipulated to abort injury and coax regeneration in ethanol-damaged livers.
References
1. Adachi Y, Moore LE, Bradford BU, et al: Antibiotics prevent liver injury in rats following long-term exposure to
ethanol. Gastroenterology 108:218-224, 1995
2. Akerman P, Cote P, Yang SQ, et al: Antibodies to tumor necrosis factor alpha inhibit liver regeneration after partial
hepatectomy. Am J Physiol 263:G579-G585, 1992
3. Akerman PA, Cote PM, Yang SQ, et al: Long-term ethanol consumption alters the hepatic response to the regenerative
effects of tumor necrosis factor alpha. Hepatology 16:1212-1219, 1993
4. Arai T, Hiromatsu K, Kobayashi N, et al: IL-10 is involved in the protective effect of dibutyryl cyclic adenosine
monophosphate on endotoxin-induced inflammatory liver injury. J Immunol 155:5743-5749, 1995
5. Barsig J, Kusters S, Vogt K, et al: Lipopolysaccharide-induced interleukin-10 in mice: Role of endogenous tumor
necrosis factor-alpha. Eur J Immunol 25:2888-2893, 1995
6. Bemelmans MHA, Gouma DJ, Greve JW, et al: Cytokines, tumor necrosis factor and interleukin-6 in experimental
biliary obstruction in mice. Hepatology 15:1132-1136, 1992
7. Berg DJ, Kuhn R, Rajewsky K, et al: Interleukin-10 is a central regulator of the response to LPS in murine models of
endotoxic shock and the Shwartzman reaction but not endotoxin tolerance. J Clin Invest 96:2339-2347, 1995
8. Beyer HS, Stanley M, Theologides A: Tumor necrosis factor-alpha increases hepatic DNA and RNA and hepatocyte
mitosis. Biochemistry International 22:405-410, 1990
8A. Bhavani K, de la Monte S, Brown NV, et al: Effect of ethanol on p36 protein kinase substrate and insulin receptor
substrate 1 expression and tyrosyl phosphorylation in human hepatocellular carcinoma cells. Alcohol Clin Exp Res
19:441-446, 1995

9. BilliarTR, Curran RD, Harbrecht BG, et al: Modulation of nitrogen oxide synthesis in vivo: NG-monomethyl-L-arginine
inhibits endotoxin-induced nitrite/nitrate biosynthesis while promoting hepatic damage. J Leukoc Biol 48:565-569, 1990
10. Bird
GLA, Sheron N, Goka AKJ, et al: Increased plasma tumor necrosis factor in severe alcoholic hepatitis. Ann Intern
Med 112:917-920, 1990
11. Bissell
DM, Wang SS, Jarnagin WR, et al: Cell-specific expression of transforming growth factor-beta in rat liver.
Evidence for autocrine regulation of hepatocyte proliferation. J Clin Invest 96:447-455, 1995
12. Bucher NLR: Liver regeneration: An overview. J Gastroenterol Hepatol 6:615-624, 1991
13. CampbellKA, Wu Y, Chacko VP, et al: P31-MR spectroscopic changes during liver regeneration. J Surg Res
49:244-247, 1990
14. Car
BD, Eng VM, Schnyder B, et al: Interferon gamma receptor deficient mice are resistant to endotoxic shock. J Exp
Med 179:1437-1444, 1994
15. Carter
EA, Wands JR: Ethanol inhibits hormone stimulated hepatocyte DNA synthesis. Biochem Biophys Res Commun
128:767-774, 1985
16. ChapekarMS, Hugget AC, Thorgeirsson SS: Growth modulatory effects of a liver-derived growth inhibitor,
transforming growth factor beta 1 and recombinant tumor necrosis factor alpha, in normal and neoplastic cells. Exp Cell
Res 185:247-257, 1989
17. Coletti
LM, Remick DG, Burch GD, et al: The role of tumor necrosis factor-alpha in the pathophysiologic alterations
after hepatic ischemia/reperfusion in the rat. J Clin Invest 85:1936-1943, 1990
18. Cornell
RP, Leljequist BL, Bartizal KF: Depressed liver regeneration after partial hepatectomy of germ-free, athymic
and lipopolysaccharide-resistant mice. Hepatology 11:915-922, 1990
735
19. Cornell
RP: Gut-derived endotoxin elicits hepatotrophic factor secretion for liver regeneration. Am J Physiol
249:R551-R562, 1985
20. Cornell
RP: Restriction of gut-derived endotoxin impairs DNA synthesis for liver regeneration. Am J Physiol
249:R563-R569, 1985
21. Cressman DE, Greenbaum LE, DeAngelis RA, et al: Liver failure and defective hepatocyte regeneration in
interleukin-6-deficient mice. Science 274:1379-1383, 1996
22. Cunningham CC, Coleman WB, Spach PI: The effects of chronic ethanol consumption on hepatic mitochondrial energy
metabolism. Alcohol Alcohol 25:127-136, 1990
22A. DalkeDD, Sorrell MF, Casey CA, et al: Chronic ethanol administration impairs receptor-mediated endocytosis of
epidermal growth factor by rat hepatocytes. Hepatology 12:1085-1091, 1990
23. Decker K: Biologically active products of stimulated liver macrophages (Kupffer cells). Eur J Biochem 192:245-261,

1990
24. Decker
KF, Obolenskaya MY: Cytokines, nitric oxide synthesis and liver regeneration. J Gastroenterol Hepatol
10:S12-S17, 1995
25. Diaz-GuerraMJM, Velasco M, Martin-Sanz P, et al: Nuclear factor kB is required for the transcriptional control of type
II NO synthase in regenerating liver. Biochem J 326:791-797, 1997
26. Diehl
AM, Johns D, Yang SQ, et al: Adenovirus-mediated transfer of CCAAT/enhancer binding protein alpha identifies
a dominant anti-proliferative role for this isoform in hepatocytes. J Biol Chem 271:7343-7350, 1996
27. DiehlAM, Chacon MA, Wagner P: The effect of chronic ethanol feeding on ornithine decarboxylase activity and liver
regeneration. Hepatology 8:237-242, 1988
28. DiehlAM, Wells M, Brown ND, et al: Effect of ethanol on polyamine synthesis during liver regeneration in rats. J Clin
Invest 85:385-390, 1990
29. Diehl AM: Effects of alcohol on liver regeneration. Alcohol Health Res World 17:279-283, 1993
30. Diehl
AM, Thorgeirsson SS, Steer CJ: Ethanol inhibits liver regeneration in rats without reducing transcripts of key
protooncogenes. Gastroenterology 99:1105-1112, 1990
31. Diehl AM, Yang SQ: Regenerative changes in C/EBP alpha and C/EBP beta expression modulate binding to the C/EBP
site in the c-fos promoter. Hepatology 19:447-456, 1994
32. Diehl AM, Rai R: Regulation of signal transduction during liver regeneration. FASEB J 10:215-227, 1996
33. Diehl AM, Yin M, Fleckenstein J, et al: Tumor necrosis factor alpha induces c-Jun during the regenerative response to
liver injury. Am J Physiol 267:G552-G561, 1994
34. Diehl
AM, Yang SQ, Yin M, et al: Tumor necrosis factor alpha modulates CCAAT/Enhancer binding proteins activities
and promotes hepatocyte-specific gene expression during liver regeneration. Hepatology 22:252-261, 1995
35. D'Souze-Gburek S, Batta KP, Nikias GA, et al: Liver transplantation for jejunoileal bypass-associated cirrhosis:
Allograft histology in the setting of an intact bypassed limb. Liver Transpl Surg 3:23-27, 1997
36. DufourMC, Stinson FS, Fe Cases M: Trends in cirrhosis morbidity and mortality: United States, 1979-1988. Semin
Liver Dis 13:109-125, 1993
37. Dugay L, Couter D, Hetu C, et al: Inhibition of liver regeneration by chronic alcohol administration. Gut 23:8-13, 1982
38. Fang JWS, Bird GLA, Nakamura T, et al: Hepatocyte proliferation as an indicator of outcome in acute alcoholic
hepatitis. Lancet 343:820-823, 1991
39. Feingold
KR, Barker ME, Jones AL, et al: Localization of tumor necrosis factor-stimulated DNA synthesis in the liver.
Hepatology 13:337-339, 1991
40. Feingold
KR, Soued M, Grunfeld C: Tumor necrosis factor stimulates DNA synthesis in the liver of intact rats.
Biochem Biophys Res Commun 153:576-582, 1988
41. FelverME, Mezey E, McGuire M, et al: Plasma tumor necrosis factor alpha predicts decreased long-term survival in
severe alcoholic hepatitis. Alcohol Clin Exp Res 14:255-259, 1990

42. Flesch IEA, Hess JH, Huang S, et al: Early interleukin 12 production by macrophages
736
in response to mycobacterial infection depends on interferon gamma and tumor necrosis factor alpha. J Exp Med
181:1615-1621, 1995
43. Frank
WO, Rayes AN, Washing A, et al: Effect of acute ethanol administration upon hepatic regeneration. J Lab Clin
Med 93:402-413, 1979
44. Gerad C, Bruyens C, Marchant A, et al: Interleukin 10 reduces the release of tumor necrosis factor and prevents
lethality in experimental endotoxemia. J Exp Med 177:547-550, 1993
45. GoldinRD, Hunt NC, Clark J, et al: Apoptotic bodies in a murine model of alcoholic liver disease: Reversibility of
ethanol-induced changes. J Pathol 171:73-76, 1993
46. Goto M, Takei Y, Kawano S, et al: Tumor necrosis factor and endotoxin in the pathogenesis of liver and pulmonary
injuries after orthotopic liver transplantation in the rat. Hepatology 16:487-493, 1992
47. Hansen J, Cherwitz DL, Allen JI: The role of tumor necrosis factor alpha in acute endotoxin-induced hepatotoxicity in
ethanol-fed rats. Hepatology 20:461-474, 1994
48. Harbrecht
B, Stadler J, Demetris A, et al: Nitric oxide and prostaglandins interact to prevent hepatic damage during
murine endotoxemia. Am J Physiol 266:G1004-G1010, 1994
49. Hill
D, Marsano L, Cohen D, et al: Increased plasma interleukin-6 activity in alcoholic hepatitis. J Lab Clin Med
119:507-512, 1992
50. HishinumaI, Nagakawa J, Hirota K, et al: Tumor necrosis factor-alpha in galactosamine-induced hepatitis. Hepatology
12:1187-1191, 1990
51. HonchelR, Ray RB, Marsano L, et al: Tumor necrosis factor in alcohol enhanced endotoxin liver injury. Alcohol Clin
Exp Res 16:665-669, 1992
52. Ianaro
A, Xu D, O'Donnell CA, et al: Expression of TGF-beta in attenuated Salmonella typhimurium: Oral
administration leads to the reduction of inflammation, Il-12 and IFN-gamma, but enhancement of IL-10, in
carrageenin-induced oedema in mice. Immunology 84:8-15, 1995
53. Kawahara H, Matsuda Y, Takase S: Is apoptosis involved in alcoholic hepatitis? Alcohol Alcohol 29:113-118, 1994
54. Kim YM, Lee BS, Yi KY, et al: Upstream Nf-kappa B site is required for the maximal expression of mouse inducible
nitric oxide synthase gene in interferon-gamma plus lipopolysaccharide-induced RAW 264.7 macrophages. Biochem
Biophys Res Commun 236:655-660, 1997
55. Kim YM, Talanian RV, Billiar TR: Nitric oxide inhibits apoptosis by preventing increases in caspase-3-like activity via
two distinct mechanisms. J Biol Chem 272:31138-31148, 1997
56. KnolleP, Schlaak J, Uhrig A, et al: Human Kupffer cells secrete IL-10 in response to lipopolysaccharide (LPS)
challenge. J Hepatol 22:226-229, 1995
57. Kren BT, Trembley JH, Krarewski S, et al: Modulation of appoptosis-associated genes bcl-2, bcl-x, and bax during liver

regeneration. Cell Growth Differ 7:1633-1642, 1996
58. Kroemer G: The proto-oncogene Bcl-2 and its role in regulating apoptosis. Natl Med J 3:614-620, 1997
59. LeevyCM: In vitro studies of hepatic DNA synthesis in percutaneous liver biopsy specimens. J Lab Clin Med
61:761-779, 1963
60. LoffredaS, Rai R, Yang SQ, et al: Bile ducts and portal and central veins are major producers of tumor necrosis factor
alpha in regenerating rat liver. Gastroenterology 112:2089-2098, 1997
61. LukGD: Essential role of polyamine metabolism in hepatic regeneration: Inhibition of protein synthesis and tissue
regeneration by difluoromethylornithine in the rat. Gastroenterology 90:1261-1267, 1986
62. McClain CJ, Hill DB, Schmidt J, et al: Cytokines and alcoholic liver disease. Semin Liv Dis 13:170-182, 1993
63. McClain CJ, Cohen DA: Increased tumor necrosis factor production by monocytes in alcoholic hepatitis. Hepatology
9:349-351, 1989
64. McClainCJ, Cohen DA, Dinarello CA, et al: Serum interleukin-1 (IL-1) activity in alcoholic hepatitis. Life Sci
39:1479-1485, 1986
65. Mealy K, Wilmore DW: Tumor necrosis factor increases hepatic cell mass. Br J Surg 78:331-333, 1991
737
66. MedvedevAE, Espevik T, Ranges G, et al: Distinct roles of the two tumor necrosis factor (TNF) receptors in
modulating TNF and lymphotoxin alpha effects. J Biol Chem 271:9778-9784, 1996
67. Michalopoulos G: Liver regeneration: Molecular mechanisms of growth control. FASEB J 4:176-187, 1990
68. Mohr
S, Zech B, Lapetina EG, et al: Inhibition of caspase-3 by S-nitrosation and oxidation caused by nitric oxide.
Biochem Biophys Res Commun 238:387-391, 1997
69. Nagakawa
J, Hishinuma I, Hirota K, et al: Involvement of tumor necrosis factor alpha in the pathogenesis of activated
macrophage-mediated hepatitis. Gastroenterology 99:758-765, 1990
70. NanjiAA, Khettry U, Sadrzadeh SM: Lactobacillus feeding reduces endotoxemia and severity of experimental
alcoholic liver disease. Proc Soc Exp Biol Med 205:243-247, 1994
71. Narumi S, Wyner LM, Stoler MH: Tissue-specific expression of murine IP-10 mRNA following systemic treatment
with interferon gamma. J Leukoc Biol 52:27-33, 1992
72. Neurath
MF, Fuss I, Kelsall BL, et al: Antibodies to interleukin 12 abrogate established experimental colitis in mice. J
Exp Med 182:1281-1290, 1995
73. Nill
MR, Oberyszn TM, Ross MS, et al: Temporal sequence of pulmonary cytokine gene expression in response to
endotoxin in C3H/HeN endotoxin-sensitive and C3H/HeJ endotoxin-resistant mice. J Leukoc Biol 58:563-574, 1995
74. Obolenskaya M, Schulze-Specking A, Plaumann B, et al: Nitric oxide production by cells isolated from regenerating rat

liver. Biochem Biophys Res Commun 204:1305-1311, 1994
75. Obolenskaya MY, Vanin AF, Mordvintcex PZ, et al: EPR evidence of nitric oxide production by the regenerating rat
liver. Biochem Biophys Res Commun 202:571-576, 1994
76. Orange JS, Salazar-Mather TP, Opal SM, et al: Mechanism of interleukin 12-mediated toxicities during experimental
viral infections: Role of tumor necrosis factor and glucocorticoids. J Exp Med 181:901-914, 1995
77. Orrego H, Crossley IR, Saldiva V, et al: Long-term ethanol administration and short- and long-term liver regeneration
after partial hepatectomy. J Lab Clin Med 97:402-413, 1981
78. Pauli U: Control of tumor necrosis factor gene expression. Crit Rev Eukaryot Gene Exp 4:323-344, 1994
78A. RaiRM, Anania FA, Bagby G, et al: Proinflammatory cytokines induce hepatocyte growth factor expression in the
regenerating rat liver and in cultured rat stellate cells. Am J Physiol (in press).
79. Rai
R, Diehl AM: Review: Regulation of liver generation by proinflammatory cytokines. J Gastroenterol Hepatol
11:466-470, 1996
80. Rai RM, Yang SQ, McClain C, et al: Kupffer cell depletion by gadolinium chloride enhances liver regeneration after
partial hepatectomy in rats. Am J Physiol 270:G909-G918, 1996
81. Rai
RM, Loffreda S, Karp CL, et al: Kupffer cell depletion abolishes induction of interleukin-10 and permits sustained
over-expression of tumor necrosis factor alpha mRNA in the regenerating rat liver. Hepatology 25:889-895, 1997
81A. Rashid A, Lin HZ, Yang SQ, et al: Mitochondrial proteins that regulate apoptosis and necrosis are induced in fatty
liver. Hepatology (in press).
82. Rothschild MA, Berk PD, Lieber CS: Alcoholic liver disease. Semin Liv Dis 13, 1993
83. Ruddle N: Tumor necrosis factor (TNF-alpha) and lymphotoxin (TNF-beta). Curr Opin Immunol 4:327-332, 1992
84. Schenker
S, Perkins HS, Sorrell MF: Should patients with end-stage alcoholic liver disease have a new liver?
Hepatology 11:314-319, 1990
85. Schuchmann M, Hess S, Bufler P, et al: Functional discrepancies between tumor necrosis factor and lymphotoxin alpha
explained by trimer stability and distinct receptor interactions. Eur J Immunol 25:2183-2189, 1995
85A. Tomiya T, Ogata I, Fujiwara K: Transforming growth factor alpha levels in liver and blood correlate better than
hepatocyte growth factor with hepatocyte proliferation during liver regeneration. Am J Pathol 153:955-961, 1998
738
86. Tracey KJ, Cerami A: Tumor necrosis factor, other cytokines and disease. Ann Rev Cell Biol 9:317-343, 1993
87. Trautwein C, Rakemann T, Niehof M, et al: Acute-phase response factor, increased binding, and target gene
transcription during liver regeneration. Gastroenterology 110:1854-1862, 1996
88. Trinchieri G: Interleukin-12: A cytokine produced by antigen-presenting cells with immunoregulatory functions in the

generation of T-helper cells type 1 and cytotoxic lymphocytes. Blood 84:4008-4027, 1994
89. Trinchieri
G: Interleukin-12: A proinflammatory cytokine with immunoregulatory functions that bridge innate
resistance and antigen-specific adaptive immunity. Ann Rev Immunol 13:251-276, 1995
90. Tzung SP, Fausto N, Hockenberry DM: Expression of Bcl-2 family during liver regeneration and identification of Bcl-x
as a delayed early response gene. Am J Pathol 150:1985-1995, 1997
91. Vassalli P: The pathophysiology of tumor necrosis factor. Ann Rev Immunol 10:411-452, 1992
92. Wands JR, Carter EA, Bucher NLR, et al: Inhibition of hepatic regeneration in rats by acute and chronic ethanol
intoxication. Gastroenterology 77:528-531, 1979
93. Yacoub LK, Fogt F, Griniuviene B, et al: Apoptosis and Bcl-2 protein expression in experimental alcoholic liver
disease in the rat. Alcohol Clin Exp Res 19:854-859, 1995
94. Yamada Y, Kirillova I, Peschon JJ, et al: Initiation of liver growth by tumor necrosis factor: Deficient liver regeneration
in mice lacking type I tumor necrosis factor receptor. Proc Natl Acad Sci USA 94:1441-1446, 1997
94A. Yang
SQ, Lin HZ, Yin M, et al: Effects of chronic ethanol consumption on cytokine regulation of liver regeneration.
Am J Physiol 1998 (in press).
95. Yin
M, Yang SQ, Lin HZ, et al: Tumor necrosis factor alpha promotes nuclear localization of cytokine-inducible
CCAAT/enhancer binding protein isoforms in hepatocytes. J Biol Chem 1996; 271:17974-17978
96. Zeldin
G, Rai R, Yang SQ, et al: Effects of alcohol on cytokine-inducible transcription factors. Alcohol Clin Exp Res
20:1639-1645, 1996
97. ZhangR, zur Hansen A, Hoffmann R, et al: Rat liver macrophages express the 55 kDa tumor necrosis factor receptor:
Modulation by interferon-gamma, lipopolysaccharide and tumor necrosis factor alpha. Biochem Chem Hoppe-Sey
375:249-254, 1994


April 21, 2000
Conde Petra
Effects of prostaglandins on ethanol damage in

primary cultured rat hepatocytes.
Citation
Bibliographic Data
Abstract
Indexing Data
Effects of prostaglandins on ethanol
Copyright Notice and Disclaimer damage in primary cultured rat
hepatocytes.
Find More Articles Like This
Order a Full Text Copy of the NLM Citation ID:
Original Journal Article 98199450
Full Source Title:
Korean Journal of Internal Medicine
Publication Type:
Journal Article
Language:
English
Author Affiliation:
Department of Internal Medicine, St. Paul's Hospital, Catholic
University Medical College, Seoul, Korea.
Authors:
Yang JM; Choi SW; Kim SS; Sun HS; Park DH; Han SB; Oh
GT; Kim WM
Abstract:
OBJECTIVES: Several reports demonstrated that ethanol
administration impairs the DNA synthesis in rat hepatocytes.
Also, it has been demonstrated that prostaglandin (PG) helps
prevent membrane damage by hepatotoxic chemicals. In this
study, the authors examined PG's effects on the toxicity of
ethanol in the primary culture of rat regenerations. METHODS:
We examined two kinds of parameters, i.e., DNA synthesis and
lipid peroxidation in the primary culture of rat hepatocytes.
Hepatocytes were isolated by the collagenase perfusion method.
The rate of DNA synthesis was determined by pulse-labelling
cultured cells with [3H]-thymidine. Incorporation of

(3H)-thymidine was determined by liquid scintillation
spectrophotometer. DNA content was measured by the
fluorescence spectrophotometer. The lipid peroxidation was
assayed with spectrophotometer. RESULTS: The results were
as follows: 1) PG family (PGA1, PGD2, PGE1, PGE2, PGG2a,
PGI2 & Thromboxane B2) stimulated the DNA synthesis of
hepatocytes (especially PGD2 and PGE1), 2) ethanol decreased
DNA synthesis by clear dose-dependent manner, 3) the
combined treatment of PGD2 or PGE1, prevents the decreasing
of DNA synthesis, which was induced by ethanol, 4) in ethanol
treatment, lipid peroxidation was decreased significantly, but
PGD2, PGE1 and PGA1 were not affected, and 5) PGD2, PGE1
and PGA1 decreased lipid peroxidation with ethanol,
significantly. CONCLUSIONS: From these results, we
concluded that PG could be useful for the treatment of
degenerative liver disease and alcohol-induced liver disease in
the assumption that further studies on the action mechanisms of
PG will continue.
Major Subjects:
Ethanol / * Toxicity
Liver / * Drug Effects / Metabolism
Prostaglandins, Synthetic / * Pharmacology
Additional Subjects:
Animal
Cells, Cultured
Drug Interactions
DNA / Biosynthesis
Lipid Peroxidation / Drug Effects
Rats
Support, Non-U.S. Gov't
Chemical Compound Name:

(Prostaglandins, Synthetic); 64-17-5 (Ethanol); 9007-49-2
(DNA)
Bookmark URL: /das/journal/view/N/10152848?source=MI

Effects of prostaglandins on ethanol damage in primary cultured rat

hepatocytes.
NLM Citation ID:
98199450
Full Source Title:
Korean Journal of Internal Medicine
Publication Type:
Journal Article
Language:
English
Author Affiliation:
Department of Internal Medicine, St. Paul's Hospital, Catholic University Medical College, Seoul, Korea.
Authors:
Yang JM; Choi SW; Kim SS; Sun HS; Park DH; Han SB; Oh GT; Kim WM
Abstract:
OBJECTIVES: Several reports demonstrated that ethanol administration impairs the DNA synthesis in
rat hepatocytes. Also, it has been demonstrated that prostaglandin (PG) helps prevent membrane damage
by hepatotoxic chemicals. In this study, the authors examined PG's effects on the toxicity of ethanol in
the primary culture of rat regenerations. METHODS: We examined two kinds of parameters, i.e., DNA
synthesis and lipid peroxidation in the primary culture of rat hepatocytes. Hepatocytes were isolated by
the collagenase perfusion method. The rate of DNA synthesis was determined by pulse-labelling cultured
cells with [3H]-thymidine. Incorporation of (3H)-thymidine was determined by liquid scintillation
spectrophotometer. DNA content was measured by the fluorescence spectrophotometer. The lipid
peroxidation was assayed with spectrophotometer. RESULTS: The results were as follows: 1) PG family
(PGA1, PGD2, PGE1, PGE2, PGG2a, PGI2 & Thromboxane B2) stimulated the DNA synthesis of
hepatocytes (especially PGD2 and PGE1), 2) ethanol decreased DNA synthesis by clear dose-dependent
manner, 3) the combined treatment of PGD2 or PGE1, prevents the decreasing of DNA synthesis, which
was induced by ethanol, 4) in ethanol treatment, lipid peroxidation was decreased significantly, but
PGD2, PGE1 and PGA1 were not affected, and 5) PGD2, PGE1 and PGA1 decreased lipid peroxidation
with ethanol, significantly. CONCLUSIONS: From these results, we concluded that PG could be useful
for the treatment of degenerative liver disease and alcohol-induced liver disease in the assumption that
further studies on the action mechanisms of PG will continue.
Major Subjects:
Ethanol / * Toxicity
Liver / * Drug Effects / Metabolism
Prostaglandins, Synthetic / * Pharmacology
http://home.mdconsult.com/das/citation/body/jorg=j...rnal&source=MI&sp=10152848&sid=0/N/10152848/1.html (1 of 2) [21/04/2000 23:24:49]

Animal
Cells, Cultured
Drug Interactions
DNA / Biosynthesis
Lipid Peroxidation / Drug Effects
Rats

(Prostaglandins, Synthetic); 64-17-5 (Ethanol); 9007-49-2 (DNA)

April 21, 2000
Conde Petra
Ethanol-induced alterations of the microtubule

cytoskeleton in hepatocytes.
Citation
Bibliographic Data
Abstract
Indexing Data
Ethanol-induced alterations of the
Copyright Notice and Disclaimer microtubule cytoskeleton in hepatocytes.
Find More Articles Like This From NIH/NLM MEDLINE
NLM Citation ID:
Order a Full Text Copy of the 98236773
Original Journal Article
Full Source Title:
American Journal of Physiology
Publication Type:
Journal Article
Language:
English
Author Affiliation:
Center for Basic Research in Digestive Diseases, Mayo Clinic,
Rochester, Minnesota 55905, USA.
Authors:
Yoon Y; Torok N; Krueger E; Oswald B; McNiven MA
Abstract:
Ethanol has been predicted to alter vesicle-based protein traffic
in hepatocytes, in part, via a disruption of the microtubule (MT)
cytoskeleton. However, information on the effects of chronic
ethanol exposure on MT function in vivo is sparse. Therefore
the goal of this study was to test for ethanol-induced changes in
rat liver tubulin expression, assembly, and cellular organization,
using molecular, biochemical and morphological methods. The
results of this study showed that tubulin mRNA and protein
levels were not altered by ethanol. Tubulin, isolated from
control and ethanol-fed rats, showed similar polymerization
characteristics as assessed by calculation of the critical
concentration for assembly and morphological structure. In

contrast, the total amount of assembly-competent tubulin was

reduced in livers from ethanol-fed rats compared with control
rats when assessed by quantitative immunoblot analysis using a
tubulin antibody. In addition, we observed that MT regrowth
and organization in cultured hepatocytes treated with cold and
nocodazole was markedly impaired by chronic ethanol
exposure. In summary, these results indicate that tubulin levels
in liver are not reduced by ethanol exposure. While there is a
substantial amount of tubulin protein capable of assembling into
functional MTs in ethanol-damaged livers, a marked portion of
this tubulin is polymerization incompetent. This may explain
why these hepatocytes exhibit a reduced number of MTs with
an altered organization.
Major Subjects:
Cytoskeleton / * Drug Effects
Ethanol / * Pharmacology
Liver / Cytology / * Drug Effects / Metabolism
Microtubules / * Drug Effects
Animal
Male
Polymers / Metabolism
Rats
Rats, Sprague-Dawley
RNA, Messenger / Metabolism
Support, U.S. Gov't, P.H.S.
Time Factors
Tubulin / Genetics / Metabolism

(Polymers); (RNA, Messenger); (Tubulin); 64-17-5 (Ethanol)
Grant ID:
AA-09227 AA NIAAA

Ethanol-induced alterations of the microtubule cytoskeleton in

hepatocytes.
NLM Citation ID:
98236773
Full Source Title:
American Journal of Physiology
Publication Type:
Journal Article
Language:
English
Author Affiliation:
Center for Basic Research in Digestive Diseases, Mayo Clinic, Rochester, Minnesota 55905, USA.
Authors:
Yoon Y; Torok N; Krueger E; Oswald B; McNiven MA
Abstract:
Ethanol has been predicted to alter vesicle-based protein traffic in hepatocytes, in part, via a disruption of
the microtubule (MT) cytoskeleton. However, information on the effects of chronic ethanol exposure on
MT function in vivo is sparse. Therefore the goal of this study was to test for ethanol-induced changes in
rat liver tubulin expression, assembly, and cellular organization, using molecular, biochemical and
morphological methods. The results of this study showed that tubulin mRNA and protein levels were not
altered by ethanol. Tubulin, isolated from control and ethanol-fed rats, showed similar polymerization
characteristics as assessed by calculation of the critical concentration for assembly and morphological
structure. In contrast, the total amount of assembly-competent tubulin was reduced in livers from
ethanol-fed rats compared with control rats when assessed by quantitative immunoblot analysis using a
tubulin antibody. In addition, we observed that MT regrowth and organization in cultured hepatocytes
treated with cold and nocodazole was markedly impaired by chronic ethanol exposure. In summary, these
results indicate that tubulin levels in liver are not reduced by ethanol exposure. While there is a
substantial amount of tubulin protein capable of assembling into functional MTs in ethanol-damaged
livers, a marked portion of this tubulin is polymerization incompetent. This may explain why these
hepatocytes exhibit a reduced number of MTs with an altered organization.
Major Subjects:
Cytoskeleton / * Drug Effects
Liver / Cytology / * Drug Effects / Metabolism
Microtubules / * Drug Effects

Animal
Male
Polymers / Metabolism
Rats
RNA, Messenger / Metabolism
Time Factors
Tubulin / Genetics / Metabolism
(Polymers); (RNA, Messenger); (Tubulin); 64-17-5 (Ethanol)
Grant ID:
AA-09227 AA NIAAA

April 21, 2000
Conde Petra
Ethanol-induced retention of nascent proteins in

rat hepatocytes is accompanied by altered
distribution of the small GTP-binding protein
Citation rab2.
Bibliographic Data
Abstract
Indexing Data
Ethanol-induced retention of nascent
Copyright Notice and Disclaimer proteins in rat hepatocytes is
accompanied by altered distribution of
Find More Articles Like This the small GTP-binding protein rab2.
Order a Full Text Copy of the From NIH/NLM MEDLINE
Original Journal Article NLM Citation ID:
97060300
Full Source Title:
Journal of Clinical Investigation
Publication Type:
Journal Article
Language:
English
Author Affiliation:
Center for Basic Research in Digestive Diseases, Mayo Clinic
and Foundation, Rochester, Minnesota 55905, USA.
Authors:
Larkin JM; Oswald B; McNiven MA
Abstract:
Chronic ethanol consumption induces hepatocellular retention
of nascent proteins leading to hepatomegaly. While the
molecular mechanisms behind this impairment are undefined, it
has been predicted that protein retention results from a
disruption of vesicle-mediated secretory processes. Small
GTP-binding proteins (rab proteins) have recently been
implicated in the regulation of vesicular trafficking in
eukaryotic cells. Our objectives were to identify intracellular
sites of ethanol-induced protein retention and to determine
whether the distribution of secretory rab proteins was altered by
ethanol. Transport of hepatic proteins along the secretory

pathway in livers from control and ethanol-fed rats was
analyzed using subcellular fractionation and
immunoprecipitation in the context of in vivo pulse-chase
experiments. We show that pre-Golgi and Golgi compartments,
as well as secretory vesicles, are sites of ethanol-induced
retention of nascent soluble and transmembrane secretory
proteins. These results are supported by immunofluorescence
localization of hepatic proteins on liver sections. Further,
immunoblot analyses of hepatic subcellular fractions from
ethanol-damaged livers indicate a dramatic reduction in the
association of rab2 with a Golgi compartment as compared with
controls. In contrast, rab6 and alpha-mannosidase II, Golgi
marker proteins, appear unchanged. These studies provide a
detailed analysis of the intracellular site of ethanol-induced
protein retention in the hepatocyte and lend novel insight into a
potential mechanism behind this impairment. The effects of
ethanol exposure on rab proteins and Golgi function are
discussed.
Major Subjects:
GTP-Binding Proteins / * Metabolism
Liver / Drug Effects / * Metabolism
Proteins / * Metabolism
ras Proteins / Metabolism
Albumins / Metabolism
Animal
Biological Transport / Drug Effects
Carrier Proteins / Metabolism
Cell Compartmentation / Drug Effects
Cytoplasmic Granules / Metabolism
Fluorescent Antibody Technique, Indirect
Golgi Apparatus / Metabolism
Male
Nuclear Membrane / Metabolism
Protein Processing, Post-Translational / Drug Effects
Rats
Receptors, Polymeric Immunoglobulin / Metabolism

RNA, Messenger / Genetics

EC 3.6.1.- (rab2 GTP-Binding Protein); EC 3.6.1.- (ras
Proteins); EC 3.6.1.- (GTP-Binding Proteins); (Albumins);
(Carrier Proteins); (Proteins); (Rab6 protein); (Receptors,
Polymeric Immunoglobulin); (RNA, Messenger); 64-17-5
(Ethanol)
Grant ID:
AA-09227 AA NIAAA

Ethanol-induced retention of nascent proteins in rat hepatocytes is ac...panied by altered distribution of the small GTP-binding protein rab2.
Ethanol-induced retention of nascent proteins in rat hepatocytes is

accompanied by altered distribution of the small GTP-binding
protein rab2.
NLM Citation ID:
97060300
Full Source Title:
Journal of Clinical Investigation
Publication Type:
Journal Article
Language:
English
Author Affiliation:
Center for Basic Research in Digestive Diseases, Mayo Clinic and Foundation, Rochester, Minnesota
55905, USA.
Authors:
Larkin JM; Oswald B; McNiven MA
Abstract:
Chronic ethanol consumption induces hepatocellular retention of nascent proteins leading to
hepatomegaly. While the molecular mechanisms behind this impairment are undefined, it has been
predicted that protein retention results from a disruption of vesicle-mediated secretory processes. Small
GTP-binding proteins (rab proteins) have recently been implicated in the regulation of vesicular
trafficking in eukaryotic cells. Our objectives were to identify intracellular sites of ethanol-induced
protein retention and to determine whether the distribution of secretory rab proteins was altered by
ethanol. Transport of hepatic proteins along the secretory pathway in livers from control and ethanol-fed
rats was analyzed using subcellular fractionation and immunoprecipitation in the context of in vivo
pulse-chase experiments. We show that pre-Golgi and Golgi compartments, as well as secretory vesicles,
are sites of ethanol-induced retention of nascent soluble and transmembrane secretory proteins. These
results are supported by immunofluorescence localization of hepatic proteins on liver sections. Further,
immunoblot analyses of hepatic subcellular fractions from ethanol-damaged livers indicate a dramatic
reduction in the association of rab2 with a Golgi compartment as compared with controls. In contrast,
rab6 and alpha-mannosidase II, Golgi marker proteins, appear unchanged. These studies provide a
detailed analysis of the intracellular site of ethanol-induced protein retention in the hepatocyte and lend
novel insight into a potential mechanism behind this impairment. The effects of ethanol exposure on rab
proteins and Golgi function are discussed.
Major Subjects:
http://home.mdconsult.com/das/citation/body/jorg=journal&source=MI&sp=1044655&sid=0/N/1044655/1.html (1 of 2) [21/04/2000 23:26:57]

Ethanol-induced retention of nascent proteins in rat hepatocytes is ac...panied by altered distribution of the small GTP-binding protein rab2.
GTP-Binding Proteins / * Metabolism

Liver / Drug Effects / * Metabolism
Proteins / * Metabolism
ras Proteins / Metabolism
Albumins / Metabolism
Animal
Biological Transport / Drug Effects
Carrier Proteins / Metabolism
Cell Compartmentation / Drug Effects
Cytoplasmic Granules / Metabolism
Fluorescent Antibody Technique, Indirect
Golgi Apparatus / Metabolism
Male
Nuclear Membrane / Metabolism
Protein Processing, Post-Translational / Drug Effects
Rats
Receptors, Polymeric Immunoglobulin / Metabolism
RNA, Messenger / Genetics

EC 3.6.1.- (rab2 GTP-Binding Protein); EC 3.6.1.- (ras Proteins); EC 3.6.1.- (GTP-Binding Proteins);
(Albumins); (Carrier Proteins); (Proteins); (Rab6 protein); (Receptors, Polymeric Immunoglobulin);
(RNA, Messenger); 64-17-5 (Ethanol)
Grant ID:
AA-09227 AA NIAAA
http://home.mdconsult.com/das/citation/body/jorg=journal&source=MI&sp=1044655&sid=0/N/1044655/1.html (2 of 2) [21/04/2000 23:26:57]

MD Consult - Practice Guidelines
April 21, 2000
Conde Petra
Additional Article
MEDLINE, but was found in MD U.S. Preventive Services Task Force, Guidelines from Guide to
Consult's full-text literature database. Clinical Preventive Services, (Second Edition) 1996
1996
Copyright 1996 Williams & Wilkins (Reprinted with permission)
Full Text
Frontmatter Section I. SCREENING
-
Part I. Mental Disorders and
RECOMMENDATION
Substance Abuse
Burden of Suffering
Alcohol Use during

Pregnancy. Screening for Problem Drinking
Alcohol Use by Adolescents
and Young Adults.
Accuracy of Screening Tests This guideline is a portion of a larger document that cannot be
downloaded efficiently as a single HTML file. See other
Effectiveness of Early guidelines from the U.S. Preventive Services Task Force to read
Detection the complete document.
Nondependent Drinkers.
Pregnancy. RECOMMENDATION
Adolescents.
Screening to detect problem drinking is recommended for
Alcohol-Dependent Patients. all adult and adolescent patients. Screening should involve a
careful history of alcohol use and/or the use of standardized
Recommendations of Other
screening questionnaires (see Clinical Intervention ). Routine
Groups
measurement of biochemical markers is not recommended
Discussion in asymptomatic persons. Pregnant women should be
advised to limit or cease drinking during pregnancy.
CLINICAL INTERVENTION
Although there is insufficient evidence to prove or disprove
REFERENCES harms from light drinking in pregnancy, recommendations
that women abstain from alcohol during pregnancy may be
made on other grounds (see Clinical Intervention ). All
http://home.mdconsult.com/das/guideline/view/N/10...&stty=cow&nid=39982&tlt=Z&pos=&tag=relinfo&anc=gl (1 of 25) [21/04/2000 23:28:00]

persons who use alcohol should be counseled about the

dangers of operating a motor vehicle or performing other
potentially dangerous activities after drinking alcohol.
Burden of Suffering
Over half a million Americans are under treatment for
alcoholism, but there is growing recognition that alcoholism
(i.e., alcohol dependence) represents only one end of the
spectrum of "problem drinking."[1] Many problem drinkers have
medical or social problems attributable to alcohol (i.e., alcohol
abuse or "harmful drinking") without typical signs of
dependence,[2] ,[3] and other asymptomatic drinkers are at risk
for future problems due to chronic heavy alcohol consumption
or frequent binges (i.e., "hazardous drinking"). Heavy drinking
(more than 5 drinks per day, 5 times per week) is reported by
10% of adult men and 2% of women.[4] In large community
surveys using detailed interviews,[5] [6] [7] [8] the prevalence of
alcohol abuse and dependence in the previous year among men
was 17-24% among 18-29-year-olds, 11-14% among
30-44-year-olds, 6-8% among 45-64-year-olds, and 1-3% for
men over 65; among women in the corresponding age groups,
prevalence of abuse or dependence was 4-10%, 2-4%, 1-2%,
and less than 1%, respectively. Problem drinking is even more
common among patients seen in the primary care setting
(8-20%).[9]
Medical problems due to alcohol dependence include alcohol
withdrawal syndrome, psychosis, hepatitis, cirrhosis,
pancreatitis, thiamine deficiency, neuropathy, dementia, and
cardiomyopathy.[10] Nondependent heavy drinkers, however,
account for the majority of alcohol-related morbidity and
mortality in the general population.[1] There is a dose-response
relationship between daily alcohol consumption and elevations
in blood pressure and risk of cirrhosis, hemorrhagic stroke, and
cancers of the oropharynx, larynx, esophagus, and liver.[11] [12]
[13] A number of studies have reported a modest increase in
breast cancer among women drinking 2 drinks per day or more,

but a causal connection has not yet been proven.[14] Three large
cohort studies, involving over 500,000 men and women,
observed increasing all-cause mortality beginning at 4 drinks
per day in men[11] ,[12] and above 2 drinks per day in women.[15]
Women achieve higher blood alcohol levels than do men, due to
smaller size and slower metabolism.[11] ,[15] Compared to
nondrinkers and light drinkers, overall mortality was 30-38%
higher among men, and more than doubled among women, who
drank 6 or more drinks per day.[11] ,[12] Of the more than
100,000 deaths attributed to alcohol annually, nearly half are
due to unintentional and intentional injuries,[16] including 44%

of all traffic fatalities in 1993[17] and a substantial proportion of
deaths from fires, drownings, homicides, and suicides (see
Chapters 50, 51, 57, 58, and 59).
The social consequences of problem drinking are often as
damaging as the direct medical consequences. Nearly 20% of
drinkers report problems with friends, family, work, or police
due to drinking.[10] Persons who abuse alcohol have a higher
risk of divorce, depression, suicide, domestic violence,
unemployment, and poverty (see Chapters 49, 50, and 51).[10]
Intoxication may lead to unsafe sexual behavior that increases
the risk of sexually transmitted diseases, including human
immunodeficiency virus (HIV). Finally, an estimated 27 million
American children are at risk for abnormal psychosocial
development due to the abuse of alcohol by their parents.[25]
Moderate alcohol consumption has favorable effects on the risk
of coronary heart disease (CHD).[18] [19] [20] [21] [22] [23] CHD
incidence and mortality rates are 20-40% lower in men and
women who drink 1-2 drinks/day than in nondrinkers.[15] ,[21]
,[22] A meta-analysis of epidemiologic studies suggests little
additional benefit of drinking more than 0.5 drinks per day.[20]
The exact mechanism for the protective effect of alcohol is not
known but may involve increases in high-density lipoprotein[23]
and/or fibrinolytic mediators.[24]
Alcohol Use during Pregnancy.
The proportion of pregnant women who report drinking has

declined steadily in the U.S.[26] Recent surveys indicate 12-14%
of pregnant women continue to consume some alcohol,[27] ,[28]
with most reporting only occasional, light drinking (median: 4
drinks per month).[26] Binge drinking or daily risk drinking
(usually defined as 2 drinks per day or greater) is reported by
1-2% of pregnant women,[27] [28] [29] but higher rates (4-6%)
have been reported in some screening studies.[30] ,[31] Excessive
use of alcohol during pregnancy can produce the fetal alcohol
syndrome (FAS), a constellation of growth retardation, facial
deformities, and central nervous system dysfunction
(microcephaly, mental retardation, or behavioral
abnormalities).[32] Other infants display growth retardation or
neurologic involvement in the absence of full FAS (i.e., fetal
alcohol effects [FAE]).[10] FAS has been estimated to affect
approximately 1 in 3,000 births in the U.S. (1,200 children
annually), making it a leading treatable cause of birth defects
and mental retardation.[33] ,[34]

The level of alcohol consumption that poses a risk during

pregnancy remains controversial.[10] ,[35] FAS has only been
described in infants born to alcoholic mothers, but the variable
incidence of FAS among alcoholic women (from 3-40%)[33]
suggests that other factors (e.g., genetic, nutritional, metabolic,
or temporal) may influence the expression of FAS.[10] The
reported incidence of FAS is higher in Native Americans and
blacks than in whites.[33] ,[36] Most studies report an increased
risk of FAE among mothers who consume 14 drinks per week
or more,[35] ,[37] [38] [39] but the effects of lower levels of drinking
have been inconsistent.[35] ,[40] ,[41] Modest developmental
effects have been attributed to light drinking (7 drinks per
week) in some studies, but underreporting by heavy drinkers
and confounding effects of other important factors (nutrition,
environment, etc.) make it difficult to prove or disprove a direct
effect of light drinking.[10] ,[35] ,[42] Timing of exposure and
pattern of drinking may be important, with greater effects
proposed for exposure early in pregnancy and for frequent
binge drinking.[10]
Alcohol Use by Adolescents and Young Adults.
Use of alcohol by adolescents and young adults has declined

over the past decade, but remains a serious problem.[43] Among
12-17-year-olds surveyed in 1993, 18% had used alcohol in the
last month, and 35% in the last year.[4] In a separate 1993
survey, 45% and 33%, respectively, of male and female 12th
graders reported "binge" drinking (5 or more drinks on one
occasion) within the previous month.[44] The leading causes of
death in adolescents and young adults -- motor vehicle and
other unintentional injuries, homicides, and suicides -- are each
associated with alcohol or other drug intoxication in about half
of the cases. Driving under the influence of alcohol is more than
twice as common in adolescents than in adults.[45] Binge
drinking is especially prevalent among college students: half of
all men and roughly one third of all women report heavy
drinking within the previous 2 weeks.[43] ,[46] Most frequent
binge drinkers report numerous alcohol-related problems,
including problems with school work, unplanned or unsafe sex,
and trouble with police.[46]
Accuracy of Screening Tests
Accurately assessing patients for drinking problems during the
routine clinical encounter is difficult. The diagnostic standard
for alcohol dependence or abuse (Diagnostic and Statistical
Manual of Mental Disorders [DSM] IV)[2] requires a detailed
interview and is not feasible for routine screening. Physical
findings (hepatomegaly, skin changes, etc.) are only late

manifestations of prolonged, heavy alcohol abuse.[47] Asking the
patient about the quantity and frequency of alcohol use is an
essential component of assessing drinking problems, but it is
not sufficiently sensitive or specific by itself for screening. In
one study, drinking 12 or more drinks a week was specific
(92%) but insensitive (50%) for patients meeting DSM criteria
for an active drinking disorder.[48] The reliability of patient
report is highly variable and dependent on the patient, the
clinician, and individual circumstances. Heavy drinkers may
underestimate the amount they drink because of denial,
forgetfulness, or fear of the consequences of being diagnosed
with a drinking problem.
A variety of screening questionnaires have been developed
which focus on consequences of drinking and perceptions of
drinking behavior. The 25-question Michigan Alcoholism
Screening Test (MAST)[49] is relatively sensitive and specific
for DSM-diagnosed alcohol abuse or dependence (84-100% and
87-95%, respectively)[49] ,[50] but it is too lengthy for routine
screening. Abbreviated 10- and 13-item versions are easier to
use but are less sensitive and specific in primary care
populations (66-78% and 80%, respectively).[51] ,[52] The
four-question CAGE instrument [ is the most popular screening
test for use in primary care[53] and has good sensitivity and
specificity for alcohol abuse or dependence (74-89% and
79-95%, respectively) in both inpatients[54] ,[55] and
outpatients.[56] [57] [58] The CAGE is less sensitive for early
problem drinking or heavy drinking, however (49-73%).[58] ,[59]
Both the CAGE and MAST questionnaires share important
limitations as screening instruments in the primary care setting:
an emphasis on symptoms of dependence rather than early
drinking problems, lack of information on level and pattern of
alcohol use, and failure to distinguish current from lifetime
problems.[52]
Some of these weaknesses are addressed by the Alcohol Use
Disorders Identification Test (AUDIT), a 10-item screening
instrument developed by the World Health Organization
(WHO) in conjunction with an international intervention trial.
The AUDIT incorporates questions about drinking quantity,
frequency, and binge behavior along with questions about
consequences of drinking.[60] For the study population in which
it was derived, a score of 8 of 40 on the AUDIT had high
sensitivity and specificity for "harmful and hazardous drinking"
(92% and 94%, respectively) as assessed by more extensive
interview.[60] Validation studies have reported more variable

performance of the AUDIT. Sensitivity and specificity for

current abuse/dependence were high (96% and 96%,
respectively) in an inner-city clinic;[61] among rural outpatients,
AUDIT was less sensitive and specific (61% and 90%) for
current drinking problems but superior to the Short
MAST-13.[51] Because it focuses on drinking in the previous
year, however, AUDIT is less sensitive for past drinking
problems.[62] Further validation studies in other populations are
under way.
Brief screening tests may be less sensitive or less specific in
young persons: sensitivity of the CAGE for problems due to
alcohol among college freshmen was 42% in men and 25% in
women.[63] Only 38% of college students with an AUDIT score
of 8 or greater met DSM criteria for abuse or dependence;[64]
many of these "false-positive" results were due to drinking
patterns (frequent binge drinking) that would be considered
hazardous. Alternative screens have been developed for
adolescents, such as the Perceived-Benefit-of-Drinking scale[65]
and the Problem Oriented Screening Instrument for Teenagers
(POSIT),[66] but they have not yet been adequately validated in
the primary care setting.
Instruments that focus on alcohol dependency (e.g., CAGE or
MAST) are not sensitive for levels of drinking considered
dangerous in pregnancy.[67] Women may underreport alcohol
consumption while pregnant,[68] and direct questions about
drinking may provoke denial.[69] Brief instruments that
incorporate questions about tolerance to alcohol ("How many
drinks does it take to make you feel high?" or "How many
drinks can you hold?") were more sensitive than the CAGE
(69-79% vs. 49%) for risk-drinking in pregnancy (2 drinks per
day or greater).[30] ,[70] Women who require 3 or more drinks to
feel high, or who can drink more than 5 drinks at a time, are
likely to be at risk.[71]
Laboratory tests are generally insensitive and nonspecific for
problem drinking. Elevations in hepatocellular enzymes, such
as aspartate aminotransferase (AST), or the erythrocyte mean
corpuscular volume (MCV) are found in less than 10% and 30%
of problem drinkers, respectively.[72] Serum gamma-glutamyl
transferase (GGT) is more sensitive (33-60%) in various
studies,[54] ,[55] ,[72] but elevations in GGT may be due to other
causes (medications, trauma, diabetes, and heart, kidney, or
biliary tract disease). Even when the prevalence of problem
drinking is high (30%), the predictive value of an elevated GGT
has been estimated at only 56%.[72]

Effectiveness of Early Detection

Numerous studies demonstrate that clinicians are frequently
unaware of problem drinking by their patients.[10] Early
detection and intervention may alleviate ongoing medical and
social problems due to drinking and reduce the future risks from
excessive alcohol use.
A number of randomized trials have now demonstrated the

efficacy of brief outpatient counseling (5-15 minutes) for
nondependent problem drinkers. In four Scandinavian studies,
which enrolled patients with elevated GGT and heavy alcohol
consumption, brief counseling to reduce drinking and regular
follow-up produced significant improvements (decreased GGT
and/or decreased alcohol consumption) in treated versus control
subjects;[73] [74] [75] [76] counseling reduced reported sick days in
one study.[74] In the longest of these studies, patients receiving
counseling had fewer hospitalizations and 50% lower mortality
after 5 years.[73] Some of this benefit, however, may have been
due to the close medical follow-up (every 1-3 months) in the
intervention group rather than the initial counseling.
Additional trials have demonstrated that brief interventions can
reduce alcohol consumption in problem drinkers identified by
screening questionnaires or self-reported heavy drinking.[77] [78]
[79] Most recently, an international WHO study examined the
effects of 5 or 20 minutes of counseling about drinking in 1,500

"at-risk" male and female drinkers: >35 drinks per week for
men; >21 drinks per week for women; or intoxicated twice per
month; or self-perceived drinking problem.[80] After 9 months,
self-reported alcohol consumption among men was reduced
32-38% in the intervention groups and 10% in controls. Among
women, alcohol consumption declined significantly (>30%)
among both treated and control groups. A meta-analysis of six
brief-intervention trials estimated that interventions reduced
average alcohol consumption by 24%.[81] Although self-reported
consumption may be subject to bias, reported changes in
drinking correlated with objective measures (GGT, blood
pressure) in most studies. Two additional studies demonstrated
significant reductions in blood pressure as a result of advice to
stop drinking or substitution of nonalcoholic beer.[82] ,[83]
Pregnancy.
There are no definitive controlled trials of treatments for

excessive drinking in pregnancy.[84] In several uncontrolled
studies, a majority of heavy-drinking pregnant women who

received counseling reduced alcohol consumption,[32] ,[85] ,[86]
and reductions in drinking were associated with lower rates of
FAS.[32] ,[86] Many women spontaneously reduce their drinking
while pregnant, however, and women who continue to drink
differ in many respects from women who cut down (e.g.,
heavier drinking, poorer prenatal care and nutrition). As a
result, it is difficult to determine precisely the benefit of
screening and counseling during pregnancy. In two trials that
employed a control group, the proportions of women abstaining
or reducing consumption were similar in intervention and
control groups.[87] ,[88]
Adolescents.
A 1990 Institute of Medicine (IOM) report concluded that

specific recommendations for the treatment of alcohol problems
in young persons were impossible, due to disagreement over
what constitutes a drinking problem in adolescents, the wide
variety of interventions employed, and the absence of any
rigorous evaluation of different treatments.[1] Alcohol
interventions in adolescents have focused on primary
prevention of alcohol use.[10] Recent reviews of school-based
programs found that most effects were inconsistent, small, and
short-lived; programs that sought to develop social skills to
resist drug use seem to be more effective than programs that
emphasize factual knowledge.[89] ,[90]
Alcohol-Dependent Patients.
Patients with alcohol dependence usually receive more

intensive treatment. A 1989 report of the IOM[91] reviewed a
variety of alcohol treatment modalities and concluded that
various treatments were effective, but there was no single
superior treatment for all patients, and few treatments were
effective for the majority of patients. They found no evidence
that residential versus nonresidential programs, or long-versus
short-duration programs, were more effective for the average
patient, and no studies existed that adequately evaluated the
independent effect of Alcoholics Anonymous (AA). In a
subsequent trial among employees referred for alcohol
problems, patients who received inpatient treatment and
mandatory AA follow-up were more likely to be abstinent at
2-year follow-up (37% vs. 16%) than patients assigned to
mandatory AA only.[92]
Two short-term (12 weeks) randomized trials demonstrated a

significant benefit of naltrexone, an opioid antagonist, as an

adjunct to treatment of alcohol dependence. In one study,
patients receiving naltrexone and supportive psychotherapy had
significantly higher abstinence rates than did subjects receiving
placebo (61% vs. 19%).[93] In the second, men receiving
naltrexone reported less alcohol craving and fewer drinking
days than placebo-treated men.[94] In both trials, naltrexone
significantly reduced the likelihood of relapse (heavy drinking
or steady drinking) among subjects who did not achieve
complete abstinence. The benefits of alcohol-sensitizing agents,
however, remain uncertain.[10] Disulfiram (i.e., Antabuse) did
not improve long-term abstinence rates in a controlled trial, but
it did reduce drinking days among patients receiving the highest
dose.[95]
In a 10-year follow-up of 158 patients completing inpatient
treatment, 61% reported complete or stable remission of
alcoholism.[96] Completing an extended inpatient program was
associated with significantly lower mortality among alcoholic
patients in a second study.[97] Many such studies of alcohol
treatment, however, suffer from important methodologic
limitations: inadequate control groups, insufficient or selective
follow-up, and selection bias due to the characteristics of
patients who successfully complete voluntary treatment
programs.[91] ,[98] ,[99] Since spontaneous remission occurs in as
many as 30% of alcoholics,[100] ,[101] reduced consumption may
be inappropriately attributed to treatment. Successful treatment
is likely to represent a complex interaction of patient
motivation, treatment characteristics, and the posttreatment
environment (family support, stress, etc.).[1] ,[10] The IOM
review concluded that treatment of other life problems (e.g.,
with antidepressant medication, family or marital therapy, or
stress management) and empathetic therapists were likely to
improve treatment outcomes.[91]
Recommendations of Other Groups
There is a consensus among professional groups such as the
American Medical Association (AMA)[102] and the American
Academy of Family Physicians (AAFP)[103] that clinicians
should be alert to the signs and symptoms of alcohol abuse and
should routinely discuss patterns of alcohol use with all
patients. AAFP recommendations are under review. The
Canadian Task Force on the Periodic Health Examination
(CTF)[104] and a 1990 IOM panel[1] recommended screening
adults for problem drinking, using patient inquiry or
standardized instruments, and offering brief counseling to
nondependent problem drinkers.
The American Academy of Pediatrics (AAP),[105] AMA

Guidelines for Adolescent Preventive Services (GAPS),[106] the
Bright Futures guidelines,[107] and the AAFP[103] all recommend
careful discussion with all adolescents regarding alcohol use
and regular advice to abstain from alcohol. The AAP also
advises physicians to counsel parents regarding their own use of
alcohol in the home. Recommendations of the U.S. Surgeon
General,[108] the American College of Obstetricians and
Gynecologists,[109] and the AAP[109] ,[110] advise counseling all
women who are pregnant or planning pregnancy that drinking
can be harmful to the fetus and that abstinence is the safest
policy. The CTF recommends that all women be screened for
problem drinking and advised to reduce alcohol use during
pregnancy.[104]
Several organizations have made recommendations about "safe"
levels of alcohol consumption for nonpregnant adults. The
National Institute on Alcohol Abuse and Alcoholism,[111] the
U.S. Surgeon General,[112] and dietary guidelines produced
jointly by the U.S. Departments of Health and Human Services
and Agriculture[113] ,[114] recommend no more than 2 drinks per
day for men and 1 drink per day for nonpregnant women.
Slightly higher limits were proposed by national health
authorities in the U.K.[115]
Discussion
Alcohol problems are common in the primary care setting, but
they often go undetected by clinicians. Although imperfect,
asking patients direct questions about the quantity, frequency,
and pattern of their drinking is an important way to identify
those who are most likely to experience problems due to
alcohol. Questions about tolerance to the effects of alcohol may
circumvent denial among pregnant women and heavy drinkers.
The CAGE and other brief screening instruments are useful
supplements to the standard patient history, but they may be
less sensitive for early problems and hazardous drinking. The
AUDIT may detect a broader range of current drinking
problems, but its performance in the primary care setting needs
further evaluation. Although laboratory tests such as GGT are
not sufficiently sensitive or specific for routine screening, they
may be useful in selected high-risk patients to confirm clinical
suspicion or to motivate changes in drinking. Neither
questionnaires nor laboratory tests should be considered
diagnostic of problem drinking without more detailed
evaluation (see Clinical Intervention ).
Detecting early problem drinkers is important, because they
http://home.mdconsult.com/das/guideline/view/N/1...stty=cow&nid=39982&tlt=Z&pos=&tag=relinfo&anc=gl (10 of 25) [21/04/2000 23:28:00]

account for a large proportion of all alcohol problems and they

are more likely to respond to simple interventions than patients
with alcohol dependency. There is now good evidence that brief
counseling can reduce alcohol consumption in problem
drinkers, and several trials have also reported improved clinical
outcomes. Since the risks from alcohol rise steadily at higher
levels of consumption, reducing drinking should also benefit
heavy drinkers (i.e., hazardous drinkers) who do not yet
manifest problems due to drinking. Early attention to problem
drinking is especially important in young adults: hazardous
drinking is common, adverse effects of alcohol increase with
duration of use, and few persons initiate drinking after age
30.[116] Early detection is also important for alcohol-dependent
patients, but effective treatment requires more intensive and
sustained efforts to promote abstinence.
Uncertainties remain about optimal screening methods and
interventions during pregnancy, but screening is justified by the
strong evidence of the adverse effects of alcohol on the fetus.
Although the risks of occasional, light drinking during
pregnancy have not been established, abstinence can be
recommended as a prudent approach for pregnant women. At
the same time, women concerned about the effects of previous
moderate drinking early in pregnancy can be reassured that
important harms have not been demonstrated from such limited
exposures. Because exposure early in pregnancy may be most
important, screening and advice should be directed toward
women contemplating pregnancy and those at risk for
unintended pregnancy, not just women who are already
pregnant.
There is insufficient evidence to make precise recommendations
about desirable levels of drinking, but the strong association
between heavy alcohol use and risk of future complications
justifies advising all drinkers to drink moderately and avoid
frequent intoxication, even in the absence of current problems
(see below).
Screening to detect problem drinking and hazardous

drinking is recommended for all adult and adolescent
patients ("B" recommendation). Screening should involve a
careful history of alcohol use and/or the use of standardized
screening questionnaires. Patients should be asked to
describe the quantity, frequency, and other characteristics
of their use of wine, beer, and liquor, including frequency of

intoxication and tolerance to the effects of alcohol. One

drink is defined as 12 ounces of beer, a 5-ounce glass of
wine, or 1.5 fluid ounces (one jigger) of distilled spirits.
Brief questionnaires such as the CAGE or AUDIT may help
clinicians assess the likelihood of problem drinking or
hazardous drinking (see Table 52.1 ). Responses suggestive
of problem drinking should be confirmed with more
extensive discussions with the patient (and family members
where indicated) about patterns of use, problems related to
drinking, and symptoms of alcohol dependence.[2] Routine
measurement of biochemical markers, such as serum GGT,
is not recommended for screening purposes. Discussions
with adolescents should be approached with discretion to
establish a trusting relationship and to respect the patient's
concerns about the confidentiality of disclosed information.
All pregnant women should be screened for evidence of
problem drinking or risk drinking (2 drinks per day or
binge drinking) ("B" recommendation). Including questions
about tolerance to alcohol may improve detection of at-risk
women. All pregnant women and women contemplating
pregnancy should be informed of the harmful effects of
alcohol on the fetus and advised to limit or cease drinking.
Although there is insufficient evidence to prove or disprove
harms from occasional, light drinking during pregnancy,
abstinence from alcohol can be recommended on other
grounds: possible risk from even low-level exposure to
alcohol, lack of harm from abstaining, and prevailing expert
opinion ("C" recommendation). Women who smoke should
be advised that the risk of low birth weight is greatest for
mothers who both smoke and drink.
Patients with evidence of alcohol dependence should be
referred, where possible, to appropriate clinical specialists
or community programs specializing in the treatment of
alcohol dependence. Patients with evidence of alcohol abuse
or hazardous drinking should be offered brief advice and
counseling. Counseling should involve feedback of the
evidence of a drinking problem, discussion of the role of
alcohol in current medical or psychosocial problems, direct
advice to reduce consumption, and plans for regular
follow-up. Problems related to alcohol (e.g., physical
symptoms, behavioral or mood problems, or difficulties at
work and home) should be monitored to determine whether
further interventions are needed. There is no single
definition of "hazardous" drinking in asymptomatic
persons, but successful intervention trials have generally
defined 5 drinks per day in men, 3 drinks per day in women,

or frequent intoxication to identify persons at risk. Several
U.S. organizations have suggested lower limits for "safe"
drinking: 2 drinks per day in men and 1 drink per day in
women.[18] All persons who drink should be informed of the
dangers of driving or other potentially dangerous activities
after drinking (see Chapter 57). The use of alcohol should
be discouraged in persons younger than the legal age for
drinking ("B" recommendation), although the effectiveness
of alcohol abstinence messages in the primary care setting is
uncertain.
TABLE 52.1 -- AUDIT Structured Interview [

Score
Question 0 1 2 3 4
How often
do you have 4 or
Monthly 2-4 2-3
a drink Never more
or less times/mo times/wk
containing times/wk
alcohol?
How many
drinks do
you have on
a typical day None 1 or 2 3 or 4 5 or 6 7-9 [
when you
are
drinking?
How often
do you have
Less Daily or
6 or more
Never than Monthly Weekly almost
drinks on
monthly daily
one
occasion?

How often
during the
last year
have you
found that Less Daily or
you were Never than Monthly Weekly almost
unable to monthly daily
stop
drinking
once you
had started?
How often
during the
last year
have you
failed to do Less Daily or
what was Never than Monthly Weekly almost
normally monthly daily
expected
from you
because of
drinking?
How often
during the
last year
have you
needed a
first drink in Less Daily or
the morning Never than Monthly Weekly almost
to get monthly daily
yourself
going after a
heavy
drinking
session?
How often
during the
last year
have you Less Daily or
had a feeling Never than Monthly Weekly almost
of guilt or monthly daily
remorse
after
drinking?

How often
during the
last year
have you
been unable
to remember Less Daily or
what Never than Monthly Weekly almost
happened monthly daily
the night
before
because you
had been
drinking?
Have you or
someone
else been
Yes, but not in last Yes, during the last
injured as a Never
year (2 points) year (4 points)
result of
your
drinking?
Has a
relative,
doctor, or
other health
worker been
Yes, but not in last Yes, during the last
concerned Never
year (2 points) year (4 points)
about your
drinking or
suggested
you cut
down?
[a] Score of greater than 8 (out of 41) suggests problem drinking and
indicates need for more in-depth assessment. Cut-off of 10 points
recommended by some to provide greater specificity.
[*] 5 points if response is 10 or more drinks on a typical day.
The draft update of this chapter was prepared for the U.S.
Preventive Services Task Force by David Atkins, MD, MPH,
with contributions from materials prepared for the Canadian
Task Force on the Periodic Health Examination by Deborah L.
Craig, MPH, and Jean L. Haggerty, MSc.

REFERENCES
1. Institute
of Medicine.
Broadening the base of treatment for alcohol problems: report of a study
by a committee of the Institute of Medicine, Division of Mental Health and
Behavioral Medicine.Washington, DC: National Academy Press, 1990.
2. American Psychiatric Association.

Diagnostic and statistical manual of mental disorders. 4th ed.Washington,
DC: American Psychiatric Association, 1994.
3. World Health Organization.

International classification of diseases, 10th revision.Geneva: World
Health Organization, 1992.
4. SubstanceAbuse and Mental Health Administration.

National Household Survey on Drug Abuse: population estimates 1993.
Rockville, MD: U.S. Department of Health and Human Services,
Substance Abuse and Mental Health Administration, 1994. (Publication
no. (SMA) 94-3017.)
5. Grant
BF, Harford TC, Chou P, et al.
Epidemiologic Bulletin no.27: Prevalence of DSM-III-R alcohol abuse and
dependence: United States, 1988.Alcohol Health Res World
1991;15:91-96.
6. HelzerJE, Burnam A, McEvoy LT.

Alcohol abuse and dependence.In: Robins LN, Regier DA, eds. Psychiatric
disorders in America: the Epidemiologic Catchment Area Study.New
York:Free Press, 1991:81-115.
7. GrantBF, Harford TC, Dawson D, et al.

Prevalence of DSM-IV alcohol abuse and dependence: United States,
1992.Alcohol Health Res World 1994;18:243-248.
8. KesslerRC, McGonagle KA, Zhao S, et al.

Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in
the United States: results from the National Comorbidity Survey.Arch Gen
Psychiatry 1994;51:8-19.
9. BradleyKA.
The primary care provider's role in the prevention and management of
alcohol problems.Alcohol Health Res World 1994;18:97-104.
10. NationalInstitute on Alcohol Abuse and Alcoholism.

Alcohol and health: eighth special report to the U.S. Congress from the
Secretary of Health and Human Services. Rockville, MD: U.S.Department
of Health and Human Services, National Institutes of Health, 1993.(NIH
Publication no. 94-3699.)
11. Klatsky
AL, Armstrong AA, Friedman GD.
Alcohol and mortality.Ann Intern Med 1992;117:646-654.

12. Boffetta
P, Garfinkel L.
Alcohol drinking and mortality among men enrolled in an American
Cancer Society prospective study.Epidemiology 1990;1:342-348.
13. Anderson P, Cremona A, Paton A, et al.

The risk of alcohol.Addiction 1993;88:1493-1508.
14. RosenbergL, Metzger LS, Palmer JR.

Alcohol consumption and risk of breast cancer: a review of the
epidemiologic evidence.Epidemiol Rev 1993;15:133-168.
15. FuchsCS, Stampfer MJ, Colditz GA, et al.

Alcohol consumption and mortality among women.N Engl J Med
1995;332:1245-1250.
16. Centers
for Disease Control.
Alcohol-related mortality and years of potential life lost -- United States,
1987.MMWR 1990;39:173-178.
17. National Highway Traffic Safety Administration.

Traffic safety facts 1993: alcohol.Washington, DC: U.S. Department of
Transportation, National Highway Traffic Safety Administration, National
Center for Statistics and Analysis, 1994.
18. Bradley
KA, Donovan DM, Larson EB.
How much is too much? Advising patients about safe levels of alcohol
consumption.Arch Intern Med 1993;153:2734-2740.
19. Stampfer
MJ, Rimm EB, Walsh DC.
Commentary: alcohol, the heart, and public policy.Am J Public Health
1993;83:801-804.
20. MaclureM.
Demonstration of deductive meta-analysis: ethanol intake and risk of
myocardial infarction.Epidemiol Rev 1993;15:328-351.
21. KlatskyAL, Armstrong AA, Friedman GD.

Risk of cardiovascular mortality in alcohol drinkers, ex-drinkers and
non-drinkers.Am J Cardiol 1990;66:1237-1242.
22. StampferMJ, Colditz GA, Willett WC, et al.

A prospective study of moderate alcohol consumption and the risk of
coronary disease and stroke in women.N Engl J Med 1988;319:267-273.
23. Gaziano JM, Buring JE, Breslow JL, et al.

Moderate alcohol intake, increased levels of high-density lipoprotein and
its subfractions, and decreased risk of myocardial infarction.N Engl J Med
1993;329:1829-1835.
24. RidkerPM, Vaughan DE, Stampfer MJ, et al.

Association of moderate alcohol consumption and plasma concentration of
endogenous tissue-type plasminogen activator.JAMA 1994;272:929-933.
25. Sher KJ, ed.

Children of alcoholics: a critical appraisal of theory and research.Chicago:

University of Chicago Press, 1991.
26. SerdulaM, Williamson DF, Kendrick J, et al.

Trends in alcohol consumption by pregnant women: 1985 through
1988.JAMA 1991;265:876-879.
27. Goodwin MM, Bruce C, Zahniser SC, et al.

Behavioral risk factors before and during pregnancy.In: Wilcox LS, Marks
JS, eds. From data to action: CDC's public health surveillance for women,
infants, and children. Washington, DC:Department of Health and Human
Services, 1994:93-104.
28. Centersfor Disease Control and Prevention.

Frequent alcohol consumption among women of childbearing age --
Behavioral Risk Factor Surveillance System, 1991.MMWR
1994;43:328-329, 335.
29. Centers
for Disease Control and Prevention.
Sociodemographic and behavioral characteristics associated with alcohol
consumption during pregnancy -- United States, 1988.MMWR
1995;44:261-264.
30. Sokol
RJ, Martier SS, Ager JW.
The T-ACE questions: practical prenatal detection of risk-drinking.Am J
Obstet Gynecol 1989;160:863-870.
31. Russell
M, Martier SS, Sokol RJ, et al.
Screening for pregnancy risk-drinking.Alcohol Clin Exp Res
1994;18:1156-1161.
32. Rosett
HL, Weiner L, Edelin KC.
Treatment experience with pregnant problem drinkers.JAMA
1983;249:2029-2033.
33. AbelEA, Sokol RJ.

A revised conservative estimate of the incidence of FAS and its economic
impact.Alcohol Clin Exp Res 1991;15:514-524.

Fetal alcohol syndrome -- United States, 1979-1992.MMWR
1993;42:339-341.
35. RussellM.
Clinical implications of recent research on the fetal alcohol syndrome.Bull
NY Acad Sci 1991;67:207-222.
36. Burd L, Moffatt ME.

Epidemiology of fetal alcohol syndrome in American Indians, Alaskan
Natives, and Canadian Aboriginal peoples: a review of the
literature.Public Health Rep 1994;109:688-693.
37. Virji
SK.
The relationship between alcohol consumption during pregnancy and

infant birthweight: an epidemiologic study.Acta Obstet Gynecol Scand

1991;303-308.
38. Forrest
F, Florey CDV, Taylor D, et al.
Reported social alcohol consumption during pregnancy and infants'
development at 18 months.BMJ 1991;303:22-26.
39. Verkerk PH, Noord-Zaadstra BMV, Florey CDV, et al.

The effect of moderate maternal alcohol consumption on birth weight and
gestational age in a low risk population.Early Hum Dev 1993;32:121-129.
40. JacobsonJL, Jacobson SW, Sokol RJ, et al.

Teratogenic effects of alcohol on infant development.Alcohol Clin Exp Res
1993;17:174-183.
41. Streissguth
AP, Barr HM, Sampson PD.
Moderate prenatal alcohol exposure: effects on child IQ and learning
problems at age 7.5 years.Alcohol Clin Exp Res 1990;14:662-669.
42. KnupferG.
Abstaining for fetal health: the fiction that even light drinking is
dangerous.Br J Addict 1991;86:1063-1073.
43. NationalInstitute on Drug Abuse.

National survey results on drug use from the Monitoring the Future Study,
1975-1992. Vols 1 and 2. Rockville, MD: Department of Health and
Human Services, National Institute on Drug Abuse, 1993. (NIH
Publications nos. 93-3596 and 3597.)
44. Centers
Youth Risk Behavior Surveillance -- United States, 1993. MMWR
1995;44(SS-1):1-56.
45. Centers
Drinking and driving and binge drinking in selected states, 1982 and 1985:
the Behavioral Risk Factor Surveys. MMWR 1987;35:788-791.
46. Wechsler H, Davenport A, Dowdall G, et al.

Health and behavioral consequences of binge drinking in college: a
national survey of students at 140 campuses.JAMA 1994;272:1672-1677.
47. GlazeLW, Coggan PG.

Efficacy of an alcoholism self-report questionnaire in a residency clinic.J
Fam Pract 1987;25:60-64.
48. Buchsbaum DG, Welsh J, Buchanan RG, et al.

Screening for drinking problems by patient self-report: even "safe" levels
may indicate a problem.Arch Intern Med 1995;155:104-108.
49. Selzer
ML.
The Michigan Alcoholism Screening Test: the quest for a new diagnostic
instrument.Am J Psychiatry 1971;127:1653-1658.
50. Pokorny AD, Miller BA, Kaplan HB.

The Brief MAST: a shortened version of the Michigan Alcoholism

Screening Test.Am J Psychiatry 1972;129:342-345.
51. BarryK, Fleming M.

The Alcohol Use Disorders Identification Test (AUDIT) and the
SMAST-13: predictive validity in a rural primary care sample.Alcohol
Alcohol 1993;28:33-42.
52. Chan AWK, Pristach EA, Welte JW, et al.

Use of the TWEAK test in screening for alcoholism/heavy drinking in
three populations.Alcohol Clin Exp Res 1993;17:1188-1192.
53. EwingJA.
Detecting alcoholism: the CAGE questionnaire.JAMA
1984;252:1905-1907.
54. BernadtMW, Mumford J, Taylor C, et al.

Comparison of questionnaire and laboratory tests in the detection of
excessive drinking and alcoholism.Lancet 1982;1:325-328.
55. BushB, Shaw S, Cleary P, et al.

Screening for alcohol abuse using the CAGE questionnaire.Am J Med
1987;82:231-235.
56. King M.
At risk drinking among general practice attenders: validation of the CAGE
questionnaire.Psychol Med 1986;16:213-217.
57. Buchsbaum DG, Buchanan RG, Centor RM, et al.

Screening for alcohol abuse using CAGE scores and likelihood ratios.Ann
Intern Med 1991;115:774-777.
58. Chan AWK, Pristach EA, Welte JW.

Detection by the CAGE of alcoholism or heavy drinking in primary care
outpatients and the general population.J Subst Abuse 1994;6:123-135.
59. HaysJT, Spickard WA.

Alcoholism: early diagnosis and intervention.J Gen Intern Med
1987;2:420-427.
60. SaundersJB, Aasland OG, Babor TF, et al.

Development of the Alcohol Use Disorders Identification Test (AUDIT):
WHO collaborative project on early detection of persons with harmful
alcohol consumption --II.Addiction 1993;88:791-804.
61. IsaacsonJH, Butler R, Zacharek M, et al.

Screening with the Alcohol Use Disorders Identification Test (AUDIT) in
an inner-city population.J Gen Intern Med 1994;9:550-553.
62. SchmidtA, Barry K, Fleming MF.

Detection of problem drinkers: the Alcohol Use Disorders Identification
Test (AUDIT).Southern Med J 1995;88:52-59.
63. Werner MJ, Walker LS, Greene JW.

Screening for problem drinking among college freshmen.J Adolesc Health

1994;15:303-310.
64. Fleming MF, Barry KL, MacDonald R.

The Alcohol Use Disorders Identification Test (AUDIT) in a college
sample.Int J Addict 1991;26:1173-1185.
65. Petchers
MK, Singer MI.
Perceived-benefit-of-drinking Scale: approach to screening for adolescent
alcohol abuse.J Pediatr 1987;110:979-981.
66. National
Institute on Drug Abuse.
Problem oriented screening instrument for teenagers.In: Rahdert E, ed.
The adolescent assessment/referral system manual.Rockville,
MD:National Institute on Drug Abuse, 1990:5-9.
67. WatersonEJ, Murray-Lyon IM.

Screening for alcohol-related problems in the antenatal clinic.Alcohol
Alcohol 1989;24:21-30.
68. JacobsonSW, Jacobson JL, Sokol RJ, et al.

Maternal recall of alcohol, cocaine, and marijuana use during
pregnancy.Neurotoxicol Teratol 1991;13:535-540.
69. Ernhart
CB, Morrow-Tlucak M, Sokol RJ, et al.
Underreporting of alcohol use in pregnancy.Alcohol Clin Exp Res
1988;12:506-511.
70. Russell
Screening for pregnancy risk-drinking.Alcohol Exp Clin Res
1994;18:1156-1161.
71. Russell
M.
New assessment tools for risk drinking during pregnancy: T-ACE,
TWEAK, and others.Alcohol Health Res World 1994;18:55-61.
72. Hoeksema HL, de Bock GH.

The value of laboratory tests for the screening and recognition of alcohol
abuse in primary care patients.J Fam Pract 1993;37:268-276.
73. KristensonH, Trell E, Hood B.

Serum glutamyl-transferase in screening and continuous control of heavy
drinking in middle-aged men.Am J Epidemiol 1981;114:862-872.
74. PerssonJ, Magnusson PH.

Early intervention in patients with excessive consumption of alcohol: a
controlled study.Alcohol 1989;6:403-408.
75. Romelsjo A, Andersson L, Barrner H, et al.

A randomized study of secondary prevention of early stage problem
drinkers in primary health care.Br J Addict 1989;84:1319-1327.
76. Nilssen
O.
The Tromso Study: identification of and a controlled intervention on a

population of early-stage risk drinkers.Prev Med 1991;20:518-528.
77. ChickJ, Lloyd G, Crombie E.

Counselling problem drinkers in medical wards: a controlled study.BMJ
1985;290:965-967.
78. WallaceP, Cutler S, Haines A.

Randomized controlled trial of general practitioner intervention in patients
with excessive alcohol consumption.BMJ 1988;297:663-668.
79. HeatherN, Kissoon-Singh J, Fenton GW.

Assisted natural recovery from alcohol problems: effects of a self-help
manual with and without supplementary telephone contact.Br J Addict
1990;85:1177-1185.
80. BaborTF, Grant M, eds.

Project on identification and management of alcohol-related problems.
Report on Phase II: a randomized clinical trial of brief interventions in
primary health care.Geneva: World Health Organization, 1992.
81. Brief
interventions and alcohol use.
Effective Health Care, bulletin no. 7. Leeds, UK: Nuffield Institute for
Health Care, University of Leeds, 1993.
82. Maheswaran R, Beevers M, Beevers DG.

Effectiveness of advice to reduce alcohol consumption in hypertensive
patients.Hypertension 1992:19:79-84.
83. CoxKL, Puddey IB, Morton AR, et al.

The combined effects of aerobic exercise and alcohol restriction on blood
pressure and serum lipids: a two-way factorial study in sedentary men.J
Hypertens 1993;11:191-201.
84. Schorling JB.

The prevention of prenatal alcohol use: a critical analysis of intervention
studies.J Stud Alcohol 1993;54:261-267.
85. LarsonG.
Prevention of fetal alcohol effects.Acta Obstet Gynecol Scand
1983;62:171-178.
86. Halmesmaki E.
Alcohol counselling of 85 pregnant problem drinkers: effect on drinking
and fetal outcome.Br J Obstet Gynaecol 1988;95:243-247.
87. Waterson EJ, Murray-Lyon IM.

Preventing fetal alcohol effects: a trial of three methods of giving
information in the antenatal clinic.Health Educ Res 1990;5:53-61.
88. MebergA, Halvorsen B, Holter B, et al.

Moderate alcohol consumption -- need for intervention programs in
pregnancy?Acta Obstet Gynecol Scand 1986;65:861-864.
89. Ennett ST, Tobler NS, Ringwalt CL, et al.

How effective is drug abuse resistance education? A meta-analysis of

project DARE outcome evaluations.Am J Public Health
1994;84:1394-1401.
90. HansenWB.
School-based substance abuse prevention: a review of the state of the art in
curriculum, 1980-1990. Health Educ Res 1992;7:403-430.
91. Institute
of Medicine.
Prevention and treatment of alcohol problems: research
opportunities.Washington, DC: National Academy Press, 1989.
92. WalshDC, Hingson RW, Merrigan DM, et al.

A randomized trial of treatment options for alcohol-abusing workers.N
Engl J Med 1991;325:775-782.
93. O'MalleySS, Jaffe AJ, Chang G, et al.

Naltrexone and coping skills therapy for alcohol dependence: a controlled
study.Arch Gen Psychiatry 1992;49:881-887.
94. Volpicelli
JR, Alterman AI, Hayashida M, et al.
Naltrexone in the treatment of alcohol dependence.Arch Gen Psychiatry
1992;49:876-880.
95. Fuller
RK, Branchey L, Brightwell DR, et al.
Disulfiram treatment of alcoholism: A Veterans Administration
study.JAMA 1986;256:1449-1455.
96. Cross
GM, Morgan CW, Mooney AJ, et al.
Alcoholism treatment: a ten-year follow-up study.Alcohol Clin Exp Res
1990;14:169-173.
97. Bunn JY, Booth BM, Loveland Cook CA, et al.

The relationship between mortality and intensity of inpatient alcoholism
treatment.Am J Public Health 1994;84:211-214.
98. Thurstin
AH, Alfano AM, Sherer M.
Pretreatment MMPI profiles of A.A. members and nonmembers.J Stud
Alcohol 1986;47:468-471.
99. EmrickCD.
Alcoholics Anonymous: affiliation processes and effectiveness as
treatment.Alcohol Clin Exp Res 1987;11:416-423.
100. SmartR.
Spontaneous recovery in alcoholics: a review and analysis of the available
research.Drug Alcohol Depend 1975/76;1:277.
101. Saunders
WM, Kershaw PW.
Spontaneous remission from alcoholism: a community study.Br J Addict
1979;74:251-265.
102. American Medical Association, Council on Scientific Affairs.

Guidelines for alcoholism diagnosis, treatment and referral.Chicago:

American Medical Association, 1979.
103. American Academy of Family Physicians.

Age charts for periodic health examination. Kansas City, MO: American
Academy of Family Physicians, 1994.(Reprint no. 510.)
104. Canadian
Task Force on the Periodic Health Examination.
Canadian guide to clinical preventive health care.Ottawa: Canada
Communication Group, 1994:52-61, 826-836.
105. AmericanAcademy of Pediatrics, Committee on Adolescence.

Alcohol use and abuse: a pediatric concern.Pediatrics 1995;95:439-442.
106. American Medical Association.

Guidelines for adolescent preventive services (GAPS): recommendations
and rationale.Chicago: American Medical Association, 1994.
107. GreenM, ed.

Bright Futures: guidelines for health supervision of infants, children and
adolescents. Arlington VA: National Center for Education in Maternal and
Child Health, 1994.
108. Surgeon
General's Advisory on alcohol and pregnancy.
FDA Drug Bull 1981;11:9-10.
109. American Academy of Pediatrics and American College of

Obstetricians and Gynecologists.
Guidelines for perinatal care. 3rd ed.Washington, DC: American Academy
of Pediatrics and American College of Obstetricians and Gynecologists,
1992.
110. American Academy of Pediatrics, Committee on Substance Abuse and

Committee on Children with Disabilities.
Fetal alcohol syndrome and fetal alcohol effects.Pediatrics
1993;91:1004-1006.
111. National
Institute on Alcohol Abuse and Alcoholism.
Moderate drinking. Alcohol Alert no. 16. Bethesda: U.S.Department of
Health and Human Services, 1992.
112. U.S.
Department of Health and Human Services.
The Surgeon General's report on nutrition and health.Washington, DC:
Government Printing Office, 1988.(Publication no. PHS-88-50210.)
113. U.S.Department of Agriculture and U.S. Department of Health and

Human Services.
Dietary guidelines for Americans.Washington, DC: Department of
Agriculture, 1990.
114. DietaryGuidelines Advisory Committee.

Report of the Dietary Guidelines Advisory Committee on the dietary
guidelines for Americans, 1995, to the Secretary of Health and Human
Services and the Secretary of Agriculture. Washington, DC: U.S.
Department of Agriculture, 1995.

115. Secretaryof State for Health.

The Health of the Nation: a strategy for health in England.London: Her
Majesty's Stationery Office, 1992.
116. ChenK, Kandel D.

The natural history of drug use from adolescence to mid-thirties in a
general population sample.Am J Public Health 1995;85:41-47.
[a] C: "Have you ever felt you ought to Cut down on drinking?" A: "Have
people Annoyed you by criticizing your drinking?" G: "Have you ever felt
bad or Guilty about your drinking?" E: "Have you ever had a drink first
thing in the morning to steady your nerves or get rid of a hangover (Eye
opener)?

Bookmark URL: /das/guideline/view/N/10134884?ja=104571&PAGE=1.html&ANCHOR=top

U.S. Preventive Services Task Force, Guidelines from Guide to Clinical Preventive Services, (Second Edition) 1996
1996
Copyright 1996 Williams & Wilkins (Reprinted with permission)
Section I. SCREENING
Part I. Mental Disorders and Substance Abuse
This guideline is a portion of a larger document that cannot be downloaded efficiently as a single HTML
file. See other guidelines from the U.S. Preventive Services Task Force to read the complete document.
RECOMMENDATION
Screening to detect problem drinking is recommended for all adult and adolescent patients.
Screening should involve a careful history of alcohol use and/or the use of standardized screening
questionnaires (see Clinical Intervention ). Routine measurement of biochemical markers is not
recommended in asymptomatic persons. Pregnant women should be advised to limit or cease
drinking during pregnancy. Although there is insufficient evidence to prove or disprove harms
from light drinking in pregnancy, recommendations that women abstain from alcohol during
pregnancy may be made on other grounds (see Clinical Intervention ). All persons who use alcohol
should be counseled about the dangers of operating a motor vehicle or performing other
potentially dangerous activities after drinking alcohol.
Burden of Suffering
Over half a million Americans are under treatment for alcoholism, but there is growing recognition that
alcoholism (i.e., alcohol dependence) represents only one end of the spectrum of "problem drinking."[1]
Many problem drinkers have medical or social problems attributable to alcohol (i.e., alcohol abuse or
"harmful drinking") without typical signs of dependence,[2] ,[3] and other asymptomatic drinkers are at
risk for future problems due to chronic heavy alcohol consumption or frequent binges (i.e., "hazardous
drinking"). Heavy drinking (more than 5 drinks per day, 5 times per week) is reported by 10% of adult
men and 2% of women.[4] In large community surveys using detailed interviews,[5] [6] [7] [8] the prevalence
of alcohol abuse and dependence in the previous year among men was 17-24% among 18-29-year-olds,
11-14% among 30-44-year-olds, 6-8% among 45-64-year-olds, and 1-3% for men over 65; among
women in the corresponding age groups, prevalence of abuse or dependence was 4-10%, 2-4%, 1-2%,
and less than 1%, respectively. Problem drinking is even more common among patients seen in the
primary care setting (8-20%).[9]
Medical problems due to alcohol dependence include alcohol withdrawal syndrome, psychosis, hepatitis,
cirrhosis, pancreatitis, thiamine deficiency, neuropathy, dementia, and cardiomyopathy.[10] Nondependent
http://home.mdconsult.com/das/article/body/jorg=guideline&source=&sp=10134884&sid=0/N/104571/1.html (1 of 17) [21/04/2000 23:28:26]

heavy drinkers, however, account for the majority of alcohol-related morbidity and mortality in the
general population.[1] There is a dose-response relationship between daily alcohol consumption and
elevations in blood pressure and risk of cirrhosis, hemorrhagic stroke, and cancers of the oropharynx,
larynx, esophagus, and liver.[11] [12] [13] A number of studies have reported a modest increase in breast
cancer among women drinking 2 drinks per day or more, but a causal connection has not yet been
proven.[14] Three large cohort studies, involving over 500,000 men and women, observed increasing
all-cause mortality beginning at 4 drinks per day in men[11] ,[12] and above 2 drinks per day in women.[15]
Women achieve higher blood alcohol levels than do men, due to smaller size and slower metabolism.[11]
,[15] Compared to nondrinkers and light drinkers, overall mortality was 30-38% higher among men, and
more than doubled among women, who drank 6 or more drinks per day.[11] ,[12] Of the more than 100,000
deaths attributed to alcohol annually, nearly half are due to unintentional and intentional injuries,[16]
including 44% of all traffic fatalities in 1993[17] and a substantial proportion of deaths from fires,
drownings, homicides, and suicides (see Chapters 50, 51, 57, 58, and 59).
The social consequences of problem drinking are often as damaging as the direct medical consequences.
Nearly 20% of drinkers report problems with friends, family, work, or police due to drinking.[10] Persons
who abuse alcohol have a higher risk of divorce, depression, suicide, domestic violence, unemployment,
and poverty (see Chapters 49, 50, and 51).[10] Intoxication may lead to unsafe sexual behavior that
increases the risk of sexually transmitted diseases, including human immunodeficiency virus (HIV).
Finally, an estimated 27 million American children are at risk for abnormal psychosocial development
due to the abuse of alcohol by their parents.[25]
Moderate alcohol consumption has favorable effects on the risk of coronary heart disease (CHD).[18] [19]
[20] [21] [22] [23] CHD incidence and mortality rates are 20-40% lower in men and women who drink 1-2
drinks/day than in nondrinkers.[15] ,[21] ,[22] A meta-analysis of epidemiologic studies suggests little
additional benefit of drinking more than 0.5 drinks per day.[20] The exact mechanism for the protective
effect of alcohol is not known but may involve increases in high-density lipoprotein[23] and/or fibrinolytic
mediators.[24]
Alcohol Use during Pregnancy.
The proportion of pregnant women who report drinking has declined steadily in the U.S.[26] Recent
surveys indicate 12-14% of pregnant women continue to consume some alcohol,[27] ,[28] with most
reporting only occasional, light drinking (median: 4 drinks per month).[26] Binge drinking or daily risk
drinking (usually defined as 2 drinks per day or greater) is reported by 1-2% of pregnant women,[27] [28]
[29] but higher rates (4-6%) have been reported in some screening studies.[30] ,[31] Excessive use of alcohol
during pregnancy can produce the fetal alcohol syndrome (FAS), a constellation of growth retardation,
facial deformities, and central nervous system dysfunction (microcephaly, mental retardation, or
behavioral abnormalities).[32] Other infants display growth retardation or neurologic involvement in the
absence of full FAS (i.e., fetal alcohol effects [FAE]).[10] FAS has been estimated to affect approximately
1 in 3,000 births in the U.S. (1,200 children annually), making it a leading treatable cause of birth defects
and mental retardation.[33] ,[34]
The level of alcohol consumption that poses a risk during pregnancy remains controversial.[10] ,[35] FAS
has only been described in infants born to alcoholic mothers, but the variable incidence of FAS among
alcoholic women (from 3-40%)[33] suggests that other factors (e.g., genetic, nutritional, metabolic, or
temporal) may influence the expression of FAS.[10] The reported incidence of FAS is higher in Native

Americans and blacks than in whites.[33] ,[36] Most studies report an increased risk of FAE among mothers
who consume 14 drinks per week or more,[35] ,[37] [38] [39] but the effects of lower levels of drinking have
been inconsistent.[35] ,[40] ,[41] Modest developmental effects have been attributed to light drinking (7
drinks per week) in some studies, but underreporting by heavy drinkers and confounding effects of other
important factors (nutrition, environment, etc.) make it difficult to prove or disprove a direct effect of
light drinking.[10] ,[35] ,[42] Timing of exposure and pattern of drinking may be important, with greater
effects proposed for exposure early in pregnancy and for frequent binge drinking.[10]
Alcohol Use by Adolescents and Young Adults.
Use of alcohol by adolescents and young adults has declined over the past decade, but remains a serious
problem.[43] Among 12-17-year-olds surveyed in 1993, 18% had used alcohol in the last month, and 35%
in the last year.[4] In a separate 1993 survey, 45% and 33%, respectively, of male and female 12th graders
reported "binge" drinking (5 or more drinks on one occasion) within the previous month.[44] The leading
causes of death in adolescents and young adults -- motor vehicle and other unintentional injuries,
homicides, and suicides -- are each associated with alcohol or other drug intoxication in about half of the
cases. Driving under the influence of alcohol is more than twice as common in adolescents than in
adults.[45] Binge drinking is especially prevalent among college students: half of all men and roughly one
third of all women report heavy drinking within the previous 2 weeks.[43] ,[46] Most frequent binge
drinkers report numerous alcohol-related problems, including problems with school work, unplanned or
unsafe sex, and trouble with police.[46]
Accuracy of Screening Tests
Accurately assessing patients for drinking problems during the routine clinical encounter is difficult. The
diagnostic standard for alcohol dependence or abuse (Diagnostic and Statistical Manual of Mental
Disorders [DSM] IV)[2] requires a detailed interview and is not feasible for routine screening. Physical
findings (hepatomegaly, skin changes, etc.) are only late manifestations of prolonged, heavy alcohol
abuse.[47] Asking the patient about the quantity and frequency of alcohol use is an essential component of
assessing drinking problems, but it is not sufficiently sensitive or specific by itself for screening. In one
study, drinking 12 or more drinks a week was specific (92%) but insensitive (50%) for patients meeting
DSM criteria for an active drinking disorder.[48] The reliability of patient report is highly variable and
dependent on the patient, the clinician, and individual circumstances. Heavy drinkers may underestimate
the amount they drink because of denial, forgetfulness, or fear of the consequences of being diagnosed
with a drinking problem.
A variety of screening questionnaires have been developed which focus on consequences of drinking and
perceptions of drinking behavior. The 25-question Michigan Alcoholism Screening Test (MAST)[49] is
relatively sensitive and specific for DSM-diagnosed alcohol abuse or dependence (84-100% and 87-95%,
respectively)[49] ,[50] but it is too lengthy for routine screening. Abbreviated 10- and 13-item versions are
easier to use but are less sensitive and specific in primary care populations (66-78% and 80%,
respectively).[51] ,[52] The four-question CAGE instrument [ is the most popular screening test for use in
primary care[53] and has good sensitivity and specificity for alcohol abuse or dependence (74-89% and
79-95%, respectively) in both inpatients[54] ,[55] and outpatients.[56] [57] [58] The CAGE is less sensitive for
early problem drinking or heavy drinking, however (49-73%).[58] ,[59] Both the CAGE and MAST
questionnaires share important limitations as screening instruments in the primary care setting: an
emphasis on symptoms of dependence rather than early drinking problems, lack of information on level

and pattern of alcohol use, and failure to distinguish current from lifetime problems.[52]
Some of these weaknesses are addressed by the Alcohol Use Disorders Identification Test (AUDIT), a
10-item screening instrument developed by the World Health Organization (WHO) in conjunction with
an international intervention trial. The AUDIT incorporates questions about drinking quantity, frequency,
and binge behavior along with questions about consequences of drinking.[60] For the study population in
which it was derived, a score of 8 of 40 on the AUDIT had high sensitivity and specificity for "harmful
and hazardous drinking" (92% and 94%, respectively) as assessed by more extensive interview.[60]
Validation studies have reported more variable performance of the AUDIT. Sensitivity and specificity for
current abuse/dependence were high (96% and 96%, respectively) in an inner-city clinic;[61] among rural
outpatients, AUDIT was less sensitive and specific (61% and 90%) for current drinking problems but
superior to the Short MAST-13.[51] Because it focuses on drinking in the previous year, however, AUDIT
is less sensitive for past drinking problems.[62] Further validation studies in other populations are under
way.
Brief screening tests may be less sensitive or less specific in young persons: sensitivity of the CAGE for
problems due to alcohol among college freshmen was 42% in men and 25% in women.[63] Only 38% of
college students with an AUDIT score of 8 or greater met DSM criteria for abuse or dependence;[64]
many of these "false-positive" results were due to drinking patterns (frequent binge drinking) that would
be considered hazardous. Alternative screens have been developed for adolescents, such as the
Perceived-Benefit-of-Drinking scale[65] and the Problem Oriented Screening Instrument for Teenagers
(POSIT),[66] but they have not yet been adequately validated in the primary care setting.
Instruments that focus on alcohol dependency (e.g., CAGE or MAST) are not sensitive for levels of
drinking considered dangerous in pregnancy.[67] Women may underreport alcohol consumption while
pregnant,[68] and direct questions about drinking may provoke denial.[69] Brief instruments that
incorporate questions about tolerance to alcohol ("How many drinks does it take to make you feel high?"
or "How many drinks can you hold?") were more sensitive than the CAGE (69-79% vs. 49%) for
risk-drinking in pregnancy (2 drinks per day or greater).[30] ,[70] Women who require 3 or more drinks to
feel high, or who can drink more than 5 drinks at a time, are likely to be at risk.[71]
Laboratory tests are generally insensitive and nonspecific for problem drinking. Elevations in
hepatocellular enzymes, such as aspartate aminotransferase (AST), or the erythrocyte mean corpuscular
volume (MCV) are found in less than 10% and 30% of problem drinkers, respectively.[72] Serum
gamma-glutamyl transferase (GGT) is more sensitive (33-60%) in various studies,[54] ,[55] ,[72] but
elevations in GGT may be due to other causes (medications, trauma, diabetes, and heart, kidney, or
biliary tract disease). Even when the prevalence of problem drinking is high (30%), the predictive value
of an elevated GGT has been estimated at only 56%.[72]
Effectiveness of Early Detection
Numerous studies demonstrate that clinicians are frequently unaware of problem drinking by their
patients.[10] Early detection and intervention may alleviate ongoing medical and social problems due to
drinking and reduce the future risks from excessive alcohol use.
A number of randomized trials have now demonstrated the efficacy of brief outpatient counseling (5-15
minutes) for nondependent problem drinkers. In four Scandinavian studies, which enrolled patients with

elevated GGT and heavy alcohol consumption, brief counseling to reduce drinking and regular follow-up
produced significant improvements (decreased GGT and/or decreased alcohol consumption) in treated
versus control subjects;[73] [74] [75] [76] counseling reduced reported sick days in one study.[74] In the longest
of these studies, patients receiving counseling had fewer hospitalizations and 50% lower mortality after 5
years.[73] Some of this benefit, however, may have been due to the close medical follow-up (every 1-3
months) in the intervention group rather than the initial counseling.
Additional trials have demonstrated that brief interventions can reduce alcohol consumption in problem
drinkers identified by screening questionnaires or self-reported heavy drinking.[77] [78] [79] Most recently,
an international WHO study examined the effects of 5 or 20 minutes of counseling about drinking in
1,500 "at-risk" male and female drinkers: >35 drinks per week for men; >21 drinks per week for women;
or intoxicated twice per month; or self-perceived drinking problem.[80] After 9 months, self-reported
alcohol consumption among men was reduced 32-38% in the intervention groups and 10% in controls.
Among women, alcohol consumption declined significantly (>30%) among both treated and control
groups. A meta-analysis of six brief-intervention trials estimated that interventions reduced average
alcohol consumption by 24%.[81] Although self-reported consumption may be subject to bias, reported
changes in drinking correlated with objective measures (GGT, blood pressure) in most studies. Two
additional studies demonstrated significant reductions in blood pressure as a result of advice to stop
drinking or substitution of nonalcoholic beer.[82] ,[83]
Pregnancy.
There are no definitive controlled trials of treatments for excessive drinking in pregnancy.[84] In several
uncontrolled studies, a majority of heavy-drinking pregnant women who received counseling reduced
alcohol consumption,[32] ,[85] ,[86] and reductions in drinking were associated with lower rates of FAS.[32]
,[86] Many women spontaneously reduce their drinking while pregnant, however, and women who
continue to drink differ in many respects from women who cut down (e.g., heavier drinking, poorer
prenatal care and nutrition). As a result, it is difficult to determine precisely the benefit of screening and
counseling during pregnancy. In two trials that employed a control group, the proportions of women
abstaining or reducing consumption were similar in intervention and control groups.[87] ,[88]
Adolescents.
A 1990 Institute of Medicine (IOM) report concluded that specific recommendations for the treatment of
alcohol problems in young persons were impossible, due to disagreement over what constitutes a
drinking problem in adolescents, the wide variety of interventions employed, and the absence of any
rigorous evaluation of different treatments.[1] Alcohol interventions in adolescents have focused on
primary prevention of alcohol use.[10] Recent reviews of school-based programs found that most effects
were inconsistent, small, and short-lived; programs that sought to develop social skills to resist drug use
seem to be more effective than programs that emphasize factual knowledge.[89] ,[90]
Alcohol-Dependent Patients.
Patients with alcohol dependence usually receive more intensive treatment. A 1989 report of the IOM[91]
reviewed a variety of alcohol treatment modalities and concluded that various treatments were effective,
but there was no single superior treatment for all patients, and few treatments were effective for the
majority of patients. They found no evidence that residential versus nonresidential programs, or

long-versus short-duration programs, were more effective for the average patient, and no studies existed
that adequately evaluated the independent effect of Alcoholics Anonymous (AA). In a subsequent trial
among employees referred for alcohol problems, patients who received inpatient treatment and
mandatory AA follow-up were more likely to be abstinent at 2-year follow-up (37% vs. 16%) than
patients assigned to mandatory AA only.[92]
Two short-term (12 weeks) randomized trials demonstrated a significant benefit of naltrexone, an opioid
antagonist, as an adjunct to treatment of alcohol dependence. In one study, patients receiving naltrexone
and supportive psychotherapy had significantly higher abstinence rates than did subjects receiving
placebo (61% vs. 19%).[93] In the second, men receiving naltrexone reported less alcohol craving and
fewer drinking days than placebo-treated men.[94] In both trials, naltrexone significantly reduced the
likelihood of relapse (heavy drinking or steady drinking) among subjects who did not achieve complete
abstinence. The benefits of alcohol-sensitizing agents, however, remain uncertain.[10] Disulfiram (i.e.,
Antabuse) did not improve long-term abstinence rates in a controlled trial, but it did reduce drinking days
among patients receiving the highest dose.[95]
In a 10-year follow-up of 158 patients completing inpatient treatment, 61% reported complete or stable
remission of alcoholism.[96] Completing an extended inpatient program was associated with significantly
lower mortality among alcoholic patients in a second study.[97] Many such studies of alcohol treatment,
however, suffer from important methodologic limitations: inadequate control groups, insufficient or
selective follow-up, and selection bias due to the characteristics of patients who successfully complete
voluntary treatment programs.[91] ,[98] ,[99] Since spontaneous remission occurs in as many as 30% of
alcoholics,[100] ,[101] reduced consumption may be inappropriately attributed to treatment. Successful
treatment is likely to represent a complex interaction of patient motivation, treatment characteristics, and
the posttreatment environment (family support, stress, etc.).[1] ,[10] The IOM review concluded that
treatment of other life problems (e.g., with antidepressant medication, family or marital therapy, or stress
management) and empathetic therapists were likely to improve treatment outcomes.[91]
Recommendations of Other Groups
There is a consensus among professional groups such as the American Medical Association (AMA)[102]
and the American Academy of Family Physicians (AAFP)[103] that clinicians should be alert to the signs
and symptoms of alcohol abuse and should routinely discuss patterns of alcohol use with all patients.
AAFP recommendations are under review. The Canadian Task Force on the Periodic Health Examination
(CTF)[104] and a 1990 IOM panel[1] recommended screening adults for problem drinking, using patient
inquiry or standardized instruments, and offering brief counseling to nondependent problem drinkers.
The American Academy of Pediatrics (AAP),[105] AMA Guidelines for Adolescent Preventive Services
(GAPS),[106] the Bright Futures guidelines,[107] and the AAFP[103] all recommend careful discussion with
all adolescents regarding alcohol use and regular advice to abstain from alcohol. The AAP also advises
physicians to counsel parents regarding their own use of alcohol in the home. Recommendations of the
U.S. Surgeon General,[108] the American College of Obstetricians and Gynecologists,[109] and the AAP[109]
,[110] advise counseling all women who are pregnant or planning pregnancy that drinking can be harmful
to the fetus and that abstinence is the safest policy. The CTF recommends that all women be screened for
problem drinking and advised to reduce alcohol use during pregnancy.[104]
Several organizations have made recommendations about "safe" levels of alcohol consumption for
nonpregnant adults. The National Institute on Alcohol Abuse and Alcoholism,[111] the U.S. Surgeon

General,[112] and dietary guidelines produced jointly by the U.S. Departments of Health and Human
Services and Agriculture[113] ,[114] recommend no more than 2 drinks per day for men and 1 drink per day
for nonpregnant women. Slightly higher limits were proposed by national health authorities in the
U.K.[115]
Discussion
Alcohol problems are common in the primary care setting, but they often go undetected by clinicians.
Although imperfect, asking patients direct questions about the quantity, frequency, and pattern of their
drinking is an important way to identify those who are most likely to experience problems due to alcohol.
Questions about tolerance to the effects of alcohol may circumvent denial among pregnant women and
heavy drinkers. The CAGE and other brief screening instruments are useful supplements to the standard
patient history, but they may be less sensitive for early problems and hazardous drinking. The AUDIT
may detect a broader range of current drinking problems, but its performance in the primary care setting
needs further evaluation. Although laboratory tests such as GGT are not sufficiently sensitive or specific
for routine screening, they may be useful in selected high-risk patients to confirm clinical suspicion or to
motivate changes in drinking. Neither questionnaires nor laboratory tests should be considered diagnostic
of problem drinking without more detailed evaluation (see Clinical Intervention ).
Detecting early problem drinkers is important, because they account for a large proportion of all alcohol
problems and they are more likely to respond to simple interventions than patients with alcohol
dependency. There is now good evidence that brief counseling can reduce alcohol consumption in
problem drinkers, and several trials have also reported improved clinical outcomes. Since the risks from
alcohol rise steadily at higher levels of consumption, reducing drinking should also benefit heavy
drinkers (i.e., hazardous drinkers) who do not yet manifest problems due to drinking. Early attention to
problem drinking is especially important in young adults: hazardous drinking is common, adverse effects
of alcohol increase with duration of use, and few persons initiate drinking after age 30.[116] Early
detection is also important for alcohol-dependent patients, but effective treatment requires more intensive
and sustained efforts to promote abstinence.
Uncertainties remain about optimal screening methods and interventions during pregnancy, but screening
is justified by the strong evidence of the adverse effects of alcohol on the fetus. Although the risks of
occasional, light drinking during pregnancy have not been established, abstinence can be recommended
as a prudent approach for pregnant women. At the same time, women concerned about the effects of
previous moderate drinking early in pregnancy can be reassured that important harms have not been
demonstrated from such limited exposures. Because exposure early in pregnancy may be most important,
screening and advice should be directed toward women contemplating pregnancy and those at risk for
unintended pregnancy, not just women who are already pregnant.
There is insufficient evidence to make precise recommendations about desirable levels of drinking, but
the strong association between heavy alcohol use and risk of future complications justifies advising all
drinkers to drink moderately and avoid frequent intoxication, even in the absence of current problems
(see below).
Screening to detect problem drinking and hazardous drinking is recommended for all adult and
adolescent patients ("B" recommendation). Screening should involve a careful history of alcohol

use and/or the use of standardized screening questionnaires. Patients should be asked to describe
the quantity, frequency, and other characteristics of their use of wine, beer, and liquor, including
frequency of intoxication and tolerance to the effects of alcohol. One drink is defined as 12 ounces
of beer, a 5-ounce glass of wine, or 1.5 fluid ounces (one jigger) of distilled spirits. Brief
questionnaires such as the CAGE or AUDIT may help clinicians assess the likelihood of problem
drinking or hazardous drinking (see Table 52.1 ). Responses suggestive of problem drinking should
be confirmed with more extensive discussions with the patient (and family members where
indicated) about patterns of use, problems related to drinking, and symptoms of alcohol
dependence.[2] Routine measurement of biochemical markers, such as serum GGT, is not
recommended for screening purposes. Discussions with adolescents should be approached with
discretion to establish a trusting relationship and to respect the patient's concerns about the
confidentiality of disclosed information.
All pregnant women should be screened for evidence of problem drinking or risk drinking (2
drinks per day or binge drinking) ("B" recommendation). Including questions about tolerance to
alcohol may improve detection of at-risk women. All pregnant women and women contemplating
pregnancy should be informed of the harmful effects of alcohol on the fetus and advised to limit or
cease drinking. Although there is insufficient evidence to prove or disprove harms from occasional,
light drinking during pregnancy, abstinence from alcohol can be recommended on other grounds:
possible risk from even low-level exposure to alcohol, lack of harm from abstaining, and prevailing
expert opinion ("C" recommendation). Women who smoke should be advised that the risk of low
birth weight is greatest for mothers who both smoke and drink.
Patients with evidence of alcohol dependence should be referred, where possible, to appropriate
clinical specialists or community programs specializing in the treatment of alcohol dependence.
Patients with evidence of alcohol abuse or hazardous drinking should be offered brief advice and
counseling. Counseling should involve feedback of the evidence of a drinking problem, discussion
of the role of alcohol in current medical or psychosocial problems, direct advice to reduce
consumption, and plans for regular follow-up. Problems related to alcohol (e.g., physical
symptoms, behavioral or mood problems, or difficulties at work and home) should be monitored to
determine whether further interventions are needed. There is no single definition of "hazardous"
drinking in asymptomatic persons, but successful intervention trials have generally defined 5
drinks per day in men, 3 drinks per day in women, or frequent intoxication to identify persons at
risk. Several U.S. organizations have suggested lower limits for "safe" drinking: 2 drinks per day
in men and 1 drink per day in women.[18] All persons who drink should be informed of the dangers
of driving or other potentially dangerous activities after drinking (see Chapter 57). The use of
alcohol should be discouraged in persons younger than the legal age for drinking ("B"
recommendation), although the effectiveness of alcohol abstinence messages in the primary care
setting is uncertain.
TABLE 52.1 -- AUDIT Structured Interview [

Score
Question 0 1 2 3 4

How often do you have a drink Monthly or 4 or more

Never 2-4 times/mo 2-3 times/wk
containing alcohol? less times/wk
How many drinks do you have
on a typical day when you are None 1 or 2 3 or 4 5 or 6 7-9 [
drinking?
How often do you have 6 or Less than Daily or
Never Monthly Weekly
more drinks on one occasion? monthly almost daily
How often during the last year
have you found that you were Less than Daily or
unable to stop drinking once monthly almost daily
you had started?
have you failed to do what was Less than Daily or
normally expected from you monthly almost daily
because of drinking?
have you needed a first drink in
Less than Daily or
the morning to get yourself Never Monthly Weekly
monthly almost daily
going after a heavy drinking
session?
Less than Daily or
have you had a feeling of guilt Never Monthly Weekly
or remorse after drinking?
have you been unable to
Less than Daily or
remember what happened the Never Monthly Weekly
night before because you had
been drinking?
Have you or someone else been
Yes, but not in last year (2 Yes, during the last year (4
injured as a result of your Never
points) points)
drinking?
Has a relative, doctor, or other
health worker been concerned Yes, but not in last year (2 Yes, during the last year (4
Never
about your drinking or points) points)
suggested you cut down?
[a] Score of greater than 8 (out of 41) suggests problem drinking and indicates need for more in-depth assessment. Cut-off
of 10 points recommended by some to provide greater specificity.
[*] 5 points if response is 10 or more drinks on a typical day.

The draft update of this chapter was prepared for the U.S. Preventive Services Task Force by David
Atkins, MD, MPH, with contributions from materials prepared for the Canadian Task Force on the
Periodic Health Examination by Deborah L. Craig, MPH, and Jean L. Haggerty, MSc.
REFERENCES
1. Institute of Medicine.
Broadening the base of treatment for alcohol problems: report of a study by a committee of the Institute of Medicine,
Division of Mental Health and Behavioral Medicine.Washington, DC: National Academy Press, 1990.
2. American Psychiatric Association.

Diagnostic and statistical manual of mental disorders. 4th ed.Washington, DC: American Psychiatric Association, 1994.
3. World Health Organization.

International classification of diseases, 10th revision.Geneva: World Health Organization, 1992.
4. Substance Abuse and Mental Health Administration.

National Household Survey on Drug Abuse: population estimates 1993. Rockville, MD: U.S. Department of Health and
Human Services, Substance Abuse and Mental Health Administration, 1994. (Publication no. (SMA) 94-3017.)
5. Grant
BF, Harford TC, Chou P, et al.
Epidemiologic Bulletin no.27: Prevalence of DSM-III-R alcohol abuse and dependence: United States, 1988.Alcohol
Health Res World 1991;15:91-96.
6. Helzer JE, Burnam A, McEvoy LT.

Alcohol abuse and dependence.In: Robins LN, Regier DA, eds. Psychiatric disorders in America: the Epidemiologic
Catchment Area Study.New York:Free Press, 1991:81-115.
7. Grant BF, Harford TC, Dawson D, et al.

Prevalence of DSM-IV alcohol abuse and dependence: United States, 1992.Alcohol Health Res World 1994;18:243-248.
8. KesslerRC, McGonagle KA, Zhao S, et al.

Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in the United States: results from the National
Comorbidity Survey.Arch Gen Psychiatry 1994;51:8-19.
9. Bradley KA.
The primary care provider's role in the prevention and management of alcohol problems.Alcohol Health Res World
1994;18:97-104.
10. National
Alcohol and health: eighth special report to the U.S. Congress from the Secretary of Health and Human Services.
Rockville, MD: U.S.Department of Health and Human Services, National Institutes of Health, 1993.(NIH Publication no.
94-3699.)
11. Klatsky
AL, Armstrong AA, Friedman GD.
Alcohol and mortality.Ann Intern Med 1992;117:646-654.
12. Boffetta
P, Garfinkel L.
Alcohol drinking and mortality among men enrolled in an American Cancer Society prospective study.Epidemiology

1990;1:342-348.
13. Anderson P, Cremona A, Paton A, et al.

The risk of alcohol.Addiction 1993;88:1493-1508.
14. RosenbergL, Metzger LS, Palmer JR.

Alcohol consumption and risk of breast cancer: a review of the epidemiologic evidence.Epidemiol Rev 1993;15:133-168.
15. FuchsCS, Stampfer MJ, Colditz GA, et al.

Alcohol consumption and mortality among women.N Engl J Med 1995;332:1245-1250.
16. Centers
Alcohol-related mortality and years of potential life lost -- United States, 1987.MMWR 1990;39:173-178.
17. National Highway Traffic Safety Administration.

Traffic safety facts 1993: alcohol.Washington, DC: U.S. Department of Transportation, National Highway Traffic Safety
Administration, National Center for Statistics and Analysis, 1994.
18. Bradley
KA, Donovan DM, Larson EB.
How much is too much? Advising patients about safe levels of alcohol consumption.Arch Intern Med
1993;153:2734-2740.
19. Stampfer
MJ, Rimm EB, Walsh DC.
Commentary: alcohol, the heart, and public policy.Am J Public Health 1993;83:801-804.
20. Maclure
M.
Demonstration of deductive meta-analysis: ethanol intake and risk of myocardial infarction.Epidemiol Rev
1993;15:328-351.
21. KlatskyAL, Armstrong AA, Friedman GD.

Risk of cardiovascular mortality in alcohol drinkers, ex-drinkers and non-drinkers.Am J Cardiol 1990;66:1237-1242.
22. StampferMJ, Colditz GA, Willett WC, et al.

A prospective study of moderate alcohol consumption and the risk of coronary disease and stroke in women.N Engl J Med
1988;319:267-273.
23. Gaziano JM, Buring JE, Breslow JL, et al.

Moderate alcohol intake, increased levels of high-density lipoprotein and its subfractions, and decreased risk of myocardial
infarction.N Engl J Med 1993;329:1829-1835.
24. Ridker PM, Vaughan DE, Stampfer MJ, et al.

Association of moderate alcohol consumption and plasma concentration of endogenous tissue-type plasminogen
activator.JAMA 1994;272:929-933.
25. Sher
KJ, ed.
Children of alcoholics: a critical appraisal of theory and research.Chicago: University of Chicago Press, 1991.
26. SerdulaM, Williamson DF, Kendrick J, et al.

Trends in alcohol consumption by pregnant women: 1985 through 1988.JAMA 1991;265:876-879.
27. Goodwin MM, Bruce C, Zahniser SC, et al.

Behavioral risk factors before and during pregnancy.In: Wilcox LS, Marks JS, eds. From data to action: CDC's public
health surveillance for women, infants, and children. Washington, DC:Department of Health and Human Services,

1994:93-104.

Frequent alcohol consumption among women of childbearing age -- Behavioral Risk Factor Surveillance System,
1991.MMWR 1994;43:328-329, 335.
29. Centers
Sociodemographic and behavioral characteristics associated with alcohol consumption during pregnancy -- United States,
1988.MMWR 1995;44:261-264.
30. Sokol
RJ, Martier SS, Ager JW.
The T-ACE questions: practical prenatal detection of risk-drinking.Am J Obstet Gynecol 1989;160:863-870.
31. Russell
Screening for pregnancy risk-drinking.Alcohol Clin Exp Res 1994;18:1156-1161.
32. Rosett
HL, Weiner L, Edelin KC.
Treatment experience with pregnant problem drinkers.JAMA 1983;249:2029-2033.
33. AbelEA, Sokol RJ.

A revised conservative estimate of the incidence of FAS and its economic impact.Alcohol Clin Exp Res 1991;15:514-524.

Fetal alcohol syndrome -- United States, 1979-1992.MMWR 1993;42:339-341.
35. RussellM.
Clinical implications of recent research on the fetal alcohol syndrome.Bull NY Acad Sci 1991;67:207-222.
36. BurdL, Moffatt ME.

Epidemiology of fetal alcohol syndrome in American Indians, Alaskan Natives, and Canadian Aboriginal peoples: a
review of the literature.Public Health Rep 1994;109:688-693.
37. Virji
SK.
The relationship between alcohol consumption during pregnancy and infant birthweight: an epidemiologic study.Acta
Obstet Gynecol Scand 1991;303-308.
38. Forrest
F, Florey CDV, Taylor D, et al.
Reported social alcohol consumption during pregnancy and infants' development at 18 months.BMJ 1991;303:22-26.
39. Verkerk PH, Noord-Zaadstra BMV, Florey CDV, et al.

The effect of moderate maternal alcohol consumption on birth weight and gestational age in a low risk population.Early
Hum Dev 1993;32:121-129.
40. JacobsonJL, Jacobson SW, Sokol RJ, et al.

Teratogenic effects of alcohol on infant development.Alcohol Clin Exp Res 1993;17:174-183.
41. Streissguth
AP, Barr HM, Sampson PD.
Moderate prenatal alcohol exposure: effects on child IQ and learning problems at age 7.5 years.Alcohol Clin Exp Res
1990;14:662-669.
42. KnupferG.
Abstaining for fetal health: the fiction that even light drinking is dangerous.Br J Addict 1991;86:1063-1073.

43. NationalInstitute on Drug Abuse.

National survey results on drug use from the Monitoring the Future Study, 1975-1992. Vols 1 and 2. Rockville, MD:
Department of Health and Human Services, National Institute on Drug Abuse, 1993. (NIH Publications nos. 93-3596 and
3597.)
44. Centers
Youth Risk Behavior Surveillance -- United States, 1993. MMWR 1995;44(SS-1):1-56.
45. Centers
Drinking and driving and binge drinking in selected states, 1982 and 1985: the Behavioral Risk Factor Surveys. MMWR
1987;35:788-791.
46. WechslerH, Davenport A, Dowdall G, et al.

Health and behavioral consequences of binge drinking in college: a national survey of students at 140 campuses.JAMA
1994;272:1672-1677.
47. GlazeLW, Coggan PG.

Efficacy of an alcoholism self-report questionnaire in a residency clinic.J Fam Pract 1987;25:60-64.
48. Buchsbaum DG, Welsh J, Buchanan RG, et al.

Screening for drinking problems by patient self-report: even "safe" levels may indicate a problem.Arch Intern Med
1995;155:104-108.
49. Selzer
ML.
The Michigan Alcoholism Screening Test: the quest for a new diagnostic instrument.Am J Psychiatry
1971;127:1653-1658.
50. PokornyAD, Miller BA, Kaplan HB.

The Brief MAST: a shortened version of the Michigan Alcoholism Screening Test.Am J Psychiatry 1972;129:342-345.
51. Barry
K, Fleming M.
The Alcohol Use Disorders Identification Test (AUDIT) and the SMAST-13: predictive validity in a rural primary care
sample.Alcohol Alcohol 1993;28:33-42.
52. Chan AWK, Pristach EA, Welte JW, et al.

Use of the TWEAK test in screening for alcoholism/heavy drinking in three populations.Alcohol Clin Exp Res
1993;17:1188-1192.
53. EwingJA.
Detecting alcoholism: the CAGE questionnaire.JAMA 1984;252:1905-1907.
54. Bernadt
MW, Mumford J, Taylor C, et al.
Comparison of questionnaire and laboratory tests in the detection of excessive drinking and alcoholism.Lancet
1982;1:325-328.
55. BushB, Shaw S, Cleary P, et al.

Screening for alcohol abuse using the CAGE questionnaire.Am J Med 1987;82:231-235.
56. King M.
At risk drinking among general practice attenders: validation of the CAGE questionnaire.Psychol Med 1986;16:213-217.
57. Buchsbaum DG, Buchanan RG, Centor RM, et al.

Screening for alcohol abuse using CAGE scores and likelihood ratios.Ann Intern Med 1991;115:774-777.
58. ChanAWK, Pristach EA, Welte JW.

Detection by the CAGE of alcoholism or heavy drinking in primary care outpatients and the general population.J Subst
Abuse 1994;6:123-135.
59. HaysJT, Spickard WA.

Alcoholism: early diagnosis and intervention.J Gen Intern Med 1987;2:420-427.
60. SaundersJB, Aasland OG, Babor TF, et al.

Development of the Alcohol Use Disorders Identification Test (AUDIT): WHO collaborative project on early detection of
persons with harmful alcohol consumption --II.Addiction 1993;88:791-804.
61. Isaacson
JH, Butler R, Zacharek M, et al.
Screening with the Alcohol Use Disorders Identification Test (AUDIT) in an inner-city population.J Gen Intern Med
1994;9:550-553.
62. SchmidtA, Barry K, Fleming MF.

Detection of problem drinkers: the Alcohol Use Disorders Identification Test (AUDIT).Southern Med J 1995;88:52-59.
63. WernerMJ, Walker LS, Greene JW.

Screening for problem drinking among college freshmen.J Adolesc Health 1994;15:303-310.
64. Fleming
MF, Barry KL, MacDonald R.
The Alcohol Use Disorders Identification Test (AUDIT) in a college sample.Int J Addict 1991;26:1173-1185.
65. Petchers
MK, Singer MI.
Perceived-benefit-of-drinking Scale: approach to screening for adolescent alcohol abuse.J Pediatr 1987;110:979-981.
66. National
Institute on Drug Abuse.
Problem oriented screening instrument for teenagers.In: Rahdert E, ed. The adolescent assessment/referral system
manual.Rockville, MD:National Institute on Drug Abuse, 1990:5-9.
67. WatersonEJ, Murray-Lyon IM.

Screening for alcohol-related problems in the antenatal clinic.Alcohol Alcohol 1989;24:21-30.
68. JacobsonSW, Jacobson JL, Sokol RJ, et al.

Maternal recall of alcohol, cocaine, and marijuana use during pregnancy.Neurotoxicol Teratol 1991;13:535-540.
69. Ernhart
CB, Morrow-Tlucak M, Sokol RJ, et al.
Underreporting of alcohol use in pregnancy.Alcohol Clin Exp Res 1988;12:506-511.
70. Russell
Screening for pregnancy risk-drinking.Alcohol Exp Clin Res 1994;18:1156-1161.
71. Russell
M.
New assessment tools for risk drinking during pregnancy: T-ACE, TWEAK, and others.Alcohol Health Res World
1994;18:55-61.
72. Hoeksema HL, de Bock GH.

The value of laboratory tests for the screening and recognition of alcohol abuse in primary care patients.J Fam Pract
1993;37:268-276.

73. Kristenson
H, Trell E, Hood B.
Serum glutamyl-transferase in screening and continuous control of heavy drinking in middle-aged men.Am J Epidemiol
1981;114:862-872.
74. PerssonJ, Magnusson PH.

Early intervention in patients with excessive consumption of alcohol: a controlled study.Alcohol 1989;6:403-408.
75. Romelsjo
A, Andersson L, Barrner H, et al.
A randomized study of secondary prevention of early stage problem drinkers in primary health care.Br J Addict
1989;84:1319-1327.
76. Nilssen
O.
The Tromso Study: identification of and a controlled intervention on a population of early-stage risk drinkers.Prev Med
1991;20:518-528.
77. ChickJ, Lloyd G, Crombie E.

Counselling problem drinkers in medical wards: a controlled study.BMJ 1985;290:965-967.
78. Wallace
P, Cutler S, Haines A.
Randomized controlled trial of general practitioner intervention in patients with excessive alcohol consumption.BMJ
1988;297:663-668.
79. HeatherN, Kissoon-Singh J, Fenton GW.

Assisted natural recovery from alcohol problems: effects of a self-help manual with and without supplementary telephone
contact.Br J Addict 1990;85:1177-1185.
80. Babor TF, Grant M, eds.

Project on identification and management of alcohol-related problems. Report on Phase II: a randomized clinical trial of
brief interventions in primary health care.Geneva: World Health Organization, 1992.
81. Brief
interventions and alcohol use.
Effective Health Care, bulletin no. 7. Leeds, UK: Nuffield Institute for Health Care, University of Leeds, 1993.
82. Maheswaran R, Beevers M, Beevers DG.

Effectiveness of advice to reduce alcohol consumption in hypertensive patients.Hypertension 1992:19:79-84.
83. Cox KL, Puddey IB, Morton AR, et al.

The combined effects of aerobic exercise and alcohol restriction on blood pressure and serum lipids: a two-way factorial
study in sedentary men.J Hypertens 1993;11:191-201.
84. Schorling
JB.
The prevention of prenatal alcohol use: a critical analysis of intervention studies.J Stud Alcohol 1993;54:261-267.
85. LarsonG.
Prevention of fetal alcohol effects.Acta Obstet Gynecol Scand 1983;62:171-178.
86. Halmesmaki E.
Alcohol counselling of 85 pregnant problem drinkers: effect on drinking and fetal outcome.Br J Obstet Gynaecol
1988;95:243-247.
87. Waterson EJ, Murray-Lyon IM.

Preventing fetal alcohol effects: a trial of three methods of giving information in the antenatal clinic.Health Educ Res

1990;5:53-61.
88. MebergA, Halvorsen B, Holter B, et al.

Moderate alcohol consumption -- need for intervention programs in pregnancy?Acta Obstet Gynecol Scand
1986;65:861-864.
89. Ennett
ST, Tobler NS, Ringwalt CL, et al.
How effective is drug abuse resistance education? A meta-analysis of project DARE outcome evaluations.Am J Public
Health 1994;84:1394-1401.
90. HansenWB.
School-based substance abuse prevention: a review of the state of the art in curriculum, 1980-1990. Health Educ Res
1992;7:403-430.
91. Institute
of Medicine.
Prevention and treatment of alcohol problems: research opportunities.Washington, DC: National Academy Press, 1989.
92. Walsh
DC, Hingson RW, Merrigan DM, et al.
A randomized trial of treatment options for alcohol-abusing workers.N Engl J Med 1991;325:775-782.
93. O'MalleySS, Jaffe AJ, Chang G, et al.

Naltrexone and coping skills therapy for alcohol dependence: a controlled study.Arch Gen Psychiatry 1992;49:881-887.
94. Volpicelli
JR, Alterman AI, Hayashida M, et al.
Naltrexone in the treatment of alcohol dependence.Arch Gen Psychiatry 1992;49:876-880.
95. Fuller
RK, Branchey L, Brightwell DR, et al.
Disulfiram treatment of alcoholism: A Veterans Administration study.JAMA 1986;256:1449-1455.
96. Cross
GM, Morgan CW, Mooney AJ, et al.
Alcoholism treatment: a ten-year follow-up study.Alcohol Clin Exp Res 1990;14:169-173.
97. Bunn JY, Booth BM, Loveland Cook CA, et al.

The relationship between mortality and intensity of inpatient alcoholism treatment.Am J Public Health 1994;84:211-214.
98. Thurstin
AH, Alfano AM, Sherer M.
Pretreatment MMPI profiles of A.A. members and nonmembers.J Stud Alcohol 1986;47:468-471.
99. EmrickCD.
Alcoholics Anonymous: affiliation processes and effectiveness as treatment.Alcohol Clin Exp Res 1987;11:416-423.
100. Smart
R.
Spontaneous recovery in alcoholics: a review and analysis of the available research.Drug Alcohol Depend 1975/76;1:277.
101. Saunders
WM, Kershaw PW.
Spontaneous remission from alcoholism: a community study.Br J Addict 1979;74:251-265.
102. American Medical Association, Council on Scientific Affairs.

Guidelines for alcoholism diagnosis, treatment and referral.Chicago: American Medical Association, 1979.
103. American Academy of Family Physicians.

Age charts for periodic health examination. Kansas City, MO: American Academy of Family Physicians, 1994.(Reprint

no. 510.)
104. Canadian
Task Force on the Periodic Health Examination.
Canadian guide to clinical preventive health care.Ottawa: Canada Communication Group, 1994:52-61, 826-836.
105. AmericanAcademy of Pediatrics, Committee on Adolescence.

Alcohol use and abuse: a pediatric concern.Pediatrics 1995;95:439-442.
106. American Medical Association.

Guidelines for adolescent preventive services (GAPS): recommendations and rationale.Chicago: American Medical
Association, 1994.
107. GreenM, ed.

Bright Futures: guidelines for health supervision of infants, children and adolescents. Arlington VA: National Center for
Education in Maternal and Child Health, 1994.
108. Surgeon
General's Advisory on alcohol and pregnancy.
FDA Drug Bull 1981;11:9-10.
109. American Academy of Pediatrics and American College of Obstetricians and Gynecologists.
Guidelines for perinatal care. 3rd ed.Washington, DC: American Academy of Pediatrics and American College of
Obstetricians and Gynecologists, 1992.
110. American Academy of Pediatrics, Committee on Substance Abuse and Committee on Children with Disabilities.
Fetal alcohol syndrome and fetal alcohol effects.Pediatrics 1993;91:1004-1006.
111. National
Moderate drinking. Alcohol Alert no. 16. Bethesda: U.S.Department of Health and Human Services, 1992.
112. U.S.
Department of Health and Human Services.
The Surgeon General's report on nutrition and health.Washington, DC: Government Printing Office, 1988.(Publication no.
PHS-88-50210.)
113. U.S.Department of Agriculture and U.S. Department of Health and Human Services.
Dietary guidelines for Americans.Washington, DC: Department of Agriculture, 1990.
114. DietaryGuidelines Advisory Committee.

Report of the Dietary Guidelines Advisory Committee on the dietary guidelines for Americans, 1995, to the Secretary of
Health and Human Services and the Secretary of Agriculture. Washington, DC: U.S. Department of Agriculture, 1995.
115. Secretary
of State for Health.
The Health of the Nation: a strategy for health in England.London: Her Majesty's Stationery Office, 1992.
116. ChenK, Kandel D.

The natural history of drug use from adolescence to mid-thirties in a general population sample.Am J Public Health
1995;85:41-47.
[a] C: "Have you ever felt you ought to Cut down on drinking?" A: "Have people Annoyed you by criticizing your
drinking?" G: "Have you ever felt bad or Guilty about your drinking?" E: "Have you ever had a drink first thing in the
morning to steady your nerves or get rid of a hangover (Eye opener)?

Bookmark URL: /das/guideline/view/N/10134884?ja=104571&PAGE=1.html&ANCHOR=top

MD Consult - Patient Education
April 21, 2000
Conde Petra
Click VIEW PAGE to reformat this page before

printing.
About Patient Education Then use the print function on your browser to print
the handout.
All Topics
Conditions and Treatments
Medications
Alternative Therapies
Conditions and Treatments by Well-Connected
Category
Topics by Specialty
Adult Health Topics

Alcoholism
Pediatric Health Topics
What Is Alcoholism?
Senior Health Topics
Alcoholism is a chronic, progressive, and often fatal disease; it
Women's Health Topics
is a primary disorder and not a symptom of other diseases or
emotional problems. The chemistry of alcohol allows it to affect
nearly every type of cell in the body, including those in the
central nervous system. In the brain, alcohol interacts with
centers responsible for pleasure. After prolonged exposure to
alcohol, the brain adapts to the changes alcohol makes and
becomes dependent on it. For people with alcoholism, drinking
becomes the primary means through which they can deal with
people, work, and life. Alcohol dominates their thinking,
emotions, and actions. The severity of this disease is influenced
by factors such as genetics, psychology, culture, and response to
physical pain.
Alcoholism can develop insidiously; often there is no clear line

between problem drinking and alcoholism. In addition to
alcohol dependence, experts are now defining alcohol use by
levels of harm that it may be causing. This information is useful
to determine possible interventions at earlier stages [see Box,
below]. The only early indications of alcoholism may be the
unpleasant physical responses to withdrawal that occur during
even brief periods of abstinence. Sometimes people experience
http://home.mdconsult.com/das/patient/view/41/548...?stty=cow&nid=39982&tlt=Z&pos=&tag=relinfo&anc=pe (1 of 30) [21/04/2000 23:29:45]

long-term depression or anxiety, insomnia, chronic pain, or

personal or work stress that lead to the use of alcohol for relief,
but often no extraordinary events have occurred that account for
the drinking problem.
Alcoholics have little or no control over the quantity they drink

or the duration or frequency of their drinking. They are
preoccupied with drinking, deny their own addiction, and
continue to drink even though they are aware of the dangers.
Over time, some people become tolerant to the effects of
drinking and require more alcohol to become intoxicated,
creating the illusion that they can "hold their liquor." They have
blackouts after drinking and frequent hangovers that cause them
to miss work and other normal activities. Alcoholics might
drink alone and start early in the day. They periodically quit
drinking or switch from hard liquor to beer or wine, but these
periods rarely last. Severe alcoholics often have a history of
accidents, marital and work instability, and alcohol-related
health problems. Episodic violent and abusive incidents
involving spouses and children and a history of unexplained or
frequent accidents are often signs of drug or alcohol abuse.
Alcohol Use and Abuse
Experts now define levels of alcohol use by how harmful it is as

well as how dependent a person is on it (with a drink defined as
12-oz of beer, 6 oz of wine, or 1.5 oz of 90-proof liquor).
Moderate Drinking: equal to or less than two drinks a day for

men and equal to or less than one drink a day for women.
Heavy Drinking: more than 14 drinks per week or 4 drinks at

one sitting for men and more than seven drinks a week or three
drinks at one sitting for women. (Drinking over this amount
puts a person at risk for adverse health events.)
Hazardous Drinking: Hazardous drinking is an average

consumption of 21 drinks or more per week for men (or 7 or
more drinks per occasion at least 3 times a week) and 14 or
more drinks per week for women (or more than 5 drinks per
occasion at least 3 times a week). Hazardous drinking is
considered to place individuals at risk for adverse health events.

Harmful Drinking: Harmful drinking occurs when alcohol

consumption has actually caused physical or psychologic harm.
This is determined if there is clear evidence that alcohol is
responsible for such harm, the nature of that harm can be
identified, alcohol consumption has persisted for at least a
month or has occurred repeatedly for the past year, and the
individual is not alcohol dependent.
Alcohol Abuse: one or more of the following alcohol-related

problems over a period of one year: failure to fulfill work or
personal obligations; recurrent use in potentially dangerous
situations; problems with the law; and continued use in spite of
harm being done to social or personal relationships.
Alcohol Dependence: The individual experiences three or more

of the following alcohol-related problems over a period of one
year: increased amounts of alcohol needed to produce an effect;
withdrawal symptoms or drinking alcohol to avoid these
symptoms; drinking more over a given period than intended;
unsuccessful attempts to quit or cut down; giving up significant
leisure or work activities; continuing drinking in spite of the
knowledge of its physical or psychological harm to oneself or
others.
What Causes Alcoholism?

People have been drinking alcohol for perhaps 15,000 years.
Just drinking steadily and consistently over time can produce
dependence and cause withdrawal symptoms during periods of
abstinence; this physical dependence, however, is not the sole
cause of alcoholism. To develop alcoholism, other factors
usually come into play, including biology genetics, culture, and
psychology.
Genetic Factors
Genetic factors play a significant role in alcoholism and may
account for about half of the total risk for alcoholism. They
effect men and women equally. Researchers are investigating a
number of inherited traits that make particular individuals
susceptible to this disorder. Even if genetic factors can be
identified, however, they are unlikely to explain all cases of
alcoholism. In fact, it may be a natural lack of genetic

protection that plays a major role in alcoholism. Because
alcohol is not found easily in nature, genetic mechanisms to
protect against excessive consumption may not have evolved in
humans as they frequently have for protection against natural
threats. It is important to understand that, whether they inherit
the disorder or not, people with alcoholism are still legally
responsible for their actions. Inheriting genetic traits does not
doom a child to an alcoholic future. Environment, personality,
and emotional factors also play a strong role. [ See Who
Becomes an Alcoholic? Below.]
Brain Chemicals
Alcohol has widespread effects on the brain. Of particular
interest to researchers are a number of brain chemicals that
include gamma aminobutyric acid (GABA), dopamine,
serotonin, glutamate, norepinephrine, and opioid peptides.
Genetic factors that cause imbalances in one or more of these
brain chemicals and which may increase the risk for alcohol
dependency are under intense scrutiny. Long-term use of
alcohol itself, however, can cause adaptations that change the
brain's chemistry and cause cravings, pain on withdrawal, and
addiction. Such changes include depletion of
gamma-aminobutyric acid (which inhibits impulsivity), an
increase in glutamate (which produces an over-excited nervous
system), and higher levels of norepinephrine and corticotropin
releasing factor (which causes stress and tension). When a
person with alcoholism stops drinking, the hyperactivity in the
brain caused by these events produces an intense need to calm
down with the use of yet more alcohol. In addition, the patient
continues to seek euphoria and pleasure produced by other
chemicals. Serotonin and opioid peptides are important for
feelings of well being. A rapid release in the brain of
alcohol-induced dopamine causes euphoria, and repeated
alcohol administration increases the sensitivity to these
dopamine pathway. Animal studies indicate, however, that
heavy drinking depletes the stores of dopamine and serotonin
over time. Cravings for alcohol which lead to dependency or
relapse then appear to be produced by two effects: the need to
reduce agitation caused by an overexcited nervous system and
the desire to restore pleasurable feelings that have been reduced
by lower activity of dopamine and serotonin. One recent study
suggested that agitation may be the more important factor in
causing a relapse than mood shifts.

Levels of Response to Alcohol

A number of studies on twins and population groups at risk for
alcoholism, such as Native Americans, have suggested that
genetic factors may determine one's level of response to
alcohol. Such studies have found that people with a family
history of alcohol tend to be able to drink more to achieve a
level of intoxication than other people experience with less
liquor. In other words, they "hold their liquor" better. Experts
suggest such people may inherit a lack of those warning signals
that ordinarily make people stop drinking. Some research
suggests this factor may contribute to between 40% and 60% of
alcoholism cases related to genetic factors.
Factors Affecting Aldehyde or Alcohol

Dehydrogenase
Alcohol is metabolized in a two-stage process. It is first
converted to acetaldehyde by a chemical called alcohol
dehydrogenase (ADH), which is then converted into acetate by
another chemical called aldehyde dehydrogenase (ALDH).
Genes that affect either ADH or ALDH are under intense
scrutiny for either their protective or harmful roles or both. For
example, many Asians and possibly many Jewish people, are
less likely to become alcoholic because of a genetic factor that
makes them deficient in ALDH. In its absence, acetate, which is
toxic, builds up after drinking alcohol and rapidly leads to
flushing, dizziness, and nausea. People with this genetic
susceptibility, then, are likely to experience adverse reactions to
alcohol and therefore more likely not to become alcoholic. (This
deficiency is not completely protective against drinking,
however, particularly if there is added social pressure, such as
among college fraternity members.) Investigators are looking
for variants in the ALDH gene that may increase production of
this enzyme, which in turn may make people genetically
susceptible to alcoholism.
Social and Emotional Causes of Alcoholic Relapse
Between 80% and 90% of people treated for alcoholism relapse,

even after years of abstinence. Patients and their caregivers
should understand that relapses of alcoholism are analogous to

recurrent flare-ups of chronic physical diseases. One study

found that three factors placed a person at high risk for relapse:
frustration and anger, social pressure, and internal temptation.
Another study suggests that impaired sleep is also an important
predictor of relapse.
Mental and Emotional Stress. Alcohol blocks out emotional

pain and is often perceived as a loyal friend when human
relationships fail. It is also associated with freedom and a loss
of inhibition that offsets the tedium of daily routines. When the
alcoholic tries to quit drinking, the brain seeks to restore what it
perceives to be its equilibrium. The brain's best weapons against
abstinence are depression and anxiety (the emotional
equivalents of physical pain) that continue to tempt alcoholics
to return to drinking long after physical withdrawal symptoms
have abated. Even intelligence is no ally in this process, for the
brain will use all its powers of rationalization to persuade the
patient to return to drinking. According to an interesting 1999
study, however, although a person's IQ had no effect on
abstinence, a high verbal ability appeared to aid the alcoholic in
remaining sober. It is important to realize that any life change
may cause temporary grief and anxiety, even changes for the
better. With time and the substitution of healthier pleasures, this
emotional turmoil weakens and can be overcome.
Codependency. One of the most difficult problems facing a

person with alcoholism is being around people who are able to
drink socially without danger of addiction. A sense of isolation,
a loss of enjoyment, and the ex-drinker's belief that pity, not
respect, is guiding a friend's attitude can lead to loneliness, low
self-esteem, and a strong desire to drink. Close friends and even
intimate partners may have difficulty in changing their
responses to this newly sober person and, even worse, may
encourage a return to drinking. To preserve marriages to
alcoholics, spouses often build their own self-images on
surviving or handling their mates' difficult behavior and then
discover that they are threatened by abstinence. Friends may not
easily accept the sober, perhaps more subdued, comrade. In
such cases, separation from these "enablers" may be necessary
for survival. It is no wonder that, when faced with such losses,
even if they are temporary, a person returns to drinking. The
best course in these cases is to encourage close friends and
family members to seek help as well. Fortunately, groups such
as Al-Anon exist for this purpose.

Social and Cultural Pressures. The media portrays the

pleasures of drinking in advertising and programming. The
medical benefits of light to moderate drinking are frequently
publicized, giving ex-drinkers the spurious excuse of returning
to alcohol for their health. These messages must be
categorically ignored and acknowledged for what they are, an
industry's attempt to profit from potentially great harm to
individuals.
Who Becomes An Alcoholic?

General Risks and Age
Some population studies indicate that in a single year, between
7.4% and 9.7% of the population are dependent on alcohol, and
between 20% and 25% of Americans identify themselves as
heavy drinkers when they visit their physician. A 1996 national
survey reported that 11 million Americans are heavy drinkers
and 32 million engaged in binge drinking (five or more drinks
on one occasion) in the month previous to the survey. People
with a family history of alcoholism are more likely to begin
drinking before the age of 20 and to become alcoholic. Such
adolescent drinkers are also more apt to underestimate the
effects of drinking and to make judgment errors, such as going
on binges and driving after drinking, than young drinkers
without a family history of alcoholism. But anyone who begins
drinking in adolescence is at risk for developing alcoholism.
Currently 1.9 million young people between the ages of 12 and
20 are considered heavy drinkers and 4.4 million are binge
drinkers.
Although alcoholism usually develops in early adulthood, the

elderly are not exempt. A survey of 5,000 adults over 60
reported that 15% of men and 12% of women were hazardous
drinkers, and 9% of men and 3% of women were alcohol
dependent. In another study, the prevalence of problem drinking
was as high as 49% among nursing home patients. Alcohol also
affects the older body differently; people who maintain the
same drinking patterns as they age can easily develop alcohol
dependency without realizing it. Physicians may overlook
alcoholism when evaluating elderly patients, mistakenly
attributing the signs of alcohol abuse to the normal effects of
the aging process.

Gender
Most alcoholics are men, but the incidence of alcoholism in
women has been increasing over the past 30 years. About 9.3%
of men and 1.9% of women are heavy drinkers, and 22.8% of
men are binge drinkers compared to 8.7% of women. In
general, young women problem drinkers follow the drinking
patterns of their partners, although they tend to engage in
heavier drinking during the premenstrual period. Women tend
to become alcoholic later in life than men, and it is estimated
that 1.8 million older women suffer from alcohol addiction.
Even though heavy drinking in women usually occurs later in
life, the medical problems women develop because of the
disorder occur at about the same age as men, suggesting that
women are more susceptible to the physical toxicity of alcohol.
Family History and Ethnicity

The risk for alcoholism in sons of alcoholic fathers is 25%. The
familial link is weaker for women, but genetic factors contribute
to this disease in both genders. In one study, women with
alcoholism tended to have parents who drank. Women who
came from families with a history of emotional disorders,
rejecting parents, or early family disruption had no higher risk
for drinking than women without such backgrounds. A stable
family and psychological health were not protective in people
with a genetic risk. Unfortunately, there is no way to predict
which members of alcoholic families are most at risk for
alcoholism.
Some population groups, such as Irish and Native Americans,

have an increased incidence in alcoholism while others, such as
Jewish and Asian Americans, have lower risk. Overall, there is
no difference in alcoholic prevalence between African
Americans, whites, and Hispanic people. Although the
biological causes of such different risks are not known, certain
people in these population groups may be at higher or lower
risk because of the way they metabolize alcohol. [ See Genetic
Factors under What Causes Alcoholism? above .]

Emotional Disorders
Severely depressed or anxious people are at high risk for
alcoholism, smoking, and other forms of addiction. Major
depression, in fact, accompanies about one-third of all cases of
alcoholism. It is more common among alcoholic women (and
women in general) than men. Depression and anxiety may play
a major role in the development of alcoholism in the elderly,
who are often subject to dramatic life changes, such as
retirement, the loss of a spouse or friends, and medical
problems. Problem drinking in these cases may be due to
self-medication of the anxiety or depression. It should be noted,
however, that it is not always clear whether people with
emotional disorders are self-medicating with alcohol or whether
long-term alcoholism itself causes chemical changes that
produce anxiety and depression.
Personality Traits
Studies are finding that alcoholism is strongly related to
impulsive, excitable, and novelty-seeking behavior, and such
patterns are established early on, if not inherited. People with
attention deficit hyperactivity disorder, a condition that shares
these behaviors, have a higher risk for alcoholism. In a test of
mental functioning, alcoholics (mostly women) did not show
any deficits in thinking but they were less able to inhibit their
responses. Children who later become alcoholics or who abuse
drugs are more likely to have less fear of new situations than
others, even if there is a greater risk for harm than in
nonalcoholics. On the other hand, studies are also finding an
association between alcohol use and having social phobia, a
form of anxiety in which the individual has an intense fear of
being publicly scrutinized and humiliated. Such individuals may
use alcohol as a way to become less inhibited in public
situations.
Socioeconomic Factors
It has been long thought that alcoholism is more prevalent in
people with lower educational levels and in those who are
unemployed. A thorough 1996 study, however, reported that the
prevalence of alcoholism among adult welfare recipients was
4.3% to 8.2%, which was comparable to the 7.4% found in the
general population. There was also no difference in prevalence

between poor African Americans and poor whites. People in

low-income groups did display some tendencies that differed
from the general population. For instance, as many women as
men were heavy drinkers. Excessive drinking may be more
dangerous in lower income groups; one study found that it was
a major factor in the higher death rate of people, particularly
men, in lower socioeconomic groups compared with those in
higher groups.
Geographic Factors
Although 54% of urban adults use alcohol at least once a month
compared to 42% in nonurban areas, living in the city or the
country does not affect the risks for bingeing or heavy alcohol
use. One study reported that people in the north central US are
at highest risk for heavy drinking (6.4% heavy use and 19%
binge drinking) and those in the Northeast have the lowest risk
(4.5% heavy use and 13% binge drinking).
Smokers
Researchers are finding common genetic factors in alcohol and
nicotine addiction, which may explain, in part, why alcoholics
are often smokers. Alcoholics who smoke compound their
health problems. More alcoholics die from tobacco-related
illnesses, such as heart disease or cancer, than from chronic
liver disease, cirrhosis, or other conditions that are more
directly tied to excessive drinking.
Sugar Cravings
People who crave sugar may also be at higher risk for
alcoholism. In one recent study, 62% of male alcoholics
enjoyed a sweet sugar solution compared with only 21% of
those without a drinking problem. It is not known, however,
whether having a "sweet tooth" can be an early predictor of
alcoholism or whether alcohol abusers simply develop a taste
for sweetness as a result of their chronic alcohol abuse.
http://home.mdconsult.com/das/patient/view/41/54...stty=cow&nid=39982&tlt=Z&pos=&tag=relinfo&anc=pe (10 of 30) [21/04/2000 23:29:45]

How Serious Is Alcoholism?

About 100,000 deaths a year can be wholly or partially
attributed to drinking, and alcoholism reduces life expectancy
by 10 to 12 years. Next to smoking, it is the most common
preventable cause of death in America. Although studies
indicate that adults who drink moderately (about one drink a
day) have a lower mortality rate than their non-drinking peers,
their risk for untimely death increases with heavier drinking.
The earlier a person begins drinking heavily, the greater their
chance of developing serious illnesses later on. Alcoholism can
kill in many different ways, and, in general, people who drink
regularly have a higher rate of deaths from injury, violence, and
some cancers.
Overdose
Alcohol overdose can lead to death. This is a particular danger
for adolescents who may want to impress their friends with their
ability to drink alcohol but cannot yet gauge its effects.
Accidents, Suicide, and Murder

Alcohol plays a major role in more than half of all automobile
fatalities. Less than two drinks can impair the ability to drive.
Alcohol also increases the risk of accidental injuries from many
other causes. One study of emergency room patients found that
having had more than one drink doubled the risk of injury, and
more than four drinks increased the risk eleven times. Another
study reported that among emergency room patients who were
admitted for injuries, 47% tested positive for alcohol and 35%
were intoxicated. Of those who were intoxicated, 75% showed
evidence of chronic alcoholism. This disease is the primary
diagnosis in one quarter of all people who commit suicide, and
alcohol is implicated in 67% of all murders.
Domestic Violence and Effects on Family

Domestic violence is a common consequence of alcohol abuse.
Research suggests that for women, the most serious risk factor
for injury from domestic violence may be a history of alcohol
abuse in her male partner. Alcoholism in parents also increases
the risk for violent behavior and abuse toward their children.
Children of alcoholics tend to do worse academically than

others, have a higher incidence of depression, anxiety, and
stress and lower self-esteem than their peers. One study found
that children who were diagnosed with major depression
between the ages of six and 12 were more likely to have
alcoholic parents or relatives than were children who were not
depressed. Alcoholic households are less cohesive, have more
conflicts, and their members are less independent and
expressive than households with nonalcoholic or recovering
alcoholic parents. In addition to their own inherited risk for later
alcoholism, one study found that 41% of children of alcoholics
have serious coping problems that may be life long. Adult
children of alcoholic parents are at higher risk for divorce and
for psychiatric symptoms. One study concluded that the only
events with greater psychological impact on children are sexual
and physical abuse.
The Effect of Alcohol on Mental

Functioning
Drinking too much alcohol can cause mild neurologic problems
in anyone, including insomnia and headache (especially after
drinking red wine). Long-term alcohol use appears to have
major effects upon the hippocampus, an area in the brain
associated with learning and memory and the regulation of
emotion, sensory processing, appetite, and stress. Brain scans of
people with long-term alcoholism have shown atrophy in
different parts of the brain and reduced brain activity, which
fortunately seems to be reversible with continued abstinence. In
a 1999 study, loss of verbal memory and slower reaction times
were associated with a higher incidence of recent alcohol use
(ie, within the last 3 months). A history of lifetime alcohol use,
however, did not seem to impair mental functioning of patients
with mild to moderate alcoholism. One study that uses imaging
techniques to scan the brains of inebriated subjects suggested
that while alcohol stimulates those parts of the brain related to
reward and induces euphoria, it does not appear to impair
cognitive performance (the ability to think and reason). Except
in severe cases, any neurologic damage is not permanent and
abstinence nearly always leads to recovery of normal mental
function. Severely alcoholic patients, however, often have
co-existing psychiatric or neurologic problems, and habitual use
of alcohol eventually produces depression and confusion .

Medical Problems
Alcohol can affect the body in so many ways that researchers
are having a hard time determining exactly what the
consequences are from drinking. It is well known, however, that
chronic consumption leads to many problems, some of them
deadly. Frequent heavy drinking is associated with a higher risk
for alcohol-related medical disorders (pancreatitis, upper
gastrointestinal bleeding, nerve damage, and impotence) than is
episodic drinking or continuous drinking without intoxication.
As people age, it takes fewer drinks to become intoxicated, and
organs can be damaged by smaller amounts of alcohol than in
younger people. Also, up to one-half of the 100 most prescribed
drugs for older people react adversely with alcohol. Alcohol
abusers who require surgery also have an increased risk of
postoperative complications, including infections, bleeding,
insufficient heart and lung functions, and problems with wound
healing. Alcohol withdrawal symptoms after surgery may
impose further stress on the patient and hinder recuperation.
Liver Disorders. The liver is particularly endangered by

alcoholism. About 10% to 35% of heavy drinkers develop
alcoholic hepatitis, and 10% to 20% develop cirrhosis. In the
liver, alcohol converts to toxic chemicals, such as acetaldehyde,
which trigger the production of immune factors called
cytokines. In large amounts, these agents cause inflammation
and tissue injury and are proving to be major culprits in the
destructive process in the liver. Not eating when drinking and
consuming a variety of alcoholic beverages are also factors that
increase the risk for liver damage. People with alcoholism are
also at higher risk for hepatitis B and C, potentially chronic
liver diseases than can lead to cirrhosis and liver cancer. People
with alcoholism should be immunized against hepatitis B; they
may need a higher-than-normal dose of the vaccine for it to be
effective. [ See also Hepatitis.]
Gastrointestinal Problems. Alcohol can cause diarrhea and

hemorrhoids. Alcohol abuse can cause ulcers, particularly in
people taking the painkillers known as nonsteroidal
anti-inflammatory drugs (such as aspirin or ibuprofen). Alcohol
can contribute to serious and chronic inflammation of the
pancreas (pancreatitis) in people who are susceptible to this
condition.
Heart Disease and Stroke. The effects of alcohol on heart

disease vary depending on consumption. Evidence strongly

suggests that light to moderate alcohol, particularly grape wine,
consumption (one or two drinks a day) protects the heart. The
benefits are strongest in people at high risk for heart disease and
may be fairly small in those at low risk. Light to moderate
alcohol intake may even reduce the risk of sudden cardiac
death. Large doses of alcohol, however, can trigger irregular
heartbeats and raise blood pressure even in people with no
history of heart disease. A major study found that those who
consumed more than three alcoholic drinks a day had higher
blood pressure than teetotalers. The more alcohol someone
drank, the greater the increase in blood pressure, with binge
drinkers (people who have nine or more drinks once or twice a
week) being at greatest risk. One study found that binge
drinkers had a risk for a cardiac emergency that was two and a
half times that of nondrinkers. Alcohol abuse has also been
associated with and may actually be one cause of idiopathic
dilated cardiomyopathy, a condition in which the heart enlarges
and its muscles weaken, putting the patient at risk for heart
failure. Alcohol may also increase the risk for hemorrhagic
stroke (caused by bleeding in the brain), although, as with heart
disease, it may protect against stroke caused by narrowed
arteries.
Cancer. Alcohol may not cause cancer, but it probably does

increase the carcinogenic effects of other substances, such as
cigarette smoke. One study indicated that alcohol, in
combination with tobacco smoke, causes genetic damage that is
associated with the development of cancer in the upper airways,
the esophagus, and liver; abstaining from alcohol appeared to
reverse this damage. Moderate use of alcohol has been
associated with a higher risk for breast cancer because of
increased estrogen levels or possibly because the liver
overproduces certain carcinogenic growth factors in response to
alcohol.
Pneumonia and Other Infections. Acute alcoholism is strongly

associated with very serious pneumonia. One study on
laboratory animals suggests that alcohol specifically damages
the bacteria-fighting capability of lung cells. Chronic
alcoholism also causes changes in the immune system, although
in people without any existing medical problems these changes
do not appear to be significant.
Skin, Muscle, and Bone Disorders. Severe alcoholism is

associated with osteoporosis (loss of bone density), muscular

deterioration, skin sores, and itching. Alcohol-dependent
women seem to face a higher risk than men for damage to
muscles, including muscles of the heart, from the toxic effects
of alcohol.
Hormonal Effects. Alcoholism increases levels of the female

hormone estrogen and reduces levels of the male hormone
testosterone, factors that contribute to impotence in men.
Pregnancy and Infant Development. Even moderate amounts of

alcohol may have damaging effects on the developing fetus,
including low birth weight and an increased risk for
miscarriage. High amounts can cause fetal alcohol syndrome,
which can result in mental and growth retardation. One study
indicates a significantly higher risk for leukemia in infants of
women who drink any type of alcohol during pregnancy.
Diabetes. Moderate alcohol consumption may help protect the

hearts of adults with older-onset, also called type 2 diabetes. It
should be noted, however, that alcohol can cause hypoglycemia,
a drop in blood sugar, which is especially dangerous for people
with diabetes who are taking insulin. Intoxicated diabetics may
not be able to recognize symptoms of hypoglycemia, a
particularly hazardous condition.
Malnutrition and Wernicke-Korsakoff Syndrome. A pint of

whiskey provides about half the daily calories needed by an
adult, but it has no nutritional value. In addition to replacing
food, alcohol may also interfere with absorption of proteins,
vitamins, and other nutrients. Of particular concern in
alcoholism is a severe deficiency in the B-vitamin thiamin,
which can cause a serious condition called Wernicke-Korsakoff
syndrome. Symptoms of this syndrome include severe loss of
balance, confusion, and memory loss. Eventually, it can result
in permanent brain damage and death. Another serious
nutritional problem among alcoholics is deficiency of the B
vitamin folic acid, which can cause severe anemia.
Acute Respiratory Distress Syndrome. One study indicated that

intensive care patients with a history of alcohol abuse have a
significantly higher risk for developing acute respiratory
distress syndrome (ARDS) during hospitalization. ARDS is a
form of lung failure that can be fatal. It is can by caused by
many of the medical conditions common in chronic alcoholism,

including severe infection, trauma, blood transfusions,

pneumonia, and other serious lung conditions.
Drug Interactions. The effects of many medications are

strengthened by alcohol, while others are inhibited. Of
particular importance is its reinforcing effect on antianxiety
drugs, sedatives, antidepressants, and antipsychotic
medications. Alcohol also interacts with many drugs used by
diabetics. It interferes with drugs that prevent seizures or blood
clotting. It increases the risk for gastrointestinal bleeding in
people taking aspirin or other nonsteroidal inflammatory drugs
including ibuprofen and naproxen. In other words, taking
almost any medication should preclude drinking alcohol.
How Is Alcoholism Diagnosed?

Even when people with alcoholism experience withdrawal
symptoms, they nearly always deny the problem, leaving it up
to coworkers, friends, or relatives to recognize the symptoms
and to take the first steps toward treatment.
Family members cannot always rely on a physician to make an

initial diagnosis. Although 15% to 30% of people who are
hospitalized suffer from alcoholism or alcohol dependence,
physicians often fail to screen for the problem. In addition,
doctors themselves often cannot recognize the symptoms. In
one study, alcohol problems were detected by the physician in
less than half of patients who had them. It is particularly
difficult to diagnose alcoholism in the elderly, where symptoms
of confusion, memory loss, or falling may be attributed to the
aging process alone. Heavy drinkers may be more likely to
complain to their doctors about so-called somatization
symptoms, which are vague ailments such as joint pain,
intestinal problems, or general weakness, that have no
identifiable physical cause. Such complaints should signal the
physician to follow-up with screening tests for alcoholism.
Alcoholism is particularly less likely to be recognized in elderly
women. In fact, only 1% of older women who need treatment
for alcoholism are diagnosed accurately and treated
appropriately. Instead, they are often diagnosed with depression
and may even be prescribed anti-anxiety drugs or
antidepressants that can have dangerous interactions with
alcohol. Even when physicians identify an alcohol problem,
however, they are frequently reluctant to confront the patient

with a diagnosis that might lead to treatment for addiction.
Screening for Alcoholism

A physician who suspects alcohol abuse should ask the patient
questions about current and past drinking habits to distinguish
moderate from heavy drinking. If alcohol abuse or dependency
is indicated, the physician will usually perform a screening test.
Many are available for diagnosing alcoholism, usually either
standardized questionnaires that the patient can take on their
own or that are conducted by the physician. Because people
with alcoholism often deny their problem or otherwise attempt
to hide it, the tests are designed to elicit answers related to
problems associated with drinking rather than the amount of
liquor consumed or other specific drinking habits. The quickest
test takes only one minute; it is called the CAGE test, an
acronym for the following questions: (C) attempts to Cut down
on drinking; (A) Annoyance with criticisms about drinking; (G)
Guilt about drinking; and (E) use of alcohol as an eye-opener in
the morning. This test and another called the Self-Administered
Alcoholism Screening Test (SAAST), appear to be most useful
in detecting alcoholism in white middle-aged males. They are
not very accurate for identifying alcohol abuse in older people,
white women, and African- and Mexican-Americans. A more
effective test for such individuals may be the Alcohol Use
Disorders Identification Test (AUDIT), which asks three
questions about amount and frequency of drinking, three
questions about alcohol dependence, and four questions about
problems related to alcohol consumption. AUDIT is an
important component in screening for alcoholism in anyone,
because it is the only test specifically designed to identify
hazardous or harmful drinking. Other short screening tests are
the Michigan Alcoholism Screening Test (MAST) and The
Alcohol Dependence Scale (ADS).
Laboratory and Other Tests

Tests for alcohol levels in the blood are not useful for
diagnosing alcoholism because they reflect consumption at only
one point in time and not long-term usage. A mean corpuscular
volume (MCV) blood test is sometimes used to measure the size
of red blood cells, which increase with alcohol use over time. A
test for a factor known as carbohydrate-deficient transferrin
may prove to be fairly accurate indicator of heavy drinking. A

physical examination and other tests should be performed to

uncover any related medical problems. Sometimes the results of
tests that detect other problems, such as blood tests reporting
liver damage or low testosterone levels in men, can persuade
alcoholics to seek help.
What Are The General

Guidelines For Treating
Alcoholism?
Getting the Patient to Seek Treatment
Once a diagnosis of alcoholism is made, the next major step is
getting the patient to seek treatment. One study reported that the
main reasons alcoholics do not seek treatment are lack of
confidence in successful therapies, denial of their own
alcoholism, and the social stigma attached to the condition and
its treatment. The best approaches for motivating a patient to
seek treatment are group meetings between people with
alcoholism and their friends and family members who have
been affected by the alcoholic behavior. Using this
interventional approach, each person affected offers a
compassionate but direct and honest report describing
specifically how he or she has been specifically hurt by their
loved one's or friend's alcoholism. Children may even be
involved in this process, depending on their level of maturity
and ability to handle the situation. The family and friends
should express their affection for the patient and their intentions
for supporting the patient through recovery, but they must
strongly and consistently demand that the patient seek
treatment. Employers can be particularly effective. Their
approach should also be compassionate but strong, threatening
the employee with loss of employment if he or she does not
seek help. Some large companies provide access to inexpensive
or free treatment programs for their workers.
The alcoholic patient and everyone involved should fully

understand that alcoholism is a disease and that the responses to
this disease, need, craving, fear of withdrawal, are not character
flaws but symptoms, just as pain or discomfort are symptoms of
other illnesses. They should also realize that treatment is
difficult and sometimes painful, just as treatments for other
life-threatening diseases, such as cancer, are, but that it is the
only hope for a cure.
Treatment Options
A number of treatment options now exist for alcoholism,
including psychotherapy, medications that target brain
chemicals involved in addiction, and social support groups such
as Alcoholics Anonymous. Studies are suggesting the cognitive
therapies and medications, such as naltrexone, may be very
effective for some people. Even brief intervention by a family
doctor can be helpful for reducing alcohol intake in many heavy
drinkers.
People with mild to moderate withdrawal symptoms are usually

treated as outpatients and assigned to support groups,
counseling, or both. Inpatient treatment can be performed in a
general or psychiatric hospital or in a center dedicated to
treatment of alcohol and other substance abuse. It is
recommended for patients with a coexisting medical or
psychiatric disorder and those who may harm themselves or
others, who have not responded to conservative treatments, or
who have a disruptive home environment. A typical inpatient
regimen may include a physical and psychiatric work-up,
detoxification, treatment with medications and psychotherapy
or cognitive-behavioral therapy, and an introduction to
Alcoholics Anonymous. Because of the high cost of inpatient
care, its advantages over outpatient care are being questioned.
One study compared employed alcoholics who were either
hospitalized, treated as outpatients with compulsory attendance
at AA meetings, or allowed to choose their own treatment
option, including none at all. After two years, everyone
experienced fewer job problems, but those in the inpatient
group had significantly fewer rehospitalizations and remained
abstinent longer than people in the other two groups. Another
study analyzing drug and alcohol treatment programs found that
75% of inpatients completed therapy compared to only 18% of
outpatients. Other studies, however, have shown no difference
in results between inpatient and outpatient programs. Given the
current economic climate, it seems unlikely that most care
providers will choose inpatient treatment for alcoholics who are
not a threat to others or themselves if there are alternatives.
Treatment for the Medically or Mentally Ill Alcoholic. Severe

alcoholism is often complicated by the presence of serious
medical or mental illnesses. A program called integrated
outpatient treatment (IOT) has been designed specifically for

medically ill alcoholics. The patient visits a clinic once a month
for both intensive alcohol treatment and a physical check-up,
which includes tracking factors, such as liver function, that are
affected by drinking. Patients are motivated through discussions
of benefits and costs of drinking and by reporting any barriers
to changing their habits and learning strategies to overcome
them. One study showed that IOT significantly increased
abstinence and the number of treatment visits. IOT may even
improve survival rates. Interestingly, however drinking also
significantly decreased in a comparison group of patients who
were treated only for their medical conditions.
Treatment for patients with both alcoholism and mental illness

is particularly difficult. The greater the psychiatric distress a
person is experiencing, the more he or she is tempted to drink,
particularly in negative situations. Antidepressants or
anti-anxiety medications in addition to support services may
help people with depression and anxiety. People with
alcoholism and more severe problems, such as schizophrenia or
severe bipolar disorder probably need more intense help. AA
groups are underused by patients with a double-diagnosis of
mental illness and alcoholism because the focus of the
organization is on addiction not psychiatric problems. Some AA
members have even been known to discourage patients with
dual disorders to go off their medications. In one survey 54% of
AA members felt that such patients would do better in groups
designed especially for dual disorders. Unfortunately, no such
programs are available.
Treatment Goals
The ideal goals of long-term treatment by many physicians and
organizations such as AA are total abstinence and replacement
of the addictive patterns with satisfying, time-filling behaviors
that can fill the void in daily activity that occurs when drinking
has ceased. Because abstinence is so difficult to attain, many
professionals choose to treat alcoholism as a chronic disease;
that is, they expect and accept relapse but they aim for as long a
remission period as possible. Even reducing alcohol intake can
lower the risk for alcohol-related medical problems. Studies
suggest, however, that patients who secure total abstinence have
better survival rates, mental health, and marriages and they are
more responsible parents and employees than those who
continue to drink or relapse. There is also no way to determine

which people can stop after one drink and which ones cannot.
Alcoholics Anonymous and other alcoholic treatment groups
whose goal is strict abstinence are greatly worried by the
publicity surrounding these studies, since many people with
alcoholism are eager for an excuse to start drinking again. At
this time, seeking total abstinence is the only safe route.
What Is The Treatment For

Alcohol Withdrawal?
Symptoms of Withdrawal
When a person with alcoholism stops drinking, withdrawal
symptoms begin within six to 48 hours and peak about 24 to 35
hours after the last drink. During this period the inhibition of
brain activity caused by alcohol is abruptly reversed. Stress
hormones are over-produced and the central nervous system
becomes over-excited. Seizures occur in about 10% of adults
during withdrawal, and in about 60% of these patients, the
seizures are multiple. The time between the first and last seizure
is usually six hours or less. About 5% of alcoholic patients
experience delirium tremens, which usually develops two to
four days after the last drink. Symptoms include fever, rapid
heart beat, either high or low blood pressure, extremely
aggressive behavior, hallucinations, and other mental
disturbances. Although it is not clear if older people with
alcoholism are at higher risk for more severe symptoms than
younger patients, several studies have indicated that they may
suffer more complications during withdrawal, including
delirium, falls, and a decreased ability to perform normal
activities.
Initial Assessment
Upon entering a hospital because of alcohol withdrawal,
patients should be given a physical examination for any injuries
or medical conditions and should be treated for any potentially
serious problems, such as high blood pressure or irregular
heartbeat. The immediate goal of treatment is to calm the
patient as quickly as possible. Patients are usually given one of
the anti-anxiety drugs known as benzodiazepines, which relieve
withdrawal symptoms and help prevent progression to delirium
tremens. An injection of the B vitamin, thiamine, may be given
to prevent Wernicke-Korsakoff syndrome. Patients should be

observed for at least two hours to determine the severity of
withdrawal symptoms. Physicians may use assessment tests,
such as the Clinical Institute Withdrawal Assessment Scale
(CIWA), to help determine treatment and whether the
symptoms will progress in severity.
Treatment for Withdrawal Symptoms

About 95% of people have mild to moderate withdrawal
symptoms, including agitation, trembling, disturbed sleep, and
lack of appetite. In 15% to 20% of people with moderate
symptoms, brief seizures and hallucinations may occur, but they
do not progress to full-blown delirium tremens. Such patients
can nearly always be treated as outpatients. After being
examined and observed, the patient is usually sent home with a
four-day supply of anti-anxiety medication, scheduled for
follow-up and rehabilitation, and advised to return to the
emergency room if withdrawal symptoms become severe. If
possible, a family member or friend should support the patient
through the next few days of withdrawal.
Benzodiazepines. Benzodiazepines are anti-anxiety drugs that

inhibit nerve-cell excitability in the brain and help reduce the
risk for seizures. They also relieve withdrawal symptoms, and
make it easier for patients to remain in treatment. They include
diazepam (Valium), lorazepam (Ativan), midazolam (Versed),
and oxazepam (Serax). These drugs vary in how long they are
effective. Diazepam has a longer duration of action than
lorazepam or midazolam, for example. Typically, the physician
may give the patient an initial, or loading, intravenous dose of
diazepam with additional doses given every one to two hours
thereafter over the period of withdrawal. This regimen can
cause very heavy sedation. Lorazepam and oxazepam are easier
for the liver to metabolize than other benzodiazepines and often
prove useful for treating alcoholic patients. Some physicians
question the use of any anti-anxiety medication for mild
withdrawal symptoms. Others believe that repeated withdrawal
episodes, even mild forms, that are inadequately treated may
result in increasingly severe and frequent seizures with possible
brain damage. Benzodiazepines may be administered
intravenously or orally, depending on the severity of symptoms.
One study reported that when a single, intravenous dose,
lorazepam, was given within several hours of a first
alcohol-related seizure, it reduced the risk for subsequent ones.

Benzodiazepines are usually not prescribed for more than two

weeks or administered for more than three nights per week.
Tolerance to these drugs may develop after as little as four
weeks of daily use. Physical dependence may develop after just
three months of normal dosage. People who discontinue
benzodiazepines after taking them for long periods may
experience rebound symptoms, sleep disturbance and anxiety,
which can develop within hours or days after stopping the
medication. Some patients experience withdrawal symptoms
from the drugs, including stomach distress, sweating, and
insomnia, that can last from one to three weeks. Common side
effects are day-time drowsiness and a hung-over feeling.
Respiratory problems may be exacerbated. Benzodiazepines are
potentially dangerous when used in combination with alcohol.
They should not be used by pregnant women or nursing mothers
unless absolutely necessary.
Other Drugs for Mild to Moderate Withdrawal. Beta-blockers,

such as propranolol (Inderal) and atenolol (Tenormin), may
sometimes be used in combination with a benzodiazepine. This
class of drugs is effective in slowing heart rate and reducing
tremor. Other drugs being tested are clonidine (Catapres) and
carbamazepine (Tegretol). When used by themselves, they do
not, however, appear to be effective in reducing seizures or
delirium. Chlormethiazole, a derivative of vitamin B1, is
presently used in Europe and is showing promise in reducing
agitation and seizures.
Treatment for Delirium Tremens,

Seizures, and Other Severe Symptoms
People with symptoms of delirium tremens must be treated
immediately. Untreated delirium tremens has a fatality rate that
can be as high as 20%. Symptomatic patients are usually given
intravenous anti-anxiety medications. It is extremely important
that fluids be administered. Restraints may be necessary to
prevent injury to themselves or others.
Seizures are usually self-limited and treated with a

benzodiazepine. Intravenous phenytoin (Dilantin) along with a
benzodiazepine may be used in patients who have a history of
seizures, who have epilepsy, or in those whose seizures cannot
be controlled. Because phenytoin may lower blood pressure, the

patient's heart should be monitored during treatment. For

hallucinations or extremely aggressive behavior, antipsychotic
drugs, particularly haloperidol (Haldol), may be administered.
Lidocaine (Xylocaine) may be given to people with disturbed
heart rhythms.
What Are The Long-Term

Treatments For Alcoholism To
Prevent Relapse?
Psychotherapy and Cognitive-Behavioral
Therapy
The two usual forms of therapy for alcoholics are
cognitive-behavioral and interactional group psychotherapy
based on the Alcoholics Anonymous 12-step program. In one
study, all treatment approaches were, on average, equally
effective as long as the individual program was competently
administered. Those with fewer psychiatric problems, however,
did best with the AA approach. This confirms an earlier study in
which researchers categorized alcoholics as either Type A or
Type B. Type A individuals became alcoholic at a later age, had
less severe symptoms or fewer psychiatric problems, and had a
better outlook on life than those with Type B. The people in the
Type A group did well with the 12-step approach. They did not
do as well with cognitive-behavioral therapy. Type B people
became alcoholic at an early age, had a high family risk for
alcoholism, more severe symptoms, and a negative outlook on
life. This group did poorly with interactional group therapy but
tended to do better with cognitive-behavioral therapy. This
difference in response to the two forms of treatments held up
after two years.
Interactional Group Psychotherapy (12-Step Program ).

Alcoholics Anonymous (AA), founded in 1935, is an excellent
example of interactional group psychotherapy and remains the
most well-known program for helping people with alcoholism.
It offers a very strong support network using group meetings
open seven days a week in locations all over the world. A
buddy system, group understanding of alcoholism, and
forgiveness for relapses are AA's standard methods for building
self-worth and alleviating feelings of isolation. AA's 12-step
approach to recovery includes a spiritual component that might
deter people who lack religious convictions. Prayer and

meditation, however, have been known to be of great value in
the healing process of many diseases, even in people with no
particular religious assignation. AA emphasizes that the "higher
power" component of its program need not refer to any specific
belief system. Associated membership programs, Al-Anon and
Alateen, offer help for family members and friends.
Cognitive-Behavioral Therapy. Cognitive-behavioral therapy

uses a structured teaching approach and may be better than AA
for severe alcoholism. People with alcoholism are given
instruction and homework assignments intended to improve
their ability to cope with basic living situations, control their
behavior, and change the way they think about drinking. For
example, patients might write a history of their drinking
experiences and describe what they consider to be risky
situations. They are then assigned activities to help them cope
when exposed to "cues," places or circumstances that trigger
their desire to drink. Patients may also be given tasks that are
designed to replace drinking. An interesting and successful
example of such a program was one that enlisted patients in a
softball team; this gave them the opportunity to practice coping
skills, develop supportive relationships, and engage in healthy
alternative activities. In one study of patients with both
depression and alcoholism, this therapeutic approach achieved
47% abstinence rates after six months compared to only 13%
abstinence in patients who received standard treatments and
relaxation techniques. It appears to be especially effective when
used in combination with opioid antagonists, such as
naltrexone.
Medications to Aid in Abstinence

Opioid Antagonists. Opioid antagonists are drugs that reduce
the intoxicating effects of alcohol and the urge to drink. One of
these agents, naltrexone (ReVia), has been found to be very
effective for people with low- to moderate alcohol dependency
when used with cognitive behavioral therapy. In one 1999
study, for example, 62% of patients taking naltrexone and
undergoing such therapy did not relapse into heavy drinking
compared with 40% of patients taking a placebo (a "dummy"
pill). It does not appear to improve abstinent rates, however.
Taking the drug consistently as prescribed by the doctor is very
important for its success. The most common side effect of
naltrexone is nausea, which is usually mild and temporary. High

doses cause liver damage. The drug should not be administered

to anyone who has used narcotics within a week to 10 days. An
oral form of nalmefene, an opioid antagonist currently available
only by injection, is also proving to be effective in preventing
relapse in heavy drinkers. Nalmefene blocks more opioid
receptors than naltrexone does and may have less of an adverse
effect on the liver.
Aversion Medications. Some drugs have properties that interact

with alcohol to produce distressing side effects. Disulfiram
(Antabuse) causes flushing, headache, nausea, and vomiting if a
person drinks alcohol while taking the drug. The symptoms can
be triggered after drinking half a glass of wine or half a shot of
liquor and last from half an hour to two hours, depending on
dosage of the drug and the amount of alcohol consumed. One
dose of disulfiram is usually effective for one to two weeks.
Overdose can be dangerous, causing low blood pressure, chest
pain, shortness of breath, and even death. Studies have not
shown the use of disulfiram to have any effect on staying
abstinent, although it does reduce the frequency of drinking.
One study indicated that the drug may be more effective in
patients with spouses or other family members or caregivers,
including AA "buddies," who are close by and vigilant to ensure
that they take it. (Such support, however, probably improves the
effectiveness of any treatment.) Another aversion drug, calcium
carbimide, was withdrawn from the market.
Acamprosate. Acamprosate (Campral) calms the brain and

reduces cravings by inhibiting the transmission of the
neurotransmitter gamma aminobutyric acid (GABA). Studies in
Europe indicate that it reduces the frequency of drinking.
Although it is not clear whether it can improve abstinence, one
study reported that 60% of patients remained abstinent for 12
weeks, and in another 43% were still abstinent after nearly a
year. The drug may cause occasional diarrhea and headache. It
also can impair certain memory functions but does not alter
short-term working memory or mood. People with kidney
problems should use it cautiously. Combination therapy with
naltrexone or disulfiram may be possible.
Antidepressant and Anti-Anxiety Drugs. Depression is common

among alcohol-dependent people and can lead to a higher
relapse rate. Antidepressants may be helpful, particularly for
patients who suffer from both depression and alcoholism.
Because of their effect on serotonin, the antidepressants

selective serotonin reuptake inhibitors (SSRIs) were of

particular interest. They include fluoxetine (Prozac), sertraline
(Zoloft), paroxetine (Paxil), citalopran (Celexa), and
fluvoxamine (Luvox). Studies indicate they may be useful for
reducing alcohol intake in heavy drinkers even if they are not
depressed, although these drugs appear to have no significant
affect on alcoholism itself. Another small study reported that
people given the tricyclic antidepressant desipramine
(Norpramin, Pertofrane), whether or not they exhibited other
symptoms of depression, had fewer drinking days and a longer
period between relapses than those not taking the drug. A
unique anti-anxiety drug, buspirone (BuSpar), may also be
beneficial for alcoholics, particularly if they also suffer from
anxiety. The drug has few side effects and a low potential for
abuse. It not only reduces anxiety, but also appears to have
modest effects on alcohol cravings. In one study, alcoholics
who took it had a slow return to alcohol consumption and fewer
drinking days than those not on the drug. Another study,
however, found no significant effect on alcoholism.
Other Drugs. Under investigation are drugs that affect

dopamine, the neurotransmitter (chemical messenger in the
brain) that produces a sense of reward after drinking. Among
these, tiapride, which blocks dopamine, is showing some
modest benefits in small European studies. In one small study,
isradipine, a calcium channel blocker, reduced cravings more
effectively than naltrexone and the antidepressant paroxetine
(Paxil). Calcium channel blockers are ordinarily used to treat
high blood pressure and other medical conditions. Another drug
being investigated for withdrawal and abstinence is
gamma-hydroxybutyric acid (GHB). In one small study, 58% of
subjects remained abstinent during a six-month period. It should
be noted that GHB is sold illegally as a street drug because of
its euphoric effects at high doses, which can have serious side
effects, including seizures, coma, and respiratory arrest.
Where Else Can Help Be

Obtained For Alcoholism?
Alcoholics Anonymous, World Services, Inc., P.O. Box 459,
New York, NY 10163. Call (212-870-3400) or on the Internet
(http://www.alcoholics-anonymous.org/ )

Al-Anon Family Group Headquarters, Inc., 1600 Corporate

Landing Pkwy, Virginia Beach, VA 23454-5617. Call
(800-344-2666 in the US or 800-443-4525 in Canada) for
meetings. Or call (800-356-9996 in the US or 800-714-7498 in
Canada) for literature or on Internet
(http://www.Al-Anon-Alateen.org/ )
Al-Anon was started by the wife of the founder of Alcoholics
Anonymous to help families of alcoholics. They provide
meetings and educational material established along the lines of
those of AA. Also available through Al-Anon is Alateen, a
support fellowship for adolescents affected by people with
alcoholism. (This group is not for teenagers with drinking
problems.)
National Institute on Alcohol Abuse and Alcoholism, 6000

Executive Boulevard - Willco Building, Bethesda, Maryland
20892-7003. On the Internet (http://www.niaaa.nih.gov/ )
National Clearinghouse of Alcohol and Drug Information, PO

Box 2345, Rockville, MD 20852. Call (800-729-6686) or on the
Internet (http://www.health.org/ )
Offers many publications on alcohol and substance abuse.
Hazelden Foundation, PO Box 11, Center City, MN

55012-0011. Call (800-257-7800 or 1-651-257-4010 outside the
US) or on the Internet (http://www.hazelden.org/ )
Maintains chemical dependency treatment centers. Also
provides educational materials for adults and adolescents. Their
web site is very useful.
National Council on Alcoholism, 12 West 21 Street, New York,

NY 10010. Call (800-NCA-CALL) or (212-206-6770) or on the

Internet (http://www.ncadd.org/ ).
Their 800 number is a hotline that requires a touch-tone phone.
A recorded message provides local numbers for counseling,
help, and information after the caller keys in their zip code.
National Organization on Fetal Alcohol Syndrome, 18 C Street

North East, Washington, DC 20002
Call (202-785-4585 ) or on the Internet (http://www.nofas.org/

)
On the Internet:
Web of Addictions (http://www.well.com/user/woa/ ) has
good links and keeps up with current research.
Recovery (http://www.recovery.org/aa/ ) is a private web

site with good links to other sites on alcoholism.

printing.
Then use the print function on your browser to
print the handout.
Special Instructions:


Bookmark URL: /das/patient/view/41/5485.html/top

April 21, 2000
Conde Petra
April 14, 2000

SOB

http://home.mdconsult.com/das/stat/view/cow?nid=39981&anc=q2 (1 of 4) [21/04/2000 23:31:23]

of cirrhosis.
knowledge.




Bibliography
110(4):253-255, 1989(editorial)


Saunders







70(5):1081-1091, 1986


8(1):85-99, 1986








WB Saunders



April 21, 2000
Conde Petra
April 14, 2000
Question 2 - Cirrhosis
Cirrhosis is an alteration of the liver architecture and is characterized by the

following: (1) interconnecting fibrous scars, (2) fibrous bands, (3) parenchymal
nodules, and (4) formation of abnormal arteriovenous interconnections. Cirrhosis
has been divided to micronodular and macronodular, but these are purely
descriptive terms. The causes of cirrhosis in descending order of frequency follow.
Alcoholic cirrhosis (Laennec's cirrhosis) may be a sequela of alcoholic hepatitis.

This type of cirrhosis is micronodular and accounts for close to 50% to 70% of
cirrhosis in the Western world.Chronic hepatitis progresses to macronodular
cirrhosis. The causes of chronic active hepatitis are multiple and include HBV,
HCV, autoimmune injury, toxin exposure, and drug reactions (isoniazid,
amiodarone). Biliary cirrhosis may be either primary or secondary. Secondary
biliary cirrhosis develops after prolonged extrahepatic biliary tract obstruction that
leads to bile retention, inflammation, fibrosis, and eventual cirrhosis. Inherited
metabolic disorders can also lead to cirrhosis. These include Wilson's disease,
alpha-1-antitrypsin deficiency and hereditary hemochromatosis.
Alpha-1-antitrypsin deficiency (A1AT) is characterized by low levels of this

protease inhibitor resulting in an imbalance favoring activity of destructive enzymes
causing liver and lung injury. Hepatic congestion secondary to heart failure can
cause cirrhosis known as cardiac cirrhosis. It is an uncommon complication of right
heart failure and is mostly associated with constrictive pericarditis or rheumatic
heart disease. Cryptogenic cirrhosis accounts for 10% of cases. In this
circumstance, cirrhosis is found incidentally or presents as a complication of
end-stage liver disease. The diagnosis is one of exclusion.
The clinical presentation of early cirrhosis is variable. Initially symptoms include

anorexia, weight loss, weakness, and debilitation. Physical examination reveals a
firm, large liver, jaundice, spider angiomas, gynecomastia, leukonychia, muscle
wasting, and testicular atrophy. In the later stages of cirrhosis complications
develop with additional clinical findings, which will be discussed later. Other
physical abnormalities such as Dupuytren's contracture, xanthomas,
Kayser-Fleischer rings, and a bronze discoloration of the skin are appreciated in
specific forms of cirrhosis.
Laboratory evaluation of cirrhosis may reveal elevated bilirubin, mild transaminase

elevation, hypoalbuminemia, and prolonged prothrombin time. Diagnosis is
confirmed by percutaneous liver biopsy, which usually reveals distortion of hepatic
architecture and the presence of nodules and scar tissue.
Learn more about cirrhosis:

Proceed With Case
Alcoholic Liver Disease

Feldman: Sleisenger & Fordtran's Gastrointestinal and Liver Disease, Sixth Edition
Cirrhosis of the liver

Dambro: Griffith's 5-Minute Clinical Consult, 1999 ed.
Copyright 1999 Lippincott Williams & Wilkins, Inc.
Alcoholic Liver Disease, Cirrhosis, And Its Major Sequelae

Goldman: Cecil Textbook of Medicine, 21st Ed.
Cirrhosis
Rakel: Conn's Current Therapy 1999, 51st ed.
Cirrhosis of the Liver, Alternative Therapies

Copyright Integrative Medicine Communications 1999
Integrative Medicine Access
Cirrhosis
Well-Connected
Alcohol-Related Problems
Copyright Clinical Reference Systems 1999
Adult Health Advisor
Senior Health Advisor
Women's Health Advisor


April 21, 2000
Conde Petra
Feldman: Sleisenger & Fordtran's

Gastrointestinal and Liver Disease,
About the Publication Sixth Edition, Copyright 1998 W. B. Saunders
Company
Section 9 - Liver
71 - Alcoholic Liver Disease Feldman: Sleisenger & Fordtran's Gastrointestinal and Liver Disease,
Sixth Edition, Copyright 1998 W. B. Saunders Company
Introduction
EPIDEMIOLOGY
ETHANOL METABOLISM
1199
PATHOGENESIS OF
ALCOHOLIC LIVER INJURY Chapter 71 - Alcoholic Liver
COFACTORS INFLUENCING
THE DEVELOPMENT OF
Disease
DIAGNOSIS Jacquelyn J. Maher

COMPLICATIONS
EPIDEMIOLOGY, 1199
TREATMENT ETHANOL METABOLISM, 1199
PROGNOSIS Hepatic Metabolism, 1199
REFERENCES Gastric Metabolism, 1200
PATHOGENESIS OF ALCOHOLIC LIVER INJURY, 1201
Abnormalities Linked to Hepatic Ethanol Metabolism, 1201
Mitochondria as Targets of Ethanol Hepatotoxicity, 1202
Immune/Inflammatory Mechanisms of Liver Injury, 1202
Mechanisms of Fibrosis, 1203
Putative Mechanisms Underlying the Zonal Distribution of
Alcoholic Liver Damage, 1203
COFACTORS INFLUENCING THE DEVELOPMENT OF
ALCOHOLIC LIVER DISEASE, 1204
Hereditary Variations in Ethanol Metabolism, 1204
Gender, 1204
Diet/Nutrition, 1204
Viral Liver Disease, 1204
http://home.mdconsult.com/das/book/view/854/589.h...&stty=cow&nid=39983&tlt=Z&pos=&tag=relinfo&anc=bo (1 of 2) [21/04/2000 23:32:32]

DIAGNOSIS, 1205
Physical Findings, 1205
Laboratory Findings, 1205
Histology, 1206
Differential Diagnosis, 1207
COMPLICATIONS, 1207
TREATMENT, 1208
Nutrition, 1208
Propylthiouracil, 1208
Corticosteroids, 1209
Colchicine, 1209
Polyunsaturated Lecithin/Phosphatidylcholine, 1210
Antioxidants, 1210
Liver Transplantation, 1210
PROGNOSIS, 1210


Feldman: Sleisenger & Fordtran's Gastrointestinal and Liver Disease, Sixth Edition, Copyright 1998 W. B.
Saunders Company
Section 9 - Liver
1055
Chapter 62 - Anatomy and Developmental

Anomalies of the Liver
Ian R. Wanless
SURFACE ANATOMY, 1055

SEGMENTAL ANATOMY, 1055
VARIATIONS IN SURFACE ANATOMY, 1055
LARGE VESSELS OF THE LIVER, 1056
Portal Veins, 1056
Hepatic Veins, 1057
Hepatic Arteries, 1057
Hepatic Collateral Circulation, 1058
LYMPH VESSELS, 1058
NERVE SUPPLY, 1058
BILIARY SYSTEM, 1058
Anatomic Variations of the Biliary System, 1058
MICROANATOMY, 1058
Sinusoidal Pathology, 1059
Organization of Hepatic Parenchyma: The Acinus and the Lobule, 1059
Hepatic Microcirculation, 1059
Microcirculation in Chronic Liver Disease, 1060
The knowledge of normal morphology is basic to the understanding of pathologic processes and the
secure practice of hepatobiliary surgery. This chapter is an introduction to anatomy of the human liver
and its variations. More detailed expositions can be found elsewhere. [1] [2] [3]



Saunders Company
SURFACE ANATOMY
The normal liver occupies the right upper quadrant, extending from the fifth intercostal space in the
midclavicular line down to the right costal margin. The lower margin descends below the costal margin
during inspiration. The median liver weight is 1800 g in men and 1400 g in women. [4] [5]
Transcutaneous liver biopsy specimens are commonly obtained in the midaxillary line through the third
interspace below the upper limit of liver dullness during full expiration. This is usually in the ninth
intercostal space.
The superior, anterior, and right lateral surfaces are smooth and convex, fitting against the diaphragm.
The posterior surface has indentations from the colon, right kidney, and duodenum on the right lobe and
the stomach on the left lobe (Fig. 62-1) .
The fibrous capsule on the posterior aspect of the liver reflects onto the diaphragm and posterior
abdominal wall, leaving a "bare area" with the liver in continuity with the retroperitoneum. The liver is
supported by the peritoneal reflections, which form the coronary ligaments, right and left triangular
ligaments, and the falciform ligament (see Fig. 62-1) . The lower free edge of the falciform ligament
contains the round ligament, which is largely composed of the obliterated umbilical vein. The falciform
ligament joins the anterior surface of the liver to the diaphragm. Superiorly, the falciform ligament joins
the peritoneal reflections to the left of the vena cava.
The hepatoduodenal ligament connects the liver to the superior part of the duodenum. The free margin of
this ligament contains the hepatic artery, portal vein, bile duct, nerves, and lymphatic vessels. These
structures connect with the liver in the transverse portal fissure. The caudate lobe is posterior and the
quadrate lobe is anterior to this fissure. The quadrate lobe is further demarcated on the right by the
gallbladder and on the left by the umbilical fissure.


Saunders Company
SEGMENTAL ANATOMY
The liver has classically been divided into left and right lobes by the location of the falciform ligament.
Because this location does not correspond to the internal subdivisions of the liver, a more functional
nomenclature was developed by Hjortso and Couinaud based on the distribution of vessels and ducts
within the liver. [6] In this nomenclature, the line extending between the vena cava and the gallbladder
(Cantlie's line) demarcates the right and left livers (or hemilivers), each with independent vascular and
duct supplies. This line marks a relatively bloodless plane that is of use to the surgeon. The liver can be
further divided into eight segments, each containing a pedicle of portal vessels and ducts and drained by
hepatic veins situated in the planes (called scissura) between the segments [6] (Fig. 62-2) (Figure Not
Available) . The segments usually have no surface fissures to allow their accurate identification. The left
hemiliver is composed of the classic left lobe, plus the caudate lobe and the quadrate lobe and its superior
extension. There is considerable individual variation in the location of the segments, especially in the
right hemiliver. [7]
The common resections, based on these segmental definitions, are right hepatectomy (segments 5-8),
right lobectomy (trisegmentectomy of Starzl, segments 4-8), left hepatectomy (segments 1-4), and left
lobectomy (segments 1-3). [6]


Saunders Company
VARIATIONS IN SURFACE ANATOMY

An elongation of the right lobe (Riedel's lobe) can be mistaken for hepatomegaly. This anomaly and
minor variations in the shape of the liver explain why liver size estimated clinically correlates poorly
with more objective measures.
Deep fissures may demarcate supernumerary lobes. The left lobe is rarely attached to the right lobe by a
narrow pedicle. Accessory livers may be found in the ligaments or mesentery or on the surface of the
gallbladder, spleen, or adrenals.
Atrophy of the left lobe is usually an acquired lesion resulting from thrombosis of the portal or hepatic
veins. [8] Coarse lobulations (hepar lobatum) are a result of obliterative lesions in large and medium-sized
vessels, typically after invasion by neoplasms. [9]
1056
Figure 62-1 Posterior view of the liver. The shape of the liver is determined by molding against
adjacent organs. At the porta, the common bile duct lies to the right, the hepatic artery to the left, and
the portal vein to the rear. Variations in the location of the artery are frequent.


Saunders Company
Figure 62-1 Posterior view of the liver. The shape of the liver is determined by molding against adjacent
organs. At the porta, the common bile duct lies to the right, the hepatic artery to the left, and the portal
vein to the rear. Variations in the location of the artery are frequent.


Saunders Company
PATHOGENESIS OF ALCOHOLIC LIVER INJURY (Table 71-2)

Abnormalities Linked to Hepatic Ethanol Metabolism
Redox Alteration
ADH-mediated ethanol oxidation is accompanied by the reduction of oxidized nicotinamide-adenine

dinucleotide (NAD+ ) to reduced NAD (NADH) (see Fig. 71-2) . Excess NADH shifts the redox state of
hepatocytes, which in turn affects other NAD+ -dependent processes, including lipid and carbohydrate
metabolism. One consequence of the redox shift is hepatic steatosis. Excess reducing equivalents
provoke steatosis by stimulating fatty acid synthesis and inhibiting fatty acid oxidation; these compounds
then accumulate in the hepatocyte cytoplasm, where they are esterified and stored as triglyceride. NADH
also interferes with gluconeogenesis by limiting the availability of the substrates oxaloacetate, pyruvate,
and dihydroxyacetone-phosphate and by inhibiting the activity of gluconeogenic enzymes. In patients
with underlying carbohydrate malnutrition, disturbances in this pathway can lead to profound
hypoglycemia.
The redox shift and the resulting metabolic disturbances are acute consequences of ethanol oxidation and
thus are reversible with abstinence. Chronic ethanol consumption, however, can prolong the redox shift
by damaging hepatocyte mitochondria
TABLE 71-2 -- Proposed Mechanisms of Alcoholic Liver Injury
Mechanism Result
Toxic/Metabolic Abnormalities
Resulting from Ethanol Metabolism
Redox shift Steatosis
Oxidant stress Lipid peroxidation
Mitochondrial dysfunction
Microvesicular steatosis
Fibrosis
Acetaldehyde effects Altered protein trafficking
Autoimmune responses
Fibrosis
Immune/Inflammatory Abnormalities
Cytokine production and Kupffer cell activation Cytotoxicity
Inflammation
Fibrosis

Autoimmunity to acetaldehyde-and hydroxyethyl Inflammation

radical-modified proteins Fibrosis
and preventing reoxidation of NADH to NAD+ (to be described).
Oxidant Stress
Ethanol oxidation leads to formation of several free radical species in hepatocytes, including the
hydroxyethyl radical, the superoxide anion (O2 - ), and the hydroxyl radical (OH ). These
ethanol-induced free radicals inflict oxidative damage on a wide range of intracellular compounds. [16]
Ethanol-induced radical formation is typically attributed to ethanol oxidation by CYP2E1 (discussed
earlier); ADH may also contribute indirectly by means of the redox shift. Excess NADH promotes
mobilization of iron from ferritin, which in the reduced state can interact with hydrogen peroxide to form
hydroxyl radical. Inflammatory cells also contribute to ethanol-induced oxidant stress in the liver by
producing superoxide. Ethanol sensitizes both Kupffer cells and neutrophils to produce superoxide, both
spontaneously and in response to other activating stimuli.
Free radical attack on unsaturated lipids initiates a chain reaction of lipid peroxidation. In experimental
animals, ethanol feeding induces hepatic lipid peroxidation, an event that is associated with both acute
liver injury and fibrosis. [17] [18] [19] In animal models of ethanol-induced liver injury, lipid peroxidation is
enhanced by adding polyunsaturated fat to the diet; polyunsaturated fat is itself an inducer of CYP2E1
and has the added effect of altering hepatic membrane lipid composition in favor of peroxidation. In the
absence of polyunsaturated fat, ethanol induces hepatic lipid peroxidation but to a lesser extent. Both
ethanol-induced lipid peroxidation and liver injury can be abrogated in rats by administering ethanol
along with an inhibitor of CYP2E1. [20]
Like lipids, DNA is sensitive to oxidant stress. Mitochondrial DNA is more susceptible than nuclear
DNA to oxidative damage because of reduced protection by histone and nonhistone proteins and because
of a decreased capacity for repair. Oxidants can cause both deletions and mutations in mitochondrial
DNA; these in turn can result in mitochondrial dysfunction. Although chronic ethanol feeding does not
produce significant abnormalities in mitochondrial DNA content and transcription in rats, one study in
humans identified a mitochondrial DNA deletion in alcoholics with microvesicular steatosis. [21]
Although in humans it remains uncertain whether ethanol induces sufficient free radical production to
cause liver injury, the effects of radicals may be amplified if ethanol also reduces hepatic antioxidant
defenses. Chronic alcohol consumption indeed leads to depletion of several antioxidants in the liver,
including vitamins A and E and glutathione. Ethanol-induced vitamin E deficiency enhances hepatic lipid
peroxidation. Vitamin A depletion causes lysosomal damage. Ethanol-induced glutathione depletion,
which occurs selectively in hepatic mitochondria, impairs mitochondrial function. Ethanol depletes
hepatic mitochondrial glutathione stores preferentially by inhibiting the normal transport of glutathione
into the organelle from cytosol. S-adenosyl-methionine (SAM), a glutathione precursor, can replenish
mitochondrial glutathione stores and correct some of the functional alterations. [22] [23] Interestingly, the
beneficial effect of SAM does not appear related to its conversion to glutathione. SAM may instead act
by preventing or reversing the glutathione transport defect. [23]
1202

Acetaldehyde Effects
Although acetaldehyde is usually metabolized rapidly in the liver to acetate, its metabolism is slowed in
alcoholics and can lead to acetaldehyde accumulation. If acetaldehyde reaches high enough
concentration, it can become a substrate for the enzymes aldehyde oxidase and xanthine oxidase, which
produce free radicals. Acetaldehyde impairs mitochondrial beta-oxidation of fatty acids (to be discussed);
it can also react with specific amino acid residues on cellular proteins to form acetaldehyde-protein
adducts. Acetaldehyde-protein adducts are demonstrable in the livers of alcoholics [24] [25] and alcohol-fed
animals. [26] In most instances, the adducts localize preferentially to the pericentral zone, where liver
injury is most pronounced. Aldehyde-protein adducts may contribute directly to alcoholic liver fibrosis
by stimulating hepatic collagen synthesis (to be discussed).
If acetaldehyde forms adducts with cytoskeletal proteins such as tubulin, it can impair microtubule
assembly, which may in turn disturb critical transport processes in hepatocytes. These include both
receptor and non-receptor-mediated endocytosis as well as protein secretion. Acetaldehyde-induced
impairment of protein secretion has been implicated as the major event underlying hepatocellular
swelling ("ballooning") in alcoholic liver disease. [27] Acetaldehyde-protein adducts may also contribute
to liver injury by forming neoantigens that provoke immune responses (to be discussed).
Mitochondria as Targets of Ethanol Toxicity
Mitochondrial enlargement has long been recognized as a feature of alcoholic liver injury. Giant
mitochondria, or "megamitochondria," are present in approximately 25% of patients with alcoholic liver
disease, in comparison with 1% or fewer of patients with nonalcoholic liver disease. [28] [29]
Ethanol-induced changes in hepatic mitochondrial structure are accompanied by alterations in
mitochondrial function; for example, oxidative phosphorylation and the citric acid cycle are inhibited by
as much as 40%. Ethanol-induced impairments in mitochondrial function appear unique to the liver.
Either they are not observed or they manifest differently in mitochondria from brain and muscle. This
suggests that ethanol metabolism is a prerequisite to the type of mitochondrial injury observed in liver.
Ethanol-induced mitochondrial toxicity may be caused by alterations in mitochondrial membrane lipids,
mitochondrial DNA, or mitochondrial protein synthesis. Both acetaldehyde and free radicals are capable
of inducing mitochondrial damage; damage by the latter is accentuated by mitochondrial glutathione
depletion. The ominous nature of these changes is evidenced by their appearance also in severe
drug-induced toxicity, including that caused by tetracycline and fluoroiodoarabinouracil, among others
(see Chapter 73) . These drugs typically cause microvesicular steatosis, consistent with a reduction in
fatty acid oxidation. [30] Interestingly, microvesicular fat is now recognized with increasing frequency in
alcoholic liver disease. [31] Moreover, microvesicular steatosis, sometimes called alcoholic foamy
degeneration, is generally found in alcoholic livers containing megamitochondria. [29]
The appearance of both megamitochondria and microvesicular steatosis can portend a fatal outcome in
certain drug-induced hepatitides; however, in alcoholics they do not. The two are generally observed in
patients with mild to moderate clinical disease, suggesting that they are early features of alcohol-induced
liver injury. [28] [29]
Immune/Inflammatory Mechanisms of Liver Injury

Cytokine Effects
Several proinflammatory cytokines, because of their presence in either the liver or the plasma of patients
with alcoholic hepatitis, have been implicated in the pathogenesis of alcohol-induced hepatic
inflammation. Over 75% of patients with alcoholic hepatitis have elevated plasma levels of interleukin-1
(IL-1), interleukin-6 (IL-6), interleukin-8 (IL-8), and tumor necrosis factor-alpha (TNF-alpha). [32]
Plasma IL-8 and TNF-alpha correlate inversely with prognosis. IL-1 levels in plasma, in contrast, are
elevated to a similar degree in patients with both alcoholic hepatitis and inactive cirrhosis.
Because cytokines are often cleared by the liver, their presence in the circulation does not necessarily
indicate a role in liver disease pathogenesis. Consequently, attention has focused on confirming which
cytokines are actually produced in the alcoholic liver and regulated in a time frame commensurate with
liver injury. In rats, TNF-alpha messenger RNA is induced in the liver in response to ethanol feeding.
TNF-alpha messenger RNA appears to derive primarily from Kupffer cells, and its expression precedes
the onset of liver cell necrosis and inflammation. The precise stimulus to TNF-alpha production is
unknown, although endotoxin is a potential candidate (to be discussed). TNF-alpha may cause liver
injury either directly or indirectly by promoting leukocyte adherence and activation. TNF-alpha also
induces hepatocytes to produce the neutrophil chemoattractant IL-8.
Chronic ethanol ingestion also induces hepatic production of IL-6 and TGF-beta. As with TNF-alpha,
Kupffer cells are a major source of these cytokines; ethanol induces two- to four-fold increases in IL-6
and transforming growth factor-beta (TGF-beta) secretion. IL-6 and TGF-beta are of particular relevance
to alcohol-induced hepatic fibrosis, to be discussed.
Yet another cytokine that is up-regulated in the livers of alcoholics is IL-8. IL-8 is produced both by
hepatocytes and by Kupffer cells. Because of its potent chemoattractant activity toward neutrophils, and
because of the prominence of neutrophils in alcohol-induced hepatic inflammation, IL-8 is under active
study as a mediator of alcoholic hepatitis. In rats, chronic ethanol feeding induces IL-8 production by
hepatocytes and Kupffer cells. [33] [34] The mechanism whereby ethanol leads to hepatic IL-8 production
is unknown, although the IL-8 gene is inducible by oxidants and by TNF-alpha. Other neutrophil
chemoattractants are also produced in the alcoholic liver, including
19-hydroperoxy-,20-hydroxy-arachidic acid and 4-hydroxynonenal. Each of these compounds may
contribute to alcohol-induced hepatic inflammation.
Role of Kupffer Cells and Endotoxin
Because Kupffer cells are a major source of inflammatory and fibrogenic cytokines in the alcoholic liver,
ethanol-induced
1203
activation of these cells is believed to be an important pathophysiologic event. Indeed, when Kupffer
cells are selectively destroyed in the rat liver in vivo, subsequent ethanol feeding fails to induce hepatic
steatosis or inflammation. [35] Ethanol alone does not appear to activate Kupffer cells; however, it does
"prime" them to react more intensely to other stimuli. For example, when Kupffer cells from ethanol-fed
animals are incubated in vitro with endotoxin, they release more superoxide and produce more
TNF-alpha messenger RNA than do Kupffer cells from normal animals.

Endotoxin may be an important mediator of Kupffer cell activation in alcoholics in vivo. Chronic ethanol
ingestion induces low-level endotoxemia, presumably by increasing intestinal permeability. If portal
endotoxin encounters Kupffer cells that have been primed by ethanol, it could induce exaggerated release
of superoxide and cytokines and in turn cause liver injury. A connection between Kupffer cells and
endotoxin in alcoholic liver injury is supported by the results of two studies in rats. In one the researchers
used nonabsorbable antibiotics [36] and in the other, lactobacillus [37] to prevent endotoxemia during
ethanol feeding. Both treatments substantially reduced liver injury in the ethanol-fed animals.
Immune Responses to Altered
Hepatocellular Proteins
By forming adducts with either acetaldehyde or hydroxyethyl radicals, hepatocellular proteins can be
altered sufficiently to provoke immune responses. In this context, aldehyde- and radical-modified
proteins are referred to as neoantigens. Antibodies directed against neoantigens are detectable in the
blood of alcoholic patients; [38] [39] [40] cellular immune responses have also been documented in vitro.
The role of autoimmunity in the pathogenesis of alcoholic liver injury is still unclear, because the
location of the neoantigens does not lend itself to targeting of the immune response to the liver. Many of
the antigens either are retained within hepatocytes or immune effector cells or circulate freely. However,
some neoantigens are present on the hepatocyte surface and in the hepatic extracellular space, which
would permit local access and activity of both antibodies and immune effector cells. Results of
experiments with guinea pigs support the notion that autoantibodies are pathogenic in alcoholic liver
disease. In these studies, ethanol was fed to guinea pigs that had been immunized previously with
acetaldehyde-protein adducts. The guinea pigs with circulating autoantibodies developed signs of hepatic
injury within 40 days of ethanol feeding and exhibited hepatic fibrosis after 90 days. [41] [42]
Mechanisms of Fibrosis
Liver fibrosis is a serious and potentially irreversible consequence of chronic ethanol use. Fibrosis occurs
in only 10% to 15% of alcoholics, but among alcoholics with evidence of liver disease, it is found in
almost 50%. [31] [43] Central to the pathophysiology of alcoholic liver fibrosis is activation of hepatic
stellate cells. Stellate cells (formerly referred to as Ito cells, fat-storing cells, perisinusoidal cells, or
lipocytes) reside in the space of Disse between hepatocytes and sinusoidal endothelia. In normal liver,
stellate cells exhibit a quiescent phenotype and play an important role in hepatic vitamin A storage. In
liver injury, however, whether caused by alcohol abuse or by other toxic or infectious insults, the
phenotype of stellate cells is altered in such a way that they become actively proliferative,
myofibroblast-like cells. [44] Activated stellate cells are the principal collagen-producing cells of liver.
They are responsible for the perisinusoidal fibrosis that is characteristic of alcoholic liver disease.
The precise stimulus that initiates stellate cell activation in vivo is unknown. However, a number of
compounds present in the livers of alcoholics can perpetuate or enhance stellate cell activation and
collagen synthesis. Acetaldehyde and aldehyde-protein adducts increase stellate cell collagen synthesis in
culture. Products of lipid peroxidation are also reported to stimulate collagen synthesis, although this
remains controversial. Yet another potential stimulus to alcoholic liver fibrosis is the cytokine TGF-beta.
It is a potent inducer of stellate cell collagen synthesis in culture, and stellate cells from ethanol-fed rats
have increased sensitivity to its fibrogenic effects.

In rats, chronic ethanol feeding induces TGF-beta production by Kupffer cells. Ethanol feeding also
stimulates Kupffer cells to produce TNF-alpha and IL-6, which can promote fibrosis independently by
increasing stellate cell collagen synthesis and by inducing stellate cell proliferation. Moreover, once
stellate cells undergo activation, they may themselves produce TGF-beta, which may amplify and
perpetuate the fibrogenic process in an autocrine manner.
Putative Mechanisms Underlying the Zonal Distribution of Alcoholic Liver Damage
One of the hallmarks of alcoholic liver disease is that hepatocellular injury affects the pericentral zones
more than the periportal zones. There are several potential reasons for this; one is that the pericentral
zone is the primary site of ethanol metabolism. CYP2E1 is much more abundant in pericentral than in
periportal hepatocytes; the enzyme and its messenger RNA are concentrated in this region and maintain
their pericentral distribution even after induction by ethanol. Traditionally, ADH was also believed to be
concentrated in pericentral hepatocytes. Studies in both rats and humans have challenged this notion,
suggesting either an even lobular distribution or differences in lobular distribution, depending on age and
gender.
Another means by which ethanol may preferentially damage pericentral hepatocytes is through tissue
hypoxia. There is evidence that chronic ethanol consumption induces a hypermetabolic state in the liver
with increased oxygen consumption by liver cells. The hypermetabolic state has been postulated to
enhance the portal-to-central oxygen gradient, leaving pericentral hepatocytes in a state of relative
hypoxia. Pericentral hypoxia has been documented in human alcoholics; [45] however, it may occur only
during periods of abstinence or withdrawal. When ethanol is present in the blood stream, hepatic oxygen
consumption increases but is offset by a concomitant increase in splanchnic blood flow.
Despite the apparent compensation of increased oxygen demand with increased blood flow, it appears
that oxygen use by liver cells is adversely affected by ethanol and that this effect is concentrated in the
pericentral zone. This latter effect may lead to an exaggerated decrease in hepatic adenosine
1204
triphosphate (ATP) production and liver injury during periods of oxygen deprivation.


Saunders Company
COFACTORS INFLUENCING THE DEVELOPMENT OF ALCOHOLIC

LIVER DISEASE
Because serious liver disease develops in only a small proportion of alcoholics, [9] factors other than
ethanol itself must play a role in pathogenesis. Recent years have witnessed an intense search for
hereditary clues, whereby histocompatibility antigens have been thoroughly investigated as potential
predictors of alcoholic liver injury. Human leukocyte antigens (HLA)-A1, -A9, -A28, -B13, -B15,
-BW35, and -B40 all have been noted with increased frequency in patients with alcoholic cirrhosis. [46]
[47] [48] [49] However, none of these associations has been confirmed on repeat testing; [50] [51] [52] therefore,
researchers are investigating other areas.
Hereditary Variations in Ethanol Metabolism
Polymorphisms in ADH, CYP2E1, and ALDH are under investigation as risk factors for alcoholic liver
injury. In the case of ADH, the ADH2*2 and ADH3*1 alleles (which encode enzymes that metabolize
ethanol rapidly) have been of greatest interest. ADH2*2 and ADH3*1 are prevalent in Chinese and
Japanese populations. [53] [54] [55] In Chinese populations, ADH2*2 and ADH3*1 are actually less
common among alcoholics than among nonalcoholics, which indicates a negative association with
ethanol dependence and suggests that rapid metabolism leads to ethanol avoidance. [55] In this population,
there is no association between either ADH2*2 or ADH3*1 and alcoholic liver injury. [55] Among
northern Europeans, ADH2*2 is almost nonexistent, but ADH3*1 has an allele frequency of
approximately 55%. [56] One study of a British population reported a slightly but significantly higher
frequency of ADH3*1 among patients with alcoholic cirrhosis. [56] To date, this has not been confirmed.
[57]
A polymorphism in CYP2E1 has been identified in a region of the gene that controls transcription.
Persons with that rare allele, called c2, have higher baseline CYP2E1 activity than do persons without
this allele. [58] Although studies from the United States [59] and Europe [60] do not demonstrate an
association between the c2 allele and alcoholic liver injury, these populations have a very low baseline
frequency of c2. One study from Japan, where the c2 allele is much more common, indicates that 84% of
patients with alcoholic liver disease are either heterozygous or homozygous for c2. The c2 allele is more
than twice as common among patients with alcoholic liver disease than among patients with nonalcoholic
liver disease or healthy controls. [58] However, another group cautioned that the increase in the c2 allele
frequency is observed only in patients with mild alcoholic liver disease and not in patients with cirrhosis.
[61]
The ALDH2*2 allele has also been implicated in the development of alcoholic liver disease. ALDH2*2
homozygotes generally have an aversion to ethanol because of acetaldehyde toxicity; ALDH2*1/2*2
heterozygotes, however, do occasionally abuse ethanol and develop liver injury. One study from Japan
suggests that ALDH2*1/2*2 heterozygotes develop liver injury with higher frequency and at a lower
dose than ALDH2*1/2*1 homozygotes. [62] Similar findings have not been reported from China, despite a
larger study population. [55]

Gender
Women are highly susceptible to serious alcoholic liver injury, developing cirrhosis at a much lower
cumulative dose of ethanol than men. [6] [7] [8] Not only are women at increased risk of acquiring alcoholic
liver injury, but they also exhibit a tendency toward disease progression even during abstinence. [63] [64]
This gender-specific difference in the risk of alcoholic liver disease is unexplained. One theory
implicates gastric ADH as a causative factor. Because women have lower levels of gastric ADH activity
than do men, [12] a larger proportion of ingested ethanol may be absorbed in the liver, causing earlier
signs of liver toxicity. The validity of this theory depends largely on the relative contribution of gastric
ADH to total ethanol metabolism (discussed earlier).
Accelerated alcoholic liver injury in women may also be related to gender-specific differences in fatty
acid metabolism. If the fatty acids that accumulate in liver cells as a result of impairment of
beta-oxidation are not converted to triglyceride, they can induce liver injury. This problem may be
circumvented by diversion of the fatty acids to alternative routes of metabolism such as cytochrome
P-450 4A1-mediated omega-hydroxylation. In rats, this compensatory pathway is efficiently up-regulated
in males but not in females. Fatty acid-binding capacity is also reduced in female rats that have been fed
ethanol chronically, which may contribute to fatty acid toxicity.
Diet/Nutrition
Studies in baboons indicate that ethanol can induce liver injury despite adequate protein-calorie and
vitamin nutrition. In humans, however, alcoholic liver injury correlates strongly and inversely with
nutritional status. [65] Even if dietary factors are not the sole cause of alcoholic liver injury, they are likely
to impact its development and progression. Inadequate intake of antioxidant vitamins may enhance the
oxidant stress of ethanol metabolism (discussed earlier). On the other hand, excess intake of other
nutrients such as polyunsaturated fat may also increase ethanol-induced oxidant stress by up-regulating
CYP2E1 and permitting the accumulation in liver of substrates for ethanol-induced lipid peroxidation. [17]
Indeed, there is epidemiologic evidence that alcoholic liver disease is more common among populations
that consume large amounts of polyunsaturated fat. [66] Chronic ethanol ingestion promotes absorption of
iron from the intestine and increases hepatic iron stores. Because iron is an important catalyst of free
radical production, it may add to the oxidant stress induced by ethanol metabolism.
Viral Liver Disease
Approximately 18% to 25% of alcoholics are infected with the hepatitis C virus (HCV). [67] [68] [69] [70] In
alcoholics with liver disease, the prevalence of HCV is even higher; some studies
1205
have reported seropositivity rates of 40% or more. [67] [68] Not all groups confirm such a high prevalence
of HCV among alcoholics with liver disease; however, there is general agreement that alcoholics
coinfected with HCV develop liver injury at a younger age and at a lower cumulative dose of ethanol
than those without hepatitis C infection. [69] [70] This may be related to alcohol effects on HCV replication
or on the host immune response to the virus. [71] [72]

Viral hepatitis, whether caused by HCV or HBV, increases the incidence of chronic liver injury in
alcoholics. Epidemiologic studies suggest that viral infection and alcohol abuse pose additive, rather than
synergistic, risks for liver disease. [2] [3] [69]


Saunders Company
ETHANOL METABOLISM
Hepatic Metabolism
The liver is the primary site of ethanol metabolism. Within the liver, ethanol can be oxidized by three
enzyme systems:
Figure 71-1 Age-adjusted death rates from liver cirrhosis (all International Classification of Diseases
code 571): Death Registration, States, 1910-1932, and United States, 1933-1988. (From Grant, B. F.,
DeBakey, S., Zobeck, T. S. Liver Cirrhosis Mortality in the United States, 1973-1988. NIAAA
Surveillance Report No. 18. U.S. Department of Health and Human Services [DHHS] Pub. No.
[ADM]281-89-0001, 1991.)
* Satisfying criteria for alcohol abuse, alcohol dependence, or both, from the Diagnostic and Statistical Manual of Mental
Disorders, Third Edition, Revised ( DSM- III- R); Washington, DC, American Psychiatric Association, 1987.
1200
TABLE 71-1 -- Class I Alcohol Dehydrogenase (ADH) Genes and Their Encoded Enzyme Subunits S
ADH1:alpha ADH2:beta ADH3:gamma
alpha
ADH2*1 beta1 ADH3*1 gamma1
ADH2*2 beta2 ADH3*2 gamma2
ADH3*3 gamma3
[1]Subunits can form homodimers (e.g., alphaalpha, beta1 beta1 , gamma2 gamma2 ) or heterodimers (e.g., alphabeta1 ,
beta2 gamma1 , alphagamma3 ).
the alcohol dehydrogenases (ADH), the microsomal ethanol oxidizing system (MEOS), and catalase.
Catalase, present in peroxisomes and mitochondria, is the least used of the three pathways. The
remaining two enzyme systems oxidize the bulk of ethanol that reaches the liver.
The ADHs are cytoplasmic enzymes with numerous isoforms in human liver. [9] The class I enzymes,
which have the lowest Michaelis constant (Km) and the highest substrate specificity for ethanol, are
encoded by three separate genes designated ADH1, ADH2, and ADH3. These genes are translated into
peptide subunits named alpha, beta, and gamma, respectively; typically, the active enzymes are formed
by homodimerization of the subunits (alphaalpha, betabeta, gammagamma). Heterodimers can

sometimes form; moreover, to add to the complexity of the system, polymorphisms that have been
identified in ADH2 and ADH3 permit numerous subunit combinations (Table 71-1) . Variations in ADH
isoforms can account for significant differences in ethanol elimination rates among ethnic groups. People
of Asian descent, for example, who typically have the beta2 ADH subunit, metabolize ethanol 20% faster
than people of northern European descent, who typically possess the beta1 subunit. [10]
ADH is the enzyme responsible for alcohol metabolism when blood and tissue ethanol concentrations are
low. However, when tissue levels exceed 10 mmol/L (approximately 50 mg/dL), MEOS can also
contribute. A critical component of MEOS is cytochrome P-450 2E1 (CYP2E1); this enzyme catalyzes
not only ethanol oxidation but also the metabolism of other drugs such as acetaminophen, haloalkanes,
and nitrosamines. Chronic ethanol consumption up-regulates CYP2E1. Activity of the enzyme increases
as much as five- to ten-fold, as a result of increases in both CYP2E1 messenger RNA and protein. This is
likely to account for the more rapid elimination of ethanol observed in chronic alcoholic patients.
Although MEOS is much more likely to contribute to ethanol metabolism in alcoholics than in
occasional drinkers, its relative contribution in comparison with ADH is difficult to determine.
Nevertheless, MEOS is important to the pathogenesis of alcoholic liver injury because
CYP2E1-mediated ethanol oxidation yields reactive oxygen intermediates as by-products (Fig. 71-2) .
These oxygen-derived free radicals are capable of provoking hepatocellular damage (discussed later).
Both ADH and CYP2E1 convert ethanol to acetaldehyde (see Fig. 71-2) . Acetaldehyde is then oxidized
to acetate, primarily by a low-Km aldehyde dehydrogenase in hepatocyte mitochondria designated
ALDH2. In rare instances, acetaldehyde is oxidized by alternative pathways involving aldehyde oxidase
or xanthine oxidase. The latter enzymes form oxygen radicals in the process of acetaldehyde oxidation.
Acetaldehyde is a highly reactive and potentially toxic compound. Fortunately, equilibrium conditions in
the liver for aldehyde dehydrogenase strongly favor acetaldehyde elimination. When ALDH2 activity is
decreased, acetaldehyde accumulates in the liver and circulation; this can produce symptoms of flushing,
tachycardia, and even circulatory collapse. Approximately half of Japanese and Chinese people are
deficient in aldehyde dehydrogenase. [11] Deficiency results from inheritance of the ALDH2*2 allele,
which encodes a completely inactive enzyme.
Persons homozygous for ALDH2*2 rarely consume ethanol, because of the severe side effects of
acetaldehyde.
Gastric Metabolism
Although the liver is the primary site of alcohol metabolism, ethanol can also be oxidized in the
gastrointestinal tract. ADH isoenzymes with subunits different from those described earlier have been
identified in the stomach and intestine. Gastric ADH has been implicated in a so-called gastric first-pass
metabolism of ethanol; proponents of this pathway contend that the enzyme, by oxidizing ethanol in the
gastric mucosa, can limit ethanol delivery to the portal circulation. This in turn influences the amount of
ethanol that reaches the liver, potentially impacting the development of alcoholic liver injury.
Gastric ADH activity is reported to be lower in women than in men and to be depressed in both male and
female alcoholics. [12] Certain drugs, such as aspirin and H2 receptor blockers, also inhibit gastric ADH
activity. [13] [14] Several studies indicate that when gastric ADH activity is reduced, alcohol ingestion

results in higher blood ethanol levels. Despite these observations, the importance of gastric ADH and
gastric first-pass ethanol metabolism is debated. [15] Much of the controversy surrounds the fact that
blood ethanol concentrations, which are systemic rather than portal, are profoundly influenced by the rate
of ethanol absorption from the gut. Ethanol absorption from the gastrointestinal tract is in turn affected
by feeding.
Figure 71-2 Ethanol metabolism in the liver is catalyzed primarily by alcohol dehydrogenase (ADH)
and CYP2E1. ADH predominates at low ethanol concentration; at concentrations above 10 mmol,
CYP2E1 also contributes. Both enzymes convert ethanol to acetaldehyde. ADH-mediated ethanol
oxidation is coupled to the conversion of oxidized nicotinic-adenine dinucleotide (NAD+ ) to reduced NAD (NADH).
CYP2E1 utilizes reduced nicotinamide-adenine dinucleotide phosphate (NADPH) and forms oxygen-derived free radicals
as a by-product. Acetaldehyde is oxidized to acetate primarily by ALDH. Aldehyde oxidase and xanthine oxidase, which
produce oxyradicals, are used only rarely.
1201
During fasting, large amounts of ethanol can be delivered rapidly to the liver; this may exceed the liver's
capacity for first-pass metabolism and result in high blood ethanol concentrations. In contrast, in the fed
state, the same dose of ethanol is likely to be delivered more slowly to the liver. This would permit
greater hepatic first-pass metabolism and result in lower blood ethanol levels. Whether the blood ethanol
concentrations measured after ethanol ingestion reflect gastric ethanol metabolism or merely differences
in ethanol absorption remains unsettled.


Saunders Company
Figure 71-1 Age-adjusted death rates from liver cirrhosis (all International Classification of Diseases
code 571): Death Registration, States, 1910-1932, and United States, 1933-1988. (From Grant, B. F.,
DeBakey, S., Zobeck, T. S. Liver Cirrhosis Mortality in the United States, 1973-1988. NIAAA
Surveillance Report No. 18. U.S. Department of Health and Human Services [DHHS] Pub. No.
[ADM]281-89-0001, 1991.)


Saunders Company
EPIDEMIOLOGY
Alcohol is one of the most common drugs of abuse, with 15.3 million people in the United States
suffering from alcoholism. * Per capita alcohol consumption among the U.S. population is estimated at 10
L of pure ethanol per year; even higher amounts are reported in northern Europe, with the populations of
France and Spain reaching 16 L annually. Alcohol use is not as common in Asia as in Europe, the United
States, and Australia; however, studies indicate that both per capita ethanol consumption and alcohol
abuse are rising in Japan. [1]
In countries from around the world there is a direct correlation between liver-related mortality and per
capita ethanol consumption. [2] [3] [4] In the United States, the connection between ethanol consumption
and liver disease is clearly demonstrable from health statistics that span the era of Prohibition (1916 to
1932) (Fig. 71-1) . In the post-World War II era, cirrhosis-related mortality rates rose steadily in the
United States, peaking in 1973 and declining steadily thereafter. The current downward trend is
unexplained, because per capita alcohol consumption has remained steady over the last 20 years and is
still 20% higher than in the 1950s.
Alcoholic liver disease in the United States is estimated to affect more than 2 million people. [1] Of the
26,000 people who die from cirrhosis each year, at least 40% and perhaps as many as 90% have a history
of alcohol abuse. Liver disease appears to develop only after a "threshold" dose of ethanol has been
consumed; the threshold is estimated at 600 kg for men and 150 to 300 kg for women. [5] [6] [7] [8] In order
to achieve a 600-kg cumulative dose of ethanol, one must consume eight 12-oz beers, 1 L of wine, or 0.5
pint of distilled spirits daily for a period of 20 years. Almost all people who exceed this threshold dose of
ethanol exhibit some biochemical or histologic abnormality suggestive of liver injury. Excessive alcohol
consumption can induce a spectrum of abnormalities in the liver ranging from steatosis to alcoholic
hepatitis to hepatic fibrosis and cirrhosis. Of note, however, is that fewer than 50% of people who ingest
even the calculated threshold dose of ethanol eventually develop serious alcoholic liver disease (e.g.,
alcoholic hepatitis or fibrosis). Moreover, once the threshold dose has been exceeded, there is not
necessarily a strong correlation between the amount of ethanol consumed and the severity of liver
disease. [7] The relatively low incidence of serious liver injury in alcoholics, along with the poor
correlation between ethanol dose and disease severity, suggests that the pathogenesis of alcoholic liver
disease is likely to involve hereditary or environmental factors or both.


Saunders Company
DIAGNOSIS
Alcohol should be strongly suspected as a cause of liver disease in any patient who consumes more than
80 g of ethanol daily. Although the risk is lower among persons who drink less, confounding factors such
as gender should influence the prudent physician to consider alcoholic liver disease even at daily doses of
20 to 40 g (two drinks or more per day). The physician must also take into account the fact that patients
often under-report ethanol intake. Corroboration of the drinking history by an objective outsider is a
helpful adjunct to diagnosis.
Physical Findings
The most extensive demographic information on alcoholic liver disease in the United States comes from
studies of hospitalized patients who have been assigned a diagnosis on the basis of clinical and histologic
parameters. [31] [43] Although these studies exclude a potentially large number of patients who are
asymptomatic, [73] [74] they provide a useful guide to diagnosis. The most common physical finding in
persons with alcoholic liver disease is hepatomegaly. Liver enlargement is detectable in over 75% of
patients, regardless of disease severity. [31] [43] Jaundice and ascites are also found in approximately 60%
of patients but are more prevalent among those with severe disease (Table 71-3) (Table Not Available) .
[31] [43] The incidence of jaundice in the moderate and severe groups in Table 71-3 (Table Not Available)
is high because moderate disease was defined by a bilirubin level of more than 5 mg/dL. However, even
when patients are stratified solely by histologic criteria, the incidence of jaundice is found to increase as
disease progresses. [31]
Other clinical features are significant in patients with alcoholic liver disease. First, an unusually large
proportion of patients exhibit hepatic encephalopathy (44.6%), even with mild disease (27.3%). Chedid
and colleagues [31] confirmed this; they reported a 19% incidence of encephalopathy and a 28% incidence
of portal hypertension among patients who on biopsy were found to have nothing more than fatty liver.
Portal hypertension has been reported previously in patients with hepatic steatosis. [75] It may be related
to compression of the hepatic sinusoids by enlarged hepatocytes. [76]
Also of note is that approximately 25% of patients with alcoholic hepatitis present with fever. Only 9.2%
have demonstrable infections; this suggests a possible contribution by ethanol-induced cytokines. [32]
TABLE 71-3 -- Signs and Symptoms in Hospitalized Patients with Alcoholic Liver Disease
(Not Available)
Data from Mendenhall, C.L.Alcoholic hepatitis. Clin. Gastroenterol. 10:420, 1981.
Laboratory Findings
Seventy-five percent or more of patients with alcoholic liver disease have macrocytic anemia (Table

71-4) (Table Not Available) . Leukocytosis is also common, with a mean white blood cell count of
12,400 cells per cubic millimeter. Hepatic transaminase levels are only modestly elevated; aspartate
aminotransferase (AST) and alanine aminotransferase (ALT) levels rarely exceed 300 U/L and do not
correlate strongly with disease severity. [31] [43]
The AST/ALT ratio often exceeds 2 in patients with alcoholic liver disease (Fig. 71-3) . [31] [77] This,
together with the relatively low values of AST and ALT, helps distinguish alcoholic from viral or other
toxic liver diseases (see Chapter 67) . Although an AST/ALT ratio higher than 1 is used by some
investigators to predict alcoholic liver disease, Figure 71-3 demonstrates that in the range between 1 and
2, there is substantial overlap with viral hepatitis and postnecrotic cirrhosis. A ratio of 3 or higher
strongly suggests alcoholic liver disease.
The high AST/ALT ratio in alcoholics with liver disease
TABLE 71-4 -- Laboratory Values in Hospitalized Patients with Alcoholic Liver Disease
(Not Available)
Data from Mendenhall, C.L.Alcoholic hepatitis. Clin. Gastroenterol. 10:422, 1981. Moderate disease
was defined by bilirubin>5 mg/dl; severe disease, by bilirubin>5 mg/dl and prothrombin time>4 sec
prolonged. MCV, mean corpuscular volume; WBC, white blood cell count; AST, aspirate
aminotransaminase; ALT, alanine aminotransaminase.
1206
Figure 71-3 Serum glutamic-oxaloacetic transaminase (SGOT)/serum glutamate pyruvate transaminase

(SGPT) ratios in patients with biopsy-proven liver disease. The ratio was calculated from the serum
sample that had the highest individual transaminase value, be it SGOT or SGPT. (Reprinted from Cohen,
J. A., and Kaplan, M. M. The SGOT/SGPT ratio--An indicator of alcoholic liver disease. Dig. Dis. Sci.
24:836, 1979. with permission.)
has been attributed to pyridoxine deficiency. Low pyridoxine levels cause a reduction in the AST and
ALT content of hepatocytes, ALT being disproportionately affected. Pyridoxine-induced changes in
hepatocellular AST and ALT occur independently of alcoholic liver injury; consequently, in the setting
of pyridoxine deficiency, any insult that causes hepatocellular necrosis provokes an increase in
transaminases, in which an increase in AST predominates. This mechanism is likely to explain the high
AST/ALT ratio in alcoholics with acetaminophen poisoning. Acetaminophen can cause serious hepatic
necrosis in actively drinking patients, even at therapeutic doses (see Chapter 73) . In alcoholics,
acetaminophen hepatotoxicity is characterized by a dramatic increase in transaminases but with a high
AST/ALT ratio. [78] The fact that severe nonalcoholic liver injury can occur in alcoholics and that the
AST/ALT ratio can mimic that of alcohol-induced disease underscores the importance of scrutinizing the
transaminase values before making a diagnosis. A marked transaminase elevation (>300 U/L) in an
alcoholic, even with an AST/ALT ratio higher than 2, should raise concern about acute nonalcoholic
liver injury.

Bilirubin and prothrombin time (PT) are useful predictors of liver disease severity in alcoholics. The two
parameters have been used to stratify patients with alcoholic liver disease into mild, moderate, and severe
categories (see Tables 71-3 (Table Not Available) and 71-4) (Table Not Available) . [43] Maddrey and
colleagues [79] also used bilirubin and PT to generate a "discriminant function" that identifies patients
with significant short-term mortality:
Discriminant function = 4.6 [PT (seconds) - control] + bilirubin (mg/dL)
A discriminant function higher than 32 predicts a probability of mortality within 1 month of
approximately 50%. Calculation of a discriminant function may be unnecessary in patients with
encephalopathy. Evidence indicates that encephalopathy alone predicts a high probability of short-term
mortality. [80]
Several blood tests are under evaluation as markers of recent alcohol consumption. These include assays
of mitochondrial AST, [81] carbohydrate-deficient transferrin, [82] and antibodies against
acetaldehyde-protein adducts. [83] Collagen propeptides have also been studied, not only as markers of
alcohol consumption, but also as noninvasive markers of liver fibrosis. Interestingly, in some studies
collagen propeptides are better predictors of hepatic inflammation than of fibrosis. [84] [85] [86] [87] For the
most part, these tests remain investigational.
Histology
Liver biopsy is viewed by many physicians as the gold standard for diagnosing alcoholic liver injury.
Indeed, one biopsy study warned that as many as 20% of cases of alcoholic liver disease may be
misdiagnosed from clinical criteria alone. [88] Although the true error rate in diagnosis is probably closer
to 10%, liver biopsy is still quite useful for diagnosis and for predicting prognosis. Among the most
common histologic features of alcoholic liver disease are (1) steatosis, (2) ballooning degeneration of
hepatocytes, (3) Mallory bodies, (4) neutrophilic inflammation, and (5) pericellular fibrosis (Fig. 71-4) .
Steatosis is present in 60% to 95% of patients with alcoholic liver disease. [31] [43] [89] Fat is most
prominent pericentrally, although in severe cases it exhibits a panlobular distribution. Macrovesicular
steatosis is the rule; hepatocytes contain one or more large fat droplets that displace the nucleus to an
Figure 71-4 Light micrograph illustrating acute alcoholic liver injury. Note ballooning degeneration
(b); Mallory body (M); neutrophilic inflammatory infiltrate (open arrows).
1207
Figure 71-5 Light micrograph of alcoholic liver injury illustrating macrovesicular (open arrow) and
microvesicular (solid arrow) steatosis.

eccentric position. Microvesicular steatosis, or alcoholic foamy degeneration, [90] is also being recognized
with increasing frequency (Fig. 71-5) . One study reported microvesicular steatosis in fewer than 5% of
patients with alcoholic liver disease. [91] In a more recent report, the researchers did not estimate the
incidence exactly, but they did contend that microvesicular steatosis is a common accompaniment to
ballooning degeneration found in 73% of biopsies. [31]
Ballooning degeneration of hepatocytes is characterized by marked cell swelling with a pale appearance
to the cytoplasm. Ballooning is a nonspecific marker of hepatocyte injury and is often accompanied by
acidophil bodies. Both features are found in 60% to 90% of patients with alcoholic liver disease. [31] [89]
Also common in alcoholic liver disease are Mallory bodies; these crescent-shaped, eosinophilic
structures represent intermediate filaments that have undergone condensation. Mallory bodies are found
in 70% to 75% of patients undergoing biopsy for alcoholic liver disease. [31] [43] They often wrap around
the nucleus of hepatocytes (see Fig. 71-4) . Despite their prevalence among alcoholics, they are not
pathognomonic of alcoholic liver disease; they also appear in primary biliary cirrhosis, in Wilson's
disease, and in patients taking griseofulvin or amiodarone.
Hepatic inflammation is present in 50% to 85% of patients with alcoholic liver disease. [31] [43]
Neutrophils are a common component of the inflammatory infiltrate, distinguishing alcohol-induced
inflammation from other forms of hepatitis. Neutrophils are commonly found adjacent to ballooned
hepatocytes or cells with Mallory bodies. Mononuclear cells can also be seen; if classic piecemeal
necrosis is observed, it may suggest coexistent viral hepatitis. Some degree of periportal inflammation is
observed in 75% of patients with alcoholic liver injury, which suggests that this lesion may be caused by
ethanol as well. [31]
Fibrosis is found in 50% to 75% of patients with alcoholic liver injury. It begins with deposition of
connective tissue around the terminal hepatic venule and then extends into the hepatic parenchyma in a
pericellular manner ("chicken-wire" fibrosis). As fibrosis advances, broader septa are formed, with
central-central and central-portal bridging. The cirrhosis that evolves is micronodular.
Chedid and colleagues [31] defined four histologic categories of alcoholic liver disease for the purpose of
predicting survival: fatty liver (FL), alcoholic hepatitis (AH), cirrhosis (C), and cirrhosis with alcoholic
hepatitis (C+AH). The two main criteria used to stratify patients into the four categories were hepatic
inflammation (absent in FL and present in AH) and cirrhosis (present in C and C+AH but not in FL or
AH). Using these criteria, they found statistically significant differences in 4-year survival among the
four groups (70% of FL; 58% of AH; 49% of C; 35% of C+AH). These finding parallel data published
earlier by Orrego and colleagues. [92]
Differential Diagnosis
Patients who adamantly deny ethanol consumption can develop an illness that mimics alcoholic hepatitis
both biochemically and histologically. This syndrome, called nonalcoholic steatohepatitis (NASH), was
originally described as a postoperative complication of jejunoileal bypass surgery (see Chapter 72) .
NASH can occur in patients who are obese, diabetic, or hyperlipidemic or as a side effect of total

parenteral nutrition. It can also be induced by certain drugs, including estrogens, diethylstilbestrol,
glucocorticoids, amiodarone, and perhexilene. At present there is no diagnostic test that distinguishes
NASH from true alcoholic liver disease. The clinical features of NASH are benign; most patients lack
symptoms and maintain normal hepatic synthetic function. Liver histologic features are more ominous in
NASH; some degree of fibrosis is found in as many as 40% of cases. [93] [94] [95]
Alcoholic liver disease can be difficult to distinguish from hereditary hemochromatosis in the event of a
high serum iron saturation and siderosis on liver biopsy. In this situation hepatic iron quantitation is
useful, because it permits calculation of the hepatic iron index ([mug hepatic iron 58] age in years).
An index higher than 2 indicates hereditary hemochromatosis (see Chapter 65) .


Saunders Company
COMPLICATIONS
The complications of alcoholic liver disease are similar to those encountered in other types of chronic
liver injury. Among these are ascites, gastrointestinal hemorrhage, and encephalopathy, as well as
hypoalbuminemia and hypoprothrombinemia arising from hepatocellular dysfunction. These problems
and others are discussed in detail in Chapters 77 to 79.
Patients with alcoholic cirrhosis may or may not be at risk of acquiring hepatocellular carcinoma (HCC).
Ethanol-related induction of cytochrome P-450 enhances the metabolism of several compounds to
carcinogenic intermediates (see Chapter 78) ; this mechanism is likely to play an important role in the
pathogenesis of cancers in the luminal gastrointestinal tract. For the liver, however, there are no firm data
implicating alcohol as an independent risk factor for cancer. Two groups
1208
from Japan recently studied alcohol as part of multivariate analyses of risk factors for HCC. [96] [97] The
studies involved a total of 1712 patients, but despite these numbers the results were inconsistent. One
group found alcohol to be an independent risk factor for HCC [96] and the other did not. [97] At present,
some of the strongest risk factors for HCC in alcoholics are male gender, age, and the presence of viral
markers for either HBV or HCV. [98]


Saunders Company
TREATMENT
The mainstay of treatment for alcoholic liver disease is abstinence. Abstinence alone can substantially
improve survival chances in persons with alcoholic liver disease; this is true even if cirrhosis and portal
hypertension are present at the time of diagnosis (to be discussed). In the setting of severe alcoholic liver
disease, however, abstinence and supportive medical care may not be sufficient to improve clinical
status, and pharmacologic therapy may provide an important adjunct. Several agents have been used in
the treatment of patients with alcoholic hepatitis and alcoholic fibrosis. The rationales behind their use,
along with a general assessment of efficacy, are described as follows.
Nutrition
Malnutrition has never been proved to be a direct cause of alcoholic liver injury, but the two certainly go
hand in hand. A strong connection between malnutrition and alcoholic liver disease was demonstrated by
the Veterans Administration (VA) Cooperative Study Group on Alcoholic Hepatitis. This group assessed
the nutritional status of 305 alcoholics with and without liver disease; they found that more than 75% of
patients with severe liver disease had signs of kwashiorkor, marasmus, or both, whereas none of the
patients without liver disease exhibited these degrees of malnutrition (Table 71-5) (Table Not Available)
. [99] In another study, the VA Cooperative Study Group reported that malnutrition worsens the prognosis
of alcoholic hepatitis. They observed a 20% to 50% rate of mortality within 6 months in patients with
alcoholic hepatitis who had severe protein-calorie malnutrition, in sharp contrast to a mortality rate of 0%
to 9% in patients with mild malnutrition. [65] The outcome of severely malnourished patients improved if
their nutritional status was partially corrected during a 30-day hospitalization.
TABLE 71-5 -- Prevalence of Malnutrition in Patients with Alcoholism
(Not Available)
Data from Mendenhall, C.L, Anderson, S., Weesner, R. E., et al. Protein-calorie malnutrition
associated with alcoholic hepatitis, Veterans Administration Cooperative Study Group on Alcoholic
Hepatitis. Am. j. Med. 76:218, 1984.
[65]This concept, that nutritional support could improve the outcome of alcoholic hepatitis, led to several
controlled trials of enteral and parenteral supplements in hospitalized patients. Although individual
studies used different nutritional formulas and different routes of administration, the chosen endpoints
were similar and the results remarkably consistent. Contrary to expectations, nutritional supplements
offered little or no improvement in nutritional status, laboratory parameters of liver injury, or survival
rate [100] [101] [102] [103] [104] [105] (Table 71-6) .
Despite these negative results, enteral supplements may nonetheless be of general value in hospitalized
patients with alcoholic hepatitis. This is particularly true in anorexic patients, who consume less than
75% of their calculated energy and protein requirements per day. [106] Branched-chain amino acid

formulas (e.g., Hepatic-Aid) need not be used in favor of conventional amino acid preparations. The
former are expensive, and the latter have not been proved to cause hepatic encephalopathy, even in
patients with cirrhosis and portal hypertension. [100] [102]
Propylthiouracil
Thyroid hormone can provoke a hypermetabolic state in the liver similar to that induced by ethanol
(discussed earlier). To counteract this ethanol-induced hypermetabolic state and any associated liver
injury, several investigators have pursued antithyroid therapy with propylthiouracil (PTU). PTU was first
tested successfully in ethanol-fed rats; the positive results prompted subsequent trials of PTU in humans.
In the first human study, PTU (300 mg/day) was administered to 103 patients with varying degrees of
alcoholic liver disease. Over a treatment period of 42 days, PTU accelerated the clinical improvement of
the most severely ill patients but did not affect survival rate. [107]
An independent group conducted a second trial of PTU in 67 patients with alcoholic hepatitis. The
second study, which included only patients with severe liver disease, showed no benefit of PTU despite a
dosage and duration of therapy identical to those in the first trial. [108] In both trials, PTU exhibited a
demonstrable antithyroid effect, reflected by a rise in serum thyroid-stimulating hormone.
Despite the inconsistent outcomes of the two short-term PTU trials, a third long-term study was
conducted in outpatients with alcoholic liver disease. In this 2-year trial, patients in the treatment group
received 300 mg of PTU daily in a 3-months-on/1-month-off regimen. PTU significantly reduced
mortality from 25% to 13%. [109] As with the original short-term trial, patients with the most severe liver
disease were afforded the greatest benefit (55% versus 25% mortality in the placebo and PTU groups,
respectively). Patients with mild alcoholic liver injury did not derive any survival benefit from PTU,
insofar as their underlying mortality rate was quite low (3%). [109]
Despite the encouraging results of the long-term study, PTU has not gained acceptance as a treatment for
alcoholic liver disease. This is in part because of concerns that PTU might provoke hypothyroidism and
because of persistent questions about the patient populations in which PTU is effective. The issue of
hypothyroidism was addressed in a follow-up study that demonstrated almost no side effects in patients
taking PTU for up to 4 years. [110] However, concerns about efficacy
1209
TABLE 71-6 -- Studies Evaluating the Effect of Supplemental Enteral or Parenteral Nutrition on
Short-Term Survival in Alcoholic Hepatitis
REFERENCE YEAR N ROUTE FORMULA MORTALITY
100 1980 35 Enteral Amino acids Control: 78%
Amino acids:
100%
101 1985 57 Enteral BCAA Control: 20.6%
BCAA: 16.7%

102 1985 64 Enteral/parenteral BCAA/amino acids Control: 32%

BCAA: 33%
Amino acids:
43%
104 1987 28 Parenteral Amino acids Control: 21%
Amino acids:
7%
105 1991 54 Parenteral Amino acids Control: 19%
Amino acids:
21%
BCAA, branched-chain amino acids.
remain. Studies have shown that the ethanol-induced hypermetabolic state is transient, disappearing
within 30 days of abstinence. Thus if PTU exerts its effect by diminishing the hypermetabolic state, the
effect should manifest most greatly in patients who continue to drink. This was not the case in the
clinical trials; indeed, in order to derive benefit from PTU, patients had to either remain abstinent or
drink only modestly. [109] [110] Why PTU is most effective in these subgroups remains to be clarified.
Corticosteroids
Because of their broad anti-inflammatory and immunosuppressive properties, corticosteroids have been a
popular choice for the treatment of alcoholic hepatitis. Between 1971 and 1989, researchers published at
least 11 placebo-controlled trials in which they examined the effects of corticosteroids on patients with
acute alcoholic liver disease. [79] [111] [112] [113] [114] [115] [116] [117] [118] [119] [120] Although the studies involved a
total of 562 patients, they yielded widely disparate results. Only 4 of the 11 trials demonstrated a
reduction in short-term mortality by corticosteroids. The variable outcomes were attributed to numerous
factors, including gender, liver disease severity, renal function, nutritional status, and even geography. In
an effort to settle the controversy regarding steroid efficacy, Imperiale and McCullough performed a
meta-analysis of the 11 clinical trials. [80] Upon combining all the data, they found that corticosteroids
reduced the relative risk of short-term mortality to 0.63. When they restricted their analysis to include
only the highest quality studies, corticosteroids offered an even more impressive benefit with a reduction
in relative mortality risk to 0.41. [80] Corticosteroids were effective whether administered orally or
intravenously (as prednisolone, 40 mg/day, or methylprednisolone, 32 mg/day). In the trials
demonstrating efficacy, full-dose treatment was continued for 28 to 30 days and followed in most
instances by a 2- to 4-week taper. Serious infections sometimes occurred in corticosteroid-treated
patients, [114] [117] but they were rare and apparently no more frequent than in patients receiving placebo.
[117]
The meta-analysis confirmed a suspicion from several of the individual trials: that corticosteroids are
effective only in a subgroup of alcoholics who have the most severe liver disease. One means of
identifying these patients is the discriminant function of Maddrey (discussed earlier). [79] Encephalopathy
also identifies patients with high rates of short-term mortality; in the meta-analysis, encephalopathy was
the strongest predictor of a corticosteroid response, whereby patients lacking encephalopathy attained no
survival benefit regardless of disease severity assessed by other criteria. Ramond and colleagues [121]

challenged the concept that encephalopathy is required for a corticosteroid response. In their study of 61
patients with discriminant functions higher than 32, corticosteroids significantly improved patient
survival regardless of whether encephalopathy was present. [121]
In the meta-analysis of corticosteroids in alcoholic hepatitis, patients with gastrointestinal hemorrhage
had to be excluded from consideration in order for treatment to promote a survival advantage. This
suggests that gastrointestinal bleeding carries a high independent risk of mortality, sufficient to preclude
a corticosteroid effect. No evidence emerged from the analysis that corticosteroids enhance the risk of
gastrointestinal hemorrhage; nevertheless, the fact that patients with such bleeding respond poorly to
corticosteroids makes them suboptimal candidates for treatment. Another criterion that should be
considered during patient selection is renal function. Patients with a serum creatinine level less than or
equal to 2.0 mg/dL at randomization appear to respond well to corticosteroids. [119] Patients with
creatinine levels higher than 2.5 mg/dL fare much worse; [117] they have a high risk of progression to
renal failure, and a short-term mortality rate of 75% with or without corticosteroids.
In addition to gastrointestinal hemorrhage and renal insufficiency, other confounding illnesses should be
excluded in the selection of patients for corticosteroid therapy. These are active infection and
pancreatitis, and possibly insulin-dependent diabetes mellitus.
Colchicine
Colchicine inhibits leukocyte migration and function and has been reported to attenuate toxin-induced
liver injury in experimental animals. Despite this, the drug is of little value in the treatment of acute
alcoholic hepatitis. Colchicine was evaluated in a randomized controlled trial involving 72 patients with
serum bilirubin levels higher than 5 mg/dL; it was administered at a dose of 1 mg/day for 30 days. At the
end of the study interval, no differences were observed in either survival rate or biochemical liver test
results. [122]
1210
Colchicine also has reported antifibrotic effects and has been evaluated as a treatment for cirrhosis.
Kershenobich and colleagues [123] studied the efficacy of colchicine therapy in 100 cirrhotic patients, 45%
of whom had alcohol-induced liver disease. Colchicine (1 mg/day) or placebo was administered for up to
14 years (mean, 4.7 years). Life-table analysis indicated a significant survival benefit in patients
receiving colchicine; 5-year survival rates in the treatment and control groups were 75% and 34%,
respectively. [123] In some colchicine-treated patients, serial liver biopsies demonstrated resolution of
fibrosis.
Although the results with colchicine are impressive, they have been viewed with skepticism for several
reasons. First, almost three fourths of the study patients were designated as Child-Turcotte class A at the
time of enrollment, which would predict a 5-year survival much greater than 34% even in the absence of
drug treatment (Table 71-7) (Table Not Available) . Even if the majority of patients belonged to
Child-Turcotte class B, the 5-year survival rate in the placebo group was nonetheless lower than
expected, inasmuch as fewer than half the patients were drinkers. The poor survival of the control group
may have been related to problems other than liver disease. Of the 28 deaths in this group, 9 could not be
attributed to liver disease; 4 were attributed to nonhepatic illness, and in 5 the cause of death was
unknown. Eliminating these patients from analysis raises the survival of the placebo group to

equivalence with the colchicine group.

Another problem with Kershenobich and colleagues' study is that it had a 19% dropout rate. Although the
number of dropouts were equivalent in the treatment and control groups, concerns about statistical
analysis have been raised. These issues, along with the possibility that sampling error may have biased
the interpretation of serial liver biopsies, have reduced enthusiasm for the use of colchicine until further
data become available.
Polyunsaturated Lecithin/Phosphatidylcholine
Alterations in mitochondrial phospholipids have been noted in the livers of alcohol-fed animals and are
implicated in alcohol-induced hepatic mitochondrial dysfunction. One proposed means to ameliorate
ethanol-induced mitochondrial dysfunction is to provide supplemental phospholipids. A soybean extract
containing polyunsaturated lecithin (PUL) has putative membrane-stabilizing effects; this compound was
first tested by Lieber and colleagues [124] in the treatment of alcohol-fed baboons. In a study lasting 10
years, PUL failed to prevent
TABLE 71-7 -- Five-Year Survival in Patients with Alcoholic Cirrhosis, as a Function of Drinking
Behavior
(Not Available)
Data from Powell W. J., and Klatskin, G. Duration of survival in patients with Laennee's cirrhosis.
influence of alcohol withdrawal, and possible effects of recent changes in general management of the
disease, Am.J.Med, 44:406-420, 1968.
ethanol-induced mitochondrial abnormalities in liver cells. [124] The compound did, however,
significantly attenuate hepatic fibrosis (75% versus 0% fibrosis in control versus treated groups). PUL
appears to exert its effect on fibrosis by enhancing hepatic collagen degradation. [125]
In an effort to determine the active phospholipid species in PUL, Lieber, using pure phosphatidylcholine,
performed a second study in baboons. [126] Phosphatidylcholine was chosen because it constitutes 55% to
60% of the PUL soybean extract. Phosphatidylcholine appeared to be as effective as the more crude PUL
in preventing liver fibrosis, although only three of the eight baboons in the treatment group were
observed for more than 4 years. A controlled trial of phosphatidylcholine is now underway in humans
with alcoholic liver disease.
Antioxidants
Antioxidants have been used in animals and humans as potential treatment for alcoholic liver disease.
Disappointing results have been obtained thus far with both vitamin E and vitamin A, [127] [128] [129] [130]
although studies are ongoing. Vitamin A has a very narrow toxic-to-therapeutic ratio [128] [129] and is not
likely to be useful as a single agent. More encouraging results have been obtained with SAM. [22]
Liver Transplantation
Patients with advanced alcoholic liver disease may be candidates for liver transplantation (see Chapter
83) . Although transplantation in alcoholics has aroused great controversy, most centers will accept

alcoholics as transplantation candidates if they meet rigorous selection criteria. Patients undergoing
transplantation for alcohol-induced liver disease have 1- and 3-year survival rates, in comparison with
patients undergoing transplantation for nonalcoholic liver disease. [131] [132] Recidivism is rare, averaging
around 10%. [131] [132] [133] [134]
Identifying alcoholic patients with a potential for recidivism is a major challenge for transplantation
professionals. Although some authors believe that 6 months of confirmed pretransplantation abstinence is
a good predictor of post-transplantation sobriety, [131] [134] others argue that post-transplantation sobriety
can be predicted only from a combination of factors. [133] [135] Efforts are ongoing to define these factors
and incorporate them as standard transplantation selection criteria for alcoholics.


Saunders Company
PROGNOSIS
Alcoholic steatosis is generally viewed as a benign lesion. Although it can cause acute morbidity from
portal hypertension, the fat and its consequences should reverse with abstinence. If patients with steatosis
continue to drink, more serious liver injury may ensue; indeed, serial liver biopsies indicate that steatosis
carries a finite risk of disease progression. Marbet and colleagues [7] reported progression from steatosis
to liver fibrosis in 5 of 16 patients over an interval of 8 years. Similarly over 10 years, Sorensen and
associates [136] and Teli and colleagues [137] reported progression from steatosis to fibrosis or cirrhosis in
13% to 18% of 336 patients. Progression
1211
Figure 71-6 Cumulative survival function (life table method) comparing patients with and without
cirrhosis and with and without hepatitis. n, number of patients. Numbers in parentheses are
percentage mortality. Differences: cirrhosis with no hepatitis versus cirrhosis with hepatitis, P < .02;
noncirrhosis with hepatitis versus noncirrhosis with no hepatitis, not significant. (Reprinted from Orrego, H., Blake, J. E.,
Blendis, L. M., and Medline, A. Prognosis of alcoholic cirrhosis in the presence and absence of alcoholic hepatitis.
Gastroenterology 92:211, 1987 with permission.)
may be heralded on an initial liver biopsy by the presence of perivenular fibrosis. [138]
In patients with clinical evidence of alcoholic liver injury, the natural history of the disease is dependent
in large part on severity. Even those with mild disease have a significant risk of dying; patients who are
sick enough to be hospitalized, even if they do not have jaundice or hypoprothrombinemia, have a
1-month mortality rate close to 20%. [43] Patients with severe disease have a short-term mortality rate of
approximately 50%. [43] [79] When disease severity is assessed by histologic rather than clinical criteria,
survival is somewhat better (1-year survival rate of 60% to 70% for patients with severe disease). [31] [92]
Interestingly, hepatic inflammation and cirrhosis appear to be equally ominous histologic lesions in terms
of predicting prognosis. Cirrhosis alone is a poor prognostic indicator; indeed, Goldberg and colleagues,
[139] who assessed the outcome of 34 patients with biopsy-proven alcoholic cirrhosis but clinical signs of
only mild disease, found their 30-month mortality rate to be 29% (18% for the group without cirrhosis).
However, alcoholic hepatitis also carries a high independent risk of mortality and in some series is more
deadly than inactive cirrhosis [92] (Fig. 71-6) . In studies that have examined the natural history of
alcoholic liver disease on the basis of histologic findings at diagnosis, patients with fatty liver or the
equivalent have the best outcome (70% to 80% survival at 4 to 5 years), those with alcoholic hepatitis or
cirrhosis have an intermediate outcome (50% to 75% survival at 4 to 5 years), and those with cirrhosis
combined with alcoholic hepatitis have the worst outcome (30% to 50% survival at 4 to 5 years). When
all patients with alcoholic liver disease are viewed as a single group, the average 1-year and 5-year
survival rates are found to be approximately 80% and 50%, respectively. [31] [43] [140]
Continued drinking is another factor that significantly impacts survival in alcoholic liver disease. Powell
and Klatskin [141] showed that abstinence can improve patient outcome regardless of disease severity at

diagnosis (see Table 71-7) (Table Not Available) . Abstinence is no guarantee of improvement, however,
inasmuch as some studies indicate progression of liver injury despite cessation of ethanol intake. [63] [64]
The latter may be true particularly for women. [63] [64]
The natural history of alcoholic liver disease may be modified by specific nutritional or pharmacologic
therapy (discussed earlier). The greatest impact to date has been on short-term survival and
improvements in laboratory indices. The long-term impact of drug therapy on outcomes other than
survival, such as progression to cirrhosis, is not yet known.


Saunders Company
REFERENCES
1. Eighth Special Report to the U.S. Congress on Alcohol and Health, U.S. Department of Health and Human Services,
1993, pp. 1-35.
2. Parrish,K. M., Higuchi, S., Muramatsu, T., et al. A method for estimating alcohol-related liver cirrhosis mortality in
Japan. Int. J. Epidemiol. 20:921-926, 1991.
3. Corrao, G., Arico, S., Russo, R., et al. Alcohol consumption and non-cirrhotic chronic hepatitis: A case-control study.
Int. J. Epidemiol. 20:1037-1042, 1991.
4. Savolainen, V. T., Penttila, A., and Karhunen, P. J. Delayed increases in liver cirrhosis mortality and frequency of
alcoholic liver cirrhosis following an increment and redistribution of alcohol consumption in Finland: Evidence from
mortality statistics and autopsy survey covering 8533 cases in 1968-1988. Alcohol Clin. Exp. Res. 16:661-664, 1992.
5. Lelbach, W. K. Cirrhosis in the alcoholic and its relation to the volume of alcohol abuse. Ann. N. Y. Acad. Sci.
252:85-105, 1975.
6. Tuyns, A. J., and Pequignot, G. Greater risk of ascitic cirrhosis in females in relation to alcohol consumption. Int. J.
Epidemiol. 13:53-57, 1984.
7. Marbet, U. A., Bianchi, L., Meury, U., et al. Long-term histological evaluation of the natural history and prognostic
factors of alcoholic liver disease. J. Hepatol. 4:364-372, 1987.
8. Mezey, E., Kolman, C. J., Diehl, A. M., et al. Alcohol and dietary intake in the development of chronic pancreatitis and
liver disease in alcoholism. Am. J. Clin. Nutr. 48:148-151, 1988.
9. Bosron, W. F., Ehrig, T., and Li, T. K. Genetic factors in alcohol metabolism and alcoholism. Semin. Liver Dis.
13:126-135, 1993.
10. Hanna, J. M. Metabolic responses of Chinese, Japanese and Europeans to alcohol. Alcohol Clin. Exp. Res. 2:89-92,
1978.
11. Goedde, H. W., Agarwal, D. P., Harada, S., et al. Population genetic studies on aldehyde dehydrogenase isozyme
deficiency and alcohol sensitivity. Am. J. Hum. Genet. 35:769-772, 1983.
12. Frezza,
M., di Padova, C., Pozzato, G., et al. High blood alcohol levels in women. The role of decreased gastric alcohol
dehydrogenase activity and first-pass metabolism. N. Engl. J. Med. 322:95-99, 1990.
13. Roine,R., Gentry, R. T., Hernandez-Munoz, R., et al. Aspirin increases blood alcohol concentrations in humans after
ingestion of ethanol. JAMA 264:2406-2408, 1990.
14. Caballeria, J., Baraona, E., Deulofeu, R., et al. Effects of H2 -receptor antagonists on gastric alcohol dehydrogenase
activity. Dig. Dis. Sci. 36:1673-1679, 1991.

15. Smith, T., DeMaster, E. G., Furne, J. K., et al. First-pass gastric mucosal metabolism of ethanol is negligible in the rat.
J. Clin. Invest. 89:1801-1806, 1992.
16. Nordmann, R., Ribiere, C., and Rouach, H. Implication of free radical mechanisms in ethanol-induced cellular injury.
Free Radic. Biol. Med. 12:219-240, 1992.
1212
17. Nanji,A. A., Zhao, S., Lamb, R. G., et al. Changes in cytochromes P-450 2E1, 2B1, and 4A1, and phospholipases A
and C in the intragastric feeding rat model for alcoholic liver disease: Relationship to dietary fats and pathologic liver
injury. Alcohol Clin. Exp. Res. 18:902-908, 1994.
18. Tsukamoto, H., Horne, W., Kamimura, S., et al. Experimental liver cirrhosis induced by alcohol and iron. J. Clin.
Invest. 96:620-630, 1995.
19. Niemela, O., Parkkila, S., Yla-Herttuala, S., et al. Sequential acetaldehyde production, lipid peroxidation, and
fibrogenesis in micropig model of alcohol-induced liver disease. Hepatology 22:1208-1214, 1995.
20. Morimoto,
M., Hagbjork, A. L., Wan, Y. J., et al. Modulation of experimental alcohol-induced liver disease by
cytochrome P450 2E1 inhibitors. Hepatology 21:1610-1617, 1995.
21. Fromenty,B., Grimbert, S., Mansouri, A., et al. Hepatic mitochondrial DNA deletion in alcoholics: Association with
microvesicular steatosis. Gastroenterology 108:193-200, 1995.
22. Lieber,
C. S., Casini, A., DeCarli, L. M., et al. S-adenosyl-L-methionine attenuates alcohol-induced liver injury in the
baboon. Hepatology 11:165-172, 1990.
23. Garcia-Ruiz,C., Morales, A., Colell, A., et al. Feeding S-adenosyl-L-methionine attenuates both ethanol-induced
depletion of mitochondrial glutathione and mitochondrial dysfunction in periportal and perivenous rat hepatocytes.
Hepatology 21:207-214, 1995.
24. Niemela,O., Juvonen, T., and Parkkila, S. Immunohistochemical demonstration of acetaldehyde-modified epitopes in
human liver after alcohol consumption. J. Clin. Invest. 87:1367-1374, 1991.
25. Holstege,
A., Bedossa, P., Poynard, T., et al. Acetaldehyde-modified epitopes in liver biopsy specimens of alcoholic
and nonalcoholic patients: Localization and association with progression of liver fibrosis. Hepatology 19:367-374, 1994.
26. Lin,R. C., Zhou, F. C., Fillenwarth, M. J., et al. Zonal distribution of protein-acetaldehyde adducts in the liver of rats
fed alcohol for long periods. Hepatology 18:864-869, 1993.
27. Tuma, D. J., and Sorrell, M. F. Effects of ethanol on protein trafficking in the liver. Semin. Liver Dis. 8:69-80, 1988.
28. Uchida,T., Kronborg, I., and Peters, R. L. Giant mitochondria in the alcoholic liver diseases--Their identification,
frequency and pathologic significance. Liver 4:29-38, 1984.
29. Chedid,A., Mendenhall, C. L., Tosch, T., et al. Significance of megamitochondria in alcoholic liver disease.
Gastroenterology 90:1858-1864, 1986.
30. Fromenty, B., and Pessayre, D. Inhibition of mitochondrial beta-oxidation as a mechanism of hepatotoxicity.

Pharmacol. Ther. 67:101-154, 1995.
31. Chedid,
A., Mendenhall,, C. L., Gartside, P., et al. Prognostic factors in alcoholic liver disease. VA Cooperative Study
Group. Am. J. Gastroenterol. 86:210-216, 1991.
32. McClain, C., Hill, D., Schmidt, J., et al. Cytokines and alcoholic liver disease. Semin. Liver Dis. 13:170-182, 1993.
33. Shiratori,
Y., Takada, H., Hai, K., et al. Generation of chemotactic factor by hepatocytes isolated from chronically
ethanol-fed rats. Dig. Dis. Sci. 37:650-658, 1992.
34. Bautista,
A. P. Chronic alcohol intoxication enhances the expression of CD18 adhesion molecules on rat neutrophils
and release of a chemotactic factor by Kupffer cells. Alcohol Clin. Exp. Res. 19:285-290, 1995.
35. Adachi,
Y., Bradford, B. U., Gao, W., et al. Inactivation of Kupffer cells prevents early alcohol-induced liver injury.
Hepatology 20:453-460, 1994.
36. Adachi,Y., Moore, L. E., Bradford, B. U., et al. Antibiotics prevent liver injury in rats following long-term exposure to
ethanol. Gastroenterology 108:218-224, 1995.
37. Nanji,
A. A., Khettry, U., and Sadrzadeh, S. M. Lactobacillus feeding reduces endotoxemia and severity of
experimental alcoholic liver (disease). Proc. Soc. Exp. Biol. Med. 205:243-247, 1994.
38. Niemela, O., Klajner, F., Orrego, H., et al. Antibodies against acetaldehyde-modified protein epitopes in human
alcoholics. Hepatology 7:1210-1214, 1987.
39. Hoerner,M., Behrens, U. J., Worner, T. M., et al. The role of alcoholism and liver disease in the appearance of serum
antibodies against acetaldehyde adducts. Hepatology 8:569-574, 1988.
40. Clot,P., Bellomo, G., Tabone, M., et al. Detection of antibodies against proteins modified by hydroxyethyl free radicals
in patients with alcoholic cirrhosis. Gastroenterology 108:201-207, 1995.
41. Yokoyama, H., Ishii, H., Nagata, S., et al. Experimental hepatitis induced by ethanol after immunization with
acetaldehyde adducts. Hepatology 17:14-19, 1993.
42. Yokoyama,H., Nagata, S., Moriya, S., et al. Hepatic fibrosis produced in guinea pigs by chronic ethanol administration
and immunization with acetaldehyde adducts. Hepatology 21:1438-1442, 1995.
43. Mendenhall, C. L. Alcoholic hepatitis. Clin. Gastroenterol. 10:417-41, 1981.
44. Friedman, S. L. The cellular basis of hepatic fibrosis. N. Engl. J. Med. 328:1828-1835, 1993.
45. Hayashi,N., Kasahara, A., Kurosawa, K., et al. Oxygen supply to the liver in patients with alcoholic liver disease
assessed by organ-reflectance spectrophotometry. Gastroenterology 88:881-886, 1985.
46. Melendez,M., Vargas-Tank, L., Fuentes, C., et al. Distribution of HLA histocompatibility antigens, ABO blood groups
and Rh antigens in alcoholic liver disease. Gut 20:288-290, 1979.
47. Bell,
H., and Nordhagen, R. HLA antigens in alcoholics, with special reference to alcoholic cirrhosis. Scand. J.
Gastroenterol. 15:453-456, 1980.
48. Doffoel,M., Tongio, M. M., Gut, J. P., et al. Relationships between 34 HLA-A, HLA-B and HLA-DR antigens and
three serological markers of viral infections in alcoholic cirrhosis. Hepatology 6:457-463, 1986.

49. Monteiro,E., Alves, M. P., Santos, M. L., et al. Histocompatibility antigens: Markers of susceptibility to and protection
from alcoholic liver disease in a Portuguese population. Hepatology 8:455-458, 1988.
50. Scott,
B. B., Rajah, S. M., and Losowsky, M. S. Histocompatibility antigens in chronic liver disease. Gastroenterology
72:122-125, 1977.
51. Faizallah,R., Woodrow, J. C., Krasner, N. K., et al. Are HLA antigens important in the development of alcohol-induced
liver disease? BMJ 285:533-534, 1982.
52. Mills,
P. R., MacSween, R. N., Dick, H. M., et al. Histocompatibility antigens in patients with alcoholic liver disease in
Scotland and northeastern England: Failure to show an association. Gut 29:146-148, 1988.
53. Goedde,
H. W., Agarwal, D. P., Fritze, G., et al. Distribution of ADH2 and ALDH2 genotypes in different populations.
Hum. Genet. 88:344-346, 1992.
54. Thomasson,H. R., Edenberg, H. J., Crabb, D. W., et al. Alcohol and aldehyde dehydrogenase genotypes and alcoholism
in Chinese men. Am. J. Hum. Genet. 48:677-681, 1991.
55. Chao,Y. C., Liou, S. R., Chung, Y. Y., et al. Polymorphism of alcohol and aldehyde dehydrogenase genes and
alcoholic cirrhosis in Chinese patients. Hepatology 19:360-366, 1994.
56. Day,
C. P., Bashir, R., James, O. F., et al. Investigation of the role of polymorphisms at the alcohol and aldehyde
dehydrogenase loci in genetic predisposition to alcohol-related end-organ damage. Hepatology 14:798-801, 1991.
57. Poupon,
R. E., Nalpas, B., Coutelle, C., et al. Polymorphism of alcohol dehydrogenase, alcohol and aldehyde
dehydrogenase activities: Implication in alcoholic cirrhosis in white patients. Hepatology 15:1017-1022, 1992.
58. Tsutsumi, M., Takada, A., and Wang, J. S. Genetic polymorphisms of cytochrome P4502E1 related to the development
of alcoholic liver disease. Gastroenterology 107:1430-1435, 1994.
59. Carr,L. G., Hartleroad, J. Y., Liang, Y., et al. Polymorphism at the P450IIEI locus is not associated with alcoholic liver
disease in Caucasian men. Alcohol Clin. Exp. Res. 19:182-184, 1995.
60. Ingelman-Sundberg, M., Johansson, I., Yin, H., et al. Ethanol-inducible cytochrome P4502E1: Genetic polymorphism,
regulation, and possible role in the etiology of alcohol-induced liver disease. Alcohol 10:447-452, 1993.
61. Yamauchi, M., Maezawa, Y., Mizuhara, Y., et al. Polymorphisms in alcohol metabolizing enzyme genes and alcoholic
cirrhosis in Japanese patients: A multivariate analysis. Hepatology 22:1136-1142, 1995.
62. Enomoto,
N., Takase, S., Takada, N., et al. Alcoholic liver disease in heterozygotes of mutant and normal aldehyde
dehydrogenase-2 genes. Hepatology 13:1071-1075, 1991.
63. Galambos, J. T. Natural history of alcoholic hepatitis: 3. Histological changes. Gastroenterology 63:1026-1035, 1972.
64. Pares,A., Caballeria, J., Bruguera, M., et al. Histological course of alcoholic hepatitis. Influence of abstinence, sex and
extent of hepatic damage. J. Hepatol. 2:33-42, 1986.
65. Mendenhall, C. L., Tosch, T., Weesner, R. E., et al. VA cooperative study on alcoholic hepatitis: II. Prognostic
significance of protein-calorie malnutrition. Am. J. Clin. Nutr. 43:213-218, 1986.
66. Nanji, A. A., and French, S. W. Dietary factors and alcoholic cirrhosis. Alcohol Clin. Exp. Res. 10:271-273, 1986.

67. Pares,A., Barrera, J. M., Caballeria, J., et al. Hepatitis C virus antibodies in chronic alcoholic patients: Association with
severity of liver injury. Hepatology 12:1295-1299, 1990.
1213
68. Nalpas, B., Driss, F., Pol, S., et al. Association between HCV and HBV infection in hepatocellular carcinoma and
alcoholic liver disease. J. Hepatol. 12:70-74, 1991.
69. Mendenhall, C. L., Seeff, L., Diehl, A. M., et al. Antibodies to hepatitis B virus and hepatitis C virus in alcoholic
hepatitis and cirrhosis: Their prevalence and clinical relevance. The VA Cooperative Study Group (No. 119). Hepatology
14:581-589, 1991.
70. Caldwell,S. H., Li, X., Rourk, R. M., et al. Hepatitis C infection by polymerase chain reaction in alcoholics:
False-positive ELISA results and the influence of infection on a clinical prognostic score. Am. J. Gastroenterol.
88:1016-1021, 1993.
71. Sawada,M., Takada, A., Takase, S., et al. Effects of alcohol on the replication of hepatitis C virus. Alcohol Alcohol
Suppl. 1B:85-90, 1993.
72. Oshita,
M., Hayashi, N., Kasahara, A., et al. Increased serum hepatitis C virus RNA levels among alcoholic patients
with chronic hepatitis C. Hepatology 20:1115-1120, 1994.
73. Levi, A. J., and Chalmers, D. M. Recognition of alcoholic liver disease in a district general hospital. Gut 19:521-525,
1978.
74. Bruguera,
M., Bordas, J. M., and Rodes, J. Asymptomatic liver disease in alcoholics. Arch. Pathol. Lab. Med.
101:644-647, 1977.
75. Leevy,C. Fatty liver: A study of 270 patients with biopsy-proven fatty liver and a review of the literature. Medicine
41:249, 1962.
76. Israel,
Y., Orrego, H., Colman, J. C., et al. Alcohol-induced hepatomegaly: Pathogenesis and role in the production of
portal hypertension. Fed. Proc. 41:2472-2477, 1982.
77. Cohen,
J. A., and Kaplan, M. M. The SGOT/SGPT ratio--An indicator of alcoholic liver disease. Dig. Dis. Sci.
24:835-838, 1979.
78. Seeff,
L. B., Cuccherini, B. A., Zimmerman, H. J., et al. Acetaminophen hepatotoxicity in alcoholics. A therapeutic
misadventure. Ann. Intern. Med. 104:399-404, 1986.
79. Maddrey,W. C., Boitnott, J. K., Bedine, M. S., et al. Corticosteroid therapy of alcoholic hepatitis. Gastroenterology
75:193-199, 1978.
80. Imperiale,
T. F., and McCullough, A. J. Do corticosteroids reduce mortality from alcoholic hepatitis? A meta-analysis
of the randomized trials. Ann. Intern. Med. 113:299-307, 1990.
81. Fletcher,
L. M., Kwoh-Gain, I., Powell, E. E., et al. Markers of chronic alcohol ingestion in patients with nonalcoholic
steatohepatitis: An aid to diagnosis. Hepatology 13:455-459, 1991.

82. Rosman, A. S., Basu, P., Galvin, K., et al. Utility of carbohydrate-deficient transferrin as a marker of relapse in
alcoholic patients. Alcohol Clin. Exp. Res. 19:611-616, 1995.
83. Lin,
R. C., Shahidi, S., Kelly, T. J., et al. Measurement of hemoglobin-acetaldehyde adduct in alcoholic patients.
Alcohol Clin. Exp. Res. 17:669-674, 1993.
84. Annoni,G., Colombo, M., Cantaluppi, M. C., et al. Serum type III procollagen peptide and laminin (Lam-P1) detect
alcoholic hepatitis in chronic alcohol abusers. Hepatology 9:693-697, 1989.
85. Niemela, O., Risteli, L., Sotaniemi, E. A., et al. Aminoterminal propeptide of type III procollagen in serum in alcoholic
liver disease. Gastroenterology 85:254-259, 1983.
86. Torres-Salinas,M., Pares, A., Caballeria, J., et al. Serum procollagen type III peptide as a marker of hepatic
fibrogenesis in alcoholic hepatitis. Gastroenterology 90:1241-1246, 1986.
87. Niemela, O., Risteli, J., Blake, J. E., et al. Markers of fibrogenesis and basement membrane formation in alcoholic liver
disease. Relation to severity, presence of hepatitis, and alcohol intake. Gastroenterology 98:1612-1619, 1990.
88. Levin,D. M., Baker, A. L., Riddell, R. H., et al. Nonalcoholic liver disease. Overlooked causes of liver injury in
patients with heavy alcohol consumption. Am. J. Med. 66:429-434, 1979.
89. French,
S. W., Nash, J., Shitabata, P., et al. Pathology of alcoholic liver disease. VA Cooperative Study Group 119.
Semin. Liver Dis. 13:154-169, 1993.
90. Uchida, T., Kao, H., Quispe-Sjogren, M., and Peters, R. L. Alcoholic foamy degeneration--A pattern of acute alcoholic
injury of the liver. Gastroenterology 84:683-692, 1983.
91. Montull, S., Pares, A., Bruguera, M., et al. Alcoholic foamy degeneration in Spain. Prevalence and clinico-pathological
features. Liver 9:79-85, 1989.
92. Orrego,H., Blake, J. E., Blendis, L. M., and Medline, A. Prognosis of alcoholic cirrhosis in the presence and absence of
alcoholic hepatitis. Gastroenterology 92:208-214, 1987.
93. Diehl,
A. M., Goodman, Z., and Ishak, K. G. Alcohol-like liver disease in nonalcoholics. A clinical and histologic
comparison with alcohol-induced liver injury. Gastroenterology 95:1056-1062, 1988.
94. Powell,E. E., Cooksley, W. G., Hanson, R., et al. The natural history of nonalcoholic steatohepatitis: A follow-up study
of forty-two patients for up to 21 years. Hepatology 11:74-80, 1990.
95. Bacon,
B. R., Farahvash, M. J., Janney, C. G., and Neuschwander-Tetri, B. A. Nonalcoholic steatohepatitis: An
expanded clinical entity. Gastroenterology 107:1103-1109, 1994.
96. Ikeda,
K., Saitoh, S., Koida, I., et al. A multivariate analysis of risk factors for hepatocellular carcinogenesis: A
prospective observation of 795 patients with viral and alcoholic cirrhosis. Hepatology 18:47-53, 1993.
97. Tsukuma, H., Hiyama, T., Tanaka, S., et al. Risk factors for hepatocellular carcinoma among patients with chronic liver
disease. N. Engl. J. Med. 328:1797-1801, 1993.
98. Poynard,
T., Aubert, A., Lazizi, Y., et al. Independent risk factors for hepatocellular carcinoma in French drinkers.
Hepatology 13:896-901, 1991.

99. Mendenhall,
C. L., Anderson, S., Weesner, R. E., et al. Protein-calorie malnutrition associated with alcoholic hepatitis.
Veterans Administration Cooperative Study Group on Alcoholic Hepatitis. Am. J. Med. 76:211-222, 1984.
100. Nasrallah, S. M., and Galambos, J. T. Amino acid therapy of alcoholic hepatitis. Lancet 2:1276-1277, 1980.
101. Mendenhall, C., Bongiovanni, G., Goldberg, S., et al. VA Cooperative Study on Alcoholic Hepatitis: III. Changes in
protein-calorie malnutrition associated with 30 days of hospitalization with and without enteral nutritional therapy. JPEN
9:590-596, 1985.
102. Calvey,
H., Davis, M., and Williams, R. Controlled trial of nutritional supplementation, with and without branched
chain amino acid enrichment, in treatment of acute alcoholic hepatitis. J. Hepatol. 1:141-151, 1985.
103. Diehl,A. M., Boitnott, J. K., Herlong, H. F., et al. Effect of parenteral amino acid supplementation in alcoholic
hepatitis. Hepatology 5:57-63, 1985.
104. Achord, J. L. Malnutrition and the role of nutritional support in alcoholic liver disease. Am. J. Gastroenterol. 82:1-7,
1987.
105. Mezey,E., Caballeria, J., Mitchell, M. C., et al. Effect of parenteral amino acid supplementation on short-term and
long-term outcomes in severe alcoholic hepatitis: A randomized controlled trial. Hepatology 14:1090-1096, 1991.
106. Soberon,
S., Pauley, M. P., Duplantier, R., et al. Metabolic effects of enteral formula feeding in alcoholic hepatitis.
Hepatology 7:1204-1209, 1987.
107. Orrego, H., Kalant, H., Israel, Y., et al. Effect of short-term therapy with propylthiouracil in patients with alcoholic
liver disease. Gastroenterology 76:105-115, 1979.
108. Halle,P., Pare, P., Kaptein, E., et al. Double-blind, controlled trial of propylthiouracil in patients with severe acute
109. Orrego,
H., Blake, J. E., Blendis, L. M., et al. Long-term treatment of alcoholic liver disease with propylthiouracil. N.
Engl. J. Med. 317:1421-1427, 1987.
110. Orrego,H., Blake, J. E., Blendis, L. M., et al. Long-term treatment of alcoholic liver disease with propylthiouracil: Part
2. Influence of drop-out rates and of continued alcohol consumption in a clinical trial. J. Hepatol. 20:343-349, 1994.
111. Helman,R. A., Temko, M. H., Nye, S. W., et al. Alcoholic hepatitis. Natural history and evaluation of prednisolone
therapy. Ann. Intern. Med. 74:311-321, 1971.
112. Porter,H. P., Simon, F. R., Pope, C. E., et al. Corticosteroid therapy in severe alcoholic hepatitis. A double-blind drug
trial. N. Engl. J. Med. 284:1350-1355, 1971.
113. Campra, J. L., Hamlin, E. M., Kirshbaum, R. J., et al. Prednisone therapy of acute alcoholic hepatitis. Report of a
controlled trial. Ann. Intern. Med. 79:625-631, 1973.
114. Blitzer,
B. L., Mutchnick, M. G., Joshi, P. H., et al. Adrenocorticosteroid therapy in alcoholic hepatitis. A prospective,
double-blind randomized study. Am. J. Dig. Dis. 22:477-484, 1977.
115. Lesesne,H. R., Bozymski, E. M., and Fallon, H. J. Treatment of alcoholic hepatitis with encephalopathy. Comparison
of prednisolone with caloric supplements. Gastroenterology 74:169-173, 1978.
116. Shumaker, J. B., Resnick, R. H., Galambos, J. T., et al. A controlled trial of 6-methylprednisolone in acute alcoholic

hepatitis. With a note on published results in encephalopathic patients. Am. J. Gastroenterol. 69:443-449, 1978.
117. Depew, W., Boyer, T., Omata, M., et al. Double-blind controlled trial
1214
of prednisolone therapy in patients with severe acute alcoholic hepatitis and spontaneous encephalopathy.
118. Theodossi,A., Eddleston, A. L., and Williams, R. Controlled trial of methylprednisolone therapy in severe acute
alcoholic hepatitis. Gut 23:75-79, 1982.
119. Carithers,R. L., Herlong, H. F., Diehl, A. M., et al. Methylprednisolone therapy in patients with severe alcoholic
hepatitis. A randomized multicenter trial. Ann. Intern. Med. 110:685-690, 1989.
120. Mendenhall, C. L., Anderson, S., Garcia-Pont, P., et al. Short-term and long-term survival in patients with alcoholic
hepatitis treated with oxandrolone and prednisolone. N. Engl. J. Med. 311:1464-1470, 1984.
121. Ramond, M. J., Poynard, T., Rueff, B., et al. A randomized trial of prednisolone in patients with severe alcoholic
hepatitis. N. Engl. J. Med. 326:507-512, 1992.
122. Akriviadis,E. A., Steindel, H., Pinto, P. C., et al. Failure of colchicine to improve short-term survival in patients with
123. Kershenobich,
D., Vargas, F., Garcia-Tsao, G., et al. Colchicine in the treatment of cirrhosis of the liver. N. Engl. J.
Med. 318:1709-1713, 1988.
124. Lieber,C. S., DeCarli, L. M., Mak, K. M., et al. Attenuation of alcohol-induced hepatic fibrosis by polyunsaturated
lecithin. Hepatology 12:1390-1398, 1990.
125. Li,
J., Kim, C. I., Leo, M. A., et al. Polyunsaturated lecithin prevents acetaldehyde-mediated hepatic collagen
accumulation by stimulating collagenase activity in cultured lipocytes. Hepatology 15:373-381, 1992.
126. Lieber,
C. S., Robins, S. J., Li, J., et al. Phosphatidylcholine protects against fibrosis and cirrhosis in the baboon.
127. Sadrzadeh,
S. M., Meydani, M., Khettry, U., et al. High-dose vitamin E supplementation has no effect on
ethanol-induced pathological liver injury. J. Pharmacol. Exp. Ther. 273:455-460, 1995.
128. Leo,
M. A., Kim, C., Lowe, N., et al. Interaction of ethanol with beta-carotene: Delayed blood clearance and enhanced
hepatotoxicity. Hepatology 15:883-891, 1992.
129. Ahmed, S., Leo, M. A., and Lieber, C. S. Interactions between alcohol and beta-carotene in patients with alcoholic
liver disease. Am. J. Clin. Nutr. 60:430-436, 1994.
130. Butcher, G. P., Rhodes, J. M., Walker, R., et al. The effect of antioxidant supplementation on a serum marker of free
radical activity and abnormal serum biochemistry in alcoholic patients admitted for detoxification. J. Hepatol. 19:105-109,
1993.
131. Kumar,S., Stauber, R. E., Gavaler, J. S., et al. Orthotopic liver transplantation for alcoholic liver disease. Hepatology
11:159-164, 1990.

132. Lucey, M. R. Liver transplantation for alcoholic liver disease. Baillieres Clin. Gastroenterol. 7:717-727, 1993.
133. Gish,R. G., Lee, A. H., Keeffe, E. B., et al. Liver transplantation for patients with alcoholism and end-stage liver
disease. Am. J. Gastroenterol. 88:1337-1342, 1993.
134. Osorio,R. W., Ascher, N. L., Avery, M., et al. Predicting recidivism after orthotopic liver transplantation for alcoholic
liver disease. Hepatology 20:105-110, 1994.
135. Lucey, M. R. Liver transplantation for the alcoholic patient. Gastroenterol. Clin. North. Am. 22:243-256, 1993.
136. Sorensen,
T. I., Orholm, M., Bentsen, K. D., et al. Prospective evaluation of alcohol abuse and alcoholic liver injury in
men as predictors of development of cirrhosis. Lancet 2:241-244, 1984.
137. Teli,
M. R., Day, C. P., Burt, A. D., et al. Determinants of progression to cirrhosis or fibrosis in pure alcoholic fatty
liver. Lancet 346:987-990, 1995.
138. Worner, T. M., and Lieber, C. S. Perivenular fibrosis as precursor lesion of cirrhosis. JAMA 254:627-630, 1985.
139. Goldberg, S., Mendenhall, C., Anderson, S., et al. VA Cooperative Study on Alcoholic Hepatitis: IV. The significance
of clinically mild alcoholic hepatitis--Describing the population with minimal hyperbilirubinemia. Am. J. Gastroenterol.
81:1029-1034, 1986.
140. Orrego,
H., Israel, Y., Blake, J. E., and Medline, A. Assessment of prognostic factors in alcoholic liver disease:
Toward a global quantitative expression of severity. Hepatology 3:896-905, 1983.
141. Powell,
W. J., and Klatskin, G. Duration of survival in patients with Laennec's cirrhosis. Influence of alcohol
withdrawal, and possible effects of recent changes in general management of the disease. Am J Med 44:406-420, 1968.


Saunders Company
1215
Chapter 72 - Nonalcoholic Steatohepatitis and

Focal Fatty Liver
Patrick Okolo III

Anna Mae Diehl
NONALCOHOLIC STEATOHEPATITIS, 1215

Definition and Overview, 1215
Risk Factors and Pathogenesis, 1215
Symptoms/Signs, 1216
Laboratory Evaluation, 1216
Histologic Features, 1217
Prognosis, 1217
Present Therapy and Future Directions, 1219
Conclusion, 1219
FOCAL FATTY LIVER, 1219


April 21, 2000
Conde Petra
Dambro: Griffith's 5-Minute Clinical Consult,

1999 ed., Copyright 1999 Lippincott Williams & Wilkins,
Inc.
Expanded Topics
C Dambro: Griffith's 5-Minute Clinical Consult, 1999 ed., Copyright

1999 Lippincott Williams & Wilkins, Inc.
BASICS
DESCRIPTION
234
SIGNS AND SYMPTOMS
CAUSES Cirrhosis of the liver

RISK FACTORS
DIAGNOSIS
Richard E. Sampliner MD
DIFFERENTIAL
DIAGNOSIS
LABORATORY
PATHOLOGICAL
FINDINGS
SPECIAL TESTS BASICS

IMAGING DESCRIPTION
DIAGNOSTIC
PROCEDURES Histologically cirrhosis is defined by the presence of fibrosis
with regenerative nodules. Clinically cirrhosis presents with
TREATMENT evidence of portal hypertension, i.e. ascites, variceal bleeding,
APPROPRIATE HEALTH
hepatic encephalopathy.
CARE System(s) affected: Gastrointestinal, Cardiovascular,
GENERAL MEASURES Endocrine/Metabolic
SURGICAL MEASURES Genetics: For hereditary hemochromatosis mutation on

chromosome 6
ACTIVITY
Incidence/Prevalence in USA: Accounts for over 30,000
DIET deaths per year
PATIENT EDUCATION
http://home.mdconsult.com/das/book/view/846/123....stty=cow&nid=39983&tlt=Z&pos=&tag=relinfo&anc=bo (1 of 10) [21/04/2000 23:38:43]
MEDICATIONS
Predominant age: Etiology dependent
DRUG(S) OF CHOICE
Predominant sex: Etiology dependent
ALTERNATIVE DRUGS
SIGNS AND SYMPTOMS
FOLLOWUP
PATIENT MONITORING
The onset of the disease is often insidious with:
Fatigue
PREVENTION/AVOIDANCE
Anorexia
POSSIBLE Nausea
COMPLICATIONS
Abdominal discomfort and distention
EXPECTED Weakness and malaise
COURSE/PROGNOSIS
Signs and symptoms that are related to cirrhosis are
MISCELLANEOUS those of complications:
ASSOCIATED Hematemesis
CONDITIONS Encephalopathy
AGE-RELATED FACTORS Jaundice
Hepatomegaly
PREGNANCY
Splenomegaly
SYNONYMS Abdominal collateral circulation
ICD-9-CM Ascites
SEE ALSO Gynecomastia
Testicular atrophy
OTHER NOTES
Asterixis (liver flap)
ABBREVIATIONS Palmar erythema
REFERENCES Spider angiomas
CAUSES
Alcoholic cirrhosis
Chronic viral hepatitis, B (with/without D), C
Wilson's disease
Hemochromatosis
Alpha 1-antitrypsin deficiency
Cystic fibrosis
Autoimmune chronic hepatitis with cirrhosis
Primary biliary cirrhosis
Secondary biliary cirrhosis
Primary sclerosing cholangitis
Cardiac cirrhosis

Drug induced (other than alcohol)

Nonalcoholic steatohepatitis (NASH)
Inherited causes that may be present in infancy and
childhood:
Glycogen storage disease
Galactosemia
Fructose intolerance
Tyrosinemia
Acid cholesterol ester hydrolase deficiency
RISK FACTORS
Alcohol use
Hepatotoxic drugs
Excessive iron ingestion
DIAGNOSIS
DIFFERENTIAL DIAGNOSIS
Depends on presentation
Ascites - increased right heart pressure, hepatic vein
thrombosis, peritoneal infection or malignancy,
pancreatic disease, thyroid disease, lymphatic obstruction
Other causes of UGI bleeding
Other metabolic encephalopathies - renal,
cardiopulmonary, drug.
LABORATORY
Recognition of liver injury - elevated AST, elevated

ALT; elevated alkaline phosphatase. Note: All liver
injury tests may be normal.
Functional impairment of the liver - elevated bilirubin,
decreased albumin, elevated globulin, prolonged
prothrombin time
Etiologic screen for liver disease
Ceruloplasmin (Wilson's disease)

Iron, iron binding capacity, ferritin

(hemochromatosis)
Alpha fetoprotein (hepatocellular cancer)
HBsAg (hepatitis B)
Anti-HCV (hepatitis C)
HCV RNA by PCR to confirm activity
ANA (autoimmune hepatitis)
Anti-smooth muscle antibody (autoimmune
hepatitis)
Anti-mitochondrial antibody (AMA) (primary
biliary cirrhosis)
Alpha 1-antitrypsin (deficiency)
Serum protein electrophoresis (SPEP) -
increased IgG with any liver disease; increased
IgM with primary biliary cirrhosis (PBC)
Drugs that may alter lab results: N/A
Disorders that may alter lab results: N/A
PATHOLOGICAL FINDINGS
Fibrosis and regenerative nodules; specific

findings/patterns may indicate etiology
Quantitative liver chemistry for iron, copper
Special stains for iron, copper, bilirubin, collagen,
alpha 1-antitrypsin, hepatitis B
SPECIAL TESTS
Laparoscopic liver biopsy to reduce sampling error

Cholangiography to rule out common duct obstruction
and recognize sclerosing cholangitis
Doppler ultrasound to indicate direction of flow in and
patency of the portal and hepatic veins
Visceral angiography to determine vascular anatomy,
patency and collaterals
Esophagogastroduodenoscopy to assess varices
IMAGING
Ultrasound good for detecting bile duct dilatation and

space occupying lesions. Cannot make diagnosis of
cirrhosis based on ultrasound alone.

CT, if ultrasound technically inadequate
DIAGNOSTIC PROCEDURES
A liver biopsy establishes the diagnosis of cirrhosis

Patterns of injury as well as special stains may identify
a precise etiology such as alcoholic liver disease,
hemochromatosis, alpha 1-antitrypsin deficiency,
hepatitis B, primary biliary cirrhosis
Level of activity determined
TREATMENT
APPROPRIATE HEALTH CARE
Outpatient except for complicating emergencies:

GI bleeding
Hepatic encephalopathy
Spontaneous bacterial peritonitis
Unexplained decompensation
Renal failure
GENERAL MEASURES
Treatment designed to remove or alleviate underlying

cause of cirrhosis, prevent further liver damage and
prevent complications
Phlebotomy for hemochromatosis
Therapies involve drug treatment, dietary restrictions,
rest, other supportive measures. Adequate protein intake
for liver regeneration.
SURGICAL MEASURES
Interventional
Possible procedures for portal hypertension
include - splenorenal or portacaval anastomosis,
transjugular intrahepatic portal-systemic shunt
Transplantation - in suitable candidate
(substance free, motivated and adherent). Evaluate

prior to major decompensation.
ACTIVITY
Maintain as active as possible. With peripheral edema, leg

elevation necessary.
DIET
Adequate protein (1 gm/kg) and generous calories to

help regenerate the liver
In the presence of hepatic encephalopathy protein
restriction is necessary
In the presence of ascites salt restriction is necessary (2
gm or less/day)
In the presence of hyponatremia (Na < 130 mEq [< 130
mmol]) fluid restriction is necessary (< 1 L)
No alcohol
PATIENT EDUCATION
Pamphlets are available through: American Liver

Foundation, (800)223-0179
Additional material: National Digestive Diseases
Information Clearinghouse, Box NDDIC, Bethesda, MD
20892, (301)468-6344
MEDICATIONS
DRUG(S) OF CHOICE
235
Large esophageal varices seen at endoscopy prior to

clinical bleeding - nonselective beta blocker at a dose to
decrease the resting pulse by 25%
Ascites: spironolactone 100 mg up to 400 mg every
day as a single dose. Takes three days before onset of

action. Add furosemide 40-160 mg/day if needed. Loop

diuretics may produce a rapid diuresis and subsequent
intravascular volume depletion.
Encephalopathy: lactulose (Cholac) 15-30 mL/day to
produce 2-3 soft stools per day
Spontaneous bacterial peritonitis (SBP): cefotaxime
alone as initial treatment; norfloxacin 400 mg/day or
trimethoprim-sulfamethoxazole decreases risk for
subsequent development of SBP after treatment
Specific drug based on etiology
Wilson's disease: penicillamine, 125-250 mg qid
(give on empty stomach to avoid inactivation by
metal binding)
Autoimmune chronic hepatitis: corticosteroids
with or without azathioprine
Chronic hepatitis B and C: interferon
Hepatitis C relapse after interferon: interferon
and ribavirin
PBC: ursodiol (ursodeoxycholic acid), 12-15
mg/kg/day in a bid dose
Contraindications: Refer to manufacturer's literature
Precautions: Refer to manufacturer's literature
Significant possible interactions: Refer to manufacturer's
literature
ALTERNATIVE DRUGS
Ursodiol (ursodeoxycholic acid) for cholestatic disease
FOLLOWUP
PATIENT MONITORING
In a stable patient - yearly battery of liver tests. After

10 years, consider alpha-fetoprotein and imaging to
detect hepatocellular carcinoma.
In an unstable patient - tests may be repeated at weekly
intervals
Have patient monitor weight and maintain a daily diary
Limit use of alcohol and other liver toxins

No sharing of syringes
Safer sex
Screening of family members when a genetic disease is
recognized
Influenza and pneumococcal vaccines for cirrhosis
patients exposed to crowds
Hepatitis A and B vaccines
Liver test surveillance while on hepatotoxic drugs
(e.g., INH)
POSSIBLE COMPLICATIONS
Ascites
Jaundice
Coagulopathy
Bleeding esophageal varices
Liver failure
Carcinoma of the liver (uncommon)
Susceptibility to infections
Renal failure
EXPECTED COURSE/PROGNOSIS
A function of ongoing hepatic injury as well as residual hepatic

reserve. If a treatable cause is identified and intervention results
in cessation of liver destruction, then the prognosis may be
good.
MISCELLANEOUS
ASSOCIATED CONDITIONS
Hepatitis
Diseases and defects of the bile ducts
Cystic fibrosis
Heart failure
Hepatocellular cancer
AGE-RELATED FACTORS
Pediatric: N/A
Geriatric: Cirrhosis is one of the leading causes of death for
people over age 65
Others: N/A
PREGNANCY
Cirrhosis may decompensate during pregnancy. Higher rates of

spontaneous abortion, premature birth and perinatal death.
SYNONYMS
N/A
ICD-9-CM
571.2 Alcoholic cirrhosis of liver
571.5 Cirrhosis of liver without mention of alcohol
SEE ALSO
N/A
OTHER NOTES
N/A
ABBREVIATIONS
PBC = Primary biliary cirrhosis
UGI = upper gastrointestinal

ANA = anti-nuclear antibody
REFERENCES
Sampliner RE: The recognition of early liver disease.

Hospital Practice 1989; 53-56
Schiff L, Schiff ER, eds: Diseases of the Liver.
Philadelphia, J.B. Lippincott co., 1993
Runyon BA: Care of patients with ascites. NEJM 1994;
330:337-342
Illustrations: N/A
Internet references: http://www.5mcc.com


Dambro: Griffith's 5-Minute Clinical Consult, 1999 ed., Copyright 1999 Lippincott Williams & Wilkins, Inc.
234
Richard E. Sampliner MD
BASICS
DESCRIPTION
Histologically cirrhosis is defined by the presence of fibrosis with regenerative nodules. Clinically
cirrhosis presents with evidence of portal hypertension, i.e. ascites, variceal bleeding, hepatic
encephalopathy.
System(s) affected: Gastrointestinal, Cardiovascular, Endocrine/Metabolic
Genetics: For hereditary hemochromatosis mutation on chromosome 6
Incidence/Prevalence in USA: Accounts for over 30,000 deaths per year
Predominant age: Etiology dependent
Predominant sex: Etiology dependent
SIGNS AND SYMPTOMS
The onset of the disease is often insidious with:

Fatigue
Anorexia
Nausea
Abdominal discomfort and distention
Weakness and malaise
Signs and symptoms that are related to cirrhosis are those of complications:
Hematemesis

Encephalopathy
Jaundice
Hepatomegaly
Splenomegaly
Abdominal collateral circulation
Ascites
Gynecomastia
Testicular atrophy
Asterixis (liver flap)
Palmar erythema
Spider angiomas
CAUSES
Alcoholic cirrhosis
Chronic viral hepatitis, B (with/without D), C
Wilson's disease
Hemochromatosis
Alpha 1-antitrypsin deficiency
Cystic fibrosis
Autoimmune chronic hepatitis with cirrhosis
Primary sclerosing cholangitis
Cardiac cirrhosis
Drug induced (other than alcohol)
Nonalcoholic steatohepatitis (NASH)
Inherited causes that may be present in infancy and childhood:
Glycogen storage disease
Galactosemia
Fructose intolerance
Tyrosinemia
Acid cholesterol ester hydrolase deficiency
RISK FACTORS
Alcohol use
Hepatotoxic drugs
Excessive iron ingestion

DIAGNOSIS
DIFFERENTIAL DIAGNOSIS
Depends on presentation
Ascites - increased right heart pressure, hepatic vein thrombosis, peritoneal infection or
malignancy, pancreatic disease, thyroid disease, lymphatic obstruction
Other causes of UGI bleeding
Other metabolic encephalopathies - renal, cardiopulmonary, drug.
LABORATORY
Recognition of liver injury - elevated AST, elevated ALT; elevated alkaline phosphatase. Note:
All liver injury tests may be normal.
Functional impairment of the liver - elevated bilirubin, decreased albumin, elevated globulin,
prolonged prothrombin time
Etiologic screen for liver disease
Ceruloplasmin (Wilson's disease)
Iron, iron binding capacity, ferritin (hemochromatosis)
Alpha fetoprotein (hepatocellular cancer)
HBsAg (hepatitis B)
Anti-HCV (hepatitis C)
HCV RNA by PCR to confirm activity
ANA (autoimmune hepatitis)
Anti-smooth muscle antibody (autoimmune hepatitis)
Anti-mitochondrial antibody (AMA) (primary biliary cirrhosis)
Alpha 1-antitrypsin (deficiency)
Serum protein electrophoresis (SPEP) - increased IgG with any liver disease; increased
IgM with primary biliary cirrhosis (PBC)
Drugs that may alter lab results: N/A
Disorders that may alter lab results: N/A
PATHOLOGICAL FINDINGS
Fibrosis and regenerative nodules; specific findings/patterns may indicate etiology

Quantitative liver chemistry for iron, copper

Special stains for iron, copper, bilirubin, collagen, alpha 1-antitrypsin, hepatitis B
SPECIAL TESTS
Laparoscopic liver biopsy to reduce sampling error

Cholangiography to rule out common duct obstruction and recognize sclerosing cholangitis
Doppler ultrasound to indicate direction of flow in and patency of the portal and hepatic veins
Visceral angiography to determine vascular anatomy, patency and collaterals
Esophagogastroduodenoscopy to assess varices
IMAGING
Ultrasound good for detecting bile duct dilatation and space occupying lesions. Cannot make
diagnosis of cirrhosis based on ultrasound alone.
CT, if ultrasound technically inadequate
DIAGNOSTIC PROCEDURES
A liver biopsy establishes the diagnosis of cirrhosis

Patterns of injury as well as special stains may identify a precise etiology such as alcoholic liver
disease, hemochromatosis, alpha 1-antitrypsin deficiency, hepatitis B, primary biliary cirrhosis
Level of activity determined
TREATMENT
APPROPRIATE HEALTH CARE
Outpatient except for complicating emergencies:

GI bleeding
Unexplained decompensation
Renal failure
GENERAL MEASURES
Treatment designed to remove or alleviate underlying cause of cirrhosis, prevent further liver
damage and prevent complications

Phlebotomy for hemochromatosis

Therapies involve drug treatment, dietary restrictions, rest, other supportive measures. Adequate
protein intake for liver regeneration.
SURGICAL MEASURES
Interventional
Possible procedures for portal hypertension include - splenorenal or portacaval
anastomosis, transjugular intrahepatic portal-systemic shunt
Transplantation - in suitable candidate (substance free, motivated and adherent). Evaluate
prior to major decompensation.
ACTIVITY
Maintain as active as possible. With peripheral edema, leg elevation necessary.
DIET
Adequate protein (1 gm/kg) and generous calories to help regenerate the liver
In the presence of hepatic encephalopathy protein restriction is necessary
In the presence of ascites salt restriction is necessary (2 gm or less/day)
In the presence of hyponatremia (Na < 130 mEq [< 130 mmol]) fluid restriction is necessary (<
1 L)
No alcohol
PATIENT EDUCATION
Pamphlets are available through: American Liver Foundation, (800)223-0179

Additional material: National Digestive Diseases Information Clearinghouse, Box NDDIC,
Bethesda, MD 20892, (301)468-6344
MEDICATIONS
DRUG(S) OF CHOICE
235
Large esophageal varices seen at endoscopy prior to clinical bleeding - nonselective beta blocker

at a dose to decrease the resting pulse by 25%

Ascites: spironolactone 100 mg up to 400 mg every day as a single dose. Takes three days
before onset of action. Add furosemide 40-160 mg/day if needed. Loop diuretics may produce a
rapid diuresis and subsequent intravascular volume depletion.
Encephalopathy: lactulose (Cholac) 15-30 mL/day to produce 2-3 soft stools per day
Spontaneous bacterial peritonitis (SBP): cefotaxime alone as initial treatment; norfloxacin 400
mg/day or trimethoprim-sulfamethoxazole decreases risk for subsequent development of SBP after
treatment
Specific drug based on etiology
Wilson's disease: penicillamine, 125-250 mg qid (give on empty stomach to avoid
inactivation by metal binding)
Autoimmune chronic hepatitis: corticosteroids with or without azathioprine
Chronic hepatitis B and C: interferon
Hepatitis C relapse after interferon: interferon and ribavirin
PBC: ursodiol (ursodeoxycholic acid), 12-15 mg/kg/day in a bid dose
Contraindications: Refer to manufacturer's literature
Precautions: Refer to manufacturer's literature
Significant possible interactions: Refer to manufacturer's literature
ALTERNATIVE DRUGS
Ursodiol (ursodeoxycholic acid) for cholestatic disease
FOLLOWUP
PATIENT MONITORING
In a stable patient - yearly battery of liver tests. After 10 years, consider alpha-fetoprotein and
imaging to detect hepatocellular carcinoma.
In an unstable patient - tests may be repeated at weekly intervals
Have patient monitor weight and maintain a daily diary
Limit use of alcohol and other liver toxins

No sharing of syringes

Safer sex
Screening of family members when a genetic disease is recognized
Influenza and pneumococcal vaccines for cirrhosis patients exposed to crowds
Hepatitis A and B vaccines
Liver test surveillance while on hepatotoxic drugs (e.g., INH)
POSSIBLE COMPLICATIONS
Ascites
Jaundice
Coagulopathy
Bleeding esophageal varices
Liver failure
Carcinoma of the liver (uncommon)
Susceptibility to infections
Renal failure
EXPECTED COURSE/PROGNOSIS
A function of ongoing hepatic injury as well as residual hepatic reserve. If a treatable cause is identified
and intervention results in cessation of liver destruction, then the prognosis may be good.
MISCELLANEOUS
ASSOCIATED CONDITIONS
Hepatitis
Diseases and defects of the bile ducts
Cystic fibrosis
Heart failure
Hepatocellular cancer
AGE-RELATED FACTORS
Pediatric: N/A

Geriatric: Cirrhosis is one of the leading causes of death for people over age 65
Others: N/A
PREGNANCY
Cirrhosis may decompensate during pregnancy. Higher rates of spontaneous abortion, premature birth
and perinatal death.
SYNONYMS
N/A
ICD-9-CM
571.2 Alcoholic cirrhosis of liver
571.5 Cirrhosis of liver without mention of alcohol
SEE ALSO
N/A
OTHER NOTES
N/A
ABBREVIATIONS
PBC = Primary biliary cirrhosis
UGI = upper gastrointestinal
ANA = anti-nuclear antibody
REFERENCES
Sampliner RE: The recognition of early liver disease. Hospital Practice 1989; 53-56
Schiff L, Schiff ER, eds: Diseases of the Liver. Philadelphia, J.B. Lippincott co., 1993
Runyon BA: Care of patients with ascites. NEJM 1994; 330:337-342
Illustrations: N/A
Internet references: http://www.5mcc.com



April 21, 2000
Conde Petra

printing.
the handout.
All Topics
Medications
Copyright Integrative Medicine Communications 1999
Conditions and Treatments by Integrative Medicine Access
Category
Topics by Specialty Cirrhosis of the Liver
Adult Health Topics
Cirrhosis is irreversible chronic injury of the liver. It often has
Pediatric Health Topics no symptoms. Your health care provider will diagnose cirrhosis
Senior Health Topics based on your medical history, a physicial examination, and
laboratory tests.
Signs and Symptoms
The signs and symptoms of cirrhosis can range from an absence
of symptoms (in 10 to 20 percent of patients) to liver failure.
Cirrhosis can also have symptoms such as jaundice (yellowing
of the skin), weight loss, abdominal pain, testicular atrophy (in
men), menstrual irregularity (in women), swelling and fluid in
the abdomen, and enlarged veins.
What Causes It?
The most common cause of cirrhosis is alcoholism. Consuming
a lot of alcohol daily (32 to 48 oz. of beer, 4 to 8 oz. of liquor,
16 to 32 oz. of wine) for 10 years or more increases your
chances of developing cirrhosis. How much alcohol you drink
and for how long are more important than the type of alcohol
ingested. Between 5 and 10 percent of people in the United
States are alcoholics. Of these, 10 to 15 percent will develop
liver disease. Cirrhosis can also be caused by the ingestion of
drugs and toxins, infections, inherited medical conditions, and
cardiovascular diseases. About 10 percent of cases have no
known cause.

What to Expect at Your Provider's Office

Your health care provider will take a detailed history in order to
differentiate your liver disease from other conditions (obesity,
hepatitis, diabetes mellitus, biliary obstruction, drug toxicities,
infections, and other types of cirrhosis). Your provider will
order a complete blood count and liver function tests; in
addition he or she may order a liver biopsy. And finally, if your
cirrhosis is caused by alcoholism, your provider will strongly
urge you to stop drinking and will counsel you as to the risks
you are taking by continuing to drink. He or she may suggest
Alcoholics Anonymous as a good place to start your
rehabilitation.
Treatment Options
Your health care provider will treat you to try to slow the
progression of the cirrhosis and also to treat any complications
it causes.
Drug Therapies
Your health care provider may prescribe drugs to slow the

progression of the disease, such as Colchicine, diuretics,
Neomycin, and Lactulose.
Complementary and Alternative Therapies
Have much to offer in the treatment of liver disease.

Nutrition
B-complex: B1 (50 to 100 mg), B2 (50 mg), B3 (25 mg);
B5 (100 mg); B6 (50 to 100 mg), B12 (100 to 1,000 mcg),
folate (400 mcg per day) to reduce deficiencies common
in liver disease
Antioxidants: Vitamin C (1,000 to 3,000 mg per day),
vitamin E (400 to 800 IU per day), and selenium (200
mcg per day) reduce toxic effects of alcohol and drugs
and prevent tissue damage.
Essential fatty acids are anti-inflammatory; dietary
manipulation includes reducing animal fats and
increasing fish and nuts. A mix of omega-6 (evening
primrose) and omega-3 (flaxseed) may be best (1 tbsp. oil
per day or 1,000 to 1,500 mg per day).
Choline, lecithin, methionine (1 g each per day) for fat
absorption
Carnitine (300 mg per day) prevents fatty liver.
Glutathione (500 mg twice a day) helps remove ammonia
from the brain, a complication of cirrhosis.

Vitamin K is necessary for blood clotting; often depleted
in cirrhosis
Desiccated liver (500 mg three times per day) helps
provide nutrition to promote liver repair.
Restrict intake of protein to 45 g per day as long as a
minimum of 400 g of carbohydrates is ingested daily.
A change from animal to vegetable protein may be
helpful.
Herbs
As with any therapy, it is important to work with your health

care provider on getting your problem diagnosed before you
start any treatment. Herbs may be used as dried extracts
(capsules, powders, teas), glycerites (glycerine extracts), or
tinctures (alcohol extracts). Unless otherwise indicated, teas
should be made with 1 tsp. herb per cup of hot water. Steep
covered 5 to 10 minutes for leaf or flowers, and 10 to 20
minutes for roots. Drink 2 to 4 cups per day.
Due to the high doses required and the need to avoid alcohol,
the preferred form of these herbs is powdered.
Milk thistle (Silybum marianum): 100 mg three times per
day prevents free radical damage in the liver.
Barberry (Berberis vulgaris): 250 to 500 mg per day
corrects metabolic abnormalities in liver cirrhosis.
Catechin (Uncaria gambir): 400 mg three times per day,
is antioxidant, antiviral, and helps regenerate liver tissue.
Homeopathy
Homeopathy may be useful as a supportive therapy.

Physical Medicine
Castor oil pack. Used externally, castor oil is a powerful

anti-inflammatory. Apply oil directly to skin, cover with a clean
soft cloth (for example, flannel) and plastic wrap. Place a heat
source (hot water bottle or heating pad) over the pack and let sit
for 30 to 60 minutes. For best results, use for three consecutive
days. Apply pack over liver. Preliminary study shows immune
enhancement in healthy patients; was historically used to
stimulate liver function.
Acupuncture
May be helpful to alleviate symptoms and increase

physiological functioning.
Massage
May help alleviate stress and lymph congestion.

Following Up
Your health care provider will supervise and manage your
condition over the long term because cirrhosis can have serious
and life-threatening complications, particularly if you continue
to drink.
Special Considerations
Survival and management of cirrhosis is possible, especially
with proper treatment. Your health care provider will use
caution when prescribing medications if you have cirrhosis
because many medications cause complications in someone
with a weakened liver.

printing.
print the handout.


About Patient Education
http://home.mdconsult.com/das/patient/body/43/toc?node=9005606 [21/04/2000 23:42:41]

About Patient Education
AIDS Associated Tumors
AIDS Testing: ELISA and Western Blot Tests
AIDS and HIV Infection
Spanish
AIDS, Adolescents and
AIDS, Transmission of
Abdominal Pain from Gallstones (Biliary Colic)
Abdominal Pain, Brief Pediatric Version
Image
Spanish
Abdominal Pain, Pediatric Version
Abortion, Therapeutic
Abrasion, Corneal
Abrasions (Scrapes), Adolescent Version
Abrasions (Scrapes), Brief Pediatric Version
Spanish
Abrasions (Scrapes), Pediatric Version
Abscess, Tooth
Abuse of Older Adults
Spanish
Abuse, Emotional: The Effects on Children
Abuse, Sexual, Suggested Reading
Achilles Tendon Injury
Acne
Acne Rosacea
Acne, Adolescent
http://home.mdconsult.com/das/patient/body/43/toc?node=9005607 (1 of 6) [21/04/2000 23:43:03]

Acne, Brief Pediatric Version
Spanish
Acne, Newborn
Spanish
Acquired Immunodeficiency Syndrome HIV in Older Adults
Activities, Redirecting Child's
Activity During Illness, for Children
Adenitis, Cervical, Pediatric Version
Adenoid Surgery: Discharge Instructions
Adhesions
Adoption, Children and
Adoption, Suggested Reading
Aggressive Behavior, Preventing or Reducing a Child's
Aging, Health Changes with
Aging, Sex Related Changes and
Agoraphobia
Alcohol Dependence (Alcoholism)
Alcohol Dependence, Risk Factors for
Alcohol and Drug Abuse, Warning Signs, of Adolescent
Alcoholics, Children of
Alcoholism, Comprehensive Version
Alcoholism, Problems Related to
Allergic and Nonallergic Nasal Congestion (Rhinitis), Comprehensive Version
Allergies, Eye, Adolescent Version
Allergies, Eye, Brief Pediatric Version
Spanish
Allergies, Eye, Pediatric Version
Allergies: Book List
Allergy Diet, Corn

Allergy Diet, Egg
Allergy Diet, Milk
Allergy Diet, Peanut
Allergy Diet, Soy
Allergy Diet, Wheat
Allergy, Skin
Alopecia, Pediatric Version
Spanish
Alpha-Fetoprotein Screening
Altitude Sickness
Altitude Sickness, Pediatric Version
Alzheimer's Disease
Alzheimer's Disease, Behavior Problems Associated with
Alzheimer's Disease, Caregiver Issues
Alzheimer's Disease, Comprehensive Version
Amaurosis Fugax (Episodic Blindness)
Amebiasis, Gastrointestinal
Amniocentesis
Image
Amniocentesis: Brief Version
Spanish
Ampicillin, Amoxicillin, Rash from, Pediatric Version
Spanish
Amputation, Finger or Toe
Amyotrophic Lateral Sclerosis (ALS)
Anal Fissure, Pediatric Version
Anaphylactic Reaction, Pediatric Version
Spanish
Anaphylactoid Purpura: Admission and Discharge Instructions

Anemia
Anemia of Chronic Disease
Anemia, Comprehensive Version
Anemia, Iron Deficiency
Anemia, Iron Deficiency, Brief Pediatric Version
Spanish
Anemia, Iron Deficiency, Pediatric Version
Anemia, Sickle Cell
Anesthesia
Aneurysm, Dissecting
Aneurysmectomy, Abdominal Aortic
Anger, Suggested Reading
Angina
Angina and Coronary Artery Disease, Comprehensive Version
Angina, Effort
Angiogram, Abdominal Aortic
Angiogram, Coronary
Angiogram, Digital Subtraction
Angiography, Carotid
Angioplasty
Animal Bites, Prevention of, Children and
Spanish
Ankle Replacement, Total
Anorexia Nervosa
Anxiety
Anxiety Disorders, Comprehensive Version
Anxiety, Pediatric Version
Appendicitis
Appetite

Appetite, in Toddlers
Spanish
Arthritis
Image
Arthritis, Coping with
Arthritis, Rheumatoid
Image
Arthritis, Sexuality and
Arthrocentesis
Arthroscopy Meniscectomy
Arthroscopy, Knee
Assessment, Geriatric
Assisted Living
Assistive Devices
Asthma
Image
Spanish
Asthma in Adults, Comprehensive Version
Asthma in Children, Comprehensive Version
Asthma, Adolescent
Image
Asthma, Brief Version
Image
Spanish
Asthma, Exercise-Induced
Asthma, Peak Flow Meter, How to Monitor
Asthma, Pediatric Version
Image
Asthma, Pediatric, Environmental Control and

Asthma, Suggested Reading
Asthma: Admission and Discharge Instructions
Astigmatism
Atherosclerosis
Image
Athlete's Foot (Tinea Pedis) (for Teenagers)
Athlete's Foot, Adolescent
Athletic Amenorrhea
Atrophic Vaginitis
Attention Deficit Disorder, Pediatric Version
Spanish
Attention Deficit Disorder, Suggested Reading
Attention Deficit Hyperactivity Disorder, Comprehensive Version
Autism
Autism, Suggested Reading
Automobile Safety

April 21, 2000
Conde Petra

printing.
the handout.
All Topics
Medications
Category
Topics by Specialty
Adult Health Topics

Cirrhosis
What Is Cirrhosis?

The Liver.
The liver is the largest organ in the body, located immediately
below the diaphragm and occupying the entire upper right
quadrant of the abdomen. In the healthy adult, it weighs about
three pounds and is wedge shaped, with the upper part being
wider than the lower. The liver is rich in blood, holding about
13% of the body's supply. It is furnished with blood from two
large vessels--the portal vein and the hepatic artery (Hepatic
means liver). Blood that has circulated through the stomach,
spleen, and intestine enters the liver through the portal vein as
part of the so-called portal circulation system. The liver extracts
nutrients and toxins from this blood, which is then returned
through the hepatic vein to the right side of the heart. The
hepatic artery supplies blood to the liver directly from the heart.
The liver performs over 500 vital functions. It processes all of

the nutrients that the body requires, including proteins, glucose,
vitamins, cholesterol, and fats. It also renders harmless
potentially toxic substances, including alcohol, ammonia,
nicotine, drugs, and harmful by-products of digestion. The liver
synthesizes blood clotting factors and albumin, the major

protein in the blood. The liver also affects bilirubin, a

yellow-green pigment produced from the breakdown of
hemoglobin, the oxygen-carrying component in red blood cells.
The liver converts bilirubin into a water-soluble form, which is
then excreted into bile. Bile is a green-colored fluid that is
formed in the liver and contains, in addition to bilirubin, bile
salts, fatty acids, cholesterol and other substances. Bile travels
from the liver to the gall bladder, where it is stored until after a
meal; at that time it is secreted into the intestines to help digest
fat.
These vital processes rely on well-organized liver architecture

consisting of bile ducts, blood vessels, working liver tissue
(called the parenchyma), and supportive connective tissue. The
liver has two major lobes (with the right one much larger than
the left), which are separated from each other by walls of tough,
fibrous connective tissue. The lobes are composed of about
100,000 elegantly patterned lobules made of up liver cells and
blood vessels that form microscopic columns. Bracing the
corners of each lobule column are three small important
structures: an artery, a vein, and a bile duct. The arteries bring
oxygen-rich blood to nourish the liver cells. The blood passing
through the veins supplies the liver cells with the nutrients and
toxins that the liver cells process. Bile drains from tiny canals
around the liver cells into the corner ducts, which eventually
join to form the large common bile duct that leads from the liver
to the gall bladder. A central vein runs down through each
column, which collects the blood from the surrounding lobules
and acts as a tributary, eventually joining with other veins to
form the major hepatic veins. The hepatic veins, in turn, lead
into the inferior vena cava , the large vein that conducts the
worn-out blood from the liver back to the heart.
Cirrhosis.
Cirrhosis is an irreversible sequel to a number of disorders that
damage the liver cells and causes fibrosis (scarring). Often, this
process is accompanied by random clusters of regenerated liver
cells that develop throughout the liver, usually forming nodules
around the scarred areas. Eventually, this damaging pattern
becomes so extensive that the normal architecture of the liver is
distorted. Changes in the way blood and fluid flow in and out of
the liver also occur. The spleen overproduces a substance called
nitric oxide, which causes the blood vessels to relax and widen,
while vessels in other parts of the body, including the kidney,

narrow. The small blood vessels and bile ducts in the liver
constrict, so the blood that normally passes into the liver from
the intestine backs up through the portal vein and seeks other
routes. Twisted swollen veins called varices form in the
stomach and lower part of the esophagus. Bile builds up in the
blood stream, resulting in high levels of bilirubin, which causes
the yellowish cast in the skin called jaundice. Fluid build-up in
the abdomen (called ascites) and swelling in the arms and legs
is common. The liver enlarges in the first phases of the disease.
In advanced stages, however, the liver sometimes shrinks, a
condition called postnecrotic cirrhosis.
What Causes Cirrhosis?

Alcoholic Cirrhosis.
The liver is particularly endangered by alcohol. In the body,
alcohol breaks down into various chemicals, some of which are
very toxic in the liver. Alcoholic cirrhosis (also sometimes
referred to as portal, Laennec's, nutritional, or micronodular
cirrhosis) is the primary cause of cirrhosis in the U.S., and is
estimated to be responsible for 44% of deaths from cirrhosis in
North America. (Some believe this a low estimate. One
Canadian study found alcohol to be the major contributor to
80% of all cirrhosis deaths.) Over time, alcohol abuse leads to
increased demands for oxygen by the liver and, at the same
time, causes fat accumulation that impairs the liver's ability to
absorb oxygen. The immune system over-responds by
triggering an inflammatory process that damages and finally
kills liver cells, a condition called alcoholic hepatitis. During
the initial phase, the fat-laden liver becomes greatly enlarged,
but it eventually shrinks as web-like scars and small knots of
abnormal regenerated liver cells develop--the characteristics of
cirrhosis.
Cirrhosis from Chronic Hepatitis.

The next leading cause of cirrhosis in the U.S. is chronic
hepatitis, either hepatitis B or C. Chronic hepatitis C is the more
dangerous form and accounts for one-third of all cirrhosis cases.
Viruses or other mechanisms that cause hepatitis produce
inflammation in liver cells, resulting in their injury or
destruction. If the condition is severe enough, the cell damage

becomes progressive, building a layer of scar tissue over the

liver. In advanced cases, as with alcoholic cirrhosis, the liver
shrivels in size, a condition called postnecrotic or posthepatic
cirrhosis.
Primary Biliary Cirrhosis.

Primary biliary cirrhosis accounts for only 0.6% to 2% of
deaths from cirrhosis. It is most likely an autoimmune disease;
that is, the body's immune system attacks its own liver cells
mistaking them for foreign invaders (called antigens). In the
case of primary biliary cirrhosis, the cells under attack are in the
bile ducts. Liver cells are destroyed as the disease progresses.
Some research indicates that this autoimmune process may be
triggered in some patients by certain viruses, particularly those
known as retroviruses. Other autoimmune diseases, such as
scleroderma or Sjgren's syndrome, may also accompany this
form of cirrhosis.
Uncommon Causes of Cirrhosis.

Rare conditions that cause cirrhosis include hemochromatosis
(iron build-up in liver cells), which is fairly common in people
with diabetes and Wilson's disease (copper build-up in liver
cells). Liver damage that results in cirrhosis also may be caused
by a number of inherited diseases such as cystic fibrosis,
alpha-1 antitrypsin deficiency, galactosemia, and glycogen
storage diseases. Other rare causes include schistosomiasis (a
parasite found in the Far East, Africa, and South America),
long-term or high level exposure to certain chemicals and drugs
(e.g., arsenic, methotrexate, and toxic doses of vitamin A), and
small intestine bypass surgery (rarely, if ever, performed
anymore). Cancers that have metastasized to the liver, blood
clots in the hepatic or portal vein, or obstructions in the bile
duct can also cause changes that resemble cirrhosis.
Who Gets Cirrhosis?

Cirrhosis is the eleventh leading cause of death in the U.S. and
affects about three million Americans. The risk factors for liver
injury determine an individual's chances for cirrhosis.

People with Alcoholism.

About 10% to 35% of heavy drinkers develop alcoholic
hepatitis. After years of drinking, liver damage can be very
severe, leading to cirrhosis in about 10% to 20% of cases. Not
eating when drinking and consuming a variety of alcoholic
beverages are also factors that increase the risk for liver
damage. People with alcoholism are also at higher risk for
hepatitis B and C. People with alcoholism, in fact, should be
immunized against hepatitis B, and they may need a
higher-than-normal dose of the vaccine for it to be effective. [
See also Alcoholism.]
People with Chronic Hepatitis.

Because of blood screening, the risk for transmission for both
hepatitis B and C viruses through transfusions has dramatically
decreased since 1990. Hepatitis C can exist for decades,
however, without symptoms and nearly 300,000 people who
had transfusions before 1990 may have contracted the virus. In
addition, the causes of many cases of hepatitis C are unknown.
An estimated 10% to 60% of people originally infected with
hepatitis C patients develop the chronic form, which poses a
risk for cirrhosis of about 30%. Experts estimate that there may
be a 60% increase in the number of cirrhosis cases because of
rising numbers of people diagnosed with hepatitis C. It is not
clear, however, which hepatitis C patients will develop
cirrhosis. Alcohol use multiplies any average risk for cirrhosis
in hepatitis C patients. One large study also suggested that those
who contracted hepatitis after exposure in a hospital setting,
blood transfusion, and when the cause was unknown had a 20%
to 30% chance for cirrhosis. Those who developed hepatitis C
from drug abuse, sexual activity, and occupational exposure had
a lower risk--around 10%. The genetic type of the virus may
affect severity (type 1 being the most serious, type 2 and 3
posing less danger). A recent study suggested, however, that
regardless of genetic type, patients with a greater presence of
the virus in blood and liver tissue tended to have a more severe
case. About 3% to 5% of people infected with hepatitis B
develop the chronic form, and about half of these patients
develop cirrhosis. [ See also Hepatitis.]

Risk Factors for Primary Biliary Cirrhosis.

Up to 95% of primary biliary cirrhosis cases occur in women,
usually around age 50. Genetic factors are involved, but the
inheritance pattern is unclear.
What Are The Symptoms Of

Cirrhosis?
General Symptoms.
Fatigue and loss of energy are common early symptoms of
cirrhosis, along with loss of appetite and nausea, although many
people experience few symptoms at the onset of cirrhosis.
Spider angiomas may develop on the skin; these are
pinhead-sized red spots from which tiny blood vessels radiate.
Patients in later stages develop jaundice, a yellowish cast to the
skin and eyes, which is caused when the liver cannot process
bilirubin for elimination from the body. The palms of the hands
may be reddish and blotchy, a condition known as palmar
erythema. Patients may lose body hair. In men with alcoholic
cirrhosis, the testicles may atrophy and their breasts may
become swollen, sometimes painfully.
Symptoms of Complications.
A swollen belly is a sign of ascites, a condition that occurs
when fluid accumulates in the abdomen and the most common
major complication of cirrhosis. Fever, abdominal pain, and
tenderness when the belly is pressed indicate that the fluid is
infected. (Infection may occur, however, without any
symptoms.) Forgetfulness, unresponsiveness, and trouble
concentrating may be early symptoms of hepatic
encephalopathy, which is damage to the brain caused by
build-up of toxins. Sudden changes in the patient's mental state,
including agitation or confusion, may indicate an emergency
condition. Other symptoms include bad fruity-smelling breath
and tremor. Late stage symptoms of encephalopathy are stupor
and, eventually, coma.

Symptoms Specific to Rare Cases of

Cirrhosis.
People with primary biliary cirrhosis are subject to severe,
general itching and often develop small fatty yellow lumps
called xanthomas on the eyelids, hands, and elbows. They may
have an unpleasant condition called steatorrhea, in which the
feces contain excessive fat, causing them to float and to be very
foul smelling. In the rare disorder hemochromatosis, there is
often a bronze cast to the skin, an indication of iron build-up. A
thin bronze crescent bordering the cornea is called the
Kayser-Fleisher ring and is a sign of copper-build up in people
with Wilson's disease.
How Serious Is Cirrhosis?

General Outlook for Cirrhosis.
Cirrhosis is the seventh leading cause of death by disease in the
U.S., killing over 25,000 people each year. The most serious
complications of cirrhosis are bleeding, infections, and
encephalopathy--damage to the brain. Nearly every bodily
process is affected by a damaged liver, including those of the
digestive, hormonal, and circulatory systems. Less protein is
produced by the liver, for example, which causes fluid build-up,
bleeding problems, and susceptibility to infection. Additionally,
the liver cannot detoxify harmful substances, which accumulate
and impair brain function. Cirrhosis is also a cause of liver
cancer.
Cirrhosis is irreversible, but the rate of progression can be very

slow in some patients depending on its cause and other factors.
In patients with hepatitis B, for example, the five-year survival
rate after a diagnosis of cirrhosis is 71%. For alcoholics with
cirrhosis who abstain, a survival rate of five years or more can
be as high as 85%. For those who continue drinking, the chance
for living beyond five years is no higher than 60%. Often,
however, it is difficult to determine prognosis at the time of
diagnosis because the physician is usually unable to tell when
cirrhosis first occurred.
About two-thirds of patients with primary biliary cirrhosis never

develop symptoms and can have a normal life span. Once
symptoms of liver damage, such as jaundice, occur, however,
the average survival time declines. In one study of women

diagnosed with primary biliary cirrhosis, about 36% developed
symptoms over an 11-year period, and 11% either died or
required liver transplantation. Unfortunately, researchers are
unable to determine specific risk factors that could predict who
will develop a severe condition and who won't.
Portal Hypertension and its

Complications.
In cirrhosis, liver cell damage slows down blood flow and blood
pressure therefore increases. This pressure causes a back-up of
blood through the portal vein--a condition called portal
hypertension . The effects of portal hypertension can be
widespread and serious.
Ascites. Ascites is fluid build-up in the abdomen usually caused

by portal hypertension and is such a critical event in the
progression of cirrhosis that some experts refer to the phases of
cirrhosis as preascitic and ascitic. Once ascites occurs, only half
of patients survive after two years. Some physicians even
believe that ascites signals the need for liver transplantation,
particularly in patients whose cirrhosis is not due to alcohol.
Swelling can also occur in the arms and legs and in the spleen.
(Ascites can result from other conditions, and physicians should
check for other possibly serious causes, including cancer and
infections.) Ascites itself is not fatal, but it is uncomfortable and
can reduce breathing function and urination. Of interest is
research indicating that nitric oxide is overproduced in
cirrhosis; this substance is also associated with a condition
called hyperaldosteronism. Aldosterone is a steroid hormone
produced in the adrenal gland that regulates sodium and
potassium. High amounts ( hyperaldosterone) can result in a
fluid electrolyte imbalance called hypokalemic alkalosis (low
potassium levels, overly alkaline), which in turn can cause
weakness, abnormal heart rhythms, a sense of paralysis, and, in
severe cases, encephalopathy.
Variceal Bleeding . One of the most serious repercussions of

portal hypertension is the development of varices--blood vessels
that enlarge to provide an alternative pathway for blood
diverted from the liver. These abnormal blood vessels are
thin-walled, often twisted, and are subject to high pressure.
Varices often form in the stomach and esophagus (the tube

connecting the mouth and stomach). Internal bleeding from

these varices occurs in 20% to 30% of all cirrhosis patients. The
risk of death from a single episode can reach 70%. Recurrence
is common within two weeks of the first episode, but after six
weeks, the risk for recurrence is the same as for patients who
have not had a bleeding event. Factors that predict bleeding
include ascites, encephalopathy, large veins, and marked liver
cell abnormalities. It should be noted that moderate to severe
exercise may also increase the risk for variceal bleeding.
(Taking beta-blockers may reduce this risk.) Some researchers
believe that bacterial infection may be a primary factor in
triggering a bleeding episode.
Bleeding Disorders.
Gastrointestinal (GI) bleeding can occur from abnormal blood
clotting, often caused by deficiencies in vitamin K, low levels
of clotting proteins, and low counts of platelets (the blood cells
that normally initiate the clotting process).
Infections.
Abdominal infection occurs in up to 25% of patients with
cirrhosis within a year of diagnosis. At high risk are patients
whose tests results show very low protein levels and very high
bilirubin levels.
Encephalopathy.
Encephalopathy (damage to the brain) causes mental confusion
and, in worst cases, coma and death. The development of
encephalopathy is often precipitated by other problems,
including gastrointestinal bleeding, constipation, excessive
dietary protein, infection, surgery, or dehydration. No single
toxin accounts for the mental effects of encephalopathy. A
combination of conditions causes this serious complication,
such as the build-up in the blood of harmful intestinal toxins,
particularly ammonia, and an imbalance of amino acids that
effect the central nervous system.

Liver Cancer.
Cirrhosis greatly increases the risk for liver cancer, regardless
of the cause of cirrhosis.
Osteoporosis.
Primary biliary cirrhosis is associated with reduced bone
growth, partly because of the liver's inability to process vitamin
D and calcium and also from some of its treatments. As a result,
osteoporosis occurs in 20% to 30% of patients. Bone loss is also
a complication of liver disease in alcoholics and one study
indicated that it may also be a complication of cirrhosis caused
by hepatitis.
Insulin Resistance.
Nearly all patients with cirrhosis are insulin resistant. Insulin
resistance is a primary feature in type 2 diabetes and occurs
when the body is unable to use insulin, a hormone that is
important for delivering blood sugar and amino acids into cells
and helps determine whether these nutrients will be burned for
energy or stored for future use.
Other Complications.
One study reported that nearly a quarter of patients with
cirrhosis had gallstones. They may also face a higher than
average risk for certain abnormal heart rhythms. Peptic ulcers,
sleep disorders, and respiratory problems are also more
common in people with cirrhosis than in the general population.
How Is Cirrhosis Diagnosed?

Physical Examination.
The cirrhotic liver is often enlarged. It is also firm and may
even feel rock-hard. The left side can often be felt by the
physician when pressing on the abdomen. (In advanced stages,
however, the liver may become small and shriveled.) If the
abdomen is swollen, the physician will tap the flanks and listen

for a dull thud and feel for a shifting wave of fluid in the
abdomen--indications of ascites.
Biopsy.
Some experts are now recommending biopsies for all chronic
hepatitis C patients, regardless of severity, because of the risk
for liver damage even in patients without symptoms. A liver
biopsy is the only definite method for diagnosing cirrhosis. It
also helps determine its cause, treatment possibilities, the extent
of damage, and the long-term outlook. For example, hepatitis C
patients who show no significant liver scarring when biopsied
appear to have a low risk for cirrhosis. The procedure uses a
needle inserted through the abdomen to obtain a tissue sample
from the liver. The biopsy may also be performed during
peritoneoscopy--a procedure that uses a catheter and tiny
camera to view the surface of the liver. Biopsies can be
dangerous, so they cannot be performed on patients who have
test results that indicate clotting problems, on those who have
had previous liver biopsies, or who have ascites.
Blood Tests.
A number of blood tests may be performed to measure liver
function and to help determine the severity and cause of
cirrhosis. One of the most important factors indicative of liver
damage is bilirubin, a red-yellow pigment that is normally
metabolized in the liver and then excreted in the urine. In
patients with hepatitis, the liver cannot process bilirubin, and
blood levels of this substance rise, sometimes causing jaundice.
Measurements of blood levels of certain liver enzymes are
useful for diagnosing cirrhosis. To help determine outlook,
experts may use a calculation called a discriminant function
(DF), which uses two important measurements: serum albumin
concentration and prothrombin time (PT). Serum albumin
measures protein in the blood (low levels indicate poor liver
function). The PT test measures in seconds the time it takes for
blood clots to form (the longer it takes, the greater the risk for
bleeding).
Specific Blood Tests for Primary Biliary Cirrhosis . Very high

levels of serum alkaline phosphatase, an enzyme produced in
the liver, often occur in cases of primary biliary blood cirrhosis.
In 95% of people with primary biliary cirrhosis, blood tests will

detect high levels of immune factors called mitochondrial

antibodies. High antibody levels are not necessarily indicators
of severity.
Specific Blood Tests for Chronic Hepatitis . Blood tests for

chronic hepatitis include measurements of aminotransferase
levels, bilirubin levels, and detection of blood clotting
problems. Physicians will look particularly at blood levels of
enzymes known as aspartate (AST) and alanine (ALT)
aminotransferases. Research has suggested that a high ratio of
AST to ALT in chronic hepatitis C is a strong signal for the
presence of cirrhosis. Immunoassays to detect antibodies to
specific hepatitis viruses are also performed.
Caffeine Clearance Test.

A simple, inexpensive and harmless test that measures caffeine
in saliva may prove to be an accurate method for determining
the severity of cirrhosis.
Imaging Tests.
A number of imaging tests may be used to diagnose cirrhosis
and its complications. Magnetic resonance imaging (MRI),
computed tomography (CT), and ultrasound are all imaging
techniques that are useful in detecting and defining the extent of
cirrhosis. Such tests can reveal ascites, enlarged spleen,
irregular liver surface, reversed portal vein blood flow, and liver
cancer. Sometimes they can even detect abnormally large blood
vessels in the liver. Arteriography uses dye injected into the
hepatic arteries that then shows up on x-ray. Splenoportography
used uses dye injected into the spleen, which allows the
physician to measure portal vein pressure; this procedure is
risky.
Hepatic Vein Wedge Pressure.

Hepatic vein wedge pressure involves insertion of a catheter
into the hepatic veins. The blood pressure in the veins of the
liver is then measured; the result is an indicator of portal vein
pressure. If pressure is high, cirrhosis is likely. A low
measurement is a favorable sign.

Paracentesis.
If ascites is present, paracentesis is performed. This procedure
involves using a thin needle to withdraw fluid from the
abdomen. The fluid is tested for difference factors, including
protein levels, bacteria cultures, and white blood cell counts.
Low levels of protein in the fluid and a low white blood cell
count suggest that cirrhosis is the cause of the ascites. The
appearance of the fluid is helpful in determining a cause. For
example, a cloudy fluid plus a high white blood cell count mean
an infection is present. Bloody fluid suggests the presence of a
tumor.
What Are The Treatments For

Preventing Cirrhosis In
High-Risk People?
Treatment for Alcoholism.
The only treatment for alcoholic cirrhosis is to stop drinking.
[For more information on how to stop, see Alcoholism.]
Individuals with alcoholic cirrhosis are almost always
malnourished and, therefore, require increased calories and
rigorous nutritional support, which can improve survival rates.
Corticosteroids may be useful for alcoholic hepatitis, an acute
condition in which the liver is inflamed, but these drugs are not
beneficial after cirrhosis has developed. Drugs under
investigation include propylthiouracil and colchicine, which
inhibit deposits of collagen, the critical protein building block in
connective and scar tissue. Researchers are also investigating
drugs that block factors in the immune system called
thrombaxanes, which may play an important role in the
inflammatory process that kills liver cells in alcoholic cirrhoses.
Treatment for Chronic Hepatitis B or C.

Interferon alpha (Roferon-A, Intron A) is the standard drug
currently used for both chronic viral hepatitis B and C. In those
who respond, studies are showing improved symptoms, a
normal long-term survival rate, and, in some, no return of the
disease. Unfortunately, the percentage of patients who benefit

over the long-term using this drug as sole therapy is small. In

some cases, a very short course of corticosteroids may be used
initially to boost the effect of interferon. Even when the drug is
effective, the disease nearly always recurs and requires
additional treatment. Common side effects are flu-like
symptoms that usually occur within six hours and last for
twelve. More chronic effects include depression (which can be
very severe), anxiety, amnesia, confusion, irritability, changes
in sensation, weight loss, vomiting, general weakness, and
possible negative effects on cholesterol and lipid levels.
Interferon often causes a drop in platelet and white blood cell
counts, increasing susceptibility to bacterial infections. It may
also trigger an autoimmune response, possibly causing anemia,
diabetes, lupus-like symptoms, thyroid abnormalities, or even
autoimmune hepatitis. Other interferons being tested include
recombinant or consensus type-I interferon (Infergen, CIFN)
and interferon beta.
Standard Treatments for Hepatitis B. Interferon alpha has

eliminated the virus and sustained significant remission in 25%
to 40% of patients with chronic hepatitis B. The drug is usually
taken by injection every day for 16 weeks. (It does not appear to
be effective for hepatitis D.) Unfortunately, even in hepatitis B,
the virus recurs in almost all cases, although this recurring
mutation may be weaker than the original strain. Administering
the drug for longer periods may produce sustained remission in
more patients while still being safe. Interferon beta is benefiting
many children with hepatitis B who do not respond to interferon
alpha. The nucleoside analogue lamivudine has been shown to
reduce viral count in up to 56% of hepatitis B patients who took
it as sole therapy for about a year. Lamivudine is taken orally
and has very few side effects. Some researchers are concerned,
however, that, although the drug may prevent cirrhosis it does
not protect against liver cancer, particularly in those who have
harbored the virus since childhood. Small trials of the drug
adefovir, also called GS840, are reporting significant success in
reducing viral levels by over 99%.
Standard Treatments for Hepatitis C. At this time, qualified

patients with hepatitis C are treated with interferon alpha for
about 48 weeks with the goal of reducing viral levels to zero.
Patients who respond to interferon show improved symptoms
and long-term studies indicate that nearly all remain virus-free.
Interferon also appears to help prevent progression to liver
cancer in hepatitis C (although not in hepatitis B) in patients
who have cirrhosis. In patients who respond to the drug, it may
even reverse early cirrhosis and scarring. Unfortunately, the

virus remains undetectable six months after stopping treatment
in only about 15% to 20% of patients. (One small study
indicated that interferon may be even less effective for African
Americans with hepatitis C.) Extending treatments to up to 24
months have improved early relapse rates, but the regimen is
costly and side effects are more severe.
A combination of interferon alpha (Intron-A) and the nucleoside

analogue ribavirin (Rebetol), however, is proving to be five to
ten times as effective as interferon alone in sustaining a
response. This regimen may even benefit patients with
advanced fibrosis and cirrhosis. Side effects from the
combination are similar to interferon alone [ see above ],
although they are more common, and ribavirin adds the risk for
anemia. It is not clear yet how long patients should take the
combination regimen. Taking it for 48 weeks is slightly more
beneficial than taking it for 24 weeks, although side effects are
more severe. The longer-duration treatment, then, may
eventually be reserved for patients with more advanced disease.
Early studies indicated that the later the virus was eradicated,
the poorer the chance for a sustained remission. Many experts,
then, have advised stopping treatment after 12 weeks if patients
continue to show evidence of the virus. However, recent studies
are indicating that even if the virus is not eradicated until up to
24 weeks of treatments, remission is sustained in a large
proportion of patients taking either interferon alone or the
combination. Other drug combinations and interferon forms are
under investigation.
Drugs that Boost the Immune System . A number of drugs have

been developed that boost the body's own immune system to
fight the virus. For example, thymosin (Zadaxin) is a synthetic
version of a peptide derived from the thymus gland (which
produces immune factors call T-cells). It is showing promise
alone or in combination for patients with hepatitis B and C.
Vaccines, including Hepagene, are being investigated for
treating as well as preventing hepatitis B. Researchers are also
looking at interleukin 12, an immune factor that may enhance
the immune response to hepatitis C. [For more information, see
Hepatitis.]

What Lifestyle Factors Can

Help Manage Cirrhosis?
A healthy lifestyle is important for everyone, and particularly
for people with cirrhosis. Because important antioxidant
vitamins are depleted in the cirrhotic liver, people should
maintain a diet rich in fresh fruits, vegetables, and whole grains
that contain not only vitamins A, E, and C but also other
substances important for health. Vitamin supplements
themselves are not recommended except with the advice of a
physician. Hepatitis C is sometimes associated with abnormally
high iron levels. Such patients should avoid iron-rich foods,
such as red meats, liver, and iron-fortified cereals, and should
avoid cooking with iron-coated cookware and utensils.
Restricting salt consumption to less than 2,000 mg a day is
particularly important for patients with ascites. The less salt the
better. Fluid restriction is not usually necessary, but patients
with severe ascites should discuss limiting fluid with their
physicians. High-quality dietary protein may be helpful for
patients with ascites and for repairing muscle mass, but
excessive protein loads may trigger encephalopathy. Protein
solutions have been devised that provide beneficial amino acids
without including those that increase this risk. There is no limit
on vegetable proteins, such as those from soy. Exercise
increases the risk for portal pressure and variceal bleeding; a
recent study reported that taking a beta-blocker may reduce this
risk, although patients should discuss this with their physician.
How Are The Complications Of

Cirrhosis Managed And
Treated?
Treating Ascites.
Diuretics and Lifestyle Changes. Abstaining from alcohol,
restricting sodium intake, taking diuretics, usually
spironolactone (Aldactone) and furosemide (Lasix), are
effective for relieving ascites in 90% of patients. Sometimes
stopping drinking is enough to reverse this complication.
Previously, spironolactone was usually given alone, but experts
now use it by itself only in patients with minimal fluid build-up.
Patients should be monitored carefully for excessive and too
rapid fluid loss, which can set off complications, including

hypokalemia (dangerously low potassium levels), kidney
failure, or encephalopathy. Weight loss from diuretics usually
should not exceed one or two pounds a day, but there is no limit
for patients with massive swelling. Restricting fluid is not
usually necessary unless sodium levels in the blood are very
low. Physicians often recommend bed rest for patients with
ascites, but many experts believe this is not necessary and say
that studies do not support its benefits.
Paracentesis. If diuretics are not successful and the ascites is

very tense, patients may require large-volume paracentesis. In
this procedure, albumin (protein) is administered intravenously
while large volumes of fluid are removed through a tube in the
abdomen. Research indicates that four to six liters is usually
effective and safe. If the ascites does not respond to treatments,
paracentesis may need to be repeated every two weeks or more
frequently and up to 10 liters may need to be removed. Patients
who require this are probably not complying with dietary
requirements.
Liver Transplantation. Liver transplantation should be

considered for certain patients with ascites [ see below ],
particularly since the wait time for a liver donor is now very
long.
Peritoneovenous Shunting. Peritoneovenous (LeVeen, Denver)

shunting is an older, more invasive procedure, involving
insertion of a tube, or shunt, under the skin that routes the fluid
from the abdomen into the jugular vein. This procedure does
not improve survival, however, and can have serious
complications, including infection, blood clots, and rupture of
blood vessels in the esophagus. It is now reserved for patients
who are not candidates for repeat paracentesis or liver
transplantation.
Procedures for Preventing and Treating

Variceal Bleeding.
Endoscopic Sclerotherapy. Endoscopic sclerotherapy involves
inserting a tube through the mouth and then injecting into the
esophagus agents called sclerosants, which toughen the tissue
around the blood vessels. It may help prevent recurring bleeding
from the esophagus, although not from the stomach. The

treatment is repeated over a period of two or three months.

Unfortunately, bleeding still occurs in up to 50% of patients,
and complications of the procedure are high--about 40%. It is
also an unpleasant procedure and many patients cannot tolerate
it. A major analysis reported, however, that when used with the
sclerosant polidocanol, sclerotherapy is effective in helping to
reduce the risk for bleeding, particularly in high-risk patients.
Transjugular Intrahepatic Portosystemic Shunt (TIPS). An

alternative procedure is the transjugular intrahepatic
portosystemic shunt (TIPS). This procedure uses a thin,
flexible, metal tube, or stent, threaded through a catheter in a
vein that travels from the neck to the liver. There, the stent
remains, where it expands and forces blood vessels to reroute
around the scarred liver. Blockage or closure of the shunt can
develop over time. Comparative studies have not found a
survival advantage with TIPS over sclerotherapy; TIPS appears
to be better in preventing re-bleeding only when it is performed
shortly after a bleeding episode. TIPS also poses a higher risk
for encephalopathy than sclerotherapy does. At this time,
experts generally recommend TIPS only for patients who
cannot tolerate sclerotherapy, who are unlikely or unable to
comply with the repeated procedures necessary for
sclerotherapy, or who have poor blood circulation.
Endoscopic Band Ligation. Endoscopic band ligation simply

involves wrapping latex bands around the bleeding varices,
shutting off the blood supply. There are few complications. In
some studies it requires fewer sessions than sclerotherapy,
although one study found that it was not as effective as TIPS in
preventing re-bleeding.
Balloon Tamponade. Balloon tamponade employs a tube

inserted through the nose and down through the esophagus until
it reaches the upper part of the stomach. A balloon at the tube's
end is inflated and positioned tightly against the esophageal
wall. It is usually deflated in about 24 hours. Serious
complications can occur, the most dangerous being rupture of
the esophagus. A recurrence of bleeding following this
procedure is common.

Drug Therapies for Lowering Portal

Hypertension and Preventing Variceal
Bleeding.
Beta-Blockers. Beta-blockers, including propranolol and
nadolol, used alone or with sclerotherapy, reduce the heart rate
and can lower portal vein pressure, thereby reducing bleeding.
Beta-blockers are also used as a primary approach for
prevention of recurrence. It is not yet clear if these drugs are
more effective against bleeding than sclerotherapy, but they are
inexpensive and safe. Studies are indicating that a combination
of a nitrate, such as isosorbide mononitrate, with a beta-blocker
is less expensive and may, in some cases, be even more
effective than sclerotherapy. Nitrates alone are not as effective
as beta-blockers, but may be useful in certain people. Lifetime
therapy may be necessary.
Vasopressin. Vasopressin (Pitressin) with nitroglycerine is often

used. Vasopressin poses some risk to the heart, however, and it
is not clear whether it is actually helpful.
Somatostatin and Similar Drugs. Somatostatin is a hormone

that might help prevent variceal bleeding from portal
hypertension; in one study, the drug was administered within 12
hours of bleeding it and continued for 120 hours, the success of
subsequent sclerotherapy was higher than in those who did not
take the drug. Octreotide (Sandostatin), a drug that resembles
somatostatin, has been tested with mixed results; some have no
benefits in survival rates, although one indicated that it might
improve survival rates compared to sclerotherapy after taking
the drug for a year. It may reduce the risk of early re-bleeding in
patients treated with either beta-blockers, sclerotherapy, or both.
Angiotensin II Receptor Antagonists. Drugs known as

angiotensin II receptor antagonists, including losartan (Cozaar),
are being studied for lowering portal pressure.
Treating and Preventing Abdominal

Infection (Peritonitis).
Antibiotics are administered when ascites fluid examination and
tests indicate the presence of infection. For a first episode, the
antibiotic cefotaxime is typically administered intravenously,
requiring hospitalization. Treatment usually lasts 10 days but
research indicates that five days may be sufficient for certain

patients. Some research indicates that the oral antibiotic
ofloxacin may be as effective and without complications,
allowing patients to be treated at home.
In advanced cirrhosis, the risk for serious abdominal infection is

high and the antibiotic norfloxacin is often prescribed
preventively against specific organisms that infect the
abdominal cavity. One study found that preventive antibiotics
were very cost effective in high-risk patients. Another study
reported, however, that patients who took norfloxacin became
susceptible to Staphylococcal infections, which are not
ordinarily a problem in cirrhosis, and their survival rates were
similar to patients who did not take the antibiotic. Long-term
treatments with norfloxacin or similar antibiotics may increase
the risk for fungal infections after liver transplantation. More
research is needed.
Preventing and Treating Encephalopathy.

The first step in managing encephalopathy is to treat any
precipitating cause, if known, such as bleeding, high ammonia
levels, low oxygen, infection, dehydration, or use of sedatives.
Mild encephalopathy is managed by directing therapy toward
eliminating ammonia in the intestine. The first step is to restrict
animal protein, substituting meats and dairy products with
vegetable protein, such as soy, and amino acid supplements.
Enemas, which clean out the intestine, may be effective.
Lactulose (Cephulac) and lactitol, known as disaccharides, help
lower blood ammonia levels. Antibiotics, such as
metronidazole, rifamycin, or neomycin, are effective in
reducing levels of ammonia-producing bacteria in the intestine,
although long-term use of these drugs can cause toxic side
effects. Adding non-ammonia producing bacteria, including L.
acidophilus and E. faecium , to the intestine is showing promise
as a safe and effective treatment. In some studies, taking zinc
supplements have lowered ammonia levels in some patients
who were zinc-deficient, a common problem in cirrhosis. Some
studies indicate that manganese poisoning may be partially
responsible for encephalopathy in cirrhosis. Studies are needed
to determine if drugs that remove manganese improve this
complication.

Treatment for Gastrointestinal Bleeding.

Gastrointestinal (GI) bleeding is often first treated with
medications to reduce stomach acid. Reduced clotting factors or
platelets are common causes of GI bleeding in people with
alcoholic cirrhosis. Some will respond to three days of injected
vitamin K. People with alcoholism also often require folic acid.
Transfusions of replacement clotting factors or platelets may be
needed.
What Are The Primary

Treatments For Cirrhosis?
In general, no treatments exist for the basic mechanisms that
cause cirrhosis and its complications, although some therapies
are in clinical trials.
Treatment for Primary Biliary Cirrhosis.

Colchicine, a drug that inhibits collagen (a protein in the body
the makes up scar tissue) has produced some improvement in
liver function and survival. Side effects include nausea and
gastrointestinal distress. Ursodiol or ursodeoxycholic acid
(Actigall), a drug generally used to treat gallstones, improves
symptoms and other aspects of the disease and has only minor
side effects; unfortunately, it is very expensive and does not
appear to prolong survival. A similar drug,
tauroursodeoxycholate, may be more effective. Azathioprine,
methotrexate, and cyclosporine are drugs that suppress the
immune system, and, in studies, have shown modest benefits on
survival, but they have severe side effects. One study reported
that methotrexate caused remission in a group of patients, but,
although encouraging, other studies have indicated that many
patients do not respond to this drug. Corticosteroids, which
reduce inflammation, have been helpful in improving liver
function and symptoms, but patients should take measures to
prevent osteoporosis, a side effect of both the disease and these
drugs. The itching caused by primary biliary cirrhosis can be
relieved by taking cholestyramine with meals. The drug
naltrexone relieved itching in one study. (High doses of this
drug are toxic to the liver, but the low doses used in the study
were safe.) Phototherapy, which uses light, may also reduce
itching. Because primary biliary cirrhosis affects fat absorption,
patients may need high doses or injections of important

fat-soluble vitamins, including K, D, A, and E. Zinc
supplements may also be required. For steatorrhea, agents
called medium-chain triglycerides may be helpful.
Treatment for Other Forms of Cirrhosis.

Secondary biliary cirrhosis caused by blockage in the bile ducts
can be relieved by surgery. For hemochromatosis, weekly
bleedings (phlebotomies) may be performed until iron levels are
normal, then repeated as needed. If treatment is given before
cirrhosis develops, life expectancy may be normal.
D-penicillamine is the drug most used for Wilson's disease.
Investigative Therapies for Cirrhosis.

Researchers are testing certain drugs that will redress the
imbalances in circulation that lead to portal hypertension and
ascites. Of particularly interest are agents called V2 receptor
antagonists that may reverse the dilation in blood vessels that
lead to salt and fluid retention. Animal studies are reporting
success with genetic therapies using human hepatocyte growth
factor (HGF); even if feasible, however, human trials are years
away.
Liver Transplantation.
Liver transplantation may be an option for people with primary
biliary cirrhosis, for some people with alcoholic cirrhosis
(usually those who have completely abstained from alcohol for
more than six months), and for people with chronic hepatitis. It
should be noted, however, that hepatitis B patients have a
success rate of only 50% to 60% because of recurrence. (The
success rate is higher in those who have hepatitis D.) Either
lamiviudine or monthly infusions of hepatitis B immune
globulin (HBIg) after transplantation may help prevent
recurrence of hepatitis B after liver transplantation. Hepatitis C
also commonly returns in transplanted livers and progresses to
cirrhosis within an average of 51 months in 8% of patients.
Some people with cirrhosis and small localized liver cancers
may also be suitable candidates. Patients should seek medical
centers that have performed more than 50 transplants per year
which produces better than average results. Current survival

rates after a year are now nearly 90%, mainly due to

improvements in infection prevention. For people under 60, the
five-year survival rate is about 75%. It is much lower in elderly
people. Unfortunately, the waiting time for a liver doubled
between 1991 and 1996 to 241 days. Given the large number of
people with hepatitis C, this situation will almost certainly
worsen over in future years.
Where Else Can Help For

Cirrhosis Be Found
Primary Biliary Cirrhosis Patient Support Network Box 177

Tamworth, ON KOK 360
Canada
call (613-379-2534)
National Institute of Diabetes and Digestive and Kidney
Diseases (NIDDK)
National Digestive Diseases Information Clearinghouse
2 Information Way
Bethesda, MD 20892-3570
call (301-654-3810) or on the Internet
(http://www.niddk.nih.gov/ )
Centers for Disease Control and Prevention, Hepatitis Branch

1600 Clifton Road NE.
Mail Stop G37
Atlanta, GA 30333
For a special number on hepatitis call (888-4HEPCDC) or
(888-443-7232 )
or on the Internet
(http://www.cdc.gov/ncidod/diseases/hepatitis/hepatitis.htm
) This is an important source on hepatitis.

American Association for the Study of Liver Diseases

1200 19th Street NW, Suite 300
Washington, DC 20036-2422
call (202 429-5179) or on the Internet
(http://hepar-sfgh.ucsf.edu )
This group publishes the journal Hepatology.
For good journal abstracts on the web
(http://www.hepatology.org/ ) and
(http://www.ltsjournal.org/ )
Hepatitis Foundation International

30 Sunrise Terrace
Cedar Grove, New Jersey 07009
call (800-891-0707) or on the Internet (http://www.hepfi.org
)
This organization focuses just on viral hepatitis. It provides

educational materials, offers support by phone, and gives
referral to other physicians.
American Liver Foundation

75 Maiden Lane, Suite 603
New York, NY 10038
call (800-GO LIVER) or (800-465-4837) or on the Internet
(http://www.liverfoundation.org/ )
The foundation has regional chapters, some with support
groups. The hepatitis hotline provides patient information,
brochures, and video and audio tapes.
American Gastrointestinal Association

7910 Woodmont Ave., Seventh Floor,
Bethesda, MD 20814
call (301-654-2055) or on the Internet (http://www.gastro.org/

This is an association for physicians and other professionals.

They provide names of gastroenterologists in local areas.
On the Internet
Excellent site on liver diseases from Columbia-Presbyterian
Medical Center
(http://cpmcnet.columbia.edu/dept/gi/disliv.html )
For information on organ transplantation, United Network for
Organ Sharing (http://www.unos.org/ )
The Hepatitis Information Network (http://www.hepnet.com/

)
Alcoholics Anonymous
General Service Office
475 Riverside Drive
New York, NY 10015
National Clearinghouse of Alcohol and Drug Information

PO Box 2345
Rockville, MD 20852
Call (800-729-6686) or on the Internet (http://www.health.org/
)
National Institute on Alcohol Abuse and Alcoholism

6000 Executive Boulevard - Willco Building
Bethesda, Maryland 20892-7003
On the Internet (http://www.niaaa.nih.gov/ )

National Council on Alcoholism

12 West 21 Street
New York, NY 10010
Call (800-NCA-CALL) or on the Internet
(http://www.ncadd.org/ ).
On the Internet:
Medical Center
Or for organ donation information see (www.organdonor.org
)

)
Recent Literature
AASLD Practice guidelines management of adult patients with
ascites caused by cirrhosis. Hepatology, January 1998 Vol. 27.
Combination therapy for hepatitis C infection (and other

articles). The New England Journal of Medicine, 11/19/98, Vol.
339.
Effects of propranolol on the hepatic hemodynamic response to

physical exercise in patients with cirrhosis. Hepatology 1998,
Vol. 28.

High prevalence of sleep disturbance in cirrhosis. Hepatology,

1998 Vol. 27.
Mechanisms of disease: nitric oxide as a mediator of

hemodynamic abnormalities and sodium and water retention in
cirrhosis. The New England Journal of Medicine, 8/20/98 Vol.
339.
ST/ALT ratio predicts cirrhosis in patients with chronic

hepatitis C virus infection. American Journal of
Gastroenterology, January 1998
Treatment of chronic hepatitis B infection. The New England

Journal of Medicine, 7/9/98, Vol. 338

printing.
print the handout.


April 21, 2000
Conde Petra

printing.
the handout.
All Topics
Medications
Copyright Clinical Reference Systems 1999
Conditions and Treatments by Adult Health Advisor
Senior Health Advisor
Category Women's Health Advisor
Topics by Specialty
Adult Health Topics Alcohol-Related Problems


What are alcohol-related problems?
Women's Health Topics People who drink too much alcohol are at risk for having
physical, mental, and emotional problems. This is especially
true for the elderly, who often have several other medical
problems. The amount of alcohol considered excessive varies
widely from person to person. For some people two drinks a
day can lead to problems while others can consume the same
amount without any harmful effects. Recent studies suggest that
problems with health can begin when more than two drinks per
day are consumed.
How do alcohol-related problems occur?

Excessive drinking over time can lead to tissue damage and
disease. Alcohol has a toxic or irritating effect on cells and
tissues. Also, many people who drink too much do not always
follow a nutritious diet because alcohol reduces their appetite
and replaces healthier foods.
Drinking too much alcohol leads to a loss of protein, minerals,
and vitamins, especially thiamine (vitamin B-1) and other
B-complex vitamins. Thiamine is essential for the body's
normal use of carbohydrates. Also, the nerves, muscles, heart,
and brain need thiamine to function normally.
Alcohol can disrupt the normal body chemistry in the following

ways:
Alcohol increases the acidity of the stomach, which can
damage the stomach lining. Alcohol can also damage the
lining of the small intestine and interfere with the
absorption of nutrients from the small intestine.
Alcohol decreases the amount of deep sleep, which may
mean a less restful night's sleep. Also, drinking alcohol
can make insomnia worse. Although it may help a person
to go to sleep, as the drink wears off, the brain cells
become more agitated and may cause an early
awakening. One may also wake up more often to urinate.
Alcohol can cause gout. Gout occurs when the buildup of
uric acid causes inflammation in the joints. Alcohol can
reduce the body's ability to get rid of uric acid.
What are some physical problems that

are alcohol-related?
Excessive drinking can cause cancer, liver damage, nervous
system problems, heart and circulatory problems, and other
physical and behavioral problems in the elderly.
Cancer. Drinking too much alcohol increases the risk of
cancer of the liver, pancreas, mouth, tongue, pharynx,
larynx, and esophagus. This risk is even greater if
tobacco products are also used.
Liver damage. Besides liver cancer, alcohol may cause
fatty liver, hepatitis, cirrhosis, and other liver disease.
Nervous system problems. Alcohol can create a
deficiency of thiamine, which can disrupt the functioning
of the nerves. This deficiency can cause:
Wernicke's encephalopathy, which may involve
eye movement disturbances, confusion, speech
disturbances, walking disturbances, and, over time,
coma
Korsakoff's psychosis, a chronic condition which
causes short-term memory loss and affects the
nerves in the brain and spinal cord, causing pain,
cramps, numbness, tingling, and weakness in the
legs and hands.
Heart and circulatory problems. Heart failure can be
caused by a lack of thiamine. In addition, excessive
alcohol use may contribute to hypertension (high blood

pressure) and strokes. Alcohol is toxic, or harmful, to

heart tissue and can lead to cardiomyopathy (heart
muscle disease).
Other physical problems. The irritating effects of alcohol
may cause:
gastritis (inflammation of the mucous membrane
that lines the stomach)
pancreatitis (inflammation of the pancreas)
peptic ulcer (a raw area in the lining of the

gastrointestinal tract)
blood sugar levels to increase and therefore
diabetes to worsen
interactions with medications
worsening of most other medical problems (heart

disease, kidney disease, circulation problems)
injuries from falls and other accidents.
What are some other alcohol-related

problems?
Psychiatric and emotional problems can include:
anxiety (feelings of uneasiness and fear)
depression (feelings of sadness and hopelessness, a loss

of interest in life, and a reduced feeling of well-being).
confusion (both short-term and permanent)
addiction.
Regularly drinking too much alcohol also can cause major, even
life-threatening, emotional and behavioral problems including:
Marital and family problems. People who drink heavily
can have changes in moods and emotions, causing them
to become angry and irritable. Also, they can have
personality changes, such as becoming suspicious,
jealous, or possessive. The spouses, children, and lovers
of heavy drinkers find it hard to cope with these outbursts
and changes. The stable routines of the household and
family life are at risk. Family members may also develop
emotional problems. The children are at high risk of
becoming alcoholics as adults.
Acts of violence. All people who drink too much alcohol
lose proper judgment. As drinking problems become
worse, arguing can lead to fighting, at home as well as in
the workplace and in social settings. Also, these
arguments can lead to domestic violence, child abuse,

and the loss of job and friends. A large percentage of
violent crimes, child sexual abuse, assaults, and murders
are related to alcohol abuse.
Vehicle accidents. Drunk drivers cause an estimated 50%
of the deaths from vehicle accidents in the U.S.

printing.
print the handout.


April 21, 2000
Conde Petra
April 14, 2000
Question 3 - Complications of Cirrhosis

knowledge.
The manifestations of cirrhosis are divided into those related to hepatocellular

injury (previously described) and to mechanical factors. The main complication of
cirrhosis is portal hypertension. In the normal liver the resistance to portal venous
flow is very low. However, when the liver architecture is altered (cirrhosis),
resistance increases. This increase in resistance, along with the increase in portal
venous inflow, is the mechanism for the development of portal hypertension.
Cirrhosis is the main cause of this elevated resistance, but other causes include
portal vein occlusion, hepatic vein occlusion (Budd-Chiari syndrome),
veno-occlusive disease (seen after bone marrow transplant, or chemotherapeutic
agents), schistosomiasis, alcoholic hepatitis, sarcoidosis, and other disorders that
alter normal venous architecture. Varices form because venous blood flows to
vasculature of lower resistance. These include the system connecting the portal
vein to the azygos vein, gastric veins, and hemorrhoidal veins. The increase in flow
to this system leads to varices in the submucosa of the lower esophagus and
gastric fundus. These varices may bleed when the portal pressure is 12 mm Hg
above the inferior vena cava pressure.
Variceal bleeding is the most lethal complication of portal hypertension. Clinically

the varices are silent until rupture, which most commonly presents with
hematemesis but can also present with melena or hematochezia. The bleeding
may be massive with 40% fatality for each episode and 90% chance of recurrence
within 1 year. Diagnosis is made by upper endoscopy. Endoscopy should focus on
the lower esophagus, as well as the fundus of the stomach, since 10% of patients
bleed from gastric varices. Diagnosis depends on high clinical suspicion in the right
patient, presence of hematemesis or lower GI bleeding, and nasogastric tube
recovery of heme-positive material. Management of variceal bleeding emphasizes
restoration of circulating blood volume and control of the hemorrhage.
Pharmacologic agents used to control the bleeding include vasopressin and

somatostatin. Vasopressin is a hormone derived from the posterior pituitary that
causes vasoconstriction of the splanchnic circulation. NTG should be administered
concurrently to counteract its systemic vasoconstriction with end-organ ischemia.
Other measures used to control bleeding include balloon tamponade (Sengstaken
-Blakemore tube) and endoscopic-injection sclerotherapy. It is important to
remember that balloon tamponade is only used as a temporary treatment.
Emergency shunt surgery is reserved for variceal bleeding that is refractory to the
above measures and is associated with 25% to 50% mortality.
Management of persistent or recurrent variceal bleeding may involve use of the

transjugular intrahepatic portosystemic stent (TIPS), an experimental procedure.
This stent, placed with the help of fluoroscopic guidance, shunts venous blood flow
from the high resistance portal system to the right hepatic vein. This intervention
reduces the portal venous pressure gradient (usually to less than 12 mm Hg),
reducing intravariceal pressure and preventing rupture and further bleeding. The
main complication of this procedure is encephalopathy because blood perfuses the
central nervous system without having passed through the liver first. The incidence
of encephalopathy depends on the shunt diameter (smaller shunts, 8 mm, result in
decreased incidence of encephalopathy).
Variceal bleeding is associated with high morbidity and mortality; therefore

prophylactic measures are essential. Prophylactic portosystemic shunts have been
used and decrease the rate of rebleeding, but have not been shown to alter
survival. Propranolol, a beta-adrenergic blocker, reduces mortality associated with
bleeding. This drug has been shown to prevent the first episode of bleeding and to
effect a decreased rate of rebleeding. Propranolol dosing needs to be sufficient to
reduce the resting heart rate by 25%, which results in decreased variceal wall
tension. Patient compliance is a major concern, since abrupt discontinuation can
lead to rebound hypertension with acute rebleeding.
Read more about portal hypertension and espohageal varicies:

Proceed With Case
Etiology And Pathophysiology Of Portal Hypertension

Natural History And Prognosis Of Variceal Hemorrhage

The changing spectrum of treatment for variceal bleeding.

Rikkers LF - Ann Surg - 1998 Oct; 228(4): 536-46
From NIH/NLM MEDLINE, HealthSTAR
Observation of thoracic duct morphology in portal hypertension by endoscopic

ultrasound.
Parasher VK - Gastrointest Endosc - 1998 Dec; 48(6): 588-92
NATURAL HISTORY OF PORTAL HYPERTENSION

Boyer TD - Clinics in Liver Disease - 1997 May; 1(1); 31-44
PATHOPHYSIOLOGY OF PORTAL HYPERTENSION

Gupta TK - Clinics in Liver Disease - 1997 May; 1(1); 1-12

Is portal hypertension due to liver cirrhosis a major factor in the development of

portal hypertensive gastropathy?
Bayraktar Y - Am J Gastroenterol - 1996 Mar; 91(3): 554-8
Abnormalities of gastric emptying in portal hypertension.

Balan KK - Am J Gastroenterol - 1996 Mar; 91(3): 530-4
PORTAL HYPERTENSION, VARICES, AND TRANSJUGULAR INTRAHEPATIC

PORTOSYSTEMIC SHUNTS
Conn HO - Clinics in Liver Disease - 2000 Feb; 4(1); 133-150
THE ROLE OF SURGERY IN THE TREATMENT OF PORTAL HYPERTENSION

Iannitti DA - Clinics in Liver Disease - 1997 May; 1(1); 99-114
NEW CONCEPTS IN THE PATHOGENESIS OF PORTAL HYPERTENSION

Hepatic Wounding and Stellate Cell Contractility
Rockey DC - Clinics in Liver Disease - 1997 May; 1(1); 13-29
PREFACE
LABRECQUE DR - Clinics in Liver Disease - 1997 May; 1(1); 17A-20A
Diagnosis and treatment of gastrointestinal bleeding secondary to portal

hypertension. American College of Gastroenterology Practice Parameters
Committee.
Grace ND - Am J Gastroenterol - 1997 Jul; 92(7): 1081-91
Technology Assessment Status Evaluation update: transvenous intrahepatic

portosystemic shunt (TIPS). April, 1997. ASGE. American Society for
Gastrointestinal Endoscopy.
Gastrointest Endosc - 1998 Jun; 47(6): 584-7
The role of endoscopic therapy in the management of variceal hemorrhage.

American Society for Gastrointestinal Endoscopy.
Gastrointest Endosc - 1998 Dec; 48(6): 697-8
Cirrhosis
Well-Connected


April 21, 2000
Conde Petra

Company
Section 9 - Liver
77 - Portal Hypertension and Feldman: Sleisenger & Fordtran's Gastrointestinal and Liver Disease,
Gastrointestinal Bleeding
ETIOLOGY AND
PATHOPHYSIOLOGY OF
PORTAL HYPERTENSION ETIOLOGY AND PATHOPHYSIOLOGY OF
Normal Liver Blood Flow PORTAL HYPERTENSION
Hemodynamic Alterations in Normal Liver Blood Flow
Portal Hypertension
The normal anatomy of the portal and hepatic arterial
General Principles
circulation is described in Chapter 62 (see Fig. 62-3) . Normal
Role of Increased hepatic blood flow is approximately 1500 mL/min, which
Resistance represents 15% to 20% of cardiac output. One third of this flow
and 30% to 60% of the oxygen consumed by the liver are
Portal Blood Flow
provided by the hepatic artery. Approximately two thirds of the
Primary High Portal Flow hepatic blood supply is provided by portal venous blood. [7] [11]
States. The high-pressure, well-oxygenated arterial blood mixes
completely with the low-pressure, low-oxygenated,
The Portosystemic
nutrient-rich portal venous blood within the hepatic sinusoids.
Collateral Paradox.
After perfusion into the sinusoids, blood flows sequentially into
Hyperdynamic Circulation the hepatic venules, hepatic veins, and inferior vena cava. A
of Portal Hypertension. fraction of the plasma entering the space of Disse is drained into
lymphatic vessels.
Vasoactive Mediators in the
Pathogenesis of Portal A unique feature of the normal hepatic sinusoidal
Hypertension microcirculation is its low perfusion pressure. This is attributed
Vasoactive Mediators and
to the unusually high precapillary-to-postcapillary resistance in
Splanchnic Vasodilation the liver. [11] It appears that the sinusoids are normally protected
from upstream portal perfusion pressure and accompanying
Vasoactive Mediators and fluctuations by a presinusoidal site of high resistance, which is
Vascular Resistance probably located within the terminal portal venous radicals. [15]
Intrahepatic Vascular
Because the sinusoids are lined by an endothelium that lacks a
Resistance. continuous basement membrane and contains a multitude of
Portocollateral 1285
Resistance.
large (50 to 200 nm), highly permeable fenestrae, maintenance

of a low pressure in the hepatic sinusoids is critical to the
maintenance of normal rates of sinusoidal fluid transudation
into the space of Disse (see Chapter 78) .
Another feature that is unique to the hepatic circulation is the

close interrelationship between blood flow in the portal vein
and that in the hepatic artery. When portal blood flow increases,
hepatic arterial flow decreases; when portal flow decreases,
hepatic arterial flow increases. This phenomenon has been
termed the hepatic arterial buffer response and is an
adenosine-mediated vascular reflex that ensures the
maintenance of a relatively constant state of sinusoidal
perfusion in the face of changes in portal inflow that occur, for
example, with meals. [7] [11]
Hemodynamic Alterations in Portal Hypertension
General Principles
The pathogenesis of portal hypertension involves the

relationship between portal venous blood flow and the
resistance to this blood flow within the liver (portohepatic
resistance) and within portosystemic collateral blood vessels
(portocollateral resistance) that forms during the evolution of
portal hypertension (Fig. 77-1) .
The movement of blood within a vascular system such as the

portal system is driven by a pressure difference or gradient
existing along the length of that system. The portal pressure
gradient (Delta P; i.e., the difference in pressure between the
portal and systemic venous systems) is the product of portal
venous blood flow (Q) and the vascular resistance to this flow
(R), as expressed by Ohm's law:
Delta P = Q R
In turn, R is derived by Poiseuille's law from the following
relationship:
R = 8eta l/pi r4
where eta is the coefficient of viscosity, l is the length of the
vessel, and r is the radius. It is clear from this relationship that
small changes in vessel radius will lead to disproportionate
increases in resistance and, thus, pressure. For example, the
incremental influence on resistance of increasing hematocrit,
and therefore viscosity, through a blood transfusion

administered to a patient who has recently bled from varices can

also be appreciated. [11] Normal, uncorrected pressure in the
portal vein ranges between 5 and 10 mm Hg and is influenced
by intra-abdominal pressure. Hepatic venous pressure is
similarly affected by intra-abdominal pressure and also reflects
central venous filling pressure. To eliminate the contribution of
intra-abdominal pressure and central venous pressures and thus
express portal pressure as the intrinsic pressure difference
between the portal and systemic venous compartments, it has
become standard practice to express portal pressure as a portal
pressure gradient. In practical terms, this is most often
determined in patients with cirrhosis as the hepatic venous
pressure gradient (HVPG), a technique that is similar in
principle to measurement of pulmonary capillary wedge
pressure (Fig 77-2) . [3] [4] In brief, a pressure measurement is
made with the use of a catheter wedged into a hepatic vein via
either a femoral or transjugular approach. This measurement is
termed the wedged hepatic venous pressure (WHVP). After
withdrawal of the catheter into the hepatic vein, a free hepatic
vein pressure (FHVP) is obtained. The HVPG is obtained by
subtracting the value of the WHVP from that of the FHVP.
Normally, the HVPG does not exceed 5 mm Hg. Portal
hypertension exists if the value for the HVPG exceeds this
value.
It is apparent that a rise in portal pressure could, in theory,
result from an increase in either portal flow or vascular
resistance. Historically, increased flow into the portal system
from a massively enlarged spleen was believed to be the cause
of elevated portal pressure in patients with cirrhosis (Banti's
hypothesis). This view was subsequently abandoned when the
crucial role of increased resistance to portal flow at various sites
became evident. Since the late 1970s, the pendulum has swung
again, thanks to a plethora of elegant experimental and clinical
studies that have demonstrated the presence of a hyperdynamic
circulatory state in portal hypertension with increased
splanchnic blood flow that contributes to the increase
Figure 77-1 Hemodynamic principles in portal

hypertension. The schematic summarizes the
contribution and interplay of the major hemodynamic
forces underlying the pathogenesis of portal
hypertension. (See also Fig. 77-2 and Table 77-3.)
1286

Figure 77-2 Sites of obstruction (resistance) to portal

venous flow and measurement of portal pressure. The
schematic illustrates the major locations of extrahepatic
(prehepatic and posthepatic) and intrahepatic
(presinusoidal, sinusoidal, and postsinusoidal)
obstruction. A catheter tip is also shown wedged into a
small hepatic vein for the measurement of the WHVP.
When the catheter is withdrawn into the hepatic vein, the
FHVP is obtained. HVPG = WHVP - FHVP. Direct measurement of the
PVP is accomplished intraoperatively either by catheterization of the
umbilical vein or by catheterization of the portal vein via the transjugular
or transhepatic approach. HVPG, hepatic venous pressure gradient;
WHVP, wedged hepatic venous pressure; FHVP, free hepatic vein
pressure; PVP, portal venous pressure.
in portal pressure. The relative extents to which increased
splanchnic portal inflow (the "forward" force) and resistance
(the "backward" force) are responsible for the increase in portal
pressure in any given clinical situation remain a much-debated
issue and are of considerable practical importance in the
understanding of the fundamental pathophysiology of portal
hypertension as well as in the development of rational
pharmacologic therapy.
Role of Increased Resistance
Cirrhosis, usually caused by alcohol ingestion or chronic viral

hepatitis, is the most significant cause of portal hypertension in
Western populations, but there are many other causes, most of
which are noncirrhotic; schistosomiasis is a particularly
prevalent cause in developing countries. Knowledge of the
pathology of these conditions has confirmed the existence of
resistance to portal flow at a variety of different anatomic levels
in the pathogenesis of portal hypertension. This is reflected in
the conventional classification of the causes of portal
hypertension according to the localization of the site of
maximal resistance to portal flow (see Fig. 77-2 and Table
77-1) . Regarding the three major categories of prehepatic,
intrahepatic, and posthepatic, the site of increased resistance is
usually obvious. Thus, portal vein thrombosis exemplifies
prehepatic portal hypertension, whereas inferior vena caval web
typifies posthepatic portal hypertension. For intrahepatic causes,
the site of resistance is conventionally further subdivided into
presinusoidal, sinusoidal, and postsinusoidal. Precise
classification of portal hypertension according to the
intrahepatic site of maximal resistance has been limited by the
lack of technical means to measure pressure directly within the
hepatic sinusoids. Most of the information regarding this has
been provided from the combined results of direct measurement
of pressure in the portal system and indirect estimation
1287
TABLE 77-1 -- Causes of Portal Hypertension

Primary Increased Flow
Arterioportal venous fistula
Intrahepatic
Intrasplenic
Splanchnic
Splenic capillary hemangiomatosis
Primary Increased Resistance
Prehepatic
Thrombosis/cavernous transformation of the portal vein
Splenic vein thrombosis
Intrahepatic
Presinusoidal
Schistosomiasis *
Sarcoidosis *
Myeloproliferative diseases/myelofibrosis *
Congenital hepatic fibrosis
Idiopathic portal hypertension (hepatoportal sclerosis)
Chronic arsenic hepatotoxicity
Azathioprine hepatotoxicity
Vinyl chloride hepatotoxicity
Early primary biliary cirrhosis *
Early primary sclerosing cholangitis *
Sinusoidal/mixed
Cirrhosis secondary to chronic hepatitis
Alcoholic cirrhosis
Cryptogenic cirrhosis

Methotrexate
Alcoholic hepatitis
Hypervitaminosis A
Incomplete septal fibrosis
Nodular regenerative hyperplasia
Postsinusoidal
Veno-occlusive disease
Hepatic vein thrombosis (Budd-Chiari syndrome)
Posthepatic
Inferior vena caval web
Constrictive pericarditis
Tricuspid insufficiency
Severe right heart failure
*Usually presinusoidal when early; often progressing to a
sinusoidal/mixed type of resistance whenmore advanced.
of the intrasinusoidal pressure from the WHVP, with details of

the morbid anatomy taken into consideration. [3] [11] For
example, in both prehepatic and intrahepatic presinusoidal
portal hypertension, the directly measured portal venous
pressure (PVP) is always elevated in the presence of normal
WHVP and HVPG. In sinusoidal and intrahepatic
postsinusoidal portal hypertension, the WHVP tends to
approximate or equal the directly measured portal pressure, and
the HVPG is proportionately increased. In posthepatic portal
hypertension, the WHVP equals the increased PVP. Because the
FHVP is also abnormally high in this scenario, the HVPG is
usually normal.
As a rule, the intrahepatic localization of maximal resistance
tends to be more clearly established in the noncirrhotic diseases
than in cirrhosis. [16] Thus, veno-occlusive disease, resulting
from either pyrrolizidine alkaloids or chemotherapeutic agents
(see Chapter 73) , is readily apparent as a cause of
postsinusoidal, intrahepatic portal hypertension, whereas early
hepatic schistosomiasis with its characteristic granulomatous
portal inflammation and fibrosis is a well-studied example of

presinusoidal portal hypertension. However, even such widely

accepted examples of presinusoidal portal hypertension as
schistosomiasis [16] [17] and idiopathic portal hypertension [18]
become more complicated when one considers that over time,
hepatic pathology may progress and generate a more complex
or mixed pattern of vascular resistance.
The main site of portal vascular resistance in cirrhosis has been
difficult to establish. Most measurements in nonalcoholic
cirrhosis indicate a higher value for PVP than for WHVP.
Given that the WHVP is an estimate of the intrasinusoidal
pressure, this has been variously interpreted to indicate either
the presence of a presinusoidal component of resistance,
probably related to inflammatory activity or fibrotic changes in
the portal tracts, [19] or a significant presence of intersinusoidal
anastomosis that in part decompresses the sinusoids during
measurement of wedged pressure. [3] In patients with alcoholic
liver disease, the WHVP is usually equal to the PVP, suggesting
that the site of the increase in resistance in alcoholic cirrhosis
includes the entire sinusoid and that fewer decompressive
intersinusoidal anastomoses occur in this disease, possible
reflecting a greater component of intrasinusoidal fibrosis. [20]
The pathogenesis of the increased sinusoidal resistance in
alcoholic cirrhosis also is not fully understood. Concepts
emphasizing the pathogenic role of the architectural
derangement produced by the development of fibrotic septa and
regenerative nodules have to some extent been superseded by
an emphasis on pathologic changes in the sinusoids; these
include hepatocyte enlargement, resulting from an
alcohol-induced accumulation of fat and protein that
compresses the liver sinusoids and obstructs flow, and collagen
deposition in the space of Disse (see Fig. 77-2) . [20] From the
standpoint of the location of maximal resistance in the
pathogenesis of portal hypertension, alcoholic cirrhosis is by no
means a homogeneous entity. The site of predominant
resistance may vary according to the stage, activity, and
predominant pathologic morphology. Thus, hepatocyte swelling
may be an important contributor to the partially reversible
sinusoidal portal hypertension in acute alcoholic hepatitis. [16]
On the other hand, the lesions of perivenular fibrosis and central
hyaline sclerosis that characterize some cases of alcoholic liver
disease could account for a substantial component of
postsinusoidal resistance. [3] In addition, thrombus formation in
the portal and hepatic veins appears to be common in advanced
cirrhosis [21] and would be expected to further contribute to
presinusoidal and postsinusoidal resistance, respectively.
The morphologic derangements occurring in chronic liver

disease are undoubtedly the most important factors in the
increased intrahepatic resistance. However, there may be a role
for dynamic, contractile factors that can lead to increased
vascular tone. Present in fibrous scars and perisinusoidal areas
of cirrhotic but not normal livers are contractile cells called
myofibroblasts. [22] These cells appear to develop from activated
stellate cells (lipocytes) and exhibit a contractile response to
vasoconstrictors such as endothelin, [23] [24] [25] levels of which
are increased in the blood and liver tissue of patients with
cirrhosis. [26] [27] [28] [29] The emergence of myofibroblasts in the
cirrhotic liver may account for the observation that infusion of
vasodilators into a normal liver is not associated with a change
in vascular tone, whereas in the cirrhotic liver, a similar
infusion causes a decrease in pressure. [23] Therefore, in
cirrhosis portal pressure may change in an active manner
depending on intrahepatic contractile elements and the action of
vasoactive compounds in the blood.
With the development of portosystemic collaterals during
1288
the evolution of portal hypertension, the overall resistance

encountered by total portal venous flow (RP ) is determined by
the parallel resistances offered by the portohepatic vascular
resistance (RH ) and the portocollateral resistance (RC )
according to the following relationship:
1/ RP = 1/ RH + 1/ RC
The portosystemic collateral vessels have a substantial amount
of smooth muscle and thus may show active changes in
diameter promoted by vasoactive substances. Therefore, RC is
subject to modulation resulting from changes in the diameter of
the collateral vessels, perhaps to an even greater extent than RH
, rendering this source of resistance yet another dynamic and
pharmacologically amenable site of resistance (this is discussed
in greater detail later).
Portal Blood Flow
The contribution of increased portal venous blood flow to the

pathogenesis of portal hypertension is supported by several
distinct clinical and experimental observations.
Primary High Portal Flow States.

Although uncommon, conditions leading to very high flow

states in the portal system (e.g., arterioportal vein fistulas,
splenomegaly resulting from myelofibrosis/myeloid metaplasia)
are well recognized as causes of portal hypertension. In fact, the
development of portal hypertension in these patients invariably
reflects the combined effect of increased flow as well as a
component of increased resistance in the liver. [3] [11] [16] For
example, idiopathic portal hypertension, an entity that is
common in Asia but rare in the United States, was long believed
to be a primary affliction of the spleen (accounting for the older
term tropical splenomegaly) with a marked increase in splenic
arterial and therefore splenoportal venous blood flow. It is,
however, a disease of the preterminal branches of the portal
vein (hepatoportal sclerosis), [18] and portal hypertension in
idiopathic portal hypertension is not abrogated by splenectomy.
[30] Nevertheless, the importance of the contribution from flow
in certain instances is amply illustrated by the fact that clinical

evidence of portal hypertension, including ascites and
esophageal varices, may regress dramatically after the
normalization of portal flow that follows the closure of an
arterioportal fistula. [31]
The Portosystemic Collateral Paradox.
When portal pressure reaches a critical value, portosystemic

collaterals begin to develop. In alcoholic cirrhosis, a portal
pressure gradient of 10 to 12 mm Hg appears to be necessary
for the development of esophageal varices. [3] [6] Collateral veins
are believed to develop from dilatation of embryonic channels
or redirection of flow within existing veins rather than from the
formation of new blood vessels. As collaterals form, it seems
likely that they tend to decompress the portal system and lower
portal pressure. Paradoxically, the extent of collaterals often
correlates with the degree of portal pressure, [11] and therefore
either a compensatory increase in portal inflow must occur as
collaterals form, or the resistance within the collateral bed must
be unusually high. The latter does not apply because the
vascular resistance of the collateral bed, although higher than
normal portohepatic resistance, is still lower than that of the
obstructed portal system. [6] [11] [32] Therefore, portal
hypertension is maintained during collateral formation by
increased portal inflow, and as a consequence, portal
hypertension persists even in the extreme situation in which all
portal flow escapes through collaterals.
In portal hypertension, therefore, portal venous inflow is

composed of portal blood flow through the liver plus the

volume of splanchnic blood flow that bypasses the liver through
portosystemic collaterals. In other words, portal venous inflow
is equal to the flow of portal blood through liver plus the
portocollateral blood flow. Under normal conditions, the flow
of portal blood through the liver essentially equals all of the
blood entering the splanchnic system (portal venous inflow). In
portal hypertension, perfusion of the liver by portal blood is
decreased and may be negligible. In almost 10% of patients,
flow within the portal vein may even be reversed (retrograde or
hepatofugal portal flow). [33] [34] This situation develops when
hepatic arterial blood flow encounters greater resistance to flow
in its usual anterograde course through the sinusoids than via
the path offered by the portal venous radicals back to the portal
venous circulation. This loss of hepatic arterial blood flow
(hepatic arterial steal) that occurs via collaterals is associated
with a high risk of impaired hepatic function and hepatic
encephalopathy. [33] [34] This principle (which is discussed later)
also applies to therapeutic side-to-side portosystemic shunts. [14]
Although increased portal venous inflow has been well
characterized in animal models of portal hypertension, [3] [4] [6]
[32] [35] and, albeit to a lesser extent, in patients with portal
hypertension, [3] [4] its precise contribution to the maintenance of

elevated portal pressure in patients with portal hypertension has
been controversial. In a rat portal vein stenosis model of portal
hypertension, 15% to 40% of the increase in portal pressure was
estimated to be contributed by increased portal inflow. [6] [32] In
contrast, in patients with cirrhosis, a considerably lower
estimate of 2% of increased portal pressure was calculated to
result from increased portal venous inflow. [6] [11] Complicating
the estimation of the contribution made by flow to overall portal
pressure in portal hypertension is the fact that as flow is
reduced, collateral resistance tends to increase. [6] [36] Also,
reduction in flow from one part of the splanchnic circulation is
rapidly compensated for by an increase in flow from other parts.
For example, patients with portal hypertension in whom splenic
arterial flow was temporarily interrupted by balloon occlusion
showed a fall in portal flow and pressure. These reductions,
however, were less than expected from the loss of splenic blood
flow because of a marked compensatory increase in mesenteric
venous blood flow. [37] Therefore, available evidence points to
increased resistance as the major and driving force in cirrhotic
portal hypertension and supports the role of increased
splanchnic arterial flow in contributing to the maintenance of
elevated portal pressure. [11]

Hyperdynamic Circulation of Portal Hypertension.
The increase in splanchnic blood flow in portal hypertension

occurs as a result of a more-generalized hyperdynamic
circulatory disturbance. [38] [39] [40] This association between
portal hypertension and a hyperdynamic circulatory state was
first described by Kowalski and Abelmann in 1953, [41] and its
hallmarks are increased cardiac output with reduced arterial
blood pressure. Increased cardiac output results from an
increase in heart rate and possibly increased stroke volume and
increased total blood volume; decreased blood pressure results
from a reduction in systemic vascular resistance secondary to
peripheral
1289
arterial vasodilation. The severity of the hyperkinetic

circulatory abnormalities that accompany cirrhosis correlates
with clinical indices of hepatic dysfunction, [42] although the
same abnormalities are present in patients with noncirrhotic
portal hypertension. [38] These latter patients, nevertheless, have
an extensive portosystemic collateral circulation, suggesting
that portosystemic shunting rather than decompensation of
hepatic function per se plays the key role in the development of
the hyperdynamic circulatory state. The systemic consequences
of the hyperdynamic circulation in end-stage liver disease are
complex and represent a form of multisystem organ
dysfunction. [40] Significant effects on the circulation of the
kidneys, brain, and lungs have been described. In the case of the
lungs, pulmonary vasodilation leads to arterial hypoxemia; this
is observed in approximately one third of patients with cirrhosis
in the absence of detectable cardiorespiratory diseases--the
hepatopulmonary syndrome (see Chapter 79) . In this syndrome,
reduced vascular tone of the pulmonary circulation is believed
to result in a perfusion/diffusion mismatch, which is responsible
for the arterial hypoxemia. [43]
The mechanism or mechanisms of the systemic hemodynamic
changes observed in portal hypertension are explained via two
opposing although not mutually exclusive theories (Fig. 77-3) .
The essential difference between these two theories is in the
view of the primary hemodynamic event as a reduction in
vascular resistance with a compensatory increase in cardiac
output or as an increase in cardiac output with a compensatory
reduction in systemic vascular resistance.
The peripheral vasodilation theory, [44] which is a refinement

Figure 77-3 Pathogenesis of the hyperdynamic

circulation in portal hypertension. The schematic
illustrates the pathogenesis of the increased cardiac
output according to the two major theories: primary
peripheral vasodilation ("underfill"; solid lines) and
primary sodium and water retention ("overfill"; broken
lines). (Adapted from Ready, J., and Rector, W. G.
Systemic hemodynamic changes in portal hypertension.
Semin. Gastrointest. Dis. 6:134, 1995.)
of the classic underfill hypothesis formulated to account for the
development of ascites in portal hypertension (see Chapter 78) ,
proposes that a factor or factors that are poorly defined but are
associated with cirrhosis or portosystemic shunting cause
arterial vasodilation, primarily in the splanchnic circulation.
This vasodilation then results in a decrease in the central, or
effective, circulating volume. Sodium is retained by the kidneys
in an effort to defend this volume. According to this theory,
total blood volume would be expected to be increased, which is
a well-recognized finding in patients with cirrhosis. [45] A
necessary consequence of the peripheral vasodilation would
then be an increase in cardiac output as a result of afterload
reduction, producing a hyperdynamic circulatory state.
An alternative proposal, which is closely allied to the classic
overflow theory of sodium retention in cirrhosis, rejects arterial
vasodilation as the primary event in explaining the
hyperdynamic circulation of cirrhosis. [38] [46] The overflow
theory proposes that there is a primary stimulus (hepatorenal
reflex) to sodium and water retention that is a direct
consequence of the presence of portal hypertension. [40] This
alternative view posits that as a result of this primary increase in
salt and water retention in portal hypertension, blood volume
increases and causes an increase in cardiac output. According to
this theory, peripheral vasodilation occurs as an adaptation to
these earlier events.
Neither theory completely accounts for all of the systemic
hemodynamic changes described in patients with portal
hypertension. [40] Rather, there appears to be a continuum of
changes that progress as the liver disease progresses and portal
hypertension supervenes (see Fig. 77-3) . Most important are
the data from hemodynamic studies in patients with portal
hypertension who were treated with selective and nonselective
beta-blockers that point to a role for both increased cardiac
output (beta1 receptor mediated) and splanchnic arteriolar
vasodilation (beta2 receptor mediated) in generating the increase
in portal venous inflow that contributes to the maintenance of
portal hypertension. [13]
Vasoactive Mediators in the Pathogenesis of Portal

Hypertension
An important role for vasoactive factors in the pathogenesis of

portal hypertension has emerged. Both vasodilator and
vasoconstrictor substances have been implicated, acting by
either mediating systemic and splanchnic vasodilation and
therefore portocollateral blood flow or promoting an increase in
vascular resistance within the intrahepatic and portocollateral
beds (Table 77-2) .
Vasoactive Mediators and Splanchnic Vasodilation
There is abundant evidence for increased sympathetic nervous

system tone in patients with cirrhosis, and serum
norepinephrine levels are increased. However, considerable data
indicate the attenuation of sympathetic neurotransmitter effects
in portal hypertension, in part because of down-regulation of
adrenergic receptor density and in part because of postreceptor
antagonism by opposing vasodilator influences. [10] [24] [47]
Earlier studies of cross-perfusion between portal hypertensive
and normal animals produced arteriolar vasodilation in
1290
TABLE 77-2 -- Vasoactive Mediators in Portal Hypertension

VASODILATORS VASOCONSTRICTORS
Glucagon Norepinephrine
Prostacyclin Serotonin
Substance P Endothelins
Adenosine Angiotensin II
Atrial natriuretic factor Vasopressin
Bile acids
Histamine
Vasoactive intestinal peptide
gamma-Aminobutyric acid
Leu- and Met-encephalins

Endotoxin
Tumor necrosis factor-alpha
Nitric oxide
the latter, lending support to the hypothesis that a transferable

humoral vasodilator is present in the blood in portal
hypertension. [6] Much attention has been focused on putative
vasoactive mediators responsible for the arteriolar vasodilation
in splanchnic organs that underlies the increase in portal venous
inflow. [24] [48] Investigators have postulated that endogenous
vasodilators normally present in portal blood and cleared by the
liver may escape hepatic removal as a result of either
portosystemic shunting via portosystemic collaterals or
impaired hepatocellular metabolism. A further possibility is that
liver disease and portal hypertension lead to an increase in the
production of certain vasodilators within the hepatic or
splanchnic vascular beds. These vasodilators then reach high
concentrations in the systemic circulation, leading to systemic
and splanchnic arterial vasodilation.
Several gut peptide hormones have been proposed as
vasodilator mediators in portal hypertension; glucagon has been
a prime candidate. [6] Glucagon levels are increased in
experimental models of portal hypertension and in patients with
cirrhosis. [6] [49] [50] Glucagon also impairs systemic vascular
sensitivity to norepinephrine. [50] A role for glucagon in portal
hypertension is supported by the finding of a significant
reduction in splanchnic blood flow after infusion of a
glucagon-specific antiserum. [51] However, this was not
accompanied by reduction in systemic vasodilation. In addition,
others have failed to find a correlation between the magnitude
of arterial vasodilation and circulating levels of glucagon. [52]
On the other hand, infusion of pharmacologic doses of
somatostatin or its synthetic analog octreotide, which decreases
glucagon release, produces vasoconstriction of both the
splanchnic and systemic circuits. [53] [54] Because somatostatin
also inhibits the release of several other peptide vasodilators
such as substance P, vasoactive intestinal peptide, and
calcitonin gene-related peptide, it is conceivable that its effects
on the circulation in portal hypertension may be mediated by
other peptides in addition to or apart from glucagon. Also,
somatostatin may exert a direct vasoconstrictive effect on
vascular smooth muscle. [10] Therefore, the significance of
glucagon as a mediator of the systemic vasodilation remains
inconclusive, but on the basis of available data,
hyperglucagonism may account for approximately 30% to 40%

of the splanchnic vasodilation of chronic portal hypertension.
[24] [48]
Vasoactive factors produced by the vascular endothelium have

attracted considerable attention with respect to a potential role
in the pathogenesis of portal hypertension. There is increasing
evidence implicating certain of these agents, most notably,
nitric oxide (NO) and prostacyclin, in the pathogenesis of the
circulatory abnormalities in portal hypertension.
NO is a powerful endogenous vasodilator generated in several
tissues by a constitutive vascular endothelial NO synthase
(eNOS) and an inducible NO synthase (iNOS) from the amino
acid L-arginine. [55] NO is produced constituitively by eNOS and
by liver parenchymal and nonparenchymal cells after induction
of iNOS by cytokines and endotoxin. [56] An increasing body of
evidence suggests that excessive NO biosynthesis may be
involved in the pathogenesis of the low systemic and splanchnic
vascular resistance associated with portal hypertension. [57] [58]
Tumor necrosis factor-alpha (TNF-alpha) mediates the effects
of endotoxin, a potent stimulant of inducible NO synthesis.
Studies have reported a dramatic amelioration in the
hyperdynamic circulation and the increased portal pressure in
portal hypertensive rats treated with antibody to TNF-alpha. [59]
The administration of specific NO antagonists to animals with
portal hypertension induces splanchnic and systemic
vasoconstriction, attenuating the hyperdynamic circulation. [57]
[58] In addition, inhibition of NO synthesis, at least partially,
corrects the blunted vascular responsiveness to vasoconstrictors

present in portal hypertension. [60] The finding that patients with
cirrhosis have increased serum and urinary concentrations of
nitrite and nitrate (end-products of NO oxidation) also supports
a role for NO in the genesis of the circulatory disturbances of
portal hypertension. [61]
However, NO inhibition attenuates but does not normalize the
hyperkinetic syndrome of portal hypertension. [57] [58] Also,
chronic NO inhibition delayed but did not prevent the
development of splanchnic vasodilation in experimental
animals. [62] These and other data suggest that factors in addition
to NO are involved in the vasodilatory phenomena associated
with the hyperdynamic circulation.
Several studies have supported a role for prostaglandins in the
hyperdynamic circulation of portal hypertension. [63] [64]
Prostacyclin has also been found to be increased in the portal

vein of portal hypertensive rats, [63] whereas patients with

cirrhosis have increased systemic and portal levels of
prostacyclin. [63] [64] Portal levels of prostacyclin correlate with
the degree of portal pressure elevation in these patients. The
inhibition of prostaglandin biosynthesis by indomethacin
reduced the hyperdynamic circulation and portal pressure in
patients with cirrhosis and portal hypertension. [24]
A variety of other circulating vasodilators have been evaluated,
including bile acids, histamine, adenosine, and substance P (see
Table 77-2) , without convincing evidence that they contribute
to the systemic hyperdynamic state. [10] [24] [65] [66] [67]
Vasoactive Mediators and Vascular Resistance
Intrahepatic Vascular Resistance.
As discussed, there is evidence that hepatic stellate cell

(lipocyte) contraction may contribute to a dynamic component
of increased intrahepatic resistance in portal hypertension. [21]
[22] Vasoactive mediators, both vasoconstrictors and
vasodilators, may modulate intrahepatic vascular resistance

through contraction and/or relaxation of these cells. [68] [69]
1291
The endothelins (ETs), a family of at least three homologous

21-amino acid peptides (ET-1, ET-2, and ET-3), [70] are active
contractile agonists of stellate cells. [25] [68] ET-1 release from
vascular endothelial cells is stimulated by epinephrine and
angiotensin II. [70] ET-1, in turn, increases the release of these
same pressor factors, which with vasopressin are also able to
increase intrahepatic vascular resistance. [71] Experimental data
have shown that infusion of ET-1 increases portal pressure. [72]
ET has also been shown to promote the closing of endothelial
fenestrae in normal rat liver [73] and thus may contribute to the
process of sinusoidal capillarization and increased resistance in
portal hypertension. Plasma and hepatic levels of ET are
increased in patients with cirrhosis, particularly in those with
ascites. [26] [27] [28] [29] The mechanisms responsible for increased
ET production in cirrhosis are not known, but stimulation of ET
production by transforming growth factor-beta, tumor necrosis
factor, or fluid-mechanical stress has been suggested. [27]
Although mainly studied with respect to its probable role in
arterial vasodilation in portal hypertension, NO has also been
implicated in the modulation of dynamic resistance elements

and may play a role in the regulation of the intrahepatic portal

circulation in portal hypertension with a relative deficit of NO
in the intrahepatic circulation in cirrhosis contributing to
increased intrahepatic vascular resistance. [57]
Portocollateral Resistance.
In advanced portal hypertension, the collateral circulation may

carry more than 90% of the blood entering the portal system. [48]
Under these circumstances, it is obvious that the vascular
resistance of collateral vessels may markedly influence the
overall resistance to portal blood flow and, therefore, portal
pressure. [24] The factors that modulate collateral resistance are
not well characterized; studies performed in isolated, perfused
portosystemic collateral bed suggest that NO may be important
in the control of portocollateral vascular resistance. [24]
An important vasoconstrictor role for serotonin mediated by
5-hydroxytryptamine2 receptors in the splanchnic venous
circulation has also been demonstrated. [10] [24] In portal
hypertensive animals and patients, the administration of
selective 5-hydroxytryptamine2 receptor blockers ketanserin
and ritanserin caused a significant decrease in portal pressure
without modifying systemic hemodynamics or portal inflow,
suggesting that a dynamic element of portocollateral resistance
is responsible in part for the elevated portal pressure. [24]


Saunders Company
1284
Chapter 77 - Portal Hypertension and

Nathan M. Bass
Kenneth A. Somberg
ETIOLOGY AND PATHOPHYSIOLOGY OF PORTAL HYPERTENSION, 1284

Normal Liver Blood Flow, 1284
Hemodynamic Alterations in Portal Hypertension, 1285
Vasoactive Mediators in the Pathogenesis of Portal Hypertension, 1289
ANATOMIC SITES OF COLLATERAL FORMATION AND BLEEDING, 1291
Varices, 1291
Nonvariceal Mucosal Lesions, 1292
NATURAL HISTORY AND PROGNOSIS OF VARICEAL HEMORRHAGE, 1292
Determinants of Variceal Bleeding, 1292
Natural History of Variceal Bleeding, 1294
Prognosis, 1294
DIAGNOSIS, 1294
Endoscopy, 1294
Imaging Studies, 1295
Pressure Measurement, 1295
MANAGEMENT, 1295
Principles and Goals of Management, 1295
Resuscitation, 1295
Endoscopic Therapy, 1295
Pharmacologic Therapy, 1298
Balloon Tamponade, 1300
Surgical Therapy, 1301
Transjugular Intrahepatic Portosystemic Shunt, 1302
Management of Other Sources of Hemorrhage Resulting from Portal Hypertension, 1304

Management Decisions and Liver Transplantation, 1305

Summary of Treatment Strategies, 1305
The fact that gastrointestinal bleeding could occur as a consequence of derangement of the portal
circulation was appreciated by the seventeenth century. The concept that esophageal varices develop as a
result of obstruction to portal flow in liver cirrhosis was well developed by the end of the nineteenth
century, and the term portal hypertension was introduced by Gilbert and Carnot in 1902. [2] Whipple's
dour comment in 1945, at a time of considerable innovation in the surgical therapy of portal
hypertension, remains applicable today. Although considerable advances have been made, particularly
since the 1960s, the complications of portal hypertension--gastrointestinal hemorrhage, ascites, and
portosystemic encephalopathy--continue to pose difficult challenges in the management of patients with
end-stage liver disease and remain the cause of significant morbidity and mortality. Liver transplantation
provides a highly successful cure for end-stage liver disease, but there remains the need to manage portal
hypertension in patients awaiting transplantation or those who are not considered to be transplantation
candidates. Despite the difficult nature of the problems presented by portal hypertension, substantial
progress has been achieved, particularly in the understanding of its pathophysiology. [3] [4] [5] [6] [7] [8] [9] [10]
[11] [12] The translation of advancements from this field of research into therapeutic applications has been
rapid and has fostered a growth of pharmacologic therapies. [13] Other important therapeutic options in
widespread use include endoscopic injection sclerotherapy (EST), endoscopic variceal ligation (EVL), a
variety of surgical shunt and nonshunt procedures, and the transjugular intrahepatic portosystemic shunt
(TIPS). [14] Many questions remain regarding the optimal use and timing of these therapies, whereas
continued progress in the understanding of the natural history, pathophysiology, and hemodynamic
derangements that attend portal hypertension holds promise for the development of treatments and
management strategies that will be most effective while minimizing iatrogenic morbidity and cost.


Saunders Company
ETIOLOGY AND PATHOPHYSIOLOGY OF PORTAL HYPERTENSION

Normal Liver Blood Flow
The normal anatomy of the portal and hepatic arterial circulation is described in Chapter 62 (see Fig.
62-3) . Normal hepatic blood flow is approximately 1500 mL/min, which represents 15% to 20% of
cardiac output. One third of this flow and 30% to 60% of the oxygen consumed by the liver are provided
by the hepatic artery. Approximately two thirds of the hepatic blood supply is provided by portal venous
blood. [7] [11] The high-pressure, well-oxygenated arterial blood mixes completely with the low-pressure,
low-oxygenated, nutrient-rich portal venous blood within the hepatic sinusoids. After perfusion into the
sinusoids, blood flows sequentially into the hepatic venules, hepatic veins, and inferior vena cava. A
fraction of the plasma entering the space of Disse is drained into lymphatic vessels.
A unique feature of the normal hepatic sinusoidal microcirculation is its low perfusion pressure. This is
attributed to the unusually high precapillary-to-postcapillary resistance in the liver. [11] It appears that the
sinusoids are normally protected from upstream portal perfusion pressure and accompanying fluctuations
by a presinusoidal site of high resistance, which is probably located within the terminal portal venous
radicals. [15] Because the sinusoids are lined by an endothelium that lacks a continuous basement
membrane and contains a multitude of
1285
large (50 to 200 nm), highly permeable fenestrae, maintenance of a low pressure in the hepatic sinusoids
is critical to the maintenance of normal rates of sinusoidal fluid transudation into the space of Disse (see
Chapter 78) .
Another feature that is unique to the hepatic circulation is the close interrelationship between blood flow
in the portal vein and that in the hepatic artery. When portal blood flow increases, hepatic arterial flow
decreases; when portal flow decreases, hepatic arterial flow increases. This phenomenon has been termed
the hepatic arterial buffer response and is an adenosine-mediated vascular reflex that ensures the
maintenance of a relatively constant state of sinusoidal perfusion in the face of changes in portal inflow
that occur, for example, with meals. [7] [11]
Hemodynamic Alterations in Portal Hypertension
General Principles
The pathogenesis of portal hypertension involves the relationship between portal venous blood flow and
the resistance to this blood flow within the liver (portohepatic resistance) and within portosystemic
collateral blood vessels (portocollateral resistance) that forms during the evolution of portal
hypertension (Fig. 77-1) .

The movement of blood within a vascular system such as the portal system is driven by a pressure
difference or gradient existing along the length of that system. The portal pressure gradient (Delta P; i.e.,
the difference in pressure between the portal and systemic venous systems) is the product of portal
venous blood flow (Q) and the vascular resistance to this flow (R), as expressed by Ohm's law:
Delta P = Q R
In turn, R is derived by Poiseuille's law from the following relationship:
R = 8eta l/pi r4
where eta is the coefficient of viscosity, l is the length of the vessel, and r is the radius. It is clear from
this relationship that small changes in vessel radius will lead to disproportionate increases in resistance
and, thus, pressure. For example, the incremental influence on resistance of increasing hematocrit, and
therefore viscosity, through a blood transfusion administered to a patient who has recently bled from
varices can also be appreciated. [11] Normal, uncorrected pressure in the portal vein ranges between 5 and
10 mm Hg and is influenced by intra-abdominal pressure. Hepatic venous pressure is similarly affected
by intra-abdominal pressure and also reflects central venous filling pressure. To eliminate the
contribution of intra-abdominal pressure and central venous pressures and thus express portal pressure as
the intrinsic pressure difference between the portal and systemic venous compartments, it has become
standard practice to express portal pressure as a portal pressure gradient. In practical terms, this is most
often determined in patients with cirrhosis as the hepatic venous pressure gradient (HVPG), a technique
that is similar in principle to measurement of pulmonary capillary wedge pressure (Fig 77-2) . [3] [4] In
brief, a pressure measurement is made with the use of a catheter wedged into a hepatic vein via either a
femoral or transjugular approach. This measurement is termed the wedged hepatic venous pressure
(WHVP). After withdrawal of the catheter into the hepatic vein, a free hepatic vein pressure (FHVP) is
obtained. The HVPG is obtained by subtracting the value of the WHVP from that of the FHVP.
Normally, the HVPG does not exceed 5 mm Hg. Portal hypertension exists if the value for the HVPG
exceeds this value.
It is apparent that a rise in portal pressure could, in theory, result from an increase in either portal flow or
vascular resistance. Historically, increased flow into the portal system from a massively enlarged spleen
was believed to be the cause of elevated portal pressure in patients with cirrhosis (Banti's hypothesis).
This view was subsequently abandoned when the crucial role of increased resistance to portal flow at
various sites became evident. Since the late 1970s, the pendulum has swung again, thanks to a plethora of
elegant experimental and clinical studies that have demonstrated the presence of a hyperdynamic
circulatory state in portal hypertension with increased splanchnic blood flow that contributes to the
increase
Figure 77-1 Hemodynamic principles in portal hypertension. The schematic summarizes the
contribution and interplay of the major hemodynamic forces underlying the pathogenesis of portal
hypertension. (See also Fig. 77-2 and Table 77-3.)
1286

Figure 77-2 Sites of obstruction (resistance) to portal venous flow and measurement of portal pressure.
The schematic illustrates the major locations of extrahepatic (prehepatic and posthepatic) and intrahepatic
(presinusoidal, sinusoidal, and postsinusoidal) obstruction. A catheter tip is also shown wedged into a
small hepatic vein for the measurement of the WHVP. When the catheter is withdrawn into the hepatic
vein, the FHVP is obtained. HVPG = WHVP - FHVP. Direct measurement of the PVP is accomplished
intraoperatively either by catheterization of the umbilical vein or by catheterization of the portal vein via
the transjugular or transhepatic approach. HVPG, hepatic venous pressure gradient; WHVP, wedged
hepatic venous pressure; FHVP, free hepatic vein pressure; PVP, portal venous pressure.
in portal pressure. The relative extents to which increased splanchnic portal inflow (the "forward" force)
and resistance (the "backward" force) are responsible for the increase in portal pressure in any given
clinical situation remain a much-debated issue and are of considerable practical importance in the
understanding of the fundamental pathophysiology of portal hypertension as well as in the development
of rational pharmacologic therapy.
Role of Increased Resistance
Cirrhosis, usually caused by alcohol ingestion or chronic viral hepatitis, is the most significant cause of
portal hypertension in Western populations, but there are many other causes, most of which are
noncirrhotic; schistosomiasis is a particularly prevalent cause in developing countries. Knowledge of the
pathology of these conditions has confirmed the existence of resistance to portal flow at a variety of
different anatomic levels in the pathogenesis of portal hypertension. This is reflected in the conventional
classification of the causes of portal hypertension according to the localization of the site of maximal
resistance to portal flow (see Fig. 77-2 and Table 77-1) . Regarding the three major categories of
prehepatic, intrahepatic, and posthepatic, the site of increased resistance is usually obvious. Thus, portal
vein thrombosis exemplifies prehepatic portal hypertension, whereas inferior vena caval web typifies
posthepatic portal hypertension. For intrahepatic causes, the site of resistance is conventionally further
subdivided into presinusoidal, sinusoidal, and postsinusoidal. Precise classification of portal hypertension
according to the intrahepatic site of maximal resistance has been limited by the lack of technical means
to measure pressure directly within the hepatic sinusoids. Most of the information regarding this has been
provided from the combined results of direct measurement of pressure in the portal system and indirect
estimation
1287
TABLE 77-1 -- Causes of Portal Hypertension

Primary Increased Flow
Arterioportal venous fistula
Intrahepatic
Intrasplenic
Splanchnic
Splenic capillary hemangiomatosis

Primary Increased Resistance

Prehepatic
Thrombosis/cavernous transformation of the portal vein
Splenic vein thrombosis
Intrahepatic
Presinusoidal
Schistosomiasis *
Sarcoidosis *
Myeloproliferative diseases/myelofibrosis *
Congenital hepatic fibrosis
Idiopathic portal hypertension (hepatoportal sclerosis)
Chronic arsenic hepatotoxicity
Azathioprine hepatotoxicity
Vinyl chloride hepatotoxicity
Early primary biliary cirrhosis *
Early primary sclerosing cholangitis *
Sinusoidal/mixed
Cirrhosis secondary to chronic hepatitis
Alcoholic cirrhosis
Cryptogenic cirrhosis
Methotrexate
Alcoholic hepatitis
Hypervitaminosis A
Incomplete septal fibrosis
Postsinusoidal
Hepatic vein thrombosis (Budd-Chiari syndrome)
Posthepatic
Inferior vena caval web

Constrictive pericarditis
Tricuspid insufficiency
Severe right heart failure
*Usually presinusoidal when early; often progressing to a sinusoidal/mixed type of resistance whenmore advanced.
of the intrasinusoidal pressure from the WHVP, with details of the morbid anatomy taken into
consideration. [3] [11] For example, in both prehepatic and intrahepatic presinusoidal portal hypertension,
the directly measured portal venous pressure (PVP) is always elevated in the presence of normal WHVP
and HVPG. In sinusoidal and intrahepatic postsinusoidal portal hypertension, the WHVP tends to
approximate or equal the directly measured portal pressure, and the HVPG is proportionately increased.
In posthepatic portal hypertension, the WHVP equals the increased PVP. Because the FHVP is also
abnormally high in this scenario, the HVPG is usually normal.
As a rule, the intrahepatic localization of maximal resistance tends to be more clearly established in the
noncirrhotic diseases than in cirrhosis. [16] Thus, veno-occlusive disease, resulting from either
pyrrolizidine alkaloids or chemotherapeutic agents (see Chapter 73) , is readily apparent as a cause of
postsinusoidal, intrahepatic portal hypertension, whereas early hepatic schistosomiasis with its
characteristic granulomatous portal inflammation and fibrosis is a well-studied example of presinusoidal
portal hypertension. However, even such widely accepted examples of presinusoidal portal hypertension
as schistosomiasis [16] [17] and idiopathic portal hypertension [18] become more complicated when one
considers that over time, hepatic pathology may progress and generate a more complex or mixed pattern
of vascular resistance.
The main site of portal vascular resistance in cirrhosis has been difficult to establish. Most measurements
in nonalcoholic cirrhosis indicate a higher value for PVP than for WHVP. Given that the WHVP is an
estimate of the intrasinusoidal pressure, this has been variously interpreted to indicate either the presence
of a presinusoidal component of resistance, probably related to inflammatory activity or fibrotic changes
in the portal tracts, [19] or a significant presence of intersinusoidal anastomosis that in part decompresses
the sinusoids during measurement of wedged pressure. [3] In patients with alcoholic liver disease, the
WHVP is usually equal to the PVP, suggesting that the site of the increase in resistance in alcoholic
cirrhosis includes the entire sinusoid and that fewer decompressive intersinusoidal anastomoses occur in
this disease, possible reflecting a greater component of intrasinusoidal fibrosis. [20]
The pathogenesis of the increased sinusoidal resistance in alcoholic cirrhosis also is not fully understood.
Concepts emphasizing the pathogenic role of the architectural derangement produced by the development
of fibrotic septa and regenerative nodules have to some extent been superseded by an emphasis on
pathologic changes in the sinusoids; these include hepatocyte enlargement, resulting from an
alcohol-induced accumulation of fat and protein that compresses the liver sinusoids and obstructs flow,
and collagen deposition in the space of Disse (see Fig. 77-2) . [20] From the standpoint of the location of
maximal resistance in the pathogenesis of portal hypertension, alcoholic cirrhosis is by no means a
homogeneous entity. The site of predominant resistance may vary according to the stage, activity, and
predominant pathologic morphology. Thus, hepatocyte swelling may be an important contributor to the

partially reversible sinusoidal portal hypertension in acute alcoholic hepatitis. [16] On the other hand, the
lesions of perivenular fibrosis and central hyaline sclerosis that characterize some cases of alcoholic liver
disease could account for a substantial component of postsinusoidal resistance. [3] In addition, thrombus
formation in the portal and hepatic veins appears to be common in advanced cirrhosis [21] and would be
expected to further contribute to presinusoidal and postsinusoidal resistance, respectively.
The morphologic derangements occurring in chronic liver disease are undoubtedly the most important
factors in the increased intrahepatic resistance. However, there may be a role for dynamic, contractile
factors that can lead to increased vascular tone. Present in fibrous scars and perisinusoidal areas of
cirrhotic but not normal livers are contractile cells called myofibroblasts. [22] These cells appear to
develop from activated stellate cells (lipocytes) and exhibit a contractile response to vasoconstrictors
such as endothelin, [23] [24] [25] levels of which are increased in the blood and liver tissue of patients with
cirrhosis. [26] [27] [28] [29] The emergence of myofibroblasts in the cirrhotic liver may account for the
observation that infusion of vasodilators into a normal liver is not associated with a change in vascular
tone, whereas in the cirrhotic liver, a similar infusion causes a decrease in pressure. [23] Therefore, in
cirrhosis portal pressure may change in an active manner depending on intrahepatic contractile elements
and the action of vasoactive compounds in the blood.
With the development of portosystemic collaterals during
1288
the evolution of portal hypertension, the overall resistance encountered by total portal venous flow (RP )
is determined by the parallel resistances offered by the portohepatic vascular resistance (RH ) and the
portocollateral resistance (RC ) according to the following relationship:
1/ RP = 1/ RH + 1/ RC
The portosystemic collateral vessels have a substantial amount of smooth muscle and thus may show
active changes in diameter promoted by vasoactive substances. Therefore, RC is subject to modulation
resulting from changes in the diameter of the collateral vessels, perhaps to an even greater extent than RH
, rendering this source of resistance yet another dynamic and pharmacologically amenable site of
resistance (this is discussed in greater detail later).
Portal Blood Flow
The contribution of increased portal venous blood flow to the pathogenesis of portal hypertension is
supported by several distinct clinical and experimental observations.
Primary High Portal Flow States.
Although uncommon, conditions leading to very high flow states in the portal system (e.g., arterioportal
vein fistulas, splenomegaly resulting from myelofibrosis/myeloid metaplasia) are well recognized as
causes of portal hypertension. In fact, the development of portal hypertension in these patients invariably
reflects the combined effect of increased flow as well as a component of increased resistance in the liver.
[3] [11] [16] For example, idiopathic portal hypertension, an entity that is common in Asia but rare in the
United States, was long believed to be a primary affliction of the spleen (accounting for the older term

tropical splenomegaly) with a marked increase in splenic arterial and therefore splenoportal venous
blood flow. It is, however, a disease of the preterminal branches of the portal vein (hepatoportal
sclerosis), [18] and portal hypertension in idiopathic portal hypertension is not abrogated by splenectomy.
[30] Nevertheless, the importance of the contribution from flow in certain instances is amply illustrated by
the fact that clinical evidence of portal hypertension, including ascites and esophageal varices, may
regress dramatically after the normalization of portal flow that follows the closure of an arterioportal
fistula. [31]
The Portosystemic Collateral Paradox.
When portal pressure reaches a critical value, portosystemic collaterals begin to develop. In alcoholic
cirrhosis, a portal pressure gradient of 10 to 12 mm Hg appears to be necessary for the development of
esophageal varices. [3] [6] Collateral veins are believed to develop from dilatation of embryonic channels
or redirection of flow within existing veins rather than from the formation of new blood vessels. As
collaterals form, it seems likely that they tend to decompress the portal system and lower portal pressure.
Paradoxically, the extent of collaterals often correlates with the degree of portal pressure, [11] and
therefore either a compensatory increase in portal inflow must occur as collaterals form, or the resistance
within the collateral bed must be unusually high. The latter does not apply because the vascular
resistance of the collateral bed, although higher than normal portohepatic resistance, is still lower than
that of the obstructed portal system. [6] [11] [32] Therefore, portal hypertension is maintained during
collateral formation by increased portal inflow, and as a consequence, portal hypertension persists even
in the extreme situation in which all portal flow escapes through collaterals.
In portal hypertension, therefore, portal venous inflow is composed of portal blood flow through the liver
plus the volume of splanchnic blood flow that bypasses the liver through portosystemic collaterals. In
other words, portal venous inflow is equal to the flow of portal blood through liver plus the
portocollateral blood flow. Under normal conditions, the flow of portal blood through the liver
essentially equals all of the blood entering the splanchnic system (portal venous inflow). In portal
hypertension, perfusion of the liver by portal blood is decreased and may be negligible. In almost 10% of
patients, flow within the portal vein may even be reversed (retrograde or hepatofugal portal flow). [33] [34]
This situation develops when hepatic arterial blood flow encounters greater resistance to flow in its usual
anterograde course through the sinusoids than via the path offered by the portal venous radicals back to
the portal venous circulation. This loss of hepatic arterial blood flow (hepatic arterial steal) that occurs
via collaterals is associated with a high risk of impaired hepatic function and hepatic encephalopathy. [33]
[34] This principle (which is discussed later) also applies to therapeutic side-to-side portosystemic shunts.
[14]
Although increased portal venous inflow has been well characterized in animal models of portal
hypertension, [3] [4] [6] [32] [35] and, albeit to a lesser extent, in patients with portal hypertension, [3] [4] its
precise contribution to the maintenance of elevated portal pressure in patients with portal hypertension
has been controversial. In a rat portal vein stenosis model of portal hypertension, 15% to 40% of the
increase in portal pressure was estimated to be contributed by increased portal inflow. [6] [32] In contrast,
in patients with cirrhosis, a considerably lower estimate of 2% of increased portal pressure was
calculated to result from increased portal venous inflow. [6] [11] Complicating the estimation of the
contribution made by flow to overall portal pressure in portal hypertension is the fact that as flow is
reduced, collateral resistance tends to increase. [6] [36] Also, reduction in flow from one part of the
splanchnic circulation is rapidly compensated for by an increase in flow from other parts. For example,

patients with portal hypertension in whom splenic arterial flow was temporarily interrupted by balloon
occlusion showed a fall in portal flow and pressure. These reductions, however, were less than expected
from the loss of splenic blood flow because of a marked compensatory increase in mesenteric venous
blood flow. [37] Therefore, available evidence points to increased resistance as the major and driving force
in cirrhotic portal hypertension and supports the role of increased splanchnic arterial flow in contributing
to the maintenance of elevated portal pressure. [11]
Hyperdynamic Circulation of Portal Hypertension.
The increase in splanchnic blood flow in portal hypertension occurs as a result of a more-generalized
hyperdynamic circulatory disturbance. [38] [39] [40] This association between portal hypertension and a
hyperdynamic circulatory state was first described by Kowalski and Abelmann in 1953, [41] and its
hallmarks are increased cardiac output with reduced arterial blood pressure. Increased cardiac output
results from an increase in heart rate and possibly increased stroke volume and increased total blood
volume; decreased blood pressure results from a reduction in systemic vascular resistance secondary to
peripheral
1289
arterial vasodilation. The severity of the hyperkinetic circulatory abnormalities that accompany cirrhosis
correlates with clinical indices of hepatic dysfunction, [42] although the same abnormalities are present in
patients with noncirrhotic portal hypertension. [38] These latter patients, nevertheless, have an extensive
portosystemic collateral circulation, suggesting that portosystemic shunting rather than decompensation
of hepatic function per se plays the key role in the development of the hyperdynamic circulatory state.
The systemic consequences of the hyperdynamic circulation in end-stage liver disease are complex and
represent a form of multisystem organ dysfunction. [40] Significant effects on the circulation of the
kidneys, brain, and lungs have been described. In the case of the lungs, pulmonary vasodilation leads to
arterial hypoxemia; this is observed in approximately one third of patients with cirrhosis in the absence
of detectable cardiorespiratory diseases--the hepatopulmonary syndrome (see Chapter 79) . In this
syndrome, reduced vascular tone of the pulmonary circulation is believed to result in a
perfusion/diffusion mismatch, which is responsible for the arterial hypoxemia. [43]
The mechanism or mechanisms of the systemic hemodynamic changes observed in portal hypertension
are explained via two opposing although not mutually exclusive theories (Fig. 77-3) . The essential
difference between these two theories is in the view of the primary hemodynamic event as a reduction in
vascular resistance with a compensatory increase in cardiac output or as an increase in cardiac output
with a compensatory reduction in systemic vascular resistance.
The peripheral vasodilation theory, [44] which is a refinement
Figure 77-3 Pathogenesis of the hyperdynamic circulation in portal hypertension. The schematic
illustrates the pathogenesis of the increased cardiac output according to the two major theories: primary
peripheral vasodilation ("underfill"; solid lines) and primary sodium and water retention ("overfill";
broken lines). (Adapted from Ready, J., and Rector, W. G. Systemic hemodynamic changes in portal
hypertension. Semin. Gastrointest. Dis. 6:134, 1995.)

of the classic underfill hypothesis formulated to account for the development of ascites in portal
hypertension (see Chapter 78) , proposes that a factor or factors that are poorly defined but are associated
with cirrhosis or portosystemic shunting cause arterial vasodilation, primarily in the splanchnic
circulation. This vasodilation then results in a decrease in the central, or effective, circulating volume.
Sodium is retained by the kidneys in an effort to defend this volume. According to this theory, total blood
volume would be expected to be increased, which is a well-recognized finding in patients with cirrhosis.
[45] A necessary consequence of the peripheral vasodilation would then be an increase in cardiac output as
a result of afterload reduction, producing a hyperdynamic circulatory state.

An alternative proposal, which is closely allied to the classic overflow theory of sodium retention in
cirrhosis, rejects arterial vasodilation as the primary event in explaining the hyperdynamic circulation of
cirrhosis. [38] [46] The overflow theory proposes that there is a primary stimulus (hepatorenal reflex) to
sodium and water retention that is a direct consequence of the presence of portal hypertension. [40] This
alternative view posits that as a result of this primary increase in salt and water retention in portal
hypertension, blood volume increases and causes an increase in cardiac output. According to this theory,
peripheral vasodilation occurs as an adaptation to these earlier events.
Neither theory completely accounts for all of the systemic hemodynamic changes described in patients
with portal hypertension. [40] Rather, there appears to be a continuum of changes that progress as the liver
disease progresses and portal hypertension supervenes (see Fig. 77-3) . Most important are the data from
hemodynamic studies in patients with portal hypertension who were treated with selective and
nonselective beta-blockers that point to a role for both increased cardiac output (beta1 receptor mediated)
and splanchnic arteriolar vasodilation (beta2 receptor mediated) in generating the increase in portal
venous inflow that contributes to the maintenance of portal hypertension. [13]
Vasoactive Mediators in the Pathogenesis of Portal Hypertension
An important role for vasoactive factors in the pathogenesis of portal hypertension has emerged. Both
vasodilator and vasoconstrictor substances have been implicated, acting by either mediating systemic and
splanchnic vasodilation and therefore portocollateral blood flow or promoting an increase in vascular
resistance within the intrahepatic and portocollateral beds (Table 77-2) .
Vasoactive Mediators and Splanchnic Vasodilation
There is abundant evidence for increased sympathetic nervous system tone in patients with cirrhosis, and
serum norepinephrine levels are increased. However, considerable data indicate the attenuation of
sympathetic neurotransmitter effects in portal hypertension, in part because of down-regulation of
adrenergic receptor density and in part because of postreceptor antagonism by opposing vasodilator
influences. [10] [24] [47]
Earlier studies of cross-perfusion between portal hypertensive and normal animals produced arteriolar
vasodilation in
1290

TABLE 77-2 -- Vasoactive Mediators in Portal Hypertension

VASODILATORS VASOCONSTRICTORS
Glucagon Norepinephrine
Prostacyclin Serotonin
Substance P Endothelins
Adenosine Angiotensin II
Atrial natriuretic factor Vasopressin
Bile acids
Histamine
Vasoactive intestinal peptide
gamma-Aminobutyric acid
Leu- and Met-encephalins
Endotoxin
Tumor necrosis factor-alpha
Nitric oxide
the latter, lending support to the hypothesis that a transferable humoral vasodilator is present in the blood
in portal hypertension. [6] Much attention has been focused on putative vasoactive mediators responsible
for the arteriolar vasodilation in splanchnic organs that underlies the increase in portal venous inflow. [24]
[48] Investigators have postulated that endogenous vasodilators normally present in portal blood and
cleared by the liver may escape hepatic removal as a result of either portosystemic shunting via
portosystemic collaterals or impaired hepatocellular metabolism. A further possibility is that liver disease
and portal hypertension lead to an increase in the production of certain vasodilators within the hepatic or
splanchnic vascular beds. These vasodilators then reach high concentrations in the systemic circulation,
leading to systemic and splanchnic arterial vasodilation.
Several gut peptide hormones have been proposed as vasodilator mediators in portal hypertension;
glucagon has been a prime candidate. [6] Glucagon levels are increased in experimental models of portal
hypertension and in patients with cirrhosis. [6] [49] [50] Glucagon also impairs systemic vascular sensitivity
to norepinephrine. [50] A role for glucagon in portal hypertension is supported by the finding of a
significant reduction in splanchnic blood flow after infusion of a glucagon-specific antiserum. [51]
However, this was not accompanied by reduction in systemic vasodilation. In addition, others have failed
to find a correlation between the magnitude of arterial vasodilation and circulating levels of glucagon. [52]
On the other hand, infusion of pharmacologic doses of somatostatin or its synthetic analog octreotide,
which decreases glucagon release, produces vasoconstriction of both the splanchnic and systemic
circuits. [53] [54] Because somatostatin also inhibits the release of several other peptide vasodilators such as
substance P, vasoactive intestinal peptide, and calcitonin gene-related peptide, it is conceivable that its
effects on the circulation in portal hypertension may be mediated by other peptides in addition to or apart

from glucagon. Also, somatostatin may exert a direct vasoconstrictive effect on vascular smooth muscle.
[10] Therefore, the significance of glucagon as a mediator of the systemic vasodilation remains
inconclusive, but on the basis of available data, hyperglucagonism may account for approximately 30%
to 40% of the splanchnic vasodilation of chronic portal hypertension. [24] [48]
Vasoactive factors produced by the vascular endothelium have attracted considerable attention with
respect to a potential role in the pathogenesis of portal hypertension. There is increasing evidence
implicating certain of these agents, most notably, nitric oxide (NO) and prostacyclin, in the pathogenesis
of the circulatory abnormalities in portal hypertension.
NO is a powerful endogenous vasodilator generated in several tissues by a constitutive vascular
endothelial NO synthase (eNOS) and an inducible NO synthase (iNOS) from the amino acid L-arginine.
[55] NO is produced constituitively by eNOS and by liver parenchymal and nonparenchymal cells after
induction of iNOS by cytokines and endotoxin. [56] An increasing body of evidence suggests that
excessive NO biosynthesis may be involved in the pathogenesis of the low systemic and splanchnic
vascular resistance associated with portal hypertension. [57] [58] Tumor necrosis factor-alpha (TNF-alpha)
mediates the effects of endotoxin, a potent stimulant of inducible NO synthesis. Studies have reported a
dramatic amelioration in the hyperdynamic circulation and the increased portal pressure in portal
hypertensive rats treated with antibody to TNF-alpha. [59]
The administration of specific NO antagonists to animals with portal hypertension induces splanchnic
and systemic vasoconstriction, attenuating the hyperdynamic circulation. [57] [58] In addition, inhibition of
NO synthesis, at least partially, corrects the blunted vascular responsiveness to vasoconstrictors present
in portal hypertension. [60] The finding that patients with cirrhosis have increased serum and urinary
concentrations of nitrite and nitrate (end-products of NO oxidation) also supports a role for NO in the
genesis of the circulatory disturbances of portal hypertension. [61]
However, NO inhibition attenuates but does not normalize the hyperkinetic syndrome of portal
hypertension. [57] [58] Also, chronic NO inhibition delayed but did not prevent the development of
splanchnic vasodilation in experimental animals. [62] These and other data suggest that factors in addition
to NO are involved in the vasodilatory phenomena associated with the hyperdynamic circulation.
Several studies have supported a role for prostaglandins in the hyperdynamic circulation of portal
hypertension. [63] [64] Prostacyclin has also been found to be increased in the portal vein of portal
hypertensive rats, [63] whereas patients with cirrhosis have increased systemic and portal levels of
prostacyclin. [63] [64] Portal levels of prostacyclin correlate with the degree of portal pressure elevation in
these patients. The inhibition of prostaglandin biosynthesis by indomethacin reduced the hyperdynamic
circulation and portal pressure in patients with cirrhosis and portal hypertension. [24]
A variety of other circulating vasodilators have been evaluated, including bile acids, histamine,
adenosine, and substance P (see Table 77-2) , without convincing evidence that they contribute to the
systemic hyperdynamic state. [10] [24] [65] [66] [67]
Vasoactive Mediators and Vascular Resistance
Intrahepatic Vascular Resistance.
As discussed, there is evidence that hepatic stellate cell (lipocyte) contraction may contribute to a

dynamic component of increased intrahepatic resistance in portal hypertension. [21] [22] Vasoactive
mediators, both vasoconstrictors and vasodilators, may modulate intrahepatic vascular resistance through
contraction and/or relaxation of these cells. [68] [69]
1291
The endothelins (ETs), a family of at least three homologous 21-amino acid peptides (ET-1, ET-2, and
ET-3), [70] are active contractile agonists of stellate cells. [25] [68] ET-1 release from vascular endothelial
cells is stimulated by epinephrine and angiotensin II. [70] ET-1, in turn, increases the release of these same
pressor factors, which with vasopressin are also able to increase intrahepatic vascular resistance. [71]
Experimental data have shown that infusion of ET-1 increases portal pressure. [72] ET has also been
shown to promote the closing of endothelial fenestrae in normal rat liver [73] and thus may contribute to
the process of sinusoidal capillarization and increased resistance in portal hypertension. Plasma and
hepatic levels of ET are increased in patients with cirrhosis, particularly in those with ascites. [26] [27] [28]
[29] The mechanisms responsible for increased ET production in cirrhosis are not known, but stimulation
of ET production by transforming growth factor-beta, tumor necrosis factor, or fluid-mechanical stress

has been suggested. [27]
Although mainly studied with respect to its probable role in arterial vasodilation in portal hypertension,
NO has also been implicated in the modulation of dynamic resistance elements and may play a role in the
regulation of the intrahepatic portal circulation in portal hypertension with a relative deficit of NO in the
intrahepatic circulation in cirrhosis contributing to increased intrahepatic vascular resistance. [57]
Portocollateral Resistance.
In advanced portal hypertension, the collateral circulation may carry more than 90% of the blood
entering the portal system. [48] Under these circumstances, it is obvious that the vascular resistance of
collateral vessels may markedly influence the overall resistance to portal blood flow and, therefore,
portal pressure. [24] The factors that modulate collateral resistance are not well characterized; studies
performed in isolated, perfused portosystemic collateral bed suggest that NO may be important in the
control of portocollateral vascular resistance. [24]
An important vasoconstrictor role for serotonin mediated by 5-hydroxytryptamine2 receptors in the
splanchnic venous circulation has also been demonstrated. [10] [24] In portal hypertensive animals and
patients, the administration of selective 5-hydroxytryptamine2 receptor blockers ketanserin and ritanserin
caused a significant decrease in portal pressure without modifying systemic hemodynamics or portal
inflow, suggesting that a dynamic element of portocollateral resistance is responsible in part for the
elevated portal pressure. [24]


April 21, 2000
Conde Petra

Company
Section 9 - Liver
77 - Portal Hypertension and Feldman: Sleisenger & Fordtran's Gastrointestinal and Liver Disease,
NATURAL HISTORY AND

PROGNOSIS OF VARICEAL
HEMORRHAGE NATURAL HISTORY AND PROGNOSIS OF
Determinants of Variceal VARICEAL HEMORRHAGE
Bleeding
Determinants of Variceal Bleeding
Risk of Bleeding from Other
Sites The mechanisms by which varices rupture have not been fully
elucidated. The "corrosion" hypothesis postulated that reflux of
Natural History of Variceal
gastric acid injured the mucosa of the lower part of the
Bleeding
esophagus with subsequent erosion into the submucosal varices.
Prognosis This theory has fallen out of favor because of a lack of
supportive evidence for increased gastroesophageal reflux in
patients with bleeding esophageal varices, of acid-peptic
damage to the lower esophageal mucosa in patients after acute
variceal hemorrhage, or of any benefit of long-term H2 -blocker
therapy in preventing variceal bleeding. [90]
Attention has shifted to the "explosion" theory, in which the key
event is an increase in variceal elastic wall tension to a critical
level at which explosive rupture occurs. [4] [11] [90] [91] Wall
tension (T) in a varix varies as a function of the transmural
pressure (TP), vessel radius (r), and wall thickness (w), as
represented by LaPlace's law:
T = TP r/ w
The wall tension T resists the expanding force TP r/ w, and
when the latter exceeds the former, rupture occurs. Esophageal
varices rupture most often at or near the gastroesophageal
junction; this is the area of the esophagus at which the veins are
most superficial and therefore least surrounded by supportive
tissues. This encourages the progressive widening and wall
thinning of varices at this site, a natural history that must

encourage rupture and account, to some extent, for the distal

esophagus being the most frequent site of bleeding.
The explosion theory places considerable emphasis on variceal
size, wall thickness, and transmural pressure as risk factors for
bleeding (Table 77-3) . This emphasis appears to be justified.
Several groups have confirmed that variceal size is an important
risk factor for variceal bleeding, [4] [11] [92] and several systems
have been proposed for grading varices according to their size
and the presence of "red signs" [93] (Fig. 77-5) . Red signs,
which include cherry-red spots and red wale markings (a
longitudinal, raised, red streak), are usually associated with the
most advanced grade of varices and are
1293
TABLE 77-3 -- Determinants of Variceal Bleeding

Local Factors
Variceal size--vessel radius
Variceal wall thickness--red signs
Hemodynamic Factors
Portal (intravariceal) pressure--12 mm Hg hepatic venous
pressure gradient threshold
Blood volume
Collateral blood flow (?)
Intra-abdominal pressure (?)
Severity of Liver Disease (Child Class)
Other
Salicylates and nonsteroidal anti-inflammatory agents
Continued alcohol abuse (?)
thought to represent focal weaknesses, or "blowouts," in the

variceal wall ("varices-on-varices"). They also clearly carry
additional weight as risk factors for bleeding. [92]
The relationship between portal pressure and the risk of variceal
rupture has been more subtle. Although all patients with
variceal bleeding will have increased portal pressure, there is no

clear correlation between portal pressure and the risk of

bleeding. However, several groups have confirmed that
Figure 77-5 Endoscopic grading of esophageal varices

according to Paqet. [93] Grade I: small varices without luminal
prolapse; grade II: moderate-sized varices showing luminal
prolapse with minimal obscuring of the gastroesophageal
junction; grade III: large varices showing luminal prolapse
substantially obscuring the gastroesophageal junction; and
grade IV: very large varices completely obscuring the
gastroesophageal junction. The presence of red signs ( black
spots) is noted separately. (From Paquet, K. J. Prophylactic endoscopic
sclerosing treatment of the esophageal wall in varices: A prospective
controlled trial. Endoscopy 14:4, 1982.)
there is an important threshold phenomenon with respect to the
portal pressure gradient (HVPG) and the risk of bleeding. [4] [11]
[91] Thus, it is well recognized that varices rarely bleed at an
HVPG of less than 12 mm Hg, but there is no direct correlation

between the risk of bleeding and an HVPG above this threshold
value. There may be a more quantitative relationship between
pressure and the risk of bleeding from varices depending on
where the pressure is measured (i.e., HVPG versus intravariceal
pressure) [94] [95] and when it is measured in relation to an
episode of acute variceal hemorrhage. [96]
Inasmuch as they also contribute to portal and intravariceal
pressure, total blood volume and variceal collateral blood flow
are two factors that may contribute to the risk of variceal
rupture. Variceal blood flow measured as total azygous vein
blood flow has not shown a clear correlation with the risk of
variceal bleeding. On the other hand, reduction in plasma
volume through the use of diuretics in patients with portal
hypertension lowers portal and variceal pressure, [97] [98] whereas
expansion of intravascular volume may be a significant
determinant of variceal hemorrhage and, in particular, of
ongoing acute bleeding. [99] Intra-abdominal pressure clearly
influences several hemodynamic variables in portal
hypertension, but there is no convincing evidence that tense
ascites per se, as opposed to decompensated liver disease,
increases the risk of variceal bleeding. [100] [101]
Patients with decompensated liver disease bleed more often and
have a worse prognosis than do patients with compensated liver
disease. [102] [103] The severity of the underlying liver disease
(Child-Pugh or Child-Turcotte class) correlates with variceal
size [104] but also appears to be an important independent
predictor of variceal hemorrhage and may influence bleeding
risk via several other factors, including poor nutritional state

and worsening coagulopathy. [105] [106] The North Italian

Endoscopic Club [92] studied 321 patients with cirrhosis whose
varices had not yet bled. They calculated an index using three
variables: Child's class, size of varices, and presence of "red
wale" markings. Over a 2-year follow-up period, patients with
the lowest score (mild hepatic dysfunction, small varices, and
no red signs) had a 6% risk of bleeding; in those with the
highest scores (very poor liver function, large varices, and many
red signs), the risk of bleeding increased to 76%.
Spontaneous regression of varices may occur in alcoholic
patients practicing strict abstinence, [107] and continued alcohol
abuse has been long been held to play a role in promoting the
risk of variceal bleeding in patients with alcoholic cirrhosis.
Oddly, this relationship has been surprisingly difficult to prove.
[100] [108] On the other hand, suspicions relating to a role for
aspirin and nonsteroidal anti-inflammatory drugs in the

precipitation of portal hypertensive bleeding have been
substantiated, [109] and patients with portal hypertension are best
advised to avoid these agents.
Risk of Bleeding from Other Sites
Gastric varices located in the fundus, whether isolated or in

continuity with esophageal varices, carry a particularly high risk
of bleeding. [77] [78] In patients with portal hypertensive
gastropathy, bleeding occurs more often with more severe
grades. Mild or chronic bleeding was observed in 35% of
patients with mild gastropathy versus 90% of patients with
severe
1294
gastropathy. Overt bleeding occurred in 30% of those with mild

gastropathy and 60% with severe gastropathy. [110]
Natural History of Variceal Bleeding
Gastroesophageal varices develop in 50% to 60% of cirrhotic

patients, and approximately 30% of them will experience an
episode of variceal hemorrhage within 2 years of the diagnosis
of varices. [91] [103] Variceal bleeding is reported to account for
2% to 20% of all upper gastrointestinal hemorrhage and for a
much higher proportion (50%) in patients with severe,
persistent bleeding. [111] [112]
The greatest risk of initial variceal bleeding occurs within a 6-

to 12-month period after their discovery. Beyond this time, the

risk of bleeding tends to diminish in those who have not already
bled. [11] [113] A diurnal periodicity of variceal bleeding has been
noted by several groups; bleeding episodes tend to favor the
early morning and late evening. [113] After a variceal bleed, the
risk of rebleeding is particularly high-- approximately 60% to
70% over a 24-month period. The risk of rebleeding is greatest,
however, within hours or days after an acute bleed. [102] [112] [113]
Once bleeding has occurred, a number of factors have been
identified that increase the risk of early rebleeding, including
bleeding from gastric varices, thrombocytopenia,
encephalopathy, a diagnosis of alcoholic cirrhosis, large varices,
active bleeding at time of diagnostic endoscopy, and a high
portal pressure gradient. [78] [96] [102] [113]
McCormick and associates [113] suggested that secondary
hemodynamic changes in the splanchnic circulation after a
bleed may contribute to the risk of further bleeding. These
changes include an increase in portocollateral resistance after
hypotension, increased splanchnic blood flow stimulated by
blood in the gut, and an increase in portal venous pressure as a
result of overzealous volume expansion during resuscitation.
The net effect of these changes would be an "overshoot" in
portal pressure and blood flow during resuscitation, which
would encourage continued variceal bleeding.
Prognosis
Variceal hemorrhage is the most serious complication of portal

hypertension and accounts for approximately one fifth to one
third of all deaths of cirrhotic patients. [100] [111] [112] The
mortality after a variceal bleed ranges from 40% to 70% in
various series, with an average of approximately 50% within 6
weeks. [100] As noted by several authors, this high, early
mortality may have a profound effect on the survival outcome
reported by therapeutic studies depending on the interval
between presentation and inclusion in the study. [100] [103] It has
also been the prime motivation for the development of
prophylactic therapy. The most important determinant of
survival is the patient's level of hepatic function. The prognosis
associated with variceal bleeding is generally much better in
patients without significant liver impairment, such as those with
noncirrhotic portal vein thrombosis or idiopathic portal
hypertension. [18] [114] In cirrhotic patients, the prognosis is also
worse in the presence of concomitant alcoholic hepatitis,
hepatocellular carcinoma, or portal vein thrombosis. [115] [116]
[117]



Saunders Company
NATURAL HISTORY AND PROGNOSIS OF VARICEAL

HEMORRHAGE
Determinants of Variceal Bleeding
The mechanisms by which varices rupture have not been fully elucidated. The "corrosion" hypothesis
postulated that reflux of gastric acid injured the mucosa of the lower part of the esophagus with
subsequent erosion into the submucosal varices. This theory has fallen out of favor because of a lack of
supportive evidence for increased gastroesophageal reflux in patients with bleeding esophageal varices,
of acid-peptic damage to the lower esophageal mucosa in patients after acute variceal hemorrhage, or of
any benefit of long-term H2 -blocker therapy in preventing variceal bleeding. [90]
Attention has shifted to the "explosion" theory, in which the key event is an increase in variceal elastic
wall tension to a critical level at which explosive rupture occurs. [4] [11] [90] [91] Wall tension (T) in a varix
varies as a function of the transmural pressure (TP), vessel radius (r), and wall thickness (w), as
represented by LaPlace's law:
T = TP r/ w
The wall tension T resists the expanding force TP r/ w, and when the latter exceeds the former, rupture
occurs. Esophageal varices rupture most often at or near the gastroesophageal junction; this is the area of
the esophagus at which the veins are most superficial and therefore least surrounded by supportive
tissues. This encourages the progressive widening and wall thinning of varices at this site, a natural
history that must encourage rupture and account, to some extent, for the distal esophagus being the most
frequent site of bleeding.
The explosion theory places considerable emphasis on variceal size, wall thickness, and transmural
pressure as risk factors for bleeding (Table 77-3) . This emphasis appears to be justified. Several groups
have confirmed that variceal size is an important risk factor for variceal bleeding, [4] [11] [92] and several
systems have been proposed for grading varices according to their size and the presence of "red signs" [93]
(Fig. 77-5) . Red signs, which include cherry-red spots and red wale markings (a longitudinal, raised, red
streak), are usually associated with the most advanced grade of varices and are
1293
TABLE 77-3 -- Determinants of Variceal Bleeding

Local Factors
Variceal size--vessel radius
Variceal wall thickness--red signs

Hemodynamic Factors
Portal (intravariceal) pressure--12 mm Hg hepatic venous pressure gradient threshold
Blood volume
Collateral blood flow (?)
Intra-abdominal pressure (?)
Severity of Liver Disease (Child Class)
Other
Salicylates and nonsteroidal anti-inflammatory agents
Continued alcohol abuse (?)
thought to represent focal weaknesses, or "blowouts," in the variceal wall ("varices-on-varices"). They
also clearly carry additional weight as risk factors for bleeding. [92]
The relationship between portal pressure and the risk of variceal rupture has been more subtle. Although
all patients with variceal bleeding will have increased portal pressure, there is no clear correlation
between portal pressure and the risk of bleeding. However, several groups have confirmed that
Figure 77-5 Endoscopic grading of esophageal varices according to Paqet. [93] Grade I: small varices without
luminal prolapse; grade II: moderate-sized varices showing luminal prolapse with minimal obscuring of the
gastroesophageal junction; grade III: large varices showing luminal prolapse substantially obscuring the
gastroesophageal junction; and grade IV: very large varices completely obscuring the gastroesophageal
junction. The presence of red signs ( black spots) is noted separately. (From Paquet, K. J. Prophylactic
endoscopic sclerosing treatment of the esophageal wall in varices: A prospective controlled trial. Endoscopy
14:4, 1982.)
there is an important threshold phenomenon with respect to the portal pressure gradient (HVPG) and the
risk of bleeding. [4] [11] [91] Thus, it is well recognized that varices rarely bleed at an HVPG of less than 12
mm Hg, but there is no direct correlation between the risk of bleeding and an HVPG above this threshold
value. There may be a more quantitative relationship between pressure and the risk of bleeding from
varices depending on where the pressure is measured (i.e., HVPG versus intravariceal pressure) [94] [95]
and when it is measured in relation to an episode of acute variceal hemorrhage. [96]
Inasmuch as they also contribute to portal and intravariceal pressure, total blood volume and variceal
collateral blood flow are two factors that may contribute to the risk of variceal rupture. Variceal blood
flow measured as total azygous vein blood flow has not shown a clear correlation with the risk of
variceal bleeding. On the other hand, reduction in plasma volume through the use of diuretics in patients
with portal hypertension lowers portal and variceal pressure, [97] [98] whereas expansion of intravascular
volume may be a significant determinant of variceal hemorrhage and, in particular, of ongoing acute
bleeding. [99] Intra-abdominal pressure clearly influences several hemodynamic variables in portal
hypertension, but there is no convincing evidence that tense ascites per se, as opposed to decompensated
liver disease, increases the risk of variceal bleeding. [100] [101]

Patients with decompensated liver disease bleed more often and have a worse prognosis than do patients
with compensated liver disease. [102] [103] The severity of the underlying liver disease (Child-Pugh or
Child-Turcotte class) correlates with variceal size [104] but also appears to be an important independent
predictor of variceal hemorrhage and may influence bleeding risk via several other factors, including
poor nutritional state and worsening coagulopathy. [105] [106] The North Italian Endoscopic Club [92]
studied 321 patients with cirrhosis whose varices had not yet bled. They calculated an index using three
variables: Child's class, size of varices, and presence of "red wale" markings. Over a 2-year follow-up
period, patients with the lowest score (mild hepatic dysfunction, small varices, and no red signs) had a
6% risk of bleeding; in those with the highest scores (very poor liver function, large varices, and many
red signs), the risk of bleeding increased to 76%.
Spontaneous regression of varices may occur in alcoholic patients practicing strict abstinence, [107] and
continued alcohol abuse has been long been held to play a role in promoting the risk of variceal bleeding
in patients with alcoholic cirrhosis. Oddly, this relationship has been surprisingly difficult to prove. [100]
[108] On the other hand, suspicions relating to a role for aspirin and nonsteroidal anti-inflammatory drugs
in the precipitation of portal hypertensive bleeding have been substantiated, [109] and patients with portal
hypertension are best advised to avoid these agents.
Risk of Bleeding from Other Sites
Gastric varices located in the fundus, whether isolated or in continuity with esophageal varices, carry a
particularly high risk of bleeding. [77] [78] In patients with portal hypertensive gastropathy, bleeding occurs
more often with more severe grades. Mild or chronic bleeding was observed in 35% of patients with mild
gastropathy versus 90% of patients with severe
1294
gastropathy. Overt bleeding occurred in 30% of those with mild gastropathy and 60% with severe
gastropathy. [110]
Natural History of Variceal Bleeding
Gastroesophageal varices develop in 50% to 60% of cirrhotic patients, and approximately 30% of them
will experience an episode of variceal hemorrhage within 2 years of the diagnosis of varices. [91] [103]
Variceal bleeding is reported to account for 2% to 20% of all upper gastrointestinal hemorrhage and for a
much higher proportion (50%) in patients with severe, persistent bleeding. [111] [112]
The greatest risk of initial variceal bleeding occurs within a 6- to 12-month period after their discovery.
Beyond this time, the risk of bleeding tends to diminish in those who have not already bled. [11] [113] A
diurnal periodicity of variceal bleeding has been noted by several groups; bleeding episodes tend to favor
the early morning and late evening. [113] After a variceal bleed, the risk of rebleeding is particularly high--
approximately 60% to 70% over a 24-month period. The risk of rebleeding is greatest, however, within
hours or days after an acute bleed. [102] [112] [113] Once bleeding has occurred, a number of factors have
been identified that increase the risk of early rebleeding, including bleeding from gastric varices,
thrombocytopenia, encephalopathy, a diagnosis of alcoholic cirrhosis, large varices, active bleeding at
time of diagnostic endoscopy, and a high portal pressure gradient. [78] [96] [102] [113]

McCormick and associates [113] suggested that secondary hemodynamic changes in the splanchnic
circulation after a bleed may contribute to the risk of further bleeding. These changes include an increase
in portocollateral resistance after hypotension, increased splanchnic blood flow stimulated by blood in
the gut, and an increase in portal venous pressure as a result of overzealous volume expansion during
resuscitation. The net effect of these changes would be an "overshoot" in portal pressure and blood flow
during resuscitation, which would encourage continued variceal bleeding.
Prognosis
Variceal hemorrhage is the most serious complication of portal hypertension and accounts for
approximately one fifth to one third of all deaths of cirrhotic patients. [100] [111] [112] The mortality after a
variceal bleed ranges from 40% to 70% in various series, with an average of approximately 50% within 6
weeks. [100] As noted by several authors, this high, early mortality may have a profound effect on the
survival outcome reported by therapeutic studies depending on the interval between presentation and
inclusion in the study. [100] [103] It has also been the prime motivation for the development of prophylactic
therapy. The most important determinant of survival is the patient's level of hepatic function. The
prognosis associated with variceal bleeding is generally much better in patients without significant liver
impairment, such as those with noncirrhotic portal vein thrombosis or idiopathic portal hypertension. [18]
[114] In cirrhotic patients, the prognosis is also worse in the presence of concomitant alcoholic hepatitis,
hepatocellular carcinoma, or portal vein thrombosis. [115] [116] [117]


April 21, 2000
Conde Petra
The changing spectrum of treatment for variceal

bleeding.
Citation
Bibliographic Data
Abstract
Indexing Data
The changing spectrum of treatment for
Copyright Notice and Disclaimer variceal bleeding.
Find More Articles Like This From NIH/NLM MEDLINE, HealthSTAR
NLM Citation ID:
99005003
Full Text
Full Source Title:
Frontmatter
Annals of Surgery
Methods
Publication Type:
Results Journal Article
Preoperative Data Language:
English
Operative Mortality
Author Affiliation:
Postoperative Morbidity
Department of Surgery, University of Wisconsin, Madison,
Long-term Survival USA.
Discussion Authors:
Rikkers LF
Acknowledgments
Abstract:
References
OBJECTIVE: The objective of this study was to assess the
Discussion impact of endoscopic therapy, liver transplantation, and
transjugular intrahepatic portosystemic shunt (TIPS) on patient
selection and outcome of surgical treatment for this
complication of portal hypertension, as reflected in a single
surgeon's 18-year experience with operations for variceal
hemorrhage. SUMMARY BACKGROUND DATA: Definitive
treatment of patients who bleed from portal hypertension has
been progressively altered during the past 2 decades during
which endoscopic therapy, liver transplantation, and TIPS have
successively become available as alternative treatment options
to operative portosystemic shunts and devascularization

procedures. METHODS: Two hundred sixty-three consecutive

patients who were surgically treated for portal hypertensive
bleeding between 1978 and 1996 were reviewed retrospectively.
Four Eras separated by the dates when endoscopic therapy
(January 1981), liver transplantation (July 1985), and TIPS
(January 1993) became available in our institution were
analyzed. Throughout all four Eras, a selective operative
approach, using the distal splenorenal shunt (DSRS),
nonselective shunts, and esophagogastric devascularization, was
taken. The most common indications for nonselective shunts
and esophagogastric devascularization were medically
intractable ascites and splanchnic venous thrombosis,
respectively. Most other patients received a DSRS. RESULTS:
The risk status (Child's class) of patients undergoing surgery
progressively improved (p = 0.001) throughout the 4 Eras,
whereas the need for emergency surgery declined (p = 0.002).
The percentage of nonselective shunts performed decreased
because better options to manage acute bleeding episodes
(sclerotherapy, TIPS) and advanced liver disease complicated
by ascites (liver transplantation, TIPS) became available (p =
0.009). In all Eras, the operative mortality rate was directly
related to Child's class (A, 2.7%; B, 7.5%; and C, 26.1 %) (p =
0.001). As more good-risk patients underwent operations for
variceal bleeding, the incidence of postoperative
encephalopathy decreased (p = 0.015), and long-term survival
improved (p = 0.012), especially since liver transplantation
became available to salvage patients who developed hepatic
failure after a prior surgical procedure. There were no
differences between Eras with respect to rebleeding or shunt
occlusion. Distal splenorenal shunts (p = 0.004) and
nonselective shunts (p = 0.001) were more protective against
rebleeding than was esophagogastric devascularization.
CONCLUSIONS: The sequential introduction of endoscopic
therapy, liver transplantation, and TIPS has resulted in better
selection and improved results with respect to quality and
length of survival for patients treated surgically for variceal
bleeding. Despite these innovations, portosystemic shunts and
esophagogastric devascularization remain important and
effective options for selected patients with bleeding secondary
to portal hypertension.
Major Subjects:
Esophageal and Gastric Varices / Mortality / * Surgery
Gastrointestinal Hemorrhage / Mortality / * Surgery

Adult
Aged
Female
Human
Male
Middle Age
Postoperative Complications / Epidemiology
Retrospective Studies
Survival Rate
Time Factors
Bookmark URL: /das/journal/view/N/10439611?source=HS,MI

Annals of Surgery
Volume 228 Number 4 October 1998
536
CONTENTS SCIENTIFIC PAPERS PRESENTED AT

THE AMERICAN SURGICAL ASSOCIATION
The Changing Spectrum of Treatment for Variceal Bleeding
Layton F. Rikkers MD
Department of Surgery, University of Wisconsin, Madison, Wisconsin

Presented at the 118th Annual Meeting of the American Surgical Association, April 3, 1998, Palm Beach, Florida.
Address correspondence and reprint requests to Dr. Layton Rikkers, MD, University of Wisconsin Medical School,
H4/710 Clinical Science Center, 600 Highland Avenue, Madison, WI 53792-7375.
Accepted for publication April 1998.
Objective
The objective of this study was to assess the impact of endoscopic therapy, liver transplantation,
and transjugular intrahepatic portosystemic shunt (TIPS) on patient selection and outcome of
surgical treatment for this complication of portal hypertension, as reflected in a single surgeon's
18-year experience with operations for variceal hemorrhage.
Summary Background Data
Definitive treatment of patients who bleed from portal hypertension has been progressively altered
during the past 2 decades during which endoscopic therapy, liver transplantation, and TIPS have
successively become available as alternative treatment options to operative portosystemic shunts
and devascularization procedures.
Methods
Two hundred sixty-three consecutive patients who were surgically treated for portal hypertensive
bleeding between 1978 and 1996 were reviewed retrospectively. Four Eras separated by the dates
when endoscopic therapy (January 1981), liver transplantation (July 1985), and TIPS (January
http://home.mdconsult.com/das/article/body/jorg=j...al&source=HS,MI&sp=10439611&sid=0/N/126393/1.html (1 of 18) [21/04/2000 23:50:28]

1993) became available in our institution were analyzed. Throughout all four Eras, a selective
operative approach, using the distal splenorenal shunt (DSRS), nonselective shunts, and
esophagogastric devascularization, was taken. The most common indications for nonselective
shunts and esophagogastric devascularization were medically intractable ascites and splanchnic
venous thrombosis, respectively. Most other patients received a DSRS.
Results
The risk status (Child's class) of patients undergoing surgery progressively improved (p = 0.001)
throughout the 4 Eras, whereas the need for emergency surgery declined (p = 0.002). The
percentage of nonselective shunts performed decreased because better options to manage acute
bleeding episodes (sclerotherapy, TIPS) and advanced liver disease complicated by ascites (liver
transplantation, TIPS) became available (p = 0.009). In all Eras, the operative mortality rate was
directly related to Child's class (A, 2.7%; B, 7.5%; and C, 26.1%) (p = 0.001). As more good-risk
patients underwent operations for variceal bleeding, the incidence of postoperative encephalopathy
decreased (p = 0.015), and long-term survival improved (p = 0.012), especially since liver
transplantation became available to salvage patients who developed hepatic failure after a prior
surgical procedure. There were no differences between Eras with respect to rebleeding or shunt
occlusion. Distal splenorenal shunts (p = 0.004) and nonselective shunts (p = 0.001) were more
protective against rebleeding than was esophagogastric devascularization.
Conclusions
The sequential introduction of endoscopic therapy, liver transplantation, and TIPS has resulted in
better selection and improved results with respect to quality and length of survival for patients
treated surgically for variceal bleeding. Despite these innovations, portosystemic shunts and
esophagogastric devascularization remain important and effective options for selected patients
with bleeding secondary to portal hypertension.
With the exception of a few isolated, pioneering efforts during the first half of this century, definitive
treatment for variceal bleeding began in the mid 1940s when nonselective portosystemic shunts were
introduced into clinical practice. [1] End-to-side portacaval, side-to-side portacaval, and proximal
splenorenal shunts remained the only effective alternatives for this disorder until 1967 when Warren et
al. [2] introduced the concept of selective variceal decompression by means of a distal splenorenal shunt
(DSRS), and Sugiura and Futagawa [3] reported a large series of patients undergoing esophagogastric
devascularization combined with splenectomy.
537
Surgery remained the mainstay for definitive treatment of variceal bleeding until the 1970s when
endoscopic sclerotherapy was introduced. Management of this complex clinical problem was further
altered in the 1980s when liver transplantation became widely available and nonselective beta blockade
was suggested as a pharmacologic treatment for prevention of variceal rebleeding. Finally, the 1990s has
been the decade of the transjugular intrahepatic portal systemic shunt (TIPS), which provides another
nonoperative option for the treatment of variceal hemorrhage.

Although there is controversy as to which treatment option or sequence of treatments should be used in a
given clinical setting, there is no doubt that these relatively recent innovations have had a profound
impact on the overall management of patients with bleeding secondary to portal hypertension. This study
represents a retrospective analysis of a single surgeon's experience with patients undergoing surgery for
variceal hemorrhage during a time span when endoscopic sclerotherapy, liver transplantation, and TIPS
became available as alternative treatment options for these patients. The effects of each of these
innovations on therapeutic strategy and overall results of surgical treatment for variceal bleeding are
assessed.
Methods
From January 1978 through December 1995, 263 consecutive patients underwent 273 nontransplant
operations for bleeding secondary to portal hypertension (esophageal varices, gastric varices, stomal
varices, and portal hypertensive gastropathy) by a single surgeon. Between January 1978 and June 1984,
the operations were performed at the University of Utah Medical Center (97 patients) and Salt Lake City
Veteran's Administration Medical Center (49 patients). Between July 1984 and December 1995, the
operations were conducted at the University of Nebraska Medical Center (105 patients) and the Omaha
Veteran's Administration Medical Center (12 patients). With the exception of three individuals managed
by other surgeons, all patients undergoing operations for variceal bleeding at these institutions during the
time intervals of this series are included. Ten patients underwent a second operation because of failure of
the first one.
Bleeding secondary to portal hypertension was endoscopically documented in all patients by visualizing
an actively bleeding varix or gastropathy lesion, or by identifying prominent varices and no other suspect
lesions in a patient with a recent major upper gastrointestinal hemorrhage (2 or more units of blood). The
cause of portal hypertension was determined by history, liver biopsy, visceral angiography, and hepatitis
serology.
A modified Child's classification was used to grade severity of liver disease just before surgical
intervention. [4] Two biochemical indices (serum albumin and total bilirubin) and two clinical variables
(ascites and encephalopathy) were given values of 1 to 3 and totaled to determine the Child's class (A,
4-5; B, 6-8; C, 9-12). The splanchnic venous circulation was defined preoperatively by selective visceral
angiography except in patients actively bleeding at the time of surgery. Shunt patency was assessed in all
shunt patients during the early postoperative interval by selective visceral angiography and/or duplex
ultrasonography.
An incident of postoperative encephalopathy was defined as mental confusion, clearly related by the
patient or a family member, or detection of disorientation or asterixis, or both, by a physician. Altered
mental status appearing just before death from hepatic failure was not considered encephalopathy.
To determine the impact of endoscopic sclerotherapy, liver transplantation, and TIPS on nontransplant
surgery for variceal hemorrhage, 4 Eras separated by the dates when each of these therapies became
available in our institutions were analyzed. In Eras 1, none of the therapies were available (January
1978-December 1980); in Eras 2, endoscopic sclerotherapy alone was available (January 1981-June
1985); in Eras 3, endoscopic sclerotherapy and liver transplantation were available (January 1981-June
1985); and in Eras 4, endoscopic sclerotherapy, liver transplantation, and TIPS were available (January
1993-December 1995). The dates separating the Eras do not necessarily indicate when each new

treatment was first introduced, but rather the time at which it was readily available for all patients for
whom it was thought to be indicated.
Although the indications for surgical treatment of variceal bleeding have evolved as endoscopic therapy,
liver transplantation and TIPS have successively become available, the criteria for performing each of the
operations have remained the same throughout the period of this investigation. The DSRS was selected
for 177 patients with portal blood flow toward the liver on angiography, absent or medically controllable
ascites, and compatible anatomy (splenic vein greater than 6 mm in diameter and in close relation to the
left renal vein). Although not used in patients who were actively bleeding, the DSRS was considered a
reasonable alternative in the urgent setting (within 48 hours of active bleeding) when bleeding was
temporarily controlled pharmacologically or by balloon tamponade in patients who otherwise met the
criteria for this operation. The standard DSRS as described by Warren et al. [5] was constructed;
splenopancreatic disconnection was not performed. Five patients with unfavorable left renal vein
anatomy received a distal splenocaval rather than a distal splenorenal shunt.
Nonselective shunts were chosen for 60 patients with one or more of the following: medically intractable
ascites, active bleeding at the time of surgery, anatomy incompatible with a DSRS, and bleeding stomal
varices. When intractable ascites was present, one of the varieties of side-to-side shunts was always
chosen. The following nonselective shunts were used: end-to-side portacaval shunt (26 patients),
side-to-side portacaval shunt (11 patients), interposition mesorenal shunt (10 patients) interposition
538
TABLE 1 -- Preoperative Data Based on Eras

Eras 1 Eras 3 Eras 4
(1/78- Eras 2 (7/85- (1/93-
12/80) (1/81- 6/85) 12/92) 12/95) Overall
Number of patients 70 87 59 47 263
Age (yr) 51 12 52 11 50 16 46 17 50 14
Sex (%)
Male 70 71 75 49 68
Female 30 29 25 51 32
Cause of portal hypertension
(%)
Alcoholic cirrhosis 61 62 61 26 55
Nonalcoholic cirrhosis 36 33 31 49 36
Noncirrhotic 3 5 8 26 9
Child's class (%)
A 43 26 36 77 42

B 30 52 51 23 41
C 27 22 14 0 17
Type of operation (%)
DSRS 66 60 76 72 67
Nonselective shunt 23 33 15 13 23
Devascularization 11 7 9 15 10
Elective (%) 76 76 83 98 81
Emergency (%) 24 24 17 2 19
Operations/year 23.3 19.3 7.9 15.7 14.5
Continuous data are presented as mean standard deviation.
Refer to text for statistical comparisons.
DSRS = distal splenorenal shunt.
portocaval shunt (6 patients), side-to-side splenorenal shunt (5 patients), interposition mesocaval shunt (2
patients). A synthetic graft (dacron or polytetrafluoroethylene) of 14-20 mm was used for all
interposition procedures. An additional three nonselective shunts were performed as second procedures
after failure of the primary operation.
Esophagogastric devascularization (26 patients), combined with splenectomy if not previously
accomplished, was a modification of the procedure described by Sugiura and Futagawa [3] and was
performed entirely through an abdominal incision. [6] The proximal two thirds of the stomach and distal 7
cm of esophagus were devascularized. A less complete devascularization consisting of greater gastric
curvature devascularization and ligation of the left gastric vein on the lesser curvature was accomplished
in the first eight patients. With the exception of two patients, esophageal transection was not performed.
This procedure was reserved for patients who were unshuntable because of diffuse splanchnic venous
thrombosis, usually due to a hypercoagulable disorder, for patients with a failed DSRS, and for
individuals with massive splenomegaly causing either pressure symptoms or hypersplenism. An
additional seven patients underwent devascularization to treat recurrent variceal bleeding after failure of
their primary surgery.
Significant differences between groups were determined by Fisher's exact test when computationally
feasible, otherwise by the chi-square test. The rank sums test was used for continuous variables. Survival
analysis was performed by the Kaplan and Meier method, and differences between survival curves were
assessed by the log rank test. Means one standard deviation are reported for continuous variables.
Results
Preoperative Data
Table 1 compares the 4 Eras with respect to demographic variables, causes of portal hypertension,

preoperative Child's class, procedures performed, and the frequency of emergency surgical intervention.
Table 2 provides a similar analysis based on the type of operation performed.
Progressively fewer Child's class C patients underwent surgery for variceal bleeding (Eras 1 to Eras 4) as
the other treatment modalities became available (p = 0.001). Not a single Child's class C patient was
operated on since the availability of TIPS in our institution (eras 4). The lower percentage of patients
with alcoholic cirrhosis who were managed surgically in Eras 4 was due to an increased percentage of
patients who were noncirrhotic with splanchnic venous thrombosis rather than an increase in the number
of patients with nonalcoholic cirrhosis (p = 0.001). A major benefit of TIPS (Eras 4) is that it has nearly
eliminated the need for emergency surgery (p = 0.002). In contrast, endoscopic sclerotherapy (Eras 2)
had no apparent impact on the frequency of emergency operations. The highest number of operations per
year were performed in Eras 1 before any other reliable definitive therapy for variceal bleeding was
539
TABLE 2 -- Preoperative Data Based on Operation Performed

Nonselective
DSRS Shunts Devascularization Overall
Number of patients 177 60 26 263
Age (yr) 50 14 53 10 45 15 50 14
Sex (%)
Male 68 75 50 68
Female 32 25 50 32
Cause of portal hypertension (%)
Alcoholic cirrhosis 54 75 15 55
Nonalcoholic cirrhosis 41 23 31 36
Noncirrhotic 5 2 54 9
Child's class (%)
A 50 13 50 42
B 40 43 42 41
C 10 43 8 17
Elective (%) 88 65 73 81
Emergency (%) 12 35 27 19
Operations/year 9.8 3.3 1.4 14.5
Continuous data are presented as mean standard deviation.


available. Relatively more DSRSs and fewer nonselective shunts were performed in Eras 3 and 4 than in
Eras 1 and 2 (p = 0.009).
Based on the indications for the various operations, patients receiving a nonselective shunt were more
likely to have alcoholic cirrhosis (p = 0.017), to require emergency surgery (p = 0.001), and to be in
Child's class C (p = 0.001) (Table 2) . Patients who were noncirrhotic, many of whom were not shuntable
because of diffuse splanchnic venous thrombosis, were concentrated in the devascularization group (p =
0.001). The eight patients with noncirrhotic portal hypertension in the distal splenorenal shunt group had
isolated portal vein thrombosis; five were in the pediatric age range.
Operative Mortality
An operative death was defined as one occurring within 30 days of an operation or at any time during the
same hospitalization. Operative mortality rates for Eras 1 through 4 were 7%, 15%, 7%, and 2%,
respectively, with no significant differences between eras. Because Child's Class C patients and
emergency operations were concentrated in the nonselective shunt group, the operative mortality rate was
higher after this operation (25%) than after DSRS (4.5%, p = 0.001) or esophagogastric devascularization
(0%, p = 0.005).
The operative mortality rate was directly related to the preoperative Child's class (A, 2.7%; B, 7.5%; C,
26.1%) (p = 0.001). Early postoperative death was more common after emergency operations (25%) than
after elective procedures (5%) (p = 0.0001). The operative mortality rate was not significantly different
for alcoholic (10.3%) and nonalcoholic cirrhotics (8.4%, p = 0.66); none of the patients with noncirrhotic
portal hypertension died during the early postoperative interval.
Postoperative morbidity reflected by rebleeding, shunt occlusion, and encephalopathy is presented in

Table 3 for
TABLE 3 -- Postoperative Morbidity by Eras

Eras 1 1/78- Eras 2 1/81- Eras 3 7/85- Eras 4 1/93-
12/80 6/85 12/92 12/95 Overall
Number of patients 70 87 59 47 263
Rebleeding 9 (13) 14 (16) 4 (7) 5 (11) 32 (12)
Shunt occlusion 3 (4) 6 (7) 1 (2) 2 (4) 12 (5)
Encephalopathy 15 (24) 17 (23) 4 (7) 4 (9) 40 (17)
Numbers in parentheses are percentages.

540
TABLE 4 -- Postoperative Morbidity by Operation

Nonselective
DSRS Shunts Devascularization Overall
Number of patients 177 60 26 263
Rebleeding 20 (11) 3 (5) 9 (35) 32 (12)
Shunt occlusion 9 (5) 3 (5) - 12 (5)
Encephalopathy 23 (14) 16 (36) 2 (8) 40 (17)
Numbers in parentheses are percentages.
each Eras and in Table 4 for each type of surgery. There were no significant differences between Eras
with respect to rebleeding or shunt occlusion. Both selective (p = 0.004) and nonselective shunts (p =
0.001) were more protective against rebleeding than esophagogastric devascularization, but there was no
significant difference between shunt types with respect to this variable (p = 0.21). Seven of the nine
patients with an occluded DSRS underwent a secondary operation (nonselective shunt, two patients;
devascularization procedure, five patients). Two individuals had a second procedure after a failed
nonselective shunt (nonselective shunt, one patient; devascularization procedure, one patient). Repeat
esophagogastric devascularization was performed in one patient who rebled after the first surgery. Four
patients died during the early postoperative interval after a second operation (nonselective shunt, one
patient; devascularization procedure, three patients), and another two patients rebled again. The overall
operative mortality rate for secondary operations was 40%, compared with 8.7% for primary operations.
Postoperative encephalopathy was less common in Eras 3 and 4 than in Eras 1 and 2 (p = 0.015).
Encephalopathy was more frequent after nonselective shunt than after DSRS (p = 0.002) or
esophagogastric devascularization (p = 0.01). There was no difference in encephalopathy rates after
DSRS and devascularization operations. The frequency of encephalopathy was similar for alcoholic
(20%) and nonalcoholic cirrhotics (19%); none of the patients with noncirrhotic portal hypertension
developed this postoperative complication.
Long-term Survival
Patients in Eras 1 to 4 have been followed for 6.2 6.4, 6.4 5.9, 5.5 3.7, and 3.1 1.1 years,
respectively. Surviving patients in the four eras have been followed for 17.3 3.7, 14.5 2.6, 8.3 2.5,
and 3.4 0.9 years, respectively. The survival status of all patients except two, who were lost to
follow-up 4.5 and 6 years after surgery, has been determined. All patients have been followed for a

minimum of 2 years. Thirty-nine surviving patients underwent surgery less than 5 years ago. All others
have been followed for a minimum of 5 years.
Three-year, 5-year, and 10-year survival rates for all patients are 65%, 52%, and 36%, respectively.
Figure 1 shows that long-term survival has progressively improved through the 4 Eras (p = 0.006). The
survival curves of Eras 3 and Eras 4 have been bolstered by transplantation of five and three patients,
respectively, when hepatic failure ensued. Six of these eight patients are still alive; two died in the early
posttransplant interval. Figure 2 demonstrates that the higher risk nonselective shunt patients had poorer
long-term survival compared with DSRS patients (p = 0.0004) or patients undergoing esophagogastric
devascularization (p = 0.005). Patients with noncirrhotic portal hypertension survived significantly
longer than patients with alcoholic cirrhosis (p = 0.004) and nonalcoholic cirrhotics (p = 0.02) (Figure 3)
. Survival curves of patients with alcoholic and nonalcoholic cirrhosis were not significantly different (p
= 0.10). Child's class also was an important determinant of long-term survival (p = 0.0001) (Figure 4) .
Figure 1. Survival curves according to Eras. The curve for Eras 1 is significantly different from those
for Eras 3 (p = 0.05) and Eras 4 (p = 0.0006), and the curves for eras 2 and Eras 4 also are different (p
= 0.004).
541
Figure 2. Survival curves according to type of operation. The survival curve for nonselective shunt is
significantly different from those for distal splenorenal shunt (p = 0.0004) and esophagogastric
devascularization (p = 0.005), which are not different from each other.
Discussion
During the 18-year time span of the present series, definitive treatment of patients with variceal bleeding
has changed considerably. In 1978, the only available nonoperative options, pitressin infusion and
balloon tamponade, controlled acute hemorrhage in many patients, but rebleeding rates exceeded 70%.
Endoscopic sclerotherapy was being developed overseas, but it was not widely available in the United
States. Liver transplantation was an experimental procedure confined to a few centers. Thus, the only
reliable options for preventing recurrent bleeding were a variety of
Figure 3. Survival curves according to cause of portal hypertension. The curve for noncirrhotic portal
hypertension is significantly different from those for alcoholic cirrhosis (p = 0.004) and nonalcoholic
cirrhosis (p = 0.03), which are not different from each other.

Figure 4. Survival curves according to Child's class. The curve for class A is significantly different
from those for class B (p = 0.006) and class C (p = 0.0001), which are different from each other (p =
0.005).
nontransplant operations, all of which could be classified in one of three categories: nonselective shunts,
selective shunts, and devascularization operations.
Rather than embracing a single operation and applying it to all patients who bleed from portal
hypertension, since the beginning of our program we have advocated a selective operative approach,
tailoring the procedure to the individual patient's circumstances. [7] Based on the results of several
randomized trials, [8] [9] [10] which found a lower frequency of encephalopathy after the DSRS than after
nonselective shunts, the DSRS has been used for most of the patients with compatible anatomy for this
procedure and absent or medically manageable ascites. Although not selected for patients who were
actively bleeding at the time of surgery, the DSRS has been an acceptable alternative in the urgent setting
when variceal decompression was required within 48 hours of active bleeding. [11] Using these criteria,
67% of the patients in the entire series and 75% of patients since the advent of liver transplantation
underwent this operation. The use of nonselective shunts, which have mainly been chosen for actively
bleeding patients and for those with medically intractable ascites, has progressively declined as better
options (liver transplantation, TIPS) have become available for such patients. The chief indication for
esophagogastric devascularization continues to be diffuse splanchnic venous thrombosis. Because many
patients undergoing this operation are noncirrhotic and unshuntable, the frequency of its use has not been
affected by the availability of liver transplantation or TIPS.
The principal finding of this retrospective analysis is that the overall results of operations to control
bleeding secondary to portal hypertension have progressively improved in our institution over the past 2
decades. Although the surgical techniques themselves and the indications for the various procedures have
remained relatively constant, the risk
542
status of the patients receiving them has changed considerably. Many patients with advanced liver
disease (Child's class B and C) undergo liver transplantation soon after or even before they bleed from
varices. Endoscopic treatment and TIPS for failures of endoscopic therapy serve as excellent short-term
means of preventing rebleeding for these patients. Although the long-term effectiveness of TIPS is
questionable, [12] it is probably also a reasonable alternative for patients with limited hepatic functional
reserve who are not liver transplant candidates because of continuing alcoholism, associated diseases of
other major organ systems, or advanced age. The TIPS is likely to remain functional during the relatively
brief life expectancies these patients face. In the present series, not a single Child's class C patient
underwent surgery for variceal bleeding since TIPS became available (Eras 4).
Another key factor contributing to the recent improved quality and length of survival experienced by
patients in this series is the availability of liver transplantation as a salvage operation for patients who
http://home.mdconsult.com/das/article/body/jorg=...l&source=HS,MI&sp=10439611&sid=0/N/126393/1.html (10 of 18) [21/04/2000 23:50:28]

develop hepatic failure during the late postoperative interval after a shunt or devascularization operation
to control variceal bleeding. Patients who are potential future transplant candidates need to be evaluated
at least annually so they can be promptly placed on the transplant list as soon as progressive hepatic
functional deterioration is detected. Eight patients from Eras 3 and 4 of this series have undergone liver
transplantation and six remain alive from 1 to 7 years since transplantation. Although several more
patients will almost certainly undergo transplantation in the future, for many a shunt will serve as their
final definitive procedure either because they are not transplant candidates or because their hepatic
functional reserve remains stable indefinitely. Thirty-seven patients with cirrhosis (22 alcoholic, 15
nonalcoholic) are still alive more than 10 years since undergoing surgery for variceal bleeding (DSRS, 26
patients; nonselective shunt, 8 patients; devascularization, 3 patients) and only 2 have received liver
transplants. Twenty of these 37 patients underwent surgery (DSRS, 13 patients; nonselective shunt, 6
patients; devascularization, 1 patient) more than 15 years ago.
In this series, endoscopic treatment alone (Eras 2) had less impact on the risk status of surgical
candidates and on the outcome of surgery than liver transplantation (Eras 3) and TIPS (Eras 4). Although
endoscopic therapy is quite effective for control of acutely bleeding esophageal varices, the percentage of
patients who eventually rebleed is 40% to 60% in most long-term series. [13] Although rebleeding can
often be effectively treated by repeat endoscopic sclerosis or banding, the eventual failure rate of
endoscopic therapy, defined as either death from recurrent bleeding or the need for alternative therapy, is
as high as 40%. [4] In our experience, endoscopic treatment alone did not decrease the need for
emergency surgery, but rather delayed it until endoscopic sclerotherapy failure developed. Because many
sclerotherapy failures bleed from the stomach rather than the esophagus, the only effective alternatives
for these patients during Eras 2 and Eras 3 were surgical portal decompression or esophagogastric
devascularization. In contrast, emergency surgery has nearly been eliminated since TIPS became
available and has become the preferred therapy for most bleeding patients who are not controlled by
other nonoperative means (pharmacotherapy or endoscopic treatment). Many patients in this situation are
at high risk (Child's class C) for surgical intervention.
In many institutions, operative shunts are no longer used as treatment for variceal bleeding. When the
first-line options of nonselective beta blockade or endoscopic treatment fail to control bleeding in these
hospitals, a TIPS is usually placed. The advantages of TIPS are that it is nonoperative, it effectively
decompresses the portal venous circulation during the short-term, and early complications and
procedure-related mortality are infrequent. [14] However, late TIPS failure rates are high, with thrombosis
or stenosis developing in approximately 50% of patients by 1 to 2 years. [12]
Although TIPS revisions are successful in many patients, in most series, rebleeding rates after TIPS are
considerably higher (10%-30%) than after surgically constructed shunts (<10%). [12] When patent, a TIPS
is usually a nonselective shunt with encephalopathy rates in most trials similar to those seen after a
portacaval shunt. [12] Despite these disadvantages, TIPS is an excellent option for patients whose
endoscopic treatment is unsuccessful and who require relatively short-term portal decompression while
on the waiting list for a liver transplant or whose anticipated survival is limited by their underlying liver
disease. Because it is a relatively recent innovation, no long-term outcomes comparable with those
available for surgical shunts have yet been reported for TIPS. An ongoing multiinstitutional controlled
trial comparing TIPS and DSRS in patients with Child's class A or B liver disease may determine which
of these options is preferable for long-term control of bleeding in patients with good hepatic functional
reserve. To date, the only randomized comparison of an operative shunt and TIPS demonstrated an
overall lower failure rate for patients receiving a small diameter interposition portacaval shunt. [15]

The results of the present series suggest that surgery remains an important and effective option for
selected patients who bleed secondary to portal hypertension. Since the availability of liver
transplantation and TIPS, which in our institution are used in preference to nontransplant operations for
patients who are poor operative risks, the outcomes of surgical intervention for variceal hemorrhage have
been particularly excellent. Nonselective shunts are infrequently required and emergency operations have
nearly been eliminated. The operative mortality rate (Eras 4, 2%) for these selected patients approaches
the procedure-related mortality rate observed in most TIPS series with considerably fewer patients
experiencing recurrent bleeding (11%) or encephalopathy (9%) than after the TIPS procedure. [12]
Long-term survival has been particularly impressive for patients undergoing surgery since the advent of
liver transplantation (Eras 3 and 4), especially for those who are potential liver transplantation
543
candidates and who can be salvaged by this procedure when hepatic failure develops. In an earlier study
from our institution, the 5-year survival rate for this subgroup was greater than 80%. [16]
Acknowledgments
The author gratefully acknowledges Debra Chicks, RN for assisting with patient follow-up, Dennis
Heisey, PhD for statistical analysis, and Ms. Teri Cummings for manuscript preparation.
References
1. Whipple AO. The problem of portal hypertension in relation to the hepatosplenopathies. Ann Surg 1945:122:449-475.
2. Warren WD, Zeppa R, Fomon JJ. Selective transsplenic decompression of gastroesophageal varices by distal splenorenal
shunt. Ann Surg 1967;166:437-455.
3. Sugiura M, Futagawa S. A new technique for treating esophageal varices. J Thorac Cardiovasc Surg 1973;66:677-685.
4. RikkersLF, Jin G, Burnett DA, et al. Shunt surgery versus endoscopic sclerotherapy for variceal hemorrhage: late results
of a randomized trial. Am J Surg 1992;165:27-33.
5. Warren WD, Salam AA, Hutson P, Zeppa R. Selective distal splenorenal shunt: technique and results of operation. Arch
Surg 1974:108:306-314.
6. Jin G, Rikkers LF. Transabdominal esophagogastric devascularization as treatment for variceal hemorrhage. Surgery
1996;120:641-649.
7. Rikkers LF, Soper NJ, Cormier RA. Selective operative approach for variceal hemorrhage. Am J Surg 1984;147:89-96.
8. MillikanWJ, Warren WD, Henderson JM, et al. The Emory prospective randomized trial: selective versus nonselective
shunt to control variceal bleeding. Ann Surg 1985;201:712-722.
9. Langer B, Taylor BR, Mackenzie DR, et al. Further report of a prospective randomized trial comparing distal splenorenal
shunt with end-to-side portacaval shunt. Gastroenterology 1985;88:424-429.
10. da Silva LC, Strauss E, Gayotto, et al. A randomized trial for the study of elective surgical treatment of portal

hypertension in Mansonic schistosomiasis. Ann Surg 1986;204:48-153.
11. Rikkers
LF, Jin G. Emergency shunt: role in the present management of variceal bleeding. Arch Surg
1995;130:472-477.
12. Barton
RE, Rosch J, Saxan RR, et al. TIPS: short and long-term results: a survey of 1750 patients. Semin Interven
Radiol 1995;82:813-822.
13. Snady
H. The role of sclerotherapy in the treatment of esophageal varices: personal experience and a review of
randomized trials. Am J Gastroenterology 1987;82:813-822.
14. Kamath PS, McKusick MA. Viewpoints in digestive diseases. Gastroenterology 1996;111:1700-1705.
15. Rosemurgy AS, Goode SE, Zweibel BR, et al. A prospective trial of transjugular intrahepatic portasystemic stent shunts
versus small-diameter prosthetic H-graft portacaval shunts in the treatment of bleeding varices. Ann Surg
1996;224:378-386.
16. Rikkers
LF, Jin G, Langnas AN, Shaw BW. Shunt surgery during the eras of liver transplantation. Ann Surg
1997;226:51-57.
Discussion
Dr. J. Michael Henderson (Cleveland, Ohio): I appreciate the opportunity to discuss this paper. Dr.
Rikkers has chronicled for you the surgical history of portal hypertension over the past two decades. As
he has pointed out this morning, things have changed. Sclerotherapy, liver transplant, and TIPS have
dramatically altered how we do business. But I agree there is still a role for surgery in selected patients.
You have shown us data on the increasingly good outcomes that have been achieved over time with
"conventional surgery." There is an old axiom that there is no such thing as a good cirrhotic, but you
have shown that, if you are careful and select patients well, there are good cirrhotic patients who achieve
excellent outcomes. I have three questions for you. The first relates to the need for transplant. You have
shown that in the last 106 patients the last 2 eras, 8 patients came to transplant. Can you give us some
indication of approximately how many of these patients would never need transplant? How many of
these had portal vein thrombosis and a normal liver? And, how many have contraindications to
transplant? How many do you think may ultimately come to transplant? The second question relates to
TIPS. What do you really feel about TIPS? I agree with you on the role of TIPS in keeping you in bed at
night, and replacing emergency surgery. You indicated the role of TIPS as a transplant. But do you think
there is a role for TIPS in the management of the better risk elective patient? My final question relates to,
who do you think should be doing these operations? You and I are a dying breed. As other innovations
have come along, there are few surgeons who are doing shunt surgery. It is interesting over the last
couple of years I have had more calls saying, "How do we do this?" And most of those calls are coming
from the transplant surgeons. You have worked in two institutions with strong transplant programs. And
I am interested in how you see the interactions between "the shunters and the transplanters." Should
shunt surgery be done by the liver transplant surgeons? I think this is an excellent series. I appreciate the
opportunity to review the manuscript, which presents an excellent experience.
Dr. Layton F. Rikkers (Madison, Wisconsin): Out of the 106 patients in eras 3 and 4 combined, 46 of
were future transplant candidates. The survival for that group of patients is about 95% at 2 years and
85% at 3 and 5 years. The survival is so good because these patients can be salvaged by transplantation

when they develop liver failure. The remaining patients are not transplant candidates because of active
alcoholism or compromised function of other organs such as the heart or lungs. How many patients will
eventually come to transplant? I have generally thought since the advent of the transplant era that a shunt
is a long-term bridge to transplant. However, when I look back at eras one and two, there are 37 patients
that are still alive and doing well after undergoing a shunt operation more than 10 years ago. And there is
a subset of 20 patients that are still doing well 15 years after a shunt. Thus, there are a significant number
of cirrhotic patients who maintain their hepatic functional reserve indefinitely and will probably never
require transplantation. What is the role for TIPS in 1998? I think TIPS is an excellent short-term option
when endoscopic therapy fails or is not indicated. The patient who needs a liver transplant in the near
future is well served by a TIPS as a short-term bridge to transplantation. Likewise, TIPS is a good option
for nontransplant candidates who are desperately ill, e.g., Child's class C cirrhotics, and unlikely to
survive for very long. Since approximately 50% of TIPS patients develop shunt occlusion or stenosis
within two years, TIPS may not be a good choice for better risk Child's class A or B patients when
endoscopic therapy fails. I feel that an operative shunt is presently a better option for such patients. There
is presently an ongoing randomized trial of TIPS versus distal splenorenal shunt for the long-term
management of good risk patients who fail endoscopic therapy or in whom endoscopic
544
treatment is not indicated. The results of this trial should better define the role of TIPS in this group of
patients. Finally, who should be doing shunt operations? The surgical and the medical management of
patients with liver disease is quite complex. Ideally, these patients should be referred to institutions that
have the capacity of offering all the options; endoscopic treatment, TIPS, open shunts, and transplants.
Dr. Henry A. Pitt (Milwaukee, Wisconsin): I would also like to compliment Dr. Rikkers for keeping
track of and reporting the long-term results of his personal series of portal decompressive operative
procedures over the past 18 years. As endoscopic sclerotherapy, liver transplantation, and TIPS have
been introduced, his indications for distal splenorenal shunt, nonselective shunts, and devascularization
have remained constant. The result is an interesting historical perspective of the role of portal
decompressive surgery, which is now reserved for a younger, more female, nonalcoholic, noncirrhotic
Child's A patient population requiring elective surgery. Clearly, the introduction of new procedures has
been a factor in this change. However, I wonder how much of this change is related to increased
judgment and/or decreased availability of the surgeon. Most of the information on shunts and
devascularization have been presented and published before. However, Dr. Rikkers' experience with
devascularization in 26 patients, most of whom have splanchnic venous thrombosis and relatively few
surgical options is reasonably unique. And I would like to ask more about this interesting group. First,
what were the indications for surgery in this group? Specifically, were sclerotherapy and/or beta blockers
used routinely before going to surgery? Was recurrent bleeding or a single significant bleed the
indication for surgery? In addition, you did not employ esophageal transection in 24 of these 26 patients,
but 35% of them re-bled. My last question with respect to devascularization therefore is, what have you
learned about the technical aspects of this procedure that you can share with us? Finally, you reoperated
on ten patients, but four of these patients died and two rebled. Obviously, the risk of reoperative surgery
is quite high. In the past other options may not have been available; however, now that we have liver
transplantation and TIPS, what are your indications for reoperation?
Dr. Layton F. Rikkers (Madison, Wisconsin): We have done a total of 33 devascularization operations.

The 26 reported here are primary operations. The other seven are secondary operations used after failure
of a primary operation. The main indication for a devascularization operation is diffuse splanchnic
venous thrombosis. These are often patients with hypercoagulable disorders. They are a particularly
difficult group, because they have potentially life-threatening clotting and bleeding problems, so you
need to manage them carefully. Another common indication for a devascularization operation is a failed
distal splenorenal shunt. Although the frequency of shunt thrombosis in the distal splenorenal shunt
group is less than 5%, when it does occur, a splenectomy needs to be included as a component of the
salvage operation. Therefore, a devascularization operation is a good fit for these patients. The other
subset of patients who underwent esophagogastric devascularization were those with large,
uncomfortable spleens or severe hypersplenism. Many of the patients undergoing a devascularization
operation bled from gastric rather than esophageal varices, because of splenic vein thrombosis.
Therefore, esophageal transsection was usually not done. In my opinion, there is little evidence that
esophageal transsection, which can cause significant problems (leak or stenosis), is an important
component of esophagogastric devascularization for any patient.
Dr. Josef E. Fisher (Cincinnati, Ohio): It is a lovely paper, and particularly in this area where there are so
many multi-center trials. And I personally believe that the management of the individual surgeon's
judgment has a lot to do with outcome. And the outcomes here are really extraordinary. A couple of
observations and three questions. First of all, the distribution of the patient population is rather different,
for example, than our own, which includes, while urban inner city population and referral practice, I
think the large percentage of nonalcoholic cirrhotics in this particular group, which you would expect as
the program gets a reputation for excellence in dealing with this problem, makes this slightly atypical for
a number of other series, including our own. I agree entirely with the observation that sclerotherapy does
not decrease the incident of emergency surgery. In our experience, emergency sclerotherapy does not
stop the class C, or more important the class X, the orange with four toothpicks, that you would like very
much to stop, and the latter group may not be an operative candidate anyway. So I agree entirely with
that observation. I have three questions. The first deals with the very large number of patients with
esophagogastric devascularization, not so much the group with splanchnic venous thrombosis for
hemologic disease but the eight patients apparently that failed their initial distal splenorenal shunt. That
is probably the worst group to do esophagogastric devascularization, and yet you had no mortality in that
particular group. I wonder whether you would comment on that. The second question has to do with
TIPS in the emergency situation in era 4. Obviously in TIPS you avoid general anesthesia a good part of
the time. And since general anesthesia decreases hepatic blood flow, it probably decreases mortality in
this group. Do you have any data on what the mortality was in the class C's that you did not operate on
and underwent TIPS under emergency circumstances? Are you merely displacing mortality from an
operative group to what we call a nonoperative group in the emergency class C patients? Finally, you
have a significant number of patients with intractable ascites. In that particular group you have used
nonselective shunts. In our own practice we have used the proximal splenorenal shunt in two randomized
series. There was no difference in encephalopathy between that group and the distal splenorenal shunt
group, despite the fact that the distal splenorenal shunt group was not carried out in intractable ascites. Is
this a different group of patients? Do you have enough data that you are able to tell the difference in
outcome between the intractable ascites group, which requires decompression of the liver side, as
compared to the group that qualified for distal splenorenal shunt? There is a wealth of data in the
manuscript and I urge everybody to read it. It is an extraordinarily valuable contribution to this area of
experience.

Dr. Layton F. Rikkers (Madison, Wisconsin): I would agree that this is an atypical series of patients, like
I think every series of portal hypertension patients is atypical. Because portal hypertension is caused by a
large number of heterogeneous diseases, the composition of the patient population varies considerably
from center to center. I have had a fairly high percentage of reasonably favorable, nonalcoholic patients.
Such patients tend to be referred when you work in a liver transplant center. Why did the patients
devascularized after a failed distal splenorenal shunt do so well? They didn't. There were two mortalities
in the group of five patients that underwent devascularization after a failed distal splenorenal
545
shunt. Both were done in desperate situations when there were no alternatives. TIPS has the
disadvantages of being a nonselective shunt with a high frequency of encephalopathy and a high shunt
failure rate over time. I don't want to discuss which type of shunt has less encephalopathy, selective
shunts or nonselective shunts, but if selective shunts do, they are a better option than TIPS for good risk
patients. They are also considerably more durable with respect to shunt patency. The operative mortality
rate for all patients in era four was only two percent, which is similar to the reported procedure-related
mortality after TIPS. I don't have any data on the spectrum of patients undergoing TIPS or the mortality
rate after TIPS. I suspect there is a shift of mortality away from shunt patients to TIPS patients with
advanced hepatic disease. Are patients with intractable ascites different? I suspect they are because many
of them have spontaneous reversal of portal flow. The natural history of these patients is one of quite
brief survival after the onset of ascites. Therefore, many of them now undergo transplantation rather than
a non-transplant operation. In our series, most of the patients with intractable ascites were Child's class C
and underwent a nonselective shunt. The long-term survival of these patients was quite limited.
Dr. Shunzaburo Iwatsuki (Pittsburgh, Pennsylvania): I am very pleased to know as a transplant surgeon
as well as an occasional shunt surgeon, you selected so few patients every year, 16 to 17 patients per
year, for operative treatment of variceal bleeding. I wonder if you have a rough idea how many patients
are admitted per year into your hospital for esophageal bleeding? How many patients per year in average
received TIPS? And how many patients with a history of esophageal bleeding received a liver
transplantation?
Dr. Layton F. Rikkers (Madison, Wisconsin): I do not have specific data as to the number of variceal
bleeders admitted per year or the number of TIPS done per year. In my University of Nebraska
experience, approximately 28% of adult liver transplant patients had bled from varices. In a recent
publication ( Ann Surg 1997;226:51---57), I compared transplant patients with prior variceal bleeding to
shunt patients who were future transplant candidates. The survival was superior for the shunt group, with
15% of them salvaged by transplant when they developed liver failure, as compared to the transplant
group. We concluded that patients with good hepatic reserve but persistent variceal bleeding should not
undergo transplantation but rather have a shunt operation. Thus far, only a minority of these patients have
required transplantation.
Dr. Marshall J. Orloff (San Diego, California): I appreciated the opportunity to hear this retrospective
review of Dr. Rikkers' 20-year experience with the surgical treatment of variceal bleeding, particularly
since he has had an interest in this difficult problem throughout his surgical career and has made
important contributions to our knowledge in this field. I have several comments and questions. My first
comment is to emphasize and wholeheartedly agree with his conclusion that surgery remains an
important option--I believe the most important option--for treatment of variceal bleeding. It is regrettable

that many young surgeons being trained today are not taught how to do a portal-systemic shunt, which is
the mainstay of treatment of a common disorder that causes 40,000 deaths a year in the U.S.A. Second, I
wish to thank Dr. Rikkers for not getting into a debate about which shunt is better--portacaval shunt or
distal splenorenal shunt. This debate diverts the focus from the heart of the matter, which is
decompression of the portal system by whatever means the surgeon can do well, be it direct portacaval
shunt or distal splenorenal shunt. I want to point out that only 2% of Dr. Rikkers' total series and 6% of
his patients since 1985 underwent liver transplantation (LT). Similarly, in our series of 1761 shunts, only
1.7% subsequently required LT. The thought that these patients often require LT, or that LT is a frequent
salvage operation, is a serious misconception of the facts. The slide shows our published 5-year data after
portal-systemic shunt during the same time period reviewed by Dr. Rikkers--a total of 1497 cases (Table
1) . Not included are two not-yet-published prospective, randomized studies of emergency portacaval
shunt (EPCS) versus emergency and long-term sclerotherapy involving 204 patients, soon to be
published, and our current prospective, randomized trial in progress of EPCS versus TIPS involving 60
patients to date. All of these studies were prospective, which greatly increases their validity, and our
10-year follow-up rate was 97%. The studies of EPCS involved unselected patients, all comers included,
all of whom were operated on within 8 hours. My questions are: First, how many of your 263 patients
had extrahepatic portal hypertension (EHPH)? Such patients are very different from those with cirrhosis,
since they do not have liver disease. Those who are shuntable are all very good risk patients, as you can
see from our data showing a 99% 5-year survival. Those who are unshuntable and require a
devascularization procedure also have a high long-term survival rate. It is misleading to lump these
noncirrhotic patients who have normal liver function with cirrhotics, as you have done. Second, please
comment on your apparent loss of interest in surgical therapy during the 7 years from July 1985 to
December 1992, during which you did only 8 operations per year, and the rebound to 16 operations per
year during the most recent 2-year period. What was responsible for your renewed enthusiasm, or were
these statistics a reflection of the patient referral pattern which was beyond your control? Third, what
was your percent direct 5-year rate that you personally performed, which is a critical factor in evaluating
the validity of your data? It is essential to know much more than whether or not the patients were alive.
Finally, we note that the number of alcoholics in your four periods declined from 61% to 26% and the
number of Child's C patients declined progressively from 27% to 22% to 14% to zero. You concluded, of
course, that by operating on better risk patients there has been a progressive improvement in quality and
duration of survival. My final question is, what happens to the many patients who are in Child's class C,
or are alcoholic, or have uncontrollable bleeding, and how do you justify excluding this substantial
segment of the bleeding cirrhotic population? I call to your attention the 64% 5-year survival rate that we
reported to this Association in 1992, in 94 Child's C patients, all comers included, who underwent EPCS.
Furthermore, I want to point out that endoscopic sclerotherapy (EST) is not the answer, since in our
recently completed prospective randomized study of EST (emergency and long-term) versus EPCS
involving 204 patients, the long-term failure rate of EST was 68%! As Dr. Iwatsuki pointed out so
clearly a moment ago, in order for your conclusions to be valid about the improvement in results of
surgery, and the impact of other modalities such as EST, LT, and TIPS, you have to know and tell us
how many patients with variceal bleeding entered your institution, how many of these underwent EST,
LT, and TIPS rather than being referred to you, and how many survived these other measures of therapy.
It is not
546

very meaningful to simply report the survival rate of the good risk patients who happened to end up in
your hands.
Dr. Layton F. Rikkers (Madison, Wisconsin): Most people only have two constants in their life: taxes
and death. I have three: taxes, death, and Marshall Orloff discussing my paper. But I do appreciate his
comments. Of the 263 patients in this series, only 15 to 20 of them had extrahepatic portal hypertension,
with most of those having diffuse splanchnic venous thrombosis. Those patients received a
devascularization operation. Six patients with isolated portal vein thrombosis underwent a distal
splenorenal shunt. You are correct that the lowest number of patients operated on per year was during era
three. However, eight operations rather than one were done per year. A number of factors influenced how
many patients underwent surgery. Probably a major one in era three was that I had just moved to
Nebraska and was in the process of developing a referral practice. Most patients were serially followed
by either evaluation in the clinic or by telephone contact. During the late postoperative interval, the
evaluation was incomplete for approximately 15% of patients in whom survival status alone, but not
encephalopathy or rebleeding, was determined. A major reason that so few patients eventually required
transplantation is that many of the transplant candidates were transplanted fairly soon after experiencing
variceal hemorrhage and, therefore, never underwent an operation for their bleeding. In eras 3 and 4, thus
far approximately 15% of patients who were potential transplant candidates have received a transplant;
the remaining patients continue to do quite well after control of their bleeding with a nontransplant
operation.
Professor Alan G. Johnson (Sheffield, England): I am surprised your sclerotherapy did not eliminate or
reduce emergency surgery. In our experience it virtually eliminated it. I wonder whether it is a matter of
the definition of "failed" sclerotherapy. We persist with sclerotherapy two or three times. Other clinicians
say it has failed if it does not work once and then go on to surgery. Were drugs, such as octreotide,
introduced at all during this phase? Gastric varices are a subgroup which behave differently, and should
be analyzed separately. Finally, we are still looking for the first patient we think is appropriate for TIPS.
We cannot find the niche for TIPS because the other treatments we are using are so effective.
Dr. Layton F. Rikkers (Madison, Wisconsin): Experience with sclerotherapy certainly varies from
institution to institution. We do use the drugs, including octreotide and vasopressin, for acute control of
bleeding. Nonselective beta-blockers are used to prevent rebleeding in some patients. We did a
randomized controlled trial of shunt versus sclerotherapy in the past. Sclerotherapy failure, defined as
bleeding to death or the need to perform an operation to prevent bleeding to death, eventually developed
in 40% of patients after a follow-up of approximately 8 years. Often the esophageal varices are
successfully obliterated and recurrent bleeding is from gastric varices in these patients.

Bookmark URL: /das/journal/view/N/10439611?ja=126393&PAGE=1.html&ANCHOR=top&source=HS,MI

April 21, 2000
Conde Petra
Observation of thoracic duct morphology in

portal hypertension by endoscopic ultrasound.
Citation
Bibliographic Data
Abstract
Indexing Data
Observation of thoracic duct morphology
Copyright Notice and Disclaimer in portal hypertension by endoscopic
ultrasound.
Parasher VK - Gastrointest Endosc - 1998 Dec; 48(6): 588-92
Full Text NLM Citation ID:

99069171
Frontmatter
Full Source Title:
PATIENTS AND METHODS Gastrointestinal Endoscopy
Statistical analysis Publication Type:
Discussion Language:
English
References
Author Affiliation:
About the Publication Beebe Medical Center, Lewes, Delaware; Istituto Clinico
Humanitas and Nazionale Tumori, Milan, Italy.
Authors:
Parasher VK; Meroni E; Malesci A; Spinelli P; Tommasini
MA; Markert R; Bhutani MS
Abstract:
BACKGROUND:Thoracic duct dilation has been demonstrated
in portal hypertension and hepatic cirrhosis by
lymphangiography and laparotomy and at autopsy. It is thought
to be secondary to increased hepatic lymph flow and has been
described in the absence of ascites or esophageal varices. The
aim of the present study was to observe thoracic duct
morphology by endoscopic ultrasound in various subsets of
patients with portal hypertension and hepatic cirrhosis and also
to validate existing radiologic/surgical data. METHODS:The
thoracic duct of 33 patients with cirrhosis and portal
hypertension was studied by endoscopic ultrasound. Patients

were divided into four groups: 1, patients with ascites and
esophageal varices; 2, esophageal varices without ascites; 3,
without esophageal varices or ascites; 4, extrahepatic portal
hypertension due to pancreatic malignancy. The thoracic duct
diameter was also measured in 14 control subjects (group 5).
RESULTS:When the thoracic duct diameter for the five groups
was compared with analysis of variance, significance was p <
0.0001; by pairwise comparison, group 1 differed from the
other four groups (p < 0.05). Thoracic duct dilation (5.61 mm)
was seen in group 1 patients, whereas no dilation was present in
groups 2 through 4. Additionally, thoracic duct diameter in 33
portal hypertensive and/or cirrhotic patients was significantly
different from that in the 14 control subjects (p = 0. 003).
CONCLUSION:The thoracic duct can be reliably identified by
EUS in patients with hepatic cirrhosis and portal hypertension.
Dilation of the duct is seen only in patients with hepatic
cirrhosis, ascites, and esophageal varices. No thoracic duct
dilation is present in extrahepatic portal hypertension. Contrary
to existing radiologic/surgical data, thoracic duct dilation is not
seen in all patients with hepatic cirrhosis and portal
hypertension signifying advanced disease. A dilated thoracic
duct by endoscopic ultrasound should be considered yet another
sign of portal hypertension.
Major Subjects:
Hypertension, Portal / Etiology / * Ultrasonography
Liver Cirrhosis / Complications / * Ultrasonography
Thoracic Duct / * Ultrasonography
Ascites / Etiology
Case-Control Studies
Dilatation, Pathologic / Etiology / Ultrasonography
Endosonography
Esophageal and Gastric Varices / Etiology
Female
Human
Male
Middle Age

Gastrointestinal Endoscopy
Volume 48 Number 6 December 1998
Copyright 1998 American Society for Gastrointestinal Endoscopy
ORIGINAL ARTICLES
Observation of thoracic duct morphology in portal hypertension by

endoscopic ultrasound
Vinod K. Parasher MD
Emanuele Meroni MD
Alberto Malesci MD
Pasquale Spinelli MD
Maurizio A. Tommasini MD
Ronald Markert PhD
Manoop S. Bhutani, MD
Received June 25, 1996.

For revision August 30, 1996.
Accepted June 4, 1998.
From the Beebe Medical Center, Lewes, Delaware; Istituto Clinico Humanitas and Nazionale Tumori, Milan, Italy; Wright
State University, Dayton, Ohio.
Reprint requests: Vinod K. Parasher, MD, Director of Endoscopy, Beebe Medical Center, 424 Savannah Rd., Lewes, DE
19958.
Copyright 1998 by the American Society for Gastrointestinal Endoscopy
37/1/92175
Lewes, Delaware, Milan, Italy, Dayton, Ohio
Background:Thoracic duct dilation has been demonstrated in portal hypertension and hepatic cirrhosis by
lymphangiography and laparotomy and at autopsy. It is thought to be secondary to increased hepatic lymph flow
and has been described in the absence of ascites or esophageal varices. The aim of the present study was to observe
thoracic duct morphology by endoscopic ultrasound in various subsets of patients with portal hypertension and
hepatic cirrhosis and also to validate existing radiologic/surgical data.
Methods:The thoracic duct of 33 patients with cirrhosis and portal hypertension was studied by endoscopic
ultrasound. Patients were divided into four groups: 1, patients with ascites and esophageal varices; 2, esophageal

varices without ascites; 3, without esophageal varices or ascites; 4, extrahepatic portal hypertension due to
pancreatic malignancy. The thoracic duct diameter was also measured in 14 control subjects (group 5).
Results:When the thoracic duct diameter for the five groups was compared with analysis of variance, significance
was p < 0.0001; by pairwise comparison, group 1 differed from the other four groups (p < 0.05). Thoracic duct
dilation (5.61 mm) was seen in group 1 patients, whereas no dilation was present in groups 2 through 4.
Additionally, thoracic duct diameter in 33 portal hypertensive and/or cirrhotic patients was significantly different
from that in the 14 control subjects (p = 0.003).
Conclusion:The thoracic duct can be reliably identified by EUS in patients with hepatic cirrhosis and portal
hypertension. Dilation of the duct is seen only in patients with hepatic cirrhosis, ascites, and esophageal varices. No
thoracic duct dilation is present in extrahepatic portal hypertension. Contrary to existing radiologic/surgical data,
thoracic duct dilation is not seen in all patients with hepatic cirrhosis and portal hypertension signifying advanced
disease. A dilated thoracic duct by endoscopic ultrasound should be considered yet another sign of portal
hypertension. (Gastrointest Endosc 1998;48:588-92.)
Increased formation of lymph which contains an abundance of formed elements is characteristic of hepatic cirrhosis with
portal hypertension. These formed elements consist of intact red blood cells and protein. The increase in hepatic lymph
flow is due to the obstruction of the hepatic venules as a result of cirrhosis.[1] [5] The normal flow of hepatic lymph is
toward the hilum, and there the lymphatics traverse the hepatoduodenal and hepatogastric ligaments to join the cisterna
chyli and the thoracic duct. In portal hypertension, the increased lymph flow causes dilation of the hilar lymphatics and the
thoracic duct[1] [3] [6] as shown by lymphangiography [2] and postmortem studies and at laparotomy. [6] It is noteworthy
that the dilation of the thoracic duct has been noted even in the absence of varices and ascites,[2] [6] suggesting that thoracic
duct dilation may be an early sign of portal hypertension.
Unfortunately, there are as yet no noninvasive methods to study the thoracic duct. Endoscopic ultrasound (EUS) is a
sensitive imaging technique that is increasingly being used to study intraluminal and extraluminal disorders of the
gastrointestinal tract. More recently, we have described the normal anatomy of the thoracic duct by EUS, thereby
establishing a semi-invasive method to observe this structure. The thoracic duct was identified as an anechoic 3 to 4 mm
structure that is best seen near the mid-esophagus.[7] The aim of our study was to answer the following questions: (1) Can
the thoracic duct be reliably identified in patients with hepatic cirrhosis? (2) Can EUS validate existing radiologic/surgical
data that indicate that the thoracic duct is dilated in intrahepatic portal hypertension? (3) Is the thoracic duct dilated in all
cases of hepatic cirrhosis, irrespective of the presence of ascites or varices?
PATIENTS AND METHODS

Thirty-three patients with biopsy-proven cirrhosis (etiology included alcohol, hepatitis B virus, and hepatitis C virus) were
studied prospectively. The study was approved by our institutional review board. EUS was performed with the GF UM20
(Olympus America, Inc., Melville, N.Y.) echogastroscope. There were 28 men and 5 women. The ages ranged from 45 to
73 years (mean 58.5 years). Ten patients had ascites and esophageal varices (group 1), 8 had esophageal varices without
ascites (group 2), and 10 had neither ascites nor esophageal varices (group 3). Five patients who had pancreatic
malignancy with varices due to invasion of the portal system were also studied as a group with extrahepatic portal
hypertension (group 4). Ascites was not present in these patients, but all had gastric varices and portal vein obstruction.
Ascites was diagnosed by clinical examination, transcutaneous ultrasound, EUS, or CT. Esophageal varices were
diagnosed by endoscopy and EUS. Patients with active variceal bleeding or a recent history of bleeding (less than 1
month), hepatic encephalopathy, or severe ascites (causing respiratory distress) were excluded from the study.
After examining the stomach, the echoendoscope was withdrawn gradually into the esophagus. The balloon was inflated
with deaerated water and the thoracic duct was examined in a systematic fashion. Thoracic duct diameter was recorded at
its widest part. The diameter of the thoracic duct was also measured in 14 normal control subjects (group 5) undergoing
EUS for other indications. There were 10 men and 4 women ranging in age from 42 to 81 years (mean 53.6 years). The
differentiation between the thoracic duct and varices was made by following the thoracic duct inferiorly and superiorly
over the length of the esophagus. The thoracic duct generally extended longitudinally as a continuous anechoic structure,
whereas periesophageal varices were often multiple, noncontinuous, and serpiginous.

Statistical analysis
Mean thoracic duct diameters among the 5 groups were compared by analysis of variance (ANOVA) and Tukey's multiple
range test for pairwise differences. The t-test was used to compare patients with portal hypertension and/or cirrhosis with
control subjects. Inferences were made at the 0.05 level of significance.
Results
The thoracic duct was seen in all patients as an anechoic thin-walled round structure. The duct was more clearly visualized
in the mid-esophagus where the esophagus was anterior to it, the spine posterior, the azygos vein to its right, and the aorta
to the left side of the duct. A single duct was visualized in all cases. The diameter of the thoracic duct in 14 control
subjects (group 5) ranged from 1.5 to 4.0 mm (mean 2.43 mm, Fig.1).
Fig. 1. EUS demonstrates nondilated thoracic duct in patient with hepatic cirrhosis without ascites or esophageal varices (frequency 7.5
MHz, range 6 cm). Arrowhead, Thoracic duct; large arrow, azygos vein; small arrow, aorta.
Thoracic duct dilation was noted only in patients with ascites and esophageal varices (group 1). Diameter of the thoracic
duct ranged from 5 to 7 mm (mean 5.61 mm) in this group (Fig. 2).
Fig. 2. EUS demonstrates dilated thoracic duct lying between the azygous vein and aorta in a patient with ascites and esophageal
varices (frequency 7.5 MHz, range 4 cm).
The thoracic duct was not dilated in cirrhotic patients with esophageal varices who did not have ascites (group 2). Here the
diameter ranged from 2.1 to 4.0 mm (mean 2.85 mm). Furthermore, the thoracic duct was also not dilated in cirrhotic
patients with neither esophageal varices nor ascites (group 3); the diameter ranged from 1 to 3.5 mm (mean 2.44 mm).
Also, the duct was not dilated in patients with extrahepatic portal hypertension (group 4); the diameter ranged from 2.0 to
3.3 mm (mean 2.58 mm). No complications occurred during EUS.
When the thoracic duct diameter for the five groups (Table 1) was compared with analysis of variance (ANOVA),
significance was p < 0.
Table 1. Thoracic duct diameter in patient groups and control subjects

Group 1 Group 2 Group 3 Group 4 Group 5
TDD TDD TDD TDD TDD
Patients (mm) Patients (mm) Patients (mm) Patients (mm) Patients (mm)
01. 5.0 01. 3.0 01. 1.0 01. 2.0 01. 1.5
02. 5.1 02. 4.0 02. 2.0 02. 2.0 02. 2.0
03. 6.0 03. 2.3 03. 2.3 03. 3.3 03. 3.0
04. 5.4 04. 3.0 04. 3.1 04. 3.1 04. 4.0
05. 7.0 05. 3.0 05. 1.4 05. 2.5 05. 2.1

06. 5.3 06. 2.1 06. 2.4 06. 1.5

07. 5.0 07. 2.4 07. 2.5 07. 3.0
08. 6.2 08. 3.0 08. 3.2 08. 2.4
09. 5.1 09. 3.0 09. 2.0
10. 6.0 10. 3.5 10. 2.0
11. 2.1
12. 2.4
13. 4.0
14. 2.0
Mean 5.61 mm Mean 2.85 mm Mean 2.44 mm Mean 2.58 mm Mean 2.43 mm
95% CI [5.13, 95% CI [2.35, 95% CI [1.86, 95% CI [1.83, 95% CI [1.97,
6.09] 3.35] 3.02] 3.33] 2.89]
Range 5.0-7.0 mm Range 2.1-4.0 mm Range 1-3.5 mm Range 2.0-3.3 mm Range 1.5-4.0 mm
Groups are as follows: 1, cirrhotic patients with ascites and varices; 2, cirrhotic patients with varices
only; 3, cirrhotic patients without varices or ascites; 4, patients with extrahepatic portal hypertension; 5,
represents normal control subjects. TDD, Thoracic duct diameter.
0001. Furthermore, pairwise comparisons with Tukey's multiple range test found that group 1 was significantly different (p
< 0.05) from all other groups. No other pairwise comparisons were significant. When all 33 patients were compared with
control subjects, the mean thoracic duct diameter (3.52 mm) was significantly different from that for control subjects (2.43
mm, p = 0.003).
Discussion
The main function of the lymphatic system is the collection and transport of large molecules consisting of plasma proteins,
particulate matter, tissue debris, bacteria, and foreign substances.[8] The liver lymph originates from the interstitial fluid in
the intercellular spaces of Disse. From here it extends to the limiting plates and then to the portal spaces. Eighty percent of
the hepatic lymph passes through the hilar lymphatics to the cisterna chyli and into the thoracic duct. Under normal
circumstances, half of the lymph flowing through the thoracic duct originates in the liver and half in the gastrointestinal
tract.[9] Hepatic hilar lymphatics are distended in patients with portal hypertension and discharge large volumes of lymph
when transected. The thoracic duct is greatly enlarged, and lymph flow through the thoracic duct is often 10 to 15 times
the normal flow rate.[1] [3] [6] The distended hilar encapsular lymphatics reflect the hepatic venous outflow obstruction and
increased hepatic lymph formation. Decompression of the thoracic duct either externally or by a shunt may ameliorate the
portal pressure, ascites, and esophageal varices.[3] Indeed, a peritoneal venous shunt may be considered as an artificial
thoracic duct.
Thoracic duct dilation has been shown in hepatic cirrhosis at autopsy and laparotomy and by lymphangiography.[2] [6]
Additionally, thoracic duct dilation has also been demonstrated in the setting of hepatic cirrhosis without ascites and also
in cases of hepatic cirrhosis without ascites or esophageal varices. Schieber[2] demonstrated a dilated thoracic duct
(diameter 6 to 9 mm, mean 7.5 mm) in 5 patients who had no ascites or major variceal bleeding. Normal duct diameter by
lymphangiography is reported to be 1 to 4 mm (mean 2.5 mm). The thoracic duct was not only dilated but was also
increased in length, probably because of the increased tortuosity. Dumont and Mulholland[6] showed similar findings in

one patient who had neither ascites nor varices.
We have demonstrated that the thoracic duct can be identified by EUS. It appears as an anechoic structure, 3 to 4 mm in
diameter, best seen close to mid-esophagus. [7] EUS has been used to evaluate patients with portal hypertension; findings
include esophageal and periesophageal varices, gastric and perigastric varices, dilated azygos veins, and small dilated
vessels in the gastric wall in patients with portal hypertensive gastropathy and ascites. [10] [11] This is the first study of
thoracic duct by EUS in patients with portal hypertension.
The current study demonstrates that the thoracic duct can be reliably identified in patients with cirrhosis; furthermore,
thoracic duct dilation is limited to patients who appear to have advanced disease with coexistent ascites and varices.
Thoracic duct dilation was not identified in patients who had no ascites or varices. A dilated thoracic duct in the setting of
hepatic cirrhosis and ascites is yet another sign of portal hypertension identifiable by EUS. Because of the relatively small
sample size in this study, there is a possibility of beta error. The study may lack the statistical power to detect differences
in patients with less severe disease. Because of multicenter nature of the study, interobserver variability may exist.
However, we found group 1 to be different from each of the other four groups. Larger sample sizes might eventually result
in statistically significant differences among the other four groups. If the mean value for groups 1 through 4 remained
about the same, the small differences would not be clinically significant.
Our study confirms the previous radiologic/surgical data indicating that the thoracic duct is dilated in hepatic cirrhosis;
however, it does not confirm that dilation is seen in all cases of hepatic cirrhosis and portal hypertension. The reason for
this discrepancy is not known. It can be hypothesized that lymphangiography may alter lymphatic dynamics by the
injection of contrast under pressure. This may cause spurious duct dilation.[12] It is also known that cannulation of the
thoracic duct so as to block the outflow may cause higher pressure.[8] [13] The length of the thoracic duct was not
determined in our study because the anatomic length cannot be accurately measured by EUS. Periesophageal varices may
cause confusion; however, we found that by following the thoracic duct longitudinally in a dynamic fashion, this problem
could be avoided.
There has been no study of thoracic duct diameter in extrahepatic portal hypertension. We did not find a dilated thoracic
duct in a selected group of patients with extrahepatic portal hypertension; however, none of these patients had ascites.
More patients with other causes of extrahepatic portal hypertension need to be studied before conclusions can be drawn
concerning thoracic duct diameter in this group of patients. However, studies done in patients with presinusoidal
obstruction,[5] particularly in portal vein constriction, have shown that there is no additional increase in thoracic duct
lymph flow and pressure. The increased production of lymph in the small intestine is probably balanced by a decrease in
production of lymph in the liver caused by a reduction in hepatic sinusoidol pressure and decreased sinusoidol perfusion
by portal venous blood.
In conclusion, EUS will reveal a dilated thoracic duct in cirrhotic patients with ascites and esophageal varices. In our
study, the endosonographic thoracic duct diameter in cirrhotic patients without ascites or varices and patients with
extrahepatic portal hypertension was not significantly different from that of normal control subjects. Endosonography
affords a unique method to study the thoracic duct (without manipulating it) and a dilated duct found at EUS may be a sign
of advanced cirrhosis with ascites and varices. Further, studies of the thoracic duct by EUS could result in a better
understanding of pathophysiologic and hemodynamic alterations in portal hypertension.
References
1. Dumont AE, Mulholland JH. Alteration in thoracic duct lymph flow in hepatic cirrhosis; significance in portal hypertension. Ann Surg 1962;156:668-77.
2. Schieber W. Lymphangiographic demonstration of thoracic duct dilation in portal cirrhosis. Surgery 1965;57:522-4.
3. Dumont AE, Witte MH. Contrasting patterns of thoracic duct lymph formation in hepatic cirrhosis. Surg Gynecol Obstet 1966;122:524-8. citation
4. Witte CL, Witte MH, Cole WR, Chung YC, Bliesch VR, Dumant AE. Dual origins of ascites in hepatic cirrhosis. Surg Gynecol Obstet 1969;129:1027-33.
citation
5. Witte MH, Dumont AE, Cole WR, Witte CL, Kinter K. Lymph circulation in hepatic cirrhosis: effect of portacaval shunt. Ann Intern Med 1969;70:303-10.

citation
6. Dumont AE, Mulholland JH. Flow rate and composition of thoracic duct lymph in patients with cirrhosis. N Engl J Med 1960;263:471-4.
7. Parasher VK, Meroni E, Spinelli P. Anatomy of thoracic duct: an endosonographic study. Gastrointest Endosc 1995;42:188-9. full text
8. Hyman C. Physiologic functions of the lymphatic system. Cancer Chemoth Reports 1968;52:25-30.
9. Cain JC, Grindlay JH, Bollman JL, Flock FU, Mann FC. Lymph flow from the liver and thoracic duct. Surg Gynecol Obstet 1947;85:559-62.
10. Caletti GC, Bolondi L, Zani L, Brocchi E, Guizzardi G, Labo G. Detection of portal hypertension and esophageal varices by means of endoscopic
ultrasonography. Scand J Gastroenterol 1986;21:74-7.
11. Caletti GC, Brocchi E. Baraldini M, Ferrari A, Gibilaro M, Barbara L. Assessment of portal hypertension by endoscopic ultrasonography. Gastrointest Endosc
1990;36:S21-7. abstract
12. Wallace S. Dynamics of normal and abnormal lymphatic systems as studied with contrast media. Cancer Chemoth Reports 1968;52:31-58.
13. Lee FC. Some observations on lymph pressures. Am J Physiol 1923;67:498-513.


April 21, 2000
Conde Petra
NATURAL HISTORY OF PORTAL

HYPERTENSION
Additional Article
Consult's full-text literature database. Volume 1 Number 1 May 1997
Full Text
Frontmatter 31
INCIDENCE AND GROWTH OF

VARICES PORTAL HYPERTENSION
RISK FACTORS FOR INITIAL
BLEEDING FROM VARICES
NATURAL HISTORY OF PORTAL
Pathogenesis of Variceal Bleeding
HYPERTENSION
Risk Factors for Bleeding from
Esophageal Varices
Gastric Varices and Risk of

Bleeding
Thomas D. Boyer MD
Portal Gastropathy and Risk of
Bleeding
REBLEEDING FROM From the Division of Digestive Diseases, Emory University School of
Medicine, Atlanta, Georgia
ESOPHAGEAL VARICES
CONCLUSION
A rise in pressure in the portal vein or its tributaries is
termed portal hypertension. A portal-hepatic vein gradient
ACKNOWLEDGMENT (PHVG) of more than 5 mmHg is abnormal, and a patient
with this increased value is believed to have portal
References
hypertension [38] ; however, the presence of an increase in
About the Publication pressure in the portal vein is not always associated with the
complications of portal hypertension (e.g., the development
of varices), ascites, or hepatic encephalopathy. Therefore,
the natural history of portal hypertension is better referred
to as the natural history of one of its complications. This
article reviews the natural history of gastroesophageal
varices and portal hypertensive gastropathy, and the risk
factors associated with bleeding. Recently it has become

more important to understand the natural history of

gastroesophageal varices, as many believe patients with
varices that are at highest risk for bleeding should be
treated prophylactically. [4]
Before proceeding with a discussion of the natural history of
gastroesophageal varices, a few comments on how portal
hypertension progresses are required. Portal hypertension
develops in patients with cirrhosis because of increased
resistance to the passage of blood through the liver and
because of an increase in splanchnic blood flow secondary to
vasodilatation within the splanchnic vascular bed (see
articles by Groszmann and Rockey). Many assume that the
rise in portal venous pressure is progressive and does not
wax and wane. This belief is incorrect, and portal venous
pressure may vary with daily activities; rise and fall with
changes in blood volume; or, in some patients, fall as the
liver disease improves (e.g., exercise may be associated with
an acute
32
rise in portal pressure). [12] Also, changes in blood volume

may raise or lower portal pressure. [28] [22] In patients with
cirrhosis, the increase in pressure, although largely due to
fibrosis and compression of the liver microvasculature by
the regenerative nodules, may also be due to swelling of
hepatocytes and active changes in vascular resistance
secondary to the myofibroblasts present in the cirrhotic
liver (see article by Rockey). Therefore, as the liver disease
improves, so may the portal hypertension, as has been
observed in patients with alcoholic liver disease who have
stopped drinking. [3] The fall in pressure may be associated
with disappearance of the varices. [6]
This work was supported in part by NIH grant DK50680.

Thomas D. Boyer, MD
Emory University School of Medicine
WMB 2101
1639 Pierce Drive
Atlanta, GA 30322

INCIDENCE AND GROWTH OF VARICES

In discussing the natural history of esophageal varices, it is
necessary to review how varices are classified. Several
investigators have adapted the classification system developed
by the Japanese Research Society for Portal Hypertension. [19]
The Japanese system provides four categories (fundamental
color, red color sign, form [size], and location) that quantify the
variables believed to be most important in predicting bleeding
from esophageal varices (Table 1) . Other studies use
measurement of the varices for grading, but the basic
components of each classification system are the same, and
there is good agreement between different observers as to size
of varices and presence or absence of red color signs. [1] [7]
The incidence of varices in patients with cirrhosis is
approximately 50%, with varices found most commonly in
those with poor liver function. [6] [13] [29] Medium to large varices
are present in 20% of patients with cirrhosis. [29] Patients
with other findings suggestive of portal hypertension
TABLE 1 -- ELEMENTS FOR CLASSIFICATION OF
ESOPHAGEAL VARICES
Data from Japanese Research Society for Portal Hypertension:
The general rules for recording endoscopic findings on
esophageal varices. Jpn J Surg 10:84, 1980.
Color of varices
White or blue
Red-color signs
Red wale markings *
Cherry red spots *
Hematocystic spot
Diffuse redness
Form (size)
Straight and disappear with insufflation (i.e., small)
Slightly tortuous and fill less than 1/3 lumen (i.e., medium)
Tortuous and fill > 1/3 lumen (i.e., large)
Location
Above tracheal bifurcation

At or near tracheal bifurcation

Within lower esophagus
*Graded based on extent as 1-3
33
(splenomegaly) appear to be more likely to have varices than

patients who lack manifestations of portal hypertension. [6] The
presence of varices means that the pressure in the portal vein is
increased; however, an increase in portal pressure is not always
associated with the development of varices. Garcia-Tsao and
colleagues [13] measured portal pressure in a group of patients
with cirrhosis and correlated these measurements with
endoscopic findings. Patients with large varices had a
significantly higher PHVG than patients without varices;
however, 9 of the 15 patients lacking varices had a PHVG of
more than 12 mmHg, which is considered the minimal pressure
required to develop varices that are large enough to bleed. [2] [13]
[45] Therefore, the development of portal hypertension is
necessary, but not sufficient, for the development of

gastroesophageal varices. This finding is not surprising as there
are four potential collateral circulations that can develop in a
patient with portal hypertension, and only one involves the
cardioesophageal junction. [45]
The natural progression of esophageal varices has been
examined to a limited extent. Cales et al [6] followed 84 patients
with cirrhosis for a period of 16 months following an index
endoscopy. Forty-one patients had no varices at entrance into
the study whereas 43 had small sized varices. As shown in
Table 2 , 56% of the patients without varices at entrance were
still free of varices at the end of the study whereas 24% had
developed small varices and 20% developed moderate-sized
varices. In contrast, of the 43 patients with small varices, the
varices disappeared in 16%, they remained unchanged in 42%,
and in 42% they became moderate in size. In neither group did
any of the patients develop large varices. The estimated risk of
bleeding during the 2-year period of follow-up for those
patients with no varices was less than 8% whereas it was less
than 24% for those with small varices at the initial endoscopy.
In another report, [29] new varices developed in 225 cirrhotic
patients at a rate of about 8% per year during the first 2 years of
follow-up, and the rate then declined in subsequent years. Large

varices developed in 4% of those without varices and in 25% of
those with small varices during a 6-year period of follow-up.
The lower rate of appearance in the latter study [29] may reflect
differences in the patient population examined and
TABLE 2 -- CHANGES IN ESOPHAGEAL VARICES WITH
TIME
Data from Cales P, Desmorat H, Vinel JP, et al: Incidence of
large esophageal varices in patients with cirrhosis: Application
to prophylaxis of first bleeding. Gut 31:1298, 1990.
Size End of Observation
Absent Small Medium
Variceal Size At No. of
Entry * Patients No. of Patients (%)
0 41 23 (56) 10 (24) 8 (20)
1 43 7 (16) 18 (42) 18 (42)
*Size 0-absent to 3-large.
34
the endoscopic criteria used to judge variceal size. The rate of

development of esophageal varices in primary biliary cirrhosis
(PB) has also been studied. [14] Two-hundred sixty-five patients
with PBC and no varices were followed for a median of 5.6
years; 69% had stage 3 or 4 disease on liver biopsy. Esophageal
varices developed in 31%, with the highest incidence found in
those with stage 4 histology. The varices appeared at a constant
rate of 4% per year for the group as a whole and at a rate of
12% per year in those with stage 4 disease. Based on the
previously mentioned studies, the risk of developing new
varices in a patient with cirrhosis is 4% to 12% each year.
Once present, the risk of the varices becoming large is 4% to
10% each year.
In the previously mentioned studies, guidelines for the
frequency of endoscopy needed to identify patients who
develop large varices that may bleed have been proposed. Cales
and colleagues [6] suggested that endoscopy be performed every
other year in cirrhotic patients without varices and yearly in

those with small varices. Pagliaro et al, [29] in contrast,
suggested that endoscopy be performed every other year in
cirrhotics with small varices, and at less frequent intervals in
patients without varices. The purpose in identifying patients
with varices is to treat them with beta-blockers and other drugs
in the hope of preventing the varices from bleeding. Nine to 11
patients with large varices need to be treated to prevent one
from having an initial bleed. [30] If 20% of patients with
cirrhosis have high-risk varices, then it will require 100
screening endoscopies to prevent 1 to 2 patients from bleeding
from their varices. The numbers are similar for surveillance
endoscopy. For example, performing endoscopy every other
year on 100 patients with cirrhosis who lack or have small
varices at the initial endoscopy will identify 4 to 10
individuals who have developed varices that are at risk for
bleeding. If those 4 to 10 individuals are treated
prophylactically, then at least 100 endoscopies will be required
to prevent one patient from bleeding. Based on current data, it is
difficult to recommend routine surveillance endoscopy in
patients with cirrhosis who lack other manifestations of portal
hypertension. If endoscopy is performed and no varices are
found, then follow-up endoscopy should not be performed for 3
to 5 years, if ever. If small varices are present, then a repeat
endoscopy is probably warranted in 2 to 3 years. If the varices
remain small at the second endoscopy, further surveillance is
not indicated. These recommendations are based on very
limited data, and a careful cost-benefit analysis of
well-designed prospective studies is required before routine
surveillance endoscopy can be recommended for patients with
cirrhosis.
The greatest risk of bleeding appears to be during the first 1 to 2
years following identification of the varices. [27] The author
believes that this finding makes patients who are awaiting liver
transplant ideal candidates for endoscopic screening for
high-risk varices. Most patients being evaluated for liver
transplant have advanced liver disease and have suffered from
complications of portal hypertension; therefore, they are more
likely to have varices. As they have a finite period to wait for
their liver transplant (usually less than 2 years), only a single
screening
35

endoscopy is required. Also, these patients usually have

severely compromised liver function, and bleeding may be
poorly tolerated, adding to the benefit of finding and treating
the varices before they bleed.
RISK FACTORS FOR INITIAL BLEEDING

FROM VARICES
Bleeding from varices occurs commonly once the varices have
developed. The frequency of bleeding in untreated patients with
varices who served as controls in studies in which the impact of
prophylactic portacaval shunts, drugs, and sclerotherapy on
variceal bleeding was examined was 29.6% 11.5%, with an
associated high mortality (Table 3) . In most of the studies the
greatest risk for bleeding was within the first 1 to 2 years
following identification of the varices. [27] Although a number
of factors appear to be associated with bleeding risk, the
endoscopic appearance of the varices was the variable that most
consistently predicted bleeding. The following discussion
focuses on the endoscopic features of the varices that are used
to predict bleeding.
There is a minimal level of portal hypertension required for

varices to become large enough to bleed. That level is 12
mmHg or more above free hepatic vein or Inferior Vena Cava
pressure, and has been shown in a number of studies to be a
constant finding in patients who have bled from varices [2] [13]
[45] ; however, many patients with pressures above this threshold
do not bleed and may not develop varices. [13] It is not surprising
that many patients do not develop varices in response to portal
hypertension. The collateral circulation through the
cardioesophageal junction is only one of four potential routes
that blood may take to
TABLE 3 -- RISK OF FIRST BLEED AND MORTALITY IN
UNTREATED PATIENTS WITH VARICES
No. of Bleeding Mortality
Author (ref #) Patients (%) (%) *
Resnick [36] 45 27 38
Jackso [19] 58 19 28
Conn [8] 31 29 32
Pascal [31] 112 27 36

Poynard [34] 303 35 32

VA [44] 138 17 17
DeFranchis [11] 51 33 50
Koch [21] 30 30 33
Sauerbrunch [43] 65 37 46
Santangelo [39] 46 15 24
Witzel [46] 53 57 55
Mean SEM
29.6 11.5
35.3 11.3
* Period of observation varied from 2 to 8 years.
36
circumvent the cirrhotic liver. [2] Therefore, many cirrhotic

patients may have significant blood flow through the umbilical
or retroperitoneal collaterals and little flow through their
gastroesophageal varices. This latter group lacks
gastroesophageal varices despite significant portal hypertension.
Large varices are clearly more likely to bleed than small
varices. [23] [27] The current belief is that bleeding from varices
occurs when the wall of the varix ruptures. Because varices are
elastic structures, the risk of rupture is related to wall tension.
[33] Wall tension (T) is a function of intraluminal pressure (P),
vessel radius (R), and wall thickness (W), as defined by the

following equation:
T = P R/W
The wall thickness resists the expanding force of P R, and
when the expanding forces become too great, the wall ruptures
and bleeding occurs. Based on this paradigm, small
thick-walled vessels are less likely to bleed than are large
thin-walled vessels. This theory also helps to explain why
varices are more likely to bleed in the distal compared to the
more proximal esophagus. The submucosal veins are more

superficial in the distal esophagus; therefore, it is only the

vessel wall itself that resists the expansion of the vessel. In the
more proximal esophagus the vessels are deeper, and in addition
to the vessel wall, the surrounding tissues lend support and help
resist the expanding forces. [26] Similarly, very large collaterals
within the retroperitoneum do not rupture and bleed, probably
because their walls are inherently thicker and the entire vessel is
surrounded by other tissues.
Risk Factors for Bleeding from Esophageal Varices
In addition to the endoscopic appearance of varices, a number

of other variables have also been examined as predictors of the
first bleed from esophageal varices. In a large series from Italy,
321 patients with cirrhosis and esophageal varices and no
history of bleeding were followed prospectively for a median of
23 months. [27] During that period of observation, 85 patients
(26.5%) bled from their esophageal varices. Univariate analysis
found five clinical findings (Child's class, ascites, bilirubin,
albumin, prothrombin time) and five endoscopic findings (red
wale markings, variceal size, cherry-red spots, location,
hematocystic spots) that correlated significantly with risk of
bleeding. By multivariate analysis, Child's class, size of varices,
and the presence of red wale markings were independent
predictors of bleeding (Table 4) . An index was created using
these three variables, and the equation is as follows:
Nothern Italian Endoscopic Club (NIEC) index = [0.645 C] +
(0.4365 F) + [0.3193 R] 10
where C, F, and R are Child's classification, size of varices, and
red wale markings, respectively. A patient with an NIEC index
of less than 20.0
37
TABLE 4 -- FACTORS PREDICTIVE OF BLEEDING FROM

ESOPHAGEAL VARICES
Frequency of Bleeding (%)
NIEC [27] ZOLI [48]
Child's class
A 17 14.6
B 31.1 25.6

C 38.9 17
Variceal size
Small 18.1 4.8
Medium 28.6 16
Large 48.9 52.8
Red wale
Absent 19.1
Moderate 32.9
Severe 39.3
Cherry red spots Red color
signs
Absent 23
Mild 32 10.3
Moderate 40 31.1
Severe 55 51.2
NIEC = The North Italian Endoscopic Club for the Study
and Treatment of Esophageal Varices.
had a 1-year probability of bleeding of 2% whereas a patient

with an index of 35 to 40 had a probability of bleeding of
40%. A second study, also from Italy, prospectively
followed 344 patients with cirrhosis and varices that had not
bled. [48] Their findings were similar to the previous study in
that by univariate analysis, Child-Pugh score, location of
varices, their size, and presence of red color signs also
correlated with risk of bleeding (Table 4) . By multivariate
analysis, only size of varices, gastropathy, and presence of
gastric varices were independent predictors of bleeding. They
calculated a prognostic index (PI) as follows:
PI = (size 0.0395) + (gastropathy 0.878) + (gastric varices
0.705)
Patients with a PI of less than 2.56 had a probability of bleeding
of 0 to 20% during 1 year of follow-up whereas those with a PI
of greater than 4.51 had an 81% to 100% probability of
bleeding during the first year. Both of these reports demonstrate
the value of prospective studies in defining the natural history

of esophageal varices, and can be used to establish treatment

guidelines. Patients with large varices that also have red color
signs are clearly at a high risk for bleeding in the near future,
and become candidates for prophylactic therapy with drugs such
as beta-blockers and nitrates. Patients with small varices are at
little risk of bleeding, and therapy in this group of patients
clearly is not warranted.
Gastric Varices and Risk of Bleeding
The prevalence of gastric varices in patients with cirrhosis has

varied widely, and appears to be higher in patients presenting
with
38
bleeding compared with those whose varices have never bled.

[41] The frequency of gastric varices in 344 patients with
cirrhosis who had undergone endoscopy was 6% in one study

[48] and 20% in another. [41] Some of the variation in frequency
reflects difficulty in deciding endoscopically whether or not the

patient has gastric varices. As with esophageal varices, a
number of endoscopic systems of classification for gastric
varices have been developed. The grading system developed by
Sarin and colleagues [41] provides most of the elements needed
to estimate risk of bleeding. Gastric varices can be divided into
two groups: gastroesophageal varices (GOVs) and isolated
gastric varices (IGVs). GOVs are esophageal varices that
extend across the cardioesophageal junction, and are termed
GOV1 if they extend 2 to 5 cm into the stomach, and GOV2 if
they extend into the fundus. GOV1 accounts for 75% of the
gastric varices whereas GOV2 represent 21% of all gastric
varices. Type 1 IGVs are located in the fundus of the stomach,
do not extend to the cardia, and account for only 1.6% of gastric
varices, whereas type 2 IGVs are located anywhere else in the
stomach (body, antrum, or pylorus), and are slightly more
common ( 4%) compared to IGV1. [41] This system of
classification is similar to those proposed by others. [40]
Although the finding of gastric varices in the absence of
esophageal varices is believed to be indicative of splenic vein
thrombosis, most of the patients have cirrhosis and a patent
splenic vein. [24] The impact of sclerotherapy on the
development of gastric varices also deserves mention. A group
of 227 patients who underwent sclerotherapy for esophageal
varices was followed prospectively [47] ; 138 patients had gastric
varices at the initial endoscopy ( 50% GOV1) and 72% of the

gastric varices where gone at the end of treatment; 89 patients
did not have gastric varices at the beginning of treatment and
12% developed gastric varices following treatment of their
esophageal varices. Hence, the development of gastric varices
following sclerosis of esophageal varices appears to be
relatively infrequent. It is unknown whether the gastric varices
that develop following sclerotherapy have the same or a
different risk of bleeding compared to primary gastric varices.
Sarin and colleagues [41] examined the frequency of bleeding
from different types of gastric varices, as shown in Table 5 . It
is difficult to
TABLE 5 -- RISK OF BLEEDING FROM GASTRIC
VARICES
Data from Sarin SK, Lahoti D, Saxena SP, et al: Prevalence,
classification and natural history of gastric varices: A
long-term follow-up study in 568 portal hypertension patients.
Hepatology 16:1343, 1992.
Frequency of Bleeding
Type of Varix Number (%)
GOV1 85 12
GOV2 31 55
IGV1 9 78
IGV2 22 9
Total 147 24.5
See text for abbreviations.
39
be certain of the length of follow-up, but the study was

performed over a 5-year period. Approximately one quarter of
the patients with gastric varices bled during the period of
observation. This is not dissimilar to the frequency of bleeding
from esophageal varices ( see Table 3) , although in some series
the risk of gastric varices bleeding appears to be less than the
risk for esophageal varices. Bleeding from gastric varices
tended to be more severe, however. [41] The greatest risk of
bleeding was observed in those with GOV2 and IGV1;
However, one third of the bleeding events were in patients with

GOV1. In a preliminary study, the risk of bleeding from gastric

fundal varices (GOV2 and IGV1) was examined in 117 patients
with cirrhosis [20] ; bleeding was observed in 34 patients (29%)
during a 23-month period of follow-up. The cumulative risk at
1, 3, and 5 years was 16%, 36%, and 44%, respectively. The
size of the varices, presence of red spots, and poor Child's status
were all independent predictors of bleeding. Both of these
studies suggest that fundal varices have a significant risk of
bleeding during a relatively short period of observation, and that
these patients are candidates for decompressive (surgical shunts
versus transjugular intrahepatic portosystemic shunt (TIPS)
treatment once bleeding has occurred. The number of patients
with fundal varices that have been followed for a significant
period of time is too small to recommend prophylactic
treatment, especially as the bleeding rates are quite different in
the various studies.
Portal Gastropathy and Risk of Bleeding
Portal hypertensive gastropathy is a frequent endoscopic finding

( 25%-75%) in patients with cirrhosis. There are two types of
gastropathy: portal hypertensive gastropathy (PHG) and gastric
antral vascular ectasia (GAVE). [16] [25] [32] PHG consists of a
mosaic mucosal pattern with (severe) or without (mild) red or
brown spots, and is present in the gastric fundus and body of the
stomach. [25] [32] GAVE consists of linear red stripes converging
on the pylorus (watermelon stomach) or diffuse red spots
limited to the antrum. [16] [32] The histologic appearance of
GAVE differs from PHG, and the severity of bleeding appears
to be greater in patients with GAVE. [32] In addition, GAVE can
occur in noncirrhotic patients, suggesting this lesion is not due
to vascular congestion of the antral mucosa; however, if GAVE
is not due to vascular congestion of the mucosa, then it should
be a problem in patients with cirrhosis who have undergone
portosystemic shunts. GAVE has not been reported to be a
cause for bleeding in the patients following portosystemic
shunts, suggesting that portal hypertension plays a role in its
pathogenesis. This discussion is limited to PHG but in some of
the cited works, GAVE and PHG were not separated as distinct
entities, which complicates the analysis of these studies.
In one series, PHG was found in 58 of 212 patients with
cirrhosis [9] ; the lesion was mild in 45 patients and severe in 13.
During the 46-month period of follow-up, an additional 65
individuals developed mild

40
gastropathy and 7 developed severe PHG. The likelihood of

developing gastropathy was much greater in those undergoing
sclerotherapy (49%) compared to those receiving no treatment
(26%), which is similar to other reports. [42] Primignani and
colleagues [35] in a preliminary report followed 298 patients with
PHG for a mean of 18 months; Figure 1 shows the natural
history of PHG in this group of patients. Of patients with no
gastropathy at the initial endoscopy, 48% developed PHG; 85
patients had mild gastropathy at the initial endoscopy; and in
18% it disappeared whereas in 39% it worsened; 152 patients
had severe gastropathy, and the endoscopic appearance
improved in 47%. The changes in PHG could not be correlated
with treatment. These studies suggest that PHG can improve or
worsen in patients with cirrhosis, but in most patients the
disease tends to remain the same or become progressively
worse.
Bleeding is common in patients with PHG, and may be caused
by the gastropathy or coincident varices. Twenty bleeding
episodes ascribed to the gastropathy were observed in 110
patients with mild disease whereas 24 episodes of bleeding
were observed in 20 patients with severe disease. [9] Bleeding
tends to be occult or chronic with melena, although active upper
gastrointestinal hemorrhage may occur infrequently. Patients
with active upper gastrointestinal bleeding are more likely to be
bleeding from varices or GAVE. Patients with severe PHG that
has not bled may be candidates for preventive therapy.
REBLEEDING FROM ESOPHAGEAL

VARICES
The natural history of rebleeding from esophageal varices has
been examined in a number of controlled trials in which active
treatment
Figure 1. Changes in portal hypertensive gastropathy

with time. ( Data from Primignani M et al: Portal
hypertensive gastropathy [PHG] in liver cirrhosis:
Natural history: A multicenter study of the New
Italian Endoscopic Club [NIEC]. Gastroenterology
110:A1299, 1996.)
41
(shunts, beta-blockers, sclerotherapy) was compared to no

therapy. In the studies using beta-blockers and sclerotherapy, it

is important to note that during the acute episode of bleeding
the participants may have received sclerotherapy to stop the
hemorrhage, and then were left untreated. In the shunt studies
the patients were truly untreated as beta-blocker therapy and
sclerotherapy were not being used at the time these studies were
performed; however, some control patients underwent surgery,
which may have had an impact on the results. For example, in
the study of Reynolds et al, [37] 16% of the controls underwent
surgery and 95% rebled whereas in another study, [17] 34% were
lost to surgery and only 65% rebled. The risk of rebleeding is
greatest initially following the index bleed and then decreases
with time. [15] In one series, 194 cirrhotic patients were followed
for 25 days. [5] During that time period 97 patients (50%) rebled.
Of those rebleeding, 48.4% bled within 24 hours of the index
bleed 21.6% in the next 24 hours, 14.4% on day three, and
11.3% on day 4 or 5. Thus, 70% of the patients who rebled in
the first 25 days did so within 48 hours of the index bleed. For
preventative therapy to be truly effective, therefore, it must be
given within 2 to 3 days of the index bleed.
Once patients have bled from esophageal varices, they are likely
to bleed again. As shown in Table 6 , two thirds of patients
will suffer at least one additional episode of bleeding. In one
series, 98% of the patients rebled during the period of
observation. [37] Rebleeding appears to be more common in the
Child's C class patient, but patients with preserved hepatic
function also are clearly at risk. In addition, rebleeding is
associated with significant mortality. Mortality is higher in
those with poor liver function (Table 6) .
CONCLUSION
There is an increasingly better understanding of the natural
history of gastroesophageal varices and PHG. This improved
understanding is a direct result of prospective controlled trials
in which large numbers of untreated patients were followed for
significant periods of time. Unfortunately, it has been difficult
to apply this knowledge in a cost-effective manner to the
management of many of the patients. Although
TABLE 6 -- RISK OF REBLEEDING FROM ESOPHAGEAL
VARICES
Data from D'Amico G, Pagliago L, Bosch J, et al: The
treatment of portal hypertension: A meta-analytic review.

Death
Active No. of Child Rebleeding Controls
Therapy Patients C (%) Controls (%) (%)
PCS 195 NR 76.5 14.6 56.3 10.1
beta-Blockers 371 14.5 63.3 12.7 24.2 11.2
17.8
Sclerotherapy 559 41.1 65.8 11.3 55 16
14.8
PCS = Portacaval shunt; NR = not reported.
42
prophylactic therapy is clearly warranted for patients with

high-risk varices, how those patients are identified has not been
resolved. Nor is it certain how to follow patients with known
cirrhosis in the hopes of finding large varices and preventing
their bleeding. Further studies in this area are clearly warranted.
ACKNOWLEDGMENT
The author appreciates the skilled secretarial assistance of Mr.
L.T. Tucker.
References
1. BendtsenF, Skovgaard LT, Sorensen TIA, et al: Agreement among

multiple observers on endoscopic diagnosis of esophageal varices before
bleeding. Hepatology 11:341, 1990
2. Boyer TD, Henderson JM: Portal hypertension and bleeding esophageal

varices. In Zakim D, Boyer TD (eds): Hepatology: A Textbook of Liver
Disease, ed 3. Philadelphia, WB Saunders, 1996, p 720
3. Boyer
TD, Triger DR, Horisawa M, et al: Direct transhepatic
measurement of portal vein pressure using a thin needle: Comparison with
wedged hepatic vein pressure. Gastroenterology 2:584, 1977
4. Burroughs AK, D'Heygere F, McIntyre N: Pitfalls in studies of

prophylactic therapy for variceal bleeding in cirrhotics. Hepatology
6:1407, 1986
5. Burroughs AK, Mezzanotte G, Phillips A, et al: Cirrhotics with variceal

hemorrhage: The importance of the time interval between admission and
the start of analysis for survival and rebleeding rates. Hepatology 9:801,

1989
6. Cales P, Desmorat H, Vinel JP, et al: Incidence of large oesophageal

varices in patients with cirrhosis: Application to prophylaxis of first
bleeding. Gut 31:1298, 1990
7. CalesP, Zabotto B, Meskens C, et al: Gastroesophageal endoscopic

features in cirrhosis: Observer variability, interassociations and
relationship to hepatic dysfunction. Gastroenterology 8:156, 1990
8. Conn HO, Lindenmuth WW: Prophylactic portacaval anastomosis in

cirrhotic patients with esophageal varices: A progress report of a
continuing study. N Engl J Med 272:1255, 1965
9. D'Amico G, Montalbano L, Traina M, et al: Natural history of

congestive gastropathy in cirrhosis. Gastroenterology 99:1558, 1990
10. D'AmicoG, Pagliago L, Bosch J: The treatment of portal hypertension:

A meta-analytic review. Hepatology 22:332, 1995
11. DeFranchis R, Primignani M, Arcidiacono PG, et al: Prophylactic

sclerotherapy in high-risk cirrhotics selected by endoscopic criteria: A
multicenter randomized controlled trial. Gastroenterology 101:1087, 1991
12. Garcia-Pagan JC, Santos C, Barbera JA, et al: Physical exercise

increases portal pressure in patients with cirrhosis and portal hypertension.
Gastroenterology 111:1300, 1996
13. Garcia-Tsao G, Groszmann RJ, Fisher RL, et al: Portal pressure,

presence of gastroesophageal varices and variceal bleeding. Hepatology
5:419, 1985
14. Gores GJ, Wiesner RH, Dickson ER, et al: Prospective evaluation of
esophageal varices in primary biliary cirrhosis: Development, natural
history and influence on survival. Gastroenterology 96:1552, 1989
15. GrahamDY, Smith JL: The course of patients after variceal

hemorrhage. Gastroenterology 80:800, 1981
16. Jabbari
M, Cherry R, Lough J, et al: Gastric antral vascular ectasia: The
watermelon stomach. Gastroenterology 87:1165, 1984
17. JacksonFC, Perrin EB, Felix RW, et al: A clinical investigation of the
portacaval shunt: Survival analysis of the therapeutic operation. Ann Surg
174:672, 1971
43
18. JacksonFC, Perrin EB, Smith AG, et al: A clinical investigation of the
portacaval shunt: II. Survival analysis of the prophylactic operation. Am J
Surg 115:22, 1968

19. JapaneseResearch Society for Portal Hypertension: The general rules

for recording endoscopic findings on esophageal varices. Jpn J Surg 10:84,
1980
20. Kim T, Shijo H, Kokawa H, et al: Risk factors for hemorrhage from
gastric fundal varices. Hepatology 24:209A, 1996
21. Koch H, Henning H, Grimm H, et al: Prophylactic sclerosing of

esophageal varices: Results of a prospective controlled study. Endoscopy
18:40, 1986
22. Kravetz
D, Sikuler E, Groszmann RJ: Splanchnic and systemic
hemodynamics in portal hypertensive rats during hemorrhage and blood
volume restitution. Gastroenterology 90:1232, 1986
23. Lebrec D, De Fleury P, Rueff B, et al: Portal hypertension, size of

esophageal varices and risk of gastrointestinal bleeding in alcoholic
cirrhosis. Gastroenterology 79:1139, 1980
24. LevineMS, Kieu K, Rubesin SE, et al: Isolated gastric varices: Splenic
vein obstruction or portal hypertension? Gastrointest Radiol 15:188, 1990
25. McCormack TT, Sims J, Eyre-Brook I, et al: Gastric lesions in portal

hypertension inflammatory gastritis or congestive gastropathy. Gut
26:1226, 1985
26. Noda T: Angioarchitectural study of esophageal varices with special

reference to variceal rupture. Virchows Arch [A] 404:381, 1984
27. The North Italian Endoscopic Club for the Study and Treatment of
Esophageal Varices: Prediction of the first variceal hemorrhage in patients
with cirrhosis of the liver and esophageal varices. N Engl J Med 319:983,
1988
28. Okumura H, Aramaki T, Katsuta Y, et al: Reduction in hepatic venous

pressure gradient as a consequence of volume contraction due to chronic
administration of spironolactone in patients with cirrhosis and no ascites.
Am J Gastroenterol 86:46, 1991
29. PagliaroL, D'Amico G, Pasta F, et al: Portal hypertension in cirrhosis:

Natural history. In Bosch J, Groszmann RJ (eds): Portal Hypertension:
Pathophysiology and Treatment. London, Blackwell Scientific, 1994, p 72
30. PagliaroL, D'Amico G, Sorensen TIA, et al: Prevention of first

bleeding in cirrhosis: A meta-analysis of randomized trials of nonsurgical
treatment. Ann Intern Med 117:59, 1992
31. PascalJ-P, Cales P, Multicenter Study Group: Propranolol in the

prevention of first upper gastrointestinal tract hemorrhage in patients with
cirrhosis of the liver and esophageal varices. N Engl J Med 317:856, 1987
32. PayenJ-L, Cales P, Voigt J-J, et al: Severe portal hypertensive

gastropathy and antral vascular ectasia are distinct entities in patients with
cirrhosis. Gastroenterology 108:138, 1995
33. Polio
J, Groszmann RJ: Hemodynamic factors involved in the
development and rupture of esophageal varices: A pathophysiologic
approach to treatment. Semin Liver Dis 6:318, 1986
34. Poynard T, Cales P, Pasta L, et al: Beta-adrenergic-antagonist drugs in

the prevention of gastrointestinal bleeding in patients with cirrhosis and
esophageal varices. N Engl J Med 324:1532, 1991
35. PrimignaniM, Carpinelli L, Preatoni P, et al: Portal hypertensive

gastropathy (PHG) in liver cirrhosis: Natural history: A multicenter study
of the New Italian Endoscopic Club (NIEC). Gastroenterology 110:A1299,
1996
36. ResnickRH, Chalmers TC, Ishihara AM, et al: A controlled study of

prophylactic portacaval shunt: A final report. Ann Intern Med 70:675,
1969
37. ReynoldsTB, Donovan AJ, Mikkelsen WP, et al: Results of a 12-year

randomized trial of portacaval shunt in patients with alcoholic liver disease
and bleeding varices. Gastroenterology 80:1005, 1981
38. Reynolds
TB, Redeker AG, Geller HM: Wedged hepatic vein pressure.
Am J Med 22:341, 1957
39. SantangeloWC, Dueno MI, Estes BL, et al: Prophylactic sclerotherapy

of large esophageal varices. N Engl J Med 318:814, 1988
40. SarinSK: Diagnostic issues: Portal hypertensive gastropathy and gastric

varices. In De Franchis R (ed): Portal Hypertension, vol 2. London,
Blackwell Scientific, 1996, p 30
44
41. Sarin SK, Lahoti D, Saxena SP, et al: Prevalence, classification and
natural history of gastric varices: A long-term follow-up study in 568
portal hypertension patients. Hepatology 16:1343, 1992
42. Sarin
SK, Sreenivas DV, Lahoti D, et al: Factors influencing
development of portal hypertensive gastropathy in patients with portal
hypertension. Gastroenterology 102:994, 1992
43. Sauerbruch T, Wotzka R, Kopcke W, et al: Prophylactic sclerotherapy

before the first episode of variceal hemorrhage in patients with cirrhosis. N
Engl J Med 319:8, 1988
44. The Veterans Affairs Cooperative Variceal Sclerotherapy Group:

Prophylactic sclerotherapy for esophageal varices in men with alcoholic
liver disease. N Engl J Med 324:1779, 1991

45. Viallet
A, Marleau D, Huet M, et al: Hemodynamic evaluation of
patients with intrahepatic portal hypertension. Gastroenterology 69:1297,
1975
46. Witzel
L, Wolbergs E, Merk H: Prophylactic endoscopic sclerotherapy
of oesophageal varices. Lancet I:773, 1985
47. ZanasiG, Rossi A, Grosso C, et al: The effect of endoscopic

sclerotherapy of esophageal varices on the development of gastric varices.
Endoscopy 28:234, 1996
48. ZoliM, Merkel C, Magalotti D, et al: Evaluation of a new endoscopic

index to predict first bleeding from the upper gastrointestinal tract in
patients with cirrhosis. Hepatology 24:1047, 1996



Volume 1 Number 1 May 1997
31
PORTAL HYPERTENSION
NATURAL HISTORY OF PORTAL HYPERTENSION
Thomas D. Boyer MD
From the Division of Digestive Diseases, Emory University School of Medicine, Atlanta, Georgia
A rise in pressure in the portal vein or its tributaries is termed portal hypertension. A
portal-hepatic vein gradient (PHVG) of more than 5 mmHg is abnormal, and a patient with this
increased value is believed to have portal hypertension [38] ; however, the presence of an increase in
pressure in the portal vein is not always associated with the complications of portal hypertension
(e.g., the development of varices), ascites, or hepatic encephalopathy. Therefore, the natural
history of portal hypertension is better referred to as the natural history of one of its
complications. This article reviews the natural history of gastroesophageal varices and portal
hypertensive gastropathy, and the risk factors associated with bleeding. Recently it has become
more important to understand the natural history of gastroesophageal varices, as many believe
patients with varices that are at highest risk for bleeding should be treated prophylactically. [4]
Before proceeding with a discussion of the natural history of gastroesophageal varices, a few
comments on how portal hypertension progresses are required. Portal hypertension develops in
patients with cirrhosis because of increased resistance to the passage of blood through the liver and
because of an increase in splanchnic blood flow secondary to vasodilatation within the splanchnic
vascular bed (see articles by Groszmann and Rockey). Many assume that the rise in portal venous
pressure is progressive and does not wax and wane. This belief is incorrect, and portal venous
pressure may vary with daily activities; rise and fall with changes in blood volume; or, in some
patients, fall as the liver disease improves (e.g., exercise may be associated with an acute
32
rise in portal pressure). [12] Also, changes in blood volume may raise or lower portal pressure. [28]
[22] In patients with cirrhosis, the increase in pressure, although largely due to fibrosis and
compression of the liver microvasculature by the regenerative nodules, may also be due to swelling

of hepatocytes and active changes in vascular resistance secondary to the myofibroblasts present in
the cirrhotic liver (see article by Rockey). Therefore, as the liver disease improves, so may the
portal hypertension, as has been observed in patients with alcoholic liver disease who have stopped
drinking. [3] The fall in pressure may be associated with disappearance of the varices. [6]
This work was supported in part by NIH grant DK50680.

Thomas D. Boyer, MD
Emory University School of Medicine
WMB 2101
1639 Pierce Drive
Atlanta, GA 30322
INCIDENCE AND GROWTH OF VARICES

In discussing the natural history of esophageal varices, it is necessary to review how varices are
classified. Several investigators have adapted the classification system developed by the Japanese
Research Society for Portal Hypertension. [19] The Japanese system provides four categories
(fundamental color, red color sign, form [size], and location) that quantify the variables believed to be
most important in predicting bleeding from esophageal varices (Table 1) . Other studies use measurement
of the varices for grading, but the basic components of each classification system are the same, and there
is good agreement between different observers as to size of varices and presence or absence of red color
signs. [1] [7]
The incidence of varices in patients with cirrhosis is approximately 50%, with varices found most
commonly in those with poor liver function. [6] [13] [29] Medium to large varices are present in 20% of
patients with cirrhosis. [29] Patients with other findings suggestive of portal hypertension
TABLE 1 -- ELEMENTS FOR CLASSIFICATION OF ESOPHAGEAL VARICES
Data from Japanese Research Society for Portal Hypertension: The general rules for recording
endoscopic findings on esophageal varices. Jpn J Surg 10:84, 1980.
Color of varices
White or blue
Red-color signs
Red wale markings *
Cherry red spots *
Hematocystic spot
Diffuse redness
Form (size)
Straight and disappear with insufflation (i.e., small)
Slightly tortuous and fill less than 1/3 lumen (i.e., medium)

Tortuous and fill > 1/3 lumen (i.e., large)

Location
Above tracheal bifurcation
At or near tracheal bifurcation
Within lower esophagus
*Graded based on extent as 1-3
33
(splenomegaly) appear to be more likely to have varices than patients who lack manifestations of portal
hypertension. [6] The presence of varices means that the pressure in the portal vein is increased; however,
an increase in portal pressure is not always associated with the development of varices. Garcia-Tsao and
colleagues [13] measured portal pressure in a group of patients with cirrhosis and correlated these
measurements with endoscopic findings. Patients with large varices had a significantly higher PHVG
than patients without varices; however, 9 of the 15 patients lacking varices had a PHVG of more than 12
mmHg, which is considered the minimal pressure required to develop varices that are large enough to
bleed. [2] [13] [45] Therefore, the development of portal hypertension is necessary, but not sufficient, for the
development of gastroesophageal varices. This finding is not surprising as there are four potential
collateral circulations that can develop in a patient with portal hypertension, and only one involves the
cardioesophageal junction. [45]
The natural progression of esophageal varices has been examined to a limited extent. Cales et al [6]
followed 84 patients with cirrhosis for a period of 16 months following an index endoscopy. Forty-one
patients had no varices at entrance into the study whereas 43 had small sized varices. As shown in Table
2 , 56% of the patients without varices at entrance were still free of varices at the end of the study
whereas 24% had developed small varices and 20% developed moderate-sized varices. In contrast, of the
43 patients with small varices, the varices disappeared in 16%, they remained unchanged in 42%, and in
42% they became moderate in size. In neither group did any of the patients develop large varices. The
estimated risk of bleeding during the 2-year period of follow-up for those patients with no varices was
less than 8% whereas it was less than 24% for those with small varices at the initial endoscopy. In
another report, [29] new varices developed in 225 cirrhotic patients at a rate of about 8% per year during
the first 2 years of follow-up, and the rate then declined in subsequent years. Large varices developed in
4% of those without varices and in 25% of those with small varices during a 6-year period of follow-up.
The lower rate of appearance in the latter study [29] may reflect differences in the patient population
examined and
TABLE 2 -- CHANGES IN ESOPHAGEAL VARICES WITH TIME
Data from Cales P, Desmorat H, Vinel JP, et al: Incidence of large esophageal varices in patients with
cirrhosis: Application to prophylaxis of first bleeding. Gut 31:1298, 1990.

Size End of Observation

Absent Small Medium
Variceal Size At Entry * No. of Patients No. of Patients (%)
0 41 23 (56) 10 (24) 8 (20)
1 43 7 (16) 18 (42) 18 (42)
*Size 0-absent to 3-large.
34
the endoscopic criteria used to judge variceal size. The rate of development of esophageal varices in
primary biliary cirrhosis (PB) has also been studied. [14] Two-hundred sixty-five patients with PBC and
no varices were followed for a median of 5.6 years; 69% had stage 3 or 4 disease on liver biopsy.
Esophageal varices developed in 31%, with the highest incidence found in those with stage 4 histology.
The varices appeared at a constant rate of 4% per year for the group as a whole and at a rate of
12% per year in those with stage 4 disease. Based on the previously mentioned studies, the risk of
developing new varices in a patient with cirrhosis is 4% to 12% each year. Once present, the risk of
the varices becoming large is 4% to 10% each year.
In the previously mentioned studies, guidelines for the frequency of endoscopy needed to identify
patients who develop large varices that may bleed have been proposed. Cales and colleagues [6] suggested
that endoscopy be performed every other year in cirrhotic patients without varices and yearly in those
with small varices. Pagliaro et al, [29] in contrast, suggested that endoscopy be performed every other year
in cirrhotics with small varices, and at less frequent intervals in patients without varices. The purpose in
identifying patients with varices is to treat them with beta-blockers and other drugs in the hope of
preventing the varices from bleeding. Nine to 11 patients with large varices need to be treated to prevent
one from having an initial bleed. [30] If 20% of patients with cirrhosis have high-risk varices, then it
will require 100 screening endoscopies to prevent 1 to 2 patients from bleeding from their varices.
The numbers are similar for surveillance endoscopy. For example, performing endoscopy every other
year on 100 patients with cirrhosis who lack or have small varices at the initial endoscopy will identify
4 to 10 individuals who have developed varices that are at risk for bleeding. If those 4 to 10
individuals are treated prophylactically, then at least 100 endoscopies will be required to prevent one
patient from bleeding. Based on current data, it is difficult to recommend routine surveillance endoscopy
in patients with cirrhosis who lack other manifestations of portal hypertension. If endoscopy is performed
and no varices are found, then follow-up endoscopy should not be performed for 3 to 5 years, if ever. If
small varices are present, then a repeat endoscopy is probably warranted in 2 to 3 years. If the varices
remain small at the second endoscopy, further surveillance is not indicated. These recommendations are

based on very limited data, and a careful cost-benefit analysis of well-designed prospective studies is
required before routine surveillance endoscopy can be recommended for patients with cirrhosis.
The greatest risk of bleeding appears to be during the first 1 to 2 years following identification of the
varices. [27] The author believes that this finding makes patients who are awaiting liver transplant ideal
candidates for endoscopic screening for high-risk varices. Most patients being evaluated for liver
transplant have advanced liver disease and have suffered from complications of portal hypertension;
therefore, they are more likely to have varices. As they have a finite period to wait for their liver
transplant (usually less than 2 years), only a single screening
35
endoscopy is required. Also, these patients usually have severely compromised liver function, and
bleeding may be poorly tolerated, adding to the benefit of finding and treating the varices before they
bleed.
RISK FACTORS FOR INITIAL BLEEDING FROM VARICES

Bleeding from varices occurs commonly once the varices have developed. The frequency of bleeding in
untreated patients with varices who served as controls in studies in which the impact of prophylactic
portacaval shunts, drugs, and sclerotherapy on variceal bleeding was examined was 29.6% 11.5%, with
an associated high mortality (Table 3) . In most of the studies the greatest risk for bleeding was within
the first 1 to 2 years following identification of the varices. [27] Although a number of factors appear to be
associated with bleeding risk, the endoscopic appearance of the varices was the variable that most
consistently predicted bleeding. The following discussion focuses on the endoscopic features of the
varices that are used to predict bleeding.
There is a minimal level of portal hypertension required for varices to become large enough to bleed.
That level is 12 mmHg or more above free hepatic vein or Inferior Vena Cava pressure, and has been
shown in a number of studies to be a constant finding in patients who have bled from varices [2] [13] [45] ;
however, many patients with pressures above this threshold do not bleed and may not develop varices.
[13] It is not surprising that many patients do not develop varices in response to portal hypertension. The
collateral circulation through the cardioesophageal junction is only one of four potential routes that blood
may take to
TABLE 3 -- RISK OF FIRST BLEED AND MORTALITY IN UNTREATED PATIENTS WITH
VARICES
Author (ref #) No. of Patients Bleeding (%) Mortality (%) *
Resnick [36] 45 27 38
Jackso [19] 58 19 28
Conn [8] 31 29 32
Pascal [31] 112 27 36

Poynard [34] 303 35 32

VA [44] 138 17 17
DeFranchis [11] 51 33 50
Koch [21] 30 30 33
Sauerbrunch [43] 65 37 46
Santangelo [39] 46 15 24
Witzel [46] 53 57 55
Mean SEM
29.6 11.5
35.3 11.3
* Period of observation varied from 2 to 8 years.
36
circumvent the cirrhotic liver. [2] Therefore, many cirrhotic patients may have significant blood flow
through the umbilical or retroperitoneal collaterals and little flow through their gastroesophageal varices.
This latter group lacks gastroesophageal varices despite significant portal hypertension.
Large varices are clearly more likely to bleed than small varices. [23] [27] The current belief is that bleeding
from varices occurs when the wall of the varix ruptures. Because varices are elastic structures, the risk of
rupture is related to wall tension. [33] Wall tension (T) is a function of intraluminal pressure (P), vessel
radius (R), and wall thickness (W), as defined by the following equation:
T = P R/W
The wall thickness resists the expanding force of P R, and when the expanding forces become too
great, the wall ruptures and bleeding occurs. Based on this paradigm, small thick-walled vessels are less
likely to bleed than are large thin-walled vessels. This theory also helps to explain why varices are more
likely to bleed in the distal compared to the more proximal esophagus. The submucosal veins are more
superficial in the distal esophagus; therefore, it is only the vessel wall itself that resists the expansion of
the vessel. In the more proximal esophagus the vessels are deeper, and in addition to the vessel wall, the
surrounding tissues lend support and help resist the expanding forces. [26] Similarly, very large collaterals
within the retroperitoneum do not rupture and bleed, probably because their walls are inherently thicker
and the entire vessel is surrounded by other tissues.
Risk Factors for Bleeding from Esophageal Varices

In addition to the endoscopic appearance of varices, a number of other variables have also been
examined as predictors of the first bleed from esophageal varices. In a large series from Italy, 321
patients with cirrhosis and esophageal varices and no history of bleeding were followed prospectively for
a median of 23 months. [27] During that period of observation, 85 patients (26.5%) bled from their
esophageal varices. Univariate analysis found five clinical findings (Child's class, ascites, bilirubin,
albumin, prothrombin time) and five endoscopic findings (red wale markings, variceal size, cherry-red
spots, location, hematocystic spots) that correlated significantly with risk of bleeding. By multivariate
analysis, Child's class, size of varices, and the presence of red wale markings were independent
predictors of bleeding (Table 4) . An index was created using these three variables, and the equation is as
follows:
Nothern Italian Endoscopic Club (NIEC) index = [0.645 C] + (0.4365 F) + [0.3193 R] 10
where C, F, and R are Child's classification, size of varices, and red wale markings, respectively. A
patient with an NIEC index of less than 20.0
37
TABLE 4 -- FACTORS PREDICTIVE OF BLEEDING FROM ESOPHAGEAL VARICES

Frequency of Bleeding (%)
NIEC [27] ZOLI [48]
Child's class
A 17 14.6
B 31.1 25.6
C 38.9 17
Variceal size
Small 18.1 4.8
Medium 28.6 16
Large 48.9 52.8
Red wale
Absent 19.1
Moderate 32.9
Severe 39.3
Cherry red spots Red color signs
Absent 23
Mild 32 10.3

Moderate 40 31.1
Severe 55 51.2
NIEC = The North Italian Endoscopic Club for the Study and Treatment of Esophageal Varices.
had a 1-year probability of bleeding of 2% whereas a patient with an index of 35 to 40 had a

probability of bleeding of 40%. A second study, also from Italy, prospectively followed 344 patients
with cirrhosis and varices that had not bled. [48] Their findings were similar to the previous study in that
by univariate analysis, Child-Pugh score, location of varices, their size, and presence of red color signs
also correlated with risk of bleeding (Table 4) . By multivariate analysis, only size of varices,
gastropathy, and presence of gastric varices were independent predictors of bleeding. They calculated a
prognostic index (PI) as follows:
PI = (size 0.0395) + (gastropathy 0.878) + (gastric varices 0.705)
Patients with a PI of less than 2.56 had a probability of bleeding of 0 to 20% during 1 year of follow-up
whereas those with a PI of greater than 4.51 had an 81% to 100% probability of bleeding during the first
year. Both of these reports demonstrate the value of prospective studies in defining the natural history of
esophageal varices, and can be used to establish treatment guidelines. Patients with large varices that also
have red color signs are clearly at a high risk for bleeding in the near future, and become candidates for
prophylactic therapy with drugs such as beta-blockers and nitrates. Patients with small varices are at little
risk of bleeding, and therapy in this group of patients clearly is not warranted.
Gastric Varices and Risk of Bleeding
The prevalence of gastric varices in patients with cirrhosis has varied widely, and appears to be higher in
patients presenting with
38
bleeding compared with those whose varices have never bled. [41] The frequency of gastric varices in 344
patients with cirrhosis who had undergone endoscopy was 6% in one study [48] and 20% in another. [41]
Some of the variation in frequency reflects difficulty in deciding endoscopically whether or not the
patient has gastric varices. As with esophageal varices, a number of endoscopic systems of classification
for gastric varices have been developed. The grading system developed by Sarin and colleagues [41]
provides most of the elements needed to estimate risk of bleeding. Gastric varices can be divided into
two groups: gastroesophageal varices (GOVs) and isolated gastric varices (IGVs). GOVs are esophageal
varices that extend across the cardioesophageal junction, and are termed GOV1 if they extend 2 to 5 cm
into the stomach, and GOV2 if they extend into the fundus. GOV1 accounts for 75% of the gastric
varices whereas GOV2 represent 21% of all gastric varices. Type 1 IGVs are located in the fundus of the
stomach, do not extend to the cardia, and account for only 1.6% of gastric varices, whereas type 2 IGVs
are located anywhere else in the stomach (body, antrum, or pylorus), and are slightly more common (
4%) compared to IGV1. [41] This system of classification is similar to those proposed by others. [40]

Although the finding of gastric varices in the absence of esophageal varices is believed to be indicative
of splenic vein thrombosis, most of the patients have cirrhosis and a patent splenic vein. [24] The impact
of sclerotherapy on the development of gastric varices also deserves mention. A group of 227 patients
who underwent sclerotherapy for esophageal varices was followed prospectively [47] ; 138 patients had
gastric varices at the initial endoscopy ( 50% GOV1) and 72% of the gastric varices where gone at the
end of treatment; 89 patients did not have gastric varices at the beginning of treatment and 12%
developed gastric varices following treatment of their esophageal varices. Hence, the development of
gastric varices following sclerosis of esophageal varices appears to be relatively infrequent. It is
unknown whether the gastric varices that develop following sclerotherapy have the same or a different
risk of bleeding compared to primary gastric varices.
Sarin and colleagues [41] examined the frequency of bleeding from different types of gastric varices, as
shown in Table 5 . It is difficult to
TABLE 5 -- RISK OF BLEEDING FROM GASTRIC VARICES

Data from Sarin SK, Lahoti D, Saxena SP, et al: Prevalence, classification and natural history of gastric
varices: A long-term follow-up study in 568 portal hypertension patients. Hepatology 16:1343, 1992.
Type of Varix Number Frequency of Bleeding (%)
GOV1 85 12
GOV2 31 55
IGV1 9 78
IGV2 22 9
Total 147 24.5
See text for abbreviations.
39
be certain of the length of follow-up, but the study was performed over a 5-year period. Approximately
one quarter of the patients with gastric varices bled during the period of observation. This is not
dissimilar to the frequency of bleeding from esophageal varices ( see Table 3) , although in some series
the risk of gastric varices bleeding appears to be less than the risk for esophageal varices. Bleeding from
gastric varices tended to be more severe, however. [41] The greatest risk of bleeding was observed in those
with GOV2 and IGV1; However, one third of the bleeding events were in patients with GOV1. In a
preliminary study, the risk of bleeding from gastric fundal varices (GOV2 and IGV1) was examined in
117 patients with cirrhosis [20] ; bleeding was observed in 34 patients (29%) during a 23-month period of
follow-up. The cumulative risk at 1, 3, and 5 years was 16%, 36%, and 44%, respectively. The size of the
varices, presence of red spots, and poor Child's status were all independent predictors of bleeding. Both
of these studies suggest that fundal varices have a significant risk of bleeding during a relatively short
period of observation, and that these patients are candidates for decompressive (surgical shunts versus
transjugular intrahepatic portosystemic shunt (TIPS) treatment once bleeding has occurred. The number

of patients with fundal varices that have been followed for a significant period of time is too small to
recommend prophylactic treatment, especially as the bleeding rates are quite different in the various
studies.
Portal Gastropathy and Risk of Bleeding
Portal hypertensive gastropathy is a frequent endoscopic finding ( 25%-75%) in patients with

cirrhosis. There are two types of gastropathy: portal hypertensive gastropathy (PHG) and gastric antral
vascular ectasia (GAVE). [16] [25] [32] PHG consists of a mosaic mucosal pattern with (severe) or without
(mild) red or brown spots, and is present in the gastric fundus and body of the stomach. [25] [32] GAVE
consists of linear red stripes converging on the pylorus (watermelon stomach) or diffuse red spots limited
to the antrum. [16] [32] The histologic appearance of GAVE differs from PHG, and the severity of bleeding
appears to be greater in patients with GAVE. [32] In addition, GAVE can occur in noncirrhotic patients,
suggesting this lesion is not due to vascular congestion of the antral mucosa; however, if GAVE is not
due to vascular congestion of the mucosa, then it should be a problem in patients with cirrhosis who have
undergone portosystemic shunts. GAVE has not been reported to be a cause for bleeding in the patients
following portosystemic shunts, suggesting that portal hypertension plays a role in its pathogenesis. This
discussion is limited to PHG but in some of the cited works, GAVE and PHG were not separated as
distinct entities, which complicates the analysis of these studies.
In one series, PHG was found in 58 of 212 patients with cirrhosis [9] ; the lesion was mild in 45 patients
and severe in 13. During the 46-month period of follow-up, an additional 65 individuals developed mild
40
gastropathy and 7 developed severe PHG. The likelihood of developing gastropathy was much greater in
those undergoing sclerotherapy (49%) compared to those receiving no treatment (26%), which is similar
to other reports. [42] Primignani and colleagues [35] in a preliminary report followed 298 patients with
PHG for a mean of 18 months; Figure 1 shows the natural history of PHG in this group of patients. Of
patients with no gastropathy at the initial endoscopy, 48% developed PHG; 85 patients had mild
gastropathy at the initial endoscopy; and in 18% it disappeared whereas in 39% it worsened; 152 patients
had severe gastropathy, and the endoscopic appearance improved in 47%. The changes in PHG could not
be correlated with treatment. These studies suggest that PHG can improve or worsen in patients with
cirrhosis, but in most patients the disease tends to remain the same or become progressively worse.
Bleeding is common in patients with PHG, and may be caused by the gastropathy or coincident varices.
Twenty bleeding episodes ascribed to the gastropathy were observed in 110 patients with mild disease
whereas 24 episodes of bleeding were observed in 20 patients with severe disease. [9] Bleeding tends to
be occult or chronic with melena, although active upper gastrointestinal hemorrhage may occur
infrequently. Patients with active upper gastrointestinal bleeding are more likely to be bleeding from
varices or GAVE. Patients with severe PHG that has not bled may be candidates for preventive therapy.

REBLEEDING FROM ESOPHAGEAL VARICES

The natural history of rebleeding from esophageal varices has been examined in a number of controlled
trials in which active treatment
Figure 1. Changes in portal hypertensive gastropathy with time. ( Data from Primignani M et al:
Portal hypertensive gastropathy [PHG] in liver cirrhosis: Natural history: A multicenter study of the
New Italian Endoscopic Club [NIEC]. Gastroenterology 110:A1299, 1996.)
41
(shunts, beta-blockers, sclerotherapy) was compared to no therapy. In the studies using beta-blockers and
sclerotherapy, it is important to note that during the acute episode of bleeding the participants may have
received sclerotherapy to stop the hemorrhage, and then were left untreated. In the shunt studies the
patients were truly untreated as beta-blocker therapy and sclerotherapy were not being used at the time
these studies were performed; however, some control patients underwent surgery, which may have had
an impact on the results. For example, in the study of Reynolds et al, [37] 16% of the controls underwent
surgery and 95% rebled whereas in another study, [17] 34% were lost to surgery and only 65% rebled. The
risk of rebleeding is greatest initially following the index bleed and then decreases with time. [15] In one
series, 194 cirrhotic patients were followed for 25 days. [5] During that time period 97 patients (50%)
rebled. Of those rebleeding, 48.4% bled within 24 hours of the index bleed 21.6% in the next 24 hours,
14.4% on day three, and 11.3% on day 4 or 5. Thus, 70% of the patients who rebled in the first 25 days
did so within 48 hours of the index bleed. For preventative therapy to be truly effective, therefore, it must
be given within 2 to 3 days of the index bleed.
Once patients have bled from esophageal varices, they are likely to bleed again. As shown in Table 6 ,
two thirds of patients will suffer at least one additional episode of bleeding. In one series, 98% of the
patients rebled during the period of observation. [37] Rebleeding appears to be more common in the
Child's C class patient, but patients with preserved hepatic function also are clearly at risk. In addition,
rebleeding is associated with significant mortality. Mortality is higher in those with poor liver function
(Table 6) .
CONCLUSION
There is an increasingly better understanding of the natural history of gastroesophageal varices and PHG.
This improved understanding is a direct result of prospective controlled trials in which large numbers of
untreated patients were followed for significant periods of time. Unfortunately, it has been difficult to
apply this knowledge in a cost-effective manner to the management of many of the patients. Although
TABLE 6 -- RISK OF REBLEEDING FROM ESOPHAGEAL VARICES
Data from D'Amico G, Pagliago L, Bosch J, et al: The treatment of portal hypertension: A meta-analytic
review. Hepatology 22:332, 1995.

No. of Rebleeding Death

Active Therapy Patients Child C (%) Controls (%) Controls (%)
PCS 195 NR 76.5 14.6 56.3 10.1
beta-Blockers 371 14.5 17.8 63.3 12.7 24.2 11.2
Sclerotherapy 559 41.1 14.8 65.8 11.3 55 16
PCS = Portacaval shunt; NR = not reported.
42
prophylactic therapy is clearly warranted for patients with high-risk varices, how those patients are
identified has not been resolved. Nor is it certain how to follow patients with known cirrhosis in the
hopes of finding large varices and preventing their bleeding. Further studies in this area are clearly
warranted.
ACKNOWLEDGMENT
The author appreciates the skilled secretarial assistance of Mr. L.T. Tucker.
References
1. Bendtsen F, Skovgaard LT, Sorensen TIA, et al: Agreement among multiple observers on endoscopic diagnosis of
esophageal varices before bleeding. Hepatology 11:341, 1990
2. Boyer TD, Henderson JM: Portal hypertension and bleeding esophageal varices. In Zakim D, Boyer TD (eds):
Hepatology: A Textbook of Liver Disease, ed 3. Philadelphia, WB Saunders, 1996, p 720
3. Boyer TD, Triger DR, Horisawa M, et al: Direct transhepatic measurement of portal vein pressure using a thin needle:
Comparison with wedged hepatic vein pressure. Gastroenterology 2:584, 1977
4. Burroughs AK, D'Heygere F, McIntyre N: Pitfalls in studies of prophylactic therapy for variceal bleeding in cirrhotics.
Hepatology 6:1407, 1986
5. Burroughs AK, Mezzanotte G, Phillips A, et al: Cirrhotics with variceal hemorrhage: The importance of the time interval
between admission and the start of analysis for survival and rebleeding rates. Hepatology 9:801, 1989
6. Cales P, Desmorat H, Vinel JP, et al: Incidence of large oesophageal varices in patients with cirrhosis: Application to
prophylaxis of first bleeding. Gut 31:1298, 1990
7. Cales P, Zabotto B, Meskens C, et al: Gastroesophageal endoscopic features in cirrhosis: Observer variability,
interassociations and relationship to hepatic dysfunction. Gastroenterology 8:156, 1990
8. Conn HO, Lindenmuth WW: Prophylactic portacaval anastomosis in cirrhotic patients with esophageal varices: A
progress report of a continuing study. N Engl J Med 272:1255, 1965

9. D'Amico G, Montalbano L, Traina M, et al: Natural history of congestive gastropathy in cirrhosis. Gastroenterology
99:1558, 1990
10. D'Amico G, Pagliago L, Bosch J: The treatment of portal hypertension: A meta-analytic review. Hepatology 22:332,
1995
11. DeFranchis R, Primignani M, Arcidiacono PG, et al: Prophylactic sclerotherapy in high-risk cirrhotics selected by
endoscopic criteria: A multicenter randomized controlled trial. Gastroenterology 101:1087, 1991
12. Garcia-Pagan JC, Santos C, Barbera JA, et al: Physical exercise increases portal pressure in patients with cirrhosis and
portal hypertension. Gastroenterology 111:1300, 1996
13. Garcia-TsaoG, Groszmann RJ, Fisher RL, et al: Portal pressure, presence of gastroesophageal varices and variceal
bleeding. Hepatology 5:419, 1985
14. Gores
GJ, Wiesner RH, Dickson ER, et al: Prospective evaluation of esophageal varices in primary biliary cirrhosis:
Development, natural history and influence on survival. Gastroenterology 96:1552, 1989
15. Graham DY, Smith JL: The course of patients after variceal hemorrhage. Gastroenterology 80:800, 1981
16. Jabbari
M, Cherry R, Lough J, et al: Gastric antral vascular ectasia: The watermelon stomach. Gastroenterology
87:1165, 1984
17. JacksonFC, Perrin EB, Felix RW, et al: A clinical investigation of the portacaval shunt: Survival analysis of the
therapeutic operation. Ann Surg 174:672, 1971
43
18. JacksonFC, Perrin EB, Smith AG, et al: A clinical investigation of the portacaval shunt: II. Survival analysis of the
prophylactic operation. Am J Surg 115:22, 1968
19. Japanese Research Society for Portal Hypertension: The general rules for recording endoscopic findings on esophageal
varices. Jpn J Surg 10:84, 1980
20. Kim T, Shijo H, Kokawa H, et al: Risk factors for hemorrhage from gastric fundal varices. Hepatology 24:209A, 1996
21. Koch H, Henning H, Grimm H, et al: Prophylactic sclerosing of esophageal varices: Results of a prospective controlled
study. Endoscopy 18:40, 1986
22. Kravetz
D, Sikuler E, Groszmann RJ: Splanchnic and systemic hemodynamics in portal hypertensive rats during
hemorrhage and blood volume restitution. Gastroenterology 90:1232, 1986
23. LebrecD, De Fleury P, Rueff B, et al: Portal hypertension, size of esophageal varices and risk of gastrointestinal
bleeding in alcoholic cirrhosis. Gastroenterology 79:1139, 1980
24. LevineMS, Kieu K, Rubesin SE, et al: Isolated gastric varices: Splenic vein obstruction or portal hypertension?
Gastrointest Radiol 15:188, 1990
25. McCormack TT, Sims J, Eyre-Brook I, et al: Gastric lesions in portal hypertension inflammatory gastritis or congestive

gastropathy. Gut 26:1226, 1985
26. NodaT: Angioarchitectural study of esophageal varices with special reference to variceal rupture. Virchows Arch [A]
404:381, 1984
27. The
North Italian Endoscopic Club for the Study and Treatment of Esophageal Varices: Prediction of the first variceal
hemorrhage in patients with cirrhosis of the liver and esophageal varices. N Engl J Med 319:983, 1988
28. Okumura H, Aramaki T, Katsuta Y, et al: Reduction in hepatic venous pressure gradient as a consequence of volume
contraction due to chronic administration of spironolactone in patients with cirrhosis and no ascites. Am J Gastroenterol
86:46, 1991
29. Pagliaro
L, D'Amico G, Pasta F, et al: Portal hypertension in cirrhosis: Natural history. In Bosch J, Groszmann RJ (eds):
Portal Hypertension: Pathophysiology and Treatment. London, Blackwell Scientific, 1994, p 72
30. Pagliaro L, D'Amico G, Sorensen TIA, et al: Prevention of first bleeding in cirrhosis: A meta-analysis of randomized
trials of nonsurgical treatment. Ann Intern Med 117:59, 1992
31. Pascal
J-P, Cales P, Multicenter Study Group: Propranolol in the prevention of first upper gastrointestinal tract
hemorrhage in patients with cirrhosis of the liver and esophageal varices. N Engl J Med 317:856, 1987
32. Payen J-L, Cales P, Voigt J-J, et al: Severe portal hypertensive gastropathy and antral vascular ectasia are distinct
entities in patients with cirrhosis. Gastroenterology 108:138, 1995
33. Polio
J, Groszmann RJ: Hemodynamic factors involved in the development and rupture of esophageal varices: A
pathophysiologic approach to treatment. Semin Liver Dis 6:318, 1986
34. Poynard T, Cales P, Pasta L, et al: Beta-adrenergic-antagonist drugs in the prevention of gastrointestinal bleeding in
patients with cirrhosis and esophageal varices. N Engl J Med 324:1532, 1991
35. PrimignaniM, Carpinelli L, Preatoni P, et al: Portal hypertensive gastropathy (PHG) in liver cirrhosis: Natural history:
A multicenter study of the New Italian Endoscopic Club (NIEC). Gastroenterology 110:A1299, 1996
36. ResnickRH, Chalmers TC, Ishihara AM, et al: A controlled study of prophylactic portacaval shunt: A final report. Ann
Intern Med 70:675, 1969
37. Reynolds TB, Donovan AJ, Mikkelsen WP, et al: Results of a 12-year randomized trial of portacaval shunt in patients
with alcoholic liver disease and bleeding varices. Gastroenterology 80:1005, 1981
38. Reynolds TB, Redeker AG, Geller HM: Wedged hepatic vein pressure. Am J Med 22:341, 1957
39. Santangelo WC, Dueno MI, Estes BL, et al: Prophylactic sclerotherapy of large esophageal varices. N Engl J Med
318:814, 1988
40. Sarin
SK: Diagnostic issues: Portal hypertensive gastropathy and gastric varices. In De Franchis R (ed): Portal
Hypertension, vol 2. London, Blackwell Scientific, 1996, p 30
44

41. Sarin
SK, Lahoti D, Saxena SP, et al: Prevalence, classification and natural history of gastric varices: A long-term
follow-up study in 568 portal hypertension patients. Hepatology 16:1343, 1992
42. Sarin
SK, Sreenivas DV, Lahoti D, et al: Factors influencing development of portal hypertensive gastropathy in patients
with portal hypertension. Gastroenterology 102:994, 1992
43. Sauerbruch T, Wotzka R, Kopcke W, et al: Prophylactic sclerotherapy before the first episode of variceal hemorrhage in
patients with cirrhosis. N Engl J Med 319:8, 1988
44. The
Veterans Affairs Cooperative Variceal Sclerotherapy Group: Prophylactic sclerotherapy for esophageal varices in
men with alcoholic liver disease. N Engl J Med 324:1779, 1991
45. Viallet
A, Marleau D, Huet M, et al: Hemodynamic evaluation of patients with intrahepatic portal hypertension.
46. Witzel L, Wolbergs E, Merk H: Prophylactic endoscopic sclerotherapy of oesophageal varices. Lancet I:773, 1985
47. Zanasi G, Rossi A, Grosso C, et al: The effect of endoscopic sclerotherapy of esophageal varices on the development of
gastric varices. Endoscopy 28:234, 1996
48. ZoliM, Merkel C, Magalotti D, et al: Evaluation of a new endoscopic index to predict first bleeding from the upper
gastrointestinal tract in patients with cirrhosis. Hepatology 24:1047, 1996


April 21, 2000
Conde Petra
PATHOPHYSIOLOGY OF PORTAL
HYPERTENSION
Additional Article
Full Text
Frontmatter 1
INCREASED RESISTANCE
PORTAL HYPERTENSION
Intrahepatic Resistance
Portosystemic Collateral
Resistance PATHOPHYSIOLOGY OF PORTAL
HYPERDYNAMIC CIRCULATION IN HYPERTENSION
PORTAL HYPERTENSION
Systemic Vasodilatation
Increased Circulating
Vasodilators
Tarun K. Gupta MD
Increased Endothelial Lisa Chen MD
Production of Vasodilators
Roberto J. Groszmann MD, FRCP
Nitric Oxide (NO).
From the Veteran's Affairs Medical Center, West Haven,
Prostaglandins. Connecticut; Yale University School of Medicine, New Haven,
Decreased Response to Connecticut; and the Bridgeport Hospital, Bridgeport,
Vasoconstrictors Connecticut (TKG)
Plasma Volume
SUMMARY
A grasp of the biological mechanisms involved in the
References pathogenesis of portal hypertension is essential to an
understanding of the complications of chronic liver disease
and to the development of rational therapies. This article is
an overview of the basic pathophysiologic mechanisms of
the splanchnic and systemic circulatory derangements that
lead to portal hypertension and cause the portosystemic
collaterals to develop.
The pressure in a vascular system is directly proportional to
the resistance (R) and flow (Q) in that system.
Palpha Q R
Currently, both increased vascular resistance and increased
splanchnic blood flow are believed to play key roles in the
production of elevated portal pressure (Fig. 1) . [40]

Roberto J. Groszmann, MD
Yale University School of Medicine
Hepatic Hemodynamic Laboratories/111H
VA Connecticut Healthcare System
950 Campbell Avenue
West Haven, CT 06516
Portal hypertension is associated with increased resistance to
portal blood flow. Increased vascular resistance is due to an
increase in both intrahepatic and portocollateral resistance in
comparison to the low resistance of the normal liver.
Figure 1. Factors contributing to elevated portal

pressure.
The main site of resistance in the normal liver is somewhat

controversial. The hepatic sinusoids (where stellate cells are
present), terminal hepatic venules, and portal venules have been
suggested as possible sites of resistance; however, in view of
the minimal contribution of the intrahepatic resistance to portal
pressure in the normal liver, this issue is of little importance.
The flow into the normal portal system is actively regulated by
changes in vascular resistance at the level of splanchnic
arterioles, and not by the liver itself. [17]
In portal hypertensive syndromes, increased resistance to portal
venous flow may be localized to prehepatic, posthepatic, and
intrahepatic (presinusoidal, sinusoidal, or postsinusoidal) sites.

In prehepatic and posthepatic portal hypertension,

[13] [18]
increased resistance is secondary to obstruction of portal venous

inflow or hepatic venous outflow, respectively. Unlike pre- and
posthepatic portal hypertension, the intrahepatic syndromes are
more complex and rarely can be classified according to a single
site of resistance.
An early view of vascular resistance in cirrhotic livers
postulated that portal hypertension is the consequence of a
vascular obliterative process, with scar tissue and regenerative
nodules, occluding and compressing vascular structure. [3] [33] [41]
Thus, earlier understanding of intrahepatic portal hypertension
emphasized only the role of anatomic alterations leading to
mechanical obstruction (irreversible component) in the
increased intrahepatic resistance; however, recently, it has been
shown that there is an increase in vascular tone (which normally
is negligible in the normal liver) that imposes an additional
increase in resistance (reversible component). [5]
In intrahepatic portal hypertension, there may be several areas
of obstruction, and as the disease progresses, new sites may
become involved.
For example, in hepatic schistosomiasis, the increased

intrahepatic resistance results from granulomas located in
presinusoidal areas [4] ; however, in late stages, an elevated
hepatic wedge venous pressure gradient may be observed,
reflecting increased sinusoidal resistance. [42] Chronic hepatitis
seems to have both presinusoidal and sinusoidal abnormalities
that increasingly contribute to vascular resistance as the lesion
progresses toward cirrhosis. [53] In ethanol-induced liver disease,
the intrahepatic resistance is increased because of lesions in
sinusoidal and postsinusoidal sites. The terminal hepatic vein
fibrosis (or sclerosis), encroachment on sinusoids by enlarged
hepatocytes, and collagen deposition in the perisinusoidal
region or the space of Disse, resulting in sinusoidal narrowing,
combine to produce elevated portal pressure, even in a
precirrhotic stage [27] [37] [43] [47] ; however, in advanced stages of
cirrhosis, regenerating nodules and pruning of the vascular tree
contribute to the increased vascular resistance. A variety of
other nonalcoholic forms of liver diseases cause portal
hypertension owing to increased sinusoidal vascular resistance.
In hepatic amyloidosis, resistance is increased because of the
deposition of amyloid in the space of Disse. [6] The
capillarization process and occlusion of the endothelial

fenestrae are postulated to increase the resistance of passage

across the endothelium; however, the extent of their effect on
resistance to blood flow is unknown.
The morphologic changes occurring in chronic liver diseases
are undoubtedly the most important factor involved in the
increased intrahepatic resistance; however, recent data also
suggest a role of functional factors that lead to increased
vascular tone, similar to what is seen in arterial hypertension. In
chronic liver disease, and also during acute liver injury, hepatic
stellate cells acquire contractile properties and may contribute
to the dynamic modulation of intrahepatic resistance. [22] [38]
These cells may act as pericytes, a type of cell that has been
shown to regulate blood flow in other organs. The hepatic
stellate cells, which are also the main source of collagen
synthesis, may contribute to the regulation of hepatic blood
flow at the microcirculatory level. Stellate cells are strategically
located in the sinusoids, with perisinusoidal and
interhepatocellular branching processes that contain actin-like
filaments. They also express the alpha smooth muscle actin
gene, which is characteristic of vascular smooth muscle. The
characteristics of these cells make them similar to the
myofibroblast. Myofibroblasts are intermediate in structure
between smooth muscle cells and fibroblasts.
Myofibroblast-like cells have been shown to exist in fibrous
septa around the sinusoids and terminal hepatic venules in
cirrhotic livers. [46] These cells are postulated to have a role in
the regulation of vascular resistance in the cirrhotic rat liver. [5]
The vascular endothelium synthesizes various vasodilators such
as nitric oxide (NO), prostacyclins, and hyperpolarizing factor,
and the vasoconstrictors such as endothelins and prostanoids.
[45] [54] These vasoactive substances act in a paracrine fashion on
the underlying vascular smooth muscle and modulate vascular

tone. Normal vascular tone is maintained by a delicate balance
between these vasodilatory and vasoconstrictive
substances. Perturbation of this balance leads to abnormal

vascular tone. Increased vascular tone seen in cirrhotic livers
could be due to a deficit of endothelial vasodilators, an increase
in the vasoconstrictors, or a combination of both. NO is a potent
endothelial vasodilator that has been shown to play an
important role in the modulation of intrahepatic vascular tone in
normal livers. [34] Preliminary evidence from the authors'
laboratory suggests that there may be a deficit of NO in the

cirrhotic intrahepatic microcirculation, as seen in other

hypertensive regional microcirculations. [19] Using the isolated
rat liver perfusion model, the authors have demonstrated that
there is endothelial dysfunction in the intrahepatic
microcirculation of cirrhotic livers. [20] Endothelial dysfunction
leads to an impaired release of endothelial vasodilators, which
may, in part, be responsible for the increased vascular tone
observed in cirrhotic livers. More recently, it has been shown
that the stellate cells (myofibroblasts) from cirrhotic livers
exhibit enhanced response to endothelins. [44] An imbalance
between endothelial vasodilators and vasoconstrictors can affect
the activated stellate cells (myofibroblasts) that modulate
intrahepatic vascular tone. It is possible that both a deficit of
vasodilators and an increase in vasoconstrictors may be
responsible for the increased vascular tone. In summary, there
are multiple factors that may lead to increased intrahepatic
resistance to portal blood flow. Some of these are irreversible,
like fibrosis, capillarization, and regenerating nodules, and
some are quite dynamic, like the imbalance between endothelial
factors leading to increased vascular tone.
Portosystemic Collateral Resistance
Portosystemic collaterals develop as a result of portal

hypertension. The development of collaterals is the central
pathophysiologic event that leads to variceal bleeding and
portosystemic encephalopathy in patients with portal
hypertension. Although the collateral circulation begins as a
consequence of portal hypertension, it evolves into an important
mediator of the circulatory derangements of portal hypertension
in its own right. Although the collaterals provide a route to
decompress the hypertensive portal system, their vascular
resistance remains higher than the normal intrahepatic
resistance. Hence, portosystemic collaterals do not permit a
portal decompression that will return portal pressure to normal.
The exact nature of the physiologic stimuli responsible for
initiating collateral formation remains somewhat controversial.
Propranolol and clonidine have been shown to decrease portal
pressure and ameliorate portosystemic shunting, thus
implicating increased portal pressure in the pathophysiology of
collateral formation. [21] Lee et al, [24] however, demonstrated
that collateral formation could be ameliorated by preventing an
increase in splanchnic blood flow without any decrease in portal
pressure. Thus, increased portal pressure does not appear to be
the only factor involved in the development of collaterals in
portal hypertensive states. Therefore, propranolol and clonidine
may prevent collateral formation
by decreasing flow, not by their effect on portal pressure. [21] On

the other hand, Sikuler et al [50] studied partial
portal-vein-ligated (PVL) rats on postoperative day 3, and
found no relationship between portosystemic shunting and
portal venous inflow. Therefore, the development of
portosystemic shunting does not appear to be dependent
exclusively on either portal pressure or portal venous inflow.
Studies in the PVL rat model have shown that the blockade of
NO synthesis with Nw -nitro-L-arginine (NNA) can ameliorate
portosystemic shunting, suggesting that increased NO may be
responsible for the initial collateralization of the portal system.
[24] Increased NO could result from progressive increases in
splanchnic flow and shear stress, potent stimuli known to be

present in portal hypertensive states. In addition, development
of a new portocollateral bed renders available a new endothelial
surface capable of producing NO. [24]
Mosca et al [32] determined the vascular responsiveness of
collateral vessels to various vasoconstrictors and vasodilators
using an in situ perfused animal model. By administering
norepinephrine, 5-hydroxytryptamine, isoproterenol, and
acetylcholine in the presence and absence of their respective
blockers phentolamine, propranolol, and L-NNA, they
determined that collateral vessels have functional
alpha-adrenoreceptors, 5-hydroxytryptamine receptors, and
beta-adrenoreceptors. In addition, collateral veins appear to be
as sensitive to NO as arteries. [32]
Regardless of the specific initiating stimulus, in the beginning
the mechanism of collateral formation appears to be recruitment
of preformed channels, as opposed to de novo formation of new
vessels. Portosystemic shunting can be detected almost
immediately after the induction of portal hypertension, a pattern
of timing consistent with the rapid dilatation of preformed
vessels. [32] Whether these preformed vessels expand as the
result of active dilatation or as a passive response to increased
splanchnic flow or portal pressure is unclear [32] ; however,
neoformation of collateral vessels is probably a later event.

HYPERDYNAMIC CIRCULATION IN
PORTAL HYPERTENSION
Chronic elevations in systemic and splanchnic blood flow have

been documented as key elements of the hyperdynamic
circulatory state (HCS) of portal hypertensive animals and
humans. Peripheral vasodilatation initiates the development of
the classic profile of decreased systemic vascular resistance and
mean arterial pressure plasma volume expansion, elevated
splanchnic blood flow, and elevated cardiac index that
characterize this state. [9] At least three mechanisms are believed
to contribute to this peripheral vasodilatation: (1) increased
concentrations of circulating vasodilators, (2) increased
endothelial production of local
vasodilators, and (3) decreased vascular responsiveness to

endogenous vasoconstrictors. This last mechanism is probably
due to the effects of the first two components. Increased
Circulating Vasodilators
In portal hypertensive states there is an increase in both
endothelium-dependent and independent vasodilators. Possible
causes for increased circulatory concentrations of vasodilatory
substances include increased production, decreased catabolism
secondary to impaired hepatic function, and portosystemic
shunting.
Circulating bile acids, routinely cleared by the liver, are present
in elevated concentrations when liver function is impaired. Bile
acid depletion has been shown to be associated with a decrease
in splanchnic hyperemia. Experimental evidence, however,
suggests that an increase in circulating bile acids is not essential
for maintaining the HCS in portal hypertension. [14] More
specifically, cholestyramine-induced reduction of bile acids to
concentrations seen in placebo-treated controls did not
ameliorate the hemodynamic changes of the HCS in portal
hypertensive animals. [14]
Elevated concentrations of circulating glucagon have also been
documented in both animals and humans with portal
hypertension. [16] [39] [51] Rats with portal hypertension induced
experimentally by either PVL or carbon tetrachloride induced
cirrhosis had significantly higher glucagon and insulin

concentrations compared with control rats. [16] [39] In addition,

pancreatic islet isolates from these animals exhibited a
significantly higher glucagon secretion rate in response to
glucose and arginine administration compared with controls,
although insulin secretion appeared to be impaired. [16]
Intraoperative glucagon concentrations measured from the
portal vein and inferior vena cava are significantly increased
after surgical portosystemic shunting in patients with
Budd-Chiari syndrome and slightly, but not significantly, in
cirrhotic patients. [51] Glucagon concentrations in this study,
however, did not correlate with portal pressure. Therefore,
whether glucagon plays a role in the production of the HCS
remains unclear. [51]
Increased Endothelial Production of Vasodilators
Recently, increasing evidence has pointed toward a major role
for the endothelium in the maintenance of basal vascular tone
and the development of local and generalized vasodilatation in
portal hypertension. The endothelium produces at least two
substances that are known to contribute to the development of
systemic and splanchnic vasodilatation in portal hypertension:
NO and prostaglandins. [7]
Nitric Oxide (NO).
NO, previously known as endothelial-derived relaxing factor, is

synthesized from L-arginine by the enzyme NO synthase (NOS),
which has constitutive and inducible forms in different cell
types. NO mediates its potent vasodilatory action on smooth
muscle
cells through soluble guanylate cyclase. Evidence exists that

elevated production of NO is essential to the development of the
hyperdynamic circulation. Treatment with NNA, a
physiologically inactive substrate analogue, and thus a
competitive inhibitor of NOS, prevented the development of
peripheral vasodilatation, decreased systemic arterial pressure,
plasma volume expansion, and sodium retention in a PVL rat
model.25 Similarly, chronic continuous administration of
L-NAME, another inhibitor of NOS, has recently been shown to
delay splanchnic vasodilatation, increased splanchnic blood
flow, and development of collaterals in PVL rats. [12] The
vasodilatory effects of NO in portal hypertension are not limited
to the splanchnic circulation as they also affect the systemic
circulation; in PVL rats, isolated aortic rings demonstrated
increased relaxation to the vasodilator acetylcholine

(endothelial agonist) compared with sham controls. This
increased response to acetylcholine was partly reversed with
L-NAME. [23]
Whether production of NO is a primary stimulus in the

development of vasodilatation or a subsequent phenomenon that
results from shear stress and secondarily contributes to
increased flow has yet to be determined. Vallance et al [52]
hypothesized that chronic endotoxemia in portal hypertension
and cirrhosis may up-regulate inducible NOS (iNOS), thus
causing the increased NO production that leads to splanchnic
vasodilation. [52] Tumor necrosis factor-alpha (TNF-alpha) may
play an important role in the vasodilatation by up-regulating
iNOS. Treatment with anti-TNF-alpha polyclonal antibodies
and thalidomide have been shown to ameliorate the
development of the hyperdynamic circulation in the PVL rat
model of portal hypertension. [29] [30] The relative contributions
of the constitutive and inducible forms of NOS are still under
investigation.
Prostaglandins.
Several studies in animals and humans have implicated

increased endothelial production of prostaglandins as a cause of
splanchnic vasodilatation in portal hypertensive states. Portal
vein concentrations of prostacyclin (PGI2 ), for example, have
been found to be elevated in PVL rabbits and rats, as well as in
patients with cirrhosis and Budd-Chiari syndrome. These PGI2
concentrations have correlated with portal pressure. [55] In
addition, increased concentrations of the prostaglandin
metabolite 2,3-dinor-6 keto PGF1alpha have been observed in
cirrhotic patients, and elevated gastric PGE2 synthesis has been
seen in cirrhotic humans with severe portal hypertensive
gastropathy. [7]
In the PVL animal model of portal hypertension, increased
prostaglandin production appears to have a more prominent role
in rabbits than in rats. Inhibition of prostaglandin synthesis
through pharmacologic cyclooxygenase blockade has been
shown to prevent development of the HCS in portal
hypertensive rabbits. [55] Wu et al [55] measured splanchnic flow
and portal pressure in PVL rabbits in the presence and absence
of the respective inhibitors of cyclooxygenase and NOS,
indomethacin, and L-NAME; their results were consistent with
mediation of splanchnic hyperemia predominantly by a
prostaglandin, possibly prostacyclin, with a limited role for NO

as a mediator of basal vascular tone. In addition, the effects of

NO blockade with L-NAME, and reversal of that
blockade with the naturally occurring substrate for NO

synthase, L-arginine, were not significantly different between
normal and PVL rabbits, thus implying that at least in rabbits,
increased NO production may not be responsible for the HCS
associated with portal hypertension. [55]
Similar studies performed in PVL rats, however, suggest that
both NO and prostaglandins contribute to gastric hyperemia. [7]
In both rabbits and rats, the two endothelial products appear to
cause local vasodilatation through distinct mechanisms.
Relaxation of vascular smooth muscle in response to
acetylcholine, which depends on increased NO production from
an intact endothelium, was not prevented by cyclooxygenase
blockade. [55] In PVL rats, hemodynamic measurements in the
presence or absence of indomethacin, and L-NAME
demonstrated that prostaglandins and NO do not appear to act
synergistically. [7] In addition, an in vitro endothelial study
suggests that prostacyclin release may be markedly suppressed
by NO, [11] and other researchers have found that NO inhibition
may be associated with increased prostacyclin production. [8]
Thus, both NO and prostaglandins appear to act through
separate pathways to contribute to the vasodilatation that leads
to increased splanchnic flow in portal hypertension, although
the relative contribution of each may vary among different
species.
Decreased Response to Vasoconstrictors
Basal vascular tone is regulated by the complex balance
between endogenous vasodilators and vasoconstrictors. A
blunted response to vasoconstrictors, therefore, should also
contribute to vasodilatation and, subsequently, hyperdynamic
flow. In portal hypertensive states, in vivo hyporesponsiveness
to the endogenous vasopressors norepinephrine, arginine
vasopressin, and angiotensin II, has been reported to contribute
to the HCS. [48]
This hyporeactivity to vasopressors appears to be mediated
largely by NO. In portal hypertensive rats, inhibition of NO in
isolated perfused superior mesenteric artery beds has been
shown to prevent the development of vascular hyporeactivity to
the endogenous vasoconstrictors norepinephrine and

vasopressin, [48] the exogenous alpha-agonist methoxamine, and

the receptor-independent vasoconstrictor potassium chloride. [49]
These observations are consistent with the previous hypothesis
that the decreased response to vasoconstrictors in portal
hypertension is mediated by receptor-independent mechanisms.
[51]
In portal hypertensive rats, a role for prostaglandins in

hyporesponsiveness to vasoconstrictors has not been
substantiated. In fact, cyclooxygenase inhibition with
indomethacin did not ameliorate vascular hyporeactivity in
superior mesenteric artery preparations in partial PVL rats. [49]
Therefore, at least in the rat model of portal hypertension, NO
appears to cause the vascular hyporeactivity to endogenous and
exogenous vasoconstrictors that contributes to the generalized
vasodilation seen in the HCS.
Plasma Volume
The hyperdynamic circulation is mediated in part by

vasodilatation, but this alone is not sufficient to cause the
circulation to become hyperdynamic. Plasma volume expansion
has been recognized in portal hypertension for many years. [28]
[31] [35] In the PVL rat model of portal hypertension, plasma
volume failed to expand in animals on a sodium-restricted diet

as compared to those on normal diet. Moreover, the fully
developed hyperdynamic circulation was found to be nearly
reversible by sodium restriction. [15] Subsequent studies have
demonstrated that in the PVL rats, vasodilatation, expansion of
plasma volume by sodium retention, and development of the
hyperdynamic circulation follow each other in a stepwise
fashion.1,10 Vascular resistance in the systemic circulation drops
significantly within 1 day of partial PVL, followed on day 2 by
an increase in plasma volume and a parallel progressive
increase in systemic and regional blood flows. The fully
expanded plasma volume is observed on day 4, and coincides
with a maximally hyperkinetic cardiac index. These studies
provide important evidence for the existence of several events
in the pathogenesis of the hyperdynamic circulation. These
include initial vasodilatation (induced by humoral and local
endothelial factors) followed by plasma volume expansion. The
authors have demonstrated that both octreotide, which supresses
the secretion of vasodilatory peptides, and NO blockers
decrease renal sodium retention and plasma volume expansion
by diminishing vasodilatation, thereby preventing the full
expression of the hyperdynamic circulation. [2] [26]

The studies that examine the role of vasodilatation and plasma
volume expansion in the hyperdynamic circulation provide
support for the peripheral vasodilatation hypothesis. According
to this hypothesis, portal hypertension leads to a relative
hypovolemia induced by the dilatation of the systemic and
splanchnic circulations. This results in underfilling of the
systemic vascular space and a decrease in the central blood
volume. This in turn leads to activation of the sympathetic
nervous system and the renin-angiotensin system, and a release
of antidiuretic hormone. Mediators from these systems result in
sodium and water retention by the kidneys, resulting in
increased plasma volume.
SUMMARY
Portal hypertension is a common clinical syndrome associated
with chronic liver diseases, and is characterized by a
pathological increase in portal pressure that leads to the
formation of portosystemic collaterals resulting in shunting of
portal blood into the systemic circulation. The increase in portal
pressure is due to an increase in vascular resistance and an
elevated portal blood flow. The site of increased resistance is
variable, and dependent upon the disease process. The site of
relative obstruction may be prehepatic, hepatic, or posthepatic.
There are several intrahepatic lesions that lead to increased
resistance. Some of these
10
lesions are irreversible, like fibrosis, regenerating nodules, and

capillarization of the space of Disse; however, there is a
functional component, increased vascular tone, which
contributes to increased intrahepatic resistance and is
potentially reversible. Another important factor contributing to
the increased portal pressure is elevated portal blood flow.
Peripheral vasodilatation initiates the classical profile of
decreased systemic resistance, expanded plasma volume,
elevated splanchnic blood flow, and elevated cardiac index,
which characterize the hyperdynamic circulatory state. This
hyperdynamic circulation is responsible for various
complications of portal hypertension.
References

1. AlbillosA, Colombato LA, Groszmann RJ: Vasodilatation and sodium

retention in prehepatic portal hypertension. Gastroenterology 102:931-935,
1992
2. AlbillosA, Colombato LA, Lee FY, et al: Octreotide ameliorates

vasodilatation and sodium retention in portal hypertensive rats.
3. BaldusWP, Hoffbauer FW: Vascular changes in the cirrhotic liver as

studied by injection technic. Am J Dig Dis 8:689-692, 1963
4. BekerS, Valencia-Parparcen J: Portal hypertension syndrome: A

comparative analysis of bilharzial fibrosis and hepatic cirrhosis. Am J Dig
Dis 13:1047-1054, 1968
5. Bhathal PS, Grossman HJ: Reduction of the increased portal vascular

resistance of the isolated perfused cirrhotic rat liver by vasodilators. J
Hepatol 1:325-337, 1985
6. BrionE, Brenard R, Pariente EA, et al: Sinusoidal portal hypertensin in

hepatic amyloidosis. Gut 32:227-230, 1991
7. Casadevall M, Panes J, Pique JM, et al: Involvement of nitric oxide and

prostaglandins in gastric mucosal hyperemia of portal-hypertensive
anesthetized rats. Hepatology 18:628-34, 1993
8. ClariaJ, Jimenez W, Ros J, et al: Pathogenesis of arterial hypotension in

cirrhotic rats with ascites: Role of endogenous nitric oxide. Hepatology
15:343-349, 1993
9. Colombato LA, Albillos A, Groszmann RJ: Temporal relationship of

peripheral vasodilatation, plasma volume expansion and the hyperdynamic
circulatory state in portal-hypertensive rats. Hepatology 15:323-328, 1991
11. Doni MG, Whittle BJ, Palmer RM, et al: Action of nitric oxide on the
release of prostacyclin from bovine endothelial cells in culture. Eur J
Pharm 151:1925, 1988
12. Garcia-Pagan JC, Fernandez M, Bernadich C, et al: Effect of continued

NO inhibition on portal hypertensive syndrome after portal vein stenosis in
rat. Am J Physiol 267:G984-90, 1994
13. Genecin P, Groszmann RJ: Portal hypertension. In Schiff E, Schiff L

(eds): Diseases of the Liver. Philadelphia, JB Lippincott, 1993
14. Genecin
P, Polio J, Ferraioli G, et al: Bile acids do not mediate the
hyperdynamic circulation in portal hypertensive rats. Am J Physiol
259:G21-G25, 1990
15. GenecinP, Polio J, Groszmann RJ: Na restriction blunts expansion of

plasma volume and ameliorates hyperdynamic circulation in portal
hypertension. Am J Physiol 259:G498-G503, 1990
16. Gomis R, Fernandez-Alvarez J, Pizcueta P, et al: Impaired function of

pancreatic islets from rats with portal hypertension resulting from cirrhosis
and partial portal vein ligation. Hepatology 19:1257-1261, 1994
17. Greenway CV, Stark RD: Hepatic vascular bed. Physiol Rev 51:23-65,
1971
18. Groszmann RJ, Atterbury CE: The pathophysiology of portal

hypertension: A basis for classification. Semin Liver Dis 2:177-186, 1982
19. GuptaTK, Groszmann RJ: Administration of L-arginine, the

physiological precursor of nitric oxide, reduces portal perfusion pressure
and ameliorates hepatic vascular hyperreactivity in experimental cirrhosis.
Hepatology 20:200A, 1994
11
20. Gupta TK, Chung M, Sessa WC, et al: Impaired endothelial function in
the intrahepatic circulation in cirrhosis. Hepatology 22:156A, 1995
21. Halvorsen JF, Myking AO: The porto-systemic colleteral pattern in the
rat. Eur Surg Res 6:183-195, 1974
22. Johnson SJ, Hines JE, Burt AD: Phenotypic modulation of

perisinusoidal cells following acute liver injury: A quantitative analysis.
Int J Exp Pathol 73:765-772, 1992
23. Karatapanis S, McCormick PA, Kakad S, et al: Alteration in vascular

reactivity in isolated aortic rings from portal vein-constricted rats.
Hepatology 20:1516-1521, 1994
24. LeeF-Y, Colombato LA, Albillos A, et al: Administration of

N-omega-Nitro-L-arginine ameliorates portal-systemic shunting in
portal-hypertensive rats. Gastroenterology 105:1464-1470, 1993
26. LeeF-Y, Colombato LA, Albillos A, et al: Nw -nitro-L-arginine

administration corrects vasodilatation and systemic capillary hypotension,
and ameliorates plasma volume expansion and sodium retention in portal
hypertensive rats. Hepatology 17:84-90, 1993
27. Lieber
CS, Zimmon DS, Kessler RE, et al: Portal hypertension in
experimental alcoholic liver injury. Clin Res 24:478, 1976
28. Liebermann FL, Reynolds TB: Plasma volume in cirrhosis of the liver.
J Clin Invest 46:1279-1308, 1967
29. Lopez-Talavera JC, Merrill WW, Groszmann RJ: Tumor necrosis factor
alpha: A major contributor to the hyperdynamic circulation in prehepatic
portal-hypertensive rats. Gastroenterology 108:761-767, 1995
30. Lopez-Talavera JC, Cadelina GW, Olchowski J, et al: Thalidomide

inhibits tumor necrosis factor-alpha, decreases nitric oxide synthesis and
ameliorates the hyperdynamic circulatory syndrome in portal hypertensive

rats. Hepatology 23:1616-1621, 1996
31. Maddrey WC, Boyer JL, Sen NN: Plasma volume expansion in portal
hypertension. Johns Hopkins Med J 125:171-182, 1969
32. Mosca P, Lee F-Y, Kaumann AJ, et al: Pharmacology of

portal-systemic collaterals in portal hypertensive rats: Role of
endothelium. Am J Physiol 263:G544-50, 1992
33. McIndoe AH: Vascular lesions of portal cirrhosis. Arch Pathol 5:23-42,
1928
34. MittalMK, Gupta TK, Lee FY, et al: Nitric oxide modulates vascular
tone in normal rat liver. Am J Physiol 267:G416-G422, 1994
35. Murray JF, Dawson AM, Sherlock S: Circulatory changes in chronic

liver disease. Am J Med 24:358-367, 1958
36. OmasSH, Joh T, Benoit JN: Role of bile acids in splanchnic

hemodynamics in response to chronic portal hypertension. Dig Dis Sci
36:1243-1248, 1991
37. Orrego H, Blendis LM, Crossley IR, et al: Correlation of intrahepatic

pressure with collagen in the Disse space and hepatomegaly in humans and
in the rat. Gastroenterology 80:546-556, 1981
38. Pinzani M, Failli P, Ruocco C, et al: Fat-storing cells as liver-specific

pericytes: Spatial dynamics of agonist-stimulated intracellular calcium
transients. J Clin Invest 90:642-646, 1992
39. PizcuetaMP, Casamitjana R, Bosch J, et al: Decreased systemic

vascular sensitivity to norepinephrine in portal hypertensive rats: Role of
hyperglucagonemia. Am J Physiol 258:G191-G195, 1995
40. Polio
J, Groszmann RJ: Pharmacologic control of portal hypertension.
In Boyer J, Ockner R (eds): Progress in Liver Diseases. Philadelphia, WB
Saunders, 1993, pp 231-239
41. Popper
H, Zak FG: Pathological aspects of cirrhosis. Am J Med
24:593-625, 1958
42. Ramos OL, Saad F, Leser WP: Portal hemodynamics and liver cell
function in hepatic schistosomiasis. Gastroenterology 47:241-247, 1964
43. Reynolds TB, Hidmura R, Mitchell H, et al: Portal hypertension

without cirrhosis in alcoholic liver disease. Ann Intern Med 70:497-506,
1969
44. Rockey DC, Weisiger RA: Endothelin induced contractility of stellate

cell from normal and cirrhotic rat liver: Implication for regulation of portal
pressure and resistance. Hepatology 24:233-240, 1996

45. RubanyiGM: Endothelium-derived relaxing and contracting factors. J

Cell Biochem 46:27-36, 1990
12
46. Rudolph R, McClure WJ, Woodward M: Contractile fibroblasts in

chronic alcoholic cirrhosis. Gastroenterology 76:704-709, 1979
47. Schaffner
F, Popper H: Capilarization of the hepatic sinusoids in man.
48. Sieber
CC, Groszmann RJ: Nitric oxide mediates hyporeactivity to
vasopressors in mesenteric vessels of portal hypertensive rats.
49. SieberCC, Groszmann RJ: In vitro hyporeactivity to methoxamine in

portal hypertensive rats: Reversal by nitric oxide blockade. Am J Physiol
262:G996-1001, 1992
50. Sikuler
E, Kravetz D, Groszmann RJ: Evolution of portal hypertension
and mechanisms involved in its maintenance in a rat model. Am J Physiol
248:G618-G625, 1985
51. SitzmannJV, Campbell KA, Wu Y, et al: Effect of portosystemic

shunting on PgI2 and glucagon levels in humans. Ann Surg 217: 248-252,
1993
52. Vallance P, Moncada S: Hyperdynamic circulation in cirrhosis: A role

for nitric oxide? Lancet 337:776-778, 1991
53. vanLeeuwen DJ, Howe SC, Scheuer PJ, et al: Portal hypertension in
chronic hepatitis: Relationship to morphological changes. Gut 31:339-343,
1991
54. VaneJR, Anggard EE, Botting RM: Regulatory functions of vascular

endothelium. N Engl J Med 323:27-36, 1990
55. Wu Y, Burns C, Sitzmann JV: Effects of nitric oxide and

cyclooxygenase inhibition on splanchnic hemodynamics in portal
hypertension. Hepatology 18:1416-1421, 1993



PORTAL HYPERTENSION
PATHOPHYSIOLOGY OF PORTAL HYPERTENSION
Tarun K. Gupta MD
Lisa Chen MD
Roberto J. Groszmann MD, FRCP
From the Veteran's Affairs Medical Center, West Haven, Connecticut; Yale University School of
Medicine, New Haven, Connecticut; and the Bridgeport Hospital, Bridgeport, Connecticut (TKG)
A grasp of the biological mechanisms involved in the pathogenesis of portal hypertension is

essential to an understanding of the complications of chronic liver disease and to the development
of rational therapies. This article is an overview of the basic pathophysiologic mechanisms of the
splanchnic and systemic circulatory derangements that lead to portal hypertension and cause the
portosystemic collaterals to develop.
The pressure in a vascular system is directly proportional to the resistance (R) and flow (Q) in that
system.
Palpha Q R
Currently, both increased vascular resistance and increased splanchnic blood flow are believed to
play key roles in the production of elevated portal pressure (Fig. 1) . [40]

Roberto J. Groszmann, MD
Yale University School of Medicine
Hepatic Hemodynamic Laboratories/111H
VA Connecticut Healthcare System
950 Campbell Avenue
West Haven, CT 06516

Portal hypertension is associated with increased resistance to portal blood flow. Increased vascular
resistance is due to an increase in both intrahepatic and portocollateral resistance in comparison to the
low resistance of the normal liver.
Figure 1. Factors contributing to elevated portal pressure.
The main site of resistance in the normal liver is somewhat controversial. The hepatic sinusoids (where
stellate cells are present), terminal hepatic venules, and portal venules have been suggested as possible
sites of resistance; however, in view of the minimal contribution of the intrahepatic resistance to portal
pressure in the normal liver, this issue is of little importance. The flow into the normal portal system is
actively regulated by changes in vascular resistance at the level of splanchnic arterioles, and not by the
liver itself. [17]
In portal hypertensive syndromes, increased resistance to portal venous flow may be localized to
prehepatic, posthepatic, and intrahepatic (presinusoidal, sinusoidal, or postsinusoidal) sites. [13] [18] In
prehepatic and posthepatic portal hypertension, increased resistance is secondary to obstruction of portal
venous inflow or hepatic venous outflow, respectively. Unlike pre- and posthepatic portal hypertension,
the intrahepatic syndromes are more complex and rarely can be classified according to a single site of
resistance.
An early view of vascular resistance in cirrhotic livers postulated that portal hypertension is the
consequence of a vascular obliterative process, with scar tissue and regenerative nodules, occluding and
compressing vascular structure. [3] [33] [41] Thus, earlier understanding of intrahepatic portal hypertension
emphasized only the role of anatomic alterations leading to mechanical obstruction (irreversible
component) in the increased intrahepatic resistance; however, recently, it has been shown that there is an
increase in vascular tone (which normally is negligible in the normal liver) that imposes an additional
increase in resistance (reversible component). [5]
In intrahepatic portal hypertension, there may be several areas of obstruction, and as the disease
progresses, new sites may become involved.
For example, in hepatic schistosomiasis, the increased intrahepatic resistance results from granulomas
located in presinusoidal areas [4] ; however, in late stages, an elevated hepatic wedge venous pressure
gradient may be observed, reflecting increased sinusoidal resistance. [42] Chronic hepatitis seems to have
both presinusoidal and sinusoidal abnormalities that increasingly contribute to vascular resistance as the

lesion progresses toward cirrhosis. [53] In ethanol-induced liver disease, the intrahepatic resistance is
increased because of lesions in sinusoidal and postsinusoidal sites. The terminal hepatic vein fibrosis (or
sclerosis), encroachment on sinusoids by enlarged hepatocytes, and collagen deposition in the
perisinusoidal region or the space of Disse, resulting in sinusoidal narrowing, combine to produce
elevated portal pressure, even in a precirrhotic stage [27] [37] [43] [47] ; however, in advanced stages of
cirrhosis, regenerating nodules and pruning of the vascular tree contribute to the increased vascular
resistance. A variety of other nonalcoholic forms of liver diseases cause portal hypertension owing to
increased sinusoidal vascular resistance. In hepatic amyloidosis, resistance is increased because of the
deposition of amyloid in the space of Disse. [6] The capillarization process and occlusion of the
endothelial fenestrae are postulated to increase the resistance of passage across the endothelium;
however, the extent of their effect on resistance to blood flow is unknown.
The morphologic changes occurring in chronic liver diseases are undoubtedly the most important factor
involved in the increased intrahepatic resistance; however, recent data also suggest a role of functional
factors that lead to increased vascular tone, similar to what is seen in arterial hypertension. In chronic
liver disease, and also during acute liver injury, hepatic stellate cells acquire contractile properties and
may contribute to the dynamic modulation of intrahepatic resistance. [22] [38] These cells may act as
pericytes, a type of cell that has been shown to regulate blood flow in other organs. The hepatic stellate
cells, which are also the main source of collagen synthesis, may contribute to the regulation of hepatic
blood flow at the microcirculatory level. Stellate cells are strategically located in the sinusoids, with
perisinusoidal and interhepatocellular branching processes that contain actin-like filaments. They also
express the alpha smooth muscle actin gene, which is characteristic of vascular smooth muscle. The
characteristics of these cells make them similar to the myofibroblast. Myofibroblasts are intermediate in
structure between smooth muscle cells and fibroblasts. Myofibroblast-like cells have been shown to exist
in fibrous septa around the sinusoids and terminal hepatic venules in cirrhotic livers. [46] These cells are
postulated to have a role in the regulation of vascular resistance in the cirrhotic rat liver. [5]
The vascular endothelium synthesizes various vasodilators such as nitric oxide (NO), prostacyclins, and
hyperpolarizing factor, and the vasoconstrictors such as endothelins and prostanoids. [45] [54] These
vasoactive substances act in a paracrine fashion on the underlying vascular smooth muscle and modulate
vascular tone. Normal vascular tone is maintained by a delicate balance between these vasodilatory and
vasoconstrictive
substances. Perturbation of this balance leads to abnormal vascular tone. Increased vascular tone seen in
cirrhotic livers could be due to a deficit of endothelial vasodilators, an increase in the vasoconstrictors, or
a combination of both. NO is a potent endothelial vasodilator that has been shown to play an important
role in the modulation of intrahepatic vascular tone in normal livers. [34] Preliminary evidence from the
authors' laboratory suggests that there may be a deficit of NO in the cirrhotic intrahepatic
microcirculation, as seen in other hypertensive regional microcirculations. [19] Using the isolated rat liver
perfusion model, the authors have demonstrated that there is endothelial dysfunction in the intrahepatic
microcirculation of cirrhotic livers. [20] Endothelial dysfunction leads to an impaired release of
endothelial vasodilators, which may, in part, be responsible for the increased vascular tone observed in
cirrhotic livers. More recently, it has been shown that the stellate cells (myofibroblasts) from cirrhotic
livers exhibit enhanced response to endothelins. [44] An imbalance between endothelial vasodilators and

vasoconstrictors can affect the activated stellate cells (myofibroblasts) that modulate intrahepatic
vascular tone. It is possible that both a deficit of vasodilators and an increase in vasoconstrictors may be
responsible for the increased vascular tone. In summary, there are multiple factors that may lead to
increased intrahepatic resistance to portal blood flow. Some of these are irreversible, like fibrosis,
capillarization, and regenerating nodules, and some are quite dynamic, like the imbalance between
endothelial factors leading to increased vascular tone.
Portosystemic Collateral Resistance
Portosystemic collaterals develop as a result of portal hypertension. The development of collaterals is the
central pathophysiologic event that leads to variceal bleeding and portosystemic encephalopathy in
patients with portal hypertension. Although the collateral circulation begins as a consequence of portal
hypertension, it evolves into an important mediator of the circulatory derangements of portal
hypertension in its own right. Although the collaterals provide a route to decompress the hypertensive
portal system, their vascular resistance remains higher than the normal intrahepatic resistance. Hence,
portosystemic collaterals do not permit a portal decompression that will return portal pressure to normal.
The exact nature of the physiologic stimuli responsible for initiating collateral formation remains
somewhat controversial. Propranolol and clonidine have been shown to decrease portal pressure and
ameliorate portosystemic shunting, thus implicating increased portal pressure in the pathophysiology of
collateral formation. [21] Lee et al, [24] however, demonstrated that collateral formation could be
ameliorated by preventing an increase in splanchnic blood flow without any decrease in portal pressure.
Thus, increased portal pressure does not appear to be the only factor involved in the development of
collaterals in portal hypertensive states. Therefore, propranolol and clonidine may prevent collateral
formation
by decreasing flow, not by their effect on portal pressure. [21] On the other hand, Sikuler et al [50] studied
partial portal-vein-ligated (PVL) rats on postoperative day 3, and found no relationship between
portosystemic shunting and portal venous inflow. Therefore, the development of portosystemic shunting
does not appear to be dependent exclusively on either portal pressure or portal venous inflow. Studies in
the PVL rat model have shown that the blockade of NO synthesis with Nw -nitro-L-arginine (NNA) can
ameliorate portosystemic shunting, suggesting that increased NO may be responsible for the initial
collateralization of the portal system. [24] Increased NO could result from progressive increases in
splanchnic flow and shear stress, potent stimuli known to be present in portal hypertensive states. In
addition, development of a new portocollateral bed renders available a new endothelial surface capable
of producing NO. [24]
Mosca et al [32] determined the vascular responsiveness of collateral vessels to various vasoconstrictors
and vasodilators using an in situ perfused animal model. By administering norepinephrine,
5-hydroxytryptamine, isoproterenol, and acetylcholine in the presence and absence of their respective
blockers phentolamine, propranolol, and L-NNA, they determined that collateral vessels have functional
alpha-adrenoreceptors, 5-hydroxytryptamine receptors, and beta-adrenoreceptors. In addition, collateral
veins appear to be as sensitive to NO as arteries. [32]
Regardless of the specific initiating stimulus, in the beginning the mechanism of collateral formation

appears to be recruitment of preformed channels, as opposed to de novo formation of new vessels.

Portosystemic shunting can be detected almost immediately after the induction of portal hypertension, a
pattern of timing consistent with the rapid dilatation of preformed vessels. [32] Whether these preformed
vessels expand as the result of active dilatation or as a passive response to increased splanchnic flow or
portal pressure is unclear [32] ; however, neoformation of collateral vessels is probably a later event.
HYPERDYNAMIC CIRCULATION IN PORTAL HYPERTENSION

Chronic elevations in systemic and splanchnic blood flow have been documented as key elements of the
hyperdynamic circulatory state (HCS) of portal hypertensive animals and humans. Peripheral
vasodilatation initiates the development of the classic profile of decreased systemic vascular resistance
and mean arterial pressure plasma volume expansion, elevated splanchnic blood flow, and elevated
cardiac index that characterize this state. [9] At least three mechanisms are believed to contribute to this
peripheral vasodilatation: (1) increased concentrations of circulating vasodilators, (2) increased
endothelial production of local
vasodilators, and (3) decreased vascular responsiveness to endogenous vasoconstrictors. This last
mechanism is probably due to the effects of the first two components. Increased Circulating
Vasodilators
In portal hypertensive states there is an increase in both endothelium-dependent and independent
vasodilators. Possible causes for increased circulatory concentrations of vasodilatory substances include
increased production, decreased catabolism secondary to impaired hepatic function, and portosystemic
shunting.
Circulating bile acids, routinely cleared by the liver, are present in elevated concentrations when liver
function is impaired. Bile acid depletion has been shown to be associated with a decrease in splanchnic
hyperemia. Experimental evidence, however, suggests that an increase in circulating bile acids is not
essential for maintaining the HCS in portal hypertension. [14] More specifically, cholestyramine-induced
reduction of bile acids to concentrations seen in placebo-treated controls did not ameliorate the
hemodynamic changes of the HCS in portal hypertensive animals. [14]
Elevated concentrations of circulating glucagon have also been documented in both animals and humans
with portal hypertension. [16] [39] [51] Rats with portal hypertension induced experimentally by either PVL
or carbon tetrachloride induced cirrhosis had significantly higher glucagon and insulin concentrations
compared with control rats. [16] [39] In addition, pancreatic islet isolates from these animals exhibited a
significantly higher glucagon secretion rate in response to glucose and arginine administration compared
with controls, although insulin secretion appeared to be impaired. [16] Intraoperative glucagon
concentrations measured from the portal vein and inferior vena cava are significantly increased after
surgical portosystemic shunting in patients with Budd-Chiari syndrome and slightly, but not
significantly, in cirrhotic patients. [51] Glucagon concentrations in this study, however, did not correlate
with portal pressure. Therefore, whether glucagon plays a role in the production of the HCS remains
unclear. [51]

Increased Endothelial Production of Vasodilators

Recently, increasing evidence has pointed toward a major role for the endothelium in the maintenance of
basal vascular tone and the development of local and generalized vasodilatation in portal hypertension.
The endothelium produces at least two substances that are known to contribute to the development of
systemic and splanchnic vasodilatation in portal hypertension: NO and prostaglandins. [7]
Nitric Oxide (NO).
NO, previously known as endothelial-derived relaxing factor, is synthesized from L-arginine by the
enzyme NO synthase (NOS), which has constitutive and inducible forms in different cell types. NO
mediates its potent vasodilatory action on smooth muscle
cells through soluble guanylate cyclase. Evidence exists that elevated production of NO is essential to the
development of the hyperdynamic circulation. Treatment with NNA, a physiologically inactive substrate
analogue, and thus a competitive inhibitor of NOS, prevented the development of peripheral
vasodilatation, decreased systemic arterial pressure, plasma volume expansion, and sodium retention in a
PVL rat model.25 Similarly, chronic continuous administration of L-NAME, another inhibitor of NOS,
has recently been shown to delay splanchnic vasodilatation, increased splanchnic blood flow, and
development of collaterals in PVL rats. [12] The vasodilatory effects of NO in portal hypertension are not
limited to the splanchnic circulation as they also affect the systemic circulation; in PVL rats, isolated
aortic rings demonstrated increased relaxation to the vasodilator acetylcholine (endothelial agonist)
compared with sham controls. This increased response to acetylcholine was partly reversed with
L-NAME. [23]
Whether production of NO is a primary stimulus in the development of vasodilatation or a subsequent

phenomenon that results from shear stress and secondarily contributes to increased flow has yet to be
determined. Vallance et al [52] hypothesized that chronic endotoxemia in portal hypertension and cirrhosis
may up-regulate inducible NOS (iNOS), thus causing the increased NO production that leads to
splanchnic vasodilation. [52] Tumor necrosis factor-alpha (TNF-alpha) may play an important role in the
vasodilatation by up-regulating iNOS. Treatment with anti-TNF-alpha polyclonal antibodies and
thalidomide have been shown to ameliorate the development of the hyperdynamic circulation in the PVL
rat model of portal hypertension. [29] [30] The relative contributions of the constitutive and inducible forms
of NOS are still under investigation.
Prostaglandins.
Several studies in animals and humans have implicated increased endothelial production of
prostaglandins as a cause of splanchnic vasodilatation in portal hypertensive states. Portal vein
concentrations of prostacyclin (PGI2 ), for example, have been found to be elevated in PVL rabbits and
rats, as well as in patients with cirrhosis and Budd-Chiari syndrome. These PGI2 concentrations have
correlated with portal pressure. [55] In addition, increased concentrations of the prostaglandin metabolite
2,3-dinor-6 keto PGF1alpha have been observed in cirrhotic patients, and elevated gastric PGE2 synthesis
has been seen in cirrhotic humans with severe portal hypertensive gastropathy. [7]

In the PVL animal model of portal hypertension, increased prostaglandin production appears to have a
more prominent role in rabbits than in rats. Inhibition of prostaglandin synthesis through pharmacologic
cyclooxygenase blockade has been shown to prevent development of the HCS in portal hypertensive
rabbits. [55] Wu et al [55] measured splanchnic flow and portal pressure in PVL rabbits in the presence and
absence of the respective inhibitors of cyclooxygenase and NOS, indomethacin, and L-NAME; their
results were consistent with mediation of splanchnic hyperemia predominantly by a prostaglandin,
possibly prostacyclin, with a limited role for NO as a mediator of basal vascular tone. In addition, the
effects of NO blockade with L-NAME, and reversal of that
blockade with the naturally occurring substrate for NO synthase, L-arginine, were not significantly
different between normal and PVL rabbits, thus implying that at least in rabbits, increased NO
production may not be responsible for the HCS associated with portal hypertension. [55]
Similar studies performed in PVL rats, however, suggest that both NO and prostaglandins contribute to
gastric hyperemia. [7] In both rabbits and rats, the two endothelial products appear to cause local
vasodilatation through distinct mechanisms. Relaxation of vascular smooth muscle in response to
acetylcholine, which depends on increased NO production from an intact endothelium, was not prevented
by cyclooxygenase blockade. [55] In PVL rats, hemodynamic measurements in the presence or absence of
indomethacin, and L-NAME demonstrated that prostaglandins and NO do not appear to act
synergistically. [7] In addition, an in vitro endothelial study suggests that prostacyclin release may be
markedly suppressed by NO, [11] and other researchers have found that NO inhibition may be associated
with increased prostacyclin production. [8]
Thus, both NO and prostaglandins appear to act through separate pathways to contribute to the
vasodilatation that leads to increased splanchnic flow in portal hypertension, although the relative
contribution of each may vary among different species.
Decreased Response to Vasoconstrictors
Basal vascular tone is regulated by the complex balance between endogenous vasodilators and
vasoconstrictors. A blunted response to vasoconstrictors, therefore, should also contribute to
vasodilatation and, subsequently, hyperdynamic flow. In portal hypertensive states, in vivo
hyporesponsiveness to the endogenous vasopressors norepinephrine, arginine vasopressin, and
angiotensin II, has been reported to contribute to the HCS. [48]
This hyporeactivity to vasopressors appears to be mediated largely by NO. In portal hypertensive rats,
inhibition of NO in isolated perfused superior mesenteric artery beds has been shown to prevent the
development of vascular hyporeactivity to the endogenous vasoconstrictors norepinephrine and
vasopressin, [48] the exogenous alpha-agonist methoxamine, and the receptor-independent vasoconstrictor
potassium chloride. [49] These observations are consistent with the previous hypothesis that the decreased
response to vasoconstrictors in portal hypertension is mediated by receptor-independent mechanisms. [51]
In portal hypertensive rats, a role for prostaglandins in hyporesponsiveness to vasoconstrictors has not
been substantiated. In fact, cyclooxygenase inhibition with indomethacin did not ameliorate vascular
hyporeactivity in superior mesenteric artery preparations in partial PVL rats. [49] Therefore, at least in the
rat model of portal hypertension, NO appears to cause the vascular hyporeactivity to endogenous and

exogenous vasoconstrictors that contributes to the generalized vasodilation seen in the HCS.
Plasma Volume
The hyperdynamic circulation is mediated in part by vasodilatation, but this alone is not sufficient to
cause the circulation to become hyperdynamic. Plasma volume expansion has been recognized in portal
hypertension for many years. [28] [31] [35] In the PVL rat model of portal hypertension, plasma volume
failed to expand in animals on a sodium-restricted diet as compared to those on normal diet. Moreover,
the fully developed hyperdynamic circulation was found to be nearly reversible by sodium restriction. [15]
Subsequent studies have demonstrated that in the PVL rats, vasodilatation, expansion of plasma volume
by sodium retention, and development of the hyperdynamic circulation follow each other in a stepwise
fashion.1,10 Vascular resistance in the systemic circulation drops significantly within 1 day of partial
PVL, followed on day 2 by an increase in plasma volume and a parallel progressive increase in systemic
and regional blood flows. The fully expanded plasma volume is observed on day 4, and coincides with a
maximally hyperkinetic cardiac index. These studies provide important evidence for the existence of
several events in the pathogenesis of the hyperdynamic circulation. These include initial vasodilatation
(induced by humoral and local endothelial factors) followed by plasma volume expansion. The authors
have demonstrated that both octreotide, which supresses the secretion of vasodilatory peptides, and NO
blockers decrease renal sodium retention and plasma volume expansion by diminishing vasodilatation,
thereby preventing the full expression of the hyperdynamic circulation. [2] [26]
The studies that examine the role of vasodilatation and plasma volume expansion in the hyperdynamic
circulation provide support for the peripheral vasodilatation hypothesis. According to this hypothesis,
portal hypertension leads to a relative hypovolemia induced by the dilatation of the systemic and
splanchnic circulations. This results in underfilling of the systemic vascular space and a decrease in the
central blood volume. This in turn leads to activation of the sympathetic nervous system and the
renin-angiotensin system, and a release of antidiuretic hormone. Mediators from these systems result in
sodium and water retention by the kidneys, resulting in increased plasma volume.
SUMMARY
Portal hypertension is a common clinical syndrome associated with chronic liver diseases, and is
characterized by a pathological increase in portal pressure that leads to the formation of portosystemic
collaterals resulting in shunting of portal blood into the systemic circulation. The increase in portal
pressure is due to an increase in vascular resistance and an elevated portal blood flow. The site of
increased resistance is variable, and dependent upon the disease process. The site of relative obstruction
may be prehepatic, hepatic, or posthepatic. There are several intrahepatic lesions that lead to increased
resistance. Some of these
10
lesions are irreversible, like fibrosis, regenerating nodules, and capillarization of the space of Disse;
however, there is a functional component, increased vascular tone, which contributes to increased
intrahepatic resistance and is potentially reversible. Another important factor contributing to the

increased portal pressure is elevated portal blood flow. Peripheral vasodilatation initiates the classical
profile of decreased systemic resistance, expanded plasma volume, elevated splanchnic blood flow, and
elevated cardiac index, which characterize the hyperdynamic circulatory state. This hyperdynamic
circulation is responsible for various complications of portal hypertension.
References
1. Albillos
A, Colombato LA, Groszmann RJ: Vasodilatation and sodium retention in prehepatic portal hypertension.
2. Albillos A, Colombato LA, Lee FY, et al: Octreotide ameliorates vasodilatation and sodium retention in portal
hypertensive rats. Gastroenterology 104:575-57, 1993
3. Baldus
WP, Hoffbauer FW: Vascular changes in the cirrhotic liver as studied by injection technic. Am J Dig Dis
8:689-692, 1963
4. Beker S, Valencia-Parparcen J: Portal hypertension syndrome: A comparative analysis of bilharzial fibrosis and hepatic
cirrhosis. Am J Dig Dis 13:1047-1054, 1968
5. Bhathal PS, Grossman HJ: Reduction of the increased portal vascular resistance of the isolated perfused cirrhotic rat
liver by vasodilators. J Hepatol 1:325-337, 1985
6. Brion E, Brenard R, Pariente EA, et al: Sinusoidal portal hypertensin in hepatic amyloidosis. Gut 32:227-230, 1991
7. Casadevall M, Panes J, Pique JM, et al: Involvement of nitric oxide and prostaglandins in gastric mucosal hyperemia of
portal-hypertensive anesthetized rats. Hepatology 18:628-34, 1993
8. Claria J, Jimenez W, Ros J, et al: Pathogenesis of arterial hypotension in cirrhotic rats with ascites: Role of endogenous
nitric oxide. Hepatology 15:343-349, 1993
9. Colombato LA, Albillos A, Groszmann RJ: Temporal relationship of peripheral vasodilatation, plasma volume
expansion and the hyperdynamic circulatory state in portal-hypertensive rats. Hepatology 15:323-328, 1991
11. Doni MG, Whittle BJ, Palmer RM, et al: Action of nitric oxide on the release of prostacyclin from bovine endothelial
cells in culture. Eur J Pharm 151:1925, 1988
12. Garcia-Pagan JC, Fernandez M, Bernadich C, et al: Effect of continued NO inhibition on portal hypertensive syndrome
after portal vein stenosis in rat. Am J Physiol 267:G984-90, 1994
13. Genecin P, Groszmann RJ: Portal hypertension. In Schiff E, Schiff L (eds): Diseases of the Liver. Philadelphia, JB
Lippincott, 1993
14. Genecin
P, Polio J, Ferraioli G, et al: Bile acids do not mediate the hyperdynamic circulation in portal hypertensive rats.
Am J Physiol 259:G21-G25, 1990
15. Genecin P, Polio J, Groszmann RJ: Na restriction blunts expansion of plasma volume and ameliorates hyperdynamic
circulation in portal hypertension. Am J Physiol 259:G498-G503, 1990
16. Gomis R, Fernandez-Alvarez J, Pizcueta P, et al: Impaired function of pancreatic islets from rats with portal
hypertension resulting from cirrhosis and partial portal vein ligation. Hepatology 19:1257-1261, 1994

17. Greenway CV, Stark RD: Hepatic vascular bed. Physiol Rev 51:23-65, 1971
18. Groszmann RJ, Atterbury CE: The pathophysiology of portal hypertension: A basis for classification. Semin Liver Dis
2:177-186, 1982
19. GuptaTK, Groszmann RJ: Administration of L-arginine, the physiological precursor of nitric oxide, reduces portal
perfusion pressure and ameliorates hepatic vascular hyperreactivity in experimental cirrhosis. Hepatology 20:200A, 1994
11
20. Gupta
TK, Chung M, Sessa WC, et al: Impaired endothelial function in the intrahepatic circulation in cirrhosis.
21. Halvorsen JF, Myking AO: The porto-systemic colleteral pattern in the rat. Eur Surg Res 6:183-195, 1974
22. Johnson SJ, Hines JE, Burt AD: Phenotypic modulation of perisinusoidal cells following acute liver injury: A
quantitative analysis. Int J Exp Pathol 73:765-772, 1992
23. KaratapanisS, McCormick PA, Kakad S, et al: Alteration in vascular reactivity in isolated aortic rings from portal
vein-constricted rats. Hepatology 20:1516-1521, 1994
24. LeeF-Y, Colombato LA, Albillos A, et al: Administration of N-omega-Nitro-L-arginine ameliorates portal-systemic
shunting in portal-hypertensive rats. Gastroenterology 105:1464-1470, 1993
26. Lee F-Y, Colombato LA, Albillos A, et al: Nw -nitro-L-arginine administration corrects vasodilatation and systemic
capillary hypotension, and ameliorates plasma volume expansion and sodium retention in portal hypertensive rats.
Hepatology 17:84-90, 1993
27. Lieber CS, Zimmon DS, Kessler RE, et al: Portal hypertension in experimental alcoholic liver injury. Clin Res 24:478,
1976
28. Liebermann FL, Reynolds TB: Plasma volume in cirrhosis of the liver. J Clin Invest 46:1279-1308, 1967
29. Lopez-Talavera
JC, Merrill WW, Groszmann RJ: Tumor necrosis factor alpha: A major contributor to the
hyperdynamic circulation in prehepatic portal-hypertensive rats. Gastroenterology 108:761-767, 1995
30. Lopez-Talavera JC, Cadelina GW, Olchowski J, et al: Thalidomide inhibits tumor necrosis factor-alpha, decreases nitric
oxide synthesis and ameliorates the hyperdynamic circulatory syndrome in portal hypertensive rats. Hepatology
23:1616-1621, 1996
31. Maddrey WC, Boyer JL, Sen NN: Plasma volume expansion in portal hypertension. Johns Hopkins Med J 125:171-182,
1969
32. Mosca P, Lee F-Y, Kaumann AJ, et al: Pharmacology of portal-systemic collaterals in portal hypertensive rats: Role of
endothelium. Am J Physiol 263:G544-50, 1992
33. McIndoe AH: Vascular lesions of portal cirrhosis. Arch Pathol 5:23-42, 1928

34. Mittal
MK, Gupta TK, Lee FY, et al: Nitric oxide modulates vascular tone in normal rat liver. Am J Physiol
267:G416-G422, 1994
35. Murray JF, Dawson AM, Sherlock S: Circulatory changes in chronic liver disease. Am J Med 24:358-367, 1958
36. Omas SH, Joh T, Benoit JN: Role of bile acids in splanchnic hemodynamics in response to chronic portal hypertension.
Dig Dis Sci 36:1243-1248, 1991
37. Orrego
H, Blendis LM, Crossley IR, et al: Correlation of intrahepatic pressure with collagen in the Disse space and
hepatomegaly in humans and in the rat. Gastroenterology 80:546-556, 1981
38. Pinzani M, Failli P, Ruocco C, et al: Fat-storing cells as liver-specific pericytes: Spatial dynamics of agonist-stimulated
intracellular calcium transients. J Clin Invest 90:642-646, 1992
39. Pizcueta
MP, Casamitjana R, Bosch J, et al: Decreased systemic vascular sensitivity to norepinephrine in portal
hypertensive rats: Role of hyperglucagonemia. Am J Physiol 258:G191-G195, 1995
40. Polio J, Groszmann RJ: Pharmacologic control of portal hypertension. In Boyer J, Ockner R (eds): Progress in Liver
Diseases. Philadelphia, WB Saunders, 1993, pp 231-239
41. Popper H, Zak FG: Pathological aspects of cirrhosis. Am J Med 24:593-625, 1958
42. Ramos OL, Saad F, Leser WP: Portal hemodynamics and liver cell function in hepatic schistosomiasis.
43. Reynolds
TB, Hidmura R, Mitchell H, et al: Portal hypertension without cirrhosis in alcoholic liver disease. Ann Intern
Med 70:497-506, 1969
44. Rockey DC, Weisiger RA: Endothelin induced contractility of stellate cell from normal and cirrhotic rat liver:
Implication for regulation of portal pressure and resistance. Hepatology 24:233-240, 1996
45. Rubanyi GM: Endothelium-derived relaxing and contracting factors. J Cell Biochem 46:27-36, 1990
12
46. RudolphR, McClure WJ, Woodward M: Contractile fibroblasts in chronic alcoholic cirrhosis. Gastroenterology
76:704-709, 1979
47. Schaffner F, Popper H: Capilarization of the hepatic sinusoids in man. Gastroenterology 44:239-251, 1963
48. Sieber
CC, Groszmann RJ: Nitric oxide mediates hyporeactivity to vasopressors in mesenteric vessels of portal
hypertensive rats. Gastroenterology 103:235-239, 1992
49. Sieber
CC, Groszmann RJ: In vitro hyporeactivity to methoxamine in portal hypertensive rats: Reversal by nitric oxide
blockade. Am J Physiol 262:G996-1001, 1992
50. Sikuler
E, Kravetz D, Groszmann RJ: Evolution of portal hypertension and mechanisms involved in its maintenance in
a rat model. Am J Physiol 248:G618-G625, 1985

51. SitzmannJV, Campbell KA, Wu Y, et al: Effect of portosystemic shunting on PgI2 and glucagon levels in humans. Ann
Surg 217: 248-252, 1993
52. Vallance P, Moncada S: Hyperdynamic circulation in cirrhosis: A role for nitric oxide? Lancet 337:776-778, 1991
53. van
Leeuwen DJ, Howe SC, Scheuer PJ, et al: Portal hypertension in chronic hepatitis: Relationship to morphological
changes. Gut 31:339-343, 1991
54. Vane JR, Anggard EE, Botting RM: Regulatory functions of vascular endothelium. N Engl J Med 323:27-36, 1990
55. Wu Y, Burns C, Sitzmann JV: Effects of nitric oxide and cyclooxygenase inhibition on splanchnic hemodynamics in
portal hypertension. Hepatology 18:1416-1421, 1993



Figure 1. Factors contributing to elevated portal pressure.

Bookmark URL: /das/journal/view/N/10888314?ja=97653&PAGE=I3.fig&ANCHOR=top
http://home.mdconsult.com/das/article/body/jorg=journal&source=&sp=10888314&sid=0/N/97653/I3.fig [21/04/2000 23:54:56]

April 21, 2000
Conde Petra
Is portal hypertension due to liver cirrhosis a

major factor in the development of portal
hypertensive gastropathy?
Citation
Bibliographic Data
Abstract
Indexing Data
Is portal hypertension due to liver
Copyright Notice and Disclaimer cirrhosis a major factor in the
development of portal hypertensive
Find More Articles Like This gastropathy?
Bayraktar Y - Am J Gastroenterol - 1996 Mar; 91(3): 554-8
Full Text
NLM Citation ID:
Frontmatter 96209591
INTRODUCTION
Full Source Title:
MATERIALS AND METHODS American Journal of Gastroenterology
Subjects Publication Type:
Journal Article
Procedures
Language:
Endoscopic definitions and studies
English
Liver biopsies
Author Affiliation:
Histology Department of Gastroenterology, School of Medicine,
Hacettepe University, Ankara, Turkey.
Authors:
RESULTS
Bayraktar Y; Balkanci F; Uzunalimoglu B; Gokoz A; Koseoglu
Endoscopy T; Batman F; Gurakar A; Van Thiel DH; Kayhan B
Splenoportography Abstract:
OBJECTIVES: The gastric mucosa of patients with portal
DISCUSSION hypertension frequently manifests changes in its appearance
REFERENCES that are readily identifiable by endoscopy. Many of these can be
sources of bleeding, and some imply the presence of systemic
About the Publication disease. Although portal hypertension is critical in development
of portal hypertensive gastropathy (PHG), the role that other
factors might play in its pathogenesis is uncertain. METHODS:
Four groups of subjects were studied prospectively: 37 with
portal hypertension due to cirrhosis, 26 noncirrhotic subjects
with portal hypertension due to extrahepatic portal vein

obstruction (PVO), nine cirrhotic patients with extrahepatic
PVO, and 57 control subjects. The diagnosis in each case was
based on a combination of clinical data, needle liver biopsy,
ultrasonography, splenoportography, and upper GI endoscopy.
RESULTS: Snake skin, scarlatina rash, diffuse hyperemia, and
diffuse bleeding were frequent endoscopic gastric findings in
cirrhotic patients. These findings were seen less frequently in
noncirrhotic patients with portal hypertension due to PVO than
in cirrhotic patients (p< 0.0001). The highest incidence was
seen in cirrhotic patients with PVO (P< 0.001). Positive
correlations existed among the endoscopic findings, the clinical
estimate of the cirrhosis severity (Child-Pugh grade), and the
size and appearance of esophageal varices (Beppu score). No
endoscopic findings of the gastric mucosa enabled one to
distinguish between groups. Hypergastrinemia was present in
cirrhotics with and without PVO but not in noncirrhotic patients
with portal hypertension resulting from isolated PVO.
CONCLUSION: These findings suggest that the endoscopic
findings of PHG are affected by the severity of the underlying
liver disease and the presence or absence of coexisting PVO.
There is no association between PHG and the presence of
gastric varices. Thus, the development of the gastric lesions
characteristic of PHG requires not only portal hypertension but
also some other consequence of parenchymal liver disease.
Major Subjects:
Esophageal and Gastric Varices / Diagnosis / * Etiology
Hypertension, Portal / Complications / Diagnosis / *

Etiology
Adult
Biopsy
Endoscopy, Digestive System
Fasting / Blood
Female
Gastrins / Blood
Human
Liver / Pathology / Ultrasonography
Liver Cirrhosis / Complications / Diagnosis
Male
Portography


(Gastrins)

American Journal of Gastroenterology

Volume 91 Number 3 March 1996
Copyright 1996 American College of Gastroenterology
554
Original contributions
Is Portal Hypertension Due to Liver Cirrhosis a Major Factor in the

Development of Portal Hypertensive Gastropathy?
Yusuf Bayraktar M.D.

Ferhun Balkanci M.D.
Bedri Uzunalimoglu M.D.
Aytac Gokoz M.D.
Tankut Koseoglu M.D.
Figen Batman M.D.
Ahmet Gurakar M.D.
David H. Van Thiel M.D.
Burhan Kayhan M.D.
Departments of Gastroenterology, Pathology, and Radiology, School of Medicine, Hacettepe University,
Ankara, Turkey and Oklahoma Transplant Institute, Baptist Medical Center of Oklahoma,
Oklahoma City, Oklahoma
Objectives: The gastric mucosa of patients with portal hypertension frequently manifests changes
in its appearance that are readily identifiable by endoscopy. Many of these can be sources of
bleeding, and some imply the presence of systemic disease. Although portal hypertension is critical
in development of portal hypertensive gastropathy (PHG), the role that other factors might play in
its pathogenesis is uncertain. Methods: Four groups of subjects were studied prospectively: 37 with
portal hypertension due to cirrhosis, 26 noncirrhotic subjects with portal hypertension due to
extrahepatic portal vein obstruction (PVO), nine cirrhotic patients with extrahepatic PVO, and 57
control subjects. The diagnosis in each case was based on a combination of clinical data, needle
liver biopsy, ultrasonography, splenoportography, and upper GI endoscopy. Results: Snake skin,
scarlatina rash, diffuse hyperemia, and diffuse bleeding were frequent endoscopic gastric findings
in cirrhotic patients. These findings were seen less frequently in noncirrhotic patients with portal
hypertension due to PVO than in cirrhotic patients ( p < 0.0001). The highest incidence was seen in
cirrhotic patients with PVO ( p < 0.0001). Positive correlations existed among the endoscopic
findings, the clinical estimate of the cirrhosis severity (Child-Pugh grade), and the size and
appearance of esophageal varices (Beppu score). No endoscopic findings of the gastric mucosa
http://home.mdconsult.com/das/article/body/jorg=journal&source=MI&sp=721548&sid=0/N/1975/1.html (1 of 8) [21/04/2000 23:55:59]

enabled one to distinguish between groups. Hypergastrinemia was present in cirrhotics with and
without PVO but not in noncirrhotic patients with portal hypertension resulting from isolated
PVO. Conclusion: These findings suggest that the endoscopic findings of PHG are affected by the
severity of the underlying liver disease and the presence or absence of coexisting PVO. There is no
association between PHG and the presence of gastric varices. Thus, the development of the gastric
lesions characteristic of PHG requires not only portal hypertension but also some other
consequence of parenchymal liver disease.
Reprint requests and correspondence: Yusuf Bayraktar, M.D., Oklahoma Transplant Institute, Baptist Medical Center of
Oklahoma, 3300 Northwest Expressway, Oklahoma City, OK 73112.
Received Aug. 5, 1995; accepted Nov. 1, 1995.
INTRODUCTION
Regardless of etiology, the stomach of individuals with portal hypertension frequently demonstrates endoscopic findings
that have been recognized as the entity termed "portal hypertensive gastropathy" (PHG). The recognition of these gastric
changes is a result of the observations that the stomach as well as the esophagus may be an important source of upper GI
hemorrhage in individuals with portal hypertension. In the past, these mucosal changes have been identified as erosive
gastritis [1] [2] , congestive gastropathy [3] , and a gastric mosaic pattern and more recently as PHG [4] . Little is known
about the factors responsible for these mucosal findings although most investigators believe that they are a consequence of
portal hypertension [4] [5] . Statistical significant relationships among the presence of esophageal varices, large varices, and
prior sclerotherapy and PHG have been reported [6] . The incidence and clinical appearance of the gastric mucosal
endoscopic findings ascribed to portal hypertension have been the object of several studies [4] [5] [6] [7] [8] . The prevalence
of PHG in various reported series ranges from 50 to 100% of individuals examined [4] [5] [6] [7] . Most studies assessing
PHG have been limited to individuals with cirrhosis [4] [5] [6] . Few have compared the gastric findings in noncirrhotics
with portal hypertension with those of cirrhotic individuals with portal hypertension [6] [7] .
Thus, the aim of this study was to determine whether the severity of liver disease, the cause for the portal hypertension,
and the presence or absence of gastric varices determine the endoscopic appearance of PHG and the relationships that exist
between each. In addition, the fasting serum gastrin levels in each of the four study populations were defined.
MATERIALS AND METHODS

Subjects
A total of 72 individuals with portal hypertension (PHT) and 57 control subjects were enrolled in this study. Forty six of
these 72 patients had liver cirrhosis and were selected from a total of 457 cirrhotics seen at Hacettepe University
555
Hospital between January 1987 and June 1994. Nine of the 46 cirrhotics additionally had portal vein thrombosis. Those 46
of a total of 457 cirrhotics studied were selected according to the exclusionary critieria. The list of required studies
included: liver biopsy confirming the diagnosis of cirrhosis, ultrasound and CT scan of the liver, and an endoscopy of the
upper GI tract with biopsy of the stomach. The exclusionary criteria were: any clinical or historical evidence for prior GI
bleeding; use of a drug known to produce gastric mucosal damage; and prior sclerotherapy. In the remaining 26 of the 72
patients with portal hypertension, isolated portal vein thrombosis without any evidence of liver disease was diagnosed.
These 129 subjects (72 patients and 57 controls) were divided into four subgroups as follows: group 1 consisted of 37
patients (23 male, 14 female; mean age 43 yr) with portal hypertension due to cirrhosis occurring primarily as a result of
viral hepatitis; group 2 consisted of 26 patients (15 male, 11 female; mean age 37 yr) with portal hypertension due to
extrahepatic portal vein thrombosis without coexisting cirrhosis; group 3 consisted of nine patients (three female, six male;
mean age 40 yr) with cirrhosis complicated by extrahepatic portal vein thrombosis. The controls (group 4) consisted of 57

patients (26 female, 31 male; mean age 42 yr) who complained of various upper abdominal symptoms but had no history
or clinical evidence of hepatic disease or portal hypertension. Abdominal discomfort, fullness, belching, burning, and
bloating were the major symptoms of this control group.
Procedures
All patients were endoscoped with either an Olympus model Q10 or Q20 endoscope. All of the endoscopies were done by
an endoscopist who was blinded to the nature of the study.
The diagnosis of cirrhosis in the subjects in groups 1 and 3 was based on a combination of clinical findings, laboratory
tests, and a liver biopsy. The presence or absence of portal hypertension was defined in each case using the criteria
proposed by Pagliaro et al.[9] . These include: 1) two or more of the following: splenomegaly (>13 cm of longitudinal axis
on ultrasonography); 2) thrombocytopenia (<100,000/mul); 3) leukopenia (<4,000/mul); 4) portal vein diameter greater
than 14 mm by ultrasonography; and 5) esophageal varices seen at endoscopy. Other signs of portal hypertension, such as
the presence of ascites and gastric varices, were seen frequently in both groups. All patient groups, except for the controls
(group 4), had esophageal varices demonstrated endoscopically and a portal vein greater than 14 mm in diameter if the
portal vein was found to be patent. Wedged hepatic venous pressures were not measured in any case. The severity of the
esophageal varices was recorded using Beppu's score [10] . Particular attention was paid to the following five factors: 1)
color of the varices (white or blue); 2) red color sign (positive or negative); 3) form of the varices (F1, straight varices; F2,
enlarged tortuous varices; F3, very large convoluted varices); 4) location (L1, locus inferior or lower third of esophagus;
L2, medialis or lower two-thirds; and/or L3, superior or full extent; and, 5) esophagitis (present or absent).
During this same time period, a total of 44 patients (27 male, 15 female; mean age 36 yr) with PVO diagnosed
ultrasonographically was seen. These individuals were studied prospectively. The diagnosis was confirmed by either
splenoportography or arterial portography with digital subtraction angiography in all 44. Digital subtraction angiography
of the superior mesenteric artery was obtained after selective injection of 35 ml of 50% dilute contrast medium at a rate of
8 ml/s. Images were obtained for 25 s at a rate of one frame per second during the arterial, arteriocapillary, and venous
phases of each study.
Sonographically guided liver biopsies were obtained in 38 of the 44 (86%) subjects with portal vein obstruction (PVO). In
addition to routine blood and urine analyses, serial liver injury tests and tumor markers were obtained in all. In 26 (59%)
cases, the liver was normal histologically as well as biochemically and serologically (15 male, 11 female; mean age 37 yr).
No etiology for the portal vein thrombosis was evident in these 26 cases. Nine (20%) patients with PVO had histological
evidence for cirrhosis. These nine patients were analyzed separately. The endoscopic findings of these nine patients were
compared with those present in individuals with portal hypertension but without cirrhosis.
Endoscopic definitions and studies
The terminology of the World Society of Digestive Endoscopy was used [11] . The following specific descriptions were
used: 1) confluent measles-like appearance; 2) hyperemic areas of differing sizes, of irregular form, and raised above the
surrounding mucosa [12] ; 3) snake skin mucosa consisting of a fine, white, irregular reticular pattern with different shaped
areas of edematous red mucosa; and 4) scarlatina rash consisting of a fine, pink speckling of the mucosa. The duodenal
mucosa was evaluated separately for evidence of a portal hypertensive enteropathy (edema with mucosal thickening and
loss of color) on a scale ranging from normal (0) to severe [4] .
Liver biopsies
All cirrhotic and noncirrhotic subjects underwent an ultrasonographically guided liver biopsy. The right lobe of the liver
was biopsied in each case.
Histology
All biopsy specimens were fixed in buffered formalin, processed, and stained with hematoxylin and eosin.

All values are reported as means SD. chi2 analysis and Fisher's exact test were used for the statistical analysis. A p value
< 0.05 was considered significant.
556
TABLE 1 -- Characteristics of Patients and Control Subjects

Portal
Portal Portal Hypertension
Hypertension Hypertension due to C
due to C due to PVO and PVO Control
N patients 37 26 9 57
Male 23 11 6 31
Female 14 15 3 26
Mean age (yr) 43 37 40 42
Cause of C (n)
Hepatitis B virus 23 -- 7 --
Hepatitis C virus 9 -- 2 --
Primary biliary cirrhosis 1 -- -- --
Cryptogenic 4 -- -- --
Child's grade (n)
B 21 -- 2 --
C 16 -- 7 --
C, cirrhosis; PVO, portal vein obstruction.
RESULTS
The clinical characteristics of the population studied are reported in Table 1 .
Endoscopy
The frequency of various endoscopic findings in the stomach, duodenum, and esophagus of the patients studied is reported
in Table 2 . The endoscopic descriptors used in patients with liver cirrhosis were snake skin (56.8%), scarlatina rash
(35.1%), petechia (37.8%), and hyperemia (54%). The incidence of fundic varices (21%) was lower in those with cirrhosis
than in those without cirrhosis (46%) ( p < 0.001). Both in terms of size and extension, the esophageal varices were more
prominent in patients with PVO compared with cirrhotic patients ( p < 0.001).

The finding of red signs, which is thought to be predictive factor for bleeding, was present significantly less often in
patients with PVO compared with those with cirrhosis ( p < 0.001). Cirrhotic patients with PVO had the largest and most
abnormal appearing varices. When the two cirrhotic groups were compared (those with and those without PVO), a
statistically significant difference was seen for the presence of red signs ( p < 0.001).
Splenoportography
In all subjects with PVO, including those with cirrhosis, portography demonstrated multiple collaterals and tortuous
vessels at the porta hepatis as well as a hepatofugal blood flow in the anomalous vessels. The PVO was near complete in
the cirrhotic patients compared with the patients with isolated portal venous cavernous transformation. Portography was
performed in 10 cirrhotic patients without PVO to demonstrate the pattern of blood flow through the liver in
uncomplicated cirrhosis. In all of these 10 cirrhotic cases, portal venous blood flow was evident within the liver even when
a hepatofugal flow was present.
TABLE 2 -- Frequency of Various Endoscopic Findings Present in the Stomach, Duodenum, and Esophagus of the
Patients Studied
Portal Hypertension
Cirrhosis PVO C+PVO Control
Findings n = 37 (%) n = 26 (%) n = 9 (%) n = 57 (%)
Duodenum
Ulcer 8 (21.6) 3 (11.6) 1 (11.1) --
Varices 1 (2.7) 4 (15.4) 2 (22.2) --
Duodenitis 10 (27) 5 (19.2) 3 (33.3) 6 (10.5)
Stomach
Ulcer 3 (8.1) 1 (3.8) 1 (11.1) --
Snake skin 21 (56.8) 4 (15.4) 8 (88.9) 1 (1.8)
Scarlatina rash 13 (35.1) 5 (19.2) 7 (77.8) --
Petechia 14 (37.8) 4 (15.4) 9 (100) 3 (5.3)
Hyperemia 20 (54) 4 (15.4) 7 (77.8) 2 (3.5)
Fundic varices 8 (21.6) 12 (46.2) 9 (100) --
Esophagus
Esophageal varices
Forms
F1 7 (18.9) 1 (3.8) -- --
F2 23 (61.1) 5 (19.2) -- --
F3 7 (18.9) 20 (76.9) 9 (100) --
Extent
L1 (lower 1/3) 6 (16.2) 1 (3.8) -- --

L2 (lower 2/3) 24 (64.9) 3 (11.6) -- --

L3 (full extend) 6 (16.2) 21 (80.8) 9 (100)
Color
Blue 29 (78.4) 19 (73.1) 5 (55.6) --
White 8 (21.6) 7 (26.9) 4 (44.4)
Red Signs
Cherry red spot 15 (40.5) 2 (7.7) 8 (88.9)
Hematocytic spot 13 (35.1) 1 (3.8) 7 (77.8)
Diffuse redness 11 (29.7) 1 (3.8) 6 (66.7)
Esophagitis 9 (24.3) 2 (7.7) 4 (44.4) 3 (5.3)
F1, varices flattened by insufflation; F2, varices not flattened by insufflation, separated by areas of
normal mucosa; F3, confluent varices not flattened by insufflation; PVO, portal vein obstruction.
DISCUSSION
The three macroscopic aspects that, either alone or in combination, characterize the gastric mucosa of cirrhotic patients
with portal hypertension are petechia, hyperemia, and a snake skin appearance. Papazian et al.[4] and McCormack et al.[3]
were the first to characterize the gastric mucosal appearance of individuals with cirrhosis.
By design, none of the patients in the present study had had an episode of bleeding from the upper GI tract during the 3
months before the study. Moreover, none had undergone sclerotherapy or other therapeutic procedures related to GI
bleeding either before or at the time of the endoscopic evaluation performed as part of this study. These criteria for
admission to the study were chosen to assess the gastric mucosa at its best condition and to avoid the possibility that the
presence or absence of a given endoscopic sign would have been affected by the altered hemodynamic conditions
produced by the procedure [13] .
557
A snake skin mucosa was present in 56.8% of the cirrhotic patients without PVO, in 89% of the cirrhotic patients with
portal vein thrombosis, and in 15% of the patients with PVO but without cirrhosis. Only 2% of controls had a snake skin
appearance to their mucosa. The other endoscopic signs of PHG, such as petechia, hyperemia, and scarlatina rash, were all
characteristic of cirrhosis rather than PVO. As seen in Table 2 , all of the mucosal lesions of the stomach and esophagus
had a higher incidence in cirrhotic patients compared with patients with PVO. In each case, the difference was statistically
significant. On the other hand, gastric varices and the size of the esophageal varices were more prominent in patients with
PVO with or without cirrhosis. Portal venous blood flow to the liver occurs in cirrhosis, but, in cases of PVO, the
hepatopetal flow is markedly reduced, suggesting that the portal pressure in such patients is greater than it is in cirrhotic
patients without PVO. This putative increase in portal pressure with PVO was not measured directly in the present series,
but the greater size and extent of the esophageal varices and higher incidence of gastric varices in the cases with PVO
suggest that such was the case. Thus, it appears as if the additional mucosal lesions other than those currently recognized
as consistent with a PHG require the presence of parenchymal liver disease and do not occur solely as a result of an
increase in portal pressure.
In general, the presence of gastric mucosal lesions and esophageal red signs (cherry red spots and hematocystic spots) was
associated with the clinical severity of the liver disease (Child-Pugh grade) or the Beppu's score but not the PVO per se. It

should be noted, however, that most patients with vascular lesions of the gastric mucosa were classified as Child's class C.
On the other hand, although both the size and extension of varices in the patients with PVO were advanced, red signs were
present less often in these patients than in those with cirrhosis. Thus, it appears as if liver disease is an important
determinant of the presence and development of vascular mucosal lesions located in both the gastric and esophageal
mucosa.
The mucosa of patients with portal hypertension demonstrate endoscopic signs that singularly or in combination appear to
define the presence of portal hypertension. The size and extent of the esophageal varices relate to the PVO, and the
mucosal signs on esophageal varices and characteristics of portal gastropathy are found almost exclusively in cirrhotic
patients. Future prospective studies may enable investigators to determine which factor or factors are responsible for the
development of these vascular mucosal lesions.
Consistent with this observation is the reported higher incidence of portal gastropathy in the West [11] [14] compared to its
incidence in India [10] [11] [14] [15] [16] , where the majority of the cases have extrahepatic portal hypertension. As seen in
Table 2 , vascular mucosal lesions are seen in about 50% of cirrhotic patients and in only 20% of noncirrhotic patients with
portal hypertension.
The present study material is unique because it includes patients with liver cirrhosis, PVO, and cirrhosis complicated with
portal venous occlusion. Most of the material in the Western world includes only patients with portal hypertension due to
liver cirrhosis. During an 8-yr period, 26 patients with portal venous occlusion without coexistent liver disease were
identified at our institution. In these cases, although the incidence of fundic varices and the extent and size of the
esophageal varices was great, the incidence of vascular mucosal lesions characteristic of PHG was low.
Is portal pressure the sole determinant of PHG? This has been suggested by some investigators [3] [9] . However, this seems
unlikely because not every patient with portal hypertension has evidence for [7] or develops PHG. According to Sarin et
al.[7] , variceal pressure is similar in patients with and without PHG. Therefore, in addition to pressure, other factors must
contribute to the development of the mucosal lesions characterized as PHG.
In the present study, the presence of gastric varices was related more to the portal hypertension per se rather than to the
presence of cirrhosis; gastric varices were present in 21% of the cirrhotic patients but in 45% of those with extrahepatic
portal hypertension and in 100% of those with PVO. Almost complete occlusion of the portal vein with very large
collaterals in the fundic area of the stomach were the principal radiological findings of cirrhotic patients with PVO. These
radiological findings corelated quite well with the endoscopic findings in all nine patients with cirrhosis and PVO but were
not found in cirrhotic patients without PVO.
The other important observation in our study was that the development of PHG was greatly influenced by the severity of
the liver disease. Patients with liver disease of Child's grade C had portal gastropathy significantly more often than did
those with Child's B liver disease ( p < 0.05).
Overall, the results suggest that, although increased portal pressure is a prerequisite, the development of PHG is strongly
influenced by the severity and presence of the liver disease. On the other hand, the size and extent of the varices seem to
be related to the intensity of the portal hypertension.
REFERENCES
1. Khodadoost J, Glass GBJ. Erosive gastritis and acute gastroduodenal ulcerations as source of upper gastrointestinal bleeding in liver cirrhosis. Digestion
1972;7:129-38.
2. Sarfeh IJ, Juler GL, Stemer EA, et al. Results of surgical management of hemorrhagic gastritis in patients with gastroesophageal varices. Surg Gynecol Obstet
1982;155:167-70.
3. McCormack TT, Sims J, Eyre-Brook I, et al. Gastric lesions in portal hypertension: Inflammatory gastritis or congestive gastropathy? Gut 1985;26:1226-32.
4. Papazian A, Braillan A, Dupas JL, et al. Portal hypertensive gastric mucosa: An endoscopic study. Gut 1986;27:1199-203.
5. Quintero E, Pique JM, Bombi JA, et al. Gastric mucosal vascular

558
ectasias causing bleeding in cirrhosis. Gastroenterology 1987;93:1054-61.
6. D'Amico G, Montalboro L, Traina M, et al. Natural history of congestive gastropathy in cirrhosis. Gastroenterology 1990;99:1558-64.
7. Sarin SK, Sreenivas DV, Lahoti D, et al. Factors influencing development of portal hypertensive gastropathy in patients with portal hypertension.
Gastroenterology 1992;102:994-9.
8. Sarin SK. Non-cirrhotic portal hypertension. In: Bosch J, Groszmann RJ, eds. Portal hypertension, pathophysiology and treatment. Boston: Blackwell Scientific
Publications, 1994: 27-53.
9. Pagliaro L, Spina L. The Italian programme on liver cirrhosis. Ital J Gastroenterol 1987;19:295-7.
10. Beppu K, Inokuchi K, Koyanagi N, et al. Prediction on variceal hemorrhage by esophageal endoscopy. Gastrointest Endosc 1981;27:213-8.
11. Maratka Z. Terminology, definitions and diagnostic criteria in digestive endoscopy. With the collaboration of the members of the Terminology Committee of
the World Society of Digestive Endoscopy/OMED. Scand J Gastroenterol 1984;103:1-74.
12. Taor R, Fox B, Ware J, et al. Gastritis: Gastroscopic and microscopic. Endoscopy 1975;7:209-15.
13. Kato K. Endoscopic and histological study of gastric mucosa in liver cirrhosis with special reference to cases treated with endoscopic injection sclerotherapy for
esophageal varices. Gastrointest Endosc 1987;29:2166-76.
14. Hosking SW, Kennedy HJ, Seddon I, et al. The role of propranolol in congestive gastropathy of portal hypertension. Hepatology 1987;7:437-41.
15. Tandon RK. Emergency endoscopy in acute upper gastrointestinal hemorrhage. J Assoc Physicians India 1978;26:1-6.
16. Misra SP, Dwivedi M, Misra V, et al. Endoscopic and histological appearance of the gastric mucosa in patients with portal hypertension. Gastrointest Endosc
1990;36:575-9.

Bookmark URL: /das/journal/view/N/721548?ja=1975&PAGE=1.html&ANCHOR=top&source=MI

April 21, 2000
Conde Petra
Abnormalities of gastric emptying in portal

hypertension.
Citation
Bibliographic Data
Abstract
Indexing Data
Abnormalities of gastric emptying in
Copyright Notice and Disclaimer portal hypertension.
Balan KK - Am J Gastroenterol - 1996 Mar; 91(3): 530-4
NLM Citation ID:
96209586
Full Text
Full Source Title:
Frontmatter
INTRODUCTION
Publication Type:
SUBJECTS AND METHODS Journal Article
Subjects Language:
English
Endoscopy
Author Affiliation:
GE studies
Department of Nuclear Medicine, Royal Liverpool University
STATISTICAL METHODS Hospital, United Kingdom.
RESULTS Authors:
Balan KK; Grime S; Sutton R; Critchley M; Jenkins SA
DISCUSSION
Abstract:
ACKNOWLEDGMENTS
OBJECTIVES: To investigate: 1) the rate of gastric emptying
REFERENCES of portal hypertensive patients and 2) whether alterations in
gastric emptying play any role in the development of portal
hypertensive gastropathy. METHODS: Fifty patients (37 with
esophageal varices) with cirrhosis and seven with extrahepatic
portal hypertension underwent upper GI endoscopy followed by
radionuclide gastric emptying studies using a semi-solid meal.
Twenty-six patients also under went corrected wedged hepatic
venous pressure measurement. Sixteen normal subjects
underwent gastric emptying studies only. RESULTS: Varices
were completely obliterated by sclerotherapy in 17 patients and
were patent in 27. Thirty-seven patients had portal hypertensive

gastropathy, 25 of whom had mild changes and 12 severe. No

significant difference in gastric emptying was observed between
patients with mild and severe portal hypertensive gastropathy
and between those with portal hypertensive gastropathy and a
normal gastric mucosa. There was no significant difference in
gastric emptying between normal subjects and portal
hypertensive patients although the latter group showed a
tendency for faster gastric emptying. No difference in the rate
of gastric emptying was observed between portal hypertensive
patients with intrahepatic and extrahepatic pathology. However,
patients with esophageal varices (patent and obliterated)
emptied their stomachs significantly faster than those without (p
= 0.01). There was no correlation between the rate of gastric
emptying and corrected wedged hepatic venous pressure.
CONCLUSIONS: It would appear that, although alterations in
gastric emptying are common in portal hypertension, gastric
emptying does not appear to play a causative role in the
mucosal changes characteristic of portal hypertensive
gastropathy.
Major Subjects:
Gastric Emptying / * Physiology
Hypertension, Portal / Etiology / * Physiopathology
Analysis of Variance
Chi-Square Distribution
Endoscopy, Gastrointestinal
Esophageal and Gastric Varices / Physiopathology
Female
Human
Liver Cirrhosis / Physiopathology
Male
Middle Age


530
Abnormalities of Gastric Emptying in Portal Hypertension
K. K. Balan
M.D M.R.C.P.
S. Grime M.Sc.
R. Sutton
M. Phil. F.R.C.S.
M. Critchley
M.D
F.R.C.R.
S. A. Jenkins Ph.D.
Departments of Nuclear Medicine and Surgery, Royal Liverpool University Hospital, Prescot Street,
Liverpool, United Kingdom
Objectives: To investigate: 1) the rate of gastric emptying of portal hypertensive patients and 2)
whether alterations in gastric emptying play any role in the development of portal hypertensive
gastropathy. Methods: Fifty patients (37 with esophageal varices) with cirrhosis and seven with
extrahepatic portal hypertension underwent upper GI endoscopy followed by radionuclide gastric
emptying studies using a semi-solid meal. Twenty-six patients also underwent corrected wedged
hepatic venous pressure measurement. Sixteen normal subjects underwent gastric emptying
studies only. Results: Varices were completely obliterated by sclerotherapy in 17 patients and were
patent in 27. Thirty-seven patients had portal hypertensive gastropathy, 25 of whom had mild
changes and 12 severe. No significant difference in gastric emptying was observed between patients
with mild and severe portal hypertensive gastropathy and between those with portal hypertensive
gastropathy and a normal gastric mucosa. There was no significant difference in gastric emptying
between normal subjects and portal hypertensive patients although the latter group showed a
tendency for faster gastric emptying. No difference in the rate of gastric emptying was observed
between portal hypertensive patients with intrahepatic and extrahepatic pathology. However,
patients with esophageal varices (patent and obliterated) emptied their stomachs significantly
faster than those without ( p = 0.01). There was no correlation between the rate of gastric emptying
and corrected wedged hepatic venous pressure. Conclusions: It would appear that, although
http://home.mdconsult.com/das/article/body/jorg=journal&source=HS,MI&sp=721543&sid=0/N/1970/1.html (1 of 8) [21/04/2000 23:57:37]

alterations in gastric emptying are common in portal hypertension, gastric emptying does not
appear to play a causative role in the mucosal changes characteristic of portal hypertensive
gastropathy.
Reprint requests and correspondence: Dr. K. K. Balan, M.D., M.R.C.P., Department of Nuclear Medicine, Royal Liverpool
University, Hospital, Liverpool, L7 8XP, UK.
Received May 30, 1995; accepted Nov. 6, 1995.
INTRODUCTION
Portal hypertensive gastropathy (PHG) is recognized as an important cause of nonvariceal bleeding in patients with portal
hypertension [1] [2] . Several morphological studies have shown that PHG is characterized by mucosal or submucosal
vascular dilation without inflammation [1] [3] [4] . Although portal hypertension is a prerequisite for the development of
PHG [5] [6] [7] [8] , the pathogenesis of PHG is incompletely understood. Studies in both experimental animals and in
humans have shown that the functional and ultrastructural changes that take place in the portal hypertensive gastric
mucosa make it more susceptible to the deleterious effects of various substances, including aspirin, bile acids, and alcohol
[3] [9] [10] [11] . Similarly, the portal hypertensive gastric mucosa may be more susceptible to the deleterious effects of
endogenous factors, such as acid and pepsin, perhaps as a result of increased production,or by decreased clearance from
the stomach. However, previous data on gastric motor function in portal hypertension are incomplete and confusing.
Furthermore, there are very few data on gastric emptying (GE) in patients displaying different degrees of PHG. The aim of
this study was to assess the rate of GE of portal hypertensive patients and to examine whether abnormal GE may play
causative role in the pathogenesis of PHG.
SUBJECTS AND METHODS

Subjects
The investigation was performed of 57 patients (male 33, female 24, median age 53 yr) with either biopsy-proven cirrhosis
or overt evidence of extrahepatic portal hypertension (Table 1) and of sixteen normal control subjects (male 11, female
five, median age 42 yr). None of the patients had alcoholic hepatitis, and the majority (94%) of alcoholic cirrhotics were
abstinent at the time of the study. Thirty-four of the patients had received previous injection sclerotherapy for esophageal
varices. Patients with severe cardiac, renal, and pulmonary disease were excluded from the study. Also excluded from the
study were those patients with severe hepatic dysfunction (Child's C) [12] , acute GI bleeding, severe ascites, previous
gastric surgery, or grossly abnormal clotting disorders. All medication known to influence GE was discontinued 48 h
before the study. Ethical approval was obtained from the Hospital Ethical Committee. All of the studies were conducted
after obtaining written, informed consent from the patients and volunteers, and Administration of Radioactive Substances
Advisory Committee approval was granted to carry out radionuclide GE studies.
531
TABLE 1 -- Etiology of Liver Disease and Portal Hypertension

Intrahepatic pathology
Alcoholic cirrhosis 32
Primary biliary cirrhosis 7
Cryptogenic cirrhosis 5
Hemochromatosis 2

Polycystic liver disease 1

Chronic active hepatitis 1
alpha-1 anti-trypsin deficiency 1
Trauma 1
Extrahepatic pathology
Portal vein thrombosis 3
Mesenteric vein thrombosis 1
Congenital portal vein abnormality 1
Congenital splenic vein abnormality 1
Portal vein obstruction 1
TABLE 2 -- McCormack's Classification of Portal Hypertensive Gastropathy

Mild PHG
i. A fine pink speckling or "scarlatina" type rash
ii. A superficial reddening, particularly on the surface of the rugae, giving a striped appearance
iii. A fine white reticular pattern separating areas of raised edematous mucosa resembling a "snake
skin."
Severe PHG
i. Discrete red spots analogous to the cherry red spots described in the esophagus (Inokuchi, 1980)
ii. A diffuse hemorrhagic gastritis
Endoscopy
All patients underwent an upper GI endoscopy. If esophageal varices were present, their site, size, shape, color, and extent
were recorded and subsequently graded according to the criteria laid down by the Japanese Research Society for Portal
Hypertension [13] . The macroscopic appearance of the stomach was noted, and the degree of PHG was classified
according to the criteria (Table 2) of McCormack et al.[1] . Corrected wedged hepatic venous pressure was measured in 26
of the patients.
GE studies
The rate of GE was measured according to the method described by Sagar et al.[14] and has been in use in our department
for the past 10 yr. Both patients and volunteers were studied after an overnight fast using a meal containing porridge made
from Ready-Brek and milk into which 10 MBq of [99m]technetium-bran (5 g) had been incorporated; this meal is similar
to a semi-solid meal in its GE characteristics. The complete meal included two sandwiches comprised of white bread,
butter, and cheese and a cup of tea without sugar. The total volume of the meal was 510 ml, the weight was 430 g, and the
total calorific value was 2.7 MJ.

Each patient sat at 45 on a chair in front of a large field of view gamma camera interfaced to a digital computer, and
acquisition of the data was commenced at the beginning of the meal (ingested in 10 min) at a rate of 1 frame per min
Figure 1. Analysis of GE data.
and continued for 90 min in all cases. At the end of the study, a composite image of the stomach was obtained by summing
the first 10 frames of the study. Two regions of interest (ROI) were then drawn on this image to define the stomach content
and a background area. The counts in each ROI on each frame of the study were then obtained, and a GE curve and
background curve were produced. A background corrected GE curve was derived by subtracting the individual background
curve value multiplied by the quotient of the number of cells in the gastric ROI and the number of cells in the background
ROI from each value of the gastric curve at each time interval. After correcting for radioactive decay, the GE curve was
finally displayed as a percentage emptying curve by dividing each point of the curve by the maximum curve value and
multiplying the answer by 100 (Fig. 1) . GE did not follow a constant pattern and could not therefore be fitted with a
simple function, so, in addition to the time taken for 50% GE (t1/2 ), the area under the percentage GE curve was also used
to express the GE results. According to the latter method, the smaller the area under the curve, the faster the GE and vice
versa.
STATISTICAL METHODS
A chi2 test was used to determine the association between two discrete variables, a Fisher-Irwin exact probability test was
used for 2 2 tables with small frequencies, and unpaired Student' s t test (two groups) or one-factor analysis of variance
(three or more groups) was used to compare continuous variables between groups of patients. If the data were skewed,
Mann-Whitney U or Kruskal-Wallis H tests were used. An association between two continuous variables was determined
by the Pearson correlation coefficient. A p value of less than 0.05 was considered significant.
RESULTS
Of the total of 57 patients, 44 had esophageal varices, of whom 17 had successful sclerotherapy, 17 had sclerotherapy
532
TABLE 3 -- The Results of GE (Median and Range): Comparison of GE (t1/2 ) and Area Under Curve
Patients Normals p Value
t1/2 (min) 38 47 0.6
(12-90) (22-60)
Area under curve 4365 4592 0.5
(2541-6224) (2766-5301)
Figure 2. GE in patients with and without PHG. Thick horizontal bars represent median values, and thin horizontal lines represent
interquartile ranges.

but varices remained patent, and 10 had not recieved any sclerotherapy. Five patients had gastric varices (associated with
patent esophageal varices).
Thirty-seven patients (67%) had endoscopic evidence of PHG. Twenty-five patients had mild (45%) and 12 (22%) severe
PHG. Forty-seven patients had mild (Child's A) and 10 moderate (Child's B) hepatic dysfunction. Corrected wedged
hepatic venous pressure ranged between 14 and 43 mm Hg (median 23.4). None of the patients had active peptic ulcer
disease.
The pattern of GE was very variable, ranging from a monoexponential to a double exponential. There was no relationship
between any of the patient groups and the observed pattern of GE. Because of the complexity of the emptying pattern, two
parameters, i.e., t1/2 of GE and the area under the curve, were used to compare the rate of GE between the groups.
However, the two methods of assessing the GE agreed closely (Table 3) , so only t1/2 has been used to illustrate the
difference in GE between patient groups. There was no significant difference in GE between the portal hypertensive
patients considered as a single group and normal healthy volunteers although the former tended to empty their stomachs
faster ( p = 0.2; Mann-Whitney U test). No significant difference was observed in the rate of GE (Fig. 2) between patients
with a normal gastric mucosa and those with either mild or severe PHG ( p = 0.07; Kruskal-Wallis H test). Patients with
PHG, particularly those with the severe form, however, showed a trend for faster GE than those with endoscopically
normal-looking gastric mucosa. No significant difference in GE was observed between cirrhotic and noncirrhotic portal
hypertensive patients ( p = 0.9; Mann-Whitney U test) or between
Figure 3. GE in patients with and without esophageal varices. Thick horizontal bars represent median values, and thin horizontal lines
represent interquartile ranges.
Figure 4. Correlation between GE rate and corrected wedged hepatic venous pressure.
alcoholic and nonalcoholic cirrhotics. GE was significantly faster ( p = 0.01; Mann-Whitney U test) in patients with
esophageal varices either obliterated or patent compared with those without (Fig. 3) . However, no difference was
observed in the frequency of PHG in patients with and without varices although severe PHG was absent in the latter. There
was no significant difference in GE between patients with esophageal varices who had and had not received injection
sclerotherapy. No significant difference in GE was observed between patients with different grades of varices. The rate of
GE correlated poorly ( r = 0.19; p = 0.34) with the portal pressure (Fig. 4) . There was no significant difference in GE
between patients with Child's grade A and Child's grade B liver disease. Although female cirrhotics tended to empty their
stomachs faster than male cirrhotics, the difference was not statistically significant ( p = 0.2; Mann-Whitney U test).
DISCUSSION
The results of the present study indicate that there is no significant difference in GE of a semi-solid meal between normal
subjects and portal hypertensive patients although the latter showed a tendency for faster GE. The results, however, clearly
demonstrate that the rate of GE is significantly more rapid in patients with esophageal varices than in those without

although no relationship was observed
533
between GE and either the grade of the varices or the portal pressure. These observations suggest that the presence of
varices and possibly a threshold elevation in portal pressure are the major factors responsible for the increased rate of GE.
Possibly, above the degree of portal hypertension required for the formation of esophageal varices, further increases in
portal pressure do not influence the motor function of the stomach. It could also be argued that, after the establishment of a
portal collateral circulation, the circulating levels of GI hormones and peptides reach a plateau, beyond which they cannot
further influence GE.
The faster GE of portal hypertensive patients with esophageal varices observed in our study is in agreement with previous
studies. Aprile et al.[15] not only demonstrated a faster GE of liquids but also showed an impaired gastric accommodation
to distension in portal hypertensive patients with esophageal varices. From these observations, Aprile et al.[15] concluded
that the faster GE of liquids in patients with esophageal varices might have been caused by the reduced gastric wall
compliance. This suggestion is supported by Reilly et al.[16] , who reported an accelerated rate of GE of both liquids and
solids in portal hypertensive rats compared with controls. On the basis of their results, they speculated that a decreased
gastric wall compliance in portal hypertension caused an increase in intragastric pressure with the development of a
gastroduodenal pressure gradient, which promoted the rapid emptying of liquids. They also believed that a similar
reduction in gastric wall compliance would limit antral distensibility and could lower the threshold for activation of
contractions, thereby resulting in a potentiation of solid GE. Holdeman et al.[17] studied simultaneous liquid and solid GE
and small bowel transit in 10 patients with portal hypertension and 14 control subjects. These investigators found that the
GE of solids, but not liquids, was prolonged in the portal hypertensive group compared with the controls. Small bowel
transit was, however, not significantly different between portal hypertensive patients and controls.
In normal control subjects, vagal innervation is known to maintain fundal tone and mediate receptive relaxation and
therefore plays an important role in the GE of a liquid meal. Several reports have suggested that GE of a liquid meal is
accelerated after either a highly selective vagotomy [18] [19] or truncal vagotomy [20] for peptic ulcer disease. Similarly, a
delay in the GE of solids has also been observed after highly selective and truncal vagotomies [21] [22] . The semi-solid
meal used in this study is like a slow liquid in its GE characteristics, so the faster GE observed in patients with portal
hypertension and esophageal varices could result from alterations in vagal function. A situation analogous to vagotomies
could occur in cirrhotic patients as a result of autonomic neuropathy [23] [24] . However, autonomic neuropathy also occurs
in compensated cirrhotics without varices. In patients who have received injection sclerotherapy, extravasation of
sclerosant into the mediastinum may produce a chemical vagotomy that could alter the rate of GE. However, there was no
difference in GE between patients with varices who had and had not received injection sclerotherapy. Nevertheless, it is
possible that the presence of varices exacerbates the autonomic neuropathy commonly observed in cirrhotics. In our study,
although none of the patients had any obvious signs or symptoms suggestive of autonomic neuropathy, autonomic function
was not formally assessed. Consequently, further work is required to establish whether autonomic neuropathy is more
severe in cirrhotic patients with or without varices.
In the present study, no relationship was observed between the rate of GE and the severity of liver disease. However, it
must be stressed that we excluded patients with severe liver disease who may have shown marked disturbances in GE.
Furthermore, no significant difference was observed in the rate of GE between male and female portal hypertensive
patients although normal males are known to empty their stomachs faster than normal females [25] . This situation might
have arisen because of the altered levels of female sex hormones in cirrhotic males. The reported effects of acute and
chronic alcohol administration on human GE are inconsistent; acceleration [26] , delay [27] [28] , or no effect [29] on GE after
ingestion of alcohol all have been reported in the literature. We, however, did not detect any difference in the rate of GE
between cirrhosis with alcoholic and nonalcoholic etiologies. Differences in the severity of the liver disease and the
male:female ratio may explain the conflicting observations between the present study and other reports on abnormal GE in
portal hypertensive patients. Chesta et al.[30] demonstrated a prolongation of mouth-to-cecum transit of a solid meal in a
small number of cirrhotic patients. Furthermore, most of their patients had severe liver disease whereas our patients had
mild to moderate liver disease. Similarly, two other recent studies [31] [32] , the former using both solid and liquid meals
and the latter a semi-solid meal, have also reported a delay in GE in cirrhosis. Not only did these studies include a small
number of patients, they also did not examine the relationship between GE and factors such as etiology of portal
hypertension, severity of the liver disease, level of portal pressure, grade of varices, and gender.

None of the previous studies has examined the pattern of GE of portal hypertensive patients with or without PHG and
between those with different degrees of PHG. The results of the present study indicate that there is no significant
difference in the rate of GE between patients with and without PHG. However, there was a trend for patients with PHG,
particularly for those with the severe form, to empty their stomachs more rapidly than patients with a normal gastric
mucosa. Although no firm conclusion can be drawn from these findings, it is possible that abnormalities in GE may not be
uncommon in PHG, and further studies are required to clarify the role of GE in the pathogenesis of this condition.
In conclusion, our results show that GE of a semi-solid meal is faster in portal hypertensive patients with esophageal
534
varices compared with those without varices although no linear relationship could be demonstrated between GE and portal
pressure. Furthermore, although disturbances in GE were common in cirrhotics, no definite relationship was observed
between PHG and GE. Therefore, further studies are required to evaluate the role of GE in PHG.
ACKNOWLEDGMENTS
The work described in this study has been presented in part to the autumn meeting of the British Society of Nuclear
Medicine, Sheffield, December, 1993 and was published in abstract form in Nucl Med Commun 1994;15:261.
REFERENCES
1. McCormack TT, Sims J, Eyre-Brook I, et al. Gastric lesions in portal hypertension. Inflammatory gastritis or congestive gastropathy? Gut 1985;26:1226-32.
2. Tarnawski AS, Sarfeh IJ, Stachura J, et al. Microvascular abnormalities of the portal hypertensive gastric mucosa. Hepatology 1988;8:1708-10.
3. Quintero E, Pique JM, Bombi JA, et al. Upper gastrointestinal bleeding caused by gastrointestinal vascular malformations. Incidence, diagnosis and treatment.
Dig Dis Sci 1986;31:897-905.
4. Hashizume M, Tanaka K, Inokuchi K. Morphology of gastric microcirculation in cirrhosis. Hepatology 1983;3:1008-112.
5. Lebrec D, De Fleury P, Rueff B, et al. Portal hypertension size of oesophageal varices and the risk of gastrointestinal bleeding in alcoholic cirrhosis.
6. Hosking SW, Kennedy HJ, Seddon I, et al. The role of propranolol in congestive gastropathy of portal hypertensionHepatology 1987;7:437-41.
7. Reynolds TB. Portal Hypertension. In: Schiff L, Schiff ER (Eds) Diseases of the Liver. JB Lippincott Company, Philadelphia,1987.
8. Sarfeh IJ, Malki A, Tarnawski A Mason GR, Mach T, Ivey KJ. Portal hypertension and gastric mucosal injury in rats: effects of alcohol. Gastroenterology,
1983;84:987-93.
9. Sarfeh IJ, Tarnawski A, Maeda R, et al. The gastric mucosa in portal hypertension. Effects of topical bile acid. Scand J Gastroenterol 1984;19(Suppl)92:189-94.
10. Sarfeh IJ, Tarnawski A, Hajduczek A, et al. Does portal hypertension predispose gastric mucosa to aspirin injury? Histological ultrastructural and functional
analysis Gastroenterology 1987;92:A1614 (abstract).
11. Manabe T, Suzuki T, Honjo I. Changes of gastric blood flow in experimentally induced cirrhosis of the liver. Surg Gynaecol Obstet 1978;147:753-7.
12. Child CG, Turcotte JG. Surgery and portal hypertension. In: Child CG, Dumfry JE, eds. Major problems in clinical surgery: The liver and portal hypertension,
vol 1. Philadelphia: WB Saunders, 1964.
13. Japanese Research Society for portal hypertension. The general rules for recording endoscopic findings on oesophageal varices. Jap J Surg 1980;10:84-7.
14. Sagar S, Grime JS, Little M, et al. Technetium-99 m labelled bran. A new agent for measuring gastric emptying. Clin Radiol 1983;34:275-8.
15. Aprile LRO, Meneghelli UG, Martinelli ALC. Gastric accommodation to distension and gastric emptying of liquids in patients with portal hypertension (PH).
Proceedings of the World Congress of Gastroenterology, Los Angeles, CA, 1994.

16. Reilly JA, Frost CF, Quigley EMM, et al. Gastric emptying of liquid and solids in the portal hypertensive rat. Dig Dis Sci 1990;35:781-6.
17. Holdeman KP, Galati JS, Dalrymple GV, et al. Gastric emptying and small bowel transit in patients with portal hypertension. J Nucl Med 1993;34:111P
(abstract).
18. Clarke RJ, Alexander Williams J. The effect of preserving antral innervation and of a pyloroplasty on gastric emptying after vagotomy. Arch Surg
1973;102:43-4.
19. Glysteen JJ, Burdeshaw JA, Hallenbeck GA. Gastric emptying of liquids after different vagotomies and pyloroplasty. Surg Gynaecol Obstet 1976;142:41-8.
20. Nilsson G, Yallow RS, Berson SA. Distribution of gastrin in the gastrointestinal tract of human, dog, cat and hog. In: Anderson, Almquist, Wiksell, eds.
Frontiers in gastrointestinal hormone research. Stockholm, 1973: 95-101.
21. Kelly KA, Code CF. Effect of transthoracic vagotomy on canine gastric electrical activity. Gastroenterology 1969;57:51-8.
22. Sheiner HJ, Quinlan MF, Thompson IJ. Gastric motility and emptying in normal and post-vagotomy subjects. Gut 1980;21:753-9.
23. Thuluvath PJ, Triger DR. Autonomic neuropathy and chronic liver disease. Quart J Med 1989;268:737-47.
24. Khosla SN, Sanyal S, Nand N. Autonomic function tests and clinical significance of dysautonomia in chronic liver disease. JAPI 1991;12:924-6.
25. Notival R, Carrio I, Cano L, et al. Gastric emptying of solid and liquid meals in healthy young subjects. Scand J Gastroenterol 1984;19:1107-13.
26. Kaufman SE, Kaye MD. Effect of ethanol upon gastric emptying. Gut 1979;20:688-92.
27. Barboriak JJ, Mead RC. Effect of alcohol on gastric emptying in man. Am J Clin Nutr 1970;23:1151-3.
28. Jian R, Cortot A, Ducrot F, et al. Effect of ethanol ingestion on postprandial gastric emptying and secretion biliopancreatic secretions and duodenal absorption
in man. Dig Dis Sci 1986;31:604-14.
29. Moore JG, Christian PE, Datz FL, et al. Effect of wine on gastric emptying in humans. Gastroenterology 1981;81:1072-5.
30. Chesta J, Lillo R, Defilippi C, et al. Orocecal transit time and gastric emptying in patients with cirrhosis. Gastroenterology 1991;100:A43 (abstract).
31. Galati JS, Holdeman KP, Dalrymple GV, et al. Delayed gastric emptying of both the liquid and solid components of a meal in chronic liver disease. Am J
32. Isobe H, Sakai H, Satoh M, et al. Delayed gastric emptying in patients with liver cirrhosis. Dig Dis Sci 1994;39(5):983-7.


April 21, 2000
Conde Petra
PORTAL HYPERTENSION, VARICES, AND

TRANSJUGULAR INTRAHEPATIC
Additional Article
Consult's full-text literature database. Volume 4 Number 1 February 2000
Full Text
Frontmatter 133
PORTAL HYPERTENSION
HEPATOLOGY: A CENTURY
INDICATIONS FOR
TRANSJUGULAR INTRAHEPATIC OF PROGRESS
CONTRAINDICATIONS FOR
TRANSJUGULAR INTRAHEPATIC PORTAL HYPERTENSION, VARICES, AND
COMPLICATIONS PORTOSYSTEMIC SHUNTS
References
Harold O. Conn MD
Department of Medicine, The Yale University School of Medicine, New

Haven Connecticut; and The University of Miami School of Medicine,
Miami, Florida
Harold O. Conn, MD
The Yale University School of Medicine
160 Morgan Avenue
East Haven, Connecticut 06512-4519
The longest
march starts with a single stride.
LAO-TZU

The primary topic of this article is portal hypertension and

esophagogastric varices, the primary clinical problem posed
by portal hypertension. Transjugular intrahepatic
portosystemic (TIPS) shunts is the newest and the least
invasive method of eradicating varices. This article defines
portal hypertension succinctly, describes how it gives rise to
varices and their consequences, and briefly reviews the
development, short experience with, and current status of
TIPS shunts.
Portal hypertension is the state of increased pressure in the
portal venous system and its tributaries and collaterals,
including the hepatic sinusoids. The portal venous pressure
is determined by the amount of blood that flows through the
portal venous system and the resistance that opposes the
blood flow. It can be expressed as a variant of Ohm's law, P
= FR; that is, the pressure gradient through the portal
venous system, P, is determined by F, the volume of blood
flowing through the system, and R, the resistance to that
flow. Changes in either F or R affect the pressure. The
portal venous blood flow is the total of the blood flow to the
liver, the spleen, and the splanchnic organs. In most types of
portal hypertension, both the blood flow and the resistance
to the blood flow are altered.
Resistance is determined largely by the radius of the vessel,
by the viscosity of the blood and by the length of the vessel.
This relationship, which is known as Poiseuille's law, can be
expressed as R = 8nL/r4 , where n is the coefficient of
viscosity and L is the length of the vessel, both of which are
essentially constant. Because the radius ( r) of the vessel is
expressed as its fourth power, it
134
is by far the most mathematically important component of

the formula. Thus, the smaller the vessel, the greater the
resistance to the passage of blood through that vessel. The
main sites of resistance to hepatic blood flow appear to be
small portal and hepatic veins and the sinusoids.
The key to the syndrome of portal hypertension is an
increase in plasma volume. [20] In experimental animal
models, sodium restriction has been shown to prevent
plasma expansion and the development of portal
hypertension, [21] and thus the development of varices, and
can minimize the hyperdynamic state. The second most

important factor in the development of portal hypertension

is the development of the portosystemic collateral
circulation. The complications of portal hypertension, which
are consequences of the increased portal pressure, are
mediated more by the portosystemic collaterals than by the
portal pressure itself.
PORTAL HYPERTENSION
Portal hypertension is usually caused by intrahepatic fibrosis or
nodularity that increases resistance to the egress of portal blood
from the liver. Cirrhosis is the most common disorder that
induces portal hypertension, and the cirrhosis that induces
portal hypertension may be of alcoholic, viral, drug-induced,
biliary, or some other origin. The portal hypertension of
cirrhosis, which is a form of postsinusoidal portal venous
obstruction, that is, increased resistance to portal blood outflow,
is defined as a sustained portosystemic pressure gradient of 10
to 12 mm Hg, although the precise minimal level at which
portal hypertension begins may be as low as 7 mm Hg. The
portosystemic pressure gradient is defined as the occluded
hepatic venous pressure minus the unoccluded, or free, portal
venous pressure, that is, the hepatovenous pressure gradient
(HVPG).
All patients with cirrhosis or other chronic liver disease, such as
chronic viral or alcoholic hepatitis, are at risk of developing
esophagogastric varices. These collateral vessels develop
between the hypertensive portal venous vessels and the
systemic venous system, which is under normal pressure, in an
attempt to decompress the elevated portal venous pressure.
Esophagogastric varices are prone to erode or to explode, giving
rise to hemonage of esophogastric varices (HEV), the most
serious and lethal complication of portal hypertension. Patients
with varices should be periodically screened endoscopically for
the presence and size of varices, which are proportional to the
severity of the liver disease. Even patients with Child-Pugh
class A cirrhosis should be screened for varices if the
esophageal vein diameters are enlarged (> 3 mm) or if the
HVPG is increased (> 12 mm Hg). Patients with Child-Pugh
class B or C cirrhosis usually have varices, the growth of which
parallels the increasing severity of the cirrhosis. If small varices
are found, frequent endoscopic assessments are indicated,

because such varices tend to grow in size. Furthermore, the

three most important risk factors for the initial episode of HEV
are the severity of the liver disease, the size of the varices and
the presence of endoscopic red markings described by Beppu et
al, [4] such as red spots or red wales, also known as varices on
varices that are considered signs of impending rupture.
The risks of rebleeding are extremely high in survivors of the
first hemorrhage, especially during the first week after
hemorrhage, and the risks remain high for an additional 8 to 12
weeks. There are no quantitative risk factors for recurrent
bleeding other than the severity of the liver disease as evaluated
by clinical and laboratory findings, such as the total serum
bilirubin concentration, serum albumin level, and prothrombin
times. When patients have survived for
135
3 to 6 months after an initial episode of HEV, the risks of

rebleeding are reduced, but such patients are probably always
somewhat at risk.
There does not appear to be any clear association between the
nature of the initial episodes of bleeding (e.g., oozing or
spurting) and the risk of recurrent HEV. It is widely agreed that
the need for more than two units of blood, which is usually
associated with a systolic blood pressure of less than 100 mm
Hg, a pulse rate of more than 100 beats/min or a postural
decrease of more than 20 mm Hg in systolic blood pressure
constitutes clinically significant bleeding.
Signs of instability that indicate diminished blood volume
include failure to control HEV within 6 hours, (i.e., the need for
four or more units of blood to restore the systolic blood pressure
levels to normal) and a persistent, rapid pulse rate. Recurrence
of bleeding, as demonstrated by hematemosis or by gastric
aspiration, by a decrease in the vital signs to the levels outlined
previously, or by the continued need for blood transfusions to
maintain the hematocrit above 26%, is evidence that therapy has
failed.
One third of all deaths of cirrhotic patients result from HEV, the
most common signle cause of death from cirrhosis. [19] Because
almost 90% of cirrhotic patients develop varices [8] and about
30% of them bleed from their varices, many patients are
involved, of whom about 40% die of HEV. Recurrences of
HEV are common: 70% of patients who survive the initial
hemorrhage bleed again. [26] The dilatation and bleeding result
from systemic vasodilation and a decrease in systemic vascular

resistance. It is a complex scenario involving nitric oxide (NO),
glucagon, prostaglandins, tumor necrosis factor-alpha and other
substances that give rise to volume expansion, increments in
cardiac output, and a hyperdynamic circulation. [14] The
vasodilation and hyperdynamic circulation result in the
appearance of varices, dilated intraesophageal veins with an
extensive, tortuous network of capillary vessels that provide the
vascular power of these lesions. Varices develop when the
hepatic vein pressure gradient surpasses 12 mm Hg. Bleeding
may occur at any time thereafter.
The diagnosis of varices is best made by endoscopic
examination, as is the demonstration of actively bleeding
varices. Many effective treatments are available, including
endoscopic obliteration of varices either by sclerotherapy (EST)
per se or esophageal variceal ligation (EVL). A transjugular
intrahepatic portosystemic shunt appears to be the most rapid
and effective means of doing so. [13] Both techniques
immediately stop the bleeding, at least transiently, in more than
90% of patients. [12] Pharmacologic therapies with vasopressin,
somatostatin, terlipressin, or octreotide plus nitroglycerin [22]
have all been shown to be effective.
Much research has been devoted to the prevention of the initial,
most critical episode of HEV. Beta-adrenergic blockade appears
to be the most successful form of treatment available. [11] These
studies are discussed in detail by Grace (Fig. 1) (Figure Not
Available) . [24]
Somewhat less successful but equally important are the
prevention and management of recurrent HEV by
pharmacologic, [7] or endoscopic [27] therapies or by both
combined.
The treatment of the acute bleeding episode by the use of
pharmacologic agents, such as vasopressin, [25] vasopressin plus
nitroglycerin [6] or other pressure-lowering substances,
glypressin (terlipressin), [60] ornipressin, somatostatin, [5] or
octreotide [13] is usually successful. Various means of
Endoscopic therapy, such as endoscopic sclerotherapy, [23]
endoscopic variceal ligation, [62] or TIPS shunt, [59] are
considered standard methods of controlling HEV.
The definitive therapy of HEV is portal decompressive surgery.
The standard techniques include (1) end-to-side portacaval
anastomoses (PCA), (2) side-to-side PCA, (3) mesocaval
interposition H-graft anastomoses, and (4) distal

136
Figure 1. (Figure Not Available) Randomized, controlled trials published

in peer-reviewed journals of nonselective beta-adrenergic blockers
(propranolol and nadolol) versus placebo for the prevention of first
variceal hemorrhage. These results are for bleeding from all sites. Results
for bleeding from esophageal varices alone are similar. The odds ratio for
the group is 0.52; 95% confidence interval, 0.38-0.71 ( P < 0.01). ( From
Grace ND: Management of portal hypertension. Gastroenterologist
1993:1:39; with permission.)
splenorenal anastomoses. After successful PCA, further
episodes of HEV are rare. Recent advances include
small-diameter portacaval H-grafts in which a prosthetic graft is
interposed between the portal vein and the inferior vena cava.
This technique permits the maintenance of sufficient portal
hypertension so that the liver receives adequate portal blood
flow. Comparisons of this technique with TIPS demonstrate
similar hemodynamics.
Some investigators, Orloff and colleagues among them,
strongly advocate emergency portacaval anastomoses (EPCA)
as the treatment of choice [40] that should be performed as soon
as possible after endoscopic proof of the bleeding site. It is
probably true that no other form of therapy has shown such
beneficial results. [9]
Although TIPS shunts may ultimately become the preferred
treatment, a high rate of stent stenosis may prevent it from
maintaining its early, transient success. [29] The recent studies of
Siegerstetter and Rossle and their associates, however, offer
hope that stenosis can be prevented. [57]
Transjugular intrahepatic portosystemic shunts were first used
when Rosch and colleagues, while attempting a relatively
noninvasive way of creating a transjugular,
hepatovenous-choledochal fistula, accidentally induced a
hepatovenous-portovenous fistula in a dog. [42] Recognizing the
potential therapeutic value of such an anastomosis, they
reported the creation of a transjugular portacaval shunt. For 20
years, Rosch and coworkers continued investigations that
culminated in the development of a transjugular portacaval
shunt, which was first used in a human by Richter et al in 1989.
[41]
In principle, the procedure requires that an expandable metal

stent on a coaxial catheter be inserted from the jugular vein into
a hepatic vein and then

137
through a parenchymal tract in the liver from the hepatic to the

portal vein. This procedure required the serial passage of
coaxial catheters of increasing size until a stent that extended
from the hepatic to the portal vein was implanted.
The development of TIPS shunts has progressed rapidly. Rossi
and his associates in Rome studied both balloon-expandable
and, later, self-expanding stents, focusing on the length,
diameter, flexibility, and radiopacity of the stents to select the
most appropriate materials for the individual patient. [45] They
studied nitinol and tantalum stents, the latter of which are
highly radiopaque. Polylactic and polyglycolic acid and other
absorbable materials were studied to determine the feasibility of
using temporary, absorbable stents. Such stents are intrinsically
less strong and more bulky. Efforts were made to coat the stents
with heparin, which can be covalently bonded perpendicular to
the polymeric surface. Theoretically, the use of molecules that
are recognized by endothelial cells but not by activated
circulating cells creates an artificial vascular surface that is
rapidly endothelialized. Medical-grade steel of the 300 series
(eight- or nine-element alloys) have the most desirable
properties for intravascular stenting (Table 1) . Electropolishing
removes most of these accessory elements from the surface,
leaving behind a high concentration of chromium, which forms
a thin patina of chromium oxide that prevents further oxidation.
Depending on the composition of the alloy, oxides of tantalum,
titanium, or other metals form the ultimate interface with the
host. Figure 2 (Figure Not Available) shows a time sequence for
the changes that occur after the exposure of a prosthetic,
metallic material to the luminal surface of blood vessels.
Self-expanding stents have many advantages. Their major
disadvantage is their shortening on expansion, which can
amount to 35% of the length of the unexpanded stent. Most of
the shortening occurs near the proximal end of the stent.
Such stents can be implanted in almost all patients and case
decrease the portal venous pressure gradient in half, from 19 or
20 mm Hg to 9 or 10 mm Hg.
Siegerstetter et al compared either Palmaz or Wallstents in 24
cirrhotic patients, with or without periprocedural heparin
therapy, (24 U/kg, followed by intravenous heparin for 24
hours, followed by subcutaneous heparin, 0.3 mL/d, for 1
week). [58] Stent thrombogenicity was determined
scintigraphically using Tc99m-labeled platelets. Wallstents

showed a significantly higher risk of shunt malfunction than

Palmaz stents. Heparin was needed with the wallstints but not
with the Palmaz stents. In experiments in dogs, 17% of
Wallstents in the femoral arteries occluded, but none of the
Palmaz stents occluded. The difference
TABLE 1 -- METALLIC COMPOSITION OF
INTRAVASCULAR DEVICES
Device Composition
Z stent 304 stainless steel
Palmaz stent 316-L stainless steel
Medinvent stent Mediloy (unknown composition)
Strecker stent Tantalum
Nitinol stent Nitinol
Greenfield filter 316-L stainless steel, beta3 titanium
Bird's nest filter 304 stainless steel
Gunter filter 304 stainless steel
Gianturco coils 304 stainless steel
Data from Conn HO, Palmaz J, Rosch J, et al (eds):
Transjugular intrahepatic portosystemic stent-shunts. New
York, Igaku Shoin Medical Publishers, 1996, pp 68, 151, 178;
with permission.
138
Figure 2. (Figure Not Available) Time sequence of the events that take
place after exposure of prosthetic metallic surfaces in close relationship
with the luminal surface of vessels. Warry lines=fibrinogen; horizontal
lines=intercellular matrix. (From Conn HO, Palmaz J, Rosch J, et al:
Transjugular intrahepatic portosystemic stent-shunts. New York, Igaky
Shoin Medical Publishers, 1996; with permission.)
was attributed to the superiority of the slotted steel tube stents
(Palmaz) over the woven type stents (Wallstents).
Similarly, the expanded polytetrafluoroethylene
(ePTFE)-covered stents were shown by DiSalle et al to be
superior to the uncovered Wallstents. [15]
In another recent, small series by Lotterer et al, 21 patients had
TIPS shunts implanted for HEV. [35] The investigators
accomplished their primary goal by reducing portal venous

pressure from 21 to 11 mm Hg, increasing cardiac output from

7.1 to 8.9 L/min reducting systemic vascular resistance from
990 to 660 dynes/s/cm5 , and decreasing azygos blood flow
from 474 to 375 mL/min. These hemodynamic effects were
short-lived and returned to pre-implantation levels within 3
months.
The TIPS shunt procedure can be accomplished in almost all
patients with an acceptable rate of complications. Several
groups have reported on their first 100 TIPS shunt procedures.
[33] [47] [49] The primary complications (hepatic encephalopathy
and late-onset stenosis) are considered in detail later.

Rosemurgy et al performed a randomized, controlled trial in
which all 70 patients with persistent HEV were randomly
assigned in pairs to receive either TIPS shunt or small-diameter
H-graft portacaval shunts (SDHGPCS). [43] The TIPS shunting
procedure was performed through a right jugular approach
under general anesthesia using Wallstents that were 10 mm in
diameter and 68 mm long. The portacaval shunts were ePTFE, 8
mm in diameter. The two groups were matched for age, sex,
hepatic manifestations and Child-Pugh class. In the two groups,
the percentages of elective (70%), urgent (20%), or emergent
(10%) procedures were similar. Preshunt portal pressure levels
and gradients were also similar (30-32 and 17-18 mm Hg,
respectively). Postshunt portal pressure gradients were lower
after H grafts (6 mm Hg) than after TIPS shunting (10 mm Hg).
The 30-day mortality was the same in the two groups (35%),
and the most frequent cause of death in both groups was HEV.
Inability to complete the shunts was higher with TIPS shunt
(6%) than with PCS shunt implantations
139
(0%). Recurrent HEV and shunt failures were more frequent

after TIPS shunting than after PCS shunting. The authors
concluded that the TIPS procedure was associated with more
deaths, more rebleeding, and more treatment failures than the
PCS, procedure but the differences were not statistically
significant. The authors favor PCS shunts over TIPS shunts.
Kuhn-Fulton et al compared the efficacy of 10-mm and 12-mm
Wallstents in two groups of 23 patients each with portal
hypertension of various causes. [30] The study was performed in
sequential, nonrandomized fashion, first with 23 consecutive
10-mm stents, and then with 23 consecutive 12-mm stents. A
target portosystemic gradient (PSG) of 10 mm Hg was achieved

by balloon dilation of the shunts. Maximal dilation was defined

as 12 mm for 10-mm stents and 14 mm for 12-mm stents. The
TIPS shunt placement was successful in 95% of patients. The
groups were similar in age, gender, Child-Pugh class, and
clinical indications. For the two shunts, the mean diameter was
9.2 and 11 mm, the, PSG was 10.6 and 9.6 mm Hg, and flow
velocities were 1.7 and 1.3 m/s respectively. Patency rates were
measured at 1, 3, 6, and 12 months and progressively fell from
96% to 42% with the 10-mm stents and from 78% to 56% with
the 12-mm stents. The investigators concluded that 12-mm
stents yielded significantly poorer short- and long-term results,
probably caused by the decreased radial strength of the longer
stents. Whether these differences are caused by stent shortening,
overdilation during calibration, or the learning effects inherent
in sequential studies is not known. Clearly, in the future such
studies should be randomized, or alternated, not consecutive.
Sanyal and associates compared the cell phenotypes in patients
with TIPS with stenosed stents and those without stenosis. [51]
The TIPS shunt tissue was composed of collagen and palisades
of mesenchymal cells and was lined by an endothelium.
Immunostaining showed that alpha-smooth muscle staining was
strong and uniform in mesenchymal TIPS and peri-TIPS cells.
The TIPS cells expressed high levels of type 1 procollagen in
RNA but contained less fibronectin and relatively more type III
collagen than the peri-TIPS cells. The investigators found no
differences in clinical features, time since stenting, histology,
gross morphology, or the presence of bile fistulae on cell
phenotype. They did find that smooth muscle cells (SMC) from
stenotic stents exhibited greater cell proliferation and collagen I
and III secretion than did nonstenosed stents. The significance
of these observations is not known.
Rossle of the University of Freiburg has solved the most
difficult problems involved in the creation of a TIPS shunt
locating the best site of entry to the hepatoportal venous vessels.
[46] In the past 8 years, his group has accumulated a large and
successful series of TIPS shunt implantations, representing

more than 1000 cases.
Table 2 (Table Not Available) lists Rossle's 10 rules for a
successful TIPS shunt implantation. Some of these rules are
obvious, but others are more subtle. For example, rule 2,
collaborate with experts, emphasizes that, when dealing with
problems that are complex or poorly understood, such as the
factors that lead to endothelialization, thrombosis, or stent
stenosis, it is wise to seek the help of experts in coagulation, in

endothelial proliferation, or in more narrow disciplines.

Shunt stenosis is a case in point. Stenosis of TIPS shunts is just
beginning to emerge as a specific discipline, because some
clues to its pathogenesis now suggest that it is a simple
manifestation of a much more complex phenomenon.
Saxon, et al at the Dotter Interventional Institute in Portland,
Oregon, recognized that almost half of their TIPS shunt patients
and more than three fourths of their experimental pigs with
severe stenosis of TIPS shunts exhibited evidence of a
substantial biliary fistula. [55] In subsequent studies they
speculated
140
TABLE 2 -- GUIDELINES TO THE TIPS PROCEDURE

(Not Available)
From Conn HO, Palmaz J, Rosch J, et al (eds): Transjugular
intrahepatic portosystemic stent-shunts. New York, Igaku
Shoin Medical Publishers, 1996, p 178; with permission.
the mucus in the biliary secretion was a likely inducer of

pseudointimal hyperplasia that might give rise to myoblast and
smooth muscle proliferation and, ultimately, to collagen
deposition and shunt stenosis. Indeed, they have demonstrated
that the injection of radiopaque contrast medium into stenosed
shunts may show extensive filling of the biliary tree (Fig. 3) ,
implying that the bile leakage may be the stimulus for TIPS
shunt stenosis. Thus, bile fistulae may be the origin of shunt
stenosis. These observations have been supported by the
histopathologic studies of LaBerg et al. [32]
It has been inferred that free bile in the tissue is an irritating,
inflammatory substance that may give rise to thrombosis or
stenosis. Further studies may provide important information
about this phenomenon. The way in which biliary secretion
relates to the underlying endothelialization process may well
shed light on the problem of stenosis. It must be remembered
that the endothelialization of a steel or other framework is
merely the initial, critical phase of the stenting procedure. With
incredible rapidity, the skeletal stent is covered by platelets to
which fibroblasts and myoblasts are attracted (see Fig. 2)
(Figure Not Available) , and within hours or days a fibrous

coating completely lines the stent, converting it to an

endothelium-lined tube which cannot be differentiated from the
endothelium of a normal blood vessel by electron microscopy
and which functions as a blood vessel. The mechanism that
signals the newly formed endothelium to stop proliferating may
not function properly so that the thickness of this neointima
increases until the lumen of the stent is decreased, permitting
the process of thrombosis to take over.
The following sections discuss the clinical indications (Table 3)
(Table Not Available) and contraindications (Table 4) (Table
Not Available) for TIPS shunt implantation.
INDICATIONS FOR TRANSJUGULAR

INTRAHEPATIC PORTOSYSTEMIC
SHUNTS
The primary indication for implanting a TIPS shunt is to
prevent HEV. [3] Initially TIPS shunting was considered to be a
novel, noninvasive method of creating a portosystemic
anastomosis. Indeed, a TIPS shunt can promptly reduce the
portal venous pressure so esophageal varices stop active
bleeding almost instantaneously and can prevent recurrences of
bleeding. As would be expected, a TIPS shunt also appears to
prevent bleeding from gastric and other gastrointestinal lesions,
including intestinal, anorectal, and stomal varices.
141
Figure 3. Extensive filling of the biliary tree 10 months

after transjugular intrahepatic portosystemic shunts (TIPS)
implantation after injection of contrast medium into the
occluded lumen of a TIPS shunt.
The second major indication for TIPS shunting is the treatment

of cirrhotic ascites. Portal hypertension is the driving force in
the pathogenesis of ascites, and the restoration of normal portal
pressure suppresses the formation of ascitic fluid and results in
the reduction of ascites. Because cirrhotic hydrothorax is often
an extension of ascites, TIPS shunting is a reasonable and
effective method of treating hydrothorax. Similarly, because the
hepatorenal syndrome (HRS) can also indicate ascites formation

[38] it, too, can be effectively treated by TIPS shunting. [28] More
recently, it has been demonstrated that the hepatopulmonary

syndrome (HPS), which also has its pathogenetic roots in portal
hypertension, may also be effectively treated by TIPS shunts.
The management of ascites is one of the most important and
frequent reasons for performing TIPS shunting. The treatment
of ascites is more than simply preventing the formation of
peritoneal fluid. Trotter et al studied serially the effects of TIPS
shunts in 35 patients with refractory ascites. [65] They
retrospectively analyzed the effects of TIPS shunts on body
weight, on the volume of ascites, and on the Child-Pugh score
before, 2 months after, and 9 months after implantation. The
mean portal venous pressure gradient fell from 21.5 to 8.1 mm
Hg. Ascites was completely resolved in 23 of 24 patients (96%)
within the first 2 months, during which the patients showed a
mean weight loss of 2.1 kg,
142
TABLE 3 -- INDICATIONS FOR TIPS

(Not Available)
Adapted from Conn HO, Palmaz J, Rosch J, et al (eds):
York, Igaku Shoin Medical Publishers, 1996, p 151; with
permission.
which paralleled the decrease in ascites. Between 2 months and

9 months, however, there was a mean weight gain of 5.5 kg
without the reappearance of ascites. It was thought that the
weight gain reflected an increase in lean body mass. During this
7-month period, the mean albumin concentration increased from
2.7 to 2.9 g/dL, and the mean Child-Pugh score improved from
9.7 to 8.2 ( P < 0.05). This pattern is demonstrated in a
photograph by Ochs et al (Fig. 4) . [38] Similar changes have
been reported after implantation of side-to-side portacaval
anastomoses and peritoneovenous shunts and have been
attributed largely to improved appetite and nutrition after the
delivery of the ascites.
The TIPS shunting procedure also been used effectively to treat
hepatic vein obstruction, (Budd-Chiari syndrome) and portal

vein thrombosis, which may be considered fraternal-twin

syndromes of the portal circulation.
Transjugular intrahepatic portosystemic shunts have also been
used as therapy for thoracic duct cutaneous fistulae and for bile
duct fistulae to the hepatic or portal veins. The procedure is also
considered to be a bridge to liver transplantation
TABLE 4 -- CONTRAINDICATIONS TO TIPS
(Not Available)
Adapted from Conn HO, Palmaz J, Rosch J, et al (eds):
York, Igaku Shoin Medical Publishers, 1996, p 151; with
permission.
143
Figure 4. Ascitic patient before ( A)

and after transjugular intrahepatic
portosystemic shunts (TIPS) ( B).
Note the disappearance of ascites and
the improvement in the patient's
nutritional state.
for patients who are deteriorating while awaiting liver

transplantation because of variceal bleeding or ascites. [37]
A secondary indication for TIPS shunt implantation is the relief
of protein-losing enteropathy. [63] Studies using Tc 99m-labeled
human serum albumin permit the diagnosis of the disorder and
show the extent of the bowel involved. The condition can also
be assessed by measuring the amounts of albumin, alpha1
-antitrypsin and other substances in the stool following orally
administered large amounts of water.
Transjugular intrahepatic portosystemic shunts have been used
to treat gastric, duodenal, intestinal, and stomal varices and the
angiodysplastic lesions of colopathy. [2] It is an effective
alternative therapy for patients who are not responsive to
adrenergic beta-blocker therapy, endoscopic sclerotherapy, or
ligation of varices. The TIPS shunt is also indicated for the
treatment of hypersplenism. [1]

CONTRAINDICATIONS FOR
There are a number of contraindications to TIPS shunt
implantation (Table 4) (Table Not Available) . Among the most
important contraindications is heart failure, because the
congested heart cannot handle the additional vascular load
imposed by a TIPS shunt. Similarly, the inability of the liver to
carry out its normal functions in fulminant hepatic failure or in
decompensated hepatic encephalopathy (HE) indicates that the
failing liver may not be able to deal with the additional hepatic
metabolic load imposed by the TIPS shunt. Celiac or mesenteric
arterial stenosis
144
may likewise interfere with the additional metabolic loads of

TIPS shunts. If active bacterial or fungal infections are present,
TIPS shunting may overload the body's capacities in this time
of stress. Finally, the presence of polycystic liver disease is an
absolute contraindication to implanting a TIPS shunt, to avoid
the creation of an enormous hepatoportoperitoneal anastomosis.
COMPLICATIONS
The complications of TIPS shunts, which seem almost infinite
in number and type, can be classified as the technical
misadventures that are related directly to problems encountered
in placement of the stent and as physically and temporally more
remote adverse events (Table 5) (Table Not Available) . [56] The
misadventures can include errant punctures of nearby vessels or
organs and their consequences, such as the induction of
bleeding fistulae between vascular, biliary, and lymphatic
vessels and the pneumothorax. Stents may be implanted in
incorrect positions or may migrate some time after
implantation. Complications also
TABLE 5 -- COMPLICATIONS ASSOCIATED WITH
PLACEMENT
(Not Available)

Adapted from Freedman AM, et al: Radiographics

13:1185-1210, 1993; with permission.
145
include cardiac arrhythmias that may be caused by the touch of

a catheter and may end up as myocardial infarctions or
pulmonary emboli.
The inadvertent consequences of creating an intrahepatic
parenchymal tract in which the shunt is deployed are true
technical complications. Indeed, as indicated previously, there
are reasons to believe that stenosis may be related to the leakage
of bile or mucus into this tissue tract.
Allergy to any of the substances used in performing the
procedure, ranging from anesthesia to idiosyncratic reactions to
stent materials, may give rise to a variety of syndromes, as may
too much irradiation. Renal tubular injury related to the use or
abuse of contrast media is another problem that is often
encountered.
Infections with bacteria or fungi are uncommon complications
that may involve the stent, the cardiac valves, or perhaps more
disseminated sites such as the bloodstream (e.g., enterococcal
bacteremia). Late episodes of endocarditis may be a problem,
because the risks of these infections are increased in patients
who have portosystemic anastomoses of all types.
Among the most important of the late-onset complications are
HE and stent stenosis. Somberg et al reported their considerable
experience with post-TIPS HE. [61] At the University of
California, San Francisco, TIPS shunts were successfully
emplaced in 108 patients during a 22-month period. Hepatic
encephalopathy could be accurately assessed in 77 of these
patients. New-onset HE or overt worsening of pre-existing HE
within 1 year were the endpoints observed. Eighteen of the 77
patients (23%) developed new or worsened HE. Fifteen of the
50 patients who had never experienced HE (30%) previously
developed new HE. Three of the 27 patients who had previously
experienced HE developed worsened HE (11%). Treatment of
the HE with lactulose resulted in improvement in 78% of the
patients, but in 22% the HE progressed. The risk of HE tended
to be greater in patients with nonalcoholic cirrhosis (relative
risk 9.2), in women (relative risk 3.0), and in those with
hypoalbuminemia (relative risk 2.2 for each g/dL decrease in
the albumen concentration). This large study confirms the 20%

to 30% prevalence of post-TIPS HE observed by so many
investigators.
New complications of TIPS shunting continue to emerge.
Sanyal et al reported on the occurrence in a single patient of
acute, hemolytic anemia. [50] In a subsequent prospective
systemic analysis, they compared 60 patients with TIPS shunts
and 40 control subjects who were receiving sclerotherapy. [52]
These investigators found five patients with acute hemolysis
(8%), compared with none in the control group (0%). This
self-limited disorder appears to be largely mixed-type
hemolysis which is virtually asymptomatic. The report was
followed by an editorial [48] suggesting that complete
endothelialization of the stent prevents hemolysis, but the
presence of some naked steel wires in the blood stream may
cause traumatic injury to erythrocytes and provoke the acute
hemolysis.
In 1998, Rossle et al. [48] They reviewed the state of the art after
10 years of experience with TIPS shunts. They described the
crescendo of published articles on TIPS shunts, starting with a
handful in 1988 and 1989 and surpassing 100 papers in both
1996 and 1997. They classified the more than 500 publications
by topic, assessed the reported complications, confronted the
problem of shunt stenosis (malfunction), considered the
indications (old and new) and contraindications, and predicted
the future of this procedure.
In 1997, the author of this review reported on the current status
of TIPS. [10] This update summarizes a nonrandomized clinical
trial that compared TIPS shunts with esophageal transection and
devascularization (ETDV) in 38 patients. Either Palmaz stents
or Wallstents were implanted successfully in 90% of the
patients. The overall 30-day mortality rate was higher in the
ETDV group than
146
in the TIPS group (70% versus 42%). The report shows one of
the uncertainties of performing nonrandomized, clinical
investigations on small numbers of patients.
Also cited was a consecutive series of 96 TIPS shunt
implantations performed by Coldwell et al at eight, large,
tertiary medical centers, which demonstrated that TIPS shunting
can be performed successfully at multiple large
university-affiliated hospitals in the United States with
acceptable success (100%). The mortality rates (14%), and

primary patency rates (88%) may not be within acceptable
levels, however. Hepatic encephalopathy was noted in 31% of
these patients.
The most worrisome delayed complication of TIPS
emplacement is late-onset stenosis. [54] It seems to be a very
nebulous problem with no obvious points at which it can be
attacked. Reviewing what is known about these stenoses, Rossle
et al [57] observed that they usually occur in the hepatic vein
distal to the stent. These investigators clarified the anatomical
locations of the stenoses and created a workable classification.
They distinguished two types of stenoses: type 1, the
thrombogenic type that is located within the stent itself, and
type 2, the proliferative type that is located in the draining
hepatic veins.
Type 1 stenosis is caused by thrombus formation [17] [32] and is
theoretically treatable by anticoagulation or by platelet
inhibition. [53] Type 2 stenosis is thought to be the result of
localized intimal proliferation in the hepatic veins that may be a
general characteristic of the stenting process per se. [34] Indeed,
endothelialization is characteristic of stenting in peripheral
vascular [16] and coronary artery stents [31] and is the essence of
the stenting process that converts a steel skeleton into a vascular
structure. Endothelialization is the response to injury that
attracts platelets and macrophages and releases platelet-derived
growth factor (PDGF) [44] downstream, which stimulates the
activation of smooth muscle cells. [36] Inhibitors of PDGF may,
therefore, prevent type 2 stenosis. Investigators tested this
hypothesis in a randomized, controlled trial by comparing
anticoagulation with trapidil, a potent PDGF inhibitor, [39] along
with ticlopidine, which inhibits platelet aggregation. [18] This
investigation gave remarkable results. The anticoagulation
prevented thrombosis (i.e., type 1 stenosis), and the inhibitors of
PDGF and platelet aggregation prevented the activation of the
proliferation of smooth muscle cells (type 2 stenosis). These
breakthrough findings represent the first effective, rational
forms of therapy of TIPS shunt stenosis and may lead to a
means of preventing late-onset, postimplantation stenosis.
In these studies, Siegerstetter et al studied 84 patients with TIPS
shunts. Forty-two randomly selected patients received heparin
(12-24 U/kg) at the time of implantation, followed by 1 week of
treatment with intravenous heparin and 4 weeks of treatment
with subcutaneous heparin. The other 42 patients received
trapidil plus ticlopidine, which suppressed almost all of the type

2 stenoses and also reduced the number of type 1 stenoses.

Fifty-seven percent of the patients receiving heparin developed
stenosis compared with 33% of patients receiving
antiproliferative the therapy ( P < 0.05). This trend, which was
statistically significant after 6 months, continued as long as the
treatment was continued but diminished after the therapy was
stopped. Portal hemodynamics and coagulation parameters were
similar in the two groups of patients were similar before
therapy. About half of both groups had either Palmaz or
Memotherm stents implanted, with appropriate decreases in the
portal venous pressure and flow velocity. Intention-to-treat
analysis showed that type 1 stenoses were significantly reduced
by anticoagulation therapy in both groups, and type 2 stenoses
were decreased by suppression of PDGF and platelet
aggregation. The fact that stenoses, as measured by duplex
ultrasonography, began to recur after discontinuation of the
medications suggests that long-term treatment--perhaps
147
permanent therapy--is needed. Clearly, anti-stenotic therapy is

effective. It is hoped that further refinements will improve its
efficacy and safety and eradicate the single biggest obstacle to
TIPS shunt therapy.
Another approach to the prevention of stenosis has been the
implantation of Wallstents covered with silicone to avoid the
stenosogenic effects of native substances in the parenchymal
tract. [64] Although early studies with tetrafluoropolyethylene
coating appears advantageous, [15] the silicone-covered stents
were a failure. [64]
References
1. AlvarezOA, Lopera GA, Vinaychandra P, et al: Improvement of

thrombocytopenia due to hypersplenism after transjugular intrahepatic
portosystemic shunt placement in cirrhotic patients. J Gastroenterol
91:134-137, 1996
2. BalzerC, Lotterer E, Kleber G, et al: Transjugular intrahepatic

portosystemic shunt for bleeding angiodysplasia-like lesions in
portal-hypertensive colopathy. Gastroenterology 115:167-172, 1998
abstract
3. BennerKG: Indications for and contraindications to TIPS. In Conn HO,

Palmaz J, Rosche J, et al: (eds): Transjugular Intrahepatic Portosystemic
Stent-Shunts (TIPS) New York, Igaku Shoin Medical Publishers, 1996

4. BeppuK, Inokuchi K, Koyanagi H: Prediction of variceal hemorrhage

by esophageal endoscopy. Gastrointest Endosc 27:213-223, 1981
abstract
5. Bosch J, D'Amico G, Luca A, et al: Drug therapy for variceal

hemorrhage. In Bosch J, Groszmann RJ (eds): Portal Hypertension:
Pathophysiology and Treatment. Oxford, Blackwell Scientific
Publications, 1994, pp 108-123
6. Bosch J, Groszmann RJ, Garcia-Pagan JC, et al: Association of

transdermal nitroglycerin to vasopressin infusion in the treatment of
variceal hemorrhage: A placebo-controlled trial. Hepatology 10:962-968,
1989 abstract
7. Burroughs AK, Jenkins WJ, Sherlock S: Controlled trial of propranolol

for the prevention of recurrent variceal hemorrhage in patients with
cirrhosis. N Engl J Med 309:1539-1542, 1983 abstract
8. Cales P, Desmorat H, Vinel JP: Incidence of large oesophageal varices

in patients with cirrhosis: Application to prophylaxis of first bleeding. Gut
31:1298-1302, 1990 abstract
9. Conn HO: Emergency portacaval shunts in variceal hemorrhage: A

critique of critics. HPB Surg 8:154-157, 1994
10. ConnHO: Transjugular intrahepatic portosystemic shunts (TIPS).

Current Hepatology 17:251-273, 1997
11. Conn HO, Grace ND, Bosch J: Propranolol in the prevention of the first
hemorrhage from esophageal varices: Results of a randomized double
blind, cooperative clinical trial. Hepatology 13:902-912, 1991 abstract
12. ConnHO, Palmaz J, Rosch J, et al (eds): Transjugular Intrahepatic

Portosystemic Stent-Shunts (TIPS). New York, Igaku Shoin Medical
Publishers, 1996
13. D'AmicoG, Pagliaro L, Bosch J: The treatment of portal hypertension:

A meta-analytic review. Hepatology 22:332-354, 1995 citation
14. deFranchis R, Pascal JP, Ancona E: Definitions, methodology and

therapeutic strategies in portal hypertension: A consensus development
workshop. J Hepatol 15:256-261, 1992 citation
15. DiSalleRS, Dolmatch BL: Treatment of TIPS stenosis with ePTFE

graft-covered stents. Cardiovasc Intervent Radiol 21:172-175, 1998
abstract
16. Do-dai-Do,Triller J, Walpoth BH, et al: A comparison study of

self-expandable stents vs balloon angioplasty alone in femoropopliteal

artery occlusions. Cardiovasc Intervent Radiol 15:306-312, 1992 abstract
17. Ducoin H, El-Khoury J, Rousseau H, et al: Histopathologic analysis of

transjugular intrahepatic portosystemic shunts. Hepatology 25:1064-1069,
1997 abstract
18. Flores-Runk
P, Raasch RH: Ticlopidine and antiplatelet therapy. Ann
Pharmacother 27:1090-1098, 1993 abstract
148
19. Garceau AJ, Chalmers TC, the Boston Inter-Hospital Liver Group: The
natural history of cirrhosis: I. Survival with esophageal varices. N Engl J
Med 268:469-473, 1963
20. Genecin P, Groszmann RJ: Portal hypertension. In Schiff ER (eds).

Diseases of the Liver, ed 7. Philadelphia, Lippincott, 1993, pp 935-973
21. GenecinP, Polio J, Groszmann RJ: Sodium restriction blunts expansion

of plasma volume and ameliorates hyperdynamic circulation in portal
hypertension. Am J Physiol 259:G498-G505, 1990 abstract
22. GimsonAE, Westaby D, Hegarty J: A randomized trial of vasopressin

and vasopressin plus nitroglycerin in the control of acute variceal
hemorrhage. Hepatology 6:410-413, 1986 abstract
23. Gimson
AE, Westaby D, Williams R: Endoscopic sclerotherapy in the
management of gastric variceal hemorrhage. J Hepatol 13:274-278, 1991
abstract
24. Grace ND: Diagnosis and treatment of gastrointestinal bleeding

secondary to portal hypertension. Am J Gastroenterol 92:1081-1091, 1997
full text
25. Grace ND: Variceal hemorrhage; pharmacologic approach. In

McDermott WV, Both A (eds): Surgery of the Liver. Boston, Blackwell
Scientific, 1988, pp 303-314
26. Graham D, Smith JL: The course of patients after variceal hemorrhage.
Gastroenterology 80:800-809, 1981 abstract
27. GregoryPB, The VA Cooperative Variceal Sclerotherapy Group:

Sclerotherapy for male alcoholics with cirrhosis who have bled from
esophageal varices: Results of a randomized multicenter trial. Hepatology
20:618-625, 1994 abstract
28. Guevara M, Gines P, Bandi JC, et al: Transjugular intrahepatic

portosystemic shunt in hepatorenal syndrome: Effects on renal function

and vasoactive systems [comments]. Hepatology 28:416-422, 1998
abstract
29. HaskalZJ, Pentecost MJ, Soulen MC: Transjugular intrahepatic

portosystemic shunt stenosis and revision: Early and midterm results. AJR
Am J Roentgenol 163:439-444, 1994 abstract
30. Kuhn-Fulton J, Trerotola SO, Harris VJ, et al: Transjugular intrahepatic

portosystemic shunt procedure: Efficacy of 10-mm versus 12-mm
Wallstents. Radiology 199:658-664, 1996 abstract
31. Kuntz RE, Gibson CM, Nobuyoshi M, et al: Generalized model of

restenosis after conventional balloon angioplasty, stenting and directional
atherectomy. J Am Coll Cardiol 21:15-25, 1993 abstract
32. LaBerge JM, Ferrell LD, Ring EJ, et al: Histopathologic study of
stenotic and occluded transjugular intrahepatic portosystemic shunts. J
Vasc Interv Radiol 4:779-786, 1993 abstract
33. LaBerge JM, Ring EJ, Gordon RL: Creation of transjugular intrahepatic
portosystemic shunts with the Wallstent endoprosthesis: results in 100
patients. Radiology 187:413-420, 1993 abstract
34. LindCD, Malisch TW, Chong WK, et al: Incidence of shunt occlusion
or stenosis following transjugular intrahepatic portosystemic shunt
placement. Gastroenterology 106:1277-1283, 1994 abstract
35. Lotterer
E, Wengert A, Fleig WE: Transjugular intrahepatic
portosystemic shunt: Short-term and long-term effects on hepatic and
systemic hemodynamics in patients with cirrhosis. Hepatology
29:632-639, 1999 abstract
36. Marmur JD, Poon M, Rossikhina M, et al: Induction of

PDGF-responsive genes in vascular smooth muscle. Implications for the
early response to vessel injury. Circulation 86:53-60, 1992 abstract
37. MartinL, Dondelinger RF, Trotteur G: Treatment of Budd-Chiari

syndrome by metallic stent as a bridge to liver transplantation. Cardiovasc
Intervent Radiol 18:196-199, 1995 abstract
38. OchsA: Effects of TIPS on ascites and related complications. In Conn

HO, Palmaz J, Rosche J, et al (eds): Transjugular Intrahepatic
Publishers, 1996, pp 229-267
39. OhnishiH, Yamaguchi K, Shimada S, et al: A new approach to the

treatment of atherosclerosis and trapidil as an antagonist to platelet-derived
growth factor. Life Sci 28:1641-1646, 1981 citation

40. OrloffMJ, Orloff MS, Rambotti M: Treatment of bleeding

esophagogastric varices due to extrahepatic portal hypertension: Results of
portal-systemic shunts during 35 years. J Pediatr Surg 29:142-154, 1994
abstract
149
41. Richter GM, Palmaz JC, Noldge G: The transjugular intrahepatic

portosystemic stent-shunt (TIPSS): A new nonoperative, transjugular
percutaneous procedure. Radiologe 29:406-411, 1989 abstract
42. Rosch J, Hanafee WN, Snow H: Transjugular portal venography and

radiologic portacaval shunt: An experimental study. Radiology
92:1112-1114, 1969 citation
43. Rosemurgy AB: A prospective trial of transjugular intrahepatic

portasystemic stent shunts versus small-diameter prosthetic H-graft
portacaval shunts in the treatment of bleeding varices. Ann Surg
224:378-387, 1991
44. Ross
R: Platelet-derived growth factor. Annu Rev Med 38:71-79, 1987
abstract
45. Rossi P, Maccioni F, Bazzi M: TIPS: Technical aspects and clinical

results. HIT Angiography and Interventional Radiology Topics 9:1-17,
1994
46. RossleM: The implantation and maintenance of TIPS: Rules of the

road and tricks of the trade. In Conn HO, Palmaz J, Rosche J, et al (eds):
Transjugular Intrahepatic Portosystemic Stent-Shunts (TIPS). New York,
Igaku Shoin Medical Publishers, 1996
47. RossleM, Haag K, Ochs A: The transjugular intrahepatic portosystemic

stent-shunt procedure for variceal bleeding. N Engl J Med 330:165-171,
1994 abstract
48. RossleM, Siegerstetter V, Huber M, et al: The first decade of the

transjugular intrahepatic portosystemic shunt (TIPS): State of the art. Liver
18:73-89, 1998 abstract
49. Sahagun G, Benner KG, Saxon R: Clinical outcome of 100 patients

after transjugular intrahepatic portosystemic shunt (TIPS) for variceal
hemorrhage. Am J Gastroenterol 92:1412-1416, 1997 full text
50. Sanyal AJ: Hemolytic anemia following TIPS. In Conn HO, Palmaz J,
Rosche J, et al (eds): Transjugular Intrahepatic Portosystemic Stent-Shunts
(TIPS). New York, Igaku Shoin Medical Publishers, 1996, pp 301-307

51. SanyalAJ, Contos MJ, Yager D, et al: Development of pseudointima

and stenosis after transjugular intrahepatic portasystemic shunts:
Characterization of cell phenotype and function. Hepatology 28:22-32,
1998
52. Sanyal AJ, Freedman AM, Purdun PP: The hematological consequences
of transjugular intra-hepatic portosystemic shunts. Hepatology 23:32-39,
1996 abstract
53. SauerP, Theilmann L, Herrmann S, et al: Phenprocoumon for

prevention of shunt occlusion after transjugular intrahepatic portosystemic
stent shunt: A randomized trial. Hepatology 24:1433-1436, 1996 abstract
54. SaxonRR, Barton RE, Keller FS: Stenosis and occlusion after TIPS:
Prevention and correction. In Conn HO, Palmaz J, Rosche J, et al (eds):
Transjugular Intrahepatic Portosystemic Stent-Shunts (TIPS). New York,
55. Saxon RR, Mendel-Hartvig J, Corless CL, et al: Bile duct injury as a
major cause of stenosis and occlusion in transjugular intrahepatic
portosystemic shunts: Comparative histopathologic analysis in humans and
swine. J Vasc Interv Radiol 7:487-497, 1996 abstract
56. Schiffman ML, Cole PE: Complications of TIPS implantation. In Conn

HO, Palmaz J, Rosche J, et al (eds): Transjugular Intrahepatic
Publishers, 1996
57. SiegerstetterV, Huber M, Ochs A, et al: Platelet aggregation and

platelet-derived growth factor inhibition for prevention of insufficiency of
the transjugular intrahepatic portosystemic shunt: A randomized study
comparing trapidil plus ticlopidine with heparin treatment. Hepatology
29:33-38, 1999 abstract
58. Siegerstetter
V, Krauss T, Rossle M, et al: Transjugular intrahepatic
portosystemic shunt (TIPS). Thrombogenicity in stents and its effect on
shunt patency. Acta Radiol 38:558-564, 1997 abstract
59. Simpson KJ, Chalmers N, Redhead DN: Transjugular

intrahepatic-portasystemic stent shunting for control of acute and recurrent
upper gastrointestinal haemorrhage related to portal hypertension. Gut
34:968-973, 1993 abstract
60. Soderlund C, Magnusson I, Torngren S, et al: Terlipressin

(triglycyl-lysine vasopressin) controls acute bleeding oesophageal varices.
A double-blind randomized, placebo-controlled trial. Scand J
Gastroenterol 25:622-630, 1990 abstract
61. Somberg KA, Reigler JL, LaBerge JM, et al: Hepatic encephalopathy
after transjugular intrahepatic portosystemic shunts: Incidence and risk
factors. Am J Gastroenterol 90:549-555, 1995 full text

150
62. Stiegmann GV, Goff JS, Sun JH, et al: Endoscopic ligation of
esophageal varices. Am J Surg 159:21-26, 1990 abstract
63. TakedaH, Takahashi T, Ajitsu S, et al: Protein-losing gastroenteropathy

detected by technetium-99m m-labeled human serum albumin. Am J
Gastroenterol 86:450-453, 1991 abstract
64. Tanihata
H, Saxon RR, Kubota Y, et al: Transjugular intrahepatic
portosystemic shunt with silicone-covered Wallstents: Results in a swine
model. Radiology 205:181-184, 1997 abstract
65. Trotter
JF, Suhocki PV, Rockey DC: Transjugular intrahepatic
portosystemic shunt (TIPS) in patients with refractory ascites: Effect on
body weight and Child-Pugh score. Am J Gastroenterol 93:1891-1894,
1998 abstract



Volume 4 Number 1 February 2000
133
HEPATOLOGY: A CENTURY OF PROGRESS
PORTAL HYPERTENSION, VARICES, AND TRANSJUGULAR

INTRAHEPATIC PORTOSYSTEMIC SHUNTS
Harold O. Conn MD
Department of Medicine, The Yale University School of Medicine, New Haven Connecticut; and The University of Miami
School of Medicine, Miami, Florida
Harold O. Conn, MD
The Yale University School of Medicine
160 Morgan Avenue
East Haven, Connecticut 06512-4519
The longest
march starts with a single stride.
LAO-TZU
The primary topic of this article is portal hypertension and esophagogastric varices, the primary
clinical problem posed by portal hypertension. Transjugular intrahepatic portosystemic (TIPS)
shunts is the newest and the least invasive method of eradicating varices. This article defines portal
hypertension succinctly, describes how it gives rise to varices and their consequences, and briefly
reviews the development, short experience with, and current status of TIPS shunts.
Portal hypertension is the state of increased pressure in the portal venous system and its
tributaries and collaterals, including the hepatic sinusoids. The portal venous pressure is
determined by the amount of blood that flows through the portal venous system and the resistance
that opposes the blood flow. It can be expressed as a variant of Ohm's law, P = FR; that is, the
pressure gradient through the portal venous system, P, is determined by F, the volume of blood
flowing through the system, and R, the resistance to that flow. Changes in either F or R affect the
pressure. The portal venous blood flow is the total of the blood flow to the liver, the spleen, and the
splanchnic organs. In most types of portal hypertension, both the blood flow and the resistance to

the blood flow are altered.

Resistance is determined largely by the radius of the vessel, by the viscosity of the blood and by the
length of the vessel. This relationship, which is known as Poiseuille's law, can be expressed as R =
8nL/r4 , where n is the coefficient of viscosity and L is the length of the vessel, both of which are
essentially constant. Because the radius ( r) of the vessel is expressed as its fourth power, it
134
is by far the most mathematically important component of the formula. Thus, the smaller the
vessel, the greater the resistance to the passage of blood through that vessel. The main sites of
resistance to hepatic blood flow appear to be small portal and hepatic veins and the sinusoids.
The key to the syndrome of portal hypertension is an increase in plasma volume. [20] In
experimental animal models, sodium restriction has been shown to prevent plasma expansion and
the development of portal hypertension, [21] and thus the development of varices, and can minimize
the hyperdynamic state. The second most important factor in the development of portal
hypertension is the development of the portosystemic collateral circulation. The complications of
portal hypertension, which are consequences of the increased portal pressure, are mediated more
by the portosystemic collaterals than by the portal pressure itself.
PORTAL HYPERTENSION
Portal hypertension is usually caused by intrahepatic fibrosis or nodularity that increases resistance to the
egress of portal blood from the liver. Cirrhosis is the most common disorder that induces portal
hypertension, and the cirrhosis that induces portal hypertension may be of alcoholic, viral, drug-induced,
biliary, or some other origin. The portal hypertension of cirrhosis, which is a form of postsinusoidal
portal venous obstruction, that is, increased resistance to portal blood outflow, is defined as a sustained
portosystemic pressure gradient of 10 to 12 mm Hg, although the precise minimal level at which portal
hypertension begins may be as low as 7 mm Hg. The portosystemic pressure gradient is defined as the
occluded hepatic venous pressure minus the unoccluded, or free, portal venous pressure, that is, the
hepatovenous pressure gradient (HVPG).
All patients with cirrhosis or other chronic liver disease, such as chronic viral or alcoholic hepatitis, are
at risk of developing esophagogastric varices. These collateral vessels develop between the hypertensive
portal venous vessels and the systemic venous system, which is under normal pressure, in an attempt to
decompress the elevated portal venous pressure. Esophagogastric varices are prone to erode or to
explode, giving rise to hemonage of esophogastric varices (HEV), the most serious and lethal
complication of portal hypertension. Patients with varices should be periodically screened endoscopically
for the presence and size of varices, which are proportional to the severity of the liver disease. Even
patients with Child-Pugh class A cirrhosis should be screened for varices if the esophageal vein
diameters are enlarged (> 3 mm) or if the HVPG is increased (> 12 mm Hg). Patients with Child-Pugh
class B or C cirrhosis usually have varices, the growth of which parallels the increasing severity of the

cirrhosis. If small varices are found, frequent endoscopic assessments are indicated, because such varices
tend to grow in size. Furthermore, the three most important risk factors for the initial episode of HEV are
the severity of the liver disease, the size of the varices and the presence of endoscopic red markings
described by Beppu et al, [4] such as red spots or red wales, also known as varices on varices that are
considered signs of impending rupture.
The risks of rebleeding are extremely high in survivors of the first hemorrhage, especially during the first
week after hemorrhage, and the risks remain high for an additional 8 to 12 weeks. There are no
quantitative risk factors for recurrent bleeding other than the severity of the liver disease as evaluated by
clinical and laboratory findings, such as the total serum bilirubin concentration, serum albumin level, and
prothrombin times. When patients have survived for
135
3 to 6 months after an initial episode of HEV, the risks of rebleeding are reduced, but such patients are
probably always somewhat at risk.
There does not appear to be any clear association between the nature of the initial episodes of bleeding
(e.g., oozing or spurting) and the risk of recurrent HEV. It is widely agreed that the need for more than
two units of blood, which is usually associated with a systolic blood pressure of less than 100 mm Hg, a
pulse rate of more than 100 beats/min or a postural decrease of more than 20 mm Hg in systolic blood
pressure constitutes clinically significant bleeding.
Signs of instability that indicate diminished blood volume include failure to control HEV within 6 hours,
(i.e., the need for four or more units of blood to restore the systolic blood pressure levels to normal) and a
persistent, rapid pulse rate. Recurrence of bleeding, as demonstrated by hematemosis or by gastric
aspiration, by a decrease in the vital signs to the levels outlined previously, or by the continued need for
blood transfusions to maintain the hematocrit above 26%, is evidence that therapy has failed.
One third of all deaths of cirrhotic patients result from HEV, the most common signle cause of death
from cirrhosis. [19] Because almost 90% of cirrhotic patients develop varices [8] and about 30% of them
bleed from their varices, many patients are involved, of whom about 40% die of HEV. Recurrences of
HEV are common: 70% of patients who survive the initial hemorrhage bleed again. [26] The dilatation and
bleeding result from systemic vasodilation and a decrease in systemic vascular resistance. It is a complex
scenario involving nitric oxide (NO), glucagon, prostaglandins, tumor necrosis factor-alpha and other
substances that give rise to volume expansion, increments in cardiac output, and a hyperdynamic
circulation. [14] The vasodilation and hyperdynamic circulation result in the appearance of varices, dilated
intraesophageal veins with an extensive, tortuous network of capillary vessels that provide the vascular
power of these lesions. Varices develop when the hepatic vein pressure gradient surpasses 12 mm Hg.
Bleeding may occur at any time thereafter.
The diagnosis of varices is best made by endoscopic examination, as is the demonstration of actively
bleeding varices. Many effective treatments are available, including endoscopic obliteration of varices
either by sclerotherapy (EST) per se or esophageal variceal ligation (EVL). A transjugular intrahepatic
portosystemic shunt appears to be the most rapid and effective means of doing so. [13] Both techniques
immediately stop the bleeding, at least transiently, in more than 90% of patients. [12] Pharmacologic
therapies with vasopressin, somatostatin, terlipressin, or octreotide plus nitroglycerin [22] have all been

shown to be effective.
Much research has been devoted to the prevention of the initial, most critical episode of HEV.
Beta-adrenergic blockade appears to be the most successful form of treatment available. [11] These studies
are discussed in detail by Grace (Fig. 1) (Figure Not Available) . [24]
Somewhat less successful but equally important are the prevention and management of recurrent HEV by
pharmacologic, [7] or endoscopic [27] therapies or by both combined.
The treatment of the acute bleeding episode by the use of pharmacologic agents, such as vasopressin, [25]
vasopressin plus nitroglycerin [6] or other pressure-lowering substances, glypressin (terlipressin), [60]
ornipressin, somatostatin, [5] or octreotide [13] is usually successful. Various means of Endoscopic
therapy, such as endoscopic sclerotherapy, [23] endoscopic variceal ligation, [62] or TIPS shunt, [59] are
considered standard methods of controlling HEV.
The definitive therapy of HEV is portal decompressive surgery. The standard techniques include (1)
end-to-side portacaval anastomoses (PCA), (2) side-to-side PCA, (3) mesocaval interposition H-graft
anastomoses, and (4) distal
136
Figure 1. (Figure Not Available) Randomized, controlled trials published in peer-reviewed journals of nonselective
beta-adrenergic blockers (propranolol and nadolol) versus placebo for the prevention of first variceal hemorrhage. These
results are for bleeding from all sites. Results for bleeding from esophageal varices alone are similar. The odds ratio for the
group is 0.52; 95% confidence interval, 0.38-0.71 ( P < 0.01). ( From Grace ND: Management of portal hypertension.
Gastroenterologist 1993:1:39; with permission.)
splenorenal anastomoses. After successful PCA, further episodes of HEV are rare. Recent advances
include small-diameter portacaval H-grafts in which a prosthetic graft is interposed between the portal
vein and the inferior vena cava. This technique permits the maintenance of sufficient portal hypertension
so that the liver receives adequate portal blood flow. Comparisons of this technique with TIPS
demonstrate similar hemodynamics.
Some investigators, Orloff and colleagues among them, strongly advocate emergency portacaval
anastomoses (EPCA) as the treatment of choice [40] that should be performed as soon as possible after
endoscopic proof of the bleeding site. It is probably true that no other form of therapy has shown such
beneficial results. [9]
Although TIPS shunts may ultimately become the preferred treatment, a high rate of stent stenosis may
prevent it from maintaining its early, transient success. [29] The recent studies of Siegerstetter and Rossle
and their associates, however, offer hope that stenosis can be prevented. [57]
Transjugular intrahepatic portosystemic shunts were first used when Rosch and colleagues, while
attempting a relatively noninvasive way of creating a transjugular, hepatovenous-choledochal fistula,
accidentally induced a hepatovenous-portovenous fistula in a dog. [42] Recognizing the potential
therapeutic value of such an anastomosis, they reported the creation of a transjugular portacaval shunt.
For 20 years, Rosch and coworkers continued investigations that culminated in the development of a
transjugular portacaval shunt, which was first used in a human by Richter et al in 1989. [41]
In principle, the procedure requires that an expandable metal stent on a coaxial catheter be inserted from

the jugular vein into a hepatic vein and then
137
through a parenchymal tract in the liver from the hepatic to the portal vein. This procedure required the
serial passage of coaxial catheters of increasing size until a stent that extended from the hepatic to the
portal vein was implanted.
The development of TIPS shunts has progressed rapidly. Rossi and his associates in Rome studied both
balloon-expandable and, later, self-expanding stents, focusing on the length, diameter, flexibility, and
radiopacity of the stents to select the most appropriate materials for the individual patient. [45] They
studied nitinol and tantalum stents, the latter of which are highly radiopaque. Polylactic and polyglycolic
acid and other absorbable materials were studied to determine the feasibility of using temporary,
absorbable stents. Such stents are intrinsically less strong and more bulky. Efforts were made to coat the
stents with heparin, which can be covalently bonded perpendicular to the polymeric surface.
Theoretically, the use of molecules that are recognized by endothelial cells but not by activated
circulating cells creates an artificial vascular surface that is rapidly endothelialized. Medical-grade steel
of the 300 series (eight- or nine-element alloys) have the most desirable properties for intravascular
stenting (Table 1) . Electropolishing removes most of these accessory elements from the surface, leaving
behind a high concentration of chromium, which forms a thin patina of chromium oxide that prevents
further oxidation. Depending on the composition of the alloy, oxides of tantalum, titanium, or other
metals form the ultimate interface with the host. Figure 2 (Figure Not Available) shows a time sequence
for the changes that occur after the exposure of a prosthetic, metallic material to the luminal surface of
blood vessels.
Self-expanding stents have many advantages. Their major disadvantage is their shortening on expansion,
which can amount to 35% of the length of the unexpanded stent. Most of the shortening occurs near the
proximal end of the stent.
Such stents can be implanted in almost all patients and case decrease the portal venous pressure gradient
in half, from 19 or 20 mm Hg to 9 or 10 mm Hg.
Siegerstetter et al compared either Palmaz or Wallstents in 24 cirrhotic patients, with or without
periprocedural heparin therapy, (24 U/kg, followed by intravenous heparin for 24 hours, followed by
subcutaneous heparin, 0.3 mL/d, for 1 week). [58] Stent thrombogenicity was determined scintigraphically
using Tc99m-labeled platelets. Wallstents showed a significantly higher risk of shunt malfunction than
Palmaz stents. Heparin was needed with the wallstints but not with the Palmaz stents. In experiments in
dogs, 17% of Wallstents in the femoral arteries occluded, but none of the Palmaz stents occluded. The
difference
TABLE 1 -- METALLIC COMPOSITION OF INTRAVASCULAR DEVICES
Device Composition
Z stent 304 stainless steel
Palmaz stent 316-L stainless steel
Medinvent stent Mediloy (unknown composition)

Strecker stent Tantalum

Nitinol stent Nitinol
Greenfield filter 316-L stainless steel, beta3 titanium
Bird's nest filter 304 stainless steel
Gunter filter 304 stainless steel
Gianturco coils 304 stainless steel
Data from Conn HO, Palmaz J, Rosch J, et al (eds): Transjugular intrahepatic portosystemic
stent-shunts. New York, Igaku Shoin Medical Publishers, 1996, pp 68, 151, 178; with permission.
138
Figure 2. (Figure Not Available) Time sequence of the events that take place after exposure of prosthetic metallic surfaces
in close relationship with the luminal surface of vessels. Warry lines=fibrinogen; horizontal lines=intercellular matrix.
(From Conn HO, Palmaz J, Rosch J, et al: Transjugular intrahepatic portosystemic stent-shunts. New York, Igaky Shoin
Medical Publishers, 1996; with permission.)
was attributed to the superiority of the slotted steel tube stents (Palmaz) over the woven type stents
(Wallstents).
Similarly, the expanded polytetrafluoroethylene (ePTFE)-covered stents were shown by DiSalle et al to
be superior to the uncovered Wallstents. [15]
In another recent, small series by Lotterer et al, 21 patients had TIPS shunts implanted for HEV. [35] The
investigators accomplished their primary goal by reducing portal venous pressure from 21 to 11 mm Hg,
increasing cardiac output from 7.1 to 8.9 L/min reducting systemic vascular resistance from 990 to 660
dynes/s/cm5 , and decreasing azygos blood flow from 474 to 375 mL/min. These hemodynamic effects
were short-lived and returned to pre-implantation levels within 3 months.
The TIPS shunt procedure can be accomplished in almost all patients with an acceptable rate of
complications. Several groups have reported on their first 100 TIPS shunt procedures. [33] [47] [49] The
primary complications (hepatic encephalopathy and late-onset stenosis) are considered in detail later.
Rosemurgy et al performed a randomized, controlled trial in which all 70 patients with persistent HEV
were randomly assigned in pairs to receive either TIPS shunt or small-diameter H-graft portacaval shunts
(SDHGPCS). [43] The TIPS shunting procedure was performed through a right jugular approach under
general anesthesia using Wallstents that were 10 mm in diameter and 68 mm long. The portacaval shunts
were ePTFE, 8 mm in diameter. The two groups were matched for age, sex, hepatic manifestations and
Child-Pugh class. In the two groups, the percentages of elective (70%), urgent (20%), or emergent (10%)
procedures were similar. Preshunt portal pressure levels and gradients were also similar (30-32 and 17-18
mm Hg, respectively). Postshunt portal pressure gradients were lower after H grafts (6 mm Hg) than after
TIPS shunting (10 mm Hg). The 30-day mortality was the same in the two groups (35%), and the most
frequent cause of death in both groups was HEV. Inability to complete the shunts was higher with TIPS
shunt (6%) than with PCS shunt implantations

139
(0%). Recurrent HEV and shunt failures were more frequent after TIPS shunting than after PCS
shunting. The authors concluded that the TIPS procedure was associated with more deaths, more
rebleeding, and more treatment failures than the PCS, procedure but the differences were not statistically
significant. The authors favor PCS shunts over TIPS shunts.
Kuhn-Fulton et al compared the efficacy of 10-mm and 12-mm Wallstents in two groups of 23 patients
each with portal hypertension of various causes. [30] The study was performed in sequential,
nonrandomized fashion, first with 23 consecutive 10-mm stents, and then with 23 consecutive 12-mm
stents. A target portosystemic gradient (PSG) of 10 mm Hg was achieved by balloon dilation of the
shunts. Maximal dilation was defined as 12 mm for 10-mm stents and 14 mm for 12-mm stents. The
TIPS shunt placement was successful in 95% of patients. The groups were similar in age, gender,
Child-Pugh class, and clinical indications. For the two shunts, the mean diameter was 9.2 and 11 mm,
the, PSG was 10.6 and 9.6 mm Hg, and flow velocities were 1.7 and 1.3 m/s respectively. Patency rates
were measured at 1, 3, 6, and 12 months and progressively fell from 96% to 42% with the 10-mm stents
and from 78% to 56% with the 12-mm stents. The investigators concluded that 12-mm stents yielded
significantly poorer short- and long-term results, probably caused by the decreased radial strength of the
longer stents. Whether these differences are caused by stent shortening, overdilation during calibration,
or the learning effects inherent in sequential studies is not known. Clearly, in the future such studies
should be randomized, or alternated, not consecutive.
Sanyal and associates compared the cell phenotypes in patients with TIPS with stenosed stents and those
without stenosis. [51] The TIPS shunt tissue was composed of collagen and palisades of mesenchymal
cells and was lined by an endothelium. Immunostaining showed that alpha-smooth muscle staining was
strong and uniform in mesenchymal TIPS and peri-TIPS cells. The TIPS cells expressed high levels of
type 1 procollagen in RNA but contained less fibronectin and relatively more type III collagen than the
peri-TIPS cells. The investigators found no differences in clinical features, time since stenting, histology,
gross morphology, or the presence of bile fistulae on cell phenotype. They did find that smooth muscle
cells (SMC) from stenotic stents exhibited greater cell proliferation and collagen I and III secretion than
did nonstenosed stents. The significance of these observations is not known.
Rossle of the University of Freiburg has solved the most difficult problems involved in the creation of a
TIPS shunt locating the best site of entry to the hepatoportal venous vessels. [46] In the past 8 years, his
group has accumulated a large and successful series of TIPS shunt implantations, representing more than
1000 cases.
Table 2 (Table Not Available) lists Rossle's 10 rules for a successful TIPS shunt implantation. Some of
these rules are obvious, but others are more subtle. For example, rule 2, collaborate with experts,
emphasizes that, when dealing with problems that are complex or poorly understood, such as the factors
that lead to endothelialization, thrombosis, or stent stenosis, it is wise to seek the help of experts in
coagulation, in endothelial proliferation, or in more narrow disciplines.
Shunt stenosis is a case in point. Stenosis of TIPS shunts is just beginning to emerge as a specific
discipline, because some clues to its pathogenesis now suggest that it is a simple manifestation of a much
more complex phenomenon.
Saxon, et al at the Dotter Interventional Institute in Portland, Oregon, recognized that almost half of their

TIPS shunt patients and more than three fourths of their experimental pigs with severe stenosis of TIPS
shunts exhibited evidence of a substantial biliary fistula. [55] In subsequent studies they speculated
140
TABLE 2 -- GUIDELINES TO THE TIPS PROCEDURE

(Not Available)
From Conn HO, Palmaz J, Rosch J, et al (eds): Transjugular intrahepatic portosystemic stent-shunts.
New York, Igaku Shoin Medical Publishers, 1996, p 178; with permission.
the mucus in the biliary secretion was a likely inducer of pseudointimal hyperplasia that might give rise
to myoblast and smooth muscle proliferation and, ultimately, to collagen deposition and shunt stenosis.
Indeed, they have demonstrated that the injection of radiopaque contrast medium into stenosed shunts
may show extensive filling of the biliary tree (Fig. 3) , implying that the bile leakage may be the stimulus
for TIPS shunt stenosis. Thus, bile fistulae may be the origin of shunt stenosis. These observations have
been supported by the histopathologic studies of LaBerg et al. [32]
It has been inferred that free bile in the tissue is an irritating, inflammatory substance that may give rise
to thrombosis or stenosis. Further studies may provide important information about this phenomenon.
The way in which biliary secretion relates to the underlying endothelialization process may well shed
light on the problem of stenosis. It must be remembered that the endothelialization of a steel or other
framework is merely the initial, critical phase of the stenting procedure. With incredible rapidity, the
skeletal stent is covered by platelets to which fibroblasts and myoblasts are attracted (see Fig. 2) (Figure
Not Available) , and within hours or days a fibrous coating completely lines the stent, converting it to an
endothelium-lined tube which cannot be differentiated from the endothelium of a normal blood vessel by
electron microscopy and which functions as a blood vessel. The mechanism that signals the newly
formed endothelium to stop proliferating may not function properly so that the thickness of this
neointima increases until the lumen of the stent is decreased, permitting the process of thrombosis to take
over.
The following sections discuss the clinical indications (Table 3) (Table Not Available) and
contraindications (Table 4) (Table Not Available) for TIPS shunt implantation.
INDICATIONS FOR TRANSJUGULAR INTRAHEPATIC

The primary indication for implanting a TIPS shunt is to prevent HEV. [3] Initially TIPS shunting was
considered to be a novel, noninvasive method of creating a portosystemic anastomosis. Indeed, a TIPS
shunt can promptly reduce the portal venous pressure so esophageal varices stop active bleeding almost
instantaneously and can prevent recurrences of bleeding. As would be expected, a TIPS shunt also

appears to prevent bleeding from gastric and other gastrointestinal lesions, including intestinal, anorectal,
and stomal varices.
141
Figure 3. Extensive filling of the biliary tree 10 months after transjugular intrahepatic portosystemic shunts
(TIPS) implantation after injection of contrast medium into the occluded lumen of a TIPS shunt.
The second major indication for TIPS shunting is the treatment of cirrhotic ascites. Portal hypertension is
the driving force in the pathogenesis of ascites, and the restoration of normal portal pressure suppresses
the formation of ascitic fluid and results in the reduction of ascites. Because cirrhotic hydrothorax is
often an extension of ascites, TIPS shunting is a reasonable and effective method of treating hydrothorax.
Similarly, because the hepatorenal syndrome (HRS) can also indicate ascites formation [38] it, too, can be
effectively treated by TIPS shunting. [28] More recently, it has been demonstrated that the
hepatopulmonary syndrome (HPS), which also has its pathogenetic roots in portal hypertension, may also
be effectively treated by TIPS shunts.
The management of ascites is one of the most important and frequent reasons for performing TIPS
shunting. The treatment of ascites is more than simply preventing the formation of peritoneal fluid.
Trotter et al studied serially the effects of TIPS shunts in 35 patients with refractory ascites. [65] They
retrospectively analyzed the effects of TIPS shunts on body weight, on the volume of ascites, and on the
Child-Pugh score before, 2 months after, and 9 months after implantation. The mean portal venous
pressure gradient fell from 21.5 to 8.1 mm Hg. Ascites was completely resolved in 23 of 24 patients
(96%) within the first 2 months, during which the patients showed a mean weight loss of 2.1 kg,
142
TABLE 3 -- INDICATIONS FOR TIPS

(Not Available)
Adapted from Conn HO, Palmaz J, Rosch J, et al (eds): Transjugular intrahepatic portosystemic
stent-shunts. New York, Igaku Shoin Medical Publishers, 1996, p 151; with permission.
which paralleled the decrease in ascites. Between 2 months and 9 months, however, there was a mean
weight gain of 5.5 kg without the reappearance of ascites. It was thought that the weight gain reflected an
increase in lean body mass. During this 7-month period, the mean albumin concentration increased from
2.7 to 2.9 g/dL, and the mean Child-Pugh score improved from 9.7 to 8.2 ( P < 0.05). This pattern is
demonstrated in a photograph by Ochs et al (Fig. 4) . [38] Similar changes have been reported after

implantation of side-to-side portacaval anastomoses and peritoneovenous shunts and have been attributed
largely to improved appetite and nutrition after the delivery of the ascites.
The TIPS shunting procedure also been used effectively to treat hepatic vein obstruction, (Budd-Chiari
syndrome) and portal vein thrombosis, which may be considered fraternal-twin syndromes of the portal
circulation.
Transjugular intrahepatic portosystemic shunts have also been used as therapy for thoracic duct
cutaneous fistulae and for bile duct fistulae to the hepatic or portal veins. The procedure is also
considered to be a bridge to liver transplantation
TABLE 4 -- CONTRAINDICATIONS TO TIPS
(Not Available)
Adapted from Conn HO, Palmaz J, Rosch J, et al (eds): Transjugular intrahepatic portosystemic
stent-shunts. New York, Igaku Shoin Medical Publishers, 1996, p 151; with permission.
143
Figure 4. Ascitic patient before ( A) and after transjugular intrahepatic portosystemic

shunts (TIPS) ( B). Note the disappearance of ascites and the improvement in the
patient's nutritional state.
for patients who are deteriorating while awaiting liver transplantation because of variceal bleeding or
ascites. [37]
A secondary indication for TIPS shunt implantation is the relief of protein-losing enteropathy. [63] Studies
using Tc 99m-labeled human serum albumin permit the diagnosis of the disorder and show the extent of
the bowel involved. The condition can also be assessed by measuring the amounts of albumin, alpha1
-antitrypsin and other substances in the stool following orally administered large amounts of water.
Transjugular intrahepatic portosystemic shunts have been used to treat gastric, duodenal, intestinal, and
stomal varices and the angiodysplastic lesions of colopathy. [2] It is an effective alternative therapy for
patients who are not responsive to adrenergic beta-blocker therapy, endoscopic sclerotherapy, or ligation
of varices. The TIPS shunt is also indicated for the treatment of hypersplenism. [1]

CONTRAINDICATIONS FOR TRANSJUGULAR INTRAHEPATIC

There are a number of contraindications to TIPS shunt implantation (Table 4) (Table Not Available) .
Among the most important contraindications is heart failure, because the congested heart cannot handle
the additional vascular load imposed by a TIPS shunt. Similarly, the inability of the liver to carry out its
normal functions in fulminant hepatic failure or in decompensated hepatic encephalopathy (HE) indicates
that the failing liver may not be able to deal with the additional hepatic metabolic load imposed by the
TIPS shunt. Celiac or mesenteric arterial stenosis
144
may likewise interfere with the additional metabolic loads of TIPS shunts. If active bacterial or fungal
infections are present, TIPS shunting may overload the body's capacities in this time of stress. Finally,
the presence of polycystic liver disease is an absolute contraindication to implanting a TIPS shunt, to
avoid the creation of an enormous hepatoportoperitoneal anastomosis.
COMPLICATIONS
The complications of TIPS shunts, which seem almost infinite in number and type, can be classified as
the technical misadventures that are related directly to problems encountered in placement of the stent
and as physically and temporally more remote adverse events (Table 5) (Table Not Available) . [56] The
misadventures can include errant punctures of nearby vessels or organs and their consequences, such as
the induction of bleeding fistulae between vascular, biliary, and lymphatic vessels and the pneumothorax.
Stents may be implanted in incorrect positions or may migrate some time after implantation.
Complications also
TABLE 5 -- COMPLICATIONS ASSOCIATED WITH PLACEMENT
(Not Available)
Adapted from Freedman AM, et al: Radiographics 13:1185-1210, 1993; with permission.
145
include cardiac arrhythmias that may be caused by the touch of a catheter and may end up as myocardial
infarctions or pulmonary emboli.
The inadvertent consequences of creating an intrahepatic parenchymal tract in which the shunt is
deployed are true technical complications. Indeed, as indicated previously, there are reasons to believe
that stenosis may be related to the leakage of bile or mucus into this tissue tract.
Allergy to any of the substances used in performing the procedure, ranging from anesthesia to
idiosyncratic reactions to stent materials, may give rise to a variety of syndromes, as may too much
irradiation. Renal tubular injury related to the use or abuse of contrast media is another problem that is

often encountered.
Infections with bacteria or fungi are uncommon complications that may involve the stent, the cardiac
valves, or perhaps more disseminated sites such as the bloodstream (e.g., enterococcal bacteremia). Late
episodes of endocarditis may be a problem, because the risks of these infections are increased in patients
who have portosystemic anastomoses of all types.
Among the most important of the late-onset complications are HE and stent stenosis. Somberg et al
reported their considerable experience with post-TIPS HE. [61] At the University of California, San
Francisco, TIPS shunts were successfully emplaced in 108 patients during a 22-month period. Hepatic
encephalopathy could be accurately assessed in 77 of these patients. New-onset HE or overt worsening of
pre-existing HE within 1 year were the endpoints observed. Eighteen of the 77 patients (23%) developed
new or worsened HE. Fifteen of the 50 patients who had never experienced HE (30%) previously
developed new HE. Three of the 27 patients who had previously experienced HE developed worsened
HE (11%). Treatment of the HE with lactulose resulted in improvement in 78% of the patients, but in
22% the HE progressed. The risk of HE tended to be greater in patients with nonalcoholic cirrhosis
(relative risk 9.2), in women (relative risk 3.0), and in those with hypoalbuminemia (relative risk 2.2 for
each g/dL decrease in the albumen concentration). This large study confirms the 20% to 30% prevalence
of post-TIPS HE observed by so many investigators.
New complications of TIPS shunting continue to emerge. Sanyal et al reported on the occurrence in a
single patient of acute, hemolytic anemia. [50] In a subsequent prospective systemic analysis, they
compared 60 patients with TIPS shunts and 40 control subjects who were receiving sclerotherapy. [52]
These investigators found five patients with acute hemolysis (8%), compared with none in the control
group (0%). This self-limited disorder appears to be largely mixed-type hemolysis which is virtually
asymptomatic. The report was followed by an editorial [48] suggesting that complete endothelialization of
the stent prevents hemolysis, but the presence of some naked steel wires in the blood stream may cause
traumatic injury to erythrocytes and provoke the acute hemolysis.
In 1998, Rossle et al. [48] They reviewed the state of the art after 10 years of experience with TIPS shunts.
They described the crescendo of published articles on TIPS shunts, starting with a handful in 1988 and
1989 and surpassing 100 papers in both 1996 and 1997. They classified the more than 500 publications
by topic, assessed the reported complications, confronted the problem of shunt stenosis (malfunction),
considered the indications (old and new) and contraindications, and predicted the future of this
procedure.
In 1997, the author of this review reported on the current status of TIPS. [10] This update summarizes a
nonrandomized clinical trial that compared TIPS shunts with esophageal transection and
devascularization (ETDV) in 38 patients. Either Palmaz stents or Wallstents were implanted successfully
in 90% of the patients. The overall 30-day mortality rate was higher in the ETDV group than
146
in the TIPS group (70% versus 42%). The report shows one of the uncertainties of performing
nonrandomized, clinical investigations on small numbers of patients.
Also cited was a consecutive series of 96 TIPS shunt implantations performed by Coldwell et al at eight,
large, tertiary medical centers, which demonstrated that TIPS shunting can be performed successfully at

multiple large university-affiliated hospitals in the United States with acceptable success (100%). The
mortality rates (14%), and primary patency rates (88%) may not be within acceptable levels, however.
Hepatic encephalopathy was noted in 31% of these patients.
The most worrisome delayed complication of TIPS emplacement is late-onset stenosis. [54] It seems to be
a very nebulous problem with no obvious points at which it can be attacked. Reviewing what is known
about these stenoses, Rossle et al [57] observed that they usually occur in the hepatic vein distal to the
stent. These investigators clarified the anatomical locations of the stenoses and created a workable
classification. They distinguished two types of stenoses: type 1, the thrombogenic type that is located
within the stent itself, and type 2, the proliferative type that is located in the draining hepatic veins.
Type 1 stenosis is caused by thrombus formation [17] [32] and is theoretically treatable by anticoagulation
or by platelet inhibition. [53] Type 2 stenosis is thought to be the result of localized intimal proliferation in
the hepatic veins that may be a general characteristic of the stenting process per se. [34] Indeed,
endothelialization is characteristic of stenting in peripheral vascular [16] and coronary artery stents [31] and
is the essence of the stenting process that converts a steel skeleton into a vascular structure.
Endothelialization is the response to injury that attracts platelets and macrophages and releases
platelet-derived growth factor (PDGF) [44] downstream, which stimulates the activation of smooth muscle
cells. [36] Inhibitors of PDGF may, therefore, prevent type 2 stenosis. Investigators tested this hypothesis
in a randomized, controlled trial by comparing anticoagulation with trapidil, a potent PDGF inhibitor, [39]
along with ticlopidine, which inhibits platelet aggregation. [18] This investigation gave remarkable results.
The anticoagulation prevented thrombosis (i.e., type 1 stenosis), and the inhibitors of PDGF and platelet
aggregation prevented the activation of the proliferation of smooth muscle cells (type 2 stenosis). These
breakthrough findings represent the first effective, rational forms of therapy of TIPS shunt stenosis and
may lead to a means of preventing late-onset, postimplantation stenosis.
In these studies, Siegerstetter et al studied 84 patients with TIPS shunts. Forty-two randomly selected
patients received heparin (12-24 U/kg) at the time of implantation, followed by 1 week of treatment with
intravenous heparin and 4 weeks of treatment with subcutaneous heparin. The other 42 patients received
trapidil plus ticlopidine, which suppressed almost all of the type 2 stenoses and also reduced the number
of type 1 stenoses. Fifty-seven percent of the patients receiving heparin developed stenosis compared
with 33% of patients receiving antiproliferative the therapy ( P < 0.05). This trend, which was
statistically significant after 6 months, continued as long as the treatment was continued but diminished
after the therapy was stopped. Portal hemodynamics and coagulation parameters were similar in the two
groups of patients were similar before therapy. About half of both groups had either Palmaz or
Memotherm stents implanted, with appropriate decreases in the portal venous pressure and flow velocity.
Intention-to-treat analysis showed that type 1 stenoses were significantly reduced by anticoagulation
therapy in both groups, and type 2 stenoses were decreased by suppression of PDGF and platelet
aggregation. The fact that stenoses, as measured by duplex ultrasonography, began to recur after
discontinuation of the medications suggests that long-term treatment--perhaps
147
permanent therapy--is needed. Clearly, anti-stenotic therapy is effective. It is hoped that further
refinements will improve its efficacy and safety and eradicate the single biggest obstacle to TIPS shunt
therapy.

Another approach to the prevention of stenosis has been the implantation of Wallstents covered with
silicone to avoid the stenosogenic effects of native substances in the parenchymal tract. [64] Although
early studies with tetrafluoropolyethylene coating appears advantageous, [15] the silicone-covered stents
were a failure. [64]
References
1. AlvarezOA, Lopera GA, Vinaychandra P, et al: Improvement of thrombocytopenia due to hypersplenism after
transjugular intrahepatic portosystemic shunt placement in cirrhotic patients. J Gastroenterol 91:134-137, 1996
2. Balzer C, Lotterer E, Kleber G, et al: Transjugular intrahepatic portosystemic shunt for bleeding angiodysplasia-like
lesions in portal-hypertensive colopathy. Gastroenterology 115:167-172, 1998 abstract
3. BennerKG: Indications for and contraindications to TIPS. In Conn HO, Palmaz J, Rosche J, et al: (eds): Transjugular
Intrahepatic Portosystemic Stent-Shunts (TIPS) New York, Igaku Shoin Medical Publishers, 1996
4. BeppuK, Inokuchi K, Koyanagi H: Prediction of variceal hemorrhage by esophageal endoscopy. Gastrointest Endosc
27:213-223, 1981 abstract
5. BoschJ, D'Amico G, Luca A, et al: Drug therapy for variceal hemorrhage. In Bosch J, Groszmann RJ (eds): Portal
Hypertension: Pathophysiology and Treatment. Oxford, Blackwell Scientific Publications, 1994, pp 108-123
6. BoschJ, Groszmann RJ, Garcia-Pagan JC, et al: Association of transdermal nitroglycerin to vasopressin infusion in the
treatment of variceal hemorrhage: A placebo-controlled trial. Hepatology 10:962-968, 1989 abstract
7. BurroughsAK, Jenkins WJ, Sherlock S: Controlled trial of propranolol for the prevention of recurrent variceal
hemorrhage in patients with cirrhosis. N Engl J Med 309:1539-1542, 1983 abstract
8. Cales
P, Desmorat H, Vinel JP: Incidence of large oesophageal varices in patients with cirrhosis: Application to
prophylaxis of first bleeding. Gut 31:1298-1302, 1990 abstract
9. Conn HO: Emergency portacaval shunts in variceal hemorrhage: A critique of critics. HPB Surg 8:154-157, 1994
10. Conn HO: Transjugular intrahepatic portosystemic shunts (TIPS). Current Hepatology 17:251-273, 1997
11. Conn HO, Grace ND, Bosch J: Propranolol in the prevention of the first hemorrhage from esophageal varices: Results
of a randomized double blind, cooperative clinical trial. Hepatology 13:902-912, 1991 abstract
12. ConnHO, Palmaz J, Rosch J, et al (eds): Transjugular Intrahepatic Portosystemic Stent-Shunts (TIPS). New York,
13. D'Amico G, Pagliaro L, Bosch J: The treatment of portal hypertension: A meta-analytic review. Hepatology
22:332-354, 1995 citation
14. deFranchis
R, Pascal JP, Ancona E: Definitions, methodology and therapeutic strategies in portal hypertension: A
consensus development workshop. J Hepatol 15:256-261, 1992 citation

15. DiSalle
RS, Dolmatch BL: Treatment of TIPS stenosis with ePTFE graft-covered stents. Cardiovasc Intervent Radiol
21:172-175, 1998 abstract
16. Do-dai-Do,Triller J, Walpoth BH, et al: A comparison study of self-expandable stents vs balloon angioplasty alone in
femoropopliteal artery occlusions. Cardiovasc Intervent Radiol 15:306-312, 1992 abstract
17. Ducoin
H, El-Khoury J, Rousseau H, et al: Histopathologic analysis of transjugular intrahepatic portosystemic shunts.
Hepatology 25:1064-1069, 1997 abstract
18. Flores-Runk P, Raasch RH: Ticlopidine and antiplatelet therapy. Ann Pharmacother 27:1090-1098, 1993 abstract
148
19. GarceauAJ, Chalmers TC, the Boston Inter-Hospital Liver Group: The natural history of cirrhosis: I. Survival with
esophageal varices. N Engl J Med 268:469-473, 1963
20. Genecin P, Groszmann RJ: Portal hypertension. In Schiff ER (eds). Diseases of the Liver, ed 7. Philadelphia,
Lippincott, 1993, pp 935-973
21. Genecin
P, Polio J, Groszmann RJ: Sodium restriction blunts expansion of plasma volume and ameliorates
hyperdynamic circulation in portal hypertension. Am J Physiol 259:G498-G505, 1990 abstract
22. Gimson AE, Westaby D, Hegarty J: A randomized trial of vasopressin and vasopressin plus nitroglycerin in the control
of acute variceal hemorrhage. Hepatology 6:410-413, 1986 abstract
23. GimsonAE, Westaby D, Williams R: Endoscopic sclerotherapy in the management of gastric variceal hemorrhage. J
Hepatol 13:274-278, 1991 abstract
24. Grace
ND: Diagnosis and treatment of gastrointestinal bleeding secondary to portal hypertension. Am J Gastroenterol
92:1081-1091, 1997 full text
25. GraceND: Variceal hemorrhage; pharmacologic approach. In McDermott WV, Both A (eds): Surgery of the Liver.
Boston, Blackwell Scientific, 1988, pp 303-314
26. Graham D, Smith JL: The course of patients after variceal hemorrhage. Gastroenterology 80:800-809, 1981 abstract
27. Gregory PB, The VA Cooperative Variceal Sclerotherapy Group: Sclerotherapy for male alcoholics with cirrhosis who
have bled from esophageal varices: Results of a randomized multicenter trial. Hepatology 20:618-625, 1994 abstract
28. Guevara M, Gines P, Bandi JC, et al: Transjugular intrahepatic portosystemic shunt in hepatorenal syndrome: Effects on
renal function and vasoactive systems [comments]. Hepatology 28:416-422, 1998 abstract
29. Haskal
ZJ, Pentecost MJ, Soulen MC: Transjugular intrahepatic portosystemic shunt stenosis and revision: Early and
midterm results. AJR Am J Roentgenol 163:439-444, 1994 abstract

30. Kuhn-Fulton
J, Trerotola SO, Harris VJ, et al: Transjugular intrahepatic portosystemic shunt procedure: Efficacy of
10-mm versus 12-mm Wallstents. Radiology 199:658-664, 1996 abstract
31. Kuntz RE, Gibson CM, Nobuyoshi M, et al: Generalized model of restenosis after conventional balloon angioplasty,
stenting and directional atherectomy. J Am Coll Cardiol 21:15-25, 1993 abstract
32. LaBergeJM, Ferrell LD, Ring EJ, et al: Histopathologic study of stenotic and occluded transjugular intrahepatic
portosystemic shunts. J Vasc Interv Radiol 4:779-786, 1993 abstract
33. LaBergeJM, Ring EJ, Gordon RL: Creation of transjugular intrahepatic portosystemic shunts with the Wallstent
endoprosthesis: results in 100 patients. Radiology 187:413-420, 1993 abstract
34. LindCD, Malisch TW, Chong WK, et al: Incidence of shunt occlusion or stenosis following transjugular intrahepatic
portosystemic shunt placement. Gastroenterology 106:1277-1283, 1994 abstract
35. Lotterer
E, Wengert A, Fleig WE: Transjugular intrahepatic portosystemic shunt: Short-term and long-term effects on
hepatic and systemic hemodynamics in patients with cirrhosis. Hepatology 29:632-639, 1999 abstract
36. Marmur JD, Poon M, Rossikhina M, et al: Induction of PDGF-responsive genes in vascular smooth muscle.
Implications for the early response to vessel injury. Circulation 86:53-60, 1992 abstract
37. MartinL, Dondelinger RF, Trotteur G: Treatment of Budd-Chiari syndrome by metallic stent as a bridge to liver
transplantation. Cardiovasc Intervent Radiol 18:196-199, 1995 abstract
38. OchsA: Effects of TIPS on ascites and related complications. In Conn HO, Palmaz J, Rosche J, et al (eds): Transjugular
Intrahepatic Portosystemic Stent-Shunts (TIPS). New York, Igaku Shoin Medical Publishers, 1996, pp 229-267
39. Ohnishi H, Yamaguchi K, Shimada S, et al: A new approach to the treatment of atherosclerosis and trapidil as an
antagonist to platelet-derived growth factor. Life Sci 28:1641-1646, 1981 citation
40. Orloff
MJ, Orloff MS, Rambotti M: Treatment of bleeding esophagogastric varices due to extrahepatic portal
hypertension: Results of portal-systemic shunts during 35 years. J Pediatr Surg 29:142-154, 1994 abstract
149
41. Richter
GM, Palmaz JC, Noldge G: The transjugular intrahepatic portosystemic stent-shunt (TIPSS): A new
nonoperative, transjugular percutaneous procedure. Radiologe 29:406-411, 1989 abstract
42. RoschJ, Hanafee WN, Snow H: Transjugular portal venography and radiologic portacaval shunt: An experimental
study. Radiology 92:1112-1114, 1969 citation
43. Rosemurgy AB: A prospective trial of transjugular intrahepatic portasystemic stent shunts versus small-diameter
prosthetic H-graft portacaval shunts in the treatment of bleeding varices. Ann Surg 224:378-387, 1991

44. Ross R: Platelet-derived growth factor. Annu Rev Med 38:71-79, 1987 abstract
45. Rossi
P, Maccioni F, Bazzi M: TIPS: Technical aspects and clinical results. HIT Angiography and Interventional
Radiology Topics 9:1-17, 1994
46. RossleM: The implantation and maintenance of TIPS: Rules of the road and tricks of the trade. In Conn HO, Palmaz J,
Rosche J, et al (eds): Transjugular Intrahepatic Portosystemic Stent-Shunts (TIPS). New York, Igaku Shoin Medical
Publishers, 1996
47. Rossle
M, Haag K, Ochs A: The transjugular intrahepatic portosystemic stent-shunt procedure for variceal bleeding. N
Engl J Med 330:165-171, 1994 abstract
48. Rossle M, Siegerstetter V, Huber M, et al: The first decade of the transjugular intrahepatic portosystemic shunt (TIPS):
State of the art. Liver 18:73-89, 1998 abstract
49. Sahagun G, Benner KG, Saxon R: Clinical outcome of 100 patients after transjugular intrahepatic portosystemic shunt
(TIPS) for variceal hemorrhage. Am J Gastroenterol 92:1412-1416, 1997 full text
50. SanyalAJ: Hemolytic anemia following TIPS. In Conn HO, Palmaz J, Rosche J, et al (eds): Transjugular Intrahepatic
Portosystemic Stent-Shunts (TIPS). New York, Igaku Shoin Medical Publishers, 1996, pp 301-307
51. SanyalAJ, Contos MJ, Yager D, et al: Development of pseudointima and stenosis after transjugular intrahepatic
portasystemic shunts: Characterization of cell phenotype and function. Hepatology 28:22-32, 1998
52. SanyalAJ, Freedman AM, Purdun PP: The hematological consequences of transjugular intra-hepatic portosystemic
shunts. Hepatology 23:32-39, 1996 abstract
53. SauerP, Theilmann L, Herrmann S, et al: Phenprocoumon for prevention of shunt occlusion after transjugular
intrahepatic portosystemic stent shunt: A randomized trial. Hepatology 24:1433-1436, 1996 abstract
54. Saxon RR, Barton RE, Keller FS: Stenosis and occlusion after TIPS: Prevention and correction. In Conn HO, Palmaz J,
Rosche J, et al (eds): Transjugular Intrahepatic Portosystemic Stent-Shunts (TIPS). New York, Igaku Shoin Medical
Publishers, 1996
55. Saxon RR, Mendel-Hartvig J, Corless CL, et al: Bile duct injury as a major cause of stenosis and occlusion in
transjugular intrahepatic portosystemic shunts: Comparative histopathologic analysis in humans and swine. J Vasc Interv
Radiol 7:487-497, 1996 abstract
56. Schiffman ML, Cole PE: Complications of TIPS implantation. In Conn HO, Palmaz J, Rosche J, et al (eds):
Transjugular Intrahepatic Portosystemic Stent-Shunts (TIPS). New York, Igaku Shoin Medical Publishers, 1996
57. SiegerstetterV, Huber M, Ochs A, et al: Platelet aggregation and platelet-derived growth factor inhibition for
prevention of insufficiency of the transjugular intrahepatic portosystemic shunt: A randomized study comparing trapidil
plus ticlopidine with heparin treatment. Hepatology 29:33-38, 1999 abstract
58. Siegerstetter V, Krauss T, Rossle M, et al: Transjugular intrahepatic portosystemic shunt (TIPS). Thrombogenicity in
stents and its effect on shunt patency. Acta Radiol 38:558-564, 1997 abstract

59. Simpson KJ, Chalmers N, Redhead DN: Transjugular intrahepatic-portasystemic stent shunting for control of acute and
recurrent upper gastrointestinal haemorrhage related to portal hypertension. Gut 34:968-973, 1993 abstract
60. SoderlundC, Magnusson I, Torngren S, et al: Terlipressin (triglycyl-lysine vasopressin) controls acute bleeding
oesophageal varices. A double-blind randomized, placebo-controlled trial. Scand J Gastroenterol 25:622-630, 1990
abstract
61. Somberg KA, Reigler JL, LaBerge JM, et al: Hepatic encephalopathy after transjugular intrahepatic portosystemic
shunts: Incidence and risk factors. Am J Gastroenterol 90:549-555, 1995 full text
150
62. Stiegmann GV, Goff JS, Sun JH, et al: Endoscopic ligation of esophageal varices. Am J Surg 159:21-26, 1990
abstract
63. TakedaH, Takahashi T, Ajitsu S, et al: Protein-losing gastroenteropathy detected by technetium-99m m-labeled human
serum albumin. Am J Gastroenterol 86:450-453, 1991 abstract
64. Tanihata H, Saxon RR, Kubota Y, et al: Transjugular intrahepatic portosystemic shunt with silicone-covered Wallstents:
Results in a swine model. Radiology 205:181-184, 1997 abstract
65. TrotterJF, Suhocki PV, Rockey DC: Transjugular intrahepatic portosystemic shunt (TIPS) in patients with refractory
ascites: Effect on body weight and Child-Pugh score. Am J Gastroenterol 93:1891-1894, 1998 abstract


April 21, 2000
Conde Petra
THE ROLE OF SURGERY IN THE TREATMENT

OF PORTAL HYPERTENSION
Additional Article
Full Text
Frontmatter 99
SURGICAL PROCEDURES
PORTAL HYPERTENSION
Total Portal Systemic Shunts
Partial Shunts
Selective Shunts THE ROLE OF SURGERY IN THE

TREATMENT OF PORTAL
Devascularization Procedures
HYPERTENSION
MANAGEMENT STRATEGIES
Prophylaxis Prior to the First

Variceal Bleed
Acute Bleeding Esophageal

David A. Iannitti MD
Varices
J. Michael Henderson MB,ChB,FRCS,FACS
Prevention of Recurrent Variceal
Bleeding
From the Cleveland Clinic Foundation, Cleveland, Ohio
Refractory Ascites
References J. Michael Henderson, MD
Department of General Surgery A8-418
About the Publication The Cleveland Clinic Foundation
9500 Euclid Avenue
Cleveland, OH 44095
The role of surgery in the treatment of portal hypertension

remains a complex and highly debated issue. Liver
transplantation has come of age in the 1990s and is an accepted
therapy for many patients with end-stage liver disease. The
focus of this article is on the role of other surgical options in
managing the complications of portal hypertension, particularly

variceal bleeding and refractory ascites. Several factors must be

considered when surgical options are to be entertained,
including origin and extent of liver disease, response to prior
medical treatment, and possibility of future liver
transplantation. Also, availability of and response to other
technologies, such as endoscopic treatment and radiologic
intervention (TIPS), must be factored into the treatment plan.
This article reviews the history of interventional treatments of
portal hypertension, describes the various shunting procedures,
and evaluates their role in the various clinical scenarios that
may require surgery as a treatment option.
For the past 100 years, surgery has played an integral role in the
understanding and treatment of portal hypertension.
Understanding portal anatomy and alterations in portal venous
blood flow in cirrhosis is the basis for surgical and other
interventional procedures (Fig. 1) . The initial work in this area
began with Nikolai Eck in 1877. He fashioned side-to-side
portacaval anastomoses with ligation of the distal portal vein in
eight dogs (Fig. 2) . In 1903 Eugene Vidal performed the first
end-to-side portacaval anastomosis successfully in man. Vidal
observed
100
Figure 1. Normal portal anatomy demonstrating

hepatopedal blood flow in the coronary, splenic,
superior mesenteric, inferior mesenteric, and portal
veins.
Figure 2. The end-to-side portacaval shunt or Eck

fistula. Note complete diversion of all portal blood
flow into the inferior vena cava. The distal portal
vein is ligated which maintains a high hepatic
sinusoid pressure.
101
cessation of bleeding and resolution of ascites; however, the

patient subsequently developed encephalopathy, reaccumulation

of ascites, and expired 3.5 months after surgery. Several

historical articles review these early experiences. [5] [6] [12]
Several procedures were developed in the early 20th century to
promote portalsystemic collateralization, including the
omentopexy of Drummond and Morrison, [74] splenic
transposition, first described by Talma, [28] and the visceral
abrasion techniques of Madden et al. [39] Enthusiasm for these
procedures waned as their results yielded no improvement in
patient outcome. In 1945, Allen Whipple reintroduced portal
systemic shunting for the management of complications of
portal hypertension. [73] He reported a series of portacaval and
nonselective splenorenal shunts. Linton [38] continued this work,
and in 1961 reviewed his experience in 169 patients. He
concluded that each type of shunt had equal efficacy in
controlling bleeding, end-to-side portacaval shunts did not
control ascites, and splenorenal shunting resulted in less
encephalopathy. The incidence of encephalopathy continued to
remain high, and progression of hepatic failure accelerated
following total, nonselective shunting procedures. These
findings led to the development of selective shunting, including
the distal splenorenal shunt by Warren [71] (Fig. 3) and the
coronary-caval shunt by Inokuchi [30] (Fig. 4) . These shunting
procedures
Figure 3. The distal splenorenal shunt or Warren

shunt. The distal splenic vein is mobilized and
anastomosed to the left renal vein. This selectively
provides a low pressure outflow tract for
gastroesophageal varices. The coronary vein and
other collaterals are ligated. Note this is not a
portal-systemic shunt.
102
Figure 4. The coronary-caval shunt. The left gastric

vein is anastomosed to the inferior vena cava
providing decompression of gastroesophageal
varices. Splenectomy is included with this procedure.
maintain portal hypertension and hepatic perfusion while

selectively decompressing gastroesophageal varices effectively.
The past two decades have seen major changes in managing
patients with cirrhosis and portal hypertension. Further

understanding of the pathophysiology of this disease process

has led to the use of newer and more efficacious medications.
[52] [55] [66] [69] Medical pharmacotherapy presently includes the
use of vasopressin alone or in combination with nitroglycerine;

propranolol; and, more recently, octreotide and isosorbide
mononitrate. [36] [42] [70]
In the 1970s endoscopic sclerotherapy, initially described in

1939, was reintroduced and became widely accepted. This
modality of treatment became the standard for treating acute
variceal bleeding. Sclerotherapy combined with pharmacologic
reduction of portal hypertension using beta-blockers also
became the primary treatment for acute-bleeding esophageal
varices and prevention of recurrent bleeding. [33] [37] Improved
endoscopic techniques using elastic band ligation have proved
to be more efficacious with less procedure-related
complications. [68]
The introduction of the transjugular intrahepatic portal systemic

shunt (TIPS) has rapidly gained worldwide acceptance in
creating a functional, total side-to-side shunt using a minimally
invasive technique. [35] [58] TIPS patency rates have improved
with aggressive monitoring
103
and radiographic interventional procedures. The accepted

indications for TIPS are acute variceal bleeding that cannot be
successfully controlled with medical treatment, including
sclerotherapy, and recurrent variceal bleeding in patients who
are refractory or intolerant to conventional medical
management, including sclerotherapy and pharmacologic
therapy, and are not candidates for surgery. TIPS may be useful
for the treatment of refractory ascites, and its use has been
reported for acute Budd-Chiari syndrome. [61]
The 1990s have also seen a further development in the surgical
treatment of portal hypertension. Surgical shunt procedures now
include partial portal systemic shunts and pancreatic
disconnection for distal splenorenal shunts. [7] [26] The
introduction of liver transplantation in the 1980s provided
definitive treatment of the underlying liver disease, and is the
ultimate treatment for portal hypertension. [20] [64] Liver
transplantation has improved with lower morbidity and
mortality and improved survival with the development of
newer, more effective immunosuppressant medications, such as

tacrolimus (FK506), and improved management schemes for
control of infection.
Presently, it is difficult to decide the optimal treatment for a

given patient with such options available. Careful patient
evaluation with an understanding of the underlying liver
disease, portal anatomy, possible need for transplantation, and
availability of appropriate technology and expertise must be
considered for the treatment plan. [24] [27] [34]
SURGICAL PROCEDURES
Decompression of gastroesophageal varices controls bleeding,
but has not been widely accepted to treat all patients with
variceal bleeding. This is primarily because the first type of
shunts were total portal systemic shunts, and diversion of portal
blood flow was detrimental to liver function; however, other
types of operative shunts have been introduced, and in the
1990s attention has focused back on decompression because of
TIPS. In evaluating the role of decompression, the effect of
different types of shunts on portal perfusion is key. [8] [11] [21] [54]
Total portalsystemic shunts divert the entire blood flow of the

portal system into the systemic circulation. Total shunts can be
achieved in several ways. The classic end-to-side portacaval
anastomosis (the Eck fistula) (see Fig. 2) clearly provides
complete diversion of all portal flow; however, the procedure
does not provide outflow of the liver sinusoids, and as such, it
cannot be used to treat ascites. The side-to-side portacaval
anastomosis of 1.2 cm or greater also results in a total portal
systemic shunt, but because the portal vein remains intact, it
acts as an outflow from the liver sinusoids and thus controls
ascites (Fig. 5) . Any portacaval
104
Figure 5. The side-to-side portacaval shunt. This is a

total shunt providing complete diversion of all portal
blood flow as well as decompression of the hepatic
sinusoids.

shunt requires dissection of the hepatic hilus to free the portal

vein and will make subsequent liver transplantation more
difficult.
Total portal systemic shunting can also be accomplished using
large-caliber (16-22-mm) interposition graft placed from the
portal or superior mesenteric vein to the inferior vena cava, left
renal vein, or right atrium (Fig. 6) . Although some of these
shunts do require dissection of the hepatic hilus, they are more
easily divided should liver transplantation be needed. The major
disadvantage of an interposition graft is the primary patency is
less than that of a side-to-side portacaval vein-to-vein
anastomosis. Recent radiographic techniques, however,
including balloon dilatation and infusion of thrombolytic
agents, have increased the primary, primary assisted, and
secondary patency of these grafts.
Nonselective or total shunts eliminate portal hypertension,
effectively control and prevent variceal bleeding, and with the
exception of the end-to-side portacaval shunt, are useful in
treating ascites. Disadvantages are related to diversion of all
portal flow, which leads to an increased incidence of
encephalopathy and accelerated progression of the underlying
liver disease.
Partial Shunts
Partial shunts create a side-to-side communication of limited

size between the portal and systemic circulation, lower the
portal pressure,
105
Figure 6. The interposition shunt. This is a functional

side-to-side total shunt. An interposition graft
(16-22-mm polytetrafluoroethylene [PTFE]) can be
placed from the portal or superior mesenteric vein to
the inferior vena cava, left renal vein, or right atrium.
and maintain some hepatic portal perfusion. This concept of

partial shunting was begun by Bismuth, [2] who observed an
initial gradient across a 15-mm side-to-side portacaval
anastamosis. At this size, these eventually became total shunts
with reversal of portal flow. Johansen [31] continued this work

by creating smaller, 10- to 12-mm side-to-side portacaval

anastamoses, and although significant pressure gradients could
be measured initially between the portal and systemic
circulations, these shunts became total shunts eventually.
Recently, Sarfeh [59] has popularized the use of small-diameter,
reinforced polytetrafluoroethylene (PTFE) interposition grafts
between the portal vein and inferior vena cava. [59] An 8-mm
graft reduces portal pressure to 12 mmHg, which is low enough
to control bleeding, while maintaining hepatic portal perfusion
in 80% of subjects. When a 10-mm graft is used, hepatic portal
perfusion is only maintained in 20% of subjects. The coronary
vein, gastroepiploic veins, and other collateral vessels must be
ligated to decrease inflow to the esophageal varices.
Partial shunting results in lowered portal pressure while
maintaining some prograde hepatic venous flow and are
effective treatment for bleeding esophageal varices with a lower
rate of encephalopathy than total shunts. The disadvantages
include dissection of the hepatic hilus and an early thrombosis
rate of 16%; however, with modern
106
radiologic interventional techniques, long-term patency can be

maintained.
Selective Shunts
Selective shunts provide selective decompression of

gastroesophageal varices while maintaining portal hypertension
and portal venous perfusion. These are not portal systemic
shunts because portal blood flow is not diverted. Generally
there are two types of selective shunts: the distal splenorenal
shunt-devised and popularized by Warren, [71] and the
infrequently used coronary-caval shunt of Inokuchi. [30]
The coronary-caval shunt includes splenectomy and involves
dissection and disconnection of the coronary (left gastric) vein
and anastomosis to the inferior vena cava (see Fig. 4) . This
procedure is technically difficult, usually requires interposition
vein grafting, and is rarely used.
The distal splenorenal shunt, which selectively decompresses
the spleen and gastroesophageal junction, has gained worldwide
acceptance. The procedure requires disconnection of the splenic
vein from the superior mesenteric vein, dissection of this vein
out from the pancreas, and anastomosis to the left renal vein in
an end-to-side fashion (see Fig. 3) . This provides a

low-pressure outlow tract for gastroesophageal varices. Portal
systemic collateral vessels, including the left gastric vein,
gastroepiploic veins, and vessels within the splenocolic
ligament, are ligated.
The distal splenorenal shunt has several advantages over portal
systemic shunting for the decompression of gastroesophageal
varices. Hepatopedal portal vein flow is maintained, which
decreases the risk of encephalopathy and accelerated liver
failure. Esophageal and gastric varices are effectively
decompressed, and the dissection is within the lesser sac, distant
to hepatic hilus. Long-term patency is excellent in this
high-flow vein-to-vein anastomosis. [23] Although portal vein
thrombosis has been recorded in 5% to 10% of patients, hepatic
portal perfusion is maintained in 90% of patients early after
shunt. In nonalcoholic patients, good hepatic portal perfusion is
maintained long term, whereas in alcoholics it is lost in 16% to
50%, depending on how much splenic vein is dissected out of
the pancreas. Loss of portal flow occurs through the
development of intrapancreatic portal systemic collateralization
known as the pancreatic siphon.[48] Improved maintenance of
portal perfusion in alcoholic patients has been achieved by the
addition of splenopancreatic disconnection to the distal
splenorenal shunt procedure. [26] [72] In this modification, the
splenic vein is completely disconnected from the posterior
pancreas to the splenic hilum. Other modifications, such as
transposition of the splenic vein [47] or interposition grafting to
the inferior vena cava, can be made in patients with a
retro-aortic left renal vein (present in 4% of the population).
Distal splenorenal shunting effectively treats bleeding
gastroesophageal varicies, portal hypertensive gastropathy, and
prevents recurrent bleeding in over 90% of patients. [3] Because
portal hypertension is maintained, it cannot be used to treat
ascites.
107
Devascularization (nonshunting) procedures of the stomach and

lower esophagus were proposed by Gray and Witesell in 1950.
[16] Several investigators performed various devascularization
procedures with the modifications of variceal ligation,

esophageal transection, and splenectomy. In 1973, Sugiura and
Futagawa [67] reported a two-stage procedure involving both

transthoracic and transabdominal approaches for interruption of

gastroesophageal varices while still maintaining flow through
other portal-azygous collaterals. Yammamoto et al [75] modified
this procedure to be performed through the transabdominal
approach only.
The effectiveness of devascularization procedures for the
control of variceal bleeding appears to be related to the extent
of the procedure. For maximal efficacy, a devascularization
procedure usually includes esophageal transection and
reanastamosis (which can be achieved with a circular stapling
device), truncal vagotomy, pyloroplasty, splenectomy, and
ligation of perigastric and periesophageal vessels to at least 7
cm up the esophagus. This provides interruption of inflow to
gastroesphageal varices with maintenance of portal
hypertension and hepatic portal perfusion. Devascularization
procedures are effective in controlling actively bleeding
gastroesophageal varices; there is a low operative mortality and
rebleeding rate. [29] Outside of Japan, most centers report up to a
37% rebleeding rate. [10] [46] Most surgeons can perform
devascularization procedures, so they remain an acceptable
alternative in an emergency and for "unshuntable" patients. [4]
Splenectomy alone in treating bleeding varices has a limited
role. [18] Splenectomy may be curative in patients with bleeding
gastric varices from isolated splenic vein thrombosis; however,
splenectomy alone is not indicated in patients with portal
hypertension secondary to cirrhosis.
This section briefly describes management options at the major
time points of presentation for patients with portal hypertension
and varices.
Prophylaxis Prior to the First Variceal Bleed
Esophageal varices are present at the time of diagnosis of

cirrhosis in 30% of patients with compensated and 60% of
patients with decompensated disease. Untreated varices tend to
progressively increase in size before they eventually rupture and
bleed. There is a 5% per year risk of bleeding in unselected
patients with cirrhosis, this risk increases to 20% to 30% in
patients who already have varices. [11]
Data in the 1990s support the prevention or prophylaxis of a
first variceal bleed with pharmacologic treatment. [52]
Propranolol effectively

108
lowers variceal pressure and hepatic venous pressure gradient.

[14] [15] A recent multicenter trial of patients with cirrhosis and
endoscopically proven esophageal varices compared the use of

propranolol or placebo prior to the first variceal bleed. The
propranolol group had a significantly lower incidence of
bleeding from esophageal varices (4%) versus the placebo
group (22%), and portal hypertensive gastropathy, although
there was no difference in survival between the two groups. [9]
Meta-analysis of multiple studies supports this management
strategy.
Endoscopic sclerotherapy or banding should not be undertaken
as a prophylactic measure in patients with portal hypertension
and varices. Reports are conflicting as to their efficacy, [17] [51]
but their use cannot be supported outside randomized trials.
Presently, surgery is not indicated as a treatment prophylaxis for
variceal bleeding. [17]
Acute Bleeding Esophageal Varices
Approximately 30% of patients with cirrhosis and esophageal

varices experience variceal hemorrhage, usually within 1 year
of diagnosis. This is a major, potentially life-threatening
complication, with mortality rates ranging from 30% to more
than 50% in most studies. [11] Gastric varices are less likely to
bleed; however, when bleeding occurs, it is more extensive. [60]
The status of hepatic function is the major determinant in
survival. [40]
The initial management of a patient with bleeding esophageal
varices consists of adequate volume replacement with blood,
blood products, and colloid solutions. Pharmacologic portal
pressure reduction can be acutely achieved with medications
such as vasopressin and nitroglycerine, or octreotide. [55] Early
endoscopy is employed to identify and confirm the source of
bleeding and to control bleeding with sclerotherapy or banding.
When these measures are carried out, control of variceal
hemorrhage can be achieved in about 90% of patients. [25]
Urgent or emergent shunting may be required in the 10% of
patients whose acute bleeding is not controlled by endoscopic
measures. [50] [53] This should be performed with a TIPS
procedure because failure to control acute bleeding
endoscopically almost always occurs in poor risk patients. In
some good-risk patients, operative shunting procedures may be
used. [32] [57] The largest experience of emergency portacaval

shunts has recently been reported by Orloff et al. [45] He

reviewed 340 unselected cirrhotic patients undergoing
emergency side-to-side portacaval shunts. All patients had
control of bleeding and long-term patency of their shunts.
Remarkably, the recent group of 220 patients had a 15%
incidence of hepatic failure and 9% incidence of
encephalopathy. Selective and nonselective shunts effectively
stop bleeding from gastroesophageal varices, and have
acceptable incidence of side effects. The choice of shunt
ultimately remains with the involved surgeon.
In the case of "unshuntable" patients or unfamiliarity with shunt
procedures, nonshunting devascularization procedures can be
used. [4]
109
Devascularization adequately controls bleeding and can be

performed by most surgeons. This procedure is associated with
a low rate of encephalopathy and hepatic failure; however,
rebleeding rates remain high (about 40%). Devascularization is
a reasonable alternative in an emergency situation.
Prevention of Recurrent Variceal Bleeding
The natural history of patients with cirrhosis who experience a

variceal bleed is that 47% to 84% of these patients will
experience rebleeding within 1 to 2 years if they do not receive
treatment for the portal hypertension or obliteration of the
varices. [44] The mortality of a recurrent variceal hemorrhage
ranges from 20% to 70%. Advanced liver disease, variceal size,
and continued alcohol use are related to increased rebleeding
risk. [11]
Following the initial resuscitation and control of acute bleeding,
patient evaluation is the key to making definitive treatment
choices. Several questions must be answered in this time period.
First, evaluation should address the status of liver function.
Does the patient have end-stage liver disease that can be only
managed by transplantation? If so, is the patient a transplant
candidate? Does the patient have normal hepatic function with
an extra hepatic cause of portal hypertension, such as portal
vein thrombosis? The next series of questions relate to the
varices. This involves endoscopy and a hemodynamic study
with Doppler ultrasound or angiography to evaluate the patency
of the portal venous system. Finally, the question must be
asked: what resources are available for definitive treatment?

These include endoscopists capable of performing sclerotherapy

and elastic band ligation, interventional radiologists skilled in
TIPS, surgeons familiar with various shunting and nonshunting
procedures, and liver transplant capability and availability.
The first line of treatment for prevention of recurrent bleeding
in patients with portal hypertension is by pharmacologic and
endoscopic treatment. [14] [36] When these measures fail to
prevent recurrent variceal bleeding, variceal decompression
should be considered. TIPS, surgical shunts, and liver
transplantation are the decompression options.
Patients with poor hepatic function are potential transplant
candidates. They should be fully evaluated for transplant, and if
candidates, the only issue is that of timing. In transplant
candidates with recurrent bleeding not controlled by endoscopy,
TIPS is a reasonable option. Equally, TIPS, in patients who fail
medical and endoscopic therapy and have poor liver function is
an option in patients who are not transplant candidates. The
current data on TIPS can be summarized as follows:
Technical success of TIPS is achieved in approximately
90% of patients, resulting in a portal systemic pressure
gradient of < 12 mmHg.
Failure of TIPS is related to shunt stenosis and occlusion.
110
Primary patency at 1 year ranges from 50%-65%, and

primary-assisted patency is about 85% at 1 year. Careful
follow-up with stent monitoring results in a 25%-50%
reintervention rate to maintain patency. One-year
rebleeding rate after TIPS is 18%-20%. [19] [35] Long-term
patency and rebleeding rates are now being evaluated.
Patients with good or adequate hepatic function and failing

medical and endoscopic treatment are candidates for surgical
decompression. Distal splenorenal shunts have been proved to
be more efficacious in prevention of rebleeding from
gastroesophageal varices than endoscopic sclerotherapy in a
meta-analysis of four randomized clinical trials. [62] [63] Distal
splenorenal shunts are effective in long-term control of variceal
bleeding in over 90% of patients. Portal perfusion is maintained

in more than 90% of patients with nonalcoholic liver disease

and 50% to 84% of patients with alcoholic liver disease.
Modifying the distal splenorenal shunt with pancreatic
disconnection of the splenic vein in the latter group minimizes
the development of intrapancreatic portal systemic collaterals
[26] and maintain better portal flow. Postshunt encephalopathy
rates remain less than 10%, and selective shunting does not
accelerate the natural history of the underlying liver disease. [23]
Distal splenorenal shunting has been compared prospectively to
partial portalsystemic shunting in a recent study. [41] Both shunts
are equally efficacious in control of rebleeding; however,
encephalopathy and shunt thrombosis rates were higher in
patients who recieved partial, 10-mm interposition mesocaval
shunts.
The choice between distal splenorenal shunts and TIPS in
patients with good hepatic function is a topic of current interest.
Distal splenorenal shunts have proved long-term patency and
efficacy in this patient group. The popularity and use of TIPS
has increased in recent years. Retrospective review of patients
undergoing surgical shunting or TIPS reveals lower rates of
mortality, encephalopathy, and rebleeding in the surgical shunt
groups. TIPS had a 12% early thrombosis, 41% stenosis, about
20% rebleeding, and 29% encephalopathy rates. [1] [8] [11]
Surgical shunts overall have about 5% variceal rebleeding, and
encephalopathy from 10% to 40%, depending on the type of
shunt. Presently there is an ongoing multicenter, prospective,
randomized trial comparing distal splenorenal shunts to TIPS in
patients with good hepatic function.
Devascularization procedures are generally used in unshuntable
patients with thrombosed splenic and portal venous systems.
Devascularization in Child's A patients results in 4% early and
6% rebleeding rates in Japan, [29] and a 3% encephalopathy rate.
[46] This procedure is a good alternative in otherwise
unshuntable patients.
Refractory Ascites
Ascites remain a difficult problem in patients with cirrhosis and

portal hypertension. Medical treatment includes dietary salt
restriction
111
and diuretic therapy, which can usually improve or resolve

ascites in 90% of patients. Ascites that are not responsive to diet

and diuretics is termed refractory. When this occurs, repeat

large-volume paracentesis can provide temporary relief, but is
associated with a high recurrence rate. Several interventional
procedures, including peritoneovenous shunts, surgical portal
systemic shunts, and TIPS, have been used in these patients.
These procedures are all associated with high morbidity,
mortality, and failure rates. Liver transplantation remains the
definitive treatment for patients with cirrhosis, portal
hypertension, and refractory ascites.
Patients with poor hepatic function and refractory ascites are
best treated by liver transplantation. As a bridge to transplant,
treatment may include TIPS, but complications remain high and
include a 30-day mortality of 4% to 8%, and an encephalopathy
rate of 20% to 30%. TIPS improves ascites in over 90% of
patients that survive and can be used as a temporary measure
until transplant. [43] TIPS has also been used for cirrhotic
hydrothorax and hepatorenal syndrome.
Surgical side-to-side portal systemic shunts are also effective in
treating refractory ascites in patients with portal hypertension.
In patients with good hepatic function and a history of variceal
bleeding, a functional side-to-side total shunt may be indicated.
End-to-side portacaval and distal splenorenal shunts do not help
ascites. The major problem with portal systemic shunting is the
high incidence of encephalopathy. Of patients following a total
shunt for refractory ascites, 40% to 50% will develop
encephalopathy within 1 year.
Peritoneovenous shunts are another treatment that has been used
for patients with refractory ascites. These shunts are usually
indicated for patients who have no history of variceal bleeding
and reasonable liver function. Patients with a high degree of
liver dysfunction who undergo peritoneovenous shunting have a
higher rate of postprocedure complications. These include
disseminated intravascular coagulopathy, dilutional
coagulapathy, gastrointestinal bleeding, congestive heart failure,
shunt infection, and shunt malfunction. Although
peritoneovenous shunting has not been shown to improve
survival compared with repeat paracentesis, it can provide
better control of ascites and better quality of life. Improvements
in respiratory function, activity level, food intake, and weight
loss can be seen following peritoneovenous shunting in
carefully selected patients. In the 1990s, however, the indication
for peritoneovenous shunting is rare as most of these patients
are transplant candidates. [8] [13]

References
1. BeckerYT, Reed G, Lind CD: The role of elective operation in the

treatment of portal hypertension. Am Surg 62:171, 1996
2. Bismuth H, Franco D, Hepp J: Portal-systemic shunt in hepatic cirrhosis:

Does the type of shunt decisively influence the clinical result? Ann Surg
179:209, 1974
112
3. Boyer TD, Kokenes DD, Hertzler G, et al: Effect of distal splenorenal

shunt on survival of patients with primary biliary cirrhosis. Hepatology
20:1484, 1994
4. CapsMT, Helton WS, Johansen K: Left-upper-quadrant

devascularization for "unshuntable" portal hypertension. Arch Surg
131:834, 1996
5. ChandlerJG: The history of surgical treatment of portal hypertension.

Arch Surg 128:925, 1993
6. Child CG: Eck's fistula. Surg Gynecol Obstet 96:375, 1953
7. Collins
JC, Sarfeh IJ: Surgical management of portal hypertension. West
J Med 162:527, 1995
8. Collins
JC, Rypins EB, Sarfeh IJ: Narrow-diameter portacaval shunts for
management of variceal bleeding. World J Surg 18:211, 1994
9. Conn HO, Grace ND, Bosch J, et al: Propranolol in the prevention of the
first hemorrhage from esophagogastric varices: A multicenter, randomized
clinical trial. Hepatology 13:902, 1991
10. DagenaisM, Langer B, Taylor BR, et al: Experience with radical

esophagogastric devascularization procedures (Sugiura) for variceal
bleeding outside Japan. World J Surg 18:222, 1994
11. D'AmicoG, Pagliaro L, Bosch J: The treatment of portal hypertension:

A meta-analytic review. Hepatology 22:332, 1995
12. DonovanAJ, Covey PC: Early history of portacaval shunt in humans.

Surg Gynecol Obstet 147:423, 1978
13. Elcheroth J, Vons C, Franco D: Role of surgical therapy in management

of intractable ascites. World J Surg 18:241, 1994
14. Feu F, Bordas JM, Garcia-Pagan JC, et al: Double-blind investigation

of the effects of propranolol and placebo on the pressure of esophageal

varices in patients with portal hypertension. Hepatology 13:917, 1991
15. FeuF, Bordas JM, Lucy A, et al: Reduction of variceal pressure by

propranolol: Comparison of the effects on portal pressure and azygos
blood flow in patients with cirrhosis. Hepatology 18:1082, 1993
16. Gray
HK, Whitsell FB: Hemorrhage from esophageal varices. Ann Surg
132:798, 1950
17. GreigJD, Garden OJ, Carter DC: Prophylactic treatment of patients

with esophageal varices: Is it ever indicated? World J Surg 18:176, 1994
18. GusbergRJ, Peterec SM, Sumpio BE, et al: Splenomegaly and variceal
bleeding: Hemodynamic basis and treatment implications.
Hepato-Gastroenterol 41:572, 1994
19. Helton
WS, Belshaw A, Althaus S, et al: Critical appraisal of the
angiographic portacaval shunt (TIPS). Am J Surg 165:566, 1993
20. HendersonJM: Liver transplant for portal hypertension. Gastroenterol

Clin North Am 21:197, 1992
21. Henderson JM: Portal hypertension and shunt surgery. Adv Surg
26:233, 1993
22. Henderson JM: The distal splenorenal shunt in the management of

bleeding varices. J Royal Coll Surg Edinburg 27:130, 1982
23. Henderson
JM: The role of distal splenorenal shunt for long-term
24. Henderson JM, Warren WD: Portal hypertension. Curr Prob Surg
25:151, 1988
25. Henderson JM, Carey WD, Vogt DP, et al: Management of variceal
bleeding in the 1990's. Clev Clin J Med 60:431, 1993
26. Henderson JM, Warren WD, Millikan WJ, et al: Distal splenorenal
shunt with splenopancreatic disconnection: A 4-year assessment. Ann Surg
210:332, 1989
27. Hermann RE, Henderson JM, Vogt DP, et al: Fifty years of surgery for
portal hypertension at the Cleveland Clinic Foundation. Ann Surg
221:459, 1995
28. Holman E: Implantation of the spleen in the abdominal wall for portal
obstruction: A suggested operation for hepatic cirrhosis. West J Surg
Obstet Gynecol 41:255, 1933
29. Idezuki
Y, Kokudo N, Sanjo K, et al: Sugiura procedure for
management of variceal bleeding in Japan. World J Surg 18:216, 1994
30. Inokuchi K: A selective portacaval shunt. Lancet 2:51, 1968

31. JohansenK: Partial portal decompression for variceal hemorrhage. Am

J Surg 157:479, 1989
113
32. Johansen K, Helton WS: Portal hypertension and bleeding esophageal

varices. Ann Vasc Surg 6:553, 1992
33. Johnston GW, Rodgers HW: A review of 15 years' experience in the

use of sclerotherapy in the control of acute hemorrhage from oesophageal
varices. Br J Surg 60:797, 1973
34. KnechtleSJ, Kalayoglu M, D'Alessandro AM, et al: Portal

hypertension: Surgical management in the 1990's. Surgery 116:687, 1994
35. LaBerge JM, Ring EJ, Gordon RL, et al: Creation of transjugular
intrahepatic portosystemic shunt with wallstent endoprosthesis: Results in
100 patients. Radiology 187:413, 1993
36. Lebrec
D: Long-term management of variceal bleeding: The place of
pharmacotherapy. World J Surg 18:229, 1994
37. Lebrec D, Wovel O, Corbic M, et al: Propranolol: A medical treatment

for portal hypertension. Lancet 2:180, 1980
38. Linton RR, Ellis DS, Geary JE: Critical comparative analysis of early
and late results of splenorenal and direct portacaval shunts performed in
169 patients with portal cirrhosis. Ann Surg 154:446, 1961
39. Madden JL, Lore JM, Gerald FP, et al: The pathogenesis of ascites and
a consideration of its treatment. Surg Gynecol Obstet 99:585, 1954
40. McCormick PA, Burroughs AK: Relation between liver pathology and
prognosis in patients with portal hypertension. World J Surg 18:171, 1994
41. Mercado MA, Morales-Linares JC, Granados-Garcia J, et al: Distal

splenorenal shunt versus 10 mm low-diameter mesocaval shunt for
variceal hemorrhage. Am J Surg 171:591, 1996
42. MorillasRM, Planas R, Cabre E, et al: Propranolol plus

isosorbide-5-mononitrate for portal hypertension in cirrhosis: Long-term
hemodynamic and renal effects. Hepatology 20:1502, 1994
43. OchsA, Rossle M, Haag K, et al: The transjugular intrahepatic

portosystemic stent-shunt procedure for refractory ascites. N Engl J Med
332:1192, 1995
44. Orloff
MJ, Duguay LA, Kosta LD: Criteria for selection of patients for
emergency portacaval shunts. Am J Surg 134:146, 1977

45. Orloff
MJ, Orloff MS, Orloff SL, et al: Three decades of experience
with emergency portacaval shunt for acutely bleeding esophageal varices
in 400 unselected patients with cirrhosis of the liver. J Am Coll Surg
180:257, 1995
46. OrozcoH, Mercado MA, Takahashi T, et al: Elective treatment of

bleeding varices with the Sugiura operation over 10 years. Am J Surg
163:585, 1992
47. Orozco H, Mercado MA, Takahashi T, et al: Selective splenocaval

shunt for bleeding portal hypertension: Fifteen-year evaluation period.
Surgery 113:260, 1993
48. Orozco H, Takahashi T, Mercado MA, et al: Intermittent changes in

portal venous flow in relation with spleno-pancreatic collaterals--"the
pancreatic siphon"--after a selective shunt in a patient with idiopathic
portal hypertension: Case report. Surgery 111:105, 1992
49. PagliaroL, Burroughs AK, Sorensen T, et al: Therapeutic controversies

and randomized controlled trials (RCT's): Prevention of bleeding and
re-bleeding in cirrhosis. Gastroenterol Int 2:71, 1989
50. PottsJR, Henderson JM, Millikan WJ, et al: Emergency distal

splenorenal shunts for variceal hemorrhage refractory to nonoperative
control. Am J Surg 148:813, 1984
51. PotziR, Bauer P, Schofl R, et al: Prophylactic endoscopic sclerotherapy

of esophageal varices in liver cirrhosis: Long-term follow up and final
results of a multicenter prospective controlled trial in Vienna. Endoscopy
25:287, 1993
52. ReichenJ: Liver function and pharmacological considerations in

pathogenesis and treatment of portal hypertension. Hepatology 11:1066,
1990
53. Rikkers
LF, Jin G: Surgical management of acute variceal hemorrhage.
World J Surg 18:193, 1994
54. Rikkers
LF, Sorrell WT, Jin G: Which portosystemic shunt is best?
Gastroenterol Clin North Am 21:179, 1992
55. Roberts
LR, Kamath PS: Pathophysiology and treatment of variceal
hemorrhage. Mayo Clin Proc 71:973, 1996
114
56. Rosemurgy AS, Goode SE, Camps M: The effect of small-diameter

H-graft portacaval shunts on portal blood flow. Am J Surg 171:154, 1996
57. Rosemurgy AS, Goode SE, Zwiebel BR: A prospective trial of

transjugular intrahepatic portasystemic stent shunts versus small-diameter

prosthetic H-graft portacaval shunts in the treatment of bleeding varices.
Ann Surg 224:378, 1996
58. RossleM, Haag K, Ochs A, et al: The transjugular intrahepatic

portosystemic stent shunt procedure for variceal bleeding. N Engl J Med
330:165, 1994
59. Sarfeh
IJ, Rypins EB, Mason GR: A systematic appraisal of portacaval
H-Graft diameters: Clinical and hemodynamic perspectives. Ann Surg
204:356, 1986
60. SarinSK, Lahoti D, Saxena SP, et al: Prevelence, classification and

natural history of gastric varices: A long-term follow-up study in 568
portal hypertension patients. Hepatology 16:1343, 1992
61. Shiffman ML, Jeffers L, Hoofnagle JH, et al: The role of transjugular
intrahepatic portosystemic shunt for treatment of portal hypertension and
its complications: A conference sponsored by the national digestive
diseases advisory board. Hepatology 22:1591, 1995
62. SpinaGP, Henderson JM, Rikkers LF, et al: Distal spleno-renal shunt
versus endoscopic sclerotherapy in the prevention of variceal bleeding: A
meta-analysis of 4 randomized clinical trials. J Hepatol 16:338, 1992
63. Spina GP, Santambrogio R, Opochere E, et al: Distal splenorenal shunt

versus endoscopic sclerotherapy in the prevention of variceal rebleeding.
First stage of a randomized, controlled trial. Ann Surg 211:178, 1990
64. Starzl
TE, Demetris AJ, Van Thiel DH: Medical progress: Liver
transplantation. N Engl J Med 321:1014, 1989
65. Steigmann G, Goff J, Michaletz-Grody P, et al: Endoscopic

sclerotherapy as compared with endoscopic ligation for bleeding
esophageal varices. N Engl J Med 326:1527, 1992
66. Stipa
S, Balducci G, Ziparo V, et al: Total shunting and elective
67. Sugiura M, Futagawa S: A new technique for treating esophageal

varices. J Thorac Cardiovasc Surg 66:677, 1973
68. Terblanche J, Steigmann GV, Krige JE, et al: Long-term management

of variceal bleeding: The place of varix injection and ligation. World J
Surg 18:185, 1994
69. VorobioffJ, Groszmann RJ, Picabea E, et al: Prognostic value of

hepatic venous pressure gradient measurements in alcoholic cirrhosis: A
10-year prospective study. Gastroenterology 111:701, 1996
70. VorobioffJ, Picabea E, Gamen M, et al: Propranolol compared with

propranolol plus isosorbide dinitrate in portal-hypertensive patients:
Long-term hemodynamic and renal effects. Hepatology 20:1502, 1994

71. Warren WD, Zeppa R, Foman JS: Selective transplenic decompression

of gastroesophageal varices by distal splenorenal shunt. Ann Surg
166:437, 1967
72. WarrenWD, Millikan WJ, Henderson JM, et al: Splenopancreatic

disconnection: Improved selectivity of distal splenorenal shunt. Ann Surg
204:346, 1986
73. WhippleAO: The problem of portal hypertension in relation to the

hepatosplenopathies. Ann Surg 122:449, 1945
74. White S: Discussion of surgical treatment of ascites secondary to

vascular cirrhosis of the liver. Br Med J 2:1287, 1906
75. Yamamoto S, Hidemura R, Sanada M, et al: The late results of terminal

esophago-proximal gastrectomy (TEPG) with extensive devascularization
and splenectomy for bleeding esophageal varices in cirrhosis. Surgery
80:106, 1976



99
PORTAL HYPERTENSION
THE ROLE OF SURGERY IN THE TREATMENT OF PORTAL

HYPERTENSION
David A. Iannitti MD
J. Michael Henderson MB,ChB,FRCS,FACS
From the Cleveland Clinic Foundation, Cleveland, Ohio

J. Michael Henderson, MD
Department of General Surgery A8-418
The Cleveland Clinic Foundation
9500 Euclid Avenue
Cleveland, OH 44095
The role of surgery in the treatment of portal hypertension remains a complex and highly debated issue.
Liver transplantation has come of age in the 1990s and is an accepted therapy for many patients with
end-stage liver disease. The focus of this article is on the role of other surgical options in managing the
complications of portal hypertension, particularly variceal bleeding and refractory ascites. Several factors
must be considered when surgical options are to be entertained, including origin and extent of liver
disease, response to prior medical treatment, and possibility of future liver transplantation. Also,
availability of and response to other technologies, such as endoscopic treatment and radiologic
intervention (TIPS), must be factored into the treatment plan. This article reviews the history of
interventional treatments of portal hypertension, describes the various shunting procedures, and evaluates
their role in the various clinical scenarios that may require surgery as a treatment option.
For the past 100 years, surgery has played an integral role in the understanding and treatment of portal
hypertension. Understanding portal anatomy and alterations in portal venous blood flow in cirrhosis is
the basis for surgical and other interventional procedures (Fig. 1) . The initial work in this area began
with Nikolai Eck in 1877. He fashioned side-to-side portacaval anastomoses with ligation of the distal
portal vein in eight dogs (Fig. 2) . In 1903 Eugene Vidal performed the first end-to-side portacaval

anastomosis successfully in man. Vidal observed
100
Figure 1. Normal portal anatomy demonstrating hepatopedal blood flow in the coronary, splenic,
superior mesenteric, inferior mesenteric, and portal veins.
Figure 2. The end-to-side portacaval shunt or Eck fistula. Note complete diversion of all portal blood
flow into the inferior vena cava. The distal portal vein is ligated which maintains a high hepatic
sinusoid pressure.
101
cessation of bleeding and resolution of ascites; however, the patient subsequently developed
encephalopathy, reaccumulation of ascites, and expired 3.5 months after surgery. Several historical
articles review these early experiences. [5] [6] [12]
Several procedures were developed in the early 20th century to promote portalsystemic collateralization,
including the omentopexy of Drummond and Morrison, [74] splenic transposition, first described by
Talma, [28] and the visceral abrasion techniques of Madden et al. [39] Enthusiasm for these procedures
waned as their results yielded no improvement in patient outcome. In 1945, Allen Whipple reintroduced
portal systemic shunting for the management of complications of portal hypertension. [73] He reported a
series of portacaval and nonselective splenorenal shunts. Linton [38] continued this work, and in 1961
reviewed his experience in 169 patients. He concluded that each type of shunt had equal efficacy in
controlling bleeding, end-to-side portacaval shunts did not control ascites, and splenorenal shunting
resulted in less encephalopathy. The incidence of encephalopathy continued to remain high, and
progression of hepatic failure accelerated following total, nonselective shunting procedures. These
findings led to the development of selective shunting, including the distal splenorenal shunt by Warren
[71] (Fig. 3) and the coronary-caval shunt by Inokuchi [30] (Fig. 4) . These shunting procedures
Figure 3. The distal splenorenal shunt or Warren shunt. The distal splenic vein is mobilized and
anastomosed to the left renal vein. This selectively provides a low pressure outflow tract for
gastroesophageal varices. The coronary vein and other collaterals are ligated. Note this is not a
portal-systemic shunt.

102
Figure 4. The coronary-caval shunt. The left gastric vein is anastomosed to the inferior vena cava
providing decompression of gastroesophageal varices. Splenectomy is included with this procedure.
maintain portal hypertension and hepatic perfusion while selectively decompressing gastroesophageal
varices effectively.
The past two decades have seen major changes in managing patients with cirrhosis and portal
hypertension. Further understanding of the pathophysiology of this disease process has led to the use of
newer and more efficacious medications. [52] [55] [66] [69] Medical pharmacotherapy presently includes the
use of vasopressin alone or in combination with nitroglycerine; propranolol; and, more recently,
octreotide and isosorbide mononitrate. [36] [42] [70]
In the 1970s endoscopic sclerotherapy, initially described in 1939, was reintroduced and became widely
accepted. This modality of treatment became the standard for treating acute variceal bleeding.
Sclerotherapy combined with pharmacologic reduction of portal hypertension using beta-blockers also
became the primary treatment for acute-bleeding esophageal varices and prevention of recurrent
bleeding. [33] [37] Improved endoscopic techniques using elastic band ligation have proved to be more
efficacious with less procedure-related complications. [68]
The introduction of the transjugular intrahepatic portal systemic shunt (TIPS) has rapidly gained
worldwide acceptance in creating a functional, total side-to-side shunt using a minimally invasive
technique. [35] [58] TIPS patency rates have improved with aggressive monitoring
103
and radiographic interventional procedures. The accepted indications for TIPS are acute variceal
bleeding that cannot be successfully controlled with medical treatment, including sclerotherapy, and
recurrent variceal bleeding in patients who are refractory or intolerant to conventional medical
management, including sclerotherapy and pharmacologic therapy, and are not candidates for surgery.
TIPS may be useful for the treatment of refractory ascites, and its use has been reported for acute
Budd-Chiari syndrome. [61]
The 1990s have also seen a further development in the surgical treatment of portal hypertension. Surgical
shunt procedures now include partial portal systemic shunts and pancreatic disconnection for distal
splenorenal shunts. [7] [26] The introduction of liver transplantation in the 1980s provided definitive
treatment of the underlying liver disease, and is the ultimate treatment for portal hypertension. [20] [64]
Liver transplantation has improved with lower morbidity and mortality and improved survival with the
development of newer, more effective immunosuppressant medications, such as tacrolimus (FK506), and

improved management schemes for control of infection.
Presently, it is difficult to decide the optimal treatment for a given patient with such options available.
Careful patient evaluation with an understanding of the underlying liver disease, portal anatomy, possible
need for transplantation, and availability of appropriate technology and expertise must be considered for
the treatment plan. [24] [27] [34]
SURGICAL PROCEDURES
Decompression of gastroesophageal varices controls bleeding, but has not been widely accepted to treat
all patients with variceal bleeding. This is primarily because the first type of shunts were total portal
systemic shunts, and diversion of portal blood flow was detrimental to liver function; however, other
types of operative shunts have been introduced, and in the 1990s attention has focused back on
decompression because of TIPS. In evaluating the role of decompression, the effect of different types of
shunts on portal perfusion is key. [8] [11] [21] [54]
Total portalsystemic shunts divert the entire blood flow of the portal system into the systemic circulation.
Total shunts can be achieved in several ways. The classic end-to-side portacaval anastomosis (the Eck
fistula) (see Fig. 2) clearly provides complete diversion of all portal flow; however, the procedure does
not provide outflow of the liver sinusoids, and as such, it cannot be used to treat ascites. The side-to-side
portacaval anastomosis of 1.2 cm or greater also results in a total portal systemic shunt, but because the
portal vein remains intact, it acts as an outflow from the liver sinusoids and thus controls ascites (Fig. 5) .
Any portacaval
104
Figure 5. The side-to-side portacaval shunt. This is a total shunt providing complete diversion of all
portal blood flow as well as decompression of the hepatic sinusoids.
shunt requires dissection of the hepatic hilus to free the portal vein and will make subsequent liver
transplantation more difficult.
Total portal systemic shunting can also be accomplished using large-caliber (16-22-mm) interposition
graft placed from the portal or superior mesenteric vein to the inferior vena cava, left renal vein, or right
atrium (Fig. 6) . Although some of these shunts do require dissection of the hepatic hilus, they are more
easily divided should liver transplantation be needed. The major disadvantage of an interposition graft is
the primary patency is less than that of a side-to-side portacaval vein-to-vein anastomosis. Recent
radiographic techniques, however, including balloon dilatation and infusion of thrombolytic agents, have

increased the primary, primary assisted, and secondary patency of these grafts.
Nonselective or total shunts eliminate portal hypertension, effectively control and prevent variceal
bleeding, and with the exception of the end-to-side portacaval shunt, are useful in treating ascites.
Disadvantages are related to diversion of all portal flow, which leads to an increased incidence of
encephalopathy and accelerated progression of the underlying liver disease.
Partial Shunts
Partial shunts create a side-to-side communication of limited size between the portal and systemic
circulation, lower the portal pressure,
105
Figure 6. The interposition shunt. This is a functional side-to-side total shunt. An interposition graft
(16-22-mm polytetrafluoroethylene [PTFE]) can be placed from the portal or superior mesenteric
vein to the inferior vena cava, left renal vein, or right atrium.
and maintain some hepatic portal perfusion. This concept of partial shunting was begun by Bismuth, [2]
who observed an initial gradient across a 15-mm side-to-side portacaval anastamosis. At this size, these
eventually became total shunts with reversal of portal flow. Johansen [31] continued this work by creating
smaller, 10- to 12-mm side-to-side portacaval anastamoses, and although significant pressure gradients
could be measured initially between the portal and systemic circulations, these shunts became total
shunts eventually.
Recently, Sarfeh [59] has popularized the use of small-diameter, reinforced polytetrafluoroethylene
(PTFE) interposition grafts between the portal vein and inferior vena cava. [59] An 8-mm graft reduces
portal pressure to 12 mmHg, which is low enough to control bleeding, while maintaining hepatic portal
perfusion in 80% of subjects. When a 10-mm graft is used, hepatic portal perfusion is only maintained in
20% of subjects. The coronary vein, gastroepiploic veins, and other collateral vessels must be ligated to
decrease inflow to the esophageal varices.
Partial shunting results in lowered portal pressure while maintaining some prograde hepatic venous flow
and are effective treatment for bleeding esophageal varices with a lower rate of encephalopathy than total
shunts. The disadvantages include dissection of the hepatic hilus and an early thrombosis rate of 16%;
however, with modern
106
radiologic interventional techniques, long-term patency can be maintained.
Selective Shunts
Selective shunts provide selective decompression of gastroesophageal varices while maintaining portal

hypertension and portal venous perfusion. These are not portal systemic shunts because portal blood flow
is not diverted. Generally there are two types of selective shunts: the distal splenorenal shunt-devised and
popularized by Warren, [71] and the infrequently used coronary-caval shunt of Inokuchi. [30]
The coronary-caval shunt includes splenectomy and involves dissection and disconnection of the
coronary (left gastric) vein and anastomosis to the inferior vena cava (see Fig. 4) . This procedure is
technically difficult, usually requires interposition vein grafting, and is rarely used.
The distal splenorenal shunt, which selectively decompresses the spleen and gastroesophageal junction,
has gained worldwide acceptance. The procedure requires disconnection of the splenic vein from the
superior mesenteric vein, dissection of this vein out from the pancreas, and anastomosis to the left renal
vein in an end-to-side fashion (see Fig. 3) . This provides a low-pressure outlow tract for
gastroesophageal varices. Portal systemic collateral vessels, including the left gastric vein, gastroepiploic
veins, and vessels within the splenocolic ligament, are ligated.
The distal splenorenal shunt has several advantages over portal systemic shunting for the decompression
of gastroesophageal varices. Hepatopedal portal vein flow is maintained, which decreases the risk of
encephalopathy and accelerated liver failure. Esophageal and gastric varices are effectively
decompressed, and the dissection is within the lesser sac, distant to hepatic hilus. Long-term patency is
excellent in this high-flow vein-to-vein anastomosis. [23] Although portal vein thrombosis has been
recorded in 5% to 10% of patients, hepatic portal perfusion is maintained in 90% of patients early after
shunt. In nonalcoholic patients, good hepatic portal perfusion is maintained long term, whereas in
alcoholics it is lost in 16% to 50%, depending on how much splenic vein is dissected out of the pancreas.
Loss of portal flow occurs through the development of intrapancreatic portal systemic collateralization
known as the pancreatic siphon.[48] Improved maintenance of portal perfusion in alcoholic patients has
been achieved by the addition of splenopancreatic disconnection to the distal splenorenal shunt
procedure. [26] [72] In this modification, the splenic vein is completely disconnected from the posterior
pancreas to the splenic hilum. Other modifications, such as transposition of the splenic vein [47] or
interposition grafting to the inferior vena cava, can be made in patients with a retro-aortic left renal vein
(present in 4% of the population). Distal splenorenal shunting effectively treats bleeding
gastroesophageal varicies, portal hypertensive gastropathy, and prevents recurrent bleeding in over 90%
of patients. [3] Because portal hypertension is maintained, it cannot be used to treat ascites.
107
Devascularization (nonshunting) procedures of the stomach and lower esophagus were proposed by Gray
and Witesell in 1950. [16] Several investigators performed various devascularization procedures with the
modifications of variceal ligation, esophageal transection, and splenectomy. In 1973, Sugiura and
Futagawa [67] reported a two-stage procedure involving both transthoracic and transabdominal approaches
for interruption of gastroesophageal varices while still maintaining flow through other portal-azygous
collaterals. Yammamoto et al [75] modified this procedure to be performed through the transabdominal
approach only.
The effectiveness of devascularization procedures for the control of variceal bleeding appears to be
related to the extent of the procedure. For maximal efficacy, a devascularization procedure usually

includes esophageal transection and reanastamosis (which can be achieved with a circular stapling
device), truncal vagotomy, pyloroplasty, splenectomy, and ligation of perigastric and periesophageal
vessels to at least 7 cm up the esophagus. This provides interruption of inflow to gastroesphageal varices
with maintenance of portal hypertension and hepatic portal perfusion. Devascularization procedures are
effective in controlling actively bleeding gastroesophageal varices; there is a low operative mortality and
rebleeding rate. [29] Outside of Japan, most centers report up to a 37% rebleeding rate. [10] [46] Most
surgeons can perform devascularization procedures, so they remain an acceptable alternative in an
emergency and for "unshuntable" patients. [4] Splenectomy alone in treating bleeding varices has a
limited role. [18] Splenectomy may be curative in patients with bleeding gastric varices from isolated
splenic vein thrombosis; however, splenectomy alone is not indicated in patients with portal hypertension
secondary to cirrhosis.
This section briefly describes management options at the major time points of presentation for patients
with portal hypertension and varices.
Prophylaxis Prior to the First Variceal Bleed
Esophageal varices are present at the time of diagnosis of cirrhosis in 30% of patients with compensated
and 60% of patients with decompensated disease. Untreated varices tend to progressively increase in size
before they eventually rupture and bleed. There is a 5% per year risk of bleeding in unselected patients
with cirrhosis, this risk increases to 20% to 30% in patients who already have varices. [11]
Data in the 1990s support the prevention or prophylaxis of a first variceal bleed with pharmacologic
treatment. [52] Propranolol effectively
108
lowers variceal pressure and hepatic venous pressure gradient. [14] [15] A recent multicenter trial of
patients with cirrhosis and endoscopically proven esophageal varices compared the use of propranolol or
placebo prior to the first variceal bleed. The propranolol group had a significantly lower incidence of
bleeding from esophageal varices (4%) versus the placebo group (22%), and portal hypertensive
gastropathy, although there was no difference in survival between the two groups. [9] Meta-analysis of
multiple studies supports this management strategy.
Endoscopic sclerotherapy or banding should not be undertaken as a prophylactic measure in patients with
portal hypertension and varices. Reports are conflicting as to their efficacy, [17] [51] but their use cannot be
supported outside randomized trials. Presently, surgery is not indicated as a treatment prophylaxis for
variceal bleeding. [17]
Acute Bleeding Esophageal Varices
Approximately 30% of patients with cirrhosis and esophageal varices experience variceal hemorrhage,
usually within 1 year of diagnosis. This is a major, potentially life-threatening complication, with
mortality rates ranging from 30% to more than 50% in most studies. [11] Gastric varices are less likely to
bleed; however, when bleeding occurs, it is more extensive. [60] The status of hepatic function is the

major determinant in survival. [40]

The initial management of a patient with bleeding esophageal varices consists of adequate volume
replacement with blood, blood products, and colloid solutions. Pharmacologic portal pressure reduction
can be acutely achieved with medications such as vasopressin and nitroglycerine, or octreotide. [55] Early
endoscopy is employed to identify and confirm the source of bleeding and to control bleeding with
sclerotherapy or banding. When these measures are carried out, control of variceal hemorrhage can be
achieved in about 90% of patients. [25]
Urgent or emergent shunting may be required in the 10% of patients whose acute bleeding is not
controlled by endoscopic measures. [50] [53] This should be performed with a TIPS procedure because
failure to control acute bleeding endoscopically almost always occurs in poor risk patients. In some
good-risk patients, operative shunting procedures may be used. [32] [57] The largest experience of
emergency portacaval shunts has recently been reported by Orloff et al. [45] He reviewed 340 unselected
cirrhotic patients undergoing emergency side-to-side portacaval shunts. All patients had control of
bleeding and long-term patency of their shunts. Remarkably, the recent group of 220 patients had a 15%
incidence of hepatic failure and 9% incidence of encephalopathy. Selective and nonselective shunts
effectively stop bleeding from gastroesophageal varices, and have acceptable incidence of side effects.
The choice of shunt ultimately remains with the involved surgeon.
In the case of "unshuntable" patients or unfamiliarity with shunt procedures, nonshunting
devascularization procedures can be used. [4]
109
Devascularization adequately controls bleeding and can be performed by most surgeons. This procedure
is associated with a low rate of encephalopathy and hepatic failure; however, rebleeding rates remain
high (about 40%). Devascularization is a reasonable alternative in an emergency situation.
Prevention of Recurrent Variceal Bleeding
The natural history of patients with cirrhosis who experience a variceal bleed is that 47% to 84% of these
patients will experience rebleeding within 1 to 2 years if they do not receive treatment for the portal
hypertension or obliteration of the varices. [44] The mortality of a recurrent variceal hemorrhage ranges
from 20% to 70%. Advanced liver disease, variceal size, and continued alcohol use are related to
increased rebleeding risk. [11]
Following the initial resuscitation and control of acute bleeding, patient evaluation is the key to making
definitive treatment choices. Several questions must be answered in this time period. First, evaluation
should address the status of liver function. Does the patient have end-stage liver disease that can be only
managed by transplantation? If so, is the patient a transplant candidate? Does the patient have normal
hepatic function with an extra hepatic cause of portal hypertension, such as portal vein thrombosis? The
next series of questions relate to the varices. This involves endoscopy and a hemodynamic study with
Doppler ultrasound or angiography to evaluate the patency of the portal venous system. Finally, the
question must be asked: what resources are available for definitive treatment? These include
endoscopists capable of performing sclerotherapy and elastic band ligation, interventional radiologists
skilled in TIPS, surgeons familiar with various shunting and nonshunting procedures, and liver transplant

capability and availability.

The first line of treatment for prevention of recurrent bleeding in patients with portal hypertension is by
pharmacologic and endoscopic treatment. [14] [36] When these measures fail to prevent recurrent variceal
bleeding, variceal decompression should be considered. TIPS, surgical shunts, and liver transplantation
are the decompression options.
Patients with poor hepatic function are potential transplant candidates. They should be fully evaluated for
transplant, and if candidates, the only issue is that of timing. In transplant candidates with recurrent
bleeding not controlled by endoscopy, TIPS is a reasonable option. Equally, TIPS, in patients who fail
medical and endoscopic therapy and have poor liver function is an option in patients who are not
transplant candidates. The current data on TIPS can be summarized as follows:
Technical success of TIPS is achieved in approximately 90% of patients, resulting in a portal
systemic pressure gradient of < 12 mmHg.
Failure of TIPS is related to shunt stenosis and occlusion.
110
Primary patency at 1 year ranges from 50%-65%, and primary-assisted patency is about 85% at 1
year. Careful follow-up with stent monitoring results in a 25%-50% reintervention rate to maintain
patency. One-year rebleeding rate after TIPS is 18%-20%. [19] [35] Long-term patency and
rebleeding rates are now being evaluated.
Patients with good or adequate hepatic function and failing medical and endoscopic treatment are
candidates for surgical decompression. Distal splenorenal shunts have been proved to be more
efficacious in prevention of rebleeding from gastroesophageal varices than endoscopic sclerotherapy in a
meta-analysis of four randomized clinical trials. [62] [63] Distal splenorenal shunts are effective in
long-term control of variceal bleeding in over 90% of patients. Portal perfusion is maintained in more
than 90% of patients with nonalcoholic liver disease and 50% to 84% of patients with alcoholic liver
disease. Modifying the distal splenorenal shunt with pancreatic disconnection of the splenic vein in the
latter group minimizes the development of intrapancreatic portal systemic collaterals [26] and maintain
better portal flow. Postshunt encephalopathy rates remain less than 10%, and selective shunting does not
accelerate the natural history of the underlying liver disease. [23] Distal splenorenal shunting has been
compared prospectively to partial portalsystemic shunting in a recent study. [41] Both shunts are equally
efficacious in control of rebleeding; however, encephalopathy and shunt thrombosis rates were higher in
patients who recieved partial, 10-mm interposition mesocaval shunts.
The choice between distal splenorenal shunts and TIPS in patients with good hepatic function is a topic
of current interest. Distal splenorenal shunts have proved long-term patency and efficacy in this patient
group. The popularity and use of TIPS has increased in recent years. Retrospective review of patients
undergoing surgical shunting or TIPS reveals lower rates of mortality, encephalopathy, and rebleeding in
the surgical shunt groups. TIPS had a 12% early thrombosis, 41% stenosis, about 20% rebleeding, and

29% encephalopathy rates. [1] [8] [11] Surgical shunts overall have about 5% variceal rebleeding, and
encephalopathy from 10% to 40%, depending on the type of shunt. Presently there is an ongoing
multicenter, prospective, randomized trial comparing distal splenorenal shunts to TIPS in patients with
good hepatic function.
Devascularization procedures are generally used in unshuntable patients with thrombosed splenic and
portal venous systems. Devascularization in Child's A patients results in 4% early and 6% rebleeding
rates in Japan, [29] and a 3% encephalopathy rate. [46] This procedure is a good alternative in otherwise
unshuntable patients.
Refractory Ascites
Ascites remain a difficult problem in patients with cirrhosis and portal hypertension. Medical treatment
includes dietary salt restriction
111
and diuretic therapy, which can usually improve or resolve ascites in 90% of patients. Ascites that are not
responsive to diet and diuretics is termed refractory. When this occurs, repeat large-volume paracentesis
can provide temporary relief, but is associated with a high recurrence rate. Several interventional
procedures, including peritoneovenous shunts, surgical portal systemic shunts, and TIPS, have been used
in these patients. These procedures are all associated with high morbidity, mortality, and failure rates.
Liver transplantation remains the definitive treatment for patients with cirrhosis, portal hypertension, and
refractory ascites.
Patients with poor hepatic function and refractory ascites are best treated by liver transplantation. As a
bridge to transplant, treatment may include TIPS, but complications remain high and include a 30-day
mortality of 4% to 8%, and an encephalopathy rate of 20% to 30%. TIPS improves ascites in over 90%
of patients that survive and can be used as a temporary measure until transplant. [43] TIPS has also been
used for cirrhotic hydrothorax and hepatorenal syndrome.
Surgical side-to-side portal systemic shunts are also effective in treating refractory ascites in patients
with portal hypertension. In patients with good hepatic function and a history of variceal bleeding, a
functional side-to-side total shunt may be indicated. End-to-side portacaval and distal splenorenal shunts
do not help ascites. The major problem with portal systemic shunting is the high incidence of
encephalopathy. Of patients following a total shunt for refractory ascites, 40% to 50% will develop
encephalopathy within 1 year.
Peritoneovenous shunts are another treatment that has been used for patients with refractory ascites.
These shunts are usually indicated for patients who have no history of variceal bleeding and reasonable
liver function. Patients with a high degree of liver dysfunction who undergo peritoneovenous shunting
have a higher rate of postprocedure complications. These include disseminated intravascular
coagulopathy, dilutional coagulapathy, gastrointestinal bleeding, congestive heart failure, shunt infection,
and shunt malfunction. Although peritoneovenous shunting has not been shown to improve survival
compared with repeat paracentesis, it can provide better control of ascites and better quality of life.
Improvements in respiratory function, activity level, food intake, and weight loss can be seen following
peritoneovenous shunting in carefully selected patients. In the 1990s, however, the indication for

peritoneovenous shunting is rare as most of these patients are transplant candidates. [8] [13]
References
1. Becker YT, Reed G, Lind CD: The role of elective operation in the treatment of portal hypertension. Am Surg 62:171,
1996
2. Bismuth H, Franco D, Hepp J: Portal-systemic shunt in hepatic cirrhosis: Does the type of shunt decisively influence the
clinical result? Ann Surg 179:209, 1974
112
3. Boyer TD, Kokenes DD, Hertzler G, et al: Effect of distal splenorenal shunt on survival of patients with primary biliary
cirrhosis. Hepatology 20:1484, 1994
4. Caps MT, Helton WS, Johansen K: Left-upper-quadrant devascularization for "unshuntable" portal hypertension. Arch
Surg 131:834, 1996
5. Chandler JG: The history of surgical treatment of portal hypertension. Arch Surg 128:925, 1993
6. Child CG: Eck's fistula. Surg Gynecol Obstet 96:375, 1953
7. Collins JC, Sarfeh IJ: Surgical management of portal hypertension. West J Med 162:527, 1995
8. Collins
JC, Rypins EB, Sarfeh IJ: Narrow-diameter portacaval shunts for management of variceal bleeding. World J Surg
18:211, 1994
9. Conn HO, Grace ND, Bosch J, et al: Propranolol in the prevention of the first hemorrhage from esophagogastric varices:
A multicenter, randomized clinical trial. Hepatology 13:902, 1991
10. Dagenais M, Langer B, Taylor BR, et al: Experience with radical esophagogastric devascularization procedures
(Sugiura) for variceal bleeding outside Japan. World J Surg 18:222, 1994
11. D'Amico G, Pagliaro L, Bosch J: The treatment of portal hypertension: A meta-analytic review. Hepatology 22:332,
1995
12. Donovan AJ, Covey PC: Early history of portacaval shunt in humans. Surg Gynecol Obstet 147:423, 1978
13. Elcheroth J, Vons C, Franco D: Role of surgical therapy in management of intractable ascites. World J Surg 18:241,
1994
14. FeuF, Bordas JM, Garcia-Pagan JC, et al: Double-blind investigation of the effects of propranolol and placebo on the
pressure of esophageal varices in patients with portal hypertension. Hepatology 13:917, 1991
15. FeuF, Bordas JM, Lucy A, et al: Reduction of variceal pressure by propranolol: Comparison of the effects on portal
pressure and azygos blood flow in patients with cirrhosis. Hepatology 18:1082, 1993
16. Gray HK, Whitsell FB: Hemorrhage from esophageal varices. Ann Surg 132:798, 1950

17. GreigJD, Garden OJ, Carter DC: Prophylactic treatment of patients with esophageal varices: Is it ever indicated? World
J Surg 18:176, 1994
18. Gusberg RJ, Peterec SM, Sumpio BE, et al: Splenomegaly and variceal bleeding: Hemodynamic basis and treatment
implications. Hepato-Gastroenterol 41:572, 1994
19. HeltonWS, Belshaw A, Althaus S, et al: Critical appraisal of the angiographic portacaval shunt (TIPS). Am J Surg
165:566, 1993
20. Henderson JM: Liver transplant for portal hypertension. Gastroenterol Clin North Am 21:197, 1992
21. Henderson JM: Portal hypertension and shunt surgery. Adv Surg 26:233, 1993
22. Henderson JM: The distal splenorenal shunt in the management of bleeding varices. J Royal Coll Surg Edinburg
27:130, 1982
23. Henderson JM: The role of distal splenorenal shunt for long-term management of variceal bleeding. World J Surg
18:205, 1994
24. Henderson JM, Warren WD: Portal hypertension. Curr Prob Surg 25:151, 1988
25. Henderson JM, Carey WD, Vogt DP, et al: Management of variceal bleeding in the 1990's. Clev Clin J Med 60:431,
1993
26. HendersonJM, Warren WD, Millikan WJ, et al: Distal splenorenal shunt with splenopancreatic disconnection: A 4-year
assessment. Ann Surg 210:332, 1989
27. Hermann RE, Henderson JM, Vogt DP, et al: Fifty years of surgery for portal hypertension at the Cleveland Clinic
Foundation. Ann Surg 221:459, 1995
28. Holman E: Implantation of the spleen in the abdominal wall for portal obstruction: A suggested operation for hepatic
cirrhosis. West J Surg Obstet Gynecol 41:255, 1933
29. Idezuki
Y, Kokudo N, Sanjo K, et al: Sugiura procedure for management of variceal bleeding in Japan. World J Surg
18:216, 1994
30. Inokuchi K: A selective portacaval shunt. Lancet 2:51, 1968
31. Johansen K: Partial portal decompression for variceal hemorrhage. Am J Surg 157:479, 1989
113
32. Johansen K, Helton WS: Portal hypertension and bleeding esophageal varices. Ann Vasc Surg 6:553, 1992
33. Johnston
GW, Rodgers HW: A review of 15 years' experience in the use of sclerotherapy in the control of acute
hemorrhage from oesophageal varices. Br J Surg 60:797, 1973
34. Knechtle SJ, Kalayoglu M, D'Alessandro AM, et al: Portal hypertension: Surgical management in the 1990's. Surgery

116:687, 1994
35. LaBergeJM, Ring EJ, Gordon RL, et al: Creation of transjugular intrahepatic portosystemic shunt with wallstent
endoprosthesis: Results in 100 patients. Radiology 187:413, 1993
36. Lebrec D: Long-term management of variceal bleeding: The place of pharmacotherapy. World J Surg 18:229, 1994
37. Lebrec D, Wovel O, Corbic M, et al: Propranolol: A medical treatment for portal hypertension. Lancet 2:180, 1980
38. LintonRR, Ellis DS, Geary JE: Critical comparative analysis of early and late results of splenorenal and direct
portacaval shunts performed in 169 patients with portal cirrhosis. Ann Surg 154:446, 1961
39. Madden JL, Lore JM, Gerald FP, et al: The pathogenesis of ascites and a consideration of its treatment. Surg Gynecol
Obstet 99:585, 1954
40. McCormick PA, Burroughs AK: Relation between liver pathology and prognosis in patients with portal hypertension.
World J Surg 18:171, 1994
41. MercadoMA, Morales-Linares JC, Granados-Garcia J, et al: Distal splenorenal shunt versus 10 mm low-diameter
mesocaval shunt for variceal hemorrhage. Am J Surg 171:591, 1996
42. Morillas
RM, Planas R, Cabre E, et al: Propranolol plus isosorbide-5-mononitrate for portal hypertension in cirrhosis:
Long-term hemodynamic and renal effects. Hepatology 20:1502, 1994
43. Ochs A, Rossle M, Haag K, et al: The transjugular intrahepatic portosystemic stent-shunt procedure for refractory
ascites. N Engl J Med 332:1192, 1995
44. Orloff
MJ, Duguay LA, Kosta LD: Criteria for selection of patients for emergency portacaval shunts. Am J Surg
134:146, 1977
45. Orloff
MJ, Orloff MS, Orloff SL, et al: Three decades of experience with emergency portacaval shunt for acutely
bleeding esophageal varices in 400 unselected patients with cirrhosis of the liver. J Am Coll Surg 180:257, 1995
46. OrozcoH, Mercado MA, Takahashi T, et al: Elective treatment of bleeding varices with the Sugiura operation over 10
years. Am J Surg 163:585, 1992
47. Orozco H, Mercado MA, Takahashi T, et al: Selective splenocaval shunt for bleeding portal hypertension: Fifteen-year
evaluation period. Surgery 113:260, 1993
48. OrozcoH, Takahashi T, Mercado MA, et al: Intermittent changes in portal venous flow in relation with
spleno-pancreatic collaterals--"the pancreatic siphon"--after a selective shunt in a patient with idiopathic portal
hypertension: Case report. Surgery 111:105, 1992
49. Pagliaro
L, Burroughs AK, Sorensen T, et al: Therapeutic controversies and randomized controlled trials (RCT's):
Prevention of bleeding and re-bleeding in cirrhosis. Gastroenterol Int 2:71, 1989
50. Potts
JR, Henderson JM, Millikan WJ, et al: Emergency distal splenorenal shunts for variceal hemorrhage refractory to
nonoperative control. Am J Surg 148:813, 1984
51. Potzi
R, Bauer P, Schofl R, et al: Prophylactic endoscopic sclerotherapy of esophageal varices in liver cirrhosis:
Long-term follow up and final results of a multicenter prospective controlled trial in Vienna. Endoscopy 25:287, 1993

52. Reichen
J: Liver function and pharmacological considerations in pathogenesis and treatment of portal hypertension.
53. Rikkers LF, Jin G: Surgical management of acute variceal hemorrhage. World J Surg 18:193, 1994
54. Rikkers LF, Sorrell WT, Jin G: Which portosystemic shunt is best? Gastroenterol Clin North Am 21:179, 1992
55. Roberts LR, Kamath PS: Pathophysiology and treatment of variceal hemorrhage. Mayo Clin Proc 71:973, 1996
114
56. Rosemurgy AS, Goode SE, Camps M: The effect of small-diameter H-graft portacaval shunts on portal blood flow. Am
J Surg 171:154, 1996
57. Rosemurgy AS, Goode SE, Zwiebel BR: A prospective trial of transjugular intrahepatic portasystemic stent shunts
versus small-diameter prosthetic H-graft portacaval shunts in the treatment of bleeding varices. Ann Surg 224:378, 1996
58. RossleM, Haag K, Ochs A, et al: The transjugular intrahepatic portosystemic stent shunt procedure for variceal
bleeding. N Engl J Med 330:165, 1994
59. SarfehIJ, Rypins EB, Mason GR: A systematic appraisal of portacaval H-Graft diameters: Clinical and hemodynamic
perspectives. Ann Surg 204:356, 1986
60. Sarin
SK, Lahoti D, Saxena SP, et al: Prevelence, classification and natural history of gastric varices: A long-term
follow-up study in 568 portal hypertension patients. Hepatology 16:1343, 1992
61. Shiffman ML, Jeffers L, Hoofnagle JH, et al: The role of transjugular intrahepatic portosystemic shunt for treatment of
portal hypertension and its complications: A conference sponsored by the national digestive diseases advisory board.
62. SpinaGP, Henderson JM, Rikkers LF, et al: Distal spleno-renal shunt versus endoscopic sclerotherapy in the prevention
of variceal bleeding: A meta-analysis of 4 randomized clinical trials. J Hepatol 16:338, 1992
63. SpinaGP, Santambrogio R, Opochere E, et al: Distal splenorenal shunt versus endoscopic sclerotherapy in the
prevention of variceal rebleeding. First stage of a randomized, controlled trial. Ann Surg 211:178, 1990
64. Starzl TE, Demetris AJ, Van Thiel DH: Medical progress: Liver transplantation. N Engl J Med 321:1014, 1989
65. SteigmannG, Goff J, Michaletz-Grody P, et al: Endoscopic sclerotherapy as compared with endoscopic ligation for
bleeding esophageal varices. N Engl J Med 326:1527, 1992
66. Stipa
S, Balducci G, Ziparo V, et al: Total shunting and elective management of variceal bleeding. World J Surg
18:200, 1994
67. Sugiura M, Futagawa S: A new technique for treating esophageal varices. J Thorac Cardiovasc Surg 66:677, 1973
68. Terblanche J, Steigmann GV, Krige JE, et al: Long-term management of variceal bleeding: The place of varix injection
and ligation. World J Surg 18:185, 1994

69. Vorobioff J, Groszmann RJ, Picabea E, et al: Prognostic value of hepatic venous pressure gradient measurements in
alcoholic cirrhosis: A 10-year prospective study. Gastroenterology 111:701, 1996
70. VorobioffJ, Picabea E, Gamen M, et al: Propranolol compared with propranolol plus isosorbide dinitrate in
portal-hypertensive patients: Long-term hemodynamic and renal effects. Hepatology 20:1502, 1994
71. WarrenWD, Zeppa R, Foman JS: Selective transplenic decompression of gastroesophageal varices by distal
splenorenal shunt. Ann Surg 166:437, 1967
72. WarrenWD, Millikan WJ, Henderson JM, et al: Splenopancreatic disconnection: Improved selectivity of distal
splenorenal shunt. Ann Surg 204:346, 1986
73. Whipple AO: The problem of portal hypertension in relation to the hepatosplenopathies. Ann Surg 122:449, 1945
74. White S: Discussion of surgical treatment of ascites secondary to vascular cirrhosis of the liver. Br Med J 2:1287, 1906
75. Yamamoto S, Hidemura R, Sanada M, et al: The late results of terminal esophago-proximal gastrectomy (TEPG) with
extensive devascularization and splenectomy for bleeding esophageal varices in cirrhosis. Surgery 80:106, 1976



Figure 5. The side-to-side portacaval shunt. This is a total shunt providing complete diversion of all
portal blood flow as well as decompression of the hepatic sinusoids.



Figure 2. The end-to-side portacaval shunt or Eck fistula. Note complete diversion of all portal blood
flow into the inferior vena cava. The distal portal vein is ligated which maintains a high hepatic sinusoid
pressure.



Figure 4. The coronary-caval shunt. The left gastric vein is anastomosed to the inferior vena cava
providing decompression of gastroesophageal varices. Splenectomy is included with this procedure.



Figure 6. The interposition shunt. This is a functional side-to-side total shunt. An interposition graft
(16-22-mm polytetrafluoroethylene [PTFE]) can be placed from the portal or superior mesenteric vein to
the inferior vena cava, left renal vein, or right atrium.


April 22, 2000
Conde Petra
NEW CONCEPTS IN THE PATHOGENESIS OF

PORTAL HYPERTENSION Hepatic Wounding
and Stellate Cell Contractility
Additional Article
Full Text
Frontmatter 13
CELLULAR COMPONENTS OF THE

HEPATIC MICROCIRCULATION PORTAL HYPERTENSION
HEPATIC INJURY, WOUNDING,
AND STELLATE CELLS
NEW CONCEPTS IN THE PATHOGENESIS
STELLATE CELL CONTRACTILITY
OF PORTAL HYPERTENSION
ENDOTHELIN AND ENDOTHELIN Hepatic Wounding and Stellate Cell Contractility
RECEPTORS
Endothelin
ET Regulation
ET Receptors Don C. Rockey MD

ET Receptors in Stellate Cells
From the Gastroenterology Division, Duke University Medical Center,
NITRIC OXIDE AND CARBON Durham, North Carolina
MONOXIDE
Complications resulting from portal hypertension
Nitric Oxide encompass some of the most serious clinical sequelae
Carbon Monoxide confronting clinicians caring for patients with chronic liver
disease. A number of postulates attempt to explain the
HEPATIC WOUNDING AND pathogenesis of portal hypertension. In the "backward
STELLATE CELL CONTRACTILITY flow" theory, increased intrahepatic resistance leads to
ET IN LIVER INJURY elevated portal pressure; the "forward flow" hypothesis
suggests that humoral factors (directly or indirectly)
SUMMARY contribute to increased portal venous return and flow,
References
ultimately resulting in portal hypertension. Nonetheless,
portal hypertension occurs after any of a multitude of
About the Publication different parenchymal liver injuries and increased

intrahepatic resistance to blood flow is a central element,

regardless of the pathogenesis.
The precise site of altered intrahepatic resistance in patients
with parenchymal liver disease is controversial but, in
theory, may occur at any of several levels within the liver.
From a cellular standpoint, the modulation of blood flow
within the liver may be presinusoidal, sinusoidal, or
postsinusoidal. [21] [42] [52] [103] Moreover, cellular, structural,
and humoral elements acting alone or in combination at one
or more sites all
14
may play a role. It is noteworthy that some of these elements

are fixed, whereas others are dynamically regulable.
Examples of fixed components that may alter sinusoidal
blood flow include regenerative nodules and possibly
hepatocyte swelling. [42] Recent work has focused on the
stellate cell (also known as an Ito cell, perisinusoidal cell, or
lipocyte) as a dynamic modulator of sinusoidal blood flow.
This cell has been shown to contract in response to
vasoactive substances and relax in response to dilating
agents. [34] [63] [69] This article highlights new developments in
the pathogenesis of portal hypertension, with emphasis on
the role of hepatic wounding, stellate cells, and
vasoregulatory compounds.
This work was supported by grants from the NIH (DK 02124 and DK
50574).
Don Rockey, MD
Duke University Medical Center
Sands Building, Rm 334
Research Drive, Box 3083
Durham, NC 27710
CELLULAR COMPONENTS OF THE

HEPATIC MICROCIRCULATION
The intrahepatic circulatory system consists of three
microvascular components: (1) the terminal portal venule
(TPV) and hepatic arteriole, (2) the sinusoids, which correspond
to the capillary bed, and (3) the terminal hepatic venule (THV).
[5] [21] [40] [41] [52] [59] Each may modulate liver blood flow. The
major pre-sinusoidal and postsinusoidal components have been

presumed to reflect contraction of vascular smooth muscle cells

in the TPV and THV. Recent evidence suggests that cells within
the preterminal portal venule may play an important role in
local control of sinusoidal blood flow. [32] The major cellular
elements (that might modulate blood flow) residing in the
sinusoidal compartment of the liver include sinusoidal
endothelial cells and stellate cells; both cell types encircle the
sinusoid. Dynamic changes in endothelial fenestrae induced by
neuroamines that correlate with changes in sinusoidal
dimensions imply that sinusoidal endothelial cells could be
involved in blood flow regulation. [59] A prominent feature of
cirrhosis is sinusoidal endothelial capillarization, which could
also contribute to underlying increases in resistance to hepatic
blood flow. [42] The other cellular element within the sinusoid
with the potential to modulate blood flow is the stellate cell.
From an ultrastructural standpoint, this cell possesses long
cytoplasmic processes that envelope sinusoidal endothelial
cells. [97] This anatomic relationship within the sinusoid
suggests that stellate cells may be analogous to pericytes,
smooth muscle-like cells that are thought to control capillary
blood flow in a wide variety of tissues. [86] [91] Data
demonstrating that stellate cells express the muscle-specific
intermediate filament, desmin, assign them to a smooth muscle
lineage and further support this postulate (Fig. 1) . [101]
Morphologic studies have identified unmyelinated nerve fibers
in close proximity to stellate cell processes, indicating that
stellate cells may respond to neural stimuli. [10] Finally, data that
show that stellate cells express the smooth muscle- and
myofibroblast-specific protein, smooth muscle alpha actin,
support the concept that stellate cells are contractile (Fig. 1) .
15
Figure 1. Stellate cell

cytoskeleton. Stellate cells were
isolated as previously described.
[14] After 10 days in culture,
(stellate cells undergo
"activation" with progressive
culture, an event that is recapitulated in vivo during liver injury, see text
for details) cells were fixed and incubated with antidesmin and anti-
smooth muscle alpha actin antibody (linked to Texas Red and fluorescein,
respectively). A fluorescence photomicrograph of a stellate cell expressing
both desmin (A) and smooth muscle alpha actin bundles (B) is shown.
Both of these features implicate stellate cells as smooth muscle-like.
(Original magnification 360.)

HEPATIC INJURY, WOUNDING, AND

STELLATE CELLS
The general response of the liver and other organs to chronic
injury is one of wounding, a complex process ultimately
resulting in fibrosis. Regardless of the type of injury (e.g.,
viruses, toxins, drugs, and hereditary defects in the liver), the
end result is similar. Fibrogenesis in the liver is typified by
fourfold to sixfold increases in collagen (type I > III > IV)
16
as well as increases in proteoglycan content (dermatan sulfate >

chondroitin sulfate > heparan sulfate). [22] [74] [81] A central
feature of the many divergent types of liver injury is the
transformation of resident stellate cells from "quiescent" to
"activated" cells (See "Hepatic Wounding and Stellate Cell
Contractility," below). Characteristics of this transition include
both morphologic and functional changes. Morphologic
changes include loss of vitamin A, acquisition of stress bundles,
and development of prominent rough endoplasmic reticulum.
[16] [36] Importantly, stellate cells from normal livers, which are
cultured in plastic culture dishes, undergo an identical process

that recapitulates the phenomena as it occurs in vivo. Among
known functional changes associated with activation is
increased secretion of extracellular matrix proteins. In rats and
humans with liver injury, type I, III, and IV collagens;
fibronectin; laminin; and proteoglycans have been localized to
stellate cells. [15] [22] [31] Quantitative studies comparing mRNA
for types I, III and IV collagen after liver injury in stellate cells
and hepatocytes have demonstrated a greater than 50-fold
increase in stellate cells. [46] Thus, considerable evidence
supports the concept that activated stellate cells are responsible
for increased extracellular matrix deposition during liver injury.
Common to most forms of epithelial injury, including in the
liver, is the emergence of actin-containing cells termed
myofibroblasts, generally located in proximity to expanded
extracellular matrix. [13] [17] Myofibroblasts appear to be
responsible for the inherent contractility of the cirrhotic liver.
[28] Until recently, a major unresolved question has been the
origin of these cells. Cutaneous myofibroblasts appear to arise

from fibroblasts in healing wound. [17] In the liver, however,
fibroblasts are few in number and seen only in periportal areas.
[67] Myofibroblasts have smooth-muscle features and could
represent cells migrating from the muscular layer of arteries or

veins (the only sites within normal liver which contain smooth
muscle cells). The authors, and others, have recently
documented that during liver injury, stellate cells acquire
smooth muscle alpha actin, [4] [67] a protein characteristic of
smooth muscle cells or myofibroblasts. [37] Thus, activation is
characterized not only by enhanced fibrogenesis, but also by de
novo smooth muscle alpha actin expression (Fig. 1) . These data
indicate that at least one population of myofibroblasts arises
from stellate cells during liver injury. [36] [47] This feature has
functional implications, and served as the basis for studies
designed to directly examine the contractility of stellate cells.

Anatomically, stellate cells are analogous to tissue pericytes, a
cell type oriented in a pericapillary fashion in capillary beds that
appear to regulate blood flow via pericapillary constriction. [85]
[91] [97] Given the resemblance to tissue pericytes and the
expression smooth muscle-specific proteins (Fig. 1) , the

possibility that stellate cells may exhibit a contractile phenotype
has been examined in detail. Experiments in isolated human
stellate cells revealed an increase in intracellular calcium
17
with concomitant cell rounding. [63] Stellate cells induced

wrinkling of silicon membranes and contraction of thick
collagen lattices in response to putative vasoconstrictors (Fig. 2)
. [34] [72] The cell culture studies have been further extended to in
vivo microscopy, which has demonstrated that stellate cells
appear to function as liver-specific pericytes in vivo. [8] [62] [90]
[104] Additionally, Zhang and coworkers [104] colocalized
sinusoidal constriction to stellate cell autofluorescence,

suggesting that stellate cells, rather than endothelial cells, were
responsible for sinusoidal constriction. Collectively, these data
provide strong evidence that stellate cells are capable of
functioning as perisinusoidal contractile elements within the
sinusoid.
A number of vasoactive substances have been reported to
induce stellate cell contraction in various systems, including
angiotensin II, thrombin, substance P, thromboxane A2
analogues, prostaglandin F2alpha ,
Figure 2. Stellate cell contraction of collagen lattices.

Stellate cells were isolated and cultured for 5 days on

thick collagen lattices (so as to allow culture induced

activation). After introduction of serum free
conditions, serum containing medium (20% serum)
was added and lattices were detached. Hepatocytes,
cultured under similar conditions serve as a negative
control. Contraction is measured as the reduction in
initial lattice area over time. Stellate cells contract collagen lattices
rapidly, as emphasized in the inset.
18
and the endothelins (ETs) (Table 1) . [34] [35] [63] [72] [77]
Importantly, of the various vasoactive compounds identified,
the endothelins appear to be the most potent and reliable
inducers of stellate cell contraction. Thus, endothelin (ET) and
its biology are reviewed in the following section.
ENDOTHELIN AND ENDOTHELIN

RECEPTORS
Endothelin
The ETs comprise a family of potent vasoconstrictors. [76] [100]

Three unique ET peptides, each consisting of 21 amino acids,
have been identified, and termed ET-1, ET-2, and ET-3. [76] [100]
They bind to at least two different G-coupled protein receptors,
termed ET A (ETA ) and ET B (ETB ) receptors. ETs are
produced by endothelial cells, and exert paracrine effects on
adjacent smooth muscle cells; they may also have autocrine or
intracrine effects. The only type of ET that liver endothelial
cells have been shown to produce is ET-1. [76]
The ETs have been implicated in a wide variety of normal and
pathological processes. Their major function appears to be
related to the local control of vascular tone, including regulation
of basal blood pressure. [76] [100] They appear to induce
vasoconstriction via stimulation of smooth muscle cells and
vasodilation due to effects on endothelial cells. [11] [12] Although
the vasoregulatory functions of the ETs have been emphasized,
recent reports have documented promitogenic effects. [6]
Additionally, recent production of ET-1- and ET-3-deficient
mice indicate that the ETs are important regulators of
mammalian neural crest development. [9] [39]
Circulating levels of immunoreactive ET-1 have been reported
to be elevated in a number of diseases, including hypertension,
atherosclerosis, acute myocardial infarction, pulmonary fibrosis,
acute renal failure, and chronic liver disease [9] [19] [20] [39] [44] [54]
; however, the importance of elevated plasma

[57] [89]
immunoreactive ET-1 in pathologic states must be interpreted

with caution. Circulating plasma levels of ET-1 (even when
TABLE 1 -- AGENTS THAT MODULATE STELLATE
CELL CONTRACTILITY
Contract Relax
Endothelin (1, 2, 3) [35] [63] [71] Nitric oxide [68]
Angiotensin II [63] Carbon monoxide
Thrombin [63] [89]
Prostaglandin F2alpha [34]

U46619 (thromboxane A2 ) [34]
Substance P [76]
Serum [71]
These agents have been shown to directly or indirectly, in one
or more systems, cause stellate cells to contract or to relax.
19
elevated) are several orders of magnitude lower than

concentrations required to induce biologic effects of ET-1. ET-1
appears to be rapidly cleared from the circulation, implicating
little role for circulating ETs. Like other endothelium-derived
vasoactive substances (e.g., prostaglandin G2 , nitric oxide),
ETs appear to act as local paracrine or autocrine factors. Indeed,
the target cell types (i.e., those expressing ET receptors) are
found within close vicinity of cells that produce endothelins.
ET Regulation
The production of ET-1 appears to be regulated, at least in part,

at the level of mRNA transcription. [50] The expression of prepro
ET-1 is stimulated by vasopressor hormones, such as
epinephrine, angiotensin II, and vasopressin, shear stress, and
cytokines, such as transforming growth factor beta and
interleukin-1. [48] [50] [102] In vascular endothelial cells, once
ET-1 is produced, it appears to be secreted in a constitutive
manner, without regulation at the level of exocytosis. [100]
The ET peptides arise by proteolytic processing of large
precursors ( 200 amino acid residues). Intermediates termed
big ET-1, -2, and -3 (38-41 amino acids) are excised from
prepropeptides by proteases that cleave at sites containing

paired basic amino acids. Big ETs, which have little or no

biologic activity, [100] are finally cleaved at Trp-21-Val/Ile-22 to
produce mature 21-residue peptides. The putative enzyme
responsible for the specific cleavage at Trp-21 has been termed
ET-converting enzyme (ECE), and has been identified as a
neutral membrane-bound metalloprotease with Mr = 120 kD. [61]
Recently, an ECE that cleaves big ET-1 has been cloned and
termed ECE-1. [84] [99] Of note, this peptidase does not appear to
have activity for big ET-3. Analysis of the cDNA demonstrated
structural and sequence homology to neutral endopeptidase
24.11. By northern analysis, ECE has been identified in lung,
adrenal cortex, adrenal medulla, ovary, and testis. Although
northern analysis has demonstrated little or no ECE mRNA in
liver, one study detected this mRNA in a perisinusoidal location
by in situ hybridization, [99] consistent with its expression in
endothelium. The precise localization of ECE is an important
issue because it may determine the cellular target of mature
ET-1.
ET Receptors
Since the discovery of ETs, identification of ET receptors has

been an active area of study. Two major subtypes of G
protein-coupled receptors, ETA and ETB , have been definitively
identified, cloned, and shown to mediate a range of biologic
effects. [2] [79] ETA receptors are found predominantly on
vascular smooth muscle cells and are preferentially activated by
ET-1. Rank order affinities are ET-1 > ET-2 >>> ET-3; the
affinity of ET-1 for the ETA receptor is more than 100-fold that
of ET-3. [78]
20
ETB receptors are widely distributed and have equal affinity for
ET-1, ET-2, and ET-3. ETB receptor stimulation appears to
bring about divergent responses, depending on the cell type
expressing the receptor. Stimulation of ETB receptors on
endothelial cells results in nitric oxide (NO) release and
relaxation of vascular smooth muscle [75] ; however, we have
clearly shown stellate cell contraction after stimulation of ETB ,
as have others studying systemic vasoconstriction mediated by
ETB receptors on vascular smooth muscle cells. [66] [82] It has
been proposed that endothelium-dependent relaxation and
smooth muscle vasoconstriction mediated by the ETB receptor
are brought about by two different ETB receptors, termed ETB1

and ETB2 , respectively. [12] Whether smooth muscle cells (or

stellate cells) possess both ETB1 and ETB2 receptors is
unknown. Other ET receptors are emerging, with recent studies
having demonstrated unique receptors in Xenopus dermal
melanophores and Xenopus heart, termed ETC and ETAX ,
respectively. [33] [38] Additionally, a "super-high affinity"
receptor related to ETB has been proposed based on radioligand
binding studies. [87] Precise characterization of stellate cell
receptors mediating biologic responses is essential to the
implementation of therapies based on ET antagonists.
ET receptor antagonists have been the center of intense
investigation because of their potential for therapeutic
application in a wide variety of diseases, including hypertension
and coronary and cerebral vascular diseases. [11] [45] [60] [40] [41] [67]
A vast literature exists emphasizing that the antagonists have
variable biologic effects depending on the target tissue. [45]
Numerous ETA receptor, selective antagonists have been
described, the most widely studied of which is BQ-123. [27]
Several recent ETB antagonists have been described, [29] [94] but
their effects have not been widely characterized, and, in one
instance, not reproduced. [94] Recent studies have focused on a
number of mixed receptor antagonists. Interest in ET
antagonists is substantial because these agents are likely to
provide effective therapies for disease processes characterized
by abnormal ET homeostasis.
Studies of ET receptors on stellate cells indicate the presence of

both ETA and ETB receptors. Specific saturation-binding
experiments of 125 I-ET-1 to rat stellate cells, endothelial cells,
and hepatocytes in primary cultures demonstrated that the
number of receptors on stellate cells outnumbered that for
hepatocytes and endothelial cells by 83 fold and 31 fold,
respectively. [26] [88] Competitive binding studies further
demonstrated that stellate cells expressed both ETA and ETB
receptors. The presence of these receptors was confirmed by
detection of specific ETA and ETB receptor mRNAs.
Furthermore, these mRNAs were unchanged with activation,
suggesting that the effect of ET on stellate cells is regulated at
the level of ligand rather than the receptor.
From a functional standpoint, both ETA and ETB receptors
appear to mediate stellate cell contraction. [66] The ETB receptor
appears to be

21
involved additionally in regulating growth of human

myofibroblast-like stellate cells. [49] Data with regard to
modulation of receptor subtype are conflicting. Some studies
have failed to find changes in receptor expression during liver
injury or in culture, [26] whereas others report that with repeated
passage of cultured cells the relative abundance of ETA
receptors on stellate cells decreases markedly. [64] Because the
effect of various ETs on stellate cells is in theory dependent on
receptor subtype, further work is necessary to elucidate how
receptors are modulated.
NITRIC OXIDE AND CARBON MONOXIDE

Nitric Oxide
Stellate cell contraction within the hepatic sinusoid is likely to

be balanced by dilating compounds (Table 1) . Candidate
effectors, such as NO, a recently identified vasodilating
substance, are attractive. The half-life of NO is short, but
products of NO's interaction with reactive oxygen intermediates
(i.e., NO plus superoxide yields peroxynitrite) results in
products with longer half-lives, which may also exert important
local effects. NO is produced by one of three isoforms of NO
synthase (NOS). [56] Endothelial cells and neurons produce NO
from L-arginine via two distinct constitutive NOS enzymes
while a wide variety of cells express inducible NOS (iNOS). [56]
The constitutive isoforms typically produce small amounts of
NO and are regulated by calmodulin binding in the context of
changes in intracellular calcium. In contrast, iNOS is bound
tightly to calmodulin, and therefore functions in large part
independently of intracellular calcium; it produces large
amounts of NO. [56]
The amount and duration of NO production in a biologic system
depends on its enzymatic source. Although the term constitutive
implies that these isoforms are not regulated, their mRNA and
protein levels may be altered by changes in the environment,
such as hypoxia, cytokines, or stretch. [83] The inducible isoform
is stimulated by a wide array of compounds/stimuli, including
cytokines and lipopolysaccharide (LPS), which have been
emphasized as prominent inducers of iNOS. Cytokines and LPS
appear to stimulate iNOS gene transcription via autoregulation
of the NFkappa-B/Ikappa-Balpha complex. [98] iNOS regulation

is particularly complex, as recent studies indicate that iNOS

enzymatic activity is itself inhibited by NO. [96]
NO may play a role in the modulation of sinusoidal blood flow
and in the pathogenesis of portal hypertension. The NO
inhibitor, NNA ( Nw -nitro- L-arginine) increased portal pressure
in normal rat livers whereas another NO inhibitor, L-NAME (
Nw -nitro- L-arginine methyl ester), had no effects. [53] [90]
Although NO levels in the normal liver appear to be quite low
(implicating a minor role for NO), perfusion of cirrhotic livers
with L-arginine reduced portal reactivity to norepinephrine,
suggesting a defect in endogenous NO production in cirrhosis.
[23] Indeed, in some
22
forms of experimental liver injury and portal hypertension

iNOS is not induced, consistent with the possibility that a lack
of NO may be important in the perpetuation of intrahepatic
portal hypertension. [71] Further, recent data indicate that
sinusoidal endothelial cells from cirrhotic livers have an
impairment in the ability to produce NO via endothelial NOS.
[70] Thus, a relative decrease in NO may contribute to enhanced
stellate cell contractility and portal hypertension.

At a cellular level, effects of NO on stellate cells are
pronounced. Current data indicate that exogenous NO is
capable of preventing ET-induced contraction, as well as
inducing relaxation in precontracted cells. [35] [69] Additionally,
stellate cells themselves produce NO in response to
interferon-gamma and other cytokines with or without LPS, [24]
[69] and NO produced in an autocrine fashion has prominent
relaxing effects on stellate cells. [69] Essentially all liver cell

types are capable of producing NO (via inducible or constitutive
isoforms of NOS). Therefore, paracrine NO may also serve as
an important mediator of stellate cell relaxation.
Carbon Monoxide
Although NO has received a great deal of attention as a

mediator of stellate cell relaxation, Suematsu and coworkers [90]
have recently raised the possibility that locally produced CO
might also control stellate cell contractility, and thus sinusoidal
blood flow. These authors demonstrated that inhibition of CO
production (CO is produced by the oxidation of heme-by-heme
oxygenase) caused an increase in portal vascular resistance.
These data are provocative and identify another potential

modulator of stellate cell contractility.
HEPATIC WOUNDING AND STELLATE

CELL CONTRACTILITY
During the hepatic wounding response stellate cells undergo a
transition from a "quiescent" to an "activated" state (Fig. 3)
(Figure Not Available) . One of the most prominent features of
this process is the de novo acquisition of the smooth muscle
protein, smooth muscle alpha actin (Figure 3) (Figure Not
Available) . In culture, this phenomenon coincides with the
acquisition of the contractile phenotype, suggesting that smooth
muscle alpha actin confers on stellate cells the ability to
contract. In culture models, quiescent stellate cells are not
contractile, yet activated cells display enhanced contractility. [34]
[72] Such data imply that contractility may be directly related to
the activation state of stellate cells in vivo. This possibility has

been tested in two in vivo models of experimental liver injury.
[7] [73] In one, freshly isolated stellate cells isolated from livers
with advanced liver injury displayed significantly greater

contractility than those isolated from normal livers. [73]
Additionally, intravital microscopy indicated enhanced in situ
contractility of stellate cells after activation induced by chronic
ethanol consumption. [7]
23
Figure 3. (Figure Not Available) Stellate Cell Activation and

Contractility. In normal liver, stellate cells retain "quiescent" features
which include a vitamin A rich state, relatively minor collagen synthesis
and a low level of proliferation. After liver injury or during culture on
uncoated plastic, stellate cells undergo "activation", a programmed
cascade of events. The in vivo phenomenon of stellate cell activation
appears to be closely recapitulated by culture of stellate cells on standard
uncoated plastic substrata. In addition to proliferation and enhanced
extracellular matrix production, a major feature of activation, which is
emphasized, is new expression of smooth muscle alpha actin. The
acquisition of this smooth muscle protein implies a contractile phenotype
(see text for details). Of the proposed modulators of stellate cell
contractility, endothelin (ET) and nitric oxide (NO) appear to be the most
prominent. (Courtesy of Lydia V. Kibiuk)
Available data are consistent with stellate cell-mediated
alterations in sinusoidal blood flow via perisinusoidal
constriction, particularly in the injured liver; however, it is
noteworthy that myofibroblasts (i.e., activated stellate cells)
populate large fibrous bands (consisting primarily of type I
collagen) in the injured and cirrhotic liver. [4] Moreover, ET-1,
when perfused into the liver of rats with advanced fibrosis,

binds not only to cells in perisinusoidal spaces (consistent with

stellate cells) but also to cells within fibrous bands. [73] Stellate
cells express abundant quantities of the collagen-binding
integrins alpha1beta1 and alpha2beta1 and these integrins
readily mediate stellate cell contraction of type I collagen. [65]
Such information raises the possibility that stellate cells are
capable of contraction of large collagenous bands (i.e., in
response to ET), causing alteration in lobular architecture and
subsequent derangement of hepatic blood flow. This concept is
consistent with previous data demonstrating inherent
contractility of the cirrhotic liver. [28] Thus, stellate cell
contractility may contribute to intrahepatic portal hypertension
via perisinusoidal constriction, macroscopic effects, or both.
24
TABLE 2 -- CIRCULATING ET IN PATIENTS WITH

CIRRHOSIS
Study Author (yr) Type of ET Comment
Uchihara (1992) ET-1 Ascites > w/o
Moore (1992) ET-1/ET-3 ascites
Moller (1993) ET-1 HRS >
Isobe (1993) ET-1 cirrhosis only
Asbert (1993) ET-1 Proportional
Matsumoto (1994) ET-1 to Creatinine
Gerbes (1995) ET-1/ET-3 Varices > w/o
varices
Ascites > w/o
ascites
Cirrhosis >
hepatitis
ET-3
proportional to
Child-Turcotte
HRS = Hepatorenal syndrome.
ET IN LIVER INJURY
Multiple studies indicate that circulating ET levels are elevated
in patients with cirrhosis (Table 2) . * Because ET acts as a
paracrine (or autocrine) factor, and circulating ET is likely to

represent local overproduction of ET, the possibility that ET

was overproduced in the injured liver has been investigated. [25]
[43] [64] PreproET-1 mRNA has not been detectable in normal
liver but is abundant in the livers of animals with several types

of experimental liver injury. Further, intrahepatic levels of ET-1
peptide appear to correlate with the severity of disease and the
amount of ascites. [1] ET-1 appears to be localized to
perisinusoidal cells, consistent with production by endothelial
or stellate cells. [25] Additionally, during culture-induced
activation, mRNA for preproET-1 increased in stellate cells, [26]
suggesting an autocrine loop for its actions. Thus, ET produced
in an autocrine fashion may have effects on stellate cell
contractility. The finding that a novel mixed ETA /ETB
antagonist, bosentan, reduced portal pressure in rats with portal
hypertension further supports this postulate. [73]
Because stellate cells appear to be a major target of ET in liver,
and possess a marked fibrogenic capacity when activated,
recent investigation has focused on the role of ETs and stellate
cell activation. [49] [68] In cultured stellate cells, both ET-1 and
sarafotoxin S6C, a potent ETB receptor agonist, stimulated
expression of smooth muscle alpha actin, a classic feature of
activation (see Figs. 1 and 3) (Figure Not Available) . [68] [72] In
an in vivo model of liver injury, antagonism of the effects of ET
by blocking ETA /ETB receptors reduced stellate cell activation
and fibrogenesis. [68] These data indicate that not only is ET
important in mediating stellate cell contraction, but it is also an
important component of stellate cell activation.
SUMMARY
The pathogenesis of portal hypertension is multifactorial, and
appears to result from interplay between fixed and dynamically
modulable
* References [3] [18] [30] [51] [54] [55] [57] [58] [80] [92] [93] [95] .
25
elements; the stellate cell is a newly recognized example of the

latter. This perisinusoidal, pericyte-like cell has contractile
features that are most prominent after liver injury, concomitant
with their activation. These data imply an exaggerated
contractile phenotype in the cirrhotic liver. This cell may
contribute to increased intrahepatic portal hypertension via
perisinusoidal constriction of the sinusoid or by contraction of
fibrous extracellular matrix rich in type I collagen with
concomitant disruption of lobular architecture. Endothelins and

NO play a major role in the modulation of stellate cell
contractility, and are therefore important in the pathogenesis of
intrahepatic portal hypertension. These new data provide
potential areas for therapeutic intervention in this clinical entity.
References
1. Alam I, Bass NM, Gee L, et al: Hepatic endothelin-levels in chronic

liver disease correlate with disease severity and ascites. Hepatology
24:142a, 1996
2. Arai
H, Hori S, Aramori I, et al: Cloning and expression of a cDNA
encoding an endothelin receptor. Nature 348:730-732, 1990
3. Asbert M, Gines A, Gines P, et al: Circulating levels of endothelin in

cirrhosis. Gastroenterology 104:1485-91, 1993
4. BallardiniG, Fallani M, Biagini G, et al: Desmin and actin in the

identification of Ito cells and in monitoring their evolution to
myofibroblasts in experimental liver fibrosis. Virchows Arch [Cell Pathol]
56:45-49, 1988
5. Bass NM, Manning JA, Weisiger RA: Increased sinusoidal volume and
solute extraction during retrograde liver perfusion. Am J Physiol
256:G1041-G1048, 1989
6. Battistini
B, Chailler P, D'Orleans-Juste P, et al: Growth regulatory
properties of endothelins. Peptides 14:385-399, 1993
7. BauerM, Paquette NC, Zhang JX, et al: Chronic ethanol consumption

increases hepatic sinusoidal contractile response to endothelin-1 in the rat.
Hepatology 22:1565-76, 1995
8. BauerM, Zhang JX, Bauer I, et al: ET-1 induced alterations of hepatic

microcirculation: Sinusoidal and extrasinusoidal sites of action. Am J
Physiol 267:G143-G149, 1994
9. Baynash AG, Hosoda K, Giaid A, et al: Interaction of endothelin-3 with

endothelin-B receptor is essential for development of epidermal
melanocytes and enteric neurons. Cell 79:1277-1285, 1994
10. Bioulac-Sage P, Lafon ME, Saric J, et al: Nerves and perisinusoidal

cells in human liver. J Hepatol 10:105-112, 1990
11. ClozelM, Breu V, Burri K, et al: Pathophysiological role of endothelin

revealed by the first orally active endothelin receptor antagonist. Nature
365:759-761, 1993
12. Clozel
M, Gray GA, Breu V, et al: The endothelin ETB receptor
mediates both vasodilation and vasoconstriction in vivo. Biochem Biophys
Res Commun 186:867-873, 1992
13. Darby I, Skalli O, Gabbiani G: Alpha-smooth muscle actin is transiently

expressed by myofibroblasts during experimental wound healing. Lab
Invest 63:21-29, 1990
14. de Leeuw AM, McCarthy SP, Geerts A, et al: Purified rat liver
fat-storing cells in culture divide and contain collagen. Hepatology
4:392-403, 1984
15. Friedman SL: Cellular sources of collagen and regulation of collagen

production in liver. Semin Liver Dis 10:20-29, 1990
16. Friedman SL: Seminars in medicine of the Beth Israel Hospital, Boston:
The cellular basis of hepatic fibrosis: Mechanisms and treatment
strategies. N Engl J Med 328:1828-1835, 1993
17. Gabbiani G, Ryan GB, Majne G: Presence of modified fibroblasts in

granulation tissue and their possible role in wound contraction. Experientia
27:549-550, 1971
26
18. GerbesAL, Moller S, Gulberg V, et al: Endothelin-1 and -3 plasma

concentrations in patients with cirrhosis: Role of splanchnic and renal
passage and liver function. Hepatology 21:735-739, 1995
19. GiaidA, Michel RP, Stewart DJ, et al: Expression of endothelin-1 in

lungs of patients with cryptogenic fibrosing alveolitis. Lancet
341:1550-1554, 1993
20. Giaid
A, Saleh D, Yanagisawa M, et al: Endothelin-1 immunoreactivity
and mRNA in the transplanted human heart. Transplantation
59:1308-1313, 1995
21. Greenway CV, Lautt WW: Distensibility of hepatic venous resistance

sites and consequences on portal pressure. Am J Physiol 254:H452-H458,
1988
22. GressnerAM, Bachem MG: Cellular sources of noncollagenous matrix

proteins: Role of fat-storing cells in fibrogenesis. Semin Liver Dis
10:30-46, 1990
23. GuptaTK, Groszman RJ: Administration of L-arginine, the

physiological precursor of nitric oxide, reduces portal perfusion pressure
and ameliorates hepatic hyperreactivity in experimental cirrhosis.
24. Helyar L, Bundschuh DS, Laskin JD, et al: Induction of hepatic Ito cell
nitric oxide production after acute endotoxemia. Hepatology
20:1509-1515, 1994

25. HoussetC, Laudat A, Chung JJ, et al: Expression and potential role of
endothelin in cholestatic liver injury. Hepatology 22:278, 1995
26. HoussetCN, Rockey DC, Friedman SL, et al: Hepatic lipocytes: A

major target for endothelin-1. J Hepatol 22:55-60, 1995
27. Ihara
M, Ishikawa K, Fukuroda T, et al: In vitro biological profile of a
highly potent novel endothelin (ET) antagonist BQ-123 selective for the
ETA receptor. J Cardiovasc Pharmacol 20(suppl)12:S11-S14, 1992
28. Irle
C, Kocher O, Gabbiani G: Contractility of myofibroblasts during
experimental liver cirrhosis. Journal of Submicroscopic Cytology
12:209-217, 1980
29. Ishikawa K, Ihara M, Noguchi K, et al: Biochemical and

pharmacological profile of a potent and selective endothelin B-receptor
antagonist, BQ-788. Proc Natl Acad Sci U.S.A. 91:4892-4896, 1994
30. IsobeH, Satoh M, Sakai H, et al: Increased plasma endothelin-1 levels

in patients with cirrhosis and esophageal varices. J Clin Gastroenterol
17:227-230, 1993
31. JarnaginWR, Rockey DC, Koteliansky VE, et al: Expression of variant

fibronectins in wound healing: Cellular source and biological activity of
the EIIIA segment in rat hepatic fibrogenesis. J Cell Biol 127:2037-2048,
1994
32. Kaneda K, Ekataksin W, Sogawa M, et al: Endothelin-1-induced

vasoconstriction localized in the distal segment of preterminal portal
venule causes significant increase in portal pressure of rat liver.
Hepatology 24:133a, 1996
33. KarneS, Jayawickreme CK, Lerner MR: Cloning and characterization

of an endothelin-3 specific receptor (ETC receptor) from Xenopus laevis
dermal melanophores. J Biol Chem 268:19126-19133, 1993
34. Kawada N, Klein H, Decker K: Eicosanoid-mediated contractility of

hepatic stellate cells. Biochem J 285:367-371, 1992
35. Kawada N, Tran-Thi TA, Klein H, et al: The contraction of hepatic

stellate (Ito) cells stimulated with vasoactive substances: Possible
involvement of endothelin 1 and nitric oxide in the regulation of the
sinusoidal tonus. Eur J Biochem 213:815-823, 1993
36. KentG, Gay S, Inouye T, et al: Vitamin A-containing lipocytes and

formation of type III collagen in liver injury. Proc Natl Acad Sci U.S.A.
73:3719-3722, 1976
37. Kocher O, Skalli O, Bloom WS: Cytoskeleton of rat aortic smooth

muscle cells: Normal conditions and experimental intimal thickening. Lab
Invest 50:645-652, 1984
38. Kumar C, Mwangi V, Nuthulaganti P, et al: Cloning and

characterization of a novel endothelin receptor from Xenopus heart. J Biol

Chem 269:13414-13420, 1994
39. Kurihara Y, Kurihara H, Suzuki H, et al: Elevated blood pressure and

craniofacial abnormalities in mice deficient in endothelin-1. Nature
368:703-710, 1994
40. Lautt
WW, Greenway CV: Conceptual review of the hepatic vascular
bed. Hepatology 7:952-963, 1987
41. Lautt
WW, Greenway CV, Legare DJ: Effect of hepatic nerves,
norepinephrine, angiotensin, and elevated central venous pressure on
postsinusoidal resistance sites and intrahepatic pressures in cats.
Microvasc Res 33:50-61, 1987
27
42. LauttWW, Greenway CV, Legare DJ, et al: Localization of intrahepatic

portal vascular resistance. Am J Physiol 251:G375-G381, 1986
43. LeivasA, Jimenez W, Lamas S, et al: Endothelin 1 does not play a

major role in the homeostasis of arterial pressure in cirrhotic rats with
ascites. Gastroenterology 108:1842-1848, 1995
44. Lerman A, Edwards BS, Hallett JW, et al: Circulating and tissue
endothelin immunoreactivity in advanced atherosclerosis. N Engl J Med
325:997-1001, 1991
45. Luscher
TF: Do we need endothelin antagonists? Cardiovasc Res
27:2089-2093, 1993
46. Maher JJ, McGuire RF: Extracellular matrix gene expression increases
preferentially in rat lipocytes and sinusoidal endothelial cells during
hepatic fibrosis in vivo. J Clin Invest 86:1641-1648, 1990
47. MakKM, Leo MA, Lieber CS: Alcoholic liver injury in baboons:
Transformation of lipocytes to transitional cells. Gastroenterology
87:188-200, 1984
48. Malek AM, Greene AL, Izumo S: Regulation of endothelin 1 gene by

fluid shear stress is transcriptionally mediated and independent of protein
kinase C and cAMP. Proc Natl Acad Sci U.S.A. 90:5999-6003, 1993
49. Mallat
A, Fouassier L, Preaux AM, et al: Growth inhibitory properties
of endothelin-1 in human hepatic myofibroblastic Ito cells: An endothelin
B receptor-mediated pathway. J Clin Invest 96:42-49, 1995
50. Marsden PA, Brenner BM: Transcriptional regulation of the

endothelin-1 gene by TNF-alpha. Am J Physiol 262:C854-C861, 1992
51. MatsumotoH, Uemasu J, Kitano M, et al: Clinical significance of

plasma endothelin-1 in patients with chronic liver disease. Dig Dis Sci
39:2665-70, 1994
52. McCuskey RS: A dynamic and static study of hepatic arterioles and
hepatic sphincters. American Journal of Anatomy 119:455-478, 1966
53. MittalMK, Gupta TK, Lee FY, et al: Nitric oxide modulates hepatic
vascular tone in normal rat liver. Am J Physiol 267:G416-G422, 1994
54. MollerS, Emmeluth C, Henriksen JH: Elevated circulating plasma

endothelin-1 concentrations in cirrhosis. J Hepatol 19:285-290, 1993
55. MollerS, Gulberg V, Henriksen JH, et al: Endothelin-1 and

endothelin-3 in cirrhosis: Relations to systemic and splanchnic
haemodynamics. J Hepatol 23:135-144, 1995
56. Moncada
S, Higgs A: The L-arginine-nitric oxide pathway. N Engl J
Med 329:2002-2012, 1993
57. Moore
K, Wendon J, Frazer M, et al: Plasma endothelin
immunoreactivity in liver disease and the hepatorenal syndrome [see
comments]. N Engl J Med 327:1774-1778, 1992
58. Nozue T, Kobayashi A, Uemasu F, et al: Plasma endothelin-1 levels of

children with cirrhosis. J Pediatr Gastroenterol Nutr 21:220-223, 1995
59. OdaM, Azuma T, Watanabe N: Regulatory mechanism of hepatic

microcirculation: Involvement of the contraction and dilation of sinusoids
and sinusoidal fenestrae. Progress in Applied Microcirculation
127:103-128, 1990
60. OhlsteinEH, Beck GRJ, Douglas SA, et al: Nonpeptide endothelin

receptor antagonists. II: Pharmacological characterization of SB 209670. J
Pharmacol Exp Ther 271:762-768, 1994
61. Ohnaka K, Takayanagi R, Nishikawa M, et al: Purification and

characterization of a phosphoramidon-sensitive endothelin-converting
enzyme in porcine aortic endothelium. J Biol Chem 268:26759-26766,
1993
62. Okumura S, Takei Y, Kawano S, et al: Vasoactive effect of

endothelin-1 on rat liver in vivo. Hepatology 19:155-161, 1994
63. Pinzani M, Failli P, Ruocco C, et al: Fat-storing cells as liver-specific

pericytes: Spatial dynamics of agonist-stimulated intracellular calcium
transients. J Clin Invest 90:642-646, 1992
64. Pinzani M, Milani S, De Franco R, et al: Endothelin 1 is overexpressed

in human cirrhotic liver and exerts multiple effects on activated hepatic
stellate cells. Gastroenterology 110:534-48, 1996
65. Racine-Samson L, Rockey DC, Bissell DM: Expression of the

collagen-binding integrins alpha1beta1 and alpha2beta1 during activation
of rat hepatic stellate cells in vivo. Hepatology 24:133a, 1996

28
66. Rockey DC: Characterization of endothelin receptors mediating rat

hepatic stellate cell contraction. Biochem Biophys Res Commun
207:725-731, 1995
67. RockeyDC, Boyles JK, Gabbiani G, et al: Rat hepatic lipocytes express
smooth muscle actin upon activation in vivo and in culture. J Submicrosc
Cytol Pathol 24:193-203, 1992
68. Rockey DC, Chung JJ: Endothelin antagonism in experimental hepatic

fibrosis: Implications for endothelin in the pathogenesis of wound healing.
J Clin Invest 98:1381-1388, 1996
69. Rockey DC, Chung JJ: Inducible nitric oxide synthase in rat hepatic
lipocytes and the effect of nitric oxide on lipocyte contractility. J Clin
Invest 95:1199-206, 1995
70. Rockey DC, Chung JJ: Modulation of nitric oxide synthase III in
sinusoidal endothelial cells: Implications for the pathogensis of portal
hypertension. Hepatology 24:190a, 1996
71. Rockey DC, Chung JJ: Regulation of inducible nitric oxide synthase
during liver injury. Hepatology 22:249, 1995
72. Rockey DC, Housset CN, Friedman SL: Activation-dependent

contractility of rat hepatic lipocytes in culture and in vivo. J Clin Invest
92:1795-1804, 1993
73. Rockey DC, Weisiger R: Endothelin induced contractility of stellate

cells from normal and cirrhotic rat liver: Implications for regulation of
portal pressure and resistance. Hepatology 24:233-240, 1996
74. Rojkind M, Giambrone MA, Biempica L: Collagen types in normal and

cirrhotic liver. Gastroenterology 76:710-719, 1979
75. RubanyiGM: The role of endothelium in cardiovascular homeostasis

and diseases. J Cardiovasc Pharmacol 22(suppl 4):S1-14, 1993
76. Rubanyi GM, Botelho LH: Endothelins. FASEB J 5:2713-2720, 1991
77. Sakamoto M, Ueno T, Kin M, et al: Ito cell contraction in response to

endothelin-1 and substance P. Hepatology 18:978-983, 1993
78. SakuraiT, Yanagisawa M, Masaki T: Molecular characterization of

endothelin receptors. Trends Pharmacol Sci 13:103-108, 1992
79. Sakurai
T, Yanagisawa M, Takuwa Y, et al: Cloning of a cDNA
encoding a non-isopeptide-selective subtype of the endothelin receptor.
Nature 348:732-735, 1990

80. SaloJ, Francitorra A, Follo A, et al: Increased plasma endothelin in

cirrhosis: Relationship with systemic endotoxemia and response to
changes in effective blood volume. J Hepatol 22:389-98, 1995
81. Schuppan D: Structure of the extracellular matrix in normal and fibrotic

liver: Collagens and glycoproteins. Semin Liver Dis 10:1-10, 1990
82. Seo
B, Oemar BS, Siebenmann R, et al: Both ETA and ETB receptors
mediate contraction to endothelin-1 in human blood vessels. Circulation
89:1203-1208, 1994
83. Sessa
WC: The nitric oxide synthase family of proteins. J Vasc Res
31:131-143, 1994
84. Shimada K, Takahashi M, Tanzawa K: Cloning and functional

expression of endothelin-converting enzyme from rat endothelial cells. J
Biol Chem 269:18275-18278, 1994
85. Sims DE: The pericyte: A review. Tissue Cell 18:153-174, 1986
86. SimsDE: Recent advances in pericyte biology: Implications for health

and disease. Can J Cardiol 7:431-443, 1991
87. Sokolovsky M, Ambar I, Galron R: A novel subtype of endothelin

receptors. J Biol Chem 267:20551-20554, 1992
88. Stephenson K, Harvey SA, Mustafa SB, et al: Endothelin association

with the cultured rat Kupffer cell: Characterization and regulation.
Hepatology 22:896-905, 1995
89. Stewart DJ, Kubac G, Costello KB, et al: Increased plasma endothelin-1
in the early hours of acute myocardial infarction. J Am Coll 18:38-43,
1991
90. SuematsuM, Goda N, Sano T, et al: Carbon monoxide: An endogenous

modulator of sinusoidal tone in the perfused rat liver. J Clin Invest
96:2431-2437, 1995
91. TiltonRG, Kilo C, Williamson JR, et al: Differences in pericyte

contractile function in rat cardiac and skeletal muscle microvasculatures.
Microvasc Res 18:336-352, 1979
92. UchiharaM, Izumi N, Sato C, et al: Clinical significance of elevated

plasma endothelin concentration in patients with cirrhosis. Hepatology
16:95-9, 1992
29
93. Uemasu J, Matsumoto H, Kawasaki H: Increased plasma endothelin

levels in patients with liver cirrhosis. Nephron 60:380, 1992
94. UradeY, Fujitani Y, Oda K, et al: An endothelin B receptor-selective

antagonist: IRL 1038, [Cys11-Cys15]-endothelin-1(11-21). FEBS Lett
311:12-16, 1992
95. Veglio F, Pinna G, Melchio R, et al: Plasma endothelin levels in

cirrhotic subjects. J Hepatol 15:85-7, 1992
96. Vodovotz Y, Kwon NS, Pospischil M, et al: Inactivation of nitric oxide

synthase after prolonged incubation of mouse macrophages with
IFN-gamma and bacterial lipopolysaccharide. J Immunol 152:4110-4118,
1994
97. Wake K: Perisinusoidal stellate cells (fat-storing cells, interstitial cells,

lipocytes), their related structure in and around the liver sinusoids, and
vitamin A-storing cells in extrahepatic organs. Int Rev Cytol 66:303-353,
1980
98. Xie
QW, Kashiwabara Y, Nathan C: Role of transcription factor
NF-kappa B/Rel in induction of nitric oxide synthase. J Biol Chem
269:4705-4708, 1994
99. XuD, Emoto N, Giaid A, et al: ECE-1: A membrane-bound

metalloprotease that catalyzes the proteolytic activation of big
endothelin-1. Cell 78:473-485, 1994
100. Yanagisawa M: The endothelin system: A new target for therapeutic

intervention. Circulation 89:1320-1322, 1994
101. YokoiY, Namihisa T, Kuroda H, et al: Immunocytochemical detection

of desmin in fat-storing cells (Ito cells). Hepatology 4:709-714, 1984
102. Yoshizumi M, Kurihara H, Morita T, et al: Interleukin 1 increases the

production of endothelin-1 by cultured endothelial cells. Biochem Biophys
Res Commun 166:324-329, 1990
103. Zhang JX, Bauer M, Clemens MG: Vessel- and target cell-specific
actions of endothelin-1 and endothelin-3 in rat liver. Am J Physiol
269:G269-G277, 1995
104. Zhang JX, Pegoli WJ, Clemens MG: Endothelin-1 induces direct
constriction of hepatic sinusoids. Am J Physiol 266:G624-G632, 1994



13
PORTAL HYPERTENSION
NEW CONCEPTS IN THE PATHOGENESIS OF PORTAL

HYPERTENSION
Hepatic Wounding and Stellate Cell Contractility
Don C. Rockey MD
From the Gastroenterology Division, Duke University Medical Center, Durham, North Carolina
Complications resulting from portal hypertension encompass some of the most serious clinical
sequelae confronting clinicians caring for patients with chronic liver disease. A number of
postulates attempt to explain the pathogenesis of portal hypertension. In the "backward flow"
theory, increased intrahepatic resistance leads to elevated portal pressure; the "forward flow"
hypothesis suggests that humoral factors (directly or indirectly) contribute to increased portal
venous return and flow, ultimately resulting in portal hypertension. Nonetheless, portal
hypertension occurs after any of a multitude of different parenchymal liver injuries and increased
intrahepatic resistance to blood flow is a central element, regardless of the pathogenesis.
The precise site of altered intrahepatic resistance in patients with parenchymal liver disease is
controversial but, in theory, may occur at any of several levels within the liver. From a cellular
standpoint, the modulation of blood flow within the liver may be presinusoidal, sinusoidal, or
postsinusoidal. [21] [42] [52] [103] Moreover, cellular, structural, and humoral elements acting alone or
in combination at one or more sites all
14
may play a role. It is noteworthy that some of these elements are fixed, whereas others are
dynamically regulable. Examples of fixed components that may alter sinusoidal blood flow include
regenerative nodules and possibly hepatocyte swelling. [42] Recent work has focused on the stellate
cell (also known as an Ito cell, perisinusoidal cell, or lipocyte) as a dynamic modulator of sinusoidal
blood flow. This cell has been shown to contract in response to vasoactive substances and relax in
response to dilating agents. [34] [63] [69] This article highlights new developments in the pathogenesis

of portal hypertension, with emphasis on the role of hepatic wounding, stellate cells, and
vasoregulatory compounds.
This work was supported by grants from the NIH (DK 02124 and DK 50574).
Don Rockey, MD
Duke University Medical Center
Sands Building, Rm 334
Research Drive, Box 3083
Durham, NC 27710
CELLULAR COMPONENTS OF THE HEPATIC MICROCIRCULATION

The intrahepatic circulatory system consists of three microvascular components: (1) the terminal portal
venule (TPV) and hepatic arteriole, (2) the sinusoids, which correspond to the capillary bed, and (3) the
terminal hepatic venule (THV). [5] [21] [40] [41] [52] [59] Each may modulate liver blood flow. The major
pre-sinusoidal and postsinusoidal components have been presumed to reflect contraction of vascular
smooth muscle cells in the TPV and THV. Recent evidence suggests that cells within the preterminal
portal venule may play an important role in local control of sinusoidal blood flow. [32] The major cellular
elements (that might modulate blood flow) residing in the sinusoidal compartment of the liver include
sinusoidal endothelial cells and stellate cells; both cell types encircle the sinusoid. Dynamic changes in
endothelial fenestrae induced by neuroamines that correlate with changes in sinusoidal dimensions imply
that sinusoidal endothelial cells could be involved in blood flow regulation. [59] A prominent feature of
cirrhosis is sinusoidal endothelial capillarization, which could also contribute to underlying increases in
resistance to hepatic blood flow. [42] The other cellular element within the sinusoid with the potential to
modulate blood flow is the stellate cell. From an ultrastructural standpoint, this cell possesses long
cytoplasmic processes that envelope sinusoidal endothelial cells. [97] This anatomic relationship within
the sinusoid suggests that stellate cells may be analogous to pericytes, smooth muscle-like cells that are
thought to control capillary blood flow in a wide variety of tissues. [86] [91] Data demonstrating that
stellate cells express the muscle-specific intermediate filament, desmin, assign them to a smooth muscle
lineage and further support this postulate (Fig. 1) . [101] Morphologic studies have identified unmyelinated
nerve fibers in close proximity to stellate cell processes, indicating that stellate cells may respond to
neural stimuli. [10] Finally, data that show that stellate cells express the smooth muscle- and
myofibroblast-specific protein, smooth muscle alpha actin, support the concept that stellate cells are
contractile (Fig. 1) .
15
Figure 1. Stellate cell cytoskeleton. Stellate cells were isolated as previously

described. [14] After 10 days in culture, (stellate cells undergo "activation" with
progressive culture, an event that is recapitulated in vivo during liver injury, see
text for details) cells were fixed and incubated with antidesmin and anti- smooth
muscle alpha actin antibody (linked to Texas Red and fluorescein, respectively).
A fluorescence photomicrograph of a stellate cell expressing both desmin (A)
and smooth muscle alpha actin bundles (B) is shown. Both of these features implicate stellate cells as smooth muscle-like.

(Original magnification 360.)
HEPATIC INJURY, WOUNDING, AND STELLATE CELLS

The general response of the liver and other organs to chronic injury is one of wounding, a complex
process ultimately resulting in fibrosis. Regardless of the type of injury (e.g., viruses, toxins, drugs, and
hereditary defects in the liver), the end result is similar. Fibrogenesis in the liver is typified by fourfold to
sixfold increases in collagen (type I > III > IV)
16
as well as increases in proteoglycan content (dermatan sulfate > chondroitin sulfate > heparan sulfate).
[22] [74] [81] A central feature of the many divergent types of liver injury is the transformation of resident
stellate cells from "quiescent" to "activated" cells (See "Hepatic Wounding and Stellate Cell
Contractility," below). Characteristics of this transition include both morphologic and functional
changes. Morphologic changes include loss of vitamin A, acquisition of stress bundles, and development
of prominent rough endoplasmic reticulum. [16] [36] Importantly, stellate cells from normal livers, which
are cultured in plastic culture dishes, undergo an identical process that recapitulates the phenomena as it
occurs in vivo. Among known functional changes associated with activation is increased secretion of
extracellular matrix proteins. In rats and humans with liver injury, type I, III, and IV collagens;
fibronectin; laminin; and proteoglycans have been localized to stellate cells. [15] [22] [31] Quantitative
studies comparing mRNA for types I, III and IV collagen after liver injury in stellate cells and
hepatocytes have demonstrated a greater than 50-fold increase in stellate cells. [46] Thus, considerable
evidence supports the concept that activated stellate cells are responsible for increased extracellular
matrix deposition during liver injury.
Common to most forms of epithelial injury, including in the liver, is the emergence of actin-containing
cells termed myofibroblasts, generally located in proximity to expanded extracellular matrix. [13] [17]
Myofibroblasts appear to be responsible for the inherent contractility of the cirrhotic liver. [28] Until
recently, a major unresolved question has been the origin of these cells. Cutaneous myofibroblasts appear
to arise from fibroblasts in healing wound. [17] In the liver, however, fibroblasts are few in number and
seen only in periportal areas. [67] Myofibroblasts have smooth-muscle features and could represent cells
migrating from the muscular layer of arteries or veins (the only sites within normal liver which contain
smooth muscle cells). The authors, and others, have recently documented that during liver injury, stellate
cells acquire smooth muscle alpha actin, [4] [67] a protein characteristic of smooth muscle cells or
myofibroblasts. [37] Thus, activation is characterized not only by enhanced fibrogenesis, but also by de
novo smooth muscle alpha actin expression (Fig. 1) . These data indicate that at least one population of
myofibroblasts arises from stellate cells during liver injury. [36] [47] This feature has functional
implications, and served as the basis for studies designed to directly examine the contractility of stellate
cells.

Anatomically, stellate cells are analogous to tissue pericytes, a cell type oriented in a pericapillary
fashion in capillary beds that appear to regulate blood flow via pericapillary constriction. [85] [91] [97] Given
the resemblance to tissue pericytes and the expression smooth muscle-specific proteins (Fig. 1) , the

possibility that stellate cells may exhibit a contractile phenotype has been examined in detail.
Experiments in isolated human stellate cells revealed an increase in intracellular calcium
17
with concomitant cell rounding. [63] Stellate cells induced wrinkling of silicon membranes and
contraction of thick collagen lattices in response to putative vasoconstrictors (Fig. 2) . [34] [72] The cell
culture studies have been further extended to in vivo microscopy, which has demonstrated that stellate
cells appear to function as liver-specific pericytes in vivo. [8] [62] [90] [104] Additionally, Zhang and
coworkers [104] colocalized sinusoidal constriction to stellate cell autofluorescence, suggesting that
stellate cells, rather than endothelial cells, were responsible for sinusoidal constriction. Collectively,
these data provide strong evidence that stellate cells are capable of functioning as perisinusoidal
contractile elements within the sinusoid.
A number of vasoactive substances have been reported to induce stellate cell contraction in various
systems, including angiotensin II, thrombin, substance P, thromboxane A2 analogues, prostaglandin
F2alpha ,
Figure 2. Stellate cell contraction of collagen lattices. Stellate cells were isolated and cultured for 5
days on thick collagen lattices (so as to allow culture induced activation). After introduction of serum
free conditions, serum containing medium (20% serum) was added and lattices were detached.
Hepatocytes, cultured under similar conditions serve as a negative control. Contraction is measured as
the reduction in initial lattice area over time. Stellate cells contract collagen lattices rapidly, as
emphasized in the inset.
18
and the endothelins (ETs) (Table 1) . [34] [35] [63] [72] [77] Importantly, of the various vasoactive compounds
identified, the endothelins appear to be the most potent and reliable inducers of stellate cell contraction.
Thus, endothelin (ET) and its biology are reviewed in the following section.
ENDOTHELIN AND ENDOTHELIN RECEPTORS

Endothelin
The ETs comprise a family of potent vasoconstrictors. [76] [100] Three unique ET peptides, each consisting
of 21 amino acids, have been identified, and termed ET-1, ET-2, and ET-3. [76] [100] They bind to at least
two different G-coupled protein receptors, termed ET A (ETA ) and ET B (ETB ) receptors. ETs are
produced by endothelial cells, and exert paracrine effects on adjacent smooth muscle cells; they may also
have autocrine or intracrine effects. The only type of ET that liver endothelial cells have been shown to
produce is ET-1. [76]
The ETs have been implicated in a wide variety of normal and pathological processes. Their major
function appears to be related to the local control of vascular tone, including regulation of basal blood

pressure. [76] [100] They appear to induce vasoconstriction via stimulation of smooth muscle cells and
vasodilation due to effects on endothelial cells. [11] [12] Although the vasoregulatory functions of the ETs
have been emphasized, recent reports have documented promitogenic effects. [6] Additionally, recent
production of ET-1- and ET-3-deficient mice indicate that the ETs are important regulators of
mammalian neural crest development. [9] [39]
Circulating levels of immunoreactive ET-1 have been reported to be elevated in a number of diseases,
including hypertension, atherosclerosis, acute myocardial infarction, pulmonary fibrosis, acute renal
failure, and chronic liver disease [9] [19] [20] [39] [44] [54] [57] [89] ; however, the importance of elevated plasma
immunoreactive ET-1 in pathologic states must be interpreted with caution. Circulating plasma levels of
ET-1 (even when
TABLE 1 -- AGENTS THAT MODULATE STELLATE CELL CONTRACTILITY
Contract Relax
Endothelin (1, 2, 3) [35] [63] [71] Nitric oxide [68]
Angiotensin II [63] Carbon monoxide [89]
Thrombin [63]
Prostaglandin F2alpha [34]
U46619 (thromboxane A2 ) [34]
Substance P [76]
Serum [71]
These agents have been shown to directly or indirectly, in one or more systems, cause stellate cells to
contract or to relax.
19
elevated) are several orders of magnitude lower than concentrations required to induce biologic effects of
ET-1. ET-1 appears to be rapidly cleared from the circulation, implicating little role for circulating ETs.
Like other endothelium-derived vasoactive substances (e.g., prostaglandin G2 , nitric oxide), ETs appear
to act as local paracrine or autocrine factors. Indeed, the target cell types (i.e., those expressing ET
receptors) are found within close vicinity of cells that produce endothelins.
ET Regulation
The production of ET-1 appears to be regulated, at least in part, at the level of mRNA transcription. [50]
The expression of prepro ET-1 is stimulated by vasopressor hormones, such as epinephrine, angiotensin
II, and vasopressin, shear stress, and cytokines, such as transforming growth factor beta and
interleukin-1. [48] [50] [102] In vascular endothelial cells, once ET-1 is produced, it appears to be secreted in
a constitutive manner, without regulation at the level of exocytosis. [100]
The ET peptides arise by proteolytic processing of large precursors ( 200 amino acid residues).
Intermediates termed big ET-1, -2, and -3 (38-41 amino acids) are excised from prepropeptides by
proteases that cleave at sites containing paired basic amino acids. Big ETs, which have little or no

biologic activity, [100] are finally cleaved at Trp-21-Val/Ile-22 to produce mature 21-residue peptides. The
putative enzyme responsible for the specific cleavage at Trp-21 has been termed ET-converting enzyme
(ECE), and has been identified as a neutral membrane-bound metalloprotease with Mr = 120 kD. [61]
Recently, an ECE that cleaves big ET-1 has been cloned and termed ECE-1. [84] [99] Of note, this
peptidase does not appear to have activity for big ET-3. Analysis of the cDNA demonstrated structural
and sequence homology to neutral endopeptidase 24.11. By northern analysis, ECE has been identified in
lung, adrenal cortex, adrenal medulla, ovary, and testis. Although northern analysis has demonstrated
little or no ECE mRNA in liver, one study detected this mRNA in a perisinusoidal location by in situ
hybridization, [99] consistent with its expression in endothelium. The precise localization of ECE is an
important issue because it may determine the cellular target of mature ET-1.
ET Receptors
Since the discovery of ETs, identification of ET receptors has been an active area of study. Two major
subtypes of G protein-coupled receptors, ETA and ETB , have been definitively identified, cloned, and
shown to mediate a range of biologic effects. [2] [79] ETA receptors are found predominantly on vascular
smooth muscle cells and are preferentially activated by ET-1. Rank order affinities are ET-1 > ET-2 >>>
ET-3; the affinity of ET-1 for the ETA receptor is more than 100-fold that of ET-3. [78]
20
ETB receptors are widely distributed and have equal affinity for ET-1, ET-2, and ET-3. ETB receptor
stimulation appears to bring about divergent responses, depending on the cell type expressing the
receptor. Stimulation of ETB receptors on endothelial cells results in nitric oxide (NO) release and
relaxation of vascular smooth muscle [75] ; however, we have clearly shown stellate cell contraction after
stimulation of ETB , as have others studying systemic vasoconstriction mediated by ETB receptors on
vascular smooth muscle cells. [66] [82] It has been proposed that endothelium-dependent relaxation and
smooth muscle vasoconstriction mediated by the ETB receptor are brought about by two different ETB
receptors, termed ETB1 and ETB2 , respectively. [12] Whether smooth muscle cells (or stellate cells)
possess both ETB1 and ETB2 receptors is unknown. Other ET receptors are emerging, with recent studies
having demonstrated unique receptors in Xenopus dermal melanophores and Xenopus heart, termed ETC
and ETAX , respectively. [33] [38] Additionally, a "super-high affinity" receptor related to ETB has been
proposed based on radioligand binding studies. [87] Precise characterization of stellate cell receptors
mediating biologic responses is essential to the implementation of therapies based on ET antagonists.
ET receptor antagonists have been the center of intense investigation because of their potential for
therapeutic application in a wide variety of diseases, including hypertension and coronary and cerebral
vascular diseases. [11] [45] [60] [40] [41] [67] A vast literature exists emphasizing that the antagonists have
variable biologic effects depending on the target tissue. [45] Numerous ETA receptor, selective antagonists
have been described, the most widely studied of which is BQ-123. [27] Several recent ETB antagonists
have been described, [29] [94] but their effects have not been widely characterized, and, in one instance, not
reproduced. [94] Recent studies have focused on a number of mixed receptor antagonists. Interest in ET
antagonists is substantial because these agents are likely to provide effective therapies for disease
processes characterized by abnormal ET homeostasis.

Studies of ET receptors on stellate cells indicate the presence of both ETA and ETB receptors. Specific
saturation-binding experiments of 125 I-ET-1 to rat stellate cells, endothelial cells, and hepatocytes in
primary cultures demonstrated that the number of receptors on stellate cells outnumbered that for
hepatocytes and endothelial cells by 83 fold and 31 fold, respectively. [26] [88] Competitive binding studies
further demonstrated that stellate cells expressed both ETA and ETB receptors. The presence of these
receptors was confirmed by detection of specific ETA and ETB receptor mRNAs. Furthermore, these
mRNAs were unchanged with activation, suggesting that the effect of ET on stellate cells is regulated at
the level of ligand rather than the receptor.
From a functional standpoint, both ETA and ETB receptors appear to mediate stellate cell contraction. [66]
The ETB receptor appears to be
21
involved additionally in regulating growth of human myofibroblast-like stellate cells. [49] Data with
regard to modulation of receptor subtype are conflicting. Some studies have failed to find changes in
receptor expression during liver injury or in culture, [26] whereas others report that with repeated passage
of cultured cells the relative abundance of ETA receptors on stellate cells decreases markedly. [64]
Because the effect of various ETs on stellate cells is in theory dependent on receptor subtype, further
work is necessary to elucidate how receptors are modulated.
NITRIC OXIDE AND CARBON MONOXIDE

Nitric Oxide
Stellate cell contraction within the hepatic sinusoid is likely to be balanced by dilating compounds (Table
1) . Candidate effectors, such as NO, a recently identified vasodilating substance, are attractive. The
half-life of NO is short, but products of NO's interaction with reactive oxygen intermediates (i.e., NO
plus superoxide yields peroxynitrite) results in products with longer half-lives, which may also exert
important local effects. NO is produced by one of three isoforms of NO synthase (NOS). [56] Endothelial
cells and neurons produce NO from L-arginine via two distinct constitutive NOS enzymes while a wide
variety of cells express inducible NOS (iNOS). [56] The constitutive isoforms typically produce small
amounts of NO and are regulated by calmodulin binding in the context of changes in intracellular
calcium. In contrast, iNOS is bound tightly to calmodulin, and therefore functions in large part
independently of intracellular calcium; it produces large amounts of NO. [56]
The amount and duration of NO production in a biologic system depends on its enzymatic source.
Although the term constitutive implies that these isoforms are not regulated, their mRNA and protein
levels may be altered by changes in the environment, such as hypoxia, cytokines, or stretch. [83] The
inducible isoform is stimulated by a wide array of compounds/stimuli, including cytokines and
lipopolysaccharide (LPS), which have been emphasized as prominent inducers of iNOS. Cytokines and
LPS appear to stimulate iNOS gene transcription via autoregulation of the NFkappa-B/Ikappa-Balpha
complex. [98] iNOS regulation is particularly complex, as recent studies indicate that iNOS enzymatic

activity is itself inhibited by NO. [96]

NO may play a role in the modulation of sinusoidal blood flow and in the pathogenesis of portal
hypertension. The NO inhibitor, NNA ( Nw -nitro- L-arginine) increased portal pressure in normal rat
livers whereas another NO inhibitor, L-NAME ( Nw -nitro- L-arginine methyl ester), had no effects. [53] [90]
Although NO levels in the normal liver appear to be quite low (implicating a minor role for NO),
perfusion of cirrhotic livers with L-arginine reduced portal reactivity to norepinephrine, suggesting a
defect in endogenous NO production in cirrhosis. [23] Indeed, in some
22
forms of experimental liver injury and portal hypertension iNOS is not induced, consistent with the
possibility that a lack of NO may be important in the perpetuation of intrahepatic portal hypertension. [71]
Further, recent data indicate that sinusoidal endothelial cells from cirrhotic livers have an impairment in
the ability to produce NO via endothelial NOS. [70] Thus, a relative decrease in NO may contribute to
enhanced stellate cell contractility and portal hypertension.
At a cellular level, effects of NO on stellate cells are pronounced. Current data indicate that exogenous
NO is capable of preventing ET-induced contraction, as well as inducing relaxation in precontracted
cells. [35] [69] Additionally, stellate cells themselves produce NO in response to interferon-gamma and
other cytokines with or without LPS, [24] [69] and NO produced in an autocrine fashion has prominent
relaxing effects on stellate cells. [69] Essentially all liver cell types are capable of producing NO (via
inducible or constitutive isoforms of NOS). Therefore, paracrine NO may also serve as an important
mediator of stellate cell relaxation.
Carbon Monoxide
Although NO has received a great deal of attention as a mediator of stellate cell relaxation, Suematsu and
coworkers [90] have recently raised the possibility that locally produced CO might also control stellate cell
contractility, and thus sinusoidal blood flow. These authors demonstrated that inhibition of CO
production (CO is produced by the oxidation of heme-by-heme oxygenase) caused an increase in portal
vascular resistance. These data are provocative and identify another potential modulator of stellate cell
contractility.
HEPATIC WOUNDING AND STELLATE CELL CONTRACTILITY

During the hepatic wounding response stellate cells undergo a transition from a "quiescent" to an
"activated" state (Fig. 3) (Figure Not Available) . One of the most prominent features of this process is
the de novo acquisition of the smooth muscle protein, smooth muscle alpha actin (Figure 3) (Figure Not
Available) . In culture, this phenomenon coincides with the acquisition of the contractile phenotype,
suggesting that smooth muscle alpha actin confers on stellate cells the ability to contract. In culture
models, quiescent stellate cells are not contractile, yet activated cells display enhanced contractility. [34]
[72] Such data imply that contractility may be directly related to the activation state of stellate cells in
vivo. This possibility has been tested in two in vivo models of experimental liver injury. [7] [73] In one,
freshly isolated stellate cells isolated from livers with advanced liver injury displayed significantly
greater contractility than those isolated from normal livers. [73] Additionally, intravital microscopy

indicated enhanced in situ contractility of stellate cells after activation induced by chronic ethanol
consumption. [7]
23
Figure 3. (Figure Not Available) Stellate Cell Activation and Contractility. In normal liver, stellate cells retain "quiescent"
features which include a vitamin A rich state, relatively minor collagen synthesis and a low level of proliferation. After
liver injury or during culture on uncoated plastic, stellate cells undergo "activation", a programmed cascade of events. The
in vivo phenomenon of stellate cell activation appears to be closely recapitulated by culture of stellate cells on standard
uncoated plastic substrata. In addition to proliferation and enhanced extracellular matrix production, a major feature of
activation, which is emphasized, is new expression of smooth muscle alpha actin. The acquisition of this smooth muscle
protein implies a contractile phenotype (see text for details). Of the proposed modulators of stellate cell contractility,
endothelin (ET) and nitric oxide (NO) appear to be the most prominent. (Courtesy of Lydia V. Kibiuk)
Available data are consistent with stellate cell-mediated alterations in sinusoidal blood flow via
perisinusoidal constriction, particularly in the injured liver; however, it is noteworthy that myofibroblasts
(i.e., activated stellate cells) populate large fibrous bands (consisting primarily of type I collagen) in the
injured and cirrhotic liver. [4] Moreover, ET-1, when perfused into the liver of rats with advanced
fibrosis, binds not only to cells in perisinusoidal spaces (consistent with stellate cells) but also to cells
within fibrous bands. [73] Stellate cells express abundant quantities of the collagen-binding integrins
alpha1beta1 and alpha2beta1 and these integrins readily mediate stellate cell contraction of type I
collagen. [65] Such information raises the possibility that stellate cells are capable of contraction of large
collagenous bands (i.e., in response to ET), causing alteration in lobular architecture and subsequent
derangement of hepatic blood flow. This concept is consistent with previous data demonstrating inherent
contractility of the cirrhotic liver. [28] Thus, stellate cell contractility may contribute to intrahepatic portal
hypertension via perisinusoidal constriction, macroscopic effects, or both.
24
TABLE 2 -- CIRCULATING ET IN PATIENTS WITH CIRRHOSIS

Study Author (yr) Type of ET Comment
Uchihara (1992) ET-1 Ascites > w/o ascites
Moore (1992) ET-1/ET-3 HRS > cirrhosis only
Moller (1993) ET-1 Proportional to
Isobe (1993) ET-1 Creatinine
Asbert (1993) ET-1 Varices > w/o varices
Matsumoto (1994) ET-1 Ascites > w/o ascites
Gerbes (1995) ET-1/ET-3 Cirrhosis > hepatitis
ET-3 proportional to
Child-Turcotte
HRS = Hepatorenal syndrome.

ET IN LIVER INJURY
Multiple studies indicate that circulating ET levels are elevated in patients with cirrhosis (Table 2) . *
Because ET acts as a paracrine (or autocrine) factor, and circulating ET is likely to represent local
overproduction of ET, the possibility that ET was overproduced in the injured liver has been
investigated. [25] [43] [64] PreproET-1 mRNA has not been detectable in normal liver but is abundant in the
livers of animals with several types of experimental liver injury. Further, intrahepatic levels of ET-1
peptide appear to correlate with the severity of disease and the amount of ascites. [1] ET-1 appears to be
localized to perisinusoidal cells, consistent with production by endothelial or stellate cells. [25]
Additionally, during culture-induced activation, mRNA for preproET-1 increased in stellate cells, [26]
suggesting an autocrine loop for its actions. Thus, ET produced in an autocrine fashion may have effects
on stellate cell contractility. The finding that a novel mixed ETA /ETB antagonist, bosentan, reduced
portal pressure in rats with portal hypertension further supports this postulate. [73]
Because stellate cells appear to be a major target of ET in liver, and possess a marked fibrogenic capacity
when activated, recent investigation has focused on the role of ETs and stellate cell activation. [49] [68] In
cultured stellate cells, both ET-1 and sarafotoxin S6C, a potent ETB receptor agonist, stimulated
expression of smooth muscle alpha actin, a classic feature of activation (see Figs. 1 and 3) (Figure Not
Available) . [68] [72] In an in vivo model of liver injury, antagonism of the effects of ET by blocking ETA
/ETB receptors reduced stellate cell activation and fibrogenesis. [68] These data indicate that not only is
ET important in mediating stellate cell contraction, but it is also an important component of stellate cell
activation.
SUMMARY
The pathogenesis of portal hypertension is multifactorial, and appears to result from interplay between
fixed and dynamically modulable
* References [3] [18] [30] [51] [54] [55] [57] [58] [80] [92] [93] [95] .
25
elements; the stellate cell is a newly recognized example of the latter. This perisinusoidal, pericyte-like
cell has contractile features that are most prominent after liver injury, concomitant with their activation.
These data imply an exaggerated contractile phenotype in the cirrhotic liver. This cell may contribute to
increased intrahepatic portal hypertension via perisinusoidal constriction of the sinusoid or by contraction
of fibrous extracellular matrix rich in type I collagen with concomitant disruption of lobular architecture.
Endothelins and NO play a major role in the modulation of stellate cell contractility, and are therefore
important in the pathogenesis of intrahepatic portal hypertension. These new data provide potential areas
for therapeutic intervention in this clinical entity.
References
1. Alam I, Bass NM, Gee L, et al: Hepatic endothelin-levels in chronic liver disease correlate with disease severity and
ascites. Hepatology 24:142a, 1996

2. Arai
H, Hori S, Aramori I, et al: Cloning and expression of a cDNA encoding an endothelin receptor. Nature
348:730-732, 1990
3. Asbert M, Gines A, Gines P, et al: Circulating levels of endothelin in cirrhosis. Gastroenterology 104:1485-91, 1993
4. Ballardini G, Fallani M, Biagini G, et al: Desmin and actin in the identification of Ito cells and in monitoring their
evolution to myofibroblasts in experimental liver fibrosis. Virchows Arch [Cell Pathol] 56:45-49, 1988
5. Bass NM, Manning JA, Weisiger RA: Increased sinusoidal volume and solute extraction during retrograde liver
perfusion. Am J Physiol 256:G1041-G1048, 1989
6. Battistini B, Chailler P, D'Orleans-Juste P, et al: Growth regulatory properties of endothelins. Peptides 14:385-399, 1993
7. BauerM, Paquette NC, Zhang JX, et al: Chronic ethanol consumption increases hepatic sinusoidal contractile response
to endothelin-1 in the rat. Hepatology 22:1565-76, 1995
8. Bauer M, Zhang JX, Bauer I, et al: ET-1 induced alterations of hepatic microcirculation: Sinusoidal and extrasinusoidal
sites of action. Am J Physiol 267:G143-G149, 1994
9. Baynash
AG, Hosoda K, Giaid A, et al: Interaction of endothelin-3 with endothelin-B receptor is essential for
development of epidermal melanocytes and enteric neurons. Cell 79:1277-1285, 1994
10. Bioulac-Sage P, Lafon ME, Saric J, et al: Nerves and perisinusoidal cells in human liver. J Hepatol 10:105-112, 1990
11. ClozelM, Breu V, Burri K, et al: Pathophysiological role of endothelin revealed by the first orally active endothelin
receptor antagonist. Nature 365:759-761, 1993
12. ClozelM, Gray GA, Breu V, et al: The endothelin ETB receptor mediates both vasodilation and vasoconstriction in
vivo. Biochem Biophys Res Commun 186:867-873, 1992
13. Darby
I, Skalli O, Gabbiani G: Alpha-smooth muscle actin is transiently expressed by myofibroblasts during
experimental wound healing. Lab Invest 63:21-29, 1990
14. de
Leeuw AM, McCarthy SP, Geerts A, et al: Purified rat liver fat-storing cells in culture divide and contain collagen.
Hepatology 4:392-403, 1984
15. Friedman SL: Cellular sources of collagen and regulation of collagen production in liver. Semin Liver Dis 10:20-29,
1990
16. Friedman
SL: Seminars in medicine of the Beth Israel Hospital, Boston: The cellular basis of hepatic fibrosis:
Mechanisms and treatment strategies. N Engl J Med 328:1828-1835, 1993
17. Gabbiani G, Ryan GB, Majne G: Presence of modified fibroblasts in granulation tissue and their possible role in wound
contraction. Experientia 27:549-550, 1971
26
18. Gerbes AL, Moller S, Gulberg V, et al: Endothelin-1 and -3 plasma concentrations in patients with cirrhosis: Role of

splanchnic and renal passage and liver function. Hepatology 21:735-739, 1995
19. Giaid A, Michel RP, Stewart DJ, et al: Expression of endothelin-1 in lungs of patients with cryptogenic fibrosing
alveolitis. Lancet 341:1550-1554, 1993
20. Giaid
A, Saleh D, Yanagisawa M, et al: Endothelin-1 immunoreactivity and mRNA in the transplanted human heart.
Transplantation 59:1308-1313, 1995
21. Greenway CV, Lautt WW: Distensibility of hepatic venous resistance sites and consequences on portal pressure. Am J
Physiol 254:H452-H458, 1988
22. GressnerAM, Bachem MG: Cellular sources of noncollagenous matrix proteins: Role of fat-storing cells in
fibrogenesis. Semin Liver Dis 10:30-46, 1990
23. GuptaTK, Groszman RJ: Administration of L-arginine, the physiological precursor of nitric oxide, reduces portal
perfusion pressure and ameliorates hepatic hyperreactivity in experimental cirrhosis. Hepatology 20:200, 1994
24. Helyar
L, Bundschuh DS, Laskin JD, et al: Induction of hepatic Ito cell nitric oxide production after acute endotoxemia.
Hepatology 20:1509-1515, 1994
25. HoussetC, Laudat A, Chung JJ, et al: Expression and potential role of endothelin in cholestatic liver injury. Hepatology
22:278, 1995
26. Housset CN, Rockey DC, Friedman SL, et al: Hepatic lipocytes: A major target for endothelin-1. J Hepatol 22:55-60,
1995
27. Ihara
M, Ishikawa K, Fukuroda T, et al: In vitro biological profile of a highly potent novel endothelin (ET) antagonist
BQ-123 selective for the ETA receptor. J Cardiovasc Pharmacol 20(suppl)12:S11-S14, 1992
28. Irle
C, Kocher O, Gabbiani G: Contractility of myofibroblasts during experimental liver cirrhosis. Journal of
Submicroscopic Cytology 12:209-217, 1980
29. IshikawaK, Ihara M, Noguchi K, et al: Biochemical and pharmacological profile of a potent and selective endothelin
B-receptor antagonist, BQ-788. Proc Natl Acad Sci U.S.A. 91:4892-4896, 1994
30. Isobe
H, Satoh M, Sakai H, et al: Increased plasma endothelin-1 levels in patients with cirrhosis and esophageal varices.
J Clin Gastroenterol 17:227-230, 1993
31. JarnaginWR, Rockey DC, Koteliansky VE, et al: Expression of variant fibronectins in wound healing: Cellular source
and biological activity of the EIIIA segment in rat hepatic fibrogenesis. J Cell Biol 127:2037-2048, 1994
32. KanedaK, Ekataksin W, Sogawa M, et al: Endothelin-1-induced vasoconstriction localized in the distal segment of
preterminal portal venule causes significant increase in portal pressure of rat liver. Hepatology 24:133a, 1996
33. Karne S, Jayawickreme CK, Lerner MR: Cloning and characterization of an endothelin-3 specific receptor (ETC
receptor) from Xenopus laevis dermal melanophores. J Biol Chem 268:19126-19133, 1993
34. Kawada N, Klein H, Decker K: Eicosanoid-mediated contractility of hepatic stellate cells. Biochem J 285:367-371,
1992
35. Kawada N, Tran-Thi TA, Klein H, et al: The contraction of hepatic stellate (Ito) cells stimulated with vasoactive
substances: Possible involvement of endothelin 1 and nitric oxide in the regulation of the sinusoidal tonus. Eur J Biochem

213:815-823, 1993
36. Kent
G, Gay S, Inouye T, et al: Vitamin A-containing lipocytes and formation of type III collagen in liver injury. Proc
Natl Acad Sci U.S.A. 73:3719-3722, 1976
37. Kocher O, Skalli O, Bloom WS: Cytoskeleton of rat aortic smooth muscle cells: Normal conditions and experimental
intimal thickening. Lab Invest 50:645-652, 1984
38. Kumar C, Mwangi V, Nuthulaganti P, et al: Cloning and characterization of a novel endothelin receptor from Xenopus
heart. J Biol Chem 269:13414-13420, 1994
39. KuriharaY, Kurihara H, Suzuki H, et al: Elevated blood pressure and craniofacial abnormalities in mice deficient in
endothelin-1. Nature 368:703-710, 1994
40. Lautt WW, Greenway CV: Conceptual review of the hepatic vascular bed. Hepatology 7:952-963, 1987
41. Lautt
WW, Greenway CV, Legare DJ: Effect of hepatic nerves, norepinephrine, angiotensin, and elevated central
venous pressure on postsinusoidal resistance sites and intrahepatic pressures in cats. Microvasc Res 33:50-61, 1987
27
42. Lautt
WW, Greenway CV, Legare DJ, et al: Localization of intrahepatic portal vascular resistance. Am J Physiol
251:G375-G381, 1986
43. Leivas A, Jimenez W, Lamas S, et al: Endothelin 1 does not play a major role in the homeostasis of arterial pressure in
cirrhotic rats with ascites. Gastroenterology 108:1842-1848, 1995
44. Lerman A, Edwards BS, Hallett JW, et al: Circulating and tissue endothelin immunoreactivity in advanced
atherosclerosis. N Engl J Med 325:997-1001, 1991
45. Luscher TF: Do we need endothelin antagonists? Cardiovasc Res 27:2089-2093, 1993
46. Maher JJ, McGuire RF: Extracellular matrix gene expression increases preferentially in rat lipocytes and sinusoidal
endothelial cells during hepatic fibrosis in vivo. J Clin Invest 86:1641-1648, 1990
47. MakKM, Leo MA, Lieber CS: Alcoholic liver injury in baboons: Transformation of lipocytes to transitional cells.
48. MalekAM, Greene AL, Izumo S: Regulation of endothelin 1 gene by fluid shear stress is transcriptionally mediated and
independent of protein kinase C and cAMP. Proc Natl Acad Sci U.S.A. 90:5999-6003, 1993
49. Mallat
A, Fouassier L, Preaux AM, et al: Growth inhibitory properties of endothelin-1 in human hepatic
myofibroblastic Ito cells: An endothelin B receptor-mediated pathway. J Clin Invest 96:42-49, 1995
50. Marsden
PA, Brenner BM: Transcriptional regulation of the endothelin-1 gene by TNF-alpha. Am J Physiol
262:C854-C861, 1992
51. Matsumoto H, Uemasu J, Kitano M, et al: Clinical significance of plasma endothelin-1 in patients with chronic liver
disease. Dig Dis Sci 39:2665-70, 1994

52. McCuskey RS: A dynamic and static study of hepatic arterioles and hepatic sphincters. American Journal of Anatomy
119:455-478, 1966
53. Mittal
MK, Gupta TK, Lee FY, et al: Nitric oxide modulates hepatic vascular tone in normal rat liver. Am J Physiol
267:G416-G422, 1994
54. Moller
S, Emmeluth C, Henriksen JH: Elevated circulating plasma endothelin-1 concentrations in cirrhosis. J Hepatol
19:285-290, 1993
55. MollerS, Gulberg V, Henriksen JH, et al: Endothelin-1 and endothelin-3 in cirrhosis: Relations to systemic and
splanchnic haemodynamics. J Hepatol 23:135-144, 1995
56. Moncada S, Higgs A: The L-arginine-nitric oxide pathway. N Engl J Med 329:2002-2012, 1993
57. Moore
K, Wendon J, Frazer M, et al: Plasma endothelin immunoreactivity in liver disease and the hepatorenal
syndrome [see comments]. N Engl J Med 327:1774-1778, 1992
58. NozueT, Kobayashi A, Uemasu F, et al: Plasma endothelin-1 levels of children with cirrhosis. J Pediatr Gastroenterol
Nutr 21:220-223, 1995
59. Oda M, Azuma T, Watanabe N: Regulatory mechanism of hepatic microcirculation: Involvement of the contraction and
dilation of sinusoids and sinusoidal fenestrae. Progress in Applied Microcirculation 127:103-128, 1990
60. OhlsteinEH, Beck GRJ, Douglas SA, et al: Nonpeptide endothelin receptor antagonists. II: Pharmacological
characterization of SB 209670. J Pharmacol Exp Ther 271:762-768, 1994
61. Ohnaka K, Takayanagi R, Nishikawa M, et al: Purification and characterization of a phosphoramidon-sensitive

endothelin-converting enzyme in porcine aortic endothelium. J Biol Chem 268:26759-26766, 1993
62. Okumura S, Takei Y, Kawano S, et al: Vasoactive effect of endothelin-1 on rat liver in vivo. Hepatology 19:155-161,
1994
63. Pinzani M, Failli P, Ruocco C, et al: Fat-storing cells as liver-specific pericytes: Spatial dynamics of agonist-stimulated
intracellular calcium transients. J Clin Invest 90:642-646, 1992
64. PinzaniM, Milani S, De Franco R, et al: Endothelin 1 is overexpressed in human cirrhotic liver and exerts multiple
effects on activated hepatic stellate cells. Gastroenterology 110:534-48, 1996
65. Racine-Samson L, Rockey DC, Bissell DM: Expression of the collagen-binding integrins alpha1beta1 and alpha2beta1
during activation of rat hepatic stellate cells in vivo. Hepatology 24:133a, 1996
28
66. Rockey
DC: Characterization of endothelin receptors mediating rat hepatic stellate cell contraction. Biochem Biophys
Res Commun 207:725-731, 1995
67. Rockey DC, Boyles JK, Gabbiani G, et al: Rat hepatic lipocytes express smooth muscle actin upon activation in vivo
and in culture. J Submicrosc Cytol Pathol 24:193-203, 1992

68. RockeyDC, Chung JJ: Endothelin antagonism in experimental hepatic fibrosis: Implications for endothelin in the
pathogenesis of wound healing. J Clin Invest 98:1381-1388, 1996
69. Rockey DC, Chung JJ: Inducible nitric oxide synthase in rat hepatic lipocytes and the effect of nitric oxide on lipocyte
contractility. J Clin Invest 95:1199-206, 1995
70. RockeyDC, Chung JJ: Modulation of nitric oxide synthase III in sinusoidal endothelial cells: Implications for the
pathogensis of portal hypertension. Hepatology 24:190a, 1996
71. Rockey DC, Chung JJ: Regulation of inducible nitric oxide synthase during liver injury. Hepatology 22:249, 1995
72. Rockey DC, Housset CN, Friedman SL: Activation-dependent contractility of rat hepatic lipocytes in culture and in
vivo. J Clin Invest 92:1795-1804, 1993
73. Rockey DC, Weisiger R: Endothelin induced contractility of stellate cells from normal and cirrhotic rat liver:
Implications for regulation of portal pressure and resistance. Hepatology 24:233-240, 1996
74. Rojkind M, Giambrone MA, Biempica L: Collagen types in normal and cirrhotic liver. Gastroenterology 76:710-719,
1979
75. Rubanyi GM: The role of endothelium in cardiovascular homeostasis and diseases. J Cardiovasc Pharmacol 22(suppl
4):S1-14, 1993
76. Rubanyi GM, Botelho LH: Endothelins. FASEB J 5:2713-2720, 1991
77. Sakamoto M, Ueno T, Kin M, et al: Ito cell contraction in response to endothelin-1 and substance P. Hepatology
18:978-983, 1993
78. Sakurai
T, Yanagisawa M, Masaki T: Molecular characterization of endothelin receptors. Trends Pharmacol Sci
13:103-108, 1992
79. SakuraiT, Yanagisawa M, Takuwa Y, et al: Cloning of a cDNA encoding a non-isopeptide-selective subtype of the
endothelin receptor. Nature 348:732-735, 1990
80. SaloJ, Francitorra A, Follo A, et al: Increased plasma endothelin in cirrhosis: Relationship with systemic endotoxemia
and response to changes in effective blood volume. J Hepatol 22:389-98, 1995
81. Schuppan D: Structure of the extracellular matrix in normal and fibrotic liver: Collagens and glycoproteins. Semin
Liver Dis 10:1-10, 1990
82. SeoB, Oemar BS, Siebenmann R, et al: Both ETA and ETB receptors mediate contraction to endothelin-1 in human
blood vessels. Circulation 89:1203-1208, 1994
83. Sessa WC: The nitric oxide synthase family of proteins. J Vasc Res 31:131-143, 1994
84. Shimada K, Takahashi M, Tanzawa K: Cloning and functional expression of endothelin-converting enzyme from rat
endothelial cells. J Biol Chem 269:18275-18278, 1994
85. Sims DE: The pericyte: A review. Tissue Cell 18:153-174, 1986
86. Sims DE: Recent advances in pericyte biology: Implications for health and disease. Can J Cardiol 7:431-443, 1991

87. Sokolovsky M, Ambar I, Galron R: A novel subtype of endothelin receptors. J Biol Chem 267:20551-20554, 1992
88. Stephenson K, Harvey SA, Mustafa SB, et al: Endothelin association with the cultured rat Kupffer cell: Characterization
and regulation. Hepatology 22:896-905, 1995
89. Stewart DJ, Kubac G, Costello KB, et al: Increased plasma endothelin-1 in the early hours of acute myocardial
infarction. J Am Coll 18:38-43, 1991
90. Suematsu M, Goda N, Sano T, et al: Carbon monoxide: An endogenous modulator of sinusoidal tone in the perfused rat
liver. J Clin Invest 96:2431-2437, 1995
91. Tilton
RG, Kilo C, Williamson JR, et al: Differences in pericyte contractile function in rat cardiac and skeletal muscle
microvasculatures. Microvasc Res 18:336-352, 1979
92. Uchihara M, Izumi N, Sato C, et al: Clinical significance of elevated plasma endothelin concentration in patients with
cirrhosis. Hepatology 16:95-9, 1992
29
93. Uemasu J, Matsumoto H, Kawasaki H: Increased plasma endothelin levels in patients with liver cirrhosis. Nephron
60:380, 1992
94. Urade
Y, Fujitani Y, Oda K, et al: An endothelin B receptor-selective antagonist: IRL 1038,
[Cys11-Cys15]-endothelin-1(11-21). FEBS Lett 311:12-16, 1992
95. Veglio F, Pinna G, Melchio R, et al: Plasma endothelin levels in cirrhotic subjects. J Hepatol 15:85-7, 1992
96. Vodovotz
Y, Kwon NS, Pospischil M, et al: Inactivation of nitric oxide synthase after prolonged incubation of mouse
macrophages with IFN-gamma and bacterial lipopolysaccharide. J Immunol 152:4110-4118, 1994
97. Wake K: Perisinusoidal stellate cells (fat-storing cells, interstitial cells, lipocytes), their related structure in and around
the liver sinusoids, and vitamin A-storing cells in extrahepatic organs. Int Rev Cytol 66:303-353, 1980
98. Xie
QW, Kashiwabara Y, Nathan C: Role of transcription factor NF-kappa B/Rel in induction of nitric oxide synthase. J
Biol Chem 269:4705-4708, 1994
99. XuD, Emoto N, Giaid A, et al: ECE-1: A membrane-bound metalloprotease that catalyzes the proteolytic activation of
big endothelin-1. Cell 78:473-485, 1994
100. Yanagisawa M: The endothelin system: A new target for therapeutic intervention. Circulation 89:1320-1322, 1994
101. Yokoi
Y, Namihisa T, Kuroda H, et al: Immunocytochemical detection of desmin in fat-storing cells (Ito cells).
Hepatology 4:709-714, 1984
102. Yoshizumi M, Kurihara H, Morita T, et al: Interleukin 1 increases the production of endothelin-1 by cultured
endothelial cells. Biochem Biophys Res Commun 166:324-329, 1990
103. Zhang JX, Bauer M, Clemens MG: Vessel- and target cell-specific actions of endothelin-1 and endothelin-3 in rat liver.

Am J Physiol 269:G269-G277, 1995
104. Zhang
JX, Pegoli WJ, Clemens MG: Endothelin-1 induces direct constriction of hepatic sinusoids. Am J Physiol
266:G624-G632, 1994



Figure 1. Stellate cell cytoskeleton. Stellate cells were isolated as previously described. [14] After 10 days
in culture, (stellate cells undergo "activation" with progressive culture, an event that is recapitulated in
vivo during liver injury, see text for details) cells were fixed and incubated with antidesmin and anti-
smooth muscle alpha actin antibody (linked to Texas Red and fluorescein, respectively). A fluorescence
photomicrograph of a stellate cell expressing both desmin (A) and smooth muscle alpha actin bundles (B)
is shown. Both of these features implicate stellate cells as smooth muscle-like. (Original magnification
360.)



Figure 1. Stellate cell cytoskeleton. Stellate cells were isolated as previously described. [14] After 10 days
in culture, (stellate cells undergo "activation" with progressive culture, an event that is recapitulated in
vivo during liver injury, see text for details) cells were fixed and incubated with antidesmin and anti-
smooth muscle alpha actin antibody (linked to Texas Red and fluorescein, respectively). A fluorescence
photomicrograph of a stellate cell expressing both desmin (A) and smooth muscle alpha actin bundles (B)
is shown. Both of these features implicate stellate cells as smooth muscle-like. (Original magnification
360.)


April 22, 2000
Conde Petra
PREFACE
Additional Article
Full Text
PREFACE
Frontmatter
Figure . DOUGLAS R. LaBRECQUE, MD, FACP

Guest Editor
The harbinger of death in patients with cirrhosis of the liver is

the development of portal hypertension. Bleeding varices,
ascites and its complications, hepatic encephalopathy, or a
combination of two or more of the "big three" complications of
portal hypertension, is the usual proximate cause of death in
such patients. There has been little improvement in the long
term survival of these patients since the 1940s, other than the
widespread application of liver transplantation for patients with
endstage liver disease, and cirrhosis continues to be one of the
leading causes of death worldwide. However, recent advances
in our understanding of the pathophysiology of portal
hypertension and the development of more creative approaches
to its treatment over the past decade, bring with them the
promise that the new century may see prolonged survival and a
better quality of life for these patients. The rapidly evolving
recognition of the multifactorial nature of portal hypertension is
directing therapy away from simple attempts to correct fixed
mechanical problems to an attack at the cellular level on local
factors which affect blood flow and resistance to flow through
the liver. Sclerotherapy for actively bleeding varices has
http://home.mdconsult.com/das/journal/view/N/10888...=&stty=cow&nid=39984&tlt=Z&pos=&tag=relinfo&anc=ja (1 of 6) [22/04/2000 0:06:12]

evolved into banding with the promise of increased efficacy and

fewer complications at a lower cost due to the need for fewer
therapeutic sessions. Pharmacotherapy with newer agents can
reduce the risk of the first bleed or help prevent rebleeding and
may improve survival, all with fewer side effects than in the
past. The development of the transjugular intrahepatic
portosystemic shunt (TIPS) has added a new therapeutic tool to
our armamentarium in the management of the variceal bleeder
and continued surgical advances have improved the outcome for
those patients who require shunting procedures. Finally, the
dramatic improvement in the results of liver transplantation due
to improved techniques and better immunosuppression, along
with its wider availability, offers the hope of a return to a
normal, productive and long life for patients with end stage
liver failure. In this volume we have attempted to focus on the
recent advances which have been made in our understanding of
portal hypertension and their application to its therapy. We will
also look to the future and attempt to define the next stage of
therapeutic advances.
The first section examines the pathophysiology of portal
hypertension from
18A
the clinical to the molecular level. Drs. Gupta, Chen, and

Groszmann elegantly define the clinical syndrome and detail the
physiologic characteristics of the irreversible (mechanical) and
reversible (vascular tone) components of portal hypertension.
Dr. Rockey carries this to the molecular level from which much
of the most exciting new information on portal hypertension is
emanating. He details the newly identified role of the stellate
(Ito) cell, a cell whose contractile phenotype is exacerbated in
the cirrhotic liver and may contribute a dynamic and reversible
aspect to intrahepatic portal hypertension due to perisinusoidal
constriction of the sinusoid or the extracellular fibrous matrix.
The roles of the endothelins and nitric oxide in such modulation
are stressed and future directions in this important area of
research are outlined.
The second section focuses on the most dramatic and most often
fatal complication of portal hypertension, variceal bleeding.
One problem in all studies of variceal bleeding has been a lack
of good information on the natural history of varices. Dr. Boyer
has done a masterful job of analyzing the available information
on the natural history of portal hypertensive varices and
provides a strong base from which to estimate the risk of

variceal bleeding in different patient groups and the potential

benefits of therapy. In today's world of extremely cost
conscious medicine, treatment decisions are increasingly based
on the results of outcomes research and a strict cost benefit
analysis. Twenty-five per cent of cirrhotics can be expected to
suffer a variceal bleed and 20-60% will die at the time of their
first bleed, depending on the severity of their underlying
disease. Thus Drs. Gralnek and Jensen's discussion of the cost
assessment of managing variceal bleeding is particularly
pertinent and builds nicely on Dr. Boyer's chapter to help define
when it is appropriate to provide prophylaxis against variceal
bleeding, as well as the most cost effective treatments for active
variceal bleeding. Their chapter also stresses the limits of such
data and the critical need for careful cost assessments in the
future.
Some of the greatest advances in the treatment of variceal
bleeding have occurred at the pharmacological level and these
are summarized beautifully by Drs. Grace and Bhattacharya.
The routine use of nonselective beta-adrenergic blocking agents
prior to the first variceal bleed is the first truly prophylactic
therapy which has been shown to reduce morbidity and
mortality in patients with portal hypertension. Their chapter
covers the different drugs available for prevention and treatment
of variceal bleeding and reviews the literature on such therapy
in a comprehensive fashion in order to provide practical
suggestions for approaching patients with portal hypertension
and varices. One of the newest therapies for variceal bleeding is
banding, and Drs. Slosberg and Keeffe carefully review the
potential merits of the newer technique of variceal banding and
discuss why it is likely to replace variceal sclerosis because of
its lower rate of complications, requirement of fewer sessions to
achieve variceal obliteration and possible improvement in
survival. Probably the most creative and exciting new approach
to uncontrolled variceal bleeding is the transjugular intrahepatic
portosystemic shunt (TIPS). The appeal of a non-surgical
technique with a low morbidity and low mortality is obvious
and in the initial wave of enthusiasm for this procedure, it has
also been applied to the long term therapy of varices, as well as
ascites. Drs. Faust and Sorrell evaluate its potential uses in each
of these settings, as well as its important limitations, placing a
more critical face on its role in the management of portal
hypertension. The advent of medical and endoscopic therapy for
bleeding esophageal varices, along with the improved medical
regimens to prevent variceal bleeding, have reduced the role of
surgical shunts in the treatment of portal hypertension.
However, there are situations where the simpler techniques fail
and the only option is a surgical one. Doctors Iannitti and

Henderson discuss the limits and benefits of the
19A
various surgical approaches to portal hypertension, including

updates on the non-selective small diameter interposition graft
between the portal vein and inferior vena cava (Sarfeh shunt),
and the selective distal splenorenal (Warren) and
coronary-caval (Inokuchi) shunts, defining the role of surgery in
selected patients. Ultimately, all of our current attempts to
improve survival in patients with portal hypertension are based
on trying to correct the complications of damage which is itself
irreversible. In many such patients this is not enough to improve
survival, or new damage occurs due to our inability to treat the
patient's primary disease. For many patients liver
transplantation offers their only hope for improved long term
survival. Drs. Sudan and Shaw discuss the role of liver
transplantation, stressing the need to tailor therapy for variceal
bleeding so as not to compromise future liver transplantation in
patients with a strong likelihood of eventually requiring such
treatment. I conclude this section with a brief summary of what
I feel to be a reasonable approach to the treatment of varices
based on the therapies available as we approach the twenty-first
century.
The next section considers the other two main complications of
portal hypertension. Drs. Gines, Arroyo, and Rodes consider the
pathophysiology, complications, and treatment of ascites with
special emphasis on newer information concerning the use of
the "old fashioned technique" of large volume paracentesis.
They include a thorough discussion of the controversy
concerning the need, or lack thereof, for albumin infusion
following large volume paracentesis, as well as the role of TIPS
and the timing of liver transplantation in patients with
recalcitrant ascites. Practical approaches to all of the common
problems in the patient with ascites, from hyponatremia to
hepatorenal syndrome are presented. The pathogenesis of
spontaneous bacterial peritonitis and its therapy, including
prophylaxis to prevent reinfection, are considered in detail. Drs.
Schenker and Bay provide an encyclopedic review of our
current understanding of the pathophysiology of portosystemic
encephalopathy, as well as the various approaches to therapy,
stressing the most recent developments in this complex area.
One of the more interesting and confusing complications of
portal hypertension, the hepatopulmonary syndrome, is finally

revealing some of its secrets because of the work of individuals

such as Drs. Abrams and Fallon, who discuss the current status
of our understanding of this dramatic and confounding
complication of portal hypertension and the limited approaches
to therapy currently available.
One problem infrequently faced by adult gastroenterologists
and hepatologists is portal hypertension in children. Because of
the small number of pediatric gastroenterologists and even
smaller number of pediatric hepatologists, most patients are
initially discussed with adult gastroenterologists. Drs. Alonso,
Hackworth, and Whitington emphasize the most important fact
to remember in initially evaluating such children: they are not
little adults. The causes of portal hypertension in children are
mostly unique to childhood and the approaches are often
different from those used in adults. This article provides an
excellent primer for those of us who are periodically called on
to advise our colleagues on such problems in children. The
volume concludes with a wide ranging discussion of one of the
more intriguing aspects of portal hypertension, its presence in
individuals who do not have cirrhosis. Drs. Hillaire, Valla, and
Lebrec provide a comprehensive review of the topic.
I was greatly honored when asked to be Guest Editor for this
Inaugural Issue of the Clinics in Liver Disease, but almost
begged off when I was informed of the short publication time
line. However, I was extremely gratified by the interest
expressed by the outstanding group of authors who agreed to
take part in this venture. The short time line has allowed them
to include information
20A
that is unusually up to date, even for a periodical publication of

this type. Thus the reader will find the references to be very
current, including some from 1997. I can only thank the authors
for their graciousness in agreeing to work under tight time
constraints. Their enthusiasm for this volume fed mine and I am
extremely pleased with the result. Taken together these articles
provide a comprehensive summary of our understanding of
portal hypertension, its pathogenesis and therapy, as we
approach the next century. I thoroughly enjoyed reading each of
the articles and I trust that the readers will as well.
DOUGLAS R. LABRECQUE MD, FACP

Guest Editor Department of Internal Medicine

University of Iowa Hospitals and Clinics
4553 JCP Iowa City, IA 52242


April 22, 2000
Conde Petra
Diagnosis and treatment of gastrointestinal

bleeding secondary to portal hypertension.
American College of Gastroenterology Practice
Citation Parameters Committee.
Bibliographic Data
Abstract
Indexing Data
Diagnosis and treatment of
Copyright Notice and Disclaimer gastrointestinal bleeding secondary to
portal hypertension. American College of
Journal Articles On This Topic Gastroenterology Practice Parameters
Committee.
Full Text Grace ND - Am J Gastroenterol - 1997 Jul; 92(7): 1081-91

Frontmatter
NLM Citation ID:
INTRODUCTION 97363482
DEFINITIONS AND Full Source Title:
PATHOPHYSIOLOGY American Journal of Gastroenterology
METHOD Publication Type:
DIAGNOSIS Guideline; Journal Article; Practice Guideline; Review;
Review, Tutorial
Control of acute variceal bleeding
Language:
Recommendations. English
DISCUSSION Authors:
Endoscopic treatment Grace ND
Sclerotherapy. Number of References:

150
Esophageal variceal ligation.
Abstract:
Pharmacologic therapy Guidelines for clinical practice are intended to suggest
Vasopressin. preferable approaches to particular medical problems as
established by interpretation and collation of scientifically valid
Vasopressin plus nitroglycerin. research, derived from extensive review of published literature.
When data are not available that will withstand objective
Somatostatin/octreotide.
scrutiny, a recommendation may be made based on a consensus
Terlipressin (Glypressin). of experts. Guidelines are intended to apply to the clinical
situation for all physicians without regard to specialty.
Miscellaneous drugs
http://home.mdconsult.com/das/guideline/view/N/945...I&stty=cow&nid=39984&tlt=Z&pos=&tag=relinfo&anc=gl (1 of 3) [22/04/2000 0:07:00]

Treatment of patients Guidelines are intended to be flexible, not necessarily indicating

unresponsive to medical therapy the only acceptable approach, and should be distinguished from
Balloon tamponade. standards of care, which are inflexible and rarely violated.
Given the wide range of choices in any health care problem, the
Esophageal staple transection. physician should select the course best suited to the individual
Transjugular intrahepatic
patient and the clinical situation presented. These guidelines are
portasystemic stent shunt developed under the auspices of the American College of
(TIPS). Gastroenterology and its practice parameters committee. These
guidelines are also approved by the governing boards of
Conclusions American College of Gastroenterology and Practice Parameters
PREVENTION OF FIRST VARICEAL
Committee. Expert opinion is solicited from the outset for the
HEMORRHAGE document. Guidelines are reviewed in depth by the committee,
with participation from experienced clinicians and others in
Recommendations related fields. The final recommendations are based on the data
available at the time of the production of the document and may
Discussion
be updated with pertinent scientific developments at a later
Pharmacologic therapy. time. The following guidelines are intended for adults and not
for pediatric patients. OBJECTIVE: To develop practice
Endoscopic therapy.
guidelines for the management of gastrointestinal bleeding in
Prophylactic shunt surgery. adult patients with cirrhosis and portal hypertension.
METHOD: Randomized controlled trials published through
Conclusions October of 1993 were evaluated by members of the American
PREVENTION OF RECURRENT College of Gastroenterology Practice Parameters Committee.
VARICEAL HEMORRHAGE Each paper was reviewed by three members of the committee
and rated for quality of design by predetermined criteria.
Recommendations Meta-analysis of the studies for each treatment were evaluated
Discussion for both outcome and quality of design and formed the basis for
recommendations for treatment. Randomized controlled trials
Pharmacologic therapy. published between October of 1993 and August of 1995 have
been added to update and modify the recommendations. The
Endoscopic therapy
reader is referred to an excellent article by D'Amico et al. (The
Sclerotherapy. treatment of portal hypertension: A meta-analytic review.
Hepatology 1995;22:332-354), which presents most of the
Endoscopic variceal ligation.
meta-analyses reviewed by this committee. CONCLUSIONS:
Sclerotherapy versus Once esophageal varices have been established by endoscopy as
propranolol. the site of bleeding, either sclerotherapy or endoscopic variceal
ligation should be performed to control the bleeding episodes.
Combined endoscopic and Concomitant use of vasoactive drugs lowers portal pressure,
pharmacologic therapy.
potentially offers the endoscopist a clearer field in which to
Treatment for failures of work, and is the only noninvasive treatment for
medical therapy. nonesophagogastric variceal sites of bleeding related to portal
hypertension. For patients failing medical therapy, the
TIPS procedure. transjugular intrahepatic portasystemic shunt procedure is a
Total and selective shunts. reasonable alternative to an emergency surgically created shunt.
Nonselective beta-adrenergic blockers are the only proven
CONCLUSION therapy for prevention of first variceal hemorrhage. Both

REFERENCES
nonselective beta-adrenergic blockers and endoscopic variceal
About the Publication ligation (which has replaced sclerotherapy for this indication)
are effective in reducing the risk of recurrent variceal bleeding.
For patients failing these approaches, selective or total shunts
or, in selected patients, liver transplantation are appropriate
rescue procedures.
Major Subjects:
Esophageal and Gastric Varices / * Complications /
Etiology
Gastrointestinal Hemorrhage / * Diagnosis / Drug
Therapy / Etiology / * Therapy
Hypertension, Portal / Complications / * Diagnosis / *
Therapy
Acute Disease
Combined Modality Therapy
Diagnosis, Differential
Endoscopy, Digestive System
Human
Ligation
Portasystemic Shunt, Surgical
Randomized Controlled Trials
Recurrence
Sclerotherapy
Bookmark URL: /das/guideline/view/N/9451414?source=HS,MI


Volume 92 Number 7 July 1997
P1081
Practice guidelines
Diagnosis and Treatment of Gastrointestinal Bleeding Secondary to

Portal Hypertension
Norman D. Grace M.D.
Guidelines for clinical practice are intended to suggest preferable approaches to particular medical
problems as established by interpretation and collation of scientifically valid research, derived
from extensive review of published literature. When data are not available that will withstand
objective scrutiny, a recommendation may be made based on a consensus of experts. Guidelines
are intended to apply to the clinical situation for all physicians without regard to specialty.
Guidelines are intended to be flexible, not necessarily indicating the only acceptable approach, and
should be distinguished from standards of care, which are inflexible and rarely violated. Given the
wide range of choices in any health care problem, the physician should select the course best suited
to the individual patient and the clinical situation presented. These guidelines are developed under
the auspices of the American College of Gastroenterology and its practice parameters committee.
These guidelines are also approved by the governing boards of American College of
Gastroenterology and Practice Parameters Committee. Expert opinion is solicited from the outset
for the document. Guidelines are reviewed in depth by the committee, with participation from
experienced clinicians and others in related fields. The final recommendations are based on the
data available at the time of the production of the document and may be updated with pertinent
scientific developments at a later time. The following guidelines are intended for adults and not for
pediatric patients. Objective: To develop practice guidelines for the management of gastrointestinal
bleeding in adult patients with cirrhosis and portal hypertension. Method: Randomized controlled
trials published through October of 1993 were evaluated by members of the American College of
Gastroenterology Practice Parameters Committee. Each paper was reviewed by three members of
the committee and rated for quality of design by predetermined criteria. Meta-analysis of the
studies for each treatment were evaluated for both outcome and quality of design and formed the
basis for recommendations for treatment. Randomized controlled trials published between
October of 1993 and August of 1995 have been added to update and modify the recommendations.
The reader is referred to an excellent article by D'Amico et al. (The treatment of portal
hypertension: A meta-analytic review. Hepatology 1995;22:332-354), which presents most of the
http://home.mdconsult.com/das/article/body/jorg=g...line&source=HS,MI&sp=9451414&sid=0/N/67694/1.html (1 of 17) [22/04/2000 0:07:29]

meta-analyses reviewed by this committee. Conclusions: Once esophageal varices have been
established by endoscopy as the site of bleeding, either sclerotherapy or endoscopic variceal
ligation should be performed to control the bleeding episodes. Concomitant use of vasoactive drugs
lowers portal pressure, potentially offers the endoscopist a clearer field in which to work, and is the
only noninvasive treatment for nonesophagogastric variceal sites of bleeding related to portal
hypertension. For patients failing medical therapy, the transjugular intrahepatic portasystemic
shunt procedure is a reasonable alternative to an emergency surgically created shunt. Nonselective
beta-adrenergic blockers are the only proven therapy for prevention of first variceal hemorrhage.
Both nonselective beta-adrenergic blockers and endoscopic variceal ligation (which has replaced
sclerotherapy for this indication) are effective in reducing the risk of recurrent variceal bleeding.
For patients failing these approaches, selective or total shunts or, in selected patients, liver
transplantation are appropriate rescue procedures.
Received May 23, 1996.

accepted Mar. 18, 1997.
Reprint requests and correspondence: Norman D. Grace, M.D., Faulkner Hospital, Department of Gastroenterology, 1153
Centre Street, Boston, MA 02130.
INTRODUCTION
Bleeding from esophagogastric varices accounts for one-third of all deaths in patients with cirrhosis and portal
hypertension [1] . Prospective studies have shown that up to 90% of patients with cirrhosis will develop esophageal varices
[2] [3] . The risk of bleeding from esophagogastric varices is 25-35% for both alcoholic and nonalcoholic cirrhosis with the
majority of initial bleeding episodes occurring
P1082
within the first year from the time of diagnosis [5] [6] . For those patients who survive the initial episodes of bleeding, the
risk of recurrent bleeding approaches 70% with most episodes occurring within 6 months of the index bleed [7] . The
mortality for each episode of variceal bleeding is 30-50% depending on the clinical status of the patient [7] . In the past 25
years, a number of medical and surgical approaches have been proposed in an attempt to alter these grim statistics. The
purpose of this guideline is to critically review the current therapeutic options and to develop recommendations for the
treatment of bleeding secondary to portal hypertension.
DEFINITIONS AND PATHOPHYSIOLOGY

Portal hypertension is defined as an increase in the portal venous pressure gradient (portal pressure - inferior vena cava
pressure), which is a function of portal venous blood flow and intrahepatic and portocollateral resistance. In patients with
cirrhosis, portal hypertension is initiated by an increase in intrahepatic and portocollateral resistance and is exacerbated by
systemic vasodilation and a decrease in vascular resistance. Nitric oxide, and to a lesser extent, glucagon, autonomic
nervous system dysfunction (especially the adrenergic system), prostaglandins, tumor necrosis factor alpha, and other
cytokines have been implicated as humoral agents responsible for the systemic vasodilation, which leads to plasma volume
expansion, an increase in cardiac output, and a hyperdynamic circulation. The hyperdynamic circulation leads to an
increase in portal venous blood flow against relatively fixed intrahepatic and portocollateral resistance and, consequently,
portal hypertension. Hepatic resistance may be modified by changes in the perivenular and presinusoidal myofibroblasts as
well as changes in the smooth muscle component of the portocollateral vessels.
Esophageal varices are dilated intraesophageal veins associated with an extensive, dilated and tortuous capillary network

resulting in an anatomic picture of "varices upon varices." A hepatic venous pressure gradient (HVPG) 12 mm Hg is
necessary for the formation of esophagogastric varices in patients with alcoholic cirrhosis [8] .
METHOD
Only published randomized controlled trials (RCTs) were considered by the committee for developing recommendations.
The primary author reviewed each manuscript. Meta-analysis of RCTs for each treatment option in which there were a
minimum of three similarly designed trials with a sample size of at least 100 patients were constructed using the Peto
modification of the Mantel-Haentzel method [9] [10] . The 10-member committee reviewed and discussed a quality of
design scoring system proposed by Pagliaro et al.[11] . Each paper was then graded independently by three members of the
committee for each of nine variables (randomization technique, efficacy of randomization, blinding, compliance,
withdrawals, sample size, treatment of bleeding, side effects and complications, and external validity) according to the
criteria of Pagliaro et al.[11] . The scores were submitted to the primary author and collated. If there was a difference of
greater than 20% in the scores of the reviewers, the paper was discussed at a subsequent committee meeting. The final
recommendations of the committee are based on both the meta-analysis and quality of design of the studies.
DIAGNOSIS
The diagnosis of esophagogastric varices is made by endoscopy. Although a variety of grading systems have been
proposed, a recent consensus conference agreed that varices should be classified only as large or small [5] [12] .
Interobserver agreement is good for the presence and size of esophageal varices [4] . Endoscopic diagnosis of gastric
varices is subject to more variability and endoscopic ultrasound may be valuable in confirming the diagnosis [13] . Bleeding
from esophagogastric varices is determined endoscopically, either by direct visualization of bleeding or, more often, by
endoscopic stigmata of recent hemorrhage in the absence of any other source of bleeding. Although bleeding from
esophagogastric varices is the most dramatic and potentially lethal complication of portal hypertension, other sites of
bleeding include portal hypertensive gastropathy, portal hypertensive colopathy, and duodenal and anorectal varices.
Risk factors for bleeding from esophageal varices include the size of the varices (large greater than small), endoscopic red
color signs, and Child's classification. In patients with alcoholic cirrhosis, continued consumption of alcohol is associated
with a high risk of bleeding. Using these criteria, several classifications have been proposed to predict patients at risk for
variceal bleeding [2] [5] [14] . Although these classifications have been successful in identifying a group of patients at high
risk for variceal hemorrhage, they fail to identify a substantial number of patients who experience variceal bleeding and
are at the lower end of their risk classes [15] .
Control of acute variceal bleeding
Recommendations.
Resuscitative measures should include establishment of adequate intravenous access and blood volume replacement with
packed red blood cells and fresh frozen plasma as appropriate. In general, patients should be slightly undertransfused,
maintaining a hematocrit in the low 30s. A central line may be very helpful in estimating the need for volume replacement.
In patients with acute bleeding, protection of the airway by intubation may be necessary.
Once esophageal varices have been established by endoscopy as the bleeding site, sclerotherapy for the control of acute
variceal bleeding is the currently accepted therapy of choice. If bleeding is not controlled with two sessions,
P1083
alternative approaches should be sought. Endoscopic esophageal variceal ligation is a reasonable alternative.
Concomitant use of vasoactive drugs (vasopressin plus nitroglycerin, somatostatin, octreotide, terlipressin) offers several
advantages. They can be started in the emergency room when variceal bleeding is suspected and offer a potentially clearer
field for the endoscopist. Unlike endoscopic therapy, which acts locally, they lower portal pressure. They are the only

proven medical therapy for nonesophageal sites of bleeding such as portal hypertensive gastropathy or gastric varices more
than 2-3 cm below the gastroesophageal junction.
For failures of medical therapy, the transjugular intrahepatic portasystemic shunt (TIPS) offers a less invasive means of
controlling acute bleeding than an emergency surgically created shunt. In an appropriate candidate, liver transplantation is
an option but is governed by donor availability.
DISCUSSION
Endoscopic treatment
Sclerotherapy.
Over the last decade, sclerotherapy has become the treatment of choice for the control of acute variceal bleeding.
Controlled trials [15] [16] have shown sclerotherapy to be superior to balloon tamponade or vasopressin with an initial
success rate of 60-90%. Burroughs et al.[17] reported that 62% of patients had control of variceal bleeding for at least 5
days after a single session of sclerotherapy and 78% were controlled with two sessions. Additional sessions of
sclerotherapy were of diminishing value. However, three RCTs found sclerotherapy and somatostatin [18] [19] or octreotide
[20] infusion to be equally effective with fewer complications in patients treated with somatostatin or octreotide. Only one
study reported a survival benefit for patients receiving treatment with sclerotherapy when compared with balloon
tamponade [21] . Variations in technique such as the type of sclerosant, intravariceal versus paravariceal injection, or the
use of compression devices do not appear to influence the success rate. Treatment with sclerotherapy is associated with a
10-30% complication rate and 0.5-2.0% mortality. Complications include esophageal ulceration, bleeding, perforation,
bacteremia, mediastinitis, pleural effusion, and pulmonary edema. Stricture formation, often associated with impaired
esophageal motility can usually be managed with esophageal dilation.
Esophageal variceal ligation.
The use of esophageal variceal ligation offers a reasonable alternative to sclerotherapy with potentially fewer
complications. Although none of the nine RCTs comparing sclerotherapy and esophageal variceal ligation were
specifically designed to test the procedures' efficacy for the control of acute variceal bleeding, several of the studies
include a subset of patients in whom the techniques were used for this indication. In a controlled trial by Stiegmann et
al.[22] , 86% of patients with acute esophageal variceal bleeding were controlled with endoscopic ligation compared with
77% with sclerotherapy. In subsequent RCTs, Gimson et al.[23] were able to control acute variceal bleeding in 91% of
patients with variceal ligation compared with 92% for sclerotherapy, Laine et al.[24] reported an 89% success rate for both
therapies. Hou et al.[25] demonstrated 100% control of acute bleeding with ligation compared with 88% with sclerotherapy
and Lo et al.[26] found ligation successful in 80% of patients compared with a 94% success rate for sclerotherapy. A
meta-analysis of these data shows sclerotherapy and variceal ligation to have comparable efficacy for the control of acute
variceal hemorrhage. Fewer complications were reported for patients treated with variceal ligation in the studies by
Stiegmann [22] , Laine [24] , Hou [25] , and Lo [26] , but there was no difference in the complication rate reported by Gimson
[23] .
Pharmacologic therapy
Vasoconstrictors (vasopressin, somatostatin, octreotide, terlipressin) have been used clinically in the treatment of portal
hypertension. Several RCTs have combined the vasodilator nitroglycerin with vasopressin in an attempt to enhance its
efficacy and ameliorate the side effects of vasopressin. Vasoconstrictors decrease splanchnic arterial blood flow, which in
turn, decreases portal venous blood flow and pressure. Vasodilators act primarily by decreasing intrahepatic vascular
resistance or by dilation of the portocollateral circulation. A secondary but less significant mechanism involves production
of peripheral vasodilation, which produces reflex splanchnic vasoconstriction and a decrease in blood flow.
Vasopressin.
Randomized control trials have shown that vasopressin successfully controls acute variceal bleeding in 52% of patients,
compared with 18% for placebo but is associated with a rebleeding rate of up to 45% [27] [28] [29] [30] [31] [32] . Use of
vasopressin is also associated with significant complications including myocardial and peripheral ischemia, bradycardia,

hypertension, hyponatremia, and fluid retention. None of the RCTs reported a survival benefit. Because of the high
complication rate, which is dose dependent, alternative agents have been sought.
Vasopressin plus nitroglycerin.
The addition of nitroglycerin to vasopressin has ameliorated the vasoconstrictive complications of vasopressin, allowing
the use of higher doses of vasopressin. A meta-analysis of three controlled trials shows that vasopressin plus nitroglycerin
is superior to vasopressin alone for the control of acute variceal bleeding [33] [34] [35] . One RCT reported better control of
variceal bleeding with balloon tamponade compared with vasopressin and nitroglycerin (86.5 vs 66%; p < 0.01) [36] .
Although nitroglycerin can be given sublingually or transdermally, continuous intravenous infusion is preferable because
the dose can be readily adjusted in response to peripheral blood pressure. Absorption from the transdermal route can be
quite variable.
P1084
Somatostatin/octreotide.
RCTs have shown somatostatin to be superior to the control therapy for the management of acute variceal bleeding in three
trials and comparable to the control treatment in all other trials except one. Somatostatin was more effective than
vasopressin in two trials [37] [38] and placebo in one [39] . Somatostatin has been found to be comparable to vasopressin in
five trials [40] [41] [42] [43] [44] , to Glypressin in one trial [45] , to sclerotherapy in three trials [18] [19] [20] , and to balloon
tamponade in two trials [46] [47] . Only one multicenter trial found somatostatin to be less effective than placebo and that
study reported an astonishingly high success rate of 83% for the placebo preparation [48] . All studies reported only a few
minor side effects for somatostatin and a meta-analysis failed to reveal any survival benefit.
Because somatostatin is at least as effective as vasopressin and is associated with significantly fewer complications, it may
replace vasopressin for routine clinical use. Octreotide, the synthetic longer acting analogue of somatostatin, was found to
be comparable to vasopressin in one study [49] and to balloon tamponade [50] in another. Because data for RCTs evaluating
octreotide is limited, a recommendation for its use cannot be made at this time.
Two RCTs have compared sclerotherapy plus octreotide to sclerotherapy alone for the prevention of early rebleeding from
esophageal varices. One study found the combination therapy superior to sclerotherapy alone for the prevention of early
rebleeding [51] , whereas the other study reported no benefit for the addition of octreotide to sclerotherapy [52] .
Terlipressin (Glypressin).
Terlipressin is a relatively inert analogue of lysine vasopressin which, when given in bolus form, is slowly activated with
cleavage of the N-terminal glycyl residue. In the reported trials comparing the two drugs the slow release results in lower
blood levels of lysine vasopressin resulting in less cardiotoxicity or vasoconstriction than vasopressin. Six RCTs have been
published with a variety of control groups. Three of the trials found terlipressin superior to either a placebo [53] [54] or
vasopressin [55] . However, in one of the trials comparing terlipressin to placebo [54] , balloon tamponade was used in 78%
of the patients in both groups, making it difficult to differentiate the effects of drug therapy from those of balloon
tamponade. Two trials [56] [57] reported that balloon tamponade and terlipressin were equally effective, whereas one trial
reported no difference between terlipressin and somatostatin [45] . In the trial by Soderlund et al.[53] there was a significant
difference in hospital mortality (10% for terlipressin vs 38% for placebo, p = 0.02). A RCT comparing low-dose (2-mg
bolus followed by 1 mg every 6 h) versus high-dose (4-mg bolus followed by 2 mg every 6 h) terlipressin found that the
control of bleeding and transfusion requirement was similar, suggesting no advantage for the higher dose [58] . Another
RCT [59] found terlipressin plus nitroglycerin as effective as octreotide for the control of acute variceal bleeding; however,
octreotide-treated patients had a significantly lower transfusion requirement. The combination of terlipressin and
desmopressin, a synthetic analogue of l-arginine vasopressin that has been reported to improve hemostasis in patients with
cirrhosis, was less effective than terlipressin given alone for the control of acute variceal bleeding [60] . Because of the
fewer side effects reported with its use, terlipressin may have advantages over lysine vasopressin and may be comparable
to somatostatin. However, terlipressin is not yet available in the United States.
Miscellaneous drugs
Metoclopramide and domperidone have been reported to decrease both azygous blood flow [61] and blood flow to the

submucous venous plexus in the distal esophagus and to reduce esophageal variceal pressure [62] . In a small controlled
study, Hosking et al.[63] demonstrated metoclopramide to be more effective than a placebo in the control of acute variceal
hemorrhage (91 vs 36%, p < 0.01).
Treatment of patients unresponsive to medical therapy
Seventy-five to ninety percent of patients with acute esophagogastric variceal bleeding will respond to endoscopic an/or
pharmacologic therapy. For those who fail two sessions of either sclerotherapy or variceal band ligation, alternative
treatment should be sought.
Balloon tamponade.
Patients may be temporarily controlled with balloon tamponade but long-term use of this technique is associated with an
unacceptably high complication rate and a high rebleeding rate.
Esophageal staple transection.
Transection of the distal esophagus, using a staple gun, has been shown to be successful in controlling esophageal variceal
hemorrhage. A controlled trial by Burroughs et al.[17] reported staple transection to be superior to one session and
equivalent to two sessions of sclerotherapy (88% control for transection vs 82% for sclerotherapy) with no differences in
the complication rate or survival. In a follow-up study [64] , the investigators reported that 90% of patients who rebled after
sclerotherapy responded to staple transection. In other RCTs, Huizinga et al.[65] found staple transection to be superior to
sclerotherapy for the control of acute variceal bleeding, whereas Teres et al.[66] found the procedures comparable.
Transjugular intrahepatic portasystemic stent shunt (TIPS).
This radiologic procedure, which creates an intrahepatic shunt by placing a stent connecting the hepatic and portal veins
through the liver, has gained wide popularity in the past few years. No RCTs have been published to evaluate its efficacy
in the control of acute variceal bleeding. Nevertheless, the success rate reported in several series has led to its being
proposed as the rescue procedure of choice for the control of variceal bleeding in patients failing medical therapy [67] [68]
[69] . Other options would include an emergency portasystemic shunt or liver transplantation.
P1085
Conclusions
Current practice would dictate sclerotherapy or endoscopic variceal ligation plus the use of a pharmacologic agent ( i.e.,
vasopressin and nitroglycerin, terlipressin, or somatostatin) as the initial approach for the control of acute variceal
hemorrhage.
PREVENTION OF FIRST VARICEAL HEMORRHAGE

Recommendations
Patients with established cirrhosis who are felt to be reasonable candidates for long-term pharmacologic therapy should
undergo a screening endoscopy to assess the presence of esophagogastric varices. If the initial endoscopic evaluation fails
to reveal varices, repeat endoscopy should be performed at periodic intervals, perhaps every 2 years.
Patients with large varices should be placed on pharmacologic therapy for the prevention of first variceal hemorrhage. At
present, nonselective beta-adrenergic blockers (propranolol, nadolol) are first line therapy. Long-acting nitrates (isosorbide
mono- or di-nitrate) may be a reasonable substitute in those patients who have contraindications to or are unable to tolerate
nonselective beta-blockers. This therapy should be continued indefinitely. Neither sclerotherapy nor prophylactic
portasystemic shunt surgery have any role in the prevention of first variceal hemorrhage.
Discussion

In patients with cirrhosis who are found to have esophageal varices, there is a 10-15% risk of dying from variceal
hemorrhage within 1 yr of the diagnosis of varices. This estimate is based on a 25-35% risk of bleeding from varices
within 1 yr from the time of diagnosis and a 30-50% mortality for each episode of variceal hemorrhage. Therefore, an
effective therapy to prevent initial variceal bleeding is highly desirable.
Pharmacologic therapy.
Although a number of pharmacologic agents have been shown to effectively decrease portal pressure in animal models,
only the nonselective beta-adrenergic blockers, propranolol and nadolol, have been extensively evaluated in randomized
controlled trials. beta-Blockers decrease both cardiac output (B1 effect) and splanchnic blood flow (B2 effect), which
results in a decrease in the hyperdynamic circulation and especially in portal blood flow, resulting in a decrease in portal
pressure. There are seven published RCTs comparing nonselective beta-adrenergic blockers (five propranolol, two
nadolol) with placebo for the prevention of initial variceal bleeding [70] [71] [72] [73] [74] [75] [76] . A meta-analysis of these
studies [16] shows a significant benefit for beta-adrenergic blockers in the prevention of initial variceal hemorrhage and a
trend for prevention of mortality. The studies are well designed and homogeneous. One study [75] had a three-way
randomization between propranolol, sclerotherapy, or placebo. Propranolol was significantly better than either
sclerotherapy or the placebo in prevention of first variceal hemorrhage, with no significant difference between the latter
two groups. Although there was no observed survival benefit, there was a trend for improved survival for patients treated
with propranolol. In another study [76] , there was a four-way randomization between propranolol, sclerotherapy, the
combination of propranolol and sclerotherapy, or a control group. The rate of variceal bleeding was almost identical in the
four treatment groups. However, survival was slightly better in the propranolol and control groups compared with patients
treated with sclerotherapy. Although one of two studies published in abstract only [77] showed a higher bleeding rate in
patients treated with propranolol when compared with a placebo, inclusion of these studies in a meta-analysis does not
alter the favorable outcome for beta-adrenergic blockers.
Subgroup analysis shows propranolol to be effective for both Child's A and B patients [78] . Relatively few patients with
Child's C liver disease were entered into these trials, so that recommendations for the use of beta-adrenergic blockers in
this group of patients has to be qualified. The presence of ascites is a well-known risk factor for variceal bleeding. Two
trials showed beta-adrenergic blockers to be effective in patients with ascites at the time of randomization [72] [74] , whereas
one found propranolol more effective in patients without ascites [71] . In an analysis of individual patient data from four of
the European studies, Poynard et al.[78] found beta-adrenergic blockers effective in preventing first variceal hemorrhage
for both patients with and without ascites. Eighty-two percent of patients entered into the seven published trials had large
varices at the time of randomization (only three trials included patients with small varices) [79] . Because of the limited
number of patients with small varices in these RCTs, no statement can be made about the use of beta-blockers in this group
of patients. Nonselective beta-adrenergic blockers were more likely to be effective in patients who were compliant in
taking their medication [80] . In one study, a hepatic venous pressure gradient (HVPG) of >12 mm Hg was an entry
criterion and serial measurement of the HVPG were made at 3 months, 1 year, and yearly thereafter [81] . Of the 102
patients in the study, 14 patients randomized to propranolol and 7 randomized to placebo had a decrease in the HVPG
below 12 mm Hg during follow-up and gastrointestinal hemorrhage did not occur in any of these patients.
Side effects requiring cessation of beta-adrenergic blocker therapy ranged from 3 to 27% in the reported trials. In addition,
congestive heart failure, severe chronic obstructive lung disease, asthma, and insulin dependent diabetes mellitus are felt to
be relative contraindications to the use of nonselective beta-adrenergic blockers. In patients who are unable to tolerate
beta-adrenergic blocker therapy or who have contraindications to their use, long-acting nitrates (isosorbide mono- or
dinitrate) may be a reasonable alternative. In one RCT [82] , propranolol and isosorbide mononitrate were equally effective
in the prevention of first variceal hemorrhage with no difference in survival. Therapy with beta-adrenergic
P1086
blockers should be continued indefinitely because cessation of treatment after 2 years has been associated with a
recurrence of the risk of bleeding [83] . Endoscopic therapy.
Nineteen RCTs (15 published and 4 in abstract form) have compared sclerotherapy to obliterate esophageal varices with
comparable control groups [75] [76] [84] [85] [86] [87] [88] [89] [90] [91] [92] [93] [94] [95] [96] [97] . Although the early trials were
positive, more recent studies failed to confirm this initial enthusiasm. A meta-analysis of the published trials showed a
significant benefit for sclerotherapy in the prevention of initial variceal hemorrhage [11] . However, there is an inverse

correlation between the quality of design and the reporting of favorable results. Design problems included a lack of clear
definition of a bleeding episode in nine of the studies [80] . In those studies in which bleeding was defined
semiquantitatively, none found a significant benefit for treatment with sclerotherapy. Because of the heterogeneity of the
survival data in this meta-analysis, a meaningful statement about potential survival benefit cannot be made. Two of the
three studies [11] reporting a survival benefit received poor scores for quality of design. In these two studies [85] [86] ,
patients in the control group experiencing variceal bleeding were not offered sclerotherapy to control bleeding, thus
depriving them of the most effective available treatment for control of acute bleeding. Two of the larger studies had to be
stopped because the external study monitors determined that there was a significant survival advantage for patients
randomized to the control group [76] [87] . If the early, "poorly designed" studies are excluded, then there is no advantage
for sclerotherapy either for the prevention of initial variceal bleeding or for survival. There are no published RCTs
evaluating endoscopic variceal ligation for the prevention of initial variceal bleeding.
Prophylactic shunt surgery.
Surgical creation of a portacaval shunt in patients with cirrhosis and esophageal varices is very effective in the prevention
of initial variceal hemorrhage. However, a meta-analysis [84] [98] of four RCTs [99] [100] [101] shows a survival advantage for
patients assigned to the medical control groups. Mortality in the surgical patients was related to hepatic failure and
portasystemic encephalopathy. There is no indication for portasystemic shunt surgery for the prevention of initial variceal
bleeding. One RCT has compared selective shunts (distal splenorenal) and nonshunting devascularization procedures to a
medical control group [102] . Although a survival benefit was reported for the patients randomized to surgery (22%
mortality for surgery vs 49% for medical patients), more than half of the surgical patients were treated with a nonshunt
procedure. In addition to a variety of procedures in the surgical group that were determined by surgeons' choice, patients in
the medical group received neither sclerotherapy nor pharmacologic agents for the control of acute variceal bleeding.
Conclusions
Only nonselective beta-adrenergic blockers have clearly shown a benefit for the prevention of initial variceal bleeding. All
patients with established cirrhosis should undergo screening endoscopy to assess the presence of varices. Although data
concerning the best time intervals for repeat screening sessions are lacking, several established liver centers have adopted a
policy of screening patients every 2-3 yr. Patients with large varices and/or endoscopic signs for risk of bleeding should be
placed on treatment. Whether treating patients with small esophageal varices is cost effective or clearly influences survival
is unknown. There is no indication for sclerotherapy or surgical shunts in this group of patients.
PREVENTION OF RECURRENT VARICEAL HEMORRHAGE

Recommendations
Because of the high risk of recurrent variceal hemorrhage, particularly within the first 2 months after an acute variceal
bleed [7] , therapy is indicated for all patients in an attempt to prevent rebleeding. With rare exceptions, initial therapy
should be medical, either endoscopic (sclerotherapy, esophageal variceal ligation) or pharmacologic (nonselective
beta-adrenergic blockers or long-acting nitrates). The combination of endoscopic and pharmacologic approaches may
appear logical but more data from longer term RCTs is needed before this approach can be recommended.
For patients who fail medical therapy, options would include the TIPS procedure, selective, small bore or total shunts, or
liver transplantation. The clinical status of the patient and donor availability will influence this choice.
Discussion
Unlike the prevention of initial variceal hemorrhage, about which the data supporting the use of nonselective
beta-adrenergic blockers make pharmacologic therapy the only reasonable choice, there are several good options for
prevention of recurrent variceal bleeding.
Pharmacologic therapy.
Nine published RCTs [103] [104] [105] [106] [107] [108] [109] [110] [111] have compared nonselective beta-adrenergic blockers
(propranolol, nadolol) to a placebo for the prevention of recurrent variceal bleeding. A meta-analysis of these trials shows

a significant reduction in rebleeding [15] [112] . The results are homogeneous with six trials showing a significant reduction
in recurrent bleeding and the remainder showing a trend in favor of beta-blockers. Evaluation of quality of design shows
that all the studies were of high caliber. If data from the published studies are combined, there is a reduction in bleeding
from 66% in the placebo treated patients to 44% in patients receiving beta-blockers (eight studies) [80] . However, the
reduction in risk of recurrent bleeding did not produce a significant reduction in mortality (28% placebo, 22%
beta-blockers). Thirty-seven percent of patients randomized to beta-blocker therapy died as a result of variceal bleeding
compared with 54% in the placebo treated patients (21 of 57 vs 37 of 68). The lack of a significant reduction in overall
mortality may be a reflection of variations in the
P1087
study populations including the type and severity of liver disease, the interval between the index bleed and the initiation of
treatment or it may reflect a type II error. In general, patients with relatively good liver function (Child's class A & B)
appear to be good candidates for beta-blocker therapy because very few Child's C patients were included in these trials.
Several studies have shown that a decrease in the HVPG of 20% or to below 12 mm Hg is associated with a marked
reduction in recurrent variceal bleeding, whereas treatment failure occurs frequently in patients who do not achieve a
significant decrease in HVPG [113] [114] . As previously mentioned, a significant number of patients either have
contraindications to beta-blocker treatment or cannot continue treatment because of side effects. For those patients who
can tolerate the drug and are compliant in its use, it appears to be a good choice for initial therapy to prevent recurrent
variceal bleeding.
Endoscopic therapy
Sclerotherapy.
Meta-analysis of eight RCTs [115] [116] [117] [118] [119] [120] [121] [122] comparing sclerotherapy with a medically treated
control group for the prevention of recurrent variceal bleeding shows that both rebleeding and mortality were lower in the
patients receiving sclerotherapy to obliterate varices [15] [112] . Although the studies did not receive high scores for quality
of design, the results are homogeneous with most trials showing a significant benefit for both prevention of rebleeding and
improved survival. Several quality of design issues are important in interpreting the results. A clear definition of a bleeding
episode such as a minimum transfusion requirement or drop in hematocrit is lacking in most of the trials. One study
included only bleeding episodes occurring at least 1 month after admission to the study [122] . In most of the trials, bleeding
episodes in the control patients were treated with vasoactive drugs and balloon tamponade rather than sclerotherapy. In the
two trials in which control patients received sclerotherapy acutely for all bleeding episodes, there was no significant
difference in survival [116] [121] .
Longer term follow-up of several sclerotherapy trials reveals a rebleeding rate of 50-60%, although the bleeding episodes
are generally less severe. In addition, after obliteration of esophageal varices by sclerotherapy, bleeding from other sites
such as portal hypertensive gastropathy or gastric varices increases in frequency and severity [123] [124] [125] .
Endoscopic variceal ligation.
There are five published RCTs and several studies in abstract only that have compared endoscopic variceal ligation with
sclerotherapy for the prevention of recurrent variceal bleeding [22] [23] [24] [25] [26] . A recent meta-analysis of the RCTs
concludes that treatment with variceal ligation is associated with a decreased risk of recurrent variceal bleeding, a
decreased rate of death due to bleeding, and an overall improved survival when compared with sclerotherapy [126] . In
addition, variceal ligation requires fewer sessions to obliterate esophageal varices and has fewer complications associated
with its use. The mean follow-up period in all five studies was less than 1 year. Whether the rate of rebleeding will
increase with longer follow-up in patients treated with variceal ligation remains to be determined.
Sclerotherapy versus propranolol.
Eight published RCTs [110] [127] [128] [129] [130] [131] [132] [133] have compared sclerotherapy with propranolol for the
prevention of recurrent variceal hemorrhage. Three studies favor sclerotherapy [127] [128] [133] for prevention of recurrent
variceal bleeding, one favors propranolol [129] , and four show no difference [110] [130] [131] [132] . The heterogeneity in the
results suggest that if one therapy is superior the differences are small [15] [16] . Only one RCT showed a survival benefit, in

favor of sclerotherapy [124] but the meta-analysis failed to show a survival advantage for either therapy [15] [16] .
Combined endoscopic and pharmacologic therapy.
It would seem logical to combine sclerotherapy or endoscopic variceal ligation which act locally by obliterating varices,
with nonselective beta-adrenergic blockers to reduce portal pressure in an attempt to further reduce the risk of recurrent
variceal hemorrhage. Seven RCTs have compared sclerotherapy plus propranolol to sclerotherapy alone for the prevention
of recurrent variceal bleeding [134] [135] [136] [137] [138] [139] [140] . Three of these studies [134] [135] [136] found the
combination therapy more effective then sclerotherapy, whereas four [137] [138] [139] [140] reported no difference. None of
these trials reported a survival advantage for either therapy. However, most of the trials were designed so that propranolol
was given over a short term until esophageal varices were obliterated and was then stopped. Because the mean follow-up
in most of these studies was less than a year, there is limited information as to whether combined therapy might decrease
the risk of late recurrent bleeding seen in many of the sclerotherapy trials. Two RCTs compared the combination of
sclerotherapy plus propranolol with propranolol given alone [140] [141] . One study [140] demonstrated a significant
advantage for the combined therapy in the reduction of recurrent variceal bleeding, whereas the second trial [141] showed a
trend in favor of combined therapy. A recent RCT found the combination of nadolol and isosorbide-5-mononitrate to be
superior to sclerotherapy for the prevention of recurrent variceal bleeding [114] . The combination was especially effective
when the HVPG decreased by 20% or fell below 12 mm Hg. Although more trials, especially with longer follow-up,
are needed to further assess the potential benefit of combination therapy, the data to date support a role for combining
endoscopic and pharmacologic approaches.
Treatment for failures of medical therapy.
Although both endoscopic and pharmacologic therapy reduce the risk of recurrent variceal hemorrhage, clinically
significant hemorrhage usually defined as a transfusion requirement of two or more units of blood and/or the development
of postural signs associated with the bleeding episode will recur in 30-40% of patients. Rescue procedures include
radiologically or surgically created shunts and liver transplantation.
P1088
TIPS procedure.
Evaluation of the potential role for TIPS in the prevention of recurrent variceal bleeding is limited by the lack of published
RCTs. In the only published trial [142] , TIPS was found to be more effective than sclerotherapy in the prevention of
recurrent variceal bleeding, but this finding did not result in a survival benefit. Eight abstracts comparing TIPS and
sclerotherapy for the prevention of recurrent variceal bleeding have been presented with only one finding sclerotherapy to
be more effective than the TIPS procedure for prevention of recurrent bleeding [143] . However, this trial dealt with all
patients who had a major variceal bleed, not just those failing medical therapy. In two large studies, TIPS was felt to be
effective in patients who had failed medical therapy [67] [69] [144] . In long-term follow-up of patients treated with TIPS,
there appears to be a 30-40% incidence of portasystemic encephalopathy, which in some cases, is unresponsive to medical
therapy. In addition, there is a significant risk of shunt thrombosis or stenosis, requiring dilation of the stent or,
occasionally, insertion of a new stent. However, because only one RCT has been published in the peer reviewed literature,
a recommendation about the use of TIPS would be premature.
Total and selective shunts.
Portasystemic shunts are very effective at reducing the incidence of recurrent variceal hemorrhage to approximately 10%,
but are complicated by a high incidence of hepatic encephalopathy and hepatic failure. Three RCTs [145] [146] [147] dealing
almost exclusively with patients with alcoholic cirrhosis reported a slight survival advantage for patients receiving a
portacaval shunt when compared with a medical control group, whereas a fourth study [148] failed to show a difference.
Subsequently, the distal splenorenal (selective) shunt was introduced in an attempt to decrease the risk of postoperative
hepatic encephalopathy while preserving prograde portal blood flow and surgical decompression of the variceal system.
Seven RCTs have compared the distal splenorenal shunt (DSRS) with a variety of total shunts [149] . In the four trials
comparing the DSRS with a portacaval shunt, only one report found less encephalopathy in patients receiving the DSRS.
The three trials in which an interposition graft or proximal splenorenal shunt was used as the control operation found less
encephalopathy in patients treated with the DSRS. Further analysis of these studies reveals that the DSRS is associated
with less postoperative encephalopathy in patients with nonalcoholic liver disease. In patients with alcoholic cirrhosis, the

operations are comparable. None of these trials reported a survival advantage for either operation. For selected patients,
especially those with decompensated liver disease, liver transplantation may be the procedure of choice but donor
availability is a limiting factor [150] .
CONCLUSION
beta-Adrenergic blockers have demonstrated efficacy in the prevention of recurrent variceal hemorrhage in patients with
well-compensated cirrhosis. Variceal ligation is preferable to sclerotherapy as the endoscopic technique of choice, based
on a greater reduction in recurrent variceal bleeding, fewer complications and an improved survival when compared with
sclerotherapy. For patients failing medical treatment, a variety of therapies are currently under investigation. However,
only decompressive shunt surgery (selective, partial or total) have been extensively evaluated and shown to be beneficial
in preventing recurrent bleeding with a marginal survival benefit. Liver transplantation is an appropriate alternative for
properly selected patients.
REFERENCES
1. Garceau AJ, Chalmers TC, the Boston Inter-Hospital Liver Group. The natural history of cirrhosis: I. Survival with esophageal varices. N Engl J Med
1963;268:469-73.
2. Cales P, Desmorat H, Vinel JP, et al. Incidence of large oesophageal varices in patients with cirrhosis: Application to prophylaxis of first bleeding. Gut
1990;31:1298-1302.
3. Christensen E, Fauerholdt L, Schlichting P, et al. Aspects of the natural history of gastrointestinal bleeding in cirrhosis and the effects of prednisone.
4. Cales P, Zabotto B, Meskens C, et al. Gastroesophageal endoscopic features in cirrhosis: observer variability, interassociations, and relationships to hepatic
dysfunction. Gastroenterology 1990;98:156-62.
5. The North Italian Endoscopic Club for the Study and Treatment of Esophageal Varices. Prediction of the first variceal hemorrhage in patients with cirrhosis of
the liver and esophageal varices. a prospective multicenter study. N Engl J Med 1988;319:983-9.
6. Gores GJ, Wiesner RH, Dickson ER, et al. Prospective evaluation of esophageal varices in primary biliary cirrhosis: Development, natural history and influence
in survival. Gastroenterology 1989;96:1552-9.
7. Graham D, Smith JL. The course of patients after variceal hemorrhage. Gastroenterology 1981;80:800-9.
8. Garcia-Tsao G, Groszman RJ, Fisher Rl, et al. Portal pressure presence of gastroesophageal varices and variceal bleeding. Hepatology 1985;5:419-24.
9. Buyse M, Firket P, Newdick M, et al. Data analysis for clinical medicine: The quantitative approach to patient care. In: Chalmers TC, ed. Data analysis for
clinical medicine: The quantitative approach to patient care in gastroenterology. Rome: International University Press, 1988.
10. Mantel N, Haentzel W. Statistical aspects of the analysis of data from retrospective studies of disease. J Natl Cancer Inst 1959;22:719-48.
11. Pagliaro L, D'Amico G, Sorensen TIA, et al. Prevention of first bleeding in cirrhosis: A meta-analysis of randomized trials of nonsurgical treatment. Ann Intern
Med 1992;117:59-70.
12. de Franchis R, Pascal JP, Ancona E, et al. Definitions methodology and therapeutic strategies in portal hypertension: A consensus development workshop. J
Hepatol 1992;15:256-61.
13. Caletti G, Brocchi E, Baraldini M, et al. Assessment of portal hypertension by endoscopic ultrasonography. Gastrointest Endosc 1990;36:521-7.
14. Dagradi AE: The natural history of esophageal varices in patients with alcoholic liver cirrhosis. an endoscopic and clinical study. Am J Med 1972;57:520-40.
15. D'Amico G, Pagliaro L, Bosch J. The treatment of portal hypertension: A meta-analytic review. Hepatology 1995;22:332-54.
16. Grace ND. Management of portal hypertension. Gastroenterologist 1993;1:39-58.
17. Burroughs AK, Hamilton G, Phillips A, et al. A comparison of sclerotherapy with staple transection of the esophagus for the emergency control of bleeding
from esophageal varices. N Engl J Med 1989;321:857-62.

18. Planas R, Quer JC, Boix J, et al. A prospective randomized trial
P1089
comparing somatostatin and sclerotherapy in the treatment of acute variceal bleeding. Hepatology 1994;20:370-5.
19. Shields R, Jenkins SA, Baxter JN, et al. A prospective randomized controlled trial comparing the efficacy of somatostatin with injection sclerotherapy in the
control of bleeding esophageal varices. J Hepatol 1992;15:128-137.
20. Sung JYJ Chung SCS Lai CW et al. Octreotide infusion or emergency sclerotherapy for variceal hemorrhage. Lancet 1993;342:637-41.
21. Paquet KJ, Feussner H. Endoscopic sclerosis and esophageal balloon tamponade in acute hemorrhage from esophagogastric varices: A prospective controlled
randomized trial. Hepatology 1985;5:580-3.
22. Stiegmann GV, Goff JS, Michaletz-Onody PA, et al. Endoscopic sclerotherapy as compared with endoscopic ligation for bleeding esophageal varices. N Engl J
Med 1992;326:1527-32.
23. Gimson AES, Ramage JK, Panos MZ, et al. Randomized trial of variceal banding ligation versus injection sclerotherapy for bleeding esophageal varices. Lancet
1993;342:391-4.
24. Laine L, El-Newihi HM, Migikovsky B, et al. Endoscopic ligation compared with sclerotherapy for the treatment of bleeding esophageal varices. Ann Intern
Med 1993;119:1-7.
25. Hou MC, Liu HC, Kuo BIT, et al. Comparison of endoscopic variceal injection sclerotherapy and ligation for the treatment of esophageal variceal hemorrhage:
A prospective randomized trial. Hepatology 1995;21:1517-22.
26. Lo GH, Lai KH, Cheng JS, et al. A prospective randomized trial of sclerotherapy versus ligation in the management of bleeding esophageal varices. Hepatology
1995;22:466-71.
27. Grace ND. Variceal hemorrhage; pharmacologic approach. In: McDermott WV, Bothe A, eds. Surgery of the liver. Boston: Blackwell Scientific Publications,
1988:303-14.
28. Conn HO, Ramsby GR, Starer EM, et al. Intra-arterial vasopressin in the treatment of upper gastrointestinal hemorrhage: A prospective controlled trial.
29. Mallory A, Schaeffer JU, Cohen JR, et al. Selective intra-arterial vasopressin infusion for upper gastrointestinal tract hemorrhage: A controlled trial. Arch Surg
1980;115:30-2.
30. Fogel MR, Knauer CM, Andres LL, et al. Continuous intravenous vasopressin in active upper gastrointestinal bleeding: A placebo-controlled trial. Ann Intern
Med 1982;96:565-9.
31. Merigan TP, Plotkin GR, Davidson CS: Effect of intravenously administered posterior pituitary extract on hemorrhage from bleeding esophageal varices. N
Engl J Med 1962;266:134.
32. Chojkier M, Groszmann RJ, Atterburg CE, et al: A controlled comparison of continuous intra-arterial and intravenous infusions of vasopressin in hemorrhage
from esophageal varices. Gastroenterology 1979;77:540-6.
33. Gimson AE, Westaby D, Hegarty J, et al. A randomized trial of vasopressin and vasopressin plus nitroglycerin in the control of acute variceal hemorrhage.
Hepatology 1986;6:410-3.
34. Tsai YT, Lay CS, Lai KH, et al. Controlled trial of vasopressin plus nitroglycerin versus vasopressin alone in the treatment of bleeding esophageal varices.
35. Bosch J, Groszmann RJ, Garcia-Pagan JC, et al. Association of transdermal nitroglycerin to vasopressin infusion in the treatment of variceal hemorrhage: A
placebo-controlled trial. Hepatology 1989;10:962-8.
36. Teres J, Planas R, Panes J, et al. Vasopressin/nitroglycerin infusion versus esophageal tamponade in the treatment of acute variceal bleeding: A randomized
controlled trial. Hepatology 1990;11:964-8.
37. Jenkins SA, Baxter JN, Corbett W, et al. A prospective randomized controlled clinical trial comparing somatostatin and vasopressin in controlling acute variceal
hemorrhage. Br Med J 1985;290:275-8.
38. Bagarani M, Albertini V, Anza M, et al. Effect of somatostatin in controlling bleeding from esophageal varices. Ital J Surg Sci 1987;17:21-6.
39. Burroughs AK, Mccormick PA, Hughes MD, et al. Randomized double-blind placebo-controlled trial of somatostatin for variceal bleeding. emergency control

and prevention of early variceal bleeding. Gastroenterology 1990;99:1388-95.
40. Kravetz D, Bosch J, Teres J, et al. Comparison of intravenous somatostatin and vasopressin infusions in treatment of acute variceal hemorrhage. Hepatology
1984;4:442-6.
41. Cardona C, Vida F, Balanzo J, et al. Eficacia terapeutica de la somatostatina versus vasopressina mas nitroglycerina en la hemorragia activa por varices
esofagogastrica. Gastroenterol Hepatol 1989;12:30-4.
42. Saari A, Klvilaakso E, Inberg M, et al. Comparison of somatostatin and vasopressin in bleeding esophageal varices. Am J Gastroenterol 1990;85:804-7.
43. Hsia HC, Lee FY, Tsai YT, et al. Comparison of somatostatin and vasopressin in the control of acute esophageal variceal hemorrhage: A randomized controlled
study. Chin J Gastroenterol 1990;7:71-8.
44. Huang CC, Sheen IS, Chu CM, et al. A prospective randomized controlled trial of somatostatin and vasopressin in the management of acute bleeding
esophageal varices. Chang Gung Med J 1992;15:78-83.
45. Walker S, Kreichgauer HP, Bode JC. Terlipressin versus somatostatin in bleeding esophageal varices: A controlled, double blind study. Hepatology
1992;15:1023-30.
46. Jaramillo JL, de la Mata M, Mino G, et al. Somatostatin versus Sengstaken balloon tamponade for primary hemostasis of bleeding esophageal varices: A
randomized pilot study. J Hepatol 1991;12:100-5.
47. Avgerinos A, Klonis C, Rekoumis G, et al. A prospective randomized trial comparing somatostatin balloon tamponade and the combination of both methods in
the management of acute variceal hemorrhage. J Hepatol 1991;13:78-83.
48. Valenzuela JE, Schubert T, Fogel MR, et al. A multicenter randomized double-blind trial of somatostatin in the management of acute hemorrhage from
esophageal varices. Hepatology 1989;10:958-61.
49. Hwang SJ, Lin HC, Chang CF, et al. A randomized controlled trial comparing octreotide and vasopressin in the control of acute esophageal variceal bleeding. J
Hepatol 1992;16:320-5.
50. McKee R. A study of octreotide in esophageal varices. Digestion 1990;45:60-5.
51. Besson I, Ingrand P, Person B, et al. Sclerotherapy with or without octreotide for acute variceal bleeding. N Engl J Med 1995;333:555-60.
52. Primignani M, Andreoni B, Carpinelli L, et al. Sclerotherapy plus octreotide versus sclerotherapy alone in the prevention of early rebleeding from esophageal
varices: A randomized double-blind placebo-controlled multicenter trial. Hepatology 1995;21:1322-7.
53. Soderlund C, Magnusson I, Torngren S, et al. Terlipressin (triglycyl-lysine vasopressin) controls acute bleeding esophageal varices: A double-blind randomized
placebo-controlled trial. Scand J Gastroenterol 1990;25:622-30.
54. Walker S, Stiehl A, Raedsch R, et al. Terlipressin in bleeding esophageal varices: A placebo-controlled double-blind study. Hepatology 1986;6:112-5.
55. Freeman JG, Cobden I, Lishman AH, et al. Controlled trial of terlipressin (Glypressin) versus vasopressin in the early treatment of esophageal varices. Lancet
1982;2:66-8.
56. Colin R, Giuli N, Czernichow P, et al. Prospective comparison of Glypressin tamponade and their association in the treatment of bleeding esophageal varices.
In: Lebrec D, Blei AT, eds. Vasopressin analogs and portal hypertension . Paris: J. Libbey Eurotext, 1987:149-53.
57. Fort E, Sautereau D, Silvain C, et al. A randomized trial of terlipressin plus nitroglycerin versus balloon tamponade in the control of acute variceal hemorrhage.
Hepatology 1990;11:678-81.
58. Chang TT, Lee FY, Tsai YT, et al. A randomized controlled study of low-dose and high-dose terlipressin in the control of acute esophageal variceal
hemorrhage. J Gastroenterol Hepatol 1991;6:481-4.
59. Silvain C, Carpentier S, Sautereau D, et al. Terlipressin plus transdermal nitroglycerin versus octreotide in the control of acute bleeding from esophageal
varices: A multicenter randomized trial. Hepatology 1993;18:61-5.
60. de Franchis R, Arcidiacono PG, Carpinelli L, et al. Randomized controlled trial of desmopressin plus terlipressin versus terlipressin alone for the treatment of
acute variceal hemorrhage in cirrhotic patients: A multicenter, double-blind study. Hepatology 1993;18:1102-7.
61. Mastai R, Grande L, Bosch J, et al. Effects of metoclopramide and domperidone on azygos venous blood flow in patients with cirrhosis and portal hypertension.
62. Kleber G, Sauerbruch T, Fischer G, et al. Reduction of transmural esophageal variceal pressure by metoclopramide. J Hepatol 1991;12:362-6.

P1090
63. Hosking SW, Doss W, El-Zeiny H, et al. Pharmacologic constriction of the lower esophageal sphincter: A simple method of arresting variceal hemorrhage. Gut
1988;29:1098-102.
64. McCormick PA, Kaye GL, Greenslade L, et al. Esophageal staple transection as a salvage procedure after failure of acute injection sclerotherapy. Hepatology
1992;15:403-6.
65. Huizinga WK, Angorn IB, Baker LW. Esophageal transection versus injection sclerotherapy in the management of bleeding esophageal varices in patients at
high risk. Surg Gynecol Obstet 1985;160:539-46.
66. Teres J, Baroni R, Bordas JM, et al. Randomized trial of portacaval shunt stapling transection and endoscopic sclerotherapy in uncontrolled variceal bleeding. J
Hepatol 1987;4:159-67.
67. Rossle M, Haag K, Ochs A, et al. The transjugular intrahepatic portosystemic stent-shunt procedure for variceal bleeding. N Engl J Med 1994;330:165-71.
68. Grace ND. The side-to-side portacaval shunt revisited. N Engl J Med 1994;330:208-9.
69. Laberge JM, Ring EJ, Gordon R, et al. Creation of transjugular intrahepatic portosystemic shunts with the Wallstent endoprosthesis: Results in 100 patients.
Radiology 1993;187:413-20.
70. Pascal JP, Cales P, and a Multicenter Study Group. Propranolol in the prevention of first upper gastrointestinal tract hemorrhage in patients with cirrhosis of the
liver and esophageal varices. N Engl J Med 1987;317:856-61.
71. The Italian Multicenter Project for Propranolol in Prevention of Bleeding. Propranolol prevents first gastrointestinal bleeding in nonascitic cirrhotic patients:
Final report of a multicenter randomized trial. J Hepatology 1989;9:75-83.
72. Ideo G, Bellati G, Fesce E, et al. Nadolol can prevent the first gastrointestinal bleeding in cirrhosis: A prospective, randomized study. Hepatology 1988;8:6-9.
73. Lebrec D, Poynard T, Capron JP, et al. Nadolol for prophylaxis of gastrointestinal bleeding in patients with cirrhosis: A randomized trial. J Hepatol
1988;7:118-25.
74. Conn HO, Grace ND, Bosch J, et al. Propranolol in the prevention of the first hemorrhage from esophageal varices: Results of a randomized double blind
cooperative clinical trial. Hepatology 1991;13:902-12.
75. Andreani T, Poupon RE, Balkau B, et al. Preventive therapy of first gastrointestinal bleeding in patients with cirrhosis: Results of a controlled trial comparing
propranolol, endoscopic sclerotherapy and placebo. Hepatology 1990;12:1412-9.
76. The Prova Study Group. Prophylaxis of first hemorrhage from esophageal varices by sclerotherapy propranolol or both in cirrhotic patients: A randomized
multicenter trial. Hepatology 1991;14:1016-24.
77. Colman J, Jones P, Finch C, et al. Propranolol in the prevention of variceal hemorrhage in alcoholic cirrhotic patients. Hepatology 1990;12:851.
78. Poynard T, Cales P, Pasta L, et al. Beta-adrenergic-antagonist drugs in the prevention of gastrointestinal bleeding in patients with cirrhosis and esophageal
varices: An analysis of data and prognostic factors in 589 patients from four randomised clinical trials. N Engl J Med 1991;324:1532-8.
79. Grace ND. Prophylaxis of first variceal bleeding. drug treatment. In: Holstege A, Hahn EG, Scholmerich J, eds. Portal hypertension (Folk Symposium 79).
Dordrecht, The Netherlands: Kluwer Academic Publishers, 1995:269-77.
80. Lopes GM, Grace ND: Gastroesophageal varices: Prevention of bleeding and rebleeding. In: Friedman M, ed. Gastroenterology clinics of North America:
Gastrointestinal bleeding, vol I, Philadelphia: WB Saunders, 1993:801-20.
81. Groszmann RJ, Bosch J, Grace ND, et al. Hemodynamic events in a prospective randomized trial of propranolol versus placebo in the prevention of a first
variceal hemorrhage. Gastroenterology 1990;99:1401-7.
82. Angelico M, Carli R, Piat C, et al. Isosorbide-5-mononitrate versus propranolol in the prevention of first bleeding in cirrhosis. Gastroenterology
1993;104:1460-5.
83. Grace ND, Conn HO, Groszmann RJ, et al. Propranolol for prevention of first variceal hemorrhage: A life-time commitment? Hepatology 1990;12:407.
84. Grace ND. Prevention of initial variceal hemorrhage. In: Groszmann RJ, Grace ND, eds. Gastroenterology Clinics of North America: Complications of portal
hypertension: esophagogastric varices and ascites. Philadelphia: WB Saunders, 1992:149-61.

85. Paquet HJ: Prophylactic endoscopic sclerosing treatment of the esophageal wall in varices: A prospective controlled randomized trial. Endoscopy 1982;14:4-5.
86. Witzel L, Wolbergs E, Merki H. Prophylactic endoscopic sclerotherapy of esophageal varices: A prospective controlled trial. Lancet 1985;1:773-5.
87. The Veterans Affair Cooperative Variceal Sclerotherapy Group. Prophylactic sclerotherapy for esophageal varices in alcoholic liver disease: A randomized,
single-blind, multicenter clinical trial. N Engl J Med 1991;324:1779-84.
88. Koch H, Hennings H, Grimm H, et al. Prophylactic sclerosing of esophageal varices: Results of a prospective controlled study. Endoscopy 1986;18:40-3.
89. Kobe VE, Schentke KU: Endoskopische sklerotherapic (EST) von Osophagusvarizen-studien sum technik und sum krankheitsverlauf. Dtsch Z Verdau
Stoffwecks Krankh 1987;47:151-7.
90. Wordehoff D, Spech HJ. Prophylaktishe oesophagusvarizen sklerosierung. Dtsch Med Wochenschr 1987;112:947-51.
91. Piai G, Cipolletta L, Claar M, et al. Prophylactic sclerotherapy of high-risk esophageal varices: Results of a multicentric prospective controlled trial. Hepatology
1988;8:1495-500.
92. Santangelo WC, Dueno MI, Estes BL, et al. Prophylactic sclerotherapy of large esophageal varices. N Engl J Med 1988;318:814-8.
93. Sauerbruch T, Wotzka R, Kopcke W, et al. Prophylactic sclerotherapy before the first episode of variceal hemorrhage in patients with cirrhosis. N Engl J Med
1988;319:8-14.
94. Potzi R, Bauer P, Reichel W, et al. Prophylactic endoscopic sclerotherapy of esophageal varices in liver cirrhosis: A multicenter prospective controlled
randomised trial in Vienna. Gut 1989;30:873-9.
95. Russo A, Giannone G, Magnano A, et al. Prophylactic sclerotherapy in nonalcoholic liver cirrhosis: Preliminary results of a prospective controlled randomised
trial. World J Surg 1989;13:149-53.
96. Triger DR, Smart HL, Hosking SW, et al. Prophylactic sclerotherapy for esophageal varices: Long term results of a single center trial. Hepatology
1991;13:117-23.
97. de Franchis R, Primignani M, Arcidiacono PG, et al. Prophylactic sclerotherapy (ST) in high-risk cirrhotics selected by endoscopic criteria. A multicenter
randomized controlled trial. Gastroenterology 1991;101:1087-93.
98. Conn HO. The rational evaluation and management of portal hypertension. In: Schaffner F, Sherlock S, Leevy CM, eds. The liver and its diseases. New York:
Intercontinental, 1974:289-306.
99. Conn HO, Lindenmuth WW, May CJ, et al. Prophylactic portacaval anastomosis: A tale of two studies. Medicine 1972;51:27-40.
100. Jackson FC, Perrin EB, Smith AG, et al. A clinical investigation of the portacaval shunt: Survival analysis of the prophylactic operation. Am J Surg
1968;11:22-42.
101. Resnick RH, Chalmers TC, Ishihara AM, et al. A controlled study of the prophylactic portacaval shunt: A final report. Ann Intern Med 1969;70:675-88.
102. Inokuchi K, Cooperative Study Group OF Portal Hypertension of Japan. Improved survival after prophylactic portal non-decompression surgery for
esophageal varices: A randomized clinical trial. Hepatology 1990;12:1-6.
103. Lebrec D, Poynard T, Bernuau J, et al. A randomized controlled study of propranolol for prevention of recurrent gastrointestinal bleeding in patients with
cirrhosis: A final report. Hepatology 1984;4:355-8.
104. Villeneuve JP, Pomier-Layrargues G, Infante-Rivard C, et al. Propranolol for the prevention of recurrent variceal hemorrhage: A controlled trial. Hepatology
1986;6:1239-43.
105. Queuniet AM, Czernichow P, Lerebours E, et al. Etude controllee du propranolol dans la prevention des recidives hemorragiques chez les patients cirrhotiqes.
Gastroenterol Clin Biol 1987;11:41-47.
106. Gatta A, Merkel C, Sacerdoti D, et al. Nadolol for prevention of variceal rebleeding in cirrhosis: A controlled clinical trial. Digestion 1987;37:22-8.
107. Colombo M, de Franchis R, Tommasini M, et al. Beta blockade prevents recurrent gastrointestinal bleeding in well-compensated patients with alcoholic
cirrhosis: A multicenter randomized controlled trial. Hepatology 1989;9:433-8.
108. Sheen IS, Chen TY, Liaw YF. Randomized controlled study of

P1091
propranolol for prevention of recurrent esophageal varices bleeding in patients with cirrhosis. Liver 1989;9:1-5.
109. Garden OJ, Mills PR, Birnie GG, et al. Propranolol in the prevention of recurrent variceal hemorrhage in cirrhotic patients: A controlled trial.
110. Rossi V, Cales P, Pascal B, et al. Prevention of recurrent variceal bleeding in alcoholic cirrhotic patients: Prospective controlled trial of propranolol and
sclerotherapy. J Hepatol 1991;12:283-9.
111. Burroughs AK, Jenkins WJ, Sherlock S, et al. Controlled trial of propranolol for the prevention of recurrent variceal hemorrhage in patients with cirrhosis. N
Engl J Med 1983;309:1539-42.
112. Burroughs AK, Mccormick PA: Prevention of variceal rebleeding. In: Groszmann RJ, Grace ND, eds. Gastroenterology clinics of North America:
Complications of portal hypertension: Esophagogastric varices and ascites. Philadelphia: WB Saunders, 1992:119-47.
113. Feu F, Garcia-Pagan JC, Bosch J, et al. Relation between portal pressure response to pharmacologic therapy and risk of recurrent variceal hemorrhage in
patients with cirrhosis Lancet 1995;346:1056-9.
114. Villanueva C, Balanzo J, Novella MT, et al. Nadolol plus isosorbide mononitrate compared with sclerotherapy for the prevention of variceal rebleeding. N
Engl J Med 1996;334:1624-9.
115. Barsoum MS, Bolous FI, El-Rooby AA, et al. Tamponade and injection sclerotherapy in the management of bleeding esophageal varices. Br J Surg
1982;69:76-8.
116. Burroughs AK, Mccormick PA, Siringo S, et al. Prospective randomized trial of long term sclerotherapy for variceal rebleeding using the same protocol to
treat rebleeding in all patients: Final report. Hepatology 1989;10:579 (abstract).
117. The Copenhagen Esophageal Varices Sclerotherapy Project. Sclerotherapy after first variceal hemorrhage in cirrhosis: A randomized multicenter trial. N Engl
J Med 1984;311:1594-600.
118. Gregory PB, The Va Cooperative Variceal Sclerotherapy Group, et al. Sclerotherapy for male alcoholics with cirrhosis who have bled from esophageal
varices: Results of a randomized multicenter trial. Hepatology 1994;20:618-25.
119. Korula J, Balart LA, Radvan G, et al. A prospective randomized controlled trial of chronic esophageal variceal sclerotherapy. Hepatology 1985;5:584-9.
120. Soderlund C, Ihre T. Endoscopic sclerotherapy versus conservative management of bleeding esophageal varices: A 5 year prospective controlled trial of
emergency and long term treatment. Acta Chir Scand 1985;151:449-56.
121. Terblanche J, Bornman PC, Kahn D, et al. Failure of repeated injection sclerotherapy to improve long-term survival after esophageal variceal bleeding: A five
year prospective controlled clinical trial. Lancet 1983;2:1328-32.
122. Westaby D, Macdougall BR, Williams R. Improved survival following injection sclerotherapy for esophageal varices: Final analysis of a controlled trial.
123. Henderson J, Kutner MH, Millikan WJ, et al. Endoscopic variceal sclerosis compared with distal splenorenal shunt to prevent recurrent variceal bleeding in
cirrhosis: A prospective randomized trial. Ann Intern Med 1990;122:262-9.
124. D'Amico G, Montalbarro L, Traina M, et al. Natural history of congestive gastropathy. Gastroenterology 1990;99:1558-64.
125. Tanoue K, Hashizume M, Wada H, et al. Effects of endoscopic injection sclerotherapy on portal hypertensive gastropathy: A prospective study. Gastrointest
Endosc 1992;38:582-5.
126. Laine L, Cook D. Endoscopic ligation compared with sclerotherapy for treatment of esophageal variceal bleeding: A meta-analysis. Ann Intern Med
1995;123:280-7.
127. Dasarathy S, Divinedi M, Bharagava DK, et al. A prospective randomized trial comparing repeated endoscopic sclerotherapy and propranolol in
decompensated (Child's B and C) cirrhosis patients. Hepatology 1992;16:89-94.
128. Teres J, Bosch J, Bordas JM. Propranolol versus sclerotherapy in preventing variceal rebleeding: A randomized controlled trial. Gastroenterology
1993;105:1508-14.
129. Dollet JM, Champigneulle B, Patris A, et al. Sclerotherapic endoscopique contrepropranolol apres hemorrhagic par rupture des varices oesophargrennes chez
le cirrhotique: Resultats a 4 an d'une etude randomisee. Gastroenterol Clin Biol 1988;12:234-9.
130. Westaby D, Polson RJ, Gimson AES, et al. A controlled trial of oral propranolol compared with injection sclerotherapy for long term management of variceal

bleeding. Hepatology 1990;11:353-9.
131. Fleig WF, Stange EF, Hunecke R, et al. Prevention of recurrent bleeding in cirrhosis with recent variceal hemorrhage. prospective, randomized comparison of
propranolol and sclerotherapy. Hepatology 1987;7:355-61.
132. Martin T, Taupignon A, Lavignolle A, et al. Prevention des recidives hemorragiques chez des malades atteints de cirrhose: Resultats d'une etude controle
comparant propranolol et sclerose endoscopique. Gastroenterol Clin Biol 1991;15:833-7.
133. Alexandrino PT, Alves MM, Correia JP. Propranolol or endoscopic sclerotherapy in the prevention of recurrence of variceal bleeding: A prospective
randomized controlled trial. J Hepatol 1988;7:175-85.
134. Vinel JP, Lamouliate H, Cales P, et al. Propranolol reduces the rebleeding rate during endoscopic sclerotherapy before variceal obliteration. Gastroenterology
1992;102:1760-1763.
132A. Jensen LS, Krapup N. Propranolol in prevention of rebleeding from esophageal varices during the course of endoscopic sclerotherapy. Scand J Gastroenterol
1989;24:339-45.
135. Avgerinos A, Rekoumis G, Klonis C, et al. Propranolol in the prevention of recurrent upper gastrointestinal bleeding in patients with cirrhosis undergoing
endoscopic sclerotherapy: A randomized controlled trial. J Hepatol 1993;19:301-11.
136. Westaby D, Melia W, Hegarty J, et al. Use of propranolol to reduce the rebleeding rate during injection sclerotherapy prior to variceal obliteration. Hepatology
1986;6:673-5.
137. Lundell L, Leth R, Lind T, et al. Evaluation of propranolol for prevention of recurrent bleeding from esophageal varices between sclerotherapy sessions. Acta
Chir Scand 1990;156:711-5.
138. Vicker C, Rhodes J, Chesner I, et al. Prevention of rebleeding from esophageal varices: Two year follow-up of a prospective controlled trial of propranolol in
addition to sclerotherapy. J Hepatol 1994;21:91-87.
139. Acharya KS, Dasarthry S, Saksena S, et al. A prospective randomized study to evaluate propranolol in patients undergoing long-term endoscopic
sclerotherapy. J Hepatol 1993;19:291-300.
140. O'Connor KW, Lehman G, Yune H, et al. Comparison of three nonsurgical treatments for bleeding esophageal varices. Gastroenterology 1989;96:899-906.
141. Ink O, Martin T, Poynard T, et al. Does elective sclerotherapy improve the efficacy of long-term propranolol for prevention of recurrent bleeding in patients
with severe cirrhosis? A prospective multicenter randomized trial. Hepatology 1992;16:912-9.
142. Cabrera J, Maynar M, Granados R, et al. Transjugular intrahepatic portosystemic shunt versus sclerotherapy in the elective treatment of variceal hemorrhage.
143. Sanyal AJ, Freedman AM, Purdum PP, et al. Transjugular intrahepatic portosystemic shunt (TIPS) versus sclerotherapy for prevention of recurrent variceal
hemorrhage: A randomized prospective trial. Gastroenterology 1994;106:A975.
144. La Berge JM, Somberg KA, Lake JR, et al. Two-year outcome following transjugular intrahepatic portosystemic shunt for variceal bleeding: Results in 90
patients. Gastroenterology 1995;108:1143-51.
145. Jackson FC, Perrin EB, Smith AG, et al. A clinical investigation of the portacaval shunt. II. Survival analysis of the therapeutic operation. Ann Surg
1971;174:672-701.
146. Resnick RH, Iber FL, Ishihara A, et al. A controlled study of the therapeutic portacaval shunt. Gastroenterology 1974;67:843-57.
147. Reynolds RB, Donavan AJ, Mikkelsen WP, et al. Results of a 12-year randomized trial of portacaval shunt in patients with alcoholic liver disease and bleeding
varices. Gastroenterology 1981;80:1005-11.
148. Rueff B, Degos F, Degos JD, et al. A controlled study of therapeutic portacaval shunt in alcoholic cirrhosis. Lancet 1976;1:655-9.
149. Grace ND, Conn HO, Resnick RH, et al. Distal splenorenal versus portal-systemic shunts after hemorrhage from varices: A randomized controlled trial.
Hepatology 1988;8:1475-81.
150. Henderson JM, Gilmore GT, Hooks MA, et al. Selective shunt in the management of variceal bleeding in the era of liver transplantation. Ann Surg
1992;216:248-55.

Bookmark URL: /das/guideline/view/N/9451414?ja=67694&PAGE=1.html&ANCHOR=top&source=HS,MI

April 22, 2000
Conde Petra
Technology Assessment Status Evaluation

update: transvenous intrahepatic portosystemic
shunt (TIPS). April, 1997. ASGE. American
Citation Society for Gastrointestinal Endoscopy.
Bibliographic Data
Indexing Data
Copyright Notice and Disclaimer
Technology Assessment Status
Evaluation update: transvenous
intrahepatic portosystemic shunt (TIPS).
Journal Articles On This Topic
April, 1997. ASGE. American Society for
Gastrointestinal Endoscopy.
Full Text
Gastrointest Endosc - 1998 Jun; 47(6): 584-7
Frontmatter From NIH/NLM MEDLINE, HealthSTAR
INTRODUCTION NLM Citation ID:
98309619
TIPS FOR BLEEDING
Full Source Title:
TIPS FOR ASCITES
COMPLICATIONS
Publication Type:
SUMMARY OF INDICATIONS AND Guideline; Journal Article; Practice Guideline; Review;
CONTRAINDICATIONS Review, Tutorial
REFERENCES Language:
English
Number of References:
24
Major Subjects:
Ascites / * Surgery
Gastrointestinal Hemorrhage / * Surgery
Portasystemic Shunt, Transjugular Intrahepatic / Adverse

Effects / * Methods
* Technology Assessment, Biomedical
Human
Randomized Controlled Trials
Treatment Outcome


Volume 47 Number 6 June 1998
584
Guidelines
Technology Assessment Status Evaluation: Transvenous

Intrahepatic Portosystemic Shunt (TIPS)
David L. Carr-Locke MD, Chair

Stanley M. Branch MD
William J. Byrne MD
Mitchell I. Conn MD
Karen Laing RN, CGRN
Douglas B. Nelson MD
Bret T. Petersen MD
Eduard Phillips MD
Irving Waxman MD
APRIL, 1997
Prepared by: Technology Assessment Committee
For reprints please contact: American Society for Gastrointestinal Endoscopy 13 Elm Street Manchester, MA 01944
INTRODUCTION
A Technology Assessment Report on transvenous (transjugular) intrahepatic portosystemic shunts (TIPS)
was published in 1993[1] . Publications since then include long-term data from uncontrolled studies,
randomized trials, a National Digestive Diseases Advisory Board conference[2] and several reviews[3] [4] .
The summary of a 1996 NIH Consensus Conference on TIPS should soon be available. This document
provides an update on the current status of TIPS.
http://home.mdconsult.com/das/article/body/jorg=g...ne&source=HS,MI&sp=10302543&sid=0/N/113906/1.html (1 of 6) [22/04/2000 0:08:39]

TIPS FOR BLEEDING

Hemodynamic studies during and after TIPS correlated reduction of the portosystemic venous pressure
gradient to less than 12 mm Hg with reduction in rebleeding risk, and the recurrence of portal
hypertension with rebleeding[5] . This is analogous to prior findings for both surgical shunting and
pharmacotherapy. Recurrence of portal hypertension, with or without bleeding, has been described in up
to 90% of patients two years after TIPS[5] .
Several uncontrolled studies have now provided medium to long term follow-up on the use of TIPS for
variceal hemorrhage. Among 90 patients followed for an average of 2.2 years, rebleeding occurred in
26% at one year and 32% at two years. Rebleeding was noted only among the 31% and 47% with shunt
stenosis or occlusion at 1 and 2 years, respectively. Percutaneous shunt revision was successful in 95% of
attempts [6] . A survey of 10 institutions treating 1750 patients reported 97% (range 92-99%) technical
procedural success and 91% (50 - 100%) control in the subset of patients with acute hemorrhage.
Recurrent hemorrhage occurred in 17% (2-26%) within one year and 20.7% (5-32%) by two years [7] .
A randomized controlled trial of 63 patients evaluated TIPS versus endoscopic sclerotherapy for
prevention of recurrent esophageal variceal bleeding [8] . During a mean follow-up period of 15 months
the TIPS group exhibited less rebleeding (23% vs 52.6%, p < 0.02), lower transfusion requirements (p <
0.03) and fewer hospital days for rebleeding (p < 0.05). Encephalopathy was more common following
TIPS (33% vs 13%, p < 0.05) but other complications and mortality were not different. On multivariate
analysis only the treatment modality was an independent predictor of variceal rebleeding. Both age and
treatment modality were predictors for the development of encephalopathy.
Additional randomized controlled trials published in abstract form include four comparing TIPS to
endoscopic variceal sclerotherapy alone [9] [10] [11] [12] , three comparing TIPS to sclerotherapy plus beta
blockade [13] [14] [15] , and one comparing TIPS to endoscopic band ligation [16] . Six of eight studies
reported significantly diminished rates of rebleeding following TIPS, one showed a trend in the same
direction and one found no difference. One study each showed significantly higher and lower mortality
rates following TIPS. The others showed no differences in mortality. Four out of seven reported
significantly higher rates of encephalopathy after TIPS than after endoscopic therapy. Approximately 5%
of the TIPS procedures in the randomized studies were technical failures. Rebleeding after endoscopic
management commonly required crossover to TIPS, while rebleeding after TIPS was usually
585
managed by angiography with shunt dilation or revision.

One study compared TIPS with surgical shunts [17] . Seventy patients who failed or were not candidates
for endoscopic sclerotherapy were randomized to TIPS vs. surgical placement of a small diameter
portacaval H-graft. Major variceal rebleeding occurred in 11% of TIPS patients and in none of the
surgical patients. Shunt failure from all causes was significantly more common following TIPS than
following surgery (57% vs 26%, p < 0.02). There were no significant differences in encephalopathy or
mortality.
While surveillance doppler ultrasonography has been used to detect overt shunt occlusion it may be
insensitive for partial stenosis and shunt dysfunction [5] . Angiographic assessment should be performed

for all episodes of recurrent portal hypertensive bleeding [2] [3] [4] .
TIPS FOR ASCITES

Placement of an intrahepatic shunt results in an improvement in glomerular filtration rate, a decline in
pathologically elevated levels of renin, aldosterone, and epinephrine, and an increase in atrial natriuretic
factor [18] [19] . Sustained natriuresis, as a result of these combined neuro-hormonal alterations, is
associated with control of ascites.
Uncontrolled studies have suggested a role for the use of TIPS in the treatment of medically intractable
ascites [2] . A complete response was seen in 74% and partial response in 18% of 50 patients after 3
months in one recent study [18] . A randomized controlled trial compared TIPS to large volume
paracentesis. TIPS was successful in 10 of 13 patients and paracentesis was performed in 12 patients [19] .
TIPS was superior for control of ascites only in Child-Pugh class B patients, but there was no associated
improvement in survival. Among eight class C patients the TIPS group had a statistically greater risk of
death at 1 and 2 years. In a randomized study reported in abstract form TIPS resulted in a significant
reduction in the need for paracentesis and in overall treatment failures. There was a trend toward fewer
hospital days following TIPS, but no difference in survival [20] .
A randomized trial comparing TIPS to peritoneovenous shunting for intractable ascites in 22 patients has
been reported in abstract form [21] . The two procedures were equivalent in controlling ascites and
survival. TIPS patients required a greater number of ICU days but fewer repeat interventions to maintain
shunt patency.
Improvement in ascites following TIPS may be delayed or incomplete in older patients and in those with
pre-existing renal dysfunction [22] . Response has not been predictable by pretreatment serum creatinine
levels nor by pre or post treatment portal-venous pressure gradients [18] .
COMPLICATIONS
Technical failure occurs in 5% of patients and early procedural morbidity and mortality are 10% and
1-2%, respectively [2] [5] . Acute complications are primarily bleeding, infection, and aggravation of
cardiopulmonary dysfunction. Functionally important shunt stenosis occurs in 20 - 50% of patients over
1-2 years, however with serial intravascular revisions patency and function can be maintained in the
majority [3] [6] [23] . Hepatic encephalopathy occurs in up to 30% of patients after TIPS. Predictive factors
have included older age, severity of disease and prior history of encephalopathy [24] . Most
encephalopathy is medically controllable and angiographic revision or occlusion may be useful in the
most severe or intractable cases. Other chronic complications include portal or splenic vein thrombosis
(1-15%), worsening of hepatic function (1-5%), and chronic hemolysis (1-3%).
SUMMARY OF INDICATIONS AND CONTRAINDICATIONS

Since the last technology assessment report on TIPS in 1993, the data has confirmed prior impressions
regarding the frequency of encephalopathy, early shunt dysfunction, and requirements for repeated
intervention. This limits its optimal application to those who are candidates for more definitive therapy,
i.e. transplantation. Bleeding indications continue to be: 1) active variceal bleeding that cannot be
controlled acutely with medical and endoscopic therapy, and 2) prevention of recurrent variceal

586
bleeding in patients who are refractory or intolerant to combined endoscopic and pharmacologic
management. TIPS is not recommended for primary prevention in those who have not bled and there is
insufficient data to support TIPS as the initial therapy for active variceal bleeding or for elective
secondary Prevention of bleeding.
The role of TIPS for therapy of intractable ascites is still unclear. The efficacy of TIPS for relieving
ascites has been substantiated however there may be considerable risk in some subsets of patients.
TIPS has been described for therapy of hepatic hydrothorax, hepato-pulmonary syndrome, Budd-Chiari
syndrome, portal hypertensive gastropathy and chylous fistula complicating cirrhosis.
TIPS is contraindicated in those with right-sided heart failure and elevated central venous pressure,
polycystic liver disease, severe hepatic failure, or hepatic neoplasms within the likely shunt path.
Relative contraindications include active intrahepatic or systemic infection, severe hepatic
encephalopathy inadequately controlled by medical therapy, and portal vein thrombosis.
REFERENCES
1. Technology Assessment Committee of the ASGE. Technology assessment status evaluation: transjugular intrahepatic
portosystemic shunt. Gastrointest Endosc 1993;39:889-91.
2. Shiffman ML, Jeffers L, Hoofnagle JH, Tralka TS. The role of transjugular intrahepatic portosystemic shunt for
treatment of portal hypertension and its complications: a conference sponsored by the National Digestive Diseases
Advisory Board. Hepatology 1995;22:1591-7.
3. Kamath PS, McKusick MA. Viewpoints in digestive diseases: transvenous intrahepatic portosystemic shunts.
4. Grace ND. TIPS: The long and the short of It. Gastroenterology 1997;112:1040-3.
5. Sanyal AJ, Freedman AM, Luketic VA, Purdum PP, Shiffman ML, DeMeo J, et al. The natural history of portal
hypertension after transjugular intrahepatic portosystemic shunt. Gastroenterology 1997;112:889-98.
6. LaBerge JM, Somberg KA, Lake JR, Gordon RL, Kerlan RK, Ascher NL, et al. Two-year outcome following
transjugular intrahepatic portosystemic shunt for variceal bleeding: results in 90 patients. Gastroenterology
1995;108:1143-51.
7. Barton RE, Rosch J, Saxon RR, Lakin PC, Petersen BD, Keller FS. TIPS: short- and long-term results: a survey of 1750
patients. Semin in Intervent Radiol 1995;12:364-7.
8. Cabrera J, Maynar M, Granados R, Gorriz E, Reyes R, Pulido-Duque JM, et al. Transjugular intrahepatic portosystemic
shunt versus sclerotherapy in the elective treatment of variceal hemorrhage. Gastroenterology 1996;110:832-9.
9. Sanyal AJ, Freedman AM, Purdum PP, Luketic VA, Shiffman ML, Tisnado J, et al. Transjugular intrahepatic
portosystemic shunt (TIPS) vs sclerotherapy for prevention of recurrent variceal hemorrhage: a randomized prospective
trial. [abstract] Gastroenterology 1994;106:A975.

10. CelloJP, Ring EJ, Olcott E, Koch J, Gordon R, Sandhu F, et al. Transjugular intrahepatic portosystemic shunt (TIPS) vs
sclerotherapy (ES) for variceal hemorrhage (VH) [abstract]. Gastroenterology 1995;108:A1045.
11. MerliM, Riggio O, Capocaccia L, Rossi P, Salerno F, De Franchis R, et al. TIPS vs sclerotherapy for prevention of
variceal of bleeding: results of a randomized controlled trial [abstract]. Gastroenterology 1996;110:A1265.
12. Garcia-Villarreal
L, Martinez-Lagares F, Sierra A, Guevara C, Hernandez-Cabrero T, et al. TIPS vs sclerotherapy
(SCL) for the prevention of variceal rebleeding. Preliminary results of a randomized study [abstract]. Hepatology
1996;24:208A.
587
13. RossleM, Deibert P, Haag K, Ochs A, Siegerstetter V, Langer M. TIPS versus sclerotherapy and beta-blockade:
preliminary results of a randomized study in patients with recurrent variceal hemorrhage [abstract]. Hepatology
1994;20:107A.
14. Groupe d'Etude des Anastomoses Intra-Hepatiques. TIPS vs sclerotherapy + propranolol in the prevention of variceal
rebleeding: preliminary results of a multicenter randomized trial [abstract]. Hepatology 1995;22:297A.
15. SauerP, Theilmann L, Benz C, Roeren T, Richter G, Stremmel W, et al. Transjugular intrahepatic portosystemic stent
shunt (TIPS) vs sclerotherapy in the prevention of variceal rebleeding: a randomized study [abstract]. Gastroenterology
1996;110:A1313.
16. Jalan
R, Forrest EH, Stanley AJ, Redhead DN, Dillon JF, Finlayson NDC, et al. TIPPS vs variceal band ligation for the
prevention of variceal rebleeding in cirrhosis: a randomized controlled study [abstract]. Hepatology 1996;24:247A.
17. Rosemurgy AS, Goode SE, Zwiebeel BR, Black TJ, Brady PG. A prospective trial of transjugular intrahepatic
portosystemic stent shunts versus small-diameter prosthetic H-graft portacaval shunts in the treatment of bleeding varices.
Ann Surg 1996;224:378-86.
18. Ochs A, Rossle M, Haag K, Hauenstein KH, Derbert, Siegersteter V, et al. The transjugular intrahepatic portosystemic
stent-shunt procedure for refractory ascites. N Engl J Med 1995;332:1192-7.
19. LebrecD, Giuily N, Hadengue A, Vilgrain V, Moreau R, Poynard T, et al. Transjugular intrahepatic portosystemic
shunts: comparison with paracentesis in patients with cirrhosis and refractory ascites: a randomized trial. J Hepatol
1996;25:135-44.
20. Ochs A, Gerbes AL, Haag K, Holl J, Hauenstein KH, Waggershausen T, et al. TIPS and paracentesis for the treatment
of refractory ascites (RA): interim analysis of a randomized controlled trial [abstract]. Hepatology 1995;22:297A.
21. Zervos EE, Goode SE, Wright T, Rosemurgy AS. A prospective randomized trial comparing TIPS and peritoneovenous
shunt in the treatment of intractable ascites [abstract]. Gastroenterology 1996;110:A1368.
22. Wong F, Sniderman K, Liu P, Blendis L. The mechanism of the initial natriuresis after transjugular intrahepatic
portosystemic shunt. Gastroenterology 1997;112:899-907.
23. LindCD, Malisch TW, Chong WK, Richards WO, Pinson CW, Meranze SG, et al. Incidence of shunt occlusion or
stenosis following transjugular intrahepatic portosystemic shunt placement. Gastroenterology 1994;106:1277-83.
24. Sanyal AJ, Freedman AM, Shiffman ML, Purdum PP, Luketic VA, Cheatham AK. Portosystemic encephalopathy after

transjugular intrahepatic portosystemic shunt: results of a prospective controlled study. Hepatology 1994;20:46-55.


April 22, 2000
Conde Petra
The role of endoscopic therapy in the

management of variceal hemorrhage. American
Society for Gastrointestinal Endoscopy.
Citation
Bibliographic Data
Indexing Data
The role of endoscopic therapy in the
management of variceal hemorrhage.
American Society for Gastrointestinal
Endoscopy.
Gastrointest Endosc - 1998 Dec; 48(6): 697-8
Full Text From NIH/NLM MEDLINE, HealthSTAR
Frontmatter NLM Citation ID:
Endoscopic Sclerotherapy (EST) 99069202
Endoscopic Variceal Ligation (EVL) Full Source Title:

References
Publication Type:
About the Publication Guideline; Journal Article; Practice Guideline
Language:
English
Major Subjects:
Esophageal and Gastric Varices / * Therapy
Gastrointestinal Hemorrhage / * Therapy
* Hemostasis, Endoscopic
* Sclerotherapy
Human
Ligation
http://home.mdconsult.com/das/guideline/view/N/104612...S,MI&stty=cow&nid=39984&tlt=Z&pos=&tag=relinfo&anc=gl [22/04/2000 0:09:38]

The role of endoscopic therapy in the management of variceal

hemorrhage
Variceal bleeding is a common and serious complication of portal hypertension. Mortality following the
index hemorrhage in cirrhotics has been reported to be as high as 50%, with a 30% mortality associated
with subsequent bleeding episodes[1] . The optimal managment of patients with variceal bleeding requires
a multidiciplinary approach by a team that includes endoscopists, interventional radiologists and
surgeons. A variety of management options presently exist for patients presenting with variceal bleeding.
The purpose of this guideline is to provide an updated, practical strategy for the specific use of
endoscopy in the management of patients with variceal hemorrhage.
Endoscopic Sclerotherapy (EST)

Until recently, injection sclerotherapy has been considered the endoscopic treatment of choice for
managing esophageal variceal hemorrhage. Endoscopic sclerotherapy is successful in controlling active
bleeding in over 90% of patients[2] and it has proven useful in reducing the frequency and severity of
recurrent variceal hemorrhage (secondary prophylaxis)[3] . Esophageal EST may also be appropriate in
the patient in whom acute bleeding has ceased, when no other cause for hemorrhage can be identified.
Controlled trials of esophageal EST for the primary prevention of initial esophageal variceal bleeding
have not demonstrated a benefit. Esophageal EST is not indicated in patients without a history of variceal
bleeding[4] [8] . Gastric varices which are in continuity with esophageal varices may be treated with EST
below the esophagogastric junction. Isolated gastric varices, however, tend to form an intermingling
network and are therefore less amenable to EST.
Sclerotherapy may be performed by injection of the sclerosant directly into the varix (intravariceal) to
produce thrombosis, or adjacent to the varix (paravariceal) to induce submucosal fibrosis and obliteration
of deeper perforating vessels. In practice, a combination of both techniques may be utilized during the
same session. Techniques of sclerotherapy vary and are operator-dependent. Several sclerosants (sodium
tetradecyl sulfate, sodium morrhuate, ethanolamine oleate, polidocanol and ethanol) have been employed
at varying concentrations and volumes. Sclerotherapy schedules have also varied considerably. More
frequent treatment intervals achieve more rapid variceal obliteration than less frequent treatments, but are
associated with greater mucosal ulceration[9] [11] .
Complications of EST are common and are occasionally life-threatening[3] . Fever, retrosternal
discomfort and dysphagia frequently occur acutely and usually resolve within 48 hours.
Injection-induced bleeding, post-injection esophageal ulceration with delayed bleeding and esophageal
perforation may also occur. Dysphagia may occur after EST due to esophageal ulceration, dysmotility or
a stricture. Dysphagia related to post- sclerotherapy strictures usually responds to dilation. Other
potential complications include mediastinitis, pleural effusion, bronchoesophageal fistula and the adult
respiratory distress syndrome. Bacteremia occurs frequently after EST and patients with a mechanical

prosthetic cardiac valve, a history of endocarditis, a surgicially created systemic-pulmonary shunt or a

synthetic vascular graft (placed within the previous year) should receive antibiotic prophylaxis prior to
the procedure[12] .
Endoscopic Variceal Ligation (EVL)

Endoscopic variceal ligation was initially introduced by Steigmann in 1986 with the anticipation that it
would be similiarly efficacious to EST, but associated with a lower incidence of complications[13] . The
results from six randomized, prospective trials which have directly compared EVL and EST[14] [19]
suggest that this is indeed the case. Endoscopic variceal ligation is superior to EST in terms of
eradicating varices more rapidly[15] [16] [18] with less rebleeding[15] [17] [18] and with fewer complications [14]
[16] [19] . Two of the trials have shown a survival advantage in patients treated with EVL[14] [18] . A recent
meta-analysis has confirmed the superiority of EVL, as compared to EST, for all the major outcomes of
rebleeding, local complications, time to variceal obliteration and survival[20] . In contrast to EST, EVL
appears to rarely induce bacteremia[21] . Recurrence of esophageal varices may occur more frequently in
those treated with EVL[19] and regular endoscopic surveillance remains a critical aspect of management.
Gastric varices should not be treated with EVL alone, as serious complications have been reported to
occur in this setting. Other practical problems associated with the use of EVL include the difficulty of
performing the procedure in the presence of active hemorrhage related to poor visibility, as well as the
need to utilize an overtube to facilitate repeated passage of the endoscope for band reloading. Experience
with multiple-shot firing devices is increasing[22] [23] and this type of apparatus may overcome the
problems associated with band reloading and the subsequent need for an overtube. Endoscopic variceal
ligation offers some advantages over sclerotherapy in appropriate patients.[24] [26]
References
1. Christensen E, Fauerholdt L, Schlichting P, et al. Aspects of the natural history of gastrointestinal bleeding in cirrhosis
and the effect of prednisolone. Gastroenterology 1981; 81: 944-52. abstract
2. Paquet K-J, Feussner H. Endoscopic sclerosis and esophageal balloon-tamponade in acute hemorrhage from esophageal
varices: a prospective controlled randomized trial. Hepatology 1985;5: 580-3. abstract
3. Roberts LR, Kamath PS. Pathophysiology and treatment of variceal hemorrhage. Mayo Clin Proc 1996; 71: 973-83.
abstract
4. Terblanche J. Sclerotherapy for prophylaxis of variceal bleeding. Lancet 1986; 1: 961-3. citation
5. Paquet KJ. Prophylactic endoscopic sclerosing treatment of esophageal wall varices - prospective controlled randomized
trial. Endoscopy 1982; 14: 4-5. abstract
6. Witzel L, Wolbergs E, Merku H. Prophylactic endoscopic sclerotherapy of esophageal varices. A prospective controlled
trial. Lancet 1985; 1: 773-5. abstract
7. The
Veterans Affairs Cooperative Variceal Sclerotherapy Group. Prophylactic sclerotherapy for esophageal varices in
men with alcoholic liver disease: a randomized single-blind, multicenter clinical trial. N Engl J Med 1991; 324: 1779-84.

abstract
8. Fardy
JM, Laupacis A. A meta-analysis of prophylactic endoscopic sclerotherapy for esophageal varices. Am J
Gastroenterol 1994; 89: 1938-48. abstract
9. Higashi H, Kitano S, Hashizume M, et al. A prospective randomized trial of schedules for sclerosing esophageal varices:
1-versus 2-week intervals. Hepatogastroenterol 1989; 36: 337-40.
10. Westaby D, Melia WM, MacDougall BRD, et al. Injection therapy for esophageal varices; a prospective randomized
trial of different treatment schedules. Gut 1984; 25: 129-32. abstract
11. Sarin
SK, Sachdev G, Nanda RN et al. Comparison of the two time schedules for endoscopic sclerotherapy: a
prospective randomized controlled study. Gut. 1986; 27: 710-4. abstract
12. ASGE Guideline. Antibiotic Prophylaxis for Gastrointestinal Endoscopy. #1028. Gastrointest Endosc 1995; 42; 630-5.
abstract
13. Steigmann GV, Cambre T, Sun JH. A new endoscopic elastic band ligating device. Gastrointest Endosc 1986; 32:
230-3. citation
14. SteigmannGV, Goff JS, Michaletz OP, et al. Endoscopic sclerotherapy as compared with endoscopic ligation for
bleeding esophageal varices. N Engl J Med 1992; 326: 1527-32. abstract
15. Gimson A, Ramage JK, Panos MZ, et al. Randomized trial of variceal banding ligation versus injection sclerotherapy
for bleeding esophageal varices. Lancet 1993; 342: 391-4. abstract
16. LaineL, El-Newhi HM, Migikovsky B, et al. Endoscopic ligation compared with sclerotherapy for the treatment of
bleeding esophageal varices. Ann Intern Med 1993; 119: 1-7. abstract
17. HouM-C, Lin H-C, Kuo BI-T, et al. Comparison of endoscopic variceal injection therapy and ligation for the treatment
of esophageal variceal hemorrhage: a prospective randomized trial. Hepatology 1995; 21: 1517-22. abstract
18. LoGH, Lai KH, Cheng JS, et al. A prospective, randomized trial of sclerotherapy versus ligation in the management of
bleeding esophageal varices. Hepatology 1995; 22: 466-71. abstract
19. Baroncini D, Milandri GL, Borioni D, et al. A prospective randomized trial of sclerotherapy versus ligation in the
elective treatment of bleeding esophageal varices. Endoscopy 1997; 29: 235-40. abstract
20. Laine
L, Cook D. Endoscopic ligation compared with sclerotherapy for the treatment of esophageal variceal bleeding: a
meta analysis. Ann Intern Med 1995; 123: 280-7. abstract
21. Tseng
CC, Green RM, Burke SK, et al. Bacteremia after endoscopic band ligation of esophageal varices. Gatrointest
Endosc 1992; 38: 336-7.
22. SaeedZA. Endoscopic esophagogastric variceal ligation with a six-shot multiple ligation device. Am J Gastroenterol
1995; 90: 1540.

23. HashizumeM, Sugimachi K, Kishihara F, et al. A serial transparent endoscopic elastic band ligator. Gastrointest
Endosc 1995; 42: 169-70. full text
24. Laine
L. Ligation: Endoscopic treatment of choice for patients with bleeding esophageal varices? Hepatology 1995; 22:
663-5. citation
25. Saeed
Z. Endoscopic therapy of bleeding esophageal varices: ligation is still the best. Gastroenterology 1996; 110:
635-8. citation
26. Hayes PC. The coming age of band ligation for esophageal varices. BMJ 1996; 312: 1111-2. citation


April 22, 2000
Conde Petra
Aspects of the natural history of gastrointestinal

bleeding in cirrhosis and the effect of
prednisone.
Citation
Bibliographic Data
Abstract
Indexing Data
Aspects of the natural history of
Copyright Notice and Disclaimer gastrointestinal bleeding in cirrhosis and
the effect of prednisone.
Christensen E - Gastroenterology - 1981 Nov; 81(5): 944-52
Full Source Title:
Gastroenterology
Publication Type:
Clinical Trial; Journal Article; Randomized Controlled Trial
Language:
English
Authors:
Christensen E; Fauerholdt L; Schlichting P; Juhl E; Poulsen H;
Tygstrup N
Abstract:
The natural history of gastrointestinal bleeding in cirrhosis has
been studied using prospectively collected data of 532 patients
included in a randomized clinical trial with a regular follow-up
of up to 12 yr. Of the total 199 patients who experienced
gastrointestinal bleeding, 95 (48%) bled from esophageal or
gastric varices, 67 (34%) bled from peptic ulcer or gastritis, and
37 (18%) had either insufficient evidence of the source (33) or
mixed sources (4). In the total group of patients the cumulative
percentage of patients in whom varices had been demonstrated
of patients in whom varices had been demonstrated by
radiography increased from 12 to 90 in 10 yr, while that of
bleeding from varices increased from 7 to 40. In 104 patients
who bled for the first time during the trial period (trial bleeding
http://home.mdconsult.com/das/journal/view/O/...p=6518296&PAGE=1.html&ANCHOR=abs&SOURCE=HS,MI (1 of 2) [22/04/2000 0:10:23]

patients) the median number of bleeding episodes was one

(range 1-8). In these patients the fatality from bleeding from
varices was 82%. The risk of rebleeding from varices was 81%,
and 4 yr after the first bleeding the cumulative survival had
decreased to less than 10%. Rebleeding was significantly less
frequent and survival significantly higher in patients bleeding
from sources other than varices. Prednisone reduced the
occurrence rate of varices, bleeding from varices, and death
from bleeding varices in nonalcoholic females without ascites,
40% of whom fulfilled the histologic criteria of chronic active
hepatitis. Prednisone significantly increased the occurrence rate
of varices inpatient with ascites and of bleeding from varices in
alcoholic patients. Prednisone significantly increased the
occurrence rate of peptic ulcer in males and in patients without
chronic active hepatitis.
Major Subjects:
Gastrointestinal Hemorrhage / * Etiology / Prevention &
Control
Liver Cirrhosis / * Complications
Prednisone / Adverse Effects / * Therapeutic Use
Clinical Trials
Esophageal and Gastric Varices / Complications
Female
Hepatitis / Complications
Human
Male
Peptic Ulcer Hemorrhage / Complications
Random Allocation
Risk

53-03-2 (Prednisone)
Bookmark URL: /das/journal/view/O/6518296

April 22, 2000
Conde Petra
Endoscopic sclerosis and esophageal balloon

tamponade in acute hemorrhage from
esophagogastric varices: a prospective
Citation controlled randomized trial.
Bibliographic Data
Abstract
Indexing Data
Endoscopic sclerosis and esophageal
Copyright Notice and Disclaimer balloon tamponade in acute hemorrhage
from esophagogastric varices: a
Find More Articles Like This prospective controlled randomized trial.
Paquet KJ - Hepatology - 1985 Jul-Aug; 5(4): 580-3
Order a Full Text Copy of the From NIH/NLM MEDLINE, HealthSTAR
85259432
Full Source Title:
Hepatology
Publication Type:
Language:
English
Authors:
Paquet KJ; Feussner H
Abstract:
A prospective randomized controlled clinical trial was
performed in 43 consecutive histologically proved cirrhotic
patients with endoscopically proved actively bleeding
esophageal varices. Twenty-two were randomly selected to
have esophageal tamponade with the Sengstaken-Blakemore
tube, and 21 were selected to have endoscopic sclerosis of the
esophageal wall. The two groups were similar in demographic,
clinical and laboratory data. Bleeding was controlled by the
Sengstaken-Blakemore tube in 16 of 22 patients (73%) and by
endoscopic sclerosis in 20 of 21 (95%). Among those controlled
by the Sengstaken-Blakemore tube, seven (44%) rebled and
three (43%) were again controlled by the
Sengstaken-Blakemore tube; in the endoscopic sclerosis group,

four (20%) rebled and three (75%) were controlled. Thus,

hemorrhage was definitively controlled in 52% of patients and
66% of bleeding episodes in the Sengstaken-Blakemore tube
group and in 90% of patients and 92% of bleeding episodes in
the endoscopic sclerosis group. The definite control of
hemorrhage was significantly better in the endoscopic sclerosis
group (p less than 0.01). The Sengstaken-Blakemore tube
patients received no definitive therapy after bleeding had been
controlled. Within 30 days, six patients (27%) in the
Sengstaken-Blakemore tube group had died compared to 2
(10%) in the endoscopic sclerosis group which is statistically
significant (p less than 0.01) in favor of endoscopic sclerosis.
The frequency of complications was similar in the two groups.
Endoscopic sclerosis patients received serial endoscopic
sclerosis after bleeding had been stopped during the whole
period of follow-up.(ABSTRACT TRUNCATED AT 250
WORDS)
Major Subjects:
Esophageal and Gastric Varices / * Complications /
Mortality
Gastrointestinal Hemorrhage / Etiology / Mortality / *
Therapy
Intubation / * Instrumentation
Sclerosing Solutions / * Therapeutic Use
Clinical Trials
Comparative Study
Endoscopy
Esophagus
Female
Human
Male
Prospective Studies
Random Allocation

(Sclerosing Solutions)
Bookmark URL: /das/journal/view/O/7553780

April 22, 2000
Conde Petra
Pathophysiology and treatment of variceal

hemorrhage.
Citation
Bibliographic Data
Abstract
Indexing Data
Pathophysiology and treatment of
Copyright Notice and Disclaimer variceal hemorrhage.
Roberts LR - Mayo Clin Proc - 1996 Oct; 71(10): 973-83
Find More Articles Like This From NIH/NLM MEDLINE
NLM Citation ID:
Order a Full Text Copy of the 96417988
Full Source Title:
Mayo Clinic Proceedings
Publication Type:
Journal Article; Review; Review, Tutorial
Language:
English
Author Affiliation:
Division of Gastroenterology and Internal Medicine, Mayo
Clinic Rochester, Minnesota 55905, USA.
Authors:
Roberts LR; Kamath PS
63
Abstract:
Portal hypertension results from increases in portal flow and
portal vascular resistance. Factors increasing portal blood flow
are predominantly humoral. Resistance to portal flow has a
fixed component due to distortion of the vasculature by
cirrhotic nodules and a variable component that is related to
vasoactive substances. Varices result from an increase in portal
pressure. Factors predicting the risk of variceal bleeding include
continued alcohol use, poor liver function, large varices, and red
wale markings on varices at endoscopy. Octreotide is probably
http://home.mdconsult.com/das/journal/view/N/...10?sp=955610&PAGE=1.html&ANCHOR=abs&SOURCE=MI (1 of 2) [22/04/2000 0:12:03]

the drug of choice for pharmacologic management of bleeding

esophageal varices. Propranolol has an established role in the
prevention of variceal hemorrhage, and variceal band ligation
may be the preferred endoscopic technique. Transjugular
intrahepatic portosystemic shunts have emerged as an important
treatment for patients in whom pharmacologic and endoscopic
therapies have failed and are an effective bridge to liver
transplantation.
Major Subjects:
Hemorrhage / Drug Therapy / * Physiopathology / *
Therapy
Liver / * Blood Supply
Varicose Veins / * Physiopathology / * Therapy
Animal
Hemostatics / Therapeutic Use
Human
Hypertension, Portal / Complications / Physiopathology /

Therapy
Ligation
Liver Cirrhosis / Complications / Physiopathology
Portal System / Physiology
Portasystemic Shunt, Surgical
Regional Blood Flow
Sclerotherapy

(Hemostatics)
Bookmark URL: /das/journal/view/N/955610
http://home.mdconsult.com/das/journal/view/N/...10?sp=955610&PAGE=1.html&ANCHOR=abs&SOURCE=MI (2 of 2) [22/04/2000 0:12:03]

April 22, 2000
Conde Petra

printing.
the handout.
All Topics
Medications
Category
Topics by Specialty
Adult Health Topics

Cirrhosis
What Is Cirrhosis?

The Liver.
The liver is the largest organ in the body, located immediately
below the diaphragm and occupying the entire upper right
quadrant of the abdomen. In the healthy adult, it weighs about
three pounds and is wedge shaped, with the upper part being
wider than the lower. The liver is rich in blood, holding about
13% of the body's supply. It is furnished with blood from two
large vessels--the portal vein and the hepatic artery (Hepatic
means liver). Blood that has circulated through the stomach,
spleen, and intestine enters the liver through the portal vein as
part of the so-called portal circulation system. The liver extracts
nutrients and toxins from this blood, which is then returned
through the hepatic vein to the right side of the heart. The
hepatic artery supplies blood to the liver directly from the heart.
The liver performs over 500 vital functions. It processes all of

the nutrients that the body requires, including proteins, glucose,
vitamins, cholesterol, and fats. It also renders harmless
potentially toxic substances, including alcohol, ammonia,
nicotine, drugs, and harmful by-products of digestion. The liver
synthesizes blood clotting factors and albumin, the major

protein in the blood. The liver also affects bilirubin, a

yellow-green pigment produced from the breakdown of
hemoglobin, the oxygen-carrying component in red blood cells.
The liver converts bilirubin into a water-soluble form, which is
then excreted into bile. Bile is a green-colored fluid that is
formed in the liver and contains, in addition to bilirubin, bile
salts, fatty acids, cholesterol and other substances. Bile travels
from the liver to the gall bladder, where it is stored until after a
meal; at that time it is secreted into the intestines to help digest
fat.
These vital processes rely on well-organized liver architecture

consisting of bile ducts, blood vessels, working liver tissue
(called the parenchyma), and supportive connective tissue. The
liver has two major lobes (with the right one much larger than
the left), which are separated from each other by walls of tough,
fibrous connective tissue. The lobes are composed of about
100,000 elegantly patterned lobules made of up liver cells and
blood vessels that form microscopic columns. Bracing the
corners of each lobule column are three small important
structures: an artery, a vein, and a bile duct. The arteries bring
oxygen-rich blood to nourish the liver cells. The blood passing
through the veins supplies the liver cells with the nutrients and
toxins that the liver cells process. Bile drains from tiny canals
around the liver cells into the corner ducts, which eventually
join to form the large common bile duct that leads from the liver
to the gall bladder. A central vein runs down through each
column, which collects the blood from the surrounding lobules
and acts as a tributary, eventually joining with other veins to
form the major hepatic veins. The hepatic veins, in turn, lead
into the inferior vena cava , the large vein that conducts the
worn-out blood from the liver back to the heart.
Cirrhosis.
Cirrhosis is an irreversible sequel to a number of disorders that
damage the liver cells and causes fibrosis (scarring). Often, this
process is accompanied by random clusters of regenerated liver
cells that develop throughout the liver, usually forming nodules
around the scarred areas. Eventually, this damaging pattern
becomes so extensive that the normal architecture of the liver is
distorted. Changes in the way blood and fluid flow in and out of
the liver also occur. The spleen overproduces a substance called
nitric oxide, which causes the blood vessels to relax and widen,
while vessels in other parts of the body, including the kidney,

narrow. The small blood vessels and bile ducts in the liver
constrict, so the blood that normally passes into the liver from
the intestine backs up through the portal vein and seeks other
routes. Twisted swollen veins called varices form in the
stomach and lower part of the esophagus. Bile builds up in the
blood stream, resulting in high levels of bilirubin, which causes
the yellowish cast in the skin called jaundice. Fluid build-up in
the abdomen (called ascites) and swelling in the arms and legs
is common. The liver enlarges in the first phases of the disease.
In advanced stages, however, the liver sometimes shrinks, a
condition called postnecrotic cirrhosis.
What Causes Cirrhosis?

Alcoholic Cirrhosis.
The liver is particularly endangered by alcohol. In the body,
alcohol breaks down into various chemicals, some of which are
very toxic in the liver. Alcoholic cirrhosis (also sometimes
referred to as portal, Laennec's, nutritional, or micronodular
cirrhosis) is the primary cause of cirrhosis in the U.S., and is
estimated to be responsible for 44% of deaths from cirrhosis in
North America. (Some believe this a low estimate. One
Canadian study found alcohol to be the major contributor to
80% of all cirrhosis deaths.) Over time, alcohol abuse leads to
increased demands for oxygen by the liver and, at the same
time, causes fat accumulation that impairs the liver's ability to
absorb oxygen. The immune system over-responds by
triggering an inflammatory process that damages and finally
kills liver cells, a condition called alcoholic hepatitis. During
the initial phase, the fat-laden liver becomes greatly enlarged,
but it eventually shrinks as web-like scars and small knots of
abnormal regenerated liver cells develop--the characteristics of
cirrhosis.
Cirrhosis from Chronic Hepatitis.

The next leading cause of cirrhosis in the U.S. is chronic
hepatitis, either hepatitis B or C. Chronic hepatitis C is the more
dangerous form and accounts for one-third of all cirrhosis cases.
Viruses or other mechanisms that cause hepatitis produce
inflammation in liver cells, resulting in their injury or
destruction. If the condition is severe enough, the cell damage

becomes progressive, building a layer of scar tissue over the

liver. In advanced cases, as with alcoholic cirrhosis, the liver
shrivels in size, a condition called postnecrotic or posthepatic
cirrhosis.
Primary Biliary Cirrhosis.

Primary biliary cirrhosis accounts for only 0.6% to 2% of
deaths from cirrhosis. It is most likely an autoimmune disease;
that is, the body's immune system attacks its own liver cells
mistaking them for foreign invaders (called antigens). In the
case of primary biliary cirrhosis, the cells under attack are in the
bile ducts. Liver cells are destroyed as the disease progresses.
Some research indicates that this autoimmune process may be
triggered in some patients by certain viruses, particularly those
known as retroviruses. Other autoimmune diseases, such as
scleroderma or Sjgren's syndrome, may also accompany this
form of cirrhosis.
Uncommon Causes of Cirrhosis.

Rare conditions that cause cirrhosis include hemochromatosis
(iron build-up in liver cells), which is fairly common in people
with diabetes and Wilson's disease (copper build-up in liver
cells). Liver damage that results in cirrhosis also may be caused
by a number of inherited diseases such as cystic fibrosis,
alpha-1 antitrypsin deficiency, galactosemia, and glycogen
storage diseases. Other rare causes include schistosomiasis (a
parasite found in the Far East, Africa, and South America),
long-term or high level exposure to certain chemicals and drugs
(e.g., arsenic, methotrexate, and toxic doses of vitamin A), and
small intestine bypass surgery (rarely, if ever, performed
anymore). Cancers that have metastasized to the liver, blood
clots in the hepatic or portal vein, or obstructions in the bile
duct can also cause changes that resemble cirrhosis.
Who Gets Cirrhosis?

Cirrhosis is the eleventh leading cause of death in the U.S. and
affects about three million Americans. The risk factors for liver
injury determine an individual's chances for cirrhosis.

People with Alcoholism.

About 10% to 35% of heavy drinkers develop alcoholic
hepatitis. After years of drinking, liver damage can be very
severe, leading to cirrhosis in about 10% to 20% of cases. Not
eating when drinking and consuming a variety of alcoholic
beverages are also factors that increase the risk for liver
damage. People with alcoholism are also at higher risk for
hepatitis B and C. People with alcoholism, in fact, should be
immunized against hepatitis B, and they may need a
higher-than-normal dose of the vaccine for it to be effective. [
See also Alcoholism.]
People with Chronic Hepatitis.

Because of blood screening, the risk for transmission for both
hepatitis B and C viruses through transfusions has dramatically
decreased since 1990. Hepatitis C can exist for decades,
however, without symptoms and nearly 300,000 people who
had transfusions before 1990 may have contracted the virus. In
addition, the causes of many cases of hepatitis C are unknown.
An estimated 10% to 60% of people originally infected with
hepatitis C patients develop the chronic form, which poses a
risk for cirrhosis of about 30%. Experts estimate that there may
be a 60% increase in the number of cirrhosis cases because of
rising numbers of people diagnosed with hepatitis C. It is not
clear, however, which hepatitis C patients will develop
cirrhosis. Alcohol use multiplies any average risk for cirrhosis
in hepatitis C patients. One large study also suggested that those
who contracted hepatitis after exposure in a hospital setting,
blood transfusion, and when the cause was unknown had a 20%
to 30% chance for cirrhosis. Those who developed hepatitis C
from drug abuse, sexual activity, and occupational exposure had
a lower risk--around 10%. The genetic type of the virus may
affect severity (type 1 being the most serious, type 2 and 3
posing less danger). A recent study suggested, however, that
regardless of genetic type, patients with a greater presence of
the virus in blood and liver tissue tended to have a more severe
case. About 3% to 5% of people infected with hepatitis B
develop the chronic form, and about half of these patients
develop cirrhosis. [ See also Hepatitis.]

Risk Factors for Primary Biliary Cirrhosis.

Up to 95% of primary biliary cirrhosis cases occur in women,
usually around age 50. Genetic factors are involved, but the
inheritance pattern is unclear.
What Are The Symptoms Of

Cirrhosis?
General Symptoms.
Fatigue and loss of energy are common early symptoms of
cirrhosis, along with loss of appetite and nausea, although many
people experience few symptoms at the onset of cirrhosis.
Spider angiomas may develop on the skin; these are
pinhead-sized red spots from which tiny blood vessels radiate.
Patients in later stages develop jaundice, a yellowish cast to the
skin and eyes, which is caused when the liver cannot process
bilirubin for elimination from the body. The palms of the hands
may be reddish and blotchy, a condition known as palmar
erythema. Patients may lose body hair. In men with alcoholic
cirrhosis, the testicles may atrophy and their breasts may
become swollen, sometimes painfully.
Symptoms of Complications.
A swollen belly is a sign of ascites, a condition that occurs
when fluid accumulates in the abdomen and the most common
major complication of cirrhosis. Fever, abdominal pain, and
tenderness when the belly is pressed indicate that the fluid is
infected. (Infection may occur, however, without any
symptoms.) Forgetfulness, unresponsiveness, and trouble
concentrating may be early symptoms of hepatic
encephalopathy, which is damage to the brain caused by
build-up of toxins. Sudden changes in the patient's mental state,
including agitation or confusion, may indicate an emergency
condition. Other symptoms include bad fruity-smelling breath
and tremor. Late stage symptoms of encephalopathy are stupor
and, eventually, coma.

Symptoms Specific to Rare Cases of

Cirrhosis.
People with primary biliary cirrhosis are subject to severe,
general itching and often develop small fatty yellow lumps
called xanthomas on the eyelids, hands, and elbows. They may
have an unpleasant condition called steatorrhea, in which the
feces contain excessive fat, causing them to float and to be very
foul smelling. In the rare disorder hemochromatosis, there is
often a bronze cast to the skin, an indication of iron build-up. A
thin bronze crescent bordering the cornea is called the
Kayser-Fleisher ring and is a sign of copper-build up in people
with Wilson's disease.
How Serious Is Cirrhosis?

General Outlook for Cirrhosis.
Cirrhosis is the seventh leading cause of death by disease in the
U.S., killing over 25,000 people each year. The most serious
complications of cirrhosis are bleeding, infections, and
encephalopathy--damage to the brain. Nearly every bodily
process is affected by a damaged liver, including those of the
digestive, hormonal, and circulatory systems. Less protein is
produced by the liver, for example, which causes fluid build-up,
bleeding problems, and susceptibility to infection. Additionally,
the liver cannot detoxify harmful substances, which accumulate
and impair brain function. Cirrhosis is also a cause of liver
cancer.
Cirrhosis is irreversible, but the rate of progression can be very

slow in some patients depending on its cause and other factors.
In patients with hepatitis B, for example, the five-year survival
rate after a diagnosis of cirrhosis is 71%. For alcoholics with
cirrhosis who abstain, a survival rate of five years or more can
be as high as 85%. For those who continue drinking, the chance
for living beyond five years is no higher than 60%. Often,
however, it is difficult to determine prognosis at the time of
diagnosis because the physician is usually unable to tell when
cirrhosis first occurred.
About two-thirds of patients with primary biliary cirrhosis never

develop symptoms and can have a normal life span. Once
symptoms of liver damage, such as jaundice, occur, however,
the average survival time declines. In one study of women

diagnosed with primary biliary cirrhosis, about 36% developed
symptoms over an 11-year period, and 11% either died or
required liver transplantation. Unfortunately, researchers are
unable to determine specific risk factors that could predict who
will develop a severe condition and who won't.
Portal Hypertension and its

Complications.
In cirrhosis, liver cell damage slows down blood flow and blood
pressure therefore increases. This pressure causes a back-up of
blood through the portal vein--a condition called portal
hypertension . The effects of portal hypertension can be
widespread and serious.
Ascites. Ascites is fluid build-up in the abdomen usually caused

by portal hypertension and is such a critical event in the
progression of cirrhosis that some experts refer to the phases of
cirrhosis as preascitic and ascitic. Once ascites occurs, only half
of patients survive after two years. Some physicians even
believe that ascites signals the need for liver transplantation,
particularly in patients whose cirrhosis is not due to alcohol.
Swelling can also occur in the arms and legs and in the spleen.
(Ascites can result from other conditions, and physicians should
check for other possibly serious causes, including cancer and
infections.) Ascites itself is not fatal, but it is uncomfortable and
can reduce breathing function and urination. Of interest is
research indicating that nitric oxide is overproduced in
cirrhosis; this substance is also associated with a condition
called hyperaldosteronism. Aldosterone is a steroid hormone
produced in the adrenal gland that regulates sodium and
potassium. High amounts ( hyperaldosterone) can result in a
fluid electrolyte imbalance called hypokalemic alkalosis (low
potassium levels, overly alkaline), which in turn can cause
weakness, abnormal heart rhythms, a sense of paralysis, and, in
severe cases, encephalopathy.
Variceal Bleeding . One of the most serious repercussions of

portal hypertension is the development of varices--blood vessels
that enlarge to provide an alternative pathway for blood
diverted from the liver. These abnormal blood vessels are
thin-walled, often twisted, and are subject to high pressure.
Varices often form in the stomach and esophagus (the tube

connecting the mouth and stomach). Internal bleeding from

these varices occurs in 20% to 30% of all cirrhosis patients. The
risk of death from a single episode can reach 70%. Recurrence
is common within two weeks of the first episode, but after six
weeks, the risk for recurrence is the same as for patients who
have not had a bleeding event. Factors that predict bleeding
include ascites, encephalopathy, large veins, and marked liver
cell abnormalities. It should be noted that moderate to severe
exercise may also increase the risk for variceal bleeding.
(Taking beta-blockers may reduce this risk.) Some researchers
believe that bacterial infection may be a primary factor in
triggering a bleeding episode.
Bleeding Disorders.
Gastrointestinal (GI) bleeding can occur from abnormal blood
clotting, often caused by deficiencies in vitamin K, low levels
of clotting proteins, and low counts of platelets (the blood cells
that normally initiate the clotting process).
Infections.
Abdominal infection occurs in up to 25% of patients with
cirrhosis within a year of diagnosis. At high risk are patients
whose tests results show very low protein levels and very high
bilirubin levels.
Encephalopathy.
Encephalopathy (damage to the brain) causes mental confusion
and, in worst cases, coma and death. The development of
encephalopathy is often precipitated by other problems,
including gastrointestinal bleeding, constipation, excessive
dietary protein, infection, surgery, or dehydration. No single
toxin accounts for the mental effects of encephalopathy. A
combination of conditions causes this serious complication,
such as the build-up in the blood of harmful intestinal toxins,
particularly ammonia, and an imbalance of amino acids that
effect the central nervous system.

Liver Cancer.
Cirrhosis greatly increases the risk for liver cancer, regardless
of the cause of cirrhosis.
Osteoporosis.
Primary biliary cirrhosis is associated with reduced bone
growth, partly because of the liver's inability to process vitamin
D and calcium and also from some of its treatments. As a result,
osteoporosis occurs in 20% to 30% of patients. Bone loss is also
a complication of liver disease in alcoholics and one study
indicated that it may also be a complication of cirrhosis caused
by hepatitis.
Insulin Resistance.
Nearly all patients with cirrhosis are insulin resistant. Insulin
resistance is a primary feature in type 2 diabetes and occurs
when the body is unable to use insulin, a hormone that is
important for delivering blood sugar and amino acids into cells
and helps determine whether these nutrients will be burned for
energy or stored for future use.
Other Complications.
One study reported that nearly a quarter of patients with
cirrhosis had gallstones. They may also face a higher than
average risk for certain abnormal heart rhythms. Peptic ulcers,
sleep disorders, and respiratory problems are also more
common in people with cirrhosis than in the general population.
How Is Cirrhosis Diagnosed?

Physical Examination.
The cirrhotic liver is often enlarged. It is also firm and may
even feel rock-hard. The left side can often be felt by the
physician when pressing on the abdomen. (In advanced stages,
however, the liver may become small and shriveled.) If the
abdomen is swollen, the physician will tap the flanks and listen

for a dull thud and feel for a shifting wave of fluid in the
abdomen--indications of ascites.
Biopsy.
Some experts are now recommending biopsies for all chronic
hepatitis C patients, regardless of severity, because of the risk
for liver damage even in patients without symptoms. A liver
biopsy is the only definite method for diagnosing cirrhosis. It
also helps determine its cause, treatment possibilities, the extent
of damage, and the long-term outlook. For example, hepatitis C
patients who show no significant liver scarring when biopsied
appear to have a low risk for cirrhosis. The procedure uses a
needle inserted through the abdomen to obtain a tissue sample
from the liver. The biopsy may also be performed during
peritoneoscopy--a procedure that uses a catheter and tiny
camera to view the surface of the liver. Biopsies can be
dangerous, so they cannot be performed on patients who have
test results that indicate clotting problems, on those who have
had previous liver biopsies, or who have ascites.
Blood Tests.
A number of blood tests may be performed to measure liver
function and to help determine the severity and cause of
cirrhosis. One of the most important factors indicative of liver
damage is bilirubin, a red-yellow pigment that is normally
metabolized in the liver and then excreted in the urine. In
patients with hepatitis, the liver cannot process bilirubin, and
blood levels of this substance rise, sometimes causing jaundice.
Measurements of blood levels of certain liver enzymes are
useful for diagnosing cirrhosis. To help determine outlook,
experts may use a calculation called a discriminant function
(DF), which uses two important measurements: serum albumin
concentration and prothrombin time (PT). Serum albumin
measures protein in the blood (low levels indicate poor liver
function). The PT test measures in seconds the time it takes for
blood clots to form (the longer it takes, the greater the risk for
bleeding).
Specific Blood Tests for Primary Biliary Cirrhosis . Very high

levels of serum alkaline phosphatase, an enzyme produced in
the liver, often occur in cases of primary biliary blood cirrhosis.
In 95% of people with primary biliary cirrhosis, blood tests will

detect high levels of immune factors called mitochondrial

antibodies. High antibody levels are not necessarily indicators
of severity.
Specific Blood Tests for Chronic Hepatitis . Blood tests for

chronic hepatitis include measurements of aminotransferase
levels, bilirubin levels, and detection of blood clotting
problems. Physicians will look particularly at blood levels of
enzymes known as aspartate (AST) and alanine (ALT)
aminotransferases. Research has suggested that a high ratio of
AST to ALT in chronic hepatitis C is a strong signal for the
presence of cirrhosis. Immunoassays to detect antibodies to
specific hepatitis viruses are also performed.
Caffeine Clearance Test.

A simple, inexpensive and harmless test that measures caffeine
in saliva may prove to be an accurate method for determining
the severity of cirrhosis.
Imaging Tests.
A number of imaging tests may be used to diagnose cirrhosis
and its complications. Magnetic resonance imaging (MRI),
computed tomography (CT), and ultrasound are all imaging
techniques that are useful in detecting and defining the extent of
cirrhosis. Such tests can reveal ascites, enlarged spleen,
irregular liver surface, reversed portal vein blood flow, and liver
cancer. Sometimes they can even detect abnormally large blood
vessels in the liver. Arteriography uses dye injected into the
hepatic arteries that then shows up on x-ray. Splenoportography
used uses dye injected into the spleen, which allows the
physician to measure portal vein pressure; this procedure is
risky.
Hepatic Vein Wedge Pressure.

Hepatic vein wedge pressure involves insertion of a catheter
into the hepatic veins. The blood pressure in the veins of the
liver is then measured; the result is an indicator of portal vein
pressure. If pressure is high, cirrhosis is likely. A low
measurement is a favorable sign.

Paracentesis.
If ascites is present, paracentesis is performed. This procedure
involves using a thin needle to withdraw fluid from the
abdomen. The fluid is tested for difference factors, including
protein levels, bacteria cultures, and white blood cell counts.
Low levels of protein in the fluid and a low white blood cell
count suggest that cirrhosis is the cause of the ascites. The
appearance of the fluid is helpful in determining a cause. For
example, a cloudy fluid plus a high white blood cell count mean
an infection is present. Bloody fluid suggests the presence of a
tumor.
What Are The Treatments For

Preventing Cirrhosis In
High-Risk People?
Treatment for Alcoholism.
The only treatment for alcoholic cirrhosis is to stop drinking.
[For more information on how to stop, see Alcoholism.]
Individuals with alcoholic cirrhosis are almost always
malnourished and, therefore, require increased calories and
rigorous nutritional support, which can improve survival rates.
Corticosteroids may be useful for alcoholic hepatitis, an acute
condition in which the liver is inflamed, but these drugs are not
beneficial after cirrhosis has developed. Drugs under
investigation include propylthiouracil and colchicine, which
inhibit deposits of collagen, the critical protein building block in
connective and scar tissue. Researchers are also investigating
drugs that block factors in the immune system called
thrombaxanes, which may play an important role in the
inflammatory process that kills liver cells in alcoholic cirrhoses.
Treatment for Chronic Hepatitis B or C.

Interferon alpha (Roferon-A, Intron A) is the standard drug
currently used for both chronic viral hepatitis B and C. In those
who respond, studies are showing improved symptoms, a
normal long-term survival rate, and, in some, no return of the
disease. Unfortunately, the percentage of patients who benefit

over the long-term using this drug as sole therapy is small. In

some cases, a very short course of corticosteroids may be used
initially to boost the effect of interferon. Even when the drug is
effective, the disease nearly always recurs and requires
additional treatment. Common side effects are flu-like
symptoms that usually occur within six hours and last for
twelve. More chronic effects include depression (which can be
very severe), anxiety, amnesia, confusion, irritability, changes
in sensation, weight loss, vomiting, general weakness, and
possible negative effects on cholesterol and lipid levels.
Interferon often causes a drop in platelet and white blood cell
counts, increasing susceptibility to bacterial infections. It may
also trigger an autoimmune response, possibly causing anemia,
diabetes, lupus-like symptoms, thyroid abnormalities, or even
autoimmune hepatitis. Other interferons being tested include
recombinant or consensus type-I interferon (Infergen, CIFN)
and interferon beta.
Standard Treatments for Hepatitis B. Interferon alpha has

eliminated the virus and sustained significant remission in 25%
to 40% of patients with chronic hepatitis B. The drug is usually
taken by injection every day for 16 weeks. (It does not appear to
be effective for hepatitis D.) Unfortunately, even in hepatitis B,
the virus recurs in almost all cases, although this recurring
mutation may be weaker than the original strain. Administering
the drug for longer periods may produce sustained remission in
more patients while still being safe. Interferon beta is benefiting
many children with hepatitis B who do not respond to interferon
alpha. The nucleoside analogue lamivudine has been shown to
reduce viral count in up to 56% of hepatitis B patients who took
it as sole therapy for about a year. Lamivudine is taken orally
and has very few side effects. Some researchers are concerned,
however, that, although the drug may prevent cirrhosis it does
not protect against liver cancer, particularly in those who have
harbored the virus since childhood. Small trials of the drug
adefovir, also called GS840, are reporting significant success in
reducing viral levels by over 99%.
Standard Treatments for Hepatitis C. At this time, qualified

patients with hepatitis C are treated with interferon alpha for
about 48 weeks with the goal of reducing viral levels to zero.
Patients who respond to interferon show improved symptoms
and long-term studies indicate that nearly all remain virus-free.
Interferon also appears to help prevent progression to liver
cancer in hepatitis C (although not in hepatitis B) in patients
who have cirrhosis. In patients who respond to the drug, it may
even reverse early cirrhosis and scarring. Unfortunately, the

virus remains undetectable six months after stopping treatment
in only about 15% to 20% of patients. (One small study
indicated that interferon may be even less effective for African
Americans with hepatitis C.) Extending treatments to up to 24
months have improved early relapse rates, but the regimen is
costly and side effects are more severe.
A combination of interferon alpha (Intron-A) and the nucleoside

analogue ribavirin (Rebetol), however, is proving to be five to
ten times as effective as interferon alone in sustaining a
response. This regimen may even benefit patients with
advanced fibrosis and cirrhosis. Side effects from the
combination are similar to interferon alone [ see above ],
although they are more common, and ribavirin adds the risk for
anemia. It is not clear yet how long patients should take the
combination regimen. Taking it for 48 weeks is slightly more
beneficial than taking it for 24 weeks, although side effects are
more severe. The longer-duration treatment, then, may
eventually be reserved for patients with more advanced disease.
Early studies indicated that the later the virus was eradicated,
the poorer the chance for a sustained remission. Many experts,
then, have advised stopping treatment after 12 weeks if patients
continue to show evidence of the virus. However, recent studies
are indicating that even if the virus is not eradicated until up to
24 weeks of treatments, remission is sustained in a large
proportion of patients taking either interferon alone or the
combination. Other drug combinations and interferon forms are
under investigation.
Drugs that Boost the Immune System . A number of drugs have

been developed that boost the body's own immune system to
fight the virus. For example, thymosin (Zadaxin) is a synthetic
version of a peptide derived from the thymus gland (which
produces immune factors call T-cells). It is showing promise
alone or in combination for patients with hepatitis B and C.
Vaccines, including Hepagene, are being investigated for
treating as well as preventing hepatitis B. Researchers are also
looking at interleukin 12, an immune factor that may enhance
the immune response to hepatitis C. [For more information, see
Hepatitis.]

What Lifestyle Factors Can

Help Manage Cirrhosis?
A healthy lifestyle is important for everyone, and particularly
for people with cirrhosis. Because important antioxidant
vitamins are depleted in the cirrhotic liver, people should
maintain a diet rich in fresh fruits, vegetables, and whole grains
that contain not only vitamins A, E, and C but also other
substances important for health. Vitamin supplements
themselves are not recommended except with the advice of a
physician. Hepatitis C is sometimes associated with abnormally
high iron levels. Such patients should avoid iron-rich foods,
such as red meats, liver, and iron-fortified cereals, and should
avoid cooking with iron-coated cookware and utensils.
Restricting salt consumption to less than 2,000 mg a day is
particularly important for patients with ascites. The less salt the
better. Fluid restriction is not usually necessary, but patients
with severe ascites should discuss limiting fluid with their
physicians. High-quality dietary protein may be helpful for
patients with ascites and for repairing muscle mass, but
excessive protein loads may trigger encephalopathy. Protein
solutions have been devised that provide beneficial amino acids
without including those that increase this risk. There is no limit
on vegetable proteins, such as those from soy. Exercise
increases the risk for portal pressure and variceal bleeding; a
recent study reported that taking a beta-blocker may reduce this
risk, although patients should discuss this with their physician.
How Are The Complications Of

Cirrhosis Managed And
Treated?
Treating Ascites.
Diuretics and Lifestyle Changes. Abstaining from alcohol,
restricting sodium intake, taking diuretics, usually
spironolactone (Aldactone) and furosemide (Lasix), are
effective for relieving ascites in 90% of patients. Sometimes
stopping drinking is enough to reverse this complication.
Previously, spironolactone was usually given alone, but experts
now use it by itself only in patients with minimal fluid build-up.
Patients should be monitored carefully for excessive and too
rapid fluid loss, which can set off complications, including

hypokalemia (dangerously low potassium levels), kidney
failure, or encephalopathy. Weight loss from diuretics usually
should not exceed one or two pounds a day, but there is no limit
for patients with massive swelling. Restricting fluid is not
usually necessary unless sodium levels in the blood are very
low. Physicians often recommend bed rest for patients with
ascites, but many experts believe this is not necessary and say
that studies do not support its benefits.
Paracentesis. If diuretics are not successful and the ascites is

very tense, patients may require large-volume paracentesis. In
this procedure, albumin (protein) is administered intravenously
while large volumes of fluid are removed through a tube in the
abdomen. Research indicates that four to six liters is usually
effective and safe. If the ascites does not respond to treatments,
paracentesis may need to be repeated every two weeks or more
frequently and up to 10 liters may need to be removed. Patients
who require this are probably not complying with dietary
requirements.
Liver Transplantation. Liver transplantation should be

considered for certain patients with ascites [ see below ],
particularly since the wait time for a liver donor is now very
long.
Peritoneovenous Shunting. Peritoneovenous (LeVeen, Denver)

shunting is an older, more invasive procedure, involving
insertion of a tube, or shunt, under the skin that routes the fluid
from the abdomen into the jugular vein. This procedure does
not improve survival, however, and can have serious
complications, including infection, blood clots, and rupture of
blood vessels in the esophagus. It is now reserved for patients
who are not candidates for repeat paracentesis or liver
transplantation.
Procedures for Preventing and Treating

Variceal Bleeding.
Endoscopic Sclerotherapy. Endoscopic sclerotherapy involves
inserting a tube through the mouth and then injecting into the
esophagus agents called sclerosants, which toughen the tissue
around the blood vessels. It may help prevent recurring bleeding
from the esophagus, although not from the stomach. The

treatment is repeated over a period of two or three months.

Unfortunately, bleeding still occurs in up to 50% of patients,
and complications of the procedure are high--about 40%. It is
also an unpleasant procedure and many patients cannot tolerate
it. A major analysis reported, however, that when used with the
sclerosant polidocanol, sclerotherapy is effective in helping to
reduce the risk for bleeding, particularly in high-risk patients.
Transjugular Intrahepatic Portosystemic Shunt (TIPS). An

alternative procedure is the transjugular intrahepatic
portosystemic shunt (TIPS). This procedure uses a thin,
flexible, metal tube, or stent, threaded through a catheter in a
vein that travels from the neck to the liver. There, the stent
remains, where it expands and forces blood vessels to reroute
around the scarred liver. Blockage or closure of the shunt can
develop over time. Comparative studies have not found a
survival advantage with TIPS over sclerotherapy; TIPS appears
to be better in preventing re-bleeding only when it is performed
shortly after a bleeding episode. TIPS also poses a higher risk
for encephalopathy than sclerotherapy does. At this time,
experts generally recommend TIPS only for patients who
cannot tolerate sclerotherapy, who are unlikely or unable to
comply with the repeated procedures necessary for
sclerotherapy, or who have poor blood circulation.
Endoscopic Band Ligation. Endoscopic band ligation simply

involves wrapping latex bands around the bleeding varices,
shutting off the blood supply. There are few complications. In
some studies it requires fewer sessions than sclerotherapy,
although one study found that it was not as effective as TIPS in
preventing re-bleeding.
Balloon Tamponade. Balloon tamponade employs a tube

inserted through the nose and down through the esophagus until
it reaches the upper part of the stomach. A balloon at the tube's
end is inflated and positioned tightly against the esophageal
wall. It is usually deflated in about 24 hours. Serious
complications can occur, the most dangerous being rupture of
the esophagus. A recurrence of bleeding following this
procedure is common.

Drug Therapies for Lowering Portal

Hypertension and Preventing Variceal
Bleeding.
Beta-Blockers. Beta-blockers, including propranolol and
nadolol, used alone or with sclerotherapy, reduce the heart rate
and can lower portal vein pressure, thereby reducing bleeding.
Beta-blockers are also used as a primary approach for
prevention of recurrence. It is not yet clear if these drugs are
more effective against bleeding than sclerotherapy, but they are
inexpensive and safe. Studies are indicating that a combination
of a nitrate, such as isosorbide mononitrate, with a beta-blocker
is less expensive and may, in some cases, be even more
effective than sclerotherapy. Nitrates alone are not as effective
as beta-blockers, but may be useful in certain people. Lifetime
therapy may be necessary.
Vasopressin. Vasopressin (Pitressin) with nitroglycerine is often

used. Vasopressin poses some risk to the heart, however, and it
is not clear whether it is actually helpful.
Somatostatin and Similar Drugs. Somatostatin is a hormone

that might help prevent variceal bleeding from portal
hypertension; in one study, the drug was administered within 12
hours of bleeding it and continued for 120 hours, the success of
subsequent sclerotherapy was higher than in those who did not
take the drug. Octreotide (Sandostatin), a drug that resembles
somatostatin, has been tested with mixed results; some have no
benefits in survival rates, although one indicated that it might
improve survival rates compared to sclerotherapy after taking
the drug for a year. It may reduce the risk of early re-bleeding in
patients treated with either beta-blockers, sclerotherapy, or both.
Angiotensin II Receptor Antagonists. Drugs known as

angiotensin II receptor antagonists, including losartan (Cozaar),
are being studied for lowering portal pressure.
Treating and Preventing Abdominal

Infection (Peritonitis).
Antibiotics are administered when ascites fluid examination and
tests indicate the presence of infection. For a first episode, the
antibiotic cefotaxime is typically administered intravenously,
requiring hospitalization. Treatment usually lasts 10 days but
research indicates that five days may be sufficient for certain

patients. Some research indicates that the oral antibiotic
ofloxacin may be as effective and without complications,
allowing patients to be treated at home.
In advanced cirrhosis, the risk for serious abdominal infection is

high and the antibiotic norfloxacin is often prescribed
preventively against specific organisms that infect the
abdominal cavity. One study found that preventive antibiotics
were very cost effective in high-risk patients. Another study
reported, however, that patients who took norfloxacin became
susceptible to Staphylococcal infections, which are not
ordinarily a problem in cirrhosis, and their survival rates were
similar to patients who did not take the antibiotic. Long-term
treatments with norfloxacin or similar antibiotics may increase
the risk for fungal infections after liver transplantation. More
research is needed.
Preventing and Treating Encephalopathy.

The first step in managing encephalopathy is to treat any
precipitating cause, if known, such as bleeding, high ammonia
levels, low oxygen, infection, dehydration, or use of sedatives.
Mild encephalopathy is managed by directing therapy toward
eliminating ammonia in the intestine. The first step is to restrict
animal protein, substituting meats and dairy products with
vegetable protein, such as soy, and amino acid supplements.
Enemas, which clean out the intestine, may be effective.
Lactulose (Cephulac) and lactitol, known as disaccharides, help
lower blood ammonia levels. Antibiotics, such as
metronidazole, rifamycin, or neomycin, are effective in
reducing levels of ammonia-producing bacteria in the intestine,
although long-term use of these drugs can cause toxic side
effects. Adding non-ammonia producing bacteria, including L.
acidophilus and E. faecium , to the intestine is showing promise
as a safe and effective treatment. In some studies, taking zinc
supplements have lowered ammonia levels in some patients
who were zinc-deficient, a common problem in cirrhosis. Some
studies indicate that manganese poisoning may be partially
responsible for encephalopathy in cirrhosis. Studies are needed
to determine if drugs that remove manganese improve this
complication.

Treatment for Gastrointestinal Bleeding.

Gastrointestinal (GI) bleeding is often first treated with
medications to reduce stomach acid. Reduced clotting factors or
platelets are common causes of GI bleeding in people with
alcoholic cirrhosis. Some will respond to three days of injected
vitamin K. People with alcoholism also often require folic acid.
Transfusions of replacement clotting factors or platelets may be
needed.
What Are The Primary

Treatments For Cirrhosis?
In general, no treatments exist for the basic mechanisms that
cause cirrhosis and its complications, although some therapies
are in clinical trials.
Treatment for Primary Biliary Cirrhosis.

Colchicine, a drug that inhibits collagen (a protein in the body
the makes up scar tissue) has produced some improvement in
liver function and survival. Side effects include nausea and
gastrointestinal distress. Ursodiol or ursodeoxycholic acid
(Actigall), a drug generally used to treat gallstones, improves
symptoms and other aspects of the disease and has only minor
side effects; unfortunately, it is very expensive and does not
appear to prolong survival. A similar drug,
tauroursodeoxycholate, may be more effective. Azathioprine,
methotrexate, and cyclosporine are drugs that suppress the
immune system, and, in studies, have shown modest benefits on
survival, but they have severe side effects. One study reported
that methotrexate caused remission in a group of patients, but,
although encouraging, other studies have indicated that many
patients do not respond to this drug. Corticosteroids, which
reduce inflammation, have been helpful in improving liver
function and symptoms, but patients should take measures to
prevent osteoporosis, a side effect of both the disease and these
drugs. The itching caused by primary biliary cirrhosis can be
relieved by taking cholestyramine with meals. The drug
naltrexone relieved itching in one study. (High doses of this
drug are toxic to the liver, but the low doses used in the study
were safe.) Phototherapy, which uses light, may also reduce
itching. Because primary biliary cirrhosis affects fat absorption,
patients may need high doses or injections of important

fat-soluble vitamins, including K, D, A, and E. Zinc
supplements may also be required. For steatorrhea, agents
called medium-chain triglycerides may be helpful.
Treatment for Other Forms of Cirrhosis.

Secondary biliary cirrhosis caused by blockage in the bile ducts
can be relieved by surgery. For hemochromatosis, weekly
bleedings (phlebotomies) may be performed until iron levels are
normal, then repeated as needed. If treatment is given before
cirrhosis develops, life expectancy may be normal.
D-penicillamine is the drug most used for Wilson's disease.
Investigative Therapies for Cirrhosis.

Researchers are testing certain drugs that will redress the
imbalances in circulation that lead to portal hypertension and
ascites. Of particularly interest are agents called V2 receptor
antagonists that may reverse the dilation in blood vessels that
lead to salt and fluid retention. Animal studies are reporting
success with genetic therapies using human hepatocyte growth
factor (HGF); even if feasible, however, human trials are years
away.
Liver transplantation may be an option for people with primary
biliary cirrhosis, for some people with alcoholic cirrhosis
(usually those who have completely abstained from alcohol for
more than six months), and for people with chronic hepatitis. It
should be noted, however, that hepatitis B patients have a
success rate of only 50% to 60% because of recurrence. (The
success rate is higher in those who have hepatitis D.) Either
lamiviudine or monthly infusions of hepatitis B immune
globulin (HBIg) after transplantation may help prevent
recurrence of hepatitis B after liver transplantation. Hepatitis C
also commonly returns in transplanted livers and progresses to
cirrhosis within an average of 51 months in 8% of patients.
Some people with cirrhosis and small localized liver cancers
may also be suitable candidates. Patients should seek medical
centers that have performed more than 50 transplants per year
which produces better than average results. Current survival

rates after a year are now nearly 90%, mainly due to

improvements in infection prevention. For people under 60, the
five-year survival rate is about 75%. It is much lower in elderly
people. Unfortunately, the waiting time for a liver doubled
between 1991 and 1996 to 241 days. Given the large number of
people with hepatitis C, this situation will almost certainly
worsen over in future years.
Where Else Can Help For

Cirrhosis Be Found
Primary Biliary Cirrhosis Patient Support Network Box 177

Tamworth, ON KOK 360
Canada
call (613-379-2534)
National Institute of Diabetes and Digestive and Kidney
Diseases (NIDDK)
National Digestive Diseases Information Clearinghouse
2 Information Way
Bethesda, MD 20892-3570
(http://www.niddk.nih.gov/ )
Centers for Disease Control and Prevention, Hepatitis Branch

1600 Clifton Road NE.
Mail Stop G37
Atlanta, GA 30333
For a special number on hepatitis call (888-4HEPCDC) or
(888-443-7232 )
or on the Internet
(http://www.cdc.gov/ncidod/diseases/hepatitis/hepatitis.htm
) This is an important source on hepatitis.

American Association for the Study of Liver Diseases

1200 19th Street NW, Suite 300
Washington, DC 20036-2422
call (202 429-5179) or on the Internet
(http://hepar-sfgh.ucsf.edu )
This group publishes the journal Hepatology.
For good journal abstracts on the web
(http://www.hepatology.org/ ) and
(http://www.ltsjournal.org/ )
Hepatitis Foundation International

30 Sunrise Terrace
Cedar Grove, New Jersey 07009
call (800-891-0707) or on the Internet (http://www.hepfi.org
)
This organization focuses just on viral hepatitis. It provides

educational materials, offers support by phone, and gives
referral to other physicians.
American Liver Foundation

75 Maiden Lane, Suite 603
New York, NY 10038
call (800-GO LIVER) or (800-465-4837) or on the Internet
(http://www.liverfoundation.org/ )
The foundation has regional chapters, some with support
groups. The hepatitis hotline provides patient information,
brochures, and video and audio tapes.
American Gastrointestinal Association

7910 Woodmont Ave., Seventh Floor,
Bethesda, MD 20814
call (301-654-2055) or on the Internet (http://www.gastro.org/

This is an association for physicians and other professionals.

They provide names of gastroenterologists in local areas.
On the Internet
Medical Center
For information on organ transplantation, United Network for
Organ Sharing (http://www.unos.org/ )

)
Alcoholics Anonymous
General Service Office
475 Riverside Drive
New York, NY 10015
National Clearinghouse of Alcohol and Drug Information

PO Box 2345
Rockville, MD 20852
Call (800-729-6686) or on the Internet (http://www.health.org/
)
National Institute on Alcohol Abuse and Alcoholism

6000 Executive Boulevard - Willco Building
Bethesda, Maryland 20892-7003
On the Internet (http://www.niaaa.nih.gov/ )

National Council on Alcoholism

12 West 21 Street
New York, NY 10010
Call (800-NCA-CALL) or on the Internet
(http://www.ncadd.org/ ).
On the Internet:
Medical Center
Or for organ donation information see (www.organdonor.org
)

)
Recent Literature
AASLD Practice guidelines management of adult patients with
ascites caused by cirrhosis. Hepatology, January 1998 Vol. 27.
Combination therapy for hepatitis C infection (and other

articles). The New England Journal of Medicine, 11/19/98, Vol.
339.
Effects of propranolol on the hepatic hemodynamic response to

physical exercise in patients with cirrhosis. Hepatology 1998,
Vol. 28.

High prevalence of sleep disturbance in cirrhosis. Hepatology,

1998 Vol. 27.
Mechanisms of disease: nitric oxide as a mediator of

hemodynamic abnormalities and sodium and water retention in
cirrhosis. The New England Journal of Medicine, 8/20/98 Vol.
339.
ST/ALT ratio predicts cirrhosis in patients with chronic

hepatitis C virus infection. American Journal of
Gastroenterology, January 1998
Treatment of chronic hepatitis B infection. The New England

Journal of Medicine, 7/9/98, Vol. 338

printing.
print the handout.


April 22, 2000
Conde Petra
April 14, 2000

SOB


of cirrhosis.
knowledge.




Bibliography
110(4):253-255, 1989(editorial)


Saunders







70(5):1081-1091, 1986


8(1):85-99, 1986








WB Saunders



April 22, 2000
Conde Petra
April 14, 2000
Question 4 - Ascites
wave, shifting dullness, and prominent flanks. With what entity are these
physical findings consistent and what are the associated complications?
Ascites refers to the accumulation of fluid in the peritoneal cavity. Ascites is

appreciated on physical examination by detecting shifting dullness, bulging flanks,
or a fluid wave. Cirrhosis is a common cause of ascites, but other causes include
pancreatitis, a ruptured intraabdominal viscus, right heart failure, nephrotic
syndrome, trauma, and Meigs' syndrome. In cirrhosis multiple factors lead to
ascites, and these include: low albumin with loss of oncotic force within the
vascular and interstitial spaces, renal retention of sodium and water, and portal
hypertension with increased hepatic lymph production and transudation. Ascites
resulting from cirrhosis is most commonly a transudate with protein content less
than 1.1 g/dl. The diagnosis of ascites needs to be confirmed by ultrasound unless
its presence is unquestionable by physical examination. Other imaging studies
include CT scan and plain view of the abdomen (characteristic ground-glass
appearance).
Management of uncomplicated ascites involves the following steps: (1) restriction

of sodium and water (2 g Na/day and 2000 ml/day water) and (2) use of diuretics,
specifically spironolactone. This is an aldosterone antagonist that has been proven
to be more effective than loop diuretics because of its independence on blood flow
for presentation to the nephron in patients with cirrhosis/ascites. Aldactone
(spironolactone) is started at a dose of 50 to 100 mg bid and increased to 400 mg
daily. If diuresis is not adequate with the 200 mg/day dose, then furosemide can be
added (usual starting dose 40 mg/day). It is important to remember that
spironolactone prevents the renal excretion of potassium and therefore should not
be used in patients with renal insufficiency.
scites that is difficult to control by standard management is divided into two

categories. In the first group, unresponsive ascites, diuretic use does not result in
adequate natriuresis. In this group of patients the diuretic metolazone (Zaroxolyn)
may be cautiously tried because it acts on the proximal tubule by inhibiting sodium
reabsorption. In the second group, refractory ascites, natriuresis is achieved only in
expense of a reduction in GFR. In this group a 4- to 8-week waiting period is
necessary before labeling the patient refractory. During this waiting period
reversible insults (alcoholic hepatitis, acute tubular necrosis, bacterial peritonitis)
should be resolved before the diagnosis is accepted. Once ascites is confirmed to
be refractory, peritoneovenous shunting should be considered. The LeVeen shunt
is a subcutaneous stent that results in the transfer of fluid from the peritoneal cavity
to the internal jugular vein via a one-way valve.
Large-volume paracentesis can also be used in the management of ascites. In

many studies removal of large fluid volume, 4 to 6 L/day, resulted in shorter
hospital stays and fewer complications than traditional diuretic therapy. In

refractory ascites, large-volume paracentesis is equally as effective as the LeVeen
shunt. One of the potential complications of this procedure is a decrease in the
intravascular volume. It is, therefore, recommended that during large-volume
paracentesis patients receive an intravenous infusion of albumin at the rate of 10
g/L of fluid removed. Finally, truly refractory ascites may be an indication for liver
transplantation.
Spontaneous bacterial peritonitis (SBP) is an infection of the abdominal peritoneum

and occurs in 10% to 25% of cirrhotic patients. The exact pathogenesis is
unknown. The most likely mechanism involves the hematogenous seeding of the
peritoneal cavity with enteric bacteria (translocation). These organisms enter the
portal venous system via collaterals, avoiding the reticuloendothelial system of the
liver. Another contributing factor is the low level of protein found in ascitic fluid. This
reflects low complement levels (low opsonic activity), which in turn leads to
decreased phagocytosis of bacteria. The incidence of SBP is higher in patients
with ascitic fluid protein levels below 1.0 g/dl. The most common bacteria
responsible are Escherichia coli, Klebsiella pneumoniae, Streptococcus
pneumoniae, alpha-hemolytic streptococci and group D streptococci. Clinically,
patients present with fever and/or abdominal pain; however, it is important to
remember that one third of patients are completely asymptomatic. Laboratory data
usually reveal a peripheral leukocytosis. Diagnosis is made by obtaining ascitic
fluid for analysis. Direct inoculation of blood culture vials (broth) with 10 ml of fluid
increased culture yield. Patients with suggestive clinical picture and fluid containing
greater than 500 WBC/ml with 50% or higher granulocytes should be started on
antibiotics before culture results confirm the presence of bacteria. Treatment of
SBP involves intravenous administration of antibiotics astreonam + vancomycin or
cefotaxime, with activity versus gram-negative bacilli. Follow-up examination of the
fluid at 48 hours should be done to verify improvement. Prophylactic therapy for
SBP is controversial. One antibiotic that has proven to decrease the incidence of
recurrent SBP is norfloxacin (400 mg/day). This fluoroquinolone acts by selectively
eliminating gram-negative bacilli from the intestinal flora, which is followed by a
reduction in the rate of SBP.
Read more about ascites and its treatment:

Proceed With Case
Ascites and Spontaneous Bacterial Peritonitis

Complications: Ascites And Edema

Goroll: Primary Care Medicine, 3rd ed.
Copyright 1995 Lippincott-Raven Publishers
Prevention And Treatment Of Complications: Ascites

Rakel: Conn's Current Therapy 1999, 51st ed.

Outcome of 100 patients after transjugular intrahepatic portosystemic shunt for

variceal hemorrhage [see comments]
Sahagun G - Am J Gastroenterol - 1997 Sep; 92(9): 1444-52
Patient selection is important in studying the impact of large-volume paracentesis

on intravascular volume [editorial; comment]
Runyon BA - Am J Gastroenterol - 1997 Mar; 92(3): 371-3
Hemoperitoneum in patients with ascites.

Akriviadis EA - Am J Gastroenterol - 1997 Apr; 92(4): 567-75
From NIH/NLM MEDLINE; NCI CANCERLIT
CLINICAL ISSUES IN THE MANAGEMENT OF ALCOHOLIC LIVER DISEASE

Steindl PE - Clinics in Liver Disease - 1998 Nov; 2(4); 765-779
Cardiovascular, renal, and neurohumoral responses to single large-volume

paracentesis in patients with cirrhosis and diuretic-resistant ascites [see
comments]
Peltekian KM - Am J Gastroenterol - 1997 Mar; 92(3): 394-9
Clinical practice guidelines for the management of cirrhotic patients with ascites.
Committee on Ascites of the Italian Association for the Study of the Liver.
Salerno F - Ital J Gastroenterol Hepatol - 1999 Oct; 31(7): 626-34
Furosemide
Mosby's GenRx, 10th ed.
Copyright 2000 Mosby, Inc.
Spironolactone


April 22, 2000
Conde Petra

Company
Section 9 - Liver
78 - Ascites and Spontaneous Feldman: Sleisenger & Fordtran's Gastrointestinal and Liver Disease,
Bacterial Peritonitis
Introduction
PATHOGENESIS OF ASCITES
FORMATION 1310
CLINICAL FEATURES OF
ASCITES Chapter 78 - Ascites and
COMPLICATIONS OF ASCITES Spontaneous Bacterial
TREATMENT OF ASCITES Peritonitis
SUMMARY OF TREATMENT OF
CIRRHOTIC ASCITES
Bruce A. Runyon
PROGNOSIS
REFERENCES PATHOGENESIS OF ASCITES FORMATION, 1310

Liver Disease, 1310
Noncirrhotic Ascites, 1310
CLINICAL FEATURES OF ASCITES, 1311
History, 1311
Physical Examination, 1311
Diagnosis, 1312
Abdominal Paracentesis, 1312
Ascitic Fluid Analysis, 1315
COMPLICATIONS OF ASCITES, 1321
Infection, 1321
Pathogenesis, 1321
Treatment, 1324
Tense Ascites, 1326
Pleural Effusions, 1326
Abdominal Wall Hernias, 1327

TREATMENT OF ASCITES, 1327
Low Albumin Gradient Ascites, 1327
High Albumin Gradient Ascites, 1327
SUMMARY OF TREATMENT OF CIRRHOTIC ASCITES,
1331
PROGNOSIS, 1331
"Ascites" is of Greek derivation (askos) and refers to a bag or
sack. The word is a noun and describes pathologic accumulation
of fluid within the peritoneal cavity. The adjective, ascitic, is
used in conjunction with the word fluid to describe the liquid
per se.


Saunders Company
1310
Chapter 78 - Ascites and Spontaneous Bacterial

Peritonitis
Bruce A. Runyon
PATHOGENESIS OF ASCITES FORMATION, 1310

Liver Disease, 1310
Noncirrhotic Ascites, 1310
CLINICAL FEATURES OF ASCITES, 1311
History, 1311
Physical Examination, 1311
Diagnosis, 1312
Abdominal Paracentesis, 1312
Ascitic Fluid Analysis, 1315
COMPLICATIONS OF ASCITES, 1321
Infection, 1321
Pathogenesis, 1321
Treatment, 1324
Tense Ascites, 1326
Pleural Effusions, 1326
Abdominal Wall Hernias, 1327
TREATMENT OF ASCITES, 1327
Low Albumin Gradient Ascites, 1327
High Albumin Gradient Ascites, 1327
SUMMARY OF TREATMENT OF CIRRHOTIC ASCITES, 1331
PROGNOSIS, 1331
"Ascites" is of Greek derivation (askos) and refers to a bag or sack. The word is a noun and describes
pathologic accumulation of fluid within the peritoneal cavity. The adjective, ascitic, is used in
conjunction with the word fluid to describe the liquid per se.



Saunders Company
PATHOGENESIS OF ASCITES FORMATION

Liver Disease
Ascites forms in the setting of cirrhosis as the result of the sequence of events detailed in Figure 78-1 .
The most recent theory regarding the formation of ascites, the peripheral arterial vasodilation hypothesis,
has proposed that both of the older hypotheses (i.e., the underfill and the overflow theories) are correct,
but that each is operative at a different stage. [1] [2] [3] [4] The first abnormality that develops en route to
fluid retention appears to be portal hypertension. The portal pressure increases above a critical threshold,
and nitric oxide levels
Figure 78-1 Pathogenesis of ascites formation in the setting of cirrhosis.
increase. [5] [6] Nitric oxide leads to vasodilation. [6] As the state of vasodilation worsens, plasma levels of
vasoconstrictor, sodium-retentive hormones increase, and renal function deteriorates; ascites develops
(i.e., decompensation occurs).
The explanation of the neurohumoral excitation, which is characteristic of volume depletion, in the
setting of volume overload in the cirrhotic patient with ascites may be connected with volume sensors.
Animals have very sophisticated systems for detection and preservation of vascular perfusion pressures
and intravascular osmolality. However, the organism's ability to detect changes in intravascular volume
status (especially volume overload) is limited and is linked to pressure receptors. This may partially
explain the paradox of dramatic volume overload in the face of sympathetic nervous traffic and hormone
levels that are indicative of intravascular volume depletion.
Noncirrhotic Ascites
The mechanism of fluid retention in patients with malignancy-related ascites depends on the location of
the tumor. Peritoneal carcinomatosis appears to cause ascites by "exudation" of proteinaceous fluid from
tumor cells lining the peritoneum. Extracellular fluid enters the peritoneal cavity to re-establish oncotic
balance. [7] Fluid accumulates in patients with massive liver metastases due to portal hypertension caused
by stenosis or occlusion of portal veins by tumor nodules or tumor emboli. [8] In patients with
hepatocellular carcinoma, ascites forms owing to the underlying cirrhosis-related portal hypertension or
tumor-induced portal vein thrombosis. Chylous ascites due to malignant lymphoma appears to be caused
by lymph node obstruction by a tumor and by rupture of chyle-containing lymphatics.
Ascites can complicate high-output or low-output heart failure or nephrotic syndrome. As in cirrhosis,
there appears to be decreased effective arterial blood volume and subsequent activation of the

vasopressin, renin-aldosterone, and sympathetic nervous systems. [9] This leads to renal vasoconstriction
and sodium and water retention. Fluid then weeps from the congested hepatic sinusoids as lymph, as in
cirrhotic ascites.
1311
Tuberculosis, chlamydia infection, and coccidioidomycosis probably cause ascites due to exudation of
proteinaceous fluid, such as in peritoneal carcinomatosis. Spontaneous bacterial peritonitis (SBP) does
not appear to cause fluid to accumulate; infection develops only with pre-existing ascites.
In patients with pancreatic or biliary ascites, fluid forms by leakage of pancreatic juice or bile into the
peritoneal cavity or by a "chemical burn" of the peritoneum. After abdominal surgery, especially
extensive retroperitoneal dissection, lymphatics may be transected and leak lymph for variable periods of
time. [10] [11] The formation of ascites in this condition is similar to that of malignant chylous ascites (i.e.,
lymphatic leak).


Saunders Company
Figure 78-1 Pathogenesis of ascites formation in the setting of cirrhosis.


Saunders Company
CLINICAL FEATURES OF ASCITES

History
Most patients ( 80%) with ascites in the United States have cirrhosis. The two most common causes of
cirrhosis are alcohol and hepatitis C; many patients have both. [12] In approximately 20% of patients with
ascites, there is a nonhepatic cause of fluid retention (Table 78-1) (Table Not Available) . Ascites
frequently develops as part of the patient's first "decompensation" of alcoholic liver disease. Ascites can
develop early in alcoholic liver disease in the precirrhotic, alcoholic hepatitis stage. At this stage, portal
hypertension and the resultant predisposition to
TABLE 78-1 -- Causes of Ascites
(Not Available)
Reported by permission from the American College of Physicians. Adapted from Runyon, B. A.,
Montano, A. A., Akriviadis, E. A., et al. The serum-ascities albumin grasient is superior to the
exudate-transudate concept in the differential diagnosis of ascites. Ann. Intern. Med. 117:215. 1992.
sodium retention are reversible with abstinence. [13] Patients with precirrhotic alcoholic liver disease may
lose their predisposition to fluid retention when they reduce or cease alcohol consumption. In contrast,
patients who develop ascites in the setting of nonalcoholic liver disease tend to be persistently troubled
with fluid retention thereafter, probably because of the late stage at which ascites forms in nonalcoholic
liver disease and the lack of effective therapy other than liver transplantation.
Patients with ascites should also be questioned about risk factors other than alcohol (e.g., intravenous
drug use, transfusions, homosexuality, acupuncture, tattoos, ear piercing, country of origin). Quite
commonly, the cause of ascites in a middle-aged or elderly woman is viral cirrhosis caused by a remote,
almost forgotten blood transfusion. Another cause of "cryptogenic" cirrhosis and ascites in elderly
women is nonalcoholic steatohepatitis (NASH) from lifelong obesity. [14]
Patients with a long history of stable cirrhosis and sudden development of ascites should be suspected of
harboring a hepatocellular carcinoma as the precipitating cause of decompensation.
Patients with ascites who have a history of cancer should be suspected of having malignancy-related
ascites. However, cancer in the past does not guarantee a malignant cause (e.g., patients with
tobacco-related lung cancer and alcohol abuse may have cirrhotic ascites). Breast, lung, colon, and
pancreatic primaries are regularly complicated by ascites. [8] Abdominal pain may be helpful in
differentiating patients with malignancy-related ascites from patients with cirrhotic ascites. The former
condition is frequently painful, whereas the latter is usually not, unless there is superimposed bacterial
peritonitis or alcoholic hepatitis.
A past history of heart failure can be helpful in raising the possibility of cardiac ascites. Alcoholics who

develop ascites may have alcoholic cardiomyopathy or alcoholic liver disease, but they do not usually
have both.
Tuberculous peritonitis is usually manifested by a fever and abdominal pain. Many of these patients are
recent immigrants from an endemic area. In the United States, half of the patients with tuberculous
peritonitis have underlying alcoholic cirrhosis as a second cause for ascites formation.
Patients with acute hemorrhagic pancreatitis or a ruptured pancreatic duct from chronic pancreatitis or
trauma may develop ascites. A small percentage of patients on hemodialysis develop troublesome
ascites. Chlamydia Fitz-Hugh-Curtis syndrome may cause inflammatory ascites in a sexually active
woman. [15] Patients who develop ascites and anasarca in the setting of diabetes should be suspected of
having nephrotic ascites. Ascites developing in a patient with signs and symptoms of myxedema should
prompt measurement of thyroid function. Serositis in connective tissue diseases may be complicated by
ascites. [16]
Physical Examination
On the basis of the history and the appearance of the abdomen, the diagnosis of ascites is readily
suspected and easily confirmed on a physical examination. The presence of a full, bulging abdomen
should lead to percussion of the flanks. If the amount of flank dullness is greater than usual (i.e., if the
percussed air-fluid level is higher than that normally found on
1312
the lateral aspect of the abdomen with the patient supine), then "shifting" should be checked for. If there
is no flank dullness, there is no reason to check for shifting. Approximately 1500 mL of fluid must be
present before dullness is detected. [17] If no flank dullness is present, the patient has less than a 10%
chance of having ascites. [17] It is not worth testing for a fluid wave. [17]
An obese abdomen may be diffusely dull to percussion, and an abdominal ultrasound may be required to
determine if fluid is present. Ultrasonographic scans can detect as little as 100 mL of fluid in the
abdomen. [18] Gaseous distention of the bowel, a thick panniculus, and an ovarian mass can mimic
ascites. Gaseous distention should be readily apparent on percussion. Ovarian masses usually cause
tympanitic flanks with central dullness.
The presence of palmar erythema, large pulsatile vascular spiders, large abdominal wall collateral veins,
or fetor hepaticus are very suggestive of the presence of parenchymal liver disease and portal
hypertension. The presence of large veins on the patient's back suggests blockage of the inferior vena
cava. An immobile mass in the umbilicus, the Sister Mary Joseph nodule, is very suggestive of peritoneal
carcinomatosis.
The neck veins of patients with ascites should always be specifically examined. Constrictive pericarditis
is one of the few curable causes of ascites. Most patients with cardiac ascites have impressive jugular
venous distention. Some patients have no visible jugular venous distention but have such high central
venous pressures that their bulging forehead veins rise to the tops of their skulls. When patients with liver
disease have peripheral edema, it is usually in the lower extremities and may occasionally involve the
abdominal wall. Nephrotic patients and patients with cardiac failure may have total body edema (e.g.,
anasarca).

Diagnosis
Although the diagnosis of ascites is suspected on the basis of the history and physical examination, final
confirmation is based on successful abdominal paracentesis. The diagnosis of the cause of ascites
formation is based on the results of the history, physical examination, and ascitic fluid analysis. In
general, few other tests are required. Radiologic and endoscopic procedures usually add little information
in the initial evaluation of the patient with ascites.
Abdominal Paracentesis
Indications
Abdominal paracentesis with appropriate ascitic fluid analysis is probably the most rapid and
cost-effective method of diagnosing the cause of ascites. Also, because of the 10% to 27% prevalence of
ascitic fluid infection at the time of the patient's admission to the hospital, a surveillance tap performed at
that time may detect an unexpected infection. [19] Not all patients with ascitic fluid infection are
symptomatic; many have very subtle symptoms (e.g., mild hepatic encephalopathy so that only the
family notices the patient's change in mental status). Detection of infection at an early asymptomatic
stage may reduce mortality. Therefore, the author advocates sampling ascitic fluid in all inpatients and
outpatients with new onset ascites and in all patients admitted to the hospital with ascites (i.e., a tap at the
time of each hospitalization). Paracentesis should be repeated in patients (whether in the hospital or not)
who develop signs or symptoms or have abnormal laboratory values suggestive of infection (e.g.,
hypotension, abdominal pain or tenderness, fever, encephalopathy, renal failure, acidosis, or peripheral
leukocytosis).
Contraindications
There are few contraindications to paracentesis. Coagulopathy is a potential contraindication; however,

most patients with cirrhotic ascites have coagulopathy. If mild coagulopathy were considered to be a
contraindication to paracentesis, few cirrhotic patients would undergo this procedure. In the author's
opinion, coagulopathy should preclude paracentesis only when there is clinically evident fibrinolysis or
clinically evident disseminated intravascular coagulation. [20] These conditions occur in fewer than one in
1000 taps. There is no data-supported cut-off of coagulation parameters beyond which paracentesis
should be avoided. Even after multiple paracenteses, patients with severe prolongation of prothrombin
time do not usually develop bloody ascites. Cirrhotic patients who do not have clinically obvious
coagulopathy simply do not bleed excessively from needlesticks unless a blood vessel is entered. [20]
Studies regarding complications of paracentesis in patients with ascites documented no deaths or
infections caused by the paracentesis. [20] [21] There were no episodes of hemoperitoneum or bowel entry
by the paracentesis needle. Complications included only approximately 2% abdominal wall hematomas,
despite the fact that 71% of the patients had an abnormal prothrombin time; 21% had a prothrombin time
that was prolonged by 5 seconds more. [20] Complication rates may be higher when paracentesis is
performed by inexperienced operators.
Some physicians give blood products (fresh-frozen plasma or platelets) routinely before paracentesis in
cirrhotic patients with coagulopathy, presumably in order to prevent hemorrhagic complications. This

policy is not supported by data. Since a hematoma that requires a transfusion develops in only
approximately 1% of patients who undergo paracentesis without prophylactic transfusions of plasma or
platelets, approximately 100 to 200 U of fresh-frozen plasma or platelets would have to be given to
prevent transfusion of approximately 2 U of red blood cells. Also, despite our screening efforts, there
remains a finite risk of post-transfusion hepatitis. Since most patients with cirrhosis and ascites die when
they develop superimposed acute hepatitis, a small but real risk of death from hepatitis superimposed on
cirrhosis is not an acceptable alternative to a trivial reduction in the risk of bleeding.
On the basis of the complications (reported in the older literature) of paracentesis performed with
large-bore trocars, many physicians have avoided diagnostic paracentesis in the evaluation of the patient
with ascites. However, because of the documented safety of this procedure and the frequency of ascitic
fluid infection, paracentesis is now being performed more frequently.
1313
Patient Position, Choice of Needle Entry Site, and Needle
The volume of fluid in the abdomen and the thickness of the abdominal wall determine in part the
positioning of the patient in preparation for the procedure. Patients with large-volume ascites and thin
abdominal walls can be successfully "tapped" in the supine position with the head of the bed or
examining table slightly elevated. Patients with less fluid can be placed in the lateral decubitus position
and tapped in the midline, or these patients can be tapped in the right or left lower quadrant while in the
supine position (see later). Patients with very small amounts of fluid may be tapped successfully only in
the face-down position or with ultrasound guidance. [10]
A midline insertion site may be preferable to a lower quadrant insertion site for a diagnostic tap in a
nonobese patient, unless there is a midline surgical scar. [20] The midline caudad to the umbilicus is
avascular unless there is an unusual collateral vein there. A large collateral is frequently located in the
midline cephalad to the umbilicus in the setting of portal hypertension; this area should be avoided.
Therefore, needles inserted in the midline should be placed caudad to the umbilicus.
Obese patients pose special problems. Using ultrasound guidance, the author has found that the
abdominal wall in the midline is usually substantially thicker than that of the lower quadrants in obese
patients. The abdominal wall may be thicker than the length of even a 3.5-inch paracentesis needle. Also,
it is frequently difficult to be certain of the presence or absence of ascites in the obese patient. Ultrasound
guidance with determination of the thickness of the abdominal wall in the midline versus the lower
quadrants is helpful in successfully obtaining ascitic fluid.
Surgical scars also warrant special comment regarding the selection of a site for needle insertion. The
bowel may be adherent to the peritoneal surface of the abdomen near a scar, and needles inserted there
may enter the bowel. [20] The needle must be placed several centimeters from the scar. A long midline
scar precludes a midline paracentesis. Some physicians were taught to choose the right lower quadrant to
avoid the sigmoid colon or spleen or the left lower quadrant to avoid the cecum or liver. Neither of these
recommendations is supported by data. When a midline site is inappropriate because of a scar or small
volume of ascites, the author prefers a site two fingerbreadths cephalad and two fingerbreadths medial to
the anterior superior iliac spine. The quadrant that is duller to percussion is selected as the site for needle
entry. In the patient with multiple abdominal scars, ultrasound guidance may be required. [20]

The author prefers to use standard metal 1.5-inch needles--22-gauge for diagnostic taps and 16-gauge for
therapeutic taps. Obese patients may require spinal needles (i.e., 3.5-inch needles). Recently, 15-gauge
and 17-gauge five-hole 3.25-inch needles have been produced specifically for therapeutic abdominal
paracentesis; these needles may replace spinal needles for paracentesis of obese patients. Steel needles or
blunt steel cannulas with removable steel stylets are preferable to plastic-sheathed cannulas because of:
(1) the risk of the plastic sheath shearing off into the peritoneal cavity, and (2) the tendency of the plastic
sheath to kink and obstruct the flow of fluid after the cannula is removed. Metal needles do not puncture
the bowel unless the bowel is adherent to the scar or there is severe gaseous distention. Unless the needle
is allowed to drift subcutaneously, the needle (or blunt steel cannula) can be left in the abdomen during a
therapeutic tap for over one hour without injury. Larger bore needles or cannulas may permit more rapid
fluid removal, but they leave larger defects if they inadvertently enter the vessels or the bowel.
Technique of Diagnostic Paracentesis
Drapes, gown, hat, and mask are optional equipment, but sterile gloves should be used when performing
this procedure. The skin is disinfected with an iodine solution. The skin and subcutaneous tissue should
be infiltrated with a local anesthetic. The sterile package insert enclosing the gloves can be used as a
sterile field on which to place syringes, needles, gauze, and so forth. If sterile gloves are not used,
cultures of ascitic fluid frequently grow skin contaminants.
In order to prevent leakage of fluid after the needle is withdrawn, the needle is inserted using a Z tract.
This is accomplished by displacing (with one gloved hand) the skin approximately 2 cm in relation to the
deep abdominal wall and then slowly inserting the paracentesis needle mounted on the syringe held in the
other hand. The skin is not released until the needle has penetrated the peritoneum and fluid flows. This
procedure requires that the hand holding the syringe stabilize it and retract its plunger simultaneously. A
steady hand and experience are needed. When the Z tract is used, and the needle is finally removed, the
skin resumes its original position and seals the pathway made by the needle. If the needle is not inserted
using this technique, the fluid leaks out more easily because its pathway is straight. The needle should be
advanced slowly in about 5-mm increments. If the needle is inserted slowly, the operator can see blood if
a vessel is entered; the needle can then be withdrawn immediately before further damage is done. Slow
insertion of the needle also allows the bowel to move away from the needle, thus avoiding puncture of
the bowel. The syringe that is attached to the needle should be aspirated intermittently during insertion. If
continuous suction is applied, bowel or omentum may be drawn to the end of the needle as soon as the
needle enters the peritoneal cavity, occluding flow and resulting in what appears to be an unsuccessful
tap. The author's technique involves inserting the needle in small increments, then aspirating the syringe
for a few seconds while the needle is stationary, then advancing, then aspirating, and so on until the
peritoneum is entered and fluid is aspirated. A slow insertion also allows time for the elastic peritoneum
to tent over and be pierced by the needle. The most common causes for an unsuccessful paracentesis are:
(1) continuous aspiration during insertion of the needle, and (2) rapid insertion and withdrawal of the
needle before the peritoneum is pierced.
Once fluid is flowing, the needle should be stabilized in order to ensure a steady flow. It is not unusual
for the flow to cease intermittently. When this happens, the syringe is removed from the needle, and the
needle is twisted a few degrees. If flow does not restart, the needle is twisted a bit more. If flow does not
restart, the needle is inserted in 1- to 2-mm increments until fluid drips from the needle hub. The syringe
is then reattached, and fluid is aspirated. Occasionally, fluid cannot be aspirated but drips from the needle

hub. In
1314
Figure 78-2 Differential diagnosis of ascites.
1315
this situation, fluid is allowed to drip into a sterile container for collection, as in a lumbar puncture.
Ascitic Fluid Analysis
Gross Appearance
Non-neutrocytic (i.e., neutrophil count <250/mm3 [0.25 109 /L]) ascitic fluid is transparent and usually
slightly yellow (Fig. 78-2) . Very low protein ascitic fluid may have no pigment and resemble water. The
opacity of most cloudy ascitic fluid specimens is caused by neutrophils. The presence of neutrophils
leads to a shimmering effect when a glass tube of the fluid is held in front of a light. Fluid with absolute
neutrophil counts of less than 1000/mm3 (1.0 109 /L) may be almost clear. Fluids with counts over
5000/mm3 (5.0 109 /L) are quite cloudy, and counts over 50,000/mm3 (50.0 109 /L) resemble
mayonnaise.
Specimens of ascitic fluid are frequently tinged with blood or are frankly bloody. A red blood cell count
of 10,000/mm3 (10.0 109 /L) is the threshold for a pink appearance; smaller concentrations result in
clear or turbid fluid. Ascitic fluid with a red blood cell count higher than 20,000/mm3 (20.0 109 /L) is
distinctly red. Many ascitic fluid specimens are bloody due to a traumatic tap; these specimens are
streaked with blood and frequently clot unless the fluid is transferred to the anticoagulant tube (for the
cell count) immediately. In contrast, nontraumatic or remotely traumatic blood-tinged ascitic fluid is
homogeneous and does not clot, because it has already clotted and the clot has lysed. Some patients with
portal hypertension have bloody hepatic lymph leading to bloody ascitic fluid, perhaps because of
rupture of high-pressure lymphatics. Samples from patients with hepatocellular carcinoma are regularly
bloody, but only about 10% of samples from patients with peritoneal carcinomatosis are red. [8] Although
many physicians have the impression that tuberculosis results in bloody ascites, less than 5% of
tuberculous samples are hemorrhagic in the author's experience.
Ascitic fluid is frequently lipid-laden. Lipid opacifies fluid. The degree of "opalescence" of ascitic fluid
ranges from slightly cloudy fluid to completely opaque chylous fluid. Most opaque milky fluid has a
triglyceride concentration higher than 200 mg/dL (2.26 mmol/L) and usually greater than 1000 mg/dL
(11.3 mmol/L). Fluids that have the appearance of dilute skim milk have a concentration between 100
mg/dL (1.13 mmol/L) and 200 mg/dL (2.26 mmol/L). A minority of cirrhotic ascitic fluid samples are

not transparent but not frankly milky. These "opalescent" samples have slightly elevated triglyceride
concentrations ranging from 50 mg/dL (0.56 mmol/L) to 200 mg/dL (2.26 mmol/L). [22] The opacity of
these fluids does not have the shimmering characteristics of ascitic fluid with an elevated white blood
cell (WBC) count. The lipids usually layer out in the refrigerator over a 48- to 72-hour interval. In
contrast with older published reports, most patients with chylous or opalescent ascites have cirrhosis. [22]
[23]
Dark-brown fluid with a bilirubin concentration higher than that of serum usually indicates biliary
perforation. [24] Deeply jaundiced patients have bile-stained ascitic fluid, but the bilirubin level and the
degree of pigmentation to the eye are less than those of corresponding serum. Pancreatic ascites may be
pigmented, owing to the effect of pancreatic enzymes on the red blood cells. The red blood cells may
have to be centrifuged in order to reveal the discolored supernatant. The degree of pigmentation ranges
from tea colored to jet black, such as in hemorrhagic pancreatitis. Black ascites may also be found in the
setting of malignant melanoma.
Ascitic Fluid Tests
The practice of ordering every conceivable body fluid test on every specimen of ascitic fluid is expensive
(higher than $1000 per specimen) and can be more confusing than helpful, especially when unexpectedly
abnormal results are encountered. An algorithmic approach to the analysis of ascitic fluid works quite
well in this author's experience (see Fig. 78-2) . The basic concept is that screening tests are performed
on the initial specimen. Additional testing is performed (only when necessary) based on the results of the
screening tests. Further testing usually necessitates another paracentesis; however, because most
specimens consist of uncomplicated cirrhotic ascites, no further testing is usually needed in most
specimens.
Based on cost analysis, this author has developed a list of routine, optional, unusual, and unhelpful tests
(Table 78-2) . Each of these tests is discussed in detail in order of decreasing importance later. The cell
count is the single most helpful test for ascitic fluid. Only about 10 muL are required for a standard
manual hemocytometer count; therefore, if only one drop of fluid can be obtained, it should be sent for a
cell count. With patience during the performance of the tap, more fluid can usually be obtained. The fluid
should be submitted in an anticoagulant tube (i.e., containing ethylenediaminetetraacetic acid) to prevent
clotting. Because the decision to begin empirical antibiotic treatment of suspected ascitic fluid infection
is based mainly on the absolute neutrophil count (which should have a turn-around time of a few
minutes) rather than the culture (which takes 12 to 48 hours to demonstrate growth), the cell count is
more important than the culture in the early approach to these patients with regard to the detection and
treatment of ascitic fluid infection.
TABLE 78-2 -- Ascitic Fluid Laboratory Data
ROUTINE OPTIONAL UNUSUAL UNHELPFUL
Cell count Total protein Tuberculosis smear and pH
culture
Albumin Glucose Cytology Lactate
Culture in blood culture Lactate dehydrogenase Triglyceride Cholesterol
bottles

Amylase Bilirubin Fibronectin

Gram stain alpha1 -Antitrypsin
Glycosaminoglycans
1316
Cell Count.
Surprisingly, ascitic fluid cell counts have not been standardized. Some laboratories count mesothelial
cells in addition to WBCs and label the sum "nucleated cells." The usefulness of mesothelial cell counts
is not clear. The WBC count in uncomplicated cirrhotic ascites is usually less than 500 cells/mm3 (0.5
109 /L) [25] (see Fig. 78-2) . However, during diuresis in patients with cirrhosis and ascites, the cells can
concentrate more than 1000 cells/mm3 (1.0 109 /L). [26] Before a patient can be diagnosed as having a
diuresis-related elevation of ascitic fluid WBC count, a prediuresis count must be available and the result
must be normal. There must be a predominance of lymphocytes, and there must be no unexplained
clinical signs or symptoms (e.g., fever or abdominal pain).
The upper limit absolute polymorphonuclear leukocyte (PMN) count in uncomplicated cirrhotic ascitic
fluid is usually 250/mm3 (0.25 109 /L). [25] The short survival time of PMNs results in the relative
stability in the absolute PMN count during diuresis. [26] Therefore, the 250 cells/mm3 (0.25 109 /L)
"cut-off" remains reliable, even at the end of diuresis.
Any inflammatory process can result in an elevated ascitic fluid WBC count. SBP is the most common
cause of inflammation of ascitic fluid and the most common cause of an elevated WBC count. The total
WBC count and the absolute PMN count are elevated with SBP; PMNs usually constitute greater than
70% of the total WBC count. Also, in tuberculous peritonitis and peritoneal carcinomatosis there is
frequently an elevated total WBC count but usually with a predominance of lymphocytes. [8]
Most bloody ascites is the result of a slightly traumatic tap. Leakage of blood into the peritoneal cavity
leads to an elevated ascitic fluid WBC count. Because neutrophils predominate in blood, the ascitic fluid
differential may be altered by contamination of ascitic fluid with blood. To correct for this, one PMN is
subtracted from the absolute ascitic fluid PMN count for every 250 red blood cells [26] (see Fig. 78-2) . If
the leakage of blood is remote in time, the PMNs will have lysed and the corrected PMN count will be a
negative number. If the corrected PMN count in a bloody specimen is equal to or more than 250
cells/mm3 (0.25 109 /L), the patient must be assumed to be infected.
Exudate/Transudate.
Before the 1980s, the ascitic fluid total protein concentration was used to classify ascites into exudates (
2.5 g/dL [25 g/L]) and transudates (<2.5 g/dL [25 g/L]). Unfortunately, this form of classification
does not work well in ascitic fluid, and these terms, as applied to ascitic fluid, were never carefully
defined or validated. Attempts at using combinations of lactate dehydrogenase (LDH) and serum-ascitic
fluid ratios of LDH and protein have not been shown to accurately classify ascitic fluid into exudates and
transudates either. [27]

Serum-Ascites Albumin Gradient.
The serum-ascites albumin gradient (SAAG) has been proved in multiple studies to categorize ascites
better than the total protein concentration and better than other parameters [27] [28] [29] (Table 78-3) . The
SAAG is based on oncotic-hydrostatic balance. [28] Portal hypertension results in an abnormally high
hydrostatic pressure gradient between the portal bed and ascitic fluid. There must be a similarly large
difference between ascitic fluid and intravascular oncotic forces. [28] Albumin exerts more oncotic force
per gram than do other proteins. The difference between the serum
TABLE 78-3 -- Classification of Ascites by Serum-Ascites Concentration Gradient
HIGH GRADIENT ( 1.1 g/dL [ 11 g/L] ) LOW GRADIENT (< 1.1 g/dL [ 11 g/L] )
Cirrhosis Peritoneal carcinomatosis
Alcoholic hepatitis Tuberculous peritonitis
Cardiac ascites Pancreatic ascites
"Mixed" ascites Bowel obstruction or infarction
Massive liver metastases Biliary ascites
Fulminant hepatic failure Nephrotic syndrome
Budd-Chiari syndrome Postoperative lymphatic leak
Portal vein thrombosis Serositis in connective tissue diseases
Myxedema
Fatty liver of pregnancy
and ascitic fluid albumin concentration correlates directly with portal pressure. [28]
Calculation of the SAAG involves measuring the albumin concentration of serum and ascitic fluid
specimens and simply subtracting the ascitic fluid value from the serum value. Unless there is laboratory
error, the serum has always the larger value. It is a subtraction not a ratio. If the SAAG is more than 1.1
g/dL (11 g/L), the patient has portal hypertension, with approximately 97% accuracy.[29] Also, if the
serum albumin-ascitic fluid total protein is equal to or greater than 1.1 g/dL (11 g/L), the patient has
portal hypertension, because ascitic fluid albumin cannot be greater than ascitic fluid total protein.
Conversely, if the SAAG is less than 1.1 g/dL (11 g/L), the patient does not have portal hypertension,
with approximately 97% accuracy. The albumin gradient does not explain the pathogenesis of ascites
formation, nor does it explain where the albumin came from (i.e., liver or bowel). It simply gives the
physician an accurate indirect index of portal pressure. This test is approximately 97% accurate despite
ascitic fluid infection, diuresis, therapeutic paracentesis, albumin infusion, and etiology of liver disease.
[29]
Measurement of ascitic fluid albumin has been routine in some laboratories for almost 15 years.

However, prior to sending ascitic fluid for the first time to a specific laboratory for measurement of
albumin concentration, the interested physician should discuss the matter with the laboratory chemist.
Accuracy of the albumin assay at low albumin concentrations (e.g., <1 g/dL [10 g/L]) is important,
because many patients with ascites have a serum albumin level in the 2 g/dL (20 g/L) range and an
ascitic fluid albumin level in the 0 to 1 g/dL (0 to 10 g/L) range. If the albumin assay is not accurate at a
low range, errors will occur. Also, if a cirrhotic patient has a serum albumin level less than 1.1 g/dL (11
g/L), which occurs in fewer than 1% of patients with cirrhotic ascites, the gradient will be falsely low.
There are other situations in which accuracy decreases, such as when specimens are not obtained
simultaneously. The specimens should be obtained on the same day--preferably within the same hour.
Both serum and ascitic fluid albumin concentrations change over time; however, these values change in
parallel such that the difference is stable. Peripheral hypotension may result in a decrease in the portal
pressure and a narrowing of the albumin gradient. Lipid interferes with the albumin assay, such that
chylous ascites may have a falsely high albumin gradient.
1317
Peripheral hyperglobulinemia (>5 g/dL [50 g/L]) leads to a high ascitic fluid globulin concentration and
can narrow the albumin gradient by contributing to the oncotic forces. A narrowed gradient due to high
globulin occurs in only approximately 1% of ascitic fluid specimens. To correct the SAAG in the setting
of a high serum globulin, the uncorrected SAAG is multiplied by the following: (0.16) (serum globulin
[in g/dL] + 2.5). [30]
Another problem with the exudate/transudate system of classification is that no provision is made for
patients with two causes for ascites formation (i.e., "mixed" ascites). Most of these patients have portal
hypertension due to cirrhosis plus another cause of ascites formation (e.g., tuberculosis or peritoneal
carcinomatosis). [29] Approximately 5% of patients with ascites have mixed ascites (see Table 78-1)
(Table Not Available) . The albumin gradient is high ( 1.1 g/dL [11 g/L]) in mixed ascites, as a
reflection of the underlying portal hypertension. [29]
The presence of a high albumin gradient does not diagnose cirrhosis; it simply indicates the presence of
portal hypertension. There are many causes of portal hypertension other than cirrhosis (see Tables 78-1
(Table Not Available) and 78-3) . A low albumin gradient does not diagnose peritoneal carcinomatosis.
Although peritoneal carcinomatosis is the most common cause of a low albumin gradient, there are other
causes (see Table 78-3) . The albumin gradient need only be performed on the first paracentesis in a
given patient; it need not be repeated in subsequent specimens if the first value is definitive. If the first
value is borderline (e.g., 1 or 1.1 g/dL [10 or 11 g/L]), a repeat paracentesis and analysis are usually
definitive.
High albumin gradientand low albumin gradient should replace the terms transudative and exudative in
the classification of ascites. [27] [28] [29] [30]
Culture.
The technique of ascitic fluid culture has undergone a dramatic change based on recently published data.
[19] [31] [32] The older method of culture assumed that most episodes of ascitic fluid infection were
polymicrobial with high colony counts, such as in surgical peritonitis. However, the most common

bacterial infection of ascitic fluid, SBP, is monomicrobial with a very low bacterial concentration
(median colony count of 1 organism/mL). [31] The conventional method of culture was designed to detect
bacteria in the setting of high colony count polymicrobial infections and consists of inoculation (in the
microbiology laboratory) of each of three agar plates and some broth with a few drops of fluid. This
method of culturing ascitic fluid as if it were urine or stool is predictably insensitive in detecting low
colony count monomicrobial infections. SBP is more like bacteremia in terms of numbers of bacteria
present. It is predictable that culturing ascitic fluid as if it were blood would result in a high yield.
In fact, the sensitivity of a culture in detecting bacterial growth in neutrocytic ascites (i.e., ascitic fluid
with a PMN count 250 cells/mm3 [0.25 109 /L]) varies dramatically depending on the method of
culture used. In the published studies, "conventional" cultures have been found to detect bacterial growth
in approximately 50% of neutrocytic samples, whereas bedside inoculation of blood culture bottles with
ascitic fluid detects growth in approximately 80%. [19] Six prospective studies have demonstrated the
superiority of the blood culture bottle method. [32] Also, bedside inoculation has been shown to be
superior to delayed laboratory inoculation of blood culture bottles with ascitic fluid. [32] Gene probes are
now commercially available to detect bacteremia; hopefully, they will also lead to rapid (30-minute),
accurate detection of organisms in ascitic fluid. [33] However, a culture is required for an assessment of
susceptibility.
Total Protein.
The antiquated exudate/transudate system of ascitic fluid classification, which is based on ascitic fluid
total protein concentration, is problematic. The protein concentration in cirrhotic ascites is almost
entirely determined by the serum protein concentration and the portal pressure. [28] A cirrhotic patient
with a relatively high serum protein concentration will have a relatively high ascitic fluid protein
concentration. Because of this relationship, almost 20% of uncomplicated samples of cirrhotic ascites
fluid have greater than 2.5 g/dL (25 g/L) of protein. Ascitic fluid total protein concentration does not
increase during SBP; it remains stable before, during, and after infection. [34] In fact, patients with the
lowest protein ascites are the most predisposed to develop spontaneous peritonitis. [35] During a 10-kg
diuresis, ascitic fluid total protein doubles, such that 67% of patients with cirrhotic ascites develop a
protein level greater than 2.5 g/dL (25 g/L) at the end of diuresis. [26] Almost one third of patients with
malignant ascites have massive liver metastases or hepatocellular carcinoma as the cause of ascites
formation; their ascitic fluid is low in protein. [8] Cardiac ascites has an ascitic fluid protein concentration
greater than 2.5 g/dL (25 g/L). [36]
Therefore, the exudate/transudate method of classification of ascites places many cirrhotic patients and
all cardiac patients with ascites in the exudate category. Many patients with malignant ascites and
essentially all patients with spontaneously infected ascites are classified in the transudate category. This
method of classification is not very useful. In contrast, the albumin gradient classifies fluid by the
presence or absence of portal hypertension; it is much more physiologic and intuitive. [28] [29] The albumin
gradient classifies cardiac ascites in the high albumin gradient category, similar to cirrhotic ascites. The
high SAAG of cardiac ascites is presumably caused by high right-sided heart pressures.
Measurement of the combination of ascitic fluid total protein, glucose, and LDH has been found to be
valuable in distinguishing SBP from gut perforation into ascites [37] (Fig. 78-3) . Patients who have
neutrocytic ascitic fluid and in whom there is a clinical suspicion of bacterial peritonitis (rather than
peritoneal carcinomatosis or tuberculous peritonitis) and who meet two of the following three criteria are

likely to have surgical peritonitis. Surgical peritonitis warrants an immediate radiologic evaluation to
determine if gut perforation into ascites has occurred with total protein more than 1 g/dL (10 g/L),
glucose less than 50 mg/dL (2.8 mmol/L), and LDH higher than the upper limit of normal for serum. [37]
Glucose.
The glucose molecule is small enough to diffuse readily into body fluid cavities. Therefore, the ascitic
fluid glucose concentration is similar to that of serum unless glucose is being consumed by ascitic fluid
WBCs or bacteria. [37] The ascitic fluid glucose in early detected SBP is similar to that of sterile fluid. [37]
However, in late detected SBP and in the setting of gut perforation into ascitic fluid, the glucose level
usually drops to 0 mg/dL (0 mmol/L) because of large numbers of stimulated neutrophils and bacteria.
[37]
Lactate Dehydrogenase.
LDH enters ascitic fluid by diffusion from blood and by release from disintegrating ascitic fluid WBCs.
The LDH molecule is too large to readily enter ascitic
1318
Figure 78-3 Algorithm for differentiating spontaneous from secondary bacterial peritonitis in patients
with neutrocytic ascites (i.e., neutrophil count 250 cells/mm3 [0.25 109 /L]) in the absence of
hemorrhage into ascites, tuberculosis, peritoneal carcinomatosis, or pancreatitis. U/S, ultrasound;
LDH, lactate dehydrogenase. (From Akriviadis, E. A., and Runyon, B. A. The value of an algorithm in
differentiating spontaneous from secondary bacterial peritonitis. Gastroenterology 98:127, 1990. Copyright 1990 by the
American Gastroenterological Association.)
fluid from blood. [37] The ascitic fluid concentration of LDH is usually less than half of the serum level in
uncomplicated cirrhotic ascites. In SBP, the ascitic fluid LDH level rises because of neutrophil release of
LDH, such that the ascitic fluid concentration is greater than that of serum; in secondary peritonitis, the
LDH level rises even higher than in SBP and may be several times higher than in serum. [37]
Amylase.
The ascitic fluid amylase concentration in uncomplicated cirrhotic ascites is usually half that of the
serum value, approximately 50 U/L. [38] In patients with pancreatitis or gut perforation (with release of
luminal amylase into the fluid), the ascitic fluid amylase concentrations are markedly elevated, usually
greater than 2000 U/L and approximately five-fold greater than simultaneous serum values. [37] [38]
Gram Stain.
Gram stains of body fluids demonstrate bacteria only when there are more than 10,000 bacteria/mL. The
median concentration of bacteria in SBP is only 1 organism/mL, similar to the colony count in
bacteremia. [31] Requesting an ascitic fluid Gram stain for detection of SBP is analogous to requesting a
Gram stain of blood for the detection of bacteremia. Bacteria are present only when an overwhelming
infection is present, such as in advanced SBP or asplenic pneumococcal sepsis. The Gram stain of ascitic
fluid is most helpful in the diagnosis of free perforation of the gut into ascites. In this setting sheets of
many different bacteria are found. The Gram stain of the centrifuged sediment of 50 mL of ascites is only

10% sensitive in visualizing bacteria in SBP. [31]

Smear and Culture for Tuberculosis.
The direct smear of ascitic fluid for detection of mycobacteria is almost never positive, because of the
rarity of tuberculous peritonitis and because of the low concentration of mycobacteria in ascitic fluid in
tuberculous peritonitis. [39] [40] The older literature on tuberculous peritonitis suggests a culture of 1 L of
fluid. However, the largest centrifuge tube found in most laboratories has a capacity of 50 mL. In
general, only one 50-mL aliquot of fluid is centrifuged, and the pellet is cultured. In contrast with
approximately 50% sensitivity of ascitic fluid mycobacterial culture with optimal processing,
laparoscopy with histology and culture of peritoneal biopsies is approximately 100% sensitive in
detecting tuberculous peritonitis. [40] Tuberculous peritonitis can be easily confused with SBP because of
pain and fever, and because one half of patients with tuberculous peritonitis have cirrhosis. However, the
negative result of the bacterial culture and the predominance of mononuclear cells in the differential
should provide clues with regard to the true nature of the peritonitis. DNA probes are now available for
detection of mycobacteria; this technology will probably replace older methods of detection. [41]
However, cultures will be required to determine susceptibility.
Cytology.
The older literature of ascitic fluid cytology is confusing. In the past it was assumed that malignant
ascites was caused only by peritoneal carcinomatosis. Massive liver metastases and hepatocellular
carcinoma superimposed on cirrhosis were not recognized as causes of "malignant ascites." Also, the
results of cytology were not compared with a "gold standard" diagnosis such as autopsy, laparotomy, or
laparoscopy in the older studies. Cytology was reported to be only about 60% sensitive in detecting
"malignant ascites." [42] However, cytology should only be expected to detect a malignancy when tumor
cells line the peritoneal cavity and exfoliate into the ascitic fluid (i.e., peritoneal carcinomatosis).
Cytology should not be expected to detect a tumor when the peritoneum is not involved (e.g., hepatoma
or massive liver metastases causing portal hypertension and resultant formation of ascites or a malignant
lymphoma causing ascites by lymph node obstruction). [8] A study that did involve a gold standard
diagnosis regarding the location and type of tumor causing ascites formation has shown that only about
two thirds of patients with malignancy-related ascites have peritoneal carcinomatosis. [8] Essentially
100% of patients with peritoneal carcinomatosis were reported to have positive cytologies. [8] The
remaining one third of patients with massive liver metastases, chylous ascites due to lymphoma, or
hepatocellular carcinoma have negative cytologies. [8] Therefore, the cytology is approximately 100%
sensitive in detecting peritoneal carcinomatosis but does not detect all ascites due to cancer, because
about one third of malignancy-related ascites is caused by conditions other than peritoneal
carcinomatosis. Cytologies that are carefully
1319
performed should not have false-positive results; the author has never encountered an example. Because
hepatoma rarely metastasizes to the peritoneum, a positive cytology in this setting is unusual enough to
be the subject of a case report. [43] Measurement of the serum alpha-fetoprotein concentration (which is
always higher in serum than in ascitic fluid) may be valuable in detecting hepatocellular carcinoma;
alpha-fetoprotein is much more sensitive than is cytology in this regard. [8] Malignancy-related ascites
may have an elevated PMN count, presumably because dying tumor cells may attract neutrophils into the

fluid. [8] [44] This may cause confusion with SBP; however, there is usually a predominance of
lymphocytes in malignancy-related ascites.
Triglyceride.
A triglyceride level should be measured on opalescent or frankly milky ascitic fluid (see Fig. 78-2) . By
definition, chylous ascites has a triglyceride concentration higher than 200 mg/dL (2.26 mmol/L) and
greater than the serum level; usually the level is higher than 1000 mg/dL (11.3 mmol/L). [45] The sterile
cirrhotic ascitic fluid specimens that are slightly cloudy without an elevated cell count (i.e., opalescent)
have an elevated triglyceride concentration--64 40 mg/dL (0.72 0.45 mmol/L) compared with 18 9
mg/dL (0.20 0.10 mmol/L) for clear cirrhotic ascites. [22]
Bilirubin.
Ascitic fluid that is dark brown should be tested for bilirubin concentration. An ascitic fluid bilirubin
level greater than 6 mg/dL (102 mumol/L) and greater than the serum level of bilirubin suggests biliary
or upper gut perforation into ascites. [24] [37]
Tests That Are Seldom Helpful.
Tests that had been proposed to be helpful in the analysis of ascitic fluid but were subsequently shown to
be unhelpful include pH, lactate, fibronectin, and cholesterol. The studies that attempted to validate the
pH and lactate included small numbers of patients and used a suboptimal culture technique. In the two
largest and most recent studies, which did not have some of the problems of the earlier studies, the ascitic
fluid pH and lactate were not found to be helpful. [46] [47] The pH was found to have no impact on
decision-making regarding the use of empirical antibiotics. [46]
A number of ascitic fluid tests, including fibronectin and cholesterol, had been proposed as useful tests in
detecting "malignant ascites." The basic premise of these studies is that the ascitic fluid cytology is
insensitive. Unfortunately, the design of these studies was problematic. Several subgroups of
malignancy-related ascites (e.g., massive liver metastases, hepatoma with cirrhosis) were not
acknowledged, and appropriate control groups (patients with ascites caused by conditions other than
cirrhosis or peritoneal carcinomatosis) were not included. Other studies have demonstrated that the
subgroup of patients with massive liver metastases do not have abnormally elevated ascitic fluid
fibronectin or cholesterol concentrations. [48] [49] Therefore, patients with malignancy-related ascites and
negative cytologies also usually have negative "humoral tests of malignancy." Furthermore, patients with
high-protein noncirrhotic ascites essentially always have false-positive levels of fibronectin and
cholesterol. [8] [48] [49]
Carcinoembryonic antigen (CEA) in ascitic fluid was proposed as a helpful test in detecting malignant
ascites. [50] However, the study that attempted to validate this test had some of the flaws of the studies
mentioned earlier. More studies, with subgrouping of patients and good sensitivity and specificity, are
required before ascitic fluid CEA can be considered validated as a useful test.
Adenosine deaminase has been proposed as a useful test in detecting peritoneal tuberculosis. However, in
the United States, where more than half of patients with tuberculous peritonitis have underlying cirrhosis,
adenosine deaminase has been found to be too insensitive to be helpful. [51]

Differential Diagnosis
Although cirrhosis is the cause of ascites formation in most patients evaluated by internists,
approximately 20% of patients have a cause other than liver disease (see Table 78-1) (Table Not
Available) . Approximately 5% of patients with ascites have two causes of ascites formation (i.e.,
"mixed" ascites). [29] Usually these patients have cirrhosis and also one other cause (e.g., peritoneal
carcinomatosis or peritoneal tuberculosis) (see Table 78-1) (Table Not Available) . Because tuberculosis
is curable, potentially fatal, and frequently occurs in the cirrhotic patient with baseline ascites, the
physician must not assume that a febrile alcoholic patient has only liver disease as the cause of ascites
formation, if the ascitic fluid analysis is atypical. For example, if the ascitic fluid lymphocyte count is
unusually high, peritoneal tuberculosis may be present. Interpretation of ascitic fluid analysis is difficult
in patients with mixed ascites but crucial for appropriate diagnosis and treatment. Additionally, liver
diseases other than cirrhosis (e.g., fulminant hepatic failure) can cause ascites (see Table 78-1) (Table
Not Available) .
An algorithm regarding the differential diagnosis of ascites is shown in Figure 78-2 . This evaluation
strategy is applicable for most patients with ascites, including many of those etiologies listed in Table
78-1 (Table Not Available) . However, not every patient (including patients with rare causes of ascites)
can be readily categorized in such an algorithm. Many patients with enigmatic ascites are found
eventually to have two or three causes for ascites formation (e.g., heart failure and diabetic nephropathy).
In this setting the sum of predisposing factors leads to sodium and water retention when each individual
factor might not be severe enough by itself to cause fluid overload.
Most patients with ascites will be found to have cirrhosis as the cause. Cirrhotic ascites, especially
low-protein cirrhotic ascites, is frequently complicated by SBP (see later). Other forms of ascites are so
rarely complicated by spontaneous peritonitis that they are reported as individual cases or small series.
[52] [53] [54]
The gut can perforate into any form of ascites, cirrhotic or otherwise. The ascitic fluid analysis in gut
perforation is dramatically different than in SBP [37] (see Fig. 78-3) . The distinction between SBP and
surgical peritonitis in the patient with cirrhosis is crucial to the survival of the patient. SBP is treated with
antibiotics alone, whereas surgical peritonitis is treated with antibiotics and emergent surgical
intervention.
Cancer causes less than 10% of ascites (see Table 78-1) (Table Not Available) . Not all
malignancy-related ascites is caused by peritoneal carcinomatosis; the characteristics of the ascitic fluid
and the treatments vary depending on the pathophysiology of ascites formation (e.g., peritoneal
carcinomatosis versus massive liver metastases) [8] (Table 78-4 ; see also section on Ascitic Fluid
Analysis).
Heart failure causes less than 5% of ascites. Cardiac ascites is characterized by a high albumin gradient, a
high protein
1320

TABLE 78-4 -- Classification of Malignancy-Related Ascites

Peritoneal carcinomatosis
Massive liver metastases
Peritoneal carcinomatosis with massive liver metastases
Hepatocellular carcinoma
Malignant lymph node obstruction
Malignant Budd-Chiari syndrome (tumor emboli in hepatic veins)
concentration, and a normal blood hematocrit. [36] Patients with cardiac ascites usually have alcoholic
cardiomyopathy, with cardiomegaly on chest film and four-chamber enlargement on an echocardiogram.
Heart failure may mimic cirrhosis, including the presence of small nonbleeding esophageal varices and
hepatic encephalopathy. [55] Cirrhotic ascites has a high albumin gradient similar to that of cardiac ascites
but is low in protein, and patients with cirrhosis and ascites tend to have a mean blood hematocrit of
32%. [36]
In the United States, tuberculous peritonitis is generally a disease of Asian and Latin American
immigrants to the West coast and of poor blacks on the East coast and in the Midwest. It was a rare
disease between 1955 and 1985 but is increasing in prevalence because of acquired immunodeficiency
syndrome (AIDS). Half of the patients with tuberculous peritonitis have underlying cirrhosis (i.e.,
"mixed" ascites). Although patients with liver disease do not appear to be unusually predisposed to the
hepatotoxicity of antituberculous drugs, they tolerate drug toxicity less well than do patients with normal
livers. Therefore, the diagnosis of mixed tuberculous ascites and cirrhotic ascites is very important.
Underdiagnosis and lack of antituberculous therapy of this patient group can lead to unnecessary deaths
from uncontrolled tuberculosis, whereas overdiagnosis and overtreatment of suspected but unproven
tuberculous peritonitis in the cirrhotic patient could lead to unnecessary deaths from the hepatotoxicity of
isoniazid. If the clinical setting (e.g., a febrile immigrant from an area where tuberculosis is endemic) and
initial ascitic fluid analysis (high lymphocyte count) are suggestive of tuberculosis, the author performs
an urgent laparoscopy with histology and culture of peritoneal biopsies. If the peritoneum has the typical
"millet seed" and "violin string" appearance, antituberculous therapy is started the same day as the
laparoscopy. Blind peritoneal biopsy could be performed in the patient without cirrhosis. However, in the
cirrhotic patient, the predictable presence of peritoneal collateral veins makes a blind biopsy potentially
very hazardous. A laparoscopically guided biopsy is preferable. Suspected tuberculous peritonitis is one
of the few remaining indications for diagnostic laparoscopy. [56] Peritoneal coccidioidomycosis can
mimic tuberculous peritonitis, including its appearance at laparoscopy, and can occur in patients without
AIDS. [57] There has been a resurgence of tuberculous peritonitis coincident with the AIDS epidemic. [58]
The sensitivity of cytology in detecting peritoneal carcinomatosis and of ultrasound-guided biopsy in
detecting focal liver lesions has obviated the need for laparoscopy in detecting a tumor for all practical
purposes. [8]
Pancreatic ascites, which is an unusual condition, is detected in patients with clinically obvious severe
acute pancreatitis or in patients with a history of chronic pancreatitis. For these reasons, it is not
necessary to order an ascitic fluid amylase level on all ascitic fluid samples, only in patients in whom

pancreatitis is suspected or when the initial ascitic fluid is not diagnostic (see Table 78-2) . Patients with
alcohol-related pancreatic ascites may also have underlying alcoholic cirrhosis. Pancreatic ascites is
frequently neutrocytic and can also be complicated by bacterial infection. Patients with an ascitic fluid
neutrophil count equal to or greater than 250 cells/mm3 (0.25 109 /L) warrant empirical antibiotic
coverage, at least until the cause of the elevated neutrophil count is explained.
Nephrogenous ascites is a poorly understood form of ascites that develops in patients who receive
hemodialysis. [59] On careful evaluation most of these patients are found to have another cause for ascites
formation, usually cirrhosis from alcohol abuse or hepatitis C. The presence of a second cause for fluid
overload explains why these patients have ascites, whereas the majority of dialysis patients do not.
Chlamydia peritonitis should be suspected in sexually active young women with fever and neutrocytic,
high-protein, low-gradient ascites and no evidence of liver disease. [15] This infection rapidly responds to
oral doxycycline and is one of the few curable causes of ascites formation.
Although nephrotic syndrome used to be a common cause of ascites formation in children, it is quite rare
in adults. [60] This fluid is usually low in protein with a low albumin gradient and can be complicated by
SBP.
Some patients develop pathologic accumulations of fluid in the peritoneal cavity due to leakage from a
ruptured viscus (e.g., bile ascites in the setting of a ruptured gallbladder). [24] The ascitic fluid analysis
can be crucial to making a preoperative diagnosis of this condition (see earlier section on Ascitic Fluid
Analysis and Fig. 78-3) .
Chylous ascites develops when intra-abdominal lymphatics containing chyle are ruptured. The older
literature suggests that this form of ascites is caused by a malignancy in almost 90% of patients. [45] In
contrast, cirrhosis is the cause of chylous ascites in more than 90% of the patients whom the author has
encountered [23] [29] (see Table 78-1) (Table Not Available) . The high lymphatic flow and pressure must
be the cause of rupture in patients with cirrhosis. Retroperitoneal surgery and radical pelvic surgery in
patients with cancer can transect lymphatics and can also lead to chylous ascites.
Causes of ascites excluded from or not encountered in the series described in Table 78-1 (Table Not
Available) include ambulatory peritoneal dialysis fluid, Budd-Chiari syndrome, myxedema ascites,
ascites associated with connective tissue diseases, postoperative ascites, and rare causes. The iatrogenic
form of ascites associated with peritoneal dialysis is usually not under the care of gastroenterologists.
Although Budd-Chiari syndrome is regularly complicated by ascites, hepatic vein thrombosis itself is
rare enough that it causes less than 0.1% of ascites. Ascites in patients with myxedema appears to be of
cardiac origin, related to the heart failure that these patients have. [61] Treatment of the thyroid
insufficiency cures the fluid retention. Serositis with ascites formation may complicate systemic lupus
erythematosus. [16]
Development of ascites after abdominal surgery (in particular after inappropriate cholecystectomy in the
setting of asymptomatic gallstones and abnormal liver tests) is a common mode of presentation of
previously undiagnosed cirrhosis. [11] Resection of hepatocellular carcinoma in the setting of cirrhosis
regularly leads to decompensation; unfortunately, this
1321

frequently starts a downward spiral course and fatal outcome. [62]

Aggressive hormone administration to induce ovulation can lead to ascites from ovarian hyperstimulation
syndrome. [63] Other rare causes of ascites that have been reported include POEMS syndrome (
polyneuropathy, organomegaly, endocrinopathy, M component, and skin changes), and fluorouracil
chemotherapy for colon cancer. [64] [65]


Saunders Company
COMPLICATIONS OF ASCITES
Infection
Ascitic fluid infection can be classified into five categories (including three spontaneous categories)
based on culture, PMN count, and the presence or absence of a surgical source of infection (Table 78-5) .
An abdominal paracentesis must be performed and ascitic fluid must be analyzed before a confident
diagnosis of ascitic fluid infection can be made. A clinical diagnosis of infected ascitic fluid without a
paracentesis is not adequate.
Subtypes
Of the three subtypes of spontaneous ascitic fluid infection, the prototype is SBP. This diagnosis is made
when there is a positive ascitic fluid culture and there is an elevated ascitic fluid absolute PMN count
(i.e., 250 cells/mm3 [0.25 109 /L]) without an evident intra-abdominal surgically treatable source of
infection. [19] When Correia and Conn coined the term spontaneous bacterial peritonitis in 1975, their
goal was to distinguish this form of infection from surgical peritonitis. [66] This is a very important
distinction. Therefore, although many patients with SBP have a focus of infection (e.g., urinary tract
infection or pneumonia), they are labeled as having SBP unless the focus requires surgical intervention
(e.g., ruptured viscus). I have not encountered a convincing polymicrobial case of SBP. All of the
patients with SBP whose ascitic fluid cultures were initially growing more than one organism were
eventually found to have surgical peritonitis or an erroneous culture (e.g., a pathogen plus a contaminant
or two colony morphologies of one species of bacteria that were initially misinterpreted as two bacteria).
The criteria for a diagnosis of monomicrobial non-neutrocytic bacterascites (MNB) include a positive
ascitic fluid culture for a single organism, an ascitic fluid PMN count of fewer than 250 cells/mm3 (0.25
109 /L), and no evident intra-abdominal surgically treatable source of infection. [67] In the older
literature, MNB was either grouped with SBP or called asymptomatic bacterascites. Because many
patients
TABLE 78-5 -- Classification of Ascitic Fluid Infection
Spontaneous ascitic fluid infection
Monomicrobial non-neutrocytic bacterascites
Culture-negative neutrocytic ascites
Secondary bacterial peritonitis
Polymicrobial bacterascites (needle perforation of the bowel)

with bacterascites have symptoms, the modifier "asymptomatic" seems inappropriate.

Culture-negative neutrocytic ascites (CNNA) is diagnosed when (1) the ascitic fluid culture grows no
bacteria, (2) the ascitic fluid PMN count is equal to or greater than 250 cells/mm3 (0.25 109 /L), (3) no
antibiotics have been given (not even a single dose), and (4) there is no other explanation for an elevated
PMN count (e.g., hemorrhage into ascites, peritoneal carcinomatosis, tuberculosis, or pancreatitis). [68]
This variant of ascitic fluid infection is seldom diagnosed when sensitive culture methods are used. [31] [32]
Secondary bacterial peritonitis is diagnosed when (1) the ascitic fluid culture is positive (usually for
multiple organisms), (2) the PMN count is equal to or greater than 250 cells/mm3 (0.25 109 /L), and (3)
there is an identified intra-abdominal surgically treatable primary source of infection (e.g., perforated
gut, perinephric abscess). [37] The importance of distinguishing this variant from SBP is that secondary
peritonitis is surgical peritonitis and usually requires emergency surgical intervention.
Polymicrobial bacterascites is diagnosed when (1) multiple organisms are seen on Gram stain or cultured
from ascitic fluid, and (2) the PMN count is less than 250 cells/mm3 (0.25 109 /L). [69] This diagnosis
should be suspected when (1) the paracentesis is traumatic or unusually difficult because of ileus, or (2)
stool or air is aspirated into the paracentesis syringe. Polymicrobial bacterascites is essentially diagnostic
of gut perforation by the paracentesis needle.
Clinical Setting
The spontaneous variants of ascitic fluid infection (i.e., SBP, CNNA, and MNB) occur only in the setting
of severe liver disease. The liver disease is usually chronic (cirrhosis) but may be acute (fulminant
hepatic failure) or subacute (alcoholic hepatitis). All forms of cirrhosis have been reported to be
complicated by spontaneous ascitic fluid infection. Spontaneous infection of noncirrhotic ascites is rare
enough to be the subject of case reports. [52] [53] [54]
Essentially all patients with SBP have an elevated serum bilirubin and an abnormal prothrombin time,
usually Childs-Pugh B or C. [19] Ascites appears to be a prerequisite to the development of SBP. It is
unlikely for SBP to precede the development of ascites. This infection usually develops at the time of
maximum ascites volume.
Secondary bacterial peritonitis and polymicrobial bacterascites can develop in ascites of any type. The
only prerequisite, in addition to the presence of ascites, for development of the former infection is the
presence of an intra-abdominal surgical source of infection. [37] The latter infection occurs due to needle
entry of the bowel during attempted paracentesis. [69]
Pathogenesis
During the past decade the elusive source of SBP has become more clear and the pathogenesis of the
spontaneous forms of ascitic fluid infection have been partially elucidated (Fig. 78-4) . The body of
currently available evidence suggests that the spontaneous forms of ascitic fluid infection are the result of
an overgrowth of a specific organism in the gut and translocation
1322

Figure 78-4 Current proposed pathogenesis of spontaneous ascitic fluid

infection. CNNA, culture-negative neutrocytic ascites.
of that microbe from the gut to the mesenteric lymph nodes, with resulting spontaneous bacteremia and
subsequent colonization of susceptible ascitic fluid [70] [71] [72] (see Fig. 78-4) .
Once bacteria enter the fluid in the abdomen, by whatever route, a battle ensues between the organism's
virulence factors and the host's immune defenses. [34] [35] [67] [73] [91] The ascitic fluid protein concentration
does not change with the development of spontaneous infection. [34] Low-protein concentration ascitic
fluid (e.g., <1 g/dL [10 g/L]) has been shown to be particularly susceptible to SBP. [35] The endogenous
antimicrobial activity (opsonic activity) of human ascitic fluid correlates directly with the fluid's protein
concentration. [73] Patients with deficient ascitic fluid opsonic activity have been shown to be predisposed
to SBP. [74] Patients with detectable ascitic fluid opsonic activity appear to be protected from SBP unless
a particularly virulent organism (e.g., Salmonella) is involved. [54] [73] [74]
Recent studies in both patients and animals with cirrhosis demonstrate that MNB is common. [67] [75] In
both humans and rats, most episodes of bacterascites resolve without antibiotic treatment. [67] [75] The
fluid frequently becomes sterile without an evident PMN response. Apparently the host's defense
mechanisms are able to eradicate the invading bacteria on most occasions. It is probable that uncontrolled
infection develops only when the defenses are particularly weak or the organism is particularly virulent
(see Fig. 78-4) . Bacterascites is probably much more common than SBP. It is conceivable that patients
with cirrhotic ascites regularly have their ascitic fluid colonized by bacteria and almost just as regularly
resolve the colonization. The entry of PMNs into the fluid probably signals failure of the the peritoneal
macrophages to control the infection. [76] The majority of MNB episodes appear to resolve in cirrhotic
rats and humans, whereas SBP is frequently fatal. In summary, MNB probably represents an early stage
of ascitic fluid infection that can resolve to sterile non-neutrocytic ascites or progress to SBP.
Most episodes of CNNA are diagnosed by using insensitive culture methods where there are insufficient
numbers of bacteria to reach the threshold of detectability. [31] Blood culture bottles can detect a single
organism in the cultured aliquot of fluid, whereas the conventional method of culture probably requires at
least 100 organisms/mL. [31] However, even when optimal culture methods are used, a small percentage
of patients grow no bacteria in their neutrocytic ascitic fluid. [31] A study of rapid sequential paracenteses
(before the initiation of antibiotic treatment) in patients with CNNA demonstrated that in most cases the
PMN count dropped spontaneously and the culture remained negative in the second specimen. [77A] In the
setting of sensitive culture technique, CNNA probably represents spontaneously resolving SBP in which
the paracentesis is performed after all bacteria have been killed by host defenses, but before the PMN
count has normalized.
The pathogenesis of secondary bacterial peritonitis is no mystery compared with the pathogenesis of
SBP. When the gut perforates into the ascitic fluid, billions of bacteria flood in. When there is secondary
peritonitis without a frank perforation, bacteria cross inflamed tissue planes and enter ascites. The
pathogenesis of polymicrobial bacterascites is also apparent. [69] The paracentesis needle enters the
bowel, and bowel contents are released.

Signs and Symptoms
Although 87% of patients with SBP reported in a series were symptomatic at the time of diagnosis of
infection, the symptoms or signs of infection were often subtle, such as a slight change in mental status.
[67] Without prompt paracentesis, diagnosis and treatment of infected ascites may be delayed. Delays in
the initiation of therapy usually result in death of the patient. The signs and symptoms manifested in all
five variants of ascitic fluid infection are listed in Table 78-6 .
Prevalence
Prior to the 1980s, abdominal paracentesis was not performed regularly because of fear of complications
of the procedure and because the utility of ascitic fluid analysis in the differential diagnosis of ascites
was not fully recognized. Now, routine paracentesis is performed at the time of admission of many
patients with ascites. Paracenteses on a patient's routine admission to the hospital have provided data
regarding the prevalence of ascitic fluid infection. Overall approximately 10% of patients with ascites are
infected at the time of their admission to the hospital; of the subgroup of patients with cirrhotic ascites,
about 27% are infected. [19] [35] Of patients with culture-positive ascitic fluid, about two thirds will have
neutrocytic ascitic fluid (i.e., SBP), and one third will have monomicrobial bacterascites. [67] The
prevalence of CNNA among a series of neutrocytic ascitic fluid is largely dependent on the culture
technique (see earlier). Polymicrobial bacterascites occurs in only about one in 1000 paracenteses.
Secondary bacterial peritonitis occurs in only up to 2% of patients with ascites at the time of their
admission to the hospital. [19] [37]
Flora
Escherichia coli, streptococci (mostly pneumococci), and Klebsiella cause most episodes of SBP and
monomicrobial
1323
TABLE 78-6 -- Signs and Symptoms of Ascitic Fluid Infection *

SECONDARY POLYMICROBIAL
SBP BACTERASCITES CNNA PERITONITIS BACTERASCITES
Fever 68 57 50 33 10
Abdominal pain 49 32 72 67 10
Tender abdomen 39 32 44 50 10
Rebound 10 5 0 17 0
Mental status 54 50 61 33 0
CNNA, culture-negative neutrocytic ascites.
*Data from references [37] [67] [68] [69] . Data presented as a percentage of the total number of patients in that group.

bacterascites (Table 78-7) . CNNA is, by definition, culture-negative. Polymicrobial bacterascites is, by
definition, polymicrobial. The most apparent difference between the spontaneous forms of ascitic fluid
infection and the secondary forms (secondary peritonitis and polymicrobial bacterascites) is that the
former are always monomicrobial and the latter are usually polymicrobial. Although older papers report
approximately 6% prevalence of anaerobes among SBP flora, this is probably a reflection of the presence
of unrecognized cases of secondary bacterial peritonitis. In a recent series, anaerobes caused
approximately 1% of SBP and monomicrobial bacterascites. [31] [67]
Risk Factors
Patients with cirrhosis are unusually predisposed to bacterial infections because of multiple defects in
their immune defense. The concept that cirrhosis is a form of acquired immunodeficiency (in the generic
sense) is rather new. In a published prospective study more than 40% of consecutive cirrhotic patients
had bacterial infection at the time of their admission to the hospital or developed infection during
hospitalization. [77B]
Low ascitic fluid total protein and the phagocyte (both motile and stationary) dysfunction associated with
cirrhosis were discussed earlier as risk factors for bacterial infection.
TABLE 78-7 -- Flora of Ascitic Fluid Infection
MONOMICROBIAL SECONDARY
NON-NEUTROCYTIC BACTERIAL
ORGANISM SBP BACTERASCITES PERITONITIS
Monomicrobial
Escherichia coli 37 27 20
Klebsiella pneumoniae 17 11 7
Pneumococcus 12 9 0
Streptococcus viridans 9 2 0
Staphylococcus aureus 0 7 13
Miscellaneous gram-negative 10 14 7
Miscellaneous gram-positive 14 30 0
Polymicrobial 1 0 53
*Data reported as a percentage of the total in that group. Data from references [37] [67] , [77A] .
Paracentesis itself has been proposed as a risk factor in causing ascitic fluid infection. This theoretical
risk has not been substantiated in prospective studies of paracentesis complications. [20] SBP is
statistically more likely to be diagnosed at the time of the first paracentesis compared with subsequent

taps. [20] No cases of needle-induced ascitic fluid infection have been reported unless the bowel is entered
by the paracentesis needle. [20] [69] Fortunately, this occurs only once in about 1000 taps. Skin flora (e.g.,
Staphylococcus aureus) would be expected to be isolated if poor paracentesis technique were the cause
of many cases of SBP; yet skin flora are very seldom isolated. [31] The most likely setting in which to
expect iatrogenic peritonitis is when the paracentesis needle enters the bowel during a difficult
paracentesis.
Gastrointestinal hemorrhage is an under-recognized risk factor for development of spontaneous
bacteremia and SBP; there is approximately a 40% cumulative probability of infection during a single
hospitalization for bleeding. [78A] The risk appears to peak 48 hours after the initiation of hemorrhage.
The high risk of infection is probably mediated by a shock-induced increase in translocation of bacteria
from the gut to extra-intestinal sites. [78B] Urinary tract infections are also an under-recognized risk factor
for the development of SBP. [79]
Diagnosis
A timely diagnosis of ascitic fluid infection requires a high index of suspicion of infection and a low
threshold for performing a paracentesis. Clinical deterioration in any manner in a patient with ascites
should raise suspicion of infection and prompt a tap, especially if there is fever or abdominal pain. If the
ascitic fluid PMN count is elevated, the working diagnosis is ascitic fluid infection until proven
otherwise. Although peritoneal carcinomatosis, pancreatitis, hemorrhage into ascites, pancreatitis, and
tuberculosis can lead to an elevated ascitic fluid PMN count, most cases of neutrocytic ascites are due to
infection. A majority of PMNs in the WBC differential lends further credence to the diagnosis of
infection. In the setting of peritoneal carcinomatosis, pancreatitis, and tuberculosis, there is usually not a
predominance of neutrophils. An elevated absolute ascitic fluid PMN count with a predominance of
neutrophils in a clinical setting compatible with infection (e.g., a febrile cirrhotic patient) should prompt
empirical antibiotic therapy (Table 78-8 ; see also later).
Although SBP is approximately six times as common as surgical peritonitis in the patient with ascites,
secondary peritonitis
1324
TABLE 78-8 -- Indications for Empirical Antibiotic Therapy of Suspected Spontaneous Ascitic Fluid
Infection
Ascitic fluid neutrophil count 250/mm3 (0.25 109 /L)

Convincing signs and symptoms of infection
should be considered in any patient with neutrocytic ascites. Clinical signs and symptoms do not separate
patients with secondary peritonitis from those with SBP [37] (see Fig. 78-3) . Even with free perforation of
the colon into ascitic fluid, patients do not develop a classic surgical abdomen. Peritoneal signs require
contact of inflamed visceral and parietal peritoneal surfaces. This does not happen when there is a large
volume of fluid separating these surfaces. Gut perforation can be suspected and pursued if a specimen is

neutrocytic and meets two of the following three criteria (see Fig. 78-3) : total protein greater than 1 g/dL
(10 g/L), glucose less than 50 mg/dL (2.8 mmol/L), and LDH higher than the upper limit of normal for
serum. [37] Also, essentially all ascitic fluids culture multiple organisms in the setting of a perforated
viscus, except for gallbladder rupture, which is usually monomicrobial. [23] Brown ascitic fluid with a
bilirubin concentration higher than 6 mg/dL (102 mumol/L) and greater than the serum level is indicative
of biliary or upper gut perforation into ascites. [23] An ascitic fluid amylase that is greater than five-fold
that of serum may also be indicative of gut rupture (except gallbladder rupture) and release of luminal
amylase. [23] [37]
The initial ascitic fluid analysis is very helpful in delineating which patients are likely to have a ruptured
viscus (see Fig. 78-3) . These patients need a radiologic evaluation to confirm and localize the site of
rupture, within minutes of detection of neutrocytic ascitic fluid. This should include plain and upright
abdominal films and water-soluble contrast studies of the upper and lower gut. If perforation is
documented, emergency surgical intervention is the next step. Timing is crucial; after septic shock
occurs, death is almost certain. Antibiotic therapy without surgical intervention in the treatment of a
ruptured viscus is predictably unsuccessful.
In contrast with patients with perforation peritonitis, patients with nonperforation secondary peritonitis
tend not to have a diagnostic initial ascitic fluid analysis. [37] It is less urgent to make the diagnosis of
secondary peritonitis in patients without free perforation. Therefore, there may be time to evaluate the
response of the PMN count and ascitic fluid culture to treatment. The best time to perform a single repeat
paracentesis to assess the response is after 48 hr of treatment; by 48 hr essentially every patient with SBP
who has been treated with an appropriate antibiotic will have a PMN count lower than the pretreatment
value and the culture will be negative. [37] Before 48 hr of treatment, the PMN count may rise higher than
baseline in SBP and in secondary peritonitis. [37] The culture remains positive in secondary peritonitis and
becomes rapidly negative in SBP [37] (see Fig. 78-3) . Antibiotics alone cannot control secondary
peritonitis, but medical therapy rapidly cures SBP. [37]
Treatment
Patients with ascitic fluid PMN counts equal to or higher than 250 cells/mm3 (0.25 109 /L) in a clinical
setting compatible with ascitic fluid infection should receive empirical antibiotics [19] (Table 78-9 ; see
also Table 78-8) . Patients with hemorrhage into ascites, peritoneal carcinomatosis, pancreatic ascites, or
tuberculous peritonitis may have an elevated PMN count that is not related to SBP. These patients
usually do not require empirical treatment; however, if they do receive treatment, their PMN count
usually fluctuates randomly, compared with the dramatic reduction in PMN count that is typical of
patients with SBP. If the situation is initially unclear, the physician should err on the side of
overtreatment (with a non-nephrotoxic antibiotic). Usually patients with uninfected neutrocytic ascitic
fluid (except those with hemorrhage) have a predominance of lymphocytes in their ascitic fluid
differential; this helps to distinguish them from those with SBP, in which PMNs predominate. Patients
with bloody ascites should have a "corrected" PMN count calculated (as discussed earlier). Antibiotic
therapy of patients with bloody ascites is not necessary unless their corrected PMN count is equal to or
greater than 250 cells/mm3 (0.25 109 /L).
Empirical antibiotic treatment of patients with bacterascites must be individualized. Many episodes
resolve without treatment. [67] However, the 22% to 43% hospitalization mortality of monomicrobial

bacterascites is at least partially caused by infection. [67] [80] Therefore, treatment appears to be warranted
in many patients. By definition, the PMN count is less than 250 cells/mm3 (0.25 109 /L) in this variant
of ascitic fluid infection. Because of this, the PMN count cannot be the only parameter on which the
decision about empirical therapy is based. Most patients with MNB who do not resolve the colonization
and progress to SBP have signs or symptoms of
TABLE 78-9 -- Treatment of Subtypes of Ascitic Fluid Infection
DIAGNOSIS TREATMENT
Spontaneous bacterial peritonitis 5 days of intravenous antibiotic to which the
organism is highly susceptible (e.g., cefotaxime, 2
g every 8 hr empirically, followed by more
narrow-spectrum therapy after susceptibility
results are available)
Monomicrobial non-neutrocytic bacterascites 5 days of intravenous antibiotic therapy to which
the organism is highly susceptible, if the patient is
symptomatic or persistently culture-positive
Not all patients with bacterascites require
treatment
Culture-negative neutrocytic ascites 5 days of intravenous third-generation
cephalosporin (e.g., cefotaxime, 2 g every 8 hr)
Secondary bacterial peritonitis Surgical intervention, plus approximately two
weeks of intravenous cephalosporin therapy (e.g.,
cefotaxime, 2 g every 8 hr) plus an antianaerobic
drug such as metronidazole
Polymicrobial bacterascites Intravenous third-generation cephalosporin (e.g.,
cefotaxime 2 g every 8 hr) plus an antianaerobic
drug such as metronidazole. The duration is
determined by the clinical response and by serial
ascitic fluid polymorphonuclear leukocyte counts
and cultures
1325
infection at the time of the tap that document bacterascites. [67] Therefore, patients with cirrhotic ascites
who have convincing signs or symptoms of infection should receive treatment, regardless of the PMN
count in ascitic fluid. Empirical treatment can be discontinued after only two to three days if the culture
demonstrates no growth. Asymptomatic patients may not need treatment at all. [67] [80] Patients without
clinical evidence of infection should undergo repeat paracentesis for cell count and culture, once it is
known that the initial culture is positive. If the PMN count has risen above 250/mm3 (0.25 109 /L) in
the follow-up ascitic fluid analysis or if signs or symptoms of infection have developed, treatment should
be started. Patients who have developed no PMN response and no clinical evidence of infection usually

have no growth in their second specimen and do not require treatment. [67] These are the patients who
have eradicated the colonization by their own immune defenses.
The physician does not initially know that the result of the culture is destined to be negative in the patient
with CNNA; therefore, empirical treatment should be given. When the preliminary report demonstrates
no growth, it is helpful to repeat the paracentesis after 48 hr of therapy to assess the response of the PMN
count to antibiotics. A dramatic decline in PMN count (always below the baseline pretreatment value and
frequently more than 80% reduction) confirms a response to treatment and probably warrants a few more
days of therapy. [81] A stable PMN count, especially if there is a predominance of
lymphocytes/monocytes, indicates that a nonbacterial (or mycobacterial) cause of the neutrocytosis is
present. It is appropriate to send fluid for cytology and culture for tuberculosis. Because insensitive
culture techniques result in negative cultures, one of the most important methods to reduce the
prevalence of CNNA in a hospital that is still using the conventional method of culture is to convince the
microbiology laboratory to accept and process ascitic fluid submitted in blood culture bottles. [31]
The ascitic fluid Gram stain is most helpful in detecting secondary peritonitis, in which multiple different
bacterial forms are seen. The Gram stain is not of much value in the choice of empirical antibiotic for
treatment of spontaneous infection. This author has found that the Gram stain did not allow a narrowed
spectrum of antibiotic coverage in even one patient out of approximately 500 with SBP. Only
approximately 10% of Gram stains demonstrate organisms in SBP. [31] A Gram stain indicative of
secondary peritonitis requires coverage of anaerobic flora in addition to coverage of aerobic and
facultative anaerobic flora, as well as an emergency search for a source of the bacteria [37] (see Fig. 78-3
and Table 78-9) . Therefore, a positive Gram stain may lead to broader spectrum coverage rather than to
narrower spectrum coverage. Choosing very narrow spectrum coverage (e.g., penicillin) based on a
misinterpreted Gram stain may lead to the patient's death from uncontrolled infection before it is known
that the isolated organism is resistant to the chosen antibiotic.
Relatively broad-spectrum therapy is warranted in patients with suspected ascitic fluid infection until the
results of susceptibility testing are available. After sensitivities are known, the spectrum of coverage can
usually be narrowed. The antibiotics that have been recommended for empirical treatment have changed
during the past several years. In 1978 the combination of ampicillin and gentamicin was promoted. This
recommendation was based on no susceptibility testing or efficacy data. In recent years it has become
apparent that gentamicin has an unpredictable volume of distribution in patients with ascites and that the
serum creatinine (and even the creatinine clearance) is a poor index of the glomerular filtration rate in
patients with ascites. [82] [83] Therefore, it is difficult to give appropriate loading and maintenance doses of
gentamicin to this patient population. To use aminoglycosides without reaching toxic serum levels
requires that frequent "stat" ascitic fluid and serum levels of the drug be obtained. There are no
data-supported guidelines to follow here. In this author's experience, even if overdosing is avoided, most
patients treated with aminoglycosides develop nephrotoxicity. [84] [85] There is no evidence that newer
aminoglycosides are less problematic than is gentamicin. Dialysis is not usually used to treat hepatorenal
syndrome. However, when renal failure is thought to be iatrogenic, dialysis is frequently initiated.
Cirrhotic ascites patients, with their baseline coagulopathy and hypotension, are problematic dialysis
patients.
A number of nonaminoglycoside antibiotic options are now available. Aztreonam is a monobactam that
has been used in SBP. Unfortunately it has little gram-positive coverage, and an unacceptable 19%
superinfection rate has been documented with this drug. [86] If it is used as empirical therapy, a drug that

covers gram-positive bacteria must also be used. The first- and second-generation cephalosporins cover
approximately 80% of the flora of SBP. [87] Organisms that are resistant to the empirical antibiotic may
cause the patient's death before susceptibility testing results are available. Cefotaxime, a third-generation
cephalosporin, has been shown to be superior to ampicillin plus tobramycin in a controlled trial. [88] This
drug covered 98% of the flora, was more efficacious, and did not result in superinfection or
nephrotoxicity. [88] Cefotaxime or a similar third-generation cephalosporin appears to be the treatment of
choice for suspected SBP. [19] Anaerobic coverage is not needed, nor is coverage for Pseudomonas or
Staphylococcus.[31] Dosing of cefotaxime, 2 g intravenously every 8 hr, has been shown to result in
excellent ascitic fluid levels (20-fold killing power after one dose). [89] In patients with serum creatinine
levels greater than 3 mg/dL, the dosing interval could be extended to 12 hr. [89] Neither a loading dose nor
an intraperitoneal dose appears to be necessary or appropriate. Oral ofloxacin has been reported in a
controlled trial to be as effective as parenteral cefotaxime in treatment of SBP in patients who are not
vomiting and who are not in shock. [90] Confirmatory trials are awaited before oral treatment of this
life-threatening infection can be enthusiastically recommended.
After susceptibility testing results are available, more narrow-spectrum treatment can usually be
substituted for the broad-spectrum drug (e.g., pneumococci are usually sensitive to penicillin, and most
E. coli species are usually sensitive to ampicillin).
Most infectious disease subspecialists recommend 10 to 14 days of therapy for life-threatening
infections. However, there are no data to support this duration of treatment for spontaneous ascitic fluid
infection. The ascitic fluid culture becomes sterile after one dose of cefotaxime in 86% of patients. [37]
After 48 hr of therapy, the ascitic fluid PMN count is always less than the pretreatment value in patients
with spontaneous ascitic fluid infection treated with appropriate antibiotics; frequently, there is an 80%
reduction at 48 hr. [37] A randomized controlled trial involving 100 patients has demonstrated that
1326
five days of treatment is as efficacious as ten days in the treatment of SBP and CNNA. [91]
Because of the predictable presence of anaerobes, patients suspected of having secondary peritonitis
require broader spectrum empirical antibiotic coverage than do those with SBP, in addition to an
emergency evaluation to assess the need for surgical intervention (see earlier and Table 78-8 and Fig.
78-3) . Cefotaxime plus metronidazole appears to provide excellent initial empirical therapy for
suspected secondary peritonitis.
Polymicrobial bacterascites (needle perforation of the bowel) is relatively well tolerated. Only one in ten
patients with needle perforation of the gut into ascitic fluid developed peritonitis in the study that
reported this subset of ascitic fluid infection. [69] The single episode of paracentesis-related peritonitis
was not fatal. It appears that patients with low-protein ascitic fluid are at most risk for developing a PMN
response and clinical peritonitis related to needle perforation of the gut. [69] Most of the patients with
higher protein ascites (e.g., >1 g/dL [11g/L]) did not even receive antibiotics and yet did well. However,
many physicians would probably feel uncomfortable withholding antibiotic treatment if needle
perforation is suspected. If a decision is made to treat the patient, anaerobic coverage should be included
(e.g., cefotaxime and metronidazole) (see Table 78-9) . Whether treatment is given or not, a follow-up
paracentesis is helpful (if it can be performed safely) in following the PMN count and culture. If a
decision is made not to treat and the number of organisms does not decrease or a PMN response occurs

in the second specimen, antibiotic treatment should be initiated (see Table 78-9) .
Prognosis
In the past, 48% to 95% of patients with spontaneous ascitic fluid infection died during the
hospitalization in which the diagnosis was made, despite antibiotic treatment. [19] [91] The most recent
series report the lowest mortality. This is probably a reflection of earlier detection and treatment of
infection in the 1990s as well as avoidance of nephrotoxic antibiotics. In the older series, about half of
patients with SBP died of the infection despite antibiotic treatment; now fewer than 5% of patients die of
infection if timely and appropriate antibiotics are used. [91] However, even now, many patients are cured
of their infection and yet die of liver or renal failure or gastrointestinal bleeding, because of the severity
of the underlying liver disease. Because spontaneous ascitic fluid infection is a good marker of end-stage
liver disease, it has been proposed as an indication for liver transplantation in a patient who has no
contraindications.
In order to maximize survival, it is important to perform paracentesis on all patients with ascites at
hospital admission so that infection can be diagnosed and treated early. In addition, paracentesis should
be repeated during hospitalization if any deterioration occurs--including pain, fever, mental status
change, renal failure, acidosis, peripheral leukocytosis, or gastrointestinal bleeding. In the past a delay in
diagnosis was, at least in part, responsible for the excessive mortality. If the physician waits until the
patient develops convincing signs and symptoms of infection before performing a paracentesis, the
infection is likely to be very advanced by the time that the diagnosis is made. There have been no
reported survivors of SBP when the diagnosis was made after the creatinine had risen above 4 mg/dL
(350 mumol/L) or after the patient had developed shock.
The mortality for secondary peritonitis in the patient hospitalized with ascites, when the condition is
diagnosed early and treated with emergency laparotomy, is in the same range as that of SBP,
approximately 50%. [37] Without surgical intervention, mortality is approximately 100%.
Prevention
The identification of risk factors for development of SBP (including ascitic fluid protein concentration <1
g/dL, variceal hemorrhage, and prior episode of SBP) has led to controlled trials of prophylactic
antibiotics. [35] [92] [93] [94] Norfloxacin, 400 mg/day orally, has been reported to successfully prevent SBP
in inpatients with low-protein ascites and in patients with prior SBP. [92] [93] Norfloxacin, 400 mg orally
twice per day for 7 days, helps to prevent infection in patients with variceal hemorrhage. [94] However,
oral antibiotics do not prolong survival and do select resistant gut flora, which can subsequently cause
spontaneous infection. [92] [93] [94] [95] Restricted use of these agents to inpatients only, with discontinuation
of the drug at the time of discharge, may be the best compromise in preventing infection without
selecting resistant flora, according to a randomized trial. [96]
Parenteral antibiotics to prevent sclerotherapy-related infections do not appear to be warranted, based on
a controlled trial. [97] It is the active bleeding that appears to be the risk factor for infection, not
sclerotherapy.
Tense Ascites

Some patients with ascites do not seek medical attention until they can no longer breathe or eat
comfortably because of the pressure that the intra-abdominal fluid exerts on their diaphragms. Tense
ascites requires urgent therapeutic paracentesis. Contrary to old folklore, tense ascites can be drained
without untoward hemodynamic effects. [98] [99] [100] Total paracentesis, even more than 20 liters, has been
demonstrated to be safe. [100] In the setting of tense ascites, therapeutic paracentesis improves venous
return and hemodynamics. [99] The myth of paracentesis-related hemodynamic disasters (probably
coincidences) was based on observations of small numbers of patients.
Pleural Effusions
"Sympathetic" pleural effusions are common in patients with cirrhotic ascites. They are usually unilateral
and right-sided but occasionally may be bilateral with right greater than left. A unilateral left-sided
effusion suggests tuberculosis. [101] A large effusion is referred to as hepatic hydrothorax.[102] Most
carefully studied patients with hepatic hydrothorax have been shown to have a small defect in the right
hemidiaphragm. Occasionally this develops acutely with sudden shortness of breath as the abdomen
decompresses. With large diaphragmatic
1327
defects, ascites may be undetectable on clinical examination despite a large pleural effusion.
The most common symptom associated with hepatic hydrothorax is shortness of breath. Infection of this
fluid occurs, usually as a result of SBP and transmission of bacteria across the diaphragm. [103] The
analysis of uncomplicated hepatic hydrothorax fluid shows that it resembles that of ascites but is not
identical because the pleural fluid is subject to different hydrostatic pressures than is that of the portal
bed. The total protein is higher (by approximately 1 g/dL [10 g/L]) in the pleural fluid than in ascites.
Treatment of hepatic hydrothorax had been very difficult to accomplish before the availability of the
transjugular intrahepatic portasystemic stent-shunt (TIPS). [102] These effusions tend to occur in patients
who are the least compliant or most refractory. Some authors have recommended chest tube insertion and
tetracycline sclerosis. However, chest tubes inserted to treat hepatic hydrothorax usually become very
difficult to remove. [104] Clamping the tube may cause a recurrence of shortness of breath and may lead to
a fluid leak around the tube's insertion site. Direct surgical repair of the defect can be considered, but
typically these patients are poor operative candidates. A peritoneovenous shunt can be considered in the
patient with hepatic hydrothorax and large-volume ascites, but the shunt usually clots after a short time.
Sodium restriction and diuretics are the safest and most effective first-line therapy for hepatic
hydrothorax. TIPS has been reported to be successful and is reasonable second-line treatment. [102] If the
patient is a transplant candidate, the best approach may be to proceed with this procedure.
Abdominal Wall Hernias
Abdominal wall hernias are common in patients with ascites. They are usually umbilical or incisional but
occasionally are inguinal. There is little published information about these hernias. In one study, almost
20% of cirrhotic patients with ascites were found to have umbilical hernias at the time of admission to
the hospital. [105] During four years of follow-up, the majority developed complications. [105] Because of
this high complication rate, surgical treatment should be considered electively in all patients with hernias
and ascites. Ascites should be medically removed preoperatively, because the hernia recurs in 73% of

patients who have ascites at the time of hernia repair but only in 14% of patients who have no ascites at
the time of repair. [106] Surgery should be performed semiemergently for skin ulceration, crusting, or
black discoloration. Emergency surgery should be performed for incarceration or rupture. Rupture is the
most feared complication of umbilical hernias.


Saunders Company
TREATMENT OF ASCITES
Appropriate treatment of the patient with ascites depends on the cause of fluid retention. An accurate
diagnosis regarding the etiology of ascites is very important. The SAAG is helpful diagnostically as well
as in therapeutic decision-making. Patients with a low SAAG do not usually have portal hypertension
and do not respond to salt restriction and diuretics (except nephrotic syndrome). Conversely, patients
with a high SAAG have portal hypertension and are usually responsive to these measures. [7]
Low Albumin Gradient Ascites
Peritoneal carcinomatosis is the most common form of low albumin gradient ascites. [8] Peripheral edema
in these patients responds to diuretics. Edema-free patients treated with diuretics lose only intravascular
volume without a loss of ascites. [7] The mainstay of treatment of nonovarian peritoneal carcinomatosis is
outpatient therapeutic paracentesis. [7] Patients with peritoneal carcinomatosis usually live only for a few
months. Patients with an ovarian malignancy are an exception to this rule; these patients may respond
well to surgical debulking and chemotherapy.
Ascites caused by tuberculous peritonitis (without cirrhosis) is cured by antituberculous therapy.
Diuretics do not speed weight loss unless the patient has underlying portal hypertension from cirrhosis.
Pancreatic ascites may resolve spontaneously, may require endoscopic stenting or operative intervention,
or may respond to somatostatin therapy. [107] A postoperative lymphatic leak from distal splenorenal
shunt or radical lymphadenectomy may also resolve spontaneously but may occasionally require surgical
intervention or peritoneovenous shunting. Chlamydia peritonitis is cured by tetracycline therapy. [15]
Lupus ascites may respond to steroids. [16] Dialysis ascites may respond to aggressive dialysis. [59]
High Albumin Gradient Ascites
Alcohol with or without hepatitis C is the most common cause of liver disease that leads to high albumin
gradient ascites (see Table 78-1) (Table Not Available) . One of the most important steps in treating this
form of ascites is to treat the underlying liver disease by convincing the patient to stop drinking alcohol.
In a period of months, abstinence can result in healing of the reversible component of alcoholic liver
disease. [13] Ascites may resolve or become more responsive to medical therapy during this time. Patients
with other forms of treatable liver disease (e.g., autoimmune chronic active hepatitis, iron-storage
disease, or Wilson's disease) should receive specific therapy for those diseases. Specific therapy may
improve the patient's overall liver function and increase ease of management of the patient's ascites.
However, these diseases are less reversible than is alcoholic liver disease, and by the time that ascites is
present, these patients may be better candidates for liver transplantation than for protracted medical
therapy.
Hospitalization.
Outpatient treatment of patients with small-volume ascites can be attempted initially. However, patients

with large-volume ascites and those who are resistant to outpatient treatment usually require
hospitalization for definitive diagnosis and management of their liver disease as well as their fluid
overload. [108] Many of these patients also have gut hemorrhage, encephalopathy, infection, or
hepatocellular carcinoma. An intensive period of inpatient education and treatment is required to
convince the patient that the diet and diuretics are actually effective and worth the effort that it will take
to follow the regimen at home.
1328
Precipitating Cause.
It may be valuable to determine the precipitating cause of ascites formation (e.g., dietary indiscretion or
noncompliance with diuretics). Further diet education may help to prevent future hospitalizations for
ascites. Ascites may accumulate because of the saline infusions given perioperatively or given in the
treatment of variceal hemorrhage. Ascites that develops postoperatively or with variceal hemorrhage may
resolve without need for long-term treatment.
Diet Education.
Fluid loss and weight change are directly related to sodium balance in patients with portal
hypertension-related ascites. In the presence of avid renal retention of sodium, dietary sodium restriction
is essential. The patient and the food preparer should be educated about a sodium-restricted diet by a
dietitian. Severe sodium-restricted diets (e.g., 500 mg or 22 mmol/day of sodium) are feasible (but not
palatable) in an inpatient setting but are unrealistic for outpatients. The dietary sodium restriction that the
author recommends for inpatients and outpatients is 2 g/day (88 mmol/day).
Fluid Restriction.
An indiscriminate fluid restriction in treating the patient with cirrhotic ascites is inappropriate;
hypernatremia may result. It is the sodium restriction, not the fluid restriction, that results in weight loss;
fluid follows sodium passively. The chronic hyponatremia usually seen in patients with cirrhotic ascites
is seldom morbid. Attempts to rapidly correct hyponatremia in this setting can lead to more
complications than occur with hyponatremia itself. Severe hyponatremia (e.g., serum sodium <120
mmol/L) does warrant fluid restriction in the patient with cirrhotic ascites. Cirrhotic patients do not
usually have symptoms of hyponatremia until the sodium level is below 110 mmol/L or unless the
decline in sodium is very rapid. Indiscriminate fluid restriction serves only to alienate patients, nurses,
and dietitians.
No Bed Rest.
Although it is traditional to order bed rest, there are no controlled trials to support this practice. Upright
posture may aggravate the plasma renin elevation found in most cirrhotic patients with ascites.
Theoretically, this may increase sodium avidity. However, having successfully treated hundreds of
patients without ordering bed rest, it is difficult for the author to believe that bed rest is necessary. These
emaciated patients may develop decubitus ulcers if they remain in bed.
Urine Sodium Excretion.

Measurement of 24-hour urinary sodium excretion is a helpful parameter to follow in patients with portal
hypertension-related ascites. Completeness of collection can be assessed by measurement of urinary
creatinine. Cirrhotic men should excrete 15 to 20 mg of creatinine per kilogram of body weight per day,
and women should excrete 10 to 15 mg/kg/day. Excretion of less creatinine is indicative of an incomplete
collection. Only the 10% to 15% of patients who have significant spontaneous natriuresis can be
considered for dietary sodium restriction alone (i.e., without diuretics). [108] However, when given a
choice, most patients would prefer to take some diuretics and have a more liberal sodium intake rather
than take no pills and have a more severe sodium restriction. Compliant patients, including outpatients,
can collect complete 24-hr specimens, contrary to popular belief. Total collections are preferable to
"spot" specimens (i.e., sodium concentration in a random specimen) because sodium excretion may not
be uniform throughout the day and because a 24-hr urine volume must be available in addition to the spot
concentration in order to estimate the 24-hour excretion from the spot concentration.
Because urine is the most important route of excretion of sodium in the absence of diarrhea or
hyperthermia, and because dietary intake is the only source of nonparenteral sodium, dietary intake and
urinary excretion should be roughly equivalent if the weight is stable. Nonurinary sodium losses are less
than 10 mmol/day in these patients. [109] If body weight is not declining satisfactorily, this may be due to
inadequate natriuresis or failure to properly restrict sodium intake, or both. Monitoring 24-hr urinary
sodium excretion and daily weight will clarify this issue. If the patient's weight increases despite urinary
losses in excess of prescribed dietary intake, one can assume that the patient is eating more sodium than
his or her prescribed diet. If the patient is losing weight, the urinary sodium loss per day must exceed the
intake.
Avoid Use of Urinary Bladder Catheters.
Many physicians promptly insert a bladder catheter in inpatients with cirrhosis in order to more
accurately monitor urine output. Unfortunately, these immunocompromised patients regularly have
urinary tract infections at the time of admission. [79] Urethral trauma from catheter insertion in the setting
of cystitis can lead to bacteremia. Prolonged catheterization predictably leads to cystitis and possibly to
urosepsis in these patients. The author only uses these catheters in the intensive care unit setting; these
portals of entry for bacteria should be removed as soon as possible. Twenty-four-hour urine specimens
can be collected completely without catheters.
Diuretics.
Spironolactone is the mainstay of treatment of cirrhotic ascites, but it is very slow to increase natriuresis.
Single-agent diuretic therapy with spironolactone usually requires two weeks before weight loss begins.
Although single-agent spironolactone has been shown to be superior to single-agent furosemide, [110] this
author prefers to start spironolactone and furosemide together on the patient's first or second day in the
hospital, beginning with 100 mg and 40 mg, respectively, in a single morning dose [108] (Fig. 78-5)
(Figure Not Available) . Amiloride, 10 mg/day, can be substituted for spironolactone; amiloride is less
available and more expensive than spironolactone but is more rapidly effective and does not cause
gynecomastia. The half-life of spironolactone in normal control patients is approximately 24 hr but is
markedly prolonged in patients with cirrhosis; almost one month is required to reach steady state. [111]
There is no reason to dose the drug multiple times per day in view of its long half-life. A loading dose
may be appropriate but has not been studied. Single daily doses are most appropriate and enhance

compliance; 25-, 50-, and 100-mg spironolactone tablets are available. If the combination of 100 mg/day
of spironolactone (or 10 mg of amiloride) and 40 mg/day of furosemide orally is ineffective in increasing
urinary sodium or decreasing body weight, the doses of both drugs should be simultaneously increased as
needed (e.g., 200 plus 80, then 300 plus 120, and finally 400 mg/day of spironolactone and 160 mg/day
of furosemide). Starting both drugs at once speeds the onset of diuresis, in the author's experience.
Slowly "working up" to 400 to 600 mg/day of spironolactone before adding furosemide delays diuresis
and results in hyperkalemia.
The 100:40 ratio of spironolactone to furosemide in the daily dose usually maintains normokalemia. The
ratio of spironolactone
1329
Figure 78-5 (Figure Not Available) Treatment of patients with cirrhosis and ascites. TIPS, transjugular intrahepatic
portasystemic stent-shunt. (Adapted from Runyon, B. A. Care of patients with ascites. N. Engl. J. Med. 330:337, 1994.
Reproduced with permission from the Massachusetts Medical Society.)
and furosemide can be adjusted to correct serum potassium problems. Occasionally, an alcoholic patient
with no food intake will have hypokalemia at the time of admission and for a variable interval thereafter.
These patients should receive single-agent spironolactone until their serum potassium normalizes; then
furosemide is added. Combined with a sodium-restricted diet, the spironolactone and furosemide regimen
has been demonstrated in a study that screened almost 4000 patients to achieve a successful diuresis in
more than 90% of cirrhotic patients. [112]
Intravenous diuretics cause acute decreases in glomerular filtration in these patients and should be
avoided. [113] If rapid weight loss is desired, therapeutic paracenteses should be performed (see later).
There is no limit to the daily weight loss of patients who have massive edema. Once the edema has
resolved, 0.5 kg/day is probably a reasonable maximum. [114] Encephalopathy, serum sodium less than
120 mmol/L despite fluid restriction, or serum creatinine greater than 2 mg/dL (180 mumol/L) should
result in cessation of diuretics and reassessment of the situation. Potassium abnormalities are almost
never prohibitive because of the ability to adjust the ratio of the diuretics. Patients with parenchymal
renal disease (e.g., diabetic nephropathy) usually require higher doses of furosemide and lower doses of
spironolactone than mentioned earlier. Patients who develop complications of a careful attempt at
diuretic treatment usually fail first-line treatment and require second-line therapy. Prostaglandin
inhibitors (e.g., nonsteroidal anti-inflammatory drugs) should be avoided in patients with ascites because
they curtail diuresis, may promote renal failure, and cause gastrointestinal bleeding. [115]
An attempt to minimize the quantity of fluid can be beneficial to the patient in terms of comfort as well
as in the prevention of hepatic hydrothorax and hernias. Also, a diuresis-related concentration of ascitic
fluid has been shown to increase the fluid's opsonic activity ten-fold, and theoretically, may be valuable
in preventing spontaneous ascitic fluid infection. [116]
In the past, patients with ascites frequently occupied hospital beds for prolonged periods of time because
of confusion regarding their diagnosis and treatment and because of iatrogenic problems. Although a dry
abdomen is a reasonable ultimate goal, it should not be a prerequisite for the patient's discharge from the
hospital. Patients who are stable, with ascites as their major problem, can be discharged to the clinic after
it has been determined that they are responding to their medical regimen. This usually requires less than
seven inpatient days. In order for patients to be discharged early from the hospital, they must be seen in

the outpatient setting promptly (e.g., in approximately seven more days).
No Sodium Bicarbonate.
A minority of patients with cirrhosis and ascites develop a mild renal tubular acidosis. Many
nephrologists recommend oral sodium bicarbonate administration in this setting. Clearly, such treatment
dramatically increases sodium intake and cannot be advocated in the absence of outcome data supporting
its use.
Outpatient Management.
After discharge, the author usually re-evaluates patients in one or two weeks in the outpatient clinic.
Body weight, orthostatic symptoms, serum electrolytes, urea, and creatinine are monitored.
Twenty-four-hour urine specimens can be collected to assist with management decisions. It has been the
author's experience that compliant outpatients are successful in collecting complete specimens, when
adequate written instructions are provided. Subsequent frequency of follow-up is determined by the
patient's response to treatment and by the stability of the patient. The author usually evaluates these
patients every two to four weeks until it is clear that they are responding to treatment and not developing
problems. Intensive outpatient follow-up helps to prevent subsequent hospitalizations.
Diuretic doses and diet sodium intake are adjusted to achieve weight loss and a negative sodium balance.
Patients who are gaining fluid weight despite diuretics should not be labeled as diuretic-resistant until it
is documented that they are compliant with the diet. Urine sodium monitoring provides insight into
compliance. Patients who are excreting more than 88 mmol/day of sodium in the urine should be losing
weight if they are consuming less than 88 mmol/day of sodium. This author has encountered patients
who were labeled as diuretic-resistant when they were dramatically noncompliant with the diet and were
gaining weight while excreting more than 500 mmol/day of sodium in the urine. Most patients who are
initially thought to be diuretic-resistant are eventually found to be noncompliant with the diet in the
author's experience. Education about diet is crucial for successful treatment of such patients. Truly
diuretic-resistant patients excrete nearly sodium-free urine despite a maximum dose of diuretics.
During long-term follow-up, abstinent alcoholics may become
1330
more sensitive to diuretics and may even be tapered off diuretics.
Refractory Ascites.
Refractory ascites is defined as fluid overload unresponsive to salt restriction and high-dose diuretic
treatment. The failure may be manifested by minimal to no weight loss despite diuretics or by the
development of complications of diuretics. Several studies have shown that less than 10% of patients
with cirrhotic ascites are refractory to standard medical therapy.[110] [112]
In the 1960s, portacaval shunts were used for treatment of refractory ascites, but operative hemorrhagic
complications and portasystemic encephalopathy led to abandonment of this approach. In Europe in the
1970s, the Paris pump was used to ultrafilter ascitic fluid and then to reinfuse it intravenously.
Unfortunately, this was complicated by disseminated intravascular coagulation and was abandoned.

Viable options for patients refractory to routine medical therapy include peritoneovenous shunt,
therapeutic paracentesis, and liver transplantation (see Fig. 78-5) (Figure Not Available) . In the
mid-1970s, the peritoneovenous shunt was promoted as a new "physiologic" treatment in the
management of ascites. Reports of shunt failure, fatal complications of shunt insertion, and randomized
trials demonstrating no survival advantage have led to relegation of this procedure to third-line therapy of
cirrhotic ascites [112] (see Fig. 78-5) (Figure Not Available) .
Therapeutic abdominal paracentesis is one of the oldest medical procedures. In the 1980s after 2000
years of use, scientific data regarding large-volume paracentesis were reported. [117] [118] [119] [120] Patients
were documented to tolerate large volume taps very well, just as the patients did in the 1940s and much
earlier. In one large randomized controlled trial, therapeutic paracentesis plus colloid infusion was shown
to lead to fewer minor (asymptomatic) changes in electrolytes and serum creatinine than diuretic therapy.
[117] However, no differences in morbidity or mortality could be demonstrated. [117] Therapeutic
paracentesis now appears to be the first-line therapy of patients with tense ascites and second-line
therapy for cirrhotic patients who are refractory to diuretics [108] (see Fig. 78-5) (Figure Not Available) .
Colloid Replacement.
One controversial issue regarding therapeutic paracentesis is that of colloid replacement. In one study
patients with tense ascites were randomized to receive albumin (10 g/L of fluid removed) versus no
albumin, after therapeutic paracentesis. [120] The group that received no albumin developed statistically
significantly more (asymptomatic) changes in electrolytes, plasma renin, and serum creatinine than did
the albumin group, but no more clinical morbidity or mortality. Although another study has documented
that the subset of patients who develop a postparacentesis rise in plasma renin have decreased life
expectancy, there has been no study demonstrating decreased survival in patients who are given no
plasma expander compared with patients given albumin after paracentesis. [121] Despite this finding, the
authors of these two studies recommend a routine infusion of albumin after therapeutic paracentesis. [120]
[121] However, infusions of albumin markedly increase albumin degradation, and albumin is very
expensive. [122] [123] In a study performed more than 30 years ago, 58% of infused albumin was accounted
for by increased degradation, and a 15% increase in serum albumin led to a 39% increase in degradation.
[122] Increasing albumin concentration in cell culture media has been shown to decrease the synthesis of
albumin. [124] Because of the cost (i.e., $2 to $25/g or $ 100 to $ 1250/tap), it is difficult to justify the
expense of a routine infusion of albumin based on the data at hand.
At the University of Southern California Liver Unit, approximately 500 therapeutic paracenteses are
performed each year. Therapeutic paracentesis without colloid infusion has been routine for decades
without apparent problems due to volume depletion. A pilot study performed at this unit has
demonstrated no difference in morbidity, frequency of hepatorenal syndrome, or mortality in patients
receiving albumin versus no albumin after chronic large volume taps in patients with diuretic-resistant
ascites. [125] Part of the confusion regarding this issue relates to study design. In the study from
Barcelona, patients with tense ascites could be entered into the trial of albumin versus no albumin; 31%
of these patients were not even receiving diuretics. [120] In contrast, the University of Southern California
study included only patients with diuretic-resistant ascites. [125] It seems more appropriate to study the
population that really needs chronic paracenteses (i.e., the diuretic-resistant group) rather than patients
with tense ascites.

This author's approach to patients with tense ascites is to take off enough fluid (4 to 6 L) to relieve
intra-abdominal pressure and to rely on diuretics to remove the remainder. To remove all of the fluid by
paracentesis when most of it could be removed with diuretics seems inappropriate. In addition, the
patient with early cirrhosis seems to be more sensitive to volume changes (e.g., with therapeutic
paracentesis) than is the patient with advanced cirrhosis; this may also help to explain the differences
between the aforementioned studies. [126] Patients with early cirrhosis and diuretic-sensitive ascites
should be treated with diuretics and not with paracentesis; these patients may be more sensitive to
paracentesis-related volume depletion. [126] Chronic therapeutic paracenteses should be reserved for the
10% of patients who fail diuretic treatment.
Other studies have compared less expensive alternatives with albumin. However, there were no
differences in electrolyte imbalance or clinically relevant complications between groups. [127] In addition,
the most recent studies advocate giving one half of the plasma expander immediately after the tap and the
other half after six hours. [121] [127] This approach converts an otherwise simple outpatient procedure into
an all-day clinic visit or a brief hospitalization! This seems unwarranted. More convincing data involving
appropriate groups of patients and regarding clinically relevant issues rather than asymptomatic
laboratory abnormalities are required before albumin or other alternatives can be recommended.
No Dopamine.
Nephrologists occasionally recommend dopamine infusion in the treatment of refractory ascites.

However, this drug does not even increase glomerular filtration rate in the setting of cirrhosis. [128] Its use
cannot be advocated in the absence of supportive data.
Transjugular Intrahepatic Portasystemic Stent-Shunt.
TIPS is a side-to-side portacaval shunt that is placed by an interventional radiologist under local
anesthesia. It was first used for treatment of refractory variceal bleeding but has also been advocated for
diuretic-resistant ascites. [129] TIPS has been received with great enthusiasm in the 1990s, similar to the
enthusiasm for the peritoneovenous shunt in the 1970s. Just as with peritoneovenous shunting, TIPS was
overused until serious complications and suboptimal efficacy were reported.
1331
Randomized trials in diuretic-resistant patients are needed before its position in the algorithm of
treatment of patients with ascites can be finalized (see Fig. 78-5) (Figure Not Available) . A recent
National Institutes of Health Consensus Conference has listed TIPS as "unproven but promising" for
treatment of refractory ascites. [130]
Finally, orthotopic liver transplantation should be considered in the treatment options of patients with
ascites. In areas where waiting times for transplantation are 12 to 18 months, transplant must be
considered very early after first evidence of decompensation. Once patients become refractory to routine
medical therapy, they should be prioritized for transplant. The 12-month survival of patients with ascites
refractory to medical therapy is only 25%. [131] Transplantation has a 12-month survival three times this
value.



Saunders Company
SUMMARY OF TREATMENT OF CIRRHOTIC ASCITES

The mainstay of therapy of patients with cirrhotic ascites is dietary sodium restriction and diuretics (see
Fig. 78-5) (Figure Not Available) . Standard medical therapy is effective in 90% of patients. Therapeutic
paracentesis should be performed acutely in patients with tense ascites and as a second-line chronic
treatment of the 10% of patients who are refractory to medical therapy. Data are currently insufficient to
justify routine colloid replacement after therapeutic paracentesis. Use of colloid has been shown to
prevent laboratory abnormalities; however, withholding colloid has not been shown to result in increased
morbidity or mortality. If a physician feels compelled to give post-paracentesis colloid while awaiting
more data, use of the less expensive dextran would seem appropriate. TIPS is a second- or third-line
treatment that should be reserved for diuretic-resistant patients who are not transplant candidates or who
are awaiting transplant or patients with hepatic hydrothorax. Peritoneovenous shunting is a third-line
therapy and should be reserved for special circumstances, such as ascites due to a surgical lymphatic tear
that does not respond to medical therapy and for patients with ascites refractory to medical treatment
with circumstances that preclude chronic therapeutic paracenteses. Liver transplantation is a second- or
third-line therapy for patients with cirrhosis and ascites. The selection and timing of transplantation are
dependent on local or regional expertise and on organ availability.


Saunders Company
PROGNOSIS
Cirrhotic ascites is associated with significant morbidity and mortality, which is partly related to the
severe underlying liver disease and partly due to the ascites per se. Half of patients in whom cirrhosis is
detected prior to the development of "decompensation" (i.e., prior to the development of ascites,
jaundice, encephalopathy, or gastrointestinal hemorrhage) develop ascites in ten years. [132] Once ascites
is present, the expected mortality is approximately 50% in just two years. [133] With transplantation, the
patient's survival time is dramatically improved.


Saunders Company
REFERENCES
1. Schrier, R. W., Arroyo, V., Bernardi, M., et al. Peripheral arterial vasodilation hypothesis: A proposal for the initiation of
renal sodium and water retention in cirrhosis. Hepatology 8:1151, 1988.
2. Sherlock, S., and Shaldon, S. The aetiology and management of ascites in patients with hepatic cirrhosis: A review. Gut
4:95, 1963.
3. Lieberman, F. L., and Reynolds, T. B. Plasma volume in cirrhosis of the liver: Its relation to portal hypertension, ascites,
and renal failure. J. Clin. Invest. 46:1297, 1967.
4. Lieberman, F. L., Denison, E. K., and Reynolds, T. B. The relationship of plasma volume, portal hypertension, ascites,
and renal sodium retention in cirrhosis: The overflow theory of ascites formation. Ann. N. Y. Acad. Sci. 70:202, 1970.
5. Morali, G. A., Sniderman, K. W., Deitel, K. M., et al. Is sinusoidal portal hypertension a necessary factor for the
development of ascites? J. Hepatol. 16:249, 1992.
6. Sogni,P., Moreau, R., Gadano, A., et al. The role of nitric oxide in the hyperdynamic circulatory syndrome associated
with portal hypertension. J. Hepatol. 23:218, 1995.
7. Pockros, P. J., Esrason, K. T., Nguyen, C., et al. Mobilization of malignant ascites with diuretics is dependent on ascitic
fluid characteristics. Gastroenterology 103:1302, 1992.
8. Runyon, B. A., Hoefs, J. C., and Morgan, T. R. Ascitic fluid analysis in malignancy-related ascites. Hepatology 8:1104,
1988.
9. Schrier, R. W. Pathogenesis of sodium and water retention in high-output and low-output cardiac failure, nephrotic
syndrome, cirrhosis, and pregnancy. N. Engl. J. Med. 319:1065, 1988.
10. Miedema,
E. B., Bissada, N. K., Finkbeiner, A. E., et al. Chylous ascites complicating retroperitoneal
lymphadenectomy for testis tumors: Management with peritoneovenous shunting. J. Urol. 120:377, 1978.
11. Brown, M. W., and Burk, R. F. Development of intractable ascites following upper abdominal surgery in patients with
cirrhosis. Am. J. Med. 80:879, 1986.
12. Kodali,
V. P., Gordon, S. C., Silverman, A. L., et al. Cryptogenic liver disease in the United States: Further evidence for
non-A, non-B, and non-C hepatitis. Am. J. Gastroenterol. 89:1836, 1994.
13. Reynolds, T. B., Geller, H. M., Kuzma, O. T., et al. Spontaneous decrease in portal pressure with clinical improvement
in cirrhosis. N. Engl. J. Med. 263:734, 1960.
14. Powell,E. E., Cooksley, W. G. E., Hanson, R., et al. The natural history of nonalcoholic steatohepatitis: A follow-up
study of 42 patients for up to 21 years. Hepatology 11:74, 1990.
15. Muller-Schoop, J. W., Wang, S. P., Munzinger, J., et al. Chlamydia trachomatosis as possible cause of peritonitis and

perihepatitis in young women. BMJ 1:1022, 1978.
16. Wilkins, K. W., and Hoffman, G. S. Massive ascites in systemic lupus erythematosus. J. Rheumatol. 12:571, 1985.
17. Cattau,E. I., Benjamin, S. B., Knuff, T. E., et al. The accuracy of the physical exam in the diagnosis of suspected
ascites. JAMA 247:1164, 1982.
18. Goldberg, B. B., Goodman, G. A., and Clearfield, H. R. Evaluation of ascites by ultrasound. Radiology 96:15, 1970.
19. Guarner,
C., and Runyon, B. A. Spontaneous bacterial peritonitis: Pathogenesis, diagnosis, and treatment.
Gastroenterologist 3:311, 1995.
20. Runyon, B. A. Paracentesis of ascitic fluid: A safe procedure. Arch. Intern. Med. 146:2259, 1986.
21. McVay, P. A., and Toy, P. T. C. Y. Lack of increased bleeding after paracentesis and thoracentesis in patients with mild
coagulation abnormalities. Transfusion 13:164, 1991.
22. Runyon, B. A., Akriviadis, E. A., and Keyser, A. J. The opacity of portal hypertension-related ascites correlates with the
fluid's triglyceride concentration. Am. J. Clin. Pathol. 96:142, 1991.
23. Rector, W. G. Spontaneous chylous ascites of cirrhosis. J. Clin. Gastroenterol. 6:369, 1984.
24. Runyon,B. A. Ascitic fluid bilirubin concentration as a key to the diagnosis of choleperitoneum. J. Clin. Gastroenterol.
9:543, 1987.
25. Hoefs, J. C. Diagnostic paracentesis: A potent clinical tool. Gastroenterology 98:230, 1990.
26. Hoefs,
J. C. Increase in ascites WBC and protein concentrations during diuresis in patients with chronic liver disease.
27. Runyon, B. A. Ascites. In Yamada, T., Alpers, D., Owyang, C., et al. (eds.), Textbook of Gastroenterology, 2nd ed.
Philadelphia, J. B. Lippincott, 1995, pp. 927-952.
28. Hoefs, J. C. Serum protein concentration and portal pressure determine
1332
the ascitic fluid protein concentration in patients with chronic liver disease. J. Lab. Clin. Med. 102:260, 1983.
29. Runyon, B. A., Montano, A. A., Akriviadis, E. A., et al. The serum-ascites albumin gradient is superior to the
exudate-transudate concept in the differential diagnosis of ascites. Ann. Intern. Med. 117:215, 1992.
30. Hoefs,J. C. Globulin correction of the albumin gradient: Correlation with measured serum to ascites colloid osmotic
gradient. Hepatology 16:396, 1992.
31. Runyon,B. A., Canawati, H. N., and Akriviadis, E. A. Optimization of ascitic fluid culture technique. Gastroenterology
95:1351, 1988.
32. Runyon, B. A., Antillon, M. R., Akriviadis, E. A., et al. Bedside inoculation of blood culture bottles with ascitic fluid is
superior to delayed inoculation in the detection of spontaneous bacterial peritonitis. J. Clin. Microbiol. 28:2811, 1990.

33. Davis,
T. E., and Fuller, D. D. Direct identification of bacterial isolates in blood cultures by using a DNA probe. J. Clin.
Microbiol. 29:2193, 1991.
34. Runyon,B. A., and Hoefs, J. C. Ascitic fluid analysis before, during, and after spontaneous bacterial peritonitis.
35. Runyon,B. A. Low-protein-concentration ascitic fluid is predisposed to spontaneous bacterial peritonitis.

Gastroenterology 91:1343, 1986.
36. Runyon, B. A. Cardiac ascites: A characterization. J. Clin. Gastroenterol. 10:410, 1988.
37. Akriviadis,E. A., and Runyon, B. A. The value of an algorithm in differentiating spontaneous from secondary bacterial
peritonitis. Gastroenterology 98:127, 1990.
38. Runyon, B. A. Amylase levels in ascitic fluid. J. Clin. Gastroenterol. 9:172, 1987.
39. Manohar, A., Simjee, A. A., and Pettengill, K. E. Symptoms and investigative findings in 145 patients with tuberculous
peritonitis diagnosed by peritoneoscopy and biopsy over a five year period. Gut 31:1130, 1990.
40. Shakil,A. O., Korula, J., Murray, N. G. B., et al. Diagnostic features of tuberculous peritonitis and presence of chronic
liver disease: A case control study. Am. J. Med. 100:179, 1996.
41. Altamirano,M., Kelly, M. T., Wong, A., et al. Characterization of a DNA probe for detection of Mycobacterium
tuberculosis complex in clinical samples by polymerase chain reaction. J. Clin. Microbiol. 30:2173, 1992.
42. Cardozo,P. L. A critical evaluation of 3000 cytologic analyses of pleural fluid, ascitic fluid, and pericardial fluid. Acta
Cytol. 10:455, 1966.
43. Chetty, R., Learmonth, G. M., and Taylor, D. A. Giant cell hepatocellular carcinoma. Cytopathology 1:233, 1990.
44. Wang, S.-S., Lu, C.-W., Chao, Y., et al. Malignancy-related ascites: A diagnostic pitfall of spontaneous bacterial
peritonitis by ascitic fluid polymorphonuclear cell count. J. Hepatol. 20:79, 1994.
45. Press,
O. W., Press, N. O., and Kaufman, S. D. Evaluation and management of chylous ascites. Ann. Intern. Med.
96:358, 1982.
46. Runyon, B. A., and Antillon, M. R. Ascitic fluid pH and lactate: Insensitive and nonspecific tests in detecting ascitic
fluid infection. Hepatology 13:929, 1991.
47. Albillos,
A., Cuervas-Mons, V., Millan, I., et al. Ascitic fluid polymorphonuclear cell count and serum to ascites
albumin gradient in the diagnosis of bacterial peritonitis. Gastroenterology 98:134, 1990.
48. Runyon, B. A. Elevated ascitic fluid fibronectin: A non-specific finding. J. Hepatol. 3:219, 1986.
49. Runyon,B. A. Malignancy-related ascites and ascitic fluid "humoral tests of malignancy." [Editorial]. J. Clin.
Gastroenterol. 18:94, 1994.
50. Loewenstein,M. S., Rittgers, R. A., Feinerman, A. E., et al. CEA assay of ascites and detection of malignancy. Ann.
Intern. Med. 88:635, 1978.
51. Hillebrand,
D. J., Runyon, B. A., Yasmineh, W., et al. Ascitic fluid adenosine deaminase insensitivity in detecting
tuberculous peritonitis. Hepatology 24:1408, 1996.

52. Runyon, B. A. Spontaneous bacterial peritonitis associated with cardiac ascites. Am. J. Gastroenterol. 79:796, 1984.
53. Kurtz,
R. C., and Bronzo, R. L. Does spontaneous bacterial peritonitis occur in malignant ascites? Am. J. Gastroenterol.
77:146, 1982.
54. Wolfe,G. M., and Runyon, B. A. Spontaneous Salmonella infection of high protein non-cirrhotic ascites. J. Clin.
55. Arora,
A., Seth, S., Acharya, S. K., et al. Hepatic coma as a presenting feature of constrictive pericarditis. Am. J.
56. DeGroen, P. C., Rakela, J., Moore, C., et al. Diagnostic laparoscopy in gastroenterology: A 14-year experience. Dig.
Dis. Sci. 32:677, 1987.
57. Weisman, I. M., Moreno, A. J., Parker, A. L., et al. Gastrointestinal dissemination of coccidioidomycosis. Am. J.
58. Cappell,M. S., and Shetty, V. A multicenter, case-controlled study of the clinical presentation and etiology of ascites
and of the safety and clinical efficacy of diagnostic abdominal paracentesis in HIV seropositive patients. Am. J.
59. Mauk,P. M., Schwartz, J. T., Lowe, J. E., et al. Diagnosis and course of nephrogenic ascites. Arch. Intern. Med.
148:1577, 1988.
60. Ackerman, Z. Ascites in nephrotic syndrome: Incidence, patients' characteristics and complications. J. Clin.
61. Mauer,K., and Manzione, N. C. Usefulness of the serum-ascites albumin gradient in separating transudative from
exudative ascites: Another look. Dig. Dis. Sci. 33:1208, 1988.
62. Ikeda,
Y., Kanematsu, T., Matsumata, T., et al. Liver resection and intractable postoperative ascites.
Hepatogastroenterology 40:14, 1993.
63. Schenker, J. G., and Weinstein, D. Ovarian hyperstimulation syndrome: A current survey. Fertil. Steril. 30:255, 1978.
64. Case Record. N. Engl. J. Med. 327:1014, 1992.
65. Kemeny, N., Seiter, K., Martin, D., et al. A new syndrome: Ascites, hyperbilirubinemia, and hypoalbuminemia after
biochemical modulation of fluorouracil with N-phosphonacetyl-L-aspartate (PALA). Ann. Intern. Med. 115:946, 1991.
66. Correia,
J. P., and Conn, H. O. Spontaneous bacterial peritonitis in cirrhosis: Endemic or epidemic. Med. Clin. North
Am. 59:963, 1975.
67. Runyon,B. A. Monomicrobial nonneutrocytic bacterascites: A variant of spontaneous bacterial peritonitis. Hepatology
12:710, 1990.
68. Runyon,B. A., and Hoefs, J. C. Culture-negative neutrocytic ascites: A variant of spontaneous bacterial peritonitis.
69. Runyon, B. A., Canawati, H. N., and Hoefs, J. C. Polymicrobial bacterascites: A unique entity in the spectrum of
infected ascitic fluid. Arch. Intern. Med. 146:2173, 1986.

70. Berg,
R. D., and Garlington, A. W. Translocation of certain indigenous bacteria from the gastrointestinal tract to the
mesenteric lymph nodes and other organs in a gnotobiotic mouse model. Infect. Immun. 23:403, 1979.
71. Guarner,
C., Runyon, B. A., Young, S., et al. Intestinal bacterial overgrowth and bacterial translocation in an
experimental model of cirrhosis. Hepatology 22:166A, 1995.
72. Runyon, B. A., Squier, S. U., and Borzio, M. Translocation of gut bacteria in rats with cirrhosis to mesenteric lymph
nodes partially explains the pathogenesis of spontaneous bacterial peritonitis. J. Hepatol. 21:792, 1994.
73. Runyon, B. A., Morrissey, R., Hoefs, J. C., et al. Opsonic activity of human ascitic fluid: A potentially important
protective mechanism against spontaneous bacterial peritonitis. Hepatology 5:634, 1985.
74. Runyon,B. A. Patients with deficient ascitic fluid opsonic activity are predisposed to spontaneous bacterial peritonitis.
75. Runyon,B. A., Sugano, S., Kanel, G., et al. A rodent model of cirrhosis and spontaneous bacterial peritonitis.
76. Dunn,D. L., Barke, R. A., Knight, N. B., et al. Role of resident macrophages, peripheral neutrophils, and
translymphatic absorption in bacterial clearance from the peritoneal cavity. Infect. Immun. 49:257, 1985.
77A. McHutchison, J. G., and Runyon, B. A. Spontaneous bacterial peritonitis. In Surawicz, C. M., and Owen, R. L. (eds.),
Gastrointestinal and Hepatic Infections. Philadelphia, W. B. Saunders, 1994, p. 455.
77B. Caly, W. R., and Strauss, E. A prospective study of bacterial infections in patients with cirrhosis. J. Hepatol. 18:353,
1993.
78A. Bernard, B., Cadranel, J.-F., Valla, D., et al. Prognostic significance of bacterial infection in bleeding cirrhotic
patients: A prospective study. Gastroenterology 108:1828, 1995.
78B. Hamdani, R., Young, S., and Runyon, B. A. Increased bacterial translocation in cirrhotic rats undergoing hemorrhagic
shock. Gastroenterology 106:A904, 1994.
79. Ho,H., Guerra, L. G., Zuckerman, M. J., et al. Urinary tract infection: A predisposing factor for spontaneous bacterial
peritonitis. Gastroenterology 98:A593, 1990.
80. Pelletier,
G., Lesur, G., Ink, O., et al. Asymptomatic bacterascites: Is it spontaneous bacterial peritonitis? Hepatology
14:112, 1991.
81. Runyon,B. A. Four days of antibiotics can be effective therapy of culture-negative neutrocytic ascites or of delayed
growth culture-positive spontaneous peritonitis. Hepatology 6:1139, 1986.
82. Gill, M. A., and Kern, J. W. Altered gentamicin distribution in ascitic patients. Am. J. Hosp. Pharm. 36:1704, 1979.
83. Papadakis,M. A., and Arieff, A. I. Unpredictability of clinical evaluation of renal function in cirrhosis: Prospective
study. Am. J. Med. 82:945, 1987.
1333

84. Cabrera, J., Arroyo, V., Ballesta, A. M., et al. Aminoglycoside nephrotoxicity in cirrhosis. Gastroenterology 82:97,
1982.
85. Moore,
R. D., Smith, C. R., and Lietman, P. S. Increased risk of renal dysfunction due to interaction of liver disease and
aminoglycosides. Am. J. Med. 80:1093, 1986.
86. Ariza,J., Gudiol, F., Dolz, C., et al. Evaluation of aztreonam in the treatment of spontaneous bacterial peritonitis in
patients with cirrhosis. Hepatology 6:906, 1986.
87. Runyon, B. A. Susceptibility of spontaneous bacterial peritonitis isolates to cefotaxime. Presented at the 17th Annual
International Congress on Chemotherapy, Berlin, Germany, June 24, 1991.
88. Felisart,
J., Rimola, A., Arroyo, V., et al. Randomized comparative study of efficacy and nephrotoxicity of ampicillin
plus tobramycin versus cefotaxime in cirrhotics with severe infections. Hepatology 5:457, 1985.
89. Runyon, B. A., Akriviadis, E. A., Sattler, F. R., et al. Ascitic fluid and serum cefotaxime and desacetylcefotaxime levels
in patients treated for bacterial peritonitis. Dig. Dis. Sci. 36:1782, 1991.
90. Navasa, M., Planas, R., Clemente, G., et al. Oral ofloxacin versus intravenous cefotaxime in the treatment of
non-complicated spontaneous bacterial peritonitis (SBP) in cirrhosis: Results of a multicenter, prospective, randomized
trial. J. Hepatol. 21:S11, 1994.
91. Runyon,B. A., McHutchison, J. G., Antillon, M. R., et al. Short-course vs long-course antibiotic treatment of
spontaneous bacterial peritonitis: a randomized controlled trial of 100 patients. Gastroenterology 100:1737, 1991.
92. Soriano, G., Teixedo, M., Guarner, C., et al. Selective intestinal decontamination prevents spontaneous bacterial
peritonitis. Gastroenterology 100:477, 1991.
93. Gines,P., Rimola, A., Planas, R., et al. Norfloxacin prevents spontaneous bacterial peritonitis recurrence in cirrhosis:
Results of a double-blind, placebo-controlled trial. Hepatology 12:716, 1990.
94. Soriano,
G., Guarner, C., Tomas, A., et al. Norfloxacin prevents bacterial infection in cirrhotics with gastrointestinal
hemorrhage. Gastroenterology 103:1267, 1992.
95. Runyon, B. A., Borzio, M., Young, S., et al. Effect of selective bowel decontamination with norfloxacin on spontaneous
bacterial peritonitis, translocation, and survival in an animal model of cirrhosis. Hepatology 21:1719, 1995.
96. Novella,
M. T., Soriano, G., Gana, J., et al. Prophylaxis of the first spontaneous bacterial peritonitis in cirrhotic patients.
Hepatology 20:115A, 1994.
97. Rolando,N., Gimson, A., Philpott-Howard, J., et al. Infectious sequelae after endoscopic sclerotherapy of oesophageal
varices: Role of antibiotic prophylaxis. J. Hepatol. 18:290, 1993.
98. Reynolds, T. B. Therapeutic paracentesis: Have we come full circle? Gastroenterology 93:386, 1987.
99. Guazzi,
M., Polese, A., Magrini, F., et al. Negative influences of ascites on the cardiac function of cirrhotic patients.
Am. J. Med. 59:165, 1975.
100. Tito,L., Gines, P., Arroyo, V., et al. Total paracentesis associated with intravenous albumin management of patients
with cirrhosis and ascites. Gastroenterology 98:146, 1990.

101. Mirouze,D., Juttner, H. U., and Reynolds, T. B. Left pleural effusion in patients with chronic liver disease and ascites:
Prospective study of 22 cases. Dig. Dis. Sci. 26:984, 1981.
102. Strauss,
R. M., Martin, L. G., Kaufman, S. L., et al. Transjugular intrahepatic portal systemic shunt for the
management of symptomatic cirrhotic hydrothorax. Am. J. Gastroenterol. 89:1520, 1994.
103. Xiol,
X., Castellote, J., Baliellas, C., et al. Spontaneous bacterial empyema in cirrhotic patients: Analysis of eleven
cases. Hepatology 11:365, 1990.
104. Runyon, B. A., Greenblatt, M., and Ming, R. H. C. Hepatic hydrothorax is a relative contraindication to chest tube
insertion. Am. J. Gastroenterol. 81:566, 1986.
105. Belghiti, J., Rueff, B., and Fekete, F. Umbilical hernia in cirrhotic patients with ascites. Gastroenterology 84:1363,
1983.
106. Runyon,B. A., and Juler, G. L. Natural history of umbilical hernias in patients with and without ascites. Am. J.
107. Oktedalen, O., Nygaard, K., and Osnes, M. Somatostatin in the treatment of pancreatic ascites. Gastroenterology
99:1520, 1990.
108. Runyon, B. A. Care of patients with ascites. N. Engl. J. Med. 330:337, 1994.
109. Eisenmenger, W. J., Blondheim, S. H., Bongiovanni, A. M., and Kunkel, H. G. Electrolyte studies on patients with
cirrhosis of the liver. J. Clin. Invest. 29:1491, 1950.
110. Perez-Ayuso, R. M., Arroyo, V., Planas, R., et al. Randomized comparative study of efficacy of furosemide vs.
spironolactone in nonazotemic cirrhosis with ascites. Gastroenterology 84:961, 1983.
111. Sungaila,I., Bartle, W. R., Walker, S. E., et al. Spironolactone pharmacokinetics and pharmacodynamics in patients
with cirrhotic ascites. Gastroenterology 102:1680, 1992.
112. Stanley,
M. M., Ochi, S., Lee, K. K., et al. Peritoneovenous shunting as compared with medical treatment in patients
with alcoholic cirrhosis and massive ascites. N. Engl. J. Med. 321:1632, 1989.
113. Daskalopoulos, G., Laffi, G., Morgan, T., et al. Immediate effects of furosemide on renal hemodynamics in chronic
liver disease with ascites. Gastroenterology 92:1859, 1987.
114. Pockros,P. J., and Reynolds, T. B. Rapid diuresis in patients with ascites from chronic liver disease: The importance of
peripheral edema. Gastroenterology 90:1827, 1986.
115. Mirouze, D., Zipser, R. D., and Reynolds, T. B. Effect of inhibitors of prostaglandin synthesis on induced diuresis in
cirrhosis. Hepatology 3:50, 1983.
116. Runyon, B. A., Antillon, M. R., and Montano, A. A. Effect of diuresis versus therapeutic paracentesis on ascitic fluid
opsonic activity and serum complement. Gastroenterology 97:158, 1989.
117. Gines,P., Arroyo, V., Quintero, E., et al. Comparison of paracentesis and diuretics in the treatment of cirrhotics with
tense ascites: Results of a randomized study. Gastroenterology 93:234, 1987.
118. Kao,
H. W., Rakov, N. E., Savage, E., et al. The effect of large volume paracentesis on plasma volume--A cause of
hypovolemia? Hepatology 5:403, 1985.

119. Pinto, P. C., Amerian, J., and Reynolds, T. B. Large-volume paracentesis in nonedematous patients with tense ascites:
Its effect on intravascular volume. Hepatology 8:207, 1988.
120. Gines,
P., Tito, L., Arroyo, V., et al. Randomized comparative study of therapeutic paracentesis with and without
intravenous albumin in cirrhosis. Gastroenterology 94:1493, 1988.
121. Gines,
P., Fernandez-Esparrach, G., Monescillo, A., et al. Randomized trial comparing albumin, dextran 70, and
polygeline in cirrhotic patients with ascites treated by paracentesis. Gastroenterology 111:1002, 1996.
122. Rothschild, M., Oratz, M., Evans, C., et al. Alterations in albumin metabolism after serum and albumin infusions. J.
Clin. Invest. 43:1874, 1964.
123. Wilkinson, P., and Sherlock, S. The effect of repeated albumin infusions in patients with cirrhosis. Lancet ii:1125,
1962.
124. Pietrangelo,
A., Panduro, A., Chowdhury, J. R., et al. Albumin gene expression is down-regulated by albumin or
macromolecule infusion in the rat. J. Clin. Invest. 89:1755, 1992.
125. Antillon,
M. R., and Runyon, B. A. Postparacentesis plasma expansion prevents asymptomatic laboratory
abnormalities but does it have any impact on morbidity or mortality? Gastroenterology 101:1455, 1991.
126. Moller, S., Bendtsen, F., and Henriksen, J. H. Effect of volume expansion on systemic hemodynamics and central and
arterial blood volume in cirrhosis. Gastroenterology 109:1917, 1995.
127. Planas, R., Gines, P., Arroyo, V., et al. Dextran-70 versus albumin as plasma expanders in cirrhotic patients with tense
ascites treated with total paracentesis. Gastroenterology 99:1736, 1990.
128. Bacq,Y., Gaudin, C., Hadengue, A., et al. Systemic, splanchnic and renal hemodynamic effects of a dopaminergic
dose of dopamine in patients with cirrhosis. Hepatology 14:483, 1991.
129. Ochs, A., Rossle, M., Haag, K., et al. The transjugular intrahepatic portosystemic stent-shunt procedure for refractory
ascites. N. Engl. J. Med. 332:1192, 1995.
130. Shiffman,
M. L., Jeffers, L., Hoofnagle, J. H., et al. The role of TIPS for treatment of portal hypertension and its
complications. Hepatology 22:1591, 1995.
131. Bories,
P., Garcia-Compean, D., Michel, H., et al. The treatment of refractory ascites by the LeVeen shunt: A
multi-center controlled trial (57 patients). J. Hepatol. 3:212, 1986.
132. Gines,
P., Quintero, E., Arroyo, V., et al. Compensated cirrhosis: Natural history and prognostic factors. Hepatology
7:12, 1987.
133. D'Amico,
G., Morabito, A., Pagliaro, L., and Marubini, E. Survival and prognostic indicators in compensated and
decompensated cirrhosis. Dig. Dis. Sci. 31:468, 1986.


April 22, 2000
Conde Petra
Goroll: Primary Care Medicine, 3rd ed., Copyright

1995 Lippincott-Raven Publishers
Part 5 - Gastrointestinal Problems
71 - Management of Cirrhosis and Goroll: Primary Care Medicine, 3rd ed., Copyright 1995
Lippincott-Raven Publishers
Chronic Liver Failure
Introduction
CLINICAL PRESENTATION AND

COURSE Chapter 71 - Management of
Clinical presentation Cirrhosis and
Clinical Sequelae and Chronic Liver Failure
Prognosis.
PRINCIPLES OF MANAGEMENT The best treatment is prevention, with emphasis on reducing

alcohol consumption (see Chapter 228) , limiting occupational
Cirrhosis and Its Underlying hepatotoxin exposure, and preventing parenteral transmission of
Etiologies hepatitis B and C (see Chapters 57 and 58) . The overall goal is
Alcoholic cirrhosis to minimize hepatocellular injury and the risk of progressing to
chronic liver failure and cirrhosis.
Cirrhosis represents an irreversible state of chronic liver injury.
Secondary biliary cirrhosis However, the cirrhotic patient may be kept comfortable, active,
Hemochromatosis and independent if precipitants of hepatocellular injury can be
eliminated and complications can be prevented. The
Wilson's disease responsibility for long-term management of patients with
Chronic active hepatitis cirrhosis and chronic liver failure often rests with the primary
care physician, who needs to be capable of dealing with such
Complications potential difficulties as ascites, peripheral edema,
encephalopathy, infection, bleeding, renal dysfunction, and
Ascites and edema
electrolyte imbalances.
Variceal Bleeding.
CLINICAL PRESENTATION AND COURSE
Coagulopathy Clinical presentation
Role of Liver may be rather dramatic when ascites, encephalopathy, or brisk
Transplantation. variceal bleeding brings the patient to medical attention. More
subtle manifestations include splenomegaly, a firm liver edge,
or such signs of hepatocellular failure as jaundice, palmar
http://home.mdconsult.com/das/book/view/106/80.ht...&stty=cow&nid=39985&tlt=Z&pos=&tag=relinfo&anc=bo (1 of 17) [22/04/2000 0:19:41]
THERAPEUTIC
erythema, Dupuytren's contractures, spider angiomata, parotid
RECOMMENDATIONS AND
MONITORING
and lacrimal gland hypertrophy, gynecomastia, testicular
atrophy, loss of axillary and pubic hair, and clubbing.
General Measures Hypogonadism and feminization are particularly prominent in
male patients with cirrhosis due to alcoholism or
Management of Ascites
hemochromatosis. Most findings are nonspecific and do not
Encephalopathy reflect the underlying etiology. More specific is the bronze
appearance of the skin in patients with hemochromatosis or the
Prevention of Variceal Bleeding appearance of antimitochondrial antibodies (anti-M2) in those
and Bleeding due to Clotting
with primary biliary cirrhosis. Nonspecific abnormalities in
Factor Deficiency
routine liver function tests are common, but the best measures
INDICATIONS FOR REFERRAL of hepatocellular function are the serum albumin concentration
AND ADMISSION and prothrombin time (PT). The PT is the first to become
abnormal because of the short serum half-life (as little as 7
PATIENT EDUCATION AND
days) of the clotting factors that determine it.
SUPPORT
Bibliography Clinical Sequelae and Prognosis.
Portal hypertension, fluid retention, and encephalopathy are the

major sequelae of cirrhosis; they lead to varices, ascites,
hypersplenism, peripheral edema, and altered mental status.
Variceal bleeding occurs in 20 percent to 30 percent of all
cirrhotic patients, one-third of whom die during the initial
hospitalization, one-third rebleed within 6 weeks, and one-third
survive 1 year or more. The principal causes of death in patients
with cirrhosis are variceal bleeding, encephalopathy, and
infection. In addition, patients with cirrhosis, especially those
with chronic hepatitis B infection, are at increased risk for
hepatocellular carcinoma.
Prognosis is determined by the nature, severity, and activity of
the underlying illness. For example, continued alcohol
consumption in the context of alcoholic hepatitis is associated
with an 80 percent chance of developing cirrhosis, whereas
abstinence lowers the risk to 15 percent. Even after alcoholic
cirrhosis has developed, survival continues to be affected by
alcohol ingestion. Five-year survival is 60 percent to 85 percent
in those who abstain, compared with 40 percent to 60 percent
for those who continue to drink. Onset of jaundice or ascites
further decreases 5-year survival (to 30% in drinkers). A
discriminant function (DF) using the serum bilirubin and
prothrombin time (DF = 4.6[PT - control] + bilirubin [mg per
100 mL]) has been developed to predict survival. A DF greater
than 32 indicates a short-term mortality risk of 35 percent. In a
recent study of patients with symptomatic primary biliary
cirrhosis, the average length of survival from the onset of
symptoms was about 12 years, whereas the survival of
asymptomatic patients did not differ from that of a control

population matched for age and sex. The prognosis of patients
with postnecrotic cirrhosis is difficult to assess because it is
hard to date its onset; the cirrhosis develops insidiously over
years from subclinical chronic active hepatitis.
Irrespective of etiology, development of ascites,
encephalopathy, hyperbilirubinemia, hypergammaglobulinemia
(from bypass of the hepatic reticuloendothelial system), and
hypoalbuminemia are poor prognostic signs, as is decreased
liver size. Worsening renal function is associated with a 33
percent 2-year mortality.
PRINCIPLES OF MANAGEMENT
Cirrhosis and Its Underlying Etiologies
Alcoholic cirrhosis
requires complete abstinence from further alcohol intake,

because prognosis is markedly worsened
408
by continued drinking (see above). Attention to good nutrition,

daily multiple vitamin supplements (including 1 mg of folic
acid), and correction of any iron deficiency or electrolyte
deficits are important supportive measures. The search
continues for agents that might halt hepatic fibrosis and
promote hepatocyte regeneration. Glucocorticosteroids and
portacaval shunting have failed to demonstrate an improvement
in survival, although corticosteroids do improve short-term
survival in patients with acute alcoholic hepatitis. A
randomized, placebo-controlled, long-term study of colchicine
revealed a doubling of 5- and 10-year survival.The drug is an
inhibitor of collagen deposition. However, confirmatory data
are needed before it or other drugs found experimentally to
improve survival ( e.g., propylthiouracil ) can be recommended.
Liver transplantation is sometimes considered in patients who
have become totally and permanently abstinent.
can cause severe pruritus, which may be relieved by

cholestyramine in a dose of 4 g orally with meals. Because of
decreased fat absorption from low intestinal bile-salt
concentrations, these patients are particularly prone to develop

deficiencies of the fat-soluble vitamins. They may require

supplemental vitamin K (10 mg subcutaneously [SC] every 4
weeks), vitamin D (50,000 U orally two to three times a week
or 100,000 U intramuscularly every 4 weeks) with oral calcium
(1 g daily), and vitamin A (25,000 U orally per day). Night
blindness unresponsive to vitamin A may be due to zinc
deficiency, which is treated with oral zinc sulfate (220 mg per
day). For patients with steatorrhea, medium-chain triglyceride
preparations often help. Azathioprine and colchicine may have a
beneficial effect on survival and are well tolerated.
Corticosteroids and penicillamine do not improve survival and
cause serious adverse side effects. Liver transplantation has
proven to be a viable option.
can be halted by relieving or bypassing the obstruction to bile

flow.
Hemochromatosis
is treated with weekly phlebotomies of 500 mL until the serum

iron and ferritin levels fall to normal; then they are performed
as needed.
Wilson's disease
responds to d -penicillamine therapy, which should be

administered in conjunction with a hepatologist experienced in
using the drug, because of its potential for serious side effects.
Chronic active hepatitis
may benefit from corticosteroid therapy or interferon alfa,

depending on the etiology (see Chapter 70) .
Complications
Ascites and edema
result from increased portal pressure, hypoalbuminemia,

secondary hyperaldosteronism, and impaired free water
clearance. Its presence is strongly suggested by the clinical
findings of a fluid wave, shifting dullness, and peripheral
edema. Abdominal ultrasound can be used for confirmation and
to rule out venoocclusive disease involving the hepatic veins or
the hepatic region of the inferior vena cava (Budd-Chiari
syndrome). Although ascites is not a hazard, gross ascites can

cause abdominal discomfort and respiratory compromise; under

such circumstances, it ought to be treated.
Diagnostic paracentesis is indicated before treatment is
initiated in patients with new onset of ascites, worsening hepatic
function, fever, or increasing encephalopathy to exclude
infection and malignancy. Although it used to be felt that
differentiating exudative from transudative ascitic fluid could
aid in identifying infection and malignancy, such a
differentiation has proven insufficiently sensitive and specific to
be reliable. Fluid protein concentrations in excess of 2.5 g per
deciliter are seen not only in conditions causing exudates, but
also in patients with transudative processes subjected to
diuresis. Moreover, the ascitic fluid protein concentration of
patients with spontaneous bacterial peritonitis (SBP) is typically
less than 1 g per deciliter. However, the serum-to-ascites
albumin gradient (serum albumin concentration minus the
ascites albumin concentration) can help differentiate portal
hypertension from other causes of ascites, especially when the
ascitic protein concentration is high. A gradient of greater than
1.1 mg per deciliter is indicative of portal hypertension; ascites
in the presence of a gradient less than 1.1 suggests a mechanism
other than portal hypertension. Ascitic fluid should also be sent
for cytologic examination, cell count, and culture. A leukocyte
count in excess of 2500 per milliliter is strongly suggestive of
SBP in the patient with abdominal pain and fever. Culture is
positive in over 80 percent of cases when 10 mL aliquots are
injected into three blood culture bottles; Gram's stain is usually
negative.
Management of ascites due to portal hypertension begins with
reduction in sodium intake. Prescribing a 2 g sodium diet is a
reasonable compromise between maximizing sodium restriction
and dietary palatability. Adequate nutrition is critical. In the
absence of encephalopathy, a daily protein intake of at least 50
g is recommended. Excessive water intake should be prohibited,
and free water should be restricted to 1500 mL per day if
hyponatremia ensues. An effective program of salt and water
restriction requires a cooperative patient and a conscientious
family. A dietitian can provide invaluable assistance. The
patient should be instructed to check his or her weight daily.
Measuring abdominal girth is an unreliable index of fluid loss
because of variations due to gaseous distention of the
gastrointestinal (GI) tract. About 15 percent of patients will
respond to sodium and fluid restrictions alone. Bedrest is of no
added benefit.

Diuretic therapy is indicated if a diuresis has not occurred

spontaneously after a full week of salt restriction.
Spironolactone is the agent of first choice, because it inhibits
the hyperaldosteronism of portal hypertension. Being a specific
aldosterone antagonist, it counters the hypokalemic alkalosis
commonly seen in cirrhotic patients with ascites. Its diuretic
action is mild and unlikely to cause rapid intravascular volume
depletion. The initial dosage of spironolactone is 100 mg a day
orally in divided doses. If diuresis does not follow within 1
week, the daily dosage may be increased by 100 mg every 4 or
5 days to a maximum of 400 mg daily (higher doses may cause
hyperkalemic acidosis). It is useful to monitor urinary
electrolyte concentrations, because diuresis should follow a
significant rise in urinary sodium and a fall in urinary
potassium. If natriuresis and diuresis do not
409
occur on a maximal dosage of spironolactone, a loop diuretic

such as furosemide (starting at 20 to 40 mg per day) may be
added. Bumetanide or metolazone may be added to the diuretic
program, but such potent diuretics must be used with extreme
care to avoid precipitating renal failure, hypokalemia, and
encephalopathy.
Monitoring BUN, creatinine, and electrolytes is critical during
treatment of ascites. However, some cirrhotic patients with
renal impairment do not manifest an elevated serum creatinine
(believed related to reduced creatine synthesis by the liver).
This reduces the sensitivity of the serum creatinine in some
cirrhotic patients and can complicate monitoring of their renal
status. Other measures of renal function may be required.
The maximum amount of ascitic fluid that can be mobilized in
24 hours is about 700 to 900 mL, although peripheral edema
can be mobilized at a faster rate. This translates to a daily
weight loss of 0.5 kg in patients with ascites alone and of 1 kg
in those with both ascites and peripheral edema. Daily weight
loss in excess of these amounts suggests overdiuresis, with its
attendant risk of intravascular volume depletion leading to
hepatorenal syndrome and encephalopathy. A falling urine
output accompanied by orthostatic signs (rise in pulse and fall
in blood pressure on change from supine to standing) suggest
inadequate intravascular volume. An oral fluid challenge of
several hundred milliliters of isotonic fluid can confirm the
presence of hypovolemia by inducing a temporary increase in
urine output.

Some patients with incapacitating ascites are truly refractory to

sodium restriction and diuretics. For them, large-volume
paracentesis and peritoneovenous (LeVeen) shunting become
therapeutic considerations. Removal of more than 1 to 2 L of
ascitic fluid at a time by paracentesis used to be considered
unwise, because of the risk of precipitating serious intravascular
volume depletion from the shift of intravascular fluid into the
emptied peritoneal cavity. Moreover, protein depletion and
increase in renin activity and aldosterone secretion might ensue.
By infusing intravenous albumin (6 to 8 g per liter of ascitic
fluid removed) at the time of paracentesis, large volumes (5 to 6
L) of ascitic fluid can be removed without precipitating adverse
reactions. Results are comparable to those achieved by shunting
procedures, although admissions for reaccumulation of fluid are
more frequent (every 10 to 30 days). The best candidates are
those with peripheral edema and relatively well-preserved renal
function.
Before the development of large-volume paracentesis,
peritoneovenous (LeVeen) shunting was the principal means of
dealing with refractory ascites. Although effective in treating
ascites, shunt placement does not improve survival. Moreover,
insertion can precipitate disseminated intravascular coagulation.
Infection and shunt obstruction are common complications.
Compared to large-volume paracentesis, LeVeen shunting
requires fewer readmissions for treatment of ascites, but an
overall similar number of hospitalizations due to shunt
occlusions. Its potentially serious complications and the advent
of large-volume paracentesis has led to a decline in use of
LeVeen shunts.
is thought to be produced by one or more intestinally derived

toxic substances that escape hepatic detoxification as a result of
portasystemic shunting and hepatocellular dysfunction.
Candidates include ammonia, benzodiazepine-like substances,
mercaptans, phenol, neuroinhibitors, phenol, and short-chain
fatty acids. Elevations of arterial and venous ammonia levels
usually, but not always, correlate with the presence of hepatic
encephalopathy. Venous levels may be falsely elevated when a
tourniquet is left on too long at the time of blood drawing.
Ammonia levels are useful in following the clinical state of
individual patients. Important precipitating factors include GI
bleeding, excessive dietary protein intake, hypokalemic
alkalosis, infection, constipation, use of sedative or hypnotic

drugs, surgical procedures, and volume depletion resulting from

diuresis or paracentesis. Precipitants are identified in about 50
percent of patients; the prognosis is usually better in those with
an identifiable contributory factor than in those in whom the
onset of encephalopathy is associated only with worsening
hepatic function.
Mild encephalopathy may be managed on an ambulatory basis.
Aside from excluding GI bleeding, avoiding and correcting
fluid and electrolyte disturbances, and discontinuing
tranquilizers and sedatives, the mainstay of therapy is
restriction of dietary protein intake to 30 to 40 g per day, while
maintaining a daily caloric intake of 1500 kcal. Vegetable
sources of protein are preferred over animal proteins because
their metabolism produces less ammonia. Sometimes, oral
amino acid supplements (high in branched-chain varieties and
low in aromatic amines) are added to prevent negative nitrogen
balance. Their efficacy in treatment of encephalopathy appears
to be transient.
Simple gut cleansing with enemas or cathartics is effective
when bleeding, constipation, or a large dietary protein intake
has led to encephalopathy. Patients should be monitored with
routine mental status examinations that include five-point star
and signature testing and examination for asterixis.
Specific therapy for encephalopathy begins with lactulose, a
synthetic, nonabsorbable disaccharide, which is metabolized to
organic acids by enteric bacteria, causing an osmotic catharsis.
In addition, lactulose suppresses the growth of
ammonia-forming bacteria in favor of lactose-fermenting
organisms. The initial dosage for patients with mild
encephalopathy is 15 to 30 mL orally every 4 to 6 hours, with
adjustments thereafter to produce two to three loose stools a
day. Side effects of oral lactulose include diarrhea and
abdominal distress, which usually resolve after a reduction in
dosage.
Lactulose is as effective as neomycin, a poorly absorbed,
broad-spectrum aminoglycoside antibiotic, which predated
lactulose and is more toxic. It acts by decreasing the intestinal
concentration of ammonia-forming bacteria. Prolonged use of
doses of 4 g per day or more, especially in patients with renal
insufficiency, may lead to ototoxicity and nephrotoxicity. The
recommended maximum oral dose is 1 g twice daily. The drug
also causes malabsorption. In patients who fail to respond to
either lactulose or neomycin, the two may be given together; in
some patients, the effect appears to be additive. Metronidazole
may be used instead of neomycin; it is less toxic.
Variceal Bleeding.
Primary and secondary prophylaxis against bleeding are the

principal concerns of the primary
410
care physician. In patients with known varices, risk of bleeding

is high. Independent risk factors include marked hepatocellular
dysfunction, ascites, encephalopathy, large varices, and
presence of dilated venules on the varices. Risk of a first bleed
is as high as 65 percent in the first year. Because clinical and
endoscopic risk factors are independent and therefore not
predictive of one another, some authorities recommend
endoscopic assessment of all cirrhotic patients to determine risk
and candidacy for treatment.
Patients judged to be at high risk are reasonable candidates for
beta-blocker therapy . When prescribed in dosages that produce
beta-blockade (a reduction in heart rate of about 25%), these
agents can lower portal venous pressure and decrease the risk of
variceal bleeding by about 50 percent. The rate of death from
hemorrhage is also reduced. Although disappointing for
secondary prophylaxis, beta-blockers have demonstrated
efficacy for prevention of a first variceal bleed and should be
considered for use in high-risk patients. Other agents are being
investigated for their ability to further reduce portal pressure.
Initial results with clonidine have been promising.
Injection sclerotherapy (involving endoscopic injection of
sclerosing agents) is used to control acute variceal bleeding and,
when repeated several times over 2 to 3 months, to prevent
recurrences. Controlled trials have shown it to be superior to
placebo in preventing repeat variceal hemorrhage, but having
little impact on the risk of all-cause upper GI bleeding, which
remains at close to 40 percent. Unresolved is its ability to
prevent first bleeding. Like portasystemic shunt surgery (see
below), sclerotherapy may only substitute one form of
morbidity and mortality for another. Sclerotherapy has largely
replaced portasystemic shunt surgery, which is now reserved for
those who bleed despite repeated sclerotherapies.
Portasystemic shunt surgery remains the final treatment option
for patients with recurrent variceal hemorrhage due to portal
hypertension. While reducing the risk hemorrhage, shunt
surgery increases the risk of encephalopathy. The likelihood of
recurrent variceal bleeding is significantly reduced after

portasystemic shunting, but long-term survival is not. Reasons
for lack of improvement in survival include high operative
mortality rates and no benefit to hepatic function, the principal
determinant of survival. Consequently, portasystemic shunt
surgery is not indicated in patients with varices who have never
bled. Selective shunt procedures were developed to reduce the
risk of encephalopathy, a common consequence of shunt
surgery. For example, the selective distal splenorenal shunt
attempts to preserve portal blood flow in patients with
demonstrable preoperative portal perfusion. Its incidence of
postoperative encephalopathy is lower than with conventional
portacaval shunting, but long-term benefit is reduced as
collateral channels develop. Side-to-side portacaval or proximal
splenorenal shunts are the most effective operations for
decompressing hepatic sinusoids and reducing ascites, but they
still risk precipitating encephalopathy. For patients awaiting
liver transplantation, the transjugular intrahepatic
portasystemic shunt has proven safe and effective.
Coagulopathy
results from reductions in vitamin K-dependent clotting factors

(II, VII, IX, and X) secondary to decreased hepatic protein
synthesis and increased plasma proteolytic activity. In addition,
bile-salt deficiency, neomycin therapy, and malnutrition may
contribute to malabsorption of vitamin K, and hypersplenism
may account for thrombocytopenia. If a patient with cirrhosis is
discovered to have a prolonged prothrombin time, a trial of
vitamin K 10 mg subcutaneously daily for 3 days will correct
hypoprothrombinemia caused by bile-salt deficiency, neomycin,
or malnutrition but not hypoprothrombinemia related only to
hepatocellular disease. In the absence of bleeding, measures to
correct abnormal coagulation parameters are generally not
indicated.
Role of Liver Transplantation.
In the face of terminal hepatocellular failure, liver

transplantation becomes a consideration. In properly selected
patients, 5-year survival may be as high as 85 percent. The best
candidates are those who are highly motivated, emotionally
stable, and willing to comply with a medical program.
Postnecrotic cirrhosis, primary biliary cirrhosis, and primary
sclerosing cholangitis are among the primary indications.
Alcohol-induced liver disease is a relative contraindication, as
is hepatitis B liver disease, hepatocellular carcinoma, and renal

failure. Absolute contraindications are acquired

immunodeficiency syndrome (AIDS), extrahepatic sepsis,
metastatic cancer, and severe cardiopulmonary disease.
THERAPEUTIC RECOMMENDATIONS
AND MONITORING
General Measures
The patient should maintain a caloric intake of at least

2000 to 3000 kcal per day.
Use of alcohol or other hepatotoxic agents must be
prohibited.
The patient should avoid tranquilizers and sedatives.
Monitor prothrombin time, serum albumin, and
bilirubin to assess the severity and progression of
hepatocellular dysfunction.
Check stools at each visit for evidence of occult
bleeding.
Check for asterixis and other signs of encephalopathy
at each visit.
Check the abdomen for evidence of ascites (shifting
dullness, fluid wave, bulging flanks). Ultrasound
examination is useful to confirm the presence of ascites
and rule out venoocclusive disease.
Perform a diagnostic paracentesis in patients with the

new onset of ascites or clinical deterioration in the setting
of preexisting ascites. The fluid should be sent for cell
count and differential, total protein and albumin
concentrations, culture, and cytologic examination.
Instruct patients with ascites to restrict daily sodium
intake to no more than 2 g and to consume at least 50 g of
protein per day. Consult with a dietitian and provide
patient and family with specific menus and food lists.
Restrict fluid intake to 1500 mL when there is marked
hyponatremia (serum sodium concentration less than 125
mEq per liter).
411
If salt restriction does not result in diuresis, begin

spironolactone 100 mg daily in divided doses. If

natriuresis and diuresis do not occur after 1 week,
increase the daily dose of spironolactone by 100 mg
every 4 to 5 days to a maximum of 400 mg per day.
If spironolactone alone is ineffective in causing
diuresis, add furosemide 20 to 40 mg per day to the
regimen and cautiously increase dosage as necessary.
Adjust diuretic dose so that no more than 0.5 kg of
fluid (approximately 1 lb) is lost per day in patients with
ascites alone, and no more than 1 kg per day (2 lb) in
those with both ascites and peripheral edema. Halt
diuretics at the first sign of intravascular volume
depletion.
Consider daily potassium supplementation (20 to 40
mEq KCl elixir) in patients receiving furosemide;
administer cautiously, if at all, to patients concurrently
taking a potassium-sparing diuretic such as
spironolactone.
Monitor serum potassium, BUN, creatinine, daily
weight, and postural signs to avoid inducing intravascular
volume depletion, renal failure, hypokalemia, and
encephalopathy. Be aware that, in some patients, the
serum creatinine may be falsely normal and remain
within normal limits despite worsening renal function.
Consider large volume paracentesis (5 to 6 L) with
concurrent intravenous albumin infusion (6 to 8 g per
liter of fluid removed) for patients with refractory ascites
that is disabling. Admit for the procedure.
Encephalopathy
At the first sign of encephalopathy, restrict dietary

protein intake to 20 to 30 g per day. Obtain dietary
consultation to construct a diet emphasizing plant protein
over animal protein. Consider use of an oral supplement
rich in branched-chain amino acids if protein intake is
insufficient.
Monitor mental status and check for asterixis; use
five-point star or signature testing. Monitor venous
ammonia levels; in drawing blood for a determination,
avoid prolonged tourniquet application.
When protein restriction fails to control
encephalopathy, begin oral lactulose, 15 to 30 mL every
4 to 6 hours, with subsequent adjustments in the dosage
to allow two to three soft stools a day. Add oral
neomycin 1 g bid or metronidazole 250 mg tid if

lactulose alone does not suffice.
Prevention of Variceal Bleeding and Bleeding due to

Clotting Factor Deficiency
Begin a beta-blocker (eg, propranolol 80 mg per day)

for primary prevention in patients with risk factors for
variceal bleeding (marked hepatocellular dysfunction,
ascites, encephalopathy, large varices, and presence of
dilated venules on the varices).
Consider sclerotherapy and shunt procedures for
prevention of recurrent variceal bleeding.
Monitor prothrombin time and platelet count.
Administer vitamin K (10 mg SC daily for 3 days) if
there is prolongation of prothrombin time due to
drug-induced bile-salt malabsorption, neomycin, or
malnutrition. Platelet transfusions are unwarranted unless
there is active bleeding in the context of a very low
platelet count (see Chapter 81) .
INDICATIONS FOR REFERRAL AND

ADMISSION
Prompt hospitalization is required for patients with GI bleeding,
worsening encephalopathy, increasing azotemia, signs of
peritoneal irritation, or unexplained fever. Intractable ascites
may respond to elective admission for large-volume
paracentesis. Decisions about the management of refractory
ascites, encephalopathy, variceal bleeding, and uncommon
etiologies of cirrhosis (eg, primary biliary cirrhosis, Wilson's
disease, hemochromatosis) are best made in consultation with a
gastroenterologist skilled in treating liver disease. The same
pertains to candidacy for liver transplantation. When urine
output falls in the absence of a clear-cut explanation, a
nephrologic consultation may be of considerable help,
especially because creatinine level may not adequately reflect
renal function.
PATIENT EDUCATION AND SUPPORT

It should be emphasized to the patient and family that prognosis
can often be greatly improved and symptoms lessened by
careful adherence to the prescribed medical program. In
particular, dietary discipline and omission of alcohol are central
to a successful outcome and should be stressed. Many of these
patients are chronic alcoholics with low self-esteem. A

nonjudgmental, sympathetic physician can be instrumental in
providing support, raising self-esteem, and improving the
chances of compliance (see Chapter 228) . Depression is a
frequent accompaniment of the later stages of chronic liver
disease and is manifested by failure to comply with the medical
regimen and outright expressions of wanting to die. Treatment
is very difficult. Antidepressant drugs may cause oversedation
and thus are risky. There are no simple measures, but the
physician's concern and support can help enormously (see
Chapter 227) .
A.H.G.
ANNOTATED BIBLIOGRAPHY
Black M, Friedman AC. Ultrasound examination in the patient
with ascites. Ann Intern Med 1989;110:253. (Editorial urging
its routine use in patients with new onset of ascites, both for its
confirmation and to rule out a venoocclusive etiology.)
Borowsky SA, Strome S, Lott E. Continued heavy drinking and
survival in alcoholic cirrhotics. Gastroenterology 1981;80:1405.
(80 percent of those who continued to drink heavily died an
average of 7.2 months after discharge; 95 percent of abstainers
were still alive at 14 months.)
Boyer JL, Ransohoff DF. Is colchicine effective therapy for
cirrhosis? N Engl J Med 1988;318:1751. (Urges caution in
interpreting the very encouraging data that are emerging; see
Kershenobich below.)
Campra JL, Reynolds TB. Effectiveness of high-dose
spironolactone therapy in patients with chronic liver disease and
relatively refractory ascites. Dig Dis Sci 1978;23:1025. (Doses
as high as 600 mg per day found effective in patients with
ascites presumed to be "refractory.")
Christensen E, Neuberger J, Crowe J, et al. Beneficial effect of
azathioprine and prediction of prognosis in primary biliary
cirrhosis final results of an international trial. Gastroenterology
1985;89:1084. (Improved survival demonstrated.)
412
Conn HO, Leevy CM, Vlahcevic ZR, et al. Comparison of

lactulose and neomycin in the treatment of chronic

portal-systemic encephalopathy; a double-blind controlled trial.

Gastroenterology 1977;72:573. (Found to be equally effective
and free of significant toxicity at the doses used.)
Epstein M. Treatment of refractory ascites. N Engl J Med
1989;321:1675. (Editorial summarizing data on LeVeen
shunting and large volume paracentesis; argues that
paracentesis may be preferable.)
Fraser CL, Arieff AI. Hepatic encephalopathy. N Engl J Med
1985;313:865. (Comprehensive review; 147 references.)
Gines P, Arroyo V, Vargas V, et al. Paracentesis with
intravenous infusion of albumin as compared with
peritoneovenous shunting in cirrhosis with refractory ascites. N
Engl J Med 1991;325:829. (Randomized comparison showing
equal effectiveness and similar total hospital days.)
Graham DY, Smith JL. The course of patients after variceal
hemorrhage. Gastroenterology 1981;80:800. (One-third died in
the initial hospitalization, one-third rebled within 6 weeks, and
one-third survived at least 1 year. Long-term survival after the
initial 2 weeks no different from that of unselected cirrhotics
without bleeding.)
Kaplan MM. Primary biliary cirrhosis. N Engl J Med
1987;316:521. (Authoritative and very useful review for the
generalist reader; 124 references.)
Kershenobich D, Vargas F, Garcia-Tsao G, et al. Colchicine in
the treatment of cirrhosis of the liver. N Engl J Med
1988;318:1709. (Randomized, placebo-controlled study; 14
years of follow-up; survival doubled.)
North Italian Endoscopic Club. Prediction of the first variceal
hemorrhage in patients with cirrhosis of the liver and
esophageal varices. N Engl J Med 1988;319:983. (Multivariate
analysis identifying risk factors predictive of bleeding.)
Orrego H, Blake JE, Blendis LM, et al. Long-term treatment of
alcoholic liver disease with propylthiouracil. N Engl J Med
1987;317:1421. (Long-term, double-blind, randomized trial
demonstrating reduction in mortality; about 50 percent of
patients had cirrhosis.)
Pagliaro L, D'Amico G, Sorensen TIA, et al. Prevention of first
bleeding in cirrhosis: A meta-analysis of randomized trials of
nonsurgical treatment. Ann Intern Med 1992;117:59.
(Beta-blockers prevent first bleeding in high-risk patients;
sclerotherapy is of unproven efficacy for primary prophylaxis.)
Papadakis MA, Arieff AI. Unpredictability of clinical

evaluation of renal function in cirrhosis. Am J Med
1987;82:945. (Some cirrhotics with reduced glomerular
filtration have normal serum creatinine levels, which may
remain within normal limits despite worsening of renal
function.)
Powell WJ Jr, Klatskin G. Duration of survival in patients with
Laennec's cirrhosis. Am J Med 1968;44:406. (Classic paper
showing improved survival in cirrhotic patients who abstained
from alcohol compared with those who continued to drink
heavily.)
Poynard T, Cales P, Pasta L, et al. Beta-adrenergic antagonist
drugs in the prevention of gastrointestinal bleeding in patients
with cirrhosis and esophageal varices. N Engl J Med
1991;324:1532. (Analysis of pooled data showing reduction in
risk of first bleeding and mortality from hemorrhage.)
Rector WG, Jr, Reynolds TB. Superiority of serum-ascites
albumin differences in "exudative" ascites. Am J Med
1984;77:83. (Better than total protein for differential
diagnosis.)
Rikkers LF. Operations for management of esophageal variceal
hemorrhage. West J Med 1982;136:107. (Critical review of the
various types of portasystemic shunt procedures and the
indications for their use.)
Ring EJ, Lake JR, Roberts JP, et al. Using transjugular
intrahepatic portosystemic shunts to control variceal bleeding
before liver transplantation. Ann Intern Med 1992;116:304.
(Prospective, uncontrolled trial; found safe and effective.)
Shear L, Ching S, Gabuzda GJ. Compartmentalization of ascites
and edema in patients with hepatic cirrhosis. N Engl J Med
1970;282:1391. (Classic, elegant study; maximum rate at which
ascites can be absorbed was 930 mL per 24 hours; concludes
that the therapeutic aim in patients with ascites should be
weight loss of no more than 0.5 kg daily.)
Stanley MM, Ochi S, Lee KK, et al. Peritoneovenous shunting
as compared with medical treatment in patients with alcoholic
cirrhosis and massive ascites. N Engl J Med 1989;321:1632.
(Ascites alleviated, but survival unchanged.)
Starzl TE, Demetris AJ, Van Thiel D. Liver transplantation. N
Engl J Med 1989;321:1014,1092. (Comprehensive review.)
Williams JW, Simel DL. Does this patient have ascites? How to
divine fluid in the abdomen. JAMA 1992;267:2645. (Best

review of the diagnostic utility of history and physical
examination methods.)


Goroll: Primary Care Medicine, 3rd ed., Copyright 1995 Lippincott-Raven Publishers
Chapter 71 - Management of Cirrhosis and

Chronic Liver Failure
The best treatment is prevention, with emphasis on reducing alcohol consumption (see Chapter 228) ,
limiting occupational hepatotoxin exposure, and preventing parenteral transmission of hepatitis B and C
(see Chapters 57 and 58) . The overall goal is to minimize hepatocellular injury and the risk of
progressing to chronic liver failure and cirrhosis.
Cirrhosis represents an irreversible state of chronic liver injury. However, the cirrhotic patient may be
kept comfortable, active, and independent if precipitants of hepatocellular injury can be eliminated and
complications can be prevented. The responsibility for long-term management of patients with cirrhosis
and chronic liver failure often rests with the primary care physician, who needs to be capable of dealing
with such potential difficulties as ascites, peripheral edema, encephalopathy, infection, bleeding, renal
dysfunction, and electrolyte imbalances.
CLINICAL PRESENTATION AND COURSE

Clinical presentation
may be rather dramatic when ascites, encephalopathy, or brisk variceal bleeding brings the patient to
medical attention. More subtle manifestations include splenomegaly, a firm liver edge, or such signs of
hepatocellular failure as jaundice, palmar erythema, Dupuytren's contractures, spider angiomata, parotid
and lacrimal gland hypertrophy, gynecomastia, testicular atrophy, loss of axillary and pubic hair, and
clubbing. Hypogonadism and feminization are particularly prominent in male patients with cirrhosis due
to alcoholism or hemochromatosis. Most findings are nonspecific and do not reflect the underlying
etiology. More specific is the bronze appearance of the skin in patients with hemochromatosis or the
appearance of antimitochondrial antibodies (anti-M2) in those with primary biliary cirrhosis. Nonspecific
abnormalities in routine liver function tests are common, but the best measures of hepatocellular function
are the serum albumin concentration and prothrombin time (PT). The PT is the first to become abnormal
because of the short serum half-life (as little as 7 days) of the clotting factors that determine it.
Clinical Sequelae and Prognosis.
Portal hypertension, fluid retention, and encephalopathy are the major sequelae of cirrhosis; they lead to
varices, ascites, hypersplenism, peripheral edema, and altered mental status. Variceal bleeding occurs in
20 percent to 30 percent of all cirrhotic patients, one-third of whom die during the initial hospitalization,
one-third rebleed within 6 weeks, and one-third survive 1 year or more. The principal causes of death in
patients with cirrhosis are variceal bleeding, encephalopathy, and infection. In addition, patients with
cirrhosis, especially those with chronic hepatitis B infection, are at increased risk for hepatocellular
carcinoma.

Prognosis is determined by the nature, severity, and activity of the underlying illness. For example,
continued alcohol consumption in the context of alcoholic hepatitis is associated with an 80 percent
chance of developing cirrhosis, whereas abstinence lowers the risk to 15 percent. Even after alcoholic
cirrhosis has developed, survival continues to be affected by alcohol ingestion. Five-year survival is 60
percent to 85 percent in those who abstain, compared with 40 percent to 60 percent for those who
continue to drink. Onset of jaundice or ascites further decreases 5-year survival (to 30% in drinkers). A
discriminant function (DF) using the serum bilirubin and prothrombin time (DF = 4.6[PT - control] +
bilirubin [mg per 100 mL]) has been developed to predict survival. A DF greater than 32 indicates a
short-term mortality risk of 35 percent. In a recent study of patients with symptomatic primary biliary
cirrhosis, the average length of survival from the onset of symptoms was about 12 years, whereas the
survival of asymptomatic patients did not differ from that of a control population matched for age and
sex. The prognosis of patients with postnecrotic cirrhosis is difficult to assess because it is hard to date
its onset; the cirrhosis develops insidiously over years from subclinical chronic active hepatitis.
Irrespective of etiology, development of ascites, encephalopathy, hyperbilirubinemia,
hypergammaglobulinemia (from bypass of the hepatic reticuloendothelial system), and hypoalbuminemia
are poor prognostic signs, as is decreased liver size. Worsening renal function is associated with a 33
percent 2-year mortality.
PRINCIPLES OF MANAGEMENT
Cirrhosis and Its Underlying Etiologies
Alcoholic cirrhosis
requires complete abstinence from further alcohol intake, because prognosis is markedly worsened
408
by continued drinking (see above). Attention to good nutrition, daily multiple vitamin supplements
(including 1 mg of folic acid), and correction of any iron deficiency or electrolyte deficits are important
supportive measures. The search continues for agents that might halt hepatic fibrosis and promote
hepatocyte regeneration. Glucocorticosteroids and portacaval shunting have failed to demonstrate an
improvement in survival, although corticosteroids do improve short-term survival in patients with acute
alcoholic hepatitis. A randomized, placebo-controlled, long-term study of colchicine revealed a doubling
of 5- and 10-year survival.The drug is an inhibitor of collagen deposition. However, confirmatory data
are needed before it or other drugs found experimentally to improve survival ( e.g., propylthiouracil ) can
be recommended. Liver transplantation is sometimes considered in patients who have become totally and
permanently abstinent.
can cause severe pruritus, which may be relieved by cholestyramine in a dose of 4 g orally with meals.
Because of decreased fat absorption from low intestinal bile-salt concentrations, these patients are
particularly prone to develop deficiencies of the fat-soluble vitamins. They may require supplemental
vitamin K (10 mg subcutaneously [SC] every 4 weeks), vitamin D (50,000 U orally two to three times a
week or 100,000 U intramuscularly every 4 weeks) with oral calcium (1 g daily), and vitamin A (25,000

U orally per day). Night blindness unresponsive to vitamin A may be due to zinc deficiency, which is
treated with oral zinc sulfate (220 mg per day). For patients with steatorrhea, medium-chain triglyceride
preparations often help. Azathioprine and colchicine may have a beneficial effect on survival and are
well tolerated. Corticosteroids and penicillamine do not improve survival and cause serious adverse side
effects. Liver transplantation has proven to be a viable option.
can be halted by relieving or bypassing the obstruction to bile flow.
Hemochromatosis
is treated with weekly phlebotomies of 500 mL until the serum iron and ferritin levels fall to normal; then
they are performed as needed.
Wilson's disease
responds to d -penicillamine therapy, which should be administered in conjunction with a hepatologist

experienced in using the drug, because of its potential for serious side effects.
Chronic active hepatitis
may benefit from corticosteroid therapy or interferon alfa, depending on the etiology (see Chapter 70) .
Complications
Ascites and edema
result from increased portal pressure, hypoalbuminemia, secondary hyperaldosteronism, and impaired
free water clearance. Its presence is strongly suggested by the clinical findings of a fluid wave, shifting
dullness, and peripheral edema. Abdominal ultrasound can be used for confirmation and to rule out
venoocclusive disease involving the hepatic veins or the hepatic region of the inferior vena cava
(Budd-Chiari syndrome). Although ascites is not a hazard, gross ascites can cause abdominal discomfort
and respiratory compromise; under such circumstances, it ought to be treated.
Diagnostic paracentesis is indicated before treatment is initiated in patients with new onset of ascites,
worsening hepatic function, fever, or increasing encephalopathy to exclude infection and malignancy.
Although it used to be felt that differentiating exudative from transudative ascitic fluid could aid in
identifying infection and malignancy, such a differentiation has proven insufficiently sensitive and
specific to be reliable. Fluid protein concentrations in excess of 2.5 g per deciliter are seen not only in
conditions causing exudates, but also in patients with transudative processes subjected to diuresis.
Moreover, the ascitic fluid protein concentration of patients with spontaneous bacterial peritonitis (SBP)
is typically less than 1 g per deciliter. However, the serum-to-ascites albumin gradient (serum albumin
concentration minus the ascites albumin concentration) can help differentiate portal hypertension from
other causes of ascites, especially when the ascitic protein concentration is high. A gradient of greater
than 1.1 mg per deciliter is indicative of portal hypertension; ascites in the presence of a gradient less
than 1.1 suggests a mechanism other than portal hypertension. Ascitic fluid should also be sent for
cytologic examination, cell count, and culture. A leukocyte count in excess of 2500 per milliliter is

strongly suggestive of SBP in the patient with abdominal pain and fever. Culture is positive in over 80
percent of cases when 10 mL aliquots are injected into three blood culture bottles; Gram's stain is usually
negative.
Management of ascites due to portal hypertension begins with reduction in sodium intake. Prescribing a 2
g sodium diet is a reasonable compromise between maximizing sodium restriction and dietary
palatability. Adequate nutrition is critical. In the absence of encephalopathy, a daily protein intake of at
least 50 g is recommended. Excessive water intake should be prohibited, and free water should be
restricted to 1500 mL per day if hyponatremia ensues. An effective program of salt and water restriction
requires a cooperative patient and a conscientious family. A dietitian can provide invaluable assistance.
The patient should be instructed to check his or her weight daily. Measuring abdominal girth is an
unreliable index of fluid loss because of variations due to gaseous distention of the gastrointestinal (GI)
tract. About 15 percent of patients will respond to sodium and fluid restrictions alone. Bedrest is of no
added benefit.
Diuretic therapy is indicated if a diuresis has not occurred spontaneously after a full week of salt
restriction. Spironolactone is the agent of first choice, because it inhibits the hyperaldosteronism of portal
hypertension. Being a specific aldosterone antagonist, it counters the hypokalemic alkalosis commonly
seen in cirrhotic patients with ascites. Its diuretic action is mild and unlikely to cause rapid intravascular
volume depletion. The initial dosage of spironolactone is 100 mg a day orally in divided doses. If
diuresis does not follow within 1 week, the daily dosage may be increased by 100 mg every 4 or 5 days
to a maximum of 400 mg daily (higher doses may cause hyperkalemic acidosis). It is useful to monitor
urinary electrolyte concentrations, because diuresis should follow a significant rise in urinary sodium and
a fall in urinary potassium. If natriuresis and diuresis do not
409
occur on a maximal dosage of spironolactone, a loop diuretic such as furosemide (starting at 20 to 40 mg

per day) may be added. Bumetanide or metolazone may be added to the diuretic program, but such potent
diuretics must be used with extreme care to avoid precipitating renal failure, hypokalemia, and
encephalopathy.
Monitoring BUN, creatinine, and electrolytes is critical during treatment of ascites. However, some
cirrhotic patients with renal impairment do not manifest an elevated serum creatinine (believed related to
reduced creatine synthesis by the liver). This reduces the sensitivity of the serum creatinine in some
cirrhotic patients and can complicate monitoring of their renal status. Other measures of renal function
may be required.
The maximum amount of ascitic fluid that can be mobilized in 24 hours is about 700 to 900 mL, although
peripheral edema can be mobilized at a faster rate. This translates to a daily weight loss of 0.5 kg in
patients with ascites alone and of 1 kg in those with both ascites and peripheral edema. Daily weight loss
in excess of these amounts suggests overdiuresis, with its attendant risk of intravascular volume
depletion leading to hepatorenal syndrome and encephalopathy. A falling urine output accompanied by
orthostatic signs (rise in pulse and fall in blood pressure on change from supine to standing) suggest
inadequate intravascular volume. An oral fluid challenge of several hundred milliliters of isotonic fluid
can confirm the presence of hypovolemia by inducing a temporary increase in urine output.

Some patients with incapacitating ascites are truly refractory to sodium restriction and diuretics. For
them, large-volume paracentesis and peritoneovenous (LeVeen) shunting become therapeutic
considerations. Removal of more than 1 to 2 L of ascitic fluid at a time by paracentesis used to be
considered unwise, because of the risk of precipitating serious intravascular volume depletion from the
shift of intravascular fluid into the emptied peritoneal cavity. Moreover, protein depletion and increase in
renin activity and aldosterone secretion might ensue. By infusing intravenous albumin (6 to 8 g per liter
of ascitic fluid removed) at the time of paracentesis, large volumes (5 to 6 L) of ascitic fluid can be
removed without precipitating adverse reactions. Results are comparable to those achieved by shunting
procedures, although admissions for reaccumulation of fluid are more frequent (every 10 to 30 days).
The best candidates are those with peripheral edema and relatively well-preserved renal function.
Before the development of large-volume paracentesis, peritoneovenous (LeVeen) shunting was the
principal means of dealing with refractory ascites. Although effective in treating ascites, shunt placement
does not improve survival. Moreover, insertion can precipitate disseminated intravascular coagulation.
Infection and shunt obstruction are common complications. Compared to large-volume paracentesis,
LeVeen shunting requires fewer readmissions for treatment of ascites, but an overall similar number of
hospitalizations due to shunt occlusions. Its potentially serious complications and the advent of
large-volume paracentesis has led to a decline in use of LeVeen shunts.
is thought to be produced by one or more intestinally derived toxic substances that escape hepatic
detoxification as a result of portasystemic shunting and hepatocellular dysfunction. Candidates include
ammonia, benzodiazepine-like substances, mercaptans, phenol, neuroinhibitors, phenol, and short-chain
fatty acids. Elevations of arterial and venous ammonia levels usually, but not always, correlate with the
presence of hepatic encephalopathy. Venous levels may be falsely elevated when a tourniquet is left on
too long at the time of blood drawing. Ammonia levels are useful in following the clinical state of
individual patients. Important precipitating factors include GI bleeding, excessive dietary protein intake,
hypokalemic alkalosis, infection, constipation, use of sedative or hypnotic drugs, surgical procedures,
and volume depletion resulting from diuresis or paracentesis. Precipitants are identified in about 50
percent of patients; the prognosis is usually better in those with an identifiable contributory factor than in
those in whom the onset of encephalopathy is associated only with worsening hepatic function.
Mild encephalopathy may be managed on an ambulatory basis. Aside from excluding GI bleeding,
avoiding and correcting fluid and electrolyte disturbances, and discontinuing tranquilizers and sedatives,
the mainstay of therapy is restriction of dietary protein intake to 30 to 40 g per day, while maintaining a
daily caloric intake of 1500 kcal. Vegetable sources of protein are preferred over animal proteins because
their metabolism produces less ammonia. Sometimes, oral amino acid supplements (high in
branched-chain varieties and low in aromatic amines) are added to prevent negative nitrogen balance.
Their efficacy in treatment of encephalopathy appears to be transient.
Simple gut cleansing with enemas or cathartics is effective when bleeding, constipation, or a large
dietary protein intake has led to encephalopathy. Patients should be monitored with routine mental status
examinations that include five-point star and signature testing and examination for asterixis.
Specific therapy for encephalopathy begins with lactulose, a synthetic, nonabsorbable disaccharide,
which is metabolized to organic acids by enteric bacteria, causing an osmotic catharsis. In addition,

lactulose suppresses the growth of ammonia-forming bacteria in favor of lactose-fermenting organisms.

The initial dosage for patients with mild encephalopathy is 15 to 30 mL orally every 4 to 6 hours, with
adjustments thereafter to produce two to three loose stools a day. Side effects of oral lactulose include
diarrhea and abdominal distress, which usually resolve after a reduction in dosage.
Lactulose is as effective as neomycin, a poorly absorbed, broad-spectrum aminoglycoside antibiotic,
which predated lactulose and is more toxic. It acts by decreasing the intestinal concentration of
ammonia-forming bacteria. Prolonged use of doses of 4 g per day or more, especially in patients with
renal insufficiency, may lead to ototoxicity and nephrotoxicity. The recommended maximum oral dose is
1 g twice daily. The drug also causes malabsorption. In patients who fail to respond to either lactulose or
neomycin, the two may be given together; in some patients, the effect appears to be additive.
Metronidazole may be used instead of neomycin; it is less toxic.
Variceal Bleeding.
Primary and secondary prophylaxis against bleeding are the principal concerns of the primary
410
care physician. In patients with known varices, risk of bleeding is high. Independent risk factors include
marked hepatocellular dysfunction, ascites, encephalopathy, large varices, and presence of dilated
venules on the varices. Risk of a first bleed is as high as 65 percent in the first year. Because clinical and
endoscopic risk factors are independent and therefore not predictive of one another, some authorities
recommend endoscopic assessment of all cirrhotic patients to determine risk and candidacy for treatment.
Patients judged to be at high risk are reasonable candidates for beta-blocker therapy . When prescribed in
dosages that produce beta-blockade (a reduction in heart rate of about 25%), these agents can lower
portal venous pressure and decrease the risk of variceal bleeding by about 50 percent. The rate of death
from hemorrhage is also reduced. Although disappointing for secondary prophylaxis, beta-blockers have
demonstrated efficacy for prevention of a first variceal bleed and should be considered for use in
high-risk patients. Other agents are being investigated for their ability to further reduce portal pressure.
Initial results with clonidine have been promising.
Injection sclerotherapy (involving endoscopic injection of sclerosing agents) is used to control acute
variceal bleeding and, when repeated several times over 2 to 3 months, to prevent recurrences. Controlled
trials have shown it to be superior to placebo in preventing repeat variceal hemorrhage, but having little
impact on the risk of all-cause upper GI bleeding, which remains at close to 40 percent. Unresolved is its
ability to prevent first bleeding. Like portasystemic shunt surgery (see below), sclerotherapy may only
substitute one form of morbidity and mortality for another. Sclerotherapy has largely replaced
portasystemic shunt surgery, which is now reserved for those who bleed despite repeated sclerotherapies.
Portasystemic shunt surgery remains the final treatment option for patients with recurrent variceal
hemorrhage due to portal hypertension. While reducing the risk hemorrhage, shunt surgery increases the
risk of encephalopathy. The likelihood of recurrent variceal bleeding is significantly reduced after
portasystemic shunting, but long-term survival is not. Reasons for lack of improvement in survival
include high operative mortality rates and no benefit to hepatic function, the principal determinant of
survival. Consequently, portasystemic shunt surgery is not indicated in patients with varices who have

never bled. Selective shunt procedures were developed to reduce the risk of encephalopathy, a common
consequence of shunt surgery. For example, the selective distal splenorenal shunt attempts to preserve
portal blood flow in patients with demonstrable preoperative portal perfusion. Its incidence of
postoperative encephalopathy is lower than with conventional portacaval shunting, but long-term benefit
is reduced as collateral channels develop. Side-to-side portacaval or proximal splenorenal shunts are the
most effective operations for decompressing hepatic sinusoids and reducing ascites, but they still risk
precipitating encephalopathy. For patients awaiting liver transplantation, the transjugular intrahepatic
portasystemic shunt has proven safe and effective.
Coagulopathy
results from reductions in vitamin K-dependent clotting factors (II, VII, IX, and X) secondary to
decreased hepatic protein synthesis and increased plasma proteolytic activity. In addition, bile-salt
deficiency, neomycin therapy, and malnutrition may contribute to malabsorption of vitamin K, and
hypersplenism may account for thrombocytopenia. If a patient with cirrhosis is discovered to have a
prolonged prothrombin time, a trial of vitamin K 10 mg subcutaneously daily for 3 days will correct
hypoprothrombinemia caused by bile-salt deficiency, neomycin, or malnutrition but not
hypoprothrombinemia related only to hepatocellular disease. In the absence of bleeding, measures to
correct abnormal coagulation parameters are generally not indicated.
Role of Liver Transplantation.
In the face of terminal hepatocellular failure, liver transplantation becomes a consideration. In properly
selected patients, 5-year survival may be as high as 85 percent. The best candidates are those who are
highly motivated, emotionally stable, and willing to comply with a medical program. Postnecrotic
cirrhosis, primary biliary cirrhosis, and primary sclerosing cholangitis are among the primary indications.
Alcohol-induced liver disease is a relative contraindication, as is hepatitis B liver disease, hepatocellular
carcinoma, and renal failure. Absolute contraindications are acquired immunodeficiency syndrome
(AIDS), extrahepatic sepsis, metastatic cancer, and severe cardiopulmonary disease.
THERAPEUTIC RECOMMENDATIONS AND MONITORING

General Measures
The patient should maintain a caloric intake of at least 2000 to 3000 kcal per day.
Use of alcohol or other hepatotoxic agents must be prohibited.
The patient should avoid tranquilizers and sedatives.
Monitor prothrombin time, serum albumin, and bilirubin to assess the severity and progression
of hepatocellular dysfunction.
Check stools at each visit for evidence of occult bleeding.
Check for asterixis and other signs of encephalopathy at each visit.
Check the abdomen for evidence of ascites (shifting dullness, fluid wave, bulging flanks).
Ultrasound examination is useful to confirm the presence of ascites and rule out venoocclusive
disease.

Perform a diagnostic paracentesis in patients with the new onset of ascites or clinical
deterioration in the setting of preexisting ascites. The fluid should be sent for cell count and
differential, total protein and albumin concentrations, culture, and cytologic examination.
Instruct patients with ascites to restrict daily sodium intake to no more than 2 g and to consume
at least 50 g of protein per day. Consult with a dietitian and provide patient and family with
specific menus and food lists.
Restrict fluid intake to 1500 mL when there is marked hyponatremia (serum sodium
concentration less than 125 mEq per liter).
411
If salt restriction does not result in diuresis, begin spironolactone 100 mg daily in divided doses.
If natriuresis and diuresis do not occur after 1 week, increase the daily dose of spironolactone by
100 mg every 4 to 5 days to a maximum of 400 mg per day.
If spironolactone alone is ineffective in causing diuresis, add furosemide 20 to 40 mg per day to
the regimen and cautiously increase dosage as necessary.
Adjust diuretic dose so that no more than 0.5 kg of fluid (approximately 1 lb) is lost per day in
patients with ascites alone, and no more than 1 kg per day (2 lb) in those with both ascites and
peripheral edema. Halt diuretics at the first sign of intravascular volume depletion.
Consider daily potassium supplementation (20 to 40 mEq KCl elixir) in patients receiving
furosemide; administer cautiously, if at all, to patients concurrently taking a potassium-sparing
diuretic such as spironolactone.
Monitor serum potassium, BUN, creatinine, daily weight, and postural signs to avoid inducing
intravascular volume depletion, renal failure, hypokalemia, and encephalopathy. Be aware that, in
some patients, the serum creatinine may be falsely normal and remain within normal limits despite
worsening renal function.
Consider large volume paracentesis (5 to 6 L) with concurrent intravenous albumin infusion (6
to 8 g per liter of fluid removed) for patients with refractory ascites that is disabling. Admit for the
procedure.
Encephalopathy
At the first sign of encephalopathy, restrict dietary protein intake to 20 to 30 g per day. Obtain
dietary consultation to construct a diet emphasizing plant protein over animal protein. Consider
use of an oral supplement rich in branched-chain amino acids if protein intake is insufficient.
Monitor mental status and check for asterixis; use five-point star or signature testing. Monitor
venous ammonia levels; in drawing blood for a determination, avoid prolonged tourniquet
application.
When protein restriction fails to control encephalopathy, begin oral lactulose, 15 to 30 mL every
4 to 6 hours, with subsequent adjustments in the dosage to allow two to three soft stools a day.
Add oral neomycin 1 g bid or metronidazole 250 mg tid if lactulose alone does not suffice.

Prevention of Variceal Bleeding and Bleeding due to Clotting Factor Deficiency
Begin a beta-blocker (eg, propranolol 80 mg per day) for primary prevention in patients with
risk factors for variceal bleeding (marked hepatocellular dysfunction, ascites, encephalopathy,
large varices, and presence of dilated venules on the varices).
Consider sclerotherapy and shunt procedures for prevention of recurrent variceal bleeding.
Monitor prothrombin time and platelet count. Administer vitamin K (10 mg SC daily for 3 days)
if there is prolongation of prothrombin time due to drug-induced bile-salt malabsorption,
neomycin, or malnutrition. Platelet transfusions are unwarranted unless there is active bleeding in
the context of a very low platelet count (see Chapter 81) .
INDICATIONS FOR REFERRAL AND ADMISSION

Prompt hospitalization is required for patients with GI bleeding, worsening encephalopathy, increasing
azotemia, signs of peritoneal irritation, or unexplained fever. Intractable ascites may respond to elective
admission for large-volume paracentesis. Decisions about the management of refractory ascites,
encephalopathy, variceal bleeding, and uncommon etiologies of cirrhosis (eg, primary biliary cirrhosis,
Wilson's disease, hemochromatosis) are best made in consultation with a gastroenterologist skilled in
treating liver disease. The same pertains to candidacy for liver transplantation. When urine output falls in
the absence of a clear-cut explanation, a nephrologic consultation may be of considerable help, especially
because creatinine level may not adequately reflect renal function.
PATIENT EDUCATION AND SUPPORT

It should be emphasized to the patient and family that prognosis can often be greatly improved and
symptoms lessened by careful adherence to the prescribed medical program. In particular, dietary
discipline and omission of alcohol are central to a successful outcome and should be stressed. Many of
these patients are chronic alcoholics with low self-esteem. A nonjudgmental, sympathetic physician can
be instrumental in providing support, raising self-esteem, and improving the chances of compliance (see
Chapter 228) . Depression is a frequent accompaniment of the later stages of chronic liver disease and is
manifested by failure to comply with the medical regimen and outright expressions of wanting to die.
Treatment is very difficult. Antidepressant drugs may cause oversedation and thus are risky. There are no
simple measures, but the physician's concern and support can help enormously (see Chapter 227) .
A.H.G.
ANNOTATED BIBLIOGRAPHY
Black M, Friedman AC. Ultrasound examination in the patient with ascites. Ann Intern Med
1989;110:253. (Editorial urging its routine use in patients with new onset of ascites, both for its
confirmation and to rule out a venoocclusive etiology.)
Borowsky SA, Strome S, Lott E. Continued heavy drinking and survival in alcoholic cirrhotics.
Gastroenterology 1981;80:1405. (80 percent of those who continued to drink heavily died an average of
7.2 months after discharge; 95 percent of abstainers were still alive at 14 months.)

Boyer JL, Ransohoff DF. Is colchicine effective therapy for cirrhosis? N Engl J Med 1988;318:1751.
(Urges caution in interpreting the very encouraging data that are emerging; see Kershenobich below.)
Campra JL, Reynolds TB. Effectiveness of high-dose spironolactone therapy in patients with chronic
liver disease and relatively refractory ascites. Dig Dis Sci 1978;23:1025. (Doses as high as 600 mg per
day found effective in patients with ascites presumed to be "refractory.")
Christensen E, Neuberger J, Crowe J, et al. Beneficial effect of azathioprine and prediction of prognosis
in primary biliary cirrhosis final results of an international trial. Gastroenterology 1985;89:1084.
(Improved survival demonstrated.)
412
Conn HO, Leevy CM, Vlahcevic ZR, et al. Comparison of lactulose and neomycin in the treatment of
chronic portal-systemic encephalopathy; a double-blind controlled trial. Gastroenterology 1977;72:573.
(Found to be equally effective and free of significant toxicity at the doses used.)
Epstein M. Treatment of refractory ascites. N Engl J Med 1989;321:1675. (Editorial summarizing data
on LeVeen shunting and large volume paracentesis; argues that paracentesis may be preferable.)
Fraser CL, Arieff AI. Hepatic encephalopathy. N Engl J Med 1985;313:865. (Comprehensive review;
147 references.)
Gines P, Arroyo V, Vargas V, et al. Paracentesis with intravenous infusion of albumin as compared with
peritoneovenous shunting in cirrhosis with refractory ascites. N Engl J Med 1991;325:829. (Randomized
comparison showing equal effectiveness and similar total hospital days.)
Graham DY, Smith JL. The course of patients after variceal hemorrhage. Gastroenterology 1981;80:800.
(One-third died in the initial hospitalization, one-third rebled within 6 weeks, and one-third survived at
least 1 year. Long-term survival after the initial 2 weeks no different from that of unselected cirrhotics
without bleeding.)
Kaplan MM. Primary biliary cirrhosis. N Engl J Med 1987;316:521. (Authoritative and very useful
review for the generalist reader; 124 references.)
Kershenobich D, Vargas F, Garcia-Tsao G, et al. Colchicine in the treatment of cirrhosis of the liver. N
Engl J Med 1988;318:1709. (Randomized, placebo-controlled study; 14 years of follow-up; survival
doubled.)
North Italian Endoscopic Club. Prediction of the first variceal hemorrhage in patients with cirrhosis of
the liver and esophageal varices. N Engl J Med 1988;319:983. (Multivariate analysis identifying risk
factors predictive of bleeding.)
Orrego H, Blake JE, Blendis LM, et al. Long-term treatment of alcoholic liver disease with
propylthiouracil. N Engl J Med 1987;317:1421. (Long-term, double-blind, randomized trial
demonstrating reduction in mortality; about 50 percent of patients had cirrhosis.)
Pagliaro L, D'Amico G, Sorensen TIA, et al. Prevention of first bleeding in cirrhosis: A meta-analysis of
randomized trials of nonsurgical treatment. Ann Intern Med 1992;117:59. (Beta-blockers prevent first

bleeding in high-risk patients; sclerotherapy is of unproven efficacy for primary prophylaxis.)

Papadakis MA, Arieff AI. Unpredictability of clinical evaluation of renal function in cirrhosis. Am J Med
1987;82:945. (Some cirrhotics with reduced glomerular filtration have normal serum creatinine levels,
which may remain within normal limits despite worsening of renal function.)
Powell WJ Jr, Klatskin G. Duration of survival in patients with Laennec's cirrhosis. Am J Med
1968;44:406. (Classic paper showing improved survival in cirrhotic patients who abstained from alcohol
compared with those who continued to drink heavily.)
Poynard T, Cales P, Pasta L, et al. Beta-adrenergic antagonist drugs in the prevention of gastrointestinal
bleeding in patients with cirrhosis and esophageal varices. N Engl J Med 1991;324:1532. (Analysis of
pooled data showing reduction in risk of first bleeding and mortality from hemorrhage.)
Rector WG, Jr, Reynolds TB. Superiority of serum-ascites albumin differences in "exudative" ascites.
Am J Med 1984;77:83. (Better than total protein for differential diagnosis.)
Rikkers LF. Operations for management of esophageal variceal hemorrhage. West J Med 1982;136:107.
(Critical review of the various types of portasystemic shunt procedures and the indications for their use.)
Ring EJ, Lake JR, Roberts JP, et al. Using transjugular intrahepatic portosystemic shunts to control
variceal bleeding before liver transplantation. Ann Intern Med 1992;116:304. (Prospective, uncontrolled
trial; found safe and effective.)
Shear L, Ching S, Gabuzda GJ. Compartmentalization of ascites and edema in patients with hepatic
cirrhosis. N Engl J Med 1970;282:1391. (Classic, elegant study; maximum rate at which ascites can be
absorbed was 930 mL per 24 hours; concludes that the therapeutic aim in patients with ascites should be
weight loss of no more than 0.5 kg daily.)
Stanley MM, Ochi S, Lee KK, et al. Peritoneovenous shunting as compared with medical treatment in
patients with alcoholic cirrhosis and massive ascites. N Engl J Med 1989;321:1632. (Ascites alleviated,
but survival unchanged.)
Starzl TE, Demetris AJ, Van Thiel D. Liver transplantation. N Engl J Med 1989;321:1014,1092.
(Comprehensive review.)
Williams JW, Simel DL. Does this patient have ascites? How to divine fluid in the abdomen. JAMA
1992;267:2645. (Best review of the diagnostic utility of history and physical examination methods.)


April 22, 2000
Conde Petra
Rakel: Conn's Current Therapy 1999, 51st ed.,

Section 6 - THE DIGESTIVE SYSTEM
123 - CIRRHOSIS Rakel: Conn's Current Therapy 1999, 51st ed., Copyright 1999 W. B.
Saunders Company
Introduction
DIAGNOSIS
CAUSES OF CIRRHOSIS: Chapter 123 - CIRRHOSIS

EVALUATION AND SPECIFIC
THERAPY
Alcoholic Cirrhosis PETER M. ROSENBERG M.D.

LAWRENCE S. FRIEDMAN M.D.
Cirrhosis Due to Chronic Viral
Hepatitis Massachusetts General Hospital Boston, Massachusetts
Primary Biliary Cirrhosis
Hemochromatosis Cirrhosis is the final common pathway in the course of a variety of

Wilson's Disease chronic, progressive liver diseases. Strictly defined, cirrhosis is a
histologic diagnosis, based on the presence of hepatocellular necrosis,
Autoimmune Hepatitis fibrosis, and regenerative nodules. From a clinical point of view,
PREVENTION AND TREATMENT however, cirrhosis represents a constellation of symptoms, signs, and
OF COMPLICATIONS laboratory abnormalities reflecting underlying hepatocellular
dysfunction, portal hypertension, and portosystemic shunting.
Variceal Bleeding Typically, cirrhosis follows a slow, insidious course and often
Ascites
culminates in dramatic and fatal complications. The cornerstone of
management is the prevention and treatment of these complications.
Spontaneous Bacterial
Peritonitis The prognosis for patients with cirrhosis varies according to clinical
stage and etiology. It is clearly best for patients with well-compensated
Hepatorenal Syndrome disease. For example, in patients with well-compensated cirrhosis due
Hepatic Hydrothorax
to chronic hepatitis C, the 10-year survival rate has been estimated to
be approximately 80%. Regardless of etiology, prognosis correlates
Hepatic Encephalopathy with the Child-Pugh classification (Table 1) . Mean survival periods
Hepatopulmonary Syndrome
for patients with class A, class B, and class C cirrhosis are 40, 32, and
8 months, respectively. Whereas patients with class A cirrhosis may
Coagulopathy benefit from treatment of the underlying chronic liver disease, patients
with class B or C cirrhosis should generally be evaluated for liver
Susceptibility to Infection
transplantation.
Hepatocellular Carcinoma

ORTHOTOPIC LIVER
TRANSPLANTATION DIAGNOSIS
In clinical practice, the diagnosis of cirrhosis is usually based on the
presence of suggestive symptoms, signs, and laboratory abnormalities.
Possible symptoms and signs include fatigue, nausea, vomiting,
diarrhea, amenorrhea, easy bruising, jaundice, increased abdominal
girth, leg swelling, evidence of upper gastrointestinal bleeding, and
confusion. Other physical signs include jaundice, spider
TABLE 123-1 -- Child-Pugh Classification of Cirrhosis
Score
Feature 1 2 3
Encephalopathy 0 1-2 3-4
(stage)
Ascites Absent Slight Poorly
controlled
Bilirubin <2.0 2.0-3.0 >3.0
(mg/dL)
Albumin >3.5 2.8-3.5 <2.8
(gm/dL)
Prothrombin 1.0-4.0 4.0-6.0 >6.0
time (seconds
prolonged)
Each feature is assigned 1, 2, or 3 points.
Class A: 5-6 points Class B: 7-9 points Class C: 10-15 points
angiomata, palmar erythema, ascites, peripheral edema, a firm and

nodular liver edge, splenomegaly, caput medusae, gynecomastia,
testicular atrophy, and gastric or esophageal varices (on endoscopic
examination). Characteristic laboratory abnormalities include an
elevated serum bilirubin level, decreased serum albumin level, elevated
serum globulin levels, an elevated prothrombin time (PT), and
thrombocytopenia.
If at least several of these features are present in a patient, the diagnosis
of cirrhosis is usually secure. Their absence, however, does not exclude
a diagnosis of cirrhosis, especially in patients with compensated
disease. In such cases, the disease is subclinical, and liver biopsy may
be necessary for definitive diagnosis. Ultrasonography with Doppler
flow studies may show abnormalities in liver size, surface patterns, and
homogeneity and may detect splenomegaly, portal vein enlargement,
caudate lobe enlargement, and the presence of collateral circulation, all
of which are suggestive of cirrhosis. However, ultrasonography is not
as sensitive as liver biopsy and fails to provide information on the

underlying cause of cirrhosis. Various serum markers of fibrosis have

been reported to aid in the noninvasive diagnosis of cirrhosis, and
recently, an elevated serum hyaluronate level has been suggested to be
a sensitive and specific marker. However, this test is not widely
available and has not been validated in a large population.
It should be noted that overt clinical signs of portal hypertension,
portosystemic shunting, and hepatic synthetic dysfunction may also be
seen with acute alcoholic hepatitis and, less commonly, with flares of
chronic hepatitis B, in the absence of cirrhosis. Without a liver biopsy,
distinguishing decompensation due to one of these acute processes (or
both) from cirrhosis may be impossible. When percutaneous liver
biopsy is precluded by marked coagulopathy (PT prolonged by at least
3 seconds or a platelet count of 75,000 per mm3 or less), a transjugular
approach may be used for biopsy.
CAUSES OF CIRRHOSIS: EVALUATION AND

SPECIFIC THERAPY
The causes of cirrhosis are numerous (Table 2) . In the United States,
the majority of cases can be attributed to alcohol or hepatitis C virus
infection, or both. Despite the discovery of hepatitis C virus in 1989,
many cases of cirrhosis are still considered cryptogenic. It has recently
been suggested that nonalcoholic steatohepatitis (NASH) may account
for many of these cases, because cryptogenic cirrhosis is found in a
substantial number of obese diabetic patients with
hypertriglyceridemia.
Often, the cause of cirrhosis can be determined by the history and
noninvasive testing alone. Standard work-up should include serologic
tests for viral hepatitis, including assays for hepatitis B surface antigen
and antibody to hepatitis C; determination of serum iron, total
iron-binding capacity (TIBC), ferritin, ceruloplasmin, and alpha1
-antitrypsin levels; assays for antinuclear, anti-smooth muscle, and
antimitochondrial antibodies; and, as noted previously, an abdominal
ultrasound scan with Doppler flow studies. If the presence and cause of
cirrhosis in a given patient can
461
TABLE 123-2 -- Causes of Cirrhosis in Adults in the United

States
Chronic viral hepatitis Drugs
Hepatitis C Methotrexate
Hepatitis B (with or without Isoniazid
delta co-infection) Methyldopa

Alcohol Amiodarone
Cryptogenic cirrhosis Halogenated hydrocarbons
Nonalcoholic steatohepatitis Hypervitaminosis A
Autoimmune diseases Nonviral infections
Primary biliary cirrhosis Schistosomiasis
Autoimmune hepatitis Brucellosis
Primary sclerosing Syphilis
cholangitis Nonautoimmune biliary
Metabolic diseases disease
Hemochromatosis Chronic biliary obstruction
Wilson's disease Biliary atresia
Alpha1 -antitrypsin Cystic fibrosis
deficiency
Galactosemia Chronic graft-versus-host
disease
Hereditary fructose Vascular disorders
intolerance
Tyrosinemia Chronic Budd-Chiari
Glycogen storage disease syndrome
(types III and IV) Veno-occlusive disease
Abetalipoproteinemia Chronic heart failure
Sarcoidosis
Jejunoileal bypass
be inferred from results of this noninvasive work-up a liver biopsy may

not be needed. If, however, the results are negative or ambiguous, liver
biopsy should be performed. In suspected primary sclerosing
cholangitis, endoscopic retrograde cholangiopancreatography should
be performed; anti-neutrophil cytoplasmic antibody (p-ANCA) may be
detected in 70% of affected persons.
Knowledge of the exact cause of cirrhosis in a patient may allow
treatment of the underlying disorder. Although cirrhosis is generally
thought to be irreversible, treatment of the underlying disease may
slow the progression of cirrhosis, ameliorate the clinical course, reduce
the risk of hepatocellular carcinoma, and, in rare cases, lead to
regression of cirrhosis. In general, such treatment is most effective and
best tolerated when initiated in the precirrhotic or Child class A phase.
Regardless of the cause of cirrhosis, treatment with colchicine to
inhibit hepatic fibrogenesis may be considered. In a long-term study, 5-
and 10-year survival rates for patients with early-stage cirrhosis were

found to be significantly higher for those who received oral colchicine,

in a dose of 1 mg daily, than for those who received placebo. In the
future, more specific antifibrotic agents will probably become
available.
Alcoholic Cirrhosis
Alcoholic cirrhosis typically results from repeated bouts of alcoholic

hepatitis, either overt or subclinical. Acute alcoholic hepatitis is
generally self-limited, and abstinence can clearly prevent progression
to cirrhosis. Treatment with corticosteroids may be beneficial for
selected patients with severe alcoholic hepatitis. Even once cirrhosis
develops, abstinence from alcohol improves the prognosis. In patients
with decompensated alcoholic cirrhosis who continue to drink, the
5-year survival rate is less than 50%, but in those who remain
abstinent, survival is improved and the incidence of liver-related
complications decreased. Both ascites and esophageal varices have
been observed to regress and portal pressure to decrease with
abstinence. Therefore, every effort should be made to enter patients
with alcoholic cirrhosis into organized alcohol counseling programs.
Nutrition also plays a role in therapy for the alcoholic patient with
cirrhosis. Adequate caloric intake should be ensured, and vitamin and
electrolyte deficiencies should be corrected. It is generally advisable to
maintain patients on a multivitamin, folic acid 1 mg, and thiamine 100
mg daily. Hypomagnesemia and hypophosphatemia are common and
should be corrected, initially with intravenous therapy.
Cirrhosis Due to Chronic Viral Hepatitis
Worldwide, chronic hepatitis B is probably the most common cause of

cirrhosis and liver-related death. The active, or replicative, phase of
chronic hepatitis B is characterized by the presence in serum of the
hepatitis B e antigen (HBeAg) and hepatitis B virus (HBV) DNA.
Standard treatment for patients with chronic hepatitis B is
administration of interferon alfa-2b (Intron-A), 5 million units
subcutaneously (SC) daily or 10 million units SC three times per week,
for 4 months (see the article "Acute and Chronic Viral Hepatitis" for
more detail on treatment). Response to interferon in hepatitis B is
typically accompanied by a mild flare of hepatitis. In HBeAg-positive
patients with compensated cirrhosis, response to interferon treatment
has been shown to be associated with improved survival and decreased
incidence of hepatic decompensation. However, interferon should not
be used in patients with decompensated liver disease, because the
treatment-induced flare of hepatitis may precipitate liver failure. More
promising for patients with advanced liver disease caused by chronic
hepatitis B are nucleoside analogues such as lamivudine, although
these drugs are not yet approved for treatment of hepatitis B. These
drugs may also prove effective in diminishing the risk of recurrent
hepatitis B after liver transplantation.

Interferon therapy, alone or with ribavirin, has not been shown to

improve the prognosis of hepatitis C-related cirrhosis. In fact, the
presence of cirrhosis predicts a poor response to interferon, although
therapy may be considered in patients with well-compensated cirrhosis
and active inflammation on liver biopsy. There is some evidence that
such therapy reduces the risk of hepatocellular carcinoma.
Unfortunately, there are as yet no therapeutic alternatives to interferon
for treatment of hepatitis C. Therefore, the only treatment for advanced
hepatitis C-related cirrhosis is liver transplantation, if indicated.
462
Primary biliary cirrhosis (PBC) is a chronic cholestatic disease caused

by immunologic reactivity against the epithelium of the intrahepatic
bile ductules. The disorder typically affects middle-aged women and
follows a slow but progressive course, characterized eventually by
pruritus, jaundice, steatorrhea, fat-soluble vitamin deficiencies, and
ultimately end-stage liver disease. Laboratory hallmarks of PBC are an
elevated serum alkaline phosphatase level, the presence of
antimitochondrial antibody, and hypercholesterolemia. Prognosis is
closely related to the serum bilirubin level as well as to the patient's
age, presence of edema, prolongation of PT, and variceal bleeding.
At present, the mainstay of treatment for PBC is administration of
ursodeoxycholic acid, or ursodiol (Actigall), * given orally (PO) in a
daily dose of 10 to 15 mg/kg, generally in divided doses.
Ursodeoxycholic acid is a hydrophilic bile acid that is absorbed from
the gastrointestinal tract after oral ingestion and displaces more toxic
bile acids from the enterohepatic circulation and thus from the bile
pool. It also improves bile flow and appears to modulate the humoral
immune response. Treatment with ursodeoxycholic acid has been
shown to reduce pruritus, improve biochemical parameters (especially
serum bilirubin levels), delay progression to cirrhosis, and improve
survival, as measured by time to liver transplantation. However, the
drug has not been shown to induce histologic improvement in patients
with PBC. The survival benefit has been shown even in patients with
stage IV disease (i.e., cirrhosis). Therefore, ursodeoxycholic acid is
clearly indicated in the treatment of cirrhosis due to PBC.
Other pharmacologic agents for treatment of PBC that have been
studied include colchicine, * prednisone, D-penicillamine, *
azathioprine, * cyclosporine, * chlorambucil, * and methotrexate. *
Colchicine, * 0.6 mg twice daily (bid), may slow the rate of progression
to cirrhosis but has not been shown to improve survival. Dramatic
results have been reported with long-term methotrexate therapy (in
weekly "pulses" of 15 mg) in a small series of patients, and complete
histologic remission was observed in several patients. Further trials are

needed to define the subgroup likely to benefit from methotrexate

therapy, and the drug should still be considered investigational. Trials
of the other agents have been disappointing, and the toxic effects of
these drugs appear to outweigh their benefits.
Several unique complications of PBC may require specific treatment.
Pruritus may be treated with cholestyramine (Questran) in doses of 4
grams PO in water or juice, bid or three times daily (tid); colestipol, 5
grams bid or tid; or oral antihistamines such as hydroxyzine (Atarax),
in a dose of 25 mg four times a day as needed. Additional treatments
for refractory pruritus include rifampin (Rifadin), * 300 mg PO bid;
phenobarbital, * 2 to 4 mg per kg per day, with a target serum level of
10 mum per mL; and phototherapy. The oral opiate antagonist
naltrexone (ReVia) has shown particular promise in the treatment of
pruritus associated with cholestasis. Patients with PBC also require
supplementation to compensate for malabsorption of fat-soluble
vitamins. In general, calcium, 1000-1500 mg daily, and vitamin D,
50,000 units PO per week, should be prescribed. Vitamin K, 10 mg SC
daily for 2 to 3 days and then 10 mg SC per month should be given to
patients with a prolonged PT, and vitamin A, 25,000 units PO per day,
to those with night blindness. The value of bisphosphonates and
estrogen in the prevention and treatment of osteoporosis associated
with PBC is uncertain.
Hemochromatosis
Hereditary hemochromatosis is an autosomal recessive disorder

characterized by excessive intestinal absorption of iron, with resulting
iron deposition in the liver, heart, pituitary, and pancreas and eventual
cirrhosis, dilated cardiomyopathy, panhypopituitarism, and diabetes
mellitus. Initial testing for suspected hemochromatosis should include
determination of serum iron level, TIBC, and ferritin level; a fasting
transferrin saturation (defined as the serum iron level divided by the
TIBC) of greater than 50% is suggestive of hemochromatosis.
However, the transferrin saturation and serum ferritin level are also
elevated in up to half of patients with chronic viral hepatitis, alcoholic
liver disease, and NASH. Definitive diagnosis of hemochromatosis has
traditionally been made by liver biopsy with quantitative hepatic iron
determination, which should be performed in all patients with an
elevated transferrin saturation or serum ferritin level. The recent
identification of a single mutation (HFE) in a human leukocyte antigen
(HLA)-like gene (HLA-H) has made it possible to perform genetic
testing for hemochromatosis. Two copies of the HFE gene mutation
are present in more than 90% of patients with biochemical and
histologic evidence of hemochromatosis.
Identifying hemochromatosis as the underlying cause of cirrhosis in a
given patient is of great importance, because treatment with iron
depletion therapy ameliorates the liver dysfunction and other systemic
manifestations of hemochromatosis, even in patients who already have

cirrhosis. In rare cases, phlebotomy therapy has been reported to

reverse cirrhosis. Phlebotomy with removal of 500 mL of whole blood
should be performed weekly until the transferrin saturation falls below
50% and the ferritin level below 50 ng per mL, or until the hemoglobin
level falls to 10 mg per dL. Preoperative iron depletion is particularly
important in optimizing outcome after liver transplantation. For
patients with cirrhosis who are too ill or anemic to tolerate phlebotomy,
iron chelation
*Not FDA approved for this indication
Exceeds dosage recommended by the manufacturer.
463
therapy with deferoxamine (Desferal) may be considered but is less

effective than phlebotomy. Finally, first-degree relatives of patients
with hemochromatosis should undergo screening with serum iron
studies and, if the proband has the HFE gene, genetic testing.
Wilson's Disease
Wilson's disease is an uncommon disease, inherited as an autosomal

recessive trait, resulting in copper accumulation in the liver, central
nervous system (CNS), and kidneys. The disorder tends to manifest in
childhood or early adulthood. The diagnosis is made on the basis of
detection of Kayser-Fleischer rings in the cornea, a serum
ceruloplasmin level of less than 20 mg per dL, an elevated urinary
copper level, and an elevated hepatic copper concentration on liver
biopsy. However, Kayser-Fleischer rings may be absent, particularly in
patients without neuropsychiatric features of Wilson's disease. Medical
therapy with D-penicillamine (Cuprimine), 1-2 grams daily in four
divided doses, is effective in preventing copper accumulation but has
not been shown to induce regression of established cirrhosis. Side
effects of D-penicillamine include rash, fever, bone marrow
suppression, and glomerulonephritis and may necessitate a reduction in
dosage or use of an alternative agent such as trientine hydrochloride
(Syprine), 750 to 2000 mg per day in two to four divided doses. Zinc
may be used for maintenance therapy after copper depletion has been
achieved or in pregnant and presymptomatic patients. End-stage liver
disease due to Wilson's disease carries an especially poor prognosis,
and liver transplantation should be strongly considered in such cases;
copper levels return to normal after transplantation.
Autoimmune Hepatitis
Classic (type 1) autoimmune hepatitis typically affects women in the

third to fifth decades of life and is often associated with high globulin
levels and the presence of antinuclear and anti-smooth muscle
antibodies in serum. The disease is generally very sensitive to therapy
with corticosteroids, which is indicated in patients with symptoms,

serum aminotransferase elevations at least 10 times the upper limit of

normal (or 5 times the upper limit of normal and a serum globulin level
at least twice the upper limit of normal), or histologic evidence of
severe chronic hepatitis. Prednisone should be initiated at a dose of 40
to 60 mg daily for 1 to 2 weeks and then tapered over several weeks to
a maintenance dose of 10 to 15 mg. Azathioprine (Imuran), * in a dose
of 50 to 100 mg daily, is useful in allowing a lower dose of prednisone
in these patients. Maintenance therapy should be continued until
biochemical and histologic remission occurs, which may take more
than 18 months. Thereafter, therapy may be discontinued, but relapse is
common. In patients who experience repeated relapses long-term
suppressive therapy should be used; in some cases, long-term therapy
with azathioprine alone in a dose of 2 mg per kg per day, may maintain
remission. Treatment of autoimmune hepatitis should be initiated even
if cirrhosis has already supervened, and complete regression of
established cirrhosis has been reported. Liver transplantation should be
considered for patients with progressive liver dysfunction that does not
respond to immunosuppressive therapy; clinical recurrence of
autoimmune hepatitis is uncommon after transplantation.
PREVENTION AND TREATMENT OF

COMPLICATIONS
Variceal Bleeding
Bleeding from ruptured esophageal or gastric varices is one of the most

common and potentially catastrophic complications of cirrhosis.
Prevention and treatment of variceal bleeding are addressed in the
article "Bleeding Esophageal Varices."
Ascites
Ascites is the most common and often the earliest major complication
of cirrhosis and occurs in 50% of patients within 10 years of the
diagnosis of compensated cirrhosis. Pathogenetic factors include (1)
increased hydrostatic pressure in the portal circulation due to distortion
of the hepatic vasculature, (2) decreased plasma oncotic pressure due
to impaired albumin synthesis, (3) transudation of lymph from the liver
surface due to hepatic sinusoidal obstruction, and (4) renal fluid
retention mediated by activation of the renin-angiotensin-aldosterone
system. Ascites signals a poor prognosis in chronic liver disease and is
associated with a 2-year survival rate of approximately 50%. Morbidity
may result directly from tense abdominal distention; thus, findings may
include abdominal discomfort, anorexia, early satiety, respiratory
compromise, and renal insufficiency. Massive ascites may also lead to
the formation of inguinal or umbilical hernias. When the hernias
become large and the overlying skin becomes excoriated,
life-threatening rupture of the hernia may occur, necessitating urgent
surgical repair. Superinfection of ascites may result in spontaneous

bacterial peritonitis (see later on), and translocation of ascitic fluid

across the diaphragm may lead to hepatic hydrothorax (see later on).
Diagnostic paracentesis should be performed in any patient with new
or worsening ascites, as well as in any patient with established ascites
in whom abdominal pain, fever, or encephalopathy develops. When
performed with a 21-gauge needle, paracentesis is safe even in patients
with significant coagulopathy. The fluid obtained should be sent for
cell and differential counts; total protein, albumin, amylase, and
triglyceride levels; and cytologic studies
464
and culture (by direct inoculation into blood culture bottles at the
bedside). A serum albumin level should be obtained on the same day.
Ascitic fluid glucose and lactate dehydrogenase (LDH) concentrations
and a Gram's stain are useful only when secondary bacterial peritonitis
is a diagnostic consideration. When tuberculous ascites is a possibility,
an acid-fast stain should be obtained and a specimen submitted for
mycobacterial cultures. Cirrhotic ascites is typified by a low total
protein concentration, less than 2.5 grams per dL and often less than 1
gram per dL, and a high serum-ascites albumin gradient. The
serum-ascites albumin gradient correlates directly with portal pressure,
and gradients of 1.1 grams per dL or greater signify the presence of
portal hypertension.
Therapy of cirrhotic ascites begins with dietary modification. Initial
restriction of sodium chloride intake to 2 grams per day results in
resolution of ascites in some patients and is crucial to the effectiveness
of diuretics in the remainder. Patients with new ascites and normal
renal function may respond to salt restriction alone. Counseling by a
dietitian on how to implement such a diet leads to improved
compliance. Fluid restriction is unnecessary unless significant
hyponatremia (serum sodium concentration of less than 120 mEq per
liter) develops.
The majority of patients with cirrhotic ascites require diuretic therapy
in addition to dietary salt restriction. Such therapy requires vigilant
monitoring to avoid electrolyte disturbances and volume depletion.
Before initiation of diuretic therapy, the patient's baseline body weight,
serum electrolyte values, and urinary sodium and potassium
concentrations should be measured and tests of renal function
conducted. In patients with hyperkalemia, such as those with type IV
renal tubular acidosis related to diabetes mellitus, cautious treatment, if
indicated at all, with potassium-sparing diuretics is in order. Patients
with abnormal renal function at baseline must be monitored carefully
in order to avoid precipitating prerenal azotemia or hepatorenal
syndrome. In most patients with cirrhosis, the urinary potassium
concentration exceeds that of sodium owing to hypersecretion of

aldosterone. In these patients, the goal of diuretic therapy is to induce

natriuresis and reduce kaliuresis, so that the ratio of urinary sodium to
potassium concentration is greater than 1.
The first diuretic prescribed for most patients should be spironolactone
(Aldactone), a direct aldosterone receptor antagonist and
potassium-sparing agent. The starting dose is 100 mg per day which
may be given initially in divided doses and after several days as a
single dose. The dose may be increased to a maximum of 400 to 600
mg per day, but dose changes should be made no more often than every
4 days, owing to the long half-life of the drug. The dosage should be
considered optimal once urine sodium concentration exceeds urine
potassium concentration. The most common complication of
spironolactone therapy is hyperkalemia, which is usually avoidable if
serum electrolytes are monitored. Spironolactone may cause painful
gynecomastia because it is an androgen receptor antagonist. If this
occurs, triamterene (Dyrenium) may be substituted in a dose of 50 mg
bid, with increases to 150 mg bid.
Optimal diuresis should result in weight loss of approximately 0.5 kg
per day; weight loss of up to 1.0 kg per day is well tolerated in patients
with concomitant peripheral edema. Although spironolactone
antagonizes the sodium-retaining effect of aldosterone, additional
stimuli to sodium retention remain, and effective diuresis often requires
addition of another agent, usually a loop diuretic such as furosemide
(Lasix). In patients with large-volume ascites, it is reasonable to begin
administration of furosemide simultaneously with spironolactone.
Otherwise, furosemide can be withheld until it has been established
that the dosage of spironolactone has been optimized without adequate
diuresis. The usual starting dose of furosemide is 40 mg once daily,
and the dose may be doubled every several days to a maximum of 160
mg per day if necessary. When given intravenously (IV), furosemide
diminishes renal perfusion significantly in patients with cirrhosis;
therefore, this route of administration should be avoided. Occasionally,
the addition of a third diuretic, such as hydrochlorothiazide, may be
considered.
Failure to respond to medical therapy defines refractory ascites; failure
of ascites to resolve despite maximal doses of diuretics (400 mg of
spironolactone and 160 mg of furosemide daily) defines
"diuretic-resistant" ascites, whereas failure due to intolerance of
maximal doses of diuretics defines "diuretic-intractable" ascites.
Surreptitious salt ingestion should be ruled out in patients who fail to
lose weight despite adequate natriuresis. In patients with
diuretic-resistant or diuretic-intractable ascites, the next therapeutic
option is generally large-volume paracentesis, which should be initial
therapy in any patient with respiratory compromise from tense ascites.
In this procedure, 5 or more liters of fluid is removed using an
18-gauge needle connected by phlebotomy or peritoneal dialysis tubing
to vacuum bottles. This procedure has been shown to be safe and

effective in relieving tense ascites. Complete paracentesis, in which the

entire volume of ascites is drained, has also been shown to be safe and
effective. Although our approach is controversial, we prefer to
administer approximately 10 grams of salt-poor albumin IV per liter of
ascites removed in order to minimize the risk of renal insufficiency and
hyponatremia. Large-volume paracentesis may be repeated regularly,
with careful monitoring of hemodynamics, electrolytes, and renal
function. For hospitalized patients with tenuous renal function, tense
ascites, and significant hypoalbuminemia, infusion of salt-poor
albumin without paracentesis probably enhances diuresis over the short
term without a detrimental hemodynamic effect. In such patients,
infusions of 25 to 50 grams per day should be continued until a target
serum albumin of 3.0 grams per dL is reached. However, excessive
albumin administration may increase the risk of variceal bleeding by
increasing intravascular volume.
465
Other therapies for refractory ascites include insertion of a

peritoneovenous (Le Veen) shunt and placement of a transjugular
intrahepatic portosystemic shunt (TIPS). A LeVeen shunt connects the
peritoneal cavity directly to the internal jugular vein, allowing
decompression of ascites by drainage into the vascular system.
However, the use of such shunts is associated with a high incidence of
serious complications, most notably disseminated intravascular
coagulation (DIC) and sepsis, and such devices have fallen out of
favor. Placement of a TIPS is performed by an interventional
radiologist and may lead to dramatic resolution of ascites in as many as
75% of patients in whom the condition is refractory to diuretics.
Natriuresis may be delayed up to 4 weeks after a TIPS insertion.
Unfortunately, shunt thrombosis is a common early complication, and
stent occlusion due to pseudointimal hyperplasia is nearly universal in
the first 1 to 2 years after placement of a TIPS. Therefore, frequent
shunt revisions are usually necessary. In addition, the use of a TIPS
poses an increased risk of portosystemic encephalopathy and has not
been shown to reduce overall mortality. In selected patients, refractory
ascites is an indication for liver transplantation (see later on).
Spontaneous Bacterial Peritonitis
An ominous complication in patients with cirrhotic ascites is

spontaneous bacterial peritonitis (SBP), which is associated with a
short-term mortality rate of approximately 50%, and a 1-year mortality
rate of 62% even after recovery from a first episode. SBP results from
hematogenous superinfection of ascites stemming from prolonged
bacteremia due to decreased reticuloendothelial (RE) function,
portosystemic shunting, and deficiency of complement and
immunoglobulins in ascitic fluid. Because the ascitic fluid total protein
concentration correlates directly with the opsonic activity of ascitic

fluid, patients with an ascitic fluid total protein concentration of less

than 1 gram per dL are at highest risk of developing SBP. The most
common pathogens in SBP are gram-negative enteric bacteria,
especially Escherichia coli and Klebsiella pneumoniae, but
streptococci, including pneumococci, and also enterococci account for
a significant minority of cases. In contrast, anaerobic bacteria rarely
cause SBP.
The manifestations of SBP may be protean. Abdominal pain and fever
are typical but often absent. Abdominal tenderness, sometimes with
peritoneal signs, is present in approximately half the patients. In many
cases, the only clue to the presence of the infection is worsening
encephalopathy, renal function, or ascites. In fact, up to one third of
affected patients may be completely asymptomatic. Therefore, a high
index of suspicion for SBP is necessary, and the clinician should have
a low threshold for performing diagnostic paracentesis. Paracentesis
should be performed in any patient with ascites and fever, new
abdominal pain, or new encephalopathy.
The diagnosis of SBP is based on an ascitic fluid polymorphonuclear
neutrophil (PMN) count of 500 per mm3 or higher (or 250 per mm3 or
higher if the patient has any symptoms of SBP) and absence of an
underlying intra-abdominal focus of infection, whether or not the
ascitic fluid culture is positive. At least 10 mL of fluid should be
inoculated directly into blood culture bottles at the bedside to
maximize the sensitivity of the culture. Culture-negative neutrocytic
ascites, defined as ascites with a fluid PMN count of 500 cells per mm3
or higher in the absence of a positive fluid culture and recent antibiotic
therapy, should be treated as for SBP. On the other hand,
monomicrobial non-neutrocytic "bacterascites," defined as ascites with
positive results on fluid culture but an ascitic fluid PMN count of less
than 250 per mm3 , is generally transient and often does not require
treatment.
The antibiotic of choice for treatment of SBP is the third-generation
cephalosporin cefotaxime (Claforan) in a dose of 2 grams IV every 8
hours. Alternatives include ceftriaxone (Rocephin),
ampicillin-sulbactam (Unasyn), ticarcillin-clavulanic acid (Timentin),
and piperacillin-tazobactam (Zosyn). Recently, oral ofloxacin (Floxin)
has been shown to be highly effective for treatment of SBP, but
additional studies are needed before oral antibiotic therapy alone can
be recommended. Use of aminoglycosides should be avoided, because
patients with cirrhosis are particularly prone to
aminoglycoside-induced nephrotoxicity. Antibiotic therapy should be
started immediately after paracentesis, pending culture results. If SBP
is confirmed, antibiotics should be continued for at least 5 days. So
long as the patient improves clinically, another paracentesis procedure
is not necessary. If the patient does not improve by 48 hours,
paracentesis should be repeated; if the ascitic PMN count has not fallen
by 50%, antibiotic coverage should be broadened and other causes of

peritonitis should be considered. Secondary bacterial peritonitis should

also be considered when the initial ascitic fluid PMN count is greater
than 10,000 per mm3 , the ascitic fluid glucose level is less than 50 mg
per dL, or the culture is positive for more than one organism.
Prophylactic administration of oral nonabsorbable antibiotics has
proved effective in preventing SBP in high-risk patients. The goal of
prophylaxis is selective elimination of gram-negative bacteria from the
intestinal flora. The most effective antibiotic studied has been
norfloxacin (Noroxin), * in a dose of 400 mg daily. High-risk patients
who may warrant prophylactic therapy include those with a previous
episode of SBP, with a recent episode of gastrointestinal hemorrhage,
or awaiting liver transplantation, and possibly those with an extremely
low ascitic fluid total protein concentration (less than 1 gram per dL).
Because candidal overgrowth is a possible side effect of prophylactic
antibiotic therapy, liver transplant candidates on long-term antibiotic
prophylaxis should also receive either clotrimazole (Mycelex),
466
1 troche tid, or nystatin (Mycostatin), swish and swallow 10 mL, tid.
Hepatorenal Syndrome
Progressive renal failure often heralds the cirrhotic patient's demise and
may be the primary cause of death. Although the pathogenesis is not
completely understood, hepatorenal syndrome (HRS) results from renal
cortical hypoperfusion related to the complex hemodynamic alterations
induced by cirrhosis. Portal hypertension results in decreased systemic
vascular resistance and relative hypovolemia as well as increased renal
vascular resistance due to increased sympathetic nervous system output
and elevated levels of various renal arteriolar vasoconstrictors,
including angiotensin, vasopressin, endothelin, and leukotrienes. The
effect of these renal vasoconstrictors may be offset by secretion of
prostaglandins and kallikreins, but the compensatory mechanisms may
eventually be overwhelmed, resulting in HRS. Nevertheless, kidneys
from patients with HRS remain structurally normal and can even be
used as donor organs for transplantation.
Major criteria for the diagnosis of HRS include advanced hepatic
failure and portal hypertension; a low glomerular filtration rate, as
indicated by a serum creatinine level of greater than 1.5 mg per dL or a
24-hour creatinine clearance of less than 40 mL per minute; absence of
shock, sepsis, drug-induced nephrotoxicity, and gastrointestinal or
diuretic-induced fluid losses; lack of improvement in renal function
following diuretic withdrawal and expansion of plasma volume with
1.5 liters of normal saline; and absence of significant proteinuria or
ultrasonographic evidence of obstructive or parenchymal renal disease.
Additional criteria that support the diagnosis of HRS, but which may

not be present in all cases, include oliguria (urine volume of less than
500 mL per day), low urine sodium concentration (less than 10 mEq
per liter), and urine osmolality greater than plasma osmolality.
Fractional excretion of sodium is characteristically less than 1%.
The only effective treatment of HRS is liver transplantation. Short of
transplantation, intravascular volume should be optimized using blood
products or albumin, diuretics should be withheld, and any other
potentially aggravating drugs (such as nonsteroidal anti-inflammatory
drugs and aminoglycosides) should be avoided. Intravenous infusion of
"renal-dose" dopamine (1 to 3 mug per kg per minute) may help to
maintain urine output. Intravenous infusion of ornipressin (6
International Units per hour), a vasoconstrictor that does not increase
renal vascular resistance, has been reported to be useful in stabilizing
renal function in patients with HRS awaiting liver transplantation. The
use of TIPS may also be beneficial but has not been studied
systematically in patients with HRS. Dialysis may be necessary to
maintain intravascular volume, electrolyte levels, and acid-base
homeostasis, but survival is poor once this stage is reached.
Hepatic Hydrothorax
A pleural effusion due to portal hypertension, also known as hepatic

hydrothorax, affects 5 to 10% of patients with cirrhosis. Most cases
occur in patients with poorly controlled ascites, but a pleural effusion
may occur in the absence of ascites. The effusion is typically
right-sided and may be massive. Diagnosis requires exclusion of other
causes of pleural effusion, including cardiac or pulmonary disease.
Diagnostic thoracentesis should be performed on any new pleural
effusion in a cirrhotic patient; low ascitic fluid values of LDH and total
protein as well as negative results of microbiologic studies are
consistent with hepatic hydrothorax. In addition, analogous to SBP,
"spontaneous bacterial empyema" may occur owing to superinfection
of chronic hepatic hydrothorax. Therefore, diagnostic thoracentesis
should be considered in any cirrhotic patient with a pleural effusion
and encephalopathy or fever of unknown source.
The pathogenesis of hepatic hydrothorax appears to be related to small
diaphragmatic defects that allow passage of ascitic fluid into the
thorax. Therefore, treatment is generally the same as that for
ascites--namely, salt restriction, diuretics, and large-volume
paracentesis when appropriate. Serial therapeutic thoracenteses may be
performed, but rapid reaccumulation of fluid usually follows. Chemical
pleurodesis has been attempted, but results are poor owing to
reaccumulation of fluid before the desired effect can be achieved. In
refractory cases, the use of TIPS is often successful and has been
proposed as the treatment of choice. In selected cases, thoracoscopic
closure of diaphragmatic defects has been used with excellent results,
but this approach is unlikely to be necessary with the advent of TIPS.

Hepatic Encephalopathy
Cirrhosis may result in disturbance of CNS function by several

mechanisms. The diagnostic and pathophysiologic hallmark of hepatic
encephalopathy is hyperammonemia, which results from reduced
hepatic clearance of gut-derived ammonia from portal venous blood
because of both portosystemic shunting and hepatocellular dysfunction.
Effects of ammonia on the CNS include alterations in membrane
transport, decreases in oxidation-reduction potential and energy stores,
and depletion of the excitatory neurotransmitter glutamate in neurons.
When the serum ammonia concentration increases acutely,
accumulation of glutamine in astrocytes can result in lethal cerebral
edema. Other factors that may contribute to hepatic encephalopathy
include gut-derived mercaptans, elevated serum levels of aromatic
amino acids, increased sensitivity to gamma-aminobutyric acid
(GABA) and endogenous benzodiazepine-like compounds, zinc
deficiency, and deposition of manganese in the basal ganglia.
Clinical manifestations of hepatic encephalopathy vary considerably
depending in part on the severity of hepatic dysfunction and
portosystemic shunting
467
TABLE 123-3 -- Clinical Stages of Hepatic Encephalopathy

Findings
Personality
Level of and Neurologic EEG
Grade Consciousness Intellect Abnormalities Abnormalities
0 Normal No None None
abnormality
1 Reversed sleep Forgetful, Tremor, Slowing 5-cps
pattern, agitated, apraxia, triphasic
restless irritable, incoordination, waves
mildly poor
confused handwriting
2 Lethargic Disoriented Asterixis, Slowing 5-cps
to time, dysarthria, triphasic
amnesia, ataxia, waves
disinhibited, hyporeflexia
poor
judgment
3 Somnolent Disoriented Asterixis, Slowing 5-cps
to place, rigidity, triphasic
aggressive hyperreflexia waves

4 Comatose None Decerebrate Slow 2- to

3-cps delta
activity
Abbreviations: EEG = electroencephalogram; cps = counts per
second.
(Table 3) . Patients with mild encephalopathy may present clinically

with reversal of the sleep-wake cycle. Mild personality changes may
also be noted, but asterixis is often absent. Patients with moderately
severe encephalopathy experience either drowsiness or agitation, and
asterixis is usually present. Severe acute encephalopathy is marked by
somnolence, deep confusion, asterixis, and hyper-reflexia. At this
stage, airway protection is compromised, and endotracheal intubation
should be considered to prevent aspiration pneumonia. Coma may
ultimately supervene, with decerebrate posturing followed by death.
Patients with chronic encephalopathy often display personality
changes, dementia, memory loss, and occasionally movement
disorders.
Most episodes of acute hepatic encephalopathy are precipitated by a
potentially reversible factor. Precipitants include dehydration,
constipation, excessive dietary protein intake, hypokalemia, metabolic
alkalosis, sepsis (especially SBP), gastrointestinal bleeding, renal
failure, and use of benzodiazepines or other sedatives. Identification
and removal of the relevant precipitant(s) are the first steps in the
treatment of acute encephalopathy. Other causes of mental status
changes, such as subdural hematoma, alcohol withdrawal, Wernicke's
encephalopathy, hypoglycemia, and drug or alcohol intoxication,
should be ruled out.
Further treatment of acute encephalopathy centers on facilitating net
ammonia excretion. Dietary protein should be restricted, initially to as
little as 20 grams per day. Lactulose syrup should be administered in
doses of 30 mL PO every hour until diarrhea occurs and then every 4 to
6 hours as needed to achieve two to four loose bowel movements per
day. Lactulose is a nonabsorbable disaccharide that passes unchanged
into the colon, where it is fermented by colonic flora. It reduces
ammonia absorption by producing diarrhea and by acidifying the
colonic lumen, thereby trapping ammonia as nondiffusible ammonium
ion. In patients with severely depressed levels of consciousness, a
nasogastric tube may be required for the administration of lactulose. If
ileus is present, lactulose may be administered as a retention enema,
using 300 mL of lactulose suspended in 700 mL of water. In patients
who fail to respond to lactulose alone, oral neomycin, * 6 grams per day
initially and then 1 gram bid, or metronidazole (Flagyl), * 250 mg tid,
may be added. However, neomycin is absorbed by the intestine to
some extent in patients with cirrhosis and may thus cause
nephrotoxicity and ototoxicity.

Other treatments may be beneficial in selected cases. Administration of

the short-acting benzodiazepine antagonist flumazenil (Romazicon) *
leads to improvement in up to 40% of patients unresponsive to
lactulose, but the response is transient, the drug must be administered
parenterally, and the medication is expensive. Oral sodium benzoate or
phenylacetate may facilitate renal excretion of ammonia, and ornithine
aspartate, which enhances ureagenesis, may also lower the serum
ammonia level, but these agents are not widely used for treatment of
hepatic encephalopathy due to cirrhosis. Zinc deficiency, if present,
should be corrected, because several enzymes in the urea cycle are
zinc-dependent. Finally, Helicobacter pylori, a urea-splitting bacterium
that commonly colonizes the stomach, may add in small measure to the
generation of ammonia in the gastrointestinal tract, and anecdotal
reports suggest that encephalopathy may lessen following eradication
of H. pylori.
Prevention of recurrent encephalopathy requires extensive counseling
and close monitoring of the patient. Dietary protein should be
restricted, but usually to no less than 60 to 80 grams per day, so that
protein malnutrition does not worsen. Formal consultation with a
dietitian is essential to achieving this goal. Lactulose should be
continued at a dose of 15 to 30 mL bid or tid as needed to achieve two
or three loose stools per day. Excessive diuresis and use of sedative
medications should be avoided.
Hepatopulmonary syndrome (HPS) is defined as the presence of an

increased alveolar-arterial oxygen gradient and intrapulmonary
vascular dilatations in a patient with chronic liver disease. Affected
persons
468
may become markedly debilitated owing to severe hypoxemia (partial

pressure of oxygen [Pa O2 ] of less than 50 mm Hg) and may require
continuous oxygen supplementation. Typically, oxygenation worsens
as the patient moves from the supine to the standing position
(orthodeoxia). The pathophysiology of the syndrome relates to
intrapulmonary right-to-left shunting through vascular dilatations of
varying size. Type I HPS is characterized by diffuse small precapillary
dilatations scattered throughout the lung parenchyma, whereas type II
HPS is characterized by discrete, large pulmonary arteriovenous
malformations (AVMs). The diagnosis of HPS may be made by
contrast echocardiography, radiolabeled albumin scanning, or
pulmonary angiography, but only angiography differentiates type I
from type II HPS. Angiography is potentially therapeutic in the case of
type II HPS; large AVMs can sometimes be ablated by spring-coil

embolization. Unfortunately, for most patients with HPS there is no

effective therapy. However, recent experience indicates that HPS is
reversible after liver transplantation so long as pulmonary hypertension
has not developed. Therefore, patients with debilitating hypoxemia
from HPS should be considered candidates for liver transplantation.
Coagulopathy
The liver synthesizes fibrinogen and coagulation factors II, V, VII, IX,
and X; factors II, VII, IX, and X require vitamin K for their activity.
Deficiency of one or more of these clotting factors in cirrhosis
typically results in prolongation of the prothrombin time (PT); with
severe synthetic dysfunction, the activated partial thromboplastin time
(aPTT) may also be prolonged. Caution must be used in differentiating
the coagulopathy of liver disease from DIC; in both conditions, PT and
aPTT are elevated, the platelet count is decreased, fibrinogen levels are
decreased, and fibrin split products are elevated. The most useful
laboratory parameter for differentiating the two processes is the factor
VIII level; because factor VIII is not synthesized by the liver, levels are
preserved in cirrhosis but markedly decreased in DIC.
Additional hematologic consequences of chronic liver disease include
anemia, leukopenia, and thrombocytopenia. Hypersplenism due to
portal hypertension is the most common mechanism underlying these
abnormalities, although leukopenia and thrombocytopenia do not
resolve consistently after placement of a TIPS. Iron deficiency due to
chronic or recurrent gastrointestinal bleeding may also contribute to
anemia. Folic acid deficiency, which is common in alcoholic patients,
may also contribute to both anemia and thrombocytopenia.
Polycythemia should alert the clinician to the possibility of an occult
Coagulopathy should be corrected, if possible, in any cirrhotic patient
with bleeding or before an invasive procedure. In general, invasive
procedures, including percutaneous liver biopsy, should not be
performed if the PT is prolonged by more than 3 seconds or if the
platelet count is less than 75,000 per mm3 . Administration of vitamin
K, 10 mg SC daily for 1 to 3 days, can lower the PT in patients with
vitamin K deficiency, including those with predominantly cholestatic
disease such as PBC or primary sclerosing cholangitis, those who are
malnourished (especially alcoholic patients), and those receiving
chronic antibiotic therapy. However, vitamin K does not reverse
coagulopathy due to impaired hepatocellular function. Moreover,
because the effect of vitamin K depends on synthesis of new clotting
factors, the onset of action is slow. In most patients, therefore,
fresh-frozen plasma (FFP) is needed to reverse significant
coagulopathy, and FFP should be administered in doses of 1 to 2 units
per hour IV until bleeding stops or the PT is corrected. If coagulopathy
fails to respond to large quantities of FFP and if the fibrinogen level
falls below 100 mg per dL, cryoprecipitate should be administered.

Platelet infusions are also indicated in patients with active bleeding or

before invasive procedures when the platelet count is less than 75,000
per mm3 . Desmopressin (DDAVP), in a dose of 0.3 mug per kg
diluted in 50 mL of normal saline and infused over 30 minutes, may
also be used to improve the bleeding time and aPTT; the drug, an
analogue of vasopressin, increases factor VIII and von Willebrand
factor activities.
Blood products carry a risk, albeit small, of viral transmission or
transfusion reactions, as well as overexpansion of the intravascular
space and precipitation of variceal bleeding. Therefore, they should not
be administered to nonbleeding patients with cirrhosis and laboratory
evidence of coagulopathy.
Cirrhosis induces a relatively immunocompromised state owing to

impaired RE function. Malnutrition, when present, further impairs both
humoral and cell-mediated immunity. Therefore, several steps should
be taken to protect cirrhotic patients from infection. Polyvalent
pneumococcal vaccine (Pneumovax) should be administered to all
patients with cirrhosis. Hemophilus b conjugate vaccine is also
recommended by some authorities, although Haemophilus influenzae
serotype b is an uncommon cause of respiratory infections in adults.
Influenza vaccine should be administered yearly. Finally, vaccination
against hepatitis A (Havrix or Vaqta) and hepatitis B (Engerix-B or
Recombivax HB) should be offered to seronegative patients, because
acute viral hepatitis is more likely to lead to hepatic failure in a patient
with already compromised hepatic function.
Patients with cirrhosis of any cause are at increased risk of

hepatocellular carcinoma (HCC). The risk varies with the cause of
cirrhosis; patients with chronic viral hepatitis and hemochromatosis are
at
469
especially high risk, while those with primary biliary cirrhosis and
Wilson's disease are at lowest risk. A diagnosis of HCC should be
considered in any patient with cirrhosis in whom clinical
decompensation is observed. Specifically, HCC may underlie variceal
bleeding, worsening ascites, encephalopathy, or cachexia in a
previously well-compensated patient. We screen all patients with
cirrhosis for HCC using ultrasonography and serum alpha-fetoprotein
testing every 6 months, although the cost-effectiveness of this strategy
is uncertain. To evaluate focal hepatic lesions detected by ultrasound
studies, magnetic resonance imaging and computed tomography (CT)
with arterial-phase imaging after the administration of IV contrast are

useful. Definitive diagnosis may be obtained by ultrasound- or

CT-guided needle biopsy. In patients with cirrhosis, resection of the
tumor may be precluded by limited hepatic reserve. Therefore, liver
transplantation should be considered in any patient with cirrhosis and
HCC, provided that the tumor is smaller than 5 cm in diameter and
confined to the liver or, if several lesions are present, the largest is no
more than 3 cm in diameter and there is no invasion of the portal vein.
In patients who are not candidates for surgery or transplantation,
percutaneous ethanol injection to ablate small tumors may be
considered. For large, unresectable tumors, chemoembolization via the
hepatic artery may provide palliation.
ORTHOTOPIC LIVER TRANSPLANTATION

For cirrhosis of most causes, orthotopic liver transplantation is the only
curative therapy. Proper timing of transplantation is crucial; patients
with far-advanced cirrhosis and several serious complications have a
higher mortality rate than do patients with relatively compensated
cirrhosis who undergo liver transplantation, whereas patients with fully
compensated cirrhosis do not require transplantation. In general,
patients with cirrhosis should be referred to a transplant center at the
first sign of decompensation or when death due to liver disease is
expected within several years.
In preparation for liver transplantation, the following tests should be
performed: identification of ABO blood type, determination of human
immunodeficiency virus (HIV) status, measurement of
cytomegalovirus (CMV) and varicella-zoster virus titers, tuberculin
skin test, abdominal ultrasound scan with Doppler flow studies,
pulmonary function tests, and measurement of serum alpha-fetoprotein
level. If HCC is present, CT scanning of the chest and abdomen is
necessary to rule out extrahepatic spread. In addition, formal
psychiatric evaluation should be obtained. In patients who are
CMV-seronegative, filtered or CMV-safe blood products should be
used to avoid transfusion-borne infection, because CMV infection has
been shown to increase mortality after liver transplantation.
Absolute contraindications to liver transplantation include sepsis,
extrahepatic malignancy, advanced cardiac or pulmonary disease,
acquired immunodeficiency syndrome (AIDS), and inability to
understand the procedure and post-transplant regimen. Relative
contraindications include age greater than 70 years, active alcohol or
drug use, HCC, mesenteric vein thrombosis, hepatobiliary sepsis, HIV
infection, pulmonary hypertension, advanced chronic renal
insufficiency, and severe malnutrition.
Five-year survival rates after liver transplantation are approximately
70%, and most post-transplant deaths occur in the first 6 months and
are due to perioperative complications, acute graft rejection, or sepsis.
After transplantation, patients are maintained on lifelong

immunosuppression. The standard regimen consists of cyclosporine or

tacrolimus (FK 506, Prograf), azathioprine, and prednisone. Patients
must be willing and able to comply with a complicated medical
regimen and intensive follow-up. The original liver disease may recur
in the transplanted organ. In particular, the risk of recurrent hepatitis B
after liver transplantation is high, but can be reduced with hepatitis B
immune globulin (HBIG) or treatment with the oral nucleoside
analogue lamivudine. Hepatitis C almost invariably recurs after
transplantation but follows a relatively benign course in many patients.


Rakel: Conn's Current Therapy 1999, 51st ed., Copyright 1999 W. B. Saunders Company
Section 6 - THE DIGESTIVE SYSTEM
457
Chapter 122 - CHOLELITHIASIS AND

CHOLECYSTITIS
ROBERT V. REGE M.D.

Northwestern University Medical School and Veterans Affairs Chicago Health Care System,
Lakeside Division Chicago, Illinois
Gallstones are very common, occurring in 11 to 15% of women and 3 to 11% of men younger than age
50. They are even more common in individuals with obesity, diabetes mellitus, ileal disease, or a family
history of gallstones; in certain ethnic groups (e.g., Native Americans); and in patients on parenteral
nutrition or undergoing rapid weight loss. Gallstones can cause severe abdominal pain and lead to serious
complications, such as acute cholecystitis, obstructive jaundice, pancreatitis, gallstone ileus, and
ascending cholangitis. Gallstones account for more than 500,000 operations and several billion dollars in
health care expenditures in the United States each year.
Although the general principles of gallstone management have not changed appreciably, methods of
treatment have. Laparoscopic cholecystectomy, lithotripsy, gallstone dissolution, endoscopic retrograde
management of bile duct stones, and percutaneous approaches to the biliary tract now play a role in the
treatment of gallstones, and a comprehensive approach to biliary tract disease requires a team of
well-trained surgeons, interventional gastroenterologists, and interventional radiologists. Moreover, the
evolution of laparoscopic cholecystectomy has necessitated the reeducation and retraining of surgeons.
Nonetheless, cholecystectomy remains the treatment of choice for gallstones, and gallstone disease
remains the purview of the surgeon.
TYPES OF GALLSTONES
Gallstones are classified as cholesterol, mixed, and pigment gallstones by gross and compositional
analysis. Selected patients with single, nearly pure cholesterol (but not pigment) gallstones may be
treated with gallstone dissolution and/or extracorporeal lithotripsy. Cholesterol and mixed gallstones are
both composed primarily of cholesterol. Together, they account for 80% of gallstones in the West. Two

types of pigment gallstones, black pigment and calcium bilirubinate, account for 10 to 27% of all
gallstones. Black pigment gallstones, composed mainly of an amorphous bilirubin polymer and calcium
salts, develop when there is excessive bilirubin in bile. Most patients with pigment gallstones do not have
predisposing factors, but black pigment gallstones are more common in the elderly and in patients with
hemolytic anemia or cirrhosis. Calcium bilirubinate stones usually form behind biliary strictures or in
bile containing bacteria or parasites. They are mainly composed of calcium bilirubinate, free fatty acids,
and up to 10% cholesterol.
Asymptomatic Gallstones
Most gallstones remain asymptomatic for many years; symptoms and complications develop in only 1 to
2% of patients per year. Observation of patients with asymptomatic gallstones is currently recommended
since the risk of observation is less than the risk of prophylactic surgery. However, certain patients with
gallstones, including children, patients with congenital hemolytic anemia, and those with large (>2.5 cm)
gallstones or nonfunctioning gallbladders, may benefit from prophylactic cholecystectomy. In addition,
symptoms develop in more than 36% of morbidly obese patients who have undergone barosurgery, and
20% of colectomy patients with gallstones develop symptoms within 5 years. Prophylactic
cholecystectomy adds no significant morbidity or mortality to either of these operations, and it should be
considered for these patients during their primary operation. Incidental cholecystectomy during other
abdominal operations may not be prudent in some situations (e.g., when prosthetic material is required)
and must be left to the discretion of the surgeon. In the past, it was taught that diabetic patients were
more likely to develop gallstones, complications of gallstones, and complications after emergent/urgent
biliary tract surgery. However, recent studies have demonstrated similar outcomes in diabetic and
nondiabetic patients, and prophylactic operation for diabetic patients is no longer recommended.
Symptomatic Gallstones
Right upper quadrant and epigastric pain 15 to 60 minutes after meals is quite specific for gallstone
disease. The pain often occurs after ingesting fatty foods, onions, cabbage, spicy foods, or dairy products.
It lasts from 20 minutes to several hours; pain lasting longer suggests acute cholecystitis or acute
pancreatitis. Severe episodes of pain associated with nausea and vomiting have been called "biliary
colic,"
458
although the pain is constant, not colicky. Patients with symptomatic gallstones are said to have chronic
cholecystitis. They can expect continued episodes of pain, which often increase in frequency and
severity. Complications develop more frequently in symptomatic than in asymptomatic patients, and as
many as 40 to 50% require operative treatment within 2 to 5 years. The risks incurred by merely
following symptomatic patients is higher than the risk of operation, and cholecystectomy is clearly
indicated.
Patients who present with vague, mild pain, indigestion, flatulence, and nausea without vomiting are now
classified as being mildly symptomatic, although these symptoms are not specific for gallstones, as they
can be caused by many other disorders of the gastrointestinal tract. Mildly symptomatic disease is much
like asymptomatic disease; only 1 to 2% of patients per year require cholecystectomy. Operation should

be recommended with care if only mild symptoms are present. Cholecystectomy may, however, give
relief in 50 to 70% of patients with persistent symptoms, especially if they significantly interfere with
lifestyle. Patients must be warned that cholecystectomy may not relieve their symptoms as they may be
due to other disorders.
CHRONIC CHOLECYSTITIS
Diagnosis and treatment of chronic cholecystitis are frequently straightforward. Biliary colic combined
with the presence of gallstones on an imaging study is sufficient for the diagnosis. Ultrasonography is the
study of choice, successfully demonstrating gallstones in 90 to 95% of patients. Plain films of the
abdomen and computed tomography (CT) scan are not very sensitive, demonstrating gallstones in only
20% of patients. However, positive findings on these studies are very reliable. In some individuals,
gallstone disease mimics another abdominal disease, or other gastrointestinal disorders mimic gallstone
disease, making diagnosis challenging. Extensive testing may be required to exclude gastroesophageal
reflux, peptic ulcer disease, hepatitis, pancreatitis, intestinal pathology, and malignant tumors of the
stomach, bile duct, duodenum, or pancreas. Chronic acalculous cholecystitis and cholesterolosis of the
gallbladder may cause symptoms identical to those of gallstones, but specific tests do not exist for these
disorders. An oral cholecystogram may reveal no, or only faint, visualization of the gallbladder,
demonstrating impaired gallbladder function. Impaired gallbladder emptying measured with ultrasound
or gallbladder scintigraphy (HIDA scan) after administration of cholecystokinin may also suggest
gallbladder disease in some of these patients.
Cholecystectomy is the treatment of choice for patients with symptomatic cholelithiasis. This operation
carries a mortality rate of less than 0.3% and relieves symptoms in 95% of patients. When performed by
experienced surgeons, laparoscopic cholecystectomy can be successful in more than 95% of patients, and
the rate of complications, including bile duct injury, is comparable to that with open cholecystectomy.
The laparoscopic approach is not safe or prudent in 5 to 10% of patients, and the surgeon should convert
to open operation. Conversion of the operation to open cholecystectomy should not be considered a
failure.
Common bile ducts stones are present in as many as 10 to 20% of patients undergoing cholecystectomy.
Only about 4% of patients without any preoperative risk factors will have "silent" common bile duct
stones. Intraoperative cholangiography can be performed either routinely or selectively with the
laparoscope, but it should always be used in patients at high risk for choledocholithiasis (Table 1) and
whenever biliary anatomy is not well-defined by operative dissection. In contrast, dissolution therapy and
extracorporeal shock wave lithotripsy eliminate gallstones in less than 50% of selected patients, and at
least half of these patients experience recurrence. If used at all, these methods of treatment should be
limited to patients with single, "nearly pure" cholesterol gallstones less than 2 cm in diameter.
ACUTE CHOLECYSTITIS
Acute cholecystitis is the most common complication of gallstones, occurring in about 10% of patients. It
is caused by cystic duct obstruction by a gallstone, tumor, or swelling. Bacteria grow in bile, and
infection develops in the gallbladder wall. If unchecked, acute cholecystitis can lead to gallbladder
perforation, pericholecystic abscess, or peritonitis. Patients present with unremitting right upper quadrant
pain, fever, elevated white blood cell count, nausea and vomiting, and gallstones. Ultrasonography is the

most common confirming test in demonstrating the presence of gallstones. It may also show a thickened
gallbladder wall, a gallbladder "rim" sign, or an ultrasonographic Murphy's sign (pain when the
gallbladder is compressed with the ultrasound probe). Gallbladder scintigraphy (HIDA scan) is helpful in
patients with less typical signs and symptoms. HIDA scans are very sensitive and specific for acute
cholecystitis as the gallbladder cannot be visualized in 98% of patients due to cystic duct obstruction.
Gallbladder visualization occurs in a few patients with acute acalculous cholecystitis (false negative), and
the test is not reliable in critically ill patients and patients on parenteral nutrition who have not emptied
their gallbladders. Percutaneous aspiration of
TABLE 122-1 -- Factors That Increase the Risk of Choledocholithiasis
History of acute jaundice
History of pancreatitis
Dilated biliary tree
Ultrasound--Dilated intrahepatic ducts Common bile duct >7 mm
CT scan--Dilated intra- and extrahepatic ducts
Elevated serum bilirubin, liver enzymes, or amylase
Episode of acute cholangitis
459
the gallbladder with Gram's stain examination and cultures may be helpful in these patients.
Patients with acute cholecystitis should be admitted to the hospital, given nothing by mouth, and treated
with broad-spectrum parenteral antibiotics. A second-generation cephalosporin or a combination of
ampicillin/sulbactam (Unasyn) and an aminoglycoside will suffice and should be continued
postoperatively. A nasogastric tube is placed if there is persistent vomiting. Operation is performed in the
next 24 to 72 hours. The complication rate and chance of successfully performing the procedure are not
improved by delaying cholecystectomy for 6 weeks. Perforation of the gallbladder occurs in some
patients despite clinical improvement, and 20 to 30% of patients develop recurrent symptoms while
waiting for operation. Delayed treatment is indicated in patients with medical conditions precluding
operation, especially if improvement is expected during the waiting period. Critically ill patients who do
not respond to medical therapy and who are not candidates for operation are treated with percutaneous
cholecystostomy.
Acute cholecystitis is not a contraindication to laparoscopic cholecystectomy. The procedure can be
performed safely, although conversion to open operation is necessary in 10 to 30% of patients. It is
prudent to convert to an open operation if dissection does not progress or if biliary tract anatomy cannot
be defined by intraoperative cholangiography. Laparoscopic cholecystectomy is more likely to be
successful when performed within 3 days of the onset of symptoms. Conversion rates are higher later
when edema is replaced by inflammatory tissue and fibrosis or when gangrene is present.
About 5% of patients with acute cholecystitis do not have gallstones. Often these patients are critically

ill. The diagnosis is frequently delayed because signs and symptoms of cholecystitis cannot be elicited
due to the patient's medical condition, ultrasound may be unremarkable, and the HIDA scan is not
reliable in patients who have not eaten and emptied their gallbladder. Urgent cholecystectomy or
percutaneous cholecystostomy is warranted in patients with acalculous cholecystitis as this disease may
rapidly progress to gangrene and perforation.
OTHER COMPLICATIONS OF GALLSTONES

Gangrenous cholecystitis, gallbladder empyema, emphysematous cholecystitis, and gallbladder
perforation represent advanced presentations of acute cholecystitis. Patients with these complications
often present with higher than expected white blood cell counts, severe abdominal pain often associated
with abdominal guarding or rebound tenderness, high fever, and failure to respond to medical therapy.
Abdominal radiography or CT scan shows air in the gallbladder wall or lumen with emphysematous
cholecystitis. Ultrasound or CT scan demonstrates pericholecystic fluid, subhepatic fluid, or abscess with
gallbladder perforation. Urgent abdominal exploration is indicated as these complications are associated
with significant mortality and morbidity.
Acute gallstone pancreatitis results when a gallstone passes into and through the common bile duct.
Attacks range from very mild and transient to life-threatening. Patients present with severe epigastric
pain, left upper quadrant pain, elevated serum amylase or lipase levels, gallstones, and no other obvious
causes for pancreatitis. If the attack resolves quickly, only 20 to 30% of patients will have a retained
gallstone in their common bile duct. Cholecystectomy with intraoperative cholangiography in these
patients is reasonable from both a medical and a cost perspective. The patients with retained stones may
have them removed at surgery or postoperatively using endoscopic retrograde cholangiopancreatography
(ERCP). Common bile duct stones are present in 50 to 80% of patients with persistent symptoms,
elevated liver tests, and jaundice, and this group of patients should undergo ERCP and stone removal
before cholecystectomy.
Choledocholithiasis also causes obstructive jaundice. Gallstones are more likely to cause both jaundice
and pain; bile duct stricture and malignancies are more likely to cause "silent" jaundice. Ultrasonography
is used to document stones in the gallbladder and dilated intra- and extrahepatic ducts. It will
occasionally demonstrate the stone in the common bile duct. The CT scan also demonstrates dilated
intra- and extrahepatic bile ducts and excludes mass lesions in the pancreas. The cause of jaundice is best
ascertained by directly imaging the biliary tract. ERCP is the procedure of choice because it is successful
in 90 to 95% of patients and has low complication rates. Percutaneous transhepatic cholangiography is
reserved for patients in whom ERCP cannot visualize the proximal biliary tract. It is successful and safe
in more than 90% of patients with dilated intrahepatic ducts but is more likely to result in complications,
such as bile leakage and hemorrhage.
Choledocholithiasis is most often treated endoscopically before operation. Endoscopic approaches are
successful in more than 95% of patients when performed by an experienced interventional
gastroenterologist. Gallstones that cannot be removed because they are too large or because of anatomic
considerations are removed operatively. Common duct exploration currently can be performed using
laparoscopic as well as open techniques.
Gallstone ileus refers to intestinal obstruction caused by a gallstone that has eroded from the gallbladder
into the intestinal tract. The gallstone usually causes obstruction at the ileocecal valve or in the sigmoid

colon. By necessity, a fistula is present between the gallbladder and the stomach, duodenum, or colon.
Besides dilated loops of bowel, abdominal films demonstrate air in the biliary tract. Occasionally, the
gallstone can be seen if it is calcified. Treatment should relieve the bowel obstruction by removing the
gallstone. The fistula is treated later,
460
although this is often not necessary if the patient is asymptomatic.


Rakel: Conn's Current Therapy 1999, 51st ed., Copyright 1999 W. B. Saunders Company
Chapter 123 - CIRRHOSIS
PETER M. ROSENBERG M.D.

LAWRENCE S. FRIEDMAN M.D.
Massachusetts General Hospital Boston, Massachusetts
Cirrhosis is the final common pathway in the course of a variety of chronic, progressive liver diseases.
Strictly defined, cirrhosis is a histologic diagnosis, based on the presence of hepatocellular necrosis,
fibrosis, and regenerative nodules. From a clinical point of view, however, cirrhosis represents a
constellation of symptoms, signs, and laboratory abnormalities reflecting underlying hepatocellular
dysfunction, portal hypertension, and portosystemic shunting. Typically, cirrhosis follows a slow,
insidious course and often culminates in dramatic and fatal complications. The cornerstone of
management is the prevention and treatment of these complications.
The prognosis for patients with cirrhosis varies according to clinical stage and etiology. It is clearly best
for patients with well-compensated disease. For example, in patients with well-compensated cirrhosis
due to chronic hepatitis C, the 10-year survival rate has been estimated to be approximately 80%.
Regardless of etiology, prognosis correlates with the Child-Pugh classification (Table 1) . Mean survival
periods for patients with class A, class B, and class C cirrhosis are 40, 32, and 8 months, respectively.
Whereas patients with class A cirrhosis may benefit from treatment of the underlying chronic liver
disease, patients with class B or C cirrhosis should generally be evaluated for liver transplantation.
DIAGNOSIS
In clinical practice, the diagnosis of cirrhosis is usually based on the presence of suggestive symptoms,
signs, and laboratory abnormalities. Possible symptoms and signs include fatigue, nausea, vomiting,
diarrhea, amenorrhea, easy bruising, jaundice, increased abdominal girth, leg swelling, evidence of upper
gastrointestinal bleeding, and confusion. Other physical signs include jaundice, spider
TABLE 123-1 -- Child-Pugh Classification of Cirrhosis
Score
Feature 1 2 3
Encephalopathy (stage) 0 1-2 3-4
Ascites Absent Slight Poorly controlled
Bilirubin (mg/dL) <2.0 2.0-3.0 >3.0
Albumin (gm/dL) >3.5 2.8-3.5 <2.8

Prothrombin time 1.0-4.0 4.0-6.0 >6.0

(seconds prolonged)
Each feature is assigned 1, 2, or 3 points.
Class A: 5-6 points Class B: 7-9 points Class C: 10-15 points
angiomata, palmar erythema, ascites, peripheral edema, a firm and nodular liver edge, splenomegaly,
caput medusae, gynecomastia, testicular atrophy, and gastric or esophageal varices (on endoscopic
examination). Characteristic laboratory abnormalities include an elevated serum bilirubin level,
decreased serum albumin level, elevated serum globulin levels, an elevated prothrombin time (PT), and
thrombocytopenia.
If at least several of these features are present in a patient, the diagnosis of cirrhosis is usually secure.
Their absence, however, does not exclude a diagnosis of cirrhosis, especially in patients with
compensated disease. In such cases, the disease is subclinical, and liver biopsy may be necessary for
definitive diagnosis. Ultrasonography with Doppler flow studies may show abnormalities in liver size,
surface patterns, and homogeneity and may detect splenomegaly, portal vein enlargement, caudate lobe
enlargement, and the presence of collateral circulation, all of which are suggestive of cirrhosis. However,
ultrasonography is not as sensitive as liver biopsy and fails to provide information on the underlying
cause of cirrhosis. Various serum markers of fibrosis have been reported to aid in the noninvasive
diagnosis of cirrhosis, and recently, an elevated serum hyaluronate level has been suggested to be a
sensitive and specific marker. However, this test is not widely available and has not been validated in a
large population.
It should be noted that overt clinical signs of portal hypertension, portosystemic shunting, and hepatic
synthetic dysfunction may also be seen with acute alcoholic hepatitis and, less commonly, with flares of
chronic hepatitis B, in the absence of cirrhosis. Without a liver biopsy, distinguishing decompensation
due to one of these acute processes (or both) from cirrhosis may be impossible. When percutaneous liver
biopsy is precluded by marked coagulopathy (PT prolonged by at least 3 seconds or a platelet count of
75,000 per mm3 or less), a transjugular approach may be used for biopsy.
CAUSES OF CIRRHOSIS: EVALUATION AND SPECIFIC THERAPY

The causes of cirrhosis are numerous (Table 2) . In the United States, the majority of cases can be
attributed to alcohol or hepatitis C virus infection, or both. Despite the discovery of hepatitis C virus in
1989, many cases of cirrhosis are still considered cryptogenic. It has recently been suggested that
nonalcoholic steatohepatitis (NASH) may account for many of these cases, because cryptogenic cirrhosis
is found in a substantial number of obese diabetic patients with hypertriglyceridemia.
Often, the cause of cirrhosis can be determined by the history and noninvasive testing alone. Standard
work-up should include serologic tests for viral hepatitis, including assays for hepatitis B surface antigen
and antibody to hepatitis C; determination of serum iron, total iron-binding capacity (TIBC), ferritin,
ceruloplasmin, and alpha1 -antitrypsin levels; assays for antinuclear, anti-smooth muscle, and
antimitochondrial antibodies; and, as noted previously, an abdominal ultrasound scan with Doppler flow
studies. If the presence and cause of cirrhosis in a given patient can

461
TABLE 123-2 -- Causes of Cirrhosis in Adults in the United States

Chronic viral hepatitis Drugs
Hepatitis C Methotrexate
Hepatitis B (with or without delta co-infection) Isoniazid
Methyldopa
Alcohol Amiodarone
Cryptogenic cirrhosis Halogenated hydrocarbons
Nonalcoholic steatohepatitis Hypervitaminosis A
Autoimmune diseases Nonviral infections
Primary biliary cirrhosis Schistosomiasis
Autoimmune hepatitis Brucellosis
Primary sclerosing cholangitis Syphilis
Metabolic diseases Nonautoimmune biliary disease
Hemochromatosis Chronic biliary obstruction
Wilson's disease Biliary atresia
Alpha1 -antitrypsin deficiency Cystic fibrosis
Galactosemia Chronic graft-versus-host disease
Hereditary fructose intolerance Vascular disorders
Tyrosinemia Chronic Budd-Chiari syndrome
Glycogen storage disease (types III and IV) Veno-occlusive disease
Abetalipoproteinemia Chronic heart failure
Sarcoidosis
Jejunoileal bypass
be inferred from results of this noninvasive work-up a liver biopsy may not be needed. If, however, the
results are negative or ambiguous, liver biopsy should be performed. In suspected primary sclerosing
cholangitis, endoscopic retrograde cholangiopancreatography should be performed; anti-neutrophil
cytoplasmic antibody (p-ANCA) may be detected in 70% of affected persons.
Knowledge of the exact cause of cirrhosis in a patient may allow treatment of the underlying disorder.
Although cirrhosis is generally thought to be irreversible, treatment of the underlying disease may slow
the progression of cirrhosis, ameliorate the clinical course, reduce the risk of hepatocellular carcinoma,

and, in rare cases, lead to regression of cirrhosis. In general, such treatment is most effective and best
tolerated when initiated in the precirrhotic or Child class A phase. Regardless of the cause of cirrhosis,
treatment with colchicine to inhibit hepatic fibrogenesis may be considered. In a long-term study, 5- and
10-year survival rates for patients with early-stage cirrhosis were found to be significantly higher for
those who received oral colchicine, in a dose of 1 mg daily, than for those who received placebo. In the
future, more specific antifibrotic agents will probably become available.
Alcoholic Cirrhosis
Alcoholic cirrhosis typically results from repeated bouts of alcoholic hepatitis, either overt or subclinical.
Acute alcoholic hepatitis is generally self-limited, and abstinence can clearly prevent progression to
cirrhosis. Treatment with corticosteroids may be beneficial for selected patients with severe alcoholic
hepatitis. Even once cirrhosis develops, abstinence from alcohol improves the prognosis. In patients with
decompensated alcoholic cirrhosis who continue to drink, the 5-year survival rate is less than 50%, but in
those who remain abstinent, survival is improved and the incidence of liver-related complications
decreased. Both ascites and esophageal varices have been observed to regress and portal pressure to
decrease with abstinence. Therefore, every effort should be made to enter patients with alcoholic
cirrhosis into organized alcohol counseling programs.
Nutrition also plays a role in therapy for the alcoholic patient with cirrhosis. Adequate caloric intake
should be ensured, and vitamin and electrolyte deficiencies should be corrected. It is generally advisable
to maintain patients on a multivitamin, folic acid 1 mg, and thiamine 100 mg daily. Hypomagnesemia
and hypophosphatemia are common and should be corrected, initially with intravenous therapy.
Cirrhosis Due to Chronic Viral Hepatitis
Worldwide, chronic hepatitis B is probably the most common cause of cirrhosis and liver-related death.
The active, or replicative, phase of chronic hepatitis B is characterized by the presence in serum of the
hepatitis B e antigen (HBeAg) and hepatitis B virus (HBV) DNA. Standard treatment for patients with
chronic hepatitis B is administration of interferon alfa-2b (Intron-A), 5 million units subcutaneously (SC)
daily or 10 million units SC three times per week, for 4 months (see the article "Acute and Chronic Viral
Hepatitis" for more detail on treatment). Response to interferon in hepatitis B is typically accompanied
by a mild flare of hepatitis. In HBeAg-positive patients with compensated cirrhosis, response to
interferon treatment has been shown to be associated with improved survival and decreased incidence of
hepatic decompensation. However, interferon should not be used in patients with decompensated liver
disease, because the treatment-induced flare of hepatitis may precipitate liver failure. More promising for
patients with advanced liver disease caused by chronic hepatitis B are nucleoside analogues such as
lamivudine, although these drugs are not yet approved for treatment of hepatitis B. These drugs may also
prove effective in diminishing the risk of recurrent hepatitis B after liver transplantation.
Interferon therapy, alone or with ribavirin, has not been shown to improve the prognosis of hepatitis
C-related cirrhosis. In fact, the presence of cirrhosis predicts a poor response to interferon, although
therapy may be considered in patients with well-compensated cirrhosis and active inflammation on liver
biopsy. There is some evidence that such therapy reduces the risk of hepatocellular carcinoma.
Unfortunately, there are as yet no therapeutic alternatives to interferon for treatment of hepatitis C.
Therefore, the only treatment for advanced hepatitis C-related cirrhosis is liver transplantation, if
indicated.

462
Primary biliary cirrhosis (PBC) is a chronic cholestatic disease caused by immunologic reactivity against
the epithelium of the intrahepatic bile ductules. The disorder typically affects middle-aged women and
follows a slow but progressive course, characterized eventually by pruritus, jaundice, steatorrhea,
fat-soluble vitamin deficiencies, and ultimately end-stage liver disease. Laboratory hallmarks of PBC are
an elevated serum alkaline phosphatase level, the presence of antimitochondrial antibody, and
hypercholesterolemia. Prognosis is closely related to the serum bilirubin level as well as to the patient's
age, presence of edema, prolongation of PT, and variceal bleeding.
At present, the mainstay of treatment for PBC is administration of ursodeoxycholic acid, or ursodiol
(Actigall), * given orally (PO) in a daily dose of 10 to 15 mg/kg, generally in divided doses.
Ursodeoxycholic acid is a hydrophilic bile acid that is absorbed from the gastrointestinal tract after oral
ingestion and displaces more toxic bile acids from the enterohepatic circulation and thus from the bile
pool. It also improves bile flow and appears to modulate the humoral immune response. Treatment with
ursodeoxycholic acid has been shown to reduce pruritus, improve biochemical parameters (especially
serum bilirubin levels), delay progression to cirrhosis, and improve survival, as measured by time to liver
transplantation. However, the drug has not been shown to induce histologic improvement in patients with
PBC. The survival benefit has been shown even in patients with stage IV disease (i.e., cirrhosis).
Therefore, ursodeoxycholic acid is clearly indicated in the treatment of cirrhosis due to PBC.
Other pharmacologic agents for treatment of PBC that have been studied include colchicine, * prednisone,
D-penicillamine, * azathioprine, * cyclosporine, * chlorambucil, * and methotrexate. * Colchicine, * 0.6 mg
twice daily (bid), may slow the rate of progression to cirrhosis but has not been shown to improve
survival. Dramatic results have been reported with long-term methotrexate therapy (in weekly "pulses" of
15 mg) in a small series of patients, and complete histologic remission was observed in several patients.
Further trials are needed to define the subgroup likely to benefit from methotrexate therapy, and the drug
should still be considered investigational. Trials of the other agents have been disappointing, and the
toxic effects of these drugs appear to outweigh their benefits.
Several unique complications of PBC may require specific treatment. Pruritus may be treated with
cholestyramine (Questran) in doses of 4 grams PO in water or juice, bid or three times daily (tid);
colestipol, 5 grams bid or tid; or oral antihistamines such as hydroxyzine (Atarax), in a dose of 25 mg
four times a day as needed. Additional treatments for refractory pruritus include rifampin (Rifadin), * 300
mg PO bid; phenobarbital, * 2 to 4 mg per kg per day, with a target serum level of 10 mum per mL; and
phototherapy. The oral opiate antagonist naltrexone (ReVia) has shown particular promise in the
treatment of pruritus associated with cholestasis. Patients with PBC also require supplementation to
compensate for malabsorption of fat-soluble vitamins. In general, calcium, 1000-1500 mg daily, and
vitamin D, 50,000 units PO per week, should be prescribed. Vitamin K, 10 mg SC daily for 2 to 3 days
and then 10 mg SC per month should be given to patients with a prolonged PT, and vitamin A, 25,000
units PO per day, to those with night blindness. The value of bisphosphonates and estrogen in the
prevention and treatment of osteoporosis associated with PBC is uncertain.

Hemochromatosis
Hereditary hemochromatosis is an autosomal recessive disorder characterized by excessive intestinal

absorption of iron, with resulting iron deposition in the liver, heart, pituitary, and pancreas and eventual
cirrhosis, dilated cardiomyopathy, panhypopituitarism, and diabetes mellitus. Initial testing for suspected
hemochromatosis should include determination of serum iron level, TIBC, and ferritin level; a fasting
transferrin saturation (defined as the serum iron level divided by the TIBC) of greater than 50% is
suggestive of hemochromatosis. However, the transferrin saturation and serum ferritin level are also
elevated in up to half of patients with chronic viral hepatitis, alcoholic liver disease, and NASH.
Definitive diagnosis of hemochromatosis has traditionally been made by liver biopsy with quantitative
hepatic iron determination, which should be performed in all patients with an elevated transferrin
saturation or serum ferritin level. The recent identification of a single mutation (HFE) in a human
leukocyte antigen (HLA)-like gene (HLA-H) has made it possible to perform genetic testing for
hemochromatosis. Two copies of the HFE gene mutation are present in more than 90% of patients with
biochemical and histologic evidence of hemochromatosis.
Identifying hemochromatosis as the underlying cause of cirrhosis in a given patient is of great
importance, because treatment with iron depletion therapy ameliorates the liver dysfunction and other
systemic manifestations of hemochromatosis, even in patients who already have cirrhosis. In rare cases,
phlebotomy therapy has been reported to reverse cirrhosis. Phlebotomy with removal of 500 mL of
whole blood should be performed weekly until the transferrin saturation falls below 50% and the ferritin
level below 50 ng per mL, or until the hemoglobin level falls to 10 mg per dL. Preoperative iron
depletion is particularly important in optimizing outcome after liver transplantation. For patients with
cirrhosis who are too ill or anemic to tolerate phlebotomy, iron chelation
Exceeds dosage recommended by the manufacturer.
463
therapy with deferoxamine (Desferal) may be considered but is less effective than phlebotomy. Finally,
first-degree relatives of patients with hemochromatosis should undergo screening with serum iron studies
and, if the proband has the HFE gene, genetic testing.
Wilson's Disease
Wilson's disease is an uncommon disease, inherited as an autosomal recessive trait, resulting in copper
accumulation in the liver, central nervous system (CNS), and kidneys. The disorder tends to manifest in
childhood or early adulthood. The diagnosis is made on the basis of detection of Kayser-Fleischer rings
in the cornea, a serum ceruloplasmin level of less than 20 mg per dL, an elevated urinary copper level,
and an elevated hepatic copper concentration on liver biopsy. However, Kayser-Fleischer rings may be
absent, particularly in patients without neuropsychiatric features of Wilson's disease. Medical therapy
with D-penicillamine (Cuprimine), 1-2 grams daily in four divided doses, is effective in preventing
copper accumulation but has not been shown to induce regression of established cirrhosis. Side effects of
D-penicillamine include rash, fever, bone marrow suppression, and glomerulonephritis and may
necessitate a reduction in dosage or use of an alternative agent such as trientine hydrochloride (Syprine),
750 to 2000 mg per day in two to four divided doses. Zinc may be used for maintenance therapy after

copper depletion has been achieved or in pregnant and presymptomatic patients. End-stage liver disease
due to Wilson's disease carries an especially poor prognosis, and liver transplantation should be strongly
considered in such cases; copper levels return to normal after transplantation.
Autoimmune Hepatitis
Classic (type 1) autoimmune hepatitis typically affects women in the third to fifth decades of life and is
often associated with high globulin levels and the presence of antinuclear and anti-smooth muscle
antibodies in serum. The disease is generally very sensitive to therapy with corticosteroids, which is
indicated in patients with symptoms, serum aminotransferase elevations at least 10 times the upper limit
of normal (or 5 times the upper limit of normal and a serum globulin level at least twice the upper limit
of normal), or histologic evidence of severe chronic hepatitis. Prednisone should be initiated at a dose of
40 to 60 mg daily for 1 to 2 weeks and then tapered over several weeks to a maintenance dose of 10 to 15
mg. Azathioprine (Imuran), * in a dose of 50 to 100 mg daily, is useful in allowing a lower dose of
prednisone in these patients. Maintenance therapy should be continued until biochemical and histologic
remission occurs, which may take more than 18 months. Thereafter, therapy may be discontinued, but
relapse is common. In patients who experience repeated relapses long-term suppressive therapy should
be used; in some cases, long-term therapy with azathioprine alone in a dose of 2 mg per kg per day, may
maintain remission. Treatment of autoimmune hepatitis should be initiated even if cirrhosis has already
supervened, and complete regression of established cirrhosis has been reported. Liver transplantation
should be considered for patients with progressive liver dysfunction that does not respond to
immunosuppressive therapy; clinical recurrence of autoimmune hepatitis is uncommon after
transplantation.
PREVENTION AND TREATMENT OF COMPLICATIONS

Variceal Bleeding
Bleeding from ruptured esophageal or gastric varices is one of the most common and potentially
catastrophic complications of cirrhosis. Prevention and treatment of variceal bleeding are addressed in
the article "Bleeding Esophageal Varices."
Ascites
Ascites is the most common and often the earliest major complication of cirrhosis and occurs in 50% of
patients within 10 years of the diagnosis of compensated cirrhosis. Pathogenetic factors include (1)
increased hydrostatic pressure in the portal circulation due to distortion of the hepatic vasculature, (2)
decreased plasma oncotic pressure due to impaired albumin synthesis, (3) transudation of lymph from the
liver surface due to hepatic sinusoidal obstruction, and (4) renal fluid retention mediated by activation of
the renin-angiotensin-aldosterone system. Ascites signals a poor prognosis in chronic liver disease and is
associated with a 2-year survival rate of approximately 50%. Morbidity may result directly from tense
abdominal distention; thus, findings may include abdominal discomfort, anorexia, early satiety,
respiratory compromise, and renal insufficiency. Massive ascites may also lead to the formation of
inguinal or umbilical hernias. When the hernias become large and the overlying skin becomes excoriated,
life-threatening rupture of the hernia may occur, necessitating urgent surgical repair. Superinfection of
ascites may result in spontaneous bacterial peritonitis (see later on), and translocation of ascitic fluid

across the diaphragm may lead to hepatic hydrothorax (see later on).
Diagnostic paracentesis should be performed in any patient with new or worsening ascites, as well as in
any patient with established ascites in whom abdominal pain, fever, or encephalopathy develops. When
performed with a 21-gauge needle, paracentesis is safe even in patients with significant coagulopathy.
The fluid obtained should be sent for cell and differential counts; total protein, albumin, amylase, and
triglyceride levels; and cytologic studies
464
and culture (by direct inoculation into blood culture bottles at the bedside). A serum albumin level should
be obtained on the same day. Ascitic fluid glucose and lactate dehydrogenase (LDH) concentrations and
a Gram's stain are useful only when secondary bacterial peritonitis is a diagnostic consideration. When
tuberculous ascites is a possibility, an acid-fast stain should be obtained and a specimen submitted for
mycobacterial cultures. Cirrhotic ascites is typified by a low total protein concentration, less than 2.5
grams per dL and often less than 1 gram per dL, and a high serum-ascites albumin gradient. The
serum-ascites albumin gradient correlates directly with portal pressure, and gradients of 1.1 grams per dL
or greater signify the presence of portal hypertension.
Therapy of cirrhotic ascites begins with dietary modification. Initial restriction of sodium chloride intake
to 2 grams per day results in resolution of ascites in some patients and is crucial to the effectiveness of
diuretics in the remainder. Patients with new ascites and normal renal function may respond to salt
restriction alone. Counseling by a dietitian on how to implement such a diet leads to improved
compliance. Fluid restriction is unnecessary unless significant hyponatremia (serum sodium
concentration of less than 120 mEq per liter) develops.
The majority of patients with cirrhotic ascites require diuretic therapy in addition to dietary salt
restriction. Such therapy requires vigilant monitoring to avoid electrolyte disturbances and volume
depletion. Before initiation of diuretic therapy, the patient's baseline body weight, serum electrolyte
values, and urinary sodium and potassium concentrations should be measured and tests of renal function
conducted. In patients with hyperkalemia, such as those with type IV renal tubular acidosis related to
diabetes mellitus, cautious treatment, if indicated at all, with potassium-sparing diuretics is in order.
Patients with abnormal renal function at baseline must be monitored carefully in order to avoid
precipitating prerenal azotemia or hepatorenal syndrome. In most patients with cirrhosis, the urinary
potassium concentration exceeds that of sodium owing to hypersecretion of aldosterone. In these
patients, the goal of diuretic therapy is to induce natriuresis and reduce kaliuresis, so that the ratio of
urinary sodium to potassium concentration is greater than 1.
The first diuretic prescribed for most patients should be spironolactone (Aldactone), a direct aldosterone
receptor antagonist and potassium-sparing agent. The starting dose is 100 mg per day which may be
given initially in divided doses and after several days as a single dose. The dose may be increased to a
maximum of 400 to 600 mg per day, but dose changes should be made no more often than every 4 days,
owing to the long half-life of the drug. The dosage should be considered optimal once urine sodium
concentration exceeds urine potassium concentration. The most common complication of spironolactone
therapy is hyperkalemia, which is usually avoidable if serum electrolytes are monitored. Spironolactone
may cause painful gynecomastia because it is an androgen receptor antagonist. If this occurs, triamterene

(Dyrenium) may be substituted in a dose of 50 mg bid, with increases to 150 mg bid.

Optimal diuresis should result in weight loss of approximately 0.5 kg per day; weight loss of up to 1.0 kg
per day is well tolerated in patients with concomitant peripheral edema. Although spironolactone
antagonizes the sodium-retaining effect of aldosterone, additional stimuli to sodium retention remain, and
effective diuresis often requires addition of another agent, usually a loop diuretic such as furosemide
(Lasix). In patients with large-volume ascites, it is reasonable to begin administration of furosemide
simultaneously with spironolactone. Otherwise, furosemide can be withheld until it has been established
that the dosage of spironolactone has been optimized without adequate diuresis. The usual starting dose
of furosemide is 40 mg once daily, and the dose may be doubled every several days to a maximum of
160 mg per day if necessary. When given intravenously (IV), furosemide diminishes renal perfusion
significantly in patients with cirrhosis; therefore, this route of administration should be avoided.
Occasionally, the addition of a third diuretic, such as hydrochlorothiazide, may be considered.
Failure to respond to medical therapy defines refractory ascites; failure of ascites to resolve despite
maximal doses of diuretics (400 mg of spironolactone and 160 mg of furosemide daily) defines
"diuretic-resistant" ascites, whereas failure due to intolerance of maximal doses of diuretics defines
"diuretic-intractable" ascites. Surreptitious salt ingestion should be ruled out in patients who fail to lose
weight despite adequate natriuresis. In patients with diuretic-resistant or diuretic-intractable ascites, the
next therapeutic option is generally large-volume paracentesis, which should be initial therapy in any
patient with respiratory compromise from tense ascites. In this procedure, 5 or more liters of fluid is
removed using an 18-gauge needle connected by phlebotomy or peritoneal dialysis tubing to vacuum
bottles. This procedure has been shown to be safe and effective in relieving tense ascites. Complete
paracentesis, in which the entire volume of ascites is drained, has also been shown to be safe and
effective. Although our approach is controversial, we prefer to administer approximately 10 grams of
salt-poor albumin IV per liter of ascites removed in order to minimize the risk of renal insufficiency and
hyponatremia. Large-volume paracentesis may be repeated regularly, with careful monitoring of
hemodynamics, electrolytes, and renal function. For hospitalized patients with tenuous renal function,
tense ascites, and significant hypoalbuminemia, infusion of salt-poor albumin without paracentesis
probably enhances diuresis over the short term without a detrimental hemodynamic effect. In such
patients, infusions of 25 to 50 grams per day should be continued until a target serum albumin of 3.0
grams per dL is reached. However, excessive albumin administration may increase the risk of variceal
bleeding by increasing intravascular volume.
465
Other therapies for refractory ascites include insertion of a peritoneovenous (Le Veen) shunt and
placement of a transjugular intrahepatic portosystemic shunt (TIPS). A LeVeen shunt connects the
peritoneal cavity directly to the internal jugular vein, allowing decompression of ascites by drainage into
the vascular system. However, the use of such shunts is associated with a high incidence of serious
complications, most notably disseminated intravascular coagulation (DIC) and sepsis, and such devices
have fallen out of favor. Placement of a TIPS is performed by an interventional radiologist and may lead
to dramatic resolution of ascites in as many as 75% of patients in whom the condition is refractory to
diuretics. Natriuresis may be delayed up to 4 weeks after a TIPS insertion. Unfortunately, shunt
thrombosis is a common early complication, and stent occlusion due to pseudointimal hyperplasia is
nearly universal in the first 1 to 2 years after placement of a TIPS. Therefore, frequent shunt revisions are

usually necessary. In addition, the use of a TIPS poses an increased risk of portosystemic encephalopathy
and has not been shown to reduce overall mortality. In selected patients, refractory ascites is an
indication for liver transplantation (see later on).
Spontaneous Bacterial Peritonitis
An ominous complication in patients with cirrhotic ascites is spontaneous bacterial peritonitis (SBP),
which is associated with a short-term mortality rate of approximately 50%, and a 1-year mortality rate of
62% even after recovery from a first episode. SBP results from hematogenous superinfection of ascites
stemming from prolonged bacteremia due to decreased reticuloendothelial (RE) function, portosystemic
shunting, and deficiency of complement and immunoglobulins in ascitic fluid. Because the ascitic fluid
total protein concentration correlates directly with the opsonic activity of ascitic fluid, patients with an
ascitic fluid total protein concentration of less than 1 gram per dL are at highest risk of developing SBP.
The most common pathogens in SBP are gram-negative enteric bacteria, especially Escherichia coli and
Klebsiella pneumoniae, but streptococci, including pneumococci, and also enterococci account for a
significant minority of cases. In contrast, anaerobic bacteria rarely cause SBP.
The manifestations of SBP may be protean. Abdominal pain and fever are typical but often absent.
Abdominal tenderness, sometimes with peritoneal signs, is present in approximately half the patients. In
many cases, the only clue to the presence of the infection is worsening encephalopathy, renal function, or
ascites. In fact, up to one third of affected patients may be completely asymptomatic. Therefore, a high
index of suspicion for SBP is necessary, and the clinician should have a low threshold for performing
diagnostic paracentesis. Paracentesis should be performed in any patient with ascites and fever, new
abdominal pain, or new encephalopathy.
The diagnosis of SBP is based on an ascitic fluid polymorphonuclear neutrophil (PMN) count of 500 per
mm3 or higher (or 250 per mm3 or higher if the patient has any symptoms of SBP) and absence of an
underlying intra-abdominal focus of infection, whether or not the ascitic fluid culture is positive. At least
10 mL of fluid should be inoculated directly into blood culture bottles at the bedside to maximize the
sensitivity of the culture. Culture-negative neutrocytic ascites, defined as ascites with a fluid PMN count
of 500 cells per mm3 or higher in the absence of a positive fluid culture and recent antibiotic therapy,
should be treated as for SBP. On the other hand, monomicrobial non-neutrocytic "bacterascites," defined
as ascites with positive results on fluid culture but an ascitic fluid PMN count of less than 250 per mm3 ,
is generally transient and often does not require treatment.
The antibiotic of choice for treatment of SBP is the third-generation cephalosporin cefotaxime (Claforan)
in a dose of 2 grams IV every 8 hours. Alternatives include ceftriaxone (Rocephin), ampicillin-sulbactam
(Unasyn), ticarcillin-clavulanic acid (Timentin), and piperacillin-tazobactam (Zosyn). Recently, oral
ofloxacin (Floxin) has been shown to be highly effective for treatment of SBP, but additional studies are
needed before oral antibiotic therapy alone can be recommended. Use of aminoglycosides should be
avoided, because patients with cirrhosis are particularly prone to aminoglycoside-induced nephrotoxicity.
Antibiotic therapy should be started immediately after paracentesis, pending culture results. If SBP is
confirmed, antibiotics should be continued for at least 5 days. So long as the patient improves clinically,
another paracentesis procedure is not necessary. If the patient does not improve by 48 hours, paracentesis
should be repeated; if the ascitic PMN count has not fallen by 50%, antibiotic coverage should be
broadened and other causes of peritonitis should be considered. Secondary bacterial peritonitis should
also be considered when the initial ascitic fluid PMN count is greater than 10,000 per mm3 , the ascitic

fluid glucose level is less than 50 mg per dL, or the culture is positive for more than one organism.
Prophylactic administration of oral nonabsorbable antibiotics has proved effective in preventing SBP in
high-risk patients. The goal of prophylaxis is selective elimination of gram-negative bacteria from the
intestinal flora. The most effective antibiotic studied has been norfloxacin (Noroxin), * in a dose of 400
mg daily. High-risk patients who may warrant prophylactic therapy include those with a previous episode
of SBP, with a recent episode of gastrointestinal hemorrhage, or awaiting liver transplantation, and
possibly those with an extremely low ascitic fluid total protein concentration (less than 1 gram per dL).
Because candidal overgrowth is a possible side effect of prophylactic antibiotic therapy, liver transplant
candidates on long-term antibiotic prophylaxis should also receive either clotrimazole (Mycelex),
466
1 troche tid, or nystatin (Mycostatin), swish and swallow 10 mL, tid.
Hepatorenal Syndrome
Progressive renal failure often heralds the cirrhotic patient's demise and may be the primary cause of
death. Although the pathogenesis is not completely understood, hepatorenal syndrome (HRS) results
from renal cortical hypoperfusion related to the complex hemodynamic alterations induced by cirrhosis.
Portal hypertension results in decreased systemic vascular resistance and relative hypovolemia as well as
increased renal vascular resistance due to increased sympathetic nervous system output and elevated
levels of various renal arteriolar vasoconstrictors, including angiotensin, vasopressin, endothelin, and
leukotrienes. The effect of these renal vasoconstrictors may be offset by secretion of prostaglandins and
kallikreins, but the compensatory mechanisms may eventually be overwhelmed, resulting in HRS.
Nevertheless, kidneys from patients with HRS remain structurally normal and can even be used as donor
organs for transplantation.
Major criteria for the diagnosis of HRS include advanced hepatic failure and portal hypertension; a low
glomerular filtration rate, as indicated by a serum creatinine level of greater than 1.5 mg per dL or a
24-hour creatinine clearance of less than 40 mL per minute; absence of shock, sepsis, drug-induced
nephrotoxicity, and gastrointestinal or diuretic-induced fluid losses; lack of improvement in renal
function following diuretic withdrawal and expansion of plasma volume with 1.5 liters of normal saline;
and absence of significant proteinuria or ultrasonographic evidence of obstructive or parenchymal renal
disease. Additional criteria that support the diagnosis of HRS, but which may not be present in all cases,
include oliguria (urine volume of less than 500 mL per day), low urine sodium concentration (less than
10 mEq per liter), and urine osmolality greater than plasma osmolality. Fractional excretion of sodium is
characteristically less than 1%.
The only effective treatment of HRS is liver transplantation. Short of transplantation, intravascular
volume should be optimized using blood products or albumin, diuretics should be withheld, and any
other potentially aggravating drugs (such as nonsteroidal anti-inflammatory drugs and aminoglycosides)
should be avoided. Intravenous infusion of "renal-dose" dopamine (1 to 3 mug per kg per minute) may
help to maintain urine output. Intravenous infusion of ornipressin (6 International Units per hour), a
vasoconstrictor that does not increase renal vascular resistance, has been reported to be useful in
stabilizing renal function in patients with HRS awaiting liver transplantation. The use of TIPS may also

be beneficial but has not been studied systematically in patients with HRS. Dialysis may be necessary to
maintain intravascular volume, electrolyte levels, and acid-base homeostasis, but survival is poor once
this stage is reached.
Hepatic Hydrothorax
A pleural effusion due to portal hypertension, also known as hepatic hydrothorax, affects 5 to 10% of
patients with cirrhosis. Most cases occur in patients with poorly controlled ascites, but a pleural effusion
may occur in the absence of ascites. The effusion is typically right-sided and may be massive. Diagnosis
requires exclusion of other causes of pleural effusion, including cardiac or pulmonary disease. Diagnostic
thoracentesis should be performed on any new pleural effusion in a cirrhotic patient; low ascitic fluid
values of LDH and total protein as well as negative results of microbiologic studies are consistent with
hepatic hydrothorax. In addition, analogous to SBP, "spontaneous bacterial empyema" may occur owing
to superinfection of chronic hepatic hydrothorax. Therefore, diagnostic thoracentesis should be
considered in any cirrhotic patient with a pleural effusion and encephalopathy or fever of unknown
source.
The pathogenesis of hepatic hydrothorax appears to be related to small diaphragmatic defects that allow
passage of ascitic fluid into the thorax. Therefore, treatment is generally the same as that for
ascites--namely, salt restriction, diuretics, and large-volume paracentesis when appropriate. Serial
therapeutic thoracenteses may be performed, but rapid reaccumulation of fluid usually follows. Chemical
pleurodesis has been attempted, but results are poor owing to reaccumulation of fluid before the desired
effect can be achieved. In refractory cases, the use of TIPS is often successful and has been proposed as
the treatment of choice. In selected cases, thoracoscopic closure of diaphragmatic defects has been used
with excellent results, but this approach is unlikely to be necessary with the advent of TIPS.
Cirrhosis may result in disturbance of CNS function by several mechanisms. The diagnostic and
pathophysiologic hallmark of hepatic encephalopathy is hyperammonemia, which results from reduced
hepatic clearance of gut-derived ammonia from portal venous blood because of both portosystemic
shunting and hepatocellular dysfunction. Effects of ammonia on the CNS include alterations in
membrane transport, decreases in oxidation-reduction potential and energy stores, and depletion of the
excitatory neurotransmitter glutamate in neurons. When the serum ammonia concentration increases
acutely, accumulation of glutamine in astrocytes can result in lethal cerebral edema. Other factors that
may contribute to hepatic encephalopathy include gut-derived mercaptans, elevated serum levels of
aromatic amino acids, increased sensitivity to gamma-aminobutyric acid (GABA) and endogenous
benzodiazepine-like compounds, zinc deficiency, and deposition of manganese in the basal ganglia.
Clinical manifestations of hepatic encephalopathy vary considerably depending in part on the severity of
hepatic dysfunction and portosystemic shunting
467

Findings
Level of Personality and Neurologic EEG
Grade Consciousness Intellect Abnormalities Abnormalities
0 Normal No abnormality None None
1 Reversed sleep Forgetful, agitated, Tremor, apraxia, Slowing 5-cps
pattern, restless irritable, mildly incoordination, triphasic waves
confused poor handwriting
2 Lethargic Disoriented to Asterixis, Slowing 5-cps
time, amnesia, dysarthria, ataxia, triphasic waves
disinhibited, poor hyporeflexia
judgment
3 Somnolent Disoriented to Asterixis, rigidity, Slowing 5-cps
place, aggressive hyperreflexia triphasic waves
4 Comatose None Decerebrate Slow 2- to 3-cps
delta activity
Abbreviations: EEG = electroencephalogram; cps = counts per second.
(Table 3) . Patients with mild encephalopathy may present clinically with reversal of the sleep-wake
cycle. Mild personality changes may also be noted, but asterixis is often absent. Patients with moderately
severe encephalopathy experience either drowsiness or agitation, and asterixis is usually present. Severe
acute encephalopathy is marked by somnolence, deep confusion, asterixis, and hyper-reflexia. At this
stage, airway protection is compromised, and endotracheal intubation should be considered to prevent
aspiration pneumonia. Coma may ultimately supervene, with decerebrate posturing followed by death.
Patients with chronic encephalopathy often display personality changes, dementia, memory loss, and
occasionally movement disorders.
Most episodes of acute hepatic encephalopathy are precipitated by a potentially reversible factor.
Precipitants include dehydration, constipation, excessive dietary protein intake, hypokalemia, metabolic
alkalosis, sepsis (especially SBP), gastrointestinal bleeding, renal failure, and use of benzodiazepines or
other sedatives. Identification and removal of the relevant precipitant(s) are the first steps in the treatment
of acute encephalopathy. Other causes of mental status changes, such as subdural hematoma, alcohol
withdrawal, Wernicke's encephalopathy, hypoglycemia, and drug or alcohol intoxication, should be ruled
out.
Further treatment of acute encephalopathy centers on facilitating net ammonia excretion. Dietary protein
should be restricted, initially to as little as 20 grams per day. Lactulose syrup should be administered in
doses of 30 mL PO every hour until diarrhea occurs and then every 4 to 6 hours as needed to achieve two
to four loose bowel movements per day. Lactulose is a nonabsorbable disaccharide that passes
unchanged into the colon, where it is fermented by colonic flora. It reduces ammonia absorption by
producing diarrhea and by acidifying the colonic lumen, thereby trapping ammonia as nondiffusible
ammonium ion. In patients with severely depressed levels of consciousness, a nasogastric tube may be
required for the administration of lactulose. If ileus is present, lactulose may be administered as a

retention enema, using 300 mL of lactulose suspended in 700 mL of water. In patients who fail to
respond to lactulose alone, oral neomycin, * 6 grams per day initially and then 1 gram bid, or
metronidazole (Flagyl), * 250 mg tid, may be added. However, neomycin is absorbed by the intestine to
some extent in patients with cirrhosis and may thus cause nephrotoxicity and ototoxicity.
Other treatments may be beneficial in selected cases. Administration of the short-acting benzodiazepine
antagonist flumazenil (Romazicon) * leads to improvement in up to 40% of patients unresponsive to
lactulose, but the response is transient, the drug must be administered parenterally, and the medication is
expensive. Oral sodium benzoate or phenylacetate may facilitate renal excretion of ammonia, and
ornithine aspartate, which enhances ureagenesis, may also lower the serum ammonia level, but these
agents are not widely used for treatment of hepatic encephalopathy due to cirrhosis. Zinc deficiency, if
present, should be corrected, because several enzymes in the urea cycle are zinc-dependent. Finally,
Helicobacter pylori, a urea-splitting bacterium that commonly colonizes the stomach, may add in small
measure to the generation of ammonia in the gastrointestinal tract, and anecdotal reports suggest that
encephalopathy may lessen following eradication of H. pylori.
Prevention of recurrent encephalopathy requires extensive counseling and close monitoring of the
patient. Dietary protein should be restricted, but usually to no less than 60 to 80 grams per day, so that
protein malnutrition does not worsen. Formal consultation with a dietitian is essential to achieving this
goal. Lactulose should be continued at a dose of 15 to 30 mL bid or tid as needed to achieve two or three
loose stools per day. Excessive diuresis and use of sedative medications should be avoided.
Hepatopulmonary syndrome (HPS) is defined as the presence of an increased alveolar-arterial oxygen

gradient and intrapulmonary vascular dilatations in a patient with chronic liver disease. Affected persons
468
may become markedly debilitated owing to severe hypoxemia (partial pressure of oxygen [Pa O2 ] of less
than 50 mm Hg) and may require continuous oxygen supplementation. Typically, oxygenation worsens
as the patient moves from the supine to the standing position (orthodeoxia). The pathophysiology of the
syndrome relates to intrapulmonary right-to-left shunting through vascular dilatations of varying size.
Type I HPS is characterized by diffuse small precapillary dilatations scattered throughout the lung
parenchyma, whereas type II HPS is characterized by discrete, large pulmonary arteriovenous
malformations (AVMs). The diagnosis of HPS may be made by contrast echocardiography, radiolabeled
albumin scanning, or pulmonary angiography, but only angiography differentiates type I from type II
HPS. Angiography is potentially therapeutic in the case of type II HPS; large AVMs can sometimes be
ablated by spring-coil embolization. Unfortunately, for most patients with HPS there is no effective
therapy. However, recent experience indicates that HPS is reversible after liver transplantation so long as
pulmonary hypertension has not developed. Therefore, patients with debilitating hypoxemia from HPS
should be considered candidates for liver transplantation.
Coagulopathy

The liver synthesizes fibrinogen and coagulation factors II, V, VII, IX, and X; factors II, VII, IX, and X
require vitamin K for their activity. Deficiency of one or more of these clotting factors in cirrhosis
typically results in prolongation of the prothrombin time (PT); with severe synthetic dysfunction, the
activated partial thromboplastin time (aPTT) may also be prolonged. Caution must be used in
differentiating the coagulopathy of liver disease from DIC; in both conditions, PT and aPTT are elevated,
the platelet count is decreased, fibrinogen levels are decreased, and fibrin split products are elevated. The
most useful laboratory parameter for differentiating the two processes is the factor VIII level; because
factor VIII is not synthesized by the liver, levels are preserved in cirrhosis but markedly decreased in
DIC.
Additional hematologic consequences of chronic liver disease include anemia, leukopenia, and
thrombocytopenia. Hypersplenism due to portal hypertension is the most common mechanism underlying
these abnormalities, although leukopenia and thrombocytopenia do not resolve consistently after
placement of a TIPS. Iron deficiency due to chronic or recurrent gastrointestinal bleeding may also
contribute to anemia. Folic acid deficiency, which is common in alcoholic patients, may also contribute
to both anemia and thrombocytopenia. Polycythemia should alert the clinician to the possibility of an
occult hepatocellular carcinoma.
Coagulopathy should be corrected, if possible, in any cirrhotic patient with bleeding or before an
invasive procedure. In general, invasive procedures, including percutaneous liver biopsy, should not be
performed if the PT is prolonged by more than 3 seconds or if the platelet count is less than 75,000 per
mm3 . Administration of vitamin K, 10 mg SC daily for 1 to 3 days, can lower the PT in patients with
vitamin K deficiency, including those with predominantly cholestatic disease such as PBC or primary
sclerosing cholangitis, those who are malnourished (especially alcoholic patients), and those receiving
chronic antibiotic therapy. However, vitamin K does not reverse coagulopathy due to impaired
hepatocellular function. Moreover, because the effect of vitamin K depends on synthesis of new clotting
factors, the onset of action is slow. In most patients, therefore, fresh-frozen plasma (FFP) is needed to
reverse significant coagulopathy, and FFP should be administered in doses of 1 to 2 units per hour IV
until bleeding stops or the PT is corrected. If coagulopathy fails to respond to large quantities of FFP and
if the fibrinogen level falls below 100 mg per dL, cryoprecipitate should be administered. Platelet
infusions are also indicated in patients with active bleeding or before invasive procedures when the
platelet count is less than 75,000 per mm3 . Desmopressin (DDAVP), in a dose of 0.3 mug per kg diluted
in 50 mL of normal saline and infused over 30 minutes, may also be used to improve the bleeding time
and aPTT; the drug, an analogue of vasopressin, increases factor VIII and von Willebrand factor
activities.
Blood products carry a risk, albeit small, of viral transmission or transfusion reactions, as well as
overexpansion of the intravascular space and precipitation of variceal bleeding. Therefore, they should
not be administered to nonbleeding patients with cirrhosis and laboratory evidence of coagulopathy.
Cirrhosis induces a relatively immunocompromised state owing to impaired RE function. Malnutrition,

when present, further impairs both humoral and cell-mediated immunity. Therefore, several steps should
be taken to protect cirrhotic patients from infection. Polyvalent pneumococcal vaccine (Pneumovax)
should be administered to all patients with cirrhosis. Hemophilus b conjugate vaccine is also
recommended by some authorities, although Haemophilus influenzae serotype b is an uncommon cause

of respiratory infections in adults. Influenza vaccine should be administered yearly. Finally, vaccination
against hepatitis A (Havrix or Vaqta) and hepatitis B (Engerix-B or Recombivax HB) should be offered
to seronegative patients, because acute viral hepatitis is more likely to lead to hepatic failure in a patient
with already compromised hepatic function.
Patients with cirrhosis of any cause are at increased risk of hepatocellular carcinoma (HCC). The risk
varies with the cause of cirrhosis; patients with chronic viral hepatitis and hemochromatosis are at
469
especially high risk, while those with primary biliary cirrhosis and Wilson's disease are at lowest risk. A
diagnosis of HCC should be considered in any patient with cirrhosis in whom clinical decompensation is
observed. Specifically, HCC may underlie variceal bleeding, worsening ascites, encephalopathy, or
cachexia in a previously well-compensated patient. We screen all patients with cirrhosis for HCC using
ultrasonography and serum alpha-fetoprotein testing every 6 months, although the cost-effectiveness of
this strategy is uncertain. To evaluate focal hepatic lesions detected by ultrasound studies, magnetic
resonance imaging and computed tomography (CT) with arterial-phase imaging after the administration
of IV contrast are useful. Definitive diagnosis may be obtained by ultrasound- or CT-guided needle
biopsy. In patients with cirrhosis, resection of the tumor may be precluded by limited hepatic reserve.
Therefore, liver transplantation should be considered in any patient with cirrhosis and HCC, provided
that the tumor is smaller than 5 cm in diameter and confined to the liver or, if several lesions are present,
the largest is no more than 3 cm in diameter and there is no invasion of the portal vein. In patients who
are not candidates for surgery or transplantation, percutaneous ethanol injection to ablate small tumors
may be considered. For large, unresectable tumors, chemoembolization via the hepatic artery may
provide palliation.
ORTHOTOPIC LIVER TRANSPLANTATION

For cirrhosis of most causes, orthotopic liver transplantation is the only curative therapy. Proper timing
of transplantation is crucial; patients with far-advanced cirrhosis and several serious complications have
a higher mortality rate than do patients with relatively compensated cirrhosis who undergo liver
transplantation, whereas patients with fully compensated cirrhosis do not require transplantation. In
general, patients with cirrhosis should be referred to a transplant center at the first sign of
decompensation or when death due to liver disease is expected within several years.
In preparation for liver transplantation, the following tests should be performed: identification of ABO
blood type, determination of human immunodeficiency virus (HIV) status, measurement of
cytomegalovirus (CMV) and varicella-zoster virus titers, tuberculin skin test, abdominal ultrasound scan
with Doppler flow studies, pulmonary function tests, and measurement of serum alpha-fetoprotein level.
If HCC is present, CT scanning of the chest and abdomen is necessary to rule out extrahepatic spread. In
addition, formal psychiatric evaluation should be obtained. In patients who are CMV-seronegative,
filtered or CMV-safe blood products should be used to avoid transfusion-borne infection, because CMV
infection has been shown to increase mortality after liver transplantation.

Absolute contraindications to liver transplantation include sepsis, extrahepatic malignancy, advanced

cardiac or pulmonary disease, acquired immunodeficiency syndrome (AIDS), and inability to understand
the procedure and post-transplant regimen. Relative contraindications include age greater than 70 years,
active alcohol or drug use, HCC, mesenteric vein thrombosis, hepatobiliary sepsis, HIV infection,
pulmonary hypertension, advanced chronic renal insufficiency, and severe malnutrition.
Five-year survival rates after liver transplantation are approximately 70%, and most post-transplant
deaths occur in the first 6 months and are due to perioperative complications, acute graft rejection, or
sepsis. After transplantation, patients are maintained on lifelong immunosuppression. The standard
regimen consists of cyclosporine or tacrolimus (FK 506, Prograf), azathioprine, and prednisone. Patients
must be willing and able to comply with a complicated medical regimen and intensive follow-up. The
original liver disease may recur in the transplanted organ. In particular, the risk of recurrent hepatitis B
after liver transplantation is high, but can be reduced with hepatitis B immune globulin (HBIG) or
treatment with the oral nucleoside analogue lamivudine. Hepatitis C almost invariably recurs after
transplantation but follows a relatively benign course in many patients.


April 22, 2000
Conde Petra
Outcome of 100 patients after transjugular

intrahepatic portosystemic shunt for variceal
hemorrhage [see comments]
Citation
Bibliographic Data
Abstract
Indexing Data
Outcome of 100 patients after
Copyright Notice and Disclaimer transjugular intrahepatic portosystemic
shunt for variceal hemorrhage [see
Find More Articles Like This comments]
Sahagun G - Am J Gastroenterol - 1997 Sep; 92(9): 1444-52
Full Text
NLM Citation ID:
INTRODUCTION
Comment:
METHODS Am J Gastroenterol 1997 Sep;92(9):1412-6
Study design Full Source Title:
Patients
Publication Type:
Procedures and techniques
Journal Article
Clinical assessment
Language:
Definitions English
Hepatic encephalopathy and Author Affiliation:
ascites. Department of Medicine, Oregon Health Sciences University,
Portland 97201, USA.
Shunt stenosis.
Authors:
Patency.
Sahagun G; Benner KG; Saxon R; Barton RE; Rabkin J; Keller
Recurrent upper GI FS; Rosch J
hemorrhage.
Abstract:
Duration of follow-up. OBJECTIVES: One hundred consecutive patients with
recurrent or refractory acute variceal hemorrhage treated with a
transjugular intrahepatic portosystemic shunt (TIPS) from June
RESULTS 1990 to June 1993 at Oregon Health Sciences University or the
Portland Veterans Affairs Medical Center were evaluated to
Complications
assess shunt patency and clinical outcome, including
http://home.mdconsult.com/das/journal/view/N/99174...I&stty=cow&nid=39985&tlt=Z&pos=&tag=relinfo&anc=ja (1 of 3) [22/04/2000 0:23:25]

Follow-up duration and outcome
complications of TIPS, rebleeding, and survival. METHODS:
Mortality Success of shunt placement, reduction in portal pressure,
complications, survival, recurrent hemorrhage, severity of
Indications for liver transplantation ascites, hepatic encephalopathy before and after TIPS, and
Hemorrhage shunt patency were assessed in each patient. RESULTS: The

mean follow-up period was 17.7 months (range, 0.1-56.7
Clinical characteristics months). TIPS was successfully completed in all patients, with
Shunt patency and stenosis a mean reduction in portosystemic gradient from 24 to 11 mm
Hg. Major complications occurred in 11 patients, including one
DISCUSSION death. Survival after TIPS was 85% at 30 days, 71% at 1 yr, and
56% at 2 yr. Variceal bleeding stopped within 24 hours after
REFERENCES
TIPS in all eight patients with active hemorrhage. Recurrent
About the Publication variceal hemorrhage occurred in 18 patients at a mean of 4.3
months (range, 1-713 days) after TIPS. The cumulative rate of
recurrent variceal bleeding was 20% at 1 yr and 25% at 2 yr
after TIPS. Recurrent variceal bleeding was associated with
shunt stenosis or occlusion in all patients with endoscopically
documented variceal hemorrhage, which was successfully
managed by reopening obstructed shunts and performing
variceal embolization. The prevalence of ascites was
significantly reduced among surviving patients evaluated 3
months after TIPS (67 vs 25%, p < 0.005). Three months after
TIPS, the incidence of new or worsening hepatic
encephalopathy was 20%, but encephalopathy improved in an
equal proportion of patients. Seventy-three of 77 (95%) shunts
examined for patency were open at the last follow-up
examination. However, most shunts required intervention to
maintain patency, and only 48% (37 of 77) were primarily
patent at a mean of 168 days (range, 2-538 days) of follow-up.
Shunt stenosis or occlusion, as determined by venography,
became increasingly frequent with longer follow-up (52% at 3-9
months and 70% at 9-15 months). CONCLUSIONS: TIPS is
effective in lowering elevated portal pressures in patients with
refractory variceal hemorrhage, has acceptable postprocedure
complication and mortality rates, ameliorates ascites, and in, a
minority of patients, worsens encephalopathy. Shunt stenosis
occurs in the majority of patients but can be effectively treated
by interventional techniques to maintain patency. The incidence
of recurrent variceal hemorrhage is low and is associated with
shunt stenosis or occlusion.
Major Subjects:
Esophageal and Gastric Varices / * Surgery / Therapy
Gastrointestinal Hemorrhage / * Surgery / Therapy
* Portasystemic Shunt, Transjugular Intrahepatic /

Adverse Effects
Acute Disease
Adult
Aged
Aged, 80 and over
Ascites / Surgery
Cause of Death
Constriction, Pathologic / Etiology / Therapy
Embolization, Therapeutic
Evaluation Studies
Female
Follow-Up Studies
Hepatic Encephalopathy / Surgery
Human
Hypertension, Portal / Surgery
Incidence
Male
Middle Age
Portal Pressure
Prevalence
Prospective Studies
Recurrence
Survival Rate
Treatment Outcome
Vascular Patency


Volume 92 Number 9 September 1997
P1444
ORIGINAL CONTRIBUTIONS
Outcome of 100 Patients after Transjugular Intrahepatic

Portosystemic Shunt for Variceal Hemorrhage
Geronimo Sahagun M.D

Kent G. Benner M.D.
Richard Saxon M.D.
Robert E. Barton M.D.
John Rabkin M.D.
Frederick S. Keller M.D.
Josef Rosch M.D.
Departments of Medicine, Surgery, and Radiology, Oregon Health Sciences University, the Dotter
Interventional Institute, and the Veterans Affairs Medical Center, Portland, Oregon
Objectives: One hundred consecutive patients with recurrent or refractory acute variceal
hemorrhage treated with a transjugular intrahepatic portosystemic shunt (TIPS) from June 1990
to June 1993 at Oregon Health Sciences University or the Portland Veterans Affairs Medical
Center were evaluated to assess shunt patency and clinical outcome, including complications of
TIPS, rebleeding, and survival. Methods: Success of shunt placement, reduction in portal pressure,
complications, survival, recurrent hemorrhage, severity of ascites, hepatic encephalopathy before
and after TIPS, and shunt patency were assessed in each patient. Results: The mean follow-up
period was 17.7 months (range, 0.1-56.7 months). TIPS was successfully completed in all patients,
with a mean reduction in portosystemic gradient from 24 to 11 mm Hg. Major complications
occurred in 11 patients, including one death. Survival after TIPS was 85% at 30 days, 71% at 1 yr,
and 56% at 2 yr. Variceal bleeding stopped within 24 hours after TIPS in all eight patients with
active hemorrhage. Recurrent variceal hemorrhage occurred in 18 patients at a mean of 4.3
months (range, 1-713 days) after TIPS. The cumulative rate of recurrent variceal bleeding was
20% at 1 yr and 25% at 2 yr after TIPS. Recurrent variceal bleeding was associated with shunt

stenosis or occlusion in all patients with endoscopically documented variceal hemorrhage, which
was successfully managed by reopening obstructed shunts and performing variceal embolization.
The prevalence of ascites was significantly reduced among surviving patients evaluated 3 months
after TIPS (67 vs 25%, p < 0.005). Three months after TIPS, the incidence of new or worsening
hepatic encephalopathy was 20%, but encephalopathy improved in an equal proportion of
patients. Seventy-three of 77 (95%) shunts examined for patency were open at the last follow-up
examination. However, most shunts required intervention to maintain patency, and only 48% (37
of 77) were primarily patent at a mean of 168 days (range, 2-538 days) of follow-up. Shunt stenosis
or occlusion, as determined by venography, became increasingly frequent with longer follow-up
(52% at 3-9 months and 70% at 9-15 months). Conclusions: TIPS is effective in lowering elevated
portal pressures in patients with refractory variceal hemorrhage, has acceptable postprocedure
complication and mortality rates, ameliorates ascites, and in, a minority of patients, worsens
encephalopathy. Shunt stenosis occurs in the majority of patients but can be effectively treated by
interventional techniques to maintain patency. The incidence of recurrent variceal hemorrhage is
low and is associated with shunt stenosis or occlusion.
Reprint requests and correspondence: Kent G. Benner, M.D., Division of Gastroenterology and Hepatology, Department of
Medicine, Oregon Health Sciences University, PV310, 3181 Southwest Sam Jackson Park Road, Portland, OR 97201.
Received May 15, 1996;
accepted Feb. 4, 1997.
INTRODUCTION
Treatment options for the management of acute and recurrent variceal hemorrhage secondary to portal hypertension
include endoscopic therapy with sclerotherapy or band ligation, pharmacotherapy with pitressin or octreotide, balloon
tamponade, and emergency surgical portosystemic decompression and variceal devascularization [1] [2] [3] . Portosystemic
decompressive surgery is associated with substantial mortality in patients with advanced liver disease and may complicate
subsequent liver transplantation. Consequently, surgical portosystemic decompression is often reserved for patients in
whom medical and endoscopic treatment of gastroesophageal variceal hemorrhage have failed, patients with bleeding
ectopic varices, or patients with variceal bleeding and little evidence of hepatic decompensation [4] [5] [6] [7] [8] . The
combination of transjugular liver puncture and balloon angioplasty techniques with recently developed expandable
intravascular stents has resulted in the development of the transjugular intrahepatic portosystemic shunt (TIPS), a welcome
therapeutic alternative for patients with refractory or recurrent variceal hemorrhage [9] [10] . However, the precise
application of TIPS among the various treatment alternatives for variceal bleeding is not yet well defined.
In 1969 Rosch and colleagues [9] [11] first performed TIPS in dogs using nonexpandable coil stents, but patency was
short-lived. The recent introduction of expandable metallic
P1445
endoprostheses to buttress the transhepatic tract has facilitated prolonged patency of TIPS and has heralded its application
in patients with complications of portal hypertension [12] [13] [14] [15] . We report the clinical outcome of the first 100
consecutive patients with variceal hemorrhage treated with TIPS at the Dotter Interventional Institute at Oregon Health
Sciences University and the Portland Veterans Affairs Medical Center. The goals of this study were to define the outcome
of patients after TIPS in terms of 1) recurrent hemorrhage and patient survival; 2) clinical and biochemical indicators of
hepatic decompensation, including ascites, portosystemic encephalopathy, and Child-Pugh score; and 3) short-term and
long-term shunt patency.

METHODS
Study design
This was a prospective observational study of consecutive patients referred for TIPS placement for variceal hemorrhage at
a university medical center and a Veterans Administration medical center. Patients were referred from the larger
population of patients with variceal bleeding because of failure of endoscopic treatment, the presence of varices not
amenable to endoscopic therapy, and/or suboptimal surgical candidacy. Clinical outcomes data were derived from regular
clinic visits during the follow-up period, hospital and clinics records, and phone contact with attending physicians and
patients.
Patients
The study population included 100 consecutive patients with advanced liver disease and variceal hemorrhage refractory to
standard medical therapy who underwent TIPS placement at Oregon Health Sciences University or the Portland Veterans
Affairs Medical Center between June 1990 and June 1993. Follow-up data were gathered at regular intervals up to the time
of death or liver transplantation. Eight patients with active gastroesophageal variceal bleeding typified by ongoing blood
loss and transfusion requirements that could not be controlled by medical therapy (vasopressin with or without
nitroglycerine) and balloon tamponade or endoscopic sclerotherapy or banding underwent emergent TIPS. Eighty-eight
patients with an mean of three previous episodes of bleeding from gastroesophageal varices despite a mean of four
previous sessions of endoscopic therapy underwent TIPS on an urgent basis. This group included 66 patients with two or
more episodes of recurrent variceal bleeding over a 1-week interval during a course of endoscopic therapy. Thirty of these
patients had predominantly gastric varices, which limited options for further endoscopic treatment. Four patients
underwent TIPS because of recurrent bleeding from ectopic varices (two peristomal, one duodenal, and one rectal) that
was not amenable to endoscopic therapy. The mean interval from the time of admission for variceal hemorrhage to
performance of the TIPS procedure was 4 days.
The study population included 30 women and 70 men with a median age of 52 yr (range, 25-80 yr). The cause of liver
disease was alcohol use in 63 patients, hepatitis C in 14 patients, cryptogenic in 10 patients, primary biliary cirrhosis in six
patients, hepatitis B in three patients, and Budd-Chiari syndrome, congenital hepatic fibrosis, sclerosing cholangitis, and
hemochromatosis in one patient each. Before TIPS, assessment of hepatic compensation using the Child-Pugh
classification demonstrated that 12% of the patients were in class A; 50%, in class B; and 38%, in class C. The incidence
of ascites before TIPS was 73%, with clinically apparent ascites (2+ or 3+) present in 63% (63 of 100) of patients.
Forty-six patients were receiving diuretic therapy for ascites, and 24 patients were being treated for encephalopathy at the
time of the TIPS procedure. Hepatic encephalopathy was present in 42% (42 of 100) of patients immediately before the
TIPS procedure.
Procedures and techniques
After informed consent was obtained, the neck was prepared in a sterile manner, and the right internal jugular vein was
percutaneously punctured. A 10F sheath was placed, and a coaxial catheter needle set (RUPS-100, Cook, Inc.,
Bloomington, IN) was introduced and manipulated with a leading guidewire under fluoroscopic guidance through the
superior vena cava, right atrium, and inferior vena cava into the middle or right hepatic vein as previously described [16] .
In one patient, puncture of the left internal jugular vein was required after we were unable to enter the right side. The
guidewire was then removed, and wedged hepatic venography was performed to provide retrograde visualization of the
portal vein. The liver was subsequently punctured through the wall of the hepatic vein, and generally the right portal vein
was entered with a sharp needle inside a 5F (1.5-mm) well tapered catheter. The 5F puncture catheter was then exchanged
for a diagnostic catheter, and initial baseline portal vein pressure and a portal venogram were obtained. Dilation of the
parenchymal tract and hepatic and portal vein walls using a 10-mm-diameter and 6-cm-long angioplasty balloon catheter
(Cook) followed. Expandable metal stents were then placed. Either 10-mm-wide and 6.8-cm-long Wallstents (Schneider
USA, Minneapolis, MN) or 12-mm-wide and 6- to 7.5-cm-long Gianturco-Rosch biliary Z stents (Cook) were used to
create the TIPS. Wallstents were used in 73 patients; Z stents, in 24 patients; and both types of stent, in three patients. The
Z stents were placed early in our TIPS experience and were superseded by Wallstents because of a higher incidence of
shunt stenosis among shunts constructed with Z stents. The desired endpoints of the procedure were reduction of the

portosystemic gradient to 12 mm Hg and elimination of flow-feeding varices. After stent placement, the 10-mm
angioplasty balloon was replaced, and the intraparenchymal segment of the stent was expanded to its maximum diameter.
Portal venography and pressure measurements were then repeated to assess the shunt lumen and adequacy of portal
decompression. The
P1446
portosystemic gradient was calculated by subtracting the mean pressure in the suprahepatic inferior vena cava from the
mean portal vein pressure. Variceal embolization was performed in 12 patients. In nine patients it was performed at the
time of the TIPS procedure, usually because of residual flow to varices, and in three it was performed within 24 hours after
the initial TIPS placement because of continued active bleeding despite a patent shunt. Pre-TIPS ultrasonography was used
to assure patency of hepatic vasculature before the procedure, and pre-TIPS portal venography was performed in selected
patients when hepatic vein thrombosis was suspected. Patients received a local anesthetic at the venous puncture site in the
neck and were sedated with fentanyl and versed. Antibiotic prophylaxis (cefazolin, 1 g i.v.) was routinely used
immediately before the procedure was started. Upon completion of the procedure, the catheters and sheath were removed
from the internal jugular vein. All patients were monitored for a minimum of 24 hours and were discharged when stable.
Clinical assessment
A medical history and laboratory parameters were obtained and a physical examination was performed before the TIPS
procedure, and each patient's Child-Pugh score was defined [17] [18] . History, laboratory data, and results of physical
examination were subsequently recorded at 3, 6, 9, 12, 18, 24, 36, and 48 months after the TIPS procedure. Baseline
ultrasonography with Doppler flow studies was performed within 1 day after the TIPS procedure, at the follow-up intervals
noted above, and when clinical findings suggested recurrent portal hypertension (recurrent upper GI bleeding or new or
worsening ascites). The Doppler criteria for shunt dysfunction included reduction in mean peak flow rate through the shunt
to <90 cm/s, an interval increase or decrease in peak flow rate of 50 cm/s on follow-up studies, or absence of flow within
the shunt. Portal venography, usually approached through a right internal jugular vein, was performed at 6, 12, 18, 24, 36,
and 48 months after the TIPS procedure and when abnormal results on Doppler studies suggested shunt dysfunction.
Among patients who underwent liver transplantation or died during follow-up, examination of the explanted liver or
autopsy specimen included inspection of the shunt for patency.
Definitions
Hepatic encephalopathy and ascites.
The degree of hepatic encephalopathy was graded as follows: 0, none; 1+, mild changes in behavior, sleep pattern, or
memory; 2+, debilitating encephalopathy altering daily activities; or 3+, precoma or coma. The extent of ascites was
classified as follows: 0, absent on ultrasonography; 1+, not apparent on physical examination but detectable by
ultrasonography; 2+, detectable on physical examination but not tense; or 3+, tense or requiring repeated large volume
paracentesis.
Shunt stenosis.
Shunt stenosis was determined by measurement of the diameter of the shunt lumen and portosystemic pressure gradients in
patients who underwent follow-up venography. We defined a stenotic shunt as having a portosystemic gradient of >15 mm
Hg and/or hepatic vein or shunt lumen diameter narrowing of >50% relative to the diameter of the metallic stent by
venography.
Patency.
Shunt patency was determined by venography or examination of the liver at explantation or autopsy. Shunt patency at the
last follow-up evaluation was categorized according to the following criteria as previously described [5] [19] : primary
patency, no interventions to maintain patency after TIPS placement; assisted primary patency, interventions including
balloon dilation or additional stent placement to improve patency in a narrowed shunt; secondary patency, occluded shunts
reopened by intervention; or occluded, shunts occluded and not reopened. Duration of patency was defined as follows:

primary patency, the duration without any interventions from the time of TIPS creation; primary-assisted patency, the
duration of uninterrupted patency with or without further interventions; or secondary patency, the longest duration of
patency with or without prior occlusion.
Recurrent upper GI hemorrhage.
Recurrent hemorrhage was defined as bleeding after TIPS creation resulting in a fall in Hct and recurrent hematemesis or
melena with a nasogastric aspirate positive for blood. Variceal hemorrhage was presumed to be the site of recurrent upper
GI hemorrhage unless a nonvariceal site of hemorrhage was documented at endoscopy.
Duration of follow-up.
Follow-up was measured in days from the date of TIPS creation until the date of death, liver transplantation, or most recent
clinical visit. Patients were considered lost to follow-up if they had not shown up for clinic visits during the most recent
6-month period.
Results are expressed as the mean SD unless otherwise noted. Cumulative survival and rebleeding rates were estimated
over a 2-yr period using Kaplan-Meier analysis. Overall survival was estimated as the interval from the time of the TIPS
procedure to death, while patients were censored at the time of transplantation, the last contact before being lost to
follow-up, or the end of observation. Analysis of variance was used to compare laboratory values, hepatic encephalopathy,
ascites, and Child-Pugh score before and after TIPS in surviving patients who had not undergone transplantation.
Differences in rates of rebleeding and survival among groups were analyzed by the log-rank test.
RESULTS
The TIPS procedure was successfully completed in all 100 patients, with the time required to complete the procedure
ranging from 90 to 240 min. The mean portal pressure decreased from 34 mm Hg (range, 21-54 mm Hg) before TIPS to 21
mm Hg (range, 8-42 mm Hg) after TIPS, and
P1447
the mean portosystemic gradient dropped from 24 mm Hg (range, 7-43 mm Hg) to 11 mm Hg (range, 0-22 mm Hg).
Post-TIPS portal venography showed brisk flow through the shunt in all cases.
Complications
Major complications within 30 days of TIPS attributed to the procedure occurred in 10 patients and included pulmonary
edema (four patients), hepatic arterial injury (two patients), sepsis (two patients), portal vein thrombosis (one patient), and
contrast medium-induced acute tubular necrosis requiring dialysis (one patient). Among the two patients with hepatic
artery injury, one suffered modest intraperitoneal hemorrhage that was readily controlled without further sequelae by
subelective hepatic artery embolization. In the second patient, inadvertent laceration of the hepatic artery caused
intraperitoneal hemorrhage that was controlled by selective embolization of the right hepatic artery, but the patient died 2
days later secondary to ischemic hepatic necrosis involving the right lobe of the liver. Minor procedure-related
complications included hematoma at the puncture site in the neck in 10 patients and transient elevation of the creatinine
level to >3.0 mg/dl in five patients.
Follow-up duration and outcome
The median duration of the hospital stay after TIPS was 3 days (range, 1-63 days). The mean duration of follow-up after
TIPS was 28.9 months (range, 0.1-72.2 months). At last follow-up, 20 patients were alive and had not undergone liver
transplantation (mean follow-up, 50.5 months; range, 37.9-71.2 months), 27 patients had undergone transplantation (mean
interval to transplantation, 8.8 months; range, 0.1-37.1 months), 48 patients had died (mean interval to death, 14.1 months;
range, 0.1-51.0 months), and five patients had been lost to follow-up (mean of last contact after TIPS, 9.4 months).

Mortality
Cumulative survival analysis demonstrated survival rates of 85% at 30 days, 71% at 1 yr, and 54% at 2 yr after TIPS (Fig.
1) . When patients in Child-Pugh classes A and B were combined and compared with those in class C, who were divided
into groups with less (Child-Pugh score of 10-11) and more (Child-Pugh score of 12-15) severe liver disease, survival was
significantly worse in those in a more advanced Child-Pugh class ( p < 0.001, log-rank test) (Fig. 1) . The most common
causes of death among the 15 patients who died within 30 days after TIPS were hepatic failure and multisystem organ
failure (Table 1) . In this group of patients, the Child-Pugh class at the time of TIPS placement was class A in one patient,
class B in three patients, and class C in 11 patients. Twenty-three patients died more than 30 days after TIPS, and the
predominant cause of death in this group was hepatic failure (Table 1) . The Child-Pugh
Figure 1. Kaplan-Meier analysis of patient survival among 100 consecutive patients treated for variceal hemorrhage with TIPS. For
survival analysis the endpoint was death, and patients were censored at the time of last follow-up or transplantation, and 32 patients
remained at risk at 2 yr after TIPS placement. Results are shown for combined Child-Pugh class A and B patients (n = 62), less severely
ill class C patients (Child-Pugh score of 10-11, n = 24), and more severely ill class C patients (Child-Pugh score of 12-15, n = 14).
Survival was significantly worse among both groups of class C patients compared with class A and B patients ( p < 0.0001, log-rank
test) but was not significantly different between the groups of class C patients.
TABLE 1 -- Causes of Death after TIPS

<30 Days >30 Days
Cause of Death (n = 38) after TIPS after TIPS
Upper GI hemorrhage 2 3
Hepatic failure 5 17
Sepsis 1 2
Hepatoma 0 2
Extrahepatic malignancy 0 2
Multisystem organ failure 5 3
Hemoperitoneum 1 0
Hepatorenal syndrome 1 0
Myocardial infarction 0 1
Pulmonary edema 0 1
Pneumonia 0 1
Myelodysplastic syndrome 0 1
-- --
Total 15 33
class at the time of TIPS placement in these patients was class A in six, class B in six, and class C in 11. Ten percent of all
deaths (five of 48) were attributable to recurrent upper GI hemorrhage.

Indications for liver transplantation
The indication for liver transplantation was end-stage liver disease in all patients who underwent transplantation after
TIPS. Transplantation was expedited in one patient because of TIPS-related encephalopathy and in another patient for
persistent variceal bleeding.
P1448
Figure 2. Kaplan-Meier analysis of recurrent variceal hemorrhage among 100 consecutive patients treated for variceal hemorrhage
with TIPS. In this analysis, the endpoint was recurrent variceal hemorrhage, and patients were censored at the time of last follow-up,
transplantation, or death, and 29 patients remained free of rebleeding 2 yr after TIPS. When patients were compared by Child-Pugh
class (class A, n = 12; class B, n = 50; class C, n = 38), there was significantly more rebleeding among class C patients ( p = 0.0025,
log-rank test), but rebleeding was similar between class A and class B patients.
Hemorrhage
TIPS was performed for refractory active hemorrhage in eight patients. Active bleeding stopped immediately after TIPS in
five patients. In the remaining three patients, variceal bleeding recurred within 24 hours of TIPS placement, and follow-up
venography demonstrated a patent TIPS with adequate portal decompression; variceal embolization was then performed
with subsequent cessation of hemorrhage. Overall, recurrent variceal hemorrhage occurred in 18 patients (18%) at a mean
interval of 4.3 months (range, 1-715 days) after the procedure. Kaplan-Meier analysis demonstrated recurrent variceal
hemorrhage in 10% of patients at 30 days, 20% of patients at 1 yr, and 25% at 2 yr after TIPS. When patients were
analyzed by their Child-Pugh class at the time of TIPS, there was significantly more rebleeding among patients in class C (
p = 0.002, log-rank test), but recurrent variceal bleeding between class A and class B patients was not different (Fig. 2) .
Among the 18 patients with recurrent variceal bleeding, bleeding was documented by endoscopy in nine and was
presumed to be from varices (no endoscopic documentation) in nine, either because rebleeding occurred shortly after
initial endoscopy and within 24 hours of TIPS placement (three patients) or because of the patient's terminal condition (six
patients). In four patients, follow-up endoscopy demonstrated upper GI bleeding from nonvariceal sites, including one
ulcer related to pre-TIPS endoscopic therapy, one Mallory-Weiss tear, one gastric ulcer, and one duodenal ulcer. Portal
venography in the nine patients with endoscopically documented recurrent variceal bleeding demonstrated shunt occlusion
in five patients and stenoses with >50% narrowing of the lumen in the remaining four. Balloon dilation of the shunt,
including deployment of an additional coaxial stent in the five patients with occlusion, resulted in reestablishment of shunt
patency in all nine patients. The Child-Pugh class in patients with recurrent variceal bleeding after TIPS was class A in
one, class B in six, and class C in 11. Of the 18 patients who had recurrent variceal bleeding, five underwent
transplantation, 10 died, and three were alive at the last follow-up examination.
Clinical characteristics
Among the 79 patients who had ascites reevaluated 3 month after TIPS placement, 67% of patients (53 of 79) had ascites
at baseline, including 56% (44 of 79) in whom it was clinically apparent (2+ or 3+). In these patients, there was a
significant reduction in ascites from baseline: 25% (20 of 79) had ascites at the 3-month follow-up examination, including
13% (10 of 79) in whom it was clinically apparent ( p < 0.05). Moreover, in the 53 patients with ascites before TIPS who
were reevaluated 3 months later, ascites improved by at least one grade in 77% (41 of 53), remained unchanged in 21%
(11 of 53), and worsened in only 2% (1 of 53).
Hepatic encephalopathy was present in 35% of patients (28 of 80) evaluated 3 months after TIPS. However, because the
80 patients available for reevaluation 3 months after TIPS represented a group selected for survival free of transplantation
during that interval, it is important to note that only 33% (26 of 80) of these patients had pre-TIPS encephalopathy.
Moreover, although 20% of patients (16 of 80) assessed before and 3 months after TIPS developed new (nine patients) or
worsening (seven patients) encephalopathy, 20% (16 of 80) also had improvement in encephalopathy, and 60% were
unchanged at the 3-month follow-up evaluation. Among the surviving patients who did not undergo transplantation, the

serum albumin level, prothrombin time, and total serum bilirubin level had not changed significantly at the 3-month
follow-up examination.
Among patients who had Child-Pugh scores determined over the 3 months after TIPS, the baseline mean Child-Pugh score
of 8.4 1.7 improved to 7.3 1.5 at 3 months, a difference that was not statistically significant. The modest improvement
in the Child-Pugh score at 3 months among surviving patients who did not undergo transplantation was largely attributable
to the decrease in ascites, with no significant changes in serum albumin level, prothrombin time, bilirubin level, or hepatic
encephalopathy.
Shunt patency and stenosis
Shunt patency was determined during follow-up in 77 patients. Twenty-three patients died or underwent transplantation
before a follow-up examination could be performed. The status of shunt patency at the last follow-up assessment is
summarized in Table 2 . Among the 77 patients in whom shunt patency was assessed, shunts were patent in 73 (95%) at
the last assessment. However, 36 patients (47%) required
P1449
TABLE 2 -- Shunt Patency Assessed at Last Follow-up and Interventions Required to Maintain Patency (N = 100
patients) *
Interval after
Interventions per
TIPS
n (days) Patient
Primary patency 37 168 (2-538) 0
Primary assisted patency 26 456 (2-798) 2.0
Secondary patency 10 396 (27-1149) 3.1
Occluded 4 135 (8-825) 1.0
Not assessed 23
* Patency at last follow-up examination as assessed by venography (36), Doppler ultrasound [12] , or examination of the liver at autopsy [9] or transplantation [20]
.
Mean (range).
Mean.
Twenty-one patients died before the 6-month follow-up examination, one was lost to follow-up, and one underwent transplantation.
interventions to maintain patency of stenosed or occluded shunts, and only 37 patients (48%) had primarily patent shunts
that did not require interventions to keep them open. Among the four patients with occluded shunts, the shunts were
occluded at autopsy in two. The other two occluded shunts were detected at follow-up venography in asymptomatic
patients who had undergone previous shunt revisions, in whom it was elected not to intervene because no significant
portosystemic pressure gradient was present and no varices were visualized at venography.
Shunt stenosis became increasingly prevalent as the length of time after TIPS placement increased among patients who
underwent follow-up venography: narrowing in the parenchymal tract of the shunt or in the free hepatic vein adjacent to
the shunt was present in 52% of patients studied between 3 and 9 months after the procedure (mean interval of 174 days

after TIPS, n = 44 patients), 70% of patients examined between 9 and 15 months after TIPS (mean interval of 321 days
after TIPS, n = 27 patients), and 83% of patients studied between 21 and 27 months after TIPS (mean interval of 735 days
after TIPS, n = 6 patients). No portal vein stenosis occurred. We elected to intervene on all compromised shunts when
initially detected, even though the majority of follow-up venograms were performed as scheduled surveillance in
asymptomatic patients. In all but one patient, attempts to revise stenotic or occluded shunts by angioplasty and/or
additional stent placement were successful.
DISCUSSION
Since the first clinical experience with TIPS reported 8 yr ago, the application of TIPS for the management of refractory
variceal hemorrhage has expanded rapidly [5] [14] [15] [20] [21] [22] [23] . On the basis of the outcomes of the first 100
consecutive patients with variceal hemorrhage treated with TIPS at our institution, we draw several conclusions. First, we
confirm initial reports of short-term follow-up demonstrating that TIPS can be technically completed safely in patients
with advanced liver disease who suffer refractory acute or recurrent variceal hemorrhage and that TIPS effectively lowers
portal pressure and reduces the occurrence of early rebleeding. We also confirm observations that hepatic encephalopathy
is a complication in a minority of patients. Our intermediate-term follow-up interval, at a mean of 30 months, is the longest
follow-up interval of a large cohort of patients after TIPS placement for variceal bleeding, and our findings indicate that
TIPS continues to provide extended freedom from recurrent variceal hemorrhage in 80% of patients. An additional
important observation is that although shunt stenosis develops in most patients during follow-up, shunt patency can be
maintained in most patients with aggressive monitoring and TIPS revision. With this proactive approach, recurrent TIPS
stenosis is usually not associated with symptoms of recurrent portal hypertension. However, in the minority of patients
who do suffer recurrent variceal hemorrhage, TIPS stenosis is invariably responsible.
Our experience with respect to technical success and complications is remarkably similar to the outcomes of the initial
patients treated with TIPS and followed up for extended periods as reported by investigators from two other major centers
experienced in this technique [5] [22] . The reduction in the mean portal vein pressure after TIPS of 13 mm Hg is
comparable to that reported by others but is less than typically reported with surgical side-to-side portacaval anastomosis
[6] [8] [20] [21] . Major procedure-related complications occurred in 10% of our patients. Complications were comparable in
incidence and type to those reported by investigators from two other centers [5] [20] [21] [22] . We did not recognize other
complications that have been reported, such as significant hemolysis, hemobilia, or stent migration [24] . Our only fatal
procedure-related complication was a laceration of a branch of the hepatic artery that resulted in intra-abdominal bleeding;
the bleeding was controlled by transvascular embolization but was subsequently complicated by progressive, and
ultimately fatal, hepatic necrosis.
After TIPS we evaluated laboratory parameters indicating hepatic reserve, ascites, encephalopathy, and Child-Pugh class
to define the impact of the procedure on the clinical status among surviving patients. The clinical parameter most
significantly affected was ascites. Although 55% of our patients had clinically apparent (2+ or 3+) ascites before TIPS,
only 13% and 9% had ascites that was appreciated on physical examination at the 3- and 6-month follow-up examinations,
respectively. Our data demonstrate that although 40% of the total population had encephalopathy before TIPS, 35% of
surviving patients who were evaluated at 3 months had encephalopathy, a modest increase from the 33% incidence of
encephalopathy immediately before TIPS in this selected population. When interpreting these observations, it is important
to consider that the assessment immediately before TIPS was usually performed within days of a variceal hemorrhage,
which likely contributed to an acute exacerbation of encephalopathy and
P1450
TABLE 3 -- Comparison of Recurrent Hemorrhage and Survival Rates among Three Medical Centers Experienced with
TIPS
Recurrent
Hemorrhage
(%) Survival (%)

Site Patients 1 yr 2 yr 1 yr 2 yr
Freiburg N = 90 18 85
Class A 27% 100
Class B 46% 86
Class C 22% 73
UCSF N = 90 26 32 60 51
Class A 9% 75 75
Class B 44% 68 55
Class C 47% 49 43
Oregon N = 100 20 25 71 56
Class A 12% 12 12 83 72
Class B 50% 12 16
Class C 38% 34 44 48 23
* Child-Pugh class.
did not represent the patients' baseline status. Moreover, although only 20% of our patients were treated with lactulose for
encephalopathy before TIPS placement, nearly all patients in whom hepatic encephalopathy developed during follow-up
received lactulose.
Our observed 20% incidence of recurrent variceal hemorrhage within 1 yr and 25% incidence within 2 yr after TIPS, as
determined by Kaplan-Meier analysis, compares favorably with recurrent hemorrhage rates reported by other centers [20]
[21] [22] (Table 3) . This may represent an overestimate of recurrent variceal hemorrhage because of our assumption that all
patients with recurrent upper GI bleeding who did not undergo endoscopy were bleeding from varices; moreover, it is
likely that about one-third of these patients (perhaps two or three of the eight recurrent bleeders who did not undergo
endoscopy) were actually bleeding from nonvariceal sources. In addition, three patients with active bleeding at the time of
TIPS and who were considered to have recurrent variceal hemorrhage within 1 day after TIPS placement actually had
adequate portal decompression with a functioning TIPS at the time of rebleeding. We believe that such patients with active
variceal bleeding at the time of TIPS occasionally have transient oozing from the ruptured varix for a short period after
adequate portal decompression that may be minimized by embolization of the vessels feeding the varices. Consequently,
based on this initial experience, our current practice is to embolize the accessible collateral vessels supplying varices at the
time of the TIPS in patients who have had active bleeding within a day or two of the procedure.
The Freiburg group [22] , which followed 90 patients after TIPS who had generally less severe advanced liver disease than
the patients treated in Oregon, reported an 18% rate of recurrent bleeding at 1 yr in 32 patients who had been followed
more than 1 yr (see Table 3) . The intermediate-term follow-up of 90 patients treated with TIPS for variceal bleeding at the
University of California, San Francisco (UCSF), which included patients with somewhat more severe liver disease than
our patients (mean Child-Pugh score, 9.8; class C, 48%; see Table 3) , entailed a mean of 2.2 yr of follow-up experience
[21] . Their cumulative rebleeding rates of 26% at 1 yr and 32% at 2 yr were slightly higher than the rates we observed and
may have been due to a higher percentage of Child-Pugh class C patients in their population. In our experience, which is
similar to that of the UCSF group, recurrent variceal bleeding after successful TIPS was more prevalent in patients in
Child-Pugh class C (Fig. 2) [5] . Moreover, the rate of recurrent bleeding in the three series tended to be higher at centers
with a higher proportion of class C patients (Table 3) .

Survival of our patients after TIPS was 71% at 1 yr, a rate intermediate between the 1-yr survival rates reported by the
Freiburg (85%) and UCSF groups (60%). Our 2-yr survival rate of 54% is slightly higher than the 2-yr survival rate of
51% reported by the UCSF group (Table 3) . Although these differences are likely due to differences in the severity of
liver disease among the patient populations, it should be noted that the 1-yr survival rates among patients with similar
Child-Pugh classifications differed somewhat at the Freiburg, Oregon, and UCSF centers. We found that mortality in the
first 30 days after TIPS was attributable to liver failure and multiple organ failure syndrome and that later mortality was
predominantly due to liver failure, with few deaths primarily caused by recurrent hemorrhage. We also confirmed earlier
observations that patient survival after TIPS is inversely correlated with the increasing severity of liver disease as defined
by the Child-Pugh score. Class A and class B patients had a particularly favorable outcome after TIPS: our combined
group of class A and class B patients had survival rates of 83% at 1 yr and 72% at 2 yr. However, we noted that patients
with a Child-Pugh score of 10 or greater had a significantly poorer prognosis after TIPS ( p = 0.0001, log-rank test) (Fig.
1) ; class C patients with a Child-Pugh score 10 or 11 had a 30-day survival rate of 69% and a 1 yr survival rate of 47%,
and the severely ill class C patients (Child-Pugh score of 12-15) had comparable survival rates (71% at 30 days and a 51%
at 1 yr). This differed from the experience of the UCSF group, which reported that the class C patients with less severe
illness (Child-Pugh score of 10 or 11) had survival rates similar to those of their class A and class B patients [5] .
When reviewing rates of recurrent bleeding and survival in these TIPS series, it is important to consider that the
P1451
patients included were generally selected for placement of a TIPS because of a history of recurrent or refractory acute
variceal hemorrhage, often after repeated attempts at endoscopic therapy had failed. The survival and rebleeding results we
observed compare favorably with reported outcomes after surgical creation of portosystemic shunts for persistent and
recurrent variceal bleeding [6] [7] [8] . However, meaningful comparison of TIPS with surgical portosystemic
decompression must await results of prospective trials that control for variables affecting outcome, such as the interval
from hemorrhage to treatment and the degree of hepatic decompensation.
Assessment of shunt patency based on the last examination of the shunt during follow-up (ultrasonography, venography, or
assessment at explantation or autopsy) in 77% of our patients indicated that although the majority of shunts remain open
during follow-up (95% secondary patency), a minority of patients will maintain patent shunts without redilation and
stenting (48% primary patency). Consequently, almost half of our patients required repeated procedures in the radiology
suite to maintain patency by dilating and restenting stenotic (34%) or occluded (13%) shunts. Based on the results of
follow-up portal venography, which was most often performed as surveillance in asymptomatic patients, significant shunt
stenosis developed in more than half of our patients by 6 months after TIPS, and narrowing seemed to progress with time.
Although only one in five of our patients with stenosis or occlusion of the TIPS detected during follow-up suffered
documented recurrent variceal bleeding, shunt malfunction was invariably present in our patients with documented
recurrent variceal bleeding, a consistent observation reported by other centers [5] [21] [22] .
The correlation of recurrent variceal bleeding to TIPS malfunction suggests that monitoring and maintaining TIPS patency
can minimize bleeding and improve patient outcome; however, this has not yet been demonstrated in prospective trials.
Occasionally, narrowing and/or occlusion of the shunt occurs within days or weeks after TIPS placement because of shunt
thrombosis or technical problems related to TIPS placement. Progressive TIPS narrowing often develops over a period of
weeks to months as a result of exuberant pseudointimal formation within the stent lumen or intimal hyperplasia within the
hepatic vein adjacent to the distal end of the stent [19] . It has been suggested that administering heparin to patients without
advanced coagulopathy may reduce shunt occlusion, but this hypothesis has not been tested in prospective trials [24] .
Coating TIPS shunts with prosthetic materials used in endovascular grafts is another innovation that has been tested in
experimental animals and holds promise for minimizing stenosis within the TIPS tract [25] .
The development of shunt stenosis and occlusion has important implications for the optimal long-term application of TIPS.
The future role of TIPS in the management of variceal hemorrhage will depend on the cost and effectiveness of strategies
to detect, treat, and prevent shunt narrowing and occlusion. Based on the observed correlation of recurrent bleeding with
shunt dysfunction, we believe monitoring shunt patency and redilation of narrowed and occluded stents results in less
recurrent hemorrhage. However, only prospective evaluation, including detailed cost/efficacy analysis, will determine
whether aggressive monitoring of TIPS patency and revision of asymptomatic TIPS stenosis is superior to an expectant
approach of detecting and correcting TIPS malfunction only after symptoms of recurrent portal hypertension have

developed.
In the era of liver transplantation, it is essential to consider the transplant candidacy of patients with advanced liver disease
in weighing the options for management of variceal bleeding. For the liver transplant candidate, TIPS provides the
theoretical advantages of avoiding the risks of general anesthesia, laparotomy, potential wound-healing problems,
development of portosystemic collateral vessels within postoperative adhesions after shunt surgery, and development of
esophageal ulcerations after endoscopic therapy. However, care must be taken at the time of the TIPS procedure to avoid
placement of stents too proximal in the portal vein or too distal in the hepatic vein, which may complicate hepatectomy at
the time of transplantation.
REFERENCES
1. Matloff DS. Treatment of acute variceal bleeding. Gastroenterol Clin North Am 1992;21:103-18.
2. Westaby D. Prevention of recurrent variceal bleeding. Gastroendosc Clin North Am 1992;2:121-35.
3. Burroughs AK, McCormick PA. Prevention of variceal bleeding. Gastroendosc Clin North Am 1992;21:119-47.
4. Henderson JM, Kutner MH, Millikan WJ, et al. Endoscopic variceal sclerosis compared with distal splenorenal shunt to prevent recurrent variceal bleeding in
cirrhosis: A prospective randomized trial. Ann Intern Med 1990;112:262-9.
5. LaBerge JM, Somberg KA, Lake JR, et al. Two-year outcome for variceal bleeding: Results in 90 patients. Gastroenterology 1995;108:1143-51.
6. Cello JP, Grendell JH, Crass RA, et al. Endoscopic versus portocaval shunt in patients with severe cirrhosis and acute variceal hemorrhage. N Engl J Med
1987;316:11-5.
7. Pagliaro L, Burroughs AK, Sorensen TIA. Therapeutic controversies and randomized controlled trials (RTCs): Prevention of bleeding and rebleeding in
cirrhosis. Gastroenterol Int 1989;2:71-84.
8. Henderson JM. Surgical measures in prevention of recurrent variceal bleeding. Gastrointest Clin N Am 1992;2:151-66.
9. Rosch J, Hanafee WN, Snow H. Transjugular portal venography and radiologic portocaval shunt. Radiology 1969;92:112-4.
10. Colapinto RF, Stronell RD, Birch SJ, et al. Creation of an intrahepatic portosystemic shunt with a Gruntzig balloon catheter. Can Med Assoc J 1982;126:267-8.
11. Rosch J, Hanifee WN, Snow H, et al. Transjugular intrahepatic portosystemic shunt: An experimental work. Am J Surg 1971;121:588-92.
12. Rosch J, Uchida BS, Putnam JS, et al. Experimental intrahepatic portocaval anastomosis: Use of expandable Gianturco stents. Radiology 1987;162:481-5.
13. Richter GM, Noeldge G, Palmaz JC, et al. The transjugular intrahepatic portosystemic stent-shunt (TIPSS): Results of a pilot study. Cardiovasc Intervent Radiol
1990;13:200-7.
P1452
14. Ring EJ, Lake JR, Robert JP, et al. Using transjugular intrahepatic portosystemic shunts to control variceal bleeding before liver transplantation. Ann Intern
Med 1992;116:304-9.
15. LaBerge JM, Ring EJ, Lake JR, et al. Transjugular intrahepatic portosystemic shunts: Preliminary results in 25 patients. J Vasc Surg 1992;16:258-67.
16. Rosch J, Uchida BT, Barton RE, et al. Coaxial catheter needle for portal vein entrance: Technical note. J Vasc Interv Radiology 1993;4:145-7.
17. Pugh RNH, Murray-Lyon IM, Dawson JL, et al. Transection of the oesophagus for bleeding oesophageal varices. Br J Surg 1973;60:646-9.
18. Albers I, Hartmann H, Bircher J, et al. Superiority of the Child-Pugh classification to quantitative liver function test for assessing prognosis of liver cirrhosis.
Scand J Gastroenterol 1989;24:269-76.
19. LaBerge JM, Ferrell LD, Ring EJ, et al. Histopathologic study of stenotic and occluded transjugular intrahepatic portosystemic shunts. J Vasc Interv Radiol
1991;2:549-56.

20. Rossle M, Richter GM, Noldge G, et al. Performance of an intrahepatic portacaval shunt (PCS) using a catheter technique: A case report. Hepatology
1988;8:1348 (abstract).
21. LaBerge JM, Ring EJ, Gordon RL, et al. Creation of transjugular intrahepatic portosystemic shunts with the Wallstent endoprosthesis: Results in 100 patients.
Radiology 1993;187:413-20.
22. Rossle M, Haag K, Ochs A, et al. The transjugular intrahepatic portosystemic stent-shunt procedure for variceal bleeding. N Engl J Med 1994;330:165-71.
23. Conn HO. Transjugular intrahepatic portal-systemic shunts: The state of the art. Hepatology 1993;17:148-58.
24. Freedman AM, Sanyal AJ, Tisnado J, et al. Complications of transjugular intrahepatic portosystemic shunt: A comprehensive review. Radiographics
1993;13:1185-210.
25. Nishimine K, Saxon R, Kichikawa K, et al. Improved transjugular intrahepatic portosystemic shunt patency with PTFE-covered stent-grafts: Experimental
results in swine. Radiology 1995;196:341-7.


April 22, 2000
Conde Petra
Patient selection is important in studying the

impact of large-volume paracentesis on
intravascular volume [editorial; comment]
Citation
Bibliographic Data
Indexing Data
Patient selection is important in studying
the impact of large-volume paracentesis
on intravascular volume [editorial;
comment]
Runyon BA - Am J Gastroenterol - 1997 Mar; 92(3): 371-3
Full Text From NIH/NLM MEDLINE, HealthSTAR
Frontmatter NLM Citation ID:
About the Publication 97221407
Comment:
Am J Gastroenterol 1997 Mar;92(3):394-9
Full Source Title:
Publication Type:
Comment; Editorial
Language:
English
Authors:
Runyon BA
Major Subjects:
Ascites / Diet Therapy / Drug Therapy / * Therapy /
Urine
* Paracentesis / Adverse Effects
* Patient Selection
Administration, Oral
Albumins / Therapeutic Use
Colloids / Therapeutic Use

Diet, Sodium-Restricted
Diuretics / Administration & Dosage / Therapeutic Use
Human
Liver Cirrhosis / Physiopathology
Sodium / Urine
Water-Electrolyte Balance
(Albumins); (Colloids); (Diuretics); 7440-23-5 (Sodium)

April 22, 2000
Conde Petra
Citation
Bibliographic Data
Abstract
Indexing Data
Akriviadis EA - Am J Gastroenterol - 1997 Apr; 92(4): 567-75
From NIH/NLM MEDLINE; NCI CANCERLIT
NLM Citation ID:
97274153
Full Text Full Source Title:
Frontmatter
Publication Type:
INTRODUCTION Journal Article; Review; Review, Tutorial
SPONTANEOUS Language:
HEMOPERITONEUM IN PATIENTS
English
WITH ASCITES
Author Affiliation:
Cirrhosis with ruptured varices or
lymphatic channels
Fourth Medical Unit, University of Thessaloniki, Hippocration
Hospital, Greece.
Authors:
Metastatic liver tumors Akriviadis EA
Ovarian carcinoma Number of References:
Miscellaneous causes
87
POST-TRAUMATIC Abstract:
HEMOPERITONEUM IN PATIENTS OBJECTIVES: To review existing data on the pathophysiology
WITH ASCITES and clinical presentation of hemoperitoneum in patients with
ascites and to familiarize practicing clinicians who take care of
CONCLUSION such patients with the therapeutic options currently available for
REFERENCES management of this complication. METHODS: Relevant
English-language articles published between January 1988 and
About the Publication November 1996 were identified through MEDLINE search,
using the key words "hemoperitoneum" and "ascites." Articles
cited in the bibliographies of these articles were searched
manually. Published papers that contained data on
hemoperitoneum in general and on hemoperitoneum developing
in patients with ascites were reviewed. RESULTS: In patients

with ascites, hemoperitoneum can develop spontaneously or can
follow abdominal trauma and diagnostic or therapeutic
procedures. Spontaneous bleeding into ascites usually develops
insidiously and may not cause hemodynamic instability, even in
patients with cirrhosis. Massive acute hemoperitoneum from a
ruptured intraperitoneal varix is an unusual complication of
portal hypertension requiring prompt surgical treatment. Acute
hemoperitoneum develops in 5-15% of patients with
hepatocellular carcinoma requiring transcatheter arterial
embolization. Metastatic liver tumors cause bloody ascites
infrequently; however, this is a common complication of
ovarian carcinoma. CONCLUSIONS: Hemoperitoneum is a
severe complication in patients with ascites. When it develops
spontaneously, it is usually related to the same disease process
that caused the formation of ascites. Massive bloody ascites
develops acutely after the rupture of intra-abdominal varices or
hepatocellular carcinoma and requires aggressive interventional
management. Based on a review of published data and on
personal experience with patients suffering from end-stage liver
disease, I propose an algorithm for the evaluation and treatment
of patients with cirrhosis and hemoperitoneum.
Major Subjects:
Ascites / * Complications / Diagnosis
Hemoperitoneum / Diagnosis / * Etiology
Abdominal Injuries / Complications
Carcinoma, Hepatocellular / Complications
Esophageal and Gastric Varices / Complications
Female
Human
Liver Cirrhosis / Complications
Liver Neoplasms / Complications
Male
Bookmark URL: /das/journal/view/N/9470347?source=CL,MI


Volume 92 Number 4 April 1997
567
Clinical review
Hemoperitoneum in Patients with Ascites
Evangelos A. Akriviadis M.D.
Fourth Medical Unit, University of Thessaloniki, Hippocration Hospital, Thessaloniki, Greece
Objectives: To review existing data on the pathophysiology and clinical presentation of

hemoperitoneum in patients with ascites and to familiarize practicing clinicians who take care of
such patients with the therapeutic options currently available for management of this
complication. Methods: Relevant English-language articles published between January 1988 and
November 1996 were identified through MEDLINE search, using the key words
"hemoperitoneum" and "ascites." Articles cited in the bibliographies of these articles were
searched manually. Published papers that contained data on hemoperitoneum in general and on
hemoperitoneum developing in patients with ascites were reviewed. Results: In patients with
ascites, hemoperitoneum can develop spontaneously or can follow abdominal trauma and
diagnostic or therapeutic procedures. Spontaneous bleeding into ascites usually develops
insidiously and may not cause hemodynamic instability, even in patients with cirrhosis. Massive
acute hemoperitoneum from a ruptured intraperitoneal varix is an unusual complication of portal
hypertension requiring prompt surgical treatment. Acute hemoperitoneum develops in 5-15% of
patients with hepatocellular carcinoma requiring transcatheter arterial embolization. Metastatic
liver tumors cause bloody ascites infrequently; however, this is a common complication of ovarian
carcinoma. Conclusions: Hemoperitoneum is a severe complication in patients with ascites. When
it develops spontaneously, it is usually related to the same disease process that caused the
formation of ascites. Massive bloody ascites develops acutely after the rupture of intra-abdominal
varices or hepatocellular carcinoma and requires aggressive interventional management. Based on
a review of published data and on personal experience with patients suffering from end-stage liver
disease, I propose an algorithm for the evaluation and treatment of patients with cirrhosis and
hemoperitoneum.
http://home.mdconsult.com/das/article/body/jorg=journal&source=CL,MI&sp=9470347&sid=0/N/68938/1.html (1 of 12) [22/04/2000 0:25:27]

Reprint requests and correspondence: Evangelos Akriviadis, M.D., Rancho Los Amigos Medical Center, USC Liver Unit,
7601 E. Imperial Highway, Downey, CA 90242.
Received Aug. 20, 1996;

accepted Dec. 23, 1996.
INTRODUCTION
Trauma and nonmalignant gynecological conditions account for more than 90% of intra-abdominal hemorrhage [1] . In
male patients, the most common cause of hemoperitoneum is traumatic rupture of spleen or liver, and in women, rupture of
an ectopic pregnancy [2] . In addition to traumatic causes and gynecological diseases, other conditions associated with
hemoperitoneum include neoplastic, inflammatory, and vascular disorders [3] and rupture of an intra-abdominal viscus [4] .
In exceptional cases, hemoperitoneum develops spontaneously as a result of abnormal hemostasis [ e.g., hemophilia [5] ]
(Table 1) . In most patients, however, intraperitoneal bleeding stems from structural lesions rather than from
thrombocytopenia or depletion of clotting factors [6] .
The clinical presentation of hemoperitoneum is variable; blood is a minor peritoneal irritant, and the intensity of abdominal
pain is related to the rapidity and volume of extravasation. Peritoneal signs usually are absent. Subcutaneous hemorrhagic
infiltration of the periumbilical area (Cullen's sign) can appear as soon as several days after the occurrence of
intra-abdominal hemorrhage in patients not requiring laparotomy [7] . In the acute setting, useful diagnostic information
can be obtained by emergency laparoscopy [8] [9] .
The management of patients with hemoperitoneum can be even more challenging when this potentially life-threatening
complication develops in the presence of ascites. All of the conditions mentioned in Table 1 can cause intra-abdominal
hemorrhage in patients with ascites, and they often pose difficult diagnostic and therapeutic dilemmas that are related both
to the hemorrhage and to the severity of the underlying disease process that caused the initial formation of ascites. In
patients with ascites, the diagnosis of hemoperitoneum is confirmed by aspiration of hemorrhagic ascitic fluid via
abdominal paracentesis. This is a safe procedure [10] and should be performed on every patient with newly diagnosed
ascites. It should also be performed on patients with known ascites after an episode of esophageal
568
TABLE 1 -- Conditions Associated with Hemoperitoneum

1. Gynecological diseases
2. Abdominal traumas
3. Tumors
4. Inflammatory disorders
5. Vascular disorders
6. Ruptured viscus
7. Abnormal hemostasis
TABLE 2 -- Conditions Associated with Spontaneous and Post-traumatic Hemoperitoneum in Patients with Ascites
Spontaneous

Cirrhosis with ruptured varices or lymphatic channels

Metastatic liver tumors
Ovarian carcinoma
Hemorrhagic pancreatitis
Tuberculous peritonitis
Other inflammatory disorders
Perforated viscus
Continuous ambulatory peritoneal dialysis
Posttraumatic
Blunt abdominal trauma
Postprocedural
Liver biopsy
Fine-needle aspiration
TIPS
Laparoscopy
Paracentesis (rarely)
variceal bleeding (in view of their high risk for spontaneous bacterial peritonitis) or whenever unexplainable clinical
deterioration (such as hypotension, hepatic encephalopathy, renal impairment) [11] [12] or laboratory abnormality ( e.g.,
leukocytosis or increasing serum creatinine) occurs. Ascitic fluid becomes hemorrhagic when its erythrocyte count is
>50,000/mul [hematocrit value of approximately 0.5% [6] ]. In patients with massive hemoperitoneum, ascites becomes
grossly hemorrhagic, and its hematocrit can reach values close to 40% [13] , often exceeding that of the peripheral blood. In
cases of grossly hemorrhagic ascites, it is useful to repeat paracentesis at a distant site immediately, to exclude the
possibility of puncture of a dilated peritoneal vessel at the first tap. It has been proposed that, in the presence of tense
ascites, removal of bloody ascitic fluid should be minimal, because increased intra-abdominal pressure in tense ascites
represents an important mechanical factor for spontaneous hemostasis [3] .
In patients with ascites, hemoperitoneum can develop spontaneously or can follow abdominal traumas and the
performance of diagnostic or therapeutic procedures (Table 2) . This article is a focused review of the pathophysiological
and clinical features of hemoperitoneum in the presence of ascites. Evolving therapeutic options with regard to the
treatment of patients with this potentially lethal condition are also discussed.

SPONTANEOUS HEMOPERITONEUM IN PATIENTS WITH ASCITES

Hemorrhagic ascites is encountered in only 5% of patients with cirrhosis; it rarely develops abruptly and, thus,
infrequently causes hemodynamic instability [14] . In most cirrhotic patients, the finding of spontaneous hemorrhagic
ascites is incidental and is not associated with clinical evidence of bleeding. Even in the presence of hemodynamic
stability, however, spontaneous hemoperitoneum in cirrhotic patients may be a poor prognostic sign, because it is
associated with an increased risk for hepatorenal syndrome and encephalopathy and a high mortality rate [14] . In patients
with cirrhosis, hemoperitoneum usually develops from an identifiable structural lesion, such as hepatocellular carcinoma
(HCC), rupture of ectopic intra- and retroperitoneal varices, trauma, ovarian carcinoma, hemorrhagic pancreatitis, a
perforated viscus, or tuberculous peritonitis [6] [14] . However, in a third of the patients, no specific cause is found [14] , and
in these patients, bleeding from a small vessel or leakage of hepatic lymph presumably is the mechanism of bloody ascites.
Hemorrhagic ascites can also develop in the absence of liver disease. Spontaneous bleeding into the peritoneal cavity
under these circumstances is usually related to the disease process that caused ascites in the first place.
Cirrhosis with ruptured varices or lymphatic channels
Two mechanisms have been proposed for the development of hemoperitoneum in patients with chronic liver disease [14] .
The most obvious is intra-abdominal bleeding from an organ, a small peritoneal vessel, or an abdominal cavity varix.
Another mechanism is leakage from a dilated liver lymphatic [the erythrocyte count of liver lymph is usually high [15] ].
The hematocrit value of ascitic fluid in patients with bleeding from small peritoneal vessels or a leaking liver lymphatic is
typically less than 5% and remains stable or decreases over weeks to months [14] . A characteristic feature of bloody ascites
from ruptured liver lymphatics is that the white blood cell differential count of the ascitic fluid is predominantly
mononuclear and typically does not fit with the polymorphonuclear pattern of the peripheral blood.
Intra-abdominal hemorrhage from ruptured varices is an unusual though severe complication of portal hypertension, not
widely appreciated by many clinicians. The occurrence of extensive collateral circulation resulting from long-lasting portal
hypertension is well known [16] . The development of functional anastomoses between the portal and systemic circulation
occurs at sites where veins, draining into the two systems, are juxtaposed [17] . Such communications exist not only
through esophageal, umbilical, and rectal veins, but also where organs supplied by the splanchnic
569
circulation contact the retroperitoneum [18] . In addition, previous abdominal operations or inflammatory diseases of the
peritoneal cavity have been associated with the development of mesenteric varices by creating new portosystemic
communications in the postoperative and inflammatory adhesions between the bowel and the parietal peritoneum [17] [19] .
Ectatic veins also have been described in the gallbladder bed after portal vein thrombosis. Although bleeding from these
collateral vessels is usually intraluminal [20] , intraperitoneal hemorrhage also may occur. Bleeding has been reported from
a variety of such ectopic varices [17] [19] [21] [22] [23] . In 25% of those cases, the site of bleeding was a ruptured umbilical or
paraumbilical vein [23] .
Rupture of intra-abdominal varices was first recognized as a complication of cirrhosis in 1958 [1] . Several case reports
have been published since then, but the true frequency of this complication of portal hypertension is difficult to estimate,
because laparotomy is usually required to confirm the diagnosis. Angiography can locate the site of bleeding by
demonstrating active extravasation during the venous phase in only one-third of the cases [2] .
The clinical presentation of acute hemoperitoneum from a ruptured intra-abdominal varix includes sudden abdominal pain
and diffuse tenderness without rigidity, followed by severe hypotension in most cases. A characteristic clinical feature is
the rapid enlargement of abdominal girth, in contrast to the relatively slow accumulation of nonhemorrhagic cirrhotic
ascites. Such precipitous abdominal swelling in a cirrhotic patient, especially if associated with abdominal pain, is highly
suggestive of hemoperitoneum and demands immediate diagnostic paracentesis [18] . Ascitic fluid hematocrit is typically
>5% and may be even higher than that of the peripheral blood [17] . The age of the patient developing this complication is
usually less than 50 yr [17] [18] [23] . In contrast, hemoperitoneum from a ruptured HCC is encountered, for the most part, in
patients in the 6th decade of life [24] . A history of bleeding from gastroesophageal varices is usually absent [18] .

Despite the generally poor condition of these patients, the only definitive therapy for rupture of an intra-abdominal varix is
operative exploration with plication of the bleeding vessel. Whether this should be coupled with portosystemic
anastomosis is a matter of discussion [17] [19] , but once the diagnosis of a bleeding intra-abdominal varix has been made,
the patient's hepatic reserve and ability to withstand a major procedure are probably most important in determining
ultimate prognosis [17] . The role of the transjugular intrahepatic portosystemic shunt (TIPS) in patients with bleeding
peritoneal varices was evaluated recently in a small series of four patients, culminating in a report of promising results [25]
.
Prognosis is poor and mortality rate exceeds 75% [2] . There have been no reports of survivors among patients treated
without operation [2] . In most cases, however, death is directly attributable to hepatic failure [17] , suggesting that this
complication of portal hypertension is merely a reflection of terminal liver disease. Thus it is conceivable that the
frequency of hemoperitoneum from ruptured intraperitoneal varices may be underestimated in patients with tense ascites
and end-stage liver failure.
Rupture of HCC is rarely seen in countries with a low incidence of this primary liver tumor [26] but is a frequent
complication in those populations in which HCC commonly occurs [27] . Thus, acute hemoperitoneum caused by tumor
rupture has been reported in 5% or less of patients with HCC in Western countries [3] , 5.5% in Taiwan [24] , 2.9-14% in
Japan [13] , 12.7% in South Africa [28] , 12% in Thailand [29] , and 14.5% in Hong Kong [30] . In a study from the United
States [31] , acute hemoperitoneum due to tumor rupture was the presenting symptom in 20% of symptomatic patients with
HCC. The rate of spontaneous hemoperitoneum is high (14.3%), and the prognosis is poor, even in cases of HCC in a
noncirrhotic liver [32] . However, most patients with HCC [65% [29] to 80% [13] ] have established cirrhosis. The
association with cirrhosis is even higher [81% [13] to 93% [29] ] in patients with a ruptured HCC.
Hemoperitoneum from a ruptured HCC is often a dramatic emergency and carries an extremely grave prognosis [27] [28] [29]
. It is characteristically spontaneous, although occasionally it can be induced by blunt abdominal trauma [28] . Spontaneous
hemoperitoneum from HCC usually constitutes a life-threatening condition because of hypovolemic shock due to massive
blood loss into the peritoneal cavity, underlying liver cirrhosis, and extensive tumor growth [30] [33] [34] . Moreover, an
association with lactic acidosis and hyperphosphatemia has been reported [35] . The clinical presentation can be so dramatic
that occasionally a misdiagnosis of acute appendicitis [24] or perforation of peptic ulcer [13] is made. Spontaneous rupture
of the tumor in the peritoneal cavity is one of the common causes of death in patients with HCC [13] ; it is the terminal fatal
event in 10% of Japanese patients with HCC [36] . However, it is possible that both the rate of spontaneous rupture of HCC
and its contribution to fatality are underestimated, because invasive investigation is commonly deferred in terminal
patients with HCC who present with abdominal pain, abdominal distention, hypotension, and anemia [3] . The overall
prognosis is poor, even in patients who survive the acute phase of this complication. One of the important factors
responsible for this bad outlook appears to be the high (46%) prevalence of intraperitoneal dissemination [37] , in marked
contrast to the low (3.8%) overall intraperitoneal dissemination rate in patients with unruptured HCC [33] . Ironically, the
prognosis can be less grave when traumatic rupture of HCC occurs, leading to an earlier diagnosis of the tumor, at a time
when it might be amenable to some form of treatment [28] .
Although the mechanism(s) of spontaneous hemorrhage from HCC is still a matter of controversy [24] , the following
explanations have been offered for the propensity of this tumor or of the attenuated overlying liver tissue to rupture
570
[3] [24] [28] [29]

: 1) subcapsular tumor localization, 2) tumor necrosis, 3) regional increase of venous pressure, secondary to
obstruction of the venous outflow by direct tumor invasion, and 4) portal hypertension caused by associated cirrhosis. It
has been suggested that in patients with and without cirrhosis, the rupture of HCC may have different pathogenetic
backgrounds [38] .
An accurate diagnosis of HCC can be made by abdominal computed tomography (CT) [26] . In the presence of major active
bleeding, dynamic CT scan can show extravasation of contrast material [39] . In some patients, areas of high attenuation
representing blood clots (localized hematoma or "sentinel clot") are found around the liver, usually close to the tumor [40] .
Occasionally, an actual disruption of the wall of a peripherally located mass lesion is detected [26] . In a patient with

cirrhosis, grossly hemorrhagic ascites, and a mass lesion in the liver, the diagnosis of ruptured HCC can be made with
confidence. Angiography is not required as an additional diagnostic procedure; its sensitivity in detecting HCC is high, but
it fails to reliably document active bleeding from a ruptured HCC, showing extravasation of the contrast material from the
tumor in only 13% [37] to 23.5% [41] of the cases. However, its role is very important as a therapeutic procedure to control
bleeding. In several case reports [42] [43] , radionuclide imaging with 99m Tc-labeled red blood cells has proved useful in
diagnosing the cause of hemoperitoneum; however, its role has not been evaluated in a sufficient number of patients with
cirrhosis and hemoperitoneum.
Control of intra-abdominal bleeding from a ruptured HCC is often difficult, and the mortality rate is high [29] [30] .
Treatment is aimed at controlling the hemorrhage and resecting the abnormal liver, when possible [24] . Although an
aggressive surgical approach has been advocated [29] [30] , many patients are poor surgical candidates because of
impairment of liver function secondary to cirrhosis and/or extensive tumor replacement of the liver [41] . Surgical mortality
of patients with a ruptured HCC is 50% or higher [29] [30] [37] .
HCC is almost exclusively nourished by arterial blood, whereas the nontumorous portion of the liver is supplied from
portal blood flow. Cutting off the blood supply to the tumor by transcatheter arterial embolization (TAE) of the feeding
artery is generally well tolerated, because it usually does not cause a significant impairment of liver function [13] , and it
also reduces the degree of portal hypertension in patients with cirrhosis [24] . For this procedure to be performed, previous
confirmation of adequate portal blood flow is mandatory [24] . TAE is effective in achieving immediate hemostasis in
almost all patients, even in the presence of massive hemoperitoneum [37] [41] [44] . Its immediate mortality rate is 18% [41] ,
far less than that of surgical ligation of the hepatic artery [29] [30] . Overall hospitalization mortality of TAE is 26-29% [37]
[41] . Although TAE is a useful therapeutic modality for the management of spontaneous rupture of HCC, even in patients
with advanced stages of this malignant disease [13] , its mortality rate appears to be exceptionally high in patients with
bilirubin levels >3.0 mg/dl. Such patients often have limited hepatic reserve for ischemia and cannot tolerate transarterial
chemotherapy or embolization [37] . The presence of portal vein invasion is not a contraindication (providing that complete
obstruction is not present) and is not associated with a higher mortality rate [37] . TAE in cirrhotic patients with acute
hemoperitoneum substantially increases short-term survival [13] and is considered the procedure of choice for this
catastrophic complication of HCC [37] . It has significant advantages over the surgical approach, particularly in debilitated
patients who are poor surgical candidates. Furthermore, in patients with adequate hepatic reserve, it can be combined with
surgical resection of the tumor, 1-8 wk after control of bleeding [24] .
Metastatic liver tumors
Ascites can develop in patients with metastatic liver cancer, many of whom also have disseminated peritoneal
carcinomatosis. In contrast to HCC, metastatic liver carcinoma is a notably rare cause of hepatic rupture and massive
hemoperitoneum [45] , with fewer than 50 cases reported worldwide. This probably reflects the tendency of metastatic
tumors to be more fibrotic [46] , less vascular and invasive, and to penetrate the liver capsule less frequently than HCC [6] .
Primary sites of cancer are in the lung, pancreas, stomach, colon, gallbladder, ovaries, kidney, breast, prostate, testicles,
and nasopharynx [45] [46] [47] [48] [49] . A few cases of malignant melanoma [45] and metastatic choriocarcinoma [50] also
have been reported. Bleeding from a metastatic liver tumor can be profound and lead to death from exsanguination [45] .
Therapy is palliative rather than curative, and if surgery is undertaken, its goal should be to control hemorrhage quickly
and effectively [46] . Regardless of treatment, the prognosis is very poor [46] .
Ovarian carcinoma
Most cases of hemoperitoneum in the general population are gynecological in origin [1] . Ovarian cancer represents a
substantial portion of these cases. This cancer commonly causes ascitic fluid production, frequently of hemorrhagic
appearance, associated with multiple peritoneal tumor implants [6] . In contrast to peritoneal carcinomatosis from ovarian
tumors, other types of neoplastic deposits in the peritoneum seldom evoke bloody ascites [6] . Presentation with an acute
massive hemoperitoneum is uncommon in epithelial ovarian cancer [51] ; it is most commonly associated with germ cell or
stromal tumors [52] . Occasionally, massive bloody ascites requiring transfusions develops in cases of cancer metastatic to
the ovaries [53] , because vascular invasion predisposing to massive hemoperitoneum occurs more often with metastatic
rather than primary ovarian malignancies [53] .
571

Pancreatic ascites develops in the course of acute or chronic pancreatitis when a major pancreatic duct is ruptured and the
leaking pancreatic juice is not contained by the surrounding structures. In exceptional cases, pancreatic ascites becomes
grossly hemorrhagic, secondary to bleeding from intraperitoneal rupture of pancreatic pseudocysts [54] [55] .
Tuberculous peritonitis is an infrequent cause of hemorrhagic ascites [14] . Grossly hemorrhagic ascites develops in only a
small percentage of cases [6] . Other intraperitoneal inflammatory disorders infrequently associated with the presence of
hemorrhagic ascites include the Fitz-Hugh Curtis syndrome [hemoperitoneum due to minor trauma has been reported [56] ],
systemic lupus erythematosus [57] , primary (retractile) mesenteritis of childhood [58] , and splenic infarct associated with
sterile peritonitis [59] .
In patients with free perforation of an intra-abdominal viscus, ascites may develop within hours to days, although the
abrupt onset of acute abdominal pain associated with signs of peritoneal irritation usually leads to prompt surgical
exploration. Acute hemoperitoneum is infrequently encountered in such cases, and for the most part, is found in patients
with acute cholecystitis and perforation of the inflamed gallbladder [4] [60] [61] . Intra-abdominal hemorrhage in the case of
gallbladder perforation has been associated with anticoagulation therapy [60] . It may also occur with transhepatic
perforation of the gallbladder [61] , in which case massive subsequent bleeding from the hepatic parenchyma makes it a
life-threatening complication with a reported mortality rate of 67% [4] .
Hemorrhagic ascites is a well-recognized complication in menstruating women maintained on continuous ambulatory

peritoneal dialysis. It occurs just before or with menstruation, in which case it is attributed to retrograde menstruation [62]
[63] , but it has also been reported with ovulation [63] [64] . Occasionally, massive hemoperitoneum can be caused by a
ruptured ovarian cyst [64] [65] or a hemorrhagic luteal cyst [66] , requiring transfusions [64] or active intervention [63] .
Hemorrhagic ascites in patients maintained on continuous ambulatory peritoneal dialysis can also be caused by
catheter-related traumas [67] , sclerosing peritonitis [63] , and previous radiation therapy for an intra-abdominal malignancy
[68] .
Nodular regenerative hyperplasia is characterized histologically by diffuse involvement of the liver by hyperplastic
nodules composed of cells resembling normal hepatocytes [69] . It is a noncirrhotic lesion, unrelated to focal nodular
hyperplasia, and is associated with portal hypertension in almost 50% of cases [69] . Ascites has been described in 38%,
and hemoperitoneum with associated hypotension in 19%, of the cases [69] . The source of intra-abdominal bleeding is
commonly a ruptured large hepatic nodule that can cause exsanguination, although fatal intraperitoneal hemorrhage from a
portosystemic collateral has also been reported [70] . Treatment is by surgical resection [69] .
POST-TRAUMATIC HEMOPERITONEUM IN PATIENTS WITH

ASCITES
In the general population, more than 20% of cases of hemoperitoneum are related to trauma [17] . This figure probably is
higher among alcoholics, a substantial percentage of which develop cirrhosis and ascites. Overall bloody ascites is
associated with trauma in 18% of cirrhotic patients [14] . A traumatic cause of acute hemoperitoneum in such patients can
be entirely unsuspected, because history is commonly inadequate or unreliable. The spleen is the organ most commonly
injured by blunt abdominal trauma [71] , and the risk of splenic injury is probably even higher in cirrhotic patients with
splenomegaly. Abdominal CT is an accurate and reliable diagnostic modality for locating the site of intra-abdominal
bleeding due to trauma [26] [71] , and it can help substantially in the clinical management of such cases. Its sensitivity and
specificity in diagnosing splenic injury is above 95% [71] . Radionuclide scintigraphy is of limited value in patients with
abdominal trauma. It may visualize liver and splenic lacerations and hematomas, but it fails to provide information about
other abdominal traumatic lesions that frequently coexist in such patients [72] . Acute traumatic hemoperitoneum in
cirrhotic patients is a life-threatening condition that commonly requires prompt surgical exploration, with resultant high
mortality. Clinical evaluation is the most important factor in deciding to perform laparotomy, because CT findings fail to
determine reliably which patients can be successfully treated conservatively and which need surgery [39] .
In some cases, however, bleeding into ascites results from invasive diagnostic or therapeutic procedures, commonly
employed in patients with liver disease. Significant hemorrhage after liver biopsy reportedly occurs at a frequency of

0.21% [73] to 0.35% [74] , almost exclusively in patients with cirrhosis or malignant disease [75] . Fatal hemoperitoneum
develops in 0.11% of cases [74] . Recently, an increased frequency of fatal hemoperitoneum secondary to percutaneous
liver biopsy has been reported in patients with AIDS [76] . After liver biopsy, the risk of hemorrhage is related to
perforation of distended portal or hepatic veins or aberrant arteries and cannot be predicted by standard coagulation
screening [76] . However, risk is increased with multiple passes and in patients of advanced age or with malignant liver
disease [74] . As two potentially important safety measures, it has been advocated that the procedure be performed under
ultrasound guidance [73] [77] and with the use of a smaller caliber needle [77] . However, the use of ultrasound-guided liver
biopsy increases the cost of the procedure substantially, does not decrease mortality, and should be reserved for selected
cases [78] .
Fatal hemoperitoneum occurs after transvenous liver biopsy, as well, and is associated with liver capsule perforation
572
[79]. Post-biopsy bleeding can be controlled by emergency transcatheter arterial embolization [79] . Fatal hemoperitoneum
has also been reported after fine-needle aspiration of the liver, exclusively in patients with either malignant disease or
hemangioma of the liver [80] [81] . A strong association with chronic subclinical disseminated intravascular coagulation has
also been implicated, particularly in patients with widespread malignancies [80] . Massive and even fatal hemoperitoneum
can also occur during the placement of TIPS, resulting from traversal of the liver capsule by the needle or from
extrahepatic puncture of the portal vein during the phase of portavenous shunt stent placement [82] . Hemoperitoneum
develops infrequently (0.18%) during diagnostic laparoscopy. Major hemorrhage occurs more often (0.6%) and may be
life-threatening if laparoscopy is combined with liver biopsy [83] . However, in a cirrhotic patient who underwent
diagnostic laparoscopy without liver biopsy in our unit, hemoperitoneum was caused by puncture of an abdominal wall
varix by the trocar. Laparotomy with suture ligation of the varix led to immediate hemostasis.
Paracentesis per se is very rarely associated with bleeding complications. The overall risk for transfusion-requiring
hemorrhage is 0.2% [84] [85] , and despite a recent report of four cases of massive hemoperitoneum subsequent to
therapeutic paracentesis [25] , the risk is not considered clinically significant, even in patients with moderate [85] or severe
[86] coagulopathy. In our experience with a large number of patients suffering from decompensated liver disease and severe
coagulopathy, ascitic fluid rarely became grossly hemorrhagic, even after several repeat paracenteses. Based on this
experience, we do not recommend routine preprocedure administration of fresh frozen plasma or platelets even for patients
with severe coagulopathy. However, the risk of post-paracentesis hemorrhage may be significant in the presence of
primary fibrinolysis, which, although rare, can occur in patients with end-stage liver disease. If, under those circumstances,
a strong indication for diagnostic paracentesis is present, our policy is to perform it after improvement of the euglobulin
lysis time with Amicar. A technique to minimize the risk of causing significant bleeding has been proposed, namely, the
slow advancement of the paracentesis needle, in 5-mm increments, through the abdominal wall. If a flash of blood is
aspirated in the syringe attached to the needle, it is withdrawn to avoid further damage to the vessel [84] . To avoid
puncturing abdominal wall arterial branches or collateral vessels, paracentesis can be done at the midline, caudad to the
umbilicus, with the patient in a semi-recumbent position. In my experience, bleeding is extremely rare when paracentesis
is performed at that site, despite a recent report suggesting the frequent presence of vascular channels in that area [87] . In
view of the rare occurrence of paracentesis-induced hemorrhage in our unit, we do not use ultrasound guidance to locate an
avascular site for a safe needle stick.
CONCLUSION
Hemoperitoneum in patients with ascites poses several diagnostic and therapeutic challenges. Diagnostic considerations in
patients with spontaneous hemoperitoneum and cirrhosis include life-threatening conditions such as rupture of HCC or of
an abdominal cavity varix (Table 2) . The most common cause of spontaneous hemoperitoneum in patients without
cirrhosis is ovarian carcinoma. Post-traumatic hemoperitoneum in patients with ascites develops commonly after blunt
abdominal injuries. Because history is often unreliable, a high index of suspicion is required in such cases. Procedures
associated with the development of hemoperitoneum include biopsy and fine-needle aspiration of the liver, TIPS, and
laparoscopy. Paracentesis per se is a very rare cause of hemorrhage in the ascitic fluid.
Based on review of published data and on my personal experience with patients presenting with decompensated liver
disease, I propose an algorithm for the diagnostic evaluation and care of cirrhotic patients presenting with spontaneous

hemoperitoneum (Fig. 1) . Immediate repeat paracentesis should be performed at a distant site in all patients with grossly
hemorrhagic ascites. If distant site paracentesis yields clear fluid, the diagnosis of traumatic paracentesis is made. The
prognosis for this complication is good, and therapeutic intervention is not required, even in the presence of severe
coagulopathy. If grossly hemorrhagic ascites is aspirated from both sites, a CT scan of the abdomen, looking for HCC or
localized hematoma from an unsuspected abdominal trauma, should be done promptly. Laparotomy is commonly required
in patients with hemoperitoneum secondary to blunt abdominal trauma who present with hemodynamic instability. In
patients with hemoperitoneum and a mass lesion in the liver compatible with HCC, angiography with arterial embolization
is the rational therapeutic approach, because surgery in such patients is associated with unacceptable mortality. In patients
without CT findings suggestive of localized hematoma or HCC and ascitic fluid hematocrit <5%, the source of blood is a
small superficial peritoneal vessel or a ruptured liver lymphatic. Although hemodynamic stability is preserved in these
patients, the prognosis may be poor, particularly in patients with ruptured lymphatics, because this complication of portal
hypertension is frequently associated with the development of hepatorenal syndrome. Hence, these patients should be
considered potential candidates for liver transplantation. If ascitic fluid hematocrit is >5% in patients with a CT scan
negative for localized hematoma or HCC, the most likely cause of hemoperitoneum is a ruptured abdominal cavity varix.
These patients usually develop hemodynamic instability and require immediate therapeutic intervention with exploratory
laparotomy or TIPS (Fig. 1) . In patients with cirrhosis who develop hemoperitoneum after a diagnostic or therapeutic
procedure, the prognosis can be serious. If hypotension resistant to fluid challenge develops, an invasive procedure such as
laparotomy to suture
573
Figure 1. Algorithm of evaluation and treatment of patients with cirrhosis and spontaneous hemoperitoneum. Patients with blunt
abdominal traumas are included, because the history of injury for these patients is not always present. AF, ascitic fluid; TAE,
transarterial embolization; PMN, polymorphonuclear; MN, mononuclear; Ht, hematocrit.
a bleeding vessel and evacuate a hematoma or angiography with transarterial embolization is usually required.
REFERENCES
1. Ellis H, Griffiths PWW, McIntyre A. Haemoperitoneum: A record of 129 consecutive patients with notes on some unusual cases. Br J Surg 1958;45:606-10.
2. Fox L, Crane SA, Bidari C, et al. Intra-abdominal hemorrhage from ruptured varices. Arch Surg 1982;953-6.
3. Berera T, Ferrini M, Dal Zennaro E, et al. Spontaneous rupture of hepatocellular carcinoma. Ital J Gastroenterol 1992;24:461-3.
4. Safran D. Massive hemoperitoneum from transhepatic perforation of gallbladder. Surgery 1990;107:356 (letter).
5. Jones JJ, Kitchens CS. Spontaneous intra-abdominal hemorrhage in hemophilia. Arch Intern Med 1984;144:297-300.
6. Natelson EA, Allen TW, Riggs S, et al. Bloody ascites: Diagnostic implications. Am J Gastroenterol 1969;52:523-7.
7. Chung MA, Oung C, Szilagyi A. Cullen's sign: It doesn't always mean hemorrhagic pancreatitis. Am J Gastroenterol 1992;87:1026-8.
8. Berci G, Sackier JM, Paz-Partlow M. Emergency laparoscopy. Am J Surg 1991;161:332-5.
9. Sackier JM. Laparoscopy in the emergency setting. World J Surg 1992;16:1083-8.
10. Runyon BA. Paracentesis of ascitic fluid: A safe procedure. Arch Intern Med 1986;146:2259-61.
11. Runyon BA. Spontaneous bacterial peritonitis: An explosion of information. Hepatology 1988;8:171-5.
12. Runyon BA. Bacterial infections in patients with cirrhosis. J Hepatol 1993;18:271-2.

13. Hirai K, Kawazoe Y, Yamashita K, et al. Transcatheter arterial embolization for spontaneous rupture of hepatocellular carcinoma. Am J Gastroenterol
1986;81:275-9.
14. DeSitter L, Rector WG. The significance of bloody ascites in patients with cirrhosis. Am J Gastroenterol 1984;79:136-8.
15. Dumont AG, Mulholland JH. Flow rate and composition of thoracic duct lymph in patients with cirrhosis. N Engl J Med 1960;263:471-5.
16. Rothschild JJ, Gelernt I, Sloan W. Ruptured mesenteric varix in cirrhosis: Unusual cause for hemoperitoneum. N Engl J Med 1968;278:97-8.
17. Lyon DT, Mantia AG, Schubert TT. Hemoperitoneum from a ruptured varix in cirrhosis: Case report and literature review. Am J Gastroenterol 1979;71:611-6.
18. Shapero TF, Bourne RH, Goodall RG. Intra-abdominal bleeding from variceal vessels in cirrhosis. Gastroenterology 1978;74:128-9.
19. Moncure AC, Waltman AC, Vandersalm TL, et al. Gastrointestinal hemorrhage from adhesion-related mesenteric varices. Ann Surg 1976;183:24-9.
20. Jhung JW, Micolonghi TS. Ruptured mesenteric varices in hepatic cirrhosis. A rare cause of intraperitoneal hemorrhage. Surgery 1985;97:377-80.
21. Ragupathi K, Bloom A, Pai N. Hemoperitoneum from ruptured omental varices. J Clin Gastroenterol 1985;7:537-8.
22. Paizis B, Krespis E, Filiotou A, et al. Rupture of a periumbilical vein causing hemoperitoneum in a cirrhotic patient. Mt Sinai J Med 1986;53:123-5.
23. Goldstein AM, Gorlick N, Gibbs D, et al. Hemoperitoneum due to
574
spontaneous rupture of the umbilical vein. Am J Gastroenterol 1995;90:315-7.
24. Chen M-F, Hwang T-L, Jeng L-B, et al. Surgical treatment for spontaneous rupture of hepatocellular carcinoma. Surg Gynecol Obstet 1988;167:99-102.
25. Arnold C, Ochs A, Haag K, et al. Acute hemoperitoneum: A rare complication of large volume paracentesis and indication for the transjugular intrahepatic
portosystemic shunt (TIPS). Hepatology 1996;24(No. 4, Pt 2):479A (abstract).
26. Clarkston W, Inciardi M, Kirkpatrick S, et al. Acute hemoperitoneum from rupture of a hepatocellular carcinoma. J Clin Gastroenterol 1988;10:221-5.
27. The Liver Cancer Study Group of Japan. Primary liver cancer in Japan. Cancer 1984;54:1747-55.
28. Kew MC, Hodkinson J. Rupture of hepatocellular carcinoma as a result of blunt abdominal trauma. Am J Gastroenterol 1991;86:1083-5.
29. Chearanai O, Plegvanit U, Asavanich G, et al. Spontaneous rupture of primary hepatoma. Cancer 1983;51:1532-6.
30. Ong GB, Taw JL. Spontaneous rupture of hepatocellular carcinoma. Br Med J 1972;4:146-9.
31. Luna G, Florence L, Johansen K. Hepatocellular carcinoma: A 5-year institutional experience. Am J Surg 1985;149:591-4.
32. Smalley SR, Moertel CG, Hilton JF, et al. Hepatoma in the noncirrhotic liver. Cancer 1988;62:1414-24.
33. Nakashima T, Kojiro M. Hepatocellular carcinoma: an atlas of its pathology. In: Nakashima T, Kojiro M, eds. Hepatocellular carcinoma. Tokyo:
Springer-Verlag, 1986;117-32.
34. Nagasue N, Inokuchi K. Spontaneous and traumatic rupture of hepatoma. Br J Surg 1979;66:248-50.
35. Berenson MM, Lee RG. Lactic acidosis associated with cirrhosis, hepatoma, hemoperitoneum, and hypephosphatemia. J Clin Gastroenterol 1986;8:210-2.
36. Liver Cancer Study Group of Japan. Primary liver cancer in Japan: Clinicopathologic features and results of surgical treatment. Ann Surg 1990;211:277-87.
37. Okazaki M, Higashihara H, Koganemaru F, et al. Intraperitoneal hemorrhage from hepatocellular carcinoma: Emergency chemoembolization or embolization.
Radiology 1991;180:647-51.
38. Lin D, Liaw Y, Chu C. Hepatocellular carcinoma in noncirrhotic patients. A laparoscopic study of 92 cases in Taiwan. Cancer 1984;54:1466-8.
39. Becker CD, Spring P, Glattli A, et al. Blunt splenic trauma in adults: Can CT findings be used to determine the need for surgery? AJR 1994;162:343-7.
40. Pombo F, Arrojo L, Perez-Fontan J. Haemoperitoneum secondary to spontaneous rupture of hepatocellular carcinoma: CT diagnosis. Clin Radiol

1991;43:321-2.
41. Hsieh J-S, Huang C-J, Huang Y-S, et al. Intraperitoneal hemorrhage due to spontaneous rupture of hepatocellular carcinoma: Treatment by hepatic artery
embolization. AJR 1987;149:715-7.
42. Czarnecki DJ. Intraperitoneal hemorrhage diagnosed by technetium-99m labeled RBC imaging. Clin Nucl Med 1986;11:617-8.
43. Bajakian RL, Swayne LC, Palace FM. Scintigraphic detection of hemobilia and hemoperitoneum secondary to rupture of hepatic artery aneurysm. Clin Nucl
Med 1988;13:831-2.
44. Nouchi T, Nishimura M, Maeda M, et al. Transcatheter arterial embolization of ruptured hepatocellular carcinoma associated with liver cirrhosis. Dig Dis Sci
1984;29:1137-41.
45. Schoedel KE, Dekker A. Hemoperitoneum in the setting of metastatic cancer to the liver. A report of two cases with review of the literature. Dig Dis Sci
1992;37:153-4.
46. Urdaneta LF, Nielsen JV. Massive hemoperitoneum secondary to spontaneous rupture of hepatic metastases: Report of two cases and review of the literature. J
Surg Oncol 1986;31:104-7.
47. Ben-Baruch D, Yampolski I, Ziv Y. Massive haemoperitoneum complicating metastasis in the liver. Eur J Surg Oncol 1990;16:462-3.
48. Mittleman RE. Hepatic rupture due to metastatic lung carcinoma. Am J Clin Pathol 1987;88:506-9.
49. Cunningham LN, Ginsberg P, Manfrey S, et al. Massive hemorrhage secondary to metastatic testicular carcinoma. J Am Osteopath Assoc 1989;89:341-4.
50. Erb RE, Gibler WB. Massive hemoperitoneum following rupture of hepatic metastasis from unsuspected choriocarcinoma. Am J Emerg Med 1989;7:196-8.
51. Geisler JP, Denman BJ, Cudahy TJ, et al. Ovarian carcinoma presenting as intra-abdominal hemorrhage. Gynecol Oncol 1994;53:380-1.
52. Charters JW, Prince G, McGarry JM. Granulosa cell tumour presenting with haemoperitoneum and splenic rupture. Br J Obstet Gynecol 1989;96:735-6.
53. Nelson BE, Carcangiu ML, Chambers JT. Intra-abdominal hemorrhage with pulmonary large cell carcinoma metastatic to the ovary. Gynecol Oncol
1992;47:377-81.
54. Lefemine S, Catucci V, Sebastiani R. Spontaneous hemoperitoneum caused by hemorrhage and the rupture of pancreatic pseudocysts. Minerva Chir
1989;44:1611-6.
55. Ruotolo F, Ferrini L, Petrangeli SM, et al. Hemoperitoneum in segmental portal hypertension caused by pancreatic pseudocyst. Minerva Chir 1989;44:909-13.
56. Foster HM. Haemoperitoneum: An unusual complication of Fitz-Hugh Curtis syndrome. Aust N Z J Surg 1988;58:342-3.
57. Sonpal GM, Abramovici B. Acute abdomen in systemic lupus erythematosus with spontaneous hemoperitoneum. J Rheumatol 1987;14:636-7 (letter).
58. Davis CF, Guzzetta PC, Patterson K. Primary (retractile) mesenteritis in a child. J Pediatr Surg 1992;27:1544-5.
59. Syed A, Holley JL, Piraino B. Splenic infarct presenting as sterile peritonitis with peripheral embolic phenomena. Adv Perit Dial 1993;9:202-5.
60. Gecelter GR, Schein M. Haemoperitoneum: A complication of acute cholecystitis in a patient on anticoagulant therapy: Case report. S Afr Med J 1989;76:216.
61. Syme RG, Thomas EJ. Massive hemoperitoneum from transhepatic perforation of the gallbladder: A rare complication of cholelithiasis. Surgery
1989;105:556-9.
62. Blumenkrantz MJ, Gallagher N, Bashore RA, et al. Retrograde menstruation in women undergoing chronic peritoneal dialysis. Obstet Gynecol 1981;57:667-70.
63. Greenberg A, Bernardini J, Piraino BM, et al. Hemoperitoneum complicating chronic peritoneal dialysis: Single-center experience and literature review. Am J
Kidney Dis 1992;19:252-6.
64. Harnett JD, Gill D, Corbett L, et al. Recurrent hemoperitoneum in women receiving continuous ambulatory peritoneal dialysis. Ann Intern Med
1987;107:341-3.
65. Fraley DS, Johnston JR, Bruns FJ, et al. Rupture of ovarian cyst: massive hemoperitoneum in continuous ambulatory peritoneal dialysis patients: Diagnosis and
treatment. Am J Kidney Dis 1988;12:69-71.
66. Handa SP, Scarth H. Hemorrhagic luteal cyst with massive hemoperitoneum in a patient on CAPD. Perit Dial Int 1993;13:65-6.

67. Verrina E, Perfumo F, Zacchello G, et al. Chronic peritoneal dialysis catheters in pediatric patients: Experience of the Italian Registry of Pediatric Chronic
Peritoneal Dialysis. Perit Dial Int 1993;13(suppl 2):S254-6.
68. Hassell LH, Moore J Jr, Conklin JJ. Hemoperitoneum during continuous ambulatory peritoneal dialysis: A possible complication of radiation-induced
peritoneal injury. Clin Nephrol 1984;21:241-3.
69. Dachman AH, Ros PR, Goodman ZD, et al. Nodular regenerative hyperplasia of the liver: Clinical and radiologic observations. AJR 1987;148:717-22.
70. Ben-Ari Z, Mc Cornick AP, Jain S, et al. Spontaneous haemoperitoneum caused by ruptured varices in a patient with non-cirrhotic portal hypertension. Eur J
Gastroenterol Hepatol 1995;7:87-90.
71. Gay SB, Sistrom CL. Computed tomographic evaluation of blunt abdominal trauma. Radiol Clin North Am 1992;30:367-88.
72. Federle MP. Computed tomography of blunt abdominal trauma. Radiol Clin North Am 1983;21:461-75.
73. Caturelli E, Giacobbe A, Facciorusso D, et al. Percutaneous biopsy in diffuse liver disease: Increasing diagnostic yield and decreasing complication rate by
routine ultrasound assessment of puncture site. Am J Gastroenterol 1996;91:1318-21.
74. McGill DB, Rakela J, Zinsmeister AR, et al. A 21-year experience with major hemorrhage after percutaneous liver biopsy. Gastroenterology 1990;99:1396-400.
75. Piccinino F, Sagnelli E, Pasquale G, et al. Complications following percutaneous liver biopsy: A multicentre retrospective study on 68,276 biopsies. J Hepatol
1986;2:165-73.
76. Gordon SC, Veneri RJ, McFadden RF, et al. Major hemorrhage after percutaneous liver biopsy in patients with AIDS. Gastroenterology 1991;100:1787 (letter).
575
77. Jacobs WH, Goldberg SB, and the Patient Care Committee of the American Gastroenterological Association. Statement on outpatient percutaneous liver biopsy.
Dig Dis Sci 1989;34:322-3.
78. Stotland BR, Lichtenstein GR. Liver biopsy complications and routine ultrasound. Am J Gastroenterol 1996;91:1295-6 (editorial).
79. Braillon A, Revert R, Remond A, et al. Transcatheter embolization of liver capsule perforation during transvenous liver biopsy. Gastrointest Radiol
1986;11:277-9.
80. Edoute Y, Kaplan J, Ben-Haim SA, et al. Haemoperitoneum induced by fine-needle aspiration of liver in patients with disseminated intravascular coagulation.
Lancet 1992;339:121-2.
81. Fornari F, Civardi G, Cavanna L, et al. Complications of ultrasonically guided fine-needle abdominal biopsy. Results of a multicenter Italian study and review
of the literature. The Cooperative Italian Study Group. Scand J Gastroenterol 1989;24:949-55.
82. Freedman AM, Sanyal AJ, Tisnado J, et al. Complications of transjugular intrahepatic portosystemic shunt: A comprehensive review. Radiographics
1993;13:1185-210.
83. De Groen PC, Rakela J, Moore SC, et al. Diagnostic laparoscopy in gastroenterology: A 14-year experience. Dig Dis Sci 1987;32:677-81.
84. Runyon BA. Ascites. In: Schiff L, Schiff ER, eds. Diseases of the liver, 7th ed. Philadelphia: J. B. Lippincott, 1993:990-1015.
85. McVay PA, Toy PTCY. Lack of increased bleeding after paracentesis and thoracentesis in patients with mild coagulation abnormalities. Transfusion
1991;31:164-71.
86. Wolkin Friedman E, Sussman II. Safety of invasive procedures in patients with the coagulopathy of liver disease. Clin Lab Haematol 1989;11:199-204.
87. Caldwell SH, Oelsner DH, Coles M, et al. Subumbilical midline vascularity of the abdominal wall in portal hypertension observed at laparoscopy. Hepatology
1996;24(No 4, Pt 2):212A (abstract).

Bookmark URL: /das/journal/view/N/9470347?ja=68938&PAGE=1.html&ANCHOR=top&source=CL,MI

April 22, 2000
Conde Petra
CLINICAL ISSUES IN THE MANAGEMENT OF

Additional Article
Full Text
Frontmatter 765
MAKING THE DIAGNOSIS

Laboratory and Clinical Findings
Diagnosis of Alcohol Abuse

and Alcoholic Liver Disease in CLINICAL ISSUES IN THE MANAGEMENT
Asymptomatic Subjects
OF ALCOHOLIC LIVER DISEASE
Diagnosis of Alcoholic Liver
Disease in Symptomatic
Patients
Future Trends
Petra E. Steindl MD
Differentiation from Nonalcoholic
Steatohepatitis Peter Ferenci MD
Concurrent Hepatitis C
Department of Internal Medicine IV, Division of Gastroenterology and
Iron Overload Hepatology, University of Vienna, Vienna, Austria
NEED FOR LIVER BIOPSY Address reprint requests to
Petra E. Steindl, MD
DRUG INTERACTION OF ALCOHOL Department of Internal Medicine IV
WITH SPECIAL FOCUS ON Division of Gastroenterology and Hepatology
TYLENOL University of Vienna
Waehringer Guertel 18-20
MANAGEMENT OF ALCOHOL 1090 Vienna, Austria
WITHDRAWAL
MUSCLE CRAMPS The syndrome of alcoholic liver disease (ALD) includes a

huge variety of clinical manifestations that are the result of
OTHER ORGAN INJURY three factors--hepatocellular insufficiency, portal
Heart hypertension, and extrahepatic damage by alcohol. Overt
hepatic disease (hepatomegaly, jaundice, ascites, and portal
Pancreatic Function in Alcoholic
hypertension) may be accompanied by digestive, endocrine,
Liver Disease
hematologic, muscular, neural, and other extrahepatic
Central Nervous System manifestations.
References In this article, some of the more important issues in clinical
routine are addressed, as are some of the controversies
concerning establishment of the diagnosis.

Because of its wide spectrum of clinical presentations, alcoholic

liver disease (ALD) poses great difficulties for diagnosis. The
diagnostic challenges include the detection of alcohol abuse in
asymptomatic subjects and the differentiation of ALD from
other liver diseases or from the laboratory changes induced by
ethanol.
Chronic alcohol abuse is associated with three distinct
histologic lesions in the liver--alcoholic steatosis, alcoholic
hepatitis, and alcoholic cirrhosis. The lesions may overlap and
are not associated with distinct clinical symptoms and signs.
The whole spectrum of alcoholic liver disease, from
asymptomatic hepatomegaly to profound hepatocellular failure
with portal hypertension, therefore can be observed in relation
to any of the histologic subtypes. It also is
766
important to recognize that the degree of clinical illness

associated with the development of alcoholic liver disease can
vary enormously. No clinical feature distinguishes alcoholic
liver disease from other liver diseases. Even in patients with
established ALD, the frequency and severity of particular
symptoms vary. In general, women seem to suffer more florid
symptoms of liver disease. Stigmata of liver disease, such as
spider teleangiectasia, gynecomastia, and hepatosplenomegaly,
or signs of liver failure, such as jaundice, ascites, and
encephalopathy, occur more frequently in older patients and in
those with cirrhosis. [40]
There is no single reliable test that allows a precise diagnosis of
ALD. Ethyl alcohol (ETOH) consumption per se may result in

changes of routine laboratory tests independent from liver

disease.
1. Metabolic changes, such as hypertriglyceridemia,
hypercholesterolemia, or hyperuricemia, are common in
alcoholics but certainly occur in a large number of
nonalcoholic subjects. Even in alcoholics, these findings
are not necessarily the consequence of ethanol
consumption but of concurrent overnutrition. Lipid
abnormalities lead to hepatic fat accumulation that can be
easily detected by sonography. Hepatic steatosis of any
cause may be associated with increases in alanine
aminotransferase (ALT) and other enzymes. Increased
liver enzymes in someone consuming ethanol, therefore,
are not reliable evidence for ethanol abuse.
2. Hepatic ETOH degradation is a powerful inducer of
mitochondrial oxidation and is associated with
biochemical and morphologic changes. Increased
production and release of the mitochondrial isoenzyme of
AST (mAST) and of gamma-glutamyltranspeptidase
(GGTP), [83] therefore, reflect long-term alcohol abuse.
Increased serum AST or GGTP can occur in any liver
disease, however, and, therefore, are nonspecific. They
can occur in the absence of liver disease if the patient is
on medication with drugs that are inducing hepatic
oxidation (e.g., barbiturates).
3. Pancreatic disease frequently is seen in alcoholics.
Stenosis of the common bile duct by pancreatic disease
results in cholestasis and an increase of alcaline
phosphatose (AP) and GGTP. Abnormal GGTP,
therefore, is not a sensitive indicator of ALD. The
differentiation of chronic pancreatitis with cholestasis
from chronic pancreatitis associated with ALD is a
diagnostic challenge but has important implications for
the management of the patient.
4. Drug metabolism is altered in chronic alcohol abusers
(discussed later), increasing the potential hepatotoxicity
of drugs such as paracetamol or isoniazide. Increases in
transaminase levels, therefore, are multifactorial and,
certainly, nonspecific.
Diagnosis of Alcohol Abuse and Alcoholic Liver Disease in
Asymptomatic Subjects
The prevalence of asymptomatic ALD is unknown. The
frequency in the general population cannot be calculated from
data obtained in hospitalized patients or mortality registers.

Based on questionnaires, 10% of the adult population of two

towns in northern Italy were considered alcohol abusers. [11]
Asymptomatic chronic liver disease was common (one out of
six "healthy" adults), and ETOH was considered as a causative
factor in 23% of cases. The same authors [10] found that the
threshold levels for developing cirrhosis was 30 g/day of
ethanol. The risk increased in parallel with greater daily alcohol
intake. Drinking alcohol outside of mealtimes and drinking
multiple different alcoholic beverages further augmented the
risk of alcoholic liver damage. A careful history of the amount
767
and duration of alcohol intake, the type of drinks consumed, and

the drinking pattern, therefore, is needed to relate alcohol
consumption with ALD.
Several routinely available laboratory tests are useful in the
detection of excessive drinking but are not always reliable
indicators of liver disease. Elevated GGTP activity is common
in individuals abusing alcohol, [24] especially those drinking
more than 80 g of alcohol per day. [9] [58] [99]
Laboratory Parameters in the Diagnosis of ALD
Gamma-glutamyltransferase (GGT)
Macrocytosis (MCV )
AST > ALT; AST/ALT > 2
Mitochondrial alanine aspartate aminotransferase

(mAST)/total alanine aspartate aminotransferase (tAST)
CDT
Gamma-globulins (Ig A )
Uric acid
Lactate

Serum albumin
Prothrombin time
Triglycerides
The usefulness of GGTP to detect alcohol abuse, however, is

limited by its low specificity and sensitivity. Up to 70% of
alcohol abusers may have normal values. On the other hand,
obesity, treatment with certain drugs, and hepatobiliary diseases
are associated with increased GGTP irrespective of the extent of
alcohol consumption.
Serum transaminase levels are not markedly increased in ALD.
Serum AST levels are only raised two to six times in severe
acute alcoholic hepatitis. Levels of AST greater than 500 U/L
point to another cause. The level of AST is higher than that of
ALT and a ratio of AST to ALT greater than two is present in
about 70% of cases. [21] The mitochondrial isoenzyme of AST
(mAST) is a sensitive marker of chronic alcoholism. [73] The
ratio of mAST to total AST was significantly increased in
alcoholics but was not related to the degree of liver disease. [72]
This indicates the mAST-to-tAST ratio is a marker of chronic
alcohol abuse rather than of ALD.
Macrocytosis (increased mean cell volume [MCV]) is common
in individuals abusing alcohol. It lacks sensitivity (27%-52%)
but is reasonably specific for alcohol use greater than 50 g/d. [9]
[99] It probably reflects the toxic effect of alcohol on the
developing erythrocyte. [91] The combination of raised GGTP

activity and MCV has a sensitivity for detection of alcohol of
30% to 40%.
Serum carbohydrate-deficient transferrin (CDT) is a specific
and sensitive test of alcoholism, irrespective of complicating
liver disease, but is not readily available. [8] In patients with
alcohol intake greater than 10 g/d there is a positive correlation
between CDT and alcohol intake. [9] This topic is covered in
more depth in the article by Marchesini et al in this issue.
Frequent determination of blood ETOH at every outpatient
laboratory control is helpful in determining patients' compliance
with abstinence.
Diagnosis of Alcoholic Liver Disease in Symptomatic Patients
There are routinely available laboratory parameters that may

provide some clues that alcohol is the cause of liver disease.
Characteristic but nonspecific findings include elevations in
uric acid, lactate, and triglycerides, and reductions
768
in potassium, glucose phosphate, and magnesium. A polyclonal

hypergammaglobulinemia is present in most patients with ALD,
with a disproportionate increase in circulating IgA. [43] [96]
Serum albumin and prothrombin time decrease as liver function
declines or as the nutritional status of the patient worsens. It
should be emphasized, however, that no laboratory test is
specific for ALD.
In a cholestatic variant of alcoholic hepatitis, very high levels of
alcaline phosphatase may be present, but this is also seen in a
small proportion of patients with fatty change alone. [6] [69]
Zieve's syndrome, a combination of fatty liver (with or without
cirrhosis) with hyperlipidemia and hemolysis, is also a
characteristic but infrequent picture. [100]
Future Trends
A sophisticated tool in the diagnosis of ALD is hepatic
phosphorus-31 magnetic resonance spectroscopy, a method that
provides information on hepatic energy metabolism and
phospholipid membrane metabolism. The pattern of peak area
ratios of phosphomonoesters (PME), inorganic phosphate, and
phosphodiesters (PDE) relative to betaATP, and of PME
relative to PDE changes significantly according to the cause and
the degree of liver disease. [67] [68] In the patients with
compensated disease, those with alcoholic cirrhosis showed a
significantly lower PDE/ATP ratio, [68] although, in patients
with minimal liver injury, recent alcohol intake was associated
with a significant increase in the PDE/ATP ratio. [67]
Differentiation from Nonalcoholic Steatohepatitis
Nonalcoholic steatohepatitis (NASH) is a form of chronic liver

disease that is defined by biopsy findings and has the
appearance of alcoholic hepatitis. NASH is more common in
women than in men and it appears to be associated with obesity,
diabetes mellitus, and related metabolic abnormalities, such as
hyperlipidemia and hyperglycemia. The association with female
gender, obesity, and diabetes is not as close as suggested,
however, and an underlying condition cannot always be

discerned. In a group of 33 patients with NASH, 58% were

men, 61% were nonobese, 79% were normoglycemic, and 79%
had normal lipid levels. No metabolic abnormality was present
in 42%, and it was, therefore, concluded that NASH has to be
considered in an expanded group of patients with abnormal
liver enzymes and a negative biochemical and serologic
work-up. [5] Still, obesity is the condition most frequently
associated with NASH (69%-100% in different studies). [88]
There is no test that differentiates NASH from alcoholic
steatohepatitis. The diagnosis and differential diagnosis of
NASH currently is established on morphologic grounds alone,
[47] but the final diagnosis can be made in only the context of the
clinical history. [74] The most important feature is the exclusion

of significant alcohol consumption. How much alcohol
constitutes significant consumption is unknown. It has been
shown that steatosis develops relatively rapidly in nondrinkers
after a controlled period of alcohol consumption. [10] [55]
Several studies have demonstrated a high prevalence of

hepatitis C virus (HCV) antibody (anti-HCV) in patients with
ALD. HCV infection has an important
769
role in the progression of ALD. [19] [36] [63] [78] Concomitant HCV
infection is associated with increased severity of liver injury in
patients with ALD. [66] [71] [75] [76] In serologic surveys, [46] the
prevalence of HCV was increased in patients with ALD. Most
patients with ALD with concomitant HCV infection have
identifiable parenteral risk factors. [63] Patients with HCV
infection have more severe histologic features, [65] [87] a
decreased survival rate, [66] and an earlier development of
disease [19] than those without HCV infection. Patients with
active HCV viremia have histologic features of both chronic
hepatitis [94] and ALD. The histologic features of
anti-HCV-reactive patients without viremia resemble those of
patients without serologic evidence of HCV infection. [30]
The natural history of fibrosis progression in patients with HCV
is associated with three independent factors--age at infection
older than 40 years, daily alcohol consumption of 50 g or more,
and male gender. [80] The mean stage of fibrosis at biopsy was
significantly higher in patients with daily alcohol consumption
greater than 50 g than in those who consumed less, irrespective

of age or duration of infection. The findings confirm that

alcohol is a potential risk factor in patients with HCV and that
the combination of HCV and alcohol is likely to result in a
severer course of liver disease.
Iron Overload
The consumption of excessive amounts of alcohol affects

human iron homeostasis, and iron overload and heavy alcohol
consumption frequently are associated. [29] [41] Lipid
peroxidation induced by ethanol or iron plays a major role in
hepatic damage, both in humans and in experimental animals.
Although the exact mechanism(s) of induction of lipid
peroxidation by ethanol and iron remains to be elucidated, both
toxins can exert a synergistic effect upon hepatic lipid
peroxidation. Both major proteins of iron metabolism, ferritin
and transferrin, as well as transferrin saturation are increased by
alcohol. Hepatic iron levels are increased in a high proportion
of alcoholic subjects, [86] sometimes causing confusion in
diagnosis between alcoholic liver disease and iron-overload
disease.
The pattern of hepatic iron deposition in alcoholics, however,
differs from that in hemochromatosis. The relationship between
alcoholism and genetic hemochromatosis has been an issue of
intensive research for many decades. Hepatic siderosis is
present in 57% of patients with alcoholic liver disease, with
pronounced siderosis in 7% of cases. [44] The hepatic iron
concentration in alcoholics is lower than in genetic
hemochromatosis. [12] [50] Histologic tissue iron grading is not
reliable in distinguishing between alcoholic siderosis and
genetic hemochromatosis. There is a poor correlation between
histologic evidence of iron overload and tissue iron
concentrations. [13] [27] In a study of homozygous siblings with
genetic hemochromatosis, [2] the patients with heavy alcohol
consumption had a higher prevalence of cirrhosis than
hemochromatosis patients without heavy alcohol consumption.
Hepatic iron concentration and hepatic iron index did not differ
between patients with hemochromatosis with or without ETOH
abuse. Long-term survival was reduced in patients with heavy
alcohol consumption (mean follow-up, 9.22 years). This
suggests that chronic alcohol consumption in hemochromatosis
has an additive hepatotoxic effect despite the paucity of
histologic features of alcoholic liver disease.
Factors that might be responsible for iron overload in the
alcoholic include: increased intake and absorption of iron;

altered iron use caused by impaired erythropoiesis (folate

deficiency, myelotoxicity of alcohol); repeated bursts of
770
hemolysis (Zieve's syndrome); and iron deposition as a result of

liver damage. [42] Portosystemic shunting also increases hepatic
iron deposition, but the effect is variable. [22] Determination of
the HFE-gene (the abnormal gene in the majority of patients
with hemochromatosis [26] ) mutation status may be useful to
differentiate the various reasons of iron overload in alcoholics.
[4]
NEED FOR LIVER BIOPSY

Liver biopsy plays an important role in the assessment of the
alcoholic patient with abnormal liver function tests. First, it may
be used to establish a diagnosis of alcoholic liver disease. Up to
20% of such patients will have nonalcoholic liver disease. [51]
Secondly, it is essential in determining the stage of disease
because clinical and biochemical indices are poor predictors of
the extent and severity of liver injury. Finally, liver biopsy may
help determine the prognosis.
The limitations of laboratory tests in confirming alcoholic liver
disease suggest that biopsy should be undertaken in the majority
of suspected cases, particularly if a period of abstinence has not
been accompanied by normalization of transaminases over a
period of 3 to 6 months.
Furthermore, the diagnostic value of histologic examination has
been emphasized by several studies. [51] [92] [97] The reversibility
of changes can be predicted, based on an assessment of the
extent of the hepatitis and the amount of pericellular and
perivenular fibrosis, as well as from the degree of architectural
distortion and features of hepatitis, including Mallory bodies,
which disappear on alcohol withdrawal.
Liver biopsy is relatively safe, with an associated morbidity of
0.1% to 0.6% and mortality of 0.01% to 0.03%. [31] [33] [79] There
is also relatively small interobserver variation in the histologic
interpretation of ALD, with greater agreement in samples
containing more than six portal tracts. [7]
The biopsy should not be undertaken as a blind procedure
unless the prothrombin time is within 3 seconds of control time
and the platelet count exceeds 80,000. If clotting abnormalities

persist after adequate substitution (vitamin K, clotting factors)

or there are concerns about the safety of a blind needle
procedure, then the biopsy can be undertaken under radiologic
guidance or can be obtained via the transjugular route or at
peritonoscopy. The specimen obtained using a Menghini needle
usually is adequate.
DRUG INTERACTION OF ALCOHOL WITH

SPECIAL FOCUS ON TYLENOL
It took 20 years of research in ethanol metabolism to culminate
in the molecular elucidation of an alcohol-inducible cytochrome
P450 (P450IIE1) that is involved both with ethanol metabolism
and ethanol tolerance, and with the activation of a number of
xenobiotics. The unique ability of P450IIE1 to activate
xenobiotic agents now appears to be responsible for the
increased susceptibility of the heavy drinker to hepatotoxic
industrial solvents, commonly used drugs, and over-the-counter
medications.
Especially when a substance has hepatotoxic metabolites--for
example, CCI4--the increased activity of the hepatic
microsomal enzymes in the alcoholic patient leads to an
increased conversion of CCI4 to its toxic metabolites.
The activation of the microsomal enzymes also explains some
of the interaction
771
of ethanol with nutritional factors, such as hepatic vitamin A:

Enhanced microsomal degradation of retinoids (together with
hepatic mobilization) promotes depletion. [48] Treatment,
however, is complicated by the fact that ethanol also enhances
the toxicity of excess vitamin A. [49] The list of drugs interacting
with alcohol through the activated P450IIE1 system is very
long. For review, references 53 and 54 are recommended to the
interested reader.
The enhanced hepatotoxicity of acetaminophen (paracetamol) in
alcoholics is also caused by increased activity of the
cytochrome P450 system [54] and decreased hepatic glutathione.
Under normal conditions, glutathione conjugates with the toxic
metabolite to render it nontoxic, while most of the drug is
conjugated with glucuronic acid or sulfate. After acetaminophen
overdose or when glutathione stores are decreased (induced by

alcohol or malnutrition) the protective mechanisms fail,
resulting in severe toxicity of the active metabolite. Clinical
studies have shown that the elimination half-life of
acetaminophen is significantly shorter in alcoholics than in
healthy controls, suggesting that the increased metabolism of
the drug leads to enhanced toxicity. [32] [64]
In a report on 67 patients who developed hepatic injury after
ingestion of acetaminophen, 64% were considered to be
alcoholics. Mortality of this condition was up to 18% in this
study, but 32% in the literature. [45] It is important to note that
the majority of patients reported taking doses of the drug that
were well below the accepted toxic range. [102] The syndrome of
liver injury in these patients is relatively distinctive. The most
important clinical or biochemical characteristic is the
high--often towering--level of AST. In the aforementioned
study, levels exceeded 3000 IU in more than 90% of patients
and the highest levels were at 48,000 IU, so that this marker is
almost pathognomonic. Another laboratory hint to the diagnosis
may be an extremely long prothrombin time. Histologically, the
typical lesion is zone 3 necrosis.
The important issue regarding acetaminophen toxicity is the
physician's awareness of the diagnosis and the possibility that
severe liver injury may occur even at therapeutic doses of this
easily available drug. [25] In an alcoholic who presents with
jaundice and extremely high AST levels, therefore, a careful
history of drug use regarding paracetamol is mandatory.
MANAGEMENT OF ALCOHOL
WITHDRAWAL
The clinical symptoms of alcohol withdrawal can be classified
(Table 1) as early (24-48 hours) and late (> 48 hours) as well as
minor and major. [56] The level of autonomic hyperactivity and
the presence of delirium are the main
TABLE 1 -- SYMPTOMS OF ALCOHOL WITHDRAWAL
Early (24- 48 Hours)/Minor Late (> 48 Hours)/Major
Temperature Disorientation
Tremor Cognitive impairment
Tachycardia Systolic hypertension
Insomnia Tachycardia
Restlessness Nausea
Seizures Vomiting
Agitation
Hallucinations
772
determinants of progression from minor to major symptoms.

The "3 Ts" (temperature, tremor, and tachycardia) of the
delirium tremens may be the earliest signs of alcohol
withdrawal, although the definition true delirium tremens
necessitates the presence of cognitive impairment. Insomnia,
restlessness, agitation, nausea and vomiting, myalgias, tremor,
systolic hypertension, and tachycardia can progress to
disorientation and cognitive impairment. If alcohol
withdrawal-related seizures occur, they usually occur early
during the course of withdrawal. Hallucinosis also can occur
early, especially in patients who have consumed alcohol for a
prolonged period.
The goals of treatment are amelioration of symptoms and
prevention of complications. According to the guidelines of the
American Society of Addiction Medicine, [53] benzodiazepines
are the first-line treatment of acute alcohol withdrawal and they
can reduce the risk of seizures and delirium significantly.
Although the different benzodiazepines seem to be similarly
efficacious in reducing signs and symptoms of withdrawal,
there is some evidence that the longer-acting agents may be
more effective in preventing seizures. [39] [61] [81] Clinical
experience suggests that the longer-acting agents can pose a risk
of excess sedation in selected patient groups, however,
especially the elderly and those with poor liver function.
Another consideration in the choice of benzodiazepine is their
potential for abuse, which is greater in agents with rapid onset
of action (diazepam, alprazolam, lorazepam) than in those with
slower onset of action (chlordiazepoxide, oxazepam,
halazepam). This consideration may be relevant in an outpatient
setting or for patients with a history of benzodiazepine or other
substance abuse.
There are two approaches to the dosage of sedative drugs. In
most studies, medications were given in fixed amounts at
scheduled times (e.g., chlordiazepoxide, 50 mg every 6 hours

for 5 days). An alternative to giving medication on a fixed
schedule, known as symptom-triggered therapy, has been
developed. In this approach, the patient is monitored by means
of a structured assessment scale and given medication only
when symptoms cross a threshold of severity. This has been
shown to be as effective as the fixed-dose regimen, with the
advantage of significantly less medication and significantly
shorter duration of treatment. [57] [85] In the hospital setting,
however, prevention of alcohol withdrawal symptoms by giving
meprobamate (another substance commonly used in Europe),
for example, to any patient admitted with a history of recent
heavy alcohol consumption is practiced successfully.
Other sedative-hypnotic drugs used in alcohol withdrawal are
chlormethiazole (mainly in Europe), [14] barbital, and
tetrabamate. They are equally effective as benzodiazepines.
Several other alternatives to benzodiazepines can be used
adjunctively or, occasionally, independently. beta-adrenergic
blockers are useful for controlling blood pressure and
tachyarrhythmias, but the usual contraindications must be
considered. In addition, delirium is a known side effect of
beta-blockers and, in one study, the incidence of delirium was
even increased with propranolol. [101] The centrally acting
alpha-adrenergic agonist clonidine has been shown to be
effective in ameliorating symptoms of withdrawal (especially in
treating associated hypertension), but it should only be used in
combination with benzodiazepines. [82] Carbamazepine and
valproic acid have been widely used in Europe as
anticonvulsants, being effective in increasing the seizure
threshold, but experience in the United States is limited.
Neuroleptic agents, including the phenothiazines and the
butyrophenone haloperidol, demonstrate some effectiveness in
reducing signs and symptoms of withdrawal, especially in
agitated patients, but increase
773
the incidence of seizures compared with placebo. As a result,

they have to be used with caution.
Magnesium can be used to increase the seizure threshold if the
patient's magnesium level is low. If, however, the initial
magnesium level is normal, magnesium supplementation
probably has little associated benefit.

MUSCLE CRAMPS
Muscle cramps are a common symptom in cirrhosis and are
more prevalent in this disorder than in patients with chronic
hepatitis or congestive heart failure. The incidence of cramps is
unrelated to the amount of diuretics used, the cause of liver
disease (especially ALD), abnormalities in serum electrolytes,
or differences in Child's classification. [1] This suggests that
cramps in these patients are related specifically to the
development of cirrhosis. In Abrams' study, 36 of 92 patients
had alcoholic cirrhosis, but the incidence of cramps in the group
was not higher than in liver cirrhosis of other cause. ALD,
therefore, is not an independent risk factor for the development
of cramps. Similarly, in a study of 294 patients [3] the authors
found a higher prevalence of cramps in cirrhotic patients than in
controls and it was related to the duration of recognized
cirrhosis and the severity of liver function impairment. Their
data suggested that a reduced effective plasma volume may be
involved in the pathophysiology of cramps because there were
significant differences in mean arterial blood pressure (MAP),
plasma renin activity (PRA), urinary sodium excretion, and
plasma volume in cirrhotic patients with and without cramps. At
a multiple-regression analysis, the presence of ascites, low
values of MAP, and high values of PRA were the independent
predictive factors for the occurrence of cramps in cirrhosis.
The effects of a sustained expansion of the effective circulating
volume induced by intravenous infusion of human albumin
were compared with those of a placebo in 12 cirrhotic patients
with more than three cramp crises a week. Compared with the
placebo, albumin reduced the cramp frequency ( P < 0.01).
Albumin therefore could be beneficial in the treatment of
cramps by improving effective circulating volume.
Furthermore, taurine has been proposed as an effective agent to
treat muscle cramps in cirrhotic patients [60] but, to date, there is
no generally accepted and effective treatment of cramps, so
medication is given empirically in most cases.

OTHER ORGAN INJURY

Heart
Chronic alcoholism is one of the most important causes of

dilated cardiomyopathy, and a large proportion of chronic
alcoholics demonstrate impairment of cardiac function. [84] The
development of cardiac dysfunction is apparently related to the
total lifetime dose of ethanol. Studies in experimental animals
have demonstrated that both acute and chronic ethanol
administration impair cardiac contractility. [20] The relationship,
if any, however, between the acute effects of alcohol and the
development of irreversible cardiomyopathy remain to be
elucidated.
The clinical onset of alcoholic cardiotoxicity is insidious, with
nonspecific fatigue, chest discomfort, palpitations, or an
isolated episode of atrial fibrillation. [90]
774
As the cardiomyopathy progresses, manifestations of overt right

and left heart failure become apparent.
Treatment of alcoholic cardiomyopathy is similar to that of
other dilated cardiomyopathies. The prognosis is variable,
depending on the duration and severity of the cardiac
abnormalities. Patients with severe symptoms may progress to
congestive heart failure and die even if they stop drinking. On
the other hand, left ventricular function improved dramatically
in 14 patients with severe end-stage congestive heart failure
attributable to alcoholic cardiomyopathy after 6 months of total
abstinence. [35] In another study, [59] both left ventricular
end-diastolic dimension and ejection fraction improved rapidly
after total abstinence from alcohol intake.
Several studies indicate that women are more susceptible to the
cardiotoxic effects of alcohol. [28] [95] The prevalence of dilated
cardiomyopathy in alcoholic women was similar to alcoholic
men, but total lifetime dose and threshold dose of ethanol for
the development of cardiomyopathy were considerably lower in
women than in men. The decline in ejection fraction with
increasing alcohol dose was significantly steeper.
Pancreatic Function in Alcoholic Liver Disease
There is a great deal of evidence regarding the close

relationship between alcohol consumption and chronic

pancreatitis. Alcohol usually causes a chronic, recurrent,
calcifying pancreatitis and a period of 6 to 12 years of alcohol
consumption seems to be necessary before pancreatic symptoms
occur. [17] There also is evidence, however, that the presence of
ALD has an impact on pancreatic function. To cover this topic
in depth would fill another article, but some of the data are
given here. Some studies [23] [37] have investigated the presence
of changes in pancreatic function in alcoholic patients, with and
without ALD, to detect functional alterations and possible
association of hepatic and pancreatic disease. By comparing
cirrhotic and noncirrhotic patients, they found that the volume
of duodenal juice, bicarbonate output, and amylase output were
greater in cirrhotic than in noncirrhotic alcoholic subjects, but
not in those of nonalcoholic status. Maximal bicarbonate
concentration was decreased in ALC. [37] The authors concluded
that, in ALD, exocrine pancreatic secretion tends to increase
with severity of liver damage, but concurrence of definite
chronic pancreatitis is not correlated with the severity.
In another study, [23] very similar data were obtained. There was
also a tendency to larger volume and lower bicarbonate
concentration when the hepatic lesion was more severe.
Bicarbonate output was significantly higher in patients with
alcoholic cirrhosis. Interestingly, none of the patients had
changes consistent with chronic pancreatitis.
In conclusion, the alterations in pancreatic function parallel the
severity of the liver disease, but the occurrence of alcoholic
pancreatitis seems to be attributable to direct toxic effects of
alcohol, (although the mechanisms are far from clear) and
unrelated to ALD. Still, the association of ALD with
alcohol-induced chronic pancreatitis is very common. [34]
Central Nervous System
Chronic alcoholism results in brain damage and dysfunction,

leading to a constellation of neuropsychiatric symptoms,
including cognitive dysfunction, the
775
Wernicke-Korsakoff syndrome, [98] alcoholic cerebellar

degeneration, and alcoholic dementia. That these clinically
defined entities result from independent pathophysiologic
mechanisms is unlikely. Alcohol and its metabolite
acetaldehyde are directly neurotoxic. Alcoholics are thiamin

deficient as a result of poor diet, gastrointestinal disorders, and

liver disease. In addition, both alcohol and acetaldehyde have
direct toxic effects on thiamin-related enzymes in the liver and
brain. [16]
The relative roles of alcohol toxicity, thiamin deficiency and
cirrhosis of the liver in the pathogenesis of alcohol-related brain
damage are unclear. In alcoholics, liver disease per se results in
altered thiamin homeostasis, cognitive dysfunction, and
neuropathologic damage to astrocytes. Chronic liver disease
may also interfere with brain thiamin homeostasis and therefore
contribute to the pathogenesis of the Wernicke-Korsakoff
syndrome in chronic alcoholism. [15]
The clinical triad described by Wernicke is composed of
ophtalmoplegia, ataxia, and disturbance of consciousness and
the mental state. Ocular abnormalities consist of nystagmus
(horizontal and vertical) and weakness or paralysis of the
external rectus muscles, with impairment of conjugate gaze,
accompanied by diplopia and strabismus. [103]
The ataxia is of stance and gait and can hinder walking or
standing without support in the acute phase. The disturbances of
mental state and consciousness occur mainly as a global
confusional state (occurs in all but 10% of patients), in which
the patient is apathetic, inattentive, and with minimal
spontaneous speech. Treatment consists of parenteral thiamin.
Most patients who do not recover within 48 to 72 hours develop
Korsakoff's psychosis.
Korsakoff's psychosis is characterized by various degrees of
anterograde and retrograde amnesia, with relative preservation
of other intellectual functions. Confabulation is considered a
hallmark of Korsakoff's psychosis. In the early stages of the
disease, confusion is profound and confabulation is evident; in
the convalescent phase, the patient recalls fragments of the past
in a distorted manner. The disease is potentially reversible by
early treatment with thiamin, but recovery is incomplete in
more than 50% of cases.
Other well-recognized syndromes in alcoholic patients include
alcoholic dementia, alcohol withdrawal syndrome, central
pontine myelinolysis and alcoholic cerebellar degeneration, [93]
which is associated with cortical atrophy. [38] Brain atrophy is a
common finding in chronic alcohol abusers, but the shrinkage is
reversible with abstinence. [70] There is no clear-cut
differentiation between the clinical symptoms of
alcohol-induced brain damage and the neuropsychological

changes seen in hepatic encephalopathy. In certain situations,

however, especially in the evaluation for liver transplantation,
the differentiation between the two entities can be important.
Brain MR imaging may be helpful in the differential diagnosis
because the changes seen in Wernicke's encephalopathy [77]
differ considerably from the well-described changes [89] seen in
hepatic encephalopathy (increased signal intensity of the globus
pallidus on T1-weighted images).
References
1. Abrams GA, Concato J, Fallon MB, et al: Muscle cramps in patients

with cirrhosis. Am J Gastroenterol 91:1363, 1996
2. Adams PC, Agnew S: Alcoholism in hereditary hemochromatosis

revisited: Prevalence and clinical consequences among homozygous
siblings. Hepatology 23:724, 1996
3. Angeli
P, Albino G, Carraro P, et al: Cirrhosis and muscle cramps:
Evidence of a causal relationship. Hepatology 23:264, 1996
776
4. Bacon BR: Diagnosis and management of hemochromatosis.

5. Bacon BR, Farahvash MJ, Janney CG, et al: Nonalcoholic

steatohepatitis: An expanded clinical entity. Gastroenterology 107:1103,
1994
6. BallardH, Bernstein M, Farrar JT: Fatty liver presenting as obstructive

jaundice. Am J Med 30:196, 1961
7. Bedossa P, Poynard T, Naveau S, et al: Observer variation in assessment

of liver biopsies of alcoholic patient. Alcohol Clin Exp Res 12:173, 1988
8. BellH, Tallaksen G, Sjahei M, et al: Serum carbohydrate-deficient

transferrin as a marker for alcohol consumption in patients with chronic
liver disease. Alcohol Clin Exp Res 17:246, 1993
9. Bell
H, Tallaksen C, Try K, et al: Carbohydrate deficient transferrin and
other markers of high alcohol consumption: A study of 502 patients
admitted consecutively to a medical department. Alcohol Clin Exp Res
18:1103, 1994
10. BellentaniS, Saccocio G, Costa G, et al: Drinking habits as cofactors of

risk for alcohol induced liver damage. Gut 41:845, 1997

11. BellentaniS, Tiribelli C, Saccoccio G, et al: Prevalence of chronic liver

disease in the general population of northern Italy: The Dionysos study.
12. Bothwell TH, Charlton RW: Hemochromatosis. In Schiff L, Schiff ER

(eds): Diseases of the Liver, ed 6. Philadelphia, JB Lippincott, 1987, p
1001
13. Brissot
P, Bourel M, Herry D, et al: Assessment of liver iron content in
271 patients: A reevaluation of direct and indirect methods.
14. BurroughsAK, Morgan MY, Sherlock S: Double-blind controlled trial

of bromocriptine, chlordiazepoxide and chlormethiazole for alcohol
withdrawal symptoms. Alcohol Alcohol 20:263, 1985
15. Butterworth RF: Cerebral dysfunction in chronic alcoholism: Role of

alcoholic liver disease. Alcohol Alcohol 2(suppl):259, 1994
16. Butterworth RF: Pathophysiology of alcoholic brain damage:

Synergistic effects of ethanol, thiamine deficiency and alcoholic liver
disease. Metab Brain Dis 10:1, 1995
17. Caballeria
J, Pares A: Extrahepatic manifestations of alcoholism. In
McIntyre N, Benhamou JP, Rizetto M, et al (eds): Oxford Textbook of
Clinical Hepatology. Oxford, England, Oxford University Press, 1991, p
856
18. CaldwellS, Jeffers L, Ditomaso A, et al: Antibody to hepatitis C is

common among patients with alcoholic liver disease with and without risk
factors. Am J Gastroenterol 86:1219, 1991
19. Caldwell SH, Li X, Rourk RM, et al: Hepatitis C infection by

polymerase chain reaction in alcoholics: False-positive Elisa results and
the influence of infection on clinical prognostic score. Am J Gastroenterol
88:1016, 1993
20. CapassoJM, Li P, Guideri G, et al: Left ventricular dysfunction induced

by chronic alcohol ingestion in rats. Am J Physiol 261:H212, 1991
21. Cohen JA, Kaplan MM: The SGOT/SGPT ratio--an indicator of

alcoholic liver disease. Dig Dis Sci 24:835, 1979
22. ConnHO: Portocaval anastomosis and hepatic hemosiderin deposition:

A prospective controlled investigation. Gastroenterology 62:61, 1972
23. Del-Olmo Martinez ML, Barba-Bermejo M: Changes in pancreatic

secretion in alcoholic liver disease. Rev Esp Enferm Dig 77:197, 1990
24. DragosicsB, Ferenci P, Pesendorfer FX, et al:

Gamma-glutamyl-transpeptidase in alcoholic liver disease. Acta Med
Austriaca 3:50, 1976
25. Edwards R, Oliphant J: Paracetamol toxicity in chronic alcohol

abusers--a plea for greater consumer awareness. N Z Med J 105:174, 1992
26. FederJN, Gnirke A, Thomas W, et al: A novel MHC class I-like gene is
mutated in patients with hereditary hemochromatosis. Nat Genet 13:399,
1996
27. FerenciP, Datz CH, Kazemi-Shirazi L, et al: HLA-H mutations in

patients with chronic hepatitis C [Abstract]. J Hepatol 28(suppl 1):144,
1998
28. Fernandez-SolaJ, Estruch R, Nicolas JM, et al: Comparison of

alcoholic cardiomyopathy in women versus men. Am J Cardiol 80:481,
1997
29. Fletcher
LM: Alcohol and iron: One glass of red or more? J
Gastroenterol Hepatol 11:1039, 1996
777
30. Fong TL, Kanel GC, Conrad A, et al: Clinical significance of

concomitant hepatitis C infection in patients with alcoholic liver disease.
31. Garcia-Tsao
G, Boyer JL: Outpatient liver biopsy: How safe is it? Ann
Intern Med 118:150, 1993
32. Girre
C, Hispard S, Palombo C, et al: Increased metabolism of
acetaminophen in chronically alcoholic patients. Alcohol Clin Exp Res
17:170, 1993
33. GoldbergSJ, Mendenhall CL, Connell AM, et al: "Nonalcoholic"

chronic hepatitis in the alcoholic. Gastroenterology 72:598, 1977
34. GulloL, Casadei R, Campione O, et al: Alcoholic liver disease in

alcoholic chronic pancreatitis: A prospective study. Ital J Gastroenterol
27:69, 1995
35. GuilloP, Mansourati J, Maheu B: Long-term prognosis in patients with

alcoholic cardiomyopathy and severe heart failure after total abstinence.
Am J Cardiol 79:1276, 1997
36. HalimiC, Deny P, Gotheil C, et al: Pathogenesis of liver cirrhosis in

alcoholic patients histological evidence for hepatitis C virus responsibility.
Liver 11:329, 1991
37. Hayakawa T, Kondo T, Shibata T, et al: Exocrine pancreatic function in

chronic liver diseases. Am J Gastroenterol 86:201, 1991
38. Hayes PC: Extrahepatic manifestations of alcoholic liver disease.

Baillere's Clinical Gastroenterology 7:683, 1993

39. Hill
A, Williams D: Hazards associated with the use of benzodiazepines
in detoxification. J Subst Abuse Treat 10:449, 1993
40. HislopWS, Bouchier IAD, Allan JG, et al: Alcoholic liver disease in
Scotland and North England: Presenting features in 510 patients. QJM
52:232, 1983
41. Irving
MG, Halliday JW, Powell LW: Association between alcoholism
and increased hepatic iron stores. Alcohol Clin Exp Res 12:7, 1988
42. Ishak
KG, Zimmermann HJ, Mukunda BR: Alcoholic liver disease:
Pathologic, pathogenetic and clinical aspects: Alcohol Clin Exp Res 15:45,
1991
43. Iturriaga
H, Pereda T, Estevez A, et al: Serum immunoglobulin A
changes in alcoholic patients. Annals of Clinical Research 9:39, 1977
44. JacobovitsAW, Morgan MY, Sherlock S: Hepatic siderosis in

alcoholics. Dig Dis Sci 24:305, 1979
45. Johnston SC, Pelletier LL: Enhanced hepatotoxicity of acetaminophen

in the alcoholic patient: Two case reports and a review of the literature.
Med Balt 76:185, 1997
46. KoffRS, Dienstag JL: Extrahepatic manifestations of hepatitis C and

the association with alcoholic liver disease. Semin Liver Dis 15:101, 1995
47. LeeRG: Nonalcoholic steatohepatitis: Tightening the morphological

screws on a hepatic rambler. Hepatology 21:1742, 1995
48. Leo MA, Lieber CS: Hepatic vitamin A depletion in alcoholic liver
injury. N Engl J Med 307:597, 1982
49. Leo
MA, Lieber CS: Hypervitaminosis A: A liver lover's lament.
50. LeSage GD, Baldus WP, Fairbanks VF, et al: Haemochromatosis:

Genetic or alcohol induced? Gastroenterology 84:1471, 1983
51. Levin
DL, Baker AL, Riddel RH, et al: Non alcoholic liver disease.
Overlooked causes of liver injury in patients with heavy alcohol
consumption. Am J Med 66:429, 1979
52. Lieber
CS: Interaction of alcohol with other drugs and nutrients. Drugs
40(suppl 3):23, 1990
53. Lieber
CS: Alcohol and the liver: 1994 update. Gastroenterology
106:1085, 1994
54. LieberCS: Susceptibility to alcohol-related liver injury. Alcohol

2(suppl):315, 1994

55. Lieber
CS, Jones DP, DeCarli LM: Effects of prolonged ethanol intake:
Production of fatty liver despite adequate diet. J Clin Invest 4:1009, 1965
56. Lohr
RH: Treatment of alcohol withdrawal in hospitalized patients.
Mayo Clin Proc 70:777, 1995
57. Manikant S, Tripathi BM, Chavan BS: Loading Dose diazepam therapy
for alcohol withdrawal state. Indian J Med Res 98:170, 1993
58. Marks V: Clinical pathology of alcohol. J Clin Pathol 36:365, 1983
59. Masani F, Kato H, Sasagawa Y: An echocardiographic study of

alcoholic cardiomyopathy after total abstinence. J Cardiol 20:627, 1990
778
60. Matsuzaki Y, Tanaka N, Osuga T: Is taurine effective for treatment of

painful muscle cramps in liver cirrhosis. Am J Gastroenterol 88:1466,
1993
61. Mayo-Smith M, Bernard D: Late onset seizures in alcohol withdrawal.

Alcohol Clin Exp Res 19:656, 1995
62. Mayo-Smith MF: Pharmacological management of alcohol withdrawal:

A meta-analysis and evidence based practice guideline. JAMA 278:144,
1997
63. McHutchison JG, Leai RL, Govindarajan S, et al: Hepatitis C

antibodies in patients with alcoholic liver disease commonly have an
identifiable risk factor. J Clin Gastroenterol 15:233-235, 1992
64. Melander O, Liden A, Melander A: Pharmacokinetic interactions of

alcohol and acetylsalicylic acid. Eur J Clin Pharmacol 48:151, 1995
65. MendenhallCL, Moritz T, Rouster S, et al: Epidemiology of hepatitis C

among veterans with alcoholic liver disease. Am J Gastrotenterol 88:1022,
1993
66. Mendenhall CL, Seeff L, Diehl AM, et al: Antibodies to hepatitis B

virus and hepatitis C virus in alcoholic hepatitis and cirrhosis: Their
prevalence and clinical relevance. Hepatology 14:581, 1991
67. Menon DK, Harris M, Sargentoni J, et al: In vivo hepatic 31P magnetic
resonance spectroscopy in chronic alcohol abusers. Gastroenterology
108:776, 1995
68. Menon DK, Sargentoni J, Taylor-Robinson SD, et al: Effect of

functional grade and etiology on in vivo hepatic phosphorus-31 magnetic
resonance spectroscopy in cirrhosis: Biochemical basis of spectral
appearances. Hepatology 21:417, 1995

69. Morgan MY, Sherlock S, Scheuer PJ: Acute cholestasis, hepatic failure
and fatty liver in the alcoholic. Scand J Gastroenterol 13:299, 1978
70. Muuronen A, Bergman H, Hindmarsh T, et al: Influence of improved

drinking habits on brain atrophy and cognitive performance in alcoholic
patients: A 5-year follow-up study. Alcohol Clin Exp Res 13:137, 1989
71. NalpasB, Thiers V, Pol S, et al: Hepatitis C viremia and anti-HCV

antibodies in alcoholics. J Hepatol 14:381, 1992
72. Nalpas B, Vassault A, Charpin S, et al: Serum mitochondrial aspartate

aminotransferase as a marker of chronic alcoholism: Diagnostic value and
interpretation in a liver unit. Hepatology 6:608, 1986
73. NalpasB, Vassault A, Le Guillou A, et al: Serum activity of

mitochondrial aspartate aminotransferase: A sensitive marker of
alcoholism with or without alcoholic hepatitis. Hepatology 4:893, 1984
74. Neuschwander-Tetri
BA, Bacon BR: Nonalcoholic steatohepatitis. Med
Clin North Am 80:1147, 1996
75. Nishiguchi S, Kuroki T, Yabusako T, et al: Detection of hepatitis C

virus antibodies and hepatitis C virus RNA in patients with alcoholic liver
disease. Hepatology 14:985, 1991
76. Noguchi O, Yamaoka K, Ikeda T, et al: Clinicopathological analysis of

alcoholic liver disease complicating chronic type C hepatitis. Liver 11:225,
1991
77. Ostertun
B, Dewes W, Hanisch E, et al: MR tomography and computer
tomography of alcohol-induced brain tissue changes. Rofo Fortschr Geb
Rontgenstr Neuen Bildgeb Verfahr 152:87, 1990
78. ParesA, Barera J, Caballeria J, et al: Hepatitis C virus antibodies in

chronic alcoholic patients: Association with severity of liver injury.
79. PiccininoF, Sagnelli E, Pasquale G, et al: Complications following

percutaneous liver biopsy: A multicenter retrospective study on 68,276
biopsies. J Hepatol 2:165, 1986
80. PoynardT, Bedossa P, Opolon P: Natural history of liver fibrosis

progression in patients with chronic hepatitis C. Lancet 349:825, 1997
81. Radouco-Thomas S, Garcin F, Guay D, et al: Double-blind study on the

efficacy and safety of tetrabamate and chlordiazepoxide in the treatment of
the acute alcohol withdrawal syndrome. Prog Neuropsychopharmacol Biol
Psychiatry 13:55, 1989
82. Robinson BJ, Robinson GM, Maling TJB, et al: Is clonidine useful in
the treatment of alcohol withdrawal? Alcohol Clin Exp Res 13:95, 1989

83. RosalkiSB, Tarlow D, Rau D: Plasma gamma-glutamyl transpeptidase

elevation in patients receiving enzyme-inducing drugs. Lancet 2:376, 1971
779
84. Rubin
E, Urbano-Marquez A: Alcoholic cardiomyopathy. Alcohol Clin
Exp Res 18:111, 1994
85. SaitzR, Mayo-Smith MF, Roberts MS, et al: Individualized treatment

for alcohol withdrawal state: A randomized, double-blind controlled trial.
JAMA 272:519, 1994
86. SallieRW, Reed WD, Shilkin KB: Confirmation of the efficacy of

hepatic tissue iron index in differentiating genetic haemochromatsis from
alcoholic liver disease complicated by alcoholic haemosiderosis. Gut
32:207, 1991
87. Schiff ER: Hepatitis C and alcohol. Hepatology 26(suppl 1):39S, 1997
88. Sheth
SG, Gordon FD, Chopra S: Nonalcoholic steatohepatitis. Ann
Intern Med 126:137, 1997
89. Spahr L, Butterworth RF, Fontaine S, et al: Increased blood manganese

in cirrhotic patients: Relationship to pallidal magnetic resonance signal
hyperintensity and neurological symptoms. Hepatology 24:1116, 1996
90. Steinberg
JD, Hayden MT: Prevalence of clinically occult
cardiomyopathy in chronic alcoholism. Am Heart J 101:461, 1981
91. SullivanLW, Herbert V: Suppression of hematopoiesis by ethanol. J

Clin Invest 43:2048, 1964
92. TallyNJ, Roth A, Woods J, et al: Diagnostic value of liver biopsy in

alcoholic liver disease. J Clin Gastroenterol 10:647, 1988
93. ThomsonAD: There may yet be time to save your brain [editorial]. Br J
Alcohol Alcohol 15:89, 1980
94. Uchimura Y, Sata M, Kage-M, et al: A histopathological study of

alcoholics with chronic HCV infection: Comparison with chronic hepatitis
C and alcoholic liver disease. Liver 15:300, 1995
95. Urbano-Marquez A, Estruch R, Fernandez-Sola J, et al: The greater risk

of alcoholic cardiomyopathy and myopathy in women compared with
men. JAMA 274:149, 1995
96. VanEpps E, Husby G, Williams RC, et al: Liver disease-a prominent

cause of IgE elevation. Clin Exp Immunol 23:444, 1976
97. Van Ness MM, Diehl AM: Is liver biopsy useful in the evaluation of

patients with chronically elevated liver enzymes? Ann Intern Med

111:473, 1989
98. Wernicke C: Die akute haemorrhagische Polioencephalitis superior. In

Wernicke C (ed): Lehrbuch der Gehirnkrankheiten fuer Aerzte und
Studierende. Bdll Kassel, Fischer, 1981, p 229
99. YersinB, Nicolet JF, Dercy H, et al: Screening for excessive alcohol
drinking: Comparative value of carbohydrate deficient transferrin, gamma
glutamyl transferase and mean corpuscular volume. Arch Intern Med
155:1907, 1995
100. ZieveL: Jaundice, hyperlipidemia and hemolytic anaemia: A

heretofore unrecognized syndrome associated with alcoholic fatty liver
and cirrhosis. Ann Intern Med 48:471, 1958
101. ZilmDH, Jacob MS, MacLeod SM, et al: Propranolol and

chlordiazepoxide effect on cardiac arrhythmias during alcohol withdrawal.
102. Zimmermann HJ, Maddrey WC: Acetaminophen (Paracetamol)

hepatotoxicity with regular intake of alcohol: Analysis of instances of
therapeutic misadventure. Hepatology 22:767, 1995
103. Zubaran
C, Fernandes JG, Rodnight R: Wernicke-Korsakoff
syndrome. Postgrad Med J 73:27, 1997



765
CLINICAL ISSUES IN THE MANAGEMENT OF ALCOHOLIC LIVER

DISEASE
Petra E. Steindl MD
Peter Ferenci MD
Department of Internal Medicine IV, Division of Gastroenterology and Hepatology, University of Vienna, Vienna, Austria
Petra E. Steindl, MD
Department of Internal Medicine IV
Division of Gastroenterology and Hepatology
University of Vienna
Waehringer Guertel 18-20
1090 Vienna, Austria
The syndrome of alcoholic liver disease (ALD) includes a huge variety of clinical manifestations
that are the result of three factors--hepatocellular insufficiency, portal hypertension, and
extrahepatic damage by alcohol. Overt hepatic disease (hepatomegaly, jaundice, ascites, and portal
hypertension) may be accompanied by digestive, endocrine, hematologic, muscular, neural, and
other extrahepatic manifestations.
In this article, some of the more important issues in clinical routine are addressed, as are some of
the controversies concerning establishment of the diagnosis.


Because of its wide spectrum of clinical presentations, alcoholic liver disease (ALD) poses great
difficulties for diagnosis. The diagnostic challenges include the detection of alcohol abuse in
asymptomatic subjects and the differentiation of ALD from other liver diseases or from the laboratory
changes induced by ethanol.
Chronic alcohol abuse is associated with three distinct histologic lesions in the liver--alcoholic steatosis,
alcoholic hepatitis, and alcoholic cirrhosis. The lesions may overlap and are not associated with distinct
clinical symptoms and signs. The whole spectrum of alcoholic liver disease, from asymptomatic
hepatomegaly to profound hepatocellular failure with portal hypertension, therefore can be observed in
relation to any of the histologic subtypes. It also is
766
important to recognize that the degree of clinical illness associated with the development of alcoholic
liver disease can vary enormously. No clinical feature distinguishes alcoholic liver disease from other
liver diseases. Even in patients with established ALD, the frequency and severity of particular symptoms
vary. In general, women seem to suffer more florid symptoms of liver disease. Stigmata of liver disease,
such as spider teleangiectasia, gynecomastia, and hepatosplenomegaly, or signs of liver failure, such as
jaundice, ascites, and encephalopathy, occur more frequently in older patients and in those with cirrhosis.
[40]
There is no single reliable test that allows a precise diagnosis of ALD. Ethyl alcohol (ETOH)
consumption per se may result in changes of routine laboratory tests independent from liver disease.
1. Metabolic changes, such as hypertriglyceridemia, hypercholesterolemia, or hyperuricemia, are
common in alcoholics but certainly occur in a large number of nonalcoholic subjects. Even in
alcoholics, these findings are not necessarily the consequence of ethanol consumption but of
concurrent overnutrition. Lipid abnormalities lead to hepatic fat accumulation that can be easily
detected by sonography. Hepatic steatosis of any cause may be associated with increases in alanine
aminotransferase (ALT) and other enzymes. Increased liver enzymes in someone consuming
ethanol, therefore, are not reliable evidence for ethanol abuse.
2. Hepatic ETOH degradation is a powerful inducer of mitochondrial oxidation and is associated with
biochemical and morphologic changes. Increased production and release of the mitochondrial
isoenzyme of AST (mAST) and of gamma-glutamyltranspeptidase (GGTP), [83] therefore, reflect
long-term alcohol abuse. Increased serum AST or GGTP can occur in any liver disease, however,
and, therefore, are nonspecific. They can occur in the absence of liver disease if the patient is on
medication with drugs that are inducing hepatic oxidation (e.g., barbiturates).
3. Pancreatic disease frequently is seen in alcoholics. Stenosis of the common bile duct by pancreatic
disease results in cholestasis and an increase of alcaline phosphatose (AP) and GGTP. Abnormal
GGTP, therefore, is not a sensitive indicator of ALD. The differentiation of chronic pancreatitis
with cholestasis from chronic pancreatitis associated with ALD is a diagnostic challenge but has
important implications for the management of the patient.

4. Drug metabolism is altered in chronic alcohol abusers (discussed later), increasing the potential
hepatotoxicity of drugs such as paracetamol or isoniazide. Increases in transaminase levels,
therefore, are multifactorial and, certainly, nonspecific.
Diagnosis of Alcohol Abuse and Alcoholic Liver Disease in Asymptomatic Subjects
The prevalence of asymptomatic ALD is unknown. The frequency in the general population cannot be
calculated from data obtained in hospitalized patients or mortality registers. Based on questionnaires,
10% of the adult population of two towns in northern Italy were considered alcohol abusers. [11]
Asymptomatic chronic liver disease was common (one out of six "healthy" adults), and ETOH was
considered as a causative factor in 23% of cases. The same authors [10] found that the threshold levels for
developing cirrhosis was 30 g/day of ethanol. The risk increased in parallel with greater daily alcohol
intake. Drinking alcohol outside of mealtimes and drinking multiple different alcoholic beverages further
augmented the risk of alcoholic liver damage. A careful history of the amount
767
and duration of alcohol intake, the type of drinks consumed, and the drinking pattern, therefore, is needed
to relate alcohol consumption with ALD.
Several routinely available laboratory tests are useful in the detection of excessive drinking but are not
always reliable indicators of liver disease. Elevated GGTP activity is common in individuals abusing
alcohol, [24] especially those drinking more than 80 g of alcohol per day. [9] [58] [99]
Laboratory Parameters in the Diagnosis of ALD
Gamma-glutamyltransferase (GGT)
Macrocytosis (MCV )
AST > ALT; AST/ALT > 2
Mitochondrial alanine aspartate aminotransferase (mAST)/total alanine aspartate aminotransferase

(tAST)
CDT
Gamma-globulins (Ig A )
Uric acid
Lactate

Serum albumin
Prothrombin time
Triglycerides
The usefulness of GGTP to detect alcohol abuse, however, is limited by its low specificity and
sensitivity. Up to 70% of alcohol abusers may have normal values. On the other hand, obesity, treatment
with certain drugs, and hepatobiliary diseases are associated with increased GGTP irrespective of the
extent of alcohol consumption.
Serum transaminase levels are not markedly increased in ALD. Serum AST levels are only raised two to
six times in severe acute alcoholic hepatitis. Levels of AST greater than 500 U/L point to another cause.
The level of AST is higher than that of ALT and a ratio of AST to ALT greater than two is present in
about 70% of cases. [21] The mitochondrial isoenzyme of AST (mAST) is a sensitive marker of chronic
alcoholism. [73] The ratio of mAST to total AST was significantly increased in alcoholics but was not
related to the degree of liver disease. [72] This indicates the mAST-to-tAST ratio is a marker of chronic
alcohol abuse rather than of ALD.
Macrocytosis (increased mean cell volume [MCV]) is common in individuals abusing alcohol. It lacks
sensitivity (27%-52%) but is reasonably specific for alcohol use greater than 50 g/d. [9] [99] It probably
reflects the toxic effect of alcohol on the developing erythrocyte. [91] The combination of raised GGTP
activity and MCV has a sensitivity for detection of alcohol of 30% to 40%.
Serum carbohydrate-deficient transferrin (CDT) is a specific and sensitive test of alcoholism, irrespective
of complicating liver disease, but is not readily available. [8] In patients with alcohol intake greater than
10 g/d there is a positive correlation between CDT and alcohol intake. [9] This topic is covered in more
depth in the article by Marchesini et al in this issue.
Frequent determination of blood ETOH at every outpatient laboratory control is helpful in determining
patients' compliance with abstinence.
Diagnosis of Alcoholic Liver Disease in Symptomatic Patients
There are routinely available laboratory parameters that may provide some clues that alcohol is the cause
of liver disease. Characteristic but nonspecific findings include elevations in uric acid, lactate, and
triglycerides, and reductions
768
in potassium, glucose phosphate, and magnesium. A polyclonal hypergammaglobulinemia is present in

most patients with ALD, with a disproportionate increase in circulating IgA. [43] [96] Serum albumin and
prothrombin time decrease as liver function declines or as the nutritional status of the patient worsens. It
should be emphasized, however, that no laboratory test is specific for ALD.

In a cholestatic variant of alcoholic hepatitis, very high levels of alcaline phosphatase may be present, but
this is also seen in a small proportion of patients with fatty change alone. [6] [69] Zieve's syndrome, a
combination of fatty liver (with or without cirrhosis) with hyperlipidemia and hemolysis, is also a
characteristic but infrequent picture. [100]
Future Trends
A sophisticated tool in the diagnosis of ALD is hepatic phosphorus-31 magnetic resonance spectroscopy,
a method that provides information on hepatic energy metabolism and phospholipid membrane
metabolism. The pattern of peak area ratios of phosphomonoesters (PME), inorganic phosphate, and
phosphodiesters (PDE) relative to betaATP, and of PME relative to PDE changes significantly according
to the cause and the degree of liver disease. [67] [68] In the patients with compensated disease, those with
alcoholic cirrhosis showed a significantly lower PDE/ATP ratio, [68] although, in patients with minimal
liver injury, recent alcohol intake was associated with a significant increase in the PDE/ATP ratio. [67]
Differentiation from Nonalcoholic Steatohepatitis
Nonalcoholic steatohepatitis (NASH) is a form of chronic liver disease that is defined by biopsy findings
and has the appearance of alcoholic hepatitis. NASH is more common in women than in men and it
appears to be associated with obesity, diabetes mellitus, and related metabolic abnormalities, such as
hyperlipidemia and hyperglycemia. The association with female gender, obesity, and diabetes is not as
close as suggested, however, and an underlying condition cannot always be discerned. In a group of 33
patients with NASH, 58% were men, 61% were nonobese, 79% were normoglycemic, and 79% had
normal lipid levels. No metabolic abnormality was present in 42%, and it was, therefore, concluded that
NASH has to be considered in an expanded group of patients with abnormal liver enzymes and a
negative biochemical and serologic work-up. [5] Still, obesity is the condition most frequently associated
with NASH (69%-100% in different studies). [88]
There is no test that differentiates NASH from alcoholic steatohepatitis. The diagnosis and differential
diagnosis of NASH currently is established on morphologic grounds alone, [47] but the final diagnosis can
be made in only the context of the clinical history. [74] The most important feature is the exclusion of
significant alcohol consumption. How much alcohol constitutes significant consumption is unknown. It
has been shown that steatosis develops relatively rapidly in nondrinkers after a controlled period of
alcohol consumption. [10] [55]
Several studies have demonstrated a high prevalence of hepatitis C virus (HCV) antibody (anti-HCV) in
patients with ALD. HCV infection has an important
769
role in the progression of ALD. [19] [36] [63] [78] Concomitant HCV infection is associated with increased
severity of liver injury in patients with ALD. [66] [71] [75] [76] In serologic surveys, [46] the prevalence of
HCV was increased in patients with ALD. Most patients with ALD with concomitant HCV infection
have identifiable parenteral risk factors. [63] Patients with HCV infection have more severe histologic
features, [65] [87] a decreased survival rate, [66] and an earlier development of disease [19] than those without

HCV infection. Patients with active HCV viremia have histologic features of both chronic hepatitis [94]
and ALD. The histologic features of anti-HCV-reactive patients without viremia resemble those of
patients without serologic evidence of HCV infection. [30]
The natural history of fibrosis progression in patients with HCV is associated with three independent
factors--age at infection older than 40 years, daily alcohol consumption of 50 g or more, and male
gender. [80] The mean stage of fibrosis at biopsy was significantly higher in patients with daily alcohol
consumption greater than 50 g than in those who consumed less, irrespective of age or duration of
infection. The findings confirm that alcohol is a potential risk factor in patients with HCV and that the
combination of HCV and alcohol is likely to result in a severer course of liver disease.
Iron Overload
The consumption of excessive amounts of alcohol affects human iron homeostasis, and iron overload and
heavy alcohol consumption frequently are associated. [29] [41] Lipid peroxidation induced by ethanol or
iron plays a major role in hepatic damage, both in humans and in experimental animals. Although the
exact mechanism(s) of induction of lipid peroxidation by ethanol and iron remains to be elucidated, both
toxins can exert a synergistic effect upon hepatic lipid peroxidation. Both major proteins of iron
metabolism, ferritin and transferrin, as well as transferrin saturation are increased by alcohol. Hepatic
iron levels are increased in a high proportion of alcoholic subjects, [86] sometimes causing confusion in
diagnosis between alcoholic liver disease and iron-overload disease.
The pattern of hepatic iron deposition in alcoholics, however, differs from that in hemochromatosis. The
relationship between alcoholism and genetic hemochromatosis has been an issue of intensive research for
many decades. Hepatic siderosis is present in 57% of patients with alcoholic liver disease, with
pronounced siderosis in 7% of cases. [44] The hepatic iron concentration in alcoholics is lower than in
genetic hemochromatosis. [12] [50] Histologic tissue iron grading is not reliable in distinguishing between
alcoholic siderosis and genetic hemochromatosis. There is a poor correlation between histologic evidence
of iron overload and tissue iron concentrations. [13] [27] In a study of homozygous siblings with genetic
hemochromatosis, [2] the patients with heavy alcohol consumption had a higher prevalence of cirrhosis
than hemochromatosis patients without heavy alcohol consumption. Hepatic iron concentration and
hepatic iron index did not differ between patients with hemochromatosis with or without ETOH abuse.
Long-term survival was reduced in patients with heavy alcohol consumption (mean follow-up, 9.22
years). This suggests that chronic alcohol consumption in hemochromatosis has an additive hepatotoxic
effect despite the paucity of histologic features of alcoholic liver disease.
Factors that might be responsible for iron overload in the alcoholic include: increased intake and
absorption of iron; altered iron use caused by impaired erythropoiesis (folate deficiency, myelotoxicity of
alcohol); repeated bursts of
770
hemolysis (Zieve's syndrome); and iron deposition as a result of liver damage. [42] Portosystemic shunting
also increases hepatic iron deposition, but the effect is variable. [22] Determination of the HFE-gene (the
abnormal gene in the majority of patients with hemochromatosis [26] ) mutation status may be useful to
differentiate the various reasons of iron overload in alcoholics. [4]

NEED FOR LIVER BIOPSY

Liver biopsy plays an important role in the assessment of the alcoholic patient with abnormal liver
function tests. First, it may be used to establish a diagnosis of alcoholic liver disease. Up to 20% of such
patients will have nonalcoholic liver disease. [51] Secondly, it is essential in determining the stage of
disease because clinical and biochemical indices are poor predictors of the extent and severity of liver
injury. Finally, liver biopsy may help determine the prognosis.
The limitations of laboratory tests in confirming alcoholic liver disease suggest that biopsy should be
undertaken in the majority of suspected cases, particularly if a period of abstinence has not been
accompanied by normalization of transaminases over a period of 3 to 6 months.
Furthermore, the diagnostic value of histologic examination has been emphasized by several studies. [51]
[92] [97] The reversibility of changes can be predicted, based on an assessment of the extent of the hepatitis
and the amount of pericellular and perivenular fibrosis, as well as from the degree of architectural
distortion and features of hepatitis, including Mallory bodies, which disappear on alcohol withdrawal.
Liver biopsy is relatively safe, with an associated morbidity of 0.1% to 0.6% and mortality of 0.01% to
0.03%. [31] [33] [79] There is also relatively small interobserver variation in the histologic interpretation of
ALD, with greater agreement in samples containing more than six portal tracts. [7]
The biopsy should not be undertaken as a blind procedure unless the prothrombin time is within 3
seconds of control time and the platelet count exceeds 80,000. If clotting abnormalities persist after
adequate substitution (vitamin K, clotting factors) or there are concerns about the safety of a blind needle
procedure, then the biopsy can be undertaken under radiologic guidance or can be obtained via the
transjugular route or at peritonoscopy. The specimen obtained using a Menghini needle usually is
adequate.
DRUG INTERACTION OF ALCOHOL WITH SPECIAL FOCUS ON

TYLENOL
It took 20 years of research in ethanol metabolism to culminate in the molecular elucidation of an
alcohol-inducible cytochrome P450 (P450IIE1) that is involved both with ethanol metabolism and
ethanol tolerance, and with the activation of a number of xenobiotics. The unique ability of P450IIE1 to
activate xenobiotic agents now appears to be responsible for the increased susceptibility of the heavy
drinker to hepatotoxic industrial solvents, commonly used drugs, and over-the-counter medications.
Especially when a substance has hepatotoxic metabolites--for example, CCI4--the increased activity of
the hepatic microsomal enzymes in the alcoholic patient leads to an increased conversion of CCI4 to its
toxic metabolites.
The activation of the microsomal enzymes also explains some of the interaction
771

of ethanol with nutritional factors, such as hepatic vitamin A: Enhanced microsomal degradation of
retinoids (together with hepatic mobilization) promotes depletion. [48] Treatment, however, is
complicated by the fact that ethanol also enhances the toxicity of excess vitamin A. [49] The list of drugs
interacting with alcohol through the activated P450IIE1 system is very long. For review, references 53
and 54 are recommended to the interested reader.
The enhanced hepatotoxicity of acetaminophen (paracetamol) in alcoholics is also caused by increased
activity of the cytochrome P450 system [54] and decreased hepatic glutathione. Under normal conditions,
glutathione conjugates with the toxic metabolite to render it nontoxic, while most of the drug is
conjugated with glucuronic acid or sulfate. After acetaminophen overdose or when glutathione stores are
decreased (induced by alcohol or malnutrition) the protective mechanisms fail, resulting in severe
toxicity of the active metabolite. Clinical studies have shown that the elimination half-life of
acetaminophen is significantly shorter in alcoholics than in healthy controls, suggesting that the increased
metabolism of the drug leads to enhanced toxicity. [32] [64]
In a report on 67 patients who developed hepatic injury after ingestion of acetaminophen, 64% were
considered to be alcoholics. Mortality of this condition was up to 18% in this study, but 32% in the
literature. [45] It is important to note that the majority of patients reported taking doses of the drug that
were well below the accepted toxic range. [102] The syndrome of liver injury in these patients is relatively
distinctive. The most important clinical or biochemical characteristic is the high--often towering--level of
AST. In the aforementioned study, levels exceeded 3000 IU in more than 90% of patients and the highest
levels were at 48,000 IU, so that this marker is almost pathognomonic. Another laboratory hint to the
diagnosis may be an extremely long prothrombin time. Histologically, the typical lesion is zone 3
necrosis.
The important issue regarding acetaminophen toxicity is the physician's awareness of the diagnosis and
the possibility that severe liver injury may occur even at therapeutic doses of this easily available drug.
[25] In an alcoholic who presents with jaundice and extremely high AST levels, therefore, a careful history
of drug use regarding paracetamol is mandatory.
MANAGEMENT OF ALCOHOL WITHDRAWAL

The clinical symptoms of alcohol withdrawal can be classified (Table 1) as early (24-48 hours) and late
(> 48 hours) as well as minor and major. [56] The level of autonomic hyperactivity and the presence of
delirium are the main
TABLE 1 -- SYMPTOMS OF ALCOHOL WITHDRAWAL
Early (24- 48 Hours)/Minor Late (> 48 Hours)/Major
Temperature Disorientation
Tremor Cognitive impairment
Tachycardia Systolic hypertension
Insomnia Tachycardia

Restlessness Nausea
Seizures Vomiting
Agitation
Hallucinations
772
determinants of progression from minor to major symptoms. The "3 Ts" (temperature, tremor, and
tachycardia) of the delirium tremens may be the earliest signs of alcohol withdrawal, although the
definition true delirium tremens necessitates the presence of cognitive impairment. Insomnia,
restlessness, agitation, nausea and vomiting, myalgias, tremor, systolic hypertension, and tachycardia can
progress to disorientation and cognitive impairment. If alcohol withdrawal-related seizures occur, they
usually occur early during the course of withdrawal. Hallucinosis also can occur early, especially in
patients who have consumed alcohol for a prolonged period.
The goals of treatment are amelioration of symptoms and prevention of complications. According to the
guidelines of the American Society of Addiction Medicine, [53] benzodiazepines are the first-line
treatment of acute alcohol withdrawal and they can reduce the risk of seizures and delirium significantly.
Although the different benzodiazepines seem to be similarly efficacious in reducing signs and symptoms
of withdrawal, there is some evidence that the longer-acting agents may be more effective in preventing
seizures. [39] [61] [81] Clinical experience suggests that the longer-acting agents can pose a risk of excess
sedation in selected patient groups, however, especially the elderly and those with poor liver function.
Another consideration in the choice of benzodiazepine is their potential for abuse, which is greater in
agents with rapid onset of action (diazepam, alprazolam, lorazepam) than in those with slower onset of
action (chlordiazepoxide, oxazepam, halazepam). This consideration may be relevant in an outpatient
setting or for patients with a history of benzodiazepine or other substance abuse.
There are two approaches to the dosage of sedative drugs. In most studies, medications were given in
fixed amounts at scheduled times (e.g., chlordiazepoxide, 50 mg every 6 hours for 5 days). An alternative
to giving medication on a fixed schedule, known as symptom-triggered therapy, has been developed. In
this approach, the patient is monitored by means of a structured assessment scale and given medication
only when symptoms cross a threshold of severity. This has been shown to be as effective as the
fixed-dose regimen, with the advantage of significantly less medication and significantly shorter duration
of treatment. [57] [85] In the hospital setting, however, prevention of alcohol withdrawal symptoms by
giving meprobamate (another substance commonly used in Europe), for example, to any patient admitted
with a history of recent heavy alcohol consumption is practiced successfully.
Other sedative-hypnotic drugs used in alcohol withdrawal are chlormethiazole (mainly in Europe), [14]
barbital, and tetrabamate. They are equally effective as benzodiazepines.
Several other alternatives to benzodiazepines can be used adjunctively or, occasionally, independently.
beta-adrenergic blockers are useful for controlling blood pressure and tachyarrhythmias, but the usual
contraindications must be considered. In addition, delirium is a known side effect of beta-blockers and, in
one study, the incidence of delirium was even increased with propranolol. [101] The centrally acting

alpha-adrenergic agonist clonidine has been shown to be effective in ameliorating symptoms of

withdrawal (especially in treating associated hypertension), but it should only be used in combination
with benzodiazepines. [82] Carbamazepine and valproic acid have been widely used in Europe as
anticonvulsants, being effective in increasing the seizure threshold, but experience in the United States is
limited. Neuroleptic agents, including the phenothiazines and the butyrophenone haloperidol,
demonstrate some effectiveness in reducing signs and symptoms of withdrawal, especially in agitated
patients, but increase
773
the incidence of seizures compared with placebo. As a result, they have to be used with caution.
Magnesium can be used to increase the seizure threshold if the patient's magnesium level is low. If,
however, the initial magnesium level is normal, magnesium supplementation probably has little
associated benefit.
MUSCLE CRAMPS
Muscle cramps are a common symptom in cirrhosis and are more prevalent in this disorder than in
patients with chronic hepatitis or congestive heart failure. The incidence of cramps is unrelated to the
amount of diuretics used, the cause of liver disease (especially ALD), abnormalities in serum
electrolytes, or differences in Child's classification. [1] This suggests that cramps in these patients are
related specifically to the development of cirrhosis. In Abrams' study, 36 of 92 patients had alcoholic
cirrhosis, but the incidence of cramps in the group was not higher than in liver cirrhosis of other cause.
ALD, therefore, is not an independent risk factor for the development of cramps. Similarly, in a study of
294 patients [3] the authors found a higher prevalence of cramps in cirrhotic patients than in controls and
it was related to the duration of recognized cirrhosis and the severity of liver function impairment. Their
data suggested that a reduced effective plasma volume may be involved in the pathophysiology of
cramps because there were significant differences in mean arterial blood pressure (MAP), plasma renin
activity (PRA), urinary sodium excretion, and plasma volume in cirrhotic patients with and without
cramps. At a multiple-regression analysis, the presence of ascites, low values of MAP, and high values of
PRA were the independent predictive factors for the occurrence of cramps in cirrhosis.
The effects of a sustained expansion of the effective circulating volume induced by intravenous infusion
of human albumin were compared with those of a placebo in 12 cirrhotic patients with more than three
cramp crises a week. Compared with the placebo, albumin reduced the cramp frequency ( P < 0.01).
Albumin therefore could be beneficial in the treatment of cramps by improving effective circulating
volume. Furthermore, taurine has been proposed as an effective agent to treat muscle cramps in cirrhotic
patients [60] but, to date, there is no generally accepted and effective treatment of cramps, so medication is
given empirically in most cases.

OTHER ORGAN INJURY

Heart
Chronic alcoholism is one of the most important causes of dilated cardiomyopathy, and a large
proportion of chronic alcoholics demonstrate impairment of cardiac function. [84] The development of
cardiac dysfunction is apparently related to the total lifetime dose of ethanol. Studies in experimental
animals have demonstrated that both acute and chronic ethanol administration impair cardiac
contractility. [20] The relationship, if any, however, between the acute effects of alcohol and the
development of irreversible cardiomyopathy remain to be elucidated.
The clinical onset of alcoholic cardiotoxicity is insidious, with nonspecific fatigue, chest discomfort,
palpitations, or an isolated episode of atrial fibrillation. [90]
774
As the cardiomyopathy progresses, manifestations of overt right and left heart failure become apparent.
Treatment of alcoholic cardiomyopathy is similar to that of other dilated cardiomyopathies. The
prognosis is variable, depending on the duration and severity of the cardiac abnormalities. Patients with
severe symptoms may progress to congestive heart failure and die even if they stop drinking. On the
other hand, left ventricular function improved dramatically in 14 patients with severe end-stage
congestive heart failure attributable to alcoholic cardiomyopathy after 6 months of total abstinence. [35] In
another study, [59] both left ventricular end-diastolic dimension and ejection fraction improved rapidly
after total abstinence from alcohol intake.
Several studies indicate that women are more susceptible to the cardiotoxic effects of alcohol. [28] [95] The
prevalence of dilated cardiomyopathy in alcoholic women was similar to alcoholic men, but total lifetime
dose and threshold dose of ethanol for the development of cardiomyopathy were considerably lower in
women than in men. The decline in ejection fraction with increasing alcohol dose was significantly
steeper.
Pancreatic Function in Alcoholic Liver Disease
There is a great deal of evidence regarding the close relationship between alcohol consumption and
chronic pancreatitis. Alcohol usually causes a chronic, recurrent, calcifying pancreatitis and a period of 6
to 12 years of alcohol consumption seems to be necessary before pancreatic symptoms occur. [17] There
also is evidence, however, that the presence of ALD has an impact on pancreatic function. To cover this
topic in depth would fill another article, but some of the data are given here. Some studies [23] [37] have
investigated the presence of changes in pancreatic function in alcoholic patients, with and without ALD,
to detect functional alterations and possible association of hepatic and pancreatic disease. By comparing
cirrhotic and noncirrhotic patients, they found that the volume of duodenal juice, bicarbonate output, and
amylase output were greater in cirrhotic than in noncirrhotic alcoholic subjects, but not in those of
nonalcoholic status. Maximal bicarbonate concentration was decreased in ALC. [37] The authors
concluded that, in ALD, exocrine pancreatic secretion tends to increase with severity of liver damage, but
concurrence of definite chronic pancreatitis is not correlated with the severity.

In another study, [23] very similar data were obtained. There was also a tendency to larger volume and
lower bicarbonate concentration when the hepatic lesion was more severe. Bicarbonate output was
significantly higher in patients with alcoholic cirrhosis. Interestingly, none of the patients had changes
consistent with chronic pancreatitis.
In conclusion, the alterations in pancreatic function parallel the severity of the liver disease, but the
occurrence of alcoholic pancreatitis seems to be attributable to direct toxic effects of alcohol, (although
the mechanisms are far from clear) and unrelated to ALD. Still, the association of ALD with
alcohol-induced chronic pancreatitis is very common. [34]
Central Nervous System
Chronic alcoholism results in brain damage and dysfunction, leading to a constellation of

neuropsychiatric symptoms, including cognitive dysfunction, the
775
Wernicke-Korsakoff syndrome, [98] alcoholic cerebellar degeneration, and alcoholic dementia. That these
clinically defined entities result from independent pathophysiologic mechanisms is unlikely. Alcohol and
its metabolite acetaldehyde are directly neurotoxic. Alcoholics are thiamin deficient as a result of poor
diet, gastrointestinal disorders, and liver disease. In addition, both alcohol and acetaldehyde have direct
toxic effects on thiamin-related enzymes in the liver and brain. [16]
The relative roles of alcohol toxicity, thiamin deficiency and cirrhosis of the liver in the pathogenesis of
alcohol-related brain damage are unclear. In alcoholics, liver disease per se results in altered thiamin
homeostasis, cognitive dysfunction, and neuropathologic damage to astrocytes. Chronic liver disease
may also interfere with brain thiamin homeostasis and therefore contribute to the pathogenesis of the
Wernicke-Korsakoff syndrome in chronic alcoholism. [15]
The clinical triad described by Wernicke is composed of ophtalmoplegia, ataxia, and disturbance of
consciousness and the mental state. Ocular abnormalities consist of nystagmus (horizontal and vertical)
and weakness or paralysis of the external rectus muscles, with impairment of conjugate gaze,
accompanied by diplopia and strabismus. [103]
The ataxia is of stance and gait and can hinder walking or standing without support in the acute phase.
The disturbances of mental state and consciousness occur mainly as a global confusional state (occurs in
all but 10% of patients), in which the patient is apathetic, inattentive, and with minimal spontaneous
speech. Treatment consists of parenteral thiamin. Most patients who do not recover within 48 to 72 hours
develop Korsakoff's psychosis.
Korsakoff's psychosis is characterized by various degrees of anterograde and retrograde amnesia, with
relative preservation of other intellectual functions. Confabulation is considered a hallmark of
Korsakoff's psychosis. In the early stages of the disease, confusion is profound and confabulation is
evident; in the convalescent phase, the patient recalls fragments of the past in a distorted manner. The
disease is potentially reversible by early treatment with thiamin, but recovery is incomplete in more than
50% of cases.
Other well-recognized syndromes in alcoholic patients include alcoholic dementia, alcohol withdrawal

syndrome, central pontine myelinolysis and alcoholic cerebellar degeneration, [93] which is associated
with cortical atrophy. [38] Brain atrophy is a common finding in chronic alcohol abusers, but the
shrinkage is reversible with abstinence. [70] There is no clear-cut differentiation between the clinical
symptoms of alcohol-induced brain damage and the neuropsychological changes seen in hepatic
encephalopathy. In certain situations, however, especially in the evaluation for liver transplantation, the
differentiation between the two entities can be important. Brain MR imaging may be helpful in the
differential diagnosis because the changes seen in Wernicke's encephalopathy [77] differ considerably
from the well-described changes [89] seen in hepatic encephalopathy (increased signal intensity of the
globus pallidus on T1-weighted images).
References
1. Abrams GA, Concato J, Fallon MB, et al: Muscle cramps in patients with cirrhosis. Am J Gastroenterol 91:1363, 1996
2. Adams PC, Agnew S: Alcoholism in hereditary hemochromatosis revisited: Prevalence and clinical consequences among
homozygous siblings. Hepatology 23:724, 1996
3. AngeliP, Albino G, Carraro P, et al: Cirrhosis and muscle cramps: Evidence of a causal relationship. Hepatology
23:264, 1996
776
4. Bacon BR: Diagnosis and management of hemochromatosis. Gastroenterology 113:995, 1997
5. Bacon BR, Farahvash MJ, Janney CG, et al: Nonalcoholic steatohepatitis: An expanded clinical entity. Gastroenterology
107:1103, 1994
6. Ballard H, Bernstein M, Farrar JT: Fatty liver presenting as obstructive jaundice. Am J Med 30:196, 1961
7. Bedossa P, Poynard T, Naveau S, et al: Observer variation in assessment of liver biopsies of alcoholic patient. Alcohol
Clin Exp Res 12:173, 1988
8. Bell H, Tallaksen G, Sjahei M, et al: Serum carbohydrate-deficient transferrin as a marker for alcohol consumption in
patients with chronic liver disease. Alcohol Clin Exp Res 17:246, 1993
9. Bell H, Tallaksen C, Try K, et al: Carbohydrate deficient transferrin and other markers of high alcohol consumption: A
study of 502 patients admitted consecutively to a medical department. Alcohol Clin Exp Res 18:1103, 1994
10. Bellentani S, Saccocio G, Costa G, et al: Drinking habits as cofactors of risk for alcohol induced liver damage. Gut
41:845, 1997
11. BellentaniS, Tiribelli C, Saccoccio G, et al: Prevalence of chronic liver disease in the general population of northern
Italy: The Dionysos study. Hepatology 20:1442, 1994
12. Bothwell TH, Charlton RW: Hemochromatosis. In Schiff L, Schiff ER (eds): Diseases of the Liver, ed 6. Philadelphia,
JB Lippincott, 1987, p 1001

13. Brissot
P, Bourel M, Herry D, et al: Assessment of liver iron content in 271 patients: A reevaluation of direct and
indirect methods. Gastroenterology 80:557, 1981
14. BurroughsAK, Morgan MY, Sherlock S: Double-blind controlled trial of bromocriptine, chlordiazepoxide and
chlormethiazole for alcohol withdrawal symptoms. Alcohol Alcohol 20:263, 1985
15. ButterworthRF: Cerebral dysfunction in chronic alcoholism: Role of alcoholic liver disease. Alcohol Alcohol
2(suppl):259, 1994
16. Butterworth RF: Pathophysiology of alcoholic brain damage: Synergistic effects of ethanol, thiamine deficiency and
alcoholic liver disease. Metab Brain Dis 10:1, 1995
17. Caballeria
J, Pares A: Extrahepatic manifestations of alcoholism. In McIntyre N, Benhamou JP, Rizetto M, et al (eds):
Oxford Textbook of Clinical Hepatology. Oxford, England, Oxford University Press, 1991, p 856
18. CaldwellS, Jeffers L, Ditomaso A, et al: Antibody to hepatitis C is common among patients with alcoholic liver disease
with and without risk factors. Am J Gastroenterol 86:1219, 1991
19. Caldwell SH, Li X, Rourk RM, et al: Hepatitis C infection by polymerase chain reaction in alcoholics: False-positive
Elisa results and the influence of infection on clinical prognostic score. Am J Gastroenterol 88:1016, 1993
20. CapassoJM, Li P, Guideri G, et al: Left ventricular dysfunction induced by chronic alcohol ingestion in rats. Am J
Physiol 261:H212, 1991
21. Cohen JA, Kaplan MM: The SGOT/SGPT ratio--an indicator of alcoholic liver disease. Dig Dis Sci 24:835, 1979
22. ConnHO: Portocaval anastomosis and hepatic hemosiderin deposition: A prospective controlled investigation.
23. Del-Olmo Martinez ML, Barba-Bermejo M: Changes in pancreatic secretion in alcoholic liver disease. Rev Esp Enferm
Dig 77:197, 1990
24. DragosicsB, Ferenci P, Pesendorfer FX, et al: Gamma-glutamyl-transpeptidase in alcoholic liver disease. Acta Med
Austriaca 3:50, 1976
25. EdwardsR, Oliphant J: Paracetamol toxicity in chronic alcohol abusers--a plea for greater consumer awareness. N Z
Med J 105:174, 1992
26. Feder
JN, Gnirke A, Thomas W, et al: A novel MHC class I-like gene is mutated in patients with hereditary
hemochromatosis. Nat Genet 13:399, 1996
27. Ferenci
P, Datz CH, Kazemi-Shirazi L, et al: HLA-H mutations in patients with chronic hepatitis C [Abstract]. J
Hepatol 28(suppl 1):144, 1998
28. Fernandez-SolaJ, Estruch R, Nicolas JM, et al: Comparison of alcoholic cardiomyopathy in women versus men. Am J
Cardiol 80:481, 1997
29. Fletcher LM: Alcohol and iron: One glass of red or more? J Gastroenterol Hepatol 11:1039, 1996
777

30. FongTL, Kanel GC, Conrad A, et al: Clinical significance of concomitant hepatitis C infection in patients with
alcoholic liver disease. Hepatology 19:554, 1994
31. Garcia-Tsao G, Boyer JL: Outpatient liver biopsy: How safe is it? Ann Intern Med 118:150, 1993
32. Girre
C, Hispard S, Palombo C, et al: Increased metabolism of acetaminophen in chronically alcoholic patients. Alcohol
Clin Exp Res 17:170, 1993
33. GoldbergSJ, Mendenhall CL, Connell AM, et al: "Nonalcoholic" chronic hepatitis in the alcoholic. Gastroenterology
72:598, 1977
34. Gullo L, Casadei R, Campione O, et al: Alcoholic liver disease in alcoholic chronic pancreatitis: A prospective study.
Ital J Gastroenterol 27:69, 1995
35. Guillo P, Mansourati J, Maheu B: Long-term prognosis in patients with alcoholic cardiomyopathy and severe heart
failure after total abstinence. Am J Cardiol 79:1276, 1997
36. Halimi C, Deny P, Gotheil C, et al: Pathogenesis of liver cirrhosis in alcoholic patients histological evidence for
hepatitis C virus responsibility. Liver 11:329, 1991
37. Hayakawa T, Kondo T, Shibata T, et al: Exocrine pancreatic function in chronic liver diseases. Am J Gastroenterol
86:201, 1991
38. Hayes PC: Extrahepatic manifestations of alcoholic liver disease. Baillere's Clinical Gastroenterology 7:683, 1993
39. Hill A, Williams D: Hazards associated with the use of benzodiazepines in detoxification. J Subst Abuse Treat 10:449,
1993
40. HislopWS, Bouchier IAD, Allan JG, et al: Alcoholic liver disease in Scotland and North England: Presenting features
in 510 patients. QJM 52:232, 1983
41. Irving
MG, Halliday JW, Powell LW: Association between alcoholism and increased hepatic iron stores. Alcohol Clin
Exp Res 12:7, 1988
42. Ishak
KG, Zimmermann HJ, Mukunda BR: Alcoholic liver disease: Pathologic, pathogenetic and clinical aspects:
43. Iturriaga
H, Pereda T, Estevez A, et al: Serum immunoglobulin A changes in alcoholic patients. Annals of Clinical
Research 9:39, 1977
44. Jacobovits AW, Morgan MY, Sherlock S: Hepatic siderosis in alcoholics. Dig Dis Sci 24:305, 1979
45. JohnstonSC, Pelletier LL: Enhanced hepatotoxicity of acetaminophen in the alcoholic patient: Two case reports and a
review of the literature. Med Balt 76:185, 1997
46. Koff
RS, Dienstag JL: Extrahepatic manifestations of hepatitis C and the association with alcoholic liver disease. Semin
Liver Dis 15:101, 1995
47. Lee RG: Nonalcoholic steatohepatitis: Tightening the morphological screws on a hepatic rambler. Hepatology 21:1742,

1995
48. Leo MA, Lieber CS: Hepatic vitamin A depletion in alcoholic liver injury. N Engl J Med 307:597, 1982
49. Leo MA, Lieber CS: Hypervitaminosis A: A liver lover's lament. Hepatology 8:412, 1988
50. LeSageGD, Baldus WP, Fairbanks VF, et al: Haemochromatosis: Genetic or alcohol induced? Gastroenterology
84:1471, 1983
51. LevinDL, Baker AL, Riddel RH, et al: Non alcoholic liver disease. Overlooked causes of liver injury in patients with
heavy alcohol consumption. Am J Med 66:429, 1979
52. Lieber CS: Interaction of alcohol with other drugs and nutrients. Drugs 40(suppl 3):23, 1990
53. Lieber CS: Alcohol and the liver: 1994 update. Gastroenterology 106:1085, 1994
54. Lieber CS: Susceptibility to alcohol-related liver injury. Alcohol 2(suppl):315, 1994
55. LieberCS, Jones DP, DeCarli LM: Effects of prolonged ethanol intake: Production of fatty liver despite adequate diet. J
56. Lohr RH: Treatment of alcohol withdrawal in hospitalized patients. Mayo Clin Proc 70:777, 1995
57. ManikantS, Tripathi BM, Chavan BS: Loading Dose diazepam therapy for alcohol withdrawal state. Indian J Med Res
98:170, 1993
58. Marks V: Clinical pathology of alcohol. J Clin Pathol 36:365, 1983
59. MasaniF, Kato H, Sasagawa Y: An echocardiographic study of alcoholic cardiomyopathy after total abstinence. J
Cardiol 20:627, 1990
778
60. MatsuzakiY, Tanaka N, Osuga T: Is taurine effective for treatment of painful muscle cramps in liver cirrhosis. Am J
Gastroenterol 88:1466, 1993
61. Mayo-Smith M, Bernard D: Late onset seizures in alcohol withdrawal. Alcohol Clin Exp Res 19:656, 1995
62. Mayo-SmithMF: Pharmacological management of alcohol withdrawal: A meta-analysis and evidence based practice
guideline. JAMA 278:144, 1997
63. McHutchison
JG, Leai RL, Govindarajan S, et al: Hepatitis C antibodies in patients with alcoholic liver disease
commonly have an identifiable risk factor. J Clin Gastroenterol 15:233-235, 1992
64. Melander
O, Liden A, Melander A: Pharmacokinetic interactions of alcohol and acetylsalicylic acid. Eur J Clin
Pharmacol 48:151, 1995
65. MendenhallCL, Moritz T, Rouster S, et al: Epidemiology of hepatitis C among veterans with alcoholic liver disease.
Am J Gastrotenterol 88:1022, 1993

66. Mendenhall CL, Seeff L, Diehl AM, et al: Antibodies to hepatitis B virus and hepatitis C virus in alcoholic hepatitis and
cirrhosis: Their prevalence and clinical relevance. Hepatology 14:581, 1991
67. Menon DK, Harris M, Sargentoni J, et al: In vivo hepatic 31P magnetic resonance spectroscopy in chronic alcohol
abusers. Gastroenterology 108:776, 1995
68. Menon DK, Sargentoni J, Taylor-Robinson SD, et al: Effect of functional grade and etiology on in vivo hepatic
phosphorus-31 magnetic resonance spectroscopy in cirrhosis: Biochemical basis of spectral appearances. Hepatology
21:417, 1995
69. Morgan MY, Sherlock S, Scheuer PJ: Acute cholestasis, hepatic failure and fatty liver in the alcoholic. Scand J
Gastroenterol 13:299, 1978
70. Muuronen A, Bergman H, Hindmarsh T, et al: Influence of improved drinking habits on brain atrophy and cognitive
performance in alcoholic patients: A 5-year follow-up study. Alcohol Clin Exp Res 13:137, 1989
71. Nalpas B, Thiers V, Pol S, et al: Hepatitis C viremia and anti-HCV antibodies in alcoholics. J Hepatol 14:381, 1992
72. NalpasB, Vassault A, Charpin S, et al: Serum mitochondrial aspartate aminotransferase as a marker of chronic
alcoholism: Diagnostic value and interpretation in a liver unit. Hepatology 6:608, 1986
73. NalpasB, Vassault A, Le Guillou A, et al: Serum activity of mitochondrial aspartate aminotransferase: A sensitive
marker of alcoholism with or without alcoholic hepatitis. Hepatology 4:893, 1984
74. Neuschwander-Tetri BA, Bacon BR: Nonalcoholic steatohepatitis. Med Clin North Am 80:1147, 1996
75. Nishiguchi S, Kuroki T, Yabusako T, et al: Detection of hepatitis C virus antibodies and hepatitis C virus RNA in
patients with alcoholic liver disease. Hepatology 14:985, 1991
76. NoguchiO, Yamaoka K, Ikeda T, et al: Clinicopathological analysis of alcoholic liver disease complicating chronic
type C hepatitis. Liver 11:225, 1991
77. Ostertun
B, Dewes W, Hanisch E, et al: MR tomography and computer tomography of alcohol-induced brain tissue
changes. Rofo Fortschr Geb Rontgenstr Neuen Bildgeb Verfahr 152:87, 1990
78. ParesA, Barera J, Caballeria J, et al: Hepatitis C virus antibodies in chronic alcoholic patients: Association with
severity of liver injury. Hepatology 12:1295, 1990
79. PiccininoF, Sagnelli E, Pasquale G, et al: Complications following percutaneous liver biopsy: A multicenter
retrospective study on 68,276 biopsies. J Hepatol 2:165, 1986
80. PoynardT, Bedossa P, Opolon P: Natural history of liver fibrosis progression in patients with chronic hepatitis C.
Lancet 349:825, 1997
81. Radouco-Thomas S, Garcin F, Guay D, et al: Double-blind study on the efficacy and safety of tetrabamate and
chlordiazepoxide in the treatment of the acute alcohol withdrawal syndrome. Prog Neuropsychopharmacol Biol Psychiatry
13:55, 1989
82. RobinsonBJ, Robinson GM, Maling TJB, et al: Is clonidine useful in the treatment of alcohol withdrawal? Alcohol Clin
Exp Res 13:95, 1989

83. Rosalki
SB, Tarlow D, Rau D: Plasma gamma-glutamyl transpeptidase elevation in patients receiving enzyme-inducing
drugs. Lancet 2:376, 1971
779
84. Rubin E, Urbano-Marquez A: Alcoholic cardiomyopathy. Alcohol Clin Exp Res 18:111, 1994
85. Saitz
R, Mayo-Smith MF, Roberts MS, et al: Individualized treatment for alcohol withdrawal state: A randomized,
double-blind controlled trial. JAMA 272:519, 1994
86. Sallie
RW, Reed WD, Shilkin KB: Confirmation of the efficacy of hepatic tissue iron index in differentiating genetic
haemochromatsis from alcoholic liver disease complicated by alcoholic haemosiderosis. Gut 32:207, 1991
87. Schiff ER: Hepatitis C and alcohol. Hepatology 26(suppl 1):39S, 1997
88. Sheth SG, Gordon FD, Chopra S: Nonalcoholic steatohepatitis. Ann Intern Med 126:137, 1997
89. Spahr
L, Butterworth RF, Fontaine S, et al: Increased blood manganese in cirrhotic patients: Relationship to pallidal
magnetic resonance signal hyperintensity and neurological symptoms. Hepatology 24:1116, 1996
90. Steinberg JD, Hayden MT: Prevalence of clinically occult cardiomyopathy in chronic alcoholism. Am Heart J 101:461,
1981
91. Sullivan LW, Herbert V: Suppression of hematopoiesis by ethanol. J Clin Invest 43:2048, 1964
92. Tally
NJ, Roth A, Woods J, et al: Diagnostic value of liver biopsy in alcoholic liver disease. J Clin Gastroenterol
10:647, 1988
93. Thomson AD: There may yet be time to save your brain [editorial]. Br J Alcohol Alcohol 15:89, 1980
94. Uchimura Y, Sata M, Kage-M, et al: A histopathological study of alcoholics with chronic HCV infection: Comparison
with chronic hepatitis C and alcoholic liver disease. Liver 15:300, 1995
95. Urbano-Marquez
A, Estruch R, Fernandez-Sola J, et al: The greater risk of alcoholic cardiomyopathy and myopathy in
women compared with men. JAMA 274:149, 1995
96. VanEpps E, Husby G, Williams RC, et al: Liver disease-a prominent cause of IgE elevation. Clin Exp Immunol
23:444, 1976
97. VanNess MM, Diehl AM: Is liver biopsy useful in the evaluation of patients with chronically elevated liver enzymes?
Ann Intern Med 111:473, 1989
98. Wernicke C: Die akute haemorrhagische Polioencephalitis superior. In Wernicke C (ed): Lehrbuch der
Gehirnkrankheiten fuer Aerzte und Studierende. Bdll Kassel, Fischer, 1981, p 229
99. Yersin B, Nicolet JF, Dercy H, et al: Screening for excessive alcohol drinking: Comparative value of carbohydrate
deficient transferrin, gamma glutamyl transferase and mean corpuscular volume. Arch Intern Med 155:1907, 1995
100. Zieve L: Jaundice, hyperlipidemia and hemolytic anaemia: A heretofore unrecognized syndrome associated with

alcoholic fatty liver and cirrhosis. Ann Intern Med 48:471, 1958
101. ZilmDH, Jacob MS, MacLeod SM, et al: Propranolol and chlordiazepoxide effect on cardiac arrhythmias during
alcohol withdrawal. Alcohol Clin Exp Res 4:400, 1980
102. Zimmermann HJ, Maddrey WC: Acetaminophen (Paracetamol) hepatotoxicity with regular intake of alcohol: Analysis
of instances of therapeutic misadventure. Hepatology 22:767, 1995
103. Zubaran C, Fernandes JG, Rodnight R: Wernicke-Korsakoff syndrome. Postgrad Med J 73:27, 1997


April 22, 2000
Conde Petra
Cardiovascular, renal, and neurohumoral

responses to single large-volume paracentesis
in patients with cirrhosis and diuretic-resistant
Citation ascites [see comments]
Bibliographic Data
Abstract
Indexing Data
Cardiovascular, renal, and neurohumoral
Copyright Notice and Disclaimer responses to single large-volume
paracentesis in patients with cirrhosis
Find More Articles Like This and diuretic-resistant ascites [see
comments]
Full Text Peltekian KM - Am J Gastroenterol - 1997 Mar; 92(3): 394-9

Frontmatter
NLM Citation ID:
INTRODUCTION 97221411
METHODS Comment:
Am J Gastroenterol 1997 Mar;92(3):371-3
Study design
Laboratory analysis
Full Source Title:
Calculations
Publication Type:
Statistical analysis Journal Article
RESULTS Language:
DISCUSSION
English
ACKNOWLEDGMENTS Author Affiliation:

Department of Medicine, Toronto Hospital, Ontario, Canada.
REFERENCES
Authors:
About the Publication Peltekian KM; Wong F; Liu PP; Logan AG; Sherman M;
Blendis LM
Abstract:
OBJECTIVE: Large volume paracentesis is an effective
treatment for refractory ascites, but the need for routine infusion
of albumin or other volume expanders remains controversial.
The aim of this study was to assess the short term effects of a
single 5-L paracentesis without albumin replacement on total
central blood volume, systemic and renal hemodynamics,
sodium homeostasis, and neurohumoral factors. PATIENTS

AND METHODS: Twelve patients with biopsy-proven
cirrhosis and tense, diuretic-resistant ascites were studied before
and 48 h after a single 5-L paracentesis without albumin
infusion. Systemic hemodynamics and total central blood
volume were assessed using radionuclide angiography.
Glomerular filtration rate and effective renal plasma flow were
measured by inulin and para-aminohippurate clearances,
respectively. Lithium clearance was used as an index of
proximal tubular reabsorption of sodium. In addition, plasma
concentrations of neurohumoral factors were determined.
RESULTS: Total central blood volume was 2.41 +/- 0.33 L/m2
(mean +/- SEM) before and 2.34 +/- 0.18 L/m2 48 h after large
volume paracentesis (p = 0.76). Similarly, no differences were
detected in the cardiac index, glomerular filtration rate,
effective renal plasma flow, urinary sodium excretion,
hematocrit, plasma renin activity, or concentrations of plasma
aldosterone, norepinephrine, or atrial natriuretic factor.
CONCLUSIONS: A single large volume paracentesis without
albumin replacement causes no disturbances in systemic and
renal hemodynamics 48 h after the procedure. These results
suggest that a single 5-L paracentesis without albumin infusion
is a safe and satisfactory short term option for the management
of patients with cirrhosis and tense, diuretic-resistant ascites.
Major Subjects:
Ascites / Drug Therapy / Physiopathology / * Therapy
Heart / * Physiopathology
Kidney / * Physiopathology
Liver Cirrhosis / * Physiopathology
Neurotransmitters / Blood / * Physiology
* Paracentesis
p-Aminohippuric Acid / Pharmacokinetics
Aged
Albumins / Administration & Dosage / Therapeutic Use
Aldosterone / Blood
Atrial Natriuretic Factor / Blood
Blood Volume
Cardiac Output
Diuretics / Administration & Dosage / Therapeutic Use
Drug Resistance

Female
Glomerular Filtration Rate
Hematocrit
Homeostasis
Human
Inulin / Pharmacokinetics
Kidney Tubules, Proximal / Metabolism
Lithium / Pharmacokinetics
Male
Middle Age
Norepinephrine / Blood
Plasma Substitutes / Administration & Dosage /
Therapeutic Use
Renal Plasma Flow, Effective
Renin / Blood
Sodium / Metabolism / Pharmacokinetics / Urine
EC 3.4.23.15 (Renin); (Albumins); (Diuretics);
(Neurotransmitters); (Plasma Substitutes); 51-41-2
(Norepinephrine); 52-39-1 (Aldosterone); 61-78-9
(p-Aminohippuric Acid); 7439-93-2 (Lithium); 7440-23-5
(Sodium); 85637-73-6 (Atrial Natriuretic Factor); 9005-80-5
(Inulin)


394
Cardiovascular, Renal, and Neurohumoral Responses to Single

Large-Volume Paracentesis in Patients with Cirrhosis and
Diuretic-Resistant Ascites P
Kevork M. Peltekian M.D.

Florence Wong M.D.
Peter P. Liu M.D., Ph.D.
Alexander G. Logan M.D.
Morris Sherman M.D., Ph.D.
Laurence M. Blendis M.D.
Departments of Medicine and Nuclear Cardiology, The Toronto Hospital, and Samuel Lunenfeld
Research Institute, Mount Sinai Hospital, University of Toronto,
Toronto, Ontario, Canada
Objective: Large volume paracentesis is an effective treatment for refractory ascites, but the need
for routine infusion of albumin or other volume expanders remains controversial. The aim of this
study was to assess the short term effects of a single 5-L paracentesis without albumin replacement
on total central blood volume, systemic and renal hemodynamics, sodium homeostasis, and
neurohumoral factors. Patients and Methods: Twelve patients with biopsy-proven cirrhosis and
tense, diuretic-resistant ascites were studied before and 48 h after a single 5-L paracentesis without
albumin infusion. Systemic hemodynamics and total central blood volume were assessed using
radionuclide angiography. Glomerular filtration rate and effective renal plasma flow were
measured by inulin and para-aminohippurate clearances, respectively. Lithium clearance was used
as an index of proximal tubular reabsorption of sodium. In addition, plasma concentrations of
neurohumoral factors were determined. Results: Total central blood volume was 2.41 0.33 L/m2
(mean SEM) before and 2.34 0.18 L/m2 48 h after large volume paracentesis ( p= 0.76).
Similarly, no differences were detected in the cardiac index, glomerular filtration rate, effective
renal plasma flow, urinary sodium excretion, hematocrit, plasma renin activity, or concentrations
of plasma aldosterone, norepinephrine, or atrial natriuretic factor. Conclusions: A single large
volume paracentesis without albumin replacement causes no disturbances in systemic and renal
hemodynamics 48 h after the procedure. These results suggest that a single 5-L paracentesis

without albumin infusion is a safe and satisfactory short term option for the management of
patients with cirrhosis and tense, diuretic-resistant ascites.
Reprint requests and correspondence: Dr. L. M. Blendis, Toronto Hospital (9EN/220), 200 Elizabeth Street, Toronto,
Ontario, Canada M5G 2C4.
Received Apr. 3, 1996; accepted Sept. 4, 1996.
INTRODUCTION
In patients with cirrhosis, total paracentesis (complete mobilization of ascites in only one tap) [1] [2] and repeated large
volume paracentesis (evacuation of 4-6 L/day until complete mobilization of ascites is attained) [3] [4] are safe and effective
measures for the management of ascites. To prevent effective intravascular volume depletion and renal impairment after
these procedures, many authors advocate expanding plasma volume with albumin [5] [6] or plasma substitutes [7] [8] [9] .
Unfortunately, these expensive measures have not impacted the long term morbidity and mortality of patients with
cirrhosis and ascites [5] . Hence, management in some cirrhotic patients with tense, refractory ascites has shifted from
complete mobilization of ascites to rapid relief of discomfort by intermittent single large volume paracentesis whenever
necessary. There is no evidence that single large volume paracentesis results in adverse effects similar to those of total or
repeated large volume paracentesis. However, as a precaution, some authors have extended the recommendation for
administering albumin to all patients undergoing paracentesis [10] [11] [12] , a rather expensive measure.
Several studies have assessed plasma volume [13] [14] [15] , cardiac output [10] [16] , creatinine clearance [10] [17] , or
neurohumoral factors [10] [17] after a single large volume paracentesis without volume expansion. The results of these
studies are conflicting. Furthermore, none of these studies has measured central blood volume, the stimulus to volume
receptors and baroreceptors. Therefore, we measured total central blood volume (TCBV) by radionuclide angiography, as
well as the cardiovascular, renal, and neurohumoral responses, after a single 5-L paracentesis without albumin
Presented in part at Digestive Disease Week, May 11-14, 1994, New Orleans, LA.
395
infusion in patients with cirrhosis and tense, diuretic-resistant ascites. The radionuclide imaging technique [18] used in this
study is minimally invasive and has been extensively validated and successfully used in many studies to determine
ventricular filling and function [19] [20]
METHODS
Twelve patients (two females and 10 males) with biopsy-proven cirrhosis and tense, diuretic-resistant ascites were
recruited from the liver clinics of the Toronto Hospital. The mean age was 62 yr (range, 48-75 yr). The etiology of
cirrhosis was alcoholism in nine patients, cryptogenic in two patients, and viral hepatitis C in one patient. Diuretic
resistance was defined as failure to diurese on a combination of spironolactone 400 mg or amiloride 20 mg and furosemide
120 mg daily for 2 wk. In half of these patients, azotemia was a limiting factor for maximal diuretic therapy. Stable serum
creatinine and 24-h creatinine clearance with normal findings on urinalysis and ultrasonography of the kidneys excluded
acute renal disease. None had clinical, electrocardiographic, or radiological evidence of cardiovascular disease. Peripheral
edema was present in seven of the 12 patients.
Table 1 summarizes the prestudy characteristics of the recruited patients. The severity of hepatic decompensation was
assessed by the Child-Pugh score (grade B, eight patients; grade C, four patients) [21] . All patients abstained from alcohol
use for more than 3 months before the study. All had ascites resistant to diuretic therapy. None had undergone therapeutic
paracentesis for at least 10 days before the study. Diuretic therapy had been stopped at least 1 wk before admission. These
patients had no clinical or laboratory evidence of systemic infection, spontaneous bacterial peritonitis, or active GI
hemorrhage.
The study was performed with the approval of the University of The Toronto Human Subjects Review Committee and The
Toronto Hospital Committee for Research on Human Subjects. All patients gave informed written consent.

Study design
Patients were admitted to the Clinical Investigation Unit of The Toronto Hospital for a 1-wk equilibration period before
the study. They were maintained throughout on a caffeine-free diet containing 22 mEq sodium chloride with fluid
restricted to 1.2 L/day. Daily 24-h urine collections for sodium excretion were performed to assess sodium balance.
The evening before the study, patients received lithium 300 mg p.o. at bedtime. Lithium clearance was used as an index of
proximal tubular reabsorption of sodium [22] .
On the day of the study, at 8:00 AM, patients were given a dry, light breakfast. Then, an indwelling venous catheter was
inserted in each arm; one was used for blood sampling, and the other for inulin and para-aminohippurate (PAH)
administration by an infusion pump (Harvard Apparatus-22, Harvard, South Natick, MA). Inulin and PAH clearances
TABLE 1 -- Baseline Clinical and Laboratory Data of Patients at Enrollment in Study *

Age (yr) 63 2
Gender (n)
Female 2
Male 10
Etiology of cirrhosis (n)
Alcoholism 9
Unknown (cryptogenic) 2
Viral hepatitis C 1
Peripheral edema (n) 7
Child-Pugh score 91
A 0
B 8
C 4
Total bilirubin concentration (mumol/L) 25 4
Albumin concentration (g/L) 31 2
Prothrombin time (s) 14.8 0.8
Serum creatinine concentration (mumol/L) 124 16
6.2 0.6
Urinary sodium concentration (mEq/24 h)
* Data are expressed as mean SEM.
After being on a 22-mEq sodium-restricted diet for 1 wk.

were used to measure glomerular filtration rate (GFR) [23] and effective renal plasma flow (ERPF) [24] , respectively. After
blood and urine samples were collected for blanks, a loading infusion containing 25% inulin (60 mg/kg)
(Laevosan-Gesellschaft, Linz, Austria) and 20% PAH (8 mg/kg) (Merck, Sharp, and Dohme, West Point, PA) was started.
Then, inulin and PAH infusion rates were adjusted to maintain their respective plasma concentrations at 20 and 1.5 mg/dl.
After allowing a 1-h equilibration phase for plasma inulin and PAH to reach steady state levels, clearance measurements
were started.
Pulse rate and blood pressure were recorded hourly by an automatic cuff sphygmomanometer (Dinamap, Critikon, Tampa,
FL). Patients were given 150 ml of fluid p.o. each hour to maintain urine output.
Baseline hematocrit; plasma renin activity (PRA); and serum electrolyte, creatinine, aldosterone (Aldo), norepinephrine
(NE), and atrial natriuretic factor (ANF) concentrations were measured at the end of the 1-h equilibration phase.
Except when voiding, patients remained supine for the duration of the study. There were two clearance periods of 1 h each,
and the mean value of these two clearance periods was used in the calculations. Blood was drawn during each clearance
period for inulin, PAH, and lithium determinations. Urine samples were obtained over the same period to determine inulin,
PAH, and lithium concentrations and to measure urinary volume, electrolyte levels, and the level of urinary cyclic
guanosine 3',5'-monophosphate (UcGMP), the second messenger for ANF [25] .
At 2:00 PM on the same day, all patients underwent direct measurement of TCBV, the blood volume in the entire thorax,
including the blood volumes in the right and left lungs, by radionuclide angiography in the Nuclear Cardiology
396
Department of the Toronto Hospital. This methodology was previously described in detail [18] . Briefly, patients were
prelabeled with i.v. stannous pyrophosphate 2.5 mg, followed 30 min later by i.v. 99 mTc-pertechnetate 500 mBq. Using a
scintillation camera with a large field of view and computer acquisition (Elscint-409, Elscint, Arlington Heights, IL), gated
images of the heart and entire thorax were recorded with patients in the supine position. A sample of venous blood was
drawn to determine the counts per milliliter of blood volume for subsequent calculation of volumes of blood in various
compartments of the chest. The radiation dose of this imaging technique is less than 0.29 REM for the total body and less
than 1.95 REM for the spleen as a key target organ.
Paracentesis was then performed under local anesthesia the next morning using standard aseptic technique and a 14-gauge
needle. Over 30-60 min, a mean of 5 L (range, 4-6 L) of ascitic fluid was aspirated into 500-ml vacuum bottles. Albumin
was not administered during or after paracentesis.
All study parameters were reassessed 48 h after the paracentesis.
Laboratory analysis
Blood samples for ANF, PRA, Aldo, and NE determination were collected in prechilled tubes containing sodium
ethylene-diamine-tetraacetic acid. The tubes for ANF collection also contained aprotinin. All blood samples were
immediately centrifuged at 3000 rounds/min for 10 min at 4C. Plasma was separated, placed on ice, and then stored at
-70C until assayed.
ANF was measured by radioimmunoassay (RIA) (Peninsula Laboratories, Belmont, CA) [26] . Aldo was assayed with a
commercial RIA kit (Coat-A-Count Aldosterone Kit, Diagnostic Products, Los Angeles, CA). NE concentrations were
determined by means of high performance liquid chromatography with electrochemical detection [27] . PRA was estimated
by quantitation of angiotensin I generated in 3 h (Rianen Assay System Angiotensin I [125-I] Kit, DuPont, Wilmington,
DE). UcGMP was also measured by a commercial RIA kit (cGMP RIA kit, Amersham, Arlington Heights, IL).
Serum and urinary sodium concentrations were measured by the flame photometry method. Plasma and urine
concentrations of inulin, PAH, and lithium were determined by methods described in detail elsewhere [28] .
Calculations

Ejection fraction, end-diastolic volume, and heart rate were measured during radionuclide angiography and used in the
following calculations: cardiac output = heart rate stroke volume, where stroke volume = end-diastolic volume ejection
fraction; and systemic vascular resistance = mean arterial pressure cardiac output.
Cardiac and central vascular volume (CCVV), the blood volume in the cardiac chambers and the central vascular tree, was
derived by subtracting the blood volumes of the right and left lungs from the TCBV. Because cardiac output and central
blood volumes are affected by body size, cardiac output, TCBV, and CCVV were corrected for body surface area with the
use of the patient's height and estimated body weight without ascites and expressed per square meter.
Renal vascular resistance = mean arterial pressure effective renal blood flow, where effective renal blood flow = PAH
clearance (1 - hematocrit).
Inulin and PAH clearances were corrected for body surface area to represent GFR and ERPF, respectively, and expressed
per 1.73 m2 .
Urinary sodium excretion rate (UNaV) = urinary sodium urinary volume per minute. Filtration fraction = inulin
clearance PAH clearance. Proximal tubular reabsorption of sodium = (1 - lithium clearance) inulin clearance. Distal
delivery of sodium = serum sodium lithium clearance. Distal tubular reabsorption of sodium = (1 - UNaV) distal
delivery of sodium. Fractional excretion of sodium = UNaV (serum sodium inulin clearance) [29] .
Results are presented as mean SEM. For paired (pre- and postparacentesis) comparisons, we used the Wilcoxon signed
rank test [30] . For this test, normality and equality of variance assumptions are not needed. Statistical significance was
accepted at the 95% confidence level. All p values reported are for two-tailed probabilities.
RESULTS
All patients had diuretic-refractory ascites as defined earlier. The mean 24-h UNaV was 6.2 0.6 mmol/day on the day
before the study. The mean volume of ascites removed by paracentesis was 5.0 0.2 L. No significant differences were
detected in heart rate and mean arterial pressure before, during, and immediately after paracentesis. All patients tolerated
the procedure well and were relieved of the discomfort related to the tense ascites. No local abdominal complications were
observed. No differences were found in all the study parameters between patients with and without peripheral edema.
At 48 h after large volume paracentesis, no differences could be detected in cardiovascular response (Table 2) . There were
no statistically significant changes from baseline values for cardiac index (CI), TCBV, or CCVV.
None of these patients developed electrolyte imbalances or worsening of renal function. Significant differences could not
be detected in serum sodium level, renal parameters, or renal sodium handling before and after large volume paracentesis
(Table 3) .
When changes in effective arterial blood volume after paracentesis were assessed indirectly by measurement of
neurohumoral factors, no statistically significant differences were detected (Table 4) .
397
TABLE 2 -- Cardiovascular Response to Large Volume Paracentesis *

Baseline 48 h after Paracentesis p
MAP (mm Hg) 79 4 77 3 0.20
HR (beats/min) 79 3 80 3 0.68

CI (L/min/m2 ) 5.26 0.50 5.51 0.56 0.69

SVR (dynes/cm5 ) 819 111 787 92 0.69
TCBV (L/m2 ) 2.41 0.33 2.34 0.18 0.76
CCVV (L/m2 ) 1.55 0.15 1.62 0.15 0.40
In control subjects MAP = 87 3 mm Hg; HR = 65 3 beats/min; CI = 2.91 0.30 L/min/m2 ; SVR =
1443 121 dynes/cm5 ; TCBV = 1.29 0.10 L/m2 ; and CCVV = 0.88 0.08 L/m2 [18] .
HR, heart rate; MAP, mean arterial pressure; SVR, systemic vascular resistance.
TABLE 3 -- Serum Sodium and Renal Response to Large Volume Paracentesis *

48 h after
Baseline Paracentesis p
Serum sodium 130 1 130 2 1.00
concentration (mmol/L)
GFR (ml/m/1.73 m2 ) 35 7 38 8 0.78
ERPF (ml/m/1.73 m2 ) 350 77 318 68 0.37
Filtration fraction (%) 10.9 1.6 12.6 1.4 0.15
RVR (dynes/cm5 ) 23,043 5.702 25,304 6,742 0.78
UNaV (mumol/min) 8.8 2.0 9.9 2.4 0.37
PTRNa (%) 74.6 6.0 71.8 6.1 0.78
DTRNa (%) 98.4 0.5 98.4 0.4 0.88
FENa (%) 0.26 0.05 0.29 0.06 0.69
In control subjects GFR = 104 9 ml/min/1.73 m2 ; ERPF = 613 36 ml/min/1.73 m2 ; and RVR =
6759 528 dynes/cm5 [42] .
DTRNa, distal tubular reabsorption of sodium; FENa, fractional excretion of sodium; PTRNa, proximal
tubular reabsorption of sodium; RVR, renal vascular resistance; UNaV, urinary excretion of sodium.
TABLE 4 -- Response of Hematocrit and Neurohumoral Factors to Large Volume Paracentesis *

Baseline 48 h after Paracentesis p

Hematocrit 0.31 0.01 0.30 0.01 0.51

Aldo (pmol/L) 2975 585 3363 748 0.20
PRA (ng/L/s) 3.0 0.7 4.0 1.2 0.28
NE (nmol/L) 2.7 0.4 3.1 0.6 0.05
ANF (ng/L) 149 39 129 33 0.14
UcGMP (nmol/L) 950 266 751 146 0.18
In control subjects Aldo = 656 104 pmol/L; PRA = 1.0 0.3 ng/L/s; NE = 1.1 0.1 nmol/L; ANF =
57 7 ng/L [42] ; and UcGMP = 252 42 nmol/L.
Normal valve for hematocrit 0.40-0.54.
DISCUSSION
This study suggests that a single 5-L paracentesis does not cause effective hypovolemia at 48 h. CI, TCBV, and CCVV,
measured by radionuclide angiography, were not different before or after the paracentesis. Furthermore, indirect evidence
for changes in effective arterial blood volume could not be detected when parameters such as PRA, Aldo, and NE were
measured. No differences could be detected in ANF or its second messenger, UcGMP, suggesting no fluctuations in the
atrial stretch or venous circulation. In addition, we could not detect any worsening of renal function as assessed by
measurements of GFR, ERPF, renal vascular resistance, and renal sodium handling.
The radionuclide imaging technique used in this study has been used to determine absolute left ventricular volume [31] [32] ,
pulmonary blood volume [33] , and blood volume shifts in different vascular beds [34] . It has also been used to measure
central blood volume in cirrhotic patients with and without ascites [18] [35] . This technique allows direct measurement of
left ventricular ejection fraction and end-diastolic volume, TCBV in the entire thorax, and blood volume in right and left
lungs without resorting to geometric assumptions. The only limitation of this technique relates to errors secondary to
variability in count attenuation caused by patient differences in chest wall thickness and cardiac configuration. In general,
these count-based volumes are higher in thin patients and lower in obese patients. However, relative changes in volumes
over time in the same patient would not be affected by these interpatient differences.
Studies on the short term effects of single large volume paracentesis on systemic hemodynamics have yielded conflicting
results [11] [13] [14] [15] [16] [17] [29] [36] . In studies that measured plasma volume directly by the 125 I-labeled human serum
albumin dilution technique [13] [14] [15] , there was no evidence of hypovolemia up to 48 h after single large volume
paracentesis without administration of plasma expanders in patients with [14] or without [15] peripheral edema. There were
no changes in plasma volume, calculated from fluctuations in hematocrit, up to 5 days after large volume paracentesis
without administration of albumin [13] . One study measuring cardiac output by a dye-injection method [37] found no
changes in cardiac output 24 h and 7 days after large volume paracentesis without administration of plasma expanders [16] .
These findings were substantiated by another study that detected no differences in PRA, a neurohumoral marker of
effective arterial blood volume, 48 h and 7 days after large volume paracentesis without albumin infusion [17] .
In one study, large volume paracentesis was implicated in causing hypovolemia on the basis of indirect evidence of a rise
in PRA and Aldo 24 h after single paracentesis [11] . In contrast to all other studies [14] [15] [17] , the same study also showed
a mild but significant rise in serum creatinine and in 12-h creatinine clearance 24 h after large volume paracentesis without
albumin infusion [11] . The contradictory results of this study may have been related to discontinuation of diuretics only 24
h before paracentesis in some patients.

The current recommendations for management of refractory ascites in patients with cirrhosis are [38] 1) to administer
398
albumin i.v. at a dose of 6-8 g/L ascitic fluid removed whenever complete mobilization of ascites is to be achieved by total
or repeated large volume paracentesis and 2) to resume diuretics to prevent or delay the reaccumulation of ascites.
Unfortunately, patients who received albumin after repeated large volume paracentesis did not differ in the rate or cause of
rehospitalization, survival, or cause of death from those who did not receive albumin.
In cirrhotic patients with ascites, the incidence of diuretic-induced complications is high, especially when high doses of
diuretics are used [3] [39] . Moreover, between 5 and 10% of patients with cirrhosis hospitalized for the treatment of tense
ascites do not respond to diuretics [40] . In these diuretic-resistant patients, maximal oral diuretic therapy causes a
significantly steeper rise in serum creatinine concentration [41] . Therefore, in our opinion, it would be safer in cirrhotic
patients with diuretic-resistant ascites to stop all diuretics, to restrict dietary sodium and fluid intake to prevent the rapid
accumulation of ascites, and to manage the tense ascites by intermittent large volume paracentesis for the relief of
discomfort whenever necessary.
This study shows that there are no alterations in the cardiovascular, renal, or neurohumoral responses to single large
volume paracentesis without albumin replacement 48 h after the procedure. This procedure in cirrhotic patients with tense,
diuretic-resistant ascites does not cause effective hypovolemia. If future studies confirm the long term safety and
effectiveness of performing intermittent single 5-L paracentesis without albumin infusion, then this may be a cost-saving
alternative strategy for management of tense, diuretic-resistant ascites in patients with cirrhosis.
ACKNOWLEDGMENTS
We gratefully acknowledge the technical expertise of Ms. Yasmin Allidina, Mr. Andre Laprade, and the staff of the
Department of Nuclear Cardiology at The Toronto Hospital. We also acknowledge the expert assistance of Ms. Nancy
Law and the nursing and dietetic staff of the Clinical Investigation Unit.
K. M. Peltekian received the Schering Hepatology Fellowship at The Toronto Hospital.
REFERENCES
1. Tito L, Gines P, Arroyo V, et al. Total paracentesis associated with intravenous albumin management of patients with cirrhosis and ascites. Gastroenterology
1990;98:146-51.
2. Sola R, Vila MC, Andreu M, et al. Total paracentesis with dextran-40 vs diuretics in the treatment of ascites in cirrhosis: A randomized controlled trial. J Hepatol
1994;20:282-8.
3. Gines P, Arroyo V, Quintero E, et al. Comparison of paracentesis and diuretics in the treatment of cirrhosis with tense ascites: Results of a randomized study.
4. Salerno F, Badalamenti S, Incerti P, et al. Repeated paracentesis and i.v. albumin infusion to treat "tense" ascites in cirrhotic patients: A safe alternative therapy.
J Hepatol 1987;5:102-8.
5. Gines P, Tito L, Arroyo V, et al. Randomized comparative study of paracentesis with and without intravenous albumin in cirrhosis. Gastroenterology
1988;94:1493-1502.
6. Garcia-Compean D, Villarreal JZ, Cuevas HB, et al. Total therapeutic paracentesis (TTP) with and without intravenous albumin in the treatment of cirrhotic
tense ascites: A randomized controlled trial. Liver 1993;13:233-8.
7. Planas R, Gines P, Arroyo V, et al. Dextran-70 versus albumin as plasma expanders in cirrhotic patients with tense ascites treated with total paracentesis: Results
of a randomized study. Gastroenterology 1990;99:1736-45.
8. Salerno F, Badalamenti S, Lorenzano E, et al. Randomized comparative study of hemaccel vs albumin infusion after total paracentesis in cirrhotic patients with
refractory ascites. Hepatology 1991;13:707-13.

9. Fassio E, Terg R, Landeira G, et al. Paracentesis with dextran 70 vs paracentesis with albumin in cirrhosis with tense ascites: Results of a randomized study. J
Hepatol 1992;14:310-6.
10. Kellerman PS, Linas SL. Large volume paracentesis in the treatment of ascites. Ann Intern Med 1990;112:889-91.
11. Simon DM, McCain JR, Bonkovsky HL, et al. Effects of therapeutic paracentesis on systemic and hepatic hemodynamics and on renal and hormonal function.
12. Porayko MK, Wiesner RH. Management of ascites in patients with cirrhosis: What to do when diuretics fail. Postgrad Med 1992;92:155-66.
13. Kowalski HJ, Abelmann WH, McNeely WF. The cardiac output in patients with cirrhosis of the liver and tense ascites with observations on the effect of
paracentesis. J Clin Invest 1954;33:768-73.
14. Gordon ME. The acute effects of abdominal paracentesis in Laenec's cirrhosis upon exchanges of electrolytes and water, renal function and hemodynamics. Am
J Gastroenterol 1960;33:15-37.
15. Kao HW, Rakov NE, Savage E, et al. The effect of large volume paracentesis on plasma volume: A cause of hypovolemia? Hepatology 1985;5:403-7.
16. Pinto PC, Amerian J, Reynolds TB. Large volume paracentesis in non-edematous patients with tense ascites: Its effects on intravascular volume. Hepatology
1988;8:207-10.
17. Gentile S, Angelico M, Bologna E, et al. Clinical, biochemical, and hormonal changes after a single, large volume paracentesis in cirrhosis with ascites. Am J
18. Wong F, Liu P, Tobe S, et al. Central blood volume in cirrhosis: Measurement by radionuclide angiography. Hepatology 1994;19:312-21.
19. Goodman JM, McLaughlin PR, Plyley MJ, et al. Impaired cardiopulmonary response to exercise in moderate hypertension. Can J Cardiol 1992;8:363-71.
20. Konstam MA, Kronenberg MW, Udelson JE, et al. Effectiveness of pre-load reserve as a determinant of clinical status in patients with left ventricular systolic
dysfunction. Am J Cardiol 1992;69:1591-5.
21. Pugh RNH, Murray-Lyon IM, Dawson JL, et al. Transection of the oesophagus for bleeding oesophageal varices. Br J Surg 1973;60:646-9.
22. Thomsen K. Lithium clearance as a measure of sodium and water delivery from the proximal tubules. Kidney Int 1990;28:S10-6.
23. Walser M, Davidson DG, Orloff J. The renal clearance of alkali-stable inulin. J Clin Invest 1955;34:1520-3.
24. Brun C. A rapid method for the determination of para-aminohippuric acid in kidney function tests. J Lab Clin Med 1951;37:955-8.
25. Leung WM, Logan AG, Campbell PJ, et al. Role of atrial natriuretic peptide and urinary cGMP in the natriuretic and diuretic response to central hypovolemia
in normal human subjects. Can J Physiol Pharmacol 1987;65:2076-80.
26. Larose P, Meloche S, du Souich P, et al. Radioimmunoassay of atrial natriuretic factor: Human plasma levels. Biochem Biophys Res Commun 1985;130:553-8.
27. Matson WR, Langlais P, Volicer L, et al. n-Electrode, three dimensional liquid chromatography with electrochemical detection for determination of
neurotransmitters. Clin Chem 1984;30:1477-88.
28. Miller JA, Floras JS, Skorecki KL, et al. Renal and humoral responses to sustained cardiopulmonary baroreceptor deactivation in humans. Am J Physiol
1991;260:R642-8.
29. Epstein M. Renal sodium handling in liver disease. In: Epstein M, ed. The kidney in liver disease, 3rd ed. Baltimore, MD: Williams & Wilkins, 1988: 3-30.
399
30. McKenzie J, Schafer RL, Farber E. The student edition of Minitab for Windows software, release 9. State College, PA: Minitab, 1995.
31. Links JM, Becker JC, Shindledecker JG, et al. Measurement of absolute left ventricular volume from gated blood pool studies. Circulation 1982;65:82-90.
32. Lavie CJ, O'Keefe JH Jr, Chesebro JH, et al. Prevention of late ventricular dilatation after acute myocardial infarction by successful thrombolytic reperfusion.
Am J Cardiol 1990;66:31-6.
33. Okada RD, Pohost GM, Kirshenbaum HD, et al. Radionuclide-determined changes in pulmonary blood volume with exercise. N Engl J Med 1979;301:569-76.

34. Baccelli G, Terrani S, Pacenti P, et al. Simultaneous recording of blood volume shifts in different vascular beds in man by versatile scintigraphic methods.
Angiology 1993;44:615-21.
35. Wong F, Liu P, Allidina Y, et al. Pattern of sodium handling and its consequences in pre-ascitic cirrhosis. Gastroenterology 1995;108:1820-7.
36. Carey WD, Kohne JC, Leatherman J, et al. Ascitic fluid removal: Does it cause renal or hemodynamic decompensation? Cleve Clin Q 1983;50:397-400.
37. Hamilton WF, Riley RL, Attyah, et al. Comparison of the Fick and dye injection methods for measuring the cardiac output in man. Am J Physiol
1948;153:309-15.
38. Gines P, Arroyo V. Paracentesis in the management of cirrhotic ascites. J Hepatol 1993;17:S14-8.
39. Sherlock S, Senewiratne B, Scot A, et al. Complications of diuretic therapy in hepatic cirrhosis. Lancet 1966;1:1049-53.
40. Forns X, Gines A, Gines P, et al. Management of ascites and renal failure in cirrhosis. Semin Liver Dis 1994;14:82-96.
41. Rector WG Jr. "Diuretic-resistant" ascites: Observations on pathogenesis. Arch Intern Med 1986;146:1597-1600.
42. Wong F, Logan AG, Blendis LM. Systemic hemodynamic, forearm vascular, renal and humoral response to sustained cardiopulmonary baroreceptor
deactivation in cirrhosis. Hepatology 1995;21:717-24.


April 22, 2000
Conde Petra
Clinical practice guidelines for the management

of cirrhotic patients with ascites. Committee on
Ascites of the Italian Association for the Study
Citation of the Liver.
Bibliographic Data
Abstract
Indexing Data
Clinical practice guidelines for the
Copyright Notice and Disclaimer management of cirrhotic patients with
ascites. Committee on Ascites of the
Journal Articles On This Topic Italian Association for the Study of the
Liver.
Order a Full Text Copy of the Salerno F - Ital J Gastroenterol Hepatol - 1999 Oct; 31(7):
Original Journal Article 626-34
NLM Citation ID:
20071774
Full Source Title:
Italian Journal of Gastroenterology and Hepatology
Publication Type:
Guideline; Journal Article; Practice Guideline; Review;
Review, Tutorial
Language:
English
Author Affiliation:
Department of Internal Medicine, A. Migliavacca Centre,
IRCCS Policlinico, University of Milan, Italy.
francesco.salerno@unimi.it
Authors:
Salerno F; Angeli P; Bernardi M; Laffi G; Riggio O; Salvagnini
M
67
Abstract:
An ad hoc Committee appointed by the Italian Association for
the Study of the Liver prepared these Clinical Practice
Guidelines for the Management of Cirrhotic Patients with

Ascites. The initial evaluation of a patient with ascites should
include a history, physical evaluation, paracentesis with ascitic
fluid analysis, abdominal ultrasonography and biochemistry to
assess the severity of liver disease and renal functionality. To
improve the efficiency of the choice between the different
opportunities available in the treatment of ascites, patients can
be classified into two subgroups: patients with uncomplicated
ascites and patients with complicated ascites, including
refractory ascites, bacterial peritonitis, hyponatraemia and renal
failure. Based upon evidence emerging from controlled clinical
trials or case-control studies, satisfactory treatment for
uncomplicated ascites is represented by paracentesis,
sodium-restricted diet and diuretics, whereas the treatment of
patients with complicated ascites requires other specific
approaches. As the prognosis for most patients with ascites is
poor, the last part of the paper offers simple criteria in the
selection of patients candidates for liver transplantation. The
aim of these guidelines is to reduce inappropriate practice and
to improve efficiency in the management of patients with
ascites. The Committee holds that a periodic update will be
necessary to conform to future scientific developments.
Major Subjects:
Ascites / Etiology / Mortality / * Therapy
Liver Cirrhosis / Complications / Diagnosis / Mortality /

* Therapy
Patient Care / * Standards
Female
Human
Italy
Liver Transplantation / Standards
Male
Prognosis
Societies, Medical
Survival Rate

MD Consult - Drug Information
April 22, 2000
Conde Petra

About GenRx
Key Information
Drug Identification Guide/Pill Images Furosemide (001342)

Generic Drugs
CATEGORIES:
US Drugs
Global Drugs Indications: Edema, pulmonary, adjunct; Edema,

secondary to heart failure; Edema, secondary to hepatic
Indications cirrhosis; Edema, secondary to renal dysfunction;
FDA-Class Hypertension, essential
Pregnancy Category C
Category
Top 200 Drugs; WHO Formulary
Ingredient FDA Approved 1968 Mar
Comprehensive Index FDA DRUG CLASS: Antihypertensives; Diuretics
Supplier Information
BRAND NAMES: Aldic (Thailand); Aluzine *; Anfuramaide
(Taiwan); Apo-Frusemide (New-Zealand); Apo-Furosemide
(Canada); Aquarid (South-Africa); Aquamide *; Aquasin *;
Arasemide (Japan); Bioretic *; Cetasix (Indonesia); Dirine
(Malaysia); Disemide *; Diural (Denmark, Norway, Switzerland,
Sweden); Diuresal (Bahrain, Cyprus, Egypt, Iran, Iraq, Israel, Jordan,
Kuwait, Lebanon, Libya, Oman, Qatar, Republic-of-Yemen, Saudi-Arabia,
Syria, United-Arab-Emirates; Benin, Burkina-Faso, Ethiopia, Gambia,
Ghana, Guinea, Ivory-Coast, Kenya, Liberia, Malawi, Mali, Mauritania,
Mauritius, Morocco, Niger, Nigeria, Senegal, Seychelles, Sierra-Leone,
South-Africa, Sudan, Tanzania, Tunia, Uganda, Zambia, Zimbabwe;
Bahamas, Barbados, Belize, Bermuda, Curacao, Guyana, Jamaica,
Netherland-Antilles, Puerto-Rico, Surinam, Trinidad, Taiwan); Diurin
(New-Zealand); Diurolasa (Spain); Diusemide (Benin, Burkina-Faso,
Ethiopia, Gambia, Ghana, Guinea, Ivory-Coast, Kenya, Liberia, Malawi,
Mali, Mauritania, Mauritius, Morocco, Niger, Nigeria, Senegal,
Seychelles, Sierra-Leone, Sudan, Tanzania, Tunia, Uganda, Zambia,
Zimbabwe); Diusil *; Dranex *; Dryptal (England, Ireland,
Hong-Kong); Durafurid (Germany); Edenol (Mexico); Errolon
(Argentina); Eutensin (Japan); Fluidrol *; Franyl (Japan); Frumex
*; Furmid (Malaysia); Frusedan (Benin, Burkina-Faso, Ethiopia,
Gambia, Ghana, Guinea, Ivory-Coast, Kenya, Liberia, Malawi, Mali,
Mauritania, Mauritius, Morocco, Niger, Nigeria, Senegal, Seychelles,
http://home.mdconsult.com/das/drug/view/1/1342/top?stty=cow&nid=39985&tlt=Z&pos=&tag=relinfo&anc=dr (1 of 17) [22/04/2000 0:29:24]
Sierra-Leone, South-Africa, Sudan, Tanzania, Tunia, Uganda, Zambia,

Zimbabwe); Frusema (Philippines); Frusemid (Colombia);
Frusemide (Hong-Kong); Frusetic *; Frusid (New-Zealand);
Furanthril (Czech-Republic); Furanturil (Bulgaria); Furetic
(Thailand); Furex *; Furix (Norway, Sweden, Korea); Frumide *;
Furo-Basan (Switzerland); Furocot (US); Furodiurol *; Furomen
(Finland); Furomex (Czech-Republic); Furomide M.D. (US);
Furo-Puren (Germany); Furorese (Germany); Furosan *;
Furoside (Canada); Furosix (Indonesia); Furoter *; Furovite
(Israel); Fusid (Germany, Israel); Golan *; Hissuflux (Colombia);
Hydrex (South-Africa); Hydro *; Impugan (Denmark, Sweden,
Switzerland, Indonesia, Thailand); Kofuzon (Taiwan); Kutrix (Japan);
Lasemid *; Lasiletten (Netherlands); Lasilix (France, Morocco);
Lasix (US); Lasix Retard (Denmark, Portugal, Netherlands, Norway,
Sweden); Laxur (Chile); Liside *; Marsemide (Philippines); Naclex
(Indonesia); Nadis *; Nelsix *; Nephron (Argentina); Novosemide
(Canada); Odemase *; Odemex (Costa-Rica, Dominican-Republic,
El-Salvador, Guatemala, Nicaragua, Panama, Ecuador); Oedemex
(Bahrain, Cyprus, Egypt, Iran, Iraq, Israel, Jordan, Kuwait, Lebanon,
Libya, Oman, Qatar, Republic-of-Yemen, Saudi-Arabia, Syria,
United-Arab-Emirates, Switzerland); Promedes (Japan); Promide *;
Radisemide (Benin, Burkina-Faso, Ethiopia, Gambia, Ghana, Guinea,
Ivory-Coast, Kenya, Liberia, Malawi, Mali, Mauritania, Mauritius,
Morocco, Niger, Nigeria, Senegal, Seychelles, Sierra-Leone, South-Africa,
Sudan, Tanzania, Tunia, Uganda, Zambia, Zimbabwe; Bahrain, Cyprus,
Egypt, Iran, Iraq, Israel, Jordan, Kuwait, Lebanon, Libya, Oman, Qatar,
Republic-of-Yemen, Saudi-Arabia, Syria, United-Arab-Emirates);
Radonna *; Retep (Argentina); Rosis (Taiwan); Salinex (India);
Salurid *; Seguril (Spain); Sigasalur (Germany); Trofurit *;
Uremide (Australia, New-Zealand); Urenil (Japan); Uresix
(Indonesia); Urex (Australia, New-Zealand, Japan, Hong-Kong);
Urex-M (Australia, New-Zealand); Urian *; Uridon *; Uritol
(Canada); Yidoli *;
(International brand names outside U.S. in italics)
* Indicates foreign drugs without region specification at time of
publication.
COST OF THERAPY: $ 18.54 (Hypertension; Tablet; 40 mg;

2/day; 365 days)
HCFA JCODES: J1940 up to 20 mg IM, IV
Furosemide is a potent diuretic which, if given in excessive

amounts, can lead to a profound diuresis with water and
electrolyte depletion. Therefore, careful medical supervision
is required and dose and dose schedule must be adjusted to
the individual patient's needs. (See DOSAGE AND

ADMINISTRATION.)
DESCRIPTION:
Tablets: Lasix is a diuretic which is an anthranilic acid
derivative. Lasix for oral administration contains furosemide as
the active ingredient and the following inactive ingredients:
lactose, magnesium stearate, starch, and talc. Chemically, it is
4-chloro-N-furfuryl-5-sulfamoylanthranilic acid. Furosemide is
available as white tablets for oral administration in dosage
strengths of 20, 40 and 80 mg.
Furosemide is a white to off-white odorless crystalline powder.
It is practically insoluble in water, sparingly soluble in alcohol,
freely soluble in dilute alkali solutions and insoluble in dilute
acids.
The CAS Registry Number is 54-31-9.
Injection: Lasix injection is composed of
4-chloro-N-furfuryl-5- sulfamoylanthranilic acid, sodium
chloride for isotonicity and sodium hydroxide to adjust pH.
Lasix injection 10 mg/ml is a sterile, non-pyrogenic solution in
ampules, disposable syringes and single dose vials for
intravenous and intramuscular injection.
Oral Solution: Lasix oral solution contains furosemide as the
active ingredient and the following inactive ingredients: alcohol
11.5%, D&C yellow #10, FD&C yellow #6 as color additives,
flavors, glycerin, parabens, purified water, sorbitol; sodium
hydroxide added to adjust pH. Lasix oral solution 10 mg/ml is
an orange flavored liquid for oral administration.
CLINICAL PHARMACOLOGY:
Investigations into the mode of action of furosemide have
utilized micropuncture studies in rats, stop flow experiments in
dogs and various clearance studies in both humans and
experimental animals. It has been demonstrated that furosemide
inhibits primarily the absorption of sodium and chloride not
only in the proximal and distal tubules but also in the loop of
Henle. The high degree of efficacy is largely due to this unique
site of action. The action on the distal tubule is independent of
any inhibitory effect on carbonic anhydrase and aldosterone.

Recent evidence suggests that furosemide glucuronide is the

only or at least the major biotransformation product of
furosemide in man. Furosemide is extensively bound to plasma
proteins, mainly to albumin. Plasma concentrations ranging
from 1 to 400 mcg/ml are 91 to 99% bound in healthy
individuals. The unbound fraction averages 2.3 to 4.1% at
therapeutic concentrations.
The onset of diuresis following oral administration is within 1
hour. The peak effect occurs within the first or second hour.
The duration of diuretic effect is 6 to 8 hours.
The onset of diuresis following intravenous administration is
within 5 minutes and somewhat later after intramuscular
administration. The peak effect occurs within the first half hour.
The duration of diuretic effect is approximately 2 hours.
In fasted normal men, the mean bioavailability of furosemide
from tablets and oral solution is 64% and 60%, respectively, of
that from an intravenous injection of the drug. Although
furosemide is more rapidly absorbed from the oral solution (50
minutes) than from the tablet (87 minutes), peak plasma levels
and area under the plasma concentration-time curves do not
differ significantly. Peak plasma concentrations increase with
increasing dose but times-to-peak do not differ among doses.
The terminal half-life of furosemide is approximately 2 hours.
Significantly more furosemide is excreted in urine following the
IV injection than after the tablet or oral solution. There are no
significant differences between the two oral formulations in the
amount of unchanged drug excreted in urine.
INDICATIONS AND USAGE:

Parenteral therapy should be reserved for patients unable to take
oral medication or for patients in emergency clinical situations.
Tablets, Injection, and Oral Solution

Edema: Furosemide is indicated in adults and pediatric patients
for the treatment of edema associated with congestive heart
failure, cirrhosis of the liver, and renal disease, including the
nephrotic syndrome. Furosemide is particularly useful when an
agent with greater diuretic potential is desired.
Tablets and Oral Solution

Hypertension: Oral furosemide may be used in adults for the

treatment of hypertension alone or in combination with other

antihypertensive agents. Hypertensive patients who cannot be
adequately controlled with thiazides will probably also not be
adequately controlled with furosemide alone.
Injection
Furosemide is indicated as adjunctive therapy in acute
pulmonary edema. The intravenous administration of
furosemide is indicated when a rapid onset of diuresis is
desired, e.g., in acute pulmonary edema.
If gastrointestinal absorption is impaired or oral medication is
not practical for any reason, furosemide is indicated by the
intravenous or intramuscular route. Parenteral use should be
replaced with oral furosemide as soon as practical.
CONTRAINDICATIONS:
Furosemide is contraindicated in patients with anuria and in
patients with a history of hypersensitivity to furosemide.
WARNINGS:
In patients with hepatic cirrhosis and ascites, furosemide
therapy is best initiated in the hospital. In hepatic coma and in
states of electrolyte depletion, therapy should not be instituted
until the basic condition is improved. Sudden alterations of fluid
and electrolyte balance in patients with cirrhosis may precipitate
hepatic coma; therefore, strict observation is necessary during
the period of diuresis. Supplemental potassium chloride and, if
required, an aldosterone antagonist are helpful in preventing
hypokalemia and metabolic alkalosis.
If increasing azotemia and oliguria occur during treatment of
severe progressive renal disease, furosemide should be
discontinued.
Cases of tinnitus and reversible or irreversible hearing
impairment have been reported. Usually, reports indicate that
furosemide ototoxicity is associated with rapid injection, severe
renal impairment, doses exceeding several times the usual
recommended dose, or concomitant therapy with
aminoglycoside antibiotics, ethacrynic acid, or other ototoxic
drugs. If the physician elects to use high dose parenteral
therapy, controlled intravenous infusion is advisable (for adults,
an infusion rate not exceeding 4 mg furosemide per minute has
been used).
Injection
Pediatric Use: In premature neonates with respiratory distress
syndrome, diuretic treatment with furosemide in the first few
weeks of life may increase the risk of persistent patent ductus
arteriosus (PDA), possibly through a prostaglandin-E-mediated
process.
Hearing loss in neonates has been associated with the use of
furosemide injection (see WARNINGS).
PRECAUTIONS:
General: Excessive diuresis may cause dehydration and blood
volume reduction with circulatory collapse and possibly
vascular thrombosis and embolism, particularly in elderly
patients. As with any effective diuretic, electrolyte depletion
may occur during furosemide therapy, especially in patients
receiving higher doses and a restricted salt intake. Hypokalemia
may develop with furosemide, especially with brisk diuresis,
inadequate oral electrolyte intake, when cirrhosis is present, or
during concomitant use of corticosteroids or ACTH. Digitalis
therapy may exaggerate metabolic effects of hypokalemia,
especially myocardial effects.
All patients receiving furosemide therapy should be observed
for these signs or symptoms of fluid or electrolyte imbalance
(hyponatremia, hypochloremic alkalosis, hypokalemia,
hypomagnesemia or hypocalcemia): dryness of mouth, thirst,
weakness, lethargy, drowsiness, restlessness, muscle pains or
cramps, muscular fatigue, hypotension, oliguria, tachycardia,
arrhythmia, or gastrointestinal disturbances such as nausea and
vomiting. Increases in blood glucose and alterations in glucose
tolerance tests (with abnormalities of the fasting and 2-hour
postprandial sugar) have been observed, and rarely,
precipitation of diabetes mellitus has been reported.
Asymptomatic hyperuricemia can occur and gout may rarely be
precipitated.
Oral Solution
The sorbitol present in the vehicle may cause diarrhea
(especially in children) when higher doses of furosemide oral
solution are given.

Patients allergic to sulfonamides may also be allergic to
furosemide. The possibility exists of exacerbation or activation
of systemic lupus erythematosus.
As with many other drugs, patients should be observed
regularly for the possible occurrence of blood dyscrasias, liver
or kidney damage, or other idiosyncratic reactions.
Information for the Patient: Patients receiving furosemide
should be advised that they may experience symptoms from
excessive fluid and/or electrolyte losses. The postural
hypotension that sometimes occurs can usually be managed by
getting up slowly. Potassium supplements and/or dietary
measures may be needed to control or avoid hypokalemia.
Patients with diabetes mellitus should be told that furosemide
may increase blood glucose levels and thereby affect urine
glucose tests. The skin of some patients may be more sensitive
to the effects of sunlight while taking furosemide.
Hypertensive patients should avoid medications that may
increase blood pressure, including over-the-counter products for
appetite suppression and cold symptoms.
Laboratory Tests: Serum electrolytes (particularly potassium),
CO2, creatinine and BUN should be determined frequently
during the first few months of furosemide therapy and
periodically thereafter. Serum and urine electrolyte
determinations are particularly important when the patient is
vomiting profusely or receiving parenteral fluids. Abnormalities
should be corrected or the drug temporarily withdrawn. Other
medications may also influence serum electrolytes.
Reversible elevations of BUN may occur and are associated
with dehydration, which should be avoided, particularly in
patients with renal insufficiency.
Urine and blood glucose should be checked periodically in
diabetics receiving furosemide, even in those suspected of latent
diabetes.
Furosemide may lower serum levels of calcium (rarely cases of
tetany have been reported) and magnesium. Accordingly, serum
levels of these electrolytes should be determined periodically.
Carcinogenesis, Mutagenesis, and Impairment of Fertility:
Furosemide was tested for carcinogenicity by oral
administration in one strain of mice and one strain of rats. A

small but significantly increased incidence of mammary gland
carcinomas occurred in female mice at a dose 17.5 times the
maximum human dose of 600 mg. There were marginal
increases in uncommon tumors in male rats at a dose of 15
mg/kg (slightly greater than the maximum human dose) but not
at 30 mg/kg.
Furosemide was devoid of mutagenic activity in various strains
of Salmonella typhimurium when tested in the presence or
absence of an in vitro metabolic activation system, and
questionably positive for gene mutation in mouse lymphoma
cells in the presence of rat liver S9 at the highest dose tested.
Furosemide did not induce sister chromatid exchange in human
cells in vitro , but other studies on chromosomal aberrations in
human cells in vitro gave conflicting results. In Chinese hamster
cells it induced chromosomal damage but was questionably
positive for sister chromatid exchange. Studies on the induction
by furosemide of chromosomal aberrations in mice were
inconclusive. The urine of rats treated with this drug did not
induce gene conversion in Saccharomyces cerevisiae .
Furosemide produced no impairment of fertility in male or
female rats, at 100 mg/kg/day (the maximum effective diuretic
dose in the rat and 8 times the maximal human dose of 600
mg/day).
Pregnancy Category C: Furosemide has been shown to cause
unexplained maternal deaths and abortions in rabbits at 2, 4 and
8 times the maximal recommended human dose. There are no
adequate and well-controlled studies in pregnant women.
Furosemide should be used during pregnancy only if the
potential benefit justifies the potential risk to the fetus.
The effects of furosemide on embryonic and fetal development
and on pregnant dams were studied in mice, rats and rabbits.
Furosemide caused unexplained maternal deaths and abortions
in the rabbit at the lowest dose of 25 mg/kg (2 times the
maximal recommended human dose of 600 mg/day). In another
study, a dose of 50 mg/kg (4 times the maximal recommended
human dose of 600 mg/day) also caused maternal deaths and
abortions when administered to rabbits between Days 12 and 17
of gestation. In a third study, none of the pregnant rabbits
survived a dose of 100 mg/kg. Data from the above studies
indicate fetal lethality that can precede maternal deaths.
The results of the mouse study and one of the three rabbit

studies also showed an increased incidence and severity of

hydronephrosis (distention of the renal pelvis and, in some
cases, of the ureters) in fetuses derived from the treated dams as
compared with the incidence in fetuses from the control group.
Nursing Mothers: Because it appears in breast milk, caution
should be exercised when furosemide is administered to a
nursing mother.
Injection
Pediatric Use: Renal calcifications (from barely visible on
x-ray to staghorn) have occurred in some severely premature
infants treated with intravenous furosemide for edema due to
patent ductus arteriosus and hyaline membrane disease. The
concurrent use of chlorothiazide has been reported to decrease
hypercalciuria and dissolve some calculi.
DRUG INTERACTIONS:
Furosemide may increase the ototoxic potential of
aminoglycoside antibiotics, especially in the presence of
impaired renal function. Except in life-threatening situations,
avoid this combination.
Furosemide should not be used concomitantly with ethacrynic
acid because of the possibility of ototoxicity. Patients receiving
high doses of salicylates concomitantly with furosemide, as in
rheumatic disease, may experience salicylate toxicity at lower
doses because of competitive renal excretory sites.
Furosemide has a tendency to antagonize the skeletal muscle
relaxing effect of tubocurarine and may potentiate the action of
succinylcholine.
Lithium generally should not be given with diuretics because
they reduce lithium's renal clearance and add a high risk of
lithium toxicity.
Furosemide may add to or potentiate the therapeutic effect of
other antihypertensive drugs. Potentiation occurs with
ganglionic or peripheral adrenergic blocking drugs.
Furosemide may decrease arterial responsiveness to
norepinephrine. However, norepinephrine may still be used
effectively.
Tablets

Simultaneous administration of sucralfate and furosemide

tablets may reduce the natriuretic and antihypertensive effects
of furosemide. Patients receiving both drugs should be observed
closely to determine if the desired diuretic and/or
antihypertensive effect of furosemide is achieved. The intake of
furosemide and sucralfate should be separated by at least two
hours.

One study in six subjects demonstrated that the combination of
furosemide and acetylsalicylic acid temporarily reduced
creatinine clearance in patients with chronic renal insufficiency.
There are case reports of patients who developed increased
BUN, serum creatinine and serum potassium levels, and weight
gain when furosemide was used in conjunction with NSAIDs.
Literature reports indicate that coadministration of
indomethacin may reduce the natriuretic and antihypertensive
effects of furosemide in some patients by inhibiting
prostaglandin synthesis. Indomethacin may also affect plasma
renin levels, aldosterone excretion, and renin profile evaluation.
Patients receiving both indomethacin and furosemide should be
observed closely to determine if the desired diuretic and/or
antihypertensive effect of furosemide is achieved.
ADVERSE REACTIONS:
Adverse reactions are categorized below by organ system and
listed by decreasing severity:
Gastrointestinal System Reactions

1. Pancreatitis.
2. Jaundice (intrahepatic cholestatic juandice).
3. Anorexia.
4. Oral and gastric irritation.
5. Cramping.
6. Diarrhea.
7. Constipation.
8. Nausea.
9. Vomiting.
Systemic Hypersensitivity Reactions

1. Systemic vasculitis.
2. Interstitial nephritis.
3. Necrotizing angiitis.
Central Nervous System Reactions

1. Tinnitus and hearing loss.
2. Paresthesias.
3. Vertigo.
4. Dizziness.
5. Headache.
6. Blurred vision.
7. Xanthopsia.
Hematologic Reactions
1. Aplastic anemia (rare).
2. Thrombocytopenia.
3. Agranulocytosis (rare).
4. Hemolytic anemia.
5. Leukopenia.
6. Anemia.

Dermatologic-Hypersensitivity Reactions
1. Exfoliative dermatitis.
2. Erythema multiforme.
3. Purpura.
4. Photosensitivity.
5. Urticaria.
6. Rash.
7. Pruritus.
Cardiovascular Reaction: Orthostatic hypotension may occur

and be aggravated by alcohol, barbiturates or narcotics.
Other Reactions
1. Hyperglycemia.
2. Glycosuria.
3. Hyperuricemia.
4. Muscle spasm.
5. Weaknesses.
6. Restlessness.
7. Urinary bladder spasm.
8. Thrombophlebitis.
9. Fever.

Whenever adverse reactions are moderate or severe, furosemide

dosage should be reduced or therapy withdrawn.
OVERDOSAGE:
The principal signs and symptoms of overdose with furosemide
are dehydration, blood volume reduction, hypotension,
electrolyte imbalance, hypokalemia and hypochloremic
alkalosis, and are extensions of its diuretic action.
The acute toxicity of furosemide has been determined in mice,
rats and dogs. In all three, the oral LD50 exceeded 1000 mg/kg
body weight while the intravenous LD50 ranged from 300 to
680 mg/kg. The acute intragastric toxicity in neonatal rats is 7
to 10 times that of adult rats.
The concentration of furosemide in biological fluids associated
with toxicity or death is not known.
Treatment of overdosage is supportive and consists of
replacement of excessive fluid and electrolyte losses. Serum
electrolytes, carbon dioxide level and blood pressure should be
determined frequently. Adequate drainage must be assured in
patients with urinary bladder outlet obstruction (such as
prostatic hypertrophy).
Hemodialysis does not accelerate furosemide elimination.
DOSAGE AND ADMINISTRATION:

Tablets and Oral Solution
Edema
Therapy should be individualized according to patient response
to gain maximal therapeutic response and to determine the
minimal dose needed to maintain that response.
Adults: The usual initial dose of furosemide is 20 to 80 mg
given as a single dose. Ordinarily a prompt diuresis ensues. If
needed, the same dose can be administered 6 to 8 hours later or
the dose may be increased. The dose may be raised by 20 or 40
mg and given not sooner than 6 to 8 hours after the previous
dose until the desired diuretic effect has been obtained. This
individually determined single dose should then be given once
or twice daily (e.g., at 8 am and 2 pm). The dose of furosemide
may be carefully titrated up to 600 mg/day in patients with
clinically severe edematous states.
Edema may be most efficiently and safely mobilized by giving

furosemide on 2 to 4 consecutive days each week.

When doses exceeding 80 mg/day are given for prolonged
periods, careful clinical observation and laboratory monitoring
are particularly advisable. (See PRECAUTIONS, Laboratory
Tests.)
Pediatric Patients: The usual initial dose of oral furosemide in

pediatric patients is 2 mg/kg body weight, given as a single
dose. If the diuretic response is not satisfactory after the initial
dose, dosage may be increased by 1 or 2 mg/kg no sooner than
6 to 8 hours after the previous dose. Doses greater than 6 mg/kg
body weight are not recommended. For maintenance therapy in
pediatric patients, the dose should be adjusted to the minimum
effective level.
Hypertension
Therapy should be individualized according to the patient's
response to gain maximal therapeutic response and to determine
the minimal dose needed to maintain the therapeutic response.
Adults: The usual initial dose of furosemide for hypertension is
80 mg, usually divided into 40 mg twice a day. Dosage should
then be adjusted according to response. If response is not
satisfactory, add other antihypertensive agents.
Changes in blood pressure must be carefully monitored when
furosemide is used with other antihypertensive drugs, especially
during initial therapy. To prevent excessive drop in blood
pressure, the dosage of other agents should be reduced by at
least 50 percent when furosemide is added to the regimen. As
the blood pressure falls under the potentiating effect of
furosemide, a further reduction in dosage or even
discontinuation of other antihypertensive drugs may be
necessary.
Injection
Adults: Parenteral therapy with furosemide Injection should be
used only in patients unable to take oral medication or in
emergency situations and should be replaced with oral therapy
as soon as practical.
Edema: The usual initial dose of furosemide is 20 to 40 mg
given as a single dose, injected intramuscularly or
intravenously. The intravenous dose should be given slowly (1
to 2 minutes). Ordinarily a prompt diuresis ensues. If needed,

another dose may be administered in the same manner 2 hours

later or the dose may be increased. The dose may be raised by
20 mg and given not sooner than 2 hours after the previous dose
until the desired diuretic effect has been obtained. This
individually determined single dose should then be given once
or twice daily.
Therapy should be individualized according to patient response
to gain maximal therapeutic response to determine the minimal
dose needed to maintain that response. Close medical
supervision is necessary.
If the physician elects to use high dose parenteral therapy, add
the furosemide to either sodium chloride injection, lactated
ringer's injection, or dextrose (5%) injection after pH has been
adjusted to above 5.5, and administer as a controlled
intravenous infusion at a rate not greater than 4 mg/min.
Furosemide Injection is a buffered alkaline solution with a pH
of about 9 and the drug may precipitate at pH values below 7.
Care must be taken to ensure that the pH of the prepared
infusion solution is in the weakly alkaline to neutral range. Acid
solutions, including other parenteral medications (e.g., labetalol,
ciprofloxacin, amrinone, and milrinone) must not be
administered concurrently in the same infusion because they
may cause precipitation of the furosemide. In addition,
furosemide injection should not be added to a running
intravenous line containing any of these acidic products.
Acute Pulmonary Edema: The usual initial dose of furosemide
is 40 mg injected slowly intravenously (over 1 to 2 minutes). If
a satisfactory response does not occur within 1 hour, the dose
may be increased to 80 mg injected slowly intravenously (over
1 to 2 minutes).
If necessary, additional therapy (e.g., digitalis, oxygen) may be
administered concomitantly.
Pediatric Patients: Parenteral therapy should be used only in
patients unable to take oral medication or in emergency
situations and should be replaced with oral therapy as soon as
practical.
The usual initial dose of furosemide Injection (intravenously or
intramuscularly) in pediatric patients is 1 mg/kg body weight
and should be given slowly under close medical supervision. If
the diuretic response to the initial dose is not satisfactory,
dosage may be increased by 1 mg/kg not sooner than 2 hours
after the previous dose, until the desired diuretic effect has been

obtained. Doses greater than 6 mg/kg body weight are not

recommended.
Furosemide Injection should be inspected visually for
particulate matter and discoloration before administration. Do
not use if solution is discolored.
To insure patient safety, this needle should be handled with care
and should be destroyed and discarded if damaged in any
manner. If cannula is bent, no attempt should be made to
straighten.
To prevent needle-stick injuries, needles should not be
recapped, purposely bent, or broken by hand.
PATIENT INFORMATION:
Furosemide is a diuretic (water pill) used to treat fluid retention
and high blood pressure. Do not take this medication if you are
allergic to sulfa medicine. Notify your physician if you are
pregnant or nursing. Notify your physician if you have diabetes
mellitus. Blood glucose levels may be increased in patients with
diabetes mellitus. Take this medication early in the day.
Furosemide may be taken with or without food. Take with food
or milk if stomach upset occurs. Notify your physician if you
develop weakness, cramps, or nausea. Dizziness or
lightheadedness may occur with therapy; avoid sudden changes
in posture. Furosemide may cause increased sensitivity to
sunlight. Use sunscreens and wear protective clothing until
degree of sensitivity is determined.
HOW SUPPLIED:
Tablets
Dispense in well-closed, light-resistant containers. Exposure to
light might cause a slight discoloration. Discolored tablets
should not be dispensed.
20 mg Tablets: Lasix tablets 20 mg are supplied as white, oval,
monogrammed tablets. They are imprinted with "Lasix" on one
side and "Hoechst" on the other.
40 mg Tablets: Lasix tablets 40 mg are supplied as white,
round, monogrammed, scored tablets. They are imprinted with
"Lasix" on one side and the Hoechst logo on the other.

80 mg Tablets: Lasix tablets 80 mg are supplied as white,
round, monogrammed, facetted edge tablets. They are imprinted
with "Lasix 80" on one side and the Hoechst logo on the other.
Oral Solution
Store at controlled room temperature (59-86 F). Dispense in
light-resistant containers. Discard opened bottle after 60 days.
Injection
Storage: Store at controlled room temperature (59-86 F).
Do not use if solution is discolored.
Protect syringes from light. Do not remove syringe from
individual package until time of use.

Bookmark URL: /das/drug/view/1/1342/top

About GenRx
Credits
Editorial Review Panel
http://home.mdconsult.com/das/drug/toc?node=8770644 [22/04/2000 0:29:33]

Key Information
FDA Pregnancy Categories
DEA Schedules of Controlled Substances
Childhood Immunizations
Summary of Recommendations for Adult Immunizations
Poison Control Centers of the United States and Canada
Directory of AIDS Drug Assistance Programs
Key Addresses and Phone Numbers
Oral Solid Dosage Forms That Should Not Be Crushed
Drug Names That Look Alike And Sound Alike
Discontinued Brands
Manufacturers Listed by NDC Number
Disclaimers
Comparative Tables

Drug Identification Guide/Pill Images
1 -- 9

Generic Drugs
Ab -- Ad
Al
Am
An -- Ar
As -- Az
BC -- Be
Bi -- Bo
Br -- Bu
Ca
Ce
Ch
Ci -- Cl
Co -- Cy
D -- De
Di -- Dy
Ec -- Ep
Er -- Eu
Fa -- Fi
Fl -- Fu
Ga -- Gr
Gu
Ha -- Ho
Hy
Ib -- In
Io -- Iv
LC -- Le
http://home.mdconsult.com/das/drug/toc?node=8770694 (1 of 2) [22/04/2000 0:30:28]

Li -- Ly
MC -- Me
Mi -- My
Na -- Ne
Ni -- Ny
Pa -- Pe
Ph -- Pi
Pl -- Po
Pr -- Py
Sa -- Se
Si -- So
Sp -- Su
Ta -- Te
Th -- Ti
To
Tr -- Ty
http://home.mdconsult.com/das/drug/toc?node=8770694 (2 of 2) [22/04/2000 0:30:28]

April 22, 2000
Conde Petra

About GenRx
Key Information
Drug Identification Guide/Pill Images Spironolactone (002269)

Generic Drugs
CATEGORIES:
US Drugs
Global Drugs Indications: Adenoma, adrenal, adjunct; Edema, secondary to

heart failure; Edema, secondary to hepatic cirrhosis; Edema,
Indications secondary to nephrotic syndrome; Hyperaldosteronism,
FDA-Class primary; Hyperplasia, adrenal; Hypertension; Hypokalemia
WHO Formulary
Category
FDA Approved 1983 Dec
Ingredient FDA DRUG CLASS: Antihypertensives; Diuretics; Repl/Regs Of
Comprehensive Index Electrolytes
Supplier Information BRAND NAMES: Adultmin *; Aldactone (US); Aldopur *;

Aldospirone *; Almatol *; Berlactone *; Diatensec *; Diram *;
Esekon *; Hypazon *; Idrolattone *; Merabis *; Novospiroton *;
Osiren *; Osyrol *; Pirolacton *; Resacton *; Sincomen *; Spiractin
*; Spiroctan *; Spirolacton *; Spirolang *; Spironex *; Spirotone *;
Tensin *; Tevaspirone *; Verospiron *; Xenalon Lactabs *;
Youlactone *;
(International brand names outside U.S. in italics)
* Indicates foreign drugs without region specification at time of publication.
COST OF THERAPY: $ 42.34 (Hypertension; Tablet; 25 mg;

2/day; 365 days)
WARNING:
Spironolactone has been shown to be a tumorigen in chronic
toxicity studies in rats (see PRECAUTIONS). Spironolactone
should be used only in those conditions described in
INDICATIONS AND USAGE. Unnecessary use of this drug
should be avoided.

DESCRIPTION:
Spironolactone oral tablets contain 25 mg, 50 mg, or 100 mg of the
aldosterone antagonist spironolactone, 17-hydroxy-7alpha
-mercapto-3-oxo-17alpha -pregn-4-ene -21-carboxylic acid gamma
-lactone acetate.
Spironolactone is practically insoluble in water, soluble in alcohol,
and freely soluble in benzene and in chloroform.
Inactive ingredients include calcium sulfate, corn starch, flavor,
hydroxypropyl methylcellulose, iron oxide, magnesium stearate,
polyethylene glycol, povidone, and titanium dioxide.
CLINICAL PHARMACOLOGY:
Mechanism of Action: Spironolactone is a specific pharmacologic
antagonist of aldosterone, acting primarily through competitive
binding of receptors at the aldosterone-dependent sodium-potassium
exchange site in the distal convoluted renal tubule. Spironolactone
causes increased amounts of sodium and water to be excreted, while
potassium is retained. Spironolactone acts both as a diuretic and as
an antihypertensive drug by this mechanism. It may be given alone
or with other diuretic agents which act more proximally in the renal
tubule.
Aldosterone Antagonist Activity: Increased levels of the
mineralocorticoid, aldosterone, are present in primary and secondary
hyperaldosteronism. Edematous states in which secondary
aldosteronism is usually involved include congestive heart failure,
hepatic cirrhosis, and the nephrotic syndrome. By competing with
aldosterone for receptor sites, spironolactone provides effective
therapy for the edema and ascites in those conditions. Spironolactone
counteracts secondary aldosteronism induced by the volume
depletion and associated sodium loss caused by active diuretic
therapy.
Spironolactone is effective in lowering the systolic and diastolic
blood pressure in patients with primary hyperaldosteronism. It is also
effective in most cases of essential hypertension, despite the fact that
aldosterone secretion may be within normal limits in benign essential
hypertension.
Through its action in antagonizing the effect of aldosterone,
spironolactone inhibits the exchange of sodium for potassium in the
distal renal tubule and helps to prevent potassium loss.
Spironolactone has not been demonstrated to elevate serum uric acid,
to precipitate gout, or to alter carbohydrate metabolism.
Pharmacokinetics: Spironolactone is rapidly and extensively

metabolized. Sulfur-containing products are the predominant

metabolites and are thought to be primarily responsible, together
with spironolactone, for the therapeutic effects of the drug. The
following pharmacokinetic data found in TABLE 1 were obtained
from 12 healthy volunteers following administration of 100 mg of
spironolactone film-coated tablets daily for 15 days. On the 15th
day, spironolactone was given immediately after a low-fat breakfast
and blood was drawn thereafter.
TABLE 1
Accumulation
Factor: AUC
(0-24 hr, day Mean (SD)
15)/AUC Mean Peak Post-Steady
(0-24 hr, day Serum State
1) Concentration Half-Life
7-alpha 1.25 391 ng/ml at 13.8 hr (6.4)
-(thiomethyl) 3.2 hr (terminal)
spirolactone
(TMS)
6-beta 1.50 125 ng/ml at 15.0 hr (4.0)
-hydroxy-7-alpha 5.1 hr (terminal)
-(thiomethyl)
spirolactone
(HTMS)
Canrenone (C) 1.41 181 ng/ml at 16.5 hr (6.3)
4.3 hr (terminal)
Spironolactone 1.30 80 ng/ml at 2.6 Approximately
hr 1.4 hr (0.5)
(beta half-life)
The pharmacological activity of spironolactone metabolites in man is

not known. However, in the adrenalectomized rat the
antimineralocorticoid activities of the metabolites C, TMS, and
HTMS, relative to spironolactone, were 1.10, 1.28, and 0.32,
respectively. Relative to spironolactone, their binding affinities to
the aldosterone receptors in rat kidney slices were 0.19, 0.86, and
0.06, respectively.
In humans the potencies of TMS and 7-alpha -thiospirolactone in
reversing the effects of the synthetic mineralocorticoid,
fludrocortisone, on urinary electrolyte composition were 0.33 and
0.26, respectively, relative to spironolactone. However, since the
serum concentrations of these steroids were not determined, their

incomplete absorption and/or first-pass metabolism could not be

ruled out as a reason for their reduced in vivo activities.
Spironolactone and its metbolites are more than 90% bound to
plasma proteins. The metabolites are excreted primarily in the urine
and secondarily in bile.
The effect food on spironolactone absorption (two-100 mg
spironolactone tablets) was assessed in a single dose study of 9
healthy, drug-free volunteers. Food increased the bioavailability of
unmetabolized spironolactone by almost 100%. The clinical
importance of this finding is not known.
INDICATIONS AND USAGE:

Spironolactone is indicated in the management of:
Primary Hyperaldosteronism For: Establishing the diagnosis of

primary hyperaldosteronism by therapeutic trial.
Short-term preoperative treatment of patients with primary
hyperaldosteronism.
Long-term maintenance therapy for patients with discrete
aldosterone-producing adrenal adenomas who are judged to be
poor operative risks or who decline surgery.
Long-term maintenance therapy for patients with bilateral
micro- or macronodular adrenal hyperplasia (idiopathic
hyperaldosteronism).
Edematous conditions for patients with:

Congestive Heart Failure: For the management of edema
and sodium retention when the patient is only partially
responsive to, or is intolerant of, other therapeutic measures.
Spironolactone is also indicated for patients with congestive
heart failure taking digitalis when other therapies are
considered inappropriate.
Cirrhosis of the Liver Accompanied by Edema and/or
Ascites: Spironolactone levels may be exceptionally high in
this condition. Spironolactone is indicated for maintenance
therapy together with bed rest and the restriction of fluid and
sodium.
The Nephrotic Syndrome: For nephrotic patients when
treatment of the underlying disease, restriction of fluid and
sodium intake, and the use of other diuretics do not provide on
adequate response.

Essential Hypertension: Usually in combination with other drugs,

spironolactone is indicated for patients who cannot be treated
adequately with other agents or for whom other agents are
Hypokalemia: For the treatment of patients with hypokalemia when
other measures are considered inappropriate or inadequate.
Spironolactone is also indicated for the prophylaxis of hypokalemia
in patients taking digitalis when other measures are considered
inadequate or inappropriate.
Usage in Pregnancy: The routine uses of diuretics in an otherwise
healthy woman is inappropriate and exposes mother and fetus to
unnecessary hazard. Diuretics do not prevent development of
toxemia of pregnancy, and there is no satisfactory evidence that they
are useful in the treatment of developing toxemia.
Edema during pregnancy may arise from pathologic causes or from
the physiologic and mechanical consequences of pregnancy.
Spironolactone is indicated in pregnancy when edema is due to
pathologic causes just as it is in the absence of pregnancy (however,
see PRECAUTIONS). Dependent edema in pregnancy, resulting
from restriction of venous return by the expanded uterus, is properly
treated through elevation of the lower extremities and use of support
hose; use of diuretics to lower intravascular volume in this case is
unsupported and unnecessary. There is hypervolemia during normal
pregnancy which is not harmful to either the fetus or the mother (in
the absence of cardiovascular disease), but which is associated with
edema, including generalized edema, in the majority of pregnant
women. If this edema produces discomfort, increased recumbency
will often provide relief. In rare instances, this edema may cause
extreme discomfort which is not relieved by rest. In these cases, a
short course of diuretics may provide relief and may be appropriate.
CONTRAINDICATIONS:
Spironolactone is contraindicated for patients with anuria, acute
renal insufficiency, significant impairment of renal excretory
function, or hyperkalemia.
WARNINGS:
Potassium supplementation, either in the form of medication or as a
diet rich in potassium, should not ordinarily be given in association
with spironolactone therapy. Excessive potassium intake may cause
hyperkalemia in patients receiving spironolactone (see
PRECAUTIONS, General). Spironolactone should not be
administered concurrently with other potassium-sparing diuretics.
Spironolactone, when used with ACE inhibitors or indomethacin,

even in the presence of a diuretic, has been associatd with severe

hyperkalemia. Extreme caution should be exercised when
spironolactone is given concomitantly with these drugs.
Spironolactone should be used with caution in patients with impaired
hepatic function because minor alterations of fluid and electrolyte
balance may precipitate hepatic coma.
Lithium generally should not be given with diuretics (see DRUG
INTERACTIONS).
PRECAUTIONS:
General
All patients receiving diuretic therapy should be observed for
evidence of fluid or electrolyte imbalance, e.g., hypomagnesemia,
hyponatremia, hypochloremic alkalosis, and hyperkalemia.
Serum and urine electrolyte determinations are particularly important
when the patient is vomiting excessively or receiving parenteral
fluids. Warning signs or symptoms of fluid and electrolyte
imbalance, irrespective of cause, include dryness of the mouth, thirst,
weakness, lethargy, drowsiness, restlessness, muscle pains or
cramps, muscular fatigue, hypotension, oliguria, tachycardia, and
gastrointestinal disturbances such as nausea and vomiting.
Hyperkalemia may occur in patients with impaired renal function or
excessive potassium intake and can cause cardiac irregularities,
which may be fatal. Consequently, no potassium supplement should
ordinarily be given with spironolactone.
Concomitant administration of potassium-sparing diuretics and ACE
inhibitors or nonsteroidal anti-inflammatory drugs (NSAIDs), e.g.,
indomethacin, has been associated with severe hyperkalemia.
If hyperkalemia is suspected (warning signs include paresthesia,
muscle weakness, fatigue, flaccid paralysis of the extremities,
bradycardia and shock) an electrocardiogram (ECG) should be
obtained. However, it is important to monitor serum potassium levels
because mild hyperkalemia may not be associated with ECG
changes.
If hyperkalemia is present, spironolactone should be discontinued
immediately. With severe hyperkalemia, the clinical situation
dictates the procedures to be employed. These include the
intravenous administration of calcium chloride solution, sodium
bicarbonate solution and/or the oral or parenteral administration of
glucose with a rapid-acting insulin preparation. These are temporary
measures to be repeated as required. Cationic exchange resins such
as sodium polystyrene sulfonate may be orally or rectally
administered. Persistent hyperkalemia may require dialysis.

Reversible hyperchloremic metabolic acidosis, usually in association

with hyperkalemia, has been reported to occur in some patients with
decompensated hepatic cirrhosis, even in the presence of normal
renal function.
Dilutional hyponatremia, manifested by dryness of the mouth, thirst,
lethargy, and drowsiness, and confirmed by a low serum sodium
level, may be caused or aggravated, especially when spironolactone
is administered in combination with other diuretics, and dilutional
hyponatremia may occur in edematous patients in hot weather;
appropriate therapy is water restriction rather than administration of
sodium, except in rare instances when the hyponatremia is
life-threatening.
Spironolactone therapy may cause a transient elevation of BUN,
especially in patients with preexisting renal impairment.
Spironolactone may cause mild acidosis.
Gynecomastia may develop in association with the use of
spironolactone; physicians should be alert to its possible onset. The
development of gynecomastia appears to be related to both dosage
level and duration of therapy and is normally reversible when
spironolactone is discontinued. In rare instances some breast
enlargement may persist when spironolactone is discontinued.
Information for the Patient

Patients who receive spironolactone should be advised to avoid
potassium supplements and foods containing high levels of
potassium including salt substitutes.
Laboratory Tests
Periodic determination of serum electrolytes to detect possible
electrolyte imbalance should be done at appropriate intervals,
particularly in the elderly and those with significant renal or hepatic
impairments.
Drug/Laboratory Test Interactions

Several reports of possible interference with digoxin
radioimmunoassays by spironolactone, or its metabolites, have
appeared in the literature. Neither the extent nor the potential clinical
significance of its interference (which may be assay-specific) has
been fully established.
Carcinogenesis, Mutagenesis, and Impairment of Fertility

Carcinogenesis, mutagenesis, impairment of fertility: Orally
administered spironolactone has been shown to be a tumorigen in

dietary administration studies performed in rats, with its proliferative

effects manifested on endocrine organs and the liver. In an 18-month
study using doses of about 50, 150 and 500 mg/kg/day, there were
statistically significant increases in benign adenomas of the thyroid
and testes and, in male rats, a dose-related increase in proliferative
changes in the liver (including hepatocytomegaly and hyperplastic
nodules). In a 24-month study in which the same strain of rat was
administered doses of about 10, 30, 100 and 150 mg
spironolactone/kg/day, the range of proliferative effects included
significant increases in hepatocellular adenomas and testicular
interstitial cell tumors in males, and significant increases in thyroid
follicular cell adenomas and carcinomas in both sexes. There was
also a statistically significant, but not dose-related, increase in
benign uterine endometrial stromal polyps in females. A dose-related
(above 20 mg/kg/day) incidence of myelocytic leukemia was
observed in rats fed daily doses of potassium canrenoate (a
compound chemically similar to spironolactone and whose primary
metabolite, canrenone, is also a major product of spironolactone in
man) for a period of one year. In two year studies in the rat, oral
administration of potassium canrenoate was associated with
myelocytic leukemia and hepatic, thyroid, testicular and mammary
tumors.
Neither spironolactone nor potassium canrenoate produced
mutagenic effects in tests using bacteria or yeast. In the absence of
metabolic activation, neither spironolactone nor potassium
canrenoate has been shown to be mutagenic in mammalian tests in
vitro . In the presence of metabolic activation, spironolactone has
been reported to be negative in some mammalian mutagenicity tests
in vitro and inconclusive (but slightly positive) for mutagenicity in
other mammalian tests in vitro. In the presence of metabolic
activation, potassium canrenoate has been reported to test positive
for mutagenicity in some mammalian tests in vitro, inconclusive in
others, and negative in still others.
In a three-litter reproduction study in which female rats received
dietary doses of 15 and 50 mg spironolactone/kg/day, there were no
effects on mating and fertility, but there was a small increase in
incidence of stillborn pups at 50 mg/kg/day. When injected into
female rats (100 mg/kg/day for 7 days, i.p.), spironolactone was
found to increase the length of the estrous cycle by prolonging
diestrus during treatment and inducing constant diestrus during a two
week posttreatment observation period. These effects were
associated with retarded ovarian follicle development and a
reduction in circulating estrogen levels, which would be expected to
impair mating, fertility and fecundity. Spironolactone (100
mg/kg/day), administered i.p. to female mice during a two week
cohabitation period with untreated males, decreased the number of
mated mice that conceived (effect shown to be caused by an

inhibition of ovulation) and decreased the number of implanted

embryos in those that became pregnant (effect shown to be caused
by an inhibition of implantation), and at 200 mg/kg, also increased
the latency period to mating.
Pregnancy, Teratogenic Effects, Pregnancy Category C

Teratology studies with spironolactone have been carried out in mice
and rabbits at doses of up to 20 mg/kg/day. On a body surface area
basis, this dose in the mouse is substantially below the maximum
recommended human dose and, in the rabbit, approximates the
maximum recommended human dose. No teratogenic or other
embryotoxic effects were observed in mice, but the 20 mg/kg dose
caused an increased rate of resorption and a lower number of live
fetuses in rabbits. Because of its anti-androgenic activity and the
requirement of testosterone for male morphogenesis, spironolactone
may have the potential for adversely affecting sex differentiation of
the male during embryogenesis. When administered to rats at 200
mg/kg/day between gestation days 13 and 21 (late embryogenesis
and fetal development), feminization of male fetuses was observed.
Offspring exposed during late pregnancy to 50 and 100 mg/kg/day
doses of spironolactone exhibited changes in the reproductive tract
including dose-dependent decreases in weights of the ventral
prostate and seminal vesicle in males, ovaries and uteri that were
enlarged in females, and other indications of endocrine dysfunction,
that persisted into adulthood. There are no adequate and
well-controlled studies with spironolactone in pregnant women.
Spironolactone has known endocrine effects in animals including
progestational and antiandrogenic effects. The antiandrogenic effects
can result in apparent estrogenic side effects in humans, such as
gynecomastia. Therefore, the use of spironolactone in pregnant
women requires that the anticipated benefit be weighed against the
possible hazards to the fetus.
Nursing Mothers
Canrenone, a major (and active) metabolite of spironolactone,
appears in human breast milk. Because spironolactone has been
found to be tumorigenic in rats, a decision should be made whether
to discontinue the drug, taking into account the importance of the
drug to the mother. If use of the drug is deemed essential, an
alternative method of infant feeding should be instituted.
Pediatric Use
Safety and effectiveness in pediatric patients have not been
established.

DRUG INTERACTIONS:
ACE Inhibitors: Concomitant administration of ACE inhibitors
with potassium-sparing diuretics has been associated with severe
hyperkalemia.
Alcohol, Barbiturates, or Narcotics: Potentiation of orthostatic
hypotension may occur.
Corticosteroids, ACTH: Intensified electrolyte depletion,
particularly hypokalemia, may occur.
Pressor Amines (e.g., Norepinephrine): Spironolactone reduces the
vascular responsiveness to norepinephrine. Therefore, caution should
be exercised in the management of patients subjected to regional or
general anesthesia while they are being treated with spironolactone.
Skeletal Muscle Relaxants, Nondepolarizing (e.g.,
Tubocurarine): Possible increased responsiveness to the muscle
relaxant may result.
Lithium: Lithium generally should not be given with diuretics.
Diuretic agents reduce the renal clearance of lithium and add a high
risk of lithium toxicity.
Nonsteroidal Anti-inflammatory Drugs (NSAIDs): In some
patients, the administration of an NSAID can reduce the diuretic,
natriuretic, and antihypertensive effect of loop, potassium-sparing
and thiazide diuretics. Combination of NSAIDs, (e.g., indomethacin,
with potassium-sparing diuretics) has been associated with severe
hyperkalemia. Therefore, when spironolactone and NSAIDs are used
concomitantly, the patient should be observed closely to determine if
the desired effect of the diuretic is obtained.
Digoxin: Spironolactone has been shown to increase the half-life of
digoxin. This may result in increased serum digoxin levels and
subsequent digitalis toxicity. It may be necessary to reduce the
maintenance and digitalization doses when spironolactone is
administered, and the patient should be carefully monitored to avoid
over- or underdigitalization.
ADVERSE REACTIONS:
The following adverse reactions have been reported and, within each
category (body system), are listed in order of decreasing severity.
Digestive: Gastric bleeding, ulceration, gastritis, diarrhea and

cramping, nausea, vomiting.
Endocrine: Gynecomastia (see PRECAUTIONS), inability to
achieve or maintain erection, irregular menses or amenorrhea,
postmenopausal bleeding. Carcinoma of the breast has been reported
in patients taking spironolactone but a cause and effect relationship
has not been established.
Hematologic: Agranulocytosis.

Hypersensitivity: Fever, urticaria, maculopapular or erythematous

cutaneous eruptions, anaphylactic reactions, vasculitis.
Nervous System/Psychiatric: Mental confusion, ataxia, headache,
drowsiness, lethargy.
Liver/Biliary: A very few cases of mixed cholestatic/hepatocellular
toxicity, with one reported fatality, have been reported with
spironolactone administration.
OVERDOSAGE:
The oral LD50 of spironolactone is greater than 1000 mg/kg in mice,
rats, and rabbits.
Acute overdosage of spironolactone may be manifested by
drowsiness, mental confusion, maculopapular or erythematous rash,
nausea, vomiting, dizziness, or diarrhea. Rarely, instances of
hyponatremia, hyperkalemia, or hepatic coma may occur in patients
with severe liver disease, but these are unlikely due to acute
overdosage. Hyperkalemia may occur, especially in patients with
impaired renal function.
Treatment: Induce vomiting or evacuate the stomach by lavage.
There is no specific antidote. Treatment is supportive to maintain
hydration, electrolyte balance, and vital functions.
Patients who have renal impairment may develop
spironolactone-induced hyperkalemia. In such cases, spironolactone
should be discontinued immediately. With severe hyperkalemia, the
clinical situation dictates the procedures to be employed. These
include the intravenous administration of calcium chloride solution,
sodium bicarbonate solution and/or the oral or parenteral
administration of glucose with a rapid-acting insulin preparation.
These are temporary measures to be repeated as required. Cationic
exchange resins such as sodium polystyrene sulfonate may be orally
or rectally administered. Persistent hyperkalemia may require
dialysis.
DOSAGE AND ADMINISTRATION:

Primary Hyperaldosteronism Spironolactone may be employed as
an initial diagnostic measure to provide presumptive evidence of
primary hyperaldosteronism while patients are on normal diets.
Long Test: Spironolactone is administered at a daily dosage of 400
mg for three to four weeks. Correction of hyperkalemia and of
hypertension provides presumptive evidence for the diagnosis of
primary hyperaldosteronism.

Short Test: Spironolactone is administered at a daily dosage of 400

mg for four days. If serum potassium increases during
spironolactone administration but drops when spironolactone is
discontinued, a presumptive diagnosis of primary
hyperaldosteronism should be considered.
After the diagnosis of hyperaldosteronism has been established by
more definitive testing procedures, spironolactone may be
administered in doses of 100 to 400 mg daily in preparation for
surgery. For patients who are considered unsuitable for surgery,
spironolactone may be employed for long-term maintenance therapy
at the lowest effective dosage determined for the individual patient.
Edema in Adults: Congestive Heart Failure, Hepatic Cirrhosis or
Nephrotic Syndrome: An initial daily dosage of 100 mg of
spironolactone administered in either single or divided doses is
recommended, but may range from 25 to 200 mg daily. When given
as the sole agent for diuresis, spironolactone should be continued for
at least five days at the initial dosage level, after which it may be
adjusted to the optimal therapeutic or maintenance level
administered in either single or divided daily doses. If, after five
days, an adequate diuretic response to spironolactone has not
occurred, a second diuretic which acts more proximally in the renal
tubule may be added to the regimen. Because of the additive effect
of spironolactone when administered concurrently with such
diuretics, an enhanced diuresis usually begins on the first day of
combined treatment; combined therapy is indicated when more rapid
diuresis is desired. The dosage of spironolactone should remain
unchanged when other diuretic therapy is added.
Essential Hypertension: For adults, an initial daily dosage of 50 to
100 mg of spironolactone administered in either single or divided
doses is recommended. Spironolactone may also be given with
diuretics which act more proximally in the renal tubule or with other
antihypertensive agents. Treatment with spironolactone should be
continued for at least two weeks, since the maximum response may
not occur before this time. Subsequently, dosage should be adjusted
according to the response of the patient.
Hypokalemia: Spironolactone in a dosage ranging from 25 to 100
mg daily is useful in treating a diuretic-induced hypokalemia, when
oral potassium supplements or other potassium-sparing regimens are
HOW SUPPLIED:
Aldactone 25-mg Tablets: Round, light yellow, film coated, with

SEARLE and 1001 debossed on one side and ALDACTONE and 25

on the other side.

Aldactone 50-mg Tablets: Oval, light orange, scored, film coated,
with SEARLE and 1041 debossed on the scored side and
ALDACTONE and 50 on the other side.
Aldactone 100-mg Tablets: Round, peach colored, scored, film
coated, with SEARLE and 1031 debossed on the scored side and
ALDACTONE and 100 on the other side.
Storage: Store below 77F (25C).

Bookmark URL: /das/drug/view/1/2269/top

April 22, 2000
Conde Petra
April 14, 2000

SOB


of cirrhosis.
knowledge.




Bibliography
110(4):253-255, 1989(editorial)


Saunders







70(5):1081-1091, 1986


8(1):85-99, 1986








WB Saunders



April 22, 2000
Conde Petra
April 14, 2000
Question 5 - Hepatic Encephalopathy

What is the likely diagnosis leading to such a mental status change? What is
another complication that may lead to renal failure?
One of the most difficult complications of cirrhosis is hepatic encephalopathy. This

results from the portosystemic shunting of nitrogenous amines that can be
neurotoxic. Precipitating factors include azotemia, use of sedatives/tranquilizers,
gastrointestinal bleeding, infection, hypokalemic alkalosis, or excess protein intake.
The symptoms and signs of this disease include (1) asterixis, a flapping motion
seen when the patient is asked to hold his arms horizontally with the hands
extended at the wrists, (2) fetor hepaticus, a feculent-fruity odor of the breath, and
(3) mental status changes that vary from mood alterations to lethargy, confusion,
stupor, and finally coma. Ammonia levels may be elevated but do not correlate with
the stage of disease. Hepatic encephalopathy can be classified in four categories:
(1) agitation, (2) lethargy with asterixis, inappropriate behavior, and slurred speech,
(3) stupor with hyperactive reflexes and nystagmus, and (4) deep coma with dilated
pupils and opisthotonos.
Management of hepatic encephalopathy should emphasize correction of the

precipitating factors. Other therapeutic measures should include temporary dietary
protein restriction and use of lactulose or neomycin. Protein intake should be
limited to 1 g/kg. Lactulose works by reducing intestine intraluminal pH and
promoting diarrhea. The acidic environment created promotes protonation of
ammonia to NH4+, which is not well absorbed and is excreted in stool. Neomycin is
an aminoglycoside antibiotic that decreases urease-producing bacteria and is as
effective as lactulose. The use of neomycin is limited by attendant side effects of
ototoxicity and nephrotoxicity.
The hepatorenal syndrome is another complication of cirrhosis. This is a potentially

fatal complication and is defined by the presence of renal failure in combination
with cirrhosis. This renal failure is characterized by decrease in GFR, azotemia,
and oliguria. The exact pathogenesis is unknown, but precipitating factors include
vigorous diuretic use, large-volume paracentesis without volume repletion and
sepsis. These conditions all lead to decreased renal blood flow and GFR, both of
which might precede overt renal failure. It is important to rule out prerenal azotemia
before settling on the diagnosis of the hepatorenal syndrome.
Other complications of cirrhosis include gallstones, which form secondary to an

increase in bilirubin from hemolytic anemia and hypersplenism. Peptic ulcer
disease occurs with greater incidence in patients with cirrhosis. Hypoxia develops
secondary to two possible mechanisms. Ascitic fluid may interfere with
diaphragmatic movement, leading to the sensation of dyspnea, and retarding
adequate ventilation. The second mechanism is known as the hepatopulmonary
syndrome. This is due to the formation of abnormal arteriovenous circuits in the

lung, wherein perfusion is position dependent. Changes of position lead to

alterations of ventilation and perfusion leading to hypoxemeia. Primary liver cell
cancer is often seen in cirrhotic livers and is associated with hepatitis B and C and
less commonly with cirrhosis due to alcohol.
Another manifestation of portal hypertension is splenomegaly, which results in

thrombocytopenia and leukopenia. The decrease in platelets is clinically significant
because of the usual concomitant coagulopathy of liver disease.
Here are some citations about hepatic encephalopathy:

Proceed With Case
Treatment of hepatic encephalopathy.

Cordoba J - Am J Gastroenterol - 1997 Sep; 92(9): 1429-39
PORTAL SYSTEMIC ENCEPHALOPATHY

Schenker S - Clinics in Liver Disease - 1997 May; 1(1); 157-184
Neurospectroscopic alterations and globus pallidus hyperintensity as related

magnetic resonance markers of reversible hepatic encephalopathy [see
comments]
Pujol J - Neurology - 1996 Dec; 47(6): 1526-30
Severe recurrent hepatic encephalopathy that responded to oral branched chain

amino acids.
Chalasani N - Am J Gastroenterol - 1996 Jun; 91(6): 1266-8
Subclinical hepatic encephalopathy: how best to diagnose?

Chalasani N - Am J Gastroenterol - 1997 May; 92(5): 905-6
Markers of reversible hepatic encephalopathy [letter; comment]

Krieger D - Neurology - 1997 Oct; 49(4): 1187-8
Helicobacter pylori, hyperammonemia, and hepatic encephalopathy: is there a

correlation? [letter]
Rinaldi V - Am J Gastroenterol - 1997 Apr; 92(4): 723-4



April 22, 2000
Conde Petra

Company
Section 9 - Liver
79 - Systemic Complications of Liver Feldman: Sleisenger & Fordtran's Gastrointestinal and Liver Disease,
Disease
HEPATIC ENCEPHALOPATHY
Definition HEPATIC ENCEPHALOPATHY

Pathophysiology
Definition
Ammonia Hypothesis
Hepatic encephalopathy, or portosystemic encephalopathy,
gamma-Aminobutyric Acid
represents a reversible decrease in neurologic function caused
Hypothesis
by liver disease. [1] It occurs most notably in patients with portal
Diagnosis hypertension and shunting of blood away from the liver. In the
setting of chronic liver disease, the onset is often insidious and
Management is characterized by subtle and sometimes intermittent changes in
memory, personality, concentration, and reaction times. [2] [3]
Typically, early changes are subclinical and are recognized only
in retrospect, but latent encephalopathy can be clinically
significant. For example, early studies of compensated cirrhotic
patients without clinical evidence of encephalopathy indicated
that more than half were unfit to drive, as assessed by a battery
of psychometric tests. [3] Although these results cannot be
extended to all patients with cirrhosis, [4] the implications for
patients and for society are important in view of the prevalence
of cirrhosis.
With progression of encephalopathy, the neurologic
abnormalities become more apparent and are commonly graded
on a numerical scale reflecting increasing degrees of neurologic
dysfunction (Table 79-1) . [1] The manifestations of stage 1
encephalopathy reflect involvement of higher cortical functions,
with decreases in attention span, changes in personality,
irritability, and impaired computational and construction skills.
A change in sleep pattern with wakefulness at night and
drowsiness during the day is notable. The electroencephalogram
(EEG), if measured, is usually normal but may show subtle
slowing of the dominant frequency. Progression to stage 2 is
characterized by an exaggeration of these cortical

manifestations, with more drowsiness and lethargy, and by
appearance of movement disorders reflecting increasing
involvement of the descending reticular system or other
neurologic structures. These include tremors, incoordination,
and the onset of asterixis. [5] [6] In cooperative patients, asterixis
is commonly evaluated by asking the patient to hold the arms
extended with the wrists dorsiflexed. It is characterized by an
abrupt loss of flexor tone and a characteristic wristdrop
occurring in a periodic manner every 2 to 3 seconds.
Alternatively, the examiner can grip the patient's hand and
lightly hold the wrist in a dorsiflexed position, and the periodic
relaxations are also apparent. In the setting of mental confusion,
drowsiness, and personality changes, the presence of asterixis is
very suggestive of underlying hepatic encephalopathy. An EEG
performed
1335

CLINICAL INTELLECTUAL NEUROMUSCULAR
STAGE FUNCTION FUNCTION
Subclinical Normal examination, Subtle changes on
but work or driving psychometric or number
may be impaired connection tests
Stage 1 Impaired attention, Tremor, incoordination
irritability, apraxia
depression, or
personality change
Stage 2 Drowsiness, Asterixis, slowed or
behavioral changes, slurred speech, ataxia
poor memory and
computation, sleep
disorders
Stage 3 Confusion and Hypoactive reflexes,
disorientation, nystagmus, clonus and
somnolence, amnesia muscular rigidity

Stage 4 Stupor and coma Dilated pupils and

decerebrate posturing;
oculocephalic reflex;
absence of response to
stimuli in advanced
stages
in stage 2 usually shows slower rhythms and appearance of

triphasic waves in the frontal regions.
Progression to stage 3, defined as increasing obtundation in a
still arousable patient, or to stage 4, in which the patient is
comatose, reflects either severe bilateral cortical dysfunction or
involvement of the brainstem and reticular activating system.
Asterixis may be lost, and hyperreflexia and muscle rigidity
become apparent. [1] The EEG shows severe slowing with
frequencies in the theta and delta ranges. Even though the
clinical features may be fully reversible with treatment,
encephalopathy of this degree is generally a manifestation of
advanced liver disease and is associated with a very poor
long-term prognosis.
These clinical features of hepatic encephalopathy are
nonspecific, and there is considerable overlap in the presence
and severity of the findings. Similar manifestations can
accompany hypoxia, acidosis, drugs, or other metabolic and
toxic insults. Consequently, it is important to consider and
exclude these possibilities by appropriate drug screens and
testing. It is worth emphasizing that the neurologic
manifestations of hepatic encephalopathy are generally
symmetric. Consequently, the appearance of focal neurologic
motor or sensory abnormalities such as cranial nerve
dysfunctions or paresis should always prompt investigation for
other causes of structural neurologic disease, such as
intracranial hemorrhage.
Histologic examination of the brains of patients with chronic
liver disease who have had recurrent or chronic encephalopathy
has identified several abnormalities, the most notable of which
is a change in the number of astrocytes, which may contribute
to the neuropathologic abnormality found in hepatic
encephalopathy. [7] [8] However, it is difficult to assess whether
these changes represent a cause or effect of encephalopathy.
Evidence for and against increased signals in T1-weighted
magnetic resonance images of the globus pallidus has also been
presented. [9] However, most early cases of encephalopathy are

fully reversible with treatment, arguing against a structural basis

for encephalopathy and for a toxic or neurohumoral cause.
Pathophysiology
Despite the frequency and characteristic clinical features of

hepatic encephalopathy, the precise mechanisms involved are
not fully defined. However, decades of experience with animal
models, including dogs with a surgically created Eck fistula
(end-to-side portacaval shunt), [10] have identified the essential
elements. In the setting of portosystemic shunting where portal
blood is diverted away from the liver and into the vena cava,
ingestion of a protein meal is associated with the onset of
encephalopathy and progression to coma and death. Although
the precise mechanisms are still not established, these findings
point toward a key role for nitrogenous by-products of proteins
absorbed from the colon into the portal circulation.
This model is in some ways simplistic and does not account for
other potentially important parameters such as changes in
central neurotransmitters and the blood-brain barrier. [11] [12]
However, it fits well with clinical experience and makes no
assumptions about the precise identity of the toxin or toxins
involved. For example, whereas creation of a portacaval
anastomosis is highly effective for treatment of bleeding
associated with portal hypertension, the clinical consequence is
increased shunting and an increase in the frequency and severity
of encephalopathy. [13] Thus, recognition of these key features--
portosystemic shunting and defective hepatic clearance of
nitrogenous metabolites--continues to form the basis for the
standard treatments of hepatic encephalopathy.
Is ammonia the toxin responsible for this disorder? Yes, but it is
not the only one, and the mechanisms by which ammonia
produces neuropsychiatric abnormalities are not fully defined.
In most clinical series, elevations of blood ammonia are
detected in 60% to 80% of patients with cirrhosis and
encephalopathy, and therapy aimed at decreasing the
concentration of ammonia results in resolution of
encephalopathy. [2] [6] [14] It is clear, however, that multiple
metabolic abnormalities coexist, including changes in the
profile of circulating amino acids, mercaptans, and central
nervous system levels of dopamine and other neurotransmitters.
[12] [15] [16] These alterations are summarized in Table 79-2 and
are present to a varying extent in different clinical scenarios and

probably work in a complementary manner to modify
neurologic function in cirrhosis. [12] Even if ammonia is not the

only cause, or even the predominant cause, of this

encephalopathy it is clinically a very useful marker for
production of enteric toxins from nitrogenous substrates.
Detailed reviews of the pathogenesis of hepatic encephalopathy
have been published. [1] [2] In the following sections, emphasis is
placed on a brief review of the role of ammonia and on the
potential role of inhibitory neurotransmission through
gamma-aminobutyric acid (GABA) receptors in the central
nervous system. This selection is based on the importance of
ammonia as a guide to therapy and on the emerging support for
the GABA receptor complex as a target for newer therapies.
Other mechanisms involving changes in central
neurotransmitters or
1336
TABLE 79-2 -- Pathogenesis of Hepatic Encephalopathy

MECHANISM HYPOTHESIS
Toxins (ammonia Ammonia and mercaptans produced by
and mercaptans) action of intestinal bacteria on urea and
protein are elevated in blood and brain
as a result of defective hepatic
clearance and lead to impaired neural
function through cytotoxicity, cell
swelling, and depletion of glutamate
GABAergic Defective hepatic clearance of GABA
neurotransmission produced by intestinal bacteria,
increased neuronal GABA synthesis,
and increased production of
benzodiazepine receptor agonists leads
to neuronal inhibition through
stimulation of the GABA receptor
complex in postsynaptic membranes
False Increases in the ratio of plasma
neurotransmitters aromatic amino acids to
branched-chain amino acids increases
brain levels of aromatic amino acid
precursors for false neurotransmitters
GABA, gamma-aminobutyric acid.
circulating amino acids are also relevant and related, but their
therapeutic implications are not as well defined.
Ammonia Hypothesis
Ammonia is a key intermediate in nitrogen and protein

metabolism, and the dynamics of ammonia handling in humans
are well defined. [17] [18] The gastrointestinal tract is the primary
site of ammonia production. Nitrogenous compounds in the
colon, which include ingested proteins and secreted urea, are
degraded by bacteria and liberate ammonia that is then absorbed
into the portal circulation, where concentrations are five- to
ten-fold greater than in mixed venous blood. [18] The liver has a
very high first-pass extraction of ammonia, [17] resulting in
clearance from the portal system and prevention of ammonia
entry into the systemic circulation. Within hepatocytes,
ammonia is rapidly converted by a series of enzymatic reactions
to nontoxic glutamine and, in separate reactions, is synthesized
into urea for secretion by the kidneys (see Chapters 63 and 67) .
Although abnormalities in urea cycle enzymes occur in
congenital syndromes, enzyme deficiencies are not the major
concern in most patients with cirrhosis, as compared with
ammonia bypassing the liver through portosystemic shunting.
In addition to the role in urea transport, the kidneys represent a
site for ammonia generation and actively secrete ammonia into
the urine. [17] Indeed, there is a net increase in the concentration
of ammonia in renal veins as compared with renal arteries; and
the concentration of ammonia in the renal veins is increased by
hypokalemia and use of diuretics. [18] [19] Clinical studies
support a role for hypokalemia in precipitation of hepatic
encephalopathy through effects on renal genesis of ammonia.
[20]
After bolus injection of radiolabeled ammonia, the liver,

bladder, and brain show appreciable uptake. [17] In
encephalopathy, arterial ammonia levels increase and the rate of
brain ammonia accumulation also increases from 32 3
mumol/min to 53 7 mumol/min. [17] Because muscle is an
important site for ammonia clearance, the muscle atrophy seen
in advanced cirrhosis may contribute to the increase in brain
uptake. [17]
Although the implications of these observations regarding
ammonia metabolism, portosystemic shunting, and the
pathogenesis of hepatic encephalopathy are not fully defined, in
aggregate they indicate a clear relationship between
encephalopathy and abnormal ammonia handling. Difficulties
in the measurement and interpretation of blood ammonia levels
include (1) substantial variations in venous as compared with

arterial levels, (2) effects of exercise-induced release of
ammonia from skeletal muscle, (3) poor correlation between the
absolute value of the ammonia level and the degree of
encephalopathy, and (4) differences in the time course between
the rise in ammonia and the onset of symptoms. [21] Despite
these limitations, measures to lower arterial ammonia levels
remain a cornerstone in the management of hepatic coma. [14] [22]
[23]
Patients with cirrhosis are subject to changes in systemic fluid

and electrolyte balance by virtue of the sodium and water
retention that accompanies cirrhosis and by the frequent use of
potent diuretics. Because encephalopathy is commonly
precipitated by metabolic events, [24] it is instructive to consider
how abnormalities in acid-base and electrolyte balance
influence ammonia metabolism, with the assumption that
increases in ammonia levels increase the severity of
encephalopathy. The effects of uremia are predictable because
urea diffuses into the colon, where it is metabolized to liberate
ammonia after bacterial degradation. The effects of
hypokalemia and alkalosis are more subtle, although
hypokalemia frequently develops in cirrhotic patients as a
consequence of diuretic-induced urinary losses, diarrhea,
vomiting, and nutritional deficiencies. First, hypokalemia
increases ammonia production by the kidney. [19] [20] Second,
hypokalemia and alkalosis favor cellular uptake of ammonia. [21]
Because most of the body's potassium stores are found in the
intracellular space, lowering of potassium concentrations in the
extracellular fluid stimulates efflux of intracellular potassium
out of cells to restore extracellular concentrations. Cells
compensate for the loss of potassium by a net uptake of sodium
and hydrogen ions to maintain electroneutrality, leading to
relative alkalinization of the extracellular space and
acidification of the intracellular space. [21] Because ammonia
and the ammonium radical exist in equilibrium, the extracellular
alkalosis increases the portion of membrane-permeable
ammonia, whereas the intracellular acidosis serves to trap
ammonium within the cell. Thus, the net effect of hypokalemia
is a shift of ammonia into neurons or other cells where it exerts
its toxic effects.
Despite the strong evidence that implicates ammonia as an
important contributor to hepatic encephalopathy, the precise
cellular mechanisms involved remain elusive. This alone does
not discount the role of ammonia, but it does lead to humility
among investigators and serves as a stimulus for continued

research. Several potential mechanisms of ammonia-induced

neuronal dysfunction have been described. Ammonia has been
reported to decrease the concentration of glycogen in cultured
astrocytes, [7] impair glial-neuronal communication, [8] and
interfere with synaptic transmission. [15] Over longer periods,
sustained elevation of ammonia induces pathologic changes in
perineural astrocytes. [15] Since glycogen stores in astrocytes
represent an important energy reserve for the brain, disruption
of glial-neuronal signaling may play a role on the pathogenesis
of hepatic encephalopathy. [7] [25] Observations in animal models
of this disorder and hyperammonemia support these general
1337
conclusions, [22] although the multiple effects of ammonia and

its metabolites have not been fully resolved.
gamma-Aminobutyric Acid Hypothesis
Ammonia causes some of the signs and symptoms of hepatic

encephalopathy only after it is metabolized by glutamine
synthetase in the brain. In an animal model, portacaval shunting
leads to increases in plasma and brain ammonia concentrations,
as well as increases in brain glutamine and tryptophan as a
result of the action of glutamine synthetase. [26] Inhibition of
glutamine synthetase results in normalization of brain glutamine
concentrations and normalization of glucose consumption and
other parameters, supporting a role for glutamine synthesis in
the development of cerebral metabolic abnormalities in
hyperammonemic states. [26] Thus, ammonia alone does not
explain the central nervous system abnormalities in hepatic
encephalopathy.
Studies in humans and animal models have implicated the
GABA-receptor complex as a key contributor to neuronal
inhibition in hepatic encephalopathy. [16] [27] The
GABA-receptor complex (Fig. 79-1) is localized to postsynaptic
membranes and constitutes the principal inhibitory network in
the central nervous system. It consists of (1) a GABA-binding
site facing the extracellular surface, (2) a chloride-selective pore
that opens in response to GABA binding to permit influx of
chloride and produce membrane hyperpolarization, and (3)
closely associated barbiturate and benzodiazepine receptor sites
that potentiate the effects of GABA. The endogenous ligands
for the benzodiazepine receptor are not known.
Theoretically, increases in GABAergic transmission could

result from increased availability of extracellular GABA or

benzodiazepine receptor ligands. The liver contains high
concentrations of GABA and GABA transaminase. [28]
Consequently, liver injury disrupts GABA homeostatic
mechanisms and may contribute to the pathogenesis of hepatic
encephalopathy. In addition, ammonia combines with
alpha-ketoglutarate in the central nervous system to form
glutamate, which, in turn,
Figure 79-1 gamma-Aminobutyric acid (GABA)

receptor-chloride channel complex in the
postsynaptic membrane.
is amidated to produce GABA. Thus, increased production of

GABA would be expected to correlate with ammonia levels. [1]
[22] There is better evidence, however, for a role for endogenous
benzodiazepine receptor ligands. [16] [27] [29]

In the absence of known ligands, putative benzodiazepine
receptor agonists are identified by their competitive inhibition
of flumazenil binding. [16] [30] In both animal [31] and human
models, [30] hepatic encephalopathy is associated with an
increase in benzodiazepine receptor ligands. Similarly, there is
increased benzodiazepine-like activity in cerebrospinal fluid,
blood and urine in human hepatic encephalopathy. [2] Several
additional points merit emphasis. First, gut bacteria provide
precursors of benzodiazepine receptor ligands just as they do
ammonia. [32] Impairment of hepatic clearance in cirrhosis of
such ligands follows important parallels with the role of
ammonia, and treatment to lower ammonia would be expected
to have similar effects on benzodiazepine receptor ligands.
Second, the concentration of these ligands correlates roughly
with the stage of encephalopathy. [30] Finally, this disorder is
ameliorated in some patients by the benzodiazepine receptor
antagonist flumazenil [29] or its structurally related analogs Ro
15-3505 and Ro 15-4513. [27]
Evaluation of the efficacy of flumazenil in the treatment of
hepatic encephalopathy has been the subject of several clinical
trials. In general, infusion of flumazenil (0.4-1 mg) results in
modest but rapid improvement of the electroencephalogram and
a more delayed improvement in mental status. [33] [34] Some of
these responders had received pharmaceutical benzodiazepines.
Other studies, including blinded cross-over trials, have failed to
identify a beneficial effect of flumazenil. [35] The reasons for

these differences are not clear. However, it should be noted that

even beneficial responses are usually incomplete, without full
recovery to normal mental status, and are short-lived. This may
reflect the fact that flumazenil-like drugs are incomplete
blockers, or, more likely, that other factors such as ammonia,
mercaptans, and amino acids contribute to hepatic
encephalopathy as well. [1] These studies support a role for
benzodiazepine receptor ligands in the pathogenesis of this
disorder and suggest that flumazenil or other benzodiazepine
receptor antagonists may be useful in treatment. Clearly, they
are of benefit in reversing the effects of exogenous
benzodiazepines and may also be useful in providing prognostic
information and aiding in the differential diagnosis of coma. [27]
Although the findings in animal models are not necessarily
generalizable to all forms of liver injury, the implications are
intriguing and imply that encephalopathy is caused in part by an
increase in inhibitory neurotransmitter tone in the central
nervous system.
Diagnosis
Hepatic encephalopathy presents as a spectrum of neurologic

abnormalities, but the principal clinical features alone are each
nonspecific. Subtle impairments of memory, consciousness, and
personality are easily overlooked if the underlying liver disease
is not recognized. Alternatively, even if there have been
well-defined periods of encephalopathy, it may be difficult to
assess whether recovery has been complete. In contrast, the
clinical features of advanced encephalopathy and asterixis in a
patient with known cirrhosis and portal hypertension are
characteristic, and the combination of asterixis,
hyperammonemia,
1338
and other clinical features permits confident recognition of

portosystemic encephalopathy.
The recognition of latent encephalopathy is of particular
importance in view of the prevalence of cirrhosis. In the
absence of characteristic features, subtle abnormalities of
neuropsychiatric function are generally assessed by more
specialized neuropsychiatric testing such as the Reitan trial test,
block design and digit symbol tests, and visual reaction times.
[4] In addition, there is evidence to support the use of brainstem
auditory evoked responses and somatosensory evoked

responses. [36] These are specialized tests not readily performed
in the clinical setting. Consequently, a high index of suspicion

in patients at risk, such as those with a recent surgical or
transjugular portosystemic shunt, [13] and a beneficial response
to a therapeutic trial are more clinically useful outside research
settings.
Special attention has been paid to the cerebral functional defects
of latent hepatic encephalopathy and fitness to drive. Driving
requires complex response and spatial recognition skills; in an
early series of cirrhotic patients without clinical signs of
portosystemic encephalopathy, psychometric testing indicated
that 60% were unfit to drive and an additional 25% had
questionable driving skills. [3] More recent studies have failed to
identify any major impairment in the ability to drive in the
absence of major abnormalities on neuropsychiatric testing. [4]
Thus, decisions regarding driving should be made on a
case-by-case basis, with no clear support for strict prohibition of
driving in those with compensated cirrhosis and no overt
evidence of encephalopathy.
With progression to clinically apparent hepatic encephalopathy,
two modes of presentation are common. These include an acute
onset of encephalopathy with rapid deterioration of mental
function and coma in the absence of prior symptomatology and
a chronic, relapsing clinical course usually in patients with more
pronounced portal hypertension. In each case, it is important to
emphasize that hepatic encephalopathy in chronic cirrhosis is
usually reversible and that a precipitating cause for the
deterioration can usually be identified and corrected, as
summarized in Table 79-3 . The relative contributions of these
different factors were analyzed in 100 hospital admissions by
Fessel and Conn [24] and are illustrated in Figure 79-2 . Many of
these are readily understood on the basis of their effects on
ammonia. An increase in nitrogenous substances from azotemia
and gastrointestinal hemorrhage together accounted for almost
half of admissions. Iatrogenic causes from medications also
figured prominently. These included
TABLE 79-3 -- Common Clinical Factors That May Precipitate
Hepatic Encephalopathy in Cirrhosis
NITROGENOUS NON-NITROGENOUS
ENCEPHALOPATHY ENCEPHALOPATHY
Uremia/azotemia Sedatives, benzodiazepines
Gastrointestinal bleeding Barbiturates
Dehydration Hypoxia, hypoglycemia

Metabolic alkalosis Hypothyroidism

Hypokalemia Anemia
Constipation
Excessive dietary protein
Infection
Figure 79-2 Clinical causes of portosystemic

encephalopathy. (Data from Fessel, J. M., and Conn,
H. O. An analysis of the causes and prevention of
hepatic coma. Gastroenterology 62:191, 1972.)
precipitation of coma directly from increased sensitivity to

tranquilizers and sedatives and indirectly through hypokalemia,
dehydration, and alkalosis associated with diuretic use. The
important lesson is that identification and correction of these
causes is the cornerstone of effective therapy and that very few
patients with chronic cirrhosis develop encephalopathy due to
an irreversible loss of hepatocyte mass and synthetic capacity.
Arterial ammonia levels should be measured whenever hepatic
encephalopathy is suspected, both for diagnostic purposes and
as a general guide to treatment. Normal values do not exclude
the diagnosis and should not delay initiation of
ammonia-lowering therapy. Approximately one fourth of
patients will have non-nitrogenous causes, such as adverse
reactions to sedatives, fluid and electrolyte imbalances, or other
causes. Given the potential involvement of the GABA-receptor
complex, it is not surprising that these patients respond to
treatment in a manner similar to those with elevated ammonia
levels. Other diagnostic tests including measurement of
glutamine levels in the cerebrospinal fluid and
electroencephalography can provide important confirmation of
the clinical impression but alone are not sufficiently sensitive or
specific to define the diagnosis.
The clinical stages of hepatic encephalopathy provide a general
index of severity in the acute setting but are not sufficiently
quantitative to assess subtle changes in clinical performance.
Consequently, there is a need for reliable, reproducible tests that
can be easily administered. The trailmaking test provides a
semiquantitative measure that has been useful. In this test, the
subject connects 25 consecutively numbered circles, and the

number of seconds required to complete the task is recorded. [6]

An alternative figure-making test has also been introduced and
validated for those who cannot recognize numbers, detecting
subclinical encephalopathy in 48% of subjects. [37] None of
these measures alone is entirely satisfactory, and these
semiquantitative tests are best utilized when administered
serially to assess changes over time. The overall assessment
provided by the Portosystemic Encephalopathy Index
introduced by Conn nearly 20 years ago is based on an arbitrary
measure of the degree of abnormality of five
1339
factors--clinical assessment of mental state, trailmaking time,

EEG, asterixis, and arterial ammonia--is still not surpassed as a
clinical research tool. [6] Although complex, it emphasizes the
need for taking multiple parameters into account in the overall
assessment and diagnosis of hepatic encephalopathy.
Management
The principles involved in management of hepatic

encephalopathy are straightforward (Table 79-4) : identify and
correct the precipitating cause or causes, initiate
ammonia-lowering therapy, and minimize the potential medical
complications of cirrhosis and depressed consciousness. Among
these, careful scrutiny for and correction of the underlying
cause of the deterioration, such as bleeding, tranquilizers, or
azotemia, is the most important. These basic steps are relatively
easy and effective, with excellent recovery to basal function in
most patients in the absence of comorbid factors.
Correction of the underlying cause of the encephalopathy
depends on a careful review of potential contributors. Many of
these, such as gastrointestinal bleeding, dehydration,
hypokalemia, and azotemia, are readily apparent from the initial
physical examination and basic laboratory studies. Particular
attention should be paid to the possibility of gastrointestinal
bleeding because of the high risk in the setting of portal
hypertension and because of the need for specific therapeutic
intervention. Catabolism of blood in the intestine, which
liberates ammonia and presumably benzodiazepine receptor
ligands and other mediators as well, is a classic cause of
nitrogenous encephalopathy.
It is essential, of course, to take action as soon as potential
precipitating factors are identified. If azotemia is the cause, then

rehydration and attention to other prerenal factors are indicated.

If bleeding is the cause, then the bleeding must be controlled,
and so forth. Medications should be reviewed in detail, with
specific attention to tranquilizers and sedatives and to the
adverse effects of diuretics. When in doubt, any potential
contributing medicines should be discontinued.
TABLE 79-4 -- Treatment of Hepatic Encephalopathy
1. Identify and correct the precipitating cause(s).
a. Monitor volume status and vital signs.
b. Assess for gastrointestinal bleeding.
c. Eliminate sedatives, tranquilizers, or similar drugs.
d. Perform screening tests for hypoxia, hypoglycemia,
anemia, hypokalemia, and other potential metabolic or
endocrine factors and correction as indicated.
2. Initiate ammonia-lowering therapy.
a. Perform nasogastric lavage and administer lactulose
other cathartics or enemas to remove the source of ammonia
from the colon.
b. Minimize or eliminate dietary protein.
c. Initiate treatment with lactulose or lactitol to produce two
to four bowel movements per day.
d. Consider antibiotics to reduce intestinal bacterial counts.
e. Consider flumazenil and other benzodiazepine receptor
antagonists (see text).
3. Minimize the potential complications of cirrhosis and
depressed consciousness.
a. Provide supportive care with attention to airway and
hemodynamic and metabolic status.
Moreover, general measures to correct and maintain glucose,

oxygenation, and acid-base balance are essential.
The second step in treatment is directed toward lowering
elevated ammonia levels. This involves removing the source of
the ammonia from the intestinal tract, trapping ammonia in the
colon to prevent systemic absorption, and, in some patients,
specific therapy to decrease the number of ammonia-producing
bacteria in the colon.

In patients with gastrointestinal bleeding, removing the source

of ammonia involves elimination of blood from the
gastrointestinal tract. For hemorrhage in the upper intestine,
nasogastric lavage to remove blood, and initiation of lactulose
or other cathartics to speed the transit through the colon are
appropriate. In more chronic encephalopathy not associated
with bleeding, excessive protein ingestion or constipation may
elevate ammonia to levels sufficient to cause encephalopathy.
[24] The same treatment principles apply, including a decrease in
dietary protein to about 60 g/day, particularly if lactulose or

other measures to decrease ammonia production in the colon
are, by themselves, ineffective. In addition, there appears to be
some advantage to substituting vegetable protein for other
protein sources because of a lower rate of ammonia production.
[38] In severe encephalopathy, dietary protein should be
eliminated until there is sufficient improvement to allow

institution of a stable therapeutic regimen.
The synthetic disaccharides lactulose (1,4-galactosidofructose)
[6] and lactitol (beta-galactosidosorbitol) [39] represent the
mainstays of medical therapy of nitrogenous hepatic

encephalopathy. These agents target the production and
absorption of ammonia and benzodiazepine receptor ligands in
the gut. Lactulose was introduced about 30 years ago as a
therapy for hepatic encephalopathy based on the concept that it
would acidify the colon contents and favor trapping of
ammonium in the lumen and prevention of absorption. In the
colon, lactulose is metabolized by bacteria to release lactic,
acetic, and other organic acids, decreasing stool pH to about
5.5. [6] It is clinically effective in over 80% of patients, leading
to a decrease in serum ammonia levels and an improvement in
encephalopathy. [6] [14] [39] Treatment is well tolerated, and the
principal toxicity is abdominal cramping, diarrhea, and
flatulence. When orally administered to normal adults in
amounts up to 160 g/day, lactulose decreases fecal ammonia
production and increases fecal nitrogen excretion approximately
four-fold, owing to the increase in stool volume. [23] Thus, an
increase in bowel movements to two to four soft stools per day
is an important therapeutic goal.
Several clinical trials have evaluated the relative efficacy of
lactulose versus lactitol in treatment of hepatic encephalopathy.
In general, these drugs are equally effective, but there is a trend
toward better palatability and fewer side effects with lactitol. [39]
Antibiotics of several types including neomycin, ampicillin, and
rifaximin are also effective in lowering blood ammonia levels [6]

[14] [40]The effect on ammonia is due in large part to a decrease

in the number of colonic bacteria and a concomitant decrease in
bacterial urease and protease activity, the main enzymes
responsible for ammonia generation. [40] In addition, decreasing
colonic bacteria appears to decrease the production of
benzodiazepine receptor ligands. [32] In most patients, the
response to antibiotics is equivalent to lactulose, and in small
series it may be associated with improved patient compliance.
[6] [14] However, nonspecific use of antibiotics in the absence
1340
of an established or suspected infection raises certain concerns.

Neomycin, for example, can be absorbed systemically in
concentrations sufficient to induce ototoxicity and
nephrotoxicity, particularly when given over longer periods;
and the alterations in gut flora associated with antibiotic use can
contribute to diarrhea, malabsorption, and staphylococcal and
other overgrowth syndromes. Thus, chronic therapy with
antibiotics should be reserved for those who cannot tolerate oral
lactulose or lactitol therapy, and neomycin should be avoided.
Treatment of hepatic encephalopathy in the absence of elevated
ammonia levels follows the same principles, including a careful
review for use of sedatives or analgesics. Prolonged recovery
from sedatives given for endoscopy or other procedures is
characteristic. In addition, initiation of ammonia-lowering steps
appears to be effective, perhaps because of the effects on
GABAergic transmission [1] [32] or other agents that are derived
from the colon and contribute to the clinical syndrome.
The role of flumazenil and other benzodiazepine receptor
antagonists in treatment of hepatic encephalopathy is not yet
defined. [27] In clinical trials, evidence for [33] [34] and against [35]
clinical benefit has been presented. Even when the response was
favorable, the recovery was rarely complete and was
short-lived, owing to the pharmacokinetic properties of the
drug. Thus, much remains to be learned regarding the origin,
overall contribution, and therapy for increased GABAergic
transmission in hepatic encephalopathy. At present, therapy
with flumazenil should be limited according to guidelines of
Jones and others, which include (1) reversal of effects of
exogenous benzodiazepines, (2) aiding the differential diagnosis
of encephalopathy, and (3) providing information about
prognosis and optimizing brain function in this disorder. [1] [27]
[29] These indications are likely to evolve with additional clinical
experience and the development of more selective and effective

analogs.
There is a long history of clinical trials involving other more
experimental approaches to treatment of hepatic
encephalopathy. These include but are not limited to use of
dietary zinc, levodopa, branched-chain amino acids, and
charcoal hemoperfusion. [1] [2] Although anecdotal reports have
been encouraging, none has been of sufficient established
benefit to achieve widespread application.


Saunders Company
1334
Chapter 79 - Systemic Complications of Liver

Disease
Greg Fitz
HEPATIC ENCEPHALOPATHY, 1334

Definition, 1334
Pathophysiology, 1335
Diagnosis, 1337
Management, 1339
PULMONARY SYNDROMES, 1340
Definition, 1340
Diagnosis, 1341
Management, 1342
HEPATORENAL SYNDROME, 1343
Definition, 1343
Diagnosis, 1345
Management, 1346
ENDOCRINE DYSFUNCTION, 1347
Definition, 1347
Diagnosis, 1349
Management, 1349
COAGULATION DISORDERS, 1349
Definition, 1349
Diagnosis, 1351
Management, 1351

The liver plays a central role in regulation of other organ systems by virtue of its role in nutrition,
metabolism, and secretion of xenobiotics and endobiotics. Consequently, chronic liver disease can lead to
a broad range of systemic manifestations that may dominate the clinical course and represent principal
indications for liver transplantation. Some of these complications result from a decrease in number of
functioning hepatocytes and concomitant loss of synthetic and metabolic capacity. Others reflect the
increased pressure in the portal circulation, leading to opening of vascular collaterals and shunting of
blood away from hepatic lobules. These manifestations of cirrhosis--decreased synthetic reserve and
altered perfusion--are functionally interrelated and can change over time in response to varying
physiologic demands. Consequently, the systemic effects of cirrhosis on other organ systems are also
dynamic, with symptoms frequently developing in the absence of obvious deterioration evident on the
standard biochemical tests of liver function.
Despite the diverse organ systems affected by cirrhosis, in the early stages they share a common
mechanistic bond in that they are largely functional, representing a secondary effect of cirrhosis and not a
primary abnormality of the target organ or organs. Thus, replacing a failing liver by transplantation can
lead to full restoration of associated renal, neurologic, or other abnormalities. Moreover, specific
diagnostic features are not always present, and patients with liver disease are also susceptible to other
disease processes. In certain systemic diseases such as cystic fibrosis (pulmonary) or Wilson disease
(brain), multiple organ systems are targeted by the same pathophysiologic mechanisms. In this chapter an
overview is provided of the pathophysiology of the most common systemic manifestations of chronic
liver disease with particular focus placed on their diagnosis and management.


Saunders Company
HEPATIC ENCEPHALOPATHY
Definition
Hepatic encephalopathy, or portosystemic encephalopathy, represents a reversible decrease in neurologic

function caused by liver disease. [1] It occurs most notably in patients with portal hypertension and
shunting of blood away from the liver. In the setting of chronic liver disease, the onset is often insidious
and is characterized by subtle and sometimes intermittent changes in memory, personality, concentration,
and reaction times. [2] [3] Typically, early changes are subclinical and are recognized only in retrospect,
but latent encephalopathy can be clinically significant. For example, early studies of compensated
cirrhotic patients without clinical evidence of encephalopathy indicated that more than half were unfit to
drive, as assessed by a battery of psychometric tests. [3] Although these results cannot be extended to all
patients with cirrhosis, [4] the implications for patients and for society are important in view of the
prevalence of cirrhosis.
With progression of encephalopathy, the neurologic abnormalities become more apparent and are
commonly graded on a numerical scale reflecting increasing degrees of neurologic dysfunction (Table
79-1) . [1] The manifestations of stage 1 encephalopathy reflect involvement of higher cortical functions,
with decreases in attention span, changes in personality, irritability, and impaired computational and
construction skills. A change in sleep pattern with wakefulness at night and drowsiness during the day is
notable. The electroencephalogram (EEG), if measured, is usually normal but may show subtle slowing
of the dominant frequency. Progression to stage 2 is characterized by an exaggeration of these cortical
manifestations, with more drowsiness and lethargy, and by appearance of movement disorders reflecting
increasing involvement of the descending reticular system or other neurologic structures. These include
tremors, incoordination, and the onset of asterixis. [5] [6] In cooperative patients, asterixis is commonly
evaluated by asking the patient to hold the arms extended with the wrists dorsiflexed. It is characterized
by an abrupt loss of flexor tone and a characteristic wristdrop occurring in a periodic manner every 2 to 3
seconds. Alternatively, the examiner can grip the patient's hand and lightly hold the wrist in a dorsiflexed
position, and the periodic relaxations are also apparent. In the setting of mental confusion, drowsiness,
and personality changes, the presence of asterixis is very suggestive of underlying hepatic
encephalopathy. An EEG performed
1335

CLINICAL
STAGE INTELLECTUAL FUNCTION NEUROMUSCULAR FUNCTION
Subclinical Normal examination, but work or driving Subtle changes on psychometric or number
may be impaired connection tests

Stage 1 Impaired attention, irritability, depression, Tremor, incoordination apraxia

or personality change
Stage 2 Drowsiness, behavioral changes, poor Asterixis, slowed or slurred speech, ataxia
memory and computation, sleep disorders
Stage 3 Confusion and disorientation, somnolence, Hypoactive reflexes, nystagmus, clonus
amnesia and muscular rigidity
Stage 4 Stupor and coma Dilated pupils and decerebrate posturing;
oculocephalic reflex; absence of response
to stimuli in advanced stages
in stage 2 usually shows slower rhythms and appearance of triphasic waves in the frontal regions.
Progression to stage 3, defined as increasing obtundation in a still arousable patient, or to stage 4, in
which the patient is comatose, reflects either severe bilateral cortical dysfunction or involvement of the
brainstem and reticular activating system. Asterixis may be lost, and hyperreflexia and muscle rigidity
become apparent. [1] The EEG shows severe slowing with frequencies in the theta and delta ranges. Even
though the clinical features may be fully reversible with treatment, encephalopathy of this degree is
generally a manifestation of advanced liver disease and is associated with a very poor long-term
prognosis.
These clinical features of hepatic encephalopathy are nonspecific, and there is considerable overlap in the
presence and severity of the findings. Similar manifestations can accompany hypoxia, acidosis, drugs, or
other metabolic and toxic insults. Consequently, it is important to consider and exclude these possibilities
by appropriate drug screens and testing. It is worth emphasizing that the neurologic manifestations of
hepatic encephalopathy are generally symmetric. Consequently, the appearance of focal neurologic
motor or sensory abnormalities such as cranial nerve dysfunctions or paresis should always prompt
investigation for other causes of structural neurologic disease, such as intracranial hemorrhage.
Histologic examination of the brains of patients with chronic liver disease who have had recurrent or
chronic encephalopathy has identified several abnormalities, the most notable of which is a change in the
number of astrocytes, which may contribute to the neuropathologic abnormality found in hepatic
encephalopathy. [7] [8] However, it is difficult to assess whether these changes represent a cause or effect
of encephalopathy. Evidence for and against increased signals in T1-weighted magnetic resonance
images of the globus pallidus has also been presented. [9] However, most early cases of encephalopathy
are fully reversible with treatment, arguing against a structural basis for encephalopathy and for a toxic
or neurohumoral cause.
Pathophysiology
Despite the frequency and characteristic clinical features of hepatic encephalopathy, the precise
mechanisms involved are not fully defined. However, decades of experience with animal models,
including dogs with a surgically created Eck fistula (end-to-side portacaval shunt), [10] have identified the
essential elements. In the setting of portosystemic shunting where portal blood is diverted away from the
liver and into the vena cava, ingestion of a protein meal is associated with the onset of encephalopathy

and progression to coma and death. Although the precise mechanisms are still not established, these
findings point toward a key role for nitrogenous by-products of proteins absorbed from the colon into the
portal circulation.
This model is in some ways simplistic and does not account for other potentially important parameters
such as changes in central neurotransmitters and the blood-brain barrier. [11] [12] However, it fits well with
clinical experience and makes no assumptions about the precise identity of the toxin or toxins involved.
For example, whereas creation of a portacaval anastomosis is highly effective for treatment of bleeding
associated with portal hypertension, the clinical consequence is increased shunting and an increase in the
frequency and severity of encephalopathy. [13] Thus, recognition of these key features-- portosystemic
shunting and defective hepatic clearance of nitrogenous metabolites--continues to form the basis for the
standard treatments of hepatic encephalopathy.
Is ammonia the toxin responsible for this disorder? Yes, but it is not the only one, and the mechanisms by
which ammonia produces neuropsychiatric abnormalities are not fully defined. In most clinical series,
elevations of blood ammonia are detected in 60% to 80% of patients with cirrhosis and encephalopathy,
and therapy aimed at decreasing the concentration of ammonia results in resolution of encephalopathy. [2]
[6] [14] It is clear, however, that multiple metabolic abnormalities coexist, including changes in the profile
of circulating amino acids, mercaptans, and central nervous system levels of dopamine and other
neurotransmitters. [12] [15] [16] These alterations are summarized in Table 79-2 and are present to a varying
extent in different clinical scenarios and probably work in a complementary manner to modify neurologic
function in cirrhosis. [12] Even if ammonia is not the only cause, or even the predominant cause, of this
encephalopathy it is clinically a very useful marker for production of enteric toxins from nitrogenous
substrates.
Detailed reviews of the pathogenesis of hepatic encephalopathy have been published. [1] [2] In the
following sections, emphasis is placed on a brief review of the role of ammonia and on the potential role
of inhibitory neurotransmission through gamma-aminobutyric acid (GABA) receptors in the central
nervous system. This selection is based on the importance of ammonia as a guide to therapy and on the
emerging support for the GABA receptor complex as a target for newer therapies. Other mechanisms
involving changes in central neurotransmitters or
1336
TABLE 79-2 -- Pathogenesis of Hepatic Encephalopathy

MECHANISM HYPOTHESIS
Toxins (ammonia and mercaptans) Ammonia and mercaptans produced by action of intestinal
bacteria on urea and protein are elevated in blood and brain as a
result of defective hepatic clearance and lead to impaired neural
function through cytotoxicity, cell swelling, and depletion of
glutamate

GABAergic neurotransmission Defective hepatic clearance of GABA produced by intestinal

bacteria, increased neuronal GABA synthesis, and increased
production of benzodiazepine receptor agonists leads to neuronal
inhibition through stimulation of the GABA receptor complex in
postsynaptic membranes
False neurotransmitters Increases in the ratio of plasma aromatic amino acids to
branched-chain amino acids increases brain levels of aromatic
amino acid precursors for false neurotransmitters
GABA, gamma-aminobutyric acid.
circulating amino acids are also relevant and related, but their therapeutic implications are not as well
defined.
Ammonia Hypothesis
Ammonia is a key intermediate in nitrogen and protein metabolism, and the dynamics of ammonia
handling in humans are well defined. [17] [18] The gastrointestinal tract is the primary site of ammonia
production. Nitrogenous compounds in the colon, which include ingested proteins and secreted urea, are
degraded by bacteria and liberate ammonia that is then absorbed into the portal circulation, where
concentrations are five- to ten-fold greater than in mixed venous blood. [18] The liver has a very high
first-pass extraction of ammonia, [17] resulting in clearance from the portal system and prevention of
ammonia entry into the systemic circulation. Within hepatocytes, ammonia is rapidly converted by a
series of enzymatic reactions to nontoxic glutamine and, in separate reactions, is synthesized into urea for
secretion by the kidneys (see Chapters 63 and 67) . Although abnormalities in urea cycle enzymes occur
in congenital syndromes, enzyme deficiencies are not the major concern in most patients with cirrhosis,
as compared with ammonia bypassing the liver through portosystemic shunting.
In addition to the role in urea transport, the kidneys represent a site for ammonia generation and actively
secrete ammonia into the urine. [17] Indeed, there is a net increase in the concentration of ammonia in
renal veins as compared with renal arteries; and the concentration of ammonia in the renal veins is
increased by hypokalemia and use of diuretics. [18] [19] Clinical studies support a role for hypokalemia in
precipitation of hepatic encephalopathy through effects on renal genesis of ammonia. [20]
After bolus injection of radiolabeled ammonia, the liver, bladder, and brain show appreciable uptake. [17]
In encephalopathy, arterial ammonia levels increase and the rate of brain ammonia accumulation also
increases from 32 3 mumol/min to 53 7 mumol/min. [17] Because muscle is an important site for
ammonia clearance, the muscle atrophy seen in advanced cirrhosis may contribute to the increase in brain
uptake. [17]
Although the implications of these observations regarding ammonia metabolism, portosystemic shunting,
and the pathogenesis of hepatic encephalopathy are not fully defined, in aggregate they indicate a clear
relationship between encephalopathy and abnormal ammonia handling. Difficulties in the measurement
and interpretation of blood ammonia levels include (1) substantial variations in venous as compared with
arterial levels, (2) effects of exercise-induced release of ammonia from skeletal muscle, (3) poor
correlation between the absolute value of the ammonia level and the degree of encephalopathy, and (4)

differences in the time course between the rise in ammonia and the onset of symptoms. [21] Despite these
limitations, measures to lower arterial ammonia levels remain a cornerstone in the management of
hepatic coma. [14] [22] [23]
Patients with cirrhosis are subject to changes in systemic fluid and electrolyte balance by virtue of the
sodium and water retention that accompanies cirrhosis and by the frequent use of potent diuretics.
Because encephalopathy is commonly precipitated by metabolic events, [24] it is instructive to consider
how abnormalities in acid-base and electrolyte balance influence ammonia metabolism, with the
assumption that increases in ammonia levels increase the severity of encephalopathy. The effects of
uremia are predictable because urea diffuses into the colon, where it is metabolized to liberate ammonia
after bacterial degradation. The effects of hypokalemia and alkalosis are more subtle, although
hypokalemia frequently develops in cirrhotic patients as a consequence of diuretic-induced urinary
losses, diarrhea, vomiting, and nutritional deficiencies. First, hypokalemia increases ammonia production
by the kidney. [19] [20] Second, hypokalemia and alkalosis favor cellular uptake of ammonia. [21] Because
most of the body's potassium stores are found in the intracellular space, lowering of potassium
concentrations in the extracellular fluid stimulates efflux of intracellular potassium out of cells to restore
extracellular concentrations. Cells compensate for the loss of potassium by a net uptake of sodium and
hydrogen ions to maintain electroneutrality, leading to relative alkalinization of the extracellular space
and acidification of the intracellular space. [21] Because ammonia and the ammonium radical exist in
equilibrium, the extracellular alkalosis increases the portion of membrane-permeable ammonia, whereas
the intracellular acidosis serves to trap ammonium within the cell. Thus, the net effect of hypokalemia is
a shift of ammonia into neurons or other cells where it exerts its toxic effects.
Despite the strong evidence that implicates ammonia as an important contributor to hepatic
encephalopathy, the precise cellular mechanisms involved remain elusive. This alone does not discount
the role of ammonia, but it does lead to humility among investigators and serves as a stimulus for
continued research. Several potential mechanisms of ammonia-induced neuronal dysfunction have been
described. Ammonia has been reported to decrease the concentration of glycogen in cultured astrocytes,
[7] impair glial-neuronal communication, [8] and interfere with synaptic transmission. [15] Over longer
periods, sustained elevation of ammonia induces pathologic changes in perineural astrocytes. [15] Since
glycogen stores in astrocytes represent an important energy reserve for the brain, disruption of
glial-neuronal signaling may play a role on the pathogenesis of hepatic encephalopathy. [7] [25]
Observations in animal models of this disorder and hyperammonemia support these general
1337
conclusions, [22] although the multiple effects of ammonia and its metabolites have not been fully
resolved.
gamma-Aminobutyric Acid Hypothesis
Ammonia causes some of the signs and symptoms of hepatic encephalopathy only after it is metabolized
by glutamine synthetase in the brain. In an animal model, portacaval shunting leads to increases in
plasma and brain ammonia concentrations, as well as increases in brain glutamine and tryptophan as a
result of the action of glutamine synthetase. [26] Inhibition of glutamine synthetase results in
normalization of brain glutamine concentrations and normalization of glucose consumption and other
parameters, supporting a role for glutamine synthesis in the development of cerebral metabolic

abnormalities in hyperammonemic states. [26] Thus, ammonia alone does not explain the central nervous
system abnormalities in hepatic encephalopathy.
Studies in humans and animal models have implicated the GABA-receptor complex as a key contributor
to neuronal inhibition in hepatic encephalopathy. [16] [27] The GABA-receptor complex (Fig. 79-1) is
localized to postsynaptic membranes and constitutes the principal inhibitory network in the central
nervous system. It consists of (1) a GABA-binding site facing the extracellular surface, (2) a
chloride-selective pore that opens in response to GABA binding to permit influx of chloride and produce
membrane hyperpolarization, and (3) closely associated barbiturate and benzodiazepine receptor sites
that potentiate the effects of GABA. The endogenous ligands for the benzodiazepine receptor are not
known.
Theoretically, increases in GABAergic transmission could result from increased availability of
extracellular GABA or benzodiazepine receptor ligands. The liver contains high concentrations of
GABA and GABA transaminase. [28] Consequently, liver injury disrupts GABA homeostatic mechanisms
and may contribute to the pathogenesis of hepatic encephalopathy. In addition, ammonia combines with
alpha-ketoglutarate in the central nervous system to form glutamate, which, in turn,
Figure 79-1 gamma-Aminobutyric acid (GABA) receptor-chloride channel complex in the

postsynaptic membrane.
is amidated to produce GABA. Thus, increased production of GABA would be expected to correlate with
ammonia levels. [1] [22] There is better evidence, however, for a role for endogenous benzodiazepine
receptor ligands. [16] [27] [29]
In the absence of known ligands, putative benzodiazepine receptor agonists are identified by their
competitive inhibition of flumazenil binding. [16] [30] In both animal [31] and human models, [30] hepatic
encephalopathy is associated with an increase in benzodiazepine receptor ligands. Similarly, there is
increased benzodiazepine-like activity in cerebrospinal fluid, blood and urine in human hepatic
encephalopathy. [2] Several additional points merit emphasis. First, gut bacteria provide precursors of
benzodiazepine receptor ligands just as they do ammonia. [32] Impairment of hepatic clearance in
cirrhosis of such ligands follows important parallels with the role of ammonia, and treatment to lower
ammonia would be expected to have similar effects on benzodiazepine receptor ligands. Second, the
concentration of these ligands correlates roughly with the stage of encephalopathy. [30] Finally, this
disorder is ameliorated in some patients by the benzodiazepine receptor antagonist flumazenil [29] or its
structurally related analogs Ro 15-3505 and Ro 15-4513. [27]
Evaluation of the efficacy of flumazenil in the treatment of hepatic encephalopathy has been the subject
of several clinical trials. In general, infusion of flumazenil (0.4-1 mg) results in modest but rapid
improvement of the electroencephalogram and a more delayed improvement in mental status. [33] [34]
Some of these responders had received pharmaceutical benzodiazepines. Other studies, including blinded
cross-over trials, have failed to identify a beneficial effect of flumazenil. [35] The reasons for these
differences are not clear. However, it should be noted that even beneficial responses are usually
incomplete, without full recovery to normal mental status, and are short-lived. This may reflect the fact

that flumazenil-like drugs are incomplete blockers, or, more likely, that other factors such as ammonia,
mercaptans, and amino acids contribute to hepatic encephalopathy as well. [1] These studies support a role
for benzodiazepine receptor ligands in the pathogenesis of this disorder and suggest that flumazenil or
other benzodiazepine receptor antagonists may be useful in treatment. Clearly, they are of benefit in
reversing the effects of exogenous benzodiazepines and may also be useful in providing prognostic
information and aiding in the differential diagnosis of coma. [27] Although the findings in animal models
are not necessarily generalizable to all forms of liver injury, the implications are intriguing and imply
that encephalopathy is caused in part by an increase in inhibitory neurotransmitter tone in the central
nervous system.
Diagnosis
Hepatic encephalopathy presents as a spectrum of neurologic abnormalities, but the principal clinical
features alone are each nonspecific. Subtle impairments of memory, consciousness, and personality are
easily overlooked if the underlying liver disease is not recognized. Alternatively, even if there have been
well-defined periods of encephalopathy, it may be difficult to assess whether recovery has been
complete. In contrast, the clinical features of advanced encephalopathy and asterixis in a patient with
known cirrhosis and portal hypertension are characteristic, and the combination of asterixis,
hyperammonemia,
1338
and other clinical features permits confident recognition of portosystemic encephalopathy.

The recognition of latent encephalopathy is of particular importance in view of the prevalence of
cirrhosis. In the absence of characteristic features, subtle abnormalities of neuropsychiatric function are
generally assessed by more specialized neuropsychiatric testing such as the Reitan trial test, block design
and digit symbol tests, and visual reaction times. [4] In addition, there is evidence to support the use of
brainstem auditory evoked responses and somatosensory evoked responses. [36] These are specialized
tests not readily performed in the clinical setting. Consequently, a high index of suspicion in patients at
risk, such as those with a recent surgical or transjugular portosystemic shunt, [13] and a beneficial
response to a therapeutic trial are more clinically useful outside research settings.
Special attention has been paid to the cerebral functional defects of latent hepatic encephalopathy and
fitness to drive. Driving requires complex response and spatial recognition skills; in an early series of
cirrhotic patients without clinical signs of portosystemic encephalopathy, psychometric testing indicated
that 60% were unfit to drive and an additional 25% had questionable driving skills. [3] More recent
studies have failed to identify any major impairment in the ability to drive in the absence of major
abnormalities on neuropsychiatric testing. [4] Thus, decisions regarding driving should be made on a
case-by-case basis, with no clear support for strict prohibition of driving in those with compensated
cirrhosis and no overt evidence of encephalopathy.
With progression to clinically apparent hepatic encephalopathy, two modes of presentation are common.
These include an acute onset of encephalopathy with rapid deterioration of mental function and coma in
the absence of prior symptomatology and a chronic, relapsing clinical course usually in patients with
more pronounced portal hypertension. In each case, it is important to emphasize that hepatic
encephalopathy in chronic cirrhosis is usually reversible and that a precipitating cause for the

deterioration can usually be identified and corrected, as summarized in Table 79-3 . The relative
contributions of these different factors were analyzed in 100 hospital admissions by Fessel and Conn [24]
and are illustrated in Figure 79-2 . Many of these are readily understood on the basis of their effects on
ammonia. An increase in nitrogenous substances from azotemia and gastrointestinal hemorrhage together
accounted for almost half of admissions. Iatrogenic causes from medications also figured prominently.
These included
TABLE 79-3 -- Common Clinical Factors That May Precipitate Hepatic Encephalopathy in Cirrhosis
NITROGENOUS ENCEPHALOPATHY NON-NITROGENOUS ENCEPHALOPATHY
Uremia/azotemia Sedatives, benzodiazepines
Gastrointestinal bleeding Barbiturates
Dehydration Hypoxia, hypoglycemia
Metabolic alkalosis Hypothyroidism
Hypokalemia Anemia
Constipation
Excessive dietary protein
Infection
Figure 79-2 Clinical causes of portosystemic encephalopathy. (Data from Fessel, J. M., and Conn,
H. O. An analysis of the causes and prevention of hepatic coma. Gastroenterology 62:191, 1972.)
precipitation of coma directly from increased sensitivity to tranquilizers and sedatives and indirectly
through hypokalemia, dehydration, and alkalosis associated with diuretic use. The important lesson is
that identification and correction of these causes is the cornerstone of effective therapy and that very few
patients with chronic cirrhosis develop encephalopathy due to an irreversible loss of hepatocyte mass and
synthetic capacity.
Arterial ammonia levels should be measured whenever hepatic encephalopathy is suspected, both for
diagnostic purposes and as a general guide to treatment. Normal values do not exclude the diagnosis and
should not delay initiation of ammonia-lowering therapy. Approximately one fourth of patients will have
non-nitrogenous causes, such as adverse reactions to sedatives, fluid and electrolyte imbalances, or other
causes. Given the potential involvement of the GABA-receptor complex, it is not surprising that these
patients respond to treatment in a manner similar to those with elevated ammonia levels. Other diagnostic
tests including measurement of glutamine levels in the cerebrospinal fluid and electroencephalography
can provide important confirmation of the clinical impression but alone are not sufficiently sensitive or
specific to define the diagnosis.
The clinical stages of hepatic encephalopathy provide a general index of severity in the acute setting but

are not sufficiently quantitative to assess subtle changes in clinical performance. Consequently, there is a
need for reliable, reproducible tests that can be easily administered. The trailmaking test provides a
semiquantitative measure that has been useful. In this test, the subject connects 25 consecutively
numbered circles, and the number of seconds required to complete the task is recorded. [6] An alternative
figure-making test has also been introduced and validated for those who cannot recognize numbers,
detecting subclinical encephalopathy in 48% of subjects. [37] None of these measures alone is entirely
satisfactory, and these semiquantitative tests are best utilized when administered serially to assess
changes over time. The overall assessment provided by the Portosystemic Encephalopathy Index
introduced by Conn nearly 20 years ago is based on an arbitrary measure of the degree of abnormality of
five
1339
factors--clinical assessment of mental state, trailmaking time, EEG, asterixis, and arterial ammonia--is
still not surpassed as a clinical research tool. [6] Although complex, it emphasizes the need for taking
multiple parameters into account in the overall assessment and diagnosis of hepatic encephalopathy.
Management
The principles involved in management of hepatic encephalopathy are straightforward (Table 79-4) :
identify and correct the precipitating cause or causes, initiate ammonia-lowering therapy, and minimize
the potential medical complications of cirrhosis and depressed consciousness. Among these, careful
scrutiny for and correction of the underlying cause of the deterioration, such as bleeding, tranquilizers, or
azotemia, is the most important. These basic steps are relatively easy and effective, with excellent
recovery to basal function in most patients in the absence of comorbid factors.
Correction of the underlying cause of the encephalopathy depends on a careful review of potential
contributors. Many of these, such as gastrointestinal bleeding, dehydration, hypokalemia, and azotemia,
are readily apparent from the initial physical examination and basic laboratory studies. Particular
attention should be paid to the possibility of gastrointestinal bleeding because of the high risk in the
setting of portal hypertension and because of the need for specific therapeutic intervention. Catabolism of
blood in the intestine, which liberates ammonia and presumably benzodiazepine receptor ligands and
other mediators as well, is a classic cause of nitrogenous encephalopathy.
It is essential, of course, to take action as soon as potential precipitating factors are identified. If azotemia
is the cause, then rehydration and attention to other prerenal factors are indicated. If bleeding is the
cause, then the bleeding must be controlled, and so forth. Medications should be reviewed in detail, with
specific attention to tranquilizers and sedatives and to the adverse effects of diuretics. When in doubt,
any potential contributing medicines should be discontinued.
TABLE 79-4 -- Treatment of Hepatic Encephalopathy
1. Identify and correct the precipitating cause(s).
a. Monitor volume status and vital signs.
b. Assess for gastrointestinal bleeding.
c. Eliminate sedatives, tranquilizers, or similar drugs.

d. Perform screening tests for hypoxia, hypoglycemia, anemia, hypokalemia, and other potential
metabolic or endocrine factors and correction as indicated.
2. Initiate ammonia-lowering therapy.
a. Perform nasogastric lavage and administer lactulose other cathartics or enemas to remove the
source of ammonia from the colon.
b. Minimize or eliminate dietary protein.
c. Initiate treatment with lactulose or lactitol to produce two to four bowel movements per day.
d. Consider antibiotics to reduce intestinal bacterial counts.
e. Consider flumazenil and other benzodiazepine receptor antagonists (see text).
3. Minimize the potential complications of cirrhosis and depressed consciousness.
a. Provide supportive care with attention to airway and hemodynamic and metabolic status.
Moreover, general measures to correct and maintain glucose, oxygenation, and acid-base balance are
essential.
The second step in treatment is directed toward lowering elevated ammonia levels. This involves
removing the source of the ammonia from the intestinal tract, trapping ammonia in the colon to prevent
systemic absorption, and, in some patients, specific therapy to decrease the number of
ammonia-producing bacteria in the colon.
In patients with gastrointestinal bleeding, removing the source of ammonia involves elimination of blood
from the gastrointestinal tract. For hemorrhage in the upper intestine, nasogastric lavage to remove
blood, and initiation of lactulose or other cathartics to speed the transit through the colon are appropriate.
In more chronic encephalopathy not associated with bleeding, excessive protein ingestion or constipation
may elevate ammonia to levels sufficient to cause encephalopathy. [24] The same treatment principles
apply, including a decrease in dietary protein to about 60 g/day, particularly if lactulose or other
measures to decrease ammonia production in the colon are, by themselves, ineffective. In addition, there
appears to be some advantage to substituting vegetable protein for other protein sources because of a
lower rate of ammonia production. [38] In severe encephalopathy, dietary protein should be eliminated
until there is sufficient improvement to allow institution of a stable therapeutic regimen.
The synthetic disaccharides lactulose (1,4-galactosidofructose) [6] and lactitol (beta-galactosidosorbitol)
[39] represent the mainstays of medical therapy of nitrogenous hepatic encephalopathy. These agents
target the production and absorption of ammonia and benzodiazepine receptor ligands in the gut.
Lactulose was introduced about 30 years ago as a therapy for hepatic encephalopathy based on the
concept that it would acidify the colon contents and favor trapping of ammonium in the lumen and
prevention of absorption. In the colon, lactulose is metabolized by bacteria to release lactic, acetic, and
other organic acids, decreasing stool pH to about 5.5. [6] It is clinically effective in over 80% of patients,
leading to a decrease in serum ammonia levels and an improvement in encephalopathy. [6] [14] [39]
Treatment is well tolerated, and the principal toxicity is abdominal cramping, diarrhea, and flatulence.
When orally administered to normal adults in amounts up to 160 g/day, lactulose decreases fecal
ammonia production and increases fecal nitrogen excretion approximately four-fold, owing to the

increase in stool volume. [23] Thus, an increase in bowel movements to two to four soft stools per day is
an important therapeutic goal.
Several clinical trials have evaluated the relative efficacy of lactulose versus lactitol in treatment of
hepatic encephalopathy. In general, these drugs are equally effective, but there is a trend toward better
palatability and fewer side effects with lactitol. [39]
Antibiotics of several types including neomycin, ampicillin, and rifaximin are also effective in lowering
blood ammonia levels [6] [14] [40] The effect on ammonia is due in large part to a decrease in the number of
colonic bacteria and a concomitant decrease in bacterial urease and protease activity, the main enzymes
responsible for ammonia generation. [40] In addition, decreasing colonic bacteria appears to decrease the
production of benzodiazepine receptor ligands. [32] In most patients, the response to antibiotics is
equivalent to lactulose, and in small series it may be associated with improved patient compliance. [6] [14]
However, nonspecific use of antibiotics in the absence
1340
of an established or suspected infection raises certain concerns. Neomycin, for example, can be absorbed
systemically in concentrations sufficient to induce ototoxicity and nephrotoxicity, particularly when
given over longer periods; and the alterations in gut flora associated with antibiotic use can contribute to
diarrhea, malabsorption, and staphylococcal and other overgrowth syndromes. Thus, chronic therapy
with antibiotics should be reserved for those who cannot tolerate oral lactulose or lactitol therapy, and
neomycin should be avoided.
Treatment of hepatic encephalopathy in the absence of elevated ammonia levels follows the same
principles, including a careful review for use of sedatives or analgesics. Prolonged recovery from
sedatives given for endoscopy or other procedures is characteristic. In addition, initiation of
ammonia-lowering steps appears to be effective, perhaps because of the effects on GABAergic
transmission [1] [32] or other agents that are derived from the colon and contribute to the clinical
syndrome.
The role of flumazenil and other benzodiazepine receptor antagonists in treatment of hepatic
encephalopathy is not yet defined. [27] In clinical trials, evidence for [33] [34] and against [35] clinical benefit
has been presented. Even when the response was favorable, the recovery was rarely complete and was
short-lived, owing to the pharmacokinetic properties of the drug. Thus, much remains to be learned
regarding the origin, overall contribution, and therapy for increased GABAergic transmission in hepatic
encephalopathy. At present, therapy with flumazenil should be limited according to guidelines of Jones
and others, which include (1) reversal of effects of exogenous benzodiazepines, (2) aiding the differential
diagnosis of encephalopathy, and (3) providing information about prognosis and optimizing brain
function in this disorder. [1] [27] [29] These indications are likely to evolve with additional clinical
experience and the development of more selective and effective analogs.
There is a long history of clinical trials involving other more experimental approaches to treatment of
hepatic encephalopathy. These include but are not limited to use of dietary zinc, levodopa,
branched-chain amino acids, and charcoal hemoperfusion. [1] [2] Although anecdotal reports have been
encouraging, none has been of sufficient established benefit to achieve widespread application.



April 22, 2000
Conde Petra
Citation
Bibliographic Data
Indexing Data
Cordoba J - Am J Gastroenterol - 1997 Sep; 92(9): 1429-39
NLM Citation ID:
97461377
Full Text
Full Source Title:
Frontmatter American Journal of Gastroenterology
INTRODUCTION Publication Type:
BASIS OF THERAPY Journal Article; Review; Review, Tutorial
The gut Language:

English
Bowel cleansing.
Author Affiliation:
Diet. Department of Medicine, Veteran's Affairs Lakeside Medical
Nonabsorbable disaccharides. Center and Northwestern University, Chicago, Illinois 60611,
USA.
Antibiotics.
Authors:
The liver Cordoba J; Blei AT
The blood-brain barrier Number of References:
Branched chain amino acids. 136
The brain Major Subjects:

Hepatic Encephalopathy / Etiology / Physiopathology / *
Flumazenil.
Therapy
Bromocriptine.
Future avenues. Blood-Brain Barrier
TREATMENT OF SPECIFIC TYPES Brain Diseases / Physiopathology
Chronic Disease
Encephalopathy in fulminant
hepatic failure Human
Intestinal Diseases / Complications / Physiopathology

Acute encephalopathy in cirrhosis
Liver Diseases / Complications / Physiopathology
Chronic encephalopathy Support, Non-U.S. Gov't
Subclinical encephalopathy Support, U.S. Gov't, Non-P.H.S.
Problematic encephalopathy
DISCUSSION
ACKNOWLEDGMENTS
REFERENCES


Volume 92 Number 9 September 1997
P1429
CLINICAL REVIEWS
Treatment of Hepatic Encephalopathy
Juan Cordoba M.D.

Andres T. Blei M.D.
Department of Medicine, Veteran's Affairs Lakeside Medical Center and Northwestern University,
Chicago, Illinois
Reprint requests and correspondence: Andres T. Blei, M.D., VA Lakeside Medical Science Building, 400 E. Ontario,
Room 127, Chicago, IL 60611.
Received Dec. 18, 1996;
accepted Apr. 14, 1997.
INTRODUCTION
Hepatic encephalopathy may be defined as a disturbance in central nervous system function due to hepatic insufficiency.
This vague definition reflects the existence of a spectrum of neurological manifestations [1] , still incompletely
characterized [2] . Their common link is their association with severe forms of liver disease and potential reversibility. In
practice, the most distinctive presentation is the development of an acute confusional state that can evolve into coma (acute
encephalopathy). In patients with fulminant hepatic failure, acute encephalopathy can be complicated with brain edema,
which is rarely seen in cirrhosis [3] . In patients with chronic liver disease, acute encephalopathy is commonly associated
with a precipitating factor that dictates the course of encephalopathy. Recurrent episodes can occur in the absence of
precipitating factors (chronic recurrent encephalopathy), or neurological deficits may not completely reverse (chronic
persistent encephalopathy). However, the most frequent neurological disturbance is not evident on clinical examination:
mild cognitive abnormalities only recognizable with psychometric tests (subclinical encephalopathy).
The development of hepatic encephalopathy carries important prognostic implications. In patients with fulminant hepatic
failure, the appearance of cerebral edema can result in brain herniation and death [4] . In the case of cirrhosis, the stage of
encephalopathy is one of five elements comprising the Child-Pugh's classification, an index of prognosis [5] . Multiples
modalities have been used for the treatment of hepatic encephalopathy. However, their efficacy has been infrequently
assessed by well-designed randomized clinical trials. This gap reflects the difficulties in evaluating a drug's effect on
clinical outcome. Hepatic encephalopathy includes a wide range of neuropsychiatric signs that may be difficult to evaluate
independently. Alteration of consciousness--its most relevant manifestation--is hard to monitor and can be influenced by
concurrent factors such as infection, hypoxia, bleeding, or electrolyte disturbances. In addition, experimental models to test

new agents do not precisely mimic the clinical syndrome [6] . In spite of these limitations, a critical reappraisal of available
data renders it possible to devise a rational approach to the treatment of hepatic encephalopathy.
BASIS OF THERAPY
Different hypotheses have been proposed to explain the changes in mental state seen in patients with liver disease [7] . A
common notion is that substances derived from the gut produce deleterious effects on brain function. A large body of data
continues to point to a key role for ammonia in the pathogenesis of encephalopathy [8] . Although ammonia is released
from several tissues, most of it enters the circulation from the gastrointestinal tract [9] . Ammonia derived from colonic
bacteria and from the deamidation of glutamine in the small bowel, is absorbed via passive diffusion and undergoes a high
first-pass hepatic extraction ( ca. 0.8) [1] [10] . Liver failure and/or portosystemic shunting lead to a rise in arterial levels
and a higher rate of exposure of the brain to ammonia [11] . There have been alternative explanations regarding the nature
of the gut-derived putative toxins--among them, the existence of bacterial precursors of benzodiazepine-like substances [12]
and other products of bacterial metabolism in the colon, such as neurotoxic short chain fatty acids, phenols, and
mercaptans [13] .
In patients with liver disease, these substances reach the systemic circulation as a result of portosystemic shunting and/or
reduced hepatic clearance. Abnormalities at the level of the blood-brain barrier may facilitate their entrance into the brain.
In this regard, blood-brain barrier permeability to ammonia has been shown to be increased in patients with hepatic
encephalopathy [11] . This factor may explain the imperfect correlation between plasma ammonia and hepatic
encephalopathy [14] and exemplifies the limitation of implicating specific toxins on the basis of their plasma levels. Once
in neural tissues, a large number of neurochemical changes occur that affect multiple neurotransmitter systems, each
affected to a variable extent. Their individual role in the development of neurological manifestations is obscured by a
limited knowledge of the neurobiological basis of behavior. A complete description of these abnormalities and their
implications are beyond the scope of the present article; the reader is referred to recent reviews [15] [16] . From a
histological perspective, the lack of histological signs of neuronal death supports the notion of reversibility. The distinctive
morphological alteration is the Alzheimer type II astrocyte change. This has lead to the interpretation that abnormalities in
neurotransmission occur after injury to glial cells [17] .
P1430
Figure 1. Hepatic encephalopathy can be conceptualized as an alteration of multiple neurotransmitter systems (the figure includes only the most
relevant) that arises from the effect of gut-derived substances. In patients with liver disease, these substances reach the brain as a result of
portosystemic shunting and reduced hepatic clearance. Abnormalities at the level of the blood-brain barrier may facilitate the development of
disturbances of neurotransmission. In agreement with this notion, treatment for hepatic encephalopathy can be classified according to it main site of
action: 1) gut, 2) liver, 3) blood-brain barrier, and 4) brain. All tested agents are enclosed in boxes; inclusion in the figure does not necessarily
imply clinical efficacy. Smaller fonts indicate candidate toxins that are considered of lower physiopathological relevance. GABA,
gamma-aminobutyric acid neurotransmitter system; DA, dopamine; GLU, glutamate; 5HT, serotonin; BBB, blood-brain barrier; SCFA, short chain
fatty acids; BZD, benzodiazepines; BCAA/ AAA, branched chain/aromatic amino acids; +, activation/agonist; -,antagonist.
Irrespective of the specific mechanisms involved in the development of hepatic encephalopathy, it is possible to envision
four different sites of derangement and four general aims of therapy (Fig. 1) : 1) gut: decrease the access of toxins to the
systemic circulation, 2) liver: increase the clearance of such toxins, 3) blood-brain barrier: interfere with their entry into
brain, 4) brain: counteract abnormalities of neurotransmission. We have classified available therapies according to this
view, which can help in the selection of a drug and the design of future therapeutic strategies. In a second section, we
propose particular therapeutic approaches based on the clinical scenario in which hepatic encephalopathy develops.
The gut
Clinical experience has taught that the gut is the main source of substance(s) that cause hepatic encephalopathy.
Encephalopathy can be precipitated by constipation or by a large oral protein load (through the diet or after gastrointestinal
hemorrhage). Different therapeutic modalities are directed at the gut. Their most obvious effect is interference with the
formation and/or absorption of nitrogenous compounds arising from bacterial metabolism in the colon, especially

ammonia, an effect that has been used in multiple studies as a surrogate marker of clinical efficacy.
Bowel cleansing.
If the toxins responsible for hepatic encephalopathy arise from the gut, it follows that bowel cleansing should be a
mainstay of therapy. Furthermore, a recent study suggests that hepatic encephalopathy might be associated with slow
intestinal transit and, through this mechanism, might amplify the generation of encephalopathic substances from the gut
[18] . Unfortunately, only patients with mild encephalopathy were studied, and the presence of intestinal bacterial
colonization, which may have affected the measurement of the orocecal transit, was not excluded. Colonic cleansing
reduces the luminal content of ammonia [19] , decreases bacterial counts [20] , and lowers blood ammonia in cirrhotic
patients [19] , all regarded as beneficial effects. Various laxatives can be used, but nonabsorbable disaccharides are
preferred, because they result in additional effects that potentiate the elimination or reduce the formation of nitrogenous
compounds [21] [22] (see below). Bowel cleansing can also be achieved after irrigation of the gut via a peroral tube. By this
approach, irrigation with an isotonic solution of mannitol has been proposed to prevent encephalopathy after a
gastrointestinal hemorrhage [23] .
Diet.
Restriction of dietary protein with subsequent progressive increments to assess clinical tolerance is a classical cornerstone
of therapy. However, protracted nitrogen restriction may be harmful, as witnessed by a poor short-term prognosis with
malnutrition [24] . Furthermore, a positive nitrogenous balance may have positive effects on encephalopathy [25] by
promoting hepatic regeneration and increasing the capacity of muscle to detoxify ammonia [26] . Nonetheless,
improvement of nutritional status in patients with encephalopathy is difficult. A high protein intake (1.2 g/kg/day) may be
necessary to maintain nitrogen balance [27] . Modifying the composition of the diet and increasing its calorie:nitrogen ratio
may improve tolerance to protein. At isonitrogenous levels, vegetable and dairy products cause less encephalopathy than
meat [28] [29] . Differences in amino acid composition [30] and in the ratio of carbohydrates to total protein could explain
these effects [31] . A high calorie-to-nitrogen ratio, characteristic of casein-based and vegetable diets, reduces
gluconeogenesis and has anabolic effects on the utilization of dietary proteins [32] . The benefits of vegetable-based diets
have also been related to the presence of nonabsorbable fiber that is metabolized by colonic bacteria. Fiber increases the
elimination of nitrogen products in stool [33] , probably through a mechanism similar to that of nonabsorbable
disaccharides. If the increased intake of fiber induces flatulence, soluble forms of fiber may
P1431
be better tolerated than vegetable-based diets [34] . When protein intake is withheld for a prolonged period, standard amino
acid formulations, up to 70 g/day, can be a safe way to satisfy nitrogen requirements [35] . Enteral administration is
preferable to the parenteral route because it lessens volume overload and risks of infection [36] . Nonabsorbable
disaccharides.
Lactulose (beta-galactosido-fructose) was first introduced with the idea of promoting the growth of Lactobacillus bifidus,
which contains little urease activity and, via this mechanism, decreasing the production of ammonia in the colon [37] . This
is not its mechanism of action, still complex and incompletely understood. However, the bulk of evidence links its efficacy
to an interaction with the enteric flora and to a decrease in the generation of nitrogenous compounds in the gut [38] . The
effects of lactulose on nitrogen metabolism include an increase in fecal nitrogen excretion by facilitating the assimilation
of ammonia into bacteria, as well as by a cathartic effect [39] and a decrease in the amount of nitrogen that reaches portal
blood. As a result, plasma levels of ammonia and the total body pool of urea decrease [33] [37] , effects that are beyond an
exclusively cathartic effect [21] .
Administered orally, lactulose is not broken down by intestinal disaccharidases and reaches the cecum where it is
metabolized by enteric bacteria to lactate and acetate [40] , causing a drop in cecal pH [41] . Adequate dosing results in
acidification of the colon; stool pH (aim, pH < 6) can be used to monitor therapy. The dose--orally or via a nasogastric
tube--is titrated to produce two to three soft bowel movements daily. Patients in coma or with small bowel ileus can
receive lactulose via enema [42] . The oral dose should be adjusted during chronic treatment because of individual
variability and changes in the bacterial metabolism of disaccharides [40] . An excessively sweet taste, flatulence, diarrhea,
and abdominal cramping are the most frequent subjective complaints. If diarrhea develops, the drug should be stopped and
reinstituted later at a lower dose. Protracted diarrhea may result in hypertonic dehydration with hypernatremia [43] .

A similar mechanism of action is shared by other nonabsorbable disaccharides that are metabolized by the colonic flora [44]
. Comparisons of lactitol (beta-galactosido-sorbitol) to lactulose in randomized trials show a similar efficacy [45] and better
palatability [46] ; however, lactitol is not currently available in the United States. Lactose can be administered to patients
with known lactase deficiency [47] .
Antibiotics.
Antibiotics affect the process by which gut toxins are generated. Benefits from neomycin, the most widely tested drug [48]
[49] [50] , generally are attributed to its effects on the generation of ammonia by intestinal bacteria [51] . However, neomycin
causes many alterations in intestinal mucosal metabolism and may decrease blood ammonia via a nonbacterial effect [52] .
It is intriguing that metronidazole and vancomycin, though affecting bacterial populations other than neomycin, can also
improve encephalopathy [53] [54] . In recent years, gastric colonization by Helicobacter pylori, a urease-containing
organism, has emerged as a possible source of hyperammonemia and hepatic encephalopathy [55] . In patients with acute
alcoholic hepatitis, a positive serological test for H. pylori has been identified as an independent risk for the development
of encephalopathy [56] . However, limited data suggest that eradication of H. pylori is not associated with a decrease in
plasma ammonia [57] .
Antibiotics are not usually recommended for prolonged periods because of associated toxicity. If they are administered,
toxicity can be minimized by tapering the dose after clinical response ( i.e., neomycin 6 g/day during 2-3 days, followed by
1-2 g/day thereafter) [58] . In spite of its poor absorption (<4%) [59] , orally administered neomycin can cause auditory loss
and renal injury. However, these are rare complications, particularly if renal function is not impaired. The most important
adverse effects are intestinal malabsorption and superinfection [60] , which may manifest as sprue-like diarrhea. The effect
of neomycin on absorption is independent of its antibacterial effects [61] . Mild malabsorption of a variety of nutrients and
drugs occurs with doses as little as 3 g/day [62] . The effect of long-term therapy is unclear. Periodic control of auditory and
renal function, special nutritional care, and dose adjustment of additional drugs is recommended. Similarly, adjustment of
the initial dose of metronidazole, a drug that undergoes hepatic elimination [63] , is advised to decrease toxicity [64] .
The liver
An increase in the capacity of a diseased liver to clear the putative toxins is a desirable goal, but is difficult to attain. The
presence of portosystemic shunting, loss of liver mass, and alterations of the intrahepatic circulation limit ability to
manipulate the metabolic capacity of the cirrhotic liver. Activation of the urea cycle has been pursued as a measure to
reduce blood ammonia. Ornithine-aspartate, still under clinical evaluation, provides substrates both for ureagenesis and for
synthesis of glutamine, the two hepatic mechanisms that remove ammonia from the portal blood. The drug appears to
prevent the rise of blood ammonia after a nitrogenous load [65] . Unlike other drugs, ornithine-aspartate has the potential
advantage that it can be administered intravenously. Zinc, a cofactor of urea cycle enzymes, is frequently deficient in
cirrhosis as a result of increased urinary excretion and malnutrition [66] . Zinc supplementation (600 mg/day) has been
proposed to reduce blood ammonia and treat hepatic encephalopathy [67] . However, clinical results conflict [68] [69] [70] .
Furthermore, zinc could be effective at other target sites, as it is also involved in neurotransmission [71] .
Alternative pathways for nitrogen excretion, such as drugs used in children with urea cycle enzyme deficiencies, have also
been tested. In the liver, benzoate is conjugated with glycine to form hippuric acid, and phenylacetate is conjugated with
glutamine to form phenylacetylglutamine. Urinary excretion of these conjugates results in the removal of one and two
nitrogen atoms per molecule of drug, respectvely.
P1432
Benzoate is as efficacious as lactulose for the treatment of acute encephalopathy [72] . A note of caution is warranted with
this approach to clinical testing, because lactulose has not been shown conclusively to be superior to the removal of the
precipitating factor. Phenylacetate is malodorous but can be administered as its precursor, phenylbutyrate. A preliminary
therapeutic trial is difficult to interpret, because patients were being treated with other drugs [73] . The efficacy of these
drugs may be limited in patients with advanced liver failure by the inability to carry out the conjugation step.
The blood-brain barrier

Alterations in the permeability of the blood-brain barrier have been implicated in the pathogenesis of hepatic
encephalopathy [74] [75] and in the development of brain edema [76] . However, the experimental observations that have
linked abnormalities of brain neurotransmission to increased permeability of the blood-brain barrier have been questioned
in recent human studies [77] . Furthermore, brain edema in fulminant hepatic failure appears to arise initially from swelling
of astrocytes (cytotoxic edema) rather than from increased permeability of the blood-brain barrier (vasogenic edema) [4] .
Branched chain amino acids.
Patients with cirrhosis exhibit an imbalance in the plasma amino acid profile characterized by a decrease in branched chain
(valine, leucine, isoleucine) and an increase in aromatic amino acids (tyrosine, phenylalanine, tryptophan). It has been
proposed that this imbalance enhances the passage of aromatic amino acids, via a neutral amino acid carrier, into the brain
in exchange for glutamine generated from ammonia detoxification in astrocytes [74] . The excess of aromatic amino acids
would then be channeled in the brain to the synthesis of false neurotransmitters (octopamine, phenylethanolamine) and to
serotonin, an inhibitory neurotransmitter. Infusions and oral administration of branched chain amino acids theoretically
would correct the amino acid profile and reduce the entrance of the potentially toxic aromatic amino acids into the brain.
Although much controversy has arisen from clinical trials using branched chain amino acids, critical reviews of the
published studies [35] [78] [79] indicate no major beneficial effect, either in acute or chronic encephalopathy. Oral
formulations of branched chain amino acids may provide a source for protein in patients with chronic encephalopathy and
dietary protein intolerance [80] .
The brain
Hepatic encephalopathy is best understood as an abnormality of multiple neurotransmitter systems rather than as result of a
primary decrease in energy metabolism [15] . Alterations of such systems have been observed in experimental models and
in the brain of cirrhotic patients dying in hepatic coma. Because the number of abnormalities is large, animal studies have
been critical to evaluate potential therapeutic targets. From this experience, alterations in both inhibitory
[gamma-aminobutyric acid (GABA), opioid, serotonin] and excitatory neurotransmission (glutamate, catecholamines)
have been reported. However, few drugs have been tested clinically for their direct effect on the central nervous system.
Flumazenil.
It has been proposed that an enhanced GABA-ergic tone is present in hepatic encephalopathy [81] . Increased sensitivity to
agonists of the GABA receptor complex, such as barbiturates and benzodiazepines (BZD), is well recognized [82] . Some
patients with hepatic encephalopathy show an increase in the concentration of BZD-like substances that bind to the BZD
site of the GABA receptor and enhance its inhibitory action. Some of these substances have been identified as diazepam
and desmethyldiazepam, but the nature of the majority remains elusive. The low concentration of BZD receptor ligands
(equivalent to anxiolytic doses of diazepam) found in the brain of patients in deep stages of encephalopathy [83] suggests
that an increase in these compounds is not sufficient to explain the whole clinical syndrome. Flumazenil--a highly selective
antagonist of the BZD receptor that has been tested in encephalopathic subjects--may transiently improve the mental state
when administered intravenously to a subset of patients (25-50%) [84] [85] [86] [87] ; a source of controversy is whether such
patients had received exogenous benzodiazepines. Nevertheless, antagonists of the GABA receptor complex have resulted
in ameliorations of hepatic encephalopathy in some animal models that have not received pharmaceutical benzodiazepines
[88] [89] . New antagonists, chemically related to flumazenil but with slightly different pharmacokinetic and
pharmacodynamic properties, may be more effective for the treatment of encephalopathy [90] .
Bromocriptine.
Drugs that enhance dopaminergic neurotransmission were introduced to restore the displacement of central
neurotransmitters caused by putative false neurotransmitters. Although this hypothesis has been contested by subsequent
experimental observations [91] , recent evidence supports the existence of dopaminergic dysfunction that may arise from
the accumulation of manganese in basal ganglia [92] [93] . When targeted to improve consciousness, dopaminergic drugs
(levo-dopa, bromocriptine) did not yield satisfactory results [94] . However, they may have a role in the treatment of
extrapyramidal manifestations in patients with chronic encephalopathy. In these subjects, improvements of extrapyramidal
signs have been reported when bromocriptine was added to conventional therapy [95] . The constipation caused by
bromocriptine may be counteracted by an increased dose of nonabsorbable disaccharides.
Future avenues.
The development of drugs with the ability to correct abnormalities of neurotransmission has been one of the most exciting

advances of recent years. Encouraging results have been reported in experimental models of hepatic encephalopathy with
drugs intended to correct abnormalities of the glutamanergic [96] , serotoninergic [97] , and opioid [98] systems. High levels
of extracellular glutamate, quantitatively the main neurotransmitter in the brain,
P1433
TABLE 1 -- Management of Precipitating Factors

Precipitating factor Diagnostic Approach Management
Gastrointestinal bleeding Exam of rectal and Specific treatment
gastric content according to site of
hemorrhage
Endoscopy Bowel cleansing
Constipation Clinical history Bowel cleansing
Large protein meal Clinical history Bowel cleansing
Psychoactive drugs Clinical history Antidotes ( i.e.,
flumazenil, naltrexone)
Drug screen
Renal failure Renal function tests Withhold diuretics and
nephrotoxic drugs
Ultrasonography Specific treatment of the
cause
Electrolyte imbalance Blood electrolytes Withhold diuretics
Correct electrolyte
disturbances
Infection Cultures of blood and If suspicious,
body fluids broad-spectrum
antibiotics (avoid
aminoglycosides).
Adjust antibiotics to
microbiological studies.
Tap of ascites and
pleural effusion
Superimposed acute hepatic injury Clinical history Specific measures may
be helpful according to
cause
Liver chemistries,
ultrasonography

Liver biopsy
This table is presented as a guide for the recognition and management of the most common
precipitating factors; it does not include an exhaustive description of the diagnostic and therapeutic
procedures.
have been observed consistently in hepatic encephalopathy [99] . In a recent experimental study, memantine, an n-methyl-
d-asparate receptor (glutamate subtype receptor) antagonist, ameliorated hepatic encephalopathy and brain edema [96] .
There has been a report of similar beneficial effects produced by counteracting high extracellular glutamate in other
neurological conditions, such as brain trauma and brain ischemia [100] , and drugs that decrease the release of glutamate by
neurons have been introduced for the treatment of neurological diseases [101] . Although new psychotropic drugs for
hepatic encephalopathy are still at preclinical stage, they are the focus of current research and in the future may result in
valid therapeutic alternatives.
TREATMENT OF SPECIFIC TYPES

Encephalopathy in fulminant hepatic failure
In patients with fulminant hepatic failure, therapy is directed at counteracting multiorgan failure as well as supporting the
impaired liver function [102] . The severity of hepatic encephalopathy is a critical criterion in judging the need for
emergency liver transplantation [103] ; any possible factor affecting mental state, such as hypoglycemia, pharmacological
sedation, or intracranial hemorrhage, needs to be excluded. Although therapy for encephalopathy in fulminant hepatic
failure has not been formally evaluated, treatment is similar to that for acute encephalopathy in cirrhosis (protein
restriction and nonabsorbable disaccharides) but appears to be only marginally effective. Because patients are
hypercatabolic, monitoring of urinary urea nitrogen may assist in determining protein needs, which should be at least 0.5
g/kg body weight. Lactulose enemas may provide a quicker way to determine whether disaccharides are effective.
Patients are managed in an intensive care unit. Tracheal intubation is recommended at the onset of grade III coma to
minimize the risk of sudden respiratory arrest. Intracranial hypertension secondary to brain edema, a frequent complication
in patients with grades III-IV encephalopathy, can result in the patient's demise or in irreversible brain damage.
Management of intracranial hypertension is facilitated by monitoring with intracranial pressure transducers; epidural
transducers are the safest [104] . However, measurement of intracranial pressure does not appear to increase survival [105] .
The procedure should be reserved for patients considered for liver transplantation, in whom management of intracranial
hypertension is critical. General nursing measures to prevent rises in intracranial pressure should be established [106] .
Sustained elevations (>25 mm Hg for 5 min) or the development of pressure waves should be treated with mannitol (0.5-1
mg/kg i.v. over 30 min). If mannitol is ineffective, infusion of barbiturates can be considered; corticosteroids are
ineffective.
Acute encephalopathy in cirrhosis
The development of acute encephalopathy in a cirrhotic patient is usually associated with a precipitating factor. On
occasion, an acute change in mental state may develop spontaneously in patients with advanced cirrhosis or with large
spontaneous portosystemic shunts. Treatment of acute encephalopathy is based on the vigorous correction of the
precipitating factor (Table 1) , the implementation of nutritional measures, and the administration of nonabsorbable
disaccharides.
Dietary protein is withheld and carbohydrates are administered (>1500 kcal/day). As the clinical situation improves,
protein is introduced in increments of 10-20 g/day every 2-3 days. Patients with prolonged coma should receive nitrogen
supplements as solutions of amino acids, preferably via the enteral route [36] . Preparations of branched chain
P1434

TABLE 2 -- Controlled Trials of Drugs * for Acute Encephalopathy in Cirrhotic Patients

No. of
Treatment Comparison Trial Patients * Time * Effect
Lactulose Placebo Simmons et al.[109] 21 10 days +
(glucose)
Lactitol Lactulose Morgan et al.[110] 28 5 days =
Lactitol Lactulose Heredia et al.[111] 40 5 days =
Lactitol/lactose Cleansing 20 To clinical +
enemas enemas Uribe et al.[42] improvement
Lactitol enemas Lactose enemas Uribe et al.[42] 40 4 days =
Lactose enemas + Neomycin + Uribe et al.[22] 18 5 days =
placebo starch enemas
Neomycin Placebo Strauss et al.[112] 39 To maximal =
improvement
Neomycin + sorbitol Lactulose Atterbury et al.[49] 45 To maximal =
improvement
Neomycin + lactulose Placebo Blanc et al.[129] 80 5 days =
Sodium Benzoate Lactulose Sushma et al.[72] 74 To maximal =
improvement
Flumazenil Placebo Pomier-Layrargues 21 1h +
et al.[84]
Flumazenil Placebo Cadrenal et al.[86] 14 10 min +
Flumazenil Placebo 49 8h =
Gyr et al.[87]
* Branched chain amino acids not included.
** Total number of patients included in the study.
*** Period of time from initiation of treatment to measurement of outcome.
Effect on encephalopathy, + lower degree of encephalopathy or more rapid clinical improvement with treatment; =, no differences between treatment and control.
Analysis of the 20 first patients of the study.
All patients received lactulose.
Only patients with encephalopathy grade I-III included. Additional treatment with lactulose was permitted. Twenty-four patients were excluded from the
analysis. For patients with clinically relevant improvement (n = 5), flumazenil was associated with a better response.
amino acids do not appear to offer additional benefits over standard solutions [107] .
Lactulose has become the standard drug for the treatment of acute encephalopathy. This procedure was criticized [108] , the

effectiveness of lactulose being compared to that of placebo in only one study that included 21 patients [109] (Table 2) .
Pooled analysis of the results of other controlled studies is not possible because of methodological differences; any critical
reappraisal of the use of lactulose for acute encephalopathy would require new clinical trials. Until then, the safety profile
and large clinical experience with lactulose justifies its use [50] . Initially, patients should receive a large oral dose of
lactulose or another nonabsorbable disaccharide [110] [111] ( i.e., 50 ml of lactulose syrup every 1-2 h) or as an enema (300
ml lactulose in 1 L water). After catharsis begins, the oral dose should be adjusted (15-30 ml lactulose q.i.d.), or a regimen
of lactulose enemas should be prescribed ( i.e., every 6-8 h).
Neomycin exhibits efficacy similar to that of lactulose [49] but is rarely used as a first line of therapy because of its
potential toxicity. A recent double-blind randomized study [112] compared neomycin to placebo in patients with acute, but
mild encephalopathy ( ca. 70% stage I-II encephalopathy). Neomycin did not significantly shorten the time of regression
to a normal mental state; however, the duration of this period was quite variable, highlighting the difficulties in performing
clinical trials in such patients, as the course of the precipitant event cannot be fully controlled during randomization.
Flumazenil is a safe drug that can improve the mental status in a subset of individuals in whom "natural" or "endogenous"
benzodiazepines might participate in the development of encephalopathy. In a double-blind randomized cross-over study
[84] including a selected group of 17 cirrhotics in stage IV encephalopathy without comorbid factors, flumazenil improved
the neurological condition of six of 13 patients (46%). In most patients, the response to flumazenil occurred within a few
minutes after the bolus and was not related to detectable plasma benzodiazepines. However, the improvement was
incomplete and transient, and no differences between placebo and flumazenil were seen 24 h after starting therapy that
included lactulose for both groups. Similar characteristics in the responses to flumazenil have been reported in cirrhotic
subjects in three additional randomized trials [85] [86] [87] . In practice, because of the high consumption of pharmaceutical
benzodiazepines and the impossibility to identify a priori those patients who may respond, a therapeutical trial can be
attempted. Flumazenil is currently available as an intravenous preparation that is administered as a bolus (0.4-2.0 mg). If a
favorable response occurs, typically within a few minutes after administration, additional doses can be given. The effects
of multiple doses of flumazenil has not been evaluated. The elimination half-life (normal individuals ca. 45 min) may be
doubled in cirrhotics [113] .
Chronic encephalopathy
Patients with chronic encephalopathy should be instructed to follow a protein-restricted diet and to use nonabsorbable
disaccharides. The tolerance to protein may be improved by dairy products and vegetable-based diets [28] . Oral branched
chain amino acids are an alternative not
P1435
TABLE 3 -- Controlled Trials of Drugs * for Chronic Encephalopathy

No. of
Treatment Comparison Trial Patients Design Effect
Lactulose Placebo Elkington et 7 cross-over +
(sorbitol) al.[114]
Lactulose Placebo German et 18 parallel
(saccarose) al.[115]
Lactitol Lactulose Blanc et 77 meta-analysis =
al.[45]
Lactitol Lactulose Camma et 72 meta-analysis =
al.[46]

Neomycin + sorbitol Lactulose Conn et 33 cross-over =

al.[116]
Neomycin + MgSO4 Lactulose Orlandi et 173 parallel =
al.[48]
Zinc + lactulose Lactulose Bresci et 90 parallel =
al.[70]
Zinc + lactulose Placebo + Reding et 22 parallel
lactulose al.[67]
Zinc + lactulose Placebo + Riggio et 15 cross-over =
lactulose al.[68]
Bromocriptine Placebo Uribe et 7 cross-over =
al.[94]
Bromocriptine + lactulose Placebo + Morgan et 6 cross-over +
lactulose al.[95]
* Branched chain amino acids not included. Four additional controlled trials, including less than 10 patients, compared neomycin and lactulose [50] .
Effect on encephalopathy: +, clinical improvement with treatment; , treatment associated with improvement in psychometric tests; =, no differences between
treatment and control.
usually recommended because of cost/benefit considerations. In one study including 64 cirrhotics with mild forms of
chronic encephalopathy [80] , treatment with oral branched chain amino acids was associated with a decreased number of
acute exacerbations and with nutritional improvement. Treatment was maintained for 3 months and was complemented
with lactulose and a limited protein intake (45-65 g/day). However, other studies with a smaller sample size and of a
shorter duration have not yielded consistent results [79] .
In spite of limited data from controlled studies (Table 3) , patients with chronic encephalopathy are treated with
nonabsorbable disaccharides. For this indication, lactulose has been compared with placebo in two studies: clinical
improvements were observed in 10 of the 16 patients treated [114] [115] (62%). In addition, clinical improvement was noted
in 144 of 202 patients (71%) included in 11 controlled trials that compared lactulose to neomycin or lactitol [50] .
Neurological scores have also improved with lactulose in cross-over studies when treatment and wash-out periods were
compared [116] . Antibiotics should be reserved as an alternative for patients who have not responded to disaccharides, for
some cirrhotics showing a better response with them [54] , or for those rare subjects who neither develop diarrhea nor
acidify their stools in spite of high doses of lactulose [117] . Tests to monitor the toxicity of neomycin should be performed
periodically.
Subclinical encephalopathy
The most characteristic deficits in patients with subclinical encephalopathy are in motor and attentional skills [118] , which
may impact on their ability to perform daily activities [119] . However, many subjects are able to compensate for these
deficits, and consequences appear to be minor [120] [121] . Moderate improvement of neuropsychological tests has been
reported with nonabsorbable disaccharides [122] , vegetable protein diet [123] , branched chain amino acids [124] , and
flumazenil [125] . However, universal treatment is not recommended. A psychometric evaluation may be adequate in those
cirrhotic individuals whose occupations demand attentional and motor abilities. In patients with significant deficits, a
therapeutic program based on dietary manipulations and/or nonabsorbable disaccharides can be tried. Benefits of the
treatment, assessed by monitoring the neuropsychological response, should be weighed against secondary side effects.
Whether this approach can prevent the progression to overt encephalopathy has not been evaluated.

Problematic encephalopathy
Most patients with hepatic encephalopathy will respond to the correction of precipitating factors, protein restriction, and
appropriate doses of nonabsorbable disaccharides. Management of encephalopathy becomes problematic when, in spite of
the previously described measures, neurological manifestations do not improve or recur frequently. A thorough evaluation
should be performed to exclude other neurological disorders that may resemble hepatic encephalopathy, and occult
precipitating factors such as zinc deficiency should be sought. When a precipitating factor is not found, the exploration
should include Doppler ultrasonography to search for large spontaneous portosystemic shunts [126] , which can be
confirmed and treated with angiographic techniques [127] .
Other drugs with potential effects on mental state can be added successively to the standard measures (Table 4) . The
choice of drugs depends on the clinical setting as well as the consideration of different target sites (Table 5) . After the
addition of an antibiotic such as neomycin, stool pH should be monitored to ensure that colonic flora still metabolize the
disaccharides. In patients who maintain stool pH 6, lactulose and neomycin have additive effects in reducing gut
ammonia production [128] . The one randomized study of the effects of this combination on acute encephalopathy showed
no benefit, compared with placebo, but stool pH was not monitored [129] . Other measures that may provide benefit include
promotion of ammonia fixation with zinc supplementation [66] [69] or benzoate [72] . For patients with
P1436
TABLE 4 -- Measures to Consider in Patients with Problematic Encephalopathy (see text)

1 Search for "occult" precipitating factors
See Table 1
Assess:
Zinc deficiency (therapeutic trial?)
Large spontaneous portosystemic shunts
2 Additional drugs
See Table 5
Antibiotic (monitoring stool pH)
Promoters of ammonia metabolism (benzoate, zinc)
Flumazenil for decreased consciousness
Bromocriptine for extrapiramidal signs
3 Shunt occlusion/reduction for patients with shunt/TIPS
Consider risk of rebleeding vs. neurological manifestations
4 Liver transplantation for appropriate candidates
5 Surgical exclusion of the colon
Patients with relatively good liver function

Noncandidates for liver transplantation
TABLE 5 -- Drug Supplementation * for Problematic Encephalopathy

Target size Drug Clinical setting Dose
Gut Neomycin Acute [117] or chronic
[48] [116] 3-6 g/d
Metronidazole Chronic [53]

800 mg/d p.o.
Vancomycin Chronic [64] 2 g/d p.o.
Liver Zinc Acute [66] or chronic [67] 600 mg/d p.o.
Benzoate Acute [72] or chronic [73] 10 g/d p.o
Brain Flumazenil Acute [84] [86] 0.4-2.0 mg i.v.
Bromocriptine Chronic [95] 15 mg/d p.o.
* These drugs may be added to the treatment for patients with encephalopathy that does not respond to correction of the precipitating factor, protein restriction, and
administration of nonabsorbable disaccharides.
Dose reduction may be necessary to avoid toxicity [58] [63] .
depressed consciousness, flumazenil can be tried; some patients may benefit from long-term treatment [130] , but an oral
preparation is not commercially available. For those with severe extrapyramidal signs, bromocriptine may provide relief
[95] .
Persistent and intractable encephalopathy after portosystemic anastomosis can be treated by occluding the shunt [131] . In
patients who have undergone a transjugular intrahepatic portosystemic stent shunt, a prudent waiting time is warranted.
Most episodes of encephalopathy are concentrated during the first 2 months after the procedure and usually respond to
treatment with protein restriction and lactulose; subsequent narrowing of the stent--an untoward hemodynamic effect--may
afford protection from encephalopathy [132] . If encephalopathy becomes problematic, the stent diameter can be reduced
[133] . The risks of reintervention and rebleeding after shunt reversal should be weighed against the severity of the
neurological symptoms.
In appropriate candidates, management of encephalopathy should include an evaluation for liver transplantation. In
patients with chronic encephalopathy, the decision to proceed to liver transplantation can be difficult. Severe chronic
neuropsychological abnormalities are usually considered a contraindication for liver transplantation. However,
improvement of such manifestations have been reported after transplantation [134] [135] . In patients who are not candidates
for liver replacement, an alternative is to perform a surgical exclusion of the colon. This measure is reserved for the
occasional patient with relatively good liver function and severe neurological manifestations. Colonic bypass is preferred
because it is associated with a lower mortality than colectomy [136] .

DISCUSSION
During "the decade of the brain," as declared by the World Health Organization, our understanding of the pathogenesis of
hepatic encephalopathy has moved from a single unifying hypothesis of a gut-derived toxin to the realization that multiple
neurotransmitter systems are affected in the brain. Each one of these systems can be involved to a different extent in
different clinical scenarios. Treatment of encephalopathy is still based on "time-honored" measures that appear to decrease
the rate of exposure of the brain to the putative toxins. Optimization of these forms of therapies may result in better clinical
outcomes. However, the main focus of present and future research is the assessment of new drugs with the capacity to
correct alterations in neurotransmission. Such an approach is limited by an unsatisfactory explanation of which
neurotransmitter abnormalities underlie each one of the neurological manifestations, but it may assist investigators unravel
the pathogenesis of hepatic encephalopathy and provide better options for patients.
ACKNOWLEDGMENTS
This study was supported by a Merit Review from the Veterans Administration Research Service. Dr. Juan Cordoba was
supported by an educational grant from Asociacion Espanola para el Estudio del Higado.
We are indebted to Drs. Robert Craig and J. Donald Ostrow for their critical review of the manuscript and helpful
discussions.
REFERENCES
1. Hepatic encephalopathy. In: Sherlock S, Dooley J, eds. Diseases of the liver and the biliary system, 9th ed. Oxford: Blackwell Scientific, 1993:86-101.
2. Mullen KD, Cole M, Foley JM. Neurological deficits in "awake" cirrhotic patients on hepatic encephalopathy treatment: Missed metabolic or metal disorder?
3. Crippin JS, Gross JB Jr, Lindor KD. Increased intracranial pressure and hepatic encephalopathy in chronic liver disease. Am J Gastroenterol 1992;87:879-82.
4. Blei AT. Cerebral edema and intracranial hypertension in acute liver
P1437
failure. Distincts aspects of the same problem. Hepatology 1991;13:376-9.
5. Pugh RWH, Murray-Lyon IM, Dawson JL. Transection of the oesophagus for bleeding oesophageal varices: Final analysis of a controlled trial. Br J Surg
1983;60:646-9.
6. Blei AT, Omary R, Butterworth RF. Animal models of hepatic encephalopathy. In: Boulton AA, Baker GB, Butterworth RF, eds. Animal models of neurological
disease. Clifton, NJ: Humana, 1992:183-222.
7. Schenker S, Brady CE. Pathogenesis of hepatic encephalopathy. In: Conn HO, Bircher J, eds. Hepatic encephalopathy: Syndromes and therapies. Bloomington,
IL: Medi-Ed Press, 1994:43-61.
8. Mousseau DD, Butterworth RF. Current theories on the pathogenesis of hepatic encephalopathy. Proc Soc Exp Biol Med 1994;206:329-44.
9. Cooper JL, Plum F. Biochemistry and physiology of brain ammonia. Physiol Rev 1987;67:440-519.
10. Nomura F, Ohnishi K, Terabayashi J, et al. Effect of intrahepatic portal-systemic shunting on hepatic ammonia extraction in patients with cirrhosis. Hepatology
1994;20:1478-81.
11. Lockwood AH, Yap EW, Wong WH. Cerebral ammonia metabolism in patients with severe liver disease and minimal hepatic encephalopathy. J Cereb Blood
Flow Metab 1991;11:337-41.
12. Yurdaydin C, Walsh TJ, Engler HD, et al. Gut bacteria provide precursors of benzodiazepine receptor ligands in a rat model of hepatic encephalopathy. Brain
Res 1995;679:42-8.

13. Zieve L, Doizaki WM, Zieve J. Synergism between mercaptans and ammonia or fatty acids in the production of coma: A possible role for mercaptans in the
pathogenesis of hepatic coma. J Lab Clin Med 1974;83:16-28.
14. Stahl J. Studies of blood ammonia in liver disease. Ann Intern Med 1963;58:1-23.
15. Butterworth RF. The neurobiology of hepatic encephalopathy. Semin Liver Dis 1996;16:235-44.
16. Norenberg MD. Astrocytic-ammonia interactions in hepatic encephalopathy. Semin Liver Dis 1996;16:245-54.
17. Norenberg MD. The role of astrocytes in hepatic encephalopathy. Neurochem Pathol 1987;6:13-33.
18. Van Thiel DH, Fagiuoli S, Wright HI, et al. Gastrointestinal transit in cirrhotic patients: Effect of hepatic encephalopathy and its treatment. Hepatology
1994;19:67-71.
19. Wolpert E, Phillips SF, Summerskill WH. Ammonia production in the human colon: Effects of cleansing, neomycin and acetohydroxamic acid. N Engl J Med
1970;283:159-64.
20. Vince A, Bown R, O'Grady F, et al. The effect of perfusion on the flora of the excluded colon. Gut 1973;14:178-82.
21. Weber FL Jr, Fresard KM. Comparative effects of lactulose and magnesium sulfate on urea metabolism and nitrogen excretion in cirrhotic subjects.
22. Uribe M, Berthier JM, Lewis H, et al. Lactose enemas plus placebo tablets vs. neomycin tablets plus starch enemas in acute portal systemic encephalopathy: A
double-blind randomized controlled study. Gastroenterology 1981;81:101-6.
23. Rolachon A, Zarski JP, Lutz JM, et al. Le lavage intestinal par une solution au mannitol est-il efficace dans la prevention de l'encephalopathie hepatique
post-hemorragique chez le malade atteint de cirrhose? Resultats d'une etude prospective randomisee. Gastroenterol Clin Biol 1994;18:1057-62.
24. O'Keefe SJD, El-Zayadi AR, Carraher T, et al. Malnutrition and immuno-incompentence in patients with liver disease. Lancet 1980;2:615-7.
25. Morgan TR, Moritz TE, Mendenhall CL, ete al. Protein consumption and hepatic encephalopathy in alcoholic hepatitis. VA Cooperative Study Group. J Am
Coll Nutr 1995;14:152-8.
26. Lockwood AM, McDonald JM, Rieman RE. The dynamics of ammonia metabolism in men: Effects of liver disease and hyperammonemia. J Clin Invest
1979;63:449-60.
27. Swart GR, van den Berg JWO, van Vuure JK, et al. Minimum protein requirements in liver cirrhosis determined by nitrogen balance measurements at three
levels of protein intake. Clin Nutr 1989;8:329-36.
28. Bianchi GP, Marchesini G, Fabbri A, et al. Vegetable versus animal protein diet in cirrhotic patients with chronic encephalopathy. A randomized cross-over
comparison. J Intern Med 1993;233:385-92.
29. Fenton JC, Knight EJ, Huperson PL. Milk and cheese diet in portal-systemic encephalopathy. Lancet 1966;1:164-6.
30. Condon RE. Effect of dietary protein on symptoms and survival in dogs with an Eck fistula. Am J Surg 1971;121:107-14.
31. Zieve L, Zieve FJ. The dietary prevention of hepatic coma in Eck fistula dogs: Ammonia and the carbohydrate to protein ratio. Hepatology 1987;7:196-8.
32. Plough IC, Iber FL, Shipman ME, et al. The effects of supplementary calories on nitrogen storage at high intakes of protein in patients with chronic liver
disease. Am J Clin Nutr 1956;4:224-30.
33. Weber FL Jr, Minco D, Fresard KM, et al. Effects of vegetable diets on nitrogen metabolism in cirrhotic subjects. Gastroenterology 1985;89:538-44.
34. Mullen KD, Weber FL Jr. Role of nutrition in hepatic encephalopathy. Semin Liver Dis 1991;11:292-304.
35. Eriksson LS, Conn HO. Branched-chain amino acids in the management of hepatic encephalopathy: An analysis of variants. Hepatology 1989;10:228-46.
36. Cabre E, Gasull MA. Nutritional support in liver disease. Eur J Gastroenterol Hepatol 1995;7:528-32.
37. Bircher J, Muller J, Guggenheim P, et al. Treatment of acute hepatic encephalopathy with lactulose. Lancet 1966;1:890-3.
38. Bircher J, Ulrich J. Clinical pharmacology of lactulose, lactitol and related disaccharides. In: Conn HO, Bircher J, eds. Hepatic encephalopathy: Syndromes and
therapies. Bloomington, IL: Medi-Ed Press, 1994:194-208.
39. Mortensen PB. The effect of oral-administered lactulose on colonic nitrogen metabolism and excretion. Hepatology 1992;16:1350-6.

40. Florent C, Flourie B, Leblond A, et al. Influence of chronic lactulose ingestion on the colonic metabolism of lactulose in man (an in vivo study). J Clin Invest
1985;75:608-13.
41. Bown RL, Gibson JA, Sladen GE, et al. Effects of lactulose and other laxatives on ileal and colonic pH as measured by a radiotelemetry device. Gut
1974;15:99-104.
42. Uribe M, Campollo O, Vargas F, et al. Acidifying enemas (lactitol and lactose) vs. nonacidifying enemas (tap water) to treat acute portal-systemic
encephalopathy: A double-blind, randomized clinical trial. Hepatology 1987;7:639-43.
43. Warren SE, Mitas JA II, Swerdlin AH. Hypernatremia in hepatic failure. JAMA 1980;243:1257-60.
44. Bird SP, Hewitt D, Ratcliffe B, et al. Effects of lactulose and lactitol on protein digestion and metabolism in conventional and germ free animal models:
Relevance of the results to their use in the treatment of portosystemic encephalopathy. Gut 1990;31:1403-6.
45. Blanc P, Daures JP, Rouillon JM, et al. Lactitol or lactulose in the treatment of chronic hepatic encephalopathy: Results of a meta-analysis. Hepatology
1992;15:222-8.
46. Camma C, Fiorello F, Tine F, et al. Lactitol in treatment of chronic hepatic encephalopathy: A meta-analysis. Dig Dis Sci 1993;38:916-22.
47. Uribe M, Marquez MA, Garcia-Ramos G, et al. Treatment of chronic portal-systemic encephalopathy with lactose in lactase-deficient patients. Dig Dis Sci
1980;25:924-8.
48. Orlandi F, Freddara U, Candelaresi MT, et al. Comparison between neomycin and lactulose in 173 patients with hepatic encephalopathy: A randomized clinical
study. Dig Dis Sci 1981;26:498-506.
49. Atterbury CE, Maddrey WC, Conn HO. Neomycin-sorbitol and lactulose in the treatment of acute portal-systemic encephalopathy: A controlled, double-blind
clinical trial. Am J Dig Dis 1978;23:398-406.
50. Orlandi F, Brunelli E, Benedetti A, et al. Clinical trials of nonabsorbable disaccharide therapy in hepatic encephalopathy. In: Conn HO, Bircher J, eds. Hepatic
encephalopathy: Syndromes and therapies. Bloomington, IL: Medi-Ed Press, 1994:209-17.
51. Fisher CJ, Faloon WW. Blood ammonia levels in hepatic cirrhosis: Their control by the oral administration of neomycin. N Engl J Med 1957;256:1030-5.
52. Van Berlo CL, van Leeuwen PA, Soeters PB. Porcine intestinal ammonia liberation: Influence of food intake, lactulose and neomycin. J Hepatol 1988;7:250-7.
53. Morgan MH, Read AE, Speller DC. Treatment of hepatic encephalopathy with metronidazole. Gut 1982;23:1-7.
54. Tarao K, Ikeda T, Hayashi K, et al. Successful use of vancomycin
P1438
hydrochloride in the treatment of lactulose resistant chronic hepatic encephalopathy. Gut 1990;31:702-6.
55. Ito S, Miyaji H, Azuma T, et al. Hyperammonaemia and Helicobacter pylori. Lancet 1995;346:124-5 (letter).
56. Gubbins GP, Moritz TE, Marsano LS, et al. Helicobacter pylori is a risk factor for hepatic encephalopathy in acute alcoholic hepatitis: The ammonia hypothesis
revisited. The Veterans Administration Cooperative Study Group No. 275. Am J Gastroenterol 1993;88:1906-10.
57. Quero JC, Hartmann IJC, de Rooij F, et al. Hyperammonemia and Helicobacter pylori. Lancet 1995;346:713-4 (letter).
58. Suh B, Stephens JL, Kunin CM. Oral neomycin dosage schedules for suppresion of ammonia production by bowel flora. Antimicrob Agents Chemother
1979;16:519-22.
59. Breen KJ, Bryant RE, Levinson JD, et al. Neomycin absorption in man. Ann Intern Med 1972;76:211-8.
60. Chambers HF, Sande MA. The amoniglycosides. In: Hardman JG, Limbird LE, Molinoff PB, et al., eds. Goodman & Gilman's The pharmacological basis of
therapeutics, 9th ed. New York: McGraw-Hill, 1995:1103-21.
61. Faloon WW. Metabolic effects of nonabsorbable antibacterial agents. Am J Clin Nutr 1970;23:645-51.
62. Green PHR, Tall AR. Drugs, alcohol and malabsorption. Am J Med 1979;67:1066-76.
63. Loft S, Sonne J, Dossing M, et al. Metronidazole pharmacokinetics in patients with hepatic encephalopathy. Scand J Gastroenterol 1987;22:117-23.

64. Uhl MD, Riely CA. Metronidazole in treating portosystemic encephalopathy. Ann Intern Med 1996;124:455 (letter).
65. Staedt U, Leweling H, Gladisch R, et al. Effects of ornithine aspartate on plasma ammonia and plasma amino acids in patients with cirrhosis: A double-blind
randomized study using a four-fold crossover design. J Hepatol 1993;19:424-30.
66. Van der Rijt CC, Schalm SW, Schat H, et al. Overt hepatic encephalopathy precipitated by zinc deficiency. Gastroenterology 1991;100:1114-8.
67. Reding P, Duchateau J, Bataille C. Oral zinc supplementation improves hepatic encephalopathy: Results of a randomised controlled trial. Lancet 1984;2:493-5.
68. Riggio O, Ariosto F, Merli M, et al. Short-term oral zinc supplementation does not improve chronic hepatic encephalopathy: Results of a double-blind crossover
trial. Dig Dis Sci 1991;36:1204-8.
69. Marchesini G, Fabbri A, Bianchi G, et al. Zinc supplementation and amino acid-nitrogen metabolism in patients with advanced cirrhosis. Hepatology
1996;23:1084-92.
70. Bresci G, Parisi G, Banti S. Management of hepatic encephalopathy with oral zinc supplementation: A long-term treatment. Eur J Med 1993;2:414-6.
71. Peters S, Koh J, Choi DW. Zinc selectively blocks the action of N-methyl-D-aspartate on cortical neurons. Science 1987;236:589-93.
72. Sushma S, Dasarathy S, Tandon RK, et al. Sodium benzoate in the treatment of acute hepatic encephalopathy: A double-blind randomized trial. Hepatology
1992;16:138-44.
73. Mendenhall CL, Rouster S, Marshall L, et al. A new therapy for portal systemic encephalopathy. Am J Gastroenterol 1986;81:540-3.
74. James JH, Ziparo V, Jeppsson B, et al. Hyperammonaemia, plasma aminoacid imbalance, and blood-brain aminoacid transport: A unified theory of
portal-systemic encephalopathy. Lancet 1979;2:772-5.
75. Horowitz ME, Schafer DF, Molnar P, et al. Increased blood-brain transfer in a rabbit model of acute liver failure. Gastroenterology 1983;84:1003-11.
76. Ede RJ, Williams RW. Hepatic encephalopathy and cerebral edema. Semin Liver Dis 1986;6:107-18.
77. Knudsen GM, Schmidt J, Almdal T, et al. Passage of amino acids and glucose across the blood-brain barrier in patients with hepatic encephalopathy.
Hepatology 1993;17:987-92.
78. Naylor CD, O'Rourke K, Detsky AS, et al. Parenteral nutrition with branched-chain amino acids in hepatic encephalopathy: A meta-analysis. Gastroenterology
1989;97:1033-42.
79. Morgan MY. Branched chain amino acids in the management of chronic liver disease: Facts and fantasies. J Hepatol 1990;11:133-41.
80. Marchesini G, Dioguardi FS, Bianchi GP, et al. Long-term oral branched-chain amino acid treatment in chronic hepatic encephalopathy: A randomized
double-blind casein-controlled trial. J Hepatol 1990;11:92-101.
81. Schafer DF, Jones EA. Hepatic encephalopathy and the gamma-aminobutyric-acid neurotransmitter system. Lancet 1982;1:18-20.
82. Bakti G, Fisch HU, Karlaganis G, et al. Mechanism of the excessive sedative response of cirrhotics to benzodiacepines: Model experiments with triazolam.
83. Basile AS, Hughes RD, Harrison PM, et al. Elevated brain concentrations of 1,4-benzodiazepines in fulminant hepatic failure. N Engl J Med 1991;325:473-8.
84. Pomier-Layrargues G, Giguere JF, Lavoie J, et al. Flumazenil in cirrhotic patients in hepatic coma: A randomized double-blind placebo-controlled crossover
trial. Hepatology 1994;19:32-7.
85. Van der Rijt CC, Schalm SW, Meulstee J, et al. Flumazenil therapy for hepatic encephalopathy: A double-blind cross over study. Gastroenterol Clin Biol
1995;19:572-80.
86. Cadranel JF, el Younsi M, Pidoux B, et al. Flumazenil therapy for hepatic encephalopathy in cirrhotic patients: A double-blind pragmatic randomized, placebo
study. Eur J Gastroenterol Hepatol 1995;7:325-9.
87. Gyr K, Meier R, Haussler J, et al. Evaluation of the efficacy and safety of flumazenil in the treatment of portal systemic encephalopathy: A double blind,
randomised, placebo controlled multicentre study. Gut 1996;39:319-24.
88. Bassett ML, Mullen KD, Skolnick P, et al. Amelioration of hepatic encephalopathy by pharmacologic antagonism of the GABAA-benzodiazepine receptor
complex in a rabbit model of fulminant hepatic failure. Gastroenterology 1987;93:1069-77.

89. Basile AS, Gammal SH, Jones EA, et al. GABAA receptor complex in an experimental model of hepatic encephalopathy: Evidence for elevated levels of an
endogenous benzodiazepine receptor ligand. J Neurochem 1989;53:1057-63.
90. Jones EA, Basile AS, Yurdaydin C, et al. Do benzodiazepine ligands contribute to hepatic encephalopathy? Advan Exp Med Biol 1993;341:57-69.
91. Zieve L, Olsen RL. Can hepatic coma be caused by a reduction of brain noradrenaline or dopamine? Gut 1977;18:688-91.
92. Butterworth RF, Spahr L, Fontaine S, et al. Manganese toxicity, dopaminergic dysfunction and hepatic encephalopathy. Metab Brain Dis 1995;10:259-67.
93. Krieger D, Krieger S, Jansen O, et al. Manganese and chronic hepatic encephalopathy. Lancet 1995;346:270-4.
94. Uribe M, Farca A, Marquez MA, et al. Treatment of chronic portal systemic encephalopathy with bromocriptine: A double-blind controlled trial.
95. Morgan MY, Jakobovits AW, James IM, et al. Successful use of bromocriptine in the treatment of chronic hepatic encephalopathy. Gastroenterology
1980;78:663-70.
96. Vogels BAPM, Maas MAW, Daalhuisein J, et al. Memantine, a non-competitive NMDA-receptor antagonist, improves hyperammonemia-induced
encephalopathy and acute hepatic encephalopathy in rats. Hepatology 1997;25:820-7.
97. Ferenci P, Herneth AM, Steindl P. Newer approaches to therapy of hepatic encephalopathy. Semin Liver Dis 1996;16:329-38.
98. Yurdaydin C, Li Y, Ha JH, et al. Brain and plasma levels of opioid peptides are altered in rats with thioacetamide-induced fulminant hepatic failure:
Implications for the treatment of hepatic encephalopathy with opioid antagonists. J Pharmacol Exp Therapeut 1995;273:185-92.
99. Michalak A, Rose C, Butterworth J, et al. Neuroactive amino acids and glutamate (NMDA) receptors in frontal cortex of rats with experimental acute liver
failure. Hepatology 1996;24:908-13.
100. Bullock R. Strategies for neuroprotection with glutamate antagonists: Extrapolating from evidence taken from the first stroke and head injury studies. Ann N Y
Acad Sci 1995;765:272-8.
101. Bensimon G, Lacomblez L, Meininger V. ALS/Riluzole Study Group: A controlled trial of riluzole in amyotrophic lateral sclerosis. N Engl J Med
1994;330:585-91.
102. Munoz SJ. Difficult management problems in fulminant hepatic failure. Semin Liver Dis 1993;13:395-413.
103. O'Grady JG, Alexander GJ, Hayllar KM, et al. Early indicators of prognosis in fulminant hepatic failure. Gastroenterology 1989;97:439-45.
104. Blei AT, Olafsson S, Webster S, et al. Complications of intracranial
P1439
pressure monitoring in fulminant hepatic failure. Lancet 1993;341:157-8.
105. Keays RT, Alexander GJ, Williams R. The safety and value of extradural intracranial pressure monitors in fulminant hepatic failure. J Hepatol 1993;18:205-9.
106. Cordoba J, Blei AT. Cerebral edema and intracranial pressure monitoring. Liver Transplant Surg 1995;1:187-94.
107. Eriksson LS. The cons of BCAA in patients with cirrhosis and acute encephalopathy. In: Capocaccia L, Merli M, Riggio O, eds. Advances in hepatic
encephalopathy and metabolic nitrogen exchange. Boca Raton, FL: CRC Press, 1995:543-8.
108. Blanc P, Larrey D, Michel H. Disaccharides and antibiotics in the treatment of hepatic encephalopathy. In: Capocaccia L, Merli M, Riggio O, eds. Advances in
hepatic encephalopathy and metabolic nitrogen exchange. Boca Raton, FL: CRC Press, 1995:513-32.
109. Simmons F, Goldstein H, Boyle JD. A controlled clinical trial of lactulose in hepatic encephalopathy. Gastroenterology 1970;59:827-32.
110. Morgan MY, Hawley KE. Lactitol vs. lactulose in the treatment of acute hepatic encephalopathy in cirrhotic patients: A double-blind, randomized trial.
Hepatology 1987;7:1278-84.
111. Heredia D, Caballeria J, Arroyo V, et al. A controlled trial. J Hepatol 1987;4:293-8.
112. Strauss E, Tramote R, Silva EP, et al. Double-blind randomized clinical trial comparing neomycin and placebo in the treatment of exogenous hepatic
encephalopathy. Hepatogastroenterology 1992;39:542-5.

113. Pomier-Layrargues G, Giguere JF, Lavoie J, et al. Pharmacokinetics of benzodiacepine antagonist Ro 15-1788 in cirrhotic patients with moderate or severe
liver dysfunction. Hepatology 1989;10:969-72.
114. Elkington SG, Floch MH, Conn HO. Lactulose in the treatment of chronic portal-systemic encephalopathy: A double-blind clinical trial. N Engl J Med
1969;281:408-12.
115. Germain L, Frexinos J, Louis A, et al. Etude en double aveugle du lactulose chez 18 malades atteints d'encephalopathie hepatique chronique apres shunt
portocave. Arch Fr Mal App Dig 1973;62:293-301.
116. Conn HO, Leevy CM, Vlahcevic ZR, et al. Comparison of lactulose and neomycin in the treatment of chronic portal-systemic encephalopathy: A double blind
controlled trial. Gastroenterology 1977;72:573-83.
117. Conn HO, Lieberthal MM. Lactulose and neomycin: Combined therapy. In: Conn HO, Lieberthal MM, eds. The hepatic coma syndromes and lactulose.
Baltimore, MD: Waverly Press, 1978:340-5.
118. McMcrea M, Cordoba J, Vessey G, et al. Neuropsychological characterization and detection of subclinical hepatic encephalopathy. Arch Neurol
1996;53:758-63.
119. Schomerus H, Hamster W, Blunck H, et al. Latent portasystemic encephalopathy. I. Nature of cerebral function defects and their effect on fitness to drive. Dig
Dis Sci 1981;26:622-30.
120. Srivastava A, Mehta R, Rothke SP, et al. Fitness to drive in patients with cirrhosis and portal-systemic shunting: A pilot study evaluating driving performance.
J Hepatol 1994;21:1023-8.
121. Quero JC, Schalm SW. Subclinical hepatic encephalopathy. Semin Liver Dis 1996;16:321-8.
122. Morgan MY, Alonso M, Stanger LC. Lactitol and lactulose for the treatment of subclinical hepatic encephalopathy in cirrhotic patients: A randomised,
cross-over study. J Hepatol 1989;8:208-17.
123. De Bruijn KM, Blendis LM, Zilm DH, et al. Effect of dietary protein manipulation in subclinical portal-systemic encephalopathy. Gut 1983;24:53-60.
124. Egberts EH, Schomerus H, Hamster W, et al. Branched chain amino acids in the treatment of latent portosystemic encephalopathy. A double-blind
placebo-controlled crossover study. Gastroenterology 1985;88:887-95.
125. Gooday R, Hayes PC, Bzeizi K, et al. Benzodiazepine receptor antagonism improves reaction time in latent hepatic encephalopathy. Psychopharmacology
1995;119:295-8.
126. Ohnishi K, Sato S, Saito M, et al. Clinical and portal hemodynamic features in cirrhotic patients having a large spontaneous splenorenal and/or gastrorenal
shunt. Am J Gastroenterol 1986;81:450-5.
127. Vavasseur D, Duvoux C, Cherqui D, et al. Chronic hepatic encephalopathy due to spontaneous splenorenal shunt: Successful treatment by transhepatic shunt
embolization. Cardiovasc Intervent Radiol 1994;17:298-300.
128. Weber FL Jr, Fresard KM, Lally BR. Effects of lactulose and neomycin on urea metabolism in cirrhotic subjects. Gastroenterology 1982;82:213-7.
129. Blanc P, Daures JP, Liautard J, et al. Association lactulose-neomycine versus placebo dans le traitement de l'encephalopathie hepatique aigue: Resultats d'un
essai controle randomise. Gastroenterol Clin Biol 1994;18:1063-8.
130. Ferenci P, Grimm G, Meryn S, et al. Successful long-term treatment of portal-systemic encephalopathy by the benzodiazepine antagonist flumazenil.
131. Hanna SS, Smith RS 3d, Henderson JM, et al. Reversal of hepatic encephalopathy after occlusion of total portasystemic shunts. Am J Surg 1981;142:285-9.
132. Blei AT. Hepatic encephalopathy in the age of TIPS. Hepatology 1994;20:249-52.
133. Hauenstein KH, Haag K, Ochs A, et al. The reducing stent: Treatment for transjugular intrahepatic portosystemic shunt-induced refractory hepatic
encephalopathy and liver failure. Radiology 1995;194:175-9.
134. Larsen FS, Ranek L, Hansen BA, et al. Chronic portosystemic hepatic encephalopathy refractory to medical treatment successfully reversed by liver
transplantation. Transplant Int 1995;8:246-7.
135. Powell EE, Pender MP, Chalk JB, et al. Improvement in chronic hepatocerebral degeneration following liver transplantation. Gastroenterology
1990;98:1079-82.
136. Dagenais MH, Bernard D, Marleau D, et al. Surgical treatment of severe postshunt hepatic encephalopathy. World J Surg 1991;15:109-13.



April 22, 2000
Conde Petra
Additional Article
Full Text
Frontmatter 157
DEFINITION
PORTAL HYPERTENSION
PATHOGENESIS
Problems in Studies of Hepatic

Encephalopathy PORTAL SYSTEMIC ENCEPHALOPATHY
Current Concepts of Pathogenesis
Toxins
Ammonia.
Steven Schenker MD
Mercaptans.
Michael K. Bay MD
Indoles.
Melatonin. From the Division of Gastroenterology and Nutrition, The University of Texas Health Science Center
at San Antonio, San Antonio, Texas
Gamma Aminobutyric Acid
(GABA). The goal of this article is to update the status of portalsystemic encephalopathy
Endogenous (PSE) in the light of new data. First, PSE is defined in the context of other types of
Benzodiazepines. hepatic encephalopathy. Subsequently, current views of the pathogenesis of this
disorder are discussed, followed by an analysis of therapeutic options. Diagnosis
Other Neuroactive will not be considered, as no major new developments have recently been
Substances. documented in this area.
Manganese.
Opioids/Enkephalins. Address reprint requests to

Steven Schenker, MD
Blood-brain Barrier Division of Gastroenterology and Nutrition
Department of Medicine
Mechanisms of Neurotoxin The University of Texas Health Science Center at San Antonio
Action 7703 Floyd Curl Drive
San Antonio, TX 78282-7878
Neurotransmitter Imbalance
Concept. DEFINITION
Endogenous
Hepatic encephalopathy is a neuropsychiatric syndrome (abnormal mental state) that
Benzodiazepine-like
Substances.
develops in patients with hepatic disease or portalsystemic shunting of blood, and is due
to these hepatocirculatory disturbances. The term encephalopathy is much more
Other Neurotransmitters. comprehensive in describing the wide spectrum of mental activity seen than the very
circumscribed terminology of "hepatic coma." It is also generally agreed that hepatic
Altered Brain Energy.
encephalopathy can develop in the setting of acute (initial) and fulminant (parenchymal)
Precipitants of PSE. liver failure (generally defined as of less than 8 weeks' duration), and that which ensues

Future Investigative Needs.
in patients with chronic liver disease or spontaneous or surgically induced
TREATMENT portalsystemic shunting. The syndrome that develops in the setting of chronic liver
dysfunction is usually designated as PSE based on the concept that it is caused by
Nutrition shunting of noxious substances from the portal to the systemic blood circuits, thereafter
Antibiotics impacting on the brain. [26] [39] [45] These substances
Lactulose
158
Ornithine-Aspartate and Zinc
are presumed to be mainly nitrogenous, with ammonia as the main culprit. Some have
Benzoate and Phenylacetate subdivided this group of patients as nitrogenous PSE as opposed to chronic liver disease
Branched Chain Amino Acids
patients with encephalopathy caused by other precipitants, such as the use of sedative
and analgesic drugs, or hypoglycemia, which is termed non-nitrogenous, or pseudo-PSE.
Flumazenil [26] Although this distinction is mechanistically stimulating, in the authors' view, it
assumes more knowledge of the pathogenesis of encephalopathy than is currently

Bromocryptine
warranted. For instance, it is not known precisely by what mechanisms the
TREATMENT OF pharmacodynamic (tissue-based) effects of sedatives affect the brain of patients with
POST-TRANSJUGULAR severe liver disease. A variety of summations with underlying nitrogenous effects could
INTRAHEPATIC PORTAL be present, thus complicating a pathophysiologic definition. Until this is certain, it
SYSTEMIC SHUNTS would seem more constructive not to split the terminology.
ENCEPHALOPATHY
Other terminology refers to encephalopathy as being acute (one episode), acute
CONCULUSION recurrent, or chronic. [26] These distinctions are fine if overt mental change with clear-cut
References improvement and worsening are used. With subtle changes, picked up perhaps only by
complex psychometric assessments, the terminology may change. Different groups use
About the Publication variable psychometric test batteries; some include evoked potentials and others find a
spectral electroencephalogram (EEG) analysis as most sensitive. Thus, the exact gold
standard for what constitutes subclinical hepatic encephalopathy is still uncertain. [3] It
would seem wise, therefore, to define one's terms in the context of the diagnostic
methods used. Finally, the issue of irreversibility comes up. For most cases of
encephalopathy due to fulminant liver failure and of PSE, successful treatment results in
apparently complete reversal of altered mental state. In the rare cases of hepatocerebral
degeneration (extrapyramidal signs and other neurologic manifestations similar to those
seen with Wilson's disease) and of spastic (demyelinating) paraparesis, any
improvements noted are at best partial. [98] Moreover, there is increasing evidence that
some aspects of PSE may persist as assessed by sophisticated psychometric techniques.
[65] Whether these have structural components is uncertain at present. [13] From these
considerations, the authors have classified hepatic encephalopathy, as shown in Table 1 .
PATHOGENESIS
The mechanisms of hepatic encephalopathy have been extensively investigated. Despite
an accelerated pace of such studies in recent years and much progress in the field, [21] [22]
[42] [46] [55] the precise causes of mental changes in patients with liver disease are still
uncertain. The following discussion cites some of the problems in carrying out such
research, presents the current status of the field, and identifies future investigative needs.
159
TABLE 1 -- CLASSIFICATION OF HEPATIC ENCEPHALOPATHY

Clinical Hepatic Laboratory
Syndrome Onset Aspects Histology Reversibility Data

Fulminant Rapid Failure of Hepatic Yes

Blood
hepatic failure synthetic necrosis
sugar,
function (
PT,
BS, PT),
Ammonia
increased
intracranial
pressure in
setting of acute
(8 wks or less)
liver failure
Portalsystemic Gradual Precipitated by Cirrhosis *
encephalopathy various Yes
Ammonia
derangements
(usually)
(i.e.,
gastrointestinal
bleeding, renal
failure,
infection, etc.),
chronic liver
disease with or
without,
portalsystemic
shunting
Hepatocerebral Gradual Characteristic Cirrhosis Partial
degeneration neurologic
Ammonia
findings
related to
altered basal
ganglia
function
Myelopathy Gradual Spastic Cirrhosis Partial
paraparesis (?)
Ammonia
altered
sphincter
function
Urea cycle Gradual Neonatal Yes
enzyme Normal
onset, by Ammonia
deficiency
dietary
nitrogen
Reye's Rapid Childhood Microvesicular Yes
onset, fat
Ammonia,
post-infectious (mitochondrial
low
precipitated by damage)
bilirubin,
aspirin,
hypoglycemia, PT,
increased Blood sugar
intracranial
pressure
*Spontaneous and surgical shunting of blood from portal to systemic circulation is important.
Subtle changes in psychometric assessment may remain (i.e., chronic, subclinical PSE).
Ammonia not always elevated. May be normal if PSE precipitated by sedatives, analgesics, etc.
Occasional cases of post-liver transplantation partial recovery.

Abnormal urea cycle enzymes.

Diffuse changes present (brain and liver).
160
Problems in Studies of Hepatic Encephalopathy
A key difficulty in investigating this field in the past, and to a considerable degree at
present, has been the inability to study in detail, and with safety, human cerebral
metabolism. Animal models of hepatic encephalopathy are better counterparts of
fulminant human encephalopathy than of mental changes (PSE), which develop with
chronic liver disease in patients. Fulminant liver failure in experimental animals is
usually precipitated by the use of toxins (i.e., carbon tetrachloride or galactosamine) or
ischemia. Acute encephalopathy has been studied by administration of large doses of
ammonia. Thus, the relevance of data obtained to the encephalopathic process seen in
patients with PSE is questionable. The animal model for PSE has conventionally been
the rat with a surgical portacaval shunt and the additional administration of ammonia.
Although the presence of portacaval shunting alone may result in subtle behavioral
changes and altered sleep cycle (perhaps resembling subclinical human PSE), and
administration of ammonia will usually produce overt neurologic signs, it is generally
agreed that this animal model system does not fully reflect PSE. [90] More specifically,
cirrhosis is absent, and there is much variation in symptoms and signs of
encephalopathy, probably due to experimental differences. In addition, physiologic
studies (cerebral blood flow, oxygen extraction, glucose consumption) in these animals
are not easy to do sequentially. Thus, such data, together with cerebral biochemical
assessments, have not been systematically obtained.
The problem of animal models and their extrapolation to patients can be added to the
known heterogeneity of the brain data (structurally, biochemically, functionally) not
reflected by analysis of the whole brain or even several brain regions. Examples of such
heterogeneity are the diverse roles of various brain areas, such as the basal ganglia,
cortex and reticular activating system, and of different cells, such as neurons versus
astroglial components. It is not surprising, therefore, that the exact areas and their
interaction, which define normal consciousness (awareness), are still elusive. As
discussed elsewhere, it is generally believed, however, that the minimal areas needed for
intact awareness are the reticular activating system in the brainstem and the cortex.
These areas are believed to define, respectively, the off and on aspects and the content of
consciousness, respectively. [111] Clearly, there are many other interactions and
modifiers.
Finally, a major problem has been the determination of whether observed changes in
mental (neurologic) status and biochemical/physiologic abnormalities are casual or
causally related. Ideally, such conclusions depend on the temporal sequence, extent of
changes, reproducibility of data in various models, and perhaps most importantly, their
concomitant reversal. An example is worsening of mental state with increased nitrogen
load and reversal (amelioration) of both with treatment. Such data are difficult to obtain
in the biochemical/physiologic areas, especially as it concerns the brain.
161
The pathogenesis of PSE can be conveniently discussed sequentially under the

interrelated categories of toxins, alterations of blood-brain barrier, mechanisms of

neurotoxin action (neurotransmitter hypothesis, altered energetics), and precipitants of

PSE (nitrogenous and non-nitrogenous).
Toxins
Ammonia.
Of the various hypothetical toxins, none has as strong a basis as excess ammonia. As
shown in the following list, this is derived from both clinical and experimental data:
Ammonia is increased in arterial blood of 80%-90% of patients with PSE.
Nitrogenous products (as well as ammonia) may induce PSE in patients with
cirrhosis or portalsystemic shunting of blood.
Spinal fluid glutamine (derivative of brain ammonia metabolism) is increased in

virtually all patients with PSE.
In patients with inborn errors of urea cycle, high levels of ammonia are associated
with encephalopathy and treatment aimed at preventing/reducing the ammonia
also prevent/ameliorate the encephalopathy.
Treatment aimed at reducing influx of ammonia (and other nitrogenous

substances) into the brain benefits patients prone to PSE.
Animals with liver damage or portacaval shunts exposed to ammonia develop

neurotoxicity.
Multiple mechanisms for cerebral ammonia neurotoxicity have been established.
Two often-cited deterrents are the absence of elevated blood ammonia in every patient
with PSE and the presence of a stage of excitation prior to sedation in animals given
ammonia acutely; however, the former concern is readily explained by greater
permeability of the blood brain barrier to ammonia in cirrhotics (i.e., increased brain
ammonia), [70] as well as the need for fastidious methodology of blood ammonia assays.
[26] The latter (excitation stage) actually corresponds fairly well to the manifestations of
coma developing rapidly in patients with fulminant liver failure rather than with PSE.
Moreover, it is entirely possible, if not likely, that excess ammonia is only one of
various toxins responsible for PSE.
Mercaptans.
These substances are derived from subnormal metabolism by the damaged liver of
sulfur-containing amino acids. Classical studies by Zieve et al [144] have shown
synergism of ammonia, short-chain fatty acids, and mercaptan toxicity on the brain in
experimental animals; however, the validity of this concept has been put to question by
methodologic differences in measuring mercaptans, as well as lack of clear information
on their mechanisms of neurotoxicity. The current consensus is that their contribution to
PSE in patients is not likely to be significant. [14]
162
The metabolites (phenols) of other (aromatic) amino acids have not been recently
studied, although their levels in blood and cerebrospinal fluid (CSF) of patients with
PSE are not very high. [55] Short- and medium-chain fatty acids may induce reversible
coma in experimental animals, possibly by an effect on Na+ -K+ ATPase, but only in
concentrations much higher than those seen clinically with PSE. Moreover, the severity

of PSE and concentrations of fatty acids in blood do not seem to correlate. [91]
Indoles.
There is substantial evidence, including data in human brain and CSF, that the
neurotransmitter, serotonin, and especially the turnover of this system, are enhanced in
the setting of PSE. [1] [15] [42] Because this neurotransmitter may participate in the arousal
reaction (sleep-wake cycles), it could impact on decreased consciousness in patients
with PSE. A number of experimental studies, however, with serotonin agonists and
antagonists, in portacaval-shunted rats and those with acute hepatic failure have not
documented consistent coordinated changes of neurobehavioral behavior and serotonin
turnover [11] [140] ; this area deserves further study. It appears also that quinolinic acid, a
derivative of tryptophan, which is a potent agonist of the N-methyl-D-aspartate
(NMDA) receptor, is not a significant player in experimental post-shunt encephalopathy.
[12]
Melatonin.
In view of the altered (inverted) sleep cycle seen in many patients with mild PSE, and
the known role of melatonin in promoting sleep, this hormone was investigated. It has
been shown that portacaval shunting in rats disrupts both the pineal melatonin rhythm
and circadian locomotor activity. [143] This is probably not surprising as melatonin is
extensively and rapidly metabolized by the liver. Most recent data suggest, however,
that changes in the area of the suprachiasmatic nucleus in brain of shunted animals
rather than the concurrent changes in melatonin may be responsible for abnormal
locomotor activity and in cirrhotic patients with sleep disturbances; melatonin was not
related to these. [29] [143]
Gamma Aminobutyric Acid (GABA).
This potent inhibitory neurotransmitter was previously believed to be responsible for

PSE. The scenario was that GABA generated in the gut may escape hepatic metabolism,
pass via an altered blood-brain barrier, and impact on the brain. Current data do not
support this precise mechanism, primarily because the brain and spinal fluid
concentrations of GABA (measured accurately) are not significantly increased in
patients with PSE [21] [42] [111] ; however, it is very likely that enhanced GABA-ergic
(inhibitory) neurotransmission may have a role in PSE.
Endogenous Benzodiazepines.
There is now evidence for the presence of endogenous benzodiazepine-like substances

that may enhance GABA-ergic inhibitory brain activity. The basis for this is the
isolation of such substances from the brain of patients and experimental animals with
acute hepatic coma, [9] [10] [93] and the apparent improvement in the mental state of some
such patients given benzodiazepine antagonists (i.e., flumazenil). [96] The exact source
(plant or bacterial) of these substances is not known, and their concentration is much
lower than the amount of benzodiazepines necessary to induce significant sedation in
163
patients without liver disease. Still, their effect could well be contributory to PSE.
Interestingly, the density and affinity of these cerebral benzodiazepine receptors is not
decreased in human PSE, as would be expected if the functional effects of these
endogenous ligands were significant. The role of antagonists and partial inverse agonists
for these receptor complexes were discussed elegantly in a recent review. [42] It appears
that it is the latter group of agents (rather than pure antagonists) that improve
neurobehavior in experimental encephalopathy. The exact mechanistic interpretation of
these agents in terms of modulating inhibitory GABA/benzodiazepine-based

neurotransmission is clearly in need of further study.

Additionally, there is evidence in both experimental animals and in human PSE of
increased density in brain of peripheral-type benzodiazepine receptors. [67] These are not
part of the GABA-benzodiazepine receptor complex, but are localized in astrocytic
mitochondria. Because ammonia is detoxified to glutamine in this cell compartment, and
energy synthesis occurs in mitochondria, this links geographically these two
mechanisms.
Other Neuroactive Substances.
Noradrenaline and dopamine are other important true synaptic transmitters. It has been
postulated that the increased aromatic/branched-chain amino acid ratio in patients with
PSE may result in increased transport of the aromatic amino acids into the brain, and this
may inhibit the synthesis of catecholaminergic neurotransmitters by accumulation of
false neurotransmitters in the synaptic clefts. [21] [111] In this context, the aromatic amino
acids would function as "toxins," and depletion of true neurotransmitter-norepinephrine
and dopamine would be the result. These amino acid changes likely result from
selectively greater degradation of branched amino acids by peripheral muscle and the
catabolic effects of increased glucagon on the liver. Indeed, decreased catecholamines
have been demonstrated in the brain of animals with liver failure. Dopamine
concentrations, on the other hand, have not been diminished. [141] More important, the
concentrations of these two true neurotransmitters have not been shown to be decreased
in postmortem brain tissue of patients with PSE. [30] This is somewhat surprising as
serum prolactin levels, usually seen with low brain dopamine, are reported in patients
with PSE. [79] Of interest, in autopsies of cirrhotics, a selective decrease in dopamine
binding sites has been noted in the pallidum, associated with an increase in manganese.
[19] Other experimental studies wherein brain dopamine is depleted and octopamine
increased did not cause changes in mental status. [145] Administration of l-dopa (to raise
brain dopamine) in patients with PSE also was not therapeutically helpful, and the value
of giving branched-chain amino acids specifically to treat encephalopathy is not
generally accepted as helpful. [39] [40] Thus, the norepinephrine/dopamine depletion
concept due to presumed "toxic" effects of excess aromatic amino acids as an important
cause of PSE is at best debatable.
Manganese.
It is now well established that many patients with PSE have hyperintense signaling from
the area of the basal ganglia on T1 -weighted
164
MR imaging. [65] This is believed to be due to accumulation of manganese at that site. [64]
[97] Indeed, blood manganese correlates roughly with pallidal hyperintensity. [64] [115]
Manganese toxicity, in turn, may cause neurologic manifestations similar to those seen
in some patients with PSE; this may be due to a disturbance of dopaminergic
neurotransmission. Although this concept is novel and interesting, there has been no
established correlation of changes (onset/offset) of PSE manifestations and these MR
imaging findings in patients with overt encephalopathy. Encephalopathy, overt or as
assessed by psychometric tests, also does not correlate with MR imaging signal
hyperintensity, [32] [65] [124] although extrapyramidal symptoms (tremor) seem to. [115] The
latter, which are not common as a manifestation of PSE, would be consistent with the
localization of manganese in the basal ganglia. The possibility that some subtle chronic
neurologic changes may be due to manganese effect requires further study. [65]
Improvement in encephalopathy and decreased hyperintensity of globus pallidus has
been reported after successful hepatic transplantation. [52]
Opioids/Enkephalins.

The possible role of these neurotransmitters in PSE is virtually unstudied, although

preliminary data show increased met-enkephalins in plasma and CSF of patients with
PSE. The specificity of this finding is uncertain as similar results were found in uremia.
[142] In rats with portacaval shunts, region-selective changes in beta-endorphin and in
opioid receptors in brain have been reported. [33] The functional significance of these
findings requires further study.
Blood-brain Barrier
The blood-brain barrier is established by the endothelial cells of capillaries with foot
processes of astroglial cells in close apposition. There is no evidence of a nonspecific
(general) breakdown of the blood-brain barrier in patients with PSE. This is best
exemplified by the absence of increased protein concentration in the spinal fluid of most
of these patients. [34] Unlike in fulminant liver failure, cerebral edema is very rare in
patients with PSE. There is, however, apparently a selective change in this barrier. This
is best shown by the increased transfer of ammonia into the brain in such individuals. [70]
Recent studies using central nervous system (CNS)-derived endothelial cells have shown
that tumor necrosis factor (TNF) may enhance the fluid-phase permeability and
diffusion of ammonia by pinocytosis. [36] Interestingly, TNF may be increased in patients
with liver disease, and pinocytic activity has been shown to be increased in blood-brain
(CNS-capillary) cells of animals with experimentally induced acute liver failure. [56] [99]
Transport of certain amino acids (i.e., tryptophan) by a specific carrier may also be
altered in such patients. The transfer of neutral amino acids is increased, that of glucose
and basic amino acids is decreased, and the flux of glutamine in the reverse direction
(from brain to blood) may be increased. [24] Interestingly, in rats with a portacaval shunt,
the pericapillary astrocytes swell, become more alkaline, and are more depolarized than
in sham-operated controls. [119]
165
Thus, in patients with PSE, selective changes in the blood-brain barrier may occur and
may contribute to neurotoxic effects of some substances. Mechanisms of Neurotoxin
Action
As described previously, excess ammonia appears to be the predominant neurotoxin in
PSE. It is likely that other "toxins" contribute to this adverse effect on the brain. Of the
ones studied (and discussed previously), the GABA-ergic influences (i.e., endogenous
benzodiazepine-like substances) seem the most likely culprits. Other nitrogenous
derivatives (i.e., trimethylamine-N-oxide) may play a part, but require more study. [55]
Neurotransmitter Imbalance Concept.
Ammonia.
In formulating a mechanism of action of PSE, one needs to concentrate first on
ammonia. Ammonia normally is generated primarily from the ingestion of protein or
other nitrogenous substances. Some ammonia can also be released from exercising
muscles and kidneys (with acidosis, hypokalemia, and diuresis). Ammonia is primarily
converted in the liver to urea (and to a small extent, glutamine), and the urea (and some
ammonia) is excreted via the kidneys. [26] Some 80% of a single bolus of ammonia is
detoxified in a single passage by the liver. [26] Normally a small amount of ammonia is
taken up by resting skeletal muscle. In patients with chronic liver disease, ammonia
detoxication is greatly compromised. This may be due to damaged liver cells with
impaired ureagenesis, shunting of blood from the portal to the venous systems
effectively by passing the liver, or both factors. A decreased muscle mass may
contribute, with inability to act as a major safety valve uptake mechanism for ammonia.
This is despite a much larger than usual extraction of ammonia by muscle as compared
to the liver in cirrhosis. [69] Excess production of ammonia (i.e., gastrointestinal

hemorrhage or from increased gut urea due to concomitant renal failure) may, and often
does, compound the problem of impaired detoxication of ammonia. The excess
ammonia that reaches the systemic circulation passes into the brain via a selectively
impaired blood-brain barrier. [70] Any increase in blood pH (i.e., hypokalemic alkalosis)
increases the nonionized ammonia component (Henderson-Hasselbalch mechanism) and
drives this readily permeable ammonia down a pH gradient into the brain (acid sink
phenomenon). [26] [131] [116] Moreover, in PSE patients there is an increase in the cerebral
metabolic rate for ammonia (for uncertain reasons). [70] In the brain, especially in the
white matter, [68] ammonia is converted to glutamine by an energy-consuming reaction in
the astroglial cells. These are, of course, the cells that are responsible for the Alzheimer
type II astrocytosis of PSE. These changes refer to an increase in size and number of
protoplasmic astrocytes that have enlarged pale nuclei with irregular chromatin and
prominent nucleoli. [22]
High concentrations of ammonia are known to have a direct toxic effect via alteration of
cellular chloride channels on both excitatory and
166
inhibitory neurotransmission in the brain. [101] [102] This is apparently exerted at both the
postsynaptic and synaptic mechanisms. [100] [102] Raabe [102] has suggested that these
effects are sufficient to explain most of the toxic effects of ammonia on the brain. It is
known, however, that inhibition of ammonia metabolism (glutamine synthesis) with
methionine sulfoximine protects rodents from cerebral ammonia toxicity despite higher
ammonia levels. [130] This, and other experimental mechanistic studies, [22] [46] [51] suggest
that ammonia primarily exerts its toxic effects on the brain via its metabolism and on the
glutaminergic neurotransmitter system.
Glutamate is the major excitatory neurotransmitter in the brain. With excess ammonia,
alpha-ketoglutarate combines with ammonia to form glutamate, and this, in turn, with
addition of more ammonia, forms glutamine. Glutamine, in turn, is converted back to
glutamate and NH3 by the action of glutaminase. Although glutamine formation occurs
in astrocytes, the regulation of overall glutamate flux also involves an interrelationship
with presynaptic and postsynaptic neurons. [22] Much research involving cultured
astrocytes, experimental animals with portacaval shunts, as well as postmortem studies
with human brain of patients with PSE has suggested that in PSE the primary defect is in
the perineuronal astrocyte, which is unable to take up extracellular glutamate released
from neurons. [22] Thus, as measured by microdialysis in portacaval shunted rats, there is
excess (overflow) glutamate in the extracellular space, ammonia inhibits glutamate
uptake by cultured astrocytes, and there is downregulation (decreased density) of
postsynaptic glutamate receptors in PSE brain. [22] The latter would be expected in
response to excess neuronally released glutamate outside of the astrocytic receptors.
Very recently, the astrocytic glutamate transporter (GLT-1) has been shown to be
decreased in brains of rats with acute ischemic liver failure, [59] and preliminary data
suggest the same in portacaval-shunted rats (personal communication, 1996). This
would be consistent with impaired neuronal-astrocytic glutamate trafficking. Astrocytic
mitochondrial membranes also contain receptors for so-called peripheral-type
benzodiazepines, [22] and these are increased in brain of animals and patients with PSE.
The precise implication of these changes is uncertain, but it may relate to the toxicity of
ammonia (on mitochondria [?]) in these cells.
This defect in glutamatergic (excitatory) synaptic function in PSE has now been
extended to a specialized form of this receptor. A series of studies using portacaval
shunted rats, acute and chronic ammonia exposure, and various receptor antagonists
suggest that ammonia impairs the function of the N-methyl-D-aspartate (NMDA)
receptor, a subclass of glutamate receptors. [46] [72] [75] [85] [95] The interaction of acute
versus acute on chronic (PSE-type) ammonia toxicity is complex [51] [61] [84] and not yet
fully characterized, but a series of studies by Grau et al have tentatively proposed a

scheme of ammonia action on the astrocyte (Fig. 1) (Figure Not Available) . [46] In this
formulation ammonia inhibits the phosphorylation of protein kinase C (PKC), resulting
in increased activity of Na+/K+ ATPase and a subsequent depletion of ATP, the
ultimate energy source. [14] The
167
Figure 1. (Figure Not Available) Effects induced in brain by hyperammonemia that are mediated by
decreased PKC-mediated phosphorylation. ( From Grau E, Marcaida G, Montoliu C, et al: Effects of
Hyperammonemia on Brain Protein Kinase C Substrates Metabol. Brain Disease 11:205-216, 1996;
with permission.)
inhibition of PKC phosphorylation may also decrease the phosphorylation of the NMDA
receptor. This view is interesting in that it ties in the effects of ammonia on the
glutamate (NMDA) neurotransmitter and brain energetics; however, this concept is
hypothetical because the mechanism by which ammonia may decrease PKC
phosphorylation in unknown. [46] Moreover, NMDA receptors appear to be uniquely
neuronal in location, whereas non-NMDA (glutamate) receptors are both neuronal and
astrocytic [82] ; hence, the role of the NMDA receptor needs further assessment. Also,
extrapolation from various animal models of hepatic encephalopathy to human disease
states (acute liver failure versus PSE) will require careful analysis.
Endogenous Benzodiazepine-like Substances.
These substances (discussed as possible neurotoxins previously) may exert their effect
on the brain by acting on the GABA-benzodiazepine receptor complex. This complex
mediates neural inhibition via influx of chloride ions into postsynaptic neurons, followed
by an inhibitory potential. [8] [42] Activation of this inhibitory complex may also explain
the increased sensitivity of patients with chronic liver disease to sedatives such as
diazepam and triazolam, and may explain the beneficial (albeit transient) effects of the
benzodiazepine antagonist, flumazenil, in some of these patients. [6] [95]
Other Neurotransmitters.
The crux of this hypothesis is that weak
168
(false) neurotransmitters, such as beta-phenylethanolamine and octopamine, derived

from aromatic amino acids that flood the brain displace true neurotransmitters (i.e.,
catecholamines) in the synaptic clefts. [21] [111] Although conceptually attractive, this
hypothesis has been seriously challenged by postmortem brain data in patients with PSE,
which showed normal concentrations of catecholamines and no accumulation of
octopamine. [30] Moreover, intraventricular administration of octopamine into the brain
of experimental animals did not induce coma. [145]
Energy, initially as glucose and ultimately as ATP, is necessary for normal cerebral
function. Accordingly, numerous studies of cerebral energy stores have been carried out,
first in various forms of experimental hepatic encephalopathy, and more recently in
patients. Problems in data interpretation have been the variability in brain regions, the
difficulty in carrying out studies in patients, and analysis of information as to
cause-effect. [111] Human studies recently have used positron emission tomography and
nuclear MR spectroscopy as relatively noninvasive tools. [55]
In experimental animals, using ammonia intoxication as a model, decreased ATP stores
were detected, especially in the brain stem (reticular activating system). [112] This
decrease in energy has been most recently interpreted as the result of the coma (i.e., due
to decreased glucose use and not the initiating event). [53] [55] Of interest, portacaval
shunting in older rats (as compared to younger ones) seemed to result in greater increase
in peripheral-type benzodiazepine binding sites, accompanied by greater impairment of
locomotor function. [5] This corresponds to a greater sensitivity to PSE in older patients,
and may relate to altered energetics as these ligands are located on the mitochondrial
membrane. Other possibilities (i.e., greater glutamate carrier abnormality with age) are
also possible.
In patients, the data on cerebral glucose utilization have been variable, depending on the
severity of encephalopathy, the method used, and area of brain studied. [16] [42] [55] [70] In
general, however, a decrease in cerebral blood flow (perhaps related to low pCO2 ), and
decreased cerebral glucose oxidation was seen. [2] [16] [70] Data on brain energetics are
also variable. Some have noted impaired high-energy phosphate metabolism [7] [55] [107]
[123] using newer MR imaging technology. Others reported no abnormality in energetics.
[25] The demonstration of increased CSF lactate, which correlated with the mental state,
seems to agree with evidence of altered brain energy turnover. [139] It has been difficult
to determine if these changes in brain glucose consumption are the result of decreased
energy demands or the early cause of neurologic symptoms. Recent studies using
position emission tomography to study regional cerebral glucose metabolism have
documented decreased glucose metabolism in various parts of the brain in patients with
subclinical hepatic encephalopathy as compared to two control groups. [57] These data
suggest an early abnormality of cerebral energetics in patients with PSE, but this report
needs to be confirmed and extended with longitudinal studies
169
before, during the development of, and after the treatment of PSE. Other studies have
shown changes in phospholipid precursors (rather than ATP), suggesting altered
membrane composition. [25] [123] Finally, a substantial decrease in myoinositol has been
detected fairly consistently. [28] [49] [63] [106] The basis for this is not certain, but may
reflect compensatory osmoregulation to counterbalance an increase in glutamine. [28] The
net effect would be to prevent the formation of brain edema. The myoinositol decrease is
actually seen in cirrhotic patients in the absence of even subtle encephalopathy, and the
decrease does not correlate well with the severity of encephalopathy. [43]
Precipitants of PSE.
These are of the nitrogenous and non-nitrogenous varieties, and are listed together with
presumed mechanisms of action in Table 2 .
TABLE 2 -- COMMON PRECIPATATING FACTORS OF
HEPATIC ENCEPHALOPATHY AND POSSIBLE
MECHANISMS
Precipitating Factor Mechanism
Azotemia Increased enterohepatic
circulation of urea with
increased ammonia
production in the intestine
Direct sedative effect of
uremia
Diuretics Excessive diuresis may
induce:

Hypokalemic alkalosis,
which enhances transfer of
ammonia across the
blood-brain barrier and
increases renal vein
ammonia output.
Hypovolemia with
decreased perfusion of
vital organs, including
brain, liver, and kidneys.
Gastrointestinal bleeding Digestion of intraluminal
blood provides a substrate
for increased ammonia
production and other
nitrogenous toxins (100
mL of blood = 5-20 g of
protein).
Sedatives, tranquilizers, and Altered
analgesics pharmacokinetics--Hepatic
oxidative degradation of
many sedative drugs is
impaired in patients with
liver disease. Unless the
dose is adjusted, these
drugs then accumulate and
may exert a direct
depressant effect on the
brain.
Pharmacodynamics--The
brain in these patients
appear to be intrinsically
more sensitive to the
effects of these drugs. In
the case of
benzodiazepines, this may
relate to the presence of an
endogenous
benzodiazepine-like
substance and greater
"activity" of the GABA
system.
Excess dietary protein Protein excess provides
substrate for production of
ammonia and other
nitrogenous toxins in the
intestine.
Infection Increased tissue
catabolism leads to a
greater endogenous
nitrogen load and
increased ammonia
production. Infection may
cause dehydration and
prerenal azotemia.
Constipation Longer contact time

between nitrogenous
substances and the
ammonia-forming bacteria
may increase production
and absorption of
ammonia and other toxic
nitrogenous derivatives
Data from Schenker S: Hepatic coma: Current concepts of
pathogenesis. Viewpoints on Digestive Diseases 2:1-4, 1970;
and Hoyumpa AM, et al: Clinical conference: Hepatic
encephalopathy. Gastroenterology 76:184-195, 1979.
170
With most illnesses, treatment is optimal after the pathogenesis of the disease is defined.
PSE is unusual in this regard, as the treatment is quite good despite a very incomplete
understanding of the problem; however, therapy now is directed only to the peripheral
sites of influx and metabolism of nitrogen. With better understanding of the CNS effects
of various toxins, centrally directed management may become possible. For instance,
with the discovery that increased myoinositol and glutamine in brain may relate to
cerebral edema in fulminant failure, therapeutic approaches may ensue. In experimental
animals, this is represented by agents that block glutamine synthesis. [121]
With the inadequacy of experimental models, the use of noninvasive MR imaging and
position emission tomography techniques in longitudinal studies of patients may yield
big dividends. This may also define which findings are the initiating events. It will also
be important to determine if all types of PSE are similar as to mechanisms, and if the
astrocytic process is truly fully reversible. Although these key studies in patients are
essential, the interaction of astrocytes and neurons in terms of neurotransmitter traffic
will likely require studies in cell culture and experimental animals.
TREATMENT
The mainstay of treatment for the majority of patients with PSE has been, and continues
to be, lactulose therapy; however, given the multiple lines of evidence, each giving a
partial but not complete understanding of the pathophysiology of PSE, a variety of
treatments have been proposed and examined (to various degrees), each with a rationale
based on that particular line of evidence. A comprehensive listing of treatments of PSE,
some experimental, is as follows:
Reduce dietary protein
Switch to vegetable dietary protein
Lactulose or lactitol
Lactose in lactase deficiency
Neomycin or other antibiotics (metronidazole, rifaximin, vancomycin,

paromomycin)

Ornithine-aspartate
Oral zinc supplementation
Benzoate
Phenylacetate
Branched-chain amino acids (valine, leucine, isoleucine)
Flumazenil
Bromocryptine
TIPS shunt revision: reducing stent diameter or occlusion
Liver transplant
Colonic diversion or colectomy
171
Also, no discussion of this topic is complete without mention of general measures to be

considered in the treatment of PSE:
Discontinuance and avoidance of analgesics, sedatives, and tranquilizers
Control of gastrointestinal bleeding and correct anemia
Purging blood from gastrointestinal tract
Diagnose and treat infections (particularly, but not limited to, spontaneous
bacterial peritonitis)
Correct respiratory or metabolic alkalosis
Correct metabolic acidosis
Correct hypoxia
Correct electrolyte abnormalities: hyponatremia, hypernatremia, hypokalemia,

hyperkalemia
Prevent constipation
Provide IV glucose solution to minimize endogenous protein breakdown
Replenish multivitamin, thiamine, folate repletion

Nutrition
The time-honored approach to the treatment of PSE by nutritional means has been to
restrict dietary protein and then to gradually increase the dietary protein in small
increments while assessing for PSE to determine the threshold limits. In addition to the
quantity of dietary protein, other nutritional factors include qualitative aspects of the
diet, including meat versus vegetable protein, calorie-to-nitrogen ratio, amino acid
composition, and amount of dietary fiber.
Often the patient with PSE and chronic liver disease is also malnourished. This has
raised some compelling questions. How much dietary protein is too much? How much is
enough? Higher amounts of dietary protein may improve the malnutrition, but increase
the risk of PSE. Older studies have shown that stable cirrhotic patients were in nitrogen
balance and conserved nitrogen appropriately when dietary protein was reduced to 1.0
g/kg of body weight/day. Some, but not all, patients were able to maintain nitrogen
balance at even lower levels of dietary protein (0.5 g/kg of body weight/day). [109] [136]
[137] In contrast, another study showed that a relatively high level of dietary protein (1.2
g/kg of body weight/day was required to maintain positive nitrogen balance in stable
cirrhotics. [120] No patients were noted to manifest PSE at any time during the study,
even those given 1.4 g/kg of body weight/day of dietary protein. Although the amount of
dietary protein required to maintain positive nitrogen balance varies in the literature, it
does appear that a positive nitrogen balance can help the patient with PSE by promoting
hepatic regeneration and augmenting the ability of muscle to detoxify ammonia. [71]
Vegetable protein diets may be more beneficial (less ammoniagenic)
172
than animal protein diets. It appears that an important difference in these diets is the
substantially increased amount of fiber found in the vegetable protein diet. [135]
Vegetable protein diets resulted in improved mental status compared with animal protein
diets containing equal nitrogen and equal calories. [47] An additional controlled study by
Uribe showed vegetable protein diets more beneficial in patients with hepatic
encephalopathy than animal protein diets. [127] Fiber has been found to promote
increased fecal excretion of nitrogen, in large part due to incorporation of nitrogen by
fecal bacteria. [135] Some patients, however, find it very difficult to comply with
vegetable protein diets, in which case other sources of protein are allowed, accompanied
by the use of lactulose (or similar agent).
Antibiotics
Antibiotics have been shown to improve PSE, and the presumed mechanism is
decreased intestinal bacterial flora resulting in reduced formation of ammonia and other
potentially toxic substances. Most commonly, neomycin has been used for this purpose,
although other agents that have been used include paromomycin, rifaximin,
sulfathalidone, metronidazole, [96] and vancomycin. [122] Although neomycin is largely
not absorbed, 1% to 3% may be absorbed by either oral or enema formulation, which
can lead to ototoxicity or nephrotoxicity. [17] In addition to decreasing the activity of
bacterial flora, some evidence suggests it inhibits glutamine conversion to ammonia in
the intestinal wall. [113] Largely due to the effectiveness of lactulose and the potential for
toxicity, neomycin has been used less; however, if lactulose fails to benefit the patient,
then a trial of neomycin may be warranted. It is usually given orally, 1 g every 4 to 6
hours, or as a 1% retention enema (1 to 2 g in 100 to 200 cc of isotonic saline) two or
three times daily. If neomycin fails to provide benefit, then the combination of lactulose
and neomycin has been shown to result in additive effect, [133] and should be considered.
Lactulose requires bacterial metabolism to cause intraluminal acid production, and thus
intraluminal ammonia trapping, and although neomycin may interfere with bacterial
metabolism (overall, in about one third of patients), the combination will usually result

in continued beneficial acidification of stool. [27] If the stool pH is less than 6, then
lactulose metabolism is occurring, and continued use of the two drugs makes sense. [27]
A different antibiotic that has been studied more recently, and reported largely in the
European literature, is rifaximin. Rifaximin is a derivative of rifamycin and is virtually
unabsorbed after oral administration. [44] Three double-blind trials comparing rifaximin
(1200 mg/d) to lactulose (from 30-120 g/d) showed significantly improved mental status
with rifaximin. [20] [78] One double-blinded study comparing rifaximin and lactulose
against neomycin and lactulose showed no significant difference in the two treatments.
[35] Another studied rifaximin (1200 mg/d)
173
with neomycin (3 g/d) in double-blinded fashion, and the results showed significant
improvement in serum NH3 in the rifaximin treatment arm and improvement in EEG,
Reitan test, and severity of hepatic encephalopathy, although not statistically significant.
The side-effect profile appears to be quite favorable, with less than 1% incidence of
gastrointestinal distress (flatulence, abdominal pain, and nausea reported). Other
uncommon side effects include urticaria, weight loss, headache, leg edema, and mild
elevations in serum potassium and serum sodium.
A new development in the use of antibiotics to treat chronic encephalopathy by
eradicating gastric Helicobacter pylori has recently been described. [31] Because of the
urease produced by the organism, it is postulated that increased gastric ammonia
contributes to the development of hepatic encephalopathy. In comparing a group of
Helicobacter-positive stable cirrhotics with a comparable group of stable cirrhotics who
were Helicobacter negative, a significantly increased level of gastric ammonia was
found. After H. pylori eradication therapy (amoxicillin and omeprazole), seven patients
showed improvement of symptoms, as manifested by resolution of day/night reversal
and a trend toward improved trail test time. Eradication of H. pylori may play a role in
the treatment of hepatic encephalopathy; however, additional studies need to be done to
confirm these preliminary results.
Lactulose
Lactulose is a synthetic disaccharide (galactosidofructose) that is not absorbed by the gut

and is metabolized by colonic bacteria to organic acids. These organic acids result in a
decrease in luminal pH, which may cause ammonia to be trapped in the lumen in the
form of ammonium. Decreased luminal pH may also promote peristalsis. Moreover,
lactulose also acts as an osmotic laxative promoting the elimination of fecal nitrogen and
other fecal material. Fecal bacteria increase incorporation of nitrogen, probably due to
the influence of lactulose or its metabolites on bacterial growth. [132] [134] In vitro studies
have shown inhibition of intestinal glutamine uptake and ammonia formation with
lactulose (or neomycin) independent of bacterial metabolism. [113]
A typical dose of lactulose is 30 mL two or three times a day given orally with the goal
of two or three soft stools a day. Some patients can be instructed to titrate their dose of
lactulose to achieve this result. If lactulose cannot be given orally (such as in patients
with grade 3 or worse encephalopathy), it may be administered as an enema (300 mL of
lactulose added to 700 mL of tap water). Excessive lactulose can result in severe
diarrhea leading to hypernatremia. Lactilol (beta-galactosidofructose) is a disaccharide
similar to lactulose, which has been shown to be as effective as lactulose in the treatment
of hepatic encephalopathy [88] ; at present, it is not available in the United States. Lactilol
is not as sweet as lactulose, and is associated with less nausea and abdominal bloating.
174
In one study, lactose was shown to be an effective treatment for PSE in those who are

lactase deficient. [126]
Ornithine-aspartate provides substrate for urea formation and for synthesis of glutamine,
thereby enhancing the two major routes of ammonia fixation in the body. It has been
shown to decrease the postprandial rise in ammonia concentration in cirrhotics [117] and
may prove to play a role in the treatment of PSE. Zinc is a cofactor of urea cycle
enzymes and also plays a role in CNS neurotransmission. [4] [54] [60] [94] [138] Zinc
deficiency has been proposed as a contributory factor in the pathogenesis of hepatic
encephalopathy. [129] The clinical trials to date have shown mixed results. [18] [76] [103] [104]
Benzoate and Phenylacetate
Benzoate and phenylacetate have been used to enhance nitrogen excretion in children
with urea cycle enzyme deficiencies, and additionally have been looked at in the
treatment of adults with PSE. Benzoate is conjugated with glycine to form hippuric acid
and phenylacetate with glutamine to form phenylacetyl glutamine; both products can be
renally excreted. For each mole of benzoate, one mole of waste nitrogen is excreted, and
for each mole of phenylacetate, two moles of waste nitrogen is eliminated. In one study,
benzoate was shown to be as effective as lactulose in the treatment of acute
encephalopathy. [118] Additionally, another study showed benefit in treating patients with
chronic stable portal systemic encephalopathy; however, only eight patients were
studied. [80]

With the observation that plasma concentrations of aromatic amino acids (tyrosine,
tryptophan, phenylalanine) are increased, and those of branched-chain amino acids
(leucine, isoleucine, valine) are decreased in patients with chronic liver disease, [105] it
was postulated that this amino acid imbalance was important in the development of PSE.
Aromatic amino acids and branched-chain amino acids are transported into the brain by
the same amino acid carrier, and with increased peripheral aromatic amino acids
available for transport, more aromatic amino acids enter the brain. These aromatic amino
acids then result in the synthesis of false neurotransmitters (phenylethanolamine,
octopamine) and serotonin. By giving branched-chain amino acids either by infusion or
oral administration, less aromatic amino acids enter the brain. In experimental animals,
the infusion of tryptophan and phenylalanine into the carotid
175
results in hepatic encephalopathy, but this can be prevented if branched-chain amino

acids are given concurrently. A number of clinical trials have evaluated this treatment,
both for acute and chronic encephalopathy. Critical analysis of these studies [37] [41] [86]
[92] conclude no significant therapeutic benefit, as to encephalopathy. However, some
patients may tolerate these nutrients better than normal protein diet.
Flumazenil
Flumazenil is a benzodiazepine ligand that stabilizes the benzodiazepine receptors on
the GABA-benzodiazepine receptor complex and competitively antagonizes the binding
of other benzodiazepine ligands. With the GABA hypothesis in mind, flumazenil has
been looked at as an agent by which to treat PSE. A recent randomized, double-blind,
placebo-controlled crossover trial looked at the response of a bolus of flumazenil in
cirrhotics in stage-four hepatic encephalopathy receiving lactulose. [96] Clinical
improvement occurred in 6 of 13 treatment periods in which flumazenil was given
compared to no improvement in all treatment periods (n = 15) in which placebo was

given.
Clinical improvement after flumazenil occurred within 5 minutes of administration and
was partial in response with the best improvement from grade 4 to grade 2 hepatic
encephalopathy. The authors concluded that the majority of patients with severe hepatic
encephalopathy do not benefit significantly from flumazenil. Other studies have yielded
similar results. [23] [128] A source of controversy in assessing the true efficacy of
flumazenil has been whether some of the patients studied had received exogenous
benzodiazepines. If so, then flumazenil could have reversed the effect of exogenous
benzodiazepine as opposed to improving symptoms of PSE.
Lastly, a single case report describes a patient with PSE refractory to standard therapy
(40 g protein diet, oral neomycin and lactulose, branched-chain amino acids) after
extensive liver resection and surgical portacaval shunt who improved dramatically while
taking flumazenil 25 mg orally twice a day with resolution of PSE despite unrestricted
dietary protein. [38] Thus, flumazenil may prove to play a role in the treatment of chronic
PSE, although it should be noted that at this time an oral preparation is not available in
the United States.
Bromocryptine
The false neurotransmitter hypothesis postulates displacement of true neurotransmitters

with false neurotransmitters (such as phenolethanolamine and octopamine) resulting in
dysfunctional dopaminergic neurotransmission. Consequently the dopaminergic drugs
l-dopa and bromocryptine have been tried as treatment for hepatic encephalopathy.
l-dopa was found to be ineffective. [73] [83] Results with bromocryptine have
176
been mixed with an initial favorable report, [87] but in a controlled study no benefit was
found. [125] In another study improvement in extrapyramidal signs in patients with PSE
after bromocryptine was reported. [89]
TREATMENT OF POST-TRANSJUGULAR INTRAHEPATIC

PORTAL SYSTEMIC SHUNTS ENCEPHALOPATHY
With the recent introduction and proliferation of transjugular intrahepatic portal
systemic shunts (TIPS), the new clinical entity of post-TIPS encephalopathy emerged.
This relatively new entity deserves special consideration under therapy. Via a
transjugular approach, interventional radiologists create a tract through liver parenchyma
connecting hepatic and portal venous circulations, and maintain patency of the tract by
deploying an expandable metal stent. This procedure is analogous to side-to-side
surgical portacaval shunts, but does not carry the risks of surgery or general anesthesia.
Since its introduction in 1989, TIPS has gained general acceptance as treatment for acute
variceal bleeding that cannot be successfully controlled with medical or endoscopic
treatment, or recurrent variceal bleeding in patients who are refractory/intolerant to
pharmacologic or endoscopic therapy. Additional indications that are as yet unproved,
but promising, include treatment for refractory ascites and hepatic vein thrombosis
(Budd-Chiari syndrome). [58]
Risk factors for the development of PSE after TIPS include abnormal part B trail test,
[110] cause of liver disease other than alcohol, [113] female gender, [113] hypoalbuminemia,
[113] and older age. [108] [110] [114] Factors that have not been found to predict the
development of PSE after TIPS include shunt diameter [108] [114] and direction of portal
blood flow. [114] Some have found a history of PSE prior to TIPS as a significant risk
factor, [110] whereas others have not. [114]
The pooled results of several recent studies suggest a 24% incidence

TABLE 3 -- POST-TIPS ENCEPHALOPATHY

PSE
Patients Refractory
With to
Successful New or Worse Medical
Reference TIPS Encephalopathy Treatment
[66] LaBerge 96 17/96 (18%) 3/96
et al
(1993)
[50] Helton et 55 10/55 (18%) NR
al (1993)
[77] Martin et 45 15/40 * (38%) 10/25
al (1993)
[108] Rossle et 93 23/93 (25%) 7/93
al (1993)
[114] Somberg 77 18/77 (23%) 4/77

et al
(1995)
[110] Sanyal et 30 9/30 (30%) 1/30
al (1994)
Totals 92/391 (24%) 25/321
(7.8%)
NR = Not reported.
*Evaluated 3 months after TIPS.
Prophylactic lactulose given to patients after TIPS.
177
of new or worsening encephalopathy and a 7.8% incidence of severe/refractory PSE

after TIPS (Table 3) . These values are limited by the lack of a rigorous definition used
by all the studies, differences in duration and methods of follow-up, and differences in
patient populations. For example, in one study, prophylactic lactulose was given to
patients after TIPS and PSE was reported only in patients refractory to lactulose. [108]
The incidence of PSE is highest during the first 3 months after TIPS, with it being less
of a problem subsequently; this is believed to be due to narrowing of the stent diameter
from neointimal hyperplasia. [108]
Post-TIPS encephalopathy responds to medical treatment, including lactulose, protein
restriction, antibiotics, and branched-chain amino acids, in roughly 75% to 90% of
cases. Options for those who fail medical treatment include liver transplantation,
occlusion of the stent, or reduction of the stent diameter. [48]
CONCULUSION
In summary, at present the major known toxins for PSE appear to be ammonia and,
perhaps, the endogenous substances which act on the GABA-benzodiazepine recepter
complex. Other factors likely summate with these. Neurotransmission imbalance

appears to be the key to the mechanism of encephalopathy and the glutamate and
GABAerig-receptor systems appear to be especially affected. The role of impaired
energy metabolism is not certain. At the cellular level, the astrocyte/neuronal interaction
appears to be critical. This is a multifactorial problem, which is largely metabolic in
nature (biochemical/neurophysiologic) and is largely, if not fully, reversible. Better
biochemical characterization of the central nervous system mechanism(s) of PSE may be
relevant to other central nervous system problems (e.g., the role of the MDA receptor
complex in alcohol effects on the brain).
Although the mainstay of treatment for the majority of patients with PSE continues to be
lactulose therapy, additional treatments continue to be examined. Some treatments have
had mixed results or have been incompletely studied. These include ornithine-aspartate,
carnitine, zinc, benzoate, phenylacetate, branched chain amino acids, flumazenil, and
bromocryptine. Other treatments are generally effective but for various reasons are less
favorable than lactulose in most situations; these include reducing dietary protein,
switching to vegetable dietary protein, and antibiotics. Lactitol is roughly equivalent to
lactulose but unavailable in the United States. In the special situation of the lactase
deficient patient, lactose is effective treatment. And in the special situation of post-TIPS
encephalopathy, TIPS shunt revision can be helpful in those refractory to medical
treatment. The surgical options of colonic diversion or colectomy are associated with
high morbidity and mortality and are rarely used. Finally, in the failing liver in which
PSE is but one manifestation of end-stage disease, liver transplantation offers the hope
of effective treatment and added life.
178
References
1. Al Mardini H, Harrison EJ, Ince PG, et al: Brain indoles in human hepatic encephalopathy.
2. Almdol T, Schroeder T, Ranek L: Cerebral blood flow and liver function in patients with
encephalopathy due to acute and chronic liver diseases. Scand J Gastroenterol 24:299, 1989
3. Amodio P, Quero JC, Del Piccolo F, et al: Diagnostic tools for the detection of subclinical hepatic
encephalopathy: Comparison of standard and computerized psychometric tests with Spectral-EEG.
Metab Brain Dis 11:315, 1996
4. Assaf SY, Chung SH: Release of endogenous zinc from brain tissue during activity. Nature 308:734,
1984
5. Audet RM, Butterworth RF: Astrocyte function and cerebral energy metabolism are more severely
compromised in old adult rats compared to young adult rats following portacaval anastomosis.
6. Bakti G, Fisch HV, Karlaganis G, et al: Mechanism of the excessive sedative response of cirrhotics to
benzodiazepines: Model experiments with triazolam. Hepatology 7:629, 1987
7. Barbara L, Barbiroli B, Gaiani S, et al: Abnormal brain energy metabolism assessed by 31P-MRS in
liver cirrhosis [abstract]. Hepatogastroenterol 40:6(Abstract), 1993
8. Basile AS, Gammal SH, Mullen KD, et al: Differential responsiveness of cerebellar Purkinge neurons
in GABA and benzodiazepine receptor ligands in an animal model of hepatic encephalopathy. J
Neurosci 8:2414, 1988
9. Basile AS, Hughes RD, Harrison IM, et al: Elevated brain concentrations of 1,4-benzodiazepines in
fulminant hepatic failure. N Engl J Med 325:475, 1991
10. Basile AS, Paudl L, Jaouni T, et al: Brain concentrations of benzodiazepines are elevated in an
animal model of hepatic encephalopathy. Proc Natl Acad Sci USA 87:5263, 1990

11. Bengtsson F, Nobin A, Falck B, et al: Portocaval shunt in the rat: Selective alterations in behavior
and brain serotonin. Pharmacol Biochem Behav 24:1611, 1986
12. Bergquist PBF, Heyes MP, Apelqvist G, et al: Brain extracellular quinolinic acid in chronic
experimental hepatic encephalopathy as assessed by in vivo microdialysis: Acute effects of
L-tryptophan. Neuropsychopharmacology 15:382, 1996
13. Bernthal P, Hays A, Tarter RE, et al: Cerebral CT scan abnormalities in cholestatic and
hepatocellular disease and their relationship to neuropsychologic test performance. Hepatology 7:107,
1987
14. Blom HJ, Ferenci P, Grimm G, et al: The role of methanethiol in the pathogenesis of hepatic
encephalopathy. Hepatology 13:445, 1991
15. Borg J, Warter JM, Schlienger JL, et al: Neurotransmitter modifications in human cerebrospinal
fluid and serum during hepatic encephalopathy. J Neurolog Sci 57:343, 1982
16. Branchi-Porro G, Maiolo AT, Della-Porto P: Cerebral blood flow and metabolism before and after
portacaval shunt operation. Gut 10:894, 1969
17. Breen KJ, Bryant RE, Levinson JD, et al: Neomycin absorption in man: Studies of oral and enema
administration and effect of intestinal ulceration. Ann Intern Med 76:211, 1972
18. Bresci G, Parisi G, Banti S: Management of hepatic encephalopathy with oral zinc supplementation:
A long-term treatment. Eur J Med 2:414, 1993
19. Brouillet EP, Shinobu L, McGarbey U, et al: Manganese injection into the rat striatum produces
excitotoxic lesions by impairing energy metabolism. Exp Neurol 120:89, 1993
20. Bucci L, Palmieri GC: Double-blind, double-dummy comparison between treatment with rifaximin
and lactulose in patients with medium to severe degree hepatic encephalopathy. Curr Med Res Opin
13:1109, 1993
21. Butterworth RF: Pathogenesis and treatment of portal-systemic encephalopathy: An update. Dig Dis
Sci 37:321, 1992
22. Butterworth RF: Portal-systemic encephalopathy: A disorder of neuron-astrocytic metabolic

trafficking. Dev Neurosci 15:313, 1993
23. Cadranel JF, Younsi M, Pidoux B, et al: Flumazenil therapy for hepatic encephalopathy in cirrhotic
patients: A double-blind pragmatic randomized, placebo study. Eur J Gastroenterol Hepatol 7:325-9,
1995
179
24. Cangrano C, Cardelli-Cangrano P, James HL, et al: Brain microvessels take up large neutral amino
acids in exchange for glutamine: Cooperative roles of Na+ dependent and Na+ independent systems. J
Biol Chem 258:8949, 1983
25. Chamuleau RAFM, Bosman DK, Bovee WMMJ, et al: Chronic hepatic encephalopathy studied by
NMR spectroscopy and MRI in man [abstract]. J Hepatol 13(suppl 2):S19, 1991
26. Conn HO: Hepatic encephalopathy. In Schiff L, Schiff ER (eds): Diseases of the Liver, ed 7.
Philadelphia, JB Lippincott Co, 1993, pp 1036
27. Conn HO, Lieberthal MM (eds): The Hepatic Coma Syndromes and Lactulose. Baltimore, Williams
and Wilkins, 1978, pp 340
28. Cordoba J, Gottstein J, Blei AT: Glutamine, myo-inositol and organic brain osmolytes after
portacaval anastomosis in the rat: Implications of ammonia-induced brain edema. Hepatology 24:919,
1996
29. Cordoba J, Steindl P, Cabrera J, et al: Melatonin and sleep disturbances in cirrhosis. Hepatology

24:A1300, 1996
30. Cuilleret G, Pomier-Layrargues G, Pons F, et al: Changes in brain catecholamine levels in human
cirrhotic hepatic encephalopathy. Gut 21:565, 1980
31. Dasani BM, Sigal SH, Lieber CS: The role of Helicobacter pylori in the pathogenesis and treatment
of chronic hepatic encephalopathy [abstract]. Hepatology 22:161, 1995
32. De la Pena A, Herrero JI, Betes MT, et al: Hyperintensity of basal ganglia on T1 -weighted MRI in
liver cirrhosis is associated to hyperbilirubinemia and portal systemic shunting. Hepatology 24:1308A,
1996
33. deWade J-P, Audet RM, Leong DK, et al: Portacaval anastomosis induces region-selective
alterations of the endogenous opiod system in the rat brain. Hepatology 24:895, 1996
34. Dillon D, Schenker S: Cerebrospinal fluid protein concentration in hepatic coma [letter]. JAMA
221:507, 1972
35. Di Piazza S, Filippazzo MG, Valenza LM, et al: Rifaximine versus neomycin in the treatment of
portosystemic encephalopathy. Ital J Gastroenterol 23:403, 1991
36. Duchini A, Govindaragon S, Santucci M, et al: Effects of tumor necrosis factor-2 and interleukin-6
on fluid-phase permeability and ammonia diffusion in CNS-derived endothelial cells. J Invest Med
44:474, 1996
37. Eriksson LS, Conn HO: Branched chain amino acids in the management of hepatic encephalopathy:
An analysis of variants. Hepatology 10:228-246, 1989
38. Ferenci P, Grimm G, Meryn S: Successful long-term treatment of portal-systemic encephalopathy

by the benzodiazepine antagonist flumazenil. Gastroenterology 96:240-3, 1989
39. Ferenci P: Hepatic encephalopathy. In McIntyre N, Benhamou J-P, Bircher J, Rizzetto M, et al

(eds): Oxford Textbook for Clinical Hepatology. Oxford, Oxford University Press, 1991, pp 473
40. Ferenci P: Critical evaluation of the role of branched chain amino acids in liver disease. In Thomas
JC, Jones EA (eds): Recent Advances in Hepatology. Edinburgh, Churchill Livingstone, 1986, p 137
41. Ferenci P: Critical evaluation of the role of branched chain amino acids in liver disease. In Thomas
JC, Jones EA (eds): Recent Advances in Hepatology. Edinburgh, Churchill Livingstone, 1986, pp
137-54
42. Ferenci P, Puspok A, Steindl P: Current concepts in the pathophysiology of hepatic encephalopathy.
Eur J Clin Invest 22:573, 1992
43. Geissler A, Lock G, Frund R, et al: Cerebral abnormalities in patients with cirrhosis detected by
proton magnetic resonance spectroscopy and magnetic resonance imaging. Hepatology 25:48, 1997
44. Gillis JC, Brogden RN: Rifaximin: A review of its antibacterial activity, pharmacokinetic properties
and therapeutic potential in conditions mediated by gastrointestinal bacteria. Drugs 3:467, 1995
45. Gitlin N: Hepatic encephalopathy. In Zakim D, Boyer TD (eds): Hepatology: A Textbook of Liver
Diseases. Philadelphia, WB Saunders Co, 1996, pp 605
180
46. Grau E, Marcaida G, Montoliu C, et al: Effects of hyperammonemia on brain protein kinase C
substrates. Metab Brain Dis 11:205, 1996
47. Greenberger NJ, Carley J, Schenker S, et al: Effect of vegetable and animal protein diets in chronic
hepatic encephalopathy. Am J Dig Dis 22:845, 1977
48. Hauenstein KH, Haag K, Ochs A, et al: The reducing stent: Treatment for transjugular intrahepatic

portosystemic shunt-induced refractory hepatic encephalopathy and liver failure. Radiology 194:175,
1995
49. Haussinger D, Laubenberger J, Vom Dahl S, et al: Proton magnetic resonance spectroscopy studies
on human brain myo-inositol in hypo-osmolarity and hepatic encephalopathy. Gastroenterology
107:1475, 1994
50. Helton WS, Belshaw A, Althaus S, et al: Critical appraisal of the angiographic portacaval shunt
(TIPS). Am J Surg 165:566, 1993
51. Hermenegildo C, Marcaida G, Montoliu C, et al: NMDA receptor antagonists prevent acute
ammonia toxicity in mice. Neurochem Res 21:1237, 1996
52. Herrmann S, Krieger D, Jansen O, et al: Effect of liver transplantation on hepatic encephalopathy
and hyperintense globus pallidus on T1 -weighted MRI in liver failure. Neurology 45:A298, 1995
53. Hindfelt B, Plum F, Duffy TE: Effect of acute ammonia intoxication on cerebral metabolism in rats
with portacaval shunts. J Clin Invest 59:386, 1977
54. Howell GA, Welch MG, Frederickson CJ: Stimulation-induced uptake and release of zinc in
hippocampal slices. Nature 308:736, 1984
55. Jalan R, Seery JP, Taylor-Robinson SD: Review article: Pathogenesis and treatment of chronic
hepatic encephalopathy. Aliment Pharmacol Ther 10:681, 1996
56. Kato M, Hughes RD, Keays RT, et al: Electron microscopic study of brain capillaries in cerebral
edema from fulminant hepatic failure. Hepatology 15:1060, 1992
57. Kato A, Kaneta Y, Ohara H, et al: Regional differences of cerebral glucose metabolism in patients
with liver cirrhosis as related to the presence of subclinical hepatic encephalopathy using position
emission tomography. Hepatology 24:1307A, 1996
58. Kerlan RK Jr, LaBerge JM, Gordon RL, et al: Transjugular intrahepatic portosystemic shunts:
Current status. AJR Am J Roentgenol 164:1050, 1995
59. Knecht K, Michalak A, Rose C, et al: Decreased glutamate transporter (GLT-1) gene expression in
brain in acute liver failure. Hepatology 24:486A, 1996
60. Koh J, Choi DW: Zinc alters excitatory amino acid neurotoxicity on cortical neurons. J Neurosci
8:2164, 1988
61. Kosenko E, Kaminsky YG, Felipo V, et al: Chronic hyperammonemia prevents changes in brain
energy and ammonia metabolites induced by acute ammonium intoxication. Biochim Biophys Acta
1180:321, 1993
62. Kosenko E, Kaminsky Y, Grau E, et al: Brain ATP depletion induced by acute ammonia
intoxication in rats is mediated by activation of the NMDA receptor and Na+, K+-ATPase. J
Neurochem 63:2172, 1994
63. Kreis R, Ross BD, Farrow NA, et al: Metabolic disorders of the brain in chronic hepatic
encephalopathy detected with H-1 MR spectroscopy. Radiology 182:19, 1992
64. Krieger D, Krieger S, Jansen O, et al: Manganese and chronic hepatic encephalopathy. Lancet
346:270, 1995
65. Krieger S, Jauss M, Jansen O, et al: Neuropsychiatric profile and hyperintense globus pallidus on
T1-weighted magnetic resonance images in liver cirrhosis. Gastroenterology 111:147, 1996
66. LaBerge JM, Ring EJ, Gordon RL, et al: Creation of transjugular intrahepatic portosystemic shunts
with the Wallstent endoprosthesis: Results in 100 patients. Radiology 187:413, 1993
67. Lavoie J, Layrargues GP, Butterworth RF: Increased densities of peripheral-type benzodiazepine
receptors in brain autopsy samples from cirrhotic patients with hepatic encephalopathy. Hepatology
11:874, 1990

68. Lockwood AH, McDonald JM, Reiman RE, et al: The dynamics of ammonia metabolism in man's
effects of liver disease and hyperammonemia. J Clin Invest 63:449, 1979
69. Lockwood AH, Ginsberg MD, Rhoades HM, et al: Cerebral glucose metabolism after portacaval
shunting in the rat. J Clin Invest 78:86, 1986
70. Lockwood AH, Yap EWH, Wong W-H: Cerebral ammonia metabolism in patients
181
with severe liver disease and minimal hepatic encephalopathy. J Cerebr Blood Flow Metab 11:337,
1992
71. Lockwood AH, McDonald JM, Rieman RE: The dynamics of ammonia metabolism in men: Effects
of liver disease and hyperammonemia. J Clin Invest 63:449, 1979
72. Lombardi G, Mannaioni G, Leonardi P, et al: Ammonium acetate inhibits ionotropic receptors and
differentially affects metabotropic receptors for glutamate. J Neural Transm 97:187, 1994
73. Lunzer M: Treatment of chronic hepatic encephalopathy with levodopa. Gut 15:555, 1974
74. Marcaida G, Felipo V, Hermenegildo C, et al: Acute ammonia toxicity is mediated by the NMDA
type of glutamate receptors. FEBS 296:67, 1992
75. Marcaida G, Minana M-D, Burgal M, et al: Ammonia prevents activation of NMDA receptors by
glutamate in rat cerebellar neuronal cultures. Eur J Neurosci 7:2389, 1995
76. Marchesini G, Fabbri A, Bianchi G, et al: Zinc supplementation and amino acid-nitrogen
metabolism inpatients with advanced cirrhosis. Hepatology 23:1084, 1996
77. Martin M, Zajko AB, Orons PD, et al: Transjugular intrahepatic portosystemic shunt in the
management of variceal bleeding: Indications and clinical results. Surgery 114:719, 1993
78. Massa P, Vallerino E, Dodero M: Treatment of hepatic encephalopathy with refaximin: Double
blind, double dummy study versus lactulose. Eur J Clin Res 4:7, 1993
79. McClain GJ, Kromhout JP, Elson MK, et al: Hyperprolactinemia in portal systemic encephalopathy.
Dig Dis Sci 26:353, 1981
80. Mendenhall CL, Rouster S, Marshall L, et al: A new therapy for portal systemic encephalopathy.
Am J Gastroenterol 81:540, 1986
81. Morgan MH, Read AE, Speller DC: Treatment of hepatic encephalopathy with metronidazole. Gut
23:1, 1982
82. Michalak A, Rose C, Butterworth RF: Evidence for glutamatergic synaptic dysfunction in acute liver
failure: Loss of glutamate (non-NMDA) receptors. Hepatology 24:1299A, 1996
83. Michel H, Solere M, Granier P, et al: Treatment of cirrhotic hepatic encephalopathy with L-dopa: A
controlled trial. Gastroenterology 79:207-211, 1980
84. Minana M-D, Felipo V, Grisolfa S: Protective effect of long term ammonium ingestion against acute
ammonium intoxication. Biochem Biophys Res Commun 153:979, 1988
85. Minana M-D, Hermenegildo C, Llansola M, et al: Carnitine and choline derivatives containing a
trimethylamine group prevent ammonia toxicity in mice and glutamate toxicity in primary cultures of
neurons. J Pharmacol Exp Ther 279:194, 1996
86. Morgan MY: Branched chain amino acids in the management of chronic liver disease: Facts and
fantasies. J Hepatol 11:133-141, 1990
87. Morgan MY, Jakobovits A, Elithorn A, et al: Successful use of bromocriptine in the treatment of a
patient with chronic portal-systemic encephalopathy. N Engl J Med 296:793-4, 1977
88. Morgan MY, Hawley KE: Lactilol vs lactulose in the treatment of acute hepatic encephalopathy in

cirrhotic patients: A double-blind, randomized trial. Hepatology 7:1278, 1987
89. Morgan MY, Jakobovits AW, James IM, et al: Successful use of bromocryptine in the treatment of
chronic hepatic encephalopathy. Gastroenterology 78:663-670, 1980
90. Mullen KD, McCullough AJ: Problems with animal models of chronic liver disease: Suggestions for
improvement in standardization. Hepatology 9:500, 1989
91. Muto Y: Clinical study on the relationship of short-chain fatty acids and hepatic encephalopathy. Jpn
J Gastroenterol 63:19, 1966
92. Naylor CD, O'Rourke K, Detsky AS, et al: Parenteral nutrition with branched chain amino acids in
hepatic encephalopathy: A meta-analysis. Gastroenterology 97:1033-1042, 1989
93. Olasmaa M, Rothstein JD, Guidotti A, et al: Endogenous benzodiazepine receptor ligands in human
and animal hepatic encephalopathy. J Neurochem 55:2015, 1990
94. Peters S, Koh J, Choi DW: Zinc selectively blocks the action of N-methyl-D-aspartate on cortical
neurons. Science 236:589, 1987
95. Peterson C, Giguere J-F, Cotman CW: Selective loss of N-methyl-D-aspartate-sensitive
182
L-[3 H]glutamate binding sites in rat brain following portacaval anastomosis. J Neurochem 55:386,
1990
96. Pomier-Layrargues G, Gignier JF, Lavoie J, et al: Flumazenil in cirrhotic patients in hepatic coma:
A randomized double-blind placebo-controlled crossover trial. Hepatology 19:32, 1994
97. Pomier-Layrargues G, Spahr L, Butterworth RF: Increased manganese concentrations in pallidum of

cirrhotic patients. Lancet 345:735, 1995
98. Powell EE, Pender MP, Chalk JB, et al: Improvement in chronic hepatocerebral degeneration
following liver transplantation. Gastroenterology 98:1079, 1990
99. Ptovin M, Finlayson MH, Hinchey EJ, et al: Cerebral abnormalities in hepatectomized rats with
acute hepatic coma. Lab Invest 55:560, 1984
100. Raabe W: Ammonium ions abolish excitatory synaptic transmission between cerebellar neurons in
primary dissociated tissue culture. J Neurophysiol 68:93, 1992
101. Raabe W: Spinal seizures in ammonia intoxication. In Felipo V, Grisolia S (eds): Hepatic
Encephalopathy, Hyperammonemia and Ammonia Toxicity. New York, Plenum Press, 1994, p 55
102. Raabe W, Lin S: Ammonia, postsynaptic inhibition and CNS-energy state. Brain Res 303:67, 1984
103. Reding P, Duchateau J, Bataille C: Oral zinc supplementation improves hepatic encephalopathy:
Results of a randomised controlled trial. Lancet 2:493, 1984
104. Riggio O, Ariosto F, Merli M, et al: Short-term oral zinc supplementation does not improve chronic
hepatic encephalopathy: Results of a double-blind crossover trial. Dig Dis Sci 36:1204, 1991
105. Rosen HM, Yoshimura N, Hodgman JM, et al: Plasma amino acid patterns in hepatic
encephalopathy of differing etiology. Gastroenterology 72:483, 1977
106. Ross BD, Jacobson S, Villamil F, et al: Subclinical hepatic encephalopathy: Proton MR
spectroscopic abnormalities. Radiology 193:457, 1994
107. Ross BD, Morgan MY, Cox IJ, et al: Continuing defect in cerebral energy metabolism in patients
with persistent hepatic encephalopathy (PHE): Monitoring with 3 P MRS. Clin Sci 72(suppl 16):26P,
1987
108. Rossle M, Haag K, Ochs A, et al: The transjugular intra-hepatic portosystemic stent-shunt

procedure for variceal bleeding. N Engl J Med 330:165, 1993
109. Rudman D, Aksun S, Galambos JT, et al: Observations on nitrogen metabolism of patients with
portal cirrhosis. Am J Clin Nutr 23:1203, 1970
110. Sanyal AJ, Freedman AM, Shiffman ML, et al: Portosystemic encephalopathy after transjugular
intrahepatic portosystemic shunt: Results of a prospective controlled study. Hepatology 20:46-55, 1994
111. Schenker S, Brady CE III: Pathogenesis of hepatic encephalopathy. In Conn HO, Bircher J (eds):
Hepatic Encephalopathy: Syndromes and Therapies. Bloomington, IL, Medi-Ed Press, 1994, p 43
112. Schenker S, McCandless DW, Brosby E, et al: Studies on the intracerebral toxicity of ammonia. J
113. Soeters PB, Van Leeuwen M: Laktulose, neomyzin und ammoniakstoffwechsel. Infusion Therapie
und Klinische Ernahrung 13:186, 1986
114. Sonberg KA, Riegler JL, Doherty M, et al: Hepatic encephalopathy following transjugular
intrahepatic portosystemic shunts (TIPS): Incidence and risk factors. Hepatology 90:549, 1995
115. Spahr L, Butterworth RF, Fontaine S, et al: Increased blood manganes in cirrhotic patients:
Relationship to pallidal magnetic resonance signal hyperintensity and neurological symptoms.
116. Stabenau JR, Warren KS, Rall DP, et al: The role of pH gradient in the distribution of ammonia
between blood and cerebrospinal fluid, brain and muscle. J Clin Invest 38:373, 1959
117. Staedt U, Leweling H, Gladisch R, et al: Effects of ornithine aspartate on plasma ammonia and
plasma amino acids in patients with cirrhosis: A double-blind randomized study using a four-fold
crossover design. J Hepatol 19:424, 1993
118. Sushma S, Tandon RK, Jain S, et al: Sodium benzoate in the treatment of acute hepatic
encephalopathy: A double-blind randomized trial. Hepatology 16:138, 1992
183
119. SwainMS, Blei AT, Butterworth RF, et al: Intracellular pH rises and astrocytes swell after
portacaval anastomosis in rats. Am J Physiol 261:R1491, 1991
120. Swart GR, Van den Berg JW, van Vuure JK, et al: Minimum protein requirements in liver cirrhosis
determined by nitrogen balance measurements at three levels of protein intake. Clin Nutr 8:329, 1989
121. Takahashi H, Koehler RC, Brusilow SW, et al: Inhibition of brain glutamine accumulation prevents
cerebral edema in hyperammonemic rats. Am J Physiol 261:H825, 1991
122. Tarao K, Ikeda T, Hayashi K, et al: Successful use of vancomycin hydrochloride in the treatment of
lactulose resistant chronic hepatic encephalopathy. Gut 31:702, 1990
123. Taylor-Robinson SD, Sargentoni J, Mallalieu RJ, et al: Cerebral phosphorus-31 magnetic resonance
spectroscopy in patients with chronic hepatic encephalopathy. Hepatology 20:1173, 1994
124. Thuluvath PJ, Edurn D, Yoe NC, et al: Increased signals seen in globus pallidus in T1 -weighted
magnetic resonance imaging in cirrhotics are not suggestive of chronic hepatic encephalopathy.
125. Uribe M, Farca A, Marquez MA, et al: Treatment of chronic portal systemic encephalopathy with
bromocriptine: A double-blind controlled trial. Gastroenterology 76:1347-1351, 1979
126. Uribe M, Marquez MA, Garcia-Ramos G, et al: Treatment of chronic portal-systemic

encephalopathy with lactose in lactase-deficient patients. Dig Dis Sci 25:924, 1980
127. Uribe M, Marquez MA, Ramos GG, et al: Treatment of chronic portal-systemic encephalopathy

with vegetable and animal protein diets: A controlled crossover study. Dig Dis Sci 27:1109, 1982
128. Van der Rijt CC, Schalm SW, Meulstee J, et al: Flumazenil therapy for hepatic encephalopathy: A
double-blind crossover study. Gastroenterol Clin Biol 19:572-580, 1995
129. Van Der Rijt CCD, Schalm SW, Schat H, et al: Overt hepatic encephalopathy precipitated by zinc
deficiency. Gastroenterology 100:1114, 1991
130. Warren KS, Schenker S: Effect of an inhibitor of glutamine synthesis (methiomine sulfaximine) on
ammonia toxicity and metabolism. J Lab Clin Med 64:442, 1964
131. Warren KS, Schenker S: Effect of temperature variation on toxicity and metabolism of ammonia in
mice. J Lab Clin Med 60:291, 1962
132. Weber FL Jr: The effect of lactulose on urea metabolism and nitrogen excretion in cirrhotic
patients. Gastroenterology 77:518-523, 1979
133. Weber FL Jr, Fresard KM, Lally BR: Effects of lactulose and neomycin on urea metabolism in
cirrhotic subjects. Gastroenterology 82:213, 1982
134. Weber FL Jr, Banwell JG, Fresard KM, et al: Nitrogen in fecal bacteria, fiber and soluble fractions
of patients with cirrhosis: Effects of lactulose and lactulose plus neomycin. J Lab Clin Med 110:259,
1987
135. Weber FL, Minco D, Fresard KM, et al: Effects of vegetable diets on nitrogen metabolism in
cirrhotic subjects. Gastroenterology 89:538, 1985
136. Weber FL Jr, Reiser BJ: Relationship of plasma amino acids to nitrogen balance and
portal-systemic encephalopathy in alcoholic liver disease. Dig Dis Sci 27:103, 1982
137. Weber FL Jr, Veach GL: The importance of the small intestine in gut ammonium production in the
fasting dog. Gastroenterology 77:235, 1979
138. Wensink J, Molenaar AJ, Woroniecka UD, et al: Zinc uptake into synaptosomes. J Neuroscience
50:782-789, 1988
139. Yao H, Sadoshima S, Fujii K, et al: Cerebrospinal fluid lactate in patients with hepatic
encephalopathy. Eur Neurol 27:182, 1987
140. Yurdoydin C, Hortnagl H, Pifl CL, et al: Modulation of hepatic encephalopathy in rats due to
thioacetamide induced liver failure by serotonin agonists and antagonists. Gastroenterology 96:A675,
1989
141. Yurdoydin C, Hortnagl H, Steind P, et al: Increased serotonergic and noradrenergic activity in
hepatic encephalopathy in rats with thioacetamide induced acute liver failure. Hepatology 12:695, 1990
142. Yurdoydin C, Karavelioghi D, Yasa H, et al: Opioid ligands in human hepatic encephalopathy.
184
143. ZeePC, Mehta R, Turek F, et al: Portacaval anastomosis disrupts circadian locomotor activity and
pineal melatonin rhythms in rats. Brain Res 560:17, 1991
144. Zieve L, Doizaki WM, Zieve FJ: Synergism between mercaptans and ammonia or fatty acids in the
production of coma: A possible role for mercaptans in the pathogenesis of hepatic coma. J Lab Clin
Med 83:16, 1974
145. Zieve L, Olsen EL: Can hepatic coma be caused by a reduction of brain noradrenaline or
dopamine? Gut 18:688, 1977



157
PORTAL HYPERTENSION
Steven Schenker MD
Michael K. Bay MD
From the Division of Gastroenterology and Nutrition, The University of Texas Health Science Center at San Antonio, San
Antonio, Texas
The goal of this article is to update the status of portalsystemic encephalopathy (PSE) in the light
of new data. First, PSE is defined in the context of other types of hepatic encephalopathy.
Subsequently, current views of the pathogenesis of this disorder are discussed, followed by an
analysis of therapeutic options. Diagnosis will not be considered, as no major new developments
have recently been documented in this area.

Steven Schenker, MD
Division of Gastroenterology and Nutrition
The University of Texas Health Science Center at San Antonio
7703 Floyd Curl Drive
San Antonio, TX 78282-7878
DEFINITION
Hepatic encephalopathy is a neuropsychiatric syndrome (abnormal mental state) that develops in patients
with hepatic disease or portalsystemic shunting of blood, and is due to these hepatocirculatory
disturbances. The term encephalopathy is much more comprehensive in describing the wide spectrum of
mental activity seen than the very circumscribed terminology of "hepatic coma." It is also generally
agreed that hepatic encephalopathy can develop in the setting of acute (initial) and fulminant
(parenchymal) liver failure (generally defined as of less than 8 weeks' duration), and that which ensues in
patients with chronic liver disease or spontaneous or surgically induced portalsystemic shunting. The

syndrome that develops in the setting of chronic liver dysfunction is usually designated as PSE based on
the concept that it is caused by shunting of noxious substances from the portal to the systemic blood
circuits, thereafter impacting on the brain. [26] [39] [45] These substances
158
are presumed to be mainly nitrogenous, with ammonia as the main culprit. Some have subdivided this
group of patients as nitrogenous PSE as opposed to chronic liver disease patients with encephalopathy
caused by other precipitants, such as the use of sedative and analgesic drugs, or hypoglycemia, which is
termed non-nitrogenous, or pseudo-PSE. [26] Although this distinction is mechanistically stimulating, in
the authors' view, it assumes more knowledge of the pathogenesis of encephalopathy than is currently
warranted. For instance, it is not known precisely by what mechanisms the pharmacodynamic
(tissue-based) effects of sedatives affect the brain of patients with severe liver disease. A variety of
summations with underlying nitrogenous effects could be present, thus complicating a pathophysiologic
definition. Until this is certain, it would seem more constructive not to split the terminology.
Other terminology refers to encephalopathy as being acute (one episode), acute recurrent, or chronic. [26]
These distinctions are fine if overt mental change with clear-cut improvement and worsening are used.
With subtle changes, picked up perhaps only by complex psychometric assessments, the terminology
may change. Different groups use variable psychometric test batteries; some include evoked potentials
and others find a spectral electroencephalogram (EEG) analysis as most sensitive. Thus, the exact gold
standard for what constitutes subclinical hepatic encephalopathy is still uncertain. [3] It would seem wise,
therefore, to define one's terms in the context of the diagnostic methods used. Finally, the issue of
irreversibility comes up. For most cases of encephalopathy due to fulminant liver failure and of PSE,
successful treatment results in apparently complete reversal of altered mental state. In the rare cases of
hepatocerebral degeneration (extrapyramidal signs and other neurologic manifestations similar to those
seen with Wilson's disease) and of spastic (demyelinating) paraparesis, any improvements noted are at
best partial. [98] Moreover, there is increasing evidence that some aspects of PSE may persist as assessed
by sophisticated psychometric techniques. [65] Whether these have structural components is uncertain at
present. [13] From these considerations, the authors have classified hepatic encephalopathy, as shown in
Table 1 .
PATHOGENESIS
The mechanisms of hepatic encephalopathy have been extensively investigated. Despite an accelerated
pace of such studies in recent years and much progress in the field, [21] [22] [42] [46] [55] the precise causes of
mental changes in patients with liver disease are still uncertain. The following discussion cites some of
the problems in carrying out such research, presents the current status of the field, and identifies future
investigative needs.
159
TABLE 1 -- CLASSIFICATION OF HEPATIC ENCEPHALOPATHY

Clinical Hepatic Laboratory

Syndrome Onset Aspects Histology Reversibility Data
Fulminant hepatic Rapid Failure of Hepatic Yes
Blood
failure synthetic necrosis
sugar, PT,
function (
Ammonia
BS, PT),
increased
intracranial
pressure in
setting of acute
(8 wks or less)
liver failure
Portalsystemic Gradual Precipitated by Cirrhosis *
Yes Ammonia
encephalopathy various
derangements (usually)
(i.e.,
gastrointestinal
bleeding, renal
failure,
infection, etc.),
chronic liver
disease with or
without,
portalsystemic
shunting
Hepatocerebral Gradual Characteristic Cirrhosis Partial
Ammonia
degeneration neurologic
findings
related to
altered basal
ganglia
function
Myelopathy Gradual Spastic Cirrhosis Partial
Ammonia
paraparesis (?)
altered
sphincter
function
Urea cycle enzyme Gradual Neonatal Yes
deficiency Normal
onset, by Ammonia
dietary
nitrogen

Reye's Rapid Childhood Microvesicular Yes

onset, fat
Ammonia, low
post-infectious (mitochondrial
precipitated by damage) bilirubin,
aspirin, PT, Blood
hypoglycemia, sugar
increased
intracranial
pressure
*Spontaneous and surgical shunting of blood from portal to systemic circulation is important.
Subtle changes in psychometric assessment may remain (i.e., chronic, subclinical PSE).
Ammonia not always elevated. May be normal if PSE precipitated by sedatives, analgesics, etc.
Occasional cases of post-liver transplantation partial recovery.
Abnormal urea cycle enzymes.
Diffuse changes present (brain and liver).
160
Problems in Studies of Hepatic Encephalopathy
A key difficulty in investigating this field in the past, and to a considerable degree at present, has been
the inability to study in detail, and with safety, human cerebral metabolism. Animal models of hepatic
encephalopathy are better counterparts of fulminant human encephalopathy than of mental changes
(PSE), which develop with chronic liver disease in patients. Fulminant liver failure in experimental
animals is usually precipitated by the use of toxins (i.e., carbon tetrachloride or galactosamine) or
ischemia. Acute encephalopathy has been studied by administration of large doses of ammonia. Thus, the
relevance of data obtained to the encephalopathic process seen in patients with PSE is questionable. The
animal model for PSE has conventionally been the rat with a surgical portacaval shunt and the additional
administration of ammonia. Although the presence of portacaval shunting alone may result in subtle
behavioral changes and altered sleep cycle (perhaps resembling subclinical human PSE), and
administration of ammonia will usually produce overt neurologic signs, it is generally agreed that this
animal model system does not fully reflect PSE. [90] More specifically, cirrhosis is absent, and there is
much variation in symptoms and signs of encephalopathy, probably due to experimental differences. In
addition, physiologic studies (cerebral blood flow, oxygen extraction, glucose consumption) in these
animals are not easy to do sequentially. Thus, such data, together with cerebral biochemical assessments,
have not been systematically obtained.
The problem of animal models and their extrapolation to patients can be added to the known
heterogeneity of the brain data (structurally, biochemically, functionally) not reflected by analysis of the
whole brain or even several brain regions. Examples of such heterogeneity are the diverse roles of
various brain areas, such as the basal ganglia, cortex and reticular activating system, and of different

cells, such as neurons versus astroglial components. It is not surprising, therefore, that the exact areas and
their interaction, which define normal consciousness (awareness), are still elusive. As discussed
elsewhere, it is generally believed, however, that the minimal areas needed for intact awareness are the
reticular activating system in the brainstem and the cortex. These areas are believed to define,
respectively, the off and on aspects and the content of consciousness, respectively. [111] Clearly, there are
many other interactions and modifiers.
Finally, a major problem has been the determination of whether observed changes in mental (neurologic)
status and biochemical/physiologic abnormalities are casual or causally related. Ideally, such conclusions
depend on the temporal sequence, extent of changes, reproducibility of data in various models, and
perhaps most importantly, their concomitant reversal. An example is worsening of mental state with
increased nitrogen load and reversal (amelioration) of both with treatment. Such data are difficult to
obtain in the biochemical/physiologic areas, especially as it concerns the brain.
161
The pathogenesis of PSE can be conveniently discussed sequentially under the interrelated categories of
toxins, alterations of blood-brain barrier, mechanisms of neurotoxin action (neurotransmitter hypothesis,
altered energetics), and precipitants of PSE (nitrogenous and non-nitrogenous).
Toxins
Ammonia.
Of the various hypothetical toxins, none has as strong a basis as excess ammonia. As shown in the
following list, this is derived from both clinical and experimental data:
Ammonia is increased in arterial blood of 80%-90% of patients with PSE.
Nitrogenous products (as well as ammonia) may induce PSE in patients with cirrhosis or
portalsystemic shunting of blood.
Spinal fluid glutamine (derivative of brain ammonia metabolism) is increased in virtually all
patients with PSE.
In patients with inborn errors of urea cycle, high levels of ammonia are associated with
encephalopathy and treatment aimed at preventing/reducing the ammonia also prevent/ameliorate
the encephalopathy.
Treatment aimed at reducing influx of ammonia (and other nitrogenous substances) into the brain
benefits patients prone to PSE.
Animals with liver damage or portacaval shunts exposed to ammonia develop neurotoxicity.

Multiple mechanisms for cerebral ammonia neurotoxicity have been established.
Two often-cited deterrents are the absence of elevated blood ammonia in every patient with PSE and the
presence of a stage of excitation prior to sedation in animals given ammonia acutely; however, the
former concern is readily explained by greater permeability of the blood brain barrier to ammonia in
cirrhotics (i.e., increased brain ammonia), [70] as well as the need for fastidious methodology of blood
ammonia assays. [26] The latter (excitation stage) actually corresponds fairly well to the manifestations of
coma developing rapidly in patients with fulminant liver failure rather than with PSE. Moreover, it is
entirely possible, if not likely, that excess ammonia is only one of various toxins responsible for PSE.
Mercaptans.
These substances are derived from subnormal metabolism by the damaged liver of sulfur-containing
amino acids. Classical studies by Zieve et al [144] have shown synergism of ammonia, short-chain fatty
acids, and mercaptan toxicity on the brain in experimental animals; however, the validity of this concept
has been put to question by methodologic differences in measuring mercaptans, as well as lack of clear
information on their mechanisms of neurotoxicity. The current consensus is that their contribution to PSE
in patients is not likely to be significant. [14]
162
The metabolites (phenols) of other (aromatic) amino acids have not been recently studied, although their
levels in blood and cerebrospinal fluid (CSF) of patients with PSE are not very high. [55] Short- and
medium-chain fatty acids may induce reversible coma in experimental animals, possibly by an effect on
Na+ -K+ ATPase, but only in concentrations much higher than those seen clinically with PSE. Moreover,
the severity of PSE and concentrations of fatty acids in blood do not seem to correlate. [91]
Indoles.
There is substantial evidence, including data in human brain and CSF, that the neurotransmitter,
serotonin, and especially the turnover of this system, are enhanced in the setting of PSE. [1] [15] [42]
Because this neurotransmitter may participate in the arousal reaction (sleep-wake cycles), it could impact
on decreased consciousness in patients with PSE. A number of experimental studies, however, with
serotonin agonists and antagonists, in portacaval-shunted rats and those with acute hepatic failure have
not documented consistent coordinated changes of neurobehavioral behavior and serotonin turnover [11]
[140] ; this area deserves further study. It appears also that quinolinic acid, a derivative of tryptophan,
which is a potent agonist of the N-methyl-D-aspartate (NMDA) receptor, is not a significant player in
experimental post-shunt encephalopathy. [12]
Melatonin.
In view of the altered (inverted) sleep cycle seen in many patients with mild PSE, and the known role of
melatonin in promoting sleep, this hormone was investigated. It has been shown that portacaval shunting
in rats disrupts both the pineal melatonin rhythm and circadian locomotor activity. [143] This is probably
not surprising as melatonin is extensively and rapidly metabolized by the liver. Most recent data suggest,
however, that changes in the area of the suprachiasmatic nucleus in brain of shunted animals rather than

the concurrent changes in melatonin may be responsible for abnormal locomotor activity and in cirrhotic
patients with sleep disturbances; melatonin was not related to these. [29] [143]
Gamma Aminobutyric Acid (GABA).
This potent inhibitory neurotransmitter was previously believed to be responsible for PSE. The scenario
was that GABA generated in the gut may escape hepatic metabolism, pass via an altered blood-brain
barrier, and impact on the brain. Current data do not support this precise mechanism, primarily because
the brain and spinal fluid concentrations of GABA (measured accurately) are not significantly increased
in patients with PSE [21] [42] [111] ; however, it is very likely that enhanced GABA-ergic (inhibitory)
neurotransmission may have a role in PSE.
Endogenous Benzodiazepines.
There is now evidence for the presence of endogenous benzodiazepine-like substances that may enhance
GABA-ergic inhibitory brain activity. The basis for this is the isolation of such substances from the brain
of patients and experimental animals with acute hepatic coma, [9] [10] [93] and the apparent improvement in
the mental state of some such patients given benzodiazepine antagonists (i.e., flumazenil). [96] The exact
source (plant or bacterial) of these substances is not known, and their concentration is much lower than
the amount of benzodiazepines necessary to induce significant sedation in
163
patients without liver disease. Still, their effect could well be contributory to PSE. Interestingly, the
density and affinity of these cerebral benzodiazepine receptors is not decreased in human PSE, as would
be expected if the functional effects of these endogenous ligands were significant. The role of antagonists
and partial inverse agonists for these receptor complexes were discussed elegantly in a recent review. [42]
It appears that it is the latter group of agents (rather than pure antagonists) that improve neurobehavior in
experimental encephalopathy. The exact mechanistic interpretation of these agents in terms of
modulating inhibitory GABA/benzodiazepine-based neurotransmission is clearly in need of further
study.
Additionally, there is evidence in both experimental animals and in human PSE of increased density in
brain of peripheral-type benzodiazepine receptors. [67] These are not part of the GABA-benzodiazepine
receptor complex, but are localized in astrocytic mitochondria. Because ammonia is detoxified to
glutamine in this cell compartment, and energy synthesis occurs in mitochondria, this links
geographically these two mechanisms.
Other Neuroactive Substances.
Noradrenaline and dopamine are other important true synaptic transmitters. It has been postulated that
the increased aromatic/branched-chain amino acid ratio in patients with PSE may result in increased
transport of the aromatic amino acids into the brain, and this may inhibit the synthesis of
catecholaminergic neurotransmitters by accumulation of false neurotransmitters in the synaptic clefts. [21]
[111] In this context, the aromatic amino acids would function as "toxins," and depletion of true
neurotransmitter-norepinephrine and dopamine would be the result. These amino acid changes likely
result from selectively greater degradation of branched amino acids by peripheral muscle and the

catabolic effects of increased glucagon on the liver. Indeed, decreased catecholamines have been
demonstrated in the brain of animals with liver failure. Dopamine concentrations, on the other hand, have
not been diminished. [141] More important, the concentrations of these two true neurotransmitters have not
been shown to be decreased in postmortem brain tissue of patients with PSE. [30] This is somewhat
surprising as serum prolactin levels, usually seen with low brain dopamine, are reported in patients with
PSE. [79] Of interest, in autopsies of cirrhotics, a selective decrease in dopamine binding sites has been
noted in the pallidum, associated with an increase in manganese. [19] Other experimental studies wherein
brain dopamine is depleted and octopamine increased did not cause changes in mental status. [145]
Administration of l-dopa (to raise brain dopamine) in patients with PSE also was not therapeutically
helpful, and the value of giving branched-chain amino acids specifically to treat encephalopathy is not
generally accepted as helpful. [39] [40] Thus, the norepinephrine/dopamine depletion concept due to
presumed "toxic" effects of excess aromatic amino acids as an important cause of PSE is at best
debatable.
Manganese.
It is now well established that many patients with PSE have hyperintense signaling from the area of the
basal ganglia on T1 -weighted
164
MR imaging. [65] This is believed to be due to accumulation of manganese at that site. [64] [97] Indeed,
blood manganese correlates roughly with pallidal hyperintensity. [64] [115] Manganese toxicity, in turn,
may cause neurologic manifestations similar to those seen in some patients with PSE; this may be due to
a disturbance of dopaminergic neurotransmission. Although this concept is novel and interesting, there
has been no established correlation of changes (onset/offset) of PSE manifestations and these MR
imaging findings in patients with overt encephalopathy. Encephalopathy, overt or as assessed by
psychometric tests, also does not correlate with MR imaging signal hyperintensity, [32] [65] [124] although
extrapyramidal symptoms (tremor) seem to. [115] The latter, which are not common as a manifestation of
PSE, would be consistent with the localization of manganese in the basal ganglia. The possibility that
some subtle chronic neurologic changes may be due to manganese effect requires further study. [65]
Improvement in encephalopathy and decreased hyperintensity of globus pallidus has been reported after
successful hepatic transplantation. [52]
Opioids/Enkephalins.
The possible role of these neurotransmitters in PSE is virtually unstudied, although preliminary data
show increased met-enkephalins in plasma and CSF of patients with PSE. The specificity of this finding
is uncertain as similar results were found in uremia. [142] In rats with portacaval shunts, region-selective
changes in beta-endorphin and in opioid receptors in brain have been reported. [33] The functional
significance of these findings requires further study.
Blood-brain Barrier
The blood-brain barrier is established by the endothelial cells of capillaries with foot processes of
astroglial cells in close apposition. There is no evidence of a nonspecific (general) breakdown of the
blood-brain barrier in patients with PSE. This is best exemplified by the absence of increased protein

concentration in the spinal fluid of most of these patients. [34] Unlike in fulminant liver failure, cerebral
edema is very rare in patients with PSE. There is, however, apparently a selective change in this barrier.
This is best shown by the increased transfer of ammonia into the brain in such individuals. [70] Recent
studies using central nervous system (CNS)-derived endothelial cells have shown that tumor necrosis
factor (TNF) may enhance the fluid-phase permeability and diffusion of ammonia by pinocytosis. [36]
Interestingly, TNF may be increased in patients with liver disease, and pinocytic activity has been shown
to be increased in blood-brain (CNS-capillary) cells of animals with experimentally induced acute liver
failure. [56] [99] Transport of certain amino acids (i.e., tryptophan) by a specific carrier may also be altered
in such patients. The transfer of neutral amino acids is increased, that of glucose and basic amino acids is
decreased, and the flux of glutamine in the reverse direction (from brain to blood) may be increased. [24]
Interestingly, in rats with a portacaval shunt, the pericapillary astrocytes swell, become more alkaline,
and are more depolarized than in sham-operated controls. [119]
165
Thus, in patients with PSE, selective changes in the blood-brain barrier may occur and may contribute to
neurotoxic effects of some substances. Mechanisms of Neurotoxin Action
As described previously, excess ammonia appears to be the predominant neurotoxin in PSE. It is likely
that other "toxins" contribute to this adverse effect on the brain. Of the ones studied (and discussed
previously), the GABA-ergic influences (i.e., endogenous benzodiazepine-like substances) seem the most
likely culprits. Other nitrogenous derivatives (i.e., trimethylamine-N-oxide) may play a part, but require
more study. [55]
Neurotransmitter Imbalance Concept.
Ammonia.
In formulating a mechanism of action of PSE, one needs to concentrate first on ammonia. Ammonia
normally is generated primarily from the ingestion of protein or other nitrogenous substances. Some
ammonia can also be released from exercising muscles and kidneys (with acidosis, hypokalemia, and
diuresis). Ammonia is primarily converted in the liver to urea (and to a small extent, glutamine), and the
urea (and some ammonia) is excreted via the kidneys. [26] Some 80% of a single bolus of ammonia is
detoxified in a single passage by the liver. [26] Normally a small amount of ammonia is taken up by
resting skeletal muscle. In patients with chronic liver disease, ammonia detoxication is greatly
compromised. This may be due to damaged liver cells with impaired ureagenesis, shunting of blood from
the portal to the venous systems effectively by passing the liver, or both factors. A decreased muscle
mass may contribute, with inability to act as a major safety valve uptake mechanism for ammonia. This
is despite a much larger than usual extraction of ammonia by muscle as compared to the liver in
cirrhosis. [69] Excess production of ammonia (i.e., gastrointestinal hemorrhage or from increased gut urea
due to concomitant renal failure) may, and often does, compound the problem of impaired detoxication
of ammonia. The excess ammonia that reaches the systemic circulation passes into the brain via a
selectively impaired blood-brain barrier. [70] Any increase in blood pH (i.e., hypokalemic alkalosis)
increases the nonionized ammonia component (Henderson-Hasselbalch mechanism) and drives this
readily permeable ammonia down a pH gradient into the brain (acid sink phenomenon). [26] [131] [116]
Moreover, in PSE patients there is an increase in the cerebral metabolic rate for ammonia (for uncertain
reasons). [70] In the brain, especially in the white matter, [68] ammonia is converted to glutamine by an

energy-consuming reaction in the astroglial cells. These are, of course, the cells that are responsible for
the Alzheimer type II astrocytosis of PSE. These changes refer to an increase in size and number of
protoplasmic astrocytes that have enlarged pale nuclei with irregular chromatin and prominent nucleoli.
[22]
High concentrations of ammonia are known to have a direct toxic effect via alteration of cellular chloride
channels on both excitatory and
166
inhibitory neurotransmission in the brain. [101] [102] This is apparently exerted at both the postsynaptic and
synaptic mechanisms. [100] [102] Raabe [102] has suggested that these effects are sufficient to explain most of
the toxic effects of ammonia on the brain. It is known, however, that inhibition of ammonia metabolism
(glutamine synthesis) with methionine sulfoximine protects rodents from cerebral ammonia toxicity
despite higher ammonia levels. [130] This, and other experimental mechanistic studies, [22] [46] [51] suggest
that ammonia primarily exerts its toxic effects on the brain via its metabolism and on the glutaminergic
neurotransmitter system.
Glutamate is the major excitatory neurotransmitter in the brain. With excess ammonia,
alpha-ketoglutarate combines with ammonia to form glutamate, and this, in turn, with addition of more
ammonia, forms glutamine. Glutamine, in turn, is converted back to glutamate and NH3 by the action of
glutaminase. Although glutamine formation occurs in astrocytes, the regulation of overall glutamate flux
also involves an interrelationship with presynaptic and postsynaptic neurons. [22] Much research
involving cultured astrocytes, experimental animals with portacaval shunts, as well as postmortem
studies with human brain of patients with PSE has suggested that in PSE the primary defect is in the
perineuronal astrocyte, which is unable to take up extracellular glutamate released from neurons. [22]
Thus, as measured by microdialysis in portacaval shunted rats, there is excess (overflow) glutamate in
the extracellular space, ammonia inhibits glutamate uptake by cultured astrocytes, and there is
downregulation (decreased density) of postsynaptic glutamate receptors in PSE brain. [22] The latter
would be expected in response to excess neuronally released glutamate outside of the astrocytic
receptors. Very recently, the astrocytic glutamate transporter (GLT-1) has been shown to be decreased in
brains of rats with acute ischemic liver failure, [59] and preliminary data suggest the same in
portacaval-shunted rats (personal communication, 1996). This would be consistent with impaired
neuronal-astrocytic glutamate trafficking. Astrocytic mitochondrial membranes also contain receptors for
so-called peripheral-type benzodiazepines, [22] and these are increased in brain of animals and patients
with PSE. The precise implication of these changes is uncertain, but it may relate to the toxicity of
ammonia (on mitochondria [?]) in these cells.
This defect in glutamatergic (excitatory) synaptic function in PSE has now been extended to a
specialized form of this receptor. A series of studies using portacaval shunted rats, acute and chronic
ammonia exposure, and various receptor antagonists suggest that ammonia impairs the function of the
N-methyl-D-aspartate (NMDA) receptor, a subclass of glutamate receptors. [46] [72] [75] [85] [95] The
interaction of acute versus acute on chronic (PSE-type) ammonia toxicity is complex [51] [61] [84] and not
yet fully characterized, but a series of studies by Grau et al have tentatively proposed a scheme of
ammonia action on the astrocyte (Fig. 1) (Figure Not Available) . [46] In this formulation ammonia
inhibits the phosphorylation of protein kinase C (PKC), resulting in increased activity of Na+/K+
ATPase and a subsequent depletion of ATP, the ultimate energy source. [14] The

167
Figure 1. (Figure Not Available) Effects induced in brain by hyperammonemia that are mediated by decreased
PKC-mediated phosphorylation. ( From Grau E, Marcaida G, Montoliu C, et al: Effects of Hyperammonemia on Brain
Protein Kinase C Substrates Metabol. Brain Disease 11:205-216, 1996; with permission.)
inhibition of PKC phosphorylation may also decrease the phosphorylation of the NMDA receptor. This
view is interesting in that it ties in the effects of ammonia on the glutamate (NMDA) neurotransmitter
and brain energetics; however, this concept is hypothetical because the mechanism by which ammonia
may decrease PKC phosphorylation in unknown. [46] Moreover, NMDA receptors appear to be uniquely
neuronal in location, whereas non-NMDA (glutamate) receptors are both neuronal and astrocytic [82] ;
hence, the role of the NMDA receptor needs further assessment. Also, extrapolation from various animal
models of hepatic encephalopathy to human disease states (acute liver failure versus PSE) will require
careful analysis.
Endogenous Benzodiazepine-like Substances.
These substances (discussed as possible neurotoxins previously) may exert their effect on the brain by
acting on the GABA-benzodiazepine receptor complex. This complex mediates neural inhibition via
influx of chloride ions into postsynaptic neurons, followed by an inhibitory potential. [8] [42] Activation of
this inhibitory complex may also explain the increased sensitivity of patients with chronic liver disease to
sedatives such as diazepam and triazolam, and may explain the beneficial (albeit transient) effects of the
benzodiazepine antagonist, flumazenil, in some of these patients. [6] [95]
Other Neurotransmitters.
The crux of this hypothesis is that weak
168
(false) neurotransmitters, such as beta-phenylethanolamine and octopamine, derived from aromatic

amino acids that flood the brain displace true neurotransmitters (i.e., catecholamines) in the synaptic
clefts. [21] [111] Although conceptually attractive, this hypothesis has been seriously challenged by
postmortem brain data in patients with PSE, which showed normal concentrations of catecholamines and
no accumulation of octopamine. [30] Moreover, intraventricular administration of octopamine into the
brain of experimental animals did not induce coma. [145]
Energy, initially as glucose and ultimately as ATP, is necessary for normal cerebral function.
Accordingly, numerous studies of cerebral energy stores have been carried out, first in various forms of
experimental hepatic encephalopathy, and more recently in patients. Problems in data interpretation have
been the variability in brain regions, the difficulty in carrying out studies in patients, and analysis of
information as to cause-effect. [111] Human studies recently have used positron emission tomography and
nuclear MR spectroscopy as relatively noninvasive tools. [55]
In experimental animals, using ammonia intoxication as a model, decreased ATP stores were detected,

especially in the brain stem (reticular activating system). [112] This decrease in energy has been most
recently interpreted as the result of the coma (i.e., due to decreased glucose use and not the initiating
event). [53] [55] Of interest, portacaval shunting in older rats (as compared to younger ones) seemed to
result in greater increase in peripheral-type benzodiazepine binding sites, accompanied by greater
impairment of locomotor function. [5] This corresponds to a greater sensitivity to PSE in older patients,
and may relate to altered energetics as these ligands are located on the mitochondrial membrane. Other
possibilities (i.e., greater glutamate carrier abnormality with age) are also possible.
In patients, the data on cerebral glucose utilization have been variable, depending on the severity of
encephalopathy, the method used, and area of brain studied. [16] [42] [55] [70] In general, however, a decrease
in cerebral blood flow (perhaps related to low pCO2 ), and decreased cerebral glucose oxidation was
seen. [2] [16] [70] Data on brain energetics are also variable. Some have noted impaired high-energy
phosphate metabolism [7] [55] [107] [123] using newer MR imaging technology. Others reported no
abnormality in energetics. [25] The demonstration of increased CSF lactate, which correlated with the
mental state, seems to agree with evidence of altered brain energy turnover. [139] It has been difficult to
determine if these changes in brain glucose consumption are the result of decreased energy demands or
the early cause of neurologic symptoms. Recent studies using position emission tomography to study
regional cerebral glucose metabolism have documented decreased glucose metabolism in various parts of
the brain in patients with subclinical hepatic encephalopathy as compared to two control groups. [57]
These data suggest an early abnormality of cerebral energetics in patients with PSE, but this report needs
to be confirmed and extended with longitudinal studies
169
before, during the development of, and after the treatment of PSE. Other studies have shown changes in
phospholipid precursors (rather than ATP), suggesting altered membrane composition. [25] [123] Finally, a
substantial decrease in myoinositol has been detected fairly consistently. [28] [49] [63] [106] The basis for this
is not certain, but may reflect compensatory osmoregulation to counterbalance an increase in glutamine.
[28] The net effect would be to prevent the formation of brain edema. The myoinositol decrease is actually
seen in cirrhotic patients in the absence of even subtle encephalopathy, and the decrease does not
correlate well with the severity of encephalopathy. [43]
Precipitants of PSE.
These are of the nitrogenous and non-nitrogenous varieties, and are listed together with presumed
mechanisms of action in Table 2 .
TABLE 2 -- COMMON PRECIPATATING FACTORS OF HEPATIC ENCEPHALOPATHY AND

POSSIBLE MECHANISMS
Precipitating Factor Mechanism
Azotemia Increased enterohepatic
circulation of urea with increased
ammonia production in the
intestine
Direct sedative effect of uremia

Diuretics Excessive diuresis may induce:

Hypokalemic alkalosis, which
enhances transfer of ammonia
across the blood-brain barrier and
increases renal vein ammonia
output.
Hypovolemia with decreased
perfusion of vital organs,
including brain, liver, and
kidneys.
Gastrointestinal bleeding Digestion of intraluminal blood
provides a substrate for increased
ammonia production and other
nitrogenous toxins (100 mL of
blood = 5-20 g of protein).
Sedatives, tranquilizers, and analgesics Altered
pharmacokinetics--Hepatic
oxidative degradation of many
sedative drugs is impaired in
patients with liver disease. Unless
the dose is adjusted, these drugs
then accumulate and may exert a
direct depressant effect on the
brain.
Pharmacodynamics--The brain in
these patients appear to be
intrinsically more sensitive to the
effects of these drugs. In the case
of benzodiazepines, this may
relate to the presence of an
endogenous benzodiazepine-like
substance and greater "activity"
of the GABA system.
Excess dietary protein Protein excess provides substrate
for production of ammonia and
other nitrogenous toxins in the
intestine.

Infection Increased tissue catabolism leads

to a greater endogenous nitrogen
load and increased ammonia
production. Infection may cause
dehydration and prerenal
azotemia.
Constipation Longer contact time between
nitrogenous substances and the
ammonia-forming bacteria may
increase production and
absorption of ammonia and other
toxic nitrogenous derivatives
Data from Schenker S: Hepatic coma: Current concepts of pathogenesis. Viewpoints on Digestive
Diseases 2:1-4, 1970; and Hoyumpa AM, et al: Clinical conference: Hepatic encephalopathy.
170
With most illnesses, treatment is optimal after the pathogenesis of the disease is defined. PSE is unusual
in this regard, as the treatment is quite good despite a very incomplete understanding of the problem;
however, therapy now is directed only to the peripheral sites of influx and metabolism of nitrogen. With
better understanding of the CNS effects of various toxins, centrally directed management may become
possible. For instance, with the discovery that increased myoinositol and glutamine in brain may relate to
cerebral edema in fulminant failure, therapeutic approaches may ensue. In experimental animals, this is
represented by agents that block glutamine synthesis. [121]
With the inadequacy of experimental models, the use of noninvasive MR imaging and position emission
tomography techniques in longitudinal studies of patients may yield big dividends. This may also define
which findings are the initiating events. It will also be important to determine if all types of PSE are
similar as to mechanisms, and if the astrocytic process is truly fully reversible. Although these key
studies in patients are essential, the interaction of astrocytes and neurons in terms of neurotransmitter
traffic will likely require studies in cell culture and experimental animals.
TREATMENT
The mainstay of treatment for the majority of patients with PSE has been, and continues to be, lactulose
therapy; however, given the multiple lines of evidence, each giving a partial but not complete
understanding of the pathophysiology of PSE, a variety of treatments have been proposed and examined
(to various degrees), each with a rationale based on that particular line of evidence. A comprehensive
listing of treatments of PSE, some experimental, is as follows:

Reduce dietary protein
Switch to vegetable dietary protein
Lactulose or lactitol
Lactose in lactase deficiency
Neomycin or other antibiotics (metronidazole, rifaximin, vancomycin, paromomycin)
Ornithine-aspartate
Oral zinc supplementation
Benzoate
Phenylacetate
Branched-chain amino acids (valine, leucine, isoleucine)
Flumazenil
Bromocryptine
TIPS shunt revision: reducing stent diameter or occlusion
Liver transplant
Colonic diversion or colectomy
171
Also, no discussion of this topic is complete without mention of general measures to be considered in the
treatment of PSE:
Discontinuance and avoidance of analgesics, sedatives, and tranquilizers
Control of gastrointestinal bleeding and correct anemia
Purging blood from gastrointestinal tract

Diagnose and treat infections (particularly, but not limited to, spontaneous bacterial peritonitis)
Correct respiratory or metabolic alkalosis
Correct metabolic acidosis
Correct hypoxia
Correct electrolyte abnormalities: hyponatremia, hypernatremia, hypokalemia, hyperkalemia
Prevent constipation
Provide IV glucose solution to minimize endogenous protein breakdown
Replenish multivitamin, thiamine, folate repletion
Nutrition
The time-honored approach to the treatment of PSE by nutritional means has been to restrict dietary
protein and then to gradually increase the dietary protein in small increments while assessing for PSE to
determine the threshold limits. In addition to the quantity of dietary protein, other nutritional factors
include qualitative aspects of the diet, including meat versus vegetable protein, calorie-to-nitrogen ratio,
amino acid composition, and amount of dietary fiber.
Often the patient with PSE and chronic liver disease is also malnourished. This has raised some
compelling questions. How much dietary protein is too much? How much is enough? Higher amounts of
dietary protein may improve the malnutrition, but increase the risk of PSE. Older studies have shown that
stable cirrhotic patients were in nitrogen balance and conserved nitrogen appropriately when dietary
protein was reduced to 1.0 g/kg of body weight/day. Some, but not all, patients were able to maintain
nitrogen balance at even lower levels of dietary protein (0.5 g/kg of body weight/day). [109] [136] [137] In
contrast, another study showed that a relatively high level of dietary protein (1.2 g/kg of body weight/day
was required to maintain positive nitrogen balance in stable cirrhotics. [120] No patients were noted to
manifest PSE at any time during the study, even those given 1.4 g/kg of body weight/day of dietary
protein. Although the amount of dietary protein required to maintain positive nitrogen balance varies in
the literature, it does appear that a positive nitrogen balance can help the patient with PSE by promoting
hepatic regeneration and augmenting the ability of muscle to detoxify ammonia. [71]
Vegetable protein diets may be more beneficial (less ammoniagenic)
172
than animal protein diets. It appears that an important difference in these diets is the substantially

increased amount of fiber found in the vegetable protein diet. [135] Vegetable protein diets resulted in
improved mental status compared with animal protein diets containing equal nitrogen and equal calories.
[47] An additional controlled study by Uribe showed vegetable protein diets more beneficial in patients
with hepatic encephalopathy than animal protein diets. [127] Fiber has been found to promote increased
fecal excretion of nitrogen, in large part due to incorporation of nitrogen by fecal bacteria. [135] Some
patients, however, find it very difficult to comply with vegetable protein diets, in which case other
sources of protein are allowed, accompanied by the use of lactulose (or similar agent).
Antibiotics
Antibiotics have been shown to improve PSE, and the presumed mechanism is decreased intestinal
bacterial flora resulting in reduced formation of ammonia and other potentially toxic substances. Most
commonly, neomycin has been used for this purpose, although other agents that have been used include
paromomycin, rifaximin, sulfathalidone, metronidazole, [96] and vancomycin. [122] Although neomycin is
largely not absorbed, 1% to 3% may be absorbed by either oral or enema formulation, which can lead to
ototoxicity or nephrotoxicity. [17] In addition to decreasing the activity of bacterial flora, some evidence
suggests it inhibits glutamine conversion to ammonia in the intestinal wall. [113] Largely due to the
effectiveness of lactulose and the potential for toxicity, neomycin has been used less; however, if
lactulose fails to benefit the patient, then a trial of neomycin may be warranted. It is usually given orally,
1 g every 4 to 6 hours, or as a 1% retention enema (1 to 2 g in 100 to 200 cc of isotonic saline) two or
three times daily. If neomycin fails to provide benefit, then the combination of lactulose and neomycin
has been shown to result in additive effect, [133] and should be considered. Lactulose requires bacterial
metabolism to cause intraluminal acid production, and thus intraluminal ammonia trapping, and although
neomycin may interfere with bacterial metabolism (overall, in about one third of patients), the
combination will usually result in continued beneficial acidification of stool. [27] If the stool pH is less
than 6, then lactulose metabolism is occurring, and continued use of the two drugs makes sense. [27]
A different antibiotic that has been studied more recently, and reported largely in the European literature,
is rifaximin. Rifaximin is a derivative of rifamycin and is virtually unabsorbed after oral administration.
[44] Three double-blind trials comparing rifaximin (1200 mg/d) to lactulose (from 30-120 g/d) showed
significantly improved mental status with rifaximin. [20] [78] One double-blinded study comparing
rifaximin and lactulose against neomycin and lactulose showed no significant difference in the two
treatments. [35] Another studied rifaximin (1200 mg/d)
173
with neomycin (3 g/d) in double-blinded fashion, and the results showed significant improvement in
serum NH3 in the rifaximin treatment arm and improvement in EEG, Reitan test, and severity of hepatic
encephalopathy, although not statistically significant. The side-effect profile appears to be quite
favorable, with less than 1% incidence of gastrointestinal distress (flatulence, abdominal pain, and nausea
reported). Other uncommon side effects include urticaria, weight loss, headache, leg edema, and mild
elevations in serum potassium and serum sodium.
A new development in the use of antibiotics to treat chronic encephalopathy by eradicating gastric
Helicobacter pylori has recently been described. [31] Because of the urease produced by the organism, it is
postulated that increased gastric ammonia contributes to the development of hepatic encephalopathy. In

comparing a group of Helicobacter-positive stable cirrhotics with a comparable group of stable cirrhotics
who were Helicobacter negative, a significantly increased level of gastric ammonia was found. After H.
pylori eradication therapy (amoxicillin and omeprazole), seven patients showed improvement of
symptoms, as manifested by resolution of day/night reversal and a trend toward improved trail test time.
Eradication of H. pylori may play a role in the treatment of hepatic encephalopathy; however, additional
studies need to be done to confirm these preliminary results.
Lactulose
Lactulose is a synthetic disaccharide (galactosidofructose) that is not absorbed by the gut and is
metabolized by colonic bacteria to organic acids. These organic acids result in a decrease in luminal pH,
which may cause ammonia to be trapped in the lumen in the form of ammonium. Decreased luminal pH
may also promote peristalsis. Moreover, lactulose also acts as an osmotic laxative promoting the
elimination of fecal nitrogen and other fecal material. Fecal bacteria increase incorporation of nitrogen,
probably due to the influence of lactulose or its metabolites on bacterial growth. [132] [134] In vitro studies
have shown inhibition of intestinal glutamine uptake and ammonia formation with lactulose (or
neomycin) independent of bacterial metabolism. [113]
A typical dose of lactulose is 30 mL two or three times a day given orally with the goal of two or three
soft stools a day. Some patients can be instructed to titrate their dose of lactulose to achieve this result. If
lactulose cannot be given orally (such as in patients with grade 3 or worse encephalopathy), it may be
administered as an enema (300 mL of lactulose added to 700 mL of tap water). Excessive lactulose can
result in severe diarrhea leading to hypernatremia. Lactilol (beta-galactosidofructose) is a disaccharide
similar to lactulose, which has been shown to be as effective as lactulose in the treatment of hepatic
encephalopathy [88] ; at present, it is not available in the United States. Lactilol is not as sweet as
lactulose, and is associated with less nausea and abdominal bloating.
174
In one study, lactose was shown to be an effective treatment for PSE in those who are lactase deficient.
[126]
Ornithine-aspartate provides substrate for urea formation and for synthesis of glutamine, thereby
enhancing the two major routes of ammonia fixation in the body. It has been shown to decrease the
postprandial rise in ammonia concentration in cirrhotics [117] and may prove to play a role in the treatment
of PSE. Zinc is a cofactor of urea cycle enzymes and also plays a role in CNS neurotransmission. [4] [54]
[60] [94] [138] Zinc deficiency has been proposed as a contributory factor in the pathogenesis of hepatic
encephalopathy. [129] The clinical trials to date have shown mixed results. [18] [76] [103] [104]
Benzoate and Phenylacetate
Benzoate and phenylacetate have been used to enhance nitrogen excretion in children with urea cycle
enzyme deficiencies, and additionally have been looked at in the treatment of adults with PSE. Benzoate
is conjugated with glycine to form hippuric acid and phenylacetate with glutamine to form phenylacetyl
glutamine; both products can be renally excreted. For each mole of benzoate, one mole of waste nitrogen

is excreted, and for each mole of phenylacetate, two moles of waste nitrogen is eliminated. In one study,
benzoate was shown to be as effective as lactulose in the treatment of acute encephalopathy. [118]
Additionally, another study showed benefit in treating patients with chronic stable portal systemic
encephalopathy; however, only eight patients were studied. [80]

With the observation that plasma concentrations of aromatic amino acids (tyrosine, tryptophan,
phenylalanine) are increased, and those of branched-chain amino acids (leucine, isoleucine, valine) are
decreased in patients with chronic liver disease, [105] it was postulated that this amino acid imbalance was
important in the development of PSE. Aromatic amino acids and branched-chain amino acids are
transported into the brain by the same amino acid carrier, and with increased peripheral aromatic amino
acids available for transport, more aromatic amino acids enter the brain. These aromatic amino acids then
result in the synthesis of false neurotransmitters (phenylethanolamine, octopamine) and serotonin. By
giving branched-chain amino acids either by infusion or oral administration, less aromatic amino acids
enter the brain. In experimental animals, the infusion of tryptophan and phenylalanine into the carotid
175
results in hepatic encephalopathy, but this can be prevented if branched-chain amino acids are given
concurrently. A number of clinical trials have evaluated this treatment, both for acute and chronic
encephalopathy. Critical analysis of these studies [37] [41] [86] [92] conclude no significant therapeutic
benefit, as to encephalopathy. However, some patients may tolerate these nutrients better than normal
protein diet.
Flumazenil
Flumazenil is a benzodiazepine ligand that stabilizes the benzodiazepine receptors on the
GABA-benzodiazepine receptor complex and competitively antagonizes the binding of other
benzodiazepine ligands. With the GABA hypothesis in mind, flumazenil has been looked at as an agent
by which to treat PSE. A recent randomized, double-blind, placebo-controlled crossover trial looked at
the response of a bolus of flumazenil in cirrhotics in stage-four hepatic encephalopathy receiving
lactulose. [96] Clinical improvement occurred in 6 of 13 treatment periods in which flumazenil was given
compared to no improvement in all treatment periods (n = 15) in which placebo was given.
Clinical improvement after flumazenil occurred within 5 minutes of administration and was partial in
response with the best improvement from grade 4 to grade 2 hepatic encephalopathy. The authors
concluded that the majority of patients with severe hepatic encephalopathy do not benefit significantly
from flumazenil. Other studies have yielded similar results. [23] [128] A source of controversy in assessing
the true efficacy of flumazenil has been whether some of the patients studied had received exogenous
benzodiazepines. If so, then flumazenil could have reversed the effect of exogenous benzodiazepine as
opposed to improving symptoms of PSE.
Lastly, a single case report describes a patient with PSE refractory to standard therapy (40 g protein diet,
oral neomycin and lactulose, branched-chain amino acids) after extensive liver resection and surgical
portacaval shunt who improved dramatically while taking flumazenil 25 mg orally twice a day with

resolution of PSE despite unrestricted dietary protein. [38] Thus, flumazenil may prove to play a role in
the treatment of chronic PSE, although it should be noted that at this time an oral preparation is not
available in the United States.
Bromocryptine
The false neurotransmitter hypothesis postulates displacement of true neurotransmitters with false
neurotransmitters (such as phenolethanolamine and octopamine) resulting in dysfunctional dopaminergic
neurotransmission. Consequently the dopaminergic drugs l-dopa and bromocryptine have been tried as
treatment for hepatic encephalopathy. l-dopa was found to be ineffective. [73] [83] Results with
bromocryptine have
176
been mixed with an initial favorable report, [87] but in a controlled study no benefit was found. [125] In
another study improvement in extrapyramidal signs in patients with PSE after bromocryptine was
reported. [89]
TREATMENT OF POST-TRANSJUGULAR INTRAHEPATIC PORTAL

SYSTEMIC SHUNTS ENCEPHALOPATHY
With the recent introduction and proliferation of transjugular intrahepatic portal systemic shunts (TIPS),
the new clinical entity of post-TIPS encephalopathy emerged. This relatively new entity deserves special
consideration under therapy. Via a transjugular approach, interventional radiologists create a tract
through liver parenchyma connecting hepatic and portal venous circulations, and maintain patency of the
tract by deploying an expandable metal stent. This procedure is analogous to side-to-side surgical
portacaval shunts, but does not carry the risks of surgery or general anesthesia. Since its introduction in
1989, TIPS has gained general acceptance as treatment for acute variceal bleeding that cannot be
successfully controlled with medical or endoscopic treatment, or recurrent variceal bleeding in patients
who are refractory/intolerant to pharmacologic or endoscopic therapy. Additional indications that are as
yet unproved, but promising, include treatment for refractory ascites and hepatic vein thrombosis
(Budd-Chiari syndrome). [58]
Risk factors for the development of PSE after TIPS include abnormal part B trail test, [110] cause of liver
disease other than alcohol, [113] female gender, [113] hypoalbuminemia, [113] and older age. [108] [110] [114]
Factors that have not been found to predict the development of PSE after TIPS include shunt diameter
[108] [114] and direction of portal blood flow. [114] Some have found a history of PSE prior to TIPS as a
significant risk factor, [110] whereas others have not. [114]

The pooled results of several recent studies suggest a 24% incidence
TABLE 3 -- POST-TIPS ENCEPHALOPATHY
Patients With PSE Refractory to
Successful New or Worse Medical
Reference TIPS Encephalopathy Treatment

[66] LaBerge et al 96 17/96 (18%) 3/96

(1993)
[50] Helton et al 55 10/55 (18%) NR
(1993)
[77] Martin et al 45 15/40 * (38%) 10/25
(1993)
[108] Rossle et al 93 23/93 (25%) 7/93
(1993)
[114] Somberg et al 77 18/77 (23%) 4/77
(1995)
[110] Sanyal et al 30 9/30 (30%) 1/30
(1994)
Totals 92/391 (24%) 25/321 (7.8%)
NR = Not reported.
*Evaluated 3 months after TIPS.
Prophylactic lactulose given to patients after TIPS.
177
of new or worsening encephalopathy and a 7.8% incidence of severe/refractory PSE after TIPS (Table 3)
. These values are limited by the lack of a rigorous definition used by all the studies, differences in
duration and methods of follow-up, and differences in patient populations. For example, in one study,
prophylactic lactulose was given to patients after TIPS and PSE was reported only in patients refractory
to lactulose. [108] The incidence of PSE is highest during the first 3 months after TIPS, with it being less
of a problem subsequently; this is believed to be due to narrowing of the stent diameter from neointimal
hyperplasia. [108]
Post-TIPS encephalopathy responds to medical treatment, including lactulose, protein restriction,
antibiotics, and branched-chain amino acids, in roughly 75% to 90% of cases. Options for those who fail
medical treatment include liver transplantation, occlusion of the stent, or reduction of the stent diameter.
[48]
CONCULUSION
In summary, at present the major known toxins for PSE appear to be ammonia and, perhaps, the
endogenous substances which act on the GABA-benzodiazepine recepter complex. Other factors likely
summate with these. Neurotransmission imbalance appears to be the key to the mechanism of

encephalopathy and the glutamate and GABAerig-receptor systems appear to be especially affected. The
role of impaired energy metabolism is not certain. At the cellular level, the astrocyte/neuronal interaction
appears to be critical. This is a multifactorial problem, which is largely metabolic in nature
(biochemical/neurophysiologic) and is largely, if not fully, reversible. Better biochemical
characterization of the central nervous system mechanism(s) of PSE may be relevant to other central
nervous system problems (e.g., the role of the MDA receptor complex in alcohol effects on the brain).
Although the mainstay of treatment for the majority of patients with PSE continues to be lactulose
therapy, additional treatments continue to be examined. Some treatments have had mixed results or have
been incompletely studied. These include ornithine-aspartate, carnitine, zinc, benzoate, phenylacetate,
branched chain amino acids, flumazenil, and bromocryptine. Other treatments are generally effective but
for various reasons are less favorable than lactulose in most situations; these include reducing dietary
protein, switching to vegetable dietary protein, and antibiotics. Lactitol is roughly equivalent to lactulose
but unavailable in the United States. In the special situation of the lactase deficient patient, lactose is
effective treatment. And in the special situation of post-TIPS encephalopathy, TIPS shunt revision can be
helpful in those refractory to medical treatment. The surgical options of colonic diversion or colectomy
are associated with high morbidity and mortality and are rarely used. Finally, in the failing liver in which
PSE is but one manifestation of end-stage disease, liver transplantation offers the hope of effective
treatment and added life.
178
References
1. Al Mardini H, Harrison EJ, Ince PG, et al: Brain indoles in human hepatic encephalopathy. Hepatology 17:1033, 1993
2. Almdol T, Schroeder T, Ranek L: Cerebral blood flow and liver function in patients with encephalopathy due to acute
and chronic liver diseases. Scand J Gastroenterol 24:299, 1989
3. AmodioP, Quero JC, Del Piccolo F, et al: Diagnostic tools for the detection of subclinical hepatic encephalopathy:
Comparison of standard and computerized psychometric tests with Spectral-EEG. Metab Brain Dis 11:315, 1996
4. Assaf SY, Chung SH: Release of endogenous zinc from brain tissue during activity. Nature 308:734, 1984
5. Audet RM, Butterworth RF: Astrocyte function and cerebral energy metabolism are more severely compromised in old
adult rats compared to young adult rats following portacaval anastomosis. Hepatology 24:778A, 1996
6. Bakti G, Fisch HV, Karlaganis G, et al: Mechanism of the excessive sedative response of cirrhotics to benzodiazepines:
Model experiments with triazolam. Hepatology 7:629, 1987
7. Barbara L, Barbiroli B, Gaiani S, et al: Abnormal brain energy metabolism assessed by 31P-MRS in liver cirrhosis
[abstract]. Hepatogastroenterol 40:6(Abstract), 1993
8. Basile AS, Gammal SH, Mullen KD, et al: Differential responsiveness of cerebellar Purkinge neurons in GABA and
benzodiazepine receptor ligands in an animal model of hepatic encephalopathy. J Neurosci 8:2414, 1988
9. Basile AS, Hughes RD, Harrison IM, et al: Elevated brain concentrations of 1,4-benzodiazepines in fulminant hepatic
failure. N Engl J Med 325:475, 1991

10. Basile
AS, Paudl L, Jaouni T, et al: Brain concentrations of benzodiazepines are elevated in an animal model of hepatic
encephalopathy. Proc Natl Acad Sci USA 87:5263, 1990
11. Bengtsson
F, Nobin A, Falck B, et al: Portocaval shunt in the rat: Selective alterations in behavior and brain serotonin.
Pharmacol Biochem Behav 24:1611, 1986
12. Bergquist
PBF, Heyes MP, Apelqvist G, et al: Brain extracellular quinolinic acid in chronic experimental hepatic
encephalopathy as assessed by in vivo microdialysis: Acute effects of L-tryptophan. Neuropsychopharmacology 15:382,
1996
13. BernthalP, Hays A, Tarter RE, et al: Cerebral CT scan abnormalities in cholestatic and hepatocellular disease and their
relationship to neuropsychologic test performance. Hepatology 7:107, 1987
14. BlomHJ, Ferenci P, Grimm G, et al: The role of methanethiol in the pathogenesis of hepatic encephalopathy.
15. Borg J, Warter JM, Schlienger JL, et al: Neurotransmitter modifications in human cerebrospinal fluid and serum during
hepatic encephalopathy. J Neurolog Sci 57:343, 1982
16. Branchi-PorroG, Maiolo AT, Della-Porto P: Cerebral blood flow and metabolism before and after portacaval shunt
operation. Gut 10:894, 1969
17. Breen KJ, Bryant RE, Levinson JD, et al: Neomycin absorption in man: Studies of oral and enema administration and
effect of intestinal ulceration. Ann Intern Med 76:211, 1972
18. BresciG, Parisi G, Banti S: Management of hepatic encephalopathy with oral zinc supplementation: A long-term
treatment. Eur J Med 2:414, 1993
19. Brouillet
EP, Shinobu L, McGarbey U, et al: Manganese injection into the rat striatum produces excitotoxic lesions by
impairing energy metabolism. Exp Neurol 120:89, 1993
20. BucciL, Palmieri GC: Double-blind, double-dummy comparison between treatment with rifaximin and lactulose in
patients with medium to severe degree hepatic encephalopathy. Curr Med Res Opin 13:1109, 1993
21. Butterworth RF: Pathogenesis and treatment of portal-systemic encephalopathy: An update. Dig Dis Sci 37:321, 1992
22. Butterworth RF: Portal-systemic encephalopathy: A disorder of neuron-astrocytic metabolic trafficking. Dev Neurosci
15:313, 1993
23. CadranelJF, Younsi M, Pidoux B, et al: Flumazenil therapy for hepatic encephalopathy in cirrhotic patients: A
double-blind pragmatic randomized, placebo study. Eur J Gastroenterol Hepatol 7:325-9, 1995
179
24. Cangrano C, Cardelli-Cangrano P, James HL, et al: Brain microvessels take up large neutral amino acids in exchange
for glutamine: Cooperative roles of Na+ dependent and Na+ independent systems. J Biol Chem 258:8949, 1983
25. Chamuleau RAFM, Bosman DK, Bovee WMMJ, et al: Chronic hepatic encephalopathy studied by NMR spectroscopy

and MRI in man [abstract]. J Hepatol 13(suppl 2):S19, 1991
26. ConnHO: Hepatic encephalopathy. In Schiff L, Schiff ER (eds): Diseases of the Liver, ed 7. Philadelphia, JB
Lippincott Co, 1993, pp 1036
27. Conn HO, Lieberthal MM (eds): The Hepatic Coma Syndromes and Lactulose. Baltimore, Williams and Wilkins, 1978,
pp 340
28. Cordoba J, Gottstein J, Blei AT: Glutamine, myo-inositol and organic brain osmolytes after portacaval anastomosis in
the rat: Implications of ammonia-induced brain edema. Hepatology 24:919, 1996
29. Cordoba J, Steindl P, Cabrera J, et al: Melatonin and sleep disturbances in cirrhosis. Hepatology 24:A1300, 1996
30. Cuilleret
G, Pomier-Layrargues G, Pons F, et al: Changes in brain catecholamine levels in human cirrhotic hepatic
encephalopathy. Gut 21:565, 1980
31. Dasani
BM, Sigal SH, Lieber CS: The role of Helicobacter pylori in the pathogenesis and treatment of chronic hepatic
encephalopathy [abstract]. Hepatology 22:161, 1995
32. Dela Pena A, Herrero JI, Betes MT, et al: Hyperintensity of basal ganglia on T1 -weighted MRI in liver cirrhosis is
associated to hyperbilirubinemia and portal systemic shunting. Hepatology 24:1308A, 1996
33. deWadeJ-P, Audet RM, Leong DK, et al: Portacaval anastomosis induces region-selective alterations of the
endogenous opiod system in the rat brain. Hepatology 24:895, 1996
34. Dillon D, Schenker S: Cerebrospinal fluid protein concentration in hepatic coma [letter]. JAMA 221:507, 1972
35. Di
Piazza S, Filippazzo MG, Valenza LM, et al: Rifaximine versus neomycin in the treatment of portosystemic
encephalopathy. Ital J Gastroenterol 23:403, 1991
36. DuchiniA, Govindaragon S, Santucci M, et al: Effects of tumor necrosis factor-2 and interleukin-6 on fluid-phase
permeability and ammonia diffusion in CNS-derived endothelial cells. J Invest Med 44:474, 1996
37. ErikssonLS, Conn HO: Branched chain amino acids in the management of hepatic encephalopathy: An analysis of
variants. Hepatology 10:228-246, 1989
38. Ferenci
P, Grimm G, Meryn S: Successful long-term treatment of portal-systemic encephalopathy by the
benzodiazepine antagonist flumazenil. Gastroenterology 96:240-3, 1989
39. FerenciP: Hepatic encephalopathy. In McIntyre N, Benhamou J-P, Bircher J, Rizzetto M, et al (eds): Oxford Textbook
for Clinical Hepatology. Oxford, Oxford University Press, 1991, pp 473
40. Ferenci
P: Critical evaluation of the role of branched chain amino acids in liver disease. In Thomas JC, Jones EA (eds):
Recent Advances in Hepatology. Edinburgh, Churchill Livingstone, 1986, p 137
41. Ferenci
P: Critical evaluation of the role of branched chain amino acids in liver disease. In Thomas JC, Jones EA (eds):
Recent Advances in Hepatology. Edinburgh, Churchill Livingstone, 1986, pp 137-54
42. Ferenci
P, Puspok A, Steindl P: Current concepts in the pathophysiology of hepatic encephalopathy. Eur J Clin Invest
22:573, 1992
43. Geissler A, Lock G, Frund R, et al: Cerebral abnormalities in patients with cirrhosis detected by proton magnetic

resonance spectroscopy and magnetic resonance imaging. Hepatology 25:48, 1997
44. Gillis
JC, Brogden RN: Rifaximin: A review of its antibacterial activity, pharmacokinetic properties and therapeutic
potential in conditions mediated by gastrointestinal bacteria. Drugs 3:467, 1995
45. Gitlin
N: Hepatic encephalopathy. In Zakim D, Boyer TD (eds): Hepatology: A Textbook of Liver Diseases.
Philadelphia, WB Saunders Co, 1996, pp 605
180
46. GrauE, Marcaida G, Montoliu C, et al: Effects of hyperammonemia on brain protein kinase C substrates. Metab Brain
Dis 11:205, 1996
47. Greenberger
NJ, Carley J, Schenker S, et al: Effect of vegetable and animal protein diets in chronic hepatic
encephalopathy. Am J Dig Dis 22:845, 1977
48. HauensteinKH, Haag K, Ochs A, et al: The reducing stent: Treatment for transjugular intrahepatic portosystemic
shunt-induced refractory hepatic encephalopathy and liver failure. Radiology 194:175, 1995
49. Haussinger D, Laubenberger J, Vom Dahl S, et al: Proton magnetic resonance spectroscopy studies on human brain
myo-inositol in hypo-osmolarity and hepatic encephalopathy. Gastroenterology 107:1475, 1994
50. HeltonWS, Belshaw A, Althaus S, et al: Critical appraisal of the angiographic portacaval shunt (TIPS). Am J Surg
165:566, 1993
51. Hermenegildo
C, Marcaida G, Montoliu C, et al: NMDA receptor antagonists prevent acute ammonia toxicity in mice.
Neurochem Res 21:1237, 1996
52. Herrmann S, Krieger D, Jansen O, et al: Effect of liver transplantation on hepatic encephalopathy and hyperintense
globus pallidus on T1 -weighted MRI in liver failure. Neurology 45:A298, 1995
53. HindfeltB, Plum F, Duffy TE: Effect of acute ammonia intoxication on cerebral metabolism in rats with portacaval
shunts. J Clin Invest 59:386, 1977
54. HowellGA, Welch MG, Frederickson CJ: Stimulation-induced uptake and release of zinc in hippocampal slices. Nature
308:736, 1984
55. Jalan
R, Seery JP, Taylor-Robinson SD: Review article: Pathogenesis and treatment of chronic hepatic encephalopathy.
Aliment Pharmacol Ther 10:681, 1996
56. Kato M, Hughes RD, Keays RT, et al: Electron microscopic study of brain capillaries in cerebral edema from fulminant
hepatic failure. Hepatology 15:1060, 1992
57. Kato A, Kaneta Y, Ohara H, et al: Regional differences of cerebral glucose metabolism in patients with liver cirrhosis
as related to the presence of subclinical hepatic encephalopathy using position emission tomography. Hepatology
24:1307A, 1996
58. Kerlan
RK Jr, LaBerge JM, Gordon RL, et al: Transjugular intrahepatic portosystemic shunts: Current status. AJR Am J
Roentgenol 164:1050, 1995

59. Knecht K, Michalak A, Rose C, et al: Decreased glutamate transporter (GLT-1) gene expression in brain in acute liver
failure. Hepatology 24:486A, 1996
60. Koh J, Choi DW: Zinc alters excitatory amino acid neurotoxicity on cortical neurons. J Neurosci 8:2164, 1988
61. Kosenko E, Kaminsky YG, Felipo V, et al: Chronic hyperammonemia prevents changes in brain energy and ammonia
metabolites induced by acute ammonium intoxication. Biochim Biophys Acta 1180:321, 1993
62. Kosenko E, Kaminsky Y, Grau E, et al: Brain ATP depletion induced by acute ammonia intoxication in rats is mediated
by activation of the NMDA receptor and Na+, K+-ATPase. J Neurochem 63:2172, 1994
63. Kreis
R, Ross BD, Farrow NA, et al: Metabolic disorders of the brain in chronic hepatic encephalopathy detected with
H-1 MR spectroscopy. Radiology 182:19, 1992
64. Krieger D, Krieger S, Jansen O, et al: Manganese and chronic hepatic encephalopathy. Lancet 346:270, 1995
65. KriegerS, Jauss M, Jansen O, et al: Neuropsychiatric profile and hyperintense globus pallidus on T1-weighted magnetic
resonance images in liver cirrhosis. Gastroenterology 111:147, 1996
66. LaBergeJM, Ring EJ, Gordon RL, et al: Creation of transjugular intrahepatic portosystemic shunts with the Wallstent
endoprosthesis: Results in 100 patients. Radiology 187:413, 1993
67. LavoieJ, Layrargues GP, Butterworth RF: Increased densities of peripheral-type benzodiazepine receptors in brain
autopsy samples from cirrhotic patients with hepatic encephalopathy. Hepatology 11:874, 1990
68. Lockwood AH, McDonald JM, Reiman RE, et al: The dynamics of ammonia metabolism in man's effects of liver
disease and hyperammonemia. J Clin Invest 63:449, 1979
69. Lockwood AH, Ginsberg MD, Rhoades HM, et al: Cerebral glucose metabolism after portacaval shunting in the rat. J
70. Lockwood AH, Yap EWH, Wong W-H: Cerebral ammonia metabolism in patients
181
with severe liver disease and minimal hepatic encephalopathy. J Cerebr Blood Flow Metab 11:337, 1992
71. Lockwood
AH, McDonald JM, Rieman RE: The dynamics of ammonia metabolism in men: Effects of liver disease and
hyperammonemia. J Clin Invest 63:449, 1979
72. LombardiG, Mannaioni G, Leonardi P, et al: Ammonium acetate inhibits ionotropic receptors and differentially affects
metabotropic receptors for glutamate. J Neural Transm 97:187, 1994
73. Lunzer M: Treatment of chronic hepatic encephalopathy with levodopa. Gut 15:555, 1974
74. Marcaida G, Felipo V, Hermenegildo C, et al: Acute ammonia toxicity is mediated by the NMDA type of glutamate
receptors. FEBS 296:67, 1992
75. Marcaida G, Minana M-D, Burgal M, et al: Ammonia prevents activation of NMDA receptors by glutamate in rat
cerebellar neuronal cultures. Eur J Neurosci 7:2389, 1995

76. MarchesiniG, Fabbri A, Bianchi G, et al: Zinc supplementation and amino acid-nitrogen metabolism inpatients with
advanced cirrhosis. Hepatology 23:1084, 1996
77. MartinM, Zajko AB, Orons PD, et al: Transjugular intrahepatic portosystemic shunt in the management of variceal
bleeding: Indications and clinical results. Surgery 114:719, 1993
78. Massa P, Vallerino E, Dodero M: Treatment of hepatic encephalopathy with refaximin: Double blind, double dummy
study versus lactulose. Eur J Clin Res 4:7, 1993
79. McClainGJ, Kromhout JP, Elson MK, et al: Hyperprolactinemia in portal systemic encephalopathy. Dig Dis Sci
26:353, 1981
80. Mendenhall CL, Rouster S, Marshall L, et al: A new therapy for portal systemic encephalopathy. Am J Gastroenterol
81:540, 1986
81. Morgan MH, Read AE, Speller DC: Treatment of hepatic encephalopathy with metronidazole. Gut 23:1, 1982
82. MichalakA, Rose C, Butterworth RF: Evidence for glutamatergic synaptic dysfunction in acute liver failure: Loss of
glutamate (non-NMDA) receptors. Hepatology 24:1299A, 1996
83. MichelH, Solere M, Granier P, et al: Treatment of cirrhotic hepatic encephalopathy with L-dopa: A controlled trial.
84. Minana M-D, Felipo V, Grisolfa S: Protective effect of long term ammonium ingestion against acute ammonium
intoxication. Biochem Biophys Res Commun 153:979, 1988
85. MinanaM-D, Hermenegildo C, Llansola M, et al: Carnitine and choline derivatives containing a trimethylamine group
prevent ammonia toxicity in mice and glutamate toxicity in primary cultures of neurons. J Pharmacol Exp Ther 279:194,
1996
86. MorganMY: Branched chain amino acids in the management of chronic liver disease: Facts and fantasies. J Hepatol
11:133-141, 1990
87. Morgan MY, Jakobovits A, Elithorn A, et al: Successful use of bromocriptine in the treatment of a patient with chronic
portal-systemic encephalopathy. N Engl J Med 296:793-4, 1977
88. Morgan MY, Hawley KE: Lactilol vs lactulose in the treatment of acute hepatic encephalopathy in cirrhotic patients: A
double-blind, randomized trial. Hepatology 7:1278, 1987
89. MorganMY, Jakobovits AW, James IM, et al: Successful use of bromocryptine in the treatment of chronic hepatic
encephalopathy. Gastroenterology 78:663-670, 1980
90. MullenKD, McCullough AJ: Problems with animal models of chronic liver disease: Suggestions for improvement in
standardization. Hepatology 9:500, 1989
91. Muto Y: Clinical study on the relationship of short-chain fatty acids and hepatic encephalopathy. Jpn J Gastroenterol
63:19, 1966
92. Naylor
CD, O'Rourke K, Detsky AS, et al: Parenteral nutrition with branched chain amino acids in hepatic
encephalopathy: A meta-analysis. Gastroenterology 97:1033-1042, 1989

93. OlasmaaM, Rothstein JD, Guidotti A, et al: Endogenous benzodiazepine receptor ligands in human and animal hepatic
encephalopathy. J Neurochem 55:2015, 1990
94. Peters
S, Koh J, Choi DW: Zinc selectively blocks the action of N-methyl-D-aspartate on cortical neurons. Science
236:589, 1987
95. Peterson C, Giguere J-F, Cotman CW: Selective loss of N-methyl-D-aspartate-sensitive
182
L-[3 H]glutamate binding sites in rat brain following portacaval anastomosis. J Neurochem 55:386, 1990
96. Pomier-Layrargues G, Gignier JF, Lavoie J, et al: Flumazenil in cirrhotic patients in hepatic coma: A randomized
double-blind placebo-controlled crossover trial. Hepatology 19:32, 1994
97. Pomier-Layrargues G, Spahr L, Butterworth RF: Increased manganese concentrations in pallidum of cirrhotic patients.
Lancet 345:735, 1995
98. PowellEE, Pender MP, Chalk JB, et al: Improvement in chronic hepatocerebral degeneration following liver
transplantation. Gastroenterology 98:1079, 1990
99. Ptovin
M, Finlayson MH, Hinchey EJ, et al: Cerebral abnormalities in hepatectomized rats with acute hepatic coma.
Lab Invest 55:560, 1984
100. Raabe W: Ammonium ions abolish excitatory synaptic transmission between cerebellar neurons in primary dissociated
tissue culture. J Neurophysiol 68:93, 1992
101. Raabe W: Spinal seizures in ammonia intoxication. In Felipo V, Grisolia S (eds): Hepatic Encephalopathy,
Hyperammonemia and Ammonia Toxicity. New York, Plenum Press, 1994, p 55
102. Raabe W, Lin S: Ammonia, postsynaptic inhibition and CNS-energy state. Brain Res 303:67, 1984
103. Reding
P, Duchateau J, Bataille C: Oral zinc supplementation improves hepatic encephalopathy: Results of a
randomised controlled trial. Lancet 2:493, 1984
104. Riggio
O, Ariosto F, Merli M, et al: Short-term oral zinc supplementation does not improve chronic hepatic
encephalopathy: Results of a double-blind crossover trial. Dig Dis Sci 36:1204, 1991
105. Rosen HM, Yoshimura N, Hodgman JM, et al: Plasma amino acid patterns in hepatic encephalopathy of differing
etiology. Gastroenterology 72:483, 1977
106. Ross
BD, Jacobson S, Villamil F, et al: Subclinical hepatic encephalopathy: Proton MR spectroscopic abnormalities.
Radiology 193:457, 1994
107. RossBD, Morgan MY, Cox IJ, et al: Continuing defect in cerebral energy metabolism in patients with persistent
hepatic encephalopathy (PHE): Monitoring with 3 P MRS. Clin Sci 72(suppl 16):26P, 1987
108. RossleM, Haag K, Ochs A, et al: The transjugular intra-hepatic portosystemic stent-shunt procedure for variceal
bleeding. N Engl J Med 330:165, 1993
109. Rudman D, Aksun S, Galambos JT, et al: Observations on nitrogen metabolism of patients with portal cirrhosis. Am J
Clin Nutr 23:1203, 1970

110. Sanyal
AJ, Freedman AM, Shiffman ML, et al: Portosystemic encephalopathy after transjugular intrahepatic
portosystemic shunt: Results of a prospective controlled study. Hepatology 20:46-55, 1994
111. Schenker S, Brady CE III: Pathogenesis of hepatic encephalopathy. In Conn HO, Bircher J (eds): Hepatic
Encephalopathy: Syndromes and Therapies. Bloomington, IL, Medi-Ed Press, 1994, p 43
112. Schenker S, McCandless DW, Brosby E, et al: Studies on the intracerebral toxicity of ammonia. J Clin Invest 46:838,
1967
113. Soeters
PB, Van Leeuwen M: Laktulose, neomyzin und ammoniakstoffwechsel. Infusion Therapie und Klinische
Ernahrung 13:186, 1986
114. SonbergKA, Riegler JL, Doherty M, et al: Hepatic encephalopathy following transjugular intrahepatic portosystemic
shunts (TIPS): Incidence and risk factors. Hepatology 90:549, 1995
115. SpahrL, Butterworth RF, Fontaine S, et al: Increased blood manganes in cirrhotic patients: Relationship to pallidal
magnetic resonance signal hyperintensity and neurological symptoms. Hepatology 24:1116, 1990
116. StabenauJR, Warren KS, Rall DP, et al: The role of pH gradient in the distribution of ammonia between blood and
cerebrospinal fluid, brain and muscle. J Clin Invest 38:373, 1959
117. StaedtU, Leweling H, Gladisch R, et al: Effects of ornithine aspartate on plasma ammonia and plasma amino acids in
patients with cirrhosis: A double-blind randomized study using a four-fold crossover design. J Hepatol 19:424, 1993
118. Sushma S, Tandon RK, Jain S, et al: Sodium benzoate in the treatment of acute hepatic encephalopathy: A
double-blind randomized trial. Hepatology 16:138, 1992
183
119. Swain MS, Blei AT, Butterworth RF, et al: Intracellular pH rises and astrocytes swell after portacaval anastomosis in
rats. Am J Physiol 261:R1491, 1991
120. SwartGR, Van den Berg JW, van Vuure JK, et al: Minimum protein requirements in liver cirrhosis determined by
nitrogen balance measurements at three levels of protein intake. Clin Nutr 8:329, 1989
121. Takahashi
H, Koehler RC, Brusilow SW, et al: Inhibition of brain glutamine accumulation prevents cerebral edema in
hyperammonemic rats. Am J Physiol 261:H825, 1991
122. TaraoK, Ikeda T, Hayashi K, et al: Successful use of vancomycin hydrochloride in the treatment of lactulose resistant
chronic hepatic encephalopathy. Gut 31:702, 1990
123. Taylor-Robinson SD, Sargentoni J, Mallalieu RJ, et al: Cerebral phosphorus-31 magnetic resonance spectroscopy in
patients with chronic hepatic encephalopathy. Hepatology 20:1173, 1994
124. Thuluvath PJ, Edurn D, Yoe NC, et al: Increased signals seen in globus pallidus in T1 -weighted magnetic resonance
imaging in cirrhotics are not suggestive of chronic hepatic encephalopathy. Hepatology 21:440, 1995
125. Uribe M, Farca A, Marquez MA, et al: Treatment of chronic portal systemic encephalopathy with bromocriptine: A

double-blind controlled trial. Gastroenterology 76:1347-1351, 1979
126. UribeM, Marquez MA, Garcia-Ramos G, et al: Treatment of chronic portal-systemic encephalopathy with lactose in
lactase-deficient patients. Dig Dis Sci 25:924, 1980
127. Uribe
M, Marquez MA, Ramos GG, et al: Treatment of chronic portal-systemic encephalopathy with vegetable and
animal protein diets: A controlled crossover study. Dig Dis Sci 27:1109, 1982
128. Vander Rijt CC, Schalm SW, Meulstee J, et al: Flumazenil therapy for hepatic encephalopathy: A double-blind
crossover study. Gastroenterol Clin Biol 19:572-580, 1995
129. VanDer Rijt CCD, Schalm SW, Schat H, et al: Overt hepatic encephalopathy precipitated by zinc deficiency.
130. Warren
KS, Schenker S: Effect of an inhibitor of glutamine synthesis (methiomine sulfaximine) on ammonia toxicity
and metabolism. J Lab Clin Med 64:442, 1964
131. Warren
KS, Schenker S: Effect of temperature variation on toxicity and metabolism of ammonia in mice. J Lab Clin
Med 60:291, 1962
132. WeberFL Jr: The effect of lactulose on urea metabolism and nitrogen excretion in cirrhotic patients. Gastroenterology
77:518-523, 1979
133. WeberFL Jr, Fresard KM, Lally BR: Effects of lactulose and neomycin on urea metabolism in cirrhotic subjects.
134. Weber FL Jr, Banwell JG, Fresard KM, et al: Nitrogen in fecal bacteria, fiber and soluble fractions of patients with
cirrhosis: Effects of lactulose and lactulose plus neomycin. J Lab Clin Med 110:259, 1987
135. WeberFL, Minco D, Fresard KM, et al: Effects of vegetable diets on nitrogen metabolism in cirrhotic subjects.
136. Weber FL Jr, Reiser BJ: Relationship of plasma amino acids to nitrogen balance and portal-systemic encephalopathy in
alcoholic liver disease. Dig Dis Sci 27:103, 1982
137. WeberFL Jr, Veach GL: The importance of the small intestine in gut ammonium production in the fasting dog.
138. Wensink J, Molenaar AJ, Woroniecka UD, et al: Zinc uptake into synaptosomes. J Neuroscience 50:782-789, 1988
139. YaoH, Sadoshima S, Fujii K, et al: Cerebrospinal fluid lactate in patients with hepatic encephalopathy. Eur Neurol
27:182, 1987
140. Yurdoydin C, Hortnagl H, Pifl CL, et al: Modulation of hepatic encephalopathy in rats due to thioacetamide induced
liver failure by serotonin agonists and antagonists. Gastroenterology 96:A675, 1989
141. Yurdoydin C, Hortnagl H, Steind P, et al: Increased serotonergic and noradrenergic activity in hepatic encephalopathy
in rats with thioacetamide induced acute liver failure. Hepatology 12:695, 1990
142. Yurdoydin C, Karavelioghi D, Yasa H, et al: Opioid ligands in human hepatic encephalopathy. Hepatology 24:1309A,
1996

184
143. Zee
PC, Mehta R, Turek F, et al: Portacaval anastomosis disrupts circadian locomotor activity and pineal melatonin
rhythms in rats. Brain Res 560:17, 1991
144. Zieve
L, Doizaki WM, Zieve FJ: Synergism between mercaptans and ammonia or fatty acids in the production of
coma: A possible role for mercaptans in the pathogenesis of hepatic coma. J Lab Clin Med 83:16, 1974
145. Zieve L, Olsen EL: Can hepatic coma be caused by a reduction of brain noradrenaline or dopamine? Gut 18:688, 1977


April 22, 2000
Conde Petra
Neurospectroscopic alterations and globus

pallidus hyperintensity as related magnetic
resonance markers of reversible hepatic
Citation encephalopathy [see comments]
Bibliographic Data
Abstract
Indexing Data
Neurospectroscopic alterations and
Copyright Notice and Disclaimer globus pallidus hyperintensity as related
magnetic resonance markers of
Find More Articles Like This reversible hepatic encephalopathy [see
comments]
Full Text Pujol J - Neurology - 1996 Dec; 47(6): 1526-30

Frontmatter
NLM Citation ID:
Methods. 97120068
Results. Comment:
Neurology 1997 Oct;49(4):1187-8
White matter analysis.
Globus pallidus analysis.

Full Source Title:
Neurology
White matter versus globus
pallidus. Publication Type:
Journal Article
Discussion.
Language:
Acknowledgments English
References Author Affiliation:
About the Publication Magnetic Resonance Center of Pedralbes, Barcelona, Spain.
Authors:
Pujol J; Kulisevsky J; Moreno A; Deus J; Alonso J; Balanzo J;
Marti-Vilalta JL; Capdevila A
Abstract:
In patients with chronic hepatic encephalopathy, proton
magnetic resonance spectroscopy can be used to detect specific
metabolic abnormalities in the brain; MRI shows a hyperintense
globus pallidus on T1-weighted sequences. We investigated the
relationship between these two MR findings in a series of 25
patients with the use of quantitative data and a multiple
regression analysis model. The cerebral increase in glutamine

compounds and the decrease in myoinositol and choline
correlated separately with globus pallidus hyperintensity, and
each was complementary in accounting for this imaging
finding. Such as association suggests that spectroscopic and
imaging alterations are two different expressions of the
reversible events that occur in the brain of patients with hepatic
encephalopathy in that both disappear after liver transplantation.
Globus pallidus hyperintensity seems to be a global indicator of
the cerebral metabolic disorder, and the spectroscopic pattern
denotes the specific metabolic alterations.
Major Subjects:
Globus Pallidus / Metabolism / * Pathology
Hepatic Encephalopathy / Metabolism / * Pathology
Aged
Female
Human
Magnetic Resonance Imaging
Male
Middle Age
Nuclear Magnetic Resonance


Neurology
Copyright 1996 American Academy of Neurology
P1526
Articles
Neurospectroscopic alterations and globus pallidus hyperintensity

as related magnetic resonance markers of reversible hepatic
encephalopathy
J. Pujol MD
J. Kulisevsky MD
A. Moreno PhD
J. Deus PhD
J. Alonso PhD
J. Balanzo MD
J.L. Marti-Vilalta MD
A. Capdevila MD
Magnetic Resonance Center of Pedralbes (Drs. Pujol, Moreno, Deus, Alonso, and Capdevila)
Barcelona, Spain;
Departments of Neurology (Drs. Kulisevsky and Marti-Vilalta)
Gastroenterology (Dr. Balanzo), Sant Pau Hospital Autonomous University of Barcelona,
Barcelona, Spain.
Article abstract--
In patients with chronic hepatic encephalopathy, proton magnetic resonance spectroscopy can be
used to detect specific metabolic abnormalities in the brain; MRI shows a hyperintense globus
pallidus on T1 -weighted sequences. We investigated the relationship between these two MR
findings in a series of 25 patients with the use of quantitative data and a multiple regression
analysis model. The cerebral increase in glutamine compounds and the decrease in myoinositol and
choline correlated separately with globus pallidus hyperintensity, and each was complementary in
accounting for this imaging finding. Such an association suggests that spectroscopic and imaging
alterations are two different expressions of the reversible events that occur in the brain of patients
with hepatic encephalopathy in that both disappear after liver transplantation. Globus pallidus

hyperintensity seems to be a global indicator of the cerebral metabolic disorder, and the
spectroscopic pattern denotes the specific metabolic alterations.
Supported in part by the Institut de Recerca de l'Hospital de la Santa Creu i Sant Pau of Barcelona and
Grant FIS 95/1032.
Received February 28, 1996. Accepted in final form May 2, 1996.

Address correspondence and reprint requests to Dr. Jesus Pujol, Magnetic Resonance Center of Pedralbes, Monestir 3,
08034 Barcelona, Spain.
The prognosis and quality of life in patients with chronic liver failure depend, to a great extent, on the degree of brain
deterioration. Fluctuating hepatic encephalopathy can produce progressive brain damage. [1] Neurospectroscopy [2]
provides tools with which to detect specific metabolic alterations in the brain of patients with chronic hepatic
encephalopathy. [3] [4] [5] [6] [7] [8] [9] The pattern of biochemical alterations consists of the increase in cerebral glutamine
and the decrease in myoinositol and choline metabolites. [5] MRI, on the other hand, shows an abnormal signal increase in
the globus pallidus on T1 -weighted sequences in similar patients, which is mainly associated with the presence of a
portal-systemic shunting of blood. [10] [11] [12] [13] [14] [15] [16] [17] [18]
These magnetic resonance (MR) spectroscopic and MRI findings relate to the severity of the encephalopathy, but both are
already evident in subclinical states, thus suggesting their potential clinical usefulness as early MR markers of cerebral
repercussion. [6] [13] [14] Although both MR findings occur in the clinical context, the relationship between them remains
unclear.
We report a systematic study based on quantitative data that is aimed at establishing the relationship between MR
spectroscopic and MRI alterations present in patients with stable liver cirrhosis.
Methods.
Twenty-five consecutive outpatients with biopsy-proven liver cirrhosis (18 with Child's grade A and seven with Child's
grade B) and 25 normal control subjects with similar age and sex were included in this study. Patients who were selected
met the criteria of stable liver cirrhosis, no overt encephalopathy (fully oriented and without asterixis), no active
alcoholism, and no decompensated medical conditions. There were 11 women and 14 men age, 63.8 8.3 years; (mean
SD range, 43 to 75 years). Eight had alcoholic cirrhosis, and 17 had nonalcoholic cirrhosis. The control subjects were
volunteers who had no history of metabolic or brain disease and had been referred to our center to receive an MRI
examination of the spine for the study of cervical pain. There were 10 women and 15 men (age, 60.4 7.4 years; range, 45
to 75 years). All subjects gave informed consent.
MR studies were performed using a 1.5-T Signa System (General Electric Medical Systems, Milwaukee, WI) and a
quadrature head coil. MR spectroscopy consisted of a "stimulated echo acquisition method," or STEAM, pulse sequence
(TR/TE, 1,600/20 msec; TM, 13.7 msec; 256 acquisitions; 2,048 data points; and 2,500-Hz spectral width)
P1527
Figure 1. The volumes of interest were selected on axial images. White matter spectrum was placed in the region posterior to the
ventricular atrium in each subject (left). The smaller basal ganglia voxel contained mainly the globus pallidus (right).
preceded by three chemical shift water-suppression pulses (50-Hz bandwidth). This pulse sequence was obtained with the
single-voxel proton brain examination (PROBE/SV) system, which automatically optimizes magnetic field homogeneity
over the selected voxel and adjusts the water suppression pulses. Two single voxels placed in the parieto-occipital white

matter (6.5 ml) and in the globus pallidus (5.4 ml) were acquired in each subject. Small volumes were chosen to minimize
partial volume effects. Figure 1 shows the voxel placement.
Data processing was performed on a workstation (SPARCstation 20; Sun Microsystems, Mountain View, CA). Peak areas
were estimated with an analysis of free induction decays as described previously. [19] Thus, the amplitudes of the time
domain signals, which correspond to the area under each resonance in the frequency domain, were measured. The main
steps of this analysis included removal of residual water with use of the Hankel-Lanczos singular value decomposition
algorithm [20] and nonlinear least-squares fitting with use of the variable projection method. [21] Results are expressed as
ratios, with creatine (Cr) used as reference. Measurements were performed at the following resonances: N-acetylaspartate
(NAA) (2.01 ppm), glutamine-plus (Glx) (2.15 to 2.50 ppm), Cr (3.04 ppm), choline (Cho) (3.22 ppm), and myoinositol
(mI) (3.55 ppm). In addition, to estimate local magnetic field homogeneity, the half-height linewidth (Hz) of the water
peak was manually measured on each spectrum.
MRIs were obtained in the sagittal, axial, and oblique-coronal planes. Signal intensity of the globus pallidus was measured
on 35-mm2 regions of interest (figure 2) in a single slice of the oblique-coronal projection obtained from an
inversion-recovery sequence (TR/TE/TI, 1,500/20/650 msec; field of view, 22 cm; matrix size, 256 192 pixels; slice
thickness, 5 mm; and interslice gap, 1.5 mm). The method of measurement, which has been described previously, [13]
involved the measurement of the globus pallidus signal intensity normalized to the putamen signal ([globus pallidus -
putamen]/[globus pallidus + putamen] signal intensity ratio).
To establish the value of the signal intensity ratio in normal subjects, we used data from a control group composed of 37
healthy subjects [13] (15 women and 22 men; mean age, 59.6 22 years). The mean signal measurement from this control
group was 0.04 0.03. With mean 2 SD taken as statistical criteria for normal range, values higher than 0.10 were
considered abnormal, indicating an increased signal in the globus pallidus.
The nonparametric Mann-Whitney U test was used in comparisons between groups due to significant SD differences
Figure 2. Oblique-coronal inversion-recovery image showing the regions of interest selected to measure globus pallidus and putamen
signal intensity.
P1528
Figure 3. White matter spectrum representation from a healthy control (A) and a cirrhotic patient (B). Arrows indicate MR spectroscopy
abnormalities. (Glx was analyzed only in the 2.15- to 2.50-ppm region in this study.) (NAA = N -acetylaspartate).
in the spectroscopic measurements. Pearson's product-moment was used in the analysis of correlations, and paired
Student's t-test was used to compare white matter and pallidal measurements. Results are shown without correction for
multiple comparisons to emphasize the patterns of spectroscopic alterations and correlations that were found.
TABLE -- Neurospectroscopic pattern and correlations with MRI values

Correlation with
MRS ratios MRS ratios Comparison of GP signal
(controls) mean (patients) mean means (patients)
SD SD Z p r p
White matter *
mI/Cr 0.69 0.11 0.34 0.24 -5.0 < 0.001 -0.46 0.010
Cho/Cr 0.83 0.12 0.73 0.21 -2.1 0.031 -0.42 0.019
Glx/Cr 1.23 0.31 1.68 0.62 -2.8 0.004 0.48 0.007
NAA/Cr 1.43 0.19 1.52 0.17 -1.7 0.091 -0.25 0.110
GP
mI/Cr 0.56 0.10 0.27 0.17 -4.4 < 0.001 -0.47 0.023
Cho/Cr 0.76 0.10 0.63 0.15 -3.0 0.003 -0.15 0.270
Glx/Cr 1.22 0.29 1.81 0.57 -3.5 < 0.001 0.54 0.010
NAA/Cr 1.11 0.15 1.06 0.12 -0.8 0.447 -0.20 0.215
MRS = MR spectroscopy; GP = globus pallidus; NAA = N-acetylaspartate.
* n = 25 subjects for both study groups.
n = 17 for the control group and n = 18 for the patient group.
Results.
Mean signal measurement of the globus pallidus in patients was 0.13 0.04 contrast units. Eighteen of the 25 patients
were outside normal limits.
White matter analysis.
The spectroscopic alterations found in white matter are illustrated in figure 3 . Patients showed a large increase in Glx/Cr
and decrease in mI/Cr. The reduction in Cho/Cr was mild but significant (table 1) . The SDs of the abnormal metabolite
ratios were notably high in the patient group (see table ), thus indicating the occurrence of a wide range of severity for the
metabolic disorders observed in our patients with stable liver cirrhosis. Glx/Cr was abnormal (outside the range of mean
2 SD) in nine patients, mI/Cr was abnormal in 18, and Cho/Cr was abnormal in five.
Findings in the correlational analysis were consistent (see table 1 ). Globus pallidus signal intensity was related to the
increase in Glx/Cr and decrease in mI/Cr and Cho/Cr. In a stepwise multiple regression analysis to account for the globus
pallidus signal intensity, results were robust. Three of the four MR spectroscopic measurements correlated significantly
with globus pallidus signal intensity. At the first step, the Glx/Cr ratio entered (R2 = 0.23), followed by the Cho/Cr ratio
(additional R2 change = 0.33) and then the mI/Cr ratio (additional R2 change = 0.07). The final value was R2 = 0.80 (F =
12.7; p = 0.0001). A simple computation, which consisted of the Glx/mI+Cho ratio, expressed this strong global

relationship between spectroscopic and imaging measurements (figure 4) .
Globus pallidus analysis.
In general, spectra from the globus pallidus were of poor quality because of local magnetic susceptibility effects. Mean
SD half-height linewidth of the water peak at this location was 8.5 1.1 Hz, clearly exceeding the linewidth found in
white matter spectra (5.4 0.3 Hz, t = 19.8, p < 0.0001, n = 50 pairs). Peak overlap as a result of this nonoptimal
resolution precluded adequate measurement of peak areas, because of automated fitting failure, in pallidal spectra of seven
patients and eight control subjects. Therefore, only 18 patients and 17 control subjects were included in this analysis.
P1529
Figure 4. Plot of the globus pallidus signal measurement (contrast units) with the Glx/mI+Cho ratio in the patient group (Pearson's
coefficient correlation = 0.70, p = 0.00005).
An increase in Glx/Cr and a decrease in mI/Cr and Cho/Cr ratios were again observed in this location (see table 1 ). Glx/Cr
was outside normal limits in 11 of these 18 patients, mI/Cr was outside normal limits in 13, and Cho/Cr was outside
normal limits in six. Regarding the correlation of spectroscopic measurements with the globus pallidus signal, we found a
similar pattern to that described for white matter, except that the correlation with the Cho/Cr did not reach statistical
significance (table 1) . The multiple regression analysis results were slightly different. Glx/Cr entered in the first step to
account for the globus pallidus signal with R2 = 0.30, and mI/Cr entered in the second and last step with an additional R2
change of 0.18. The final value was R2 = 0.69 (F = 6.7; p = 0.0081).
White matter versus globus pallidus.
With the Glx/mI+Cho ratio taken as a representative measurement of the global metabolic abnormality in patients with
liver disease, we found similar spectroscopic alteration in both locations (Glx/mI+Cho ratio: white matter, 1.81 0.76;
globus pallidus, 2.20 1.08; t = 1.62, p = 0.123, n = 18 pairs).
Discussion.
The neurospectroscopic pattern that characterizes patients with liver cirrhosis consists of an increase in Glx and a
reduction in mI and Cho and represents a robust finding, proposed by Ross et al. [6] to define subclinical hepatic
encephalopathy. All three abnormalities were present to a similar extent in both the white matter and globus pallidus of our
patients who had no manifest encephalopathy and showed a wide range of severity, demonstrating a qualitatively
consistent, quantitatively variable underlying metabolic disorder in chronic liver disease.
The central finding of our report, however, was that each spectroscopic alteration correlated separately with the globus
pallidus signal increase and that, more interestingly, each was complementary in accounting for this imaging finding.
Thus, an analysis based on quantitative data, in which we used a multiple regression analysis model, allowed us to obtain
evidence of a close association between globus pallidus hyperintensity and the combination of the spectroscopic
alterations.
Pallidal hyperintensity correlated with the metabolic derangement occurring in the two locations studied. Therefore, the
results indicate that the abnormal appearance of the globus pallidus is associated more with widespread metabolic
alterations in the brain than with particular metabolic events occurring in basal ganglia. Globus pallidus hyperintensity
could reflect a general metabolic insult, which has a particular imaging expression in the basal ganglia. Poorer results in
the globus pallidus analysis of correlations may be technically accounted for by degradation of spectra because of

magnetic susceptibility effects produced by local physiological accumulation of paramagnetic iron. [22]
The alterations detected by proton spectroscopy represent direct evidence of several metabolic disorders produced by liver
failure in the brain. [5] [6] [7] Globus pallidus hyperintensity, on the other hand, is a single nonspecific MRI finding,
relatively stable through time, [23] that denotes disease severity in portal-systemic encephalopathy. [11] [13] [14] Thus,
spectroscopic and imaging findings have different natures, although both seem to be a consequence of the effect of
nondetoxified blood on the brain of patients with liver insufficiency. [2] [11] [13] The specific association found in this study
suggests that the neurospectroscopic alterations and globus pallidus hyperintensity are two different cerebral expressions
of the underlying metabolic disorder in hepatic encephalopathy.
Although liver failure can lead to definitive brain damage, [1] a basic characteristic of hepatic encephalopathy is its
potential reversibility. Spectroscopic alterations [2] [24] and globus pallidus hyperintensity [14] [25] are reversible when liver
function normalizes, as occurs with liver transplantation. Therefore, both MR features express "reversible" events
occurring in hepatic encephalopathy. The time course for recovery in these findings is different. Transplant patients
undergo complete reversal of the biochemical abnormalities within weeks of successful transplantation, [2] whereas
hyperintensity of the globus pallidus vanishes only after 10 to 20 months, [14] [25] a period in which MR spectroscopy has
long been normal.
The reversible events of hepatic encephalopathy can now be better characterized. Neurologic and neuropsychologic
examinations allow the clinician to establish current expression of the encephalopathy with relative accuracy. On a
different level, MR spectroscopy can be used to detect direct biochemical consequences of the systemic metabolic disorder
in the brain. These phenomena are specific and change within weeks. Finally, MRI can depict a signal change in the
globus pallidus that is an expression of the entire metabolic insult that reversibly alters this region for a longer period of
months.
We conclude that the strong association of globus pallidus hyperintensity with the combination of the altered metabolites,
more than with each of them individually, suggests the existence of complex metabolic
P1530
interactions in hepatic encephalopathy and supports the view of the abnormal globus pallidus as a metabolic indicator.
That these two reversible and related alterations are present in patients without overt encephalopathy suggests their
potential complementary usefulness as early markers of brain metabolic disorder.
Acknowledgments
We thank Gerald Fannon, PhD, for revising the manuscript; Aad van den Boogaart, PhD, for providing the MRUI
software; and Miquel Cabanas, PhD, for his technical assistance.
References
1. Adams RD, Victor M. Principles of neurology, 4th ed. New York: McGraw-Hill, 1989.
2. Ross B, Michaelis T. Clinical applications of magnetic resonance spectroscopy. Magn Reson Q 1994;10:191-247.
3. Kreis R, Farrow N, Ross BD. Diagnosis of hepatic encephalopathy by proton magnetic resonance spectroscopy. Lancet 1990;336:635-636.
4. Kreis R, Farrow N, Ross BD. Localized 1 H NMR spectroscopy in patients with chronic hepatic encephalopathy: analysis of changes in cerebral glutamine,
choline and inositols. NMR Biomed 1991;4:109-116.
5. Kreis R, Ross BD, Farrow NA, Ackerman Z. Metabolic disorders of the brain in chronic hepatic encephalopathy detected with H-1 MR spectroscopy. Radiology
1992;182:19-27.
6. Ross BD, Jacobson S, Villamil F, et al. Subclinical hepatic encephalopathy: proton MR spectroscopic abnormalities. Radiology 1994;193:457-463.
7. Bottomley PA. Proton MR spectroscopy for diagnosing hepatic encephalopathy. Radiology 1992;182:6-7.

8. Haussinger D, Laubenberger J, vom Dahl S, et al. Proton magnetic resonance spectroscopy studies on human brain myo-inositol in hypo-osmolarity and hepatic
encephalopathy. Gastroenterology 1994;107:1475-1480.
9. Taylor-Robinson SD, Sargentoni J, Marcus CD, Morgan MY, Bryant D. Regional variations in cerebral proton spectroscopy in patients with chronic hepatic
encephalopathy. Metab Brain Dis 1994;9:347-359.
10. Kulisevsky J, Ruscalleda J, Grau JM. MR imaging of acquired hepatocerebral degeneration. AJNR 1991;12:527-528.
11. Inoue E, Hori S, Narumi Y, et al. Portal-systemic encephalopathy: presence of basal ganglia lesions with high signal intensity on MR images. Radiology
1991;179:551-555.
12. Brunberg JA, Kanal E, Hirsch W, van Thiel DH. Chronic acquired hepatic failure: MR imaging of the brain at 1.5 T. AJNR 1991;12:909-914.
13. Kulisevsky J, Pujol J, Balanzo J, et al. Pallidal hyperintensity on magnetic resonance imaging in cirrhotic patients: clinical correlations. Hepatology
1992;16:1382-1388.
14. Pujol A, Pujol J, Graus F, et al. Hyperintense globus pallidus on T1-weighted MRI in cirrhotic patients is associated with severity of liver failure. Neurology
1993;43:65-69.
15. Kulisevsky J, Pujol J, Junque C, Deus J, Balanzo J, Capdevila A. MRI pallidal hyperintensity and brain atrophy in cirrhotic patients: two different MRI patterns
of clinical deterioration? Neurology 1993;43:2570-2573.
16. Hauser RA, Zesiewicz TA, Rosemurgy AS, Martinez C, Olanow CW. Manganese intoxification and chronic liver failure. Ann Neurol 1994;36:871-875.
17. Krieger D, Krieger S, Jansen O, Gass P, Theilmann L, Lichtnecker H. Manganese and chronic hepatic encephalopathy. Lancet 1995;346:270-274.
18. Vymazal J, Babis M, Brooks RA, et al. T1 and T2 alterations in the brains of patients with hepatic cirrhosis. AJNR 1996;17:333-336.
19. Saunders DE, Howe FA, van den Boogaart A, McLean MA, Griffiths JR, Brown MM. Continuing ischemic damage after acute middle cerebral artery infarction
in humans demonstrated by short-echo proton spectroscopy. Stroke 1995;26:1007-1013.
20. Pijnappel WWF, van den Boogaart A, de Beer R, van Ormondt D. SVD-based quantification of magnetic resonance signals. J Magn Reson 1992;97:122-134.
21. van der Veen JWC, de Beer R, Luyten PR, van Ormondt D. Accurate quantification of in vivo P-31 NMR signals using the variable projection method and prior
knowledge. Magn Reson Med 1988;6:92-98.
22. Pujol J, Junque C, Vendrell P, et al. Biological significance of iron-related magnetic resonance imaging changes in the brain. Arch Neurol 1992;49:711-717.
23. Kulisevsky J, Pujol J, Deus J, et al. Persistence of MRI hyperintensity of the globus pallidus in cirrhotic patients: a 2-year follow-up study. Neurology
1995;45:995-997.
24. Huda AS, Thomas MA, Ke Y, Bugbee ME, Strouse TB, Guze BH. H-1 MR spectroscopy in the management of hepatic encephalopathy [abstract]. Radiology
1995;197(suppl):374.
25. Pujol A, Graus F, Peri J, Mercader JM, Rimola A. Hyperintensity in the globus pallidus on T1-weighted and inversion-recovery MRI: a possible marker of
advanced liver disease. Neurology 1991;41:1526-1527.


Neurology
Figure 1. The volumes of interest were selected on axial images. White matter spectrum was placed in
the region posterior to the ventricular atrium in each subject (left). The smaller basal ganglia voxel
contained mainly the globus pallidus (right).

Bookmark URL: /das/journal/view/N/1094366?ja=63030&PAGE=I24.fig&ANCHOR=top&source=MI
http://home.mdconsult.com/das/article/body/jorg=journal&source=MI&sp=1094366&sid=0/N/63030/I24.fig [22/04/2000 0:41:06]

Neurology
Figure 2. Oblique-coronal inversion-recovery image showing the regions of interest selected to measure
globus pallidus and putamen signal intensity.

Bookmark URL: /das/journal/view/N/1094366?ja=63030&PAGE=I25.fig&ANCHOR=top&source=MI
http://home.mdconsult.com/das/article/body/jorg=journal&source=MI&sp=1094366&sid=0/N/63030/I25.fig [22/04/2000 0:41:36]

April 22, 2000
Conde Petra
Severe recurrent hepatic encephalopathy that

responded to oral branched chain amino acids.
Citation
Bibliographic Data
Abstract
Indexing Data
Severe recurrent hepatic encephalopathy
Copyright Notice and Disclaimer that responded to oral branched chain
amino acids.
Chalasani N - Am J Gastroenterol - 1996 Jun; 91(6): 1266-8
Full Text NLM Citation ID:

96237656
Frontmatter
Full Source Title:
INTRODUCTION American Journal of Gastroenterology
CASE REPORT Publication Type:
DISCUSSION Journal Article; Review; Review of Reported Cases
REFERENCES Language:
English
Author Affiliation:
Division of Digestive Diseases, Emory University School of
Medicine, Atlanta, Georgia 30322, USA.
Authors:
Chalasani N; Gitlin N
12
Abstract:
Hepatic encephalopathy is a neuropsychiatric syndrome
occurring in patients with acute or chronic liver disease. Its
pathogenesis remains unclear; however, it appears to be
multifactorial. There are several conventional treatments for this
condition, such as lactulose, neomycin, and protein restriction.
There is significant controversy regarding the role of branched
chain amino acids in the treatment of chronic hepatic
encephalopathy. We describe a patient who had hepatic

encephalopathy secondary to Budd-Chairi syndrome and a

mesoatrial shunt that failed vigorous conventional therapy. She
required multiple hospitalizations for severe recurrent
encephalopathy. The patient was considered for a colonic
exclusion procedure for the management of intractable
encephalopathy. However, branched amino acid therapy was
instituted as a last measure before the contemplated surgery,
and the patient's encephalopathy responded in dramatic fashion,
and she remained free from encephalopathy during a prolonged
follow-up.
Major Subjects:
Amino Acids, Branched-Chain / * Administration &
Dosage
Hepatic Encephalopathy / * Drug Therapy
Acute Disease
Administration, Oral
Case Report
Female
Follow-Up Studies
Human
Middle Age
Recurrence
Time Factors

(Amino Acids, Branched-Chain)


Volume 91 Number 6 June 1996
Brief case reports
SEVERE RECURRENT HEPATIC ENCEPHALOPATHY THAT

RESPONDED TO ORAL BRANCHED CHAIN AMINO ACIDS
N. Chalasani M.D.
Norman Gitlin M.D.
Division of Digestive Diseases, Emory University School of Medicine,
Atlanta, Georgia
Hepatic encephalopathy is a neuropsychiatric syndrome occurring in patients with acute or
chronic liver disease. Its pathogenesis remains unclear; however, it appears to be multifactorial.
There are several conventional treatments for this condition, such as lactulose, neomycin, and
protein restriction. There is significant controversy regarding the role of branched chain amino
acids in the treatment of chronic hepatic encephalopathy. We describe a patient who had hepatic
encephalopathy secondary to Budd-Chairi syndrome and a mesoatrial shunt that failed vigorous
conventional therapy. She required multiple hospitalizations for severe recurrent encephalopathy.
The patient was considered for a colonic exclusion procedure for the management of intractable
encephalopathy. However, branched amino acid therapy was instituted as a last measure before
the contemplated surgery, and the patient's encephalopathy responded in dramatic fashion, and
she remained free from encephalopathy during a prolonged follow-up.
Reprint requests and correspondence: Norman Gitlin, M.D., Division of Digestive Diseases, Emory University School of
Medicine, 2101 Woodruff Memorial Building, Pierce Drive, Atlanta, GA 30322.
Received Dec. 15, 1995; accepted Feb. 27, 1996.
INTRODUCTION
Hepatic encephalopathy (HE) is defined as a neuropsychiatric syndrome associated with altered mental status, elevated
plasma ammonia, and a fetor, occurring in patients with acute or chronic liver failure. Several mechanisms have been
postulated as possible causes of hepatic encephalopathy. These include elevation of plasma ammonia [1] [2] , excess gamma
amino butyric acid (GABA) [3] or activation of GABA receptor sites, excess short chain amino acids [4] , excess
mercaptans [5] , a false neurotransmitter mechanism involving octopamine, or an imbalance of aromatic to branched chain
amino acid (BCAA) ratio in the serum [6] [7] . Protein restriction, lactulose, and neomycin are the mainstay of the treatment
of HE. BCAA, colonic exclusion/colectomy, or liver transplantation may play a role. Here, we describe a patient in whom
BCAA played a major role in the management of severe recurrent hepatic encephalopathy.

CASE REPORT
A 49-yr-old female patient with a history of left iliac vein thrombosis presented to the hospital with abdominal pain and
ascites. Laboratory values included ALT 1082 IU, AST 982 IU, bilirubin 3.6 mg/dl, LDH 504 IU, albumin 2 g/dl, and
globulin 3.1 g/dl. After a detailed workup including abdominal ultrasound, abdominal CT, visceral angiography, and a
liver biopsy, subacute Budd-Chiari syndrome was diagnosed. A mesoatrial shunt was performed to relieve the portal
hypertension. Soon after her discharge
P1267
Figure 1. The clinical course of hepatic encephalopathy in response to various treatments.
from the hospital, the patient developed grade 4 HE, requiring hospitalization, and was successfully treated with oral
lactulose and protein restriction. Over the ensuing 3 months, she had recurrent episodes of severe HE, requiring a total of
six hospitalizations, despite treatment with increasing doses of oral lactulose, 3 g/day of oral neomycin, and a 20-g protein
diet. She was not considered a candidate for liver transplantation because of her difficult vascular anatomy and technical
anastomotic problems. At this stage, a colonic exclusion operation or a total colectomy was considered as a final measure
to treat the incapacitating recurrent HE. Before the contemplated surgery, treatment was started with 45 g/day of an oral
BCAA-enriched formula (Hepatic-aid, McGaw). The clinical response was rapid and sustained despite the discontinuation
of lactulose and neomycin. She remained free from HE at 3-, 6-, 9-, and 12-month follow-up (Fig. 1) . Several attempts to
wean the patient from BCAA resulted in worsening of her mental condition and early encephalopathy, indicating the
therapeutic role of the BCAA.
DISCUSSION
As early as 1957, it was reported that there was an increase in the serum levels of aromatic amino acids (AAA)
(phenylalanine, tyrosine, and tryptophan) and a decrease in the serum levels of BCAA (valine, leucine, isoleucine) in
patients with liver failure, with the normal BCAA:AAA ratio of 3.5:1.0 changing to 1.0:1.0 [6] . The excess AAA may
interfere in central nervous system neurotransmission by competitively inhibiting the normal neurotransmitters (dopamine,
norepinephrine) [7] . Based on this theory, several studies have been undertaken to evaluate the use of BCAA-enriched
formulas to restore the AAA:BCAA ratio and, thereby, to treat HE. The role of oral BCAA therapy in cirrhotic patients
with chronic HE is controversial, and studies have yielded mixed results [8] [9] [10] [11] [12] .
Regardless of the controversy surrounding the use of oral BCAA in the treatment of chronic HE, our patient has responded
dramatically to oral BCAA therapy in a prolonged follow-up. She has not been hospitalized for 1 yr, with a significant
socioeconomic impact. She avoided debilitating surgery carrying a significant morbidity and mortality, and there has been
a dramatic improvement in the quality of her life. On the basis of this experience, we believe that patients with intractable
severe recurrent HE should be considered candidates for therapy with BCAA formula if standard therapeutic options are
ineffective.
REFERENCES
1. 1. Hensen DM. Portal systemic encephalopathy and hepatic coma. Med Clin North Am 1986;70:1081-92.
2. 2. Lockwood AH, McDonald JM, Reiman RE, et al. The dynamics of
P1268

ammonia metabolism in man: Effects of liver disease and hyperammonemia. J Clin Invest 1979;63:449-60.
3. 3. Ferenci P, Schafer DF, Kleinberger G, et al. Serum levels of gamma-amino butyric acid like activity in acute and chronic hepatocellular disease. Lancet
1983;2:811-4.
4. 4. Muto Y. Clinical study on the relationship of short chain fatty acids and hepatic encephalopathy. Jpn J Gastroenterol 1966;63:19-23.
5. 5. Zieve L, Doizaki WM, Zieve FJ. Synergism between marcaptans and ammonia in the production of coma. A possible role for mercaptans in the pathogenesis
of hepatic coma. J Lab Clin Med 1974;83:16.
6. 6. Iber FL, Rosen H, Levenson SM, et al. The plasma amino acids in patients with liver failure. J Lab Clin Med 1957;50:417-525.
7. 7. Fischer JE, Baldessarini RJ. False neurotransmitters and hepatic failure. Lancet 1971;2:75-80.
8. 8. Siez A, Walker S, Czygan P, et al. Branched chain aminoacid-enriched elemental diet in patients with cirrhosis of the liver. A double blind crossover trial. Z
9. 9. Morgan MY. Branched chain amino acids in the management of chronic liver disease: Facts and fantasies. J Hepatol 1990;11:133-41.
10. 10. Marchesini G, Dioguardi FS, Bianchi GP, et al. Long-term oral branched-chain amino acid treatment in chronic hepatic encephalopathy. A randomized
double-blind casein-controlled trial. J Hepatol 1990;11:92-101.
11. 11. Naylor CD, O'Rourke K, Detsky AS, et al. Parenteral nutrition with branched-chain amino acids in hepatic encephalopathy. A meta-analysis. Gastroenterol
1989;97:1033-42.
12. 12. Eriksson LS, Conn HO. Branched-chain amino acids in the management of hepatic encephalopathy: An analysis of variants. Hepatology 1989;10:228-46.


April 22, 2000
Conde Petra
Subclinical hepatic encephalopathy: how best to

diagnose?
Citation
Bibliographic Data
Indexing Data
Subclinical hepatic encephalopathy: how
best to diagnose?
Find More Articles Like This Chalasani N - Am J Gastroenterol - 1997 May; 92(5): 905-6
NLM Citation ID:
Full Text
97293165
Frontmatter
Full Source Title:
COMMENT American Journal of Gastroenterology
REFERENCES Publication Type:
Journal Article
Language:
English
Author Affiliation:
Division of Digestive Diseases, Emory University School of
Medicine, Atlanta, GA, USA.
Authors:
Chalasani N; Gitlin N
Major Subjects:
* Electroencephalography
Hepatic Encephalopathy / * Diagnosis / Etiology /

Physiopathology / Psychology
* Neuropsychological Tests
Diagnosis, Differential
Human
Liver Cirrhosis / Complications

http://home.mdconsult.com/das/journal/view/N/9371503?...S,MI&stty=cow&nid=39986&tlt=Z&pos=&tag=relinfo&anc=ja [22/04/2000 0:43:56]


SUBCLINICAL HEPATIC ENCEPHALOPATHY: HOW BEST TO

DIAGNOSE?
Quero JC, Hartmann IJC, Meul stee J, et al.
The Clinical Diagnosis of Subclinical Hepatic Encephalopathy in Patients with Cirrhosis Using
Neuropsychological Tests and Automated Electroencephalogram Analysis
Hepatology 1996;24:556-60
ABSTRACT
Neuropsychological tests used for the assessment of subclinical hepatic encephalopathy (SHE) may
overdiagnose SHE, because scores are usually not corrected for age. The aim of this study was to
estimate the prevalence of SHE using two easily administrable psychometric tests (Number
Connection Test part A [NCT-A] and Symbol Digit Test [SDT]) with age-related normal values. In
addition, spectral electroencephalogram (EEG) was used, which is the in-house
electrophysiological method for quantifying encephalopathy.
One hundred and thirty-seven consecutive patients (mean age 49 yr, range 17-77) with cirrhosis,
without any clinical signs of encephalopathy, were screened for SHE. In addition, the Child-Pugh
score and the arterial blood ammonia were determined. Patients with concurrent use of alcohol,
benzodiazepines, or anti-epileptics were excluded.
Fifty percent of the patients had an abnormal NCT according to the standard recommended
procedure; in contrast, only 7% of the patients had an abnormal NCT when scores corrected for
age were used. Combining the results of the spectral EEG and the psychometric tests corrected for
age yielded a higher prevalence of SHE (23%) than when each test method was used alone (17 vs
10% abnormal, respectively). Severity of liver disease correlated with the presence of SHE,
because the prevalence of abnormal tests increased from 14% in Child-Pugh grade A to 45% in
Child-Pugh B or C. Age above 40 yr and an elevated blood ammonia level were significant
determinants related to an abnormal EEG. We conclude that the NCT uncorrected for age
markedly overdiagnoses SHE and, therefore, should not be used as a test for the screening of SHE.
A low prevalence of SHE in patients with Child A liver cirrhosis is found when using a
combination of spectral EEG and two psychometric tests with age-corrected normal values. Older
patients with an elevated arterial ammonia are more prone to develop SHE than younger patients
with an equal arterial ammonia concentration.
COMMENT
It seems that stable cirrhotics who have no signs of overt hepatic encephalopathy show evidence of SHE
on psychometric testing or other advanced diagnostic tests, such as evoked potentials, spectral
encephalography, or magnetic resonant spectroscopy. There is a significant controversy relating to the

optimal diagnostic tests to diagnose SHE in an ambulatory setting. Depending upon the modality and
criteria used to diagnose, the prevalence of SHE can vary from 23-70%. The advantages of psychometric
tests are that they are inexpensive and that they can be performed quickly and with ease in an outpatient
setting. In the above paper, Quero and co-authors argue that the routine psychometric tests may
overdiagnose SHE and they recommend that these tests be corrected for age and other variables, such as
literacy and linguistic skills. The significance and validity of their argument is unclear. There have been
reports in the literature suggesting that performance on the number connecting test (NCT), which is also
known as "trail test," may vary depending in the age and level of education [1] . However,
906
the impact of age and other variables on other psychometric tests or their combination is unknown.
Furthermore, earlier studies using age and other variable-matched controls found a much higher
prevalence of SHE (up to 70%) in the stable cirrhotic patients [2] [3] . Diagnostic tests, such as MR
spectroscopy or spectral encephalography, seem to have greater specificity, but their sensitivity to
diagnose SHE is unclear and these tests are impractical in an ambulatory setting. At present, there is no
gold standard for diagnosing SHE and the consensus of opinion favors using a combination of two or
more psychometric tests to diagnose SHE.
Subclinical hepatic encephalopathy is an important, often neglected, entity with major implications. It
has been demonstrated that the patients with SHE diagnosed on psychometric testing may have
diminished capabilities, such as driving motor vehicles or handling heavy machinery in the construction
industry. The patients with SHE can be treated effectively with dietary protein restriction, administration
of nonabsorbable disaccharides, such as lactulose or lactitol, or supplementation with branched chain
aminoacids.
Naga Chalasan M.D. Norman Gitlin M.D., F.A.C.G. Division of Digestive Diseases Emory University
School of Medicine
Atlanta, GA
REFERENCES
1. Zeneroli ML, Ciono G, Ventura P, et al. Interindividual variability of the number connection test. J Hepatol
1992;15:263-4.
2. Moore JW, Dunk AA, Crawford JR, et al. Neuropsychological deficits and morphological MRI brain scan abnormalities
in apparently healthy non-encephalopathic patients with cirrhosis: A controlled study. J Hepatol 1989;9:319-25.
3. Gitlin N, Lewis DC, Hinkley L. The diagnosis and prevalence of subclinical hepatic encephalopathy in apparently
healthy, ambulant, non-shunted patients with cirrhosis. J Hepatol 1986;3:75-82.


April 22, 2000
Conde Petra
Markers of reversible hepatic encephalopathy

[letter; comment]
Citation
Bibliographic Data
Indexing Data
Markers of reversible hepatic
encephalopathy [letter; comment]
Find More Articles Like This Krieger D - Neurology - 1997 Oct; 49(4): 1187-8
NLM Citation ID:
Full Text
97479651
Frontmatter
Comment:
Reply from the Authors: Neurology 1996 Dec;47(6):1526-30
About the Publication Full Source Title:
Neurology
Publication Type:
Comment; Letter
Language:
English
Authors:
Krieger D; Krieger S
Major Subjects:
Hepatic Encephalopathy / * Diagnosis
Human
Magnetic Resonance Imaging
Nuclear Magnetic Resonance

http://home.mdconsult.com/das/journal/view/N/9930893?...e=MI&stty=cow&nid=39986&tlt=Z&pos=&tag=relinfo&anc=ja [22/04/2000 0:44:44]

April 22, 2000
Conde Petra
Helicobacter pylori, hyperammonemia, and

hepatic encephalopathy: is there a correlation?
[letter]
Citation
Bibliographic Data
Indexing Data
Helicobacter pylori, hyperammonemia,
and hepatic encephalopathy: is there a
correlation? [letter]
Rinaldi V - Am J Gastroenterol - 1997 Apr; 92(4): 723-4
Full Text
NLM Citation ID:
About the Publication Full Source Title:
Publication Type:
Letter
Language:
English
Authors:
Rinaldi V; Zullo A; Diana F; Capocaccia L
Major Subjects:
Ammonia / * Blood
* Helicobacter pylori
Helicobacter Infections / * Blood / Complications
Hepatic Encephalopathy / * Etiology
Human
Liver Cirrhosis / Blood / Complications

7664-41-7 (Ammonia)
http://home.mdconsult.com/das/journal/view/N/9470388?...e=MI&stty=cow&nid=39986&tlt=Z&pos=&tag=relinfo&anc=ja [22/04/2000 0:45:22]

April 22, 2000
Conde Petra
April 14, 2000

SOB


of cirrhosis.
knowledge.




Bibliography
110(4):253-255, 1989(editorial)


Saunders







70(5):1081-1091, 1986


8(1):85-99, 1986








WB Saunders



April 22, 2000
Conde Petra
April 14, 2000
Question 6 - Liver Transplantation

Orthotopic liver transplantation (OLT) has become routine therapy for advanced
liver disease. This procedure has improved life expectancy, and most OLT
recipients are able to return to work and have a relatively normal life. This success
is due to improvement in surgical technique, effective immunosuppressive therapy,
and better guidelines regarding indications for transplantation. The most common
indication for adult liver transplantation is end-stage cirrhosis. In evaluating a
cirrhotic patient for liver transplantation, one should use the Child-Turcotte
classification. This classification uses albumin, bilirubin, control of ascites, degree
of encephalopathy, and nutrition status to determine the patient's clinical condition.
Worsening of the above parameters indicates clinical deterioration resulting in
life-threatening events. Detection of such deterioration by the Child score is
important in confirming candidacy for liver transplantation. Additional clinical events
that may signal the need for transplantation include recurrent SBP, refractory
variceal bleeding, hepatorenal syndrome, symptomatic coagulopathy, and the
nonspecific but debilitating symptoms of fatigue and weakness. The above events
are common to end-stage cirrhosis. Other more specific events that also signal the
need for transplantation are progressive, severe bone disease, seen in primary
biliary cirrhosis and primary sclerosing cholangitis, and recurrent bacterial
cholangitis, often seen in patients with primary sclerosing cholangitis. The
disease-specific indications for liver transplantation are acute liver failure, cirrhosis
from previous alcohol abuse, cirrhosis from chronic hepatitis C, cryptogenic
cirrhosis, primary biliary cirrhosis, cirrhosis from other viral hepatitis (B and D),
primary sclerosing cholangitis, cirrhosis from autoimmune chronic active hepatitis,
cirrhosis due to alpha-1-antitrypsin deficiency, Budd-Chiari syndrome, and
It is also important to consider other factors in determining candidacy for liver

transplantation. The most common disease pre-OLT is alcoholic cirrhosis.
Abstinence from alcohol is a requirement before consideration for liver
transplantation. After transplant, patients need to be compliant with
immunosuppressive medications. Patients must be emotionally capable to handle
such a life-long commitment. Family support and adequate resources are also
essential for successful liver transplantation.
Read more about evaluation of the patient for liver

transplantation:
Proceed With Case

Sabiston: Textbook of Surgery, 15th ed.
LIVER TRANSPLANTATION AT THE MILLENNIUM Past, Present, and Future

Keeffe EB - Clinics in Liver Disease - 2000 Feb; 4(1); 241-255
Split liver transplantation.

Busuttil RW - Ann Surg - 1999 Mar; 229(3): 313-21
From NIH/NLM MEDLINE, HealthSTAR; NCI CANCERLIT
SELECTION OF PATIENTS FOR LIVER TRANSPLANTATION IN 1997 AND

BEYOND
Yoshida EM - Clinics in Liver Disease - 1997 Aug; 1(2); 247-261
PREFACE
KEEFFE EB - Clinics in Liver Disease - 1997 Aug; 1(2); 13A-14A
LIVER TRANSPLANTATION FOR ALCOHOLIC LIVER DISEASE

Burke A - Clinics in Liver Disease - 1998 Nov; 2(4); 839-850
The changing spectrum of treatment for variceal bleeding.

Guidelines for surgical procedures after liver transplantation.

Testa G - Ann Surg - 1998 Apr; 227(4): 590-9
Liver transplantation. American Association for the Study of Liver Diseases.

Carithers RL Jr - Liver Transpl - 2000 Jan; 6(1): 122-35
Indications for referral and assessment in adult liver transplantation: a clinical

guideline. British Society of Gastroenterology.
Devlin J - Gut - 1999 Dec; 45 Suppl 6: VI1-VI22

Minimal criteria for placement of adults on the liver transplant waiting list: a report
of a national conference organized by the American Society of Transplant
Physicians and the American Association for the Study of Liver Diseases.
Lucey MR - Transplantation - 1998 Oct 15; 66(7): 956-62
Clinical practice guidelines for the management of cirrhotic patients with ascites.
Committee on Ascites of the Italian Association for the Study of the Liver.
Salerno F - Ital J Gastroenterol Hepatol - 1999 Oct; 31(7): 626-34
Alcoholic liver disease: proposed recommendations for the American College of

Gastroenterology [see comments]
McCullough AJ - Am J Gastroenterol - 1998 Nov; 93(11): 2022-36


April 22, 2000
Conde Petra

Company
Section 9 - Liver
83 - Liver Transplantation Feldman: Sleisenger & Fordtran's Gastrointestinal and Liver Disease,
Introduction
PATIENT SELECTION FOR

LIVER TRANSPLANTATION
1404
CONTROVERSIES IN PATIENT
SELECTION
PATIENT SELECTION
Chapter 83 - Liver
SUMMARY Transplantation
POST-TRANSPLANTATION
MANAGEMENT
John R. Lake
COMPLICATIONS RELATED TO Jerome Gournay
IMMUNOSUPPRESSION
POST-TRANSPLANTATION PATIENT SELECTION FOR LIVER TRANSPLANTATION,

SUMMARY 1404
REFERENCES Contraindications, 1404
Indications, 1405
Timing, 1406
CONTROVERSIES IN PATIENT SELECTION, 1406
Alcoholic Liver Disease, 1406
Malignant Disease, 1407
Hepatitis B, 1407
Hepatitis C, 1407
Acute Liver Failure, 1408
PATIENT SELECTION SUMMARY, 1408
POST-TRANSPLANTATION MANAGEMENT, 1408
Hepatic Allograft Dysfunction, 1408
Post-Transplantation Hepatitis, 1410
Biliary Tract Complications, 1410

Vascular Thrombosis, 1411

Recurrent Viral Disease, 1411
Recurrence of Nonviral Disease, 1411
COMPLICATIONS RELATED TO
IMMUNOSUPPRESSION, 1412
Cyclosporine/Tacrolimus Toxicity, 1412
Infections, 1413
Other Post-Transplantation Complications, 1414
POST-TRANSPLANTATION SUMMARY, 1415
Orthotopic liver transplantation (OLT) has become the
treatment of choice for many patients with end-stage liver
disease and fulminant hepatic failure. [1] No other treatment
offers an equivalent chance to restore health. In 1993, 3500
patients in the United States and as many as 10,000 patients
worldwide received transplants. [2] The survival of patients
undergoing OLT has also improved dramatically. In 1980, only
approximately 50% of recipients survived 1 year. In 1996, most
leading programs achieved a 1-year patient survival rate of
more than 85% and a 5-year survival rate of more than 75%
(Fig. 83-1) . In the United States, there were approximately
7000 patients on the waiting list for OLT as of January 1997;
this represents
Figure 83-1 Survival curves for the United Network

for Organ Sharing liver transplantation database.
a more-than-1-year "backlog" of patients in need of OLT. Most

of the Western world is experiencing a similar shortage of
donor organs. This requires transplant physicians to carefully
scrutinize the indications for OLT and the outcome of OLT for
various diseases. It has become clear since the mid-1980s that
most diseases for which OLT is performed have the potential
for recurrence. However, the presentation and clinical
importance of recurrent disease vary greatly. In this chapter, we
discuss the current indications for OLT as well as the timing of
OLT and outcome and address the common problems
encountered after transplantation, including hepatic allograft
dysfunction and problems related to immunosuppression.
Most adult patients undergoing OLT have cirrhosis and
complications of portal hypertension. [1] [2] Among the most

common diseases for which OLT is performed in adults are
alcoholic liver disease (ALD), chronic viral hepatitis, primary
biliary cirrhosis (PBC), primary sclerosing cholangitis (PSC),
and autoimmune hepatitis. [2] The most common disease
indications for OLT in pediatric patients are extrahepatic biliary
atresia and alpha1 -antitrypsin deficiency, which account for
55% and 6.6% of pediatric recipients, respectively. [2] These
disease indications have changed since 1980. In particular, the
number of patients with ALD who have been offered OLT has
increased, as has the number of patients with acute liver failure
who have undergone transplantation. In contrast, far fewer
patients with malignant disease have received transplants.
The number of transplantations performed each year is limited
by the number of available donors. This donor shortage makes
it essential to maximize the use of every potential donor. This
means not only increasing organ donation but also carefully
examining the outcome of recipients undergoing OLT. It also
has become increasingly important to lower the costs of OLT.
Two factors that affect cost are patient selection and the timing
of OLT.


Saunders Company
1404
Chapter 83 - Liver Transplantation
John R. Lake
Jerome Gournay
PATIENT SELECTION FOR LIVER TRANSPLANTATION, 1404

Contraindications, 1404
Indications, 1405
Timing, 1406
CONTROVERSIES IN PATIENT SELECTION, 1406
Alcoholic Liver Disease, 1406
Malignant Disease, 1407
Hepatitis B, 1407
Hepatitis C, 1407
Acute Liver Failure, 1408
PATIENT SELECTION SUMMARY, 1408
POST-TRANSPLANTATION MANAGEMENT, 1408
Hepatic Allograft Dysfunction, 1408
Post-Transplantation Hepatitis, 1410
Biliary Tract Complications, 1410
Vascular Thrombosis, 1411
Recurrent Viral Disease, 1411
Recurrence of Nonviral Disease, 1411
COMPLICATIONS RELATED TO IMMUNOSUPPRESSION, 1412
Cyclosporine/Tacrolimus Toxicity, 1412
Infections, 1413
Other Post-Transplantation Complications, 1414
POST-TRANSPLANTATION SUMMARY, 1415
Orthotopic liver transplantation (OLT) has become the treatment of choice for many patients with
end-stage liver disease and fulminant hepatic failure. [1] No other treatment offers an equivalent chance to

restore health. In 1993, 3500 patients in the United States and as many as 10,000 patients worldwide
received transplants. [2] The survival of patients undergoing OLT has also improved dramatically. In
1980, only approximately 50% of recipients survived 1 year. In 1996, most leading programs achieved a
1-year patient survival rate of more than 85% and a 5-year survival rate of more than 75% (Fig. 83-1) . In
the United States, there were approximately 7000 patients on the waiting list for OLT as of January 1997;
this represents
Figure 83-1 Survival curves for the United Network for Organ Sharing liver transplantation database.
a more-than-1-year "backlog" of patients in need of OLT. Most of the Western world is experiencing a
similar shortage of donor organs. This requires transplant physicians to carefully scrutinize the
indications for OLT and the outcome of OLT for various diseases. It has become clear since the
mid-1980s that most diseases for which OLT is performed have the potential for recurrence. However,
the presentation and clinical importance of recurrent disease vary greatly. In this chapter, we discuss the
current indications for OLT as well as the timing of OLT and outcome and address the common
problems encountered after transplantation, including hepatic allograft dysfunction and problems related
to immunosuppression.
Most adult patients undergoing OLT have cirrhosis and complications of portal hypertension. [1] [2]
Among the most common diseases for which OLT is performed in adults are alcoholic liver disease
(ALD), chronic viral hepatitis, primary biliary cirrhosis (PBC), primary sclerosing cholangitis (PSC), and
autoimmune hepatitis. [2] The most common disease indications for OLT in pediatric patients are
extrahepatic biliary atresia and alpha1 -antitrypsin deficiency, which account for 55% and 6.6% of
pediatric recipients, respectively. [2] These disease indications have changed since 1980. In particular, the
number of patients with ALD who have been offered OLT has increased, as has the number of patients
with acute liver failure who have undergone transplantation. In contrast, far fewer patients with
malignant disease have received transplants.
The number of transplantations performed each year is limited by the number of available donors. This
donor shortage makes it essential to maximize the use of every potential donor. This means not only
increasing organ donation but also carefully examining the outcome of recipients undergoing OLT. It
also has become increasingly important to lower the costs of OLT. Two factors that affect cost are patient
selection and the timing of OLT.


Saunders Company
PATIENT SELECTION FOR LIVER TRANSPLANTATION

Contraindications
Contraindications to OLT can be divided into two categories. Absolute contraindications (Table 83-1)
include clinical conditions
1405
TABLE 83-1 -- Absolute Contraindications to Liver Transplantation

HIV seropositivity
Extrahepatic malignancy
Advanced cardiac or pulmonary disease
Inability to comply with an immunosuppression protocol
Severe and uncontrolled extrahepatic infection
Multisystem organ failure
for which the results of OLT are so poor that it should not be offered. Relative contraindications (Table
83-2) include clinical conditions that negatively affect survival but not to such a degree that
transplantation should never be considered. Indeed, transplantation is appropriate for selected patients
with fulminant hepatic failure. The contraindications listed reflect the experience of the liver transplant
program at the University of California, San Francisco and the consensus statement on indications for
OLT from a meeting held in 1993. [3]
Human immunodeficiency virus (HIV) infection is generally considered an absolute contraindication
because the survival of patients with HIV infection has been poor. [4] There is little argument that
malignancy outside the biliary tree represents a contraindication. It has traditionally been considered that
advanced cardiopulmonary disease is an absolute contraindication. However, patients with hypoxemia
and chronic liver disease should undergo a careful evaluation to determine the etiology of the hypoxemia.
Patients with hepatopulmonary syndrome, defined as hypoxemia caused by noncardiac, right-to-left
shunting, can improve substantially after OLT, and disabled patients can be transformed into active
persons with few physical limitations. [5] Advanced cardiac disease, however, remains an absolute
contraindication to OLT unless simultaneous transplantation of the heart can be performed. [1]
Extrahepatic infection that is not controlled with the appropriate antibiotic therapy (e.g., septic shock,
septicemia, pneumonia) also remains an absolute contraindication to transplantation. Many patients
experience spontaneous bacterial peritonitis before OLT. [6] It has been shown that the peritoneal fluid is

sterile after only 48 hours of antibiotic therapy and that transplantation need not be delayed much beyond
this 48-hour period. Finally, the results of OLT when three or more organs, including the liver, have
failed are too poor to justify the use of a donor organ.
Renal insufficiency has been associated with decreased patient survival after OLT; this is independent of
the cause of renal insufficiency and of whether combined liver and kidney transplantation is performed.
Similarly, the survival of patients undergoing OLT for end-stage hemochromatosis has been
TABLE 83-2 -- Relative Contraindications to Liver Transplantation and Factors Associated with
Decreased Survival
Renal insufficiency
Hepatitis B virus infection with viremia (i.e., HBV DNA present in serum)
Primary hepatobiliary malignancy
Hemochromatosis
Fulminant hepatic failure
TABLE 83-3 -- Quality-of-Life Indications for Liver Transplantation

Cholestatic liver disease
Intractable pruritus
Metabolic bone disease with fracture
Recurrent episodes of biliary sepsis
Xanthomatous neuropathy
Chronic liver disease (with or without cholestasis)
Intractable ascites
Encephalopathy
Bleeding from esophageal or gastric varices
Fatigue
Inborn errors of metabolism
Correction of nonhepatic manifestations of metabolic liver disease
poor. This appears to be the result of increased early mortality as a result of occult cardiac disease. Many
regard the presence of hepatitis B virus (HBV) DNA or hepatitis B antigen (HBeAg) in serum (which
indicates active viral replication) to be a contraindication to OLT. The risk of recurrent HBV infection
and disease, with the potential for graft lost, may be as high as 90% in the absence of hepatitis B
immunoglobulin (HBIG) (see below). [7] Assessment of the likelihood of compliance with medical

therapy is essential because noncompliance is one of the most important causes of late allograft rejection.
The results of OLT for hepatobiliary malignancy and acute liver failure are discussed below.
Indications
Indications for OLT can be divided generally into two categories: quality-of-life indications (Table 83-3)
, which represent conditions that markedly impair the patient's quality of life, and severity of disease
indications (Table 83-4) , which identify patients at risk for death in the near (i.e., within 1 year) future.
Quality-of-Life Indications
The most common quality-of-life indication is intractable ascites, defined as ascites that is resistant to
diuretic therapy. Several recent reports have demonstrated the efficacy of the use of transjugular
intrahepatic portosystemic shunts (TIPS) in such patients. [8] Whether TIPS provides long-term relief
TABLE 83-4 -- Severity of Disease Indications for Liver Transplantation
Chronic liver disease
Hepatorenal syndrome
Recurrent spontaneous bacterial peritonitis
Serum albumin <2.5 g/dL
Prothrombin time >5 seconds prolonged
Serum bilirubin >5.0 mg/dL
Cholestatic liver disease
Serum bilirubin >10 mg/dL
Mayo Clinic model (for primary biliary cirrhosis and primarysclerosing cholangitis) predictive of
<2-year survival
1406
from intractable ascites or simply functions as a bridge to OLT requires further study.
Encephalopathy is the second most common quality-of-life indication. Patients who experience repeated
bouts of encephalopathy despite therapy with lactulose, neomycin, or both should be referred for OLT.
The differentiation between chronic encephalopathy and irreversible organic brain syndromes can be
difficult; formal psychomotor testing may be helpful in this regard.
Bleeding from gastroesophageal varices has become a less common indication for OLT, in large part
reflecting improved methods for managing recurrent variceal hemorrhage, including the use of
beta-blockers, variceal banding, and TIPS (see Chapter 77) . TIPS has been shown to be effective for
patients with bleeding refractory to other forms of therapy. [9] Unfortunately, 25% of patients develop
post-TIPS encephalopathy and as many as 60% of patients will require TIPS revision because of a

stenosis or an occlusion leading to the recurrence of portal hypertension. Furthermore, the long-term
results of TIPS remain to be defined. Nevertheless, the results of OLT for patients treated with TIPS are
substantially better than those for patients treated with surgical shunts. [10] TIPS is generally the preferred
management of variceal bleeding and intractable ascites in patients awaiting OLT. The major advantages
of TIPS are that the extrahepatic anatomy is not altered, surgery in the right upper quadrant is avoided,
and the shunt is entirely removed with the explant at the time of transplantation. [9]
Fatigue is likely one of the more troublesome quality-of-life indications because of the difficulty in
differentiating it from depression, which is common in persons with chronic disease.
Several quality-of-life indications are unique to patients with cholestatic liver disease, such as PBC, PSC,
and extrahepatic biliary atresia; the most common of these is intractable pruritus. Many patients with
cholestasis develop pruritus that fails to respond to medical therapy, including the use of ursodeoxycholic
acid and rifampicin. [11] These patients are also at risk for metabolic bone disease, which can lead to
fractures, even before the development of advanced liver disease. [12] OLT is the only therapy that has
been shown to increase bone mineralization in such patients. OLT is also an effective therapy for
xanthomatous neuropathy, a rare but disabling complication of severe cholestasis.
With the marked improvement in patient survival, it is reasonable to consider OLT for correction of
nonhepatic manifestations of certain metabolic disease in which the genetic defect is expressed in the
liver (e.g., Crigler-Najjar syndrome, severe familial hypercholesterolemia, and hereditary oxalosis).
Disease Severity
The second category of indications for OLT identify patients who have a less-than-50% chance of
surviving 1 year (see Table 83-4) . Unfortunately, the natural history of most common forms of liver
disease, such as postviral cirrhosis and ALD, is not well defined, making accurate predictions difficult.
Hepatorenal syndrome represents a very poor prognostic factor, and these patients usually require urgent
transplantation. [13] Similarly, patients with two or more episodes of recurrent spontaneous bacterial
peritonitis have a limited life expectancy. [14] However, the use of antibiotic prophylaxis has been shown
to be effective in decreasing recurrent episodes of spontaneous bacterial peritonitis (see Chapter 78) .
In contrast to most other forms of liver disease, good natural history data are available for the cholestatic
forms of liver disease, in particular, PBC and PSC (see Chapters 59 and 76) . Prognostic models have
been developed that allow prediction of survival. [15] [16] The Mayo Clinic model for PBC uses five
independent clinical variables: serum bilirubin and albumin concentrations, age, prothrombin, and the
presence or absence of peripheral edema. Serum bilirubin concentration is the most predictive of these
variables; when it reaches 10 mg/dL in patients with PBC, the chance of surviving 2 years is less than
50%. [17] Similar scoring systems have been developed for PSC. [15] As with PBC, the most predictive
variable is serum bilirubin. Thus, once a patient with either of these diseases develops clinical jaundice,
he or she should be evaluated for OLT.
Timing
The improvement in survival after OLT has changed our concepts regarding the appropriate timing of
OLT--in particular, the optimal timing in terms of survival, financial cost, rehabilitation periods, and
functional outcome. For example, a study from the Mayo Clinic demonstrated that patients who received

a transplant early in the course of the disease had higher survival rates and lower financial costs after
transplantation. [16] A second study showed a similar correlation between cost and disease severity, as
indicated by the patient's United Network for Organ Sharing (UNOS) status at the time of transplantation
(i.e., worse UNOS status was equated with greater costs). [18] It is noteworthy that the UNOS status of
adult OLT recipients has improved since 1991: 10.7% of patients who received a transplant before 1991
were in the best functional categories compared with 22.5% of the recipients in 1991 and 21.9% of the
recipients in 1992. [2]


Saunders Company
CONTROVERSIES IN PATIENT SELECTION

Several specific diseases generate controversy regarding the appropriateness of OLT as treatment. For
HBV and hepatobiliary malignancy, the issue is a relatively poor outcome after transplantation. [2] For
ALD, the issues are much different. Figure 83-2 lists 2-year patient survival by disease for patients
enrolled in the National Institute of Diabetes and Digestive and Kidney Diseases OLT Database. The best
survival results are obtained for patients with the cholestatic forms of liver disease (PBC and PSC);
excellent results are also achieved for most of the other forms of chronic liver disease. The diseases that
exhibit the worse short-term survival rates are malignancies, chronic HBV, and fulminant hepatic failure;
even though most diseases have the potential for recurrence, it is only recurrence of HBV and
malignancy that has an impact on short-term survival.
Alcoholic Liver Disease
Initially, OLT was offered to relatively few patients with ALD for several reasons. First, the effect of
alcohol abuse on other
1407
Figure 83-2 Two-year survival rates for patients in the National Institute for Diabetes and Digestive
and Kidney Diseases liver transplantation database by disease indication, including fulminant hepatic
failure (FHF), cirrhosis caused by chronic hepatitis B (CAH-B) and hepatitis C (CAH-C) virus
infection, autoimmune hepatitis (AIH), alcoholic liver disease (ALD), primary biliary cirrhosis
(PBC), primary sclerosing cholangitis (PSC), extrahepatic biliary atresia (EBA), metabolic liver
disease (MET), and hepatobiliary malignancy (MALIG).
organs (e.g., pancreas, heart, and the central venous system) could make the surgical procedure and
perioperative management more difficult while adversely affecting outcome. Second, alcohol treatment
programs lead to long-term sobriety in fewer than half of the patients. [19] Thus, it was believed that most
patients would return to alcohol abuse after restoration of his or her health with OLT. Third, there were
concerns about the ability of an alcoholic patient to comply with a relatively strict post-OLT regimen.
Finally, it was believed that individuals or families donating organs would not want these organs to go to
patients suffering from the effects of alcohol abuse.
To date, these concerns have not been borne out. [20] [21] [22] The survival of alcoholic patients undergoing
OLT is no worse than the survival of patients transplanted for non-ALD, and in several programs it is
even better. In addition, the incidence of recidivism after OLT for ALD is reported to be between 10%
and 20%. [20] [21] [22] Deaths from recurrent ALD are extremely uncommon. The percentage of alcoholic
patients who return to work and to his or her former quality of life has been comparable to that of
nonalcoholic patients. Compliance with medical therapy also appears to be similar in the two groups of
patients. [23] Finally, most people believe that organ donation should not be restricted based on the cause
of liver disease in the recipient.

Nevertheless, transplantation for ALD remains controversial. An estimated 20,000 patients die yearly of
ALD in the United States. Thus, it is possible that patients with ALD will overwhelm U.S. OLT
programs. One controversial issue is the period of time for which an alcoholic patient should be sober
before OLT is offered. In one study, the incidence of recidivism was significantly lower in patients who
had been sober for at least 6 months compared with recipients who had been sober for a shorter period.
[22] Some programs require that patients enroll in alcohol-treatment programs. However, the impact of
pretransplantation alcohol treatment on the maintenance of long-term sobriety after OLT remains to be
defined.
Malignant Disease
Initially, as many as one third of patients undergoing OLT received a transplant as therapy for malignant
disease. The belief was that patients with hepatocellular carcinoma (HCC) and cholangiocarcinoma had
disease confined to the hepatobiliary tree and that dissemination was a relatively late event.
Unfortunately, the results did not bear out this theory. The majority of patients undergoing OLT for HCC
experience recurrence. [24] [25] [26] The 3-year survival has usually been less than 50%, with the most
common cause of death being recurrent malignancy. Prognostic factors that represent poor prognosis
variables include larger tumors, vascular invasion, lymph node involvement, and cirrhosis. On the other
hand, patients with "incidental" tumors (i.e., tumors discovered incidentally during the transplantation
procedure) have a better prognosis. Several programs have recently reported excellent results with the
use of OLT for HCC when conservative selection criteria are used. For example, several programs have
shown that the survival of transplant patients who had tumors that were less than 5 cm in diameter, fewer
than three in number, and without clinical evidence of vascular or lymph node involvement have survival
rates similar to transplant patients who had nonmalignant disease. [27] In some programs, adjuvant
chemotherapy or chemoembolization is used in an attempt to improve outcome. [28] Several studies have
suggested that these additional therapies might improve results. [27] [28]
The results of OLT for the treatment of cholangiocarcinoma have been so poor (a 3-year survival rate of
<30%) that most programs recognize this as an absolute contraindication to transplantation. [25] [26]
Hepatitis B
HBV infection recurs in most patients undergoing OLT for chronic HBV in the absence of
immunoprophylaxis. [29] Furthermore, although it often takes 10 to 20 years for immunocompetent
patients with chronic HBV to develop advanced liver disease, after OLT the disease can progress to
cirrhosis or liver failure within 1 year. [29] A minority of these patients develop a characteristic histologic
lesion termed fibrosing cholestatic hepatitis, [30] which is characterized by a lack of inflammation and
hepatocytes that are swollen and in disarray. Hepatocytes express large amounts of viral material. Once
this lesion develops, the outcome is usually fatal.
Not all forms of HBV recur at the same rate. [7] [31] Patients receiving OLT for acute HBV have HBV
recurrence rates that are less than those of patients with chronic HBV. Patients who are coinfected with
hepatitis D virus (HDV) also are less likely to develop HBV reinfection, likely reflecting the suppression
of HBV replication by HDV. Patients at the highest risk for recurrence are patients with chronic HBV, in
whom HBV DNA can be detected in serum by methods other than polymerase chain reaction. Similarly,
the presence of HBeAg in serum predicts a higher rate of recurrence. [7]

The intravenous administration of high concentrations of HBIG decreases the risk of HBV recurrence
after OLT, especially in patients who have acute HBV or HDV coinfection and in patients with chronic
hepatitis who are HBV DNA negative in serum as assessed with standard methods. [31] HBIG must be
administrated regularly to maintain high titers (i.e.,
1408
>100 IU of anti-hepatitis B surface antigen [HBs]), perhaps for life; this adds substantial costs to the
transplantation procedure (approximately $36,000 in the first year). The role that antiviral drugs, such as
lamivudine (which is undergoing development and testing), will play in the prevention of reinfection
remains to be determined (see Chapter 68) .
Hepatitis C
Hepatitis C virus (HCV)-related end-stage liver disease has become the most common disease for which
OLT is performed. [2] In 1992, the proportion of recipients with HCV as the primary diagnosis was the
same as for ALD: 19.1%. However, at least 30% of the patients with a diagnosis of ALD also have HCV
infection. [32] HCV infection recurs in more than 90% of patients with pretransplantation infection.
Although recurrent HCV infection seems to have little effect on early (i.e., <5 years) graft survival, the
long-term outcome of graft infection by HCV is unknown. [33] [34] Rapid progression to cirrhosis and graft
failure has been reported. [33]
Acute Liver Failure
The results of OLT for patients with acute liver failure have improved substantially since the mid-1980s.
[35] This in large part reflects the organization of national and regional donor procurement networks,
which allow the rapid identification of potential donors for patients with this rapidly evolving form of
liver disease. The development of prognostic criteria for patients with various forms of fulminant hepatic
failure was also an important advance (see Chapter 80) . O'Grady and associates, [36] who are from the
King's College Hospital in London, developed the most commonly used prognostic scoring system. For
patients with acetaminophen-induced fulminant hepatic failure, the presence of acidemia or the
combination of marked prolongation of prothrombin time and elevation of serum creatinine in patients
with grade III or IV encephalopathy represents an indication for urgent transplantation. For patients with
acute liver failure not caused by acetaminophen, age, etiology, time course of the illness, prothrombin
time, and serum bilirubin concentrations were found to represent prognostic variables. Patients with even
two of these criteria had a less-than-10% chance of survival. These prognostic variables can be assessed
within 1 to 2 hours of the patient's arrival in a hospital and thus allow rapid determination of prognosis
and early referral of these patients to a transplantation center. The 1-year survival rate for patients
undergoing OLT for fulminant hepatic failure has improved; for example, at UCSF, it is currently more
than 90%.


Saunders Company
Figure 83-2 Two-year survival rates for patients in the National Institute for Diabetes and Digestive and
Kidney Diseases liver transplantation database by disease indication, including fulminant hepatic failure
(FHF), cirrhosis caused by chronic hepatitis B (CAH-B) and hepatitis C (CAH-C) virus infection,
autoimmune hepatitis (AIH), alcoholic liver disease (ALD), primary biliary cirrhosis (PBC), primary
sclerosing cholangitis (PSC), extrahepatic biliary atresia (EBA), metabolic liver disease (MET), and
hepatobiliary malignancy (MALIG).


Saunders Company
PATIENT SELECTION SUMMARY

The indications for OLT continue to evolve. For example, the progress in the management of OLT
recipients with HBV infection and the hope that various antiviral treatments will become available
should increase the number of patients undergoing OLT for this disease. HCV and ALD have become the
most common diseases for which OLT is performed in adults. If the opportunity for OLT were extended
more broadly to alcoholic patients, the system would be quickly overwhelmed. Assessment of the
long-term outcomes of such patients will help identify which OLT candidates will do well. Finally,
improvements in cancer therapy and the use of living-related donors will also influence patient selection
and the availability of OLT.


Saunders Company
POST-TRANSPLANTATION MANAGEMENT
Hepatic Allograft Dysfunction
Hepatic allograft dysfunction is the most important complication seen after OLT. The differential
diagnosis includes acute and chronic rejection, infections caused by hepatotropic and nonhepatotropic
viruses, biliary tract disease, drug hepatotoxicity, and ischemic liver injury related to hepatic artery
thrombosis (HAT) or "preservation injury." The various disorders that lead to allograft dysfunction occur
at different times after OLT. Liver biopsy is critical in differentiating these various causes of allograft
dysfunction because many of them have similar clinical and biochemical presentations.
One to Five Days After OLT
HAT is the most severe etiology of early allograft dysfunction (Table 83-5) . The frequency of HAT is
5% to 12% in adults and 12% to 26% in children. [1] Early HAT presents as severe liver injury, with high
transaminase activity and hypoprothrombinemia. Late HAT (i.e., >2 weeks after OLT) usually presents
with biliary strictures, cholangitis, liver abscesses, or development of secondary biliary cirrhosis because
the bile duct is supplied solely by the hepatic artery. In most cases, the development of HAT necessitates
retransplantation.
Preservation/ischemic injury after OLT is clinically manifested by mild-to-moderate cholestasis and
usually resolves over the first 1 to 3 weeks. This functional cholestasis must be distinguished from
rejection, bile duct obstruction, drug toxicity, and sepsis. The liver biopsy shows centrizonal ballooning,
necrosis, and cholestasis (Fig. 83-3) .
TABLE 83-5 -- Differential Diagnosis of Allograft Dysfunction

1- 5 Days After Transplantation
Hepatic artery thrombosis
Preservation/ischemic injury
Hyperacute rejection
5- 30 Days After Transplantation
Acute cellular rejection
Bile leaks
Azathioprine toxicity
Preservation injury
> 30 Days After Transplantation
Acute cellular rejection

Cytomegalovirus hepatitis
Biliary strictures
Hepatitis C
Hepatitis B
Recurrent primary biliary cirrhosis or primary sclerosing cholangitis
Non-B, non-C hepatitis
1409
Figure 83-3 Preservation/ischemic injury after transplantation. The liver biopsy shows ballooning,
necrosis, and cholestasis in hepatocytes around the central vein (arrow).
Although extremely uncommon, hyperacute rejection has been described after OLT. The presence of
immunoglobulin deposits (IgM and IgG) and complement (C1q and C3) is consistent with a humoral
mechanism. [37] Portal and arteriolar thromboses also may be found in the graft. The major risk factor
associated with humoral rejection is ABO incompatibility. Hyperacute rejection usually requires urgent
retransplantation. Acute cellular rejection is uncommon during the first 4 days.
Five to 30 Days After OLT
Allograft dysfunction occurring between 5 and 30 days after OLT is most commonly caused by acute
cellular rejection. [38] The differential diagnosis is shown in Table 83-5 . Biliary tract disease and drug
toxicity (e.g., azathioprine) may also present within this time frame. [39] Although most forms of viral
hepatitis are uncommon during this time period, Herpes simplex hepatitis can occur within the first
month after transplantation. Beyond 2 weeks after OLT, HAT presents usually as biliary tract disease or
occasionally as a hepatic abscess.
The appropriate workup of allograft dysfunction that occurs at 5 to 30 days after OLT is ultrasonography
with Doppler followed by a liver biopsy. If more than 21 days has passed since the patient underwent
OLT, a culture of the liver biopsy for cytomegalovirus (CMV) is also appropriate. Cholangiography
should be performed if the biopsy suggests a biliary etiology.
Acute cellular rejection is one of the most common problems faced after OLT; the incidence ranges from
50% to 70%. [38] It is most commonly seen in the first month after OLT (peak incidence is on day 10).
Rejection is broadly defined as the response of the recipient's immune system to the allograft. The major
targets of hepatic allograft rejection are the epithelial cells of the bile ducts and the endothelium of the
hepatic blood vessels. The diagnosis is usually suspected by abnormal serum transaminase and alkaline
phosphatase activities before clinical symptoms develop (i.e., jaundice or fever). [38] No clinical or

biochemical marker is specific for rejection, and the diagnosis should be confirmed by liver biopsy. The
characteristic histologic triad of acute cellular rejection includes portal or periportal hepatitis,
nonsuppurative cholangitis, and endotheliitis (Fig. 83-4) . Endotheliitis is present in a minority of cases.
The benefit of grading of acute cellular rejection remains debatable because none of the proposed
classification systems reliably predict the response to therapy. [37] Furthermore, the correlation between
histologic severity and the degree of liver dysfunction has been poor.
Acute cellular rejection is treated with increased immunosuppression, most often, additional
corticosteroids. This treatment is successful in approximately 85% of acute rejection episodes. [38]
Patients who experience steroid-resistant rejection are usually treated with the murine monoclonal
antibody OKT3; approximately 75% of patients with steroid-resistant rejection respond to this therapy. In
the case of OKT3 failure, several centers have reported 50% to 80% response rates in response to
tacrolimus rescue therapy.
Another common cause of allograft dysfunction during this period is biliary tract disease. In the early
postoperative period, T tube misplacement is most commonly the cause of biliary obstruction. Later,
strictures, stones, and ampullary dysfunction become more common causes of abnormal liver tests.
Azathioprine can produce hepatotoxicity, which in its earliest stages, resembles an ischemic lesion. [39]
Azathioprine hepatotoxicity is manifested clinically as cholestasis and an increase in serum transaminase
and alkaline phosphatase activities.
More Than 30 Days After OLT
The differential diagnosis of hepatic allograft dysfunction more than 30 days after transplantation differs
from that of
Figure 83-4 Acute cellular rejection. Manifestations of rejection include endotheliitis ( open
arrowheads, inflammatory cells adhering to the vascular endothelium), a portal infiltrate consisting of
lymphocytes, plasma cells and eosinophils, and damage to bile ducts ( solid arrow, duct at the
periphery of the portal tract).
1410
earlier dysfunction (see Table 83-5) . Acute cellular rejection is a less common cause of allograft
dysfunction during this period. In contrast, the peak incidence of CMV hepatitis occurs on day 35 and
must be considered in patients within 3 months after transplantation, particularly if the patient manifests
fever and leukopenia. The biochemical presentation of CMV hepatitis is not different from that of acute
cellular rejection; [40] both present as cholestatic hepatitis. The histologic manifestations of CMV
hepatitis include focal necrosis and neutrophilic microabscesses in the parenchyma of the liver. Biliary
obstruction and chronic rejection also become progressively more common during this period. After 2
months, it is important to consider recurrent disease in the differential diagnosis, particularly recurrent
HCV and HBV. Hepatic artery thrombosis is uncommon during this time and invariably presents as
biliary tract disease. Unusual causes of allograft dysfunction during this time include Epstein-Barr virus
(EBV) hepatitis and post-transplantation lymphoproliferative disorder (PTLD). [41]

In terms of evaluation, liver biopsy is most important. A CMV culture is appropriate if the patient is
within 2 months of transplantation. If there is a suspicion of biliary tract disease based on the liver biopsy
appearance, endoscopic retrograde cholangiography (ERCP), transhepatic cholangiography (PTC), or T
tube cholangiography is indicated. Doppler ultrasonography should be reserved for patients who show
changes consistent with ischemia on biopsy or biliary tree abnormalities.
Late episodes of cellular rejection (i.e., >6 months) often reflect a reduction in immunosuppression. This
reduction may be the result of decreased absorption (e.g., diarrhea) or poor compliance. Ductopenic
rejection reportedly affects approximately 10% to 20% of patients but has been occurring less often. [42] It
rarely occurs during the first 2 months after OLT. Ductopenic rejection is usually preceded by bouts of
cellular rejection; the diagnosis is based on histologic criteria. Ductopenic rejection is defined as a loss of
interlobular and septal bile ducts in at least 50% of portal tracts. [43] The loss of bile ducts is often
accompanied by a foam cell arteriopathy. Ductopenic rejection commonly leads to irreversible graft
failure.
Biliary tract disease during this period usually reflects strictures. [44] [45] [46] The biochemical presentation
is similar to that of acute cellular rejection. The biopsy results include bile ductular proliferation,
cholangitis, and portal tract edema. The diagnosis is made through direct visualization of the biliary
system.
Post-Transplantation Hepatitis
Recurrence of both HBV and HCV is common after OLT (see below). Other viruses that may cause
post-OLT hepatitis include EBV, herpes simplex virus (HSV), and adenovirus. The diagnosis of these
infections is usually made through histologic [47] and serologic examinations. HSV and EBV hepatitis
may respond to treatment with high-dose acyclovir, whereas the treatment of adenovirus hepatitis is
largely supportive.
Biliary Tract Complications
The biliary tree has been described as the "Achilles heel" of OLT. Biliary tract complications include
leaks, strictures, and obstruction secondary to stones and biliary sludge. [44] [46] Leaks are most common
early after OLT and occur in as many as 25% of recipients. Leaks occur at various sites, including the T
tube insertion site, the biliary anastomosis, the cut edge of the liver after reduced-size OLT, or,
occasionally, a previously unrecognized auxiliary duct. [44] Leaks after construction of a
choledochojejunostomy are usually anastomotic.
Bile leaks present with pain and fever with or without abnormal liver tests. Ultrasonography may show
duct dilatation but is most helpful in identifying a biloma. Similarly, computed tomography scanning
may identify an undrained biloma. If a biliary leak is suspected, contrast cholangiography should be
performed. If a T tube is present, visualization of the biliary tree is relatively easy. If a T tube is not
present, ERCP or PTC should be used. [44]
Eighty percent of leaks at the T tube insertion site can be managed before T tube removal by reopening
the T tube. T tube exit site leaks that occur after T tube removal are most commonly managed
endoscopically (see Chapter 61) ; [44] [48] options include placement of a nasobiliary tube for 24 to 48
hours, internal biliary stenting, and sphincterotomy. The use of cholangiography is important in deciding

how to best manage leaks after unintentional T tube removal. In the case of early, accidental T tube
removal, the rent in the biliary tree may be quite large. Under these circumstances, conservative therapy
often is not adequate, and surgical repair is the treatment of choice. In contrast to the success obtained
with nonsurgical management of leaks after choledochocholedochostomy, anastomotic leaks after a
choledochojejunostomy usually require surgical repair. Leaks often produce a biloma, which can be a
source of infection. Regardless of whether the biliary leak is treated surgically or nonsurgically, it is
important to drain the biloma at the same time. This can be accomplished nonsurgically with a
percutaneous catheter.
Three types of strictures have been described after OLT on the basis of their location: anastomotic, hilar,
and diffuse. [44] [45] [46] [49] Strictures may result from a variety of post-transplantation complications. For
example, hepatic artery ischemia, related to chronic allograft rejection or HAT, can occasionally present
as an anastomotic stricture or as diffuse strictures. Thus, determination of hepatic artery patency is
important in the evaluation of all post-OLT patients with a stricture.
The choledochocholedochostomy anastomosis is the most common site of strictures. [44] [46] [49] These
present relatively early, usually within the first 2 to 6 months. The most common presentation is an
asymptomatic rise in liver enzymes, predominantly serum alkaline phosphatase and gamma-glutamyl
transpeptidase activities. Occasionally, patients may present with signs and symptoms of cholangitis; less
often, unsuspected intrahepatic duct dilatation is found by ultrasonography. Some patients present only
with histologic evidence of bile duct obstruction (i.e., bile duct proliferation, pericholangitis, or both). If
the diagnosis is suggested on the basis of clinical or laboratory data, visualization of the biliary tree is
essential. Treatment options for anastomotic strictures include surgical conversion to a Roux-en-Y
choledochojejunostomy or balloon dilatation and stenting by an interventional radiologist or a biliary
endoscopist. [44] Anastomotic strictures are treated initially at UCSF with a period of percutaneous biliary
drainage after balloon dilatation.
The clinical presentation of an anastomotic stricture when the anastomosis is a choledochojejunostomy is
more commonly
1411
cholangitis, with obstructing sludge and stones. In contrast to patients with a

choledochocholedochostomy, the response to nonsurgical management of anastomotic strictures after a
choledochojejunostomy is better, with success rates of 70% to 80%. [46] [49]
Strictures at the bifurcation of the common hepatic duct (Fig. 83-5) are particularly common after OLT.
[44] [46] [49] This may be in part because the blood supply to this area is relatively tenuous. Hilar strictures
therefore may reflect the presence of ischemia. Patients receiving organs from ABO-incompatible donors
often develop biliary tract disease, with the hilum being the most common site of stricturing. Because
surgical access to this part of the biliary tree can be difficult, these strictures are commonly managed by
dilatation and stenting. However, unlike anastomotic strictures, these commonly require prolonged and
repeated treatments. The use of metal mesh stents may improve outcome; however, long-term results
with these endoprostheses have not been reported. [44]
Diffuse bile duct stricturing occurs in four settings: (1) late hepatic artery thrombosis, (2)
ABO-incompatible transplants, (3) prolonged cold ischemia, and (4) recurrence of sclerosing cholangitis.

The long-term outcome of diffuse biliary stricturing with conservative treatment tends to be poor, and
[45]
these patients often require retransplantation.
Vascular Thrombosis
Vascular thromboses occurring after OLT include HAT, portal vein thrombosis, and hepatic vein
thrombosis. [50] As discussed above, if HAT occurs more than 1 month after OLT, stricturing and
obstruction of the biliary tree with bacteremia, cholangitis, and hepatic abscesses may develop;
retransplantation is usually necessary.
Portal vein thrombosis is an unusual complication after OLT ( 2%). Patients usually present with
increasing ascites or variceal bleeding. Surgical evacuation of the clot with postoperative anticoagulation
may be successful if performed early. Splenorenal shunts may palliate late portal vein thromboses.
Hepatic vein thrombosis may recur in transplant patients who had Budd-Chiari syndrome; post-OLT
anticoagulation
Figure 83-5 Stricturing at the bifurcation of the common hepatic duct (arrow).
can prevent this unfortunate complication. Children receiving a reduced-size graft or a graft from a live
donor may develop hepatic venous outflow obstruction, reflecting a problem at the supracaval
anastomosis. The use of balloon dilatation has been effective in this situation.
Recurrent Viral Disease
Hepatitis B
HBV infection recurs in most patients undergoing OLT for chronic HBV in the absence of
immunoprophylaxis with HBIG. [7] [29] Recurrent HBV can progress to cirrhosis or liver failure in periods
as short as 1 year. As mentioned earlier, some patients with recurrent HBV develop the characteristic
histologic lesion fibrosing cholestatic hepatitis. [30] It is characterized by a lack of inflammation,
hepatocyte swelling and disarray, and striking intracellular expression of viral proteins. Once this lesion
develops, graft failure is virtually inevitable.
If immunoprophylaxis fails, no proven therapy exists to prevent graft loss. One study reported modest
success with prostaglandin E. Eleven of 14 patients responded with decreased serum transaminase
activity. Side effects were universal. Although such treatments offer some hope for the treatment of
recurrent HBV, the best strategy remains prevention. Other agents that are under study as treatment for
recurrent HBV include famciclovir and lamivudine.
Hepatitis C
As discussed above, HCV is the major cause of post-OLT hepatitis. [32] HCV infection recurs in most
patients with pretransplantation infection. Short-term follow-up of HCV-infected OLT recipients

suggests that 40% to 70% develop chronic hepatitis, with most showing only mild histologic changes. [32]
[33] HCV RNA levels increase markedly after OLT. [34] However, many patients with high levels of HCV
viremia have no evidence of liver damage. The relationship between the level of viremia and the degree
of hepatic damage remains controversial. Several studies have suggested that recipients reinfected with
genotype 1b have a worse outcome.
Because prophylaxis against recurrent HCV infection is not possible, attention has been focused on
therapy of recurrent hepatitis, usually with interferon-alpha. Two studies have shown that complete
responses (normalization of serum aspartate aminotransferase and alanine aminotransferase) are
relatively infrequent. Serum HCV levels decreased in both responders and nonresponders; however,
histology did not improve, and side effects were common. [51] Furthermore, in some patients, ductopenic
rejection leading to retransplantation has been reported with interferon therapy. These studies suggest
that interferon-alpha has, at best, a transient antiviral effect on OLT recipients with chronic hepatitis C
and carries substantial risk. Additional studies of the use of interferon-alpha as well as other antiviral
agents, such as ribavirin, are under way.
Recurrence of Nonviral Disease
Whether PBC and PSC recur after OLT is controversial. [49] [52] In one study, OLT recipients with the
pretransplantation diagnosis
1412
of PSC were compared with a control group of patients with the same biliary reconstruction who did not
have PSC. [49] Features of biliary obstruction were seen more commonly in patients with PSC. Fibrous
cholangitis and classic fibro-obliterative lesions were seen only in patients with PSC. These data support
the concept that PSC can recur.
Recurrent PBC has also been reported. A recent study reported the long-term follow-up of patients
transplanted for PBC and compared them with patients without PBC or PSC who survived more than 1
year after OLT. Biopsies in both groups were generally normal. [52] Bile duct epithelial damage was seen
in both groups, and there were no differences in liver chemistries. All PBC patients were persistently
positive for antimitochondrial antibody (AMA), although at a lower titer. Thus, this small study was
unable to show that recurrent PBC is an important post-OLT problem. Although PBC likely does recur
after OLT, it is not likely to be a significant problem in more than a small percentage of patients.


Saunders Company
Figure 83-3 Preservation/ischemic injury after transplantation. The liver biopsy shows ballooning,
necrosis, and cholestasis in hepatocytes around the central vein (arrow).


Saunders Company
COMPLICATIONS RELATED TO IMMUNOSUPPRESSION

Cyclosporine/Tacrolimus Toxicity
The two agents that are the cornerstone of most immunosuppressive regimens are cyclosporine and
tacrolimus. [53] [54] [55] Tacrolimus, also known by its experimental name of FK 506, is a macrolide
antibiotic, which, like cyclosporine, inhibits the proliferation of cytotoxic T lymphocytes and the
production of interleukin (IL)-2, IL-3, IL-4, granulocyte-macrophage colony-stimulating factor, tumor
necrosis factor-alpha, and interferon-gammaalpha. [53] [55] Tacrolimus is approximately 100 times as
potent as cyclosporine by weight. These agents are administered in combination with other
immunosuppressive medications, including azathioprine, mycophenolate mofetil, and corticosteroids.
Tacrolimus is often used without azathioprine and may permit doses of prednisone to be reduced or
eliminated. [55] In two large multicenter trials (in Europe and in the United States) in which tacrolimus-
and cyclosporine-based regimens were compared, the incidence of steroid-refractory rejection was
substantially lower in patients treated with tacrolimus. [55] There were no significant differences in patient
and graft survival between the cyclosporine-treated and the tacrolimus-treated patients. The side effect
profiles of cyclosporine and tacrolimus are also similar (Tables 83-6 and 83-7) .
Nephrotoxicity
The most important side effect of these agents is nephrotoxicity. [55] The acute administration of both
agents leads to a decrease in urine output, natriuresis, and glomerular filtration
TABLE 83-6 -- Common Side Effects of Cyclosporine
Renal insufficiency Gingival hyperplasia
Hypertension Hyperuricemia
Central nervous system toxicity Hyperkalemia, type IV renal tubular acidosis
Lymphoma
Hirsutism Hyperlipidemia
TABLE 83-7 -- Common Side Effects of Tacrolimus (FK 506)

Renal insufficiency "Wasting" syndrome
Neurotoxicity Diarrhea
Diabetes Pruritus

rate (GFR). This reflects the acute effects of these agents on intrarenal blood flow and is reversible.
Patients with impaired renal function before transplantation appear to be at an increased risk for
post-transplantation renal dysfunction. Despite the prevailing notion that renal function in patients with
hepatorenal syndrome is normal after successful OLT, many patients with pre-OLT functional renal
failure are at risk for developing post-OLT renal failure and commonly develop long-standing renal
insufficiency.
Acute renal failure requiring hemodialysis occurs in 10% to 20% of patients after OLT. Most often,
intraoperative hypotension in addition to drug toxicity results in renal ischemia and acute tubular
necrosis. Management includes the withdrawal of the offending drug or dose reduction. Survival in
patients requiring dialysis appears to be reduced.
Most patients develop a decrease in GFR (varying from 10% to 70%) after OLT. Chronic cyclosporine or
tacrolimus nephrotoxicity seems to be minimally reversible with discontinuation of drug. In one study,
discontinuation of cyclosporine more than 2 years after OLT in patients with renal insufficiency had no
effect on GFR and precipitated episodes of rejection in one half of the patients who previously had not
rejected. In addition to decreased GFR, both cyclosporine and tacrolimus can produce a type IV renal
tubular acidosis manifested predominantly by hyperkalemia; this responds well to mineralocorticoid
administration.
The managing physician must be aware that various drugs affect cyclosporine and tacrolimus
metabolism. [54] Some drugs increase drug blood concentration and thus may precipitate renal
insufficiency. Drugs that affect cytochrome P-450 enzyme activity and thus increase or decrease both
cyclosporine and tacrolimus drug levels are shown in Table 83-8 .
Hypertension
Between 30% and 85% of liver transplant recipients treated with cyclosporine require antihypertensive
therapy within the first year. [55] The incidence of hypertension with tacrolimus may be less.
Drug-induced hypertension seems to be, in part, dose dependent, and attempts to reduce
cyclosporine/tacrolimus
TABLE 83-8 -- Cyclosporine and Tacrolimus Drug Interactions
DECREASE CYTOCHROME P-450 INCREASE CYTOCHROME P-450
ACTIVITY/INCREASE DRUG LEVELS ACTIVITY/DECREASE DRUG LEVELS
Ketoconazole/fluconazole/itraconazole Phenytoin
Erythromycin/clarithramycin Barbiturates
Cimetidine Rifampin
Corticosteroids Alcohol
Verapamil
1413

and prednisone dosing to the absolute minimum seems reasonable. Calcium channel blockers are the
drug of choice for hypertension after OLT. If another agent is needed, beta-blockers followed by
angiotensin-converting enzyme inhibitors can be added.
Neurotoxicity
Neurotoxicity affects 15% to 20% of patients undergoing OLT. [55] It can be one of the more disabling
side effects of cyclosporine and tacrolimus. Neurologic complications seen after OLT include central
pontine myelinolysis, headache, sleep disturbance, psychosis, encephalopathy, seizures, tremors,
myoclonus, cortical blindness, hemiplegias, spastic quadripareses, speech apraxia, and coma. [56] The
etiology of the more severe forms of neurotoxicity is, in part, demyelination. Central nervous system
toxicity is often associated with dramatic computed tomography or magnetic resonance images of the
brain that vary from minor white matter changes to hemorrhagic infarcts. Temporary discontinuation of
cyclosporine or tacrolimus can result in complete resolution of symptoms and the radiologic changes.
Although most patients with neurotoxicity completely recover, some are left with residual gait or visual
or speech disturbances. In a recent study, the incidence of moderate and severe neurotoxicity in patients
transplanted for HCV was significantly higher in tacrolimus-treated patients than in cyclosporine-treated
patients. Treatment for post-OLT seizures is usually needed for only a limited period (i.e., 3 months).
Minor neurologic side effects, including headaches and tremors, are common after OLT. [55] Headaches
occur with both cyclosporine and tacrolimus and can be quite troublesome for some patients. Because the
headaches tend to be a vascular type, they may respond to migraine regimens (i.e., beta-blockers).
Infections
One of the most common clinical problems after OLT is fever, particularly during the early
post-transplantation period. Appropriate cultures include blood, urine, wound, and bile; if it has been
more than 3 weeks since the transplantation, a CMV culture should also be performed. Chest
radiography, urinalysis, and T tube cholangiography, if available, are also appropriate studies. If it is
relatively soon after OLT, particularly in patients with a spiking fever and a high white blood cell count,
a computed tomography scan is indicated to rule out intra-abdominal abscess.
Infections occurring within 1 month after OLT tend to be bacterial and may be caused by a technical
complication such as a bile leak or contamination of the wound. The organisms most commonly
responsible are Gram-positive aerobic bacteria. In the period of 1 to 6 months after OLT, opportunistic
organisms such as CMV, EBV, Pneumocystis carinii, Nocardia, and Aspergillus must be considered.
CMV infection is the most common opportunistic infection seen after transplantation, with an overall
incidence of 25% to 85% in liver transplant recipients, depending on how infection is defined. [40] Certain
forms, such as CMV pneumonitis, can be fatal; CMV infection may also increase the risk of both
superinfection with other organisms and chronic rejection. Patients who are at an increased risk of
developing CMV disease include seronegative recipients who receive an organ from a CMV-seropositive
donor, patients who receive OKT3 therapy for rejection, and patients undergoing retransplantation. CMV
infection occurs most commonly 30 to 50 days after OLT. The diagnosis of CMV disease is made by
culture (tissue or blood not urine). The use of the "shell vial" technique, which involves the use of tissue
culture and indirect immunofluorescence to detect CMV, yields results within 48 hours. CMV infection
in CMV-seronegative recipients is often symptomatic, with major organ involvement, such as hepatitis or

pneumonitis. Patients who develop CMV reactivation, in contrast, are often minimally symptomatic and
less frequently experience direct organ involvement.
The diagnosis of organ involvement is based on histology and culture. The typical cytopathic changes
include the presence of typical intranuclear inclusions bodies. Intravenous ganciclovir eradicates disease
in approximately 85% of patients. The usual dose is 5 mg/kg administered intravenously every 12 hours,
with adjustment for renal dysfunction. Resistance to this agent has been reported but seems to be rare in
OLT recipients. In patients not responding to ganciclovir, CMV hyperimmune globulin or the antiviral
foscarnet may be used. Reduced immunosuppression in conjunction with antiviral therapy is also used in
some instances. Prophylaxis with ganciclovir, with or without acyclovir or CMV hyperimmune globulin,
appears to decrease the incidence and severity of post-OLT CMV disease and is commonly used in
patients considered to be at risk. [57]
Patients with fulminant hepatic failure, patients receiving increased immunosuppression, and patients
undergoing retransplantation are more susceptible to fungal infections. [58] The incidence declines with
time after discharge from the hospital. Superficial infections involving the skin or mouth may be treated
with topical antifungals such as nystatin (Mycostatin) or clotrimazole. Candida urinary tract infections
often respond to oral fluconazole or amphotericin B bladder irrigation. Invasive infections, however,
must be treated with systemic amphotericin B, often for prolonged periods.
Aspergillus is an important opportunistic infection after OLT, commonly involving the lungs but
occasionally involving the brain. In contrast to patients with intact immune systems, immunosuppressed
patients often develop a diffuse, patchy infiltrate on chest radiography rather than a fungus ball. It is a
particular problem in patients who have undergone OLT for fulminant hepatic failure. Treatment is
generally unsatisfactory, and most patients ultimately die. Prolonged courses of amphotericin are usually
required. Itraconazole also has activity against Aspergillus.
Infections with Cryptococcus, Mucor, or Rhizopus species may also occur but are less common.
Treatment involves the combination of systemic amphotericin, surgical debridement of infected tissue
where possible, and reduced immunosuppression.
Pneumocystis carinii infections usually occur 2 to 6 months after OLT; the incidence is markedly
reduced by the use of prophylactic trimethoprim-sulfamethoxazole, inhaled pentamidine, or dapsone.
Mycobacterium tuberculosis infection may develop or reactivate after OLT and often presents atypically.
In one study, the onset varied from 2 to 56 months after OLT. Most patients
1414
presented with fever; other presentations included pulmonary infiltrates, meningitis, miliary lesions of
the peritoneum, and necrotizing intestinal granulomas. Acid-fast bacilli smears were positive. All isolates
were sensitive to the usual antimycobacterial agents, and most patients recovered with therapy.
Beyond 6 months, community-acquired infections predominate. Pneumococcus is most common.
Legionella is another important infection to consider.
It is important to focus the evaluation on likely sites and sources of infection. If the patient has received
excessive immunosuppression, the risk of opportunistic infection, such as tuberculosis, cryptococcosis,

and aspergillosis, is markedly increased.
Other Post-Transplantation Complications
Hyperglycemia
Hyperglycemia is a common complication after OLT. It occurs most commonly during the first month; it
is related to high doses of corticosteroids and is likely enhanced by both cyclosporine and tacrolimus.
Steroid tapering is the key to management of early post-OLT hyperglycemia. Only half of patients
require long-term insulin therapy after steroid tapering. Diabetes is more commonly seen with tacrolimus
therapy, with an incidence at 1 year after OLT of approximately 15%. The diabetes seen with tacrolimus
also more commonly requires insulin therapy, but it may respond early to tacrolimus dose reduction.
Bone Disease
Bone loss is maximal during the first 3 months after OLT because of the effects of high-dose
corticosteroid therapy, bed rest, and, perhaps, increased cytokine levels. [59] In patients with preexisting
osteopenia, calcium and vitamin D supplementation is often used after OLT but is of unproved benefit.
The role of pre- or post-OLT bisphosphonate or calcitonin administration in the prevention of bone
resorption is under investigation. Osteonecrosis, particularly of the hip, can also occur and usually
requires hip replacement.
Hyperlipidemia
As many as 40% of liver transplant recipients develop hyperlipidemia after OLT. Tacrolimus appears to
be associated with lower serum cholesterol and low-density lipoprotein levels compared with
cyclosporine. [55] 3-Hydroxy-3-methylglutaryl coenzyme A reductase inhibitors can be effective in
patients who require treatment. Lovastatin in combination with cyclosporine has been reported to
produce rhabdomyolysis.
Obesity
Obesity is common after OLT in adults. The average gain in body weight is approximately 30% within
the first 10 months after OLT. Early intervention is important, and formal weight reduction programs are
most successful.
Post-Transplantation Lymphoproliferative Disorder
PTLD occurs in 1% to 3% of liver transplant recipients. [41] It can occur as early as 1 month or as late as
10 years after OLT. The clinical presentation comprises a spectrum that includes an infectious
mononucleosis-like illness with fevers and lymphadenopathy, hepatitis, weight loss, and symptoms
suggestive of lymphoma. The development of PTLD is thought to reflect the unrestricted proliferation of
B cells stimulated by EBV infection. Both polyclonal and monoclonal B cell proliferation have been
described. The major risk factors for the development of this syndrome are the exposure to greater
amounts of immunosuppression and the lack of previous EBV exposure (i.e., young children). The
diagnosis is made histologically, but clinical suspicion is important. PTLD may respond to a lowered
level of immunosuppression and high doses of intravenous acyclovir or ganciclovir.

Hemolytic Anemia
Hemolytic anemia is most commonly seen in patients who receive transplants from non-ABO-identical
donors. It is usually self-limited.
Pulmonary
Pleural effusions are extremely common after OLT but are usually clinically insignificant. They typically
occur on the right side and resolve within 1 month after surgery. [5] Atelectasis is also common and is
usually right sided.
Pulmonary infiltrates occur in 12% to 50% of patients, with approximately half of these being infectious.
Early pulmonary infections are more commonly caused by bacteria, with Gram-negative organisms
predominating. Opportunistic infections tend to occur later in the postoperative course; causes include P.
carinii, Cryptococcus, Aspergillus, and Candida. CMV is the most common viral pathogen.
Bronchoscopy with bronchoalveolar lavage and, rarely, open-lung biopsy may be required to make a
diagnosis.
Hypoxemia reflecting hepatopulmonary syndrome usually takes weeks to months to resolve after
successful OLT. [5]
Gastrointestinal
Diarrhea is quite common after OLT, particularly during the initial hospitalization. [55] The evaluation
should be focused on ruling out a bacterial cause through examination of the stool for white blood cells,
culture for pathogens, and, in particular, assay for Clostridium difficile toxin. If white blood cells are
present and the C. difficile toxin assay is negative, a sigmoidoscopy is indicated. Two diagnoses that are
always worth considering include graft-versus-host disease and inflammatory bowel disease.
Graft-versus-host disease commonly presents with fever and leukopenia, followed soon by diarrhea.
Because a skin rash also is often present, a skin biopsy is usually done first; however, a sigmoidoscopic
biopsy can also be extremely helpful in making this diagnosis. Inflammatory bowel disease in patients
transplanted for sclerosing cholangitis tends to be mild after OLT. However, it is
1415
extremely important to screen patients with ulcerative colitis for colon cancer. Colon cancer is the most
common cause of death in patients transplanted for PSC who survive for more than 1 year. Some have
even recommended prophylactic colectomy in all recipients with a history of colitis for more than 20
years. Because tacrolimus is a macrolide, it also can cause diarrhea. [55] Other gastrointestinal side effects
seen with tacrolimus include nausea, vomiting, and a wasting-like syndrome.
Dermatology
Dermatologic side effects seen after OLT include hirsutism and gingival hyperplasia. [55] These are
cyclosporine related and not seen with tacrolimus. In contrast, hair loss and pruritus are more commonly
seen with tacrolimus.
Skin cancers are the most common malignancy seen after OLT. Prevention is most important through the

use of a sun blocker. Excision is the treatment of choice. Occasionally, these squamous cell cancers can
be invasive.


Saunders Company
POST-TRANSPLANTATION SUMMARY
The liver transplant recipient represents a special clinical challenge. However, most clinical problems
occurring after OLT have standard differentials and protocols for evaluation. The care of patients after
OLT is increasingly being handled by community physicians, making effective communication with the
transplant center especially important.


Saunders Company
REFERENCES
1. Starzl, T. E., Demetris, A. J., and Van Thiel, D. Liver transplantation. N. Engl. J. Med. 329:1014, 1092, 1989.
2. Belle, S. H., Kimberly, C., and Detre, KM. Trends in liver transplantation in the United States. In Terasaki, P., and
Cecka, M. (eds.), Clinical Transplant 1993. Los Angeles, University of California, Los Angeles, Tissue Typing
Laboratory, 1993, pp. 19-35.
3. Consensus statement on indications for liver transplantation. Hepatology 20:63S, 1994.
4. Rubin, R. H., Jenkins, R. L., Shaw, B. W. Jr., et al. The acquired immunodeficiency syndrome and transplantation.
Transplantation 44:1, 1987.
5. Krowka, M. J., and Cortese, D. A. Pulmonary aspects of liver disease and liver transplantation. Clin. Chest Med. 10:593,
1989.
6. Blaise, M., Pateron, D., Trinchet, J. C., et al. Systemic antibiotic therapy prevents bacterial infection in cirrhotic patients
with gastrointestinal hemorrhage. Hepatology 20:34, 1994.
7. Samuel, D., Bismuth, A., Mathieu, D., et al. Passive immunoprophylaxis after liver transplantation in HBsAg-positive
patients. Lancet 1:813, 1991.
8. Somberg, K. A., Lake, J. R., Tomlanovich, S. J., et al. Transjugular intrahepatic portosystemic stent shunts for refractory
ascites: Assessment of clinical and hemoral response in renal function. Hepatology 21:709, 1995.
9. Ring, E. J., Lake, J. R., Roberts, J. P., et al. Using transjugular intrahepatic portosystemic shunts to control variceal
bleeding before liver transplantation. Ann. Intern. Med. 116:304, 1992.
10. Brems,J. J., Hiatt, J. R., Klein, A. S., et al. Effect of a prior portasystemic shunt on subsequent liver transplantation.
Ann. Surg. 209:51, 1989.
11. Poupon,
R. E., Poupon, R., and Balkau, B. Ursodiol for the long-term treatment of primary biliary cirrhosis: The
UDCA-PBC Study Group. N. Engl. J. Med. 330:1342, 1994.
12. Porayko, M. K., Wiesner, R. H., Hay, J. E., et al. Bone disease in liver transplant recipients: Incidence, timing, and risk
factors. Transplant. Proc. 23:1462, 1991.
13. Wood, R. P., Ellis, D., and Starzl, T. E. The reversal of the hepatorenal syndrome in four pediatric patients following
successful orthotopic liver transplantation. Ann. Surg. 205:415, 1987.
14. Tito,
L., Rimola, A., Gines, P., et al. Recurrence of spontaneous bacterial peritonitis in cirrhosis: Frequency and
predictive factors. Hepatology 8:27, 1988.
15. Dickson,
E. R., Murtaugh, P. A., Wiesner, R. H., et al. Primary sclerosing cholangitis: Refinement and validation of
survival models. Gastroenterology 103:1893, 1992.

16. Wiesner, R. H., Porayko, M. K., Dickson, E. R., et al. Selection and timing of liver transplantation in primary biliary
cirrhosis and primary sclerosing cholangitis. Hepatology 16:1290, 1992.
17. Shapiro,
J. M., Smith, H., and Schaffner, F. Serum bilirubin: A prognostic factor in primary biliary cirrhosis. Gut
20:137, 1979.
18. Evans,R. W., Manninen, D. L., and Dong, F. B. An economic analysis of liver transplantation: Costs, insurance
coverage, and reimbursement. Gastroenterol. Clin. North Am. 22:451, 1993.
19. Vaillant,
G. E. What can long-term follow-up teach us about relapse and prevention of relapse in addiction? Br. J.
Addict. 83:1147, 1988.
20. Bird, G. L., O'Grady, J. G., Harvey, F. A., et al. Liver transplantation in patients with alcoholic cirrhosis: Selection
criteria and rates of survival and relapse. BMJ 301:15, 1990.
21. Lucey,M. R., Merion, R. M., Henley, K. S., et al. Selection for and outcome of liver transplantation in alcoholic liver
disease. Gastroenterology 102:1736, 1992.
22. Osorio, R. W., Ascher, N. L., Avery, M., et al. Predicting recidivism after orthotopic liver transplantation for alcoholic
liver disease. Hepatology 20:105, 1994.
23. Beresford, T. P., Schwartz, J., Wilson, D., et al. The short-term psychological health of alcoholic and non-alcoholic
liver transplant recipients. Alcohol Clin. Exp. Res. 16:996, 1992.
24. Iwatsuki,
S., Starzl, T. E., Sheahan, D. G., et al. Hepatic resection versus transplantation for hepatocellular carcinoma.
Ann. Surg. 214:221, 1991.
25. Olthoff,
K. M., Millis, J. M., Rosove, M. H., et al. Is liver transplantation justified for the treatment of hepatic
malignancies? Arch. Surg. 125:1261, 1990.
26. Ringe,B., Pichlmayr, R., Wittekind, C., et al. Surgical treatment of hepatocellular carcinoma: Experience with liver
resection and transplantation in 198 patients. World J. Surg. 15:270, 1991.
27. Venook, A. P., Ferrell, L. D., Roberts, J. P., et al. Liver transplantation for hepatocellular carcinoma: Results with
preoperative chemoembolization. Liver Transplant. Surg. 1:242, 1995.
28. Stone,
M. J., Klintmalm, G. B., Polter, D., et al. Neoadjuvant chemotherapy and liver transplantation for hepatocellular
carcinoma: A pilot study in 20 patients. Gastroenterology 104:196, 1993.
29. Todo, S., Demetris, A. J., Van Thiel, D., et al. Orthotopic liver transplantation for patients with hepatitis B virus-related
liver disease. Hepatology 13:619, 1991.
30. Davies,S. E., Portmann, B. C., O'Grady, J. G., et al. Hepatic histological findings after transplantation for chronic
hepatitis B virus infection, including a unique pattern of fibrosing cholestatic hepatitis. Hepatology 13:150, 1991.
31. Samuel, D., Muller, R., Alexander, G., et al. Liver transplantation in European patients with the hepatitis B surface
antigen. N. Engl. J. Med. 329:1842, 1993.
32. Wright, T. L., Donegan, E., Hsu, H. H., et al. Recurrent and acquired hepatitis C viral infection in liver transplant
recipients. Gastroenterology 103:317, 1992.

33. Feray,
C., Gigou, M., Samuel, D., et al. The course of hepatitis C virus infection after liver transplantation. Hepatology
20:1137, 1994.
34. Chazouilleres,
O., Kim, M., Combs, C., et al. Quantitation of hepatitis C virus RNA in liver transplant recipients.
35. Bismuth,
H., Samuel, D., Gugenheim, J., et al. Emergency liver transplantation for fulminant hepatitis. Ann. Intern.
Med. 107:337, 1987.
36. O'Grady,J. G., Alexander, G. J., Hayllar, K. M., et al. Early indicators of prognosis in fulminant hepatic failure.
37. Terminologyof chronic hepatitis, hepatic allograft rejection, and nodular lesions of the liver: Summary of
recommendations developed by an international working party, supported by the World Congresses of Gastroenterology,
Los Angeles, 1994. Am. J. Gastroenterol. 89:S177, 1994.
38. Wiesner, R. H., Ludwig, J., Krom, R. A., et al. Hepatic allograft rejection: New developments in terminology,
diagnosis, prevention, and treatment. Mayo Clin. Proc. 68:69, 1993.
39.
1416
Sterneck, M., Wiesner, R., Ascher, N., et al. Azathioprine hepatotoxicity after liver transplantation. Hepatology 14:806,
1991.
40. Stratta,
R. J., Shaeffer, M. S., Markin, R. S., et al. Cytomegalovirus infection and disease after liver transplantation: An
overview. Dig. Dis. Sci. 37:673, 1992.
41. Starzl,
T. E., Nalesnik, M. A., Porter, K. A., et al. Reversibility of lymphomas and lympho-proliferative lesions
developing under cyclosporin-steroid therapy. Lancet 1:583, 1984.
42. Wiesner,R. H., Ludwig, J., van Hoek, B., et al. Current concepts in cell-mediated hepatic allograft rejection leading to
ductopenia and liver failure. Hepatology 14:721, 1991.
43. Freese,
D. K., Snover, D. C., Sharp, H. L., et al. Chronic rejection after liver transplantation: A study of clinical,
histopathological and immunological features. Hepatology 13:882, 1991.
44. LaBerge,J. M., and Ostoff, J. W. Nonoperative management of biliary tract complications after liver transplantation.
Semin. Gastrointest. Dis. 4:170, 1993.
45. Li,
S., Stratta, R. J., Langnus, A. N., et al. Diffuse biliary tract injury after orthotopic liver transplantation. Am. J. Surg.
164:536, 1992.
46. Stratta,
R. J., Wood, R. P., Langnas, A. N., et al. Diagnosis and treatment of biliary tract complications after orthotopic
liver transplantation. Surgery 106:675, 1989.
47. Ferrell,
L. D., Wright, T. L., Roberts, J., et al. Hepatitis C viral infection in liver transplant recipients. Hepatology
16:865, 1992.
48. Osorio,R. W., Freise, C. E., Stock, P. G., et al. Nonoperative management of biliary leaks after orthotopic liver
transplantation. Transplantation 55:1074, 1993.

49. Greif,
F., Bronsther, O. L., Van Thiel, D. H., et al. The incidence, timing, and management of biliary tract
complications after orthotopic liver transplantation. Ann. Surg. 219:40, 1994.
50. Lebeau,
G., Yanaga, K., Marsh, J. W., et al. Analysis of surgical complications after 397 hepatic transplantations. Surg.
Gynecol. Obstet. 170:317, 1990.
51. Wright,T. L., Combs, C., Kim, M., et al. Interferon-alpha therapy for hepatitis C virus infection after liver
transplantation. Hepatology 20:773, 1994.
52. Gouw, A. S., Haagsma, E. B., Manns, M., et al. Is there recurrence of primary biliary cirrhosis after liver
transplantation? A clinicopathologic study in long-term survivors. J. Hepatol. 20:500, 1994.
53. Bumgardner, G. L., and Roberts, J. P. New immunosuppressive agents. Gastroenterol. Clin. North Am. 22:421, 1993.
54. Kahan,B. D. Cyclosporine nephrotoxicity: Pathogenesis, prophylaxis, therapy, and prognosis. Am. J. Kidney Dis.
85:323, 1986.
55. The
U.S. Multicenter FK 506 Liver Study Group. A comparison of tacrolimus (FK 506) and cyclosporine for
immunosuppression in liver transplantation. N. Engl. J. Med. 331:1110, 1994.
56. DeGroen, P. C., Aksamit, A. J., Rakela, J., et al. Central nervous system toxicity after liver transplantation: The role of
cyclosporine and cholesterol. N. Engl. J. Med. 317:861, 1987.
57. Freise,C. E., Pons, V., Lake, J., et al. Comparison of three regimens for cytomegalovirus prophylaxis in 147 liver
transplant recipients. Transplant. Proc. 23:1498, 1991.
58. Castaldo,P., Stratta, R. J., Wood, R. P., et al. Fungal disease in liver transplant recipients: An analysis of risk factors.
Transplant. Proc. 23:1517, 1991.
59. Hay,
J. E., Lindor, K. D., Wiesner, R. H., et al. The metabolic bone disease of primary sclerosing cholangitis.


April 22, 2000
Conde Petra
Sabiston: Textbook of Surgery, 15th ed., Copyright

1997 W. B. Saunders Company
20 - TRANSPLANTATION
VIII - LIVER TRANSPLANTATION Sabiston: Textbook of Surgery, 15th ed., Copyright 1997 W. B.
Saunders Company
INDICATIONS FOR
TRANSPLANTATION
Cholestatic Liver Diseases. VIII - LIVER TRANSPLANTATION

Alcoholic Liver Disease.
Pierre A. Clavien M.D., Ph.D.
Hepatitis C and Cryptogenic Allan D. Kirk M.D., Ph.D.
Cirrhosis.
Metabolic Diseases. INDICATIONS FOR TRANSPLANTATION
The specific indications for liver transplantation have become

Fulminant Hepatic Failure.
more standardized (Tables 20-15 and 20-16) owing to the better
Chronic Hepatitis B.
knowledge of the natural history of a number of liver diseases
Intrahepatic Malignancy. and concomitant improvement in short- and long-term results of
orthotopic liver transplantation (OLT). [2] [6] [12] As with most
Pediatric Indications.
aspects of this topic, however, significant controversy remains
SPECIFIC PATIENT SELECTION about several disease states leading to end-stage liver failure.
AND PREOPERATIVE
CONSIDERATIONS Cholestatic Liver Diseases.
Patient Selection. Cholestatic diseases, including primary biliary cirrhosis (PBC),

secondary biliary cirrhosis, and primary sclerosing cholangitis
Preoperative Preparation.
(PSC), are the diseases most successfully treated by liver
SELECTION OF DONORS transplantation. [7] [25] Operative
OPERATION ON THE DONOR TABLE 20-15 -- Adult Indications for Orthotopic Liver
Transplantations
IMMEDIATE PREOPERATIVE
Data from Belle, S. H., Beringer, K. C., and Detre, K. M.:
MANAGEMENT
Trends in liver transplantation in the United States. In Terasaki,
OPERATION ON THE P. I., and Cecka, J. M. (Eds.): Clinical Transplants. Los
RECIPIENT Angeles, UCLA Tissue Typing Laboratory, 1993.
POSTOPERATIVE Percentage
MANAGEMENT Before Percent
1991 1992
IMMUNOLOGIC MANAGEMENT Indication n=4833 n=2315
HLA Typing.
Primary cholestatic liver disease 23.7 19.1
Crossmatch and ABO
Matching. Other cirrhosis 55.9 65.2
Acute Rejection. Alcoholic 17.5 21.1
Chronic Rejection. Hepatitis C/ non-A, non-B 14.7 20.3
Immunosuppressive Autoimmune and other 16.5 18.7

Pharmacology. Hepatitis B 7.2 5.1
Antiviral Immunity and Fulminant hepatic failure 7.0 5.5
Immunosuppression.
Metabolic diseases 4.2 4.1
OUTCOMES
Malignant neoplasms 6.1 3.4
Medical Perspective.
Benign neoplasms 0.5 0.5
Patient Perspective.
Biliary atresia 0.6 0.4
Payer's Perspective.
Miscellaneous 2.0 1.8
EMERGING TECHNIQUES
Split and Reduced-Size

Transplantation.
TABLE 20-16 -- Pediatric Indications for Orthotopic Liver
Living-Related Transplantation. Transplantation
Heterotopic Liver Data from Belle, S. H., Beringer, K. C., and Detre, K. M.:
Transplantation. Trends in liver transplantation in the United States. In Terasaki,
P. I., and Cecka, J. M. (Eds.): Clinical Transplants. Los
ALTERNATIVE THERAPIES Angeles, UCLA Tissue Typing Laboratory, 1993.
Xenogeneic Support. Percentage
Before Percent
Bioartificial Liver.
1991 1992
SUMMARY Indication n=1117 n=394
Bibliography Biliary atresia 56.8 51.0
REFERENCES Fulminant hepatic failure 9.4 14.0

Cirrhosis 7.8 7.4
Hepatitis B 0.4 0.5
Hepatitis C/ non-A, non-B 1.6 1.0
Alcoholic 0.0 0.0
Autoimmune and other 5.8 5.9


survival is greater than 90%, and 5-year survival is

approximately 80%. [6] This rate is due mostly to the slow onset
of illness, which allows for preoperative preparation of the
recipient and appropriate timing of the procedure, as well as
absence of recurrence of the disease in the graft. As a result of
this success, cholestatic disease is one of the most frequent
indications for transplantation, accounting for 22% of all
transplants performed in the United States. Clearly, the timing
of operation is critical, and a substantial literature has addressed
the rate of disease progression.
The indications for transplantation in patients with PSC are
based on refractory symptoms resulting from recurrent
cholangitis or impaired synthetic function. The risk of
cholangiocarcinoma, approximately 15% in these patients, is an
462
additional relative indication for OLT. A current trend toward

earlier intervention with liver transplantation in PSC is
underway, with retrospective comparisons of transplant versus
nontransplant therapy weighing heavily in favor of the
transplant approach. [21] Clearly, however, some patients,
particularly those with predominantly extrahepatic biliary
strictures do well with conventional bypass [42] or endoscopic
procedures. [30] Thus, timing remains a matter of clinical
judgment, with most patients benefiting from
allotransplantation.
Several models of survival have been established to predict
outcome in patients with PBC. [7] [27] All prognostic scores are
based in part on the rise in serum bilirubin. In general,
transplantation is recommended when the bilirubin level
exceeds 15 mg./100 ml., although earlier transplantation for
severe associated symptoms such as pruritus or fatigue,
refractory ascites, or variceal bleeding is appropriate. Although
close monitoring of asymptomatic patients is reasonable, it is
important not to delay intervention until the clinical status of the
patient deteriorates, because OLT is the only definitive
treatment. [7]

Nineteen percent of all liver transplants performed in the United

States have been for alcoholic cirrhosis. [6] Although the
procedure was originally avoided because of fears of high
postoperative recidivism and a tendency toward medical
noncompliance, the national trend has been to transplant in this
setting more frequently, and in 1993 it was the most common
indication reported by the United Network for Organ Sharing
(UNOS) registry. Experience has demonstrated that graft and
patient survival are not significantly different from those of
other favorable indications and that disease recurrence (return to
heavy alcohol use) is approximately 12%. [29] Given these
results, combined with the extraordinary prevalence of
alcohol-related liver disease compared with other transplantable
diseases (36,000 deaths per year), transplantation for alcoholic
cirrhosis could easily exhaust the already scarce supply of
donor organs. Selection is clearly needed to exclude
comorbidities, alcohol-associated organ failure such as
cardiomyopathy, and patients with continued dependence on
alcohol. Even so, the need far outpaces the supply. Considerable
philosophical debate addresses the use of scarce resources for
individuals with self-inflicted disease. At this point, the decision
to transplant should be based on a thorough preoperative
medical evaluation supplemented by involvement in alcoholic
rehabilitation and abstinence for at least 6 months. Patients
presenting too sick to drink must be evaluated individually with
an estimate of recidivism made to optimize long-term results.
Palliative procedures are appropriate when possible. The recent
development of the transjugular intrahepatic portosystemic
shunt (TIPS) procedure has allowed many of these patients to
be stabilized without extensive surgical intervention, for more
thorough evaluation in a nonemergent setting.
Hepatitis C and Cryptogenic Cirrhosis.
Another rapidly increasing indication for OLT is hepatitis

caused by the hepatitis C virus (HCV). This is frequently
discussed in tandem with cryptogenic cirrhosis. Transplantation
is pursued for symptomatic infection only. Since the discovery
of reliable methods for detecting HCV, this pathogen has been
identified as the etiologic agent for liver failure in an increasing
number of recipients. [5] In 1996, it was associated with more
than 25% of transplants in most centers. Cryptogenic disease
accounted for an additional 12%. Excellent short-term results
have been achieved with a 3-year survival of 70% (63%
nonretransplant survival). These results have been tempered,
however, by reinfection at almost 90%, with active recurrent

hepatitis at 50%. Recurrent hepatitis has led to few deaths,
although a fourth of the reinfected patients have converted to a
chronic state. More deaths from recurrent disease can be
expected in this group, and longer follow-up is required to
determine whether results will remain acceptable. Disease
recurrence has also been seen in cryptogenic cases, again
suggesting a viral infection. Several treatments have been
studied for HCV infection, with the most promising being
parenteral therapy with interferon alfa.
Metabolic Diseases.
Several metabolic diseases have been successfully cured with

OLT, including alpha1 -antitrypsin deficiency, Wilson's disease,
hemochromatosis, Crigler-Najjar syndrome, tyrosinemia,
primary hyperoxaluria, and familial homozygous
hypercholesterolemia. Survival is uniformly excellent. The
recipient hepatectomy is usually simple, and the preoperative
state of the patient is stable. Timing of transplantation should be
such that secondary effects of the disease are not extensive.
In the absence of pre-existing liver disease, rapid loss of hepatic

function culminating in jaundice and coma is termed fulminant
hepatic failure (FHF). The etiologic agent can vary. The time
course relates inversely to the prognosis, with those progressing
to coma in less than 2 weeks having a 36% survival and those
progressing in 2 to 8 weeks having a 7% survival. Despite the
dramatic rate of progression of disease, these patients have very
acceptable results from OLT (about 60% 5-year survival). [6] [37]
[56] Predictably, survival after OLT improves with improved
health of the patient at the time of transplant. Survival for all

etiologies of FHF has doubled in the past decade, primarily due
to great strides in supportive intensive care. Unlike chronic viral
diseases, recurrence of disease for viral causes of FHF is
unusual.
Care of patients with FHF is complex, and OLT should be
considered early. Thus, prompt transfer to a liver transplantation
center is critical. In patients with rapid deterioration, supportive
care (e.g., hyperventilation, diuresis) should be supplemented
with monitoring of intracranial pressure (ICP) for best results.
Subdural ICP monitors have obviated the need for
intraventricular devices and have thus reduced the bleeding
risks of this approach. Perfusion pressures (mean arterial

pressure minus ICP) of less than 40 mm. Hg, especially when

combined with pupillary fixation, suggest that irreversible brain
injury has occurred, and OLT should be avoided. Patient
decompensation can be rapid and unpredictable. In the absence
of a suitable donor, several aggressive maneuvers can
successfully delay the onset of brain death from intracranial
pressure. These include hepatectomy with end-to-side
portacaval shunt, xenogeneic ex vivo perfusion, and bioartificial
hepatic support (see later). Donor criteria may be liberalized to
achieve timely hepatic replacement, and ABO incompatibility is
acceptable.
Transplantation for chronic hepatitis B virus (HBV) cirrhosis

remains a controversial topic. [39] [55] The discouraging
reinfection rate of over 80% of recipients associated with a high
rate of clinical hepatitis recurrence (60% at 1 year) and high
related mortality (30% at 1 year and 52% at 5 years) raises
questions about the procedure. [6] [14] Although generally
contraindicated, transplantation may be appropriate in certain
settings. Recent intense investigation in this area has identified
several factors affecting reinfection. In general, the state of viral
replication at the time of transplantation is critical in
establishing reinfection. Identification of active viral replication
(serum HBV DNA and HBeAg detection) at the time of
transplantation clearly worsens the prognosis.
Immunoprophylaxis with anti-HBV antibodies improves
outcome for replication-negative patients, but long-term results
for patients so treated remain to be seen. In general, OLT in the
presence of HBV infection should be reserved for patients
enrolled in specific viral prophylaxis protocols.
Intrahepatic Malignancy.
Predictably, transplantation for primary and metastatic cancer

has been associated with a
463
high recurrence of tumor and a poor 5-year survival. [6] [40]

Thus, there is little indication for OLT in a patient with known
malignancy. Liver transplantation is no longer an accepted
therapy for hepatic metastasis except perhaps in some patients
with rare neuroendocrine tumors. [19] Today, liver
transplantation is usually limited to patients with nonresectable
hepatoma fulfilling the following criteria: asymptomatic

hepatoma (i.e., not associated with recent weight loss, ascites,

or constitutional symptoms); tumor less than 5 cm. in diameter;
and fewer than or equal to three intrahepatic tumors. If feasible,
liver resection should always be considered first. Extensive
evaluation is required to exclude the presence of extrahepatic
spread. Transplantation for cholangiocarcinoma is even more
controversial and should probably be performed only under the
guidance of specific protocols. The role of adjuvant
chemotherapy or chemoembolization in patients undergoing
transplantation for cancer remains unknown.
The most common indication for liver transplantation in

children is biliary atresia (see Table 20-16) . This diagnosis
accounts for 55% of the pediatric recipients in the UNOS
registry. [6] [41] Although the creation of a portoenterostomy
(Kasai procedure) remains the standard initial treatment,
long-term survival without eventual hepatic failure is
uncommon. Five-year survival remains under 50%. For this
reason, many have proposed that early intervention with OLT
be considered. The results of liver transplantation in infants
have been somewhat better, with 5-year patient survival of
64%. Thus, a reasonable course of action appears to be neonatal
nontransplant surgical intervention, with transplantation
reserved for those children developing hepatic insufficiency
despite a Kasai procedure. [38] The interim growth improves the
donor pool substantially and decreases the technical difficulties
inherent in the management of small children. Five-year
survival improves to 74% at 3 to 5 years and 79% from 5 to 15
years. However, multiple reoperations and revisions after an
initial portoenterostomy should be avoided because they rarely
provide long-term disease-free survival and substantially hinder
efforts to transplant the patient. The remaining usual indications
for pediatric transplantation also occur in the adult population,
with 13% performed for metabolic diseases (see above) and
10% associated with fulminant hepatic failure secondary to
hepatitis of various etiologies.
SPECIFIC PATIENT SELECTION AND PREOPERATIVE

CONSIDERATIONS
Patient Selection.
The success of liver transplantation is closely related to the

rational selection of patients most likely to benefit from the
procedure. In fact, the single most important prognostic factor

affecting survival is the medical condition of the recipient at the

time of OLT (Fig. 20-42) . [6] As with the recipients of other
organs, patients should be without additional end-organ failure
(other than that clearly related to hepatic insufficiency) and
should be candidates for a major operative intervention. In
general, dependence on alcohol or other harmful substances
should be resolved for at least 6 months. Extrahepatic
malignancy, sepsis, and diffuse mesenteric venous thrombosis
represent absolute contraindications. Isolated portal vein
thrombosis is a relative contraindication. The patient's liver
function should be such that complications of dysfunction are
emerging, with the predicted life span of the patient managed
medically less than 2 years. With the improvements in survival
after allotransplantation realized in the past decade, it is also
appropriate to consider patients with metabolic diseases or
moderately advanced liver disease with significant alterations in
quality of life such as extreme fatigue, refractory pruritus, or
encephalopathy.
Figure 20-42 United Network for Organ Sharing

(UNOS) status and survival after orthotopic liver
transplantation (for definition of UNOS status, see
text).
Once advanced liver disease is identified, efforts to proceed

with transplantation should be initiated, because the
preoperative health of the recipient greatly affects the chance
for success. Early evaluation by a multidisciplinary team
schooled in the specific requirements for liver transplantation is
critical. In addition to surgical evaluation, the selection of
appropriate candidates for transplantation should be based on
input from medical colleagues, including a hepatologist and
infectious disease specialist. Evaluation by a social worker and
psychiatrist to establish the ability of the patient and family to
manage themselves postoperatively is important. Immediate
efforts should optimize the candidate's nutritional status and
overall medical condition. Finally, education of the patient must
be initiated early to ensure that complications of medical
noncompliance are avoided postoperatively.
For all liver recipients, thorough preoperative evaluation to
determine the antibody titer directed against hepatitis A, B, and
C as well as cytomegalovirus, human immunodeficiency virus,

Epstein-Barr virus, and herpes simplex is mandatory. In

addition, viral antigen detection is required to identify active
infection. The role of transplantation for primary viral hepatitis
was discussed previously. Active infection with
cytomegalovirus or herpesvirus requires clinical resolution
before transplantation, but viral carriers can receive transplants
given appropriate antiviral prophylaxis. Human
immunodeficiency virus infection contraindicates OLT. In
addition to the recipient's status, the donor's status for these
viruses should be ascertained before implantation, again to
allow for antiviral prophylaxis when indicated.
Several alternative therapies should be considered in the
candidate for transplantation, not necessarily to obviate the need
for transplantation but to improve the medical condition before
transplantation. Patients with good synthetic function (normal
bilirubin, normal coagulation, normal albumin), such as an
individual with Child's A cirrhosis and recurrent variceal
bleeding, may benefit from nontransplant surgical palliation
such as a Warren shunt. The TIPS procedure is also a useful
palliative step for refractory complications of portal
hypertension. Stabilization of disease with these interventions
gives more time to evaluate patients and optimize preoperative
conditions. It allows for a more thorough evaluation of a
patient's ability to comply with the posttransplant medical
regimen, particularly alcoholic patients who present abstinent
because they are too sick to
464
drink. This compliance alone may stabilize the disease in an

alcoholic patient with Child's A cirrhosis and avoid the need for
transplantation. Reduction in the degree of portal hypertension
may also minimize intraoperative bleeding during the recipient
hepatectomy. Problems with these palliative procedures are
frequent and require surveillance with particular attention to
portal vein patency. Identified candidates should be treated at a
center with transplant expertise to facilitate palliation that does
not jeopardize future transplantation. Alternative treatment
should not delay transplantation.
SELECTION OF DONORS
Immediate function of a transplanted liver is imperative. Unlike

kidney, pancreas, or, to some extent, heart transplantation, no
artificial means is readily available to support an anhepatic
patient in the event of graft failure. Without a rapid restoration
of synthetic function, death from bleeding or cerebral edema

generally ensues within 72 hours. The single most important
factor determining the early function of a liver allograft is the
viability of the donor liver. Although this seems obvious,
determining the state of the liver in a heart-beating cadaver
remains imprecise, so careful attention to the conditions of the
donor's death, the morphology and function of the organ before
harvest, and the specifics of the extirpation is critical. Failure of
a graft to function at all after a technically successful
transplantation is known as primary nonfunction (PNF). The
only treatment for PNF is retransplantation within 24 to 72
hours. Factors contributing to the development of PNF include
parenchymal insufficiency unrecognized at the time of harvest,
graft injury during the harvest or cadaver resuscitation,
preservation injury or prolonged cold ischemia, prolonged
rewarming time, and reperfusion injury after implantation. [15]
Several factors have been investigated to aid in the prediction of
PNF. The most widely noted is the estimated parenchymal fat
content. [18] [53] Donor liver biopsy specimens that show a 40%
or greater parenchymal replacement by fat have a higher chance
of PNF, and in some settings this is a reasonable indicator of the
adequacy of a donor organ. This must, however, be balanced
against the severity of the recipient's disease and the urgency of
the planned operation. Other factors used to evaluate the donor's
status include the age of the donor, the level of inotropic
support, the mechanism of the donor's death, the level of
hepatic or intra-abdominal trauma, the presence of
hypernatremia, and the biochemical studies of liver function.
No single parameter has been established absolutely governing
the acceptance of a donor for organ harvest. [11] [50] Rather,
combinations of risk factors are generally responsible for
discarding a potential donor liver. In case of doubt about the
quality of an organ, the personal inspection of the liver by an
experienced transplant surgeon is often critical for decision
about the use of an organ.
Donor use has been improved with the introduction of UW
preservation solution as mentioned earlier. Until 1987 the outer
limit for cold storage of the liver using EuroCollins solution
was about 8 hours. Use of UW solution has extended the time to
about 24 hours, allowing for better allocation of the organ and
preparation of the recipient. The specifics of this and other
preservation solutions are detailed in Part VII of this chapter.
The considerations of ABO typing and other immunologic
concerns are discussed later. An additional issue is size

compatibility. Smaller organs are easily adapted to a large

recipient, but the converse is not true. One advance has been the
use of reduced-size allografts, particularly in children. Usually
the left lateral lobe (segments 2 and 3) or left lobe (segments 2,
3, and 4) is used, allowing up to 1/10 weight mismatch. This
has been the most important factor in the success of
transplantation in small children. This technique is sometimes
used in adults when an emergency transplantation is required.
Finally, male recipients of female organs have 10% worse
survival than other sex-match combinations, according to the
1994 UNOS registry.
OPERATION ON THE DONOR
Harvest of the donor organ should be performed by an

experienced surgeon, with particular care taken to optimize the
preharvest resuscitation of the heart-beating cadaver. Because
immediate hepatic function after transplantation is required,
there is no room for error. Errors in resuscitation include
injudicious use of vasopressors, prolonged acidosis, and
hypoperfusion from hypovolemia. Visual inspection and
palpation of the liver with knowledge of the potential recipient's
status and size aid in the assessment of the appropriateness of a
donor liver.
The liver is generally procured through a midline incision from
jugular notch to pubis, including median sternotomy (Fig.
20-43) . Harvest is coordinated with the harvest teams for other
organs. Complete mobilization of the liver is required, including
division of both triangular ligaments and the falciform ligament.
The hepatogastric ligament is divided. Particular attention is
paid to the vascular supply, including preservation of aberrant
hepatic arteries (20% aberrant right from the superior
mesenteric artery, 15% aberrant left from the left gastric artery).
Unlike the relative impunity associated with arterial ligation in
nontransplanted livers, failure to complete the arterial
revascularization due to an unrecognized arterial supply is
poorly tolerated after cold storage.
Figure 20-43 The donor procedure. UW, University

of Wisconsin.

465
PNF, late biliary stricture, or intrahepatic abscess may result.

Additional care must be taken when simultaneous pancreas
harvest is performed to avoid hepatic artery injury or portal vein
transection. The suprahepatic vena cava should be preserved for
the hepatic graft, and care should be taken to avoid caval injury
during cardiac retrieval. The gallbladder is removed either in
situ or after explantation.
Once mobilization is completed, perfusion with UW solution is
initiated through the distal aorta or common iliac arteries with
ligation of the supraceliac aorta. The inferior mesenteric vein or
splenic vein is used for portal perfusion. When pancreatic
harvest is performed, some surgeons prefer to use the portal
vein for cold perfusion. UW solution is also flushed into the
biliary tree. Topical slush is rapidly applied. Hepatic extirpation
is performed after cardiopulmonary retrieval and before harvest
of the pancreas and kidney. The iliac artery and vein should be
harvested in the event that vascular reconstruction is required.
IMMEDIATE PREOPERATIVE MANAGEMENT
Patients awaiting liver transplantation are prioritized on a

national waiting list based on severity of disease as defined by
the UNOS. Status 4 patients are at home and functioning
normally, status 3 patients can be at home but require
continuous medical care. Status 2 patients are continuously
hospitalized in an acute care unit for at least 5 days or are in the
intensive care unit. Status 1 patients are in the intensive care
unit because of acute or chronic liver failure with a life
expectancy without a liver transplant of less than 7 days. Status
7 is reserved for patients taken from the active list for a
temporary medical contraindication such as sepsis. [1] These
patients can accrue points for time on the list despite not being
able to accept an organ. Strict adherence to guidelines for
proper categorization is required to allow for ethical allocation
of organs. As soon as a donor is identified, the organ is paired
with a potential recipient, who is called to the hospital for
preoperative evaluation before organ harvest.
Appropriate perioperative management of a transplant recipient
begins with a thorough preoperative physical examination
(including rectal and dental examinations) to rule out the
possibility of ongoing infection or malignancy. Routine
screening includes a complete blood cell count, electrolyte and

metabolic profile, urinalysis, and chest film. Blood is
crossmatched. Intravenous lines, including arterial lines and
pulmonary artery catheters, should be placed with strict
attention to aseptic technique because cutaneous contamination
with bacteria or fungi can cause serious postoperative
complications in the transplant patient. Selective
decontamination of the gut and mechanical preparation of the
bowel are advocated at many centers, as is a shower with an
antimicrobial soap.
For patients in the intensive care unit, supportive measures are
continued as needed. Of specific concern is the potential for
neurologic recovery in patients with advanced encephalopathy
and cerebral edema. A recent advance has been the use of
intracranial pressure monitoring in the perioperative period.
This was once shunned as a prohibitive risk for cerebral
bleeding in the coagulopathic fulminant hepatic failure patient,
but routine assessment is now possible through minimally
invasive techniques. Small-gauge catheter pressure monitors
can be placed into the subdural space without the need for
parenchymal puncture, thus decreasing the risk of the
intervention substantially. Greatly elevated intracerebral
pressures, especially in association with signs of transtentorial
herniation in a Grade 5 coma patient, suggest an irreversible
lesion that is unlikely to resolve after transplantation.
OPERATION ON THE RECIPIENT
Few surgical procedures require the fastidious attention to

technical detail required in liver transplantation. Technical
errors are translated directly into infectious complications or
marginal biliary function. Thus, transplantation should be
performed only by surgeons proficient in the procedure. In
addition, the operative environment should include experienced
nursing and ancillary support.
Intraoperative management by a knowledgeable
anesthesiologist with experience in liver transplantation is
critical for a successful technical result. The procedure presents
the challenge of maintaining homeostasis of temperature,
circulation (including oxygen-carrying capacity and coagulation
competence), gluconeogenesis, and electrolyte concentration
while establishing adequate anesthesia and paralysis with agents
not requiring hepatic function for degradation. Intraoperative
ICP monitoring is appropriate for patients with severe
encephalopathy. In procedures using a venovenous bypass, a

perfusionist is required, and procedures performed without

bypass require adequate maintenance of preload during caval
occlusion and correction of metabolic abnormalities after
release of the congested portal circulation. After the initial
function of the allograft, the most important factor predictive of
technical success is the stability of the patient intraoperatively
and his or her delivery to the intensive care unit normothermic
with adequate circulatory competence.
Successful engraftment of the organ begins with a controlled
recipient hepatectomy. This can be a formidable task in patients
with severe portal hypertension and extensive collateral
formation or in those with multiple operative interventions. In
general, extirpation follows the basic surgical guidelines of
establishing proximal and distal vascular control combined with
lysis of all ligamentous attachments. Specific technical concerns
include retaining maximal length on all vessels. Mobilization of
the common bile duct depends on the planned biliary
reconstruction (choledochocholedochostomy versus
choledochojejunostomy). Care to avoid injury to the right
adrenal vein during caval dissection is important. If venovenous
bypass is planned, cannulation of the left axillary, femoral, and
portal veins is performed.
The decision to place the recipient on bypass is routine at a
number of centers; and liberal use of this technique, originally
developed by Shaw and colleagues, [2] [3] has clearly led to
improved operative mortality. Bypass avoids mesenteric
congestion and minimizes the release of lactate and other
by-products of hypoperfusion into the portal circulation. In
addition, it improves venous return to the heart during
implantation and thus improves hemodynamic stability during
the period of caval occlusion. Bypass also diverts the portal
flow during difficult recipient hepatectomies to minimize blood
loss, particularly during dissection of the retrohepatic cava and
bare spot. [11] Despite these benefits, it is now clear that many
patients tolerate OLT without the additional manipulation
required by bypass. It has thus become policy at many centers
to employ this technique selectively after an intraoperative trial
of portal vein and vena caval occlusion.
The implantation procedure (Fig. 20-44) begins with the
suprahepatic vena caval anastomosis followed by the
infrahepatic caval anastomosis. Alternatively, the donor vena
cava can be anastomosed side to side with the recipient vena
cava if it is left in situ during the recipient hepatectomy
(piggyback technique). The operation then proceeds to the

portal anastomosis. After all venous connections, the liver is

reperfused with the suprahepatic vena cava temporarily
occluded and the infrahepatic vena cava vented to allow
washout of the hyperkalemic and adenosine-rich UW solution.
The hepatic artery anastomosis is the final vascular step in the
procedure.
466
Figure 20-44 The recipient procedure.
Some groups perform simultaneous arterial and venous

reperfusion. [43]
The biliary reconstruction remains an additional area of debate.
[54] Options include choledochocholedochostomy with or
without externalized T-tube stents. This has the advantage of

easy access for subsequent biliary manipulations or evaluation
of bile, as well as preservation of the sphincter mechanism.
Unfortunately, the anastomosis is particularly sensitive to
ischemic injury of the common duct, and complications of leak
when the T tube is removed remain vexing. [31] The incidence of
technical complications related to this method ranges from 12%
to 50% in published series, with the cumulative average being
25%. [36] These problems are obviated for the most part by use
of a Roux-en-Y choledochojejunostomy at the expense of
convenient biliary access. [8] Leak or stricture is still observed in
4% to 30% of cases (mean 14%). [36] This method is frequently
used in pediatric transplants. One report of 300 transplants
performed with a side-to-side choledochocholedochostomy
technique has reported a remarkable 2.2% technical
complication rate. [36] This technique awaits confirmation in
other centers.
POSTOPERATIVE MANAGEMENT
Management in the postoperative intensive care unit is similar

to that after any major procedure. Ventilatory support and
volume replacement are standard. Isolation is not required
beyond standard universal precautions. No sedation is given
until extubation. For unclear reasons, postoperative pain is
usually mild, and any major discomfort should alert one for

possible complications. Close monitoring of serologic liver

enzymes is critical because increasing enzymes or a failure of
enzyme values to correct rapidly suggests PNF or technical
complications such as hepatic artery thrombosis. Liberal use of
the Doppler ultrasonography and rapid return to the operating
room are mandatory in these situations because early detection
is the only factor separating a return to normal liver function
from complete graft necrosis and patient death.
Use of drains and antibiotics is no different in this operation
from that in any other major abdominal procedure.
Closed-suction drains should be used and removed early after
the threat of postoperative hemorrhage is over. Brief antibiotic
prophylaxis is appropriate with an agent with adequate skin and
biliary organism coverage. Prolonged use of prophylactic
antibiotics is contraindicated. A protocol for decontaminating
the small bowel can be used but is usually discontinued within a
few weeks of transplantation. At Duke University Medical
Center, decontamination (colistin, 100 mg./10 ml.; gentamicin,
80 mg./10 ml.; nystatin, 2 106 /10 ml.) is discontinued at the
time of normal enteral feeding, often within the first
postoperative week).
If a T tube is used in the biliary reconstruction, a T-tube
cholangiogram is obtained within a week with internalization
contingent on a normal result. Patients are discharged when
they are familiar with their medications. Patients living more
than 2 hours from the transplant center generally stay in the
vicinity for an additional 2 to 4 weeks. Close monitoring of
hepatic function and medical compliance is continued twice a
week for 4 weeks and weekly for an additional 4 weeks. After
this outpatient evaluation, patients are returned to their referring
community for chronic follow-up. It is important to establish
open lines of communication between the community physician
and the transplant center to ensure prompt recognition and
referral of postoperative complications.
IMMUNOLOGIC MANAGEMENT
The liver must be considered separately from other solid organs

with regard to immunologic management. Many
well-established concepts of donor-host interaction after kidney
or heart transplantation do not apply after hepatic
transplantation, and failure to recognize the unique properties of
this situation can lead to detrimental perioperative treatment. [35]
HLA Typing.

HLA matching is not feasible before liver transplantation.

However, because new techniques have been developed that
allow donor HLA data to be provided in a timely manner, the
future use of these data is pertinent. Although matching donors
and recipients with regard to HLA type clearly improves
outcome after kidney, heart, and pancreas transplantation, no
such correlation exists with liver transplantation. Indeed,
matching may, in fact, reduce overall survival. [33] [49] The
reasons for this lie in the dualistic nature of HLA in the
pathophysiology of liver disease. T-cell-mediated rejection of
the organ is mechanistically the same as with other organs, so
rejection is reduced with improved HLA compatibility.
However, the physiologic role of HLA is to present viral
peptides to T cells to initiate destruction of virally infected
cells. Thus, HLA compatibility potentiates
467
the inflammation during viral reinfection after transplantation

for viral hepatitis and increases the chance for clinical
recurrence of the original disease. Similarly, T-cell-mediated
autoimmune diseases (e.g., PBC) are etiologically based on
T-cell recognition of HLA presented peptides. Therefore,
recurrence of autoimmune diseases may be potentiated as well.
Further knowledge regarding specific disease states worsened
by certain HLA matches may be useful in selective typing in the
future.
Crossmatch and ABO Matching.
The lymphocytotoxic crossmatch is not used prospectively

before liver transplantation. Again, temporal concerns are most
pressing; but in this case, the value of a positive crossmatch in
predicting subsequent poor outcome from hyperacute rejection
is minimal. Indeed, hyperacute rejection is rarely seen even in
the face of documented preformed antibodies and ABO
incompatibility. [22] [24] [49] The reasons for this remain
controversial because hyperacute rejection can be readily
produced experimentally. [28] Although preformed antibodies
reduce long-term graft survival somewhat, early results appear
to be minimally affected. Grafts unmatched for ABO antigens
can cause antibody-mediated graft-versus-host disease with
mild hemolytic anemia and fever occurring between
postoperative days 5 and 12. [44] This is the result of intrahepatic
B cells that secrete antibody directed against the recipient ABO
antigens. Although it is usually self-limited, increased

immunosuppression may be required. Alternatively, some

groups suggest a decrease in immunosuppression to allow for a
limited rejection of the offending B cells. It is preferred that
rules for ABO compatibility be followed for elective
transplantation. In the emergent setting, however,
ABO-incompatible grafts can be used with acceptable results.
The 5-year survival for grafts that are ABO incompatible is
approximately 15% lower than that for grafts without ABO
discrepancy. [6] This not only reflects some immune preference
but also the patient population that receives ABO mismatched
grafts, that is, emergent transplants for FHF and PNF.
Acute Rejection.
As with other allografts, T-cell-mediated destruction of the liver

is inevitable without immunosuppressive therapy. The primary
targets for T-cell recognition are HLA antigens on the biliary
epithelium and vascular endothelium. The characteristics of this
rejection, termed acute rejection, are similar to those of kidney
or heart in that it develops in most of the cases during the first 6
months after transplantation, usually within the first 4
postoperative weeks. More than half of patients develop at least
one episode of rejection. Symptoms are nonspecific, often
including mild intermittent fever and general malaise with
alteration in liver tests. The diagnosis should be confirmed by
liver biopsy. Most episodes are readily reversible (90%), given
prompt recognition and initiation of antirejection therapy. This
contrasts sharply to chronic rejection, discussed later.
Monitoring for acute rejection is a continuous process.
Needle-core biopsy of the allograft is the best diagnostic test
(Fig. 20-45) . Histologically it appears as a predominantly
T-cell and monocyte infiltrate in the portal tracts, with
subendothelial (endotheliitis) and biliary epithelial aggregates.
Eosinophils and polymorphonuclear leukocytes are present to a
lesser degree but are more prevalent in hepatic rejection than in
infiltrate seen in other organs. Although protocol biopsies are
performed in some centers, most groups rely on monitoring of
the liver function studies and/or the serum beta2 -microglobulin
value, with biopsy used to clarify detected abnormalities.
Because a primary target in acute rejection is biliary epithelium,
it often presents initially as a cholestatic process with rapid
increases in the alkaline phosphatase and bilirubin values.
However, changes in hepatic biochemical parameters are
nonspecific and can also indicate technical or infectious
complications where alterations

Figure 20-45 Acute rejection after liver

transplantation. Portal tracts contain a mixed
inflammatory infiltrate, with mononuclear cells and
eosinophils. Bile duct damage is evident, and
endothelium of a branch of the portal vein is
disrupted and infiltrated by lymphocytes (arrow).
Hematoxylin and eosin, original magnification 325. (Courtesy of Kay
Washington, M.D., Department of Pathology, Duke University Medical
Center.)
in the immunosuppressive regimen would be ineffective or even
detrimental. Thus, liberal use of biopsy and Doppler ultrasound
evaluation of hepatic blood flow and bile duct integrity is
critical. Once acute rejection is diagnosed, rescue
immunosuppression is initiated as described later.
Chronic Rejection.
The development of liver allograft dysfunction over a period of

months to years is termed chronic rejection and, like other
allografts, is controversial and multifactorial in its etiology, and
is usually not reversible. Histologically it appears as a paucity
of bile duct epithelium without conspicuous lymphocytic
infiltration and has thus been described as the "vanishing bile
duct syndrome" (Fig. 20-46) . Additionally, an obliterative
vasculopathy can occur with parenchymal fibrosis. The time
course, histology, and refractory nature of chronic rejection
suggest that direct cell-mediated destruction is not a primary
mechanism. It is likely that the cumulative effects of mild
subclinical immune recognition by several limbs of the immune
system, and the resulting
Figure 20-46 Vanishing bile duct syndrome as a

manifestation of chronic rejection after liver
transplantation. Portal inflammation diminishes as
interlobular bile ducts disappear. There is no bile
ductular proliferation. Hematoxylin and eosin,
original magnification, 325. (Courtesy of Kay
Washington, M.D., Department of Pathology, Duke University Medical
Center.)
468
exposure to soluble factors including fibrogenic cytokines,

eventually take their toll on the fragile epithelium. Chronic
rejection often requires retransplantation.
Immunosuppressive Pharmacology.

Manipulation of the immune system is required to avoid graft

loss from rejection. Identifying a safe, effective, and minimally
immunosuppressive regimen requires a careful balance aimed at
reducing infectious and neoplastic complications without a
resultant increase in allograft rejection and/or dysfunction.
Thus, rational, selective use of several immunosuppressive
agents is required to manage successfully the broad array of
patients who are transplanted. Three general classifications of
immunosuppression are used: (1) induction therapy, a relatively
intense initial conditioning of the newly transplanted recipient;
(2) maintenance therapy, drugs given at minimal doses required
to maintain graft function; and (3) rescue therapy, heightened
immunosuppression given to reverse an episode of acute
rejection. Most transplant centers have specific protocols for
each situation. For example, the Duke University liver
transplant protocol is presented in Table 20-17 .
Several issues specific to liver transplantation should be

mentioned. Of major importance is that the liver appears to
have less immunostimulatory antigenicity than other organs.
The specifics of this perceived resistance remain somewhat
controversial but may relate to several peculiarities of hepatic
physiology. First, the portal circulation is exposed regularly to
enteric pathogens and, more importantly, to absorbed peptides,
TABLE 20-17 -- The Duke University Immunosuppressive
Protocol for Orthotopic Liver Transplantation
Induction Therapy
Methylprednisolone: Induction of anesthesia: 125 mg. IV
Postoperatively: tapered from 300 to 20
mg./ day on 5th postoperative day
Cyclosporine: Intraoperatively: continuous IV infusion (2
mg./kg./
day)
Postoperatively: continuous IV infusion (4 mg./kg./
day) on arrival in the surgical intensive care unit
On postoperative days 2-3: oral cyclosporine
(Neoral)
is started at 5 mg./kg. q12h. The IV cyclo-
sporine is decreased according to serum levels.
(Targeting postoperative levels: 250-300 ng./ml
[HPLC assay])

Azathioprine (1-1.5 mg./kg./day): Started as soon as the

patient can
tolerate oral feeding, usually by
day 2 or 3
Maintenance Therapy
Prednisone: First 3 months: 20 mg./day
3-6 months: 15 mg./day
After 6 months: 5-10 mg./day
Cyclosporine: Targeted levels: initial 10 days: 250-300
ng./ml.;
10-30 days: 200-300 ng./100 ml.; 1-6 months:
150-200 ng./ml.; 6-12 months: 120-200 ng./ml.;
and after 1 year: 80-120 ng./ml.
Azathioprine: 1-2 mg./kg., usually discontinued within 9
months of
transplantation
Rescue Therapy for Biopsy-Proven Acute Cellular
Rejection
First episode: 3-day recycling with IV methylprednisolone
500 mg./day
Second episode: repeated 3-day recycling with IV
methylprednisolone
Third episode: monoclonal antibody muromonab-CD3
(OKT3,
Orthoclone, Raritan, NJ) or antithymocyte globulin
(ATGAM, Upjohn, Kalamazoo, MI) is given for
10-15 days
Refractory rejection: cyclosporine (Neoral) is discontinued
and
FK 506 (Prograf) (target levels 10-15 ng./ml.)
is initiated
which have potential antigenicity. A generalized perihepatic

anergy has been postulated as protective in this setting to
prevent vascular thrombosis and inappropriate hepatic
inflammation. Clearly the reticuloendothelial system of the
liver--the Kupffer cells--is important in establishing appropriate
versus inappropriate presentation of portal antigen. Most of the
Kupffer cells of the liver are replaced with those of recipient
HLA type within a few weeks after transplantation. In addition,
the Kupffer cells of the donor have been described as migrating
to other recipient tissues and establishing a state of chimerism.

Whether this chimerism creates specific tolerance or is the
result of it remains to be seen. From a practical standpoint,
maintenance immunosuppression after liver transplantation can
be comparatively low, and many patients have been weaned to
little or no immunosuppression for years. Also of importance in
this regard, the liver can tolerate brief periods on
immunosuppression withdrawal if it becomes necessary to
combat a potentially lethal infection.
A recent issue of particular importance in liver transplant
immunosuppressive pharmacology has been the development of
the drug tacrolimus (FK 506). This agent has been suggested as
having particular efficacy in the setting of liver transplantation,
with truly remarkable results in early, uncontrolled trials
performed at the University of Pittsburgh. [48] Tacrolimus has a
mechanism of action similar to that of cyclosporin A and it has
been used as a replacement for this drug. A possible benefit is
the potential steroid-sparing effect of tacrolimus, exceeding that
of cyclosporine. Single-drug maintenance therapy is being
actively investigated. Multicenter, randomized, prospective
trials comparing cyclosporine to tacrolimus are underway, with
the initial 1-year follow-up demonstrating a significant
reduction in rejection episodes and a lower need for rescue
therapy with OKT3 in the tacrolimus group. [20] [52] No graft or
patient survival advantage has been demonstrated. The side
effect profile for tacrolimus differs from that of cyclosporine in
that the cosmetic complications of hirsutism, acne, and gingival
hyperplasia are not present but the potential for neurologic
abnormalities, diabetes mellitus, and renal impairment appears
to be more prevalent. An important consideration is that the
combination of cyclosporine and tacrolimus drastically
increases the incidence and severity of side effects. Thus,
therapy must be limited to one or the other, and limiting the use
of either in the early postoperative setting may benefit renal
function. Clearly, both cyclosporine and tacrolimus have proven
efficacy for maintenance immunosuppression after liver
transplantation. The specific selection of either remains to be
established by randomized investigation. As a rescue agent
during established acute rejection, neither tacrolimus nor
cyclosporine is particularly efficacious compared with
glucocorticosteroids or antilymphocyte antibody preparations.
Another practical issue unique to liver transplant
immunosuppression relates to the absorption of cyclosporine
and, to a lesser degree, tacrolimus. Drug uptake depends on the
availability of enteric bile salts. Thus, with decreased

production of bile in the initial posttransplant period or during a

rejection episode, or biliary diversion through a T tube,
cyclosporine should be administered intravenously to maintain
adequate plasma levels. A new cyclosporine formulation
(Neoral), which is bile salt independent for absorption, is now
available.
Antiviral Immunity and Immunosuppression.
The importance of viral infection in liver transplant patients

cannot be overstated. De novo infection or latent virus
reactivation of pathogens, including cytomegalovirus, herpes
simplex virus, and Epstein-Barr virus, and reinfection of
hepatotropic viruses, including hepatitis B and C, remain
serious sources of posttransplant morbidity and mortality and
are directly related
469
to the intensity of the immunosuppressive regimen employed.

In particular, the use of antilymphocyte antibody preparations,
particularly OKT3, have been implicated in increasing the
likelihood of viral infection and associated complications such
as lymphoproliferative disorders. In response to this association,
most transplant centers have incorporated antiviral prophylaxis
into their immunosuppressive protocols, with the specific
agents, duration of therapy, and dosages governed by the viral
status of the recipient and the donor organ. The relationship
between the pathophysiology of acute T-cell-mediated rejection
and the physiologic function of T cells in viral immunity is
underscored in the transplant patient on T-cell-directed
immunosuppression.
OUTCOMES
Clearly, liver transplantation is the most significant

advancement in the treatment of end-stage liver disease this
century. Diseases treated by OLT are by definition terminal
with few exceptions, and as such are lethal without hepatic
replacement. Survival of a patient with no other hope for
survival is the most obvious positive outcome. Operative
survival now exceeds 90% for first grafts. Retransplant-free
survival has improved steadily in the past 10 years and is now
73% at 1 year for all transplants reported to the UNOS registry.
The 5-year survival reflecting transplantation before ganciclovir
became available is approximately 60%, and improvements of
2% to 3% per year have been made each year since 1987.
Predicted actuarial survival for transplants performed in 1995 is

over 75%.
Survival and retransplant rates have been the principal, if not
the only, measures of outcome in the transplant literature until
recently. The technical issues of survival in the first decade of
widespread transplantation have overshadowed other outcome
measures to the point of their exclusion from the literature.
Critical evaluation of costly health care interventions has
become a fundamental priority in our society. Outcomes can be
defined either from a physician's, patient's, payer's, or society's
perspective. The medical perspective is in the traditional
purview of the clinician and clinician investigator, who use
objective and quantifiable medical parameters, whereas the
patient's perspective involves a more subjective assessment of
quality of life. Today, survival has become the norm after OLT,
and thus the next decade should clearly define outcome with
regard to each of these perspectives.
Despite the recent advances in all aspects of liver

transplantation, the procedure remains one with considerable
morbidity. [14] Most patients have some complications that
deviate from an ideal recovery, and all patients accept the trade
of their liver disease for the disease of immunosuppression.
These negative outcomes are generally remedied by prompt
recognition of problems and aggressive corrective intervention.
It is, therefore, critical that the potential obstacles arising
postoperatively are thoroughly understood. The authors have
recently categorized negative outcomes under three headings:
negative sequel, complications, and failure to cure. [16]
A negative sequel is an adverse effect inherent to the transplant
procedure. [16] Transplantation in general carries with it the risk
of lifelong immunosuppression. Patients must adapt their
lifestyle to ensure that medications are taken as prescribed and
that the immune system and organ function are monitored when
necessary. The risk of opportunistic infection and malignancy
that accompanies chronic immunosuppression persists for life.
These factors, however, are less intrusive for liver transplant
patients than for recipients of other solid organs because of their
reduced need for immunosuppression, especially after the first
transplant year.
Failure to cure refers to pre-existing conditions that remain
unchanged or recur after the transplant procedure. [16] The
likelihood of cure reflects the primary disease. Metabolic and
cholestatic diseases are generally resolved, as are the

physiologic disorders of alcoholic cirrhosis. Unfortunately, viral
infections remain generally uncured by liver replacement. As
mentioned earlier, clinically significant recurrence of hepatitis
B and hepatitis C may limit long-term cure. The cure for these
illnesses remains in more potent antiviral therapeutics.
Practically, however, viral reinfection that creates a clinically
insignificant carrier state is viewed as a cure by the patient.
Obviously, no extrahepatic malignancy can be cured by OLT,
and the potential for cure in patients with intrahepatic
malignancy is solely related to the presence or absence of
metastatic disease at the time of recipient hepatectomy.
A complication is any other negative outcome that does not fit
clearly into the definition of negative sequel or failure to cure.
Complications of some kind occur in almost all patients, but the
significance of these setbacks varies greatly. For example, acute
rejection must be considered a complication of sorts, but most
episodes are treated without significant alteration of a patient's
comfort or residual disability. Conversely, PNF at best leads to
retransplantation and at worst to death. The lack of uniform
reporting of complications makes interpretation of the results of
OLT difficult. [14] The authors have presented a classification of
complications stratified by severity [14] (Table 20-18) . With the
heightened requirement for outcomes-based research in the
current economic environment, standardized evaluation of
transplant programs with regard to complication rates will likely
increase.
Several common complications, such as acute and chronic
rejection, have been discussed. Some complications deserve
particular attention because of their seriousness and requirement
for prompt intervention. PNF presents as a complete lack of
synthetic function from the time of reperfusion. The patient
develops encephalopathy, increased intracranial pressure,
coagulopathy, hyperbilirubinemia, and hypertransaminasemia.
Aggressive supportive therapy and prompt retransplantation are
required within 72 hours. Total transplant hepatectomy with
portacaval shunt can improve the hemodynamic and metabolic
status of the patient for 24 to 36 hours if a suitable liver is not
found immediately.
Hepatic artery thrombosis remains a complication, especially in
the pediatric population. This presents as a rapid rise in serum
transaminase levels. The transplanted liver does not tolerate loss
of arterial flow, and failure to restore flow produces graft loss.
Hepatic artery stricture or stenosis generally presents as a lesser

degree of metabolic change later in the postoperative course. An

alternative presentation of dearterialization is bile leak resulting
from hepatic duct necrosis. An additional vascular complication
that is less frequent but equally devastating is early portal vein
thrombosis. Given the rapid and serious but reversible nature of
these vascular complications, any suspect change in hepatic
function requires immediate evaluation of the hepatic
vasculature by Doppler ultrasonography, followed by either
re-exploration or a confirmatory arteriogram.
Biliary complications, which occur in 15% to 30% of patients,
suggest vascular compromise. [54] Both leaks and strictures can
occur regardless of the method of reconstruction. Percutaneous
or endoscopic management is generally considered an
acceptable first alternative, but reoperation should not be
avoided for appropriate lesions at the expense of hepatic
function or cholangitis.
Today, increasing emphasis is given to patient-oriented

subjective outcomes such as quality of life and well-being.
Subjective outcomes have inherent limitations, but they are
arguably more relevant to individual patients.
470
TABLE 20-18 -- Classification of Common Complications of

Data from Clavien, P.-A, Camargo, C., Jr., Croxford, R.,
Langer, B., Levy, G., and Greig, P.: Definition and
classification of negative outcomes in organ transplantation:
Application in liver transplantation. Ann. Surg., 220:109, 1994.

Grade An alteration from ideal postoperative course with

1: complete recovery or which can be easily controlled
and which fulfills the
general characteristics, namely (1) not life
threatening; (2) not requiring use of drugs other
than immunosuppressive agents,
analgesics, antipyretic, anti-inflammatory, and
antiemetic, drugs required for urinary retention or
lower urinary tract infection,
arterial hypertension, hyperlipidemia or transient
hyperglycemia; (3) requiring only therapies that can
be performed at the
bedside; (4) postoperative bleeding requiring 3
units of blood; and (5) never associated with a
prolongation of the
intensive care unit (ICU) stay 5 days or total
hospital stay 4 weeks.
Examples:
Superficial wound infection treated without
antibiotics
Bile leak treated conservatively
Corticosteroid-responsive acute rejection
Well-controlled arterial hypertension
Grade Any complication that is potentially life-threatening
2:
or results in ICU stay 5 days or hospital stay
4 weeks, but that does
not cause residual disability or persistent diseases.
Complications requiring only use of drug therapy or
Grade
postoperative bleeding requiring 3 units of
2a
blood.
Examples:
Rejection requiring immunosuppressors not
routinely used after induction therapy (e.g., OKT3
or other antilymphocyte
drugs)
Bacterial, viral, or fungal infection requiring
antibiotic, antiviral, or antifungal therapy

Primary graft dysfunction (opening aspartate

transaminase [AST] 2000 or a transient increase
in AST levels 1000 I.U./
L., or a persistent elevated prothrombin time > 20
over 3 days)
Grade Complications requiring therapeutic interventions,
2b reoperation, readmission in the ICU, or
prolongation in the ICU stay 5
days but that do not result in residual disability.
Examples:
Primary graft dysfunction extending the ICU stay
5 days
Postoperative bleeding requiring laparotomy
Biliary problems requiring endoscopic or surgical
procedures
Grade Any complication with residual or lasting functional
3: disability or development of malignant disease
(except squamous and
basocellular cutaneous malignancies).
Complication with lasting disability that shows no
Grade evidence of progression and that has a relatively
3a low risk of graft failure and/
or death.
Examples:
De novo hepatitis C
Nonprogressive chronic rejection
Persistent bile duct stricture without evidence of
progressive liver failure or recurrent cholangitis
Complications with lasting disability that are either
Grade difficult to control or have a significant risk of
3b leading to graft failure and/or
death.
Examples:
De novo hepatitis B
Development of malignancy (e.g., sarcoma,
carcinoma, and lymphoma)

Persistent bile duct stricture not amenable to

surgical or endoscopic treatment with progressive
liver failure or recurrent
cholangitis
Progressive chronic rejection
Grade Complications that lead to retransplantation (Grade
4: 4a) or death (Grade 4b)
Well-being represents a composite of several different aspects,

including mental, physical, and social criteria.
Due to the initial formidable technical issues of OLT, analyses
of quality of life have been almost absent from the literature.
With improved results these past few years, researchers in
numerous centers are evaluating the effects of OLT on quality
of life. Successful transplantation allows a return to an active
lifestyle free from the metabolic and hematologic complications
of hepatic failure or portal hypertension. Preliminary studies
have already shown that self-image, functioning ability, and
perception of health status are significantly improved after
OLT. [32] Currently, about 60% of patients undergoing OLT
return to work within the first year, [4] a figure that continues to
increase long term. [32]
Once therapy has been shown to be effective, it is necessary to

determine whether the treatment is cost-effective in order to
compare it with other competing technologies. Analysis of
cost-effectiveness is difficult to apply to liver transplantation
because there is no alternative therapy with which the results of
the procedure can be compared. Thus, the cost issue becomes
one of establishing a monetary value for one's life. There is no
doubt that OLT is expensive. The total first-year expenses for a
new liver average $200,000, with an additional $10,000 to
$20,000 per year required thereafter. Cost for patients in the
intensive care unit at the time of transplant are 3 to 5 times
higher than those called in from home for the procedure. With
current health care reform underway, these amounts are falling
somewhat. In addition, consolidation of transplant centers in the
United States is likely to improve efficiency and reduce costs.
As new immunosuppressive agents are introduced, competition
is also likely to drive the cost down. Nonetheless, the extent to
which physicians should go to return patients to health and to
work is a matter of great societal debate. As with all other

aspects of liver transplantation, rational selection of patients is

the most important factor affecting cost.
471
EMERGING TECHNIQUES
Split and Reduced-Size Transplantation.
The remarkable ability of the liver to regenerate and support the

metabolic needs of an individual despite major resection has
allowed for correction of most conditions with partial hepatic
transplantation. [10] This is useful for major size mismatches,
especially in children, and can also help address the growing
donor organ shortage. [41] Anatomic division of the lobes of the
liver with preservation of hepatic venous and arterial, portal,
and biliary branches has become increasingly successful but
remains technically formidable. The anatomic boundaries are
based on the segmental anatomic system of Couinaud and
Bismuth. [17] Left lateral segments 2 and 3 or left lobe grafts
(segments 2, 3, and 4) can be placed in recipients who are
substantially smaller than the donor. Because of the anterior to
posterior dimensions of the right lobe (segments 5 through 8),
placement of this graft requires a recipient similar in size to the
donor.
Several major centers have initiated protocols for split and
reduced-size transplantation with excellent results. Predictably,
biliary complications are increased in this procedure, but in
children the problem of hepatic artery thrombosis is improved,
owing to the comparatively large vasculature present in the
graft when an adult liver is reduced for use in a child. Patient
survival is as good as or better than full-size grafts in children.
Use of reduced-size grafts has decreased the mortality of
pediatric recipients on the waiting list to approximately 3%.
Living-Related Transplantation.
Arising from the success of reduced-size grafting, living-related

transplantation has been initiated at selected centers. [9] [57] This
involves a reduced-size graft usually derived from a donor left
lobe (segments 2 and 3 or 2, 3, and 4). Technically, this has
been quite successful and has the benefits of reduced ischemic
time, better HLA match, and better timing of transplantation,
producing excellent graft survival. Because most of the
transplants are performed for congenital anomalies, the negative
aspects of HLA typing have not been problematic. The most

pressing concern with this procedure is ethics. Although hepatic

resection is generally safe, the mortality is not zero. One donor
perioperative death has been reported, and many argue that with
the success of reduced-size cadaveric allografts, a procedure
that places a healthy parent at risk is not necessary. The concept
of informed consent is difficult to establish because most
parents disregard personal safety when the life of their child is
at risk. One consensus hearing on the matter has approved this
procedure, with the caveat that only centers with established
success in reduced cadaveric grafts, pediatric transplantation,
and adult hepatic surgery be involved. [10]
Heterotopic Liver Transplantation.
Placement of an allograft in an anatomically altered site has the

advantages of avoiding the recipient hepatectomy (often the
most morbid portion of the procedure) and preserving the
orthotopic position for future use in the event of graft failure.
[51] Metabolic abnormalities are correctable by this approach.
Obviously, however, disorders leading to portal hypertension

are not amenable to this mode of therapy.
ALTERNATIVE THERAPIES
The development of artificial support devices has

revolutionized perioperative management in all areas of
transplantation except hepatic. Renal dialysis, ventricular assist
and intra-aortic counterpulsation devices, total parenteral
nutrition, and insulin have all helped optimize the condition of
solid organ recipients to some degree, making emergent
transplantation unusual for any organ other than the liver. The
importance of preoperative condition is clear from the survival
statistics presented in this chapter. Thus, great effort has been
directed toward finding adequate hepatic replacement.
Xenogeneic Support.
Xenotransplantation, the use of organs from other species, has

many theoretical advantages. A renewable supply of organs
subject to genetic manipulation available on an elective basis
would greatly alter the course of patients with liver failure. In
addition, the hepatotrophic viruses responsible for most
hepatitis are generally specific for human hepatocytes, so the
specter of reinfection would be abolished. Unfortunately, the
immune barriers to transspecies transplantation remain
formidable. Organs from discordant species, those
phylogenetically distant animals to which preformed natural

antibodies exist, are hyperacutely rejected. Organs from these

animals also produce plasma proteins that are similar but not
identical to their human homologues, thus raising the possibility
of antigenic proteins subject to immune clearance. Concordant
species, namely primates, are rejected in a more conventionally
acute manner, but antibody-mediated rejection occurs. Although
the immune barrier is less daunting, primates are slow-breeding
animals that could quickly become extinct if widespread use
were initiated. Of additional concern is the potential for
introducing new viruses from primates to man.
Several efforts in xenogeneic organ use have been made in the
past 5 years. Baboon livers were used to treat two patients with
hepatitis B at the University of Pittsburgh. [46] Both livers
functioned well enough to carry out the major physiologic
functions of the liver. Interpretation of the immune implications
of these procedures, however, is difficult because one patient
was infected with the human immunodeficiency virus with an
inverted T4:T8 ratio pretransplant and the other received a
concurrent baboon bone marrow transplant. An orthotopic pig
liver transplant performed at Cedars Sinai Hospital in Los
Angeles was hyperacutely rejected (L. Makowka, personal
communication, 1994).
Ex vivo perfusion with porcine livers has been successfully
employed as a bridge to orthotopic allotransplantation by the
authors' group. [13] Biochemical improvement in all measured
parameters, including reversal of cerebral edema and reduction
of coma, has been demonstrated by the authors' group and
others. [3] This approach has the advantage of being reversible
without requiring a surgical procedure. It is, however,
logistically difficult and temporary.
The ability of porcine hepatocytes to perform many of the

functions of human hepatocytes has been exploited by several
investigators by development of an ex vivo apparatus for hepatic
support consisting of porcine hepatocytes attached to a
hollow-fiber dialysis cassette. [45] Early clinical trials have
shown promise in reducing cerebral edema as well as mild
improvement in biochemical parameters. Additional
investigation in this field will determine whether the volume of
the liver can be reproduced in cellular form and whether
temporary support will allow recovery of hepatic function, thus
avoiding transplantation for some patients with viral or
toxin-induced acute liver failure.

SUMMARY
Liver transplantation has evolved in the past decade from an

experimental procedure to an accepted, effective therapy for
end-stage diseases of the liver. Extended survival of over 75%
in appropriately selected patients is now commonplace with
return to an excellent quality of life. Continued improvements
in perioperative management and operative
472
technique are being realized. The most important predictors of

success are the state of the patient at the time of transplantation
and the disease being transplanted. Early intervention once
end-stage disease is diagnosed is preferable. A critical shortage
of suitable donor organs remains the single most important
barrier to transplantation
SELECTED REFERENCES
Advances in liver transplantation. Gastroenterol. Clin. North
Am., 22:entire issue, 1993.
This is an excellent review written by experts in transplantation.
Most clinical aspects of liver transplantation are critically
discussed, including indications, immunosuppression, and
management of short- and long-term complications.
Consensus conference on indications of liver transplantation.
Hepatology, 2(Suppl.):entire issue, 1994.
Consensus conference held in Paris by a panel of medical and
surgical experts thoroughly discussed indications for liver
transplantation regarding both adult and pediatric diseases.
Neuberger, J., and Adams, D. (Eds.): Immunology of Liver
Transplantation. London, Edward Arnold, 1993.
This excellent monograph covers most of the immunological
aspects of liver transplantation, including mechanisms of
rejection, immunosuppression, and viral prophylaxis. It is
extremely well referenced.
REFERENCES
1. AnnualReport of the U.S. Scientific Registry of Transplant Recipients

and the Organ Procurement and Transplantation Network, 1994.

2. Consensusstatement on indications for liver transplantation.

Hepatology, 20:63, 1994.
3. Abouna, G., Fisher, L., Porter, K., and Andres, G.: Experience in the
treatment of hepatic failure by intermittent liver hemoperfusion. Surg.
Gynecol. Obstet., 137:141, 1963.
4. Adams, P., Ghent, C., Grant, D., and Wall, W.: Employment after liver
transplantation. Hepatology, 21:140, 1995.
5. Ascher,N. L., Lake, J. R., Emond, J., and Roberts, J.: Liver
transplantation for hepatitis C virus-related cirrhosis. Hepatology, 20:24,
1994.
6. Belle,S. H., Beringer, K. C., and Detre, K. M.: Trends in Liver

Transplantation in the United States. In Terasaki, P. I., and Cecka, J. M.
(Eds.): Clinical Transplants. Los Angeles, UCLA Tissue Typing
Laboratory, 1993, p. 19.
7. Benhamou, J.-P.: Indications for liver transplantation in primary biliary

cirrhosis. Hepatology, 20:11, 1994.
8. Bismuth, H., Castaing, D., Gugenheim, J., et al.: Roux-en-Y

hepatojejunostomy: A safe procedure for biliary anastomosis in liver
transplantation. Transplant. Proc., 19:2413, 1987.
9. Broelsch,C. E., Burdelski, M., and Rogiers, X.: Living donor for liver
transplantation. Hepatology, 20:49, 1994.
10. Broelsch,C. E., Emond, J. C., Whitington, P. F., Thistlewaite, J. R.,

Baker, A. L., and Lichtor, J. L.: Reduced sized liver transplantation. Ann.
Surg., 212:368, 1990.
11. Busuttil,R. W., Shaked, A., and Mills, J. M.: One thousand liver
transplants: The lessons learned. Ann. Surg., 219:490, 1994.
12. Calne,
R. Y.: Contraindications to liver transplantation. Hepatology,
20:3, 1994.
13. Chari,R., Collins, B. H., and Magee, J. C.: Treatment of hepatic failure
with ex-vivo pig liver perfusion followed by liver transplantation. N. Engl.
J. Med., 331:234, 1994.
14. Clavien,P.-A., Camargo, C., Jr., Croxford, R., Langer, B., Levy, G.,
and Greig, P.: Definition and classification of negative outcomes in organ
transplantation: Application in liver transplantation. Ann. Surg.,
220:109,1994.
15. Clavien,P.-A., Harvey, P. R. C., and Strasberg, S. M.: Preservation and

reperfusion injuries in liver allografts: Overview and synthesis of current
studies. Transplantation, 53:957, 1992.
16. Clavien, P.-A., Sanabria, J. R., and Strasberg, S. M.: Proposed

classification of complications of surgery with examples of utility in
cholecystectomy. Surgery, 111:518, 1992.
17. Couinaud,C.: Le Foie: Etudes Anatomiques et Chirurgicales. Paris,

Masson et Cie, 1957.
18. D'Alessandro, A. M., Kalayoglu, M., and Solinger, H. W.: The

predictive value of donor liver biopsies on the development of primary
nonfunction after orthotopic liver transplantation. Transplant Proc.,
23:1536, 1991.
19. Dousset, B., Houssin, D., Soubrane, O., Boillot, O., Baudin, F., and
Chapuis, Y.: Metastatic endocrine tumors: Is there a place for liver
transplantation? Liver Transpl. Surg., 1:111, 1995.
20. European FK506 Multicentre Liver Study Group: Randomized trial

comparing tacrolimus (FK506) and cyclosporin in prevention of allograft
rejection. Lancet, 344:423, 1994.
21. Farges,O., Malassagne, B., Sebagh, M., and Bismuth, H.: Primary
sclerosing cholangitis: Liver transplantation or biliary surgery. Surgery,
117:146, 1995.
22. Gordon, R. D., Iwatsuki, S., Esquivel, C. O., Tsakis, A., Todo, S., and
Starzl, T. E.: Liver transplantation across ABO blood groups. Surgery,
100:342, 1986.
23. Griffith,
B. P., Shaw, B. W., Hardesty, R. L., Iwatsuki, S., and Bahnson,
H. T.: Venovenous bypass without systemic anticoagulation for
transplantation of the human liver. Surg. Gynecol. Obstet., 160:271, 1985.
24. Gugenheim, J., Samuel, D., Reynes, M., and Bismuth, H.: Liver
transplantation across ABO blood group barriers. Lancet, 336:519, 1990.
25. Harrison,J., and McMaster, P.: The role of orthotopic liver

transplantation in the management of sclerosing cholangitis. Hepatology,
20:14, 1994.
26. Kalayoglu,M., Sollinger, W. H., and Stratta, R. J.: Extended

preservation of the liver for clinical transplantation. Lancet, 1:617, 1988.
27. Klion,F., Fabry, T., Palmer, M., and Schaffner, F.: Prediction of
survival in patients with primary biliary cirrhosis: Examination of the
Mayo Clinic model on a group of patients with known endpoint.
Gastroenterology, 102:310, 1992.
28. Knechtle, S. J., Kolbeck, P. S., Tsuchemntos A., et al.: Hepatic

transplantation into sensitized recipients: Demonstration of hyperacute
rejection. Transplantation, 43:8, 1987.
29. Krom, R. A. F.: Liver transplantation and alcohol: Who should get
transplants? Hepatology, 20:28, 1994.
30. Lee,J., Schutz, S., England, R., Leung, J., and Cotton, P.: Endoscopic
therapy of sclerosing cholangitis. Hepatology, 21:661, 1995.
31. Lerut,
J., Gordon, R. D., Iwatsuki, S., et al.: Biliary tract complication
in human orthotopic liver transplantation. Transplantation, 43:47, 1987.
32. Levy,
M., Jennings, L., and Abouldoud, M.: Quality of life
improvements at one, two, and five years after liver transplantation.
Transplantation, 59:515, 1995.
33. Markus, B. H., Duquesnoy, R. J., Gordon, R. D., et al.:

Histocompatibility and liver transplantation: Does HLA exert a dualistic
effect? Transplantation, 46:372, 1988.
34. NationalInstitutes of Health Consensus Development Conference

Statement. Liver transplantation. Hepatology, 4(Suppl. 1):107, 1983.
35. Neuberger,J., and Adams, D. (Eds.): Immunology of Liver

Transplantation. London, Edward Arnold, 1993.
36. Neuhaus, P., Blumhardt, G., Bechstein, W. O., Steffen, R., Platz, K.-P.,
and Keck, H.: Technique and results of biliary reconstruction using
side-to-side choledochocholedochostomy in 300 orthotopic liver
transplants. Ann. Surg., 219:426, 1994.
37. O'Grady,J. G., Schalm, S., and Williams, R.: Acute liver failure:
Redefining the syndromes. Lancet, 342:273, 1993.
38. Otte,J.-B., Goyet, J. D. V. D., and Reding, R.: Sequential treatment of

biliary atresia with Kasai portoenterostomy and liver transplantation: A
review. Hepatology, 20:41, 1994.
39. Perrillo,
R., and Mason, A.: Hepatitis B and liver transplantation:
Problems and premises. N. Engl. J. Med., 329:1885, 1993.
40. Pichlmayr, R., Weimann, A., and Ringe, B.: Indications for liver
transplantation in hepatobiliary malignancy. Hepatology, 20:33, 1994.
41. Piper,
J. B., Whitington, P. F., and Woodle, E. S.: Pediatric liver
transplantation at the University of Chicago Hospitals. In Terasaki, P. I.,
and Cecka, J. M. (Eds.): Clinical Transplants. Los Angeles, UCLA Tissue
Typing Laboratory, 1992, p. 179.
42. Pitt,
H., Thompson, H., Tompkins, R., and Longmire, W.: Primary
sclerosing cholangitis: Results of an aggressive surgical approach. Ann.
Surg., 196:259, 1982.
43. Post,S., Palma, P., Gonzalez, A., Rentsch, M., and Menger, M.: Timing
of arterialization in liver transplantation. Ann. Surg., 220:691, 1994.
44. Clavien, P. A., Camargo, C. A., Cameron, R., Washington, M. K.,

Phillips, M. J., Greig, P. D., and Levy, G. A.: Kupffer cell
erythrophagocytosis and graft-versus-host hemolysis in ABO-unmatched
liver transplantation. Gastroenterology, 110:1891, 1996.

45. Rozga,J., Podesta, L., and LaPage, E.: A bioartificial liver to treat
severe acute liver failure. Ann. Surg., 219:538, 1994.
46. Starzl,
T., Fung, J., and Tzakis, A.: Baboon-to-human liver
transplantation. Lancet, 341:65, 1993.
47. Starzl,
T. E., Marchiaro, T. L., Von Kaulla, K., et al.:
Homotransplantation of the liver in humans. Surg. Gynecol. Obstet.,
117:659, 1963
48. Starzl,
T. E., Todo, S., Fung, J., Demetris, A. J., Venkataramanan, R.,
and Jain, A.: FK 506 for human liver, kidney and pancreas transplantation.
Lancet, 2:1000, 1989.
49. Steinhoff,
G.: HLA/ABO matching. In Neuberger, J., and Adams, D.
(Eds.): Immunology of Liver Transplantation. London, Edward Arnold,
1993, p 261.
50. Strasberg,S., Howard, T., Molmenti, P., and Hertl, M.: Donor
evaluation of liver allograft. Hepatology, 20:829, 1995.
51. Terpstra,O., Schalm, S., Weimar, W., et al.: Auxiliary partial liver
transplantation for end-stage chronic liver disease. N. Engl. J. Med.,
319:1507, 1988.
52. TheUS Multicenter FK506 Liver Study Group: A comparison of

tacrolimus (FK506) and cyclosporine for immunosuppression in liver
transplantation. N. Engl. J. Med., 331:1110, 1994.
53. Todo, S., Demetris, A., VanThiel, D., Teperman, L., Fung, J., and
Starzl, T. E.: Orthotopic liver transplantation for patients with hepatitis B
virus-related liver disease. Hepatology, 13:619, 1991.
54. Vallera, R. A., Cotton, P. B., and Clavien, P.-A.: Biliary reconstruction
for
473
liver transplantation and management of biliary complications: Overview

and survey of current practices in the United States. Liver Transpl. Surg.,
1:143, 1995.
55. vanThiel, D. H., Wright, H. I., and Fagiuoli, S.: Liver transplantation
for hepatitis B virus-associated cirrhosis: A progress report. Hepatology,
20:20, 1994.
56. Williams, R., and Wendon, J.: Indictations for orthotopic liver
transplantation in fulminant liver failure. Hepatology, 20:5, 1994.
57. Yamaoka, Y., Tanaka, K., and Ozawa, K.: Liver transplantation from
living-related donors. In Terasaki, P. I., and Cecka, J. M. (Eds.): Clinical
Transplants. Los Angeles, UCLA Tissue Typing Laboratory, 1993,p. 179.



Sabiston: Textbook of Surgery, 15th ed., Copyright 1997 W. B. Saunders Company
Chapter 20 - TRANSPLANTATION
382
R. Randal Bollinger M.D., Ph.D.

Delford L. Stickel M.D.
I - HISTORICAL ASPECTS
ANCIENT ACCOUNTS OF TRANSPLANTATION
Transplantation, the removal or partial detachment of a part of the body and its implantation to the body of the same or a
different individual, has fascinated mankind for centuries. Legends of transplantation are recorded in the early written
histories of both Eastern and Western cultures. Homer, in his Iliad, describes the monstrous Chimaera, a remarkable
creature of transplanted animal parts created by the gods. This mythical hybrid animal had the heads of a lion, a goat, and
a serpent. All three of its heads breathed fire. [35] The term chimaera is now used in transplantation to describe individuals
who possess hybrid characteristics, such as the circulating cells of both donor and recipient after bone marrow
transplantation.
A Chinese document written in approximately 300 B.C. contains this legendary account of transplantation: "One day two
men, Lu and Chao, called on the surgeon Pien Ch'iao. He gave them a toxic drink and they were unconscious for three
days. Pien Ch'iao operated and opened their stomachs and explored the heart; after removing and interchanging their
organs he gave a wonderful drug and the two men went home recovered." [63]
The legend of Cosmas and Damian describes transplantation as one of the miraculous feats of these two medical martyrs.
Born in Arabia in the third century A.D. and trained in medicine in Syria, Cosmas the physician and Damian the surgeon
performed numerous miraculous healings until their martyrdom in 287 A.D. by decapitation. The miracle of the black leg
is said to have occurred posthumously in approximately 348 A.D. While an elderly parishioner with a gangrenous,
cancerous leg lay sleeping in the Basilica of Cosmas and Damian, the saints came to him and removed the diseased leg
with a saw. They replaced the destroyed tissue with the fresh leg of a Moor buried that same day in the cemetery of Saint
Peter. The new leg was attached at the thigh and ointment applied to the site. The parishioner awoke to find himself free
of pain and able to walk on his new healthy black leg. [37] Attempts at transplantation during the Middle Ages did not
always end so successfully. Tragically, in 1492 two boys were bled to death in a vain attempt to save the life of Pope
Innocent VIII by means of transfusion of young blood. [63]
The oldest evidence of grafting that could have been of some therapeutic benefit is observed in the remains of trephined
prehistoric skulls. The trephine holes were usually small, but in a Bronze Age skull a rather large defect evidently was
filled by reimplanting the removed fragment. [30] In this specimen, the cut margin showed no sign of healing, so the
operation may have been fatal. Recovery from primitive skull trephination is well documented, however, both
archeologically and in studies of primitive peoples in modern times, and it is conceivable that such trephination was
sometimes therapeutically effective.
Ancient Hindu surgeons described methods for repairing defects of the nose and ears using techniques of grafting similar
to those used in modern times. The following technique for nasal reconstruction is quoted from a translation of the
Sushruta Samhita, a document written about 700 B.C. [7] :
Now I shall deal with the process of affixing an artificial

nose. First the leaf of creeper, long and broad enough to fully cover the
hole or the severed or clipped off part should be gathered; and a patch of
living flesh equal in dimension of the preceding leaf, should be sliced off
from down upward from the region of the cheek and, after scarifying it with
a knife, swiftly adhered to the severed nose. Then the cool-headed physician
should steadily tie it up with a bandage decent to look at and perfectly
suited to the end for which it has been employed. The physician should make
sure that the adhesion of the severed parts has been fully effected and then
insert two small pipes into the nostrils to facilitate respiration and to
prevent the adhesioned flesh from hanging down. After that the adhesion part
should be dusted with the powders of sappanwood, licorice-root and bayberry
pulverized together; and the nose should be enveloped in cotton and several
times sprinkled over with a refined oil of pure sesamum....As soon as
the skin has grown together with the nose, he cuts through the connection
with the cheek.
This Indian method was lost to Western medicine until 1794 when English surgeons stationed in India described nasal
reconstruction as they had seen it performed by an Indian surgeon, the technique quite similar to that described more than
1000 years earlier. [62]
A new Western tradition of transplantation surgery arose during the Renaissance in Bologna. The sixteenth century
anatomist and surgeon Gasparo Tagliacozzi developed his technique for reconstruction using a flap of skin from the inner
aspect of the upper arm. He carved the flap of skin in the shape of the patient's nose and then stitched it to the forehead
and inner surface of the cheek, leaving a slender attachment to the arm to maintain blood supply until circulation was
re-established from the face. After this painful procedure, the patient had to sit upright with the arm alongside the face
and the head turned toward the arm for the next 3 weeks of healing; then the attachment to the arm was severed.
Tagliacozzi was successful in replacing noses cut off in combat or for punishment or destroyed by syphilis. The
technique is still in use and is known as the tagliacotian flap or the Italian method. In considering but discarding the idea
of grafting tissue donated by another individual, Tagliacozzi made the following remarkable statement: "The singular
character of the individual entirely dissuades us from attempting this work on another person. For such is the force and
the power of individuality, that if anyone should believe
383
that he could accelerate and increase the beauty of union, nay more, achieve even the least part of the operation, we
consider him plainly superstitious and badly grounded in the physical sciences." [56]
EARLY EXPERIMENTS IN TRANSPLANTATION
The Scottish surgeon John Hunter (1728-1793) is known as the father of experimental surgery because of his pioneering
research. Several of his experimental procedures involved transplantation, and some of his specimens are still preserved
in the Hunterian Museum in London. Hunter revived the practice of transplanting teeth, which had been done in ancient
Egypt, Greece, Rome, Arabia, and pre-Columbian America, as well as by Ambroise Pare of Paris in the sixteenth
century. [46] With his crude techniques of transplanting tissues without primary revascularization and without antisepsis,
Hunter was unable to distinguish allografts from autografts on the basis of graft survival. About this operation he wrote:
"Success of this operation is founded on the disposition of all living substances to unite when brought in contact with one
another, although they are of different structure and even though the circulation is carried in one of them." [10]
In other animal experiments he successfully autografted and allografted chicken testes and observed that the ends of
severed Achilles tendons grew together after suturing. A variety of connective tissue transplant procedures were
performed successfully for the first time during the eighteenth and nineteenth centuries. Foremost among these were skin
grafts and corneal transplants.
Skin Grafts.

The first well-documented report of successful free autografts of skin was in 1804 by Baronio, who experimented with
sheep, although free autografts of human skin may have been used successfully centuries before. [10] [19] [46] [62] In 1822,
Bunger reported successful use of a free full-thickness human skin autograft to repair a nasal defect. In 1870, Reverdin
reported the observation that small grafts of epidermis on a granulating surface increased in size and grew out to coalesce
with adjacent grafts. In 1886, Thiersch, in Germany, described the resurfacing of wounds with large sheets of
split-thickness skin. Such grafts are still sometimes termed Thiersch's grafts, although essentially the same procedure was
reported 14 years earlier by Ollier in France.
In 1863 Paul Bert, a student of Claude Bernard, reported that autografts, allografts, and xenografts behaved differently.
[19] [62] The significance of these observations received little attention, however; nineteenth century authors (including
Baronio and Reverdin) generally failed to observe that the results of allografts and autografts of skin were different. Skin
allografts were used to some extent clinically, as illustrated in a story by Winston Churchill of his donating a small piece
of skin to a wounded fellow officer in 1898. [19] There appear to have been three reasons for the mistaken belief that skin
allografts grew permanently, a belief still widely held as late as the third decade of the twentieth century: (1) for a week
or more skin allografts are indistinguishable from autografts; (2) it is difficult to distinguish between permanent survival
of a small skin graft and ingrowth of adjacent host skin to cover the area of a sloughed graft; and (3) corneal allografts
survive permanently.
Corneal Transplants.
Corneal xenografts attempted early in the nineteenth century were unsuccessful. A successful corneal allograft between
two gazelles was reported by Bigger in 1835 [46] [62] ; but the necessity of using a cornea from the same species was not
recognized until the period 1872 to 1880, when successful corneal allografts were reported in animals and in man.
Refinements of operative techniques, methods of preservation of grafts, and systems of graft procurement were
subsequently developed. From 1925 to 1945, corneal transplantation emerged as a widespread and generally accepted
therapeutic practice. [18] [46] [62]
TRANSPLANTATION IN THE TWENTIETH CENTURY
Although important developments in the last half of the nineteenth century, such as the use of ether and other general
anesthetics and the acceptance of Lister's principles of antiseptic surgery, were important in the progress of
transplantation, organ replacement is a development of the twentieth century. Transplantation of vascularized organs,
including the kidney, liver, heart, lung, pancreas, and intestine, was first made possible when techniques for vascular
anastomosis were developed.
The first long-functioning renal transplant was reported by Ullmann in March 1902. He transplanted kidneys into dogs
using magnesium tube stents and ligatures to make the vascular anastomoses to the carotid artery and internal jugular
vein in the neck. [59] That same year the French surgeon Carrel reported his new technique of suturing blood vessels
together using triangulation and fine silk suture material (Fig. 20-1) . [12] His revolutionary technique was rapidly applied
to the problems of organ transplantation. Between 1902 and 1912, Carrel and Guthrie of Chicago performed a large series
of animal transplantation experiments, including the transfer of blood vessels, kidneys, hearts, spleens, ovaries, thyroids,
extremities, and even the head and neck. In 1905 in his preliminary communication entitled "The Transplantation of
Organs" Carrel stated:
This operation consists of extirpating an organ with its vessels, of putting it in another region, and of uniting its vessels to
a neighboring artery and vein. If the organ is replaced in the same animal from which it was removed the operation is
called an autotransplantation. If it is placed in another animal of the same species it is called a homotransplantation,
while if it is placed into an animal of a different species, the operation is called a heterotransplantation. [13]
Terminology.
Although the terms defined by Carrel are still used occasionally, the preferred nomenclature is now allotransplantation
(allograft) for transplants between nonidentical members of the same species and xenotransplantation (xenograft) for

transplants between members of different species
Figure 20-1 Carrel's technique of vascular anastomosis. Shortly after Alexis Carrel reported his new technique of
suturing blood vessels together using triangulation and fine silk suture material, he applied the method to the
transplantation of blood vessels, hearts, spleens, kidneys, and extremities. (From Carrel, A.: La technique operatoire des
anastomoses vasculaires et la transplantation des visceres. Lyon Med., 98:859, 1902.)
384
(Table 20-1) . Graft is commonly used as a synonym for transplant and host as a synonym for recipient. The prefix iso- is
ambiguous because it is used with two distinctly different meanings, as discussed by Gorer. [27] The term isograft as
shown in Table 20-1 derives from geneticists' use of the term isogeneic in referring to genetically identical individuals,
whereas for over 70 years immunologists have used iso- to refer to immunity to antigens of blood and tissues of
genetically dissimilar individuals of the same species (e.g., isoimmune, isoantigen, isoantibody). Depending on the site of
implantation, grafts are termed orthotopic if surrounded by the same type of tissues or located in the same part of the
body after transplantation as previously; otherwise they are termed heterotopic. Heterotopic grafts are sometimes
implanted into privileged sites, such as locations that protect the graft from rejection (e.g., the anterior chamber of the
eye, the brain, the testes, or in a diffusion chamber).
Problem of Rejection.
Carrel did not understand the biologic basis for differences in graft outcome among the various types of grafts he
attempted, although by 1910 he recognized the problem of rejection:
Should an organ, extirpated from an animal and replanted
into its owner by a certain technique, continue to functionate normally, and
should it cease to functionate when transplanted into another animal by the
same technique, the physiological disturbance could not be considered as
brought about by the surgical factors. The changes undergone by the organ
would be due to the influence of the host, that is, the biological factors.
Elucidation of the biologic factors hypothesized by Carrel required several decades and is, in fact, continuing today. A
few of the milestones in transplantation immunology are cited.
Although the immunity theory of graft rejection was postulated by several authors during the first decade of the century,
a number of other theories were held by prominent authorities, such as Loeb. The immunity theory was questioned
largely because there was no direct evidence that circulating antibody--the traditional hallmark of immunity--was
involved in the rejection process. Antibodies had been demonstrated in response to allografts of tumor but not allografts
of normal tissues, and attempts to confer allograft immunity passively with serum were unsuccessful. The discovery of
cellular immunity, histocompatibility antigens, and immunologic tolerance provided important steps in the understanding
of transplant rejection.
Cellular Immunity.
In 1914, Murphy reported lymphocytic infiltrates in host tissues surrounding rejecting transplanted tumors. [19] [62] He
postulated that the small lymphocyte was responsible for that rejection, and he used radiation and treatment with benzene
to modify the process. The role of cellular immunity (lymphocytes), as distinguishable from humoral immunity
(circulating antibody), was not firmly established, however, until experiments were performed in which certain forms of
immunity were observed to be transferable
TABLE 20-1 -- Transplantation Terminology

Recent Older Relationship of Donor

Nomenclature Nomenclature and Recipient of Graft
Autogeneic graft Autograft Same individual
Isogeneic graft Isograft Same species and
genetically identical
Allogeneic graft Homograft Same species but not
genetically identical
Xenogeneic Heterograft Different species
graft
to an unimmunized subject by lymphoid cells and not by serum. [62] By 1954, these adoptive transfer experiments were
performed by Potter and then Mitchison for tumor allograft immunity in mice and by Landsteiner and Chase for
delayed-type hypersensitivity reactions in man. In 1954, Billingham, Brent, and Medawar reported the use of lymphoid
cells to transfer immunity to skin and other tissue grafts in mice. Only viable cells were capable of transferring the
immunity, a phenomenon designated adoptively acquired immunity to distinguish it from passive immunity produced by
injections of antibody.
The Second-Set Phenomenon.
In 1903, Jensen observed that a second graft did not survive as long as the first when a mouse received two grafts of a
tumor separated by an interval of several days, and he suggested that immunity was responsible for the difference. This
effect of more rapid rejection of a second graft was not always observed with transplants of tumor, however. Under
certain conditions survival of the second graft was prolonged. Casey, in 1932, termed the latter phenomenon
enhancement, and Kaliss, in 1953, reported that enhancement is transferable to normal animals by injections of serum. [62]
The effect was subsequently demonstrated to be due to an immunoglobulin, and enhancing or blocking antibodies have
been used experimentally to prolong the survival of nonneoplastic as well as neoplastic tissues.
The second-set phenomenon in human skin graft recipients was observed by Holman while treating burn patients at the
Johns Hopkins Hospital. He reported in 1924 that a second group of pinch grafts from the same donor were rejected more
rapidly than the first and that "the destroying agency is specific for each set of grafts." [34] He postulated that each group
of grafts developed its own antibody. In 1932, Shinoyi, in Japan, described the specificity of the second-set phenomenon.
Gibson and Medawar, working in England in 1943, reported similar observations with burn patients, and use of the term
second set dates to this report. [10] [19] [42] [62] In subsequent controlled experiments with rabbits, Medawar demonstrated
the immunologic specificity of the phenomenon, which was observed uniformly only when the same donor was used for
both the first set and the second set of grafts. Medawar also contrasted the histologic characteristics of first- and
second-set rejections, the first-set rejection being predominantly a cellular event, whereas both cellular and humoral
mechanisms are involved in the rejection of the second set of grafts.
Transplantation Antigens.
When immunity, both cellular and humoral, had been established as the cause of graft rejection, study was focused on the
antigens that both stimulated graft rejection and were the targets of the ensuing immune response. The antigens
responsible for graft rejection and the genetic control of these antigens have been most extensively studied in the mouse.
The influence of genetic factors was documented by Jensen, Tyzzer, and Little in the first two decades of the century. [46]
[62] In 1948, Gorer, Lyman, and Snell described H-2 as a genetic locus controlling strong histocompatibility antigens in
the mouse. Subsequently, this locus and numerous minor histocompatibility loci were characterized in great detail.
The definition of the major histocompatibility locus of man, HLA, is intertwined with the evolution of typing and
cross-matching for human blood donor selection. The work of Landsteiner during the first four decades of this century
with erythrocyte ABO and Rh antigens was necessary for blood banking and transfusion, which were done extensively
during and after World War II. The development of blood transfusion contributed to progress with the problem of graft

rejection in three respects. First, the A and B erythrocyte antigens are widely distributed in tissues and are transplantation
antigens that must be considered in the selection
385
of tissue and organ donors. Second, by analogy with typing and crossmatching for blood donor selection, one of the
major approaches to the problem of graft rejection has been tissue compatibility testing. Third, the sera of patients who
have received multiple blood transfusions frequently contain antibodies to human leukocytes. It is now known that these
are HLA antibodies, and sera from such patients were a principal source of antibodies in early studies of the HLA system.
The serologic identification of transplantation antigens began in 1952 when Dausset discovered a leukocyte antigen
responsible for transfusion reactions. [11] Payne found in 1958 that antileukocyte antibodies were frequent in the sera of
multiparous women, thus establishing a rich source of reagents for tissue typing. The new system of tissue matching was
first used to select appropriate donors and recipients by Hamburger of Paris. [31] In 1964, Payne reported the first clear
evidence that these leukocyte antigens segregated in families as a genetic system. Whereas the original serologic
identification of transplantation antigens was done by leukoagglutination, Terasaki, in 1964, introduced a much more
sensitive and specific microlymphocytotoxicity test. [57] Definition of the HLA system, the major histocompatibility gene
complex of man, has followed a series of international workshops, the first of which was organized in 1964 by Amos at
Duke University. A major advance that same year was the discovery that lymphocytes from potential donors and
recipients, when mixed together in tissue culture, undergo a vigorous proliferative response. This reaction, termed a
mixed lymphocyte culture (MLC), became, along with microlymphocytotoxicity, a major approach for histocompatibility
testing. [11]
Immunologic Tolerance.
The chimaera, an organism carrying living tissues of two or more genetically different individuals, exists not only as a
creature of Greek mythology but also naturally in dizygotic cattle twins. Owen, in 1945, reported that each of such twins
has two different types of erythrocytes, and he postulated that the marrow of each individual had become populated by
cells of both in utero when the circulation of the two placentas was mixed. [45] Owen successfully exchanged skin grafts
between the cattle twins, and in 1955 Simonson reported that kidneys as well as skin could be readily transplanted
between them. In 1953, Dunsford discovered a human twin carrying both A and O erythrocytes, but the other member of
the pair died in infancy. In 1959, Woodruff and Lennox reported successful exchange of skin grafts from dizygotic
human twins showing blood chimaerism with types A and O. [42] [62] Allografts of skin placed on a chimaera from donors
other than the chimaeric mate were rejected in the normal manner. Thus, a natural chimaera is specifically nonreactive to
the tissue antigens of its chimaeric mate. Such nonreactivity specifically limited to particular antigens is termed
immunologic tolerance. In contrast, immunosuppression is nonspecific suppression of immune responses to antigens
generally. The most common example of naturally occurring tolerance is the normal state of nonreactivity to self
antigens, that is, to the antigens of one's own body. Autoimmune diseases are the consequence of abnormal reactivity to
self antigens. Burnet conceived recognition of self antigens as one of the aspects of embryologic maturation of the
immunologic system. [42] [62]
The creation of states of acquired tolerance (i.e., induced specific immunologic tolerance) has been achieved largely by
exposure of embryonic, fetal, or early postnatal hosts to grafts that the normal adult animal would reject. Before tolerance
was defined immunologically, Murphy, in 1912, observed that rat sarcomas grew on chicken embryos but not in mature
chickens, and he noted that the chick acquired the adult capacity to reject the tumor approximately 5 days after
completion of shell life. [9] [62] In 1929, Danforth and Foster reported successful skin grafts between newly hatched Rhode
Island red and Plymouth Rock chicks (Fig. 20-2) . [42] In 1950, Cannon and Longmire reported similar observations, but
they noted additionally that the percentage of take was only 1% if the grafts were performed on chicks 3 days old and
was nil at the age of 14 days. [42] [62] A landmark article in the understanding of transplantation immunology appeared on
October 3, 1953, when Billingham, Brent, and Medawar reported their experiments on "actively acquired tolerance of
foreign cells." [8] They systematically studied the phenomenon of actively acquired tolerance between inbred strains of
mice of various ages before and after birth. It became clear that the barrier between self and non-self could be overcome
if the exposure to alloantigens occurred in the neonatal period. Grafts established on the fetus survived permanently, and

the host was tolerant of other grafts from the donor strain; grafts performed more than 1 or 2 days after birth were
rejected, and the rejection of subsequent grafts from the donor strain was accelerated. These authors also reported
breaking tolerance, that is, reversing tolerance and terminating the chimaeric state, by injecting lymphoid cells of normal
adult host-strain mice into tolerant animals. The reversal of the tolerant state in these experiments was marked by the
sloughing of long-established grafts of skin and other tissues from the donor strain.
Animals rendered tolerant prenatally or neonatally were normal except for being chimaeras and for being specifically
nonreactive to antigens of the donor. Many subsequent studies have been directed toward the objective of inducing
tolerance in the adult by methods that would be applicable to therapeutic transplantation in man. Lasting tolerance has
been produced in adult mice that were temporarily immunosuppressed at the time of initial exposure to donor antigens,
but tolerance is readily produced by this means only if the donor-recipient incompatibility is weak. Immunity, not
tolerance,
Figure 20-2 Tolerant chickens. In 1929 Danforth and Foster successfully transplanted skin
between newly hatched Plymouth Rock (light) and Rhode Island red (dark) chickens. Such grafts
take shortly after birth, but not 2 weeks later, and provide an example of actively acquired
tolerance. (From Moore, F. D.: Give and Take: The Development of Tissue Transplantation.
Philadelphia, W. B. Saunders, 1964.)
386
usually results if the incompatibility is strong. Because a uniformly effective method of producing acquired tolerance to
transplantation antigens in adult animals and humans has not yet been discovered, the progress of transplantation has
depended on the development of methods of immunosuppression.
Immunosuppression.
Total-body irradiation had been used extensively to prevent rejection of grafts in experimental animals before it was used
in the first successful human allografts from living, related donors in Paris and in Boston. [42] However, in the 4 years
between March 1958 and March 1962, of 12 potential recipients at the Peter Bent Brigham Hospital who were subjected
to total-body x-irradiation with or without marrow infusion, only 1 survived. Although the one patient with a successful
allograft lived for 25 years, irradiation as an immunosuppressive agent was judged "too blunt, nonspecific and
unpredictable." [43] Schwartz and Dameshek [50] reported in 1959 that 6-mercaptopurine blocked the capacity of rabbits to
form antibody. The animals could still react with proteins administered before or after the period of 6-mercaptopurine
treatment, suggesting an element of specificity in the suppression. Calne and Zukoski independently used the drug
successfully for canine renal transplants, and Hitchings and associates developed an imidazole derivative, azathioprine, in
1961 that could be administered conveniently and safely in an oral form. Murray, Hume, and Starzl reported clinical
successes with the new drug that same year, thus initiating the modern era of transplantation. [43]
In the 1950s, numerous authors reported the efficiency of adrenocortical steroids in reversing the manifestations of
various immunopathologic disease states. In 1963, Starzl reported that prednisone added to azathioprine produced good
results in most patients. The following year, Marchioro and associates reported the successful use of prednisone to
reverse established manifestations of renal allograft rejection. Antilymphocyte serum was demonstrated by Woodruff and
Anderson in 1963 to prolong skin allograft survival in rats and was used clinically in 1966 by Starzl. [10] [43] The
immunosuppressive properties of cyclosporine were discovered by Borel in 1972, used by him in animal studies in 1974,
and used clinically by Calne in kidney transplantation trials in 1978. [9] Whereas the indiscriminate use of
immunosuppression in the 1950s and 1960s employed modalities that affected cells and tissues in addition to the
immunocompetent cells responsible for allograft rejection, cyclosporine and, subsequently, monoclonal antibodies
allowed modulation of more defined populations of the involved cells. Although alloantigen-specific immunosuppression
remained an elusive goal of transplantation research, cyclosporine markedly improved the results of liver, heart, and
heart-lung transplantation, making them for the first time broadly applicable as therapies for end-stage organ failure.
THE ERA OF ORGAN REPLACEMENT

With the advent of chemical immunosuppression, the brief but exciting history of clinical transplantation began. For the
first time, several vascularized organs were transplanted with regular success. Foremost among these was the kidney.
Kidney Transplantation.
The technical barriers to kidney transplantation were overcome early in the twentieth century by Ullmann [59] and Carrel.
[14] In 1908 Carrel wrote: "It is to be concluded that an animal which has undergone a double nephrectomy in the grafting
of both kidneys from another animal can secrete almost normal urine with his new organs, and live in good health at least
for a few weeks. This demonstrates that it is possible to re-establish sufficiently functions of transplanted kidneys."
In 1906, Jaboulay attempted two kidney xenografts from a pig and a goat to the extremities of patients with chronic renal
failure. The kidneys failed after only 1 hour. In 1909, Unger attempted a monkey-to-human kidney transplant to save a
girl dying in renal failure. The kidney was sutured to her thigh vessels, but no urine was produced. [32] The first human
kidney allograft was performed in 1933 in the Ukraine by Voronoy. He transplanted a kidney donated from a
head-injured victim to a patient with acute renal failure from mercuric chloride poisoning. Six hours were required to
transplant the kidney to the recipient thigh vessels under local anesthesia, and the transplanted organ never functioned.
Voronoy reported six unsuccessful human renal allograft attempts between 1933 and 1949. A kidney allograft to the arm
vessels was performed by Hufnagel, Hume, and Landsteiner in Boston in 1946. The transplanted kidney functioned
transiently until the patient's own kidneys recovered, and she eventually left the hospital fully recovered. [42] Between
1950 and 1953, human kidney allografts were attempted without immunosuppression in Paris and Boston. [32] [41] Most of
these kidneys failed immediately, but one transplant recipient of Hume had life-sustaining renal function for several
months. Living-related transplantation commenced in 1953 when Michon of Paris transplanted a kidney from a mother to
her son, whose solitary kidney had been damaged in a road accident. The kidney functioned for 22 days before it was
rejected. [32] In 1954, Murray performed the first renal transplant between monozygotic twins and achieved excellent,
long-term function. [43] In March 1958, Murray, in Boston, and Hamburger, in Paris, each performed a series of human
kidney allografts using total-body irradiation for immunosuppression. [32] The modern era of immunosuppression had
begun, and the subsequent history of renal transplantation paralleled the development of immunosuppressive drugs.
Canine liver grafts were shown to function after transplantation to the pelvis by Welch in 1955. Orthotopic liver
transplantation in dogs was attempted by Cannon in 1956 and performed successfully by Moore in 1959. The first
attempt at liver allotransplantation in man was made by Starzl at the University of Colorado on March 1, 1963. [52] The
3-year-old recipient with extrahepatic biliary atresia died of hemorrhage on the day of transplantation. Ensuing attempts
in Denver, Boston, and Paris were unsuccessful until 1967, when the first extended survival of a human liver allograft
recipient was achieved by Starzl. The addition of cyclosporine immunosuppression by Calne in 1978 and then
combination therapy with cyclosporine and prednisone by Starzl in 1980, as well as better liver preservation and surgical
techniques, improved the prospects for clinical liver transplantation.
Heart Transplantation.
Carrel and Guthrie performed the first heart transplant in 1905 at the University of Chicago. [16] They transplanted a
canine heart to the neck of another dog and observed rhythmic contraction for 2 hours until coagulation occurred in the
cavities of the heart. Mann and associates, in 1933, transplanted canine hearts to the neck with more success. [40] One of
their dogs survived 8 days, allowing them to be the first to recognize cardiac allograft rejection. The first clinical heart
transplant was performed by Hardy in Jackson, Mississippi, in January 1964. [33] A 68-year-old patient in cardiogenic
shock received a chimpanzee heart when the prospective human donor became unsuitable. The small animal heart proved
inadequate to take the patient's venous return, and the recipient died after 1 hour. The first successful clinical transplant
was performed on December 3, 1967, when Dr. Christiaan Barnard, at the University of Cape Town, transplanted the
heart of a young man to a 54-year-old patient with a heart irreparably damaged by repeated myocardial infarction. [4] The
recipient lived 18 days before dying of gram-negative pneumonia. The historical foundations
387

of heart transplantation are reviewed by Griepp and Ergin. [28]
Lung Transplantation.
In contrast to the success of cardiac allografts, clinical lung transplantation proved much more difficult. The first human
lung transplant was performed by Hardy on June 11, 1963, in a patient with chronic lung disease and carcinoma of the
left lung. The patient survived 18 days before dying of renal insufficiency. Because of the difficulty of finding suitable
donors, bronchial anastomotic complications, and allograft rejection, only 38 lung transplants were performed in the first
15 years of clinical experience. [60] The longest survivor was a 23-year-old sandblaster with micronodular silicosis who
lived 10 months after lung allotransplantation. [21] The tracheobronchial anastomotic complications were initially
overcome by simultaneous transplantation of the lungs and heart, a procedure first performed by Reitz and associates at
Stanford in 1981. [49] Single-lung and eventually double-lung transplantations without the heart were made possible by
technical advances pioneered by Cooper and colleagues in Toronto. [58]
Pancreas Transplantation.
The first clinical pancreatic transplant was performed by Williams. [23] His patient, a 15-year-old boy, died in coma 3
days later. Work in this area was sporadic and unsuccessful until 1922, when Banting and associates corrected the
hyperglycemia of human diabetes mellitus by injection of bovine pancreatic extract. [3] However, clinical application of
whole pancreas transplantation in a systematic manner was not to occur for more than 40 years. In 1970, Lillehei
published the first cases in the extensive University of Minnesota series of clinical pancreatic allotransplants that began
in 1966. [39] In the original 14 cases of pancreaticoduodenal transplantation, four patients and one graft survived more
than 1 year, and one recipient was still alive in 1984. The same institution reported the first large series of human islet
allografts in 1977. However, none of 20 islet allografts led to insulin independence, and only 3 of 10 islet autografts for
chronic pancreatitis led to insulin independence. [44] The historical development of pancreatic islet transplantation is
reviewed by Downing. [23]
During the short history of clinical pancreas transplantation, the difficult problem of eliminating or draining the
pancreatic exocrine secretions was managed in several ways. Gliedman and associates attempted to anastomose the
pancreatic duct to the ureter. [26] Duct ligation was attempted by the same group as well as by Groth in 1976. [29] The
technique of injecting the pancreatic duct with a synthetic polymer to block exocrine function was reported by Dubernard
and associates in 1978. [24] Free drainage of the duct into the peritoneal cavity was investigated by Sutherland and
reported in 1979. [54] However, intractable ascites occurred in some patients, and an overall technical complication rate of
50% mandated reinvestigation of enteric drainage. [54] Bladder drainage, as popularized by Sollinger and associates, [51]
produced improved results with combined kidney-pancreas transplants. With the use of the segmental technique and a
pancreaticojejunostomy, living-related donor pancreas transplantation began in 1979. [55]
Intestine Transplantation.
Although autotransplantation of the bowel is among the most frequently used and successful forms of organ
transplantation (see Part XII in this chapter), intestinal allotransplantation has been generally unsuccessful. Clinical small
intestinal allografting was attempted in several patients after 1967 for bowel infarction, [38] repeated sepsis on total
parenteral nutrition, [1] and Gardner's syndrome with recurrent desmoid tumors of the bowel. [25] Even in the case of the
HLA identical graft, which survived 76 days, minimal useful bowel function was observed. Rejection, graft-versus-host
disease, infection, and high operative mortality historically diminished surgical enthusiasm for intestinal transplantation.
The advent of cyclosporine immunosuppression and subsequent potent drugs have reawakened interest in the field. [20] [22]
[47]
Xenotransplantation.
When the techniques of vascular anastomosis were sufficiently well known to permit successful autotransplantation and
allotransplantation, a number of xenogeneic transplants into humans were attempted. Between 1905 and 1910, several
workers, including Jaboulay in France and Unger in Germany, performed xenografts but did not document graft function.
[32] When the immunologic basis of rejection was established, renewal of interest in clinical xenotransplantation awaited

the development of new immunosuppressive measures. After the efficacy of chemical immunosuppression with
azathioprine, prednisone, and mitomycin-C was established, Reemtsma and associates, in 1963 and 1964, undertook
xenografts in patients in renal failure using kidneys from nonhuman primates. [48] Several of these cases showed
satisfactory immediate function, but all were eventually rejected within a few months. When cyclosporine became
available for clinical use, human xenografting was again attempted in 1984. [2] A baboon heart was transplanted to a child
born with a severe congenital malformation of the heart. Despite intensive immunosuppression, rejection developed and
the infant died within weeks of transplantation.
Tissue and Organ Preservation.
Along with the improved capability of transplanting tissues and organs, interest in preservation and storage of living
tissue developed. The structural integrity and the viability of the graft had to be maintained during the interval from
removal to implantation. Basically, two approaches were available: (1) methods that reduced or brought to a reversible
standstill the need for oxygen and other metabolic requirements and (2) systems that supported active metabolism. Of the
several methods that were tried to achieve long-term preservation, including freezing, only hypothermia and organ
perfusion are in general use today. In addition, chemical inhibition of metabolism in the form of cardioplegia solution is
used for cardiac transplantation and many other open-heart surgical procedures as discussed in Chapter 54
In 1908, Carrel removed an artery from one animal, preserved it with hypothermia for days, and then successfully
transplanted it to another animal. [15] The numerous other contributions of Carrel to tissue culture and ultimately organ
perfusion were reviewed by Humphries and Dennis in their "Historical Developments in Preservation." [36] Using his
newly developed media and culture techniques, Carrel was able to maintain chick embryo fibroblasts in continuous
culture for more than 25 years! His attempts at organ culture were less successful. With the use of a pump and perfusion
apparatus designed by Charles Lindbergh, organs were perfused for 20 to 40 days with normothermic serum. Although
some cells remained viable, reimplantation of the organs was not undertaken to test the effectiveness of the preservation
system. Maximal preservation of kidneys was approximately 2 days. [17]
A variety of solutions were used for continuous perfusion of organs. Humphries achieved 24-hour dog kidney
preservation using continuous perfusion with dilute blood at 10 C. Plasma protein fraction, cryoprecipitated plasma,
silicone-gel fraction of plasma, and albumin were all added to electrolyte solutions to improve preservation. Belzer and
associates, in 1967, introduced a new pump and perfusate containing lipoprotein-free serum for continuous pulsatile
perfusion at 10 C. that enabled him to consistently preserve kidneys for 72 hours. [5] Improved preservation solutions
developed by him at the University of Wisconsin extended preservation times of the kidney and ultimately of the liver
and pancreas as well. [7]
Hypothermia.
At temperatures of 0 to 4 C., tissues remain
388
viable in the absence of circulation 10 or more times longer than at normal body temperature. The simple method of
hypothermia proved useful in preserving skin, cornea, kidney, liver, heart, pancreas, and blood. Flushing an organ with a
cold perfusate, usually a balanced electrolyte solution, was used widely as a means of rapid cooling, within seconds, to
temperatures that by surface cooling were achieved only after a number of minutes and at the expense of loss of viability.
In 1960, Lapchinsky, of Moscow, reported successful reimplantation of dog kidneys and hind limbs after 24 to 28 hours
of preservation using cold storage. He perfused the kidney or limb for 1 hour with cooled whole blood, then kept the
organ cold at 2 to 4 C. until 1 hour before reimplantation, when he perfused the tissue again for 1 hour, this time with
warm blood. [36] Collins and associates, in 1969, developed a flushing solution that mimicked intracellular fluid. [18] With
this hyperkalemic and hyperosmolar solution, they flushed kidneys, kept them cold on ice, and obtained excellent
function after 30 hours of iced storage. By removing the magnesium from Collins' solution, the EuroCollins solution,
widely used by European transplant centers, was developed in 1976. EuroCollins and Sacks' solution, developed in 1978,
successfully preserved human kidneys for 50 hours or more. [36] The use of preservation solutions to flush grafts and
reduce their metabolism through hypothermia, hyperkalemia, hyponatremia, and hypocalcemia before cold storage

should be distinguished from perfusion methods designed to support metabolism by simulating as fully as possible the
normothermic internal environment of the organ. However, elements of the two approaches were often combined in the
form of continuous perfusion at 4 to 10 C.
Organ-Sharing Networks.
A natural outgrowth of the capabilities for organ preservation and tissue matching was the development of networks for
sharing kidneys on the basis of histocompatibility. For example, in 1968, Hume, at the Medical College of Virginia, in
cooperation with Amos, of Duke University, developed an organ-sharing plan to enlarge the potential recipient pool for
each new kidney that became available so that better tissue matches between donor and recipient could be obtained. The
resulting organization, named the Southeastern Organ Procurement Foundation (SEOPF), shared kidneys among nine
institutions based on computer-assisted matching of all potential recipients in that region. The SEOPF network expanded
to include 46 transplanting institutions and led in 1984 to the incorporation of the United Network for Organ Sharing
(UNOS) to facilitate organ placement throughout the United States. Similar organ-sharing networks were developed in
Europe, Scandinavia, the United Kingdom, and elsewhere. These regional networks began to cooperate in the sharing of
human organs and tissues on an international scale.
SELECTED REFERENCES
Converse, J. M., and Casson, P. R.: The historical background of transplantation. In Rapaport, F.T., and Dausset, J.
(Eds.): Human Transplantation. New York, Grune & Stratton, 1968.
The authors present a history of the principal developments in transplantation from ancient to modern times, and they
include some details not included in the other histories cited.
Griepp, R. B., and Ergin, M. D.: The history of experimental heart transplantation. Heart Transplant., 3:145, 1984.
The authors present a brief, interesting, and well-illustrated summary of the development of heart transplantation.
Hamilton, D.: Kidney transplantation: A history. In Morris, P.J. (Ed.): Kidney Transplantation: Principles and Practice.
London, Grune & Stratton, 1984.
In this chapter written for the second edition of Kidney Transplantation, the author presents an interesting account of the
evolution of human kidney transplantation from early European experiments to the modern era.
Moore, F. D.: Give and Take: The Development of Tissue Transplantation. Philadelphia, W. B. Saunders, 1964.
In this volume is presented a concise review of developments in basic biology and in medicine and surgery that apply to
therapeutic renal transplantation. Interesting aspects of historic renal transplants in Boston are described by the author,
who was there at the time and has communicated personally with scientists there and elsewhere who have made notable
contributions. For the student, this book is an informative introduction to the subject of transplantation, and for the lay
reader it is a readily understood account of some interesting developments in biology and medicine.
Starzl, T. E., Iwatsuki, S., Van Thiel, D. H., Garmer, J. C., Zitelli, B. J., Malatack, J. J., Schade, R. R., Shaw, B. W., Jr.,
Hakala, T. R., Rosenthal, J. T., and Porter, K. A.: Evolution of liver transplantation. Hepatology, 2:614, 1982.
The researcher, physician, and author who has contributed the most to the development of liver transplantation provides
an authoritative account of the development of the field.
Terasaki, P. I. (Ed.): History of Transplantation: Thirty-five Recollections. Los Angeles, UCLA Tissue Typing
Laboratory, 1991.
Personal reminiscences of pioneers who were involved in the initiation of clinical transplantation in the 1950s and 1960s
are presented in unedited form.
REFERENCES
1. Alican,
F., Hardy, J. D., Cayirli, M., Vamer, J. E., Moynihan, P. C., Turner, M. D., and Anas, P.: Intestinal transplantation: Laboratory
experience and report of a clinical case. Am. J. Surg., 121:150, 1971.

2. Bailey,
L. L., Nehlsen-Cannarell, S. L., Concepcion, W., and Jolley, W. B.: Baboon-to-human cardiac xenotransplantation in a neonate.
JAMA, 254:3321, 1985.
3. Banting, F., Best, C., Gollip, J., Campbell, W., and Fletcher, A.: Pancreas extracts in the treatment of diabetes mellitus. Can. Med. Assoc.
J., 12:141, 1922.
4. Barnard,
C. N.: A human cardiac transplant: An interim report of a successful operation performed at Groote Schuur Hospital, Cape Town.
S. Afr. Med. J., 41:1271, 1967.
5. Belzer, F. O., Ashby, B. S., and Dunphy, J. F.: Twenty-four-hour and 72-hour preservation of canine kidneys. Lancet, 2:536, 1967.
6. Belzer, F. O., Hoffman, R. M. and Southard, J. H. A new perfusate for kidney preservation. Transplantation, 33:322,1982.
7. Bhisragratna, K. K.: The Sushruta Samhita. An English translation based on the original Sanscrit text. Calcutta, 1907.
8. Billingham, R. E., Brent, L., and Medawar, P. B.: Actively acquired tolerance of foreign cells. Nature, 172:603, 1953.
9. Borel, J. F.: Cyclosporine: Historical perspectives. Transplant. Proc., 15 (Suppl. 1):2219, 1983.
10. Caine R. Y.: Renal Transplantation. London, Edward Arnold, 1967.
11. Carpenter, C. B.: Clinical histocompatibility testing: A brief historical perspective. Transplant. Proc., 13(Suppl. 1):55, 1981.
12. Carrel A.: La technique operatoire des anastomoses vasculaires et la transplantation des visceres. Lyon Med., 98:859, 1902.
13. Carrel, A.: The transplantation of organs: A preliminary communication. JAMA, 45:1645, 1905.
14. Carrel, A.: Transplantation in mass of the kidney. J. Exp. Med., 10:140, 1908.
15. Carrel, A.: Results of the transplantation of blood vessels, organs and limbs. JAMA, 51:1662, 1908.
16. Carrel, A., and Guthrie, C. C.: The transplantation of veins and organs. Am. Med., 10:1101, 1905.
17. Carrel, A., and Lindbergh, C. A.: The Culture of Organs. New York, Paul B. Hoeber, 1938.
18. Collins, G. M., Bravo-Shugarman, M., and Terasaki, P. I.: Kidney preservation for transportation. Lancet, 2:1219, 1969.
19. Converse,J. M., and Casson, P. R.: The historical aspects of transplantation. In Rapaport, P. T., and Dausset, J. (Eds.): Human
Transplantation. New York, Grune & Stratton, 1968.
20. Craddock,G. N., Nordgren, S. R., Reznick, R. K., Gilas, T., Lossing A. G., Cohen, Z., Stiller, C. R., Cullen, J. B., and Langer, B.: Small
bowel transplantation in the dog using cyclosporine. Transplantation, 35:284, 1983.
21. Dermon,F., Barbier, F., Ringoir, S., Versieck, J., Rolly, G., Berzsenyi, G., Venneire, P., and Vrints, L.: Ten-month survival after lung
homotransplantation in man. J. Thorac. Cardiovasc. Surg., 61:835, 1971.
22. Diliz-Perez,
H. S., McClure, J., Bedetti, C., Hong, H., de Santibanes, E., Shaw, B. W., Jr., Van Thiel, D., Iwatsuki, S., and Starzl, T. E.:
Successful small bowel allotransplantation in dogs with cyclosporine and prednisone. Transplantation, 37:126, 1984.
23. Downing, R.: Historical review of pancreatic islet transplantation. World J. Surg., 8:137, 1984.
24. Dubernard, J. M., Traeger, J., Neyra, P., Touraine, J. L., Tranchant, D., and Blanc-Bruant, N.: A new method of preparation of segmental
pancreatic grafts for transplantation: Trials in dogs and in man. Surgery, 84:633, 1978.
25. Fortner,
J. G., Sichuk, G., Litwin, S. D., and Beattie, E. J., Jr.: Immunological responses to an intestinal allograft with HL-A-identical
donor-recipient. Transplantation, 14:531, 1972.
26. Gliedman, M. L., Gold, M., Whittaker, J., Rifkin, H., Soberman, R., Freed, S., Tellis, V., and Veith, F. J.: Pancreatic duct to ureter

anastomosis in pancreas transplantation. Am. J. Surg., 125:245, 1973.
389
27. Gorer, P. A.: Transplantese. Ann. N.Y. Acad. Sci., 87:604, 1960.
28. Griepp, R. B., and Ergin, M. A.: The history of heart transplantation. Heart Transplant., 3:145, 1984.
29. Groth,C., Lundgren, G., Arner, P., Collste, H., Hardstedt, C., Lewander, R., and Ostman, J.: Rejection of isolated pancreatic allografts in
patients with diabetes. Surg. Gynecol. Obstet., 143:933, 1976.
30. Guthrie, D.: A History of Medicine. Philadelphia, J. B. Lippincott, 1946, p. 12.
31. Hamburger,J., Vaysse, J., Crosnier, J., Auver, J., M., and Hopper, J., Jr.: Renal homotransplantation in man after radiation of the recipient.
Am. J. Med., 32:854, 1962.
32. Hamilton,D.: Kidney transplantation: A history. In Morris, P. J. (Ed.): Kidney Transplantation: Principles and Practice. London, Grune &
Stratton, 1984.
33. Hardy,J. D., Chavez, C. M., Kurrus, F. D., Neely, W. A., Erasian, S., Turner, M. D., Fabian, L. W., and Labeckiz, T.: Heart
transplantation in man: Developmental studies and report of a case. JAMA, 188:1132, 1964.
34. Holman, E.: Protein sensitization in isoskin grafting. Gynecol. Obstet., 38:100, 1924.
35. Homer: The Iliad, Book 6.
36. Humphries,
A. L., and Dennis, A. J., Jr.. Historica1 developments in preservation. In Toledo-Pereyra, L. H. (Ed.): Basic Concepts of
Organ Procurement, Perfusion and Preservation for Transplantation. New York, Academic Press, 1982, p. 1.
37. Kahan, B. D.: Cosmas and Damian revisited. Transplant. Proc., 15:2211, 1983.
38. Lillehei,
R. C., Idezuki, Y., Feemster, J. A., Dietzman, R. H., Kelly, W. D., Merkel, F. K., Goetz, F. C., Lyons, G. W., and Manax, W. G.:
Transplantation of stomach, intestine, and pancreas: Experimental and clinical observations. Surgery, 62:721, 1967.
39. Lillehei,
R. C., Simmons, R. L., Najarian, J. S., Weil, R., Uchida, H., Ruiz, J. O., Kjellstrand, C. M., and Goetz, F. C.: Pancreaticoduodenal
allotransplantation: Experimental and clinical experience. Ann. Surg., 172:405, 1970.
40. Mann, F. C., Priestley, J. T., Markowitz, J., and Yater, W. M.: Transplantation of the intact mammalian heart. Arch. Surg., 26:219, 1933.
41. Merrill, J. P.: Early days of the artificial kidney and transplantation. Transplant. Proc., 13(Suppl. 1):4, 1981.
42. Moore, P. D.: Give and Take: The Development of Tissue Transplantation. Philadelphia, W. B. Saunders, 1964.
43. Murray, J. E.: Remembrances of the early days of renal transplantation. Transplant. Proc., 13(Suppl. 1):9, 1981.
44. Najarian,
J. S., Sutherland, D. E. R., Matas, A. J., Steffes, M. W., Simmons, R. L., and Goetz, F. C.: Human islet transplantation: A
preliminary experience. Transplant. Proc., 9:233, 1977.
45. Owen, R. D.: Immunogenetic consequences of vascular anastomoses between bovine twins. Science, 102:400, 1945.
46. Peer, L. A.: Transplantation of Tissues. Baltimore, Williams & Wilkins, 1955.
47. Raju,
S., Didlake, R. H., Gayirli, M., Turner, M. D., Grogan, J. B., and Achord, J.: Experimental small bowel transplantation utilizing
cyclosporine. Transplantation, 38:561, 1984.
48. Reemtsma, K., McCracken, B. H., Schlegel, J. U., and Pearl, M.: Heterotransplantation of the kidney: Two clinical experiences. Science,

143:700, 1964.
49. Reitz,
B. A., Wallwork, J. L., Hunt, S. A., Pennock, J. L., Billingham, M. E., Oyer, P. E., Stinson, E. B., and Shumway, N. E.: Heart-lung
transplantation: Successful therapy for patients with pulmonary vascular disease. N. Engl. J. Med., 306:557, 1982.
50. Schwartz, R., and Dameshek, W.: Drug-induced immunological tolerance. Nature, 183:1682, 1959.
51. Sollinger,H. W., Pirsch, J. D., D'Alessandro, A. M., Kalayoglu, M., and Belzer, F. O.: Advantages of bladder drainage in pancreas
transplantation: A personal view. Clin. Transplant., 4:32, 1990.
52. Starzl,
T. E., Iwatsuki, S., Van Thiel, D. H., Gartner, J. C., Zitelli, B. J., Malatack, J. J., Schade, R., R., Shaw, B. W., Jr., Hakala, T. R.,
Rosenthal, J. T., and Porter, K. A.: Evolution of liver transplantation. Hepatology, 2:614, 1982.
53. Sutherland,D. E. R., Goetz, F. C., Elick, B. A., and Najarian, J. S.: Experiments with 49 segmental pancreatic transplants in 45 diabetic
patients. Transplantation, 34:330, 1982.
54. Sutherland, D. E. R., Goetz, F. C., and Najarian, J. S.: Intraperitoneal transplantation of immediate vascularized segmental grafts without
duct ligation: A clinical trial. Transplantation, 28:485, 1979.
55. Sutherland,
D. E. R., Goetz, F. C., and Najarian, J. S.: Living related donor segmental pancreatectomy for transplantation. Transplant
Proc., 12(Suppl. 2):19, 1980.
56. Tagliacozzi, G.: De curtorum chirurgia per instionem. Venice, 1597, p. 61.
57. Terasaki, P. I., and McClelland, J. D.: Microdroplet assay of human serum cytotoxins. Nature, 204:998, 1964.
58. The Toronto Lung Transplant Group: Unilateral lung transplantation for pulmonary fibrosis. N. Engl. J. Med., 314:1140, 1986.
59. Ullmann, E.: Experimentelle Nierentransplantation. Wien. Klin. Wochenschr., 15:281, 1902.
60. Veith, F. J.: Lung transplantation. Surg. Clin. North Am., 58:357, 1978.
61. Veith, F. J., Kamholz, S. L., Mollenkopf, F. P., and Montefusco, C. M.: Lung transplantation 1983. Transplantation, 35:271, 1983.
62. Woodruff, M. F. A.: The Transplantation of Tissues and Organs. Springfield, Ill., Charles C. Thomas, 1960.
63. Worshofsky, F.: The Rebuilt Man. New York, Thomas Y. Crowell, 1965.


Sabiston: Textbook of Surgery, 15th ed., Copyright 1997 W. B. Saunders Company
VIII - LIVER TRANSPLANTATION

Pierre A. Clavien M.D., Ph.D.
Allan D. Kirk M.D., Ph.D.
INDICATIONS FOR TRANSPLANTATION
The specific indications for liver transplantation have become more standardized (Tables 20-15 and
20-16) owing to the better knowledge of the natural history of a number of liver diseases and
concomitant improvement in short- and long-term results of orthotopic liver transplantation (OLT). [2] [6]
[12] As with most aspects of this topic, however, significant controversy remains about several disease
states leading to end-stage liver failure.
Cholestatic Liver Diseases.
Cholestatic diseases, including primary biliary cirrhosis (PBC), secondary biliary cirrhosis, and primary
sclerosing cholangitis (PSC), are the diseases most successfully treated by liver transplantation. [7] [25]
Operative
TABLE 20-15 -- Adult Indications for Orthotopic Liver Transplantations
Data from Belle, S. H., Beringer, K. C., and Detre, K. M.: Trends in liver transplantation in the United
States. In Terasaki, P. I., and Cecka, J. M. (Eds.): Clinical Transplants. Los Angeles, UCLA Tissue
Typing Laboratory, 1993.
Percentage
Before 1991 Percent 1992
Indication n=4833 n=2315
Other cirrhosis 55.9 65.2
Alcoholic 17.5 21.1
Hepatitis B 7.2 5.1
Fulminant hepatic failure 7.0 5.5


TABLE 20-16 -- Pediatric Indications for Orthotopic Liver Transplantation

Data from Belle, S. H., Beringer, K. C., and Detre, K. M.: Trends in liver transplantation in the United
States. In Terasaki, P. I., and Cecka, J. M. (Eds.): Clinical Transplants. Los Angeles, UCLA Tissue
Typing Laboratory, 1993.
Percentage
Before 1991 Percent 1992
Indication n=1117 n=394
Fulminant hepatic failure 9.4 14.0
Cirrhosis 7.8 7.4
Hepatitis B 0.4 0.5
Alcoholic 0.0 0.0
survival is greater than 90%, and 5-year survival is approximately 80%. [6] This rate is due mostly to the
slow onset of illness, which allows for preoperative preparation of the recipient and appropriate timing of
the procedure, as well as absence of recurrence of the disease in the graft. As a result of this success,
cholestatic disease is one of the most frequent indications for transplantation, accounting for 22% of all
transplants performed in the United States. Clearly, the timing of operation is critical, and a substantial
literature has addressed the rate of disease progression.
The indications for transplantation in patients with PSC are based on refractory symptoms resulting from
recurrent cholangitis or impaired synthetic function. The risk of cholangiocarcinoma, approximately 15%
in these patients, is an
462

additional relative indication for OLT. A current trend toward earlier intervention with liver
transplantation in PSC is underway, with retrospective comparisons of transplant versus nontransplant
therapy weighing heavily in favor of the transplant approach. [21] Clearly, however, some patients,
particularly those with predominantly extrahepatic biliary strictures do well with conventional bypass [42]
or endoscopic procedures. [30] Thus, timing remains a matter of clinical judgment, with most patients
benefiting from allotransplantation.
Several models of survival have been established to predict outcome in patients with PBC. [7] [27] All
prognostic scores are based in part on the rise in serum bilirubin. In general, transplantation is
recommended when the bilirubin level exceeds 15 mg./100 ml., although earlier transplantation for
severe associated symptoms such as pruritus or fatigue, refractory ascites, or variceal bleeding is
appropriate. Although close monitoring of asymptomatic patients is reasonable, it is important not to
delay intervention until the clinical status of the patient deteriorates, because OLT is the only definitive
treatment. [7]
Nineteen percent of all liver transplants performed in the United States have been for alcoholic cirrhosis.
[6] Although the procedure was originally avoided because of fears of high postoperative recidivism and a
tendency toward medical noncompliance, the national trend has been to transplant in this setting more
frequently, and in 1993 it was the most common indication reported by the United Network for Organ
Sharing (UNOS) registry. Experience has demonstrated that graft and patient survival are not
significantly different from those of other favorable indications and that disease recurrence (return to
heavy alcohol use) is approximately 12%. [29] Given these results, combined with the extraordinary
prevalence of alcohol-related liver disease compared with other transplantable diseases (36,000 deaths
per year), transplantation for alcoholic cirrhosis could easily exhaust the already scarce supply of donor
organs. Selection is clearly needed to exclude comorbidities, alcohol-associated organ failure such as
cardiomyopathy, and patients with continued dependence on alcohol. Even so, the need far outpaces the
supply. Considerable philosophical debate addresses the use of scarce resources for individuals with
self-inflicted disease. At this point, the decision to transplant should be based on a thorough preoperative
medical evaluation supplemented by involvement in alcoholic rehabilitation and abstinence for at least 6
months. Patients presenting too sick to drink must be evaluated individually with an estimate of
recidivism made to optimize long-term results. Palliative procedures are appropriate when possible. The
recent development of the transjugular intrahepatic portosystemic shunt (TIPS) procedure has allowed
many of these patients to be stabilized without extensive surgical intervention, for more thorough
evaluation in a nonemergent setting.
Hepatitis C and Cryptogenic Cirrhosis.
Another rapidly increasing indication for OLT is hepatitis caused by the hepatitis C virus (HCV). This is
frequently discussed in tandem with cryptogenic cirrhosis. Transplantation is pursued for symptomatic
infection only. Since the discovery of reliable methods for detecting HCV, this pathogen has been
identified as the etiologic agent for liver failure in an increasing number of recipients. [5] In 1996, it was
associated with more than 25% of transplants in most centers. Cryptogenic disease accounted for an
additional 12%. Excellent short-term results have been achieved with a 3-year survival of 70% (63%
nonretransplant survival). These results have been tempered, however, by reinfection at almost 90%,
with active recurrent hepatitis at 50%. Recurrent hepatitis has led to few deaths, although a fourth of the

reinfected patients have converted to a chronic state. More deaths from recurrent disease can be expected
in this group, and longer follow-up is required to determine whether results will remain acceptable.
Disease recurrence has also been seen in cryptogenic cases, again suggesting a viral infection. Several
treatments have been studied for HCV infection, with the most promising being parenteral therapy with
interferon alfa.
Metabolic Diseases.
Several metabolic diseases have been successfully cured with OLT, including alpha1 -antitrypsin
deficiency, Wilson's disease, hemochromatosis, Crigler-Najjar syndrome, tyrosinemia, primary
hyperoxaluria, and familial homozygous hypercholesterolemia. Survival is uniformly excellent. The
recipient hepatectomy is usually simple, and the preoperative state of the patient is stable. Timing of
transplantation should be such that secondary effects of the disease are not extensive.
In the absence of pre-existing liver disease, rapid loss of hepatic function culminating in jaundice and
coma is termed fulminant hepatic failure (FHF). The etiologic agent can vary. The time course relates
inversely to the prognosis, with those progressing to coma in less than 2 weeks having a 36% survival
and those progressing in 2 to 8 weeks having a 7% survival. Despite the dramatic rate of progression of
disease, these patients have very acceptable results from OLT (about 60% 5-year survival). [6] [37] [56]
Predictably, survival after OLT improves with improved health of the patient at the time of transplant.
Survival for all etiologies of FHF has doubled in the past decade, primarily due to great strides in
supportive intensive care. Unlike chronic viral diseases, recurrence of disease for viral causes of FHF is
unusual.
Care of patients with FHF is complex, and OLT should be considered early. Thus, prompt transfer to a
liver transplantation center is critical. In patients with rapid deterioration, supportive care (e.g.,
hyperventilation, diuresis) should be supplemented with monitoring of intracranial pressure (ICP) for
best results. Subdural ICP monitors have obviated the need for intraventricular devices and have thus
reduced the bleeding risks of this approach. Perfusion pressures (mean arterial pressure minus ICP) of
less than 40 mm. Hg, especially when combined with pupillary fixation, suggest that irreversible brain
injury has occurred, and OLT should be avoided. Patient decompensation can be rapid and unpredictable.
In the absence of a suitable donor, several aggressive maneuvers can successfully delay the onset of brain
death from intracranial pressure. These include hepatectomy with end-to-side portacaval shunt,
xenogeneic ex vivo perfusion, and bioartificial hepatic support (see later). Donor criteria may be
liberalized to achieve timely hepatic replacement, and ABO incompatibility is acceptable.
Transplantation for chronic hepatitis B virus (HBV) cirrhosis remains a controversial topic. [39] [55] The
discouraging reinfection rate of over 80% of recipients associated with a high rate of clinical hepatitis
recurrence (60% at 1 year) and high related mortality (30% at 1 year and 52% at 5 years) raises questions
about the procedure. [6] [14] Although generally contraindicated, transplantation may be appropriate in
certain settings. Recent intense investigation in this area has identified several factors affecting
reinfection. In general, the state of viral replication at the time of transplantation is critical in establishing
reinfection. Identification of active viral replication (serum HBV DNA and HBeAg detection) at the time

of transplantation clearly worsens the prognosis. Immunoprophylaxis with anti-HBV antibodies

improves outcome for replication-negative patients, but long-term results for patients so treated remain to
be seen. In general, OLT in the presence of HBV infection should be reserved for patients enrolled in
specific viral prophylaxis protocols.
Intrahepatic Malignancy.
Predictably, transplantation for primary and metastatic cancer has been associated with a
463
high recurrence of tumor and a poor 5-year survival. [6] [40] Thus, there is little indication for OLT in a
patient with known malignancy. Liver transplantation is no longer an accepted therapy for hepatic
metastasis except perhaps in some patients with rare neuroendocrine tumors. [19] Today, liver
transplantation is usually limited to patients with nonresectable hepatoma fulfilling the following criteria:
asymptomatic hepatoma (i.e., not associated with recent weight loss, ascites, or constitutional
symptoms); tumor less than 5 cm. in diameter; and fewer than or equal to three intrahepatic tumors. If
feasible, liver resection should always be considered first. Extensive evaluation is required to exclude the
presence of extrahepatic spread. Transplantation for cholangiocarcinoma is even more controversial and
should probably be performed only under the guidance of specific protocols. The role of adjuvant
chemotherapy or chemoembolization in patients undergoing transplantation for cancer remains unknown.
The most common indication for liver transplantation in children is biliary atresia (see Table 20-16) .
This diagnosis accounts for 55% of the pediatric recipients in the UNOS registry. [6] [41] Although the
creation of a portoenterostomy (Kasai procedure) remains the standard initial treatment, long-term
survival without eventual hepatic failure is uncommon. Five-year survival remains under 50%. For this
reason, many have proposed that early intervention with OLT be considered. The results of liver
transplantation in infants have been somewhat better, with 5-year patient survival of 64%. Thus, a
reasonable course of action appears to be neonatal nontransplant surgical intervention, with
transplantation reserved for those children developing hepatic insufficiency despite a Kasai procedure.
[38] The interim growth improves the donor pool substantially and decreases the technical difficulties
inherent in the management of small children. Five-year survival improves to 74% at 3 to 5 years and
79% from 5 to 15 years. However, multiple reoperations and revisions after an initial portoenterostomy
should be avoided because they rarely provide long-term disease-free survival and substantially hinder
efforts to transplant the patient. The remaining usual indications for pediatric transplantation also occur
in the adult population, with 13% performed for metabolic diseases (see above) and 10% associated with
fulminant hepatic failure secondary to hepatitis of various etiologies.
SPECIFIC PATIENT SELECTION AND PREOPERATIVE CONSIDERATIONS
Patient Selection.
The success of liver transplantation is closely related to the rational selection of patients most likely to
benefit from the procedure. In fact, the single most important prognostic factor affecting survival is the

medical condition of the recipient at the time of OLT (Fig. 20-42) . [6] As with the recipients of other
organs, patients should be without additional end-organ failure (other than that clearly related to hepatic
insufficiency) and should be candidates for a major operative intervention. In general, dependence on
alcohol or other harmful substances should be resolved for at least 6 months. Extrahepatic malignancy,
sepsis, and diffuse mesenteric venous thrombosis represent absolute contraindications. Isolated portal
vein thrombosis is a relative contraindication. The patient's liver function should be such that
complications of dysfunction are emerging, with the predicted life span of the patient managed medically
less than 2 years. With the improvements in survival after allotransplantation realized in the past decade,
it is also appropriate to consider patients with metabolic diseases or moderately advanced liver disease
with significant alterations in quality of life such as extreme fatigue, refractory pruritus, or
encephalopathy.
Figure 20-42 United Network for Organ Sharing (UNOS) status and survival after orthotopic liver
transplantation (for definition of UNOS status, see text).
Once advanced liver disease is identified, efforts to proceed with transplantation should be initiated,
because the preoperative health of the recipient greatly affects the chance for success. Early evaluation
by a multidisciplinary team schooled in the specific requirements for liver transplantation is critical. In
addition to surgical evaluation, the selection of appropriate candidates for transplantation should be based
on input from medical colleagues, including a hepatologist and infectious disease specialist. Evaluation
by a social worker and psychiatrist to establish the ability of the patient and family to manage themselves
postoperatively is important. Immediate efforts should optimize the candidate's nutritional status and
overall medical condition. Finally, education of the patient must be initiated early to ensure that
complications of medical noncompliance are avoided postoperatively.
For all liver recipients, thorough preoperative evaluation to determine the antibody titer directed against
hepatitis A, B, and C as well as cytomegalovirus, human immunodeficiency virus, Epstein-Barr virus,
and herpes simplex is mandatory. In addition, viral antigen detection is required to identify active
infection. The role of transplantation for primary viral hepatitis was discussed previously. Active
infection with cytomegalovirus or herpesvirus requires clinical resolution before transplantation, but viral
carriers can receive transplants given appropriate antiviral prophylaxis. Human immunodeficiency virus
infection contraindicates OLT. In addition to the recipient's status, the donor's status for these viruses
should be ascertained before implantation, again to allow for antiviral prophylaxis when indicated.
Several alternative therapies should be considered in the candidate for transplantation, not necessarily to
obviate the need for transplantation but to improve the medical condition before transplantation. Patients
with good synthetic function (normal bilirubin, normal coagulation, normal albumin), such as an
individual with Child's A cirrhosis and recurrent variceal bleeding, may benefit from nontransplant
surgical palliation such as a Warren shunt. The TIPS procedure is also a useful palliative step for
refractory complications of portal hypertension. Stabilization of disease with these interventions gives
more time to evaluate patients and optimize preoperative conditions. It allows for a more thorough

evaluation of a patient's ability to comply with the posttransplant medical regimen, particularly alcoholic
patients who present abstinent because they are too sick to
464
drink. This compliance alone may stabilize the disease in an alcoholic patient with Child's A cirrhosis
and avoid the need for transplantation. Reduction in the degree of portal hypertension may also minimize
intraoperative bleeding during the recipient hepatectomy. Problems with these palliative procedures are
frequent and require surveillance with particular attention to portal vein patency. Identified candidates
should be treated at a center with transplant expertise to facilitate palliation that does not jeopardize
future transplantation. Alternative treatment should not delay transplantation.
SELECTION OF DONORS
Immediate function of a transplanted liver is imperative. Unlike kidney, pancreas, or, to some extent,
heart transplantation, no artificial means is readily available to support an anhepatic patient in the event
of graft failure. Without a rapid restoration of synthetic function, death from bleeding or cerebral edema
generally ensues within 72 hours. The single most important factor determining the early function of a
liver allograft is the viability of the donor liver. Although this seems obvious, determining the state of the
liver in a heart-beating cadaver remains imprecise, so careful attention to the conditions of the donor's
death, the morphology and function of the organ before harvest, and the specifics of the extirpation is
critical. Failure of a graft to function at all after a technically successful transplantation is known as
primary nonfunction (PNF). The only treatment for PNF is retransplantation within 24 to 72 hours.
Factors contributing to the development of PNF include parenchymal insufficiency unrecognized at the
time of harvest, graft injury during the harvest or cadaver resuscitation, preservation injury or prolonged
cold ischemia, prolonged rewarming time, and reperfusion injury after implantation. [15]
Several factors have been investigated to aid in the prediction of PNF. The most widely noted is the
estimated parenchymal fat content. [18] [53] Donor liver biopsy specimens that show a 40% or greater
parenchymal replacement by fat have a higher chance of PNF, and in some settings this is a reasonable
indicator of the adequacy of a donor organ. This must, however, be balanced against the severity of the
recipient's disease and the urgency of the planned operation. Other factors used to evaluate the donor's
status include the age of the donor, the level of inotropic support, the mechanism of the donor's death, the
level of hepatic or intra-abdominal trauma, the presence of hypernatremia, and the biochemical studies of
liver function. No single parameter has been established absolutely governing the acceptance of a donor
for organ harvest. [11] [50] Rather, combinations of risk factors are generally responsible for discarding a
potential donor liver. In case of doubt about the quality of an organ, the personal inspection of the liver
by an experienced transplant surgeon is often critical for decision about the use of an organ.
Donor use has been improved with the introduction of UW preservation solution as mentioned earlier.
Until 1987 the outer limit for cold storage of the liver using EuroCollins solution was about 8 hours. Use
of UW solution has extended the time to about 24 hours, allowing for better allocation of the organ and
preparation of the recipient. The specifics of this and other preservation solutions are detailed in Part VII
of this chapter.
The considerations of ABO typing and other immunologic concerns are discussed later. An additional
issue is size compatibility. Smaller organs are easily adapted to a large recipient, but the converse is not

true. One advance has been the use of reduced-size allografts, particularly in children. Usually the left
lateral lobe (segments 2 and 3) or left lobe (segments 2, 3, and 4) is used, allowing up to 1/10 weight
mismatch. This has been the most important factor in the success of transplantation in small children.
This technique is sometimes used in adults when an emergency transplantation is required. Finally, male
recipients of female organs have 10% worse survival than other sex-match combinations, according to
the 1994 UNOS registry.
OPERATION ON THE DONOR
Harvest of the donor organ should be performed by an experienced surgeon, with particular care taken to
optimize the preharvest resuscitation of the heart-beating cadaver. Because immediate hepatic function
after transplantation is required, there is no room for error. Errors in resuscitation include injudicious use
of vasopressors, prolonged acidosis, and hypoperfusion from hypovolemia. Visual inspection and
palpation of the liver with knowledge of the potential recipient's status and size aid in the assessment of
the appropriateness of a donor liver.
The liver is generally procured through a midline incision from jugular notch to pubis, including median
sternotomy (Fig. 20-43) . Harvest is coordinated with the harvest teams for other organs. Complete
mobilization of the liver is required, including division of both triangular ligaments and the falciform
ligament. The hepatogastric ligament is divided. Particular attention is paid to the vascular supply,
including preservation of aberrant hepatic arteries (20% aberrant right from the superior mesenteric
artery, 15% aberrant left from the left gastric artery). Unlike the relative impunity associated with arterial
ligation in nontransplanted livers, failure to complete the arterial revascularization due to an
unrecognized arterial supply is poorly tolerated after cold storage.
Figure 20-43 The donor procedure. UW, University of Wisconsin.
465
PNF, late biliary stricture, or intrahepatic abscess may result. Additional care must be taken when
simultaneous pancreas harvest is performed to avoid hepatic artery injury or portal vein transection. The
suprahepatic vena cava should be preserved for the hepatic graft, and care should be taken to avoid caval
injury during cardiac retrieval. The gallbladder is removed either in situ or after explantation.
Once mobilization is completed, perfusion with UW solution is initiated through the distal aorta or
common iliac arteries with ligation of the supraceliac aorta. The inferior mesenteric vein or splenic vein
is used for portal perfusion. When pancreatic harvest is performed, some surgeons prefer to use the portal
vein for cold perfusion. UW solution is also flushed into the biliary tree. Topical slush is rapidly applied.

Hepatic extirpation is performed after cardiopulmonary retrieval and before harvest of the pancreas and
kidney. The iliac artery and vein should be harvested in the event that vascular reconstruction is required.
IMMEDIATE PREOPERATIVE MANAGEMENT
Patients awaiting liver transplantation are prioritized on a national waiting list based on severity of
disease as defined by the UNOS. Status 4 patients are at home and functioning normally, status 3 patients
can be at home but require continuous medical care. Status 2 patients are continuously hospitalized in an
acute care unit for at least 5 days or are in the intensive care unit. Status 1 patients are in the intensive
care unit because of acute or chronic liver failure with a life expectancy without a liver transplant of less
than 7 days. Status 7 is reserved for patients taken from the active list for a temporary medical
contraindication such as sepsis. [1] These patients can accrue points for time on the list despite not being
able to accept an organ. Strict adherence to guidelines for proper categorization is required to allow for
ethical allocation of organs. As soon as a donor is identified, the organ is paired with a potential
recipient, who is called to the hospital for preoperative evaluation before organ harvest.
Appropriate perioperative management of a transplant recipient begins with a thorough preoperative
physical examination (including rectal and dental examinations) to rule out the possibility of ongoing
infection or malignancy. Routine screening includes a complete blood cell count, electrolyte and
metabolic profile, urinalysis, and chest film. Blood is crossmatched. Intravenous lines, including arterial
lines and pulmonary artery catheters, should be placed with strict attention to aseptic technique because
cutaneous contamination with bacteria or fungi can cause serious postoperative complications in the
transplant patient. Selective decontamination of the gut and mechanical preparation of the bowel are
advocated at many centers, as is a shower with an antimicrobial soap.
For patients in the intensive care unit, supportive measures are continued as needed. Of specific concern
is the potential for neurologic recovery in patients with advanced encephalopathy and cerebral edema. A
recent advance has been the use of intracranial pressure monitoring in the perioperative period. This was
once shunned as a prohibitive risk for cerebral bleeding in the coagulopathic fulminant hepatic failure
patient, but routine assessment is now possible through minimally invasive techniques. Small-gauge
catheter pressure monitors can be placed into the subdural space without the need for parenchymal
puncture, thus decreasing the risk of the intervention substantially. Greatly elevated intracerebral
pressures, especially in association with signs of transtentorial herniation in a Grade 5 coma patient,
suggest an irreversible lesion that is unlikely to resolve after transplantation.
OPERATION ON THE RECIPIENT
Few surgical procedures require the fastidious attention to technical detail required in liver
transplantation. Technical errors are translated directly into infectious complications or marginal biliary
function. Thus, transplantation should be performed only by surgeons proficient in the procedure. In
addition, the operative environment should include experienced nursing and ancillary support.
Intraoperative management by a knowledgeable anesthesiologist with experience in liver transplantation
is critical for a successful technical result. The procedure presents the challenge of maintaining
homeostasis of temperature, circulation (including oxygen-carrying capacity and coagulation
competence), gluconeogenesis, and electrolyte concentration while establishing adequate anesthesia and
paralysis with agents not requiring hepatic function for degradation. Intraoperative ICP monitoring is

appropriate for patients with severe encephalopathy. In procedures using a venovenous bypass, a
perfusionist is required, and procedures performed without bypass require adequate maintenance of
preload during caval occlusion and correction of metabolic abnormalities after release of the congested
portal circulation. After the initial function of the allograft, the most important factor predictive of
technical success is the stability of the patient intraoperatively and his or her delivery to the intensive
care unit normothermic with adequate circulatory competence.
Successful engraftment of the organ begins with a controlled recipient hepatectomy. This can be a
formidable task in patients with severe portal hypertension and extensive collateral formation or in those
with multiple operative interventions. In general, extirpation follows the basic surgical guidelines of
establishing proximal and distal vascular control combined with lysis of all ligamentous attachments.
Specific technical concerns include retaining maximal length on all vessels. Mobilization of the common
bile duct depends on the planned biliary reconstruction (choledochocholedochostomy versus
choledochojejunostomy). Care to avoid injury to the right adrenal vein during caval dissection is
important. If venovenous bypass is planned, cannulation of the left axillary, femoral, and portal veins is
performed.
The decision to place the recipient on bypass is routine at a number of centers; and liberal use of this
technique, originally developed by Shaw and colleagues, [2] [3] has clearly led to improved operative
mortality. Bypass avoids mesenteric congestion and minimizes the release of lactate and other
by-products of hypoperfusion into the portal circulation. In addition, it improves venous return to the
heart during implantation and thus improves hemodynamic stability during the period of caval occlusion.
Bypass also diverts the portal flow during difficult recipient hepatectomies to minimize blood loss,
particularly during dissection of the retrohepatic cava and bare spot. [11] Despite these benefits, it is now
clear that many patients tolerate OLT without the additional manipulation required by bypass. It has thus
become policy at many centers to employ this technique selectively after an intraoperative trial of portal
vein and vena caval occlusion.
The implantation procedure (Fig. 20-44) begins with the suprahepatic vena caval anastomosis followed
by the infrahepatic caval anastomosis. Alternatively, the donor vena cava can be anastomosed side to
side with the recipient vena cava if it is left in situ during the recipient hepatectomy (piggyback
technique). The operation then proceeds to the portal anastomosis. After all venous connections, the liver
is reperfused with the suprahepatic vena cava temporarily occluded and the infrahepatic vena cava vented
to allow washout of the hyperkalemic and adenosine-rich UW solution. The hepatic artery anastomosis is
the final vascular step in the procedure.
466
Figure 20-44 The recipient procedure.
Some groups perform simultaneous arterial and venous reperfusion. [43]

The biliary reconstruction remains an additional area of debate. [54] Options include
choledochocholedochostomy with or without externalized T-tube stents. This has the advantage of easy
access for subsequent biliary manipulations or evaluation of bile, as well as preservation of the sphincter
mechanism. Unfortunately, the anastomosis is particularly sensitive to ischemic injury of the common
duct, and complications of leak when the T tube is removed remain vexing. [31] The incidence of
technical complications related to this method ranges from 12% to 50% in published series, with the
cumulative average being 25%. [36] These problems are obviated for the most part by use of a Roux-en-Y
choledochojejunostomy at the expense of convenient biliary access. [8] Leak or stricture is still observed
in 4% to 30% of cases (mean 14%). [36] This method is frequently used in pediatric transplants. One
report of 300 transplants performed with a side-to-side choledochocholedochostomy technique has
reported a remarkable 2.2% technical complication rate. [36] This technique awaits confirmation in other
centers.
POSTOPERATIVE MANAGEMENT
Management in the postoperative intensive care unit is similar to that after any major procedure.
Ventilatory support and volume replacement are standard. Isolation is not required beyond standard
universal precautions. No sedation is given until extubation. For unclear reasons, postoperative pain is
usually mild, and any major discomfort should alert one for possible complications. Close monitoring of
serologic liver enzymes is critical because increasing enzymes or a failure of enzyme values to correct
rapidly suggests PNF or technical complications such as hepatic artery thrombosis. Liberal use of the
Doppler ultrasonography and rapid return to the operating room are mandatory in these situations
because early detection is the only factor separating a return to normal liver function from complete graft
necrosis and patient death.
Use of drains and antibiotics is no different in this operation from that in any other major abdominal
procedure. Closed-suction drains should be used and removed early after the threat of postoperative
hemorrhage is over. Brief antibiotic prophylaxis is appropriate with an agent with adequate skin and
biliary organism coverage. Prolonged use of prophylactic antibiotics is contraindicated. A protocol for
decontaminating the small bowel can be used but is usually discontinued within a few weeks of
transplantation. At Duke University Medical Center, decontamination (colistin, 100 mg./10 ml.;
gentamicin, 80 mg./10 ml.; nystatin, 2 106 /10 ml.) is discontinued at the time of normal enteral
feeding, often within the first postoperative week).
If a T tube is used in the biliary reconstruction, a T-tube cholangiogram is obtained within a week with
internalization contingent on a normal result. Patients are discharged when they are familiar with their
medications. Patients living more than 2 hours from the transplant center generally stay in the vicinity for
an additional 2 to 4 weeks. Close monitoring of hepatic function and medical compliance is continued
twice a week for 4 weeks and weekly for an additional 4 weeks. After this outpatient evaluation, patients
are returned to their referring community for chronic follow-up. It is important to establish open lines of
communication between the community physician and the transplant center to ensure prompt recognition
and referral of postoperative complications.
IMMUNOLOGIC MANAGEMENT
The liver must be considered separately from other solid organs with regard to immunologic
management. Many well-established concepts of donor-host interaction after kidney or heart

transplantation do not apply after hepatic transplantation, and failure to recognize the unique properties
of this situation can lead to detrimental perioperative treatment. [35]
HLA Typing.
HLA matching is not feasible before liver transplantation. However, because new techniques have been
developed that allow donor HLA data to be provided in a timely manner, the future use of these data is
pertinent. Although matching donors and recipients with regard to HLA type clearly improves outcome
after kidney, heart, and pancreas transplantation, no such correlation exists with liver transplantation.
Indeed, matching may, in fact, reduce overall survival. [33] [49] The reasons for this lie in the dualistic
nature of HLA in the pathophysiology of liver disease. T-cell-mediated rejection of the organ is
mechanistically the same as with other organs, so rejection is reduced with improved HLA compatibility.
However, the physiologic role of HLA is to present viral peptides to T cells to initiate destruction of
virally infected cells. Thus, HLA compatibility potentiates
467
the inflammation during viral reinfection after transplantation for viral hepatitis and increases the chance
for clinical recurrence of the original disease. Similarly, T-cell-mediated autoimmune diseases (e.g.,
PBC) are etiologically based on T-cell recognition of HLA presented peptides. Therefore, recurrence of
autoimmune diseases may be potentiated as well. Further knowledge regarding specific disease states
worsened by certain HLA matches may be useful in selective typing in the future.
Crossmatch and ABO Matching.
The lymphocytotoxic crossmatch is not used prospectively before liver transplantation. Again, temporal
concerns are most pressing; but in this case, the value of a positive crossmatch in predicting subsequent
poor outcome from hyperacute rejection is minimal. Indeed, hyperacute rejection is rarely seen even in
the face of documented preformed antibodies and ABO incompatibility. [22] [24] [49] The reasons for this
remain controversial because hyperacute rejection can be readily produced experimentally. [28] Although
preformed antibodies reduce long-term graft survival somewhat, early results appear to be minimally
affected. Grafts unmatched for ABO antigens can cause antibody-mediated graft-versus-host disease with
mild hemolytic anemia and fever occurring between postoperative days 5 and 12. [44] This is the result of
intrahepatic B cells that secrete antibody directed against the recipient ABO antigens. Although it is
usually self-limited, increased immunosuppression may be required. Alternatively, some groups suggest
a decrease in immunosuppression to allow for a limited rejection of the offending B cells. It is preferred
that rules for ABO compatibility be followed for elective transplantation. In the emergent setting,
however, ABO-incompatible grafts can be used with acceptable results. The 5-year survival for grafts
that are ABO incompatible is approximately 15% lower than that for grafts without ABO discrepancy. [6]
This not only reflects some immune preference but also the patient population that receives ABO
mismatched grafts, that is, emergent transplants for FHF and PNF.
Acute Rejection.
As with other allografts, T-cell-mediated destruction of the liver is inevitable without

immunosuppressive therapy. The primary targets for T-cell recognition are HLA antigens on the biliary
epithelium and vascular endothelium. The characteristics of this rejection, termed acute rejection, are

similar to those of kidney or heart in that it develops in most of the cases during the first 6 months after
transplantation, usually within the first 4 postoperative weeks. More than half of patients develop at least
one episode of rejection. Symptoms are nonspecific, often including mild intermittent fever and general
malaise with alteration in liver tests. The diagnosis should be confirmed by liver biopsy. Most episodes
are readily reversible (90%), given prompt recognition and initiation of antirejection therapy. This
contrasts sharply to chronic rejection, discussed later.
Monitoring for acute rejection is a continuous process. Needle-core biopsy of the allograft is the best
diagnostic test (Fig. 20-45) . Histologically it appears as a predominantly T-cell and monocyte infiltrate
in the portal tracts, with subendothelial (endotheliitis) and biliary epithelial aggregates. Eosinophils and
polymorphonuclear leukocytes are present to a lesser degree but are more prevalent in hepatic rejection
than in infiltrate seen in other organs. Although protocol biopsies are performed in some centers, most
groups rely on monitoring of the liver function studies and/or the serum beta2 -microglobulin value, with
biopsy used to clarify detected abnormalities. Because a primary target in acute rejection is biliary
epithelium, it often presents initially as a cholestatic process with rapid increases in the alkaline
phosphatase and bilirubin values. However, changes in hepatic biochemical parameters are nonspecific
and can also indicate technical or infectious complications where alterations
Figure 20-45 Acute rejection after liver transplantation. Portal tracts contain a mixed inflammatory
infiltrate, with mononuclear cells and eosinophils. Bile duct damage is evident, and endothelium of a
branch of the portal vein is disrupted and infiltrated by lymphocytes (arrow). Hematoxylin and eosin,
original magnification 325. (Courtesy of Kay Washington, M.D., Department of Pathology, Duke
University Medical Center.)
in the immunosuppressive regimen would be ineffective or even detrimental. Thus, liberal use of biopsy
and Doppler ultrasound evaluation of hepatic blood flow and bile duct integrity is critical. Once acute
rejection is diagnosed, rescue immunosuppression is initiated as described later.
Chronic Rejection.
The development of liver allograft dysfunction over a period of months to years is termed chronic
rejection and, like other allografts, is controversial and multifactorial in its etiology, and is usually not
reversible. Histologically it appears as a paucity of bile duct epithelium without conspicuous lymphocytic
infiltration and has thus been described as the "vanishing bile duct syndrome" (Fig. 20-46) . Additionally,
an obliterative vasculopathy can occur with parenchymal fibrosis. The time course, histology, and
refractory nature of chronic rejection suggest that direct cell-mediated destruction is not a primary
mechanism. It is likely that the cumulative effects of mild subclinical immune recognition by several
limbs of the immune system, and the resulting
Figure 20-46 Vanishing bile duct syndrome as a manifestation of chronic rejection after liver
transplantation. Portal inflammation diminishes as interlobular bile ducts disappear. There is no bile
ductular proliferation. Hematoxylin and eosin, original magnification, 325. (Courtesy of Kay
Washington, M.D., Department of Pathology, Duke University Medical Center.)

468
exposure to soluble factors including fibrogenic cytokines, eventually take their toll on the fragile
epithelium. Chronic rejection often requires retransplantation.
Immunosuppressive Pharmacology.
Manipulation of the immune system is required to avoid graft loss from rejection. Identifying a safe,
effective, and minimally immunosuppressive regimen requires a careful balance aimed at reducing
infectious and neoplastic complications without a resultant increase in allograft rejection and/or
dysfunction. Thus, rational, selective use of several immunosuppressive agents is required to manage
successfully the broad array of patients who are transplanted. Three general classifications of
immunosuppression are used: (1) induction therapy, a relatively intense initial conditioning of the newly
transplanted recipient; (2) maintenance therapy, drugs given at minimal doses required to maintain graft
function; and (3) rescue therapy, heightened immunosuppression given to reverse an episode of acute
rejection. Most transplant centers have specific protocols for each situation. For example, the Duke
University liver transplant protocol is presented in Table 20-17 .
Several issues specific to liver transplantation should be mentioned. Of major importance is that the liver
appears to have less immunostimulatory antigenicity than other organs. The specifics of this perceived
resistance remain somewhat controversial but may relate to several peculiarities of hepatic physiology.
First, the portal circulation is exposed regularly to enteric pathogens and, more importantly, to absorbed
peptides,
TABLE 20-17 -- The Duke University Immunosuppressive Protocol for Orthotopic Liver
Transplantation
Induction Therapy
Methylprednisolone: Induction of anesthesia: 125 mg. IV Postoperatively: tapered from
300 to 20 mg./ day on 5th postoperative day
Cyclosporine: Intraoperatively: continuous IV infusion (2 mg./kg./
day)
Postoperatively: continuous IV infusion (4 mg./kg./
day) on arrival in the surgical intensive care unit
On postoperative days 2-3: oral cyclosporine (Neoral)
is started at 5 mg./kg. q12h. The IV cyclo-
sporine is decreased according to serum levels.
(Targeting postoperative levels: 250-300 ng./ml
[HPLC assay])
Azathioprine (1-1.5 mg./kg./day): Started as soon as the patient can
tolerate oral feeding, usually by
day 2 or 3
Maintenance Therapy

Prednisone: First 3 months: 20 mg./day

3-6 months: 15 mg./day
After 6 months: 5-10 mg./day
Cyclosporine: Targeted levels: initial 10 days: 250-300 ng./ml.;
10-30 days: 200-300 ng./100 ml.; 1-6 months:
150-200 ng./ml.; 6-12 months: 120-200 ng./ml.;
and after 1 year: 80-120 ng./ml.
Azathioprine: 1-2 mg./kg., usually discontinued within 9 months of
transplantation
Rescue Therapy for Biopsy-Proven Acute Cellular Rejection
First episode: 3-day recycling with IV methylprednisolone
500 mg./day
Second episode: repeated 3-day recycling with IV
methylprednisolone
Third episode: monoclonal antibody muromonab-CD3 (OKT3,
Orthoclone, Raritan, NJ) or antithymocyte globulin
(ATGAM, Upjohn, Kalamazoo, MI) is given for
10-15 days
Refractory rejection: cyclosporine (Neoral) is discontinued and
FK 506 (Prograf) (target levels 10-15 ng./ml.)
is initiated
which have potential antigenicity. A generalized perihepatic anergy has been postulated as protective in
this setting to prevent vascular thrombosis and inappropriate hepatic inflammation. Clearly the
reticuloendothelial system of the liver--the Kupffer cells--is important in establishing appropriate versus
inappropriate presentation of portal antigen. Most of the Kupffer cells of the liver are replaced with those
of recipient HLA type within a few weeks after transplantation. In addition, the Kupffer cells of the
donor have been described as migrating to other recipient tissues and establishing a state of chimerism.
Whether this chimerism creates specific tolerance or is the result of it remains to be seen. From a
practical standpoint, maintenance immunosuppression after liver transplantation can be comparatively
low, and many patients have been weaned to little or no immunosuppression for years. Also of
importance in this regard, the liver can tolerate brief periods on immunosuppression withdrawal if it
becomes necessary to combat a potentially lethal infection.
A recent issue of particular importance in liver transplant immunosuppressive pharmacology has been
the development of the drug tacrolimus (FK 506). This agent has been suggested as having particular
efficacy in the setting of liver transplantation, with truly remarkable results in early, uncontrolled trials
performed at the University of Pittsburgh. [48] Tacrolimus has a mechanism of action similar to that of
cyclosporin A and it has been used as a replacement for this drug. A possible benefit is the potential
steroid-sparing effect of tacrolimus, exceeding that of cyclosporine. Single-drug maintenance therapy is
being actively investigated. Multicenter, randomized, prospective trials comparing cyclosporine to

tacrolimus are underway, with the initial 1-year follow-up demonstrating a significant reduction in
rejection episodes and a lower need for rescue therapy with OKT3 in the tacrolimus group. [20] [52] No
graft or patient survival advantage has been demonstrated. The side effect profile for tacrolimus differs
from that of cyclosporine in that the cosmetic complications of hirsutism, acne, and gingival hyperplasia
are not present but the potential for neurologic abnormalities, diabetes mellitus, and renal impairment
appears to be more prevalent. An important consideration is that the combination of cyclosporine and
tacrolimus drastically increases the incidence and severity of side effects. Thus, therapy must be limited
to one or the other, and limiting the use of either in the early postoperative setting may benefit renal
function. Clearly, both cyclosporine and tacrolimus have proven efficacy for maintenance
immunosuppression after liver transplantation. The specific selection of either remains to be established
by randomized investigation. As a rescue agent during established acute rejection, neither tacrolimus nor
cyclosporine is particularly efficacious compared with glucocorticosteroids or antilymphocyte antibody
preparations.
Another practical issue unique to liver transplant immunosuppression relates to the absorption of
cyclosporine and, to a lesser degree, tacrolimus. Drug uptake depends on the availability of enteric bile
salts. Thus, with decreased production of bile in the initial posttransplant period or during a rejection
episode, or biliary diversion through a T tube, cyclosporine should be administered intravenously to
maintain adequate plasma levels. A new cyclosporine formulation (Neoral), which is bile salt
independent for absorption, is now available.
Antiviral Immunity and Immunosuppression.
The importance of viral infection in liver transplant patients cannot be overstated. De novo infection or
latent virus reactivation of pathogens, including cytomegalovirus, herpes simplex virus, and Epstein-Barr
virus, and reinfection of hepatotropic viruses, including hepatitis B and C, remain serious sources of
posttransplant morbidity and mortality and are directly related
469
to the intensity of the immunosuppressive regimen employed. In particular, the use of antilymphocyte
antibody preparations, particularly OKT3, have been implicated in increasing the likelihood of viral
infection and associated complications such as lymphoproliferative disorders. In response to this
association, most transplant centers have incorporated antiviral prophylaxis into their
immunosuppressive protocols, with the specific agents, duration of therapy, and dosages governed by the
viral status of the recipient and the donor organ. The relationship between the pathophysiology of acute
T-cell-mediated rejection and the physiologic function of T cells in viral immunity is underscored in the
transplant patient on T-cell-directed immunosuppression.
OUTCOMES
Clearly, liver transplantation is the most significant advancement in the treatment of end-stage liver
disease this century. Diseases treated by OLT are by definition terminal with few exceptions, and as such
are lethal without hepatic replacement. Survival of a patient with no other hope for survival is the most
obvious positive outcome. Operative survival now exceeds 90% for first grafts. Retransplant-free
survival has improved steadily in the past 10 years and is now 73% at 1 year for all transplants reported

to the UNOS registry. The 5-year survival reflecting transplantation before ganciclovir became available
is approximately 60%, and improvements of 2% to 3% per year have been made each year since 1987.
Predicted actuarial survival for transplants performed in 1995 is over 75%.
Survival and retransplant rates have been the principal, if not the only, measures of outcome in the
transplant literature until recently. The technical issues of survival in the first decade of widespread
transplantation have overshadowed other outcome measures to the point of their exclusion from the
literature. Critical evaluation of costly health care interventions has become a fundamental priority in our
society. Outcomes can be defined either from a physician's, patient's, payer's, or society's perspective.
The medical perspective is in the traditional purview of the clinician and clinician investigator, who use
objective and quantifiable medical parameters, whereas the patient's perspective involves a more
subjective assessment of quality of life. Today, survival has become the norm after OLT, and thus the
next decade should clearly define outcome with regard to each of these perspectives.
Despite the recent advances in all aspects of liver transplantation, the procedure remains one with
considerable morbidity. [14] Most patients have some complications that deviate from an ideal recovery,
and all patients accept the trade of their liver disease for the disease of immunosuppression. These
negative outcomes are generally remedied by prompt recognition of problems and aggressive corrective
intervention. It is, therefore, critical that the potential obstacles arising postoperatively are thoroughly
understood. The authors have recently categorized negative outcomes under three headings: negative
sequel, complications, and failure to cure. [16]
A negative sequel is an adverse effect inherent to the transplant procedure. [16] Transplantation in general
carries with it the risk of lifelong immunosuppression. Patients must adapt their lifestyle to ensure that
medications are taken as prescribed and that the immune system and organ function are monitored when
necessary. The risk of opportunistic infection and malignancy that accompanies chronic
immunosuppression persists for life. These factors, however, are less intrusive for liver transplant
patients than for recipients of other solid organs because of their reduced need for immunosuppression,
especially after the first transplant year.
Failure to cure refers to pre-existing conditions that remain unchanged or recur after the transplant
procedure. [16] The likelihood of cure reflects the primary disease. Metabolic and cholestatic diseases are
generally resolved, as are the physiologic disorders of alcoholic cirrhosis. Unfortunately, viral infections
remain generally uncured by liver replacement. As mentioned earlier, clinically significant recurrence of
hepatitis B and hepatitis C may limit long-term cure. The cure for these illnesses remains in more potent
antiviral therapeutics. Practically, however, viral reinfection that creates a clinically insignificant carrier
state is viewed as a cure by the patient. Obviously, no extrahepatic malignancy can be cured by OLT, and
the potential for cure in patients with intrahepatic malignancy is solely related to the presence or absence
of metastatic disease at the time of recipient hepatectomy.
A complication is any other negative outcome that does not fit clearly into the definition of negative
sequel or failure to cure. Complications of some kind occur in almost all patients, but the significance of
these setbacks varies greatly. For example, acute rejection must be considered a complication of sorts,
but most episodes are treated without significant alteration of a patient's comfort or residual disability.
Conversely, PNF at best leads to retransplantation and at worst to death. The lack of uniform reporting of

complications makes interpretation of the results of OLT difficult. [14] The authors have presented a
classification of complications stratified by severity [14] (Table 20-18) . With the heightened requirement
for outcomes-based research in the current economic environment, standardized evaluation of transplant
programs with regard to complication rates will likely increase.
Several common complications, such as acute and chronic rejection, have been discussed. Some
complications deserve particular attention because of their seriousness and requirement for prompt
intervention. PNF presents as a complete lack of synthetic function from the time of reperfusion. The
patient develops encephalopathy, increased intracranial pressure, coagulopathy, hyperbilirubinemia, and
hypertransaminasemia. Aggressive supportive therapy and prompt retransplantation are required within
72 hours. Total transplant hepatectomy with portacaval shunt can improve the hemodynamic and
metabolic status of the patient for 24 to 36 hours if a suitable liver is not found immediately.
Hepatic artery thrombosis remains a complication, especially in the pediatric population. This presents as
a rapid rise in serum transaminase levels. The transplanted liver does not tolerate loss of arterial flow,
and failure to restore flow produces graft loss. Hepatic artery stricture or stenosis generally presents as a
lesser degree of metabolic change later in the postoperative course. An alternative presentation of
dearterialization is bile leak resulting from hepatic duct necrosis. An additional vascular complication
that is less frequent but equally devastating is early portal vein thrombosis. Given the rapid and serious
but reversible nature of these vascular complications, any suspect change in hepatic function requires
immediate evaluation of the hepatic vasculature by Doppler ultrasonography, followed by either
re-exploration or a confirmatory arteriogram.
Biliary complications, which occur in 15% to 30% of patients, suggest vascular compromise. [54] Both
leaks and strictures can occur regardless of the method of reconstruction. Percutaneous or endoscopic
management is generally considered an acceptable first alternative, but reoperation should not be avoided
for appropriate lesions at the expense of hepatic function or cholangitis.
Today, increasing emphasis is given to patient-oriented subjective outcomes such as quality of life and
well-being. Subjective outcomes have inherent limitations, but they are arguably more relevant to
individual patients.
470
TABLE 20-18 -- Classification of Common Complications of Liver Transplantation

Data from Clavien, P.-A, Camargo, C., Jr., Croxford, R., Langer, B., Levy, G., and Greig, P.: Definition
and classification of negative outcomes in organ transplantation: Application in liver transplantation.
Ann. Surg., 220:109, 1994.

Grade 1: An alteration from ideal postoperative course with complete recovery or which can be
easily controlled and which fulfills the
general characteristics, namely (1) not life threatening; (2) not requiring use of drugs
other than immunosuppressive agents,
analgesics, antipyretic, anti-inflammatory, and antiemetic, drugs required for urinary
retention or lower urinary tract infection,
arterial hypertension, hyperlipidemia or transient hyperglycemia; (3) requiring only
therapies that can be performed at the
bedside; (4) postoperative bleeding requiring 3 units of blood; and (5) never
associated with a prolongation of the
intensive care unit (ICU) stay 5 days or total hospital stay 4 weeks.
Examples:
Superficial wound infection treated without antibiotics
Bile leak treated conservatively
Corticosteroid-responsive acute rejection
Well-controlled arterial hypertension
Grade 2:
Any complication that is potentially life-threatening or results in ICU stay 5 days or
hospital stay 4 weeks, but that does
not cause residual disability or persistent diseases.
Grade 2a Complications requiring only use of drug therapy or postoperative bleeding requiring
3 units of blood.
Examples:
Rejection requiring immunosuppressors not routinely used after induction therapy
(e.g., OKT3 or other antilymphocyte
drugs)
Bacterial, viral, or fungal infection requiring antibiotic, antiviral, or antifungal therapy
Primary graft dysfunction (opening aspartate transaminase [AST] 2000 or a

transient increase in AST levels 1000 I.U./
L., or a persistent elevated prothrombin time > 20 over 3 days)
Grade 2b Complications requiring therapeutic interventions, reoperation, readmission in the ICU,
or prolongation in the ICU stay 5
days but that do not result in residual disability.
Examples:
Primary graft dysfunction extending the ICU stay 5 days

Postoperative bleeding requiring laparotomy

Biliary problems requiring endoscopic or surgical procedures
Grade 3: Any complication with residual or lasting functional disability or development of
malignant disease (except squamous and
basocellular cutaneous malignancies).
Grade 3a Complication with lasting disability that shows no evidence of progression and that has
a relatively low risk of graft failure and/
or death.
Examples:
De novo hepatitis C
Nonprogressive chronic rejection
Persistent bile duct stricture without evidence of progressive liver failure or recurrent
cholangitis
Grade 3b Complications with lasting disability that are either difficult to control or have a
significant risk of leading to graft failure and/or
death.
Examples:
De novo hepatitis B
Development of malignancy (e.g., sarcoma, carcinoma, and lymphoma)
Persistent bile duct stricture not amenable to surgical or endoscopic treatment with
progressive liver failure or recurrent
cholangitis
Progressive chronic rejection
Grade 4: Complications that lead to retransplantation (Grade 4a) or death (Grade 4b)
Well-being represents a composite of several different aspects, including mental, physical, and social
criteria.
Due to the initial formidable technical issues of OLT, analyses of quality of life have been almost absent
from the literature. With improved results these past few years, researchers in numerous centers are
evaluating the effects of OLT on quality of life. Successful transplantation allows a return to an active
lifestyle free from the metabolic and hematologic complications of hepatic failure or portal hypertension.
Preliminary studies have already shown that self-image, functioning ability, and perception of health
status are significantly improved after OLT. [32] Currently, about 60% of patients undergoing OLT return
to work within the first year, [4] a figure that continues to increase long term. [32]

Once therapy has been shown to be effective, it is necessary to determine whether the treatment is
cost-effective in order to compare it with other competing technologies. Analysis of cost-effectiveness is
difficult to apply to liver transplantation because there is no alternative therapy with which the results of
the procedure can be compared. Thus, the cost issue becomes one of establishing a monetary value for
one's life. There is no doubt that OLT is expensive. The total first-year expenses for a new liver average
$200,000, with an additional $10,000 to $20,000 per year required thereafter. Cost for patients in the
intensive care unit at the time of transplant are 3 to 5 times higher than those called in from home for the
procedure. With current health care reform underway, these amounts are falling somewhat. In addition,
consolidation of transplant centers in the United States is likely to improve efficiency and reduce costs.
As new immunosuppressive agents are introduced, competition is also likely to drive the cost down.
Nonetheless, the extent to which physicians should go to return patients to health and to work is a matter
of great societal debate. As with all other aspects of liver transplantation, rational selection of patients is
the most important factor affecting cost.
471
EMERGING TECHNIQUES
Split and Reduced-Size Transplantation.
The remarkable ability of the liver to regenerate and support the metabolic needs of an individual despite
major resection has allowed for correction of most conditions with partial hepatic transplantation. [10]
This is useful for major size mismatches, especially in children, and can also help address the growing
donor organ shortage. [41] Anatomic division of the lobes of the liver with preservation of hepatic venous
and arterial, portal, and biliary branches has become increasingly successful but remains technically
formidable. The anatomic boundaries are based on the segmental anatomic system of Couinaud and
Bismuth. [17] Left lateral segments 2 and 3 or left lobe grafts (segments 2, 3, and 4) can be placed in
recipients who are substantially smaller than the donor. Because of the anterior to posterior dimensions
of the right lobe (segments 5 through 8), placement of this graft requires a recipient similar in size to the
donor.
Several major centers have initiated protocols for split and reduced-size transplantation with excellent
results. Predictably, biliary complications are increased in this procedure, but in children the problem of
hepatic artery thrombosis is improved, owing to the comparatively large vasculature present in the graft
when an adult liver is reduced for use in a child. Patient survival is as good as or better than full-size
grafts in children. Use of reduced-size grafts has decreased the mortality of pediatric recipients on the
waiting list to approximately 3%.
Living-Related Transplantation.
Arising from the success of reduced-size grafting, living-related transplantation has been initiated at
selected centers. [9] [57] This involves a reduced-size graft usually derived from a donor left lobe
(segments 2 and 3 or 2, 3, and 4). Technically, this has been quite successful and has the benefits of
reduced ischemic time, better HLA match, and better timing of transplantation, producing excellent graft
survival. Because most of the transplants are performed for congenital anomalies, the negative aspects of
HLA typing have not been problematic. The most pressing concern with this procedure is ethics.

Although hepatic resection is generally safe, the mortality is not zero. One donor perioperative death has
been reported, and many argue that with the success of reduced-size cadaveric allografts, a procedure
that places a healthy parent at risk is not necessary. The concept of informed consent is difficult to
establish because most parents disregard personal safety when the life of their child is at risk. One
consensus hearing on the matter has approved this procedure, with the caveat that only centers with
established success in reduced cadaveric grafts, pediatric transplantation, and adult hepatic surgery be
involved. [10]
Heterotopic Liver Transplantation.
Placement of an allograft in an anatomically altered site has the advantages of avoiding the recipient
hepatectomy (often the most morbid portion of the procedure) and preserving the orthotopic position for
future use in the event of graft failure. [51] Metabolic abnormalities are correctable by this approach.
Obviously, however, disorders leading to portal hypertension are not amenable to this mode of therapy.
ALTERNATIVE THERAPIES
The development of artificial support devices has revolutionized perioperative management in all areas
of transplantation except hepatic. Renal dialysis, ventricular assist and intra-aortic counterpulsation
devices, total parenteral nutrition, and insulin have all helped optimize the condition of solid organ
recipients to some degree, making emergent transplantation unusual for any organ other than the liver.
The importance of preoperative condition is clear from the survival statistics presented in this chapter.
Thus, great effort has been directed toward finding adequate hepatic replacement.
Xenogeneic Support.
Xenotransplantation, the use of organs from other species, has many theoretical advantages. A renewable
supply of organs subject to genetic manipulation available on an elective basis would greatly alter the
course of patients with liver failure. In addition, the hepatotrophic viruses responsible for most hepatitis
are generally specific for human hepatocytes, so the specter of reinfection would be abolished.
Unfortunately, the immune barriers to transspecies transplantation remain formidable. Organs from
discordant species, those phylogenetically distant animals to which preformed natural antibodies exist,
are hyperacutely rejected. Organs from these animals also produce plasma proteins that are similar but
not identical to their human homologues, thus raising the possibility of antigenic proteins subject to
immune clearance. Concordant species, namely primates, are rejected in a more conventionally acute
manner, but antibody-mediated rejection occurs. Although the immune barrier is less daunting, primates
are slow-breeding animals that could quickly become extinct if widespread use were initiated. Of
additional concern is the potential for introducing new viruses from primates to man.
Several efforts in xenogeneic organ use have been made in the past 5 years. Baboon livers were used to
treat two patients with hepatitis B at the University of Pittsburgh. [46] Both livers functioned well enough
to carry out the major physiologic functions of the liver. Interpretation of the immune implications of
these procedures, however, is difficult because one patient was infected with the human
immunodeficiency virus with an inverted T4:T8 ratio pretransplant and the other received a concurrent
baboon bone marrow transplant. An orthotopic pig liver transplant performed at Cedars Sinai Hospital in
Los Angeles was hyperacutely rejected (L. Makowka, personal communication, 1994).

Ex vivo perfusion with porcine livers has been successfully employed as a bridge to orthotopic
allotransplantation by the authors' group. [13] Biochemical improvement in all measured parameters,
including reversal of cerebral edema and reduction of coma, has been demonstrated by the authors' group
and others. [3] This approach has the advantage of being reversible without requiring a surgical procedure.
It is, however, logistically difficult and temporary.
The ability of porcine hepatocytes to perform many of the functions of human hepatocytes has been
exploited by several investigators by development of an ex vivo apparatus for hepatic support consisting
of porcine hepatocytes attached to a hollow-fiber dialysis cassette. [45] Early clinical trials have shown
promise in reducing cerebral edema as well as mild improvement in biochemical parameters. Additional
investigation in this field will determine whether the volume of the liver can be reproduced in cellular
form and whether temporary support will allow recovery of hepatic function, thus avoiding
transplantation for some patients with viral or toxin-induced acute liver failure.
SUMMARY
Liver transplantation has evolved in the past decade from an experimental procedure to an accepted,
effective therapy for end-stage diseases of the liver. Extended survival of over 75% in appropriately
selected patients is now commonplace with return to an excellent quality of life. Continued
improvements in perioperative management and operative
472
technique are being realized. The most important predictors of success are the state of the patient at the
time of transplantation and the disease being transplanted. Early intervention once end-stage disease is
diagnosed is preferable. A critical shortage of suitable donor organs remains the single most important
barrier to transplantation
SELECTED REFERENCES
Advances in liver transplantation. Gastroenterol. Clin. North Am., 22:entire issue, 1993.
This is an excellent review written by experts in transplantation. Most clinical aspects of liver
transplantation are critically discussed, including indications, immunosuppression, and management of
short- and long-term complications.
Consensus conference on indications of liver transplantation. Hepatology, 2(Suppl.):entire issue, 1994.
Consensus conference held in Paris by a panel of medical and surgical experts thoroughly discussed
indications for liver transplantation regarding both adult and pediatric diseases.
Neuberger, J., and Adams, D. (Eds.): Immunology of Liver Transplantation. London, Edward Arnold,
1993.
This excellent monograph covers most of the immunological aspects of liver transplantation, including
mechanisms of rejection, immunosuppression, and viral prophylaxis. It is extremely well referenced.

REFERENCES
1. AnnualReport of the U.S. Scientific Registry of Transplant Recipients and the Organ Procurement and Transplantation
Network, 1994.
2. Consensus statement on indications for liver transplantation. Hepatology, 20:63, 1994.
3. Abouna, G., Fisher, L., Porter, K., and Andres, G.: Experience in the treatment of hepatic failure by intermittent liver
hemoperfusion. Surg. Gynecol. Obstet., 137:141, 1963.
4. Adams, P., Ghent, C., Grant, D., and Wall, W.: Employment after liver transplantation. Hepatology, 21:140, 1995.
5. Ascher, N. L., Lake, J. R., Emond, J., and Roberts, J.: Liver transplantation for hepatitis C virus-related cirrhosis.
6. Belle,
S. H., Beringer, K. C., and Detre, K. M.: Trends in Liver Transplantation in the United States. In Terasaki, P. I.,
and Cecka, J. M. (Eds.): Clinical Transplants. Los Angeles, UCLA Tissue Typing Laboratory, 1993, p. 19.
7. Benhamou, J.-P.: Indications for liver transplantation in primary biliary cirrhosis. Hepatology, 20:11, 1994.
8. Bismuth, H., Castaing, D., Gugenheim, J., et al.: Roux-en-Y hepatojejunostomy: A safe procedure for biliary
anastomosis in liver transplantation. Transplant. Proc., 19:2413, 1987.
9. Broelsch, C. E., Burdelski, M., and Rogiers, X.: Living donor for liver transplantation. Hepatology, 20:49, 1994.
10. Broelsch, C. E., Emond, J. C., Whitington, P. F., Thistlewaite, J. R., Baker, A. L., and Lichtor, J. L.: Reduced sized
liver transplantation. Ann. Surg., 212:368, 1990.
11. Busuttil, R. W., Shaked, A., and Mills, J. M.: One thousand liver transplants: The lessons learned. Ann. Surg., 219:490,
1994.
12. Calne, R. Y.: Contraindications to liver transplantation. Hepatology, 20:3, 1994.
13. Chari, R., Collins, B. H., and Magee, J. C.: Treatment of hepatic failure with ex-vivo pig liver perfusion followed by
liver transplantation. N. Engl. J. Med., 331:234, 1994.
14. Clavien,
P.-A., Camargo, C., Jr., Croxford, R., Langer, B., Levy, G., and Greig, P.: Definition and classification of
negative outcomes in organ transplantation: Application in liver transplantation. Ann. Surg., 220:109,1994.
15. Clavien,
P.-A., Harvey, P. R. C., and Strasberg, S. M.: Preservation and reperfusion injuries in liver allografts:
Overview and synthesis of current studies. Transplantation, 53:957, 1992.
16. Clavien, P.-A., Sanabria, J. R., and Strasberg, S. M.: Proposed classification of complications of surgery with examples
of utility in cholecystectomy. Surgery, 111:518, 1992.
17. Couinaud, C.: Le Foie: Etudes Anatomiques et Chirurgicales. Paris, Masson et Cie, 1957.
18. D'Alessandro,
A. M., Kalayoglu, M., and Solinger, H. W.: The predictive value of donor liver biopsies on the
development of primary nonfunction after orthotopic liver transplantation. Transplant Proc., 23:1536, 1991.

19. Dousset, B., Houssin, D., Soubrane, O., Boillot, O., Baudin, F., and Chapuis, Y.: Metastatic endocrine tumors: Is there a
place for liver transplantation? Liver Transpl. Surg., 1:111, 1995.
20. European FK506 Multicentre Liver Study Group: Randomized trial comparing tacrolimus (FK506) and cyclosporin in
prevention of allograft rejection. Lancet, 344:423, 1994.
21. Farges,O., Malassagne, B., Sebagh, M., and Bismuth, H.: Primary sclerosing cholangitis: Liver transplantation or
biliary surgery. Surgery, 117:146, 1995.
22. Gordon,R. D., Iwatsuki, S., Esquivel, C. O., Tsakis, A., Todo, S., and Starzl, T. E.: Liver transplantation across ABO
blood groups. Surgery, 100:342, 1986.
23. Griffith,
B. P., Shaw, B. W., Hardesty, R. L., Iwatsuki, S., and Bahnson, H. T.: Venovenous bypass without systemic
anticoagulation for transplantation of the human liver. Surg. Gynecol. Obstet., 160:271, 1985.
24. Gugenheim, J., Samuel, D., Reynes, M., and Bismuth, H.: Liver transplantation across ABO blood group barriers.
Lancet, 336:519, 1990.
25. Harrison,
J., and McMaster, P.: The role of orthotopic liver transplantation in the management of sclerosing cholangitis.
26. Kalayoglu,M., Sollinger, W. H., and Stratta, R. J.: Extended preservation of the liver for clinical transplantation.
Lancet, 1:617, 1988.
27. Klion,
F., Fabry, T., Palmer, M., and Schaffner, F.: Prediction of survival in patients with primary biliary cirrhosis:
Examination of the Mayo Clinic model on a group of patients with known endpoint. Gastroenterology, 102:310, 1992.
28. Knechtle,
S. J., Kolbeck, P. S., Tsuchemntos A., et al.: Hepatic transplantation into sensitized recipients: Demonstration
of hyperacute rejection. Transplantation, 43:8, 1987.
29. Krom, R. A. F.: Liver transplantation and alcohol: Who should get transplants? Hepatology, 20:28, 1994.
30. Lee,
J., Schutz, S., England, R., Leung, J., and Cotton, P.: Endoscopic therapy of sclerosing cholangitis. Hepatology,
21:661, 1995.
31. Lerut,
J., Gordon, R. D., Iwatsuki, S., et al.: Biliary tract complication in human orthotopic liver transplantation.
Transplantation, 43:47, 1987.
32. Levy,M., Jennings, L., and Abouldoud, M.: Quality of life improvements at one, two, and five years after liver
transplantation. Transplantation, 59:515, 1995.
33. Markus, B. H., Duquesnoy, R. J., Gordon, R. D., et al.: Histocompatibility and liver transplantation: Does HLA exert a
dualistic effect? Transplantation, 46:372, 1988.
34. NationalInstitutes of Health Consensus Development Conference Statement. Liver transplantation. Hepatology,
4(Suppl. 1):107, 1983.
35. Neuberger, J., and Adams, D. (Eds.): Immunology of Liver Transplantation. London, Edward Arnold, 1993.
36. Neuhaus, P., Blumhardt, G., Bechstein, W. O., Steffen, R., Platz, K.-P., and Keck, H.: Technique and results of biliary
reconstruction using side-to-side choledochocholedochostomy in 300 orthotopic liver transplants. Ann. Surg., 219:426,
1994.

37. O'Grady, J. G., Schalm, S., and Williams, R.: Acute liver failure: Redefining the syndromes. Lancet, 342:273, 1993.
38. Otte,J.-B., Goyet, J. D. V. D., and Reding, R.: Sequential treatment of biliary atresia with Kasai portoenterostomy and
liver transplantation: A review. Hepatology, 20:41, 1994.
39. Perrillo, R., and Mason, A.: Hepatitis B and liver transplantation: Problems and premises. N. Engl. J. Med., 329:1885,
1993.
40. Pichlmayr,
R., Weimann, A., and Ringe, B.: Indications for liver transplantation in hepatobiliary malignancy.
41. Piper,
J. B., Whitington, P. F., and Woodle, E. S.: Pediatric liver transplantation at the University of Chicago Hospitals.
In Terasaki, P. I., and Cecka, J. M. (Eds.): Clinical Transplants. Los Angeles, UCLA Tissue Typing Laboratory, 1992, p.
179.
42. Pitt,
H., Thompson, H., Tompkins, R., and Longmire, W.: Primary sclerosing cholangitis: Results of an aggressive
surgical approach. Ann. Surg., 196:259, 1982.
43. Post,
S., Palma, P., Gonzalez, A., Rentsch, M., and Menger, M.: Timing of arterialization in liver transplantation. Ann.
Surg., 220:691, 1994.
44. Clavien,P. A., Camargo, C. A., Cameron, R., Washington, M. K., Phillips, M. J., Greig, P. D., and Levy, G. A.:
Kupffer cell erythrophagocytosis and graft-versus-host hemolysis in ABO-unmatched liver transplantation.
Gastroenterology, 110:1891, 1996.
45. Rozga, J., Podesta, L., and LaPage, E.: A bioartificial liver to treat severe acute liver failure. Ann. Surg., 219:538, 1994.
46. Starzl, T., Fung, J., and Tzakis, A.: Baboon-to-human liver transplantation. Lancet, 341:65, 1993.
47. Starzl,
T. E., Marchiaro, T. L., Von Kaulla, K., et al.: Homotransplantation of the liver in humans. Surg. Gynecol.
Obstet., 117:659, 1963
48. Starzl,
T. E., Todo, S., Fung, J., Demetris, A. J., Venkataramanan, R., and Jain, A.: FK 506 for human liver, kidney and
pancreas transplantation. Lancet, 2:1000, 1989.
49. Steinhoff, G.: HLA/ABO matching. In Neuberger, J., and Adams, D. (Eds.): Immunology of Liver Transplantation.
London, Edward Arnold, 1993, p 261.
50. Strasberg, S., Howard, T., Molmenti, P., and Hertl, M.: Donor evaluation of liver allograft. Hepatology, 20:829, 1995.
51. Terpstra,O., Schalm, S., Weimar, W., et al.: Auxiliary partial liver transplantation for end-stage chronic liver disease.
N. Engl. J. Med., 319:1507, 1988.
52. The
US Multicenter FK506 Liver Study Group: A comparison of tacrolimus (FK506) and cyclosporine for
immunosuppression in liver transplantation. N. Engl. J. Med., 331:1110, 1994.
53. Todo, S., Demetris, A., VanThiel, D., Teperman, L., Fung, J., and Starzl, T. E.: Orthotopic liver transplantation for
patients with hepatitis B virus-related liver disease. Hepatology, 13:619, 1991.
54. Vallera, R. A., Cotton, P. B., and Clavien, P.-A.: Biliary reconstruction for

473
liver transplantation and management of biliary complications: Overview and survey of current practices in the United
States. Liver Transpl. Surg., 1:143, 1995.
55. vanThiel, D. H., Wright, H. I., and Fagiuoli, S.: Liver transplantation for hepatitis B virus-associated cirrhosis: A
progress report. Hepatology, 20:20, 1994.
56. Williams, R., and Wendon, J.: Indictations for orthotopic liver transplantation in fulminant liver failure. Hepatology,
20:5, 1994.
57. Yamaoka, Y., Tanaka, K., and Ozawa, K.: Liver transplantation from living-related donors. In Terasaki, P. I., and
Cecka, J. M. (Eds.): Clinical Transplants. Los Angeles, UCLA Tissue Typing Laboratory, 1993,p. 179.


April 22, 2000
Conde Petra
LIVER TRANSPLANTATION AT THE

MILLENNIUM Past, Present, and Future
Additional Article
Consult's full-text literature database. Volume 4 Number 1 February 2000
Full Text
Frontmatter 241
LIVER TRANSPLANTATION: THE

PAST HEPATOLOGY: A CENTURY
Experimental Liver OF PROGRESS
Transplantation
Early Development of
Immunosuppression LIVER TRANSPLANTATION AT THE
Human Liver Transplantation MILLENNIUM
Past, Present, and Future
Development of Cyclosporine and
Tacrolimus
Additional Developments in Liver

Transplantation
LIVER TRANSPLANTATION: THE Emmet B. Keeffe MD

PRESENT
Donor Shortage Division of Gastroenterology, Department of Medicine, Stanford

University School of Medicine; and the Liver Transplant Program,
United Network for Organ Sharing Stanford University Medical Center, Stanford, California
Uniform Listing Criteria for Chronic Address reprint requests to:

Emmet B. Keeffe, MD
Liver Disease Stanford University Medical Center
750 Welch Road, Suite 210
Solutions to the Organ Shortage
Palo Alto, CA 94304-1509
Recurrence of Disease After Liver e-mail: ekeeffe@stanford.edu
Transplantation
LIVER TRANSPLANTATION: THE Transplantation is one of the greatest achievements of

FUTURE modern medicine. In AD 2000, liver transplantation is well
established as the definitive treatment for irreversible acute
Xenotransplantation
Hepatocyte Transplantation and and chronic liver disease, and progress continues both in
Liver-Directed Gene Therapy scientific investigation and in clinical practice. The care of
liver transplant patients, once the domain of transplant
SUMMARY centers, has more recently been integrated into the
References community-based practice of medicine. The two major
challenges currently facing liver transplantation are the
About the Publication large and increasing disparity between the numbers of
available cadaver donor organs and of qualified patients
listed and waiting for transplantation and the management
of recurrent disease after liver transplantation, particularly
recurrent chronic hepatitis C. Future issues concerning liver
transplantation will probably center on the effective use of
available cadaver donor organs, including re-evaluation of
selection criteria to strike a balance between medical need
and the chance of a successful outcome, and on the
development of new technologies to expand liver
transplantation, such as cadaver split-liver transplantation,
adult-to-adult living related and unrelated liver
transplantation, xenotransplantation, hepatocyte
transplantation, and liver-directed gene therapy. This
article reviews the chronology of liver transplantation,
presents the current status of transplantation, and offers
speculation regarding the future directions of liver
transplantation.
LIVER TRANSPLANTATION: THE PAST

Experimental Liver Transplantation
The first experimental attempts at liver transplantation, in dogs,

were initiated 45 years ago, in 1955, by Welch of Albany, New
York, who described the
242
TABLE 1 -- SELECTED EVENTS IN THE EARLY

HISTORY OF LIVER TRANSPLANTATION
Year Event
1955 First experimental liver transplantation
using an auxiliary technique (Welch)

1956 First experimental orthotopic liver

transplantation (Cannon)
1963 Use of azathioprine and prednisone for
immunosuppression
1963 First human liver transplantations
(Starzl)
1966 Introduction of antilymphocyte globulin
1967 First 1-year survival after human liver
transplantation (Starzl)
1980 Introduction of cyclosporine for
immunosuppression
1983 NIH Consensus Development
Conference on liver transplantation
1987 Use of University of Wisconsin solution
for improved organ preservation
1989 Introduction of tacrolimus for
immunosuppression
insertion of an auxiliary liver engrafted heterotopically in either

the pelvis or right paravertebral gutter (Table 1) . [68] The portal
vein was anastomosed to the inferior vena cava and the hepatic
artery to the aorta or iliac artery, and no immunosuppression
was used. The first experimental liver replacement, that is,
orthotopic liver transplantation, was reported by Cannon at the
University of California at Los Angeles in 1956, but none of
those dogs survived. [11] Cannon had speculated that the liver
might play a role in rejection and that a replaced liver should
not contribute to its own destruction.
In 1958, liver transplantation was established as a primary
research focus at Peter Bent Brigham Hospital in Boston,
Massachusetts, under the direction of Moore, [39] [40] and
Northwestern University in Chicago, Illinois, under the
direction of Starzl. [55] [56] The focus of research in Boston was
the immunology of liver transplantation; in Chicago, it was liver
regeneration and hepatotrophic growth factors. Initial successes
were hampered by the technical challenges in the performance
of liver transplantation in dogs and the inevitable development
of allograft rejection. Two technical improvements ultimately
provided better survival rates; adequate preservation of the graft
with portal infusion of chilled lactated Ringer's solution leading

to core cooling, [55] and decompression by bypassing the

obstructed recipient splanchnic and systemic venous beds
during the anhepatic phase when the native liver was being
removed and donor liver grafted in its place. [40] [55]
The original saline preservative solutions were replaced by
improved solutions. The Collins solution allowed safe
preservation for 5 to 6 hours [2] [66] and later the University of
Wisconsin solution extended the preservation time 18 to 24
hours. [28] [29] [60] As experience with human orthotopic liver
transplantation evolved, venous bypasses were used less often
and now are seldom employed. [54]
Early Development of Immunosuppression
The development of the field of immunosuppression was

critically important in liver transplantation. The first attempts at
liver transplantation had been performed without
immunosuppression, and thus no long-term organ engraftment
was possible. In 1944, Medawar showed that graft rejection is
an immunological event that has both specificity and memory.
[36] [37] The initial attempts to suppress the immune system to
ameliorate rejection used corticosteroid therapy or total-body

irradiation. In the late 1950s and early 1960s 6-mercaptopurine
and azathioprine were introduced as immunosuppressants
(Table 2) . [53]
243
In animal experiments and in early human cadaver kidney

transplantation, it was learned that single-agent
immunosuppression only rarely controlled rejection but that the
combination of prednisone and azathioprine effectively
prevented rejection and allowed successful renal
transplantation. It also become evident that large boluses of
corticosteroids could be used to reverse episodes of acute
allograft rejection. These two observations by Starzl and
colleagues at the University of Colorado made clinical
transplantation possible and opened the way for human liver
transplantation. [53]
Human Liver Transplantation
The first attempted human liver transplantation was reported in

1963 by Starzl. [58] The recipient was a 3-year-old boy with
biliary atresia who had had multiple previous operations who
and died of blood loss during surgery because of uncontrollable

coagulopathy. Two other liver transplantations were carried out

in the same year, but the recipients died after 22 and 7 days,
respectively. [53] [58] In the next year, isolated attempts at human
liver transplantation were unsuccessful in Boston [37] and in
Paris. [16] These first seven human liver transplant operations
achieved patient survivals ranging from zero to 23 days. [53]
The first truly successful human liver transplantation was
performed in 1967 by Starzl at the University of Colorado. The
recipient was an 18-month-old child with a hepatocellular
carcinoma who survived more than 1 year before succumbing to
recurrent tumor. [53] Six other patients underwent liver
transplantation in 1967 and 1968, with a maximum survival of
30 months. During the next 12 years, approximately one liver
transplantation per month was performed at the University of
Colorado, and the 1-year mortality rate was greater than 50%.
The long-term survival of liver transplant recipients remained at
30%. [54] On a more encouraging note, 30 of the first 170
patients (18%) in the consecutive series of liver transplants at
the University of Colorado from 1963 through 1979 lived more
than 10 years. [53] The usual immunosuppressive regimen was
prednisone, azathioprine, and polyclonal antilymphocyte
globulin. [54]
During this period, Calne of Cambridge University and the
hepatologist Williams of King's College Hospital in London
began clinical trials in liver transplantation. [10] Their first
patient also exsanguinated, but success soon followed. Other
liver transplantation teams were established in Hanover,
Germany, in 1972, under the direction of Pichlmayr, and in
Paris in 1974, under Bismuth. By 1980, it had been shown that
liver transplantation could be accomplished,
TABLE 2 -- HISTORY OF IMMUNOSUPPRESSIVE DRUG
REGIMENS
Year Agent
1962 Azathioprine
1963 Azathioprine plus corticosteroids
1966 Polyclonal antibodies; antilymphocyte
globulin as an adjunct
1970 Cyclophosphamide substituted for
azathioprine
1978 Cyclosporine use in humans
1980 Cyclosporine plus corticosteroids

1981 Development of monoclonal antibodies

1989 Tacrolimus plus corticosteroids
1990s Development of newer agents (e.g.,
mycophenolate mofetil, rapamycin)
244
but the relatively high mortality rate seemed to indicate that the
procedure was not practical.
Development of Cyclosporine and Tacrolimus
The evolution of liver transplantation from an experimental

operation applied to a few individuals to routine surgery with
excellent graft and patient survival rates resulted from several
advances, but the development of cyclosporine has been
credited with the ultimate success and wide acceptance of liver
transplantation. The initial experimental and clinical
effectiveness of cyclosporine in transplantation was
demonstrated by Calne and colleagues in 1978 and 1979. [7] [8]
[9] Cyclosporine was the first selective method of
immunosuppression, and its use increased the survival rate of

liver transplant recipients from approximately 30% to more than
70%. [27] [57] The development of cyclosporine in the late 1970s
and early 1980s led to the acceptance of liver transplantation for
routine patient care by the National Institutes of Health
Consensus Development Conference in 1983 [42] and to the
opening of many new transplant centers.
With the initial use of tacrolimus (FK506) at the University of
Pittsburgh, [59] the 1-year graft and patient survival rates
increased further. Later direct comparison between cyclosporine
and tacrolimus in controlled trials, however, produced
equivalent results that were quite good with either agent used as
the basis of a multiple-drug immunosuppressive regimen. [1] [47]
[69] Although the merits of cyclosporine versus tacrolimus
continue to be debated, there is a clinical advantage in having

available two good agents that can be substituted as clinical
circumstances dictate.
Another advance in immunosuppression was the development
of monoclonal antibodies, particularly muromonab-CD3 that is
used to control steroid-resistant rejection and is occasionally
used for induction of immunosuppression in the setting of renal

insufficiency. [43] [48] [67] Muromonab-CD3 has now essentially

replaced the earlier antilymphocyte and antithymocyte
globulins, which could never be standardized. Other agents,
such as mycophenolate mofetil and rapamycin, are being tested
in clinical trials to determine their roles in liver transplantation.
[17]
Additional Developments in Liver Transplantation
The initial technical achievement of liver replacement and the

development of improved immunosuppressive regimens were
accompanied by many other advances that led to the current
1-year patient survival rates of 85% to 90% after liver
transplantation for most conditions (Table 3) . [49] Animal
experiments were inadequate preparation for liver
transplantation in sick patients with portal hypertension and
severe coagulopathy, many of whom had prior right upper
quadrant surgery with adhesions. The management of
hemodynamic and metabolic problems that may arise during
surgery by anesthiologists who specialize in liver
transplantation, the use of modern blood component and
coagulation factor replacement therapy, and improved surgical
methods to control operative bleeding have all improved the
outcome of liver transplantation. [12] [54] The massive blood loss
that routinely characterized earlier operations has been replaced
by minimal blood loss, and as many as 30% of liver
transplantations can be performed without blood transfusion. [6]
The introduction of venovenous bypass, which was important in
the early experience of surgeons with liver transplantation,
245
TABLE 3 -- SURVIVAL AFTER ADULT LIVER

TRANSPLANTATION BY DIAGNOSIS *
Survival (% )
Diagnosis 1-Year 4-Year 7-Year
Primary sclerosing 91 84 78
cholangitis
Primary biliary 89 84 79
cirrhosis
Autoimmune hepatitis 86 81 78
Chronic hepatitis C 86 75 67

Alcoholic liver disease 85 76 63

Cryptogenic cirrhosis 84 76 67
Chronic hepatitis B 83 71 63
Malignancy 72 43 34
Data from Seaberg EC, Belle SH, Beringer KC, et al: Liver
transplantation in the United States from 1987-1998: Updated
results from the Pitt-UNOS liver transplant registry. In Cecka
JM, Teraski PI (eds): Clinical transplants 1998. Los Angeles,
UCLA Tissue Typing Laboratory, 1999, p 17.
*United Network of Organ Sharing Database 1987-1998; n = 24,900

patients.
facilitated the development of new programs by supporting

hemodynamics during the anhepatic phase. [50]
Biliary complications were a major source of morbidity and
mortality in early liver transplantations, with as many as 50% of
patients developing one or more biliary complications. [54] A
primary duct-to-duct anastomosis can now be performed
without T-tubes or stents, a development which has reduced the
biliary complication rate to 5% to 10%. [64] The use of
radiologic and endoscopic diagnostic and therapeutic
procedures has also facilitated the management of biliary
complications with lower rates of morbidity.
Other important advances include better understanding and
diagnosis of acute and chronic allograft rejection and improved
prevention and treatment of infections that occur in the
immunosuppressed transplant recipient. [22] [52] The routine use
of posttransplantation allograft liver biopsies has allowed more
accurate diagnosis of acute rejection and identification of other
pathologic processes that can cause early allograft dysfunction.
Prophylactic regimens for early bacterial infection after liver
transplantation and for late cytomegalovirus, fungal, and
Pneumocystis infections are now routine and have substantially
reduced morbidity and mortality from these infections.

LIVER TRANSPLANTATION: THE

PRESENT
The first 3 decades of liver transplantation witnessed great
successes in the development of surgical techniques and
improvement in immunosuppressive regimens. These successes
are accompanied by a new set of problems. In particular, the
growing disparity between the availability of cadaver donor
livers and the number of potential transplant recipients on the
waiting lists in the United States and other countries has led to
increased deaths among persons on the waiting list and sicker
patients coming to transplantation. An increasing number of
transplant patients have chronic hepatitis C virus (HCV)
infection, and recurrent hepatitis C is an important cause of
morbidity and mortality that may reduce the long-term utility of
liver grafting. Newer approaches to solving these problems are
under way, but no comprehensive solution is in sight.
246
TABLE 4 -- NUMBERS OF LIVER TRANSPLANTS,

PATIENTS AND DEATHS ON THE LIVER WAITING LIST
IN THE UNITED STATES FROM 1988 TO 1997
(Not Available)
Modified from Keeffe EB: Summary of guidelines on organ
allocation and patient listing for liver transplantation. Liver
Transplant Surg 4:S108, 1998; with permission.
247
Donor Shortage
The number of liver transplantations performed in 1996, 1997,

and 1998 in the United States was relatively stable, at slightly
more than 4000 transplantations annually. Currently, however,
there are more than 12,000 potential liver transplant candidates
on the United Network for Organ Sharing (UNOS) waiting list
in the United States, and therefore, a lower percentage of the
patients with chronic liver disease who might benefit from liver
transplantation can actually undergo the procedure. This
growing discrepancy between the available donor organs and
the need for transplantation has led to scrutiny of the traditional
selection and listing criteria for liver transplantation.

The supply of donor livers is insufficient to meet the current
and future need for liver transplantation, and organ donation has
remained relatively stagnant, with only minor increases in
retrieval rates in recent years in spite of many efforts to increase
donation. The growing disparity between the number of liver
transplantations that can be performed each year and the
number of patients on the waiting list is reaching crisis
proportions in the United States. Although the number of
donors and liver transplantations increased 2.4-fold between
1988 and 1997, the number of patients on the liver
transplantation list increased 15.6-fold, and the number of
deaths increased 5.8-fold during the same period. (Table 4)
(Table Not Available) . [61] This situation will only worsen in
the coming years. Analysis of data from the past 5 years to
predict the status of liver transplantation in the year 2000,
assuming no major changes in organ availability or performance
of transplantation, reveals an even greater crisis at the turn of
the century with 20,000 or more patients awaiting a liver
transplantation (Table 5) (Table Not Available) . [30]
United Network for Organ Sharing
The disparity between the number of cadaver organs and

candidates for liver transplantation has led to re-evaluation of
UNOS allocation and distribution policies for livers and has
spurred a nationwide debate among transplantation
professionals, potential recipients, and the federal government.
[5] [23] [44] Distribution policies determine the geographical areas
within which livers are allocated, and allocation policies

determine which patients within a geographical area will
receive the available livers. [44] The historical allocation scheme
has
TABLE 5 -- GROWTH OF LIVER TRANSPLANTS AND
WAITING LIST
(Not Available)
Modified from Keeffe EB: Summary of guidelines on organ
allocation and patient listing for liver transplantation. Liver
Transplant Surg 4:S108, 1998; with permission.
248

dictated that the sickest patients who have waited the longest
receive transplantation first. In the recent past, there have been
modifications and redefinition of UNOS status based on disease
severity, but the basic principle, as shown in the box , still
holds.
UNOS Liver Status For Patients 18 Years of Age

According to Disease Severity
Status 1 Fulminant liver failure with

life expectancy <7 days,
including FHF, primary graft
nonfunction, hepatic artery
thrombosis, and acute
decompensated Wilson's
disease
Status 2A Hospitalized in ICU for
chronic liver failure with life
expectancy <7 days, with a
Child-Pugh score of 10
and one of following:
unresponsive active variceal
hemorrhage, hepatorenal
syndrome, refractory
ascites/hepatic hydrothorax,
or stage 3 or 4 hepatic
encephalopathy
Status 2B Requiring continuous
medical care, with a
Child-Pugh score of 10,
or a Child-Pugh score 7
and one of the following:
unresponsive active variceal
hemorrhage, hepatorenal
syndrome, spontaneous
bacterial peritonitis, or
refractory ascites/hepatic
hydrothorax; presence of an

Status 3 Requiring continuous

medical care, with a
Child-Pugh score of 7,
but not meeting criteria for
status 2B
Status 7 Temporarily inactive
Data from United Network of organ sharing. Implemented
July 1997; modified January 1998 and August 1998.
http://www.unos.org(updated 1999)
FHF = fulminant hepatic failure
The UNOS distribution scheme as of mid-1999 dictates that

patients at transplant programs served by a local organ
procurement organization (OPO) have the first priority for
livers obtained by that OPO. [5] [23] [44] Because of this local use
of livers, the waiting times for liver transplantation differ
considerably for patients listed in different regions in the United
States. This local policy means that a patient in one OPO who is
not in immediate danger of dying may receive a transplantation
before a sicker patient in another OPO in a nearby geographic
area. This reality has led to discussion about widening the liver
distribution area for patients who are listed as status 1, but
computer simulation modeling by UNOS showed that wider
sharing, that is, a single national waiting list, would allow
urgency (often termed justice) to prevail over medical utility.
Utility factors focus on maximizing the overall benefits of
transplantation to society, that is, giving priority to the patient
who maximizes the chances of a successful outcome by having
the least risk of dying after transplantation. Urgency, or justice,
recognizes the needs of the individual transplant patient by
giving priority to the sickest patient, who has the greatest risk
of dying before transplantation. The argument made against
current UNOS policies is that the geographically restricted
distribution scheme overrides prioritization based on medical
urgency in favor of utility and is unfair to transplant candidates
in most need of liver replacement. [23] This argument proposes a
national waiting list to equalize access
249

to liver transplantation, that is, a patient-driven rather than a

local or center-driven allocation scheme. Finally, the
transplantation community is becoming cognizant of
cost-outcome analyses and the reality of managed care, which
has transferred financial risk from insurers to providers. [20]
High-risk patients represent a significant liability to
transplantation centers in the managed care marketplace.
Appropriate and reasonable patient selection may become an
important consideration in allocation policies.
Uniform Listing Criteria for Chronic Liver Disease
At a recent consensus conference at the National Institutes of

Health organized by the American Society of Transplantation
and the American Association for the Study of Liver Diseases,
uniform minimal listing criteria were developed for general
application to patients with miscellaneous chronic liver
diseases. [33] These criteria are
Immediate need for liver transplantation
Estimated 1-year survival 90%
Child-Pugh score 7 (Child-Pugh class B or C)
Portal hypertensive bleeding or a single episode of
spontaneous bacterial peritonitis, irrespective of
Child-Pugh score
These criteria were established using the general principle that
any patient listed should have an expected 1-year survival with
general supportive care of 90% or less, which is less than
expected with liver transplantation. Large studies of the natural
history of patients with compensated cirrhosis resulting from
miscellaneous causes [25] [46] or chronic hepatitis C [21] have
shown that survival is relatively good until decompensation,
when 5-year survival rates fall to approximately 50%. Thus, the
indication for listing for liver transplantation should not be
simply the presence of cirrhosis, without decompensation, that
is, Child-Pugh class A, but should be the development of
decompensation, for example, Child-Pugh class B or C.
The current approaches to the organ shortage include increased

efforts to achieve higher rates of organ procurement, expanded
use of marginal donors, and surgical alternatives such as
cadaver split-liver and adult living donor related and unrelated
liver transplantation. Xenotransplantation may become an

option in the future.

There has been a concerted effort to increase organ donation,
and in some countries small gains have been made in the past
years. In the United States, there was 5.6% increase in cadaver
donors in 1998, the first substantial increase since 1995. [61] The
organ donation rate in the United States is approximately 20/1
million population, compared with Spain, the leading Western
nation, with donor rates of 25/1 million and Italy, the worst
Western country, with a rate of less than 10/1 million.
The expanded use of marginal donors in recent years has
included implanting donor livers from older individuals, use of
grafts with substantial fatty change, and engrafting donor grafts
from patients with mild chronic hepatitis C into recipients with
hepatitis C or from patients with a positive hepatitis B core
250
antibody (anti-HBc) into recipients with hepatitis B, and,

occasionally, into other recipients. This experience is too
limited for long-term outcome to be compared with patients
receiving grafts from uninfected donors, but it is known that
anti-HBc-positive grafts frequently transmit hepatitis B virus
(HBV), and recipients without HBV infection are, therefore, at
risk and should receive prophylaxis in the form of lamivudine
or hepatitis B immune globulin. [19] The short-term and
medium-term outcome of transplanting HCV-positive grafts
into patients with chronic hepatitis C seems to be good. [63]
Although increasing donor age is associated with poor graft
function, the organ shortage seems to justify use of these grafts,
and the overall percentage of donors over the age of 50 years
has increased substantially. [18] It seems that the judgment of the
harvesting surgeon as to whether the donor liver is good, fair, or
poor is also an important factor affecting graft survival. [26]
There were no differences in the 3-month graft survival rates
using livers from younger donors judged good versus those
considered fair or poor, but there was a significantly lower
survival rate using fair or poor grafts from donors aged 50 years
or older (61% versus 92% for organs classified as good). [26] It
has been suggested that donor/recipient gender matching should
also be considered. [4] [35] Because livers from women
transplanted into men generally have a poorer outcome,
transplantation of grafts from older women into men may be
even more likely to result to in poor graft function.
Split-liver transplantation, after unsatisfactory initial results, has
undergone a resurgence during the past 2 to 3 years. This

procedure essentially achieves two liver transplantations from a
single cadaver liver, with the right lobe usually implanted into
an adult recipient and the left lobe or left lateral segment
transplanted into a child. The organ used in split-liver
transplantation can obviously be shared between two
institutions or used at a single transplant center. Living-donor
liver transplantation for adults, using a right hepatic lobe, is also
being performed more often, and good results are being
reported from centers in the United States [34] [65] and Japan. [31]
[32] The Japanese center has also had good outcomes using
living-donor liver transplantation for high-urgency patients. [32]
Recurrence of Disease After Liver Transplantation
Chronic hepatitis C has become the most common cause of

end-stage liver disease requiring liver transplantation and
accounts for 25% to 40% of all transplantations in some centers.
[61] Hepatitis C viral infection may also be present in patients
with alcoholic liver disease and in patients undergoing liver

transplantation who are classified as having cryptogenic
cirrhosis. Most patients have had HCV infection for several
decades before the onset of liver failure.
After liver transplantation, HCV reinfection occurs in almost all
patients. Most of these patients subsequently develop chronic
hepatitis, and a few progress to cirrhosis. [45] [70] Fortunately, the
infection appears to be benign in 80% to 85% of patients on
short-term and medium-term follow-up, and survival rates are
comparable with patients receiving transplantation for nonviral
chronic liver diseases. It has been shown, however, that
transplant recipients with recurrent hepatitis C have poorer
quality of life, greater depression, and higher psychologic
distress than those without HCV infection. [51] Whether
progressive chronic hepatitis and cirrhosis will occur in most
patients with longer follow-up remains uncertain but is possible.
Numerous host and viral factors have been implicated in the
development
251
of severe recurrent hepatitis C, suggesting that the process in

multifactorial. The overlapping histologic features of HCV
infection and allograft rejection, which include portal and
parenchymal mononuclear cell infiltrates, fatty change, swollen
hepatocytes with necrosis, and occasional bile duct damage,

make distinction between these two entities difficult at times.

[45]
The best treatment of recurrent HCV infection after liver

transplantation remains uncertain. It seems that neither
interferon nor ribavirin alone is beneficial, [24] but the
combination of both agents shows promise. [3] In one study, 18
of 21 patients tolerated therapy, and 24% of patients
experienced a virologic sustained response with improved
aminotransferase levels and liver biopsy histologic scores. [3]
Whether maintenance therapy can be discontinued in patients
who have a sustained virologic response remains unknown.
Studies are in progress to determine if preemptive therapy early
after liver transplantation will alter the posttransplantation
infection rate or the severity of recurrent hepatitis C.
Other diseases can recur after liver transplantation, including
HBV infection, alcoholic liver disease, and immunologic liver
diseases, such as primary biliary cirrhosis, primary sclerosing
cholangitis, and autoimmune hepatitis. [15] The strategies for
diagnosing or preventing these entities are well established,
unlike the management challenges presented by recurrent
hepatitis C.
LIVER TRANSPLANTATION: THE FUTURE

In the immediate future, developments in liver transplantation
will probably extend from the developments of the past few
years, enlarging the efforts to increase organ donation, further
expanding split-liver and adult living related and unrelated liver
transplantation, and optimizing the use of the existing donor
pool by adjusting allocation and distribution policies and
refining selection criteria. Possible clinical developments in the
near future include xenotransplantation, hepatocyte
transplantation, and liver-directed gene therapy.
Xenotransplantation
If xenotransplantation can be made safe and affordable, it may

solve many of the current problems of access to liver
transplantation. Xenotransplantation is the engraftment of
organs obtained from one species into another species.
Xenografting is not a new concept, but it has resurfaced with
the availability of more potent immunosuppressive agents and
the critical shortage of human cadaver organs. Most

investigators are now focusing on the pig as a potential donor

based on appropriate size, unlimited supply, ability to be
genetically engineered, and the more easily controlled risk of
zoonotic infection. [41] The immunologic hurdles to
xenotransplantation are hyperacute rejection, acute vascular
rejection, and cellular rejection. Another important
consideration in xenotransplantation is the potential
transmission of infectious agents from the graft to the recipient.
It has been suggested that the movement to xenotransplantation
will occur as a step-by-step process, beginning with limited
clinical trials, using xenotransplantation initially as a bridge to
human cadaver transplantation, then implanting porcine
xenografts in patients who cannot obtain a human graft, and
finally using xenotransplantation as an alternative to
allotransplantation. [41] Before this sequence can unfold, many
critical ethical issues must be addressed. [14] [62]
252
Hepatocyte Transplantation and Liver-Directed Gene

Therapy
Liver cell transplantation is being developed to treat acute and

chronic liver failure and inherited metabolic disorders. [13] Liver
cells can be isolated from a number of species, including
humans, and then cultured or cryopreserved for future use.
Cultured cells can be directly transplanted from allogeneic
donors (a process that requires immunosuppression) or
transplanted back into an individual after being surgically
harvested and transduced in culture with a therapeutic gene for
either a defective or absent protein. [13] In the latter case,
immunosuppression is not required. Liver-directed gene therapy
can be used to replace a missing gene, express a gene that is not
normally expressed in the liver, interfere with gene expression,
disrupt an offending gene, or repair a mutated gene. [13]
Hepatocyte transplantation has already been successfully used
to treat some inherited disorders, and both nonviral and viral
vectors are being developed for gene therapy targeted at the
liver. Both technologies will probably develop into more
practical therapies.
It is hoped that in the future the need for liver transplantation
will be reduced if effective medical treatments, some of which
may employ gene-direct therapy, can be discovered to treat the
full spectrum of metabolic, viral, and immunologic liver
diseases. The late 1990s have already witnessed some advances

in the treatment of chronic hepatitis C and chronic hepatitis B,

and the next few years should bring improved treatments for
many liver diseases that have historically progressed inevitably
to cirrhosis and liver failure, with the resultant need for liver
transplantation.
SUMMARY
During the past 3 decades, liver transplantation has achieved
such acceptance that more than 12,000 qualified recipients are
listed for liver transplantation in the United States, but
unfortunately just over 4000 cadaver donor organs are available
each year. Thus, given the increasing disparity between the
number of potential recipients and available cadaver organs, the
current challenge in liver transplantation is to optimize the
outcome of liver transplantation from this limited resource.
Currently under way is re-evaluation of selection criteria to use
these 4000 cadaver liver grafts most effectively by striking the
proper balance between medical urgency and utility. In parallel
with this re-evaluation, there is ongoing expansion of cadaver
split-liver transplantation and adult living related and unrelated
liver transplantation. Hoped-for but as yet unachieved
developments in liver transplantation are xenotransplantation,
hepatocyte transplantation, and liver-directed gene therapy.
Liver transplantation has come a long way from the initial,
unsuccessful human transplantations in 1963, but many
challenges remain.
References
1. Acomparison of tacrolimus (FK506) and cyclosporine for

immunosuppression in liver transplantation. The U.S. Multicenter FK506
Liver Study Group. N Engl J Med 331:1110, 1994
2. Benichou J, Halgrimson, CG, Weil R III, et al: Canine and human liver
preservation for 6 to 18 hours by cold infusion. Transplantation 24:407,
1977 abstract
253
3. Bizollon T, Palazzo U, Ducerf C, et al: Pilot study of the combination of

interferon alfa and ribavirin as therapy of recurrent hepatitis C after liver

transplantation. Hepatology 26:500, 1997 abstract
4. Brooks BK, Levy MF, Jennings LW, et al: Influence of donor and
recipient gender on the outcome of liver transplantation. Transplantation
62:1784, 1996 abstract
5. Burdick JF, Klein AS, Harper AM: The debate over liver allocation in
the United States: The UNOS perspective. In Cecka JM, Terasaki PI (eds):
Clinical Transplants 1996. Los Angeles, UCLA Tissue Typing Laboratory,
1997, p 322
6. CacciarelliTV, Keeffe EB, Moore DH, et al: Primary liver

transplantation without transfusion of red blood cells. Surgery 120:698,
1996 abstract
7. CalneRY, Rolles K, White DJG, et al: Cyclosporin A initially as the

only immunosuppressant in 34 recipients of cadaver organs: 32 kidneys, 2
pancreases, and 2 livers. Lancet 2:1033, 1979 abstract
8. CalneRY, White DJ, Rolles K, et al: Prolonged survival of pig

orthotopic heart grafts treated with cyclosporin A. Lancet 1:1183, 1978
abstract
9. CalneRY, White DJG, Thiru S, et al: Cyclosporin A in patients

receiving renal allografts from cadaver donors. Lancet 2:1323, 1978
abstract
10. CalneRY, Williams R: Liver transplantation in man. I. Observations on

technique and organization in five cases. BMJ 4:535, 1968
11. Cannon JA: Brief report. Transplant Bull 3:7, 1956
12. Carmichael FJ, Lindop MJ, Farman JV: Anaesthesia for hepatic
transplantation: Cardiovascular and metabolic alterations and their
management. Anesth Analg 64:108, 1985 abstract
13. Chowdhury JR: Forward: Prospects of liver cell transplantation and

liver-directed gene therapy. Semin Liver Dis 19:1, 1999 abstract
14. Daar AS: Ethics of xenotransplantation: Animal issues, consent, and

likely transformation of transplant ethics. World J Surg 21:975, 1997
abstract
15. Davern
TJ, Lake JR: Recurrent disease after liver transplantation.
Semin Gastrointest Dis 9:86, 1998 abstract
16. Demirleau,Noureddine, Vignes, et al: Tentative d'homograffe

hepatique (Attempted hepatic homograft). Memoires de l'Academie de
Chirurgie 90:177, 1964

17. Denton MD, Magee CC, Sayegh MH: Immunosuppressive strategies in

transplantation. Lancet 353:1083, 1999 citation
18. DetreKM, Lombardero M, Belle S, et al: Influence of donor age on

graft survival after transplantation--United Network for Organ Sharing
Registry. Liver Transpl Surg 1:311, 1995 abstract
19. Dickson RC, Everhart JE, Lake JR, et al: Transmission of hepatitis B by
transplantation of livers from donors positive for antibody to hepatitis B
core antigen. The National Institute of Diabetes and Digestive and Kidney
Diseases liver transplantation database. Gastroenterology 113:1668, 1997
abstract
20. Evans
RW, Kitzmann DJ: The "arithmetic" of donor liver allocation. In
Cecka JM, Terasaki PI (eds): Clinical Transplants 1996. Los Angeles,
UCLA Tissue Typing Laboratory, 1997, p 338
21. FattovichG, Giustina G, Degos F, et al: Morbidity and mortality in

compensated cirrhosis type C: A retrospective follow-up study of 384
patients. Gastroenterology 112:463, 1997 abstract
22. FishmanJA, Rubin RH: Infection in organ-transplant recipients. N Engl

J Med 338:1741, 1998 citation
23. Fung
JJ, Roberts JP: The rationale for equitable liver allocation. In
Cecka JM, Terasaki PI (eds): Clinical Transplants 1996. Los Angeles,
24. Gane EJ, Lo SK, Riordan SM, et al: A randomized study comparing
ribavirin and interferon alfa monotherapy for hepatitis C recurrence after
liver transplantation. Hepatology 27:1403, 1998 abstract
25. GinesP, Quintero E, Arroyo V, et al: Compensated cirrhosis: Natural

history and prognostic factors. Hepatology 7:122, 1987 abstract
26. Hoofnagle JH, Lombardero M, Zetterman RK, et al: Donor age and
outcome of liver transplantation. Hepatology 24:89, 1996 abstract
254
27. Iwatsuki S Starzl TE, Todd S, et al: Experience in 1,000 liver

transplants under cyclosporine-steroid therapy: A survival report.
Transplant Proc 20(suppl 1):498, 1988 citation
28. Jamieson NV, Sundberg R, Lindell S, et al: Successful 24- to 30-hour

preservation of the canine liver: A preliminary report. Transplant Proc

20(suppl 1):945, 1988 citation
29. Kalayoglu M, Sollinger WH, Stratta RJ, et al: Extended preservation of

the liver for clinical transplantation. Lancet 1:617, 1988 abstract
30. Keeffe EB: Summary of guidelines on organ allocation and patient

listing for liver transplantation. Liver Transpl Surg 4(suppl 1):S108, 1998
abstract
31. Lo CM, Fan ST, Liu CL, et al: Adult-to-adult living donor liver
transplantation using extended right lobe grafts. Ann Surg 226:261, 1997
full text
32. LoCM, Fan ST, Liu CL, et al: Applicability of living donor liver
transplantation to high-urgency patients. Transplantation 67:73, 1999
abstract
33. Lucey MR, Brown KA, Everson GT, et al: Minimal criteria for
placement of adults on the liver transplant waiting list: A report of a
national conference organized by the American Society of Transplant
Physicians and the American Association for the Study of Liver Diseases.
Liver Transpl Surg 3:628, 1997 abstract
34. MarcosA, Fisher RA, Ham JM, et al: Living donor liver transplantation
(LDLTx) for adults utilizing the right lobe [abstract]. Gastroenterology
29:A27, 1999
35. MarinoIR, Doyle, Aldrighetti L, et al: Effect of donor age and sex on
the outcome of liver transplantation. Hepatology 22:1754, 1995 abstract
36. Medawar PB: Second study of behavior of skin homografts in rabbits. J

Anat 79:157, 1945
37. Medawar PB: The behavior and fate of skin autografts and skin
homografts in rabbits. J Anat 78:176, 1944
38. Moore FD, Birtch AG, Dagher F, et al: Immunosuppression and

vascular insufficiency in liver transplantation. Ann N Y Acad Sci 102:729,
1964
39. Moore FD, Smith LL, Burnap TK, et al: One-stage homotransplantation
of the liver following total hepatectomy in dogs. Transplantation Bulletin
6:103, 1959
40. Moore FD, Wheeler HB, Demissianos HV, et al: Experimental whole
organ transplantation of the liver and of the spleen. Ann Surg 152:374,
1960
41. Nagayasu T, Platt JL: Progress in xenotransplantation. Graft 1:19, 1998

42. National
Institutes of Health Consensus Development Conference
Statement: Liver transplantation-June 20-23, 1983. Hepatology
4(suppl):107S, 1984
43. Neuhaus P, Bechstein WO, Blumhardt G, et al: Comparison of

quadruple immunosuppression after liver transplantation with ATG or
IL-2 receptor antibody. Transplantation 55:1320, 1993 abstract
44. Norman DJ: Allocation of livers for liver transplantation: Ethics and
politics. Clin Liver Dis 1:281, 1997
45. PessoaMG, Wright TL: Hepatitis C infection in transplantation. Clin

Liver Dis 1:663, 1997
46. PropstA, Propst T, Zangerl G, et al: Prognosis and life expectancy in

chronic liver disease. Dig Dis Sci 40:1805, 1995 abstract
47. Randomised trial comparing tacrolimus (FK506) and cyclosporin in

prevention of liver allograft rejection. European FK506 Multicentre Liver
Study Group. Lancet 344:423, 1994 abstract
48. RedingR, Vraux H, de Ville de Goyet J, et al: Monoclonal antibodies in

prophylactic immunosuppression after liver transplantation.
Transplantation 55:534, 1993 abstract
49. Seaberg EC, Belle SH, Beringer KC, et al: Liver transplantation in the
United States from 1987-1998: Updated results from the Pitt-UNOS liver
transplant registry. In Cecka JM, Terasaki PI (eds): Clinical Transplants
1998. Los Angeles, UCLA Tissue Typing Laboratory, 1999, p 17
50. Shaw BW Jr, Martin DJ, Marquez JN, et al: Venous bypass in clinical
liver transplantation. Ann Surg 200:524, 1984 abstract
51. Singh N, Gayowski T, Wagener MM, et al: Quality of life, functional

status, and depression in male liver transplant recipients with recurrent
viral hepatitis C. Transplantation 67:69, 1999 abstract
52. Snover DC, Fruse, DK, Sharp HL, et al: Liver allograft rejection: An
analysis of the use of biopsy in determining the outcome of rejection. Am
J Surg Pathol 11:1, 1987 abstract
255
53. Starzl
TE: History of liver and other splanchnic organ transplantation.
In Busuttil RW, Klintmalm GB (eds): Transplantation of the Liver.

Philadelphia, WB Saunders, 1996, p 3
54. Starzl
TE, Iwatsuki S, Van Thiel DH, et al: Evolution of liver
transplantation. Hepatology 2:614, 1982 citation
55. StarzlTE, Kaupp HA Jr, Brock DR, et al: Reconstructive problems in

canine liver homotransplantation with special reference to the
postoperative role of hepatic venous flow. Surg Gynecol Obstet 111:733,
1960
56. Starzl
TE, Kaupp HA Jr, Brock DR, et al: Studies on the rejection of the
transplanted homologous dog liver. Surgery, Gynecology and Obstetrics
112:135, 1961
57. StarzlTE, Klintmalm GBG, Porter KA, et al: Liver transplantation with
use of cyclosporin A and prednisone. N Engl J Med 305:266, 1981
citation
58. StarzlTE, Marchioro TL, Von Kaulla KN, et al: Homotransplantation

of the liver in humans. Surgery, Gynecology and Obstetrics 117:659, 1963
59. Starzl
TE, Todo S, Fung J, et al: FK506 for human liver, kidney, and
pancreas transplantation. Lancet 2:1000, 1989 abstract
60. Todo S, Nery J, Yanaga K, et al: Extended preservation of human liver

grafts with UW solution. JAMA 261:711, 1989 abstract
61. United Network for Organ Sharing. http://www.unos.org (updated

1999)
62. Vanderpool HY: Critical ethical issues in clinical trials with

xenotransplants. Lancet 351:1347, 1998 citation
63. Vargas HE, Laskus T, Wang LF, et al: Outcome of liver transplantation
in hepatitis C virus-infected patients who received hepatitis C
virus-infected grafts. Gastroenterology 117:149, 1999 abstract
64. VerranDJ, Asfar SK, Ghent CN, et al: Biliary reconstruction without
T-tubes or stents in liver transplantation: Report of 502 consecutive cases.
Liver Transpl Surg 3:365, 1997 abstract
65. Wachs ME, Bak TE, Karrer FM, et al: Adult living donor liver
transplantation using a right hepatic lobe. Transplantation 66:1313, 1998
abstract
66. WallWJ, Calne RY, Berbertson BM, et al: Simple hypothermic

preservation for transporting human livers long distance for
transplantation. Transplantation 23:210, 1977 abstract

67. WallWJ, Ghent CN, Roy A, et al: Use of OKT3 monoclonal antibody
as induction therapy for control of rejection in liver transplantation. Dig
Dis Sci 40:52, 1995 abstract
68. WelchCS: A note on transplantation of the whole liver in dogs.

Transplantation Bulletin 2:54, 1955
69. WiesnerRH: A long-term comparison of tacrolimus (FK506) versus

cyclosporine in liver transplantation: A report of the United States FK506
Study Group. Transplantation 66:493, 1998 abstract
70. Wright TL, Donegan E, Hsu HH, et al: Recurrent and acquired hepatitis
C viral infection in liver transplant recipients. Gastroenterology 103:317,
1992 abstract



Volume 4 Number 1 February 2000
241
HEPATOLOGY: A CENTURY OF PROGRESS
LIVER TRANSPLANTATION AT THE MILLENNIUM

Past, Present, and Future
Emmet B. Keeffe MD
Division of Gastroenterology, Department of Medicine, Stanford University School of Medicine; and the Liver Transplant
Program, Stanford University Medical Center, Stanford, California
Address reprint requests to:
Emmet B. Keeffe, MD
Stanford University Medical Center
750 Welch Road, Suite 210
e-mail: ekeeffe@stanford.edu
Transplantation is one of the greatest achievements of modern medicine. In AD 2000, liver

transplantation is well established as the definitive treatment for irreversible acute and chronic
liver disease, and progress continues both in scientific investigation and in clinical practice. The
care of liver transplant patients, once the domain of transplant centers, has more recently been
integrated into the community-based practice of medicine. The two major challenges currently
facing liver transplantation are the large and increasing disparity between the numbers of
available cadaver donor organs and of qualified patients listed and waiting for transplantation and
the management of recurrent disease after liver transplantation, particularly recurrent chronic
hepatitis C. Future issues concerning liver transplantation will probably center on the effective use
of available cadaver donor organs, including re-evaluation of selection criteria to strike a balance
between medical need and the chance of a successful outcome, and on the development of new
technologies to expand liver transplantation, such as cadaver split-liver transplantation,
adult-to-adult living related and unrelated liver transplantation, xenotransplantation, hepatocyte
transplantation, and liver-directed gene therapy. This article reviews the chronology of liver
transplantation, presents the current status of transplantation, and offers speculation regarding
the future directions of liver transplantation.

LIVER TRANSPLANTATION: THE PAST

Experimental Liver Transplantation
The first experimental attempts at liver transplantation, in dogs, were initiated 45 years ago, in 1955, by
Welch of Albany, New York, who described the
242
TABLE 1 -- SELECTED EVENTS IN THE EARLY HISTORY OF LIVER TRANSPLANTATION

Year Event
1955 First experimental liver transplantation using an auxiliary technique
(Welch)
1956 First experimental orthotopic liver transplantation (Cannon)
1963 Use of azathioprine and prednisone for immunosuppression
1963 First human liver transplantations (Starzl)
1966 Introduction of antilymphocyte globulin
1967 First 1-year survival after human liver transplantation (Starzl)
1980 Introduction of cyclosporine for immunosuppression
1983 NIH Consensus Development Conference on liver transplantation
1987 Use of University of Wisconsin solution for improved organ
preservation
1989 Introduction of tacrolimus for immunosuppression
insertion of an auxiliary liver engrafted heterotopically in either the pelvis or right paravertebral gutter
(Table 1) . [68] The portal vein was anastomosed to the inferior vena cava and the hepatic artery to the
aorta or iliac artery, and no immunosuppression was used. The first experimental liver replacement, that
is, orthotopic liver transplantation, was reported by Cannon at the University of California at Los
Angeles in 1956, but none of those dogs survived. [11] Cannon had speculated that the liver might play a
role in rejection and that a replaced liver should not contribute to its own destruction.
In 1958, liver transplantation was established as a primary research focus at Peter Bent Brigham Hospital
in Boston, Massachusetts, under the direction of Moore, [39] [40] and Northwestern University in Chicago,
Illinois, under the direction of Starzl. [55] [56] The focus of research in Boston was the immunology of liver

transplantation; in Chicago, it was liver regeneration and hepatotrophic growth factors. Initial successes
were hampered by the technical challenges in the performance of liver transplantation in dogs and the
inevitable development of allograft rejection. Two technical improvements ultimately provided better
survival rates; adequate preservation of the graft with portal infusion of chilled lactated Ringer's solution
leading to core cooling, [55] and decompression by bypassing the obstructed recipient splanchnic and
systemic venous beds during the anhepatic phase when the native liver was being removed and donor
liver grafted in its place. [40] [55]
The original saline preservative solutions were replaced by improved solutions. The Collins solution
allowed safe preservation for 5 to 6 hours [2] [66] and later the University of Wisconsin solution extended
the preservation time 18 to 24 hours. [28] [29] [60] As experience with human orthotopic liver transplantation
evolved, venous bypasses were used less often and now are seldom employed. [54]
Early Development of Immunosuppression
The development of the field of immunosuppression was critically important in liver transplantation. The
first attempts at liver transplantation had been performed without immunosuppression, and thus no
long-term organ engraftment was possible. In 1944, Medawar showed that graft rejection is an
immunological event that has both specificity and memory. [36] [37] The initial attempts to suppress the
immune system to ameliorate rejection used corticosteroid therapy or total-body irradiation. In the late
1950s and early 1960s 6-mercaptopurine and azathioprine were introduced as immunosuppressants
(Table 2) . [53]
243
In animal experiments and in early human cadaver kidney transplantation, it was learned that
single-agent immunosuppression only rarely controlled rejection but that the combination of prednisone
and azathioprine effectively prevented rejection and allowed successful renal transplantation. It also
become evident that large boluses of corticosteroids could be used to reverse episodes of acute allograft
rejection. These two observations by Starzl and colleagues at the University of Colorado made clinical
transplantation possible and opened the way for human liver transplantation. [53]
Human Liver Transplantation
The first attempted human liver transplantation was reported in 1963 by Starzl. [58] The recipient was a
3-year-old boy with biliary atresia who had had multiple previous operations who and died of blood loss
during surgery because of uncontrollable coagulopathy. Two other liver transplantations were carried out
in the same year, but the recipients died after 22 and 7 days, respectively. [53] [58] In the next year, isolated
attempts at human liver transplantation were unsuccessful in Boston [37] and in Paris. [16] These first seven
human liver transplant operations achieved patient survivals ranging from zero to 23 days. [53]
The first truly successful human liver transplantation was performed in 1967 by Starzl at the University
of Colorado. The recipient was an 18-month-old child with a hepatocellular carcinoma who survived
more than 1 year before succumbing to recurrent tumor. [53] Six other patients underwent liver
transplantation in 1967 and 1968, with a maximum survival of 30 months. During the next 12 years,
approximately one liver transplantation per month was performed at the University of Colorado, and the
1-year mortality rate was greater than 50%. The long-term survival of liver transplant recipients

remained at 30%. [54] On a more encouraging note, 30 of the first 170 patients (18%) in the consecutive
series of liver transplants at the University of Colorado from 1963 through 1979 lived more than 10
years. [53] The usual immunosuppressive regimen was prednisone, azathioprine, and polyclonal
antilymphocyte globulin. [54]
During this period, Calne of Cambridge University and the hepatologist Williams of King's College
Hospital in London began clinical trials in liver transplantation. [10] Their first patient also exsanguinated,
but success soon followed. Other liver transplantation teams were established in Hanover, Germany, in
1972, under the direction of Pichlmayr, and in Paris in 1974, under Bismuth. By 1980, it had been shown
that liver transplantation could be accomplished,
TABLE 2 -- HISTORY OF IMMUNOSUPPRESSIVE DRUG REGIMENS
Year Agent
1962 Azathioprine
1963 Azathioprine plus corticosteroids
1966 Polyclonal antibodies; antilymphocyte globulin as an adjunct
1970 Cyclophosphamide substituted for azathioprine
1978 Cyclosporine use in humans
1980 Cyclosporine plus corticosteroids
1981 Development of monoclonal antibodies
1989 Tacrolimus plus corticosteroids
1990s Development of newer agents (e.g., mycophenolate mofetil,
rapamycin)
244
but the relatively high mortality rate seemed to indicate that the procedure was not practical.
Development of Cyclosporine and Tacrolimus
The evolution of liver transplantation from an experimental operation applied to a few individuals to
routine surgery with excellent graft and patient survival rates resulted from several advances, but the
development of cyclosporine has been credited with the ultimate success and wide acceptance of liver
transplantation. The initial experimental and clinical effectiveness of cyclosporine in transplantation was
demonstrated by Calne and colleagues in 1978 and 1979. [7] [8] [9] Cyclosporine was the first selective
method of immunosuppression, and its use increased the survival rate of liver transplant recipients from
approximately 30% to more than 70%. [27] [57] The development of cyclosporine in the late 1970s and
early 1980s led to the acceptance of liver transplantation for routine patient care by the National
Institutes of Health Consensus Development Conference in 1983 [42] and to the opening of many new
transplant centers.

With the initial use of tacrolimus (FK506) at the University of Pittsburgh, [59] the 1-year graft and patient
survival rates increased further. Later direct comparison between cyclosporine and tacrolimus in
controlled trials, however, produced equivalent results that were quite good with either agent used as the
basis of a multiple-drug immunosuppressive regimen. [1] [47] [69] Although the merits of cyclosporine
versus tacrolimus continue to be debated, there is a clinical advantage in having available two good
agents that can be substituted as clinical circumstances dictate.
Another advance in immunosuppression was the development of monoclonal antibodies, particularly
muromonab-CD3 that is used to control steroid-resistant rejection and is occasionally used for induction
of immunosuppression in the setting of renal insufficiency. [43] [48] [67] Muromonab-CD3 has now
essentially replaced the earlier antilymphocyte and antithymocyte globulins, which could never be
standardized. Other agents, such as mycophenolate mofetil and rapamycin, are being tested in clinical
trials to determine their roles in liver transplantation. [17]
Additional Developments in Liver Transplantation
The initial technical achievement of liver replacement and the development of improved
immunosuppressive regimens were accompanied by many other advances that led to the current 1-year
patient survival rates of 85% to 90% after liver transplantation for most conditions (Table 3) . [49] Animal
experiments were inadequate preparation for liver transplantation in sick patients with portal
hypertension and severe coagulopathy, many of whom had prior right upper quadrant surgery with
adhesions. The management of hemodynamic and metabolic problems that may arise during surgery by
anesthiologists who specialize in liver transplantation, the use of modern blood component and
coagulation factor replacement therapy, and improved surgical methods to control operative bleeding
have all improved the outcome of liver transplantation. [12] [54] The massive blood loss that routinely
characterized earlier operations has been replaced by minimal blood loss, and as many as 30% of liver
transplantations can be performed without blood transfusion. [6] The introduction of venovenous bypass,
which was important in the early experience of surgeons with liver transplantation,
245
TABLE 3 -- SURVIVAL AFTER ADULT LIVER TRANSPLANTATION BY DIAGNOSIS *

Survival (% )
Diagnosis 1-Year 4-Year 7-Year
Primary sclerosing cholangitis 91 84 78
Primary biliary cirrhosis 89 84 79
Autoimmune hepatitis 86 81 78
Chronic hepatitis C 86 75 67
Alcoholic liver disease 85 76 63
Cryptogenic cirrhosis 84 76 67

Chronic hepatitis B 83 71 63
Malignancy 72 43 34
Data from Seaberg EC, Belle SH, Beringer KC, et al: Liver transplantation in the United States from
1987-1998: Updated results from the Pitt-UNOS liver transplant registry. In Cecka JM, Teraski PI
(eds): Clinical transplants 1998. Los Angeles, UCLA Tissue Typing Laboratory, 1999, p 17.
*United Network of Organ Sharing Database 1987-1998; n = 24,900 patients.
facilitated the development of new programs by supporting hemodynamics during the anhepatic phase.
[50]
Biliary complications were a major source of morbidity and mortality in early liver transplantations, with
as many as 50% of patients developing one or more biliary complications. [54] A primary duct-to-duct
anastomosis can now be performed without T-tubes or stents, a development which has reduced the
biliary complication rate to 5% to 10%. [64] The use of radiologic and endoscopic diagnostic and
therapeutic procedures has also facilitated the management of biliary complications with lower rates of
morbidity.
Other important advances include better understanding and diagnosis of acute and chronic allograft
rejection and improved prevention and treatment of infections that occur in the immunosuppressed
transplant recipient. [22] [52] The routine use of posttransplantation allograft liver biopsies has allowed
more accurate diagnosis of acute rejection and identification of other pathologic processes that can cause
early allograft dysfunction. Prophylactic regimens for early bacterial infection after liver transplantation
and for late cytomegalovirus, fungal, and Pneumocystis infections are now routine and have substantially
reduced morbidity and mortality from these infections.
LIVER TRANSPLANTATION: THE PRESENT

The first 3 decades of liver transplantation witnessed great successes in the development of surgical
techniques and improvement in immunosuppressive regimens. These successes are accompanied by a
new set of problems. In particular, the growing disparity between the availability of cadaver donor livers
and the number of potential transplant recipients on the waiting lists in the United States and other
countries has led to increased deaths among persons on the waiting list and sicker patients coming to
transplantation. An increasing number of transplant patients have chronic hepatitis C virus (HCV)
infection, and recurrent hepatitis C is an important cause of morbidity and mortality that may reduce the
long-term utility of liver grafting. Newer approaches to solving these problems are under way, but no
comprehensive solution is in sight.
246

TABLE 4 -- NUMBERS OF LIVER TRANSPLANTS, PATIENTS AND DEATHS ON THE LIVER

WAITING LIST IN THE UNITED STATES FROM 1988 TO 1997
(Not Available)
Modified from Keeffe EB: Summary of guidelines on organ allocation and patient listing for liver
transplantation. Liver Transplant Surg 4:S108, 1998; with permission.
247
Donor Shortage
The number of liver transplantations performed in 1996, 1997, and 1998 in the United States was
relatively stable, at slightly more than 4000 transplantations annually. Currently, however, there are more
than 12,000 potential liver transplant candidates on the United Network for Organ Sharing (UNOS)
waiting list in the United States, and therefore, a lower percentage of the patients with chronic liver
disease who might benefit from liver transplantation can actually undergo the procedure. This growing
discrepancy between the available donor organs and the need for transplantation has led to scrutiny of the
traditional selection and listing criteria for liver transplantation.
The supply of donor livers is insufficient to meet the current and future need for liver transplantation, and
organ donation has remained relatively stagnant, with only minor increases in retrieval rates in recent
years in spite of many efforts to increase donation. The growing disparity between the number of liver
transplantations that can be performed each year and the number of patients on the waiting list is
reaching crisis proportions in the United States. Although the number of donors and liver transplantations
increased 2.4-fold between 1988 and 1997, the number of patients on the liver transplantation list
increased 15.6-fold, and the number of deaths increased 5.8-fold during the same period. (Table 4)
(Table Not Available) . [61] This situation will only worsen in the coming years. Analysis of data from the
past 5 years to predict the status of liver transplantation in the year 2000, assuming no major changes in
organ availability or performance of transplantation, reveals an even greater crisis at the turn of the
century with 20,000 or more patients awaiting a liver transplantation (Table 5) (Table Not Available) . [30]
United Network for Organ Sharing
The disparity between the number of cadaver organs and candidates for liver transplantation has led to
re-evaluation of UNOS allocation and distribution policies for livers and has spurred a nationwide debate
among transplantation professionals, potential recipients, and the federal government. [5] [23] [44]
Distribution policies determine the geographical areas within which livers are allocated, and allocation
policies determine which patients within a geographical area will receive the available livers. [44] The
historical allocation scheme has
TABLE 5 -- GROWTH OF LIVER TRANSPLANTS AND WAITING LIST
(Not Available)
Modified from Keeffe EB: Summary of guidelines on organ allocation and patient listing for liver
transplantation. Liver Transplant Surg 4:S108, 1998; with permission.

248
dictated that the sickest patients who have waited the longest receive transplantation first. In the recent
past, there have been modifications and redefinition of UNOS status based on disease severity, but the
basic principle, as shown in the box , still holds.
UNOS Liver Status For Patients 18 Years of Age According to Disease Severity
Status 1 Fulminant liver failure with life expectancy <7

days, including FHF, primary graft nonfunction,
hepatic artery thrombosis, and acute
decompensated Wilson's disease
Status 2A Hospitalized in ICU for chronic liver failure with
life expectancy <7 days, with a Child-Pugh score
of 10 and one of following: unresponsive
active variceal hemorrhage, hepatorenal
syndrome, refractory ascites/hepatic hydrothorax,
or stage 3 or 4 hepatic encephalopathy
Status 2B Requiring continuous medical care, with a
Child-Pugh score of 10, or a Child-Pugh
score 7 and one of the following:
unresponsive active variceal hemorrhage,
hepatorenal syndrome, spontaneous bacterial
peritonitis, or refractory ascites/hepatic
hydrothorax; presence of an hepatocellular
carcinoma.
Status 3 Requiring continuous medical care, with a
Child-Pugh score of 7, but not meeting
criteria for status 2B
Status 7 Temporarily inactive
Data from United Network of organ sharing. Implemented July 1997; modified January 1998 and
August 1998. http://www.unos.org(updated 1999)
FHF = fulminant hepatic failure

The UNOS distribution scheme as of mid-1999 dictates that patients at transplant programs served by a
local organ procurement organization (OPO) have the first priority for livers obtained by that OPO. [5] [23]
[44] Because of this local use of livers, the waiting times for liver transplantation differ considerably for
patients listed in different regions in the United States. This local policy means that a patient in one OPO
who is not in immediate danger of dying may receive a transplantation before a sicker patient in another
OPO in a nearby geographic area. This reality has led to discussion about widening the liver distribution
area for patients who are listed as status 1, but computer simulation modeling by UNOS showed that
wider sharing, that is, a single national waiting list, would allow urgency (often termed justice) to prevail
over medical utility. Utility factors focus on maximizing the overall benefits of transplantation to society,
that is, giving priority to the patient who maximizes the chances of a successful outcome by having the
least risk of dying after transplantation. Urgency, or justice, recognizes the needs of the individual
transplant patient by giving priority to the sickest patient, who has the greatest risk of dying before
transplantation. The argument made against current UNOS policies is that the geographically restricted
distribution scheme overrides prioritization based on medical urgency in favor of utility and is unfair to
transplant candidates in most need of liver replacement. [23] This argument proposes a national waiting
list to equalize access
249
to liver transplantation, that is, a patient-driven rather than a local or center-driven allocation scheme.
Finally, the transplantation community is becoming cognizant of cost-outcome analyses and the reality of
managed care, which has transferred financial risk from insurers to providers. [20] High-risk patients
represent a significant liability to transplantation centers in the managed care marketplace. Appropriate
and reasonable patient selection may become an important consideration in allocation policies.
Uniform Listing Criteria for Chronic Liver Disease
At a recent consensus conference at the National Institutes of Health organized by the American Society
of Transplantation and the American Association for the Study of Liver Diseases, uniform minimal
listing criteria were developed for general application to patients with miscellaneous chronic liver
diseases. [33] These criteria are
Immediate need for liver transplantation
Estimated 1-year survival 90%
Child-Pugh score 7 (Child-Pugh class B or C)
Portal hypertensive bleeding or a single episode of spontaneous bacterial peritonitis, irrespective
of Child-Pugh score
These criteria were established using the general principle that any patient listed should have an expected
1-year survival with general supportive care of 90% or less, which is less than expected with liver
transplantation. Large studies of the natural history of patients with compensated cirrhosis resulting from
miscellaneous causes [25] [46] or chronic hepatitis C [21] have shown that survival is relatively good until

decompensation, when 5-year survival rates fall to approximately 50%. Thus, the indication for listing
for liver transplantation should not be simply the presence of cirrhosis, without decompensation, that is,
Child-Pugh class A, but should be the development of decompensation, for example, Child-Pugh class B
or C.
The current approaches to the organ shortage include increased efforts to achieve higher rates of organ
procurement, expanded use of marginal donors, and surgical alternatives such as cadaver split-liver and
adult living donor related and unrelated liver transplantation. Xenotransplantation may become an option
in the future.
There has been a concerted effort to increase organ donation, and in some countries small gains have
been made in the past years. In the United States, there was 5.6% increase in cadaver donors in 1998, the
first substantial increase since 1995. [61] The organ donation rate in the United States is approximately
20/1 million population, compared with Spain, the leading Western nation, with donor rates of 25/1
million and Italy, the worst Western country, with a rate of less than 10/1 million.
The expanded use of marginal donors in recent years has included implanting donor livers from older
individuals, use of grafts with substantial fatty change, and engrafting donor grafts from patients with
mild chronic hepatitis C into recipients with hepatitis C or from patients with a positive hepatitis B core
250
antibody (anti-HBc) into recipients with hepatitis B, and, occasionally, into other recipients. This
experience is too limited for long-term outcome to be compared with patients receiving grafts from
uninfected donors, but it is known that anti-HBc-positive grafts frequently transmit hepatitis B virus
(HBV), and recipients without HBV infection are, therefore, at risk and should receive prophylaxis in the
form of lamivudine or hepatitis B immune globulin. [19] The short-term and medium-term outcome of
transplanting HCV-positive grafts into patients with chronic hepatitis C seems to be good. [63]
Although increasing donor age is associated with poor graft function, the organ shortage seems to justify
use of these grafts, and the overall percentage of donors over the age of 50 years has increased
substantially. [18] It seems that the judgment of the harvesting surgeon as to whether the donor liver is
good, fair, or poor is also an important factor affecting graft survival. [26] There were no differences in the
3-month graft survival rates using livers from younger donors judged good versus those considered fair
or poor, but there was a significantly lower survival rate using fair or poor grafts from donors aged 50
years or older (61% versus 92% for organs classified as good). [26] It has been suggested that
donor/recipient gender matching should also be considered. [4] [35] Because livers from women
transplanted into men generally have a poorer outcome, transplantation of grafts from older women into
men may be even more likely to result to in poor graft function.
Split-liver transplantation, after unsatisfactory initial results, has undergone a resurgence during the past
2 to 3 years. This procedure essentially achieves two liver transplantations from a single cadaver liver,
with the right lobe usually implanted into an adult recipient and the left lobe or left lateral segment
transplanted into a child. The organ used in split-liver transplantation can obviously be shared between
two institutions or used at a single transplant center. Living-donor liver transplantation for adults, using a

right hepatic lobe, is also being performed more often, and good results are being reported from centers
in the United States [34] [65] and Japan. [31] [32] The Japanese center has also had good outcomes using
living-donor liver transplantation for high-urgency patients. [32]
Recurrence of Disease After Liver Transplantation
Chronic hepatitis C has become the most common cause of end-stage liver disease requiring liver
transplantation and accounts for 25% to 40% of all transplantations in some centers. [61] Hepatitis C viral
infection may also be present in patients with alcoholic liver disease and in patients undergoing liver
transplantation who are classified as having cryptogenic cirrhosis. Most patients have had HCV infection
for several decades before the onset of liver failure.
After liver transplantation, HCV reinfection occurs in almost all patients. Most of these patients
subsequently develop chronic hepatitis, and a few progress to cirrhosis. [45] [70] Fortunately, the infection
appears to be benign in 80% to 85% of patients on short-term and medium-term follow-up, and survival
rates are comparable with patients receiving transplantation for nonviral chronic liver diseases. It has
been shown, however, that transplant recipients with recurrent hepatitis C have poorer quality of life,
greater depression, and higher psychologic distress than those without HCV infection. [51] Whether
progressive chronic hepatitis and cirrhosis will occur in most patients with longer follow-up remains
uncertain but is possible.
Numerous host and viral factors have been implicated in the development
251
of severe recurrent hepatitis C, suggesting that the process in multifactorial. The overlapping histologic
features of HCV infection and allograft rejection, which include portal and parenchymal mononuclear
cell infiltrates, fatty change, swollen hepatocytes with necrosis, and occasional bile duct damage, make
distinction between these two entities difficult at times. [45]
The best treatment of recurrent HCV infection after liver transplantation remains uncertain. It seems that
neither interferon nor ribavirin alone is beneficial, [24] but the combination of both agents shows promise.
[3] In one study, 18 of 21 patients tolerated therapy, and 24% of patients experienced a virologic sustained
response with improved aminotransferase levels and liver biopsy histologic scores. [3] Whether
maintenance therapy can be discontinued in patients who have a sustained virologic response remains
unknown. Studies are in progress to determine if preemptive therapy early after liver transplantation will
alter the posttransplantation infection rate or the severity of recurrent hepatitis C.
Other diseases can recur after liver transplantation, including HBV infection, alcoholic liver disease, and
immunologic liver diseases, such as primary biliary cirrhosis, primary sclerosing cholangitis, and
autoimmune hepatitis. [15] The strategies for diagnosing or preventing these entities are well established,
unlike the management challenges presented by recurrent hepatitis C.

LIVER TRANSPLANTATION: THE FUTURE

In the immediate future, developments in liver transplantation will probably extend from the
developments of the past few years, enlarging the efforts to increase organ donation, further expanding
split-liver and adult living related and unrelated liver transplantation, and optimizing the use of the
existing donor pool by adjusting allocation and distribution policies and refining selection criteria.
Possible clinical developments in the near future include xenotransplantation, hepatocyte transplantation,
and liver-directed gene therapy.
Xenotransplantation
If xenotransplantation can be made safe and affordable, it may solve many of the current problems of
access to liver transplantation. Xenotransplantation is the engraftment of organs obtained from one
species into another species. Xenografting is not a new concept, but it has resurfaced with the availability
of more potent immunosuppressive agents and the critical shortage of human cadaver organs. Most
investigators are now focusing on the pig as a potential donor based on appropriate size, unlimited
supply, ability to be genetically engineered, and the more easily controlled risk of zoonotic infection. [41]
The immunologic hurdles to xenotransplantation are hyperacute rejection, acute vascular rejection, and
cellular rejection. Another important consideration in xenotransplantation is the potential transmission of
infectious agents from the graft to the recipient. It has been suggested that the movement to
xenotransplantation will occur as a step-by-step process, beginning with limited clinical trials, using
xenotransplantation initially as a bridge to human cadaver transplantation, then implanting porcine
xenografts in patients who cannot obtain a human graft, and finally using xenotransplantation as an
alternative to allotransplantation. [41] Before this sequence can unfold, many critical ethical issues must
be addressed. [14] [62]
252
Hepatocyte Transplantation and Liver-Directed Gene Therapy
Liver cell transplantation is being developed to treat acute and chronic liver failure and inherited
metabolic disorders. [13] Liver cells can be isolated from a number of species, including humans, and then
cultured or cryopreserved for future use. Cultured cells can be directly transplanted from allogeneic
donors (a process that requires immunosuppression) or transplanted back into an individual after being
surgically harvested and transduced in culture with a therapeutic gene for either a defective or absent
protein. [13] In the latter case, immunosuppression is not required. Liver-directed gene therapy can be
used to replace a missing gene, express a gene that is not normally expressed in the liver, interfere with
gene expression, disrupt an offending gene, or repair a mutated gene. [13] Hepatocyte transplantation has
already been successfully used to treat some inherited disorders, and both nonviral and viral vectors are
being developed for gene therapy targeted at the liver. Both technologies will probably develop into more
practical therapies.
It is hoped that in the future the need for liver transplantation will be reduced if effective medical
treatments, some of which may employ gene-direct therapy, can be discovered to treat the full spectrum
of metabolic, viral, and immunologic liver diseases. The late 1990s have already witnessed some
advances in the treatment of chronic hepatitis C and chronic hepatitis B, and the next few years should

bring improved treatments for many liver diseases that have historically progressed inevitably to
cirrhosis and liver failure, with the resultant need for liver transplantation.
SUMMARY
During the past 3 decades, liver transplantation has achieved such acceptance that more than 12,000
qualified recipients are listed for liver transplantation in the United States, but unfortunately just over
4000 cadaver donor organs are available each year. Thus, given the increasing disparity between the
number of potential recipients and available cadaver organs, the current challenge in liver transplantation
is to optimize the outcome of liver transplantation from this limited resource. Currently under way is
re-evaluation of selection criteria to use these 4000 cadaver liver grafts most effectively by striking the
proper balance between medical urgency and utility. In parallel with this re-evaluation, there is ongoing
expansion of cadaver split-liver transplantation and adult living related and unrelated liver
transplantation. Hoped-for but as yet unachieved developments in liver transplantation are
xenotransplantation, hepatocyte transplantation, and liver-directed gene therapy. Liver transplantation
has come a long way from the initial, unsuccessful human transplantations in 1963, but many challenges
remain.
References
1. A
comparison of tacrolimus (FK506) and cyclosporine for immunosuppression in liver transplantation. The U.S.
Multicenter FK506 Liver Study Group. N Engl J Med 331:1110, 1994
2. BenichouJ, Halgrimson, CG, Weil R III, et al: Canine and human liver preservation for 6 to 18 hours by cold infusion.
253
3. BizollonT, Palazzo U, Ducerf C, et al: Pilot study of the combination of interferon alfa and ribavirin as therapy of
recurrent hepatitis C after liver transplantation. Hepatology 26:500, 1997 abstract
4. Brooks BK, Levy MF, Jennings LW, et al: Influence of donor and recipient gender on the outcome of liver
5. Burdick
JF, Klein AS, Harper AM: The debate over liver allocation in the United States: The UNOS perspective. In
Cecka JM, Terasaki PI (eds): Clinical Transplants 1996. Los Angeles, UCLA Tissue Typing Laboratory, 1997, p 322
6. Cacciarelli
TV, Keeffe EB, Moore DH, et al: Primary liver transplantation without transfusion of red blood cells.
Surgery 120:698, 1996 abstract
7. Calne RY, Rolles K, White DJG, et al: Cyclosporin A initially as the only immunosuppressant in 34 recipients of

cadaver organs: 32 kidneys, 2 pancreases, and 2 livers. Lancet 2:1033, 1979 abstract
8. CalneRY, White DJ, Rolles K, et al: Prolonged survival of pig orthotopic heart grafts treated with cyclosporin A. Lancet
1:1183, 1978 abstract
9. CalneRY, White DJG, Thiru S, et al: Cyclosporin A in patients receiving renal allografts from cadaver donors. Lancet
2:1323, 1978 abstract
10. CalneRY, Williams R: Liver transplantation in man. I. Observations on technique and organization in five cases. BMJ
4:535, 1968
11. Cannon JA: Brief report. Transplant Bull 3:7, 1956
12. Carmichael FJ, Lindop MJ, Farman JV: Anaesthesia for hepatic transplantation: Cardiovascular and metabolic
alterations and their management. Anesth Analg 64:108, 1985 abstract
13. ChowdhuryJR: Forward: Prospects of liver cell transplantation and liver-directed gene therapy. Semin Liver Dis 19:1,
1999 abstract
14. DaarAS: Ethics of xenotransplantation: Animal issues, consent, and likely transformation of transplant ethics. World J
Surg 21:975, 1997 abstract
15. Davern TJ, Lake JR: Recurrent disease after liver transplantation. Semin Gastrointest Dis 9:86, 1998 abstract
16. Demirleau,Noureddine, Vignes, et al: Tentative d'homograffe hepatique (Attempted hepatic homograft). Memoires de
l'Academie de Chirurgie 90:177, 1964
17. Denton MD, Magee CC, Sayegh MH: Immunosuppressive strategies in transplantation. Lancet 353:1083, 1999
citation
18. Detre
KM, Lombardero M, Belle S, et al: Influence of donor age on graft survival after transplantation--United Network
for Organ Sharing Registry. Liver Transpl Surg 1:311, 1995 abstract
19. Dickson RC, Everhart JE, Lake JR, et al: Transmission of hepatitis B by transplantation of livers from donors positive
for antibody to hepatitis B core antigen. The National Institute of Diabetes and Digestive and Kidney Diseases liver
transplantation database. Gastroenterology 113:1668, 1997 abstract
20. EvansRW, Kitzmann DJ: The "arithmetic" of donor liver allocation. In Cecka JM, Terasaki PI (eds): Clinical
Transplants 1996. Los Angeles, UCLA Tissue Typing Laboratory, 1997, p 338
21. FattovichG, Giustina G, Degos F, et al: Morbidity and mortality in compensated cirrhosis type C: A retrospective
follow-up study of 384 patients. Gastroenterology 112:463, 1997 abstract
22. Fishman JA, Rubin RH: Infection in organ-transplant recipients. N Engl J Med 338:1741, 1998 citation
23. FungJJ, Roberts JP: The rationale for equitable liver allocation. In Cecka JM, Terasaki PI (eds): Clinical Transplants
1996. Los Angeles, UCLA Tissue Typing Laboratory, 1997, p 325

24. Gane EJ, Lo SK, Riordan SM, et al: A randomized study comparing ribavirin and interferon alfa monotherapy for
hepatitis C recurrence after liver transplantation. Hepatology 27:1403, 1998 abstract
25. Gines
P, Quintero E, Arroyo V, et al: Compensated cirrhosis: Natural history and prognostic factors. Hepatology 7:122,
1987 abstract
26. Hoofnagle
JH, Lombardero M, Zetterman RK, et al: Donor age and outcome of liver transplantation. Hepatology 24:89,
1996 abstract
254
27. IwatsukiS Starzl TE, Todd S, et al: Experience in 1,000 liver transplants under cyclosporine-steroid therapy: A survival
report. Transplant Proc 20(suppl 1):498, 1988 citation
28. Jamieson NV, Sundberg R, Lindell S, et al: Successful 24- to 30-hour preservation of the canine liver: A preliminary
report. Transplant Proc 20(suppl 1):945, 1988 citation
29. Kalayoglu
M, Sollinger WH, Stratta RJ, et al: Extended preservation of the liver for clinical transplantation. Lancet
1:617, 1988 abstract
30. KeeffeEB: Summary of guidelines on organ allocation and patient listing for liver transplantation. Liver Transpl Surg
4(suppl 1):S108, 1998 abstract
31. LoCM, Fan ST, Liu CL, et al: Adult-to-adult living donor liver transplantation using extended right lobe grafts. Ann
Surg 226:261, 1997 full text
32. Lo
CM, Fan ST, Liu CL, et al: Applicability of living donor liver transplantation to high-urgency patients.
33. Lucey MR, Brown KA, Everson GT, et al: Minimal criteria for placement of adults on the liver transplant waiting list:
A report of a national conference organized by the American Society of Transplant Physicians and the American
Association for the Study of Liver Diseases. Liver Transpl Surg 3:628, 1997 abstract
34. Marcos A, Fisher RA, Ham JM, et al: Living donor liver transplantation (LDLTx) for adults utilizing the right lobe
[abstract]. Gastroenterology 29:A27, 1999
35. MarinoIR, Doyle, Aldrighetti L, et al: Effect of donor age and sex on the outcome of liver transplantation. Hepatology
22:1754, 1995 abstract
36. Medawar PB: Second study of behavior of skin homografts in rabbits. J Anat 79:157, 1945
37. Medawar PB: The behavior and fate of skin autografts and skin homografts in rabbits. J Anat 78:176, 1944
38. Moore FD, Birtch AG, Dagher F, et al: Immunosuppression and vascular insufficiency in liver transplantation. Ann N Y

Acad Sci 102:729, 1964
39. Moore FD, Smith LL, Burnap TK, et al: One-stage homotransplantation of the liver following total hepatectomy in
dogs. Transplantation Bulletin 6:103, 1959
40. Moore FD, Wheeler HB, Demissianos HV, et al: Experimental whole organ transplantation of the liver and of the
spleen. Ann Surg 152:374, 1960
41. Nagayasu T, Platt JL: Progress in xenotransplantation. Graft 1:19, 1998
42. National
Institutes of Health Consensus Development Conference Statement: Liver transplantation-June 20-23, 1983.
Hepatology 4(suppl):107S, 1984
43. Neuhaus
P, Bechstein WO, Blumhardt G, et al: Comparison of quadruple immunosuppression after liver transplantation
with ATG or IL-2 receptor antibody. Transplantation 55:1320, 1993 abstract
44. Norman DJ: Allocation of livers for liver transplantation: Ethics and politics. Clin Liver Dis 1:281, 1997
45. Pessoa MG, Wright TL: Hepatitis C infection in transplantation. Clin Liver Dis 1:663, 1997
46. Propst A, Propst T, Zangerl G, et al: Prognosis and life expectancy in chronic liver disease. Dig Dis Sci 40:1805, 1995
abstract
47. Randomised
trial comparing tacrolimus (FK506) and cyclosporin in prevention of liver allograft rejection. European
FK506 Multicentre Liver Study Group. Lancet 344:423, 1994 abstract
48. Reding R, Vraux H, de Ville de Goyet J, et al: Monoclonal antibodies in prophylactic immunosuppression after liver
49. SeabergEC, Belle SH, Beringer KC, et al: Liver transplantation in the United States from 1987-1998: Updated results
from the Pitt-UNOS liver transplant registry. In Cecka JM, Terasaki PI (eds): Clinical Transplants 1998. Los Angeles,
50. Shaw
BW Jr, Martin DJ, Marquez JN, et al: Venous bypass in clinical liver transplantation. Ann Surg 200:524, 1984
abstract
51. Singh N, Gayowski T, Wagener MM, et al: Quality of life, functional status, and depression in male liver transplant
recipients with recurrent viral hepatitis C. Transplantation 67:69, 1999 abstract
52. Snover
DC, Fruse, DK, Sharp HL, et al: Liver allograft rejection: An analysis of the use of biopsy in determining the
outcome of rejection. Am J Surg Pathol 11:1, 1987 abstract
255
53. Starzl TE: History of liver and other splanchnic organ transplantation. In Busuttil RW, Klintmalm GB (eds):

Transplantation of the Liver. Philadelphia, WB Saunders, 1996, p 3
54. Starzl TE, Iwatsuki S, Van Thiel DH, et al: Evolution of liver transplantation. Hepatology 2:614, 1982 citation
55. Starzl
TE, Kaupp HA Jr, Brock DR, et al: Reconstructive problems in canine liver homotransplantation with special
reference to the postoperative role of hepatic venous flow. Surg Gynecol Obstet 111:733, 1960
56. Starzl
TE, Kaupp HA Jr, Brock DR, et al: Studies on the rejection of the transplanted homologous dog liver. Surgery,
Gynecology and Obstetrics 112:135, 1961
57. Starzl
TE, Klintmalm GBG, Porter KA, et al: Liver transplantation with use of cyclosporin A and prednisone. N Engl J
Med 305:266, 1981 citation
58. Starzl
TE, Marchioro TL, Von Kaulla KN, et al: Homotransplantation of the liver in humans. Surgery, Gynecology and
Obstetrics 117:659, 1963
59. Starzl
TE, Todo S, Fung J, et al: FK506 for human liver, kidney, and pancreas transplantation. Lancet 2:1000, 1989
abstract
60. Todo
S, Nery J, Yanaga K, et al: Extended preservation of human liver grafts with UW solution. JAMA 261:711, 1989
abstract
61. United Network for Organ Sharing. http://www.unos.org (updated 1999)
62. Vanderpool HY: Critical ethical issues in clinical trials with xenotransplants. Lancet 351:1347, 1998 citation
63. VargasHE, Laskus T, Wang LF, et al: Outcome of liver transplantation in hepatitis C virus-infected patients who
received hepatitis C virus-infected grafts. Gastroenterology 117:149, 1999 abstract
64. VerranDJ, Asfar SK, Ghent CN, et al: Biliary reconstruction without T-tubes or stents in liver transplantation: Report
of 502 consecutive cases. Liver Transpl Surg 3:365, 1997 abstract
65. WachsME, Bak TE, Karrer FM, et al: Adult living donor liver transplantation using a right hepatic lobe.
66. WallWJ, Calne RY, Berbertson BM, et al: Simple hypothermic preservation for transporting human livers long
distance for transplantation. Transplantation 23:210, 1977 abstract
67. Wall WJ, Ghent CN, Roy A, et al: Use of OKT3 monoclonal antibody as induction therapy for control of rejection in
liver transplantation. Dig Dis Sci 40:52, 1995 abstract
68. Welch CS: A note on transplantation of the whole liver in dogs. Transplantation Bulletin 2:54, 1955
69. WiesnerRH: A long-term comparison of tacrolimus (FK506) versus cyclosporine in liver transplantation: A report of
the United States FK506 Study Group. Transplantation 66:493, 1998 abstract

70. WrightTL, Donegan E, Hsu HH, et al: Recurrent and acquired hepatitis C viral infection in liver transplant recipients.
Gastroenterology 103:317, 1992 abstract


April 22, 2000
Conde Petra
Citation
Bibliographic Data
Abstract
Indexing Data
Busuttil RW - Ann Surg - 1999 Mar; 229(3): 313-21
NLM Citation ID:
99174660
Full Text Full Source Title:
Annals of Surgery
Frontmatter
Publication Type:
Split Liver Transplantation Journal Article; Review; Review, Tutorial
Ex Vivo Technique Language:
In Situ Technique English
Size Matching in Split Liver Author Affiliation:

Transplantation Department of Surgery, The Dumont-UCLA Transplant Center,
University of California, Los Angeles School of Medicine,
Results of In Situ Splits 90095-7054, USA.
Split Liver Policy: Impact on Waiting
Authors:
Time
Busuttil RW; Goss JA
Application of Split Liver
Transplantation
38
References
Abstract:
About the Publication OBJECTIVE: This study reviews the indications, technical
aspects, and experience with ex vivo and in situ split liver
transplantation. BACKGROUND: The shortage of cadaveric
donor livers is the most significant factor inhibiting further
application of liver transplantation for patients with end-stage
liver disease. Pediatric recipients, although they represent only
15% to 20% of the liver transplant registrants, suffer the
greatest from the scarcity of size-matched cadaveric organs.
Split liver transplantation provides an ideal means to expand the
donor pool for both children and adults. METHODS: This
review describes the evolution of split liver transplantation from

reduced liver transplantation and living-related liver
transplantation. The two types of split liver transplantation, ex
vivo and in situ, are compared and contrasted, including the
technique, selection of patients for each procedure, and the most
current results. RESULTS: Ex vivo splitting of the liver is
performed on the bench after removal from the cadaver. It is
usually divided into two grafts: segments 2 and 3 for children,
and segments 4 to 8 for adults. Since 1990, 349 ex vivo grafts
have been reported. Until recently, graft and patient survival
rates have been lower and postoperative complication rates
higher in ex vivo split grafts than in whole organ cadaveric
transplantation. Further, the use of ex vivo split grafts has been
relegated to the elective adult patient because of the high
incidence of graft dysfunction (right graft) when placed in an
emergent patient. Reasons for the poor function of ex vivo splits
except in elective patients have focused on graft damage due to
prolonged cold ischemia times and rewarming during the long
benching procedure. In situ liver splitting is accomplished in a
manner identical to the living donor procurement. This
technique for liver splitting results in the same graft types as in
the ex vivo technique. However, graft and patient survival rates
reported for in situ split livers have exceeded 85% and 90%,
respectively, with a lower incidence of postoperative
complications, including biliary and reoperation for bleeding.
These improved results have also been observed in the urgent
patient. CONCLUSION: Splitting of the cadaveric liver
expands the donor pool of organs and may eliminate the need
for living-related donation for children. Recent experience with
the ex vivo technique, if applied to elective patients, results in
patient and graft survival rates comparable to whole-organ
transplantation, although postoperative complication rates are
higher. In situ splitting provides two grafts of optimal quality
that can be applied to the entire spectrum of transplant
recipients: it is the method of choice for expanding the cadaver
liver donor pool.
Major Subjects:
Liver Transplantation / * Methods
Adolescence
Child
Hepatectomy / Methods
Human


Tissue Donors
Bookmark URL: /das/journal/view/N/10624503?source=CL,HS,MI

Annals of Surgery
313
Review
Split Liver Transplantation
Ronald W. Busuttil MD, PhD, FACS

John A. Goss MD
Department of Surgery, Division of Liver and Pancreas Transplantation, The Dumont-UCLA Transplant Center,
University of California, Los Angeles School of Medicine, Los Angeles, California
Supported in part by the Joanne Barr Memorial Foundation and the Francis P. Torino Foundation.
Correspondence: Ronald W. Busuttil, MD, PhD, The Dumont-UCLA Transplant Center, Room 77-120 CHS, Box 957054,
10833 Le Conte Avenue, Los Angeles, CA 90095-7054.
Accepted for publication August 7, 1998.
Objective
This study reviews the indications, technical aspects, and experience with ex vivo and in situ split
liver transplantation.
Background
The shortage of cadaveric donor livers is the most significant factor inhibiting further application
of liver transplantation for patients with end-stage liver disease. Pediatric recipients, although they
represent only 15% to 20% of the liver transplant registrants, suffer the greatest from the scarcity
of size-matched cadaveric organs. Split liver transplantation provides an ideal means to expand the
donor pool for both children and adults.
Methods
This review describes the evolution of split liver transplantation from reduced liver transplantation
and living-related liver transplantation. The two types of split liver transplantation, ex vivo and in
situ, are compared and contrasted, including the technique, selection of patients for each
http://home.mdconsult.com/das/article/body/jorg=j...source=CL,HS,MI&sp=10624503&sid=0/N/138989/1.html (1 of 15) [22/04/2000 0:54:57]

procedure, and the most current results.
Results
Ex vivo splitting of the liver is performed on the bench after removal from the cadaver. It is usually
divided into two grafts: segments 2 and 3 for children, and segments 4 to 8 for adults. Since 1990,
349 ex vivo grafts have been reported. Until recently, graft and patient survival rates have been
lower and postoperative complication rates higher in ex vivo split grafts than in whole organ
cadaveric transplantation. Further, the use of ex vivo split grafts has been relegated to the elective
adult patient because of the high incidence of graft dysfunction (right graft) when placed in an
emergent patient. Reasons for the poor function of ex vivo splits except in elective patients have
focused on graft damage due to prolonged cold ischemia times and rewarming during the long
benching procedure. In situ liver splitting is accomplished in a manner identical to the living donor
procurement. This technique for liver splitting results in the same graft types as in the ex vivo
technique. However, graft and patient survival rates reported for in situ split livers have exceeded
85% and 90%, respectively, with a lower incidence of postoperative complications, including
biliary and reoperation for bleeding. These improved results have also been observed in the urgent
patient.
Conclusion
Splitting of the cadaveric liver expands the donor pool of organs and may eliminate the need for
living-related donation for children. Recent experience with the ex vivo technique, if applied to
elective patients, results in patient and graft survival rates comparable to whole-organ
transplantation, although postoperative complication rates are higher. In situ splitting provides
two grafts of optimal quality that can be applied to the entire spectrum of transplant recipients: it
is the method of choice for expanding the cadaver liver donor pool.
Because donor liver shortage has been the rate-limiting step in the expansion of hepatic transplantation,
several innovative techniques have been developed to enlarge a relatively constant pool of organs. These
recently advanced procedures have focused on using a part of the liver allograft for transplantation.
Nowhere has this effort been felt more strongly than in pediatric liver transplantation. Although liver
replacement in children accounts for only 10% to 15% of all liver transplants performed, the number of
whole-organ cadaveric grafts size-matched for this population is inadequate. Because of this quantitative
disparity between donors and recipients, the pretransplant mortality rate has historically been reported to
be as high as 25% to 50% in children. [1]
To maximize donor organ use in children and small adults, three procedures have evolved from the
fundamental principle that a component of the liver with a suitable vascular pedicle, bile duct, and
venous drainage, along with sufficient functional hepatocyte mass, can sustain hepatic function in a
patient as well as a whole organ. Reduced liver transplantation (RLT) was the wellspring for this effort,
first
314

Figure 1. Segmental anatomy of liver. Segment 1: caudate liver. Segments 2 through 4: left lobe.
Segments 5 through 8: right lobe. Split liver grafts usually comprise segments 2 and 3 and segments 4
through 8.
reported by Bismuth and Houssin in 1984. [2] Because of its segmental anatomy, the liver can be
separated into several independent anatomic units, each of which can be transplanted (Fig. 1) . In RLT,
the liver graft can be tailored on the bench to a variety of functional lobes or segments. The most
commonly employed parts of a graft used in children are segments 2 and 3 (left lateral segment) and
segments 2, 3, and 4 (left lobe). Because of size discrepancy, the extended right lobe, segments 4 to 8, is
rarely used in pediatric patients. In RLT, when either a segment 2 and 3 graft or a segment 2 to 4 graft is
used, the remaining right lobe is discarded. Results with RLT in children have been comparable to those
reported when whole-organ cadaveric grafts are used. [3] [4] [5] Further, advocates of the procedure cite the
lower incidence of hepatic arterial complications due to the larger caliber of the adult hepatic artery. [6] [7]
However, although RLT increases the number of pediatric donor organs, this technique does not increase
the total number of organs available for transplantation; indeed, it actually disadvantages the adult
recipient pool, which has grown 12.1-fold during the past 8 years. [8]
Living-related donor (LRD) liver transplantation (LRD) is a natural extension of RLT. Use of a portion
of the liver from a living donor was attempted by Raia et al in 1988 [9] and first successfully carried out
by Strong et al in 1989. [10] During the past 10 years, approximately 1000 LRD transplants have been
performed throughout the world, achieving graft and patient survival rates equivalent or better than those
observed with cadaveric whole organs or RLT. The advantages of LRD include selection of an ideal
donor in whom liver graft function is immediate, and the ability to schedule the case electively, allowing
maximal preparation of the recipient. However, the potential advantages of increased histocompatibility
between donor and recipient, favoring a lower incidence of rejection, have not been realized. LRD
transplantation has also been recently applied to adults in selected cases. [11] [12] Despite the success
observed in pediatric LRD, there are still unresolved issues concerning the risks posed to donors, who are
usually parents. To date there have been at least two known donor deaths after LRD. [13]
Split liver transplantation (SLT) is the culmination of the stepwise progression from RLT and LRD. With
this technique, a whole adult cadaveric liver is divided into two functioning allografts. This procedure
not only overcomes the drawbacks of RLT and LRD but also increases the total number of donor organs.
In fact, full development of SLT may render RLT and LRD obsolete, except in unusual or emergent
circumstances. Further, SLT may be able to provide enough liver grafts for the entire pediatric recipient
pool. This review will focus on the current status of SLT and the role of this procedure in expanding the
liver donor pool.

Split Liver Transplantation

Pichlmayr et al [14] in 1988 reported the first clinical attempt at SLT, placing a right graft into a
63-year-old woman with primary biliary cirrhosis and the left graft into a small child with biliary atresia.
One year later, Bismuth et al [15] described two patients with fulminant hepatic failure, each receiving a
split graft. Although both patients recovered from coma with improvement of liver function, death
occurred from multiple organ failure on postoperative day 20 in one, and from diffuse cytomegalovirus
disease on postoperative day 45 in the other. The authors claimed that neither poor graft function nor
technical problems contributed to the patients' death. Broelsch et al [16] reported the first series of 30 SLT
procedures in 21 children and 5 adults. In this early experience, patient survival was inferior to reported
series [17] of whole-organ cadaveric liver transplants: only 67% of the children and 20% of the adults who
received split grafts survived. Technical problems were common, with a retransplant rate of 35% and a
biliary complication rate of 27%.
Despite some skepticism about the lasting role of SLT in decreasing the donor shortage, and because of
the less-than-satisfactory results obtained in early cases, several European centers, faced with increasing
numbers of deaths of waiting-list patients because of donor scarcity, cautiously pursued the SLT option.
A collective experience of 50 donor livers, providing 100 grafts during a 5-year period, was reported
from the European Split Liver Registry by de Ville. [18] In this series, graft and patient 6-month survival
rates were stratified according to elective or urgent status of the patient. In the former situation, survival
rates for graft and patient were 80% and 88.9% for children and 72.2% and 80% for adults, respectively.
In the urgent setting, graft and patient survival rates were 61.3% and 61.3% for children and 55.6% and
67.7% for adults. Twenty split grafts were lost because of complications
315
TABLE 1 -- Review of Ex Vivo Split Liver Series

Patient Graft
Survival Survival BC
Author Year Graft # % HR (% ) (% ) (% )
Emond 1990 18 28 67 50 27
Broelsch 1990 30 40 60 43 27
Shaw 1990 10 70 50 50 40
Otte 1990 4 50 50 0
Houssin 1993 16 56 75 69 25
Sloof 1995 15 73 67
Otte 1995 29 27 71 67 17
de Ville 1995 98 33 68 62 23
Bismuth 1995 30 7 93 90 23

Bismuth 1996 27 4 79 78 22
Broelsch 1996 19 58 63 58 16
Kalayoglu 1996 12 8 91 75 25
Rela 1998 41 12 90 88 14.6
HR, high-risk patient; BC, biliary tract complication.
related to the graft itself; technical complications included hepatic artery thrombosis (11.5%), portal vein
thrombosis (4%), and biliary complications (18.7%). These results were compared with the European
Liver Transplant Registry of conventional orthotopic liver transplantation performed during the same
time period and did not show a significant difference. In fact, although adult patients receiving a split
graft electively had twice the retransplant rate as those with a whole graft, patient survival was higher in
the former group (88.9% vs. 80.3%). Children who received a split graft electively actually had lower
graft loss and retransplantation rates.
The results described from the European Split Liver Registry stimulated renewed interest in SLT, as
evidenced by more recent series reported by Azoulay, [19] Otte, [20] Kalayoglu, [21] and Rela [22] and their
colleagues. In all of these later series, selection of high-risk patients appeared to be the dominant factor
influencing poor outcome. Further, this effect was more pronounced in recipients of right grafts. As
shown in Table 1 , the median percentage of high-risk patients receiving transplants in the series that
reported clinical condition was only 28%, and several transplant centers have made specific efforts to
avoid transplanting high-risk patients with split liver grafts. This has been emphasized for adult patients
by the King's College Group, [18] who state, "Our policy is not to split a liver for an ICU-based adult
patient, but to give a full-size graft." Based on the 349 cases reported and shown in Table 1 , it appears
that although patient selection undoubtedly plays an important role in graft and patient outcome, other
factors directly or indirectly related to the splitting must also be considered to account for some of the
unique complications associated with split grafts when they are used in more urgent patients, particularly
adult recipients. Further, if SLT is to be offered to the entire spectrum of patients in need of liver
replacement, further improvement in the technique is required.
The remainder of this review will address the two basic types of SLT that been performed clinically, ex
vivo and in situ, and will compare and contrast the results.
Ex Vivo Technique
In the ex vivo split liver technique, the whole organ is retrieved and preserved with University of
Wisconsin (UW) solution according to standard techniques of multiple organ procurement. Grafts are
prepared at the recipient transplant center in an ice bath of UW solution. Predissection cholangiography
and arteriography, to delineate the anatomy more precisely, are performed in some centers [19] [23] ; others
have not found this step necessary. [21] [22] In the latter, a metal cannula is used to probe the hepatic artery
and bile duct gently to facilitate detection of aberrant anatomy. Dissection of the portal triad is performed
to separate the branches of the hepatic artery, portal vein, and right and left hepatic ducts. Opinions vary

as to which liver half should retain the entire hepatic/celiac trunk and main portal vein. In all cases, the
common bile duct is retained with the right graft (Figs. 2 and 3) .
The rationale for determining which graft should receive the major vascular pedicle is explained by the
anatomy of the components of the porta hepatis. [24] In most cases, the left portal vein and right hepatic
artery should be sectioned because they are longer, thus facilitating the anastomoses to the recipient
vessels. Absence of a portal vein bifurcation is a contraindication to liver splitting. Because
microsurgical hepatic artery reconstruction is now commonly performed, sectioning the left hepatic
artery to remain with the left graft is preferred, as is done in in situ splitting. The left hepatic duct is
preferably sectioned because it is absent in only 2% to 9% of patients. When the left hepatic duct is
absent, it immediately branches to drain segment 4 and segments 2 and 3, allowing a favorable plane of
transection between
316
Figure 2. Schematic representation of preparation of each split graft. Top view

shows the celiac trunk, main portal vein, and common bile duct with a right
graft. The left graft has the sectioned left hepatic artery, left portal vein, and left
bile duct. Bottom view shows a right graft with the sectioned right hepatic
artery, right portal vein, and common bile duct. Extension grafts are sewn to the
right portal vein and right hepatic artery on the back table. The left graft has
main vessel trunks and left hepatic duct.
segment 4 and segments 2 and 3. A single right hepatic duct is less common and divides into two ducts in
30% to 50% of cases, thus making division of the right duct more problematic.
Interposition grafts consisting of allogeneic iliac artery, splenic or superior mesenteric artery, and the
iliac vein have been used as extensions to both right and left sides. The line of parenchymal transection
extends from the confluence of the middle and left hepatic veins to approximately 0.5 to 1 cm to the right
of the umbilical fissure up to the hilar plate. Dissection is usually done with mosquito fracture technique
or scalpel transection with ligation of the intrahepatic portal triad structures. The hepatic veins are
dissected, retaining the left hepatic vein with the left graft and the right and middle hepatic veins in
continuity with the vena cava with the right graft. The cut surfaces of the grafts are often sealed with
fibrin glue, reinforced with collagen or polyglactin 910 mesh, to reduce bleeding. [19]
Implantation of the right split into an adult is accomplished in the standard orthotopic manner with or
without venovenous bypass. Despite oversewing of the left hepatic vein orifice, the suprahepatic vena
caval cuff is not compromised in width and fits easily to the recipient cuff. As discussed above, if the
right hepatic artery and right portal vein are sectioned with the right lobe, interposition vascular grafts
must be used for anastomosis to a suitable source of inflow. Biliary reconstruction is usually
choledocholedocostomy with a T tube. Decompression with a T tube prevents bile leakage from the cut
surface of the liver. [25]
The left graft is transplanted into a child or small adult using a piggyback technique with retention of the
recipient vena cava. The left hepatic vein is anastomosed to the
317

Figure 3. Schematic representation of in situ liver splitting. The liver is split between segment 4 and
segments 2 and 3 after isolation of the left hepatic artery, left portal vein, and left hepatic vein in the
heart-beating cadaver. The left hepatic duct is divided in a sharp fashion before the liver is flushed
with University of Wisconsin solution. BD, biliary duct; HA, hepatic artery; IVC, inferior vena cava;
PV, portal vein.
suprahepatic vena cava of the patient. However, because of size discrepancy, various venoplasty
maneuvers often must be performed to provide a short and patulous anastomosis that is not susceptible to
kinking. These techniques have been described by Emond et al. [26] Portal vein reconstruction must be
individualized to the recipient's anatomy. Often the portal vein of children with biliary atresia is
phlebosclerotic and of diminutive diameter. In these cases, a direct end-to-end anastomosis is
contraindicated, and anastomosis to the confluence of the splenic and superior mesenteric veins is
required. In some cases, a vein graft is needed to provide a tension-free anastomosis, but routine use of
vein grafts should be proscribed. A review of the techniques of portal vein reconstruction that apply to
both LRD and SLT in children has been reported by Saad et al. [27] Hepatic artery reconstruction in ex
vivo split grafts has varied, depending on whether the common hepatic/celiac trunk is retained with the
graft. In this setting, anastomosis is either to the hepatic artery of the recipient or to the aorta with
infrarenal iliac conduit. If the left hepatic artery is retained with the left graft, then a microsurgical
reconstruction to the proper hepatic artery of the recipient is preferred, as described by Inomoto et al. [28]
The left graft biliary tract reconstruction is uniformly via a Roux-en-Y left hepaticojejunostomy, with the
caveat that in up to 25% of cases there are two or more separate ducts to segments 2 and 3.
Although the ex vivo split, as described above, is the most widely used method to transplant two patients
with one liver, there are drawbacks to this approach. Ex vivo splitting of the liver allograft on the bench is
a lengthy procedure and thus results in a long ischemic interval. This takes on more significance if a
second recipient operating room is not available, or if a split graft is transported to another center.
Extended cold ischemic times and the required dissection and manipulation of the graft compound the
deleterious effects of ischemia alone. Prolonged cold storage has also been associated with increased
cytokine release and MHC class II antigen expression, [29] [30] leading to an increased inflammatory
response on reperfusion. During the separation process into right and left grafts, some allograft
rewarming occurs; even if slight, it has been found to be associated with increased susceptibility to
hepatic ischemic/reperfusion injury. [31] The collective impact of prolonged ischemia and rewarming
during the ex vivo split results in graft injury, which predisposes to a high incidence of poor function
unless the organ is placed in a very favorable environment. In the nonurgent patient, unfavorable
operative and recipient factors can be minimized, thus decreasing the incidence of poor graft function, as
shown by Rela et al. [22] Thus, the ex vivo technique may be relegated to the elective case, particularly in
adult recipients.

In Situ Technique
A modification of the ex vivo splitting technique is in situ splitting, an extension of the techniques
established for LRD procurement that is practiced in the heart-beating cadaver donor. At UCLA, we first
attempted in situ SLT in 1992. Our initial experience was not favorable, with only one of four grafts
surviving. Causes of failure were primary nonfunction (1), multiple organ system failure (1), and graft
torsion (1). However, after establishing an LRD program and accruing an experience of 30 cases, we
resumed the in situ split liver program in 1996. In that same year, Rogers et al [32] reported an experience
with 14 split grafts that resulted in 6-month patient and graft survival rates of 92.8% and 85.7%,
respectively. Further, these authors described a lower rate of biliary complications, intraabdominal
hemorrhage, and nonfunction of the right graft compared with other series using the ex vivo split
techniques (see Table 1) .
As with the ex vivo technique, only hemodynamically stable cadaveric multiorgan donors are considered
for in situ splitting. Standard surgical facilities for a multiorgan procurement are used, and no special
equipment is needed. Donor hospitals and other procurement teams are notified as soon as possible of the
decision to split the liver in situ, and the decision to proceed is based on unanimous agreement by the
organ teams. The procedure adds 1 to 1.5 hours to a standard multiorgan procedure.
The initial step in the procedure is to obtain control of the supraceliac and infrarenal aorta and inferior
mesenteric vein to permit rapid multiorgan cooling in the event of donor instability. If on inspection the
vascular anatomy and appearance of the liver are suitable, segments 2 and 3 of the liver are mobilized as
in an living donor procurement. [33] The left hepatic artery is mobilized throughout its length. The left
portal vein is dissected with ligation of branches to the caudate lobe (segment 1) and to segment 4.
Extrahepatic
318
Figure 4. Schematic representation of the implantation of the left lateral segment liver allograft. The
patient's vena cava is left intact. The donor left hepatic vein is sutured to the confluence of the
recipient middle and left hepatic veins after the right hepatic vein is oversewn. The donor left portal
vein is sutured to the recipient portal vein, and the donor left hepatic artery is anastomosed to the
recipient common hepatic artery in a microvascular manner without extension grafts. The biliary tract
is reconstructed using a Roux-en-Y hepaticojejunostomy. The donor and recipient falciform
ligaments are reapproximated to prevent torsion of the liver allograft. A, aorta; CA, celiac axis, IVC,
inferior vena cava; PHA, proper hepatic artery; PV, portal vein; SA, splenic artery.
isolation of the left hepatic vein is accomplished with care to ensure that the middle hepatic venous
drainage of segments 4, 5, and 8 is not obstructed. Transection of the parenchyma is performed in a line
0.5 to 1 cm to the right of the umbilical fissure, as described for ex vivo splitting, and accomplished using
electrocautery. The left hilar plate and bile ducts are divided sharply with scissors so as not to
devascularize the duct. The middle hepatic vein is retained with the right graft. On completion of the
dissection, two liver grafts are procured, each with a preserved vascular pedicle and venous drainage in a
bloodless field (Fig. 3) .
The procurement proceeds in a standard fashion with perfusion of the abdominal organs with UW
solution. After perfusion, the vascular attachments between each graft are divided, leaving the

hepatic/celiac arterial trunk, main portal vein, and common bile duct with the right graft. The right graft
is prepared on the bench in the standard manner, and the stumps of the left hepatic artery, portal vein, and
bile duct are oversewn individually. The left graft usually needs no additional tailoring before
implantation.
Concern for the viability of segments 1 and 4 after liver splitting has been voiced by several groups. [19]
[22] [31] Opinions regarding the need to resect segments 1 and 4 from the right graft because of
devascularization have ranged from always [22] to never. [25] It is clear that when dissecting the left
hepatic artery, branches to segment 4 are encountered and often ligated. These vessels appear to be more
easily identified and preserved during the in situ method. Further, because the in situ procedure is
performed in the heart-beating cadaveric donor, the perfusion of segment 4 can be readily assessed.
However, in both ex vivo and in situ splitting, segment 4 hypoperfusion is a potential pitfall and may
require treatment with segmentectomy. If there is any question about the viability of segment 1, it should
be removed at the completion of graft placement.
Implantation of the right graft procured by the in situ technique is accomplished identically to a standard
orthotopic transplant in which the main hepatic artery, portal vein, and bile ducts are retained by the right
graft. The left liver allograft is transplanted in a fashion similar to that used in LRD, [33] including
microvascular reconstruction of the hepatic artery (Fig. 4) .
Size Matching in Split Liver Transplantation

In SLT, graft size must be considered both in terms of fit into the recipient and providing sufficient
functional hepatic mass. As has been learned from the RLT experience, donor/recipient weight ratio is
useful as a guide in determining which portion of the liver is suitable for a child. In general, if the ratio
exceeds 5 to 10:1, the upper limit of graft volume that can be accommodated by a small child is a
segment 2 and 3 graft. Lesser ratios permit the use of a left or right lobe graft. [34] From a practical point
of view, in most children a left lateral segment is a good size fit. However, in 10% to 20% of cases, the
segment 2 and 3 graft is oversized after reperfusion, thus precluding abdominal closure without a
prosthesis or with skin alone. [35] Despite the guidelines offered by the donor/recipient weight ratio, there
is no substitution for an experienced surgeon's visual assessment of the recipient and the graft to be
implanted.
Several guidelines [19] [35] have been proposed for the liver size required to sustain hepatic function in a
given patient. Volumetric computed tomographic scanning of the lobe of the liver to be transplanted can
provide an accurate assessment of the
319
TABLE 2 -- UCLA Patient and Graft Survival, In Situ Split

Number Percent
Total patient survival 59 /65 90.7

Right recipient survival 29 /34 85.2

Left recipient survival 30 /31 96.7
Total graft survival 58 /72 80.5
Right graft survival 31 /36 86.0
Left graft survival 27 /36 75.0
liver volume. A ratio of graft volume to standard liver volume of >30% is essential for adequate hepatic
function. [36] Although formulaic calculation of sufficient graft volume may be possible for LRD
transplantation, it is impractical when splitting a cadaveric donor, when time and resources may not be
available. In this latter setting, a graft with a volume approximately 1% to 1.5% of the body weight of the
recipient is large enough to provide excellent hepatic reserve. Segments 2 and 3 of an adult will support a
child weighing 6 to 20 kg. For an adult recipient, a left lobe graft from a donor of the same weight will
provide approximately 30% to 35% of standard liver volume, which can sustain good liver function.
Again, surgical experience is invaluable for making these crucial assessments at the time of liver harvest.
Results of In Situ Splits

There have been only a few published reports of in situ SLT [25] [32] [37] because the procedure has been
performed only during the past 3 years. Results from these series have shown improvement over the ex
vivo experience, with increased patient and graft survival rates and a lower incidence of technical
complications associated with the hepatic artery, biliary anastomosis, and postoperative hemorrhage. [25]
The patients transplanted with an ex vivo split, as reported by the King's College Group, [22] have been the
only series that matched the in situ split results; however, of these 41 patients transplanted with ex vivo
splits, all cases were elective except for 4 children and 1 adult (12%) who were classified as emergent.
This contrasts with the in situ experience, in which there were substantially more emergent cases. In the
UCLA series, [25] 38.5% of patients were United Network for Organ Sharing (UNOS) status I (intensive
care unit) and 19.2% UNOS status 2 (hospitalized), whereas in the Hamburg series, [32] 21.4% were
UNOS status 1 and 14.2% UNOS status 2.
An expanded series of in situ split liver grafts performed since July 1, 1996, at UCLA has confirmed
these improved results, even in the face of a more urgent group of patients. As of March 1, 1998, we
have transplanted 65 patients with a total of 72 in situ split grafts. Of these, 27 (41.5%) were UNOS
status 1, 16 (24.6%) UNOS status 2, and 22 (33.8%) elective or UNOS status 3. Patient and graft survival
rates are shown in Table 2 ; patient and graft survival rates for
TABLE 3 -- Patient Survival by UNOS Status, UCLA In Situ Split

Status Number Percent
1 23 /27 85.2
2 14 /18 87.5
3 22 /22 100

children are 96.7% and 75%, respectively. Survival stratified to UNOS status is shown in Table 3 .
Complications are shown in Table 4 .
The patient survival rate (90.7%) obtained with in situ splitting is comparable to the best results (90%)
reported for ex vivo splitting. However, in the former urgent cases represented 41.5% versus 12% in the
latter. Further, the incidence of biliary complications, which seem to be inherent in the ex vivo split
technique, occurring in 14.6% [22] to 40%, [38] is dramatically reduced to <3% with the in situ split
operation. Similarly, reoperation as a result of bleeding occurs in up to 20% of patients with ex vivo
splitting and in <3% for in situ splits. To date we have encountered no necrosis of segment 4, a frequent
complication of ex vivo splitting.
Despite these excellent results, we are disturbed by an 8.3% incidence of primary graft nonfunction,
which occurred equally between left and right grafts in six patients. Four of the cases of primary
nonfunction occurred early in our experience in patients who received grafts from two donor livers that,
in retrospect, were not ideal livers for splitting. Further, half of these occurred in UNOS status 1 patients,
whose unfavorable recipient factors affected graft function, as is seen in whole-organ cadaveric grafts.
Although there are advantages to in situ splitting of the liver, as demonstrated above, the technique
places an additional burden on the donor hospital and other procurement teams by increasing the
operation time by up to 2 hours. If most procurements are regional, a preemptive educational program
that presents the benefits of in situ splitting will defuse the objections to the procedure and allow its
broader application. This has been realized in our own region, where a proposal has been made to our
two organ-procurement agencies to consider all hemodynamically stable multiorgan
TABLE 4 -- Complications of 72 In Situ Split Transplants
Complication Number Percent
Hepatic artery thrombosis 2 /72 2.8
Portal vein thrombosis 1 /72 1.4
Biliary leak or stenosis 2 /72 2.8
Reoperation for bleeding 2 /72 2.8
Primary nonfunction * 6 /72 8.3
* 50% UNOS status 1; 4 of 6 grafts harvested from same 2 donors
320
TABLE 5 -- Criteria For In Situ Split Donors

Hemodynamically stable, heart-beating, cadaveric multiorgan donor
Age 10-35?
Minimal to moderate vasopressors ( i.e., dopamine < 15 mcg/kg/min)
Hospitalization <5 days
Liver function test (excluding prothrombin time) abnormality no greater than 3 normal
Serum Na < 160 mg/dl
donor candidates between ages 10 and 30 for in situ splitting. Full realization of this policy in the greater
Los Angeles area should provide an additional 65 grafts per year.
Split Liver Policy: Impact on Waiting Time

A consensus-based policy to split all suitable donor livers would have a substantial impact on the
recipient waiting time. In the Paul Brousse experience, [19] the number of transplantable grafts was
increased by 28%. Improved use and transplant efficiency has also been reported by the Hamburg group
[32] and UCLA. [25] If these improvements in donor resource use can be extended to the national liver
donor pool, the effects would be dramatic. The number of livers suitable for splitting ranges from 15% to
25% of the available donors. In 1996, 4058 livers were transplanted in the United States. [8] Assuming
that 20% were split, an additional 811 grafts would be available for transplantation. Maximal use of this
modality would provide enough grafts to supply the entire pediatric waiting list in the United States. This
effect on pediatric transplantation has been demonstrated at UCLA. Since implementation of a policy to
split every suitable liver, we have decreased the waiting list for children younger than 1 year from 128
days to 24 days; for children older than 1 year, the waiting time has decreased from 192 days to 30 days.
Further, use of split livers will benefit the small blood group O recipient, the perennial laggard on the
transplant waiting list.
Application of Split Liver Transplantation

Based on the data analyzed in this review, we propose that all hemodynamically stable multiorgan donors
who fulfill the criteria in Table 5 should be considered as candidates for split liver donation. We believe
that the in situ technique is superior because it consistently provides two allografts of optimal quality for
both adult and pediatric recipients. Further, the grafts obtained with the in situ method have proven
successful even in urgent patients, and have demonstrated a lower incidence of complications ( e.g.,
biliary and take-back for bleeding). In addition, grafts procured using the in situ technique are likely to
be more suitable for sharing among adults because of reduced ischemic injury. However, there are some
logistical drawbacks with the in situ split, particularly in relation to donor hospital inconvenience and

reticence of other organ-procurement teams to acquiesce to the additional time taken for the procedure.
In these circumstances, standard organ procurement followed by ex vivo splitting should be performed by
a team with experience and expertise in major liver resections, RLT, and LRD. Transplantation with two
grafts obtained after ex vivo splitting will provide excellent results in the elective patient, as previously
shown. [19] [21] [22]
Split liver transplantation is now recognized as a practical and meaningful procedure that can truly
expand the donor pool. Based on the cumulative experience with this technique, it should now become a
routine part of the experienced liver transplant center's armamentarium.
References
1. Emond JC, Whitington PF, Thistlethwaite JR, et al. Transplantation of two patients with one liver: analysis of a
preliminary experience with split-liver grafting. Ann Surg 1990; 212:14-22. abstract
2. BismuthH, Houssin D. Reduced-size orthotopic liver graft for liver transplantation in children. Surgery 1984;
95:367-370. abstract
3. Emond JC, Whitington PF, Thistlethwaite RJ, et al. Reduced-size orthotopic liver transplantation: use in the management
of children with chronic liver disease. Transplantation 1989; 10:867-872.
4. Broelsch
CE, Emond JC, Thistlethwaite RJ. Liver transplantation with reduced-size donor organs. Transplantation 1989;
45:519-523.
5. Jurim O, Csete M, Gelabert H, et al. Reduced-size grafts: the solution for hepatic artery thrombosis after pediatric liver
transplantation. J Pediatric Surg 1995; 30:853-855.
6. Houssin D, Soubrane O, Boillet O, et al. Orthotopic liver transplantation with a reduced-size graft: an ideal compromise
in pediatrics. Surgery 1992; 111;532-542. abstract
7. Langas AN, Marujo WC, Inagaki M, et al. The results of reduced-size liver transplantation including split livers in
patients with end-stage liver disease. Transplantation 1992; 53:387-391. abstract
8. United Network for Organ Sharing (http://wwwew3.att.net/UNOS) UNOS Web site 1998.
9. Raia S, Nery JR, Mies S. Liver transplantation from live donors. Lancet 1988; 2:497. citation
10. Strong
RW, Lynch SV, Ong TN, et al. Successful liver transplantation from a living donor to her son. N Engl J Med
1990; 322:1505-1507. citation
11. HashikuraV, Makuuchi M, Kawasaki S, et al. Successful living-related partial liver transplantation to an adult patient.
Lancet 1994; 343:1233-1234. citation
12. KawasakiS, Makuuchi M, Matsunami H, et al. Living-related liver transplantation in adults. Ann Surg 1998;
227:269-274. full text

13. Broelsch CE. Personal communication, 1997.
14. Pichlmayr R, Ringe B, Gubernatis G, et al. Transplantation einer spenderbeber auf zwei empfanger
(splitting-transplantation): eine neue methode in der weiterentwicklung der lebersegment transplantation. Langenbecks
Arch Chir 1988; 373:127-130. abstract
15. BismuthH, Marino M, Castaing D. Emergency orthotopic liver transplantation in two patients using one donor. Br J
Surg 1989; 76:722-724. abstract
16. BroelschCE, Emond JC, Whitington PF, et al. Application of reduced-size liver transplants as split grafts, auxiliary
orthotopic grafts and living related segmental transplants. Ann Surg 1990; 214:368-377. abstract
17. Busuttil
RW, Shaked A, Millis M, et al. One thousand liver transplants: the lessons learned. Ann Surg 1994;
219:490-499. abstract
321
18. de Ville de Goyet J. Split liver transplantation in Europe 1988-1993. Transplantation 1995; 59:1371-1376. abstract
19. Azoulay
D, Astarcioglu I, Bismuth H, et al. Split liver transplantation: the Paul Brousse policy. Ann Surg 1996;
224:737-748. full text
20. Otte
JB. Is it right to develop living related liver transplantation? Do reduced and split livers not suffice to cover the
needs? Transpl Int 1995; 8:69-73. citation
21. KalayogluM, D'Alessandro AM, Knechtle JS, et al. Preliminary experience with split liver transplantation. J Am Coll
Surg 1996; 182:381-387. abstract
22. Rela
M, Voregas V, Miniesan P, et al. Split liver transplantation: King's College Hospital experience. Ann Surg 1998;
227:282-288. full text
23. Otte
JB, de Ville de Goyet J, Alberti D, et al. The concept and technique of the split liver in clinical transplantation.
Surgery 1990; 107:605-612. abstract
24. Rat
P, Paris P, Friedman S, et al. Split liver orthotopic liver transplantation: how to divide the portal pedicle. Surgery
1991; 112:522-526.
25. GossJA, Yersiz H, Shackleton CR, et al. In situ splitting of the cadaveric liver for transplantation. Transplantation
1997; 64:871-877. abstract
26. EmondJC, Heffron TG, Whitington PF, et al. Reconstruction of the hepatic vein in reduced-size hepatic transplantation.
Surg Gynecol Obstet 1997; 176:11-17.
27. Saad
S, Tanaka K, Inomata Y, et al. Portal vein reconstruction in pediatric liver transplantation from living donors. Ann
Surg 1998; 227:275-281. full text

28. Inomoto T, Nishizawa F, Sasaki H, et al. Experience with 120 microsurgical reconstructions of hepatic artery in living
related liver transplantation. Surgery 1996; 119:20-26. abstract
29. Howard TK, Klintmalm GB, Corer JB, et al. The influence of preservation injury or rejection in hepatic transplant
recipients. Transplantation 1990; 49:103-107. abstract
30. Takada M, Nadeau KC, Shaw GD, et al. The cytokine-adhesion molecule cascade in ischemia/reperfusion injury of the
rat kidney: inhibition by a soluble P-selection ligand. J. Clin Invest 1997; 99:2682-2690.
31. Hertl
M, Chartraud PB, West DD, et al. The effects of hepatic preservation at 0 degrees C compared to 5 degrees C:
influence of antiproteases and periodic flushing. Cryobiology 1994; 31:434-438. abstract
32. Rogers
X, Malago M, Gawad K, et al. In situ splitting of cadaveric livers: the ultimate expansion of the donor pool. Ann
Surg 1996; 224:331-341. full text
33. TanakaK, Uemoto S, Tkunega Y, et al. Surgical techniques and innovations in living related transplantation. Ann Surg
1993; 217:82-87. abstract
34. de
Ville de Goyet J, Otte JB. Cut-down and split liver transplantation. In Busuttil RW, Klintmalm GB, eds.
Transplantation of the liver. Philadelphia: WB Saunders; 1996:481-496.
35. Superina
RA, Strasberg SM, Grieg PD, Langer B. Early experience with reduced-size liver transplants. J Pediatr Surg
1990; 25:1157. abstract
36. Lo MC, Fan ST, Chan JKF, et al. Minimum graft volume for successful adult to adult living donor liver transplantation
for fulminant hepatic failure. Transplantation 1996; 62:696-698. abstract
37. Rogers X, Malago M, Habib N, et al. In situ splitting of the liver in heart-beating cadaveric organ donor for
transplantation in two recipients. Transplantation 1995; 59:1081-1083. abstract
38. Langnas AN, Marujo WC, Inagaki M, et al. The results of reduced-size liver transplantation, including split livers in
patients with end-stage liver disease. Transplantation 1992; 53:387-391. abstract

Bookmark URL: /das/journal/view/N/10624503?ja=138989&PAGE=1.html&ANCHOR=top&source=CL,HS,MI

April 22, 2000
Conde Petra
SELECTION OF PATIENTS FOR LIVER

TRANSPLANTATION IN 1997 AND BEYOND
Additional Article
Consult's full-text literature database. Volume 1 Number 2 August 1997
Full Text
Frontmatter 247
THE BENEFITS OF LIVER

TRANSPLANTATION Liver Transplantation
INDICATIONS FOR LIVER
TRANSPLANTATION
SELECTION OF PATIENTS FOR LIVER
General Considerations
TRANSPLANTATION IN 1997 AND
Fulminant Hepatic Failure BEYOND
Chronic Autoimmune Liver
Diseases
The "Classic" Metabolic Diseases

Affecting the Liver
Eric M. Yoshida MD
Chronic Viral Hepatitis
John R. Lake MD
Primary Hepatocellular Cancer
British Columbia Transplant Society and the Department of
CONTROVERSIAL ISSUES Medicine, the University of British Columbia, Vancouver, BC,
Age and Alcohol Canada (EMY) University of California, San Francisco (JRL)
Indolent Cancers
CONTRAINDICATIONS TO LIVER
TRANSPLANTATION
Within the past three decades, liver transplantation has
evolved from its "experimental" infancy, with a
Absolute Contraindications demoralizingly high perioperative mortality, [79] to the
"cyclosporine era" of the early 1980s, [81] by which time, the
Relative Contraindications
technical aspects of surgery had been overcome and
THE FUTURE cyclosporine dramatically improved patient and graft
survival. In the 1990s, tacrolimus (FK506) produced further
References improvement in rates of graft rejection [23] [89] and graft
survival. As the turn of the century approaches, newer
immunosuppressive agents, such as mycophenolate mofetil
(RS 61443), [62] will undoubtedly have an impact on liver
transplantation. Mirroring the favorable evolutionary
history of liver transplantation has been the increasing
number of transplant centers and the increasing number of
transplants in both the United States and Canada. For the
year ending 1993, in the United States, 94 centers performed
3411 transplants (mean 36.3; median 21) versus 58 centers
that transplanted 1713 livers in 1988. [5] North of the border,
in 1991, seven Canadian centers performed 177 liver
transplants. [54]
Liver transplantation is now accepted as the standard
surgical-medical treatment for end-stage liver disease as
well as replacement therapy for certain inborn errors of
metabolism. Indications for liver transplantation continue
to change with new developments in medical-surgical
science and this current review differs in certain areas from
the last comprehensive review. [39] What remains constant,
however, is the
248
chronic shortage of available donor organs, leading to long

waiting lists of ill patients--a problem that will only intensify
as the list of contraindicated diseases diminishes. The
transplant physician or surgeon therefore is in a unique and
at times difficult position: He or she must act in the interests
of both the individual referred for transplantation and the
interests of the other patients on the list awaiting
transplantation in the face of a scarce and "rationed"
resource. This is underscored by the fact that, in the United
States, 657 patients (7.8% of the waiting list) died awaiting
transplantation in 1994. [88]

John R. Lake, MD
Liver Transplant, Box 0780
505 Parnassus Avenue, Room M-896
San Francisco, CA 94143
THE BENEFITS OF LIVER

TRANSPLANTATION
Before 1980, when patient survival at 1 year post-transplant was

30% or less, [80] liver transplantation was viewed as a therapy of

"last resort." Currently, however, reported 1-year survival rates,
depending on the center, can be in excess of 85%. [32] Moreover,
it has been clearly demonstrated that liver transplant recipients
obtain a benefit in their quality of life compared with the
pretransplant period of chronic disease. [10] [33] [85] In a recent
survey from Duke University, [33] none of the respondents had a
Karnofsky score less than 80% (normal activity with effort,
demonstrating some signs and symptoms of disease). In fact,
85% of the respondents had Karnofsky scores of 90% to 100%.
In both Canada [3] and the United States, [33] surveys have
demonstrated that approximately 60% of liver transplant
recipients are employed. Not surprisingly, in both countries,
liver transplant recipients who were unemployed tended to be
older and on some form of social assistance before
transplantation, whereas those who were employed prior to
transplantation tended to resume working after transplantation.
Clearly, liver transplantation, in the 1990s, offers patients an
excellent chance of survival, a good quality of life
post-transplant, and, in many cases, an opportunity to resume
full and productive lives. It is then appropriate to refer patients
for liver transplantation when chronic liver disease is affecting
the quality of life (e.g., intolerable pruritis, fatigue) or patients
have early decompensated portal hypertension, rather than
waiting until the clinical situation becomes desperate.
INDICATIONS FOR LIVER

TRANSPLANTATION
The list of individual diseases for which transplantation is

indicated or beneficial is lengthy and includes both primary
hepatocellular or biliary disease and systemic metabolic
diseases (familial and sporadic), in which the liver expresses the
genetic defect but is not clinically affected (e.g., severe familial
hypercholesterolemia, [7] hereditary oxalosis, [49] [92]
Crigler-Najjar syndrome, [34] [75] etc.). Transplantation in the
latter
249
case constitutes a macroscopic form of "gene replacement"

therapy. A detailed discussion of transplantation and each form
of liver disease is beyond the scope of this article; more on the
specifics of certain diseases and transplantation is covered by

the authors of other articles.

In general, most of the patients referred for liver transplantation
suffer from chronic liver disease resulting in cirrhosis and portal
hypertension. Such patients should be referred when a major
complication of liver disease occurs (e.g., ascites, variceal
bleed) or there is clear evidence of decompensation (e.g.,
hepatic encephalopathy, hypoalbuminemia, coagulopathy)
because the long-term prognosis of such patients is poor: 1- and
6-year survivals of 60% and 21% for decompensated patients,
and 54% 6-year survival for compensated patients in a review
of 1155 patients. [14] Although variceal bleeding, hepatic
encephalopathy, and coagulopathy are obvious indications for
transplant referral, spontaneous bacterial peritonitis (SBP)
deserves discussion. SBP usually indicates significantly
decompensated liver disease (80% in one study were Child's
class C [70] ) and is associated with a significant in-hospital
mortality, mainly a result of the underlying liver disease rather
than SBP specifically. [70] SBP also has a high relapse rate--69%
1-year probability in one analysis. [87] Despite the success
reported in SBP prophylaxis studies involving cotrimoxazole [76]
and quinilone antibiotics, [27] [67] the rapid emergence of
resistant strains in fecal cultures has been reported. [20] Clearly,
an episode of SBP should be considered an indication for
transplant list activation.
Because the waiting time for transplantation for an
out-of-hospital patient may be as long as 1 to 2 years at some
centers, effective medical therapy may stabilize the patient with
cirrhosis after an acute episode of decompensation and reduce
the risk of relapse. In that regard, the increasingly use of
transjugular intrahepatic portosystemic stent shunts (TIPS)
deserves comment. Although TIPS has proved itself effective in
the management of variceal bleeding [68] and refractory ascites,
[56] like the surgically created shunt, there is a risk of postshunt
encephalopathy of approximately 25% [77] and occlusion rates

of 10% to 40% annually. [13] Moreover, TIPS can complicate
transplant surgery because of placement into the vena cava or
portal vein resulting in intimal damage at the site of anastamosis
or accidental perforation of the bile duct. [52] If TIPS placement
is considered, therefore, it should be considered as a bridge to
liver transplantation and TIPS placement should be performed
at an experienced center.
Fulminant Hepatic Failure
Acute, or the more dramatic and commonly used term,

fulminant, liver failure is defined by the onset of

encephalopathy within 8 weeks of disease onset. [42] Shorter (<2
weeks) and longer (<26 weeks) durations are termed hyperacute
and subacute liver failure, respectively. Patients are generally
healthy before the catastrophic turn of events and may be of
250
any age group. Drugs such as acetaminophen, in both overdoses

and therapeutic "misadventure" in heavy alcohol drinkers, [95]
antituberculous medications, [53] and halothane [65] are
classically implicated agents but any medication or toxin,
including herbal remedies, [94] can be potentially hepatotoxic.
Any of the viral hepatitic agents, from A to E, [24] [29] [47] [71] [72]
can present with fulminant failure. In many instances, however,
the exact causative agent is never elucidated.
Raised intracranial pressure (ICP) from cerebral edema has long
been known to complicate fulminant failure [91] and is a leading
cause of mortality. Admission to the intensive care unit and
intubation for airway protection should be considered in every
patient with stage III encephalopathy. ICP monitoring is
indicated for patients who have slipped into hepatic coma (stage
IV encephalopathy), with intravenous mannitol given as needed
to lower an elevated ICP.
Patients receiving transplants for fulminant failure have long
been reported to have a worse outcome than patients receiving
transplants for other reasons. The poorer outlook, in part, may
reflect the urgent use of suboptimal grafts (older, steatotic, split,
or size-reduced grafts) but also reflects the presence of
advanced encephalopathy. [9] A recent series reported a 55.6%
1-year survival for patients transplanted in grade 3 coma
(uncoordinated response to pain only) versus 83% for those in
coma graded 2 or better. [9]
Despite the high mortality of patients with fulminant failure, not
all patients require a transplant. Various prognostic features
have been suggested to predict which patients are likely to die
without a transplant. Bismuth and colleagues [6] suggest that a
serum factor V less than 20% of normal in patients younger
than 30 years and less than 30% of normal in those older than
30 years predicts a 90% likelihood of mortality with medical
therapy alone. The King's College Liver Unit [57] has published
widely cited laboratory and clinical cutoff points that predict a
survival less than 20% without transplantation for both
acetaminophen and nonacetaminophen-associated failure. We

would recommend that those guidelines be used with the overall

clinical trend in mind; strict adherence to the guidelines for
deciding when to activate may result in unfortunate outcomes.
Chronic Autoimmune Liver Diseases
The autoimmune liver diseases, which can be divided into

cholestatic (primary biliary cirrhosis [PBC], primary sclerosing
cholangitis [PSC]) and hepatocellular (autoimmune hepatitis)
diseases, are common indications for transplantation. PBC,
considered to be the "prototypical" transplantable disease, has a
natural history that may span more than a decade in mild cases
before terminating in end-stage cirrhosis. The natural history of
PBC can be predicted using several models; the most
well-known being the Mayo Model, [16] which incorporates the
serum bilirubin, albumin, prothrombin time, age, and presence
of edema. In
251
clinical practice, the Mayo Model can be cumbersome to use;

serum bilirubin, however, is a practical marker to follow and
patients should be referred for transplantation when the serum
bilirubin is greater than 100 mumo/L (6 mg/dL).
Transplantation for PBC has been clearly shown to favorably
interrupt the natural history of PBC, with post-transplant
survival greatly exceeding that predicted by the Mayo Model.
[45] Although the much rarer antimitochondrial
antibody-negative PBC or autoimmune cholangitis [51] [86] has

not been well described in the transplant literature, undoubtedly
the Mayo Model prognostications apply and survival
post-transplant is probably similar to that for PBC. PSC, which,
like PBC, may have a natural history spanning years, is strongly
associated with inflammatory bowel disease and, in 9% to 15%
of cases, may be complicated by cholangiocarcinoma. [43] The
Mayo Model has also been applied to PSC [93] and
transplantation, likewise, has been clearly demonstrated to
favorably interrupt the disease's natural history, compared with
both predicted outcomes and actual postsurgical biliary
drainage outcome. [25]
Patients with inflammatory bowel disease are predisposed to
colonic dysplasia [11] and colonic carcinoma has been reported
to be a common post-transplant cause of death. [55]
Cholangiocarcinoma is also well known to have an unfavorable
impact on post-transplant survival. [1] Unfortunately, no study to
date has been able to reliably identify patients who have an
incidental cholangiocarcinoma and the development of that

complication increases with duration of disease--from 13% at 5
years to 31% at 10 years in one series. [25] For those reasons, it
has been proposed to transplant patients with PSC earlier and to
diligently screen for colonic dysplasia pre- and post-transplant.
Another feature common to chronic cholestatic liver diseases is
metabolic bone disease. The post-transplant course of
cholestatic bone disease is accelerated in the short term but may
ameliorate, with remineralization, in the long term. [31] [64]
Finally, autoimmune hepatitis is also readily amenable to
transplantation. Although it is obvious that a patient with
"burned-out" autoimmune hepatitis and decompensated
cirrhosis requires transplantation, young patients dependent on
high doses of corticosteroids may also be considered for earlier
The "Classic" Metabolic Diseases Affecting the Liver
Of the metabolic disorders that affect the liver, alpha1

-antitrypsin deficiency, Wilson's disease, and hemochromatosis
are the three that commonly come to mind in the differential
diagnosis of chronic liver disease. The underlying systemic
disorder is "cured" with transplant in alpha1 -antitrypsin
deficiency [90] and Wilson's disease [73] because the metabolic
defect resides within the liver. Whether or not the liver
contributes to the enhanced intestinal absorption of iron in
hemochromatosis and would correct with liver replacement is
unresolved, with arguments in
252
the medical literature both pro [37] and con. [2] [18] No medical
therapy exists for alpha1 -antitrysin deficiency, whereas copper
chelation therapy [96] with d-penicillamine, trientine, and zinc for
Wilson's disease and regular phlebotomy [21] for
hemochromatosis should avert the need for transplantation if
instituted early. Medical therapy will not reverse end-stage
cirrhosis and both Wilson's disease and hemochromatosis have
extrahepatic manifestations, the post-transplant reversibility of
which remains controversial. The post-transplant survival of
patients with hemochromatosis has been suggested to be
significantly worse than the general liver transplant recipient
population. A 5-year analysis of Medicare transplant recipients
revealed a 1- and five-year post-transplant survival of 54% and
43% for hemochromatosis versus 79% and 69% for the general
post-transplant population. [35] The exact reasons for that are not
clear, although the systemic manifestations of iron overload,
(e.g., cardiac involvement) are generally suspected. Of interest,
despite the increased incidence of hepatoma in
hemochromatosis, a recent survey suggests that hepatoma alone
does not account for the worse post-transplant outcome. [38]
The post-transplant reversibility of the neurologic
manifestations of Wilson's disease is an issue that is
controversial. Copper chelation therapy has been demonstrated
to ameliorate the neurologic symptoms of Wilson's disease [84]
but, presumably, those patients were treated earlier in the
disease's course. Likewise, neurologic symptoms have been
reported to improve [73] and resolve [63] in some patients
post-transplant. Others have reported failure of a liver allograft
to resolve the neurologic manifestations [30] and postulate a
component of irreversible basal ganglial damage. Of interest, a
case was reported recently in which a patient with liver disease
successfully treated with copper chelation underwent
transplantation because of chelation-refractory neurologic
symptoms. [46] The patient experienced a dramatic neurologic
improvement before dying of an intra-abdominal bleed in the
early post-transplant period. Neurologic examination revealed
degenerative changes in the basal ganglia despite symptomatic
improvement, suggesting that functional improvement may not
reflect the degree of structural damage and that, in general,
post-transplant reversibility depends on degree of disease
severity, which may not be predictable.
Before leaving this section, it is prudent to mention that
Wilson's disease may present for the first time in a young
person as fulminant hepatitic failure. The features that suggest
Wilson's disease include hemolysis, mildly elevated serum
transaminases, a serum alkaline phosphatase level that is
unusually normal or below normal, grossly elevated serum
bilirubin, and severe coagulopathy. [83] In our experience, such
patients die without urgent transplantation.
Transplantation is commonly performed for end-stage liver

disease secondary to chronic hepatitis C (HCV) and hepatitis B
(HBV). Both
253
viral diseases, however, have significant allograft reinfection

rates in the untreated transplant recipient. HCV infection has

long been recognized to recur, with a recent study [26]
establishing that 5 years post-transplant, only 12% of recipients
are spared histologic evidence of reinfection. HCV allograft
reinfection also leads to the same wide spectrum of disease seen
in the nontransplant liver--from mild chronic hepatitis to
cirrhosis with clinical portal hypertension. [26] Despite the
current lack of effective antiviral therapy for allograft
reinfection, the 5-year survival of HCV seropositive allograft
recipients is not significantly different from that of allograft
recipients transplanted for other diseases. [26]
Until recently, transplantation for patients with HBV was
viewed with skepticism because of the high likelihood of
allograft reinfection and rapid development of severe liver
disease leading to poor graft survival. [15] [72] A recent European
study [72] reported a 3-year actuarial HBV reinfection rate
exceeding 80% for those who are HBV DNA positive in serum
pretransplant and approaching 60% for those who were HBV
DNA negative before transplantation. Fortunately, the
long-term administration of high-dose hepatitis B immune
globulin (HBIG) has dramatically attenuated the rate allograft
reinfection, [40] [72] making liver transplantation for HBV-related
disease feasible. The success of the antinucleoside agent,
lamivudine, in suppressing HBV replication [17] in
nontransplanted patients suggests that even better results may
be forthcoming. Whether antinucleoside agents can substitute
for HBIG post-transplant or are best used in combination
remains to be determined, given that the emergence of
lamivudine escape mutants following the post-transplant use of
lamivudine as monotherapy recently was reported. [36]
The finding of hepatocellular carcinoma (HCC) in patients with

viral hepatitis or alcoholic liver disease during the pretransplant
assessment is not uncommon. Although patients with advanced
malignancy, indicated by TMN stage III or IV disease (more
than one lobe, macroscopic invasion of major vessel in unilobe
disease, with or without local or disseminated metastases [4] or
large tumor or size alone (e.g., 8 cm) have a uniformly poor
survival post-transplant survival, [50] [60] [74] with 5-year survival
on the order of 10%, those with HCC found incidentally in the
explanted liver and those with known but small tumors have a
better outcome. The policy of many transplant centers regarding
tumor size is to accept patients with tumors up to 5 cm in

diameter and, if multifocal, no more than 3 cm. Using such

criteria, the group from Milan [48] recently reported excellent
survival in a group of unresectable patients, with an overall
actuarial survival of 75% and, in those staged correctly, 85% at
4 years. The 4-year actuarial recurrence-free survival in the
latter group was 92%. It was unclear as to whether pretransplant
chemoembolization led to any significant improvement in
outcome in that study.
254
CONTROVERSIAL ISSUES
Controversial issues in liver transplantation can be divided into
medical controversies, regarding which the debate is largely
confined to the liver transplant community, and "ethical"
controversies, for which arguments are not limited to health
care professionals but involve society as a whole. At the root of
these controversies is the issue of rational use of a limited
resource. Many of the medical controversies of yesterday have
been resolved--e.g., hepatitis B is no longer an issue (in centers
that have access to HBIG). Small hepatocellular cancers are
now known to have a good post-transplant survival, and even
certain complications of portal hypertension, such as the
hepatopulmonary syndrome, formerly thought to be a
contraindication, have been demonstrated to resolve with
transplantation. [41] Nevertheless, certain issues are likely to
remain controversial and are discussed briefly.
Age and Alcohol
Previously, advanced age was considered to be a

contraindication to liver transplantation and many centers had
arbitrary age restrictions. Although many in both the medical
and lay community are of the belief that transplantation should
be limited to younger patients, the collective transplant
experience does not support that belief. Several centers have
reported no difference in post-transplant outcome in older
versus younger patients [22] [61] and have concluded that age
should not be a contraindication. Perhaps, Sir Roy Calne said it
best when he advised physicians to consider the "biological age
rather than the chronological age." [12]
It has been clearly demonstrated that patients transplanted for
alcoholic liver disease have a post-transplant survival not
different from that for most other diseases. [8] [44] [59] [82] The

issue of liver transplantation in these patients, however, remains

a fiercely debated issue within North American society, as
evidenced by media commentary and letters to newspapers
whenever former alcoholic baseball players and Hollywood
actors receive transplants. Two central themes recur regarding
the issue of transplantation for alcoholic cirrhosis. The first
concerns the "ethical" issue of transplantation of a limited
resource for a "behaviorally" acquired disease and the second
concerns the issue of resuming alcohol use. The first issue is
based heavily on philosophical, moral, and emotional
viewpoints rather than medical evidence. It is a societal issue,
not a medical one. Post-transplant alcohol use, on the other
hand, has been well studied and it appears that most
transplanted patients do not return to drinking and even fewer
return to serious drinking. [8] [28] [44] [59] [82] Abstinence
pretransplant appears to be a predictive factor with regard to
post-transplant alcohol use and most centers require a
documented period of abstinence before transplantation, often
in association with some form of pretransplant alcohol
rehabilitation. Six months, combined
255
with a good social support system, appears to be the minimum

period of abstinence that predicts a favorable outcome. [28]
Indolent Cancers
Extensive primary fibrolamellar cancers and neuroendocrine

tumors with liver metastases are two malignancies that
occasionally are referred for transplant assessment. Because the
cancers are slow growing and indolent, it may be suggested that
patients with such tumors should receive transplants. Because
fibrolamellar hepatomas are far less common than the usual
hepatomas, the experience reported in the transplant literature is
limited. It is difficult to make any conclusive inferences from
the data because small numbers of fibrolamellar cancers are
often reported as part of the much larger group of hepatocellular
carcinomas. Nonetheless, it appears that the long-term
post-transplant prognosis of fibrolamellars, like the usual
hepatomas, may be dependent upon staging. [50] [66] Likewise,
the data regarding the transplantation of patients with hepatic,
but no extrahepatic, metastases is noninferential. Although
Pichlmayr and colleagues [60] have reported excellent 3-year
survival, in excess of 80% ( n = 11), other investigators,
likewise reporting on small groups, have been less optimistic.

One group estimated a 5-year actuarial survival of 57% based

on 11 patients [69] and another reported survival of three of eight
patients followed 6 to 55 months. [19] Hemangioendothelioma is
another rare neoplasm for which long-term survival has been
reported. [58] [60] There are undoubtedly other rare indolent
malignancies for which the issue of transplantation remains
unresolved or the answer is unknown.
CONTRAINDICATIONS TO LIVER
TRANSPLANTATION
With the growing list of indications for transplantation, it must
be kept in mind that a liver allograft is not a general panacea for
all patients with liver disease. Because donor organs constitute a
scarce and rationed resource, it is the responsibility of the
transplant team to ensure that candidates are selected carefully.
At the present time, there is probably no role for "palliative"
liver transplants for diseases likely to recur and result in graft
loss and death after only a few years. Likewise, patients with
significant concurrent medical disease, unlikely to be
ameliorated by a new liver, for which the prognosis is
unfavorable, should not be candidates. In general,
contraindications can be considered "absolute" or "relative."
Absolute Contraindications
The list of absolute contraindications for transplantation reflect

the policies of the British Columbia Transplant Society and the
University
256
of California-San Francisco and include both medical and

psychosocial conditions. Absolute oncologic medical
contraindications include extrahepatic malignancies (with,
perhaps, a few exceptions, as previously discussed), large
hepatocellular carcinomas, and cholangiocarcinomas.
Cholangiocarcinomas deserve special mention because they
were once thought to be ideal malignancies for liver
transplantation. The long-term post-transplant results proved to
be disappointing, with the 2- to three-year survival around 30%.
Some patients with cholangiocarcinoma, however, may enjoy a
good long-term, recurrence-free survival.99 It has been
suggested that the Klatskin variety of cholangiocarcinomas may
represent a subgroup that has a more favorable outcome.97
Despite that, because of the issue of limited resource utilization,

our centers and most others do not consider transplantation for

that malignancy.
The other absolute medical contraindications to transplantation
include acute ongoing infection, which may be only a
temporary contraindication pending adequate antimicrobial
therapy, human immunodeficiency virus (HIV) seropositivity,
multisystem organ failure (MSOF), and advanced
cardiopulmonary disease. It may be argued that HIV infection is
a fatal, yet chronic disease, especially in patients with favorable
CD4 counts. We note, however, that post-transplant outcome in
patients with acquired immunodeficiency syndrome (AIDS) is
poor100 and the that combined effect of an antirejection protocol
is likely to increase the likelihood of opportunistic infection and
malignancy. Moreover, the Toronto Hospital group101 has
demonstrated that, in the case of coinfection with HIV and
HBV, death from AIDS-related causes predominates. Although
the exclusion of patients with advanced cardiopulmonary
disease would appear obvious and the same medical
contraindications to any other major surgery apply, we note that
7% (3 of 41 patients) of deaths in a recent European report were
secondary to congestive heart failure.102 MSOF is likewise an
absolute contraindication, with a dismal short-term survival that
is inversely related to the number of organ systems that have
failed--from 50% survival with one-organ failure to nil survival
with four-organ failure.103
Although some may view psychosocial contraindications as
"soft," there are, in fact, compelling reasons to refuse certain
patients on those grounds. To be listed for a transplant, the
patient must be agreeable and give informed consent. That may
appear to be an obvious situation but, in reality, transplantation
is a very emotional issue involving not just the patient, but the
patient's family. It has to be clear that the patient, not just the
patient's family, wants a transplant, no matter how vocal the
family may be. The issue of compliance, for medications,
diagnostic and therapeutic tests, and clinic visits, is also of
importance. Documented, habitual noncompliance is an
absolute contraindication for transplantation because
pretransplant noncompliance has been demonstrated to predict
post-transplant graft loss.104 Active alcohol or drug abuse and
unwillingness to abstain is similarly an absolute
contraindication at most centers.
257
There are many conditions that, of themselves, do not disqualify

patients from transplant candidacy yet may negatively affect
long-term post-transplant survival, increase the operative
mortality, or make the technical aspects of surgery difficult.
Many such relative contraindications are center specific,
depending on the expertise of the surgeons and the resources
and expertise of the intensive care and transplant units. What
may not be feasible at a small center, may be possible at a larger
center with more resources and experience. Examples of such
considerations include previous abdominal surgery,
malnutrition, and significant diabetes mellitus with target organ
damage, as well as psychosocial issues such as lack of social
support.
THE FUTURE
As the list of transplantable medical conditions expands, the
number of patients seeking organs also grows, translating to
longer waiting times and increased morbidity and mortality for
those on the wait list. Although greater organ donor awareness
may increase the number of available organs, liver allografts are
likely to remain a limited and rationed resource. Recently,
tremendous resources and attention have been directed toward
xenotransplantation. In theory, the availability of
xenotransplants would take pressure off the waiting lists.
Xenotransplantation, at the current time, however, is far from
becoming a realistic generalizable alternative and introduces a
whole new set of immunologic, infectious, and ethical
problems. Perhaps, in the global sense, a partial answer to our
problems lies not in the arena of transplantation, but in the areas
of public health and primary care. Measures to reduce alcohol
dependence and alcoholic liver disease, intravenous drug abuse,
and viral hepatitis, and to introduce hepatitis B vaccination
programs and reduce vertical transmission of that disease,
would, in the decades to come, reduce the waiting lists and
make donor organs available to those with unpreventable liver
disease.
References
1. Abu-Elmagd KM, Selby R, Iwatsuki S, et al: Cholangiocarcinoma and

sclerosing cholangitis: Clinical characteristics and effect on survival after
liver transplantation. Transplant Proc 25:1124-1125, 1993
2. Adams PC, Ghent CN, Grant DR, et al: Transplantation of a donor liver

with hemochromatosis: Evidence against an inherited intrahepatic defect.

Gut 32:1082-1103, 1991
3. AdamsPC, Grant DR, Wall WJ: Employment after liver transplantation.

Hepatology 21:140-144, 1995
4. American Joint Committee on Cancer: Manual for Staging of Cancer, ed

3. Philadelphia, Lippincott, 1987, pp 87-92
5. BelleSH, Beringer C, Detre KM: Liver transplantation in the United

States: Results
258
from the National Pitt-UNOS Liver Transplant Registry. In Terasaki PI,

Cecka JM (eds): Clinical Transplants. Los Angeles, UCLA Tissue Typing
Laboratory, 1994, pp 19-35
6. Bernuau J, Samuel D, Durand F, et al: Criteria for emergency liver

transplantation in patients with acute viral hepatitis and factor V below
50% of normal: A prospective study. Hepatology 14:49A, 1991
7. Bilheimer DW, Goldstein JL, Grundy SM, et al: Liver transplantation to

provide low-density lipoprotein receptors and lower plasma cholesterol in
a child with homozygous familial hypercholesterolemia. N Engl J Med
311:1658-1664, 1984
8. BirdDL, O'Grady JG, Harvey FA, et al: Liver transplantation in patients

with alcoholic cirrhosis: Selection criteria and rates of survival and
relapse. British Medical Journal 301:15-17, 1990
9. Bismuth H, Samuel D, Castaing D, et al: Orthotopic liver transplantation

in fulminant and subfulminant hepatitis: The Paul Drousse experience.
Ann Surg 222:109-119, 1995
10. Bonsel GJ, Essink-Bot ML, Klompmaker IJ, et al: Assessment of the
quality of life before and following liver transplantation. Transplantation
53:796-800, 1992
11. BrentnallTA, Haggitt RC, Rabinovitch PS, et al: Risk and natural
history of colonic neoplasm in patients with primary sclerosing cholangitis
and ulcerative colitis. Gastroenterology 110:331-338, 1996
12. Calne
R: Contraindications to liver transplantation. Hepatology
20:3S-4S, 1994
13. Conn HO: Transjugular intrahepatic portal-systemic shunts: The state of

the art. Hepatology 17:148-158, 1993
14. D'Amico G, Morabito A, Pagilaro L, et al: Survival and prognostic

indicators in compensated and decompensated cirrhosis. Dig Dis Sci
31:468-475, 1986

15. Demetris AJ, Todo S, Van Thiel DH, et al: Evolution of hepatitis B
liver disease after hepatic replacement: Practical and theoretical
considerations. Am J Pathol 137:667-676, 1990
16. Dickson ER, Grambsch PM, Fleminng TR, et al: Prognosis in primary
biliary cirrhosis: Model for decision making. Hepatology 10:1-7, 1989
17. DienstagJL, Perrillo RP, Shiff ER, et al: A preliminary trial of

lamivudine for chronic hepatitis B infection. N Engl J Med
333:1657-1661, 1995
18. DietzO, Vogel W, Braunsperger B, et al: Liver transplantation in

idiopathic hemochromotosis. Transplant Proc 22:1512-1513, 1990
19. Dousset B, Houssin D, Soubrane O, et al: Metastatic endocrine tumors:

Is there a place for liver transplantation? Liver Transplantation and
Surgery 2:111-117, 1995
20. Dupeyron C, Mangeney N, Sedrati L, et al: Rapid emergence of

quinolone resistance in cirrhotic patients treated with norfloxacin to
prevent spontaneous bacterial peritonitis. Antimicrob Agents Chemother
38:340-344, 1994
21. Edwards
CQ, Kushner JR: Screening for hemochromotosis. N Engl J
Med 328:1616-1620, 1993
22. EmreS, Mor E, Schwartz ME, et al: Liver transplantation in patients

beyond age 60. Transplant Proc 25:1075-1076, 1993
23. European FK506 Multicentre Study Group: Randomized trial

comparing tacrolimus (FK506) and cyclosporin in prevention of liver
allograft rejection. Lancet 344:423-428, 1994
24. Farci
P, Alter HJ, Shimoda A, et al: Hepatitis C virus-associated
fulminant hepatic failure. N Engl J Med 335:631-634, 1996
25. FargesO, Malassagne B, Sebagh M, et al: Primary sclerosing

cholangitis: Liver transplantation or biliary surgery? Surgery 117:146-155,
1995
26. Gane EJ, Portmann BC, Naoumov NV, et al: Longterm outcome of
hepatitis C infection after liver transplantation. N Engl J Med
334:815-820, 1996
27. Gines P, Rimola A, Planas R, et al: Norfloxacin prevents spontaneous

bacterial peritonitis recurrence in cirrhosis: Results of a double-blind
placebo-controlled trial. Hepatology 12:716-724, 1990
28. GishRG, Lee AH, Keefe EB, et al: Liver transplantation for patients
with alcoholism and end stage liver disease. Am J Gastroenterol
88:1337-1342, 1993
29. Gordon FD, Anastopoulos H, Khettry V, et al: Hepatitis C infection: A

rare case of fulminant hepatic failure. Am J Gastroenterol 90:117-120,
1995
259
30. Guarino M, Stracciari A, D'Alessandro R, et al: No neurological

improvement after liver transplantation for Wilson's disease. Acta Neurol
Scand 92:405-408, 1995
31. Hay JE, Dickson ER, Wiesner RH, et al: Long-term effect of orthotopic
liver transplantation on the osteopenia of primary biliary cirrhosis
[abstract]. Hepatology 12:838, 1990
32. HeathResources and Services Administration, Bureau of Health

Resources Development: 1991 Report of Center-Specific Graft and Patient
Survival Rates. Rockville, MD, US Dept of Health and Human Services,
1992, pp 8-10
33. HuntCM, Tart JS, Dowdy E, et al: Effect of orthotopic liver

transplantation on employment and health status. Liver Transplantation
and Surgery 2:148-153, 1996
34. Kaufman SS, Wood RP, Shaw BJ, et al: Orthotopic liver transplantation
for type I Crigler-Najjar syndrome. Hepatology 6:1259-1262, 1986
35. Kilpe
VE, Krakauer H, Wren RE: An analysis of liver transplant
experience from 37 transplant centers as reported to Medicare.
36. Kneteman NM, Bain VG, Shapiro AMJ, et al: Prevention of hepatitis B
recurrence after liver transplantation with lamivudine [abstract]. Clin
Invest Med 19:S101, 1996
37. Kosinkinas J, Portmann B, Lombard M, et al: Persistent iron overload 4

years after inadvertent transplantation of a haemochromatotic in a liver
patient with primary biliary cirrhosis. J Hepatol 16:351-354, 1992
38. Kowdley KV, Hassanein T, Kaur S, et al: Primary liver cancer and
survival in patients undergoing liver transplantation for hemochromatosis.
Liver Transplantation and Surgery 1:237-241, 1995
39. LakeJR: Changing indications for liver transplantation. Gastroenterol

Clin North Am 22:213-229, 1993
40. LakeJR: Should liver transplantation be performed for patients with

chronic hepatitis B? Yes! Liver Transplantation and Surgery 1:260-265,
1995
41. Lange
PA, Stoller JK: The hepatopulmonary syndrome. Ann Intern
Med 122:521-529, 1995
42. Lee WM: Acute liver failure. N Engl J Med 329:1862-1872, 1993
43. Lee
YM, Kaaplan M: Primary sclerosing cholangitis. N Engl J Med
332:924-933, 1994
44. Lucey MR, Merion AM, Henley KS, et al: Selection for and outcome of
liver transplantation in alcoholic liver disease. Gastroenterology
102:1836-1841, 1992
45. Markus BH, Grambsch PM, Dickson ER, et al: Efficiency of liver
transplantation in patients with primary biliary cirrhosis. N Engl J Med
320:1709-1713, 1989
46. Mason AL, Marsh W, Alpers DH: Intractable neurological Wilson's

disease treated with orthotopic liver transplantation. Dig Dis Sci
38:1746-1750, 1993
47. Masuda CT, Shaw BW Jr, Zetterman RK, et al: Fulminant hepatic
failure with massive necrosis as a result of hepatitis A infection. J Clin
Gastroenterol 17:158-162, 1993
48. Mazzaferro V, Regalia E, Doci R, et al: Liver transplantation for the

treatment of small hepatocellular carcinomas in patients with cirrhosis. N
Engl J Med 334:693-699, 1996
49. McDonald JC, Landreneau MD, Rohr MS, et al: Reversal by liver
transplantation of the complications of primary hyperoxaluria as well as
the metabolic defect. N Engl J Med 321:1100-1103, 1989
50. McPeake JR, O'Grady JG, Zaman S, et al: Liver transplantation for
primary hepatocellular cancer: Tumor size and number determine
outcome. J Hepatol 18:226-234, 1993
51. Michieletti
P, Wanless IR, Katz A, et al: Antimitochondrial-negative
primary biliary cirrhosis: A distinct syndrome of autoimmune cholangitis.
Gut 35:260-265, 1994
52. Millis
JM, Martin P, Gomes A, et al: Transjugular intrahepatic
porto-systemic shunts: Impact on liver transplantation. Liver
Transplantation and Surgery 1:229-233, 1995
53. Mitchell I, Wendon J, Fitt S, et al: Antituberculous therapy and acute

liver failure. Lancet 345:555-556, 1995
54. Mullen MA, Kohut N, Sam M, et al: Access to adult liver

transplantation in Canada: A survey and ethical analysis. Can Med Assoc J
154:337-342, 1996
55. Narumi S, Roberts JP, Emond J, et al: Liver transplantation for

sclerosing cholangitis. Hepatology 22:451-457, 1995
56. OchsA, Rossle M, Haag K, et al: The transjugular intrahepatic

portosystemic stent-shunt procedure for refractory ascites. N Engl J Med
332:1192-1197, 1995

57. O'Grady JG, Alexander GJM, Hayllar KM, et al: Early indications of
prognosis in fulminant hepatic failure. Gastroenterology 97:439-445, 1989
260
58. Olthoff KM, Millis JM, Rosove MH, et al: Is liver transplantation
justified for treatment of hepatic malignancies? Arch Surg 125:1261-1266,
1990
59. OrsorioRA, Ascher NL, Avery M, et al: Predicting recidivism after

orthotopic liver transplantation for alcoholic liver disease. Hepatology
20:105-110, 1994
60. Pichlmayer R, Weimann A, Oldhafer KJ, et al: Role of liver

transplantation in the treatment of unresected liver cancer. World J Surg
19:807-813, 1995
61. PirschJD, D'Alessandro AM, Voss BJ, et al: Orthotopic liver

transplantation in patients 60 years of age and older. Transplantation
51:431-433, 1991
62. Platz
KP, Sollinger HW, Hullet DA, et al: RS-61443, a new potent
immunosuppressive agent. Transplantation 51:428-432, 1991
63. PolsonRJ, Rolles RY, Calne RY, et al: Reversal of severe neurological
manifestations of Wilson's disease following orthotopic liver
transplantation. Q J Med 64:685-691, 1987
64. Porayko MK, Wiesner RH, Hay JE, et al: Bone disease in liver
transplantation recipients L-incidence, timing, and risk factors. Transplant
Proc 23:1462-1465, 1991
65. Ray DC, Drummond GB: Halothane hepatitis. Br J Anaesth 67:84-99,

1991
66. RingeB, Wittekind C, Weimann A, et al: Results of hepatic resection

and transplantation for fibrolamellar carcinoma. Surg Gynecol Obstet
175:299-305, 1992
67. Rolachon A, Cordier L, Bacq Y, et al: Ciprofloxacin and long-term

prevention of spontaneous bacterial peritonitis: Results of a prospective
controlled trial. Hepatology 22:1171-1174, 1995
68. RossleM, Haag K, Ochs A, et al: The transjugular intrahepatic

portosystemic stent-shunt procedure for variceal bleeding. N Engl J Med
330:165-171, 1994
69. Routley D, Ramage JK, McPeake J, et al: Orthotopic liver

transplantation in the treatment of metastatic neuroendocrine tumors of the
liver. Liver Transplantation and Surgery 2:118-121, 1995
70. Runyon BA, McHutchinson JG, Antillon MR, et al: Short course vs
long-course antibiotic treatment of spontaneous bacterial peritonitis: A
randomized controlled study of 100 patients. Gastroenterology
100:1737-1742, 1991
71. Sallie
R, Chiyende J, Tan KC, et al: Fulminant hepatic failure resulting
from co-existent Wilson's disease and hepatitis E. Gut 35:849-853, 1994
72. SamuelD, Miller R, Alexander G, et al: Liver transplantation in

European patients with the hepatitis B surface antigen. N Engl J Med
329:1842-1847, 1993
73. SchilskyML, Scheinberg IH, Sternlieb I: Liver transplantation for

Wilson's disease: Indications and outcome. Hepatology 19:583-537, 1994
74. SelbyR, Kadry Z, Carr B, et al: Liver transplantation for hepatocellular

cancer. World J Surg 19:53-58, 1995
75. ShevellMI, Bernard B, Adelson JW, et al: Crigler-Najjar syndrome

type I: Treatment by home phototherapy followed by home orthotopic
hepatic transplantation. J Pediatr 110:429-431, 1987
76. SinghN, Gazowski T, Yu VL, et al: Trimethoprim-sulphamethoxazole

for prevention of spontaneous bacterial peritonitis in cirrhosis: A
randomized trial. Ann Intern Med 122:595-598, 1995
77. Somberg KA, Riegler JL, Doherty M, et al: Hepatic encephalopathy

following transjugular intrahepatic portosystemic shunts (TIPS): Incidence
and risk factors. Hepatology 16:122A, 1992
78. Sanchez-Urdazpal L, Czaja AJ, van Hoek B, et al: Prognostic features

and role of liver transplantation in severe corticosteroid-treated
autoimmune chronic active hepatitis. Hepatology 15:215-221, 1992
79. Starzl
TE: History of liver and other splanchnic organ transplantation.
In Busuttil RW, Klintmalm GB (eds): Transplantation of the Liver.
Philadelphia, WB Saunders Company, 1996, pp 3-22
80. Starzl
TE, Demetris AJ, Van Thiel D: Liver transplantation. N Engl J
Med 329:1014-1022, 1092-1099, 1989
81. Starzl
TE, Klintmalm GBG, Porter KA, et al: Liver transplantation with
the use of cyclosporin A and prednisone. N Engl J Med 305:266-269, 1981
82. Starzl
TE, Van Thiel D, Tzakis AG, et al: Orthotopic liver
transplantation for alcoholic cirrhosis. JAMA 260:2542, 1988
261

83. Sternlieb
I: Wilson's disease: Indications for liver transplants.
Hepatology 4:15S-17S, 1984
84. Stremmel W, Meyernose KW, Niederau C, et al: Wilson disease:

Clinical presentation, treatment and survival. Ann Intern Med
115:720-726, 1991
85. Tarter
RE, Switala J, Arria A, et al: Quality of life before and after
orthotopic liver transplantation. Arch Intern Med 151:1521-1526, 1991
86. Taylor SL, Dean PJ, Riely CA: Primary autoimmune cholangitis: An
alternative to antimitochondrial-negative primary biliary cirrhosis. Am J
Surg Pathol 18:91-99, 1994
87. TitoL, Rimola A, Gines P, et al: Recurrence of spontaneous bacterial

peritonitis in cirrhosis: Frequency and predictive factors. Hepatology
8:27-31, 1988
88. UNOS:
Reported deaths on the OPTN waiting list, 1988 to 1994.
UNOS Update 12:25, 1996
89. The
U.S. Multicenter FK506 Study Group: A comparison of tacrolimus
(FK506) and cyclosporine for immunosuppression in liver transplantation.
N Engl J Med 331:1110-1115, 1994
90. Vennarecci G, Gunson BK, Ismail T, et al: Transplantation for

end-stage liver disease related to alpha1 -antitrypsin. Transplantation
61:1488-1495, 1996
91. WareAJ, D'Agostinho A, Combes B: Cerebral edema: A major

complication of massive hepatic necrosis. Gastroenterology 61:877-884,
1971
92. WattsRWE, Rolles K, Morgan SH, et al: Successful treatment of

primary hyperoxaluria Type I by combined hepatic and renal
transplantation. Lancet 2:474-475, 1987
93. Wiesner RH, Porayko MK, Dickson ER, et al: Selection and timing of
liver transplantation in primary biliary cirrhosis and primary sclerosing
cholangitis. Hepatology 16:1290-1299, 1992
94. YoshidaEM, McLean CA, Cheng ES, et al: Chinese herbal medicine,
fulminant hepatitis and liver transplantation. Am J Gastroenterol, in press
95. Zimmerman HJ, Maddrey WC: Acetoaminophen (paracetamol)

hepatotoxicity with regular intake of alcohol: Analysis of instances of
therapeutic misadventure. Hepatology 22:767-773, 1995
96. ZuckerS, Gollan J: Copper metabolism and Wilson's disease An "ion"

recent advances. In Thomson ABR, Shaffer E (eds): Modern Concepts in
Gastroenterology, vol 3. New York, Plenum Medical Book Company,
1992, pp 223-262



Volume 1 Number 2 August 1997
247
SELECTION OF PATIENTS FOR LIVER TRANSPLANTATION IN 1997

AND BEYOND
Eric M. Yoshida MD
John R. Lake MD
British Columbia Transplant Society and the Department of Medicine, the University of British
Columbia, Vancouver, BC, Canada (EMY) University of California, San Francisco (JRL)
Within the past three decades, liver transplantation has evolved from its "experimental" infancy,
with a demoralizingly high perioperative mortality, [79] to the "cyclosporine era" of the early 1980s,
[81] by which time, the technical aspects of surgery had been overcome and cyclosporine
dramatically improved patient and graft survival. In the 1990s, tacrolimus (FK506) produced
further improvement in rates of graft rejection [23] [89] and graft survival. As the turn of the century
approaches, newer immunosuppressive agents, such as mycophenolate mofetil (RS 61443), [62] will
undoubtedly have an impact on liver transplantation. Mirroring the favorable evolutionary history
of liver transplantation has been the increasing number of transplant centers and the increasing
number of transplants in both the United States and Canada. For the year ending 1993, in the
United States, 94 centers performed 3411 transplants (mean 36.3; median 21) versus 58 centers
that transplanted 1713 livers in 1988. [5] North of the border, in 1991, seven Canadian centers
performed 177 liver transplants. [54]
Liver transplantation is now accepted as the standard surgical-medical treatment for end-stage
liver disease as well as replacement therapy for certain inborn errors of metabolism. Indications
for liver transplantation continue to change with new developments in medical-surgical science and
this current review differs in certain areas from the last comprehensive review. [39] What remains
constant, however, is the
248

chronic shortage of available donor organs, leading to long waiting lists of ill patients--a problem
that will only intensify as the list of contraindicated diseases diminishes. The transplant physician
or surgeon therefore is in a unique and at times difficult position: He or she must act in the
interests of both the individual referred for transplantation and the interests of the other patients
on the list awaiting transplantation in the face of a scarce and "rationed" resource. This is
underscored by the fact that, in the United States, 657 patients (7.8% of the waiting list) died
awaiting transplantation in 1994. [88]

John R. Lake, MD
Liver Transplant, Box 0780
505 Parnassus Avenue, Room M-896
San Francisco, CA 94143
THE BENEFITS OF LIVER TRANSPLANTATION

Before 1980, when patient survival at 1 year post-transplant was 30% or less, [80] liver transplantation
was viewed as a therapy of "last resort." Currently, however, reported 1-year survival rates, depending on
the center, can be in excess of 85%. [32] Moreover, it has been clearly demonstrated that liver transplant
recipients obtain a benefit in their quality of life compared with the pretransplant period of chronic
disease. [10] [33] [85] In a recent survey from Duke University, [33] none of the respondents had a Karnofsky
score less than 80% (normal activity with effort, demonstrating some signs and symptoms of disease). In
fact, 85% of the respondents had Karnofsky scores of 90% to 100%. In both Canada [3] and the United
States, [33] surveys have demonstrated that approximately 60% of liver transplant recipients are
employed. Not surprisingly, in both countries, liver transplant recipients who were unemployed tended to
be older and on some form of social assistance before transplantation, whereas those who were employed
prior to transplantation tended to resume working after transplantation. Clearly, liver transplantation, in
the 1990s, offers patients an excellent chance of survival, a good quality of life post-transplant, and, in
many cases, an opportunity to resume full and productive lives. It is then appropriate to refer patients for
liver transplantation when chronic liver disease is affecting the quality of life (e.g., intolerable pruritis,
fatigue) or patients have early decompensated portal hypertension, rather than waiting until the clinical
situation becomes desperate.
INDICATIONS FOR LIVER TRANSPLANTATION

The list of individual diseases for which transplantation is indicated or beneficial is lengthy and includes
both primary hepatocellular or biliary disease and systemic metabolic diseases (familial and sporadic), in
which the liver expresses the genetic defect but is not clinically affected (e.g., severe familial
hypercholesterolemia, [7] hereditary oxalosis, [49] [92] Crigler-Najjar syndrome, [34] [75] etc.).
Transplantation in the latter
249

case constitutes a macroscopic form of "gene replacement" therapy. A detailed discussion of

transplantation and each form of liver disease is beyond the scope of this article; more on the specifics of
certain diseases and transplantation is covered by the authors of other articles.
In general, most of the patients referred for liver transplantation suffer from chronic liver disease
resulting in cirrhosis and portal hypertension. Such patients should be referred when a major
complication of liver disease occurs (e.g., ascites, variceal bleed) or there is clear evidence of
decompensation (e.g., hepatic encephalopathy, hypoalbuminemia, coagulopathy) because the long-term
prognosis of such patients is poor: 1- and 6-year survivals of 60% and 21% for decompensated patients,
and 54% 6-year survival for compensated patients in a review of 1155 patients. [14] Although variceal
bleeding, hepatic encephalopathy, and coagulopathy are obvious indications for transplant referral,
spontaneous bacterial peritonitis (SBP) deserves discussion. SBP usually indicates significantly
decompensated liver disease (80% in one study were Child's class C [70] ) and is associated with a
significant in-hospital mortality, mainly a result of the underlying liver disease rather than SBP
specifically. [70] SBP also has a high relapse rate--69% 1-year probability in one analysis. [87] Despite the
success reported in SBP prophylaxis studies involving cotrimoxazole [76] and quinilone antibiotics, [27] [67]
the rapid emergence of resistant strains in fecal cultures has been reported. [20] Clearly, an episode of SBP
should be considered an indication for transplant list activation.
Because the waiting time for transplantation for an out-of-hospital patient may be as long as 1 to 2 years
at some centers, effective medical therapy may stabilize the patient with cirrhosis after an acute episode
of decompensation and reduce the risk of relapse. In that regard, the increasingly use of transjugular
intrahepatic portosystemic stent shunts (TIPS) deserves comment. Although TIPS has proved itself
effective in the management of variceal bleeding [68] and refractory ascites, [56] like the surgically created
shunt, there is a risk of postshunt encephalopathy of approximately 25% [77] and occlusion rates of 10%
to 40% annually. [13] Moreover, TIPS can complicate transplant surgery because of placement into the
vena cava or portal vein resulting in intimal damage at the site of anastamosis or accidental perforation of
the bile duct. [52] If TIPS placement is considered, therefore, it should be considered as a bridge to liver
transplantation and TIPS placement should be performed at an experienced center.
Fulminant Hepatic Failure
Acute, or the more dramatic and commonly used term, fulminant, liver failure is defined by the onset of
encephalopathy within 8 weeks of disease onset. [42] Shorter (<2 weeks) and longer (<26 weeks)
durations are termed hyperacute and subacute liver failure, respectively. Patients are generally healthy
before the catastrophic turn of events and may be of
250
any age group. Drugs such as acetaminophen, in both overdoses and therapeutic "misadventure" in heavy
alcohol drinkers, [95] antituberculous medications, [53] and halothane [65] are classically implicated agents
but any medication or toxin, including herbal remedies, [94] can be potentially hepatotoxic. Any of the
viral hepatitic agents, from A to E, [24] [29] [47] [71] [72] can present with fulminant failure. In many instances,
however, the exact causative agent is never elucidated.
Raised intracranial pressure (ICP) from cerebral edema has long been known to complicate fulminant
failure [91] and is a leading cause of mortality. Admission to the intensive care unit and intubation for

airway protection should be considered in every patient with stage III encephalopathy. ICP monitoring is
indicated for patients who have slipped into hepatic coma (stage IV encephalopathy), with intravenous
mannitol given as needed to lower an elevated ICP.
Patients receiving transplants for fulminant failure have long been reported to have a worse outcome than
patients receiving transplants for other reasons. The poorer outlook, in part, may reflect the urgent use of
suboptimal grafts (older, steatotic, split, or size-reduced grafts) but also reflects the presence of advanced
encephalopathy. [9] A recent series reported a 55.6% 1-year survival for patients transplanted in grade 3
coma (uncoordinated response to pain only) versus 83% for those in coma graded 2 or better. [9]
Despite the high mortality of patients with fulminant failure, not all patients require a transplant. Various
prognostic features have been suggested to predict which patients are likely to die without a transplant.
Bismuth and colleagues [6] suggest that a serum factor V less than 20% of normal in patients younger
than 30 years and less than 30% of normal in those older than 30 years predicts a 90% likelihood of
mortality with medical therapy alone. The King's College Liver Unit [57] has published widely cited
laboratory and clinical cutoff points that predict a survival less than 20% without transplantation for both
acetaminophen and nonacetaminophen-associated failure. We would recommend that those guidelines be
used with the overall clinical trend in mind; strict adherence to the guidelines for deciding when to
activate may result in unfortunate outcomes.
Chronic Autoimmune Liver Diseases
The autoimmune liver diseases, which can be divided into cholestatic (primary biliary cirrhosis [PBC],
primary sclerosing cholangitis [PSC]) and hepatocellular (autoimmune hepatitis) diseases, are common
indications for transplantation. PBC, considered to be the "prototypical" transplantable disease, has a
natural history that may span more than a decade in mild cases before terminating in end-stage cirrhosis.
The natural history of PBC can be predicted using several models; the most well-known being the Mayo
Model, [16] which incorporates the serum bilirubin, albumin, prothrombin time, age, and presence of
edema. In
251
clinical practice, the Mayo Model can be cumbersome to use; serum bilirubin, however, is a practical
marker to follow and patients should be referred for transplantation when the serum bilirubin is greater
than 100 mumo/L (6 mg/dL). Transplantation for PBC has been clearly shown to favorably interrupt the
natural history of PBC, with post-transplant survival greatly exceeding that predicted by the Mayo
Model. [45] Although the much rarer antimitochondrial antibody-negative PBC or autoimmune cholangitis
[51] [86] has not been well described in the transplant literature, undoubtedly the Mayo Model
prognostications apply and survival post-transplant is probably similar to that for PBC. PSC, which, like
PBC, may have a natural history spanning years, is strongly associated with inflammatory bowel disease
and, in 9% to 15% of cases, may be complicated by cholangiocarcinoma. [43] The Mayo Model has also
been applied to PSC [93] and transplantation, likewise, has been clearly demonstrated to favorably
interrupt the disease's natural history, compared with both predicted outcomes and actual postsurgical
biliary drainage outcome. [25]
Patients with inflammatory bowel disease are predisposed to colonic dysplasia [11] and colonic carcinoma
has been reported to be a common post-transplant cause of death. [55] Cholangiocarcinoma is also well

known to have an unfavorable impact on post-transplant survival. [1] Unfortunately, no study to date has
been able to reliably identify patients who have an incidental cholangiocarcinoma and the development
of that complication increases with duration of disease--from 13% at 5 years to 31% at 10 years in one
series. [25] For those reasons, it has been proposed to transplant patients with PSC earlier and to diligently
screen for colonic dysplasia pre- and post-transplant.
Another feature common to chronic cholestatic liver diseases is metabolic bone disease. The
post-transplant course of cholestatic bone disease is accelerated in the short term but may ameliorate,
with remineralization, in the long term. [31] [64]
Finally, autoimmune hepatitis is also readily amenable to transplantation. Although it is obvious that a
patient with "burned-out" autoimmune hepatitis and decompensated cirrhosis requires transplantation,
young patients dependent on high doses of corticosteroids may also be considered for earlier
The "Classic" Metabolic Diseases Affecting the Liver
Of the metabolic disorders that affect the liver, alpha1 -antitrypsin deficiency, Wilson's disease, and
hemochromatosis are the three that commonly come to mind in the differential diagnosis of chronic liver
disease. The underlying systemic disorder is "cured" with transplant in alpha1 -antitrypsin deficiency [90]
and Wilson's disease [73] because the metabolic defect resides within the liver. Whether or not the liver
contributes to the enhanced intestinal absorption of iron in hemochromatosis and would correct with liver
replacement is unresolved, with arguments in
252
the medical literature both pro [37] and con. [2] [18] No medical therapy exists for alpha1 -antitrysin
deficiency, whereas copper chelation therapy [96] with d-penicillamine, trientine, and zinc for Wilson's
disease and regular phlebotomy [21] for hemochromatosis should avert the need for transplantation if
instituted early. Medical therapy will not reverse end-stage cirrhosis and both Wilson's disease and
hemochromatosis have extrahepatic manifestations, the post-transplant reversibility of which remains
controversial. The post-transplant survival of patients with hemochromatosis has been suggested to be
significantly worse than the general liver transplant recipient population. A 5-year analysis of Medicare
transplant recipients revealed a 1- and five-year post-transplant survival of 54% and 43% for
hemochromatosis versus 79% and 69% for the general post-transplant population. [35] The exact reasons
for that are not clear, although the systemic manifestations of iron overload, (e.g., cardiac involvement)
are generally suspected. Of interest, despite the increased incidence of hepatoma in hemochromatosis, a
recent survey suggests that hepatoma alone does not account for the worse post-transplant outcome. [38]
The post-transplant reversibility of the neurologic manifestations of Wilson's disease is an issue that is
controversial. Copper chelation therapy has been demonstrated to ameliorate the neurologic symptoms of
Wilson's disease [84] but, presumably, those patients were treated earlier in the disease's course. Likewise,
neurologic symptoms have been reported to improve [73] and resolve [63] in some patients post-transplant.
Others have reported failure of a liver allograft to resolve the neurologic manifestations [30] and postulate
a component of irreversible basal ganglial damage. Of interest, a case was reported recently in which a
patient with liver disease successfully treated with copper chelation underwent transplantation because of

chelation-refractory neurologic symptoms. [46] The patient experienced a dramatic neurologic

improvement before dying of an intra-abdominal bleed in the early post-transplant period. Neurologic
examination revealed degenerative changes in the basal ganglia despite symptomatic improvement,
suggesting that functional improvement may not reflect the degree of structural damage and that, in
general, post-transplant reversibility depends on degree of disease severity, which may not be
predictable.
Before leaving this section, it is prudent to mention that Wilson's disease may present for the first time in
a young person as fulminant hepatitic failure. The features that suggest Wilson's disease include
hemolysis, mildly elevated serum transaminases, a serum alkaline phosphatase level that is unusually
normal or below normal, grossly elevated serum bilirubin, and severe coagulopathy. [83] In our
experience, such patients die without urgent transplantation.
Transplantation is commonly performed for end-stage liver disease secondary to chronic hepatitis C
(HCV) and hepatitis B (HBV). Both
253
viral diseases, however, have significant allograft reinfection rates in the untreated transplant recipient.
HCV infection has long been recognized to recur, with a recent study [26] establishing that 5 years
post-transplant, only 12% of recipients are spared histologic evidence of reinfection. HCV allograft
reinfection also leads to the same wide spectrum of disease seen in the nontransplant liver--from mild
chronic hepatitis to cirrhosis with clinical portal hypertension. [26] Despite the current lack of effective
antiviral therapy for allograft reinfection, the 5-year survival of HCV seropositive allograft recipients is
not significantly different from that of allograft recipients transplanted for other diseases. [26]
Until recently, transplantation for patients with HBV was viewed with skepticism because of the high
likelihood of allograft reinfection and rapid development of severe liver disease leading to poor graft
survival. [15] [72] A recent European study [72] reported a 3-year actuarial HBV reinfection rate exceeding
80% for those who are HBV DNA positive in serum pretransplant and approaching 60% for those who
were HBV DNA negative before transplantation. Fortunately, the long-term administration of high-dose
hepatitis B immune globulin (HBIG) has dramatically attenuated the rate allograft reinfection, [40] [72]
making liver transplantation for HBV-related disease feasible. The success of the antinucleoside agent,
lamivudine, in suppressing HBV replication [17] in nontransplanted patients suggests that even better
results may be forthcoming. Whether antinucleoside agents can substitute for HBIG post-transplant or
are best used in combination remains to be determined, given that the emergence of lamivudine escape
mutants following the post-transplant use of lamivudine as monotherapy recently was reported. [36]
The finding of hepatocellular carcinoma (HCC) in patients with viral hepatitis or alcoholic liver disease
during the pretransplant assessment is not uncommon. Although patients with advanced malignancy,
indicated by TMN stage III or IV disease (more than one lobe, macroscopic invasion of major vessel in
unilobe disease, with or without local or disseminated metastases [4] or large tumor or size alone (e.g.,

8 cm) have a uniformly poor survival post-transplant survival, [50] [60] [74] with 5-year survival on the
order of 10%, those with HCC found incidentally in the explanted liver and those with known but small
tumors have a better outcome. The policy of many transplant centers regarding tumor size is to accept
patients with tumors up to 5 cm in diameter and, if multifocal, no more than 3 cm. Using such criteria,
the group from Milan [48] recently reported excellent survival in a group of unresectable patients, with an
overall actuarial survival of 75% and, in those staged correctly, 85% at 4 years. The 4-year actuarial
recurrence-free survival in the latter group was 92%. It was unclear as to whether pretransplant
chemoembolization led to any significant improvement in outcome in that study.
254
CONTROVERSIAL ISSUES
Controversial issues in liver transplantation can be divided into medical controversies, regarding which
the debate is largely confined to the liver transplant community, and "ethical" controversies, for which
arguments are not limited to health care professionals but involve society as a whole. At the root of these
controversies is the issue of rational use of a limited resource. Many of the medical controversies of
yesterday have been resolved--e.g., hepatitis B is no longer an issue (in centers that have access to
HBIG). Small hepatocellular cancers are now known to have a good post-transplant survival, and even
certain complications of portal hypertension, such as the hepatopulmonary syndrome, formerly thought
to be a contraindication, have been demonstrated to resolve with transplantation. [41] Nevertheless, certain
issues are likely to remain controversial and are discussed briefly.
Age and Alcohol
Previously, advanced age was considered to be a contraindication to liver transplantation and many
centers had arbitrary age restrictions. Although many in both the medical and lay community are of the
belief that transplantation should be limited to younger patients, the collective transplant experience does
not support that belief. Several centers have reported no difference in post-transplant outcome in older
versus younger patients [22] [61] and have concluded that age should not be a contraindication. Perhaps, Sir
Roy Calne said it best when he advised physicians to consider the "biological age rather than the
chronological age." [12]
It has been clearly demonstrated that patients transplanted for alcoholic liver disease have a
post-transplant survival not different from that for most other diseases. [8] [44] [59] [82] The issue of liver
transplantation in these patients, however, remains a fiercely debated issue within North American
society, as evidenced by media commentary and letters to newspapers whenever former alcoholic
baseball players and Hollywood actors receive transplants. Two central themes recur regarding the issue
of transplantation for alcoholic cirrhosis. The first concerns the "ethical" issue of transplantation of a
limited resource for a "behaviorally" acquired disease and the second concerns the issue of resuming
alcohol use. The first issue is based heavily on philosophical, moral, and emotional viewpoints rather
than medical evidence. It is a societal issue, not a medical one. Post-transplant alcohol use, on the other
hand, has been well studied and it appears that most transplanted patients do not return to drinking and
even fewer return to serious drinking. [8] [28] [44] [59] [82] Abstinence pretransplant appears to be a predictive
factor with regard to post-transplant alcohol use and most centers require a documented period of

abstinence before transplantation, often in association with some form of pretransplant alcohol
rehabilitation. Six months, combined
255
with a good social support system, appears to be the minimum period of abstinence that predicts a
favorable outcome. [28]
Indolent Cancers
Extensive primary fibrolamellar cancers and neuroendocrine tumors with liver metastases are two
malignancies that occasionally are referred for transplant assessment. Because the cancers are slow
growing and indolent, it may be suggested that patients with such tumors should receive transplants.
Because fibrolamellar hepatomas are far less common than the usual hepatomas, the experience reported
in the transplant literature is limited. It is difficult to make any conclusive inferences from the data
because small numbers of fibrolamellar cancers are often reported as part of the much larger group of
hepatocellular carcinomas. Nonetheless, it appears that the long-term post-transplant prognosis of
fibrolamellars, like the usual hepatomas, may be dependent upon staging. [50] [66] Likewise, the data
regarding the transplantation of patients with hepatic, but no extrahepatic, metastases is noninferential.
Although Pichlmayr and colleagues [60] have reported excellent 3-year survival, in excess of 80% ( n =
11), other investigators, likewise reporting on small groups, have been less optimistic. One group
estimated a 5-year actuarial survival of 57% based on 11 patients [69] and another reported survival of
three of eight patients followed 6 to 55 months. [19] Hemangioendothelioma is another rare neoplasm for
which long-term survival has been reported. [58] [60] There are undoubtedly other rare indolent
malignancies for which the issue of transplantation remains unresolved or the answer is unknown.
CONTRAINDICATIONS TO LIVER TRANSPLANTATION

With the growing list of indications for transplantation, it must be kept in mind that a liver allograft is not
a general panacea for all patients with liver disease. Because donor organs constitute a scarce and
rationed resource, it is the responsibility of the transplant team to ensure that candidates are selected
carefully. At the present time, there is probably no role for "palliative" liver transplants for diseases
likely to recur and result in graft loss and death after only a few years. Likewise, patients with significant
concurrent medical disease, unlikely to be ameliorated by a new liver, for which the prognosis is
unfavorable, should not be candidates. In general, contraindications can be considered "absolute" or
"relative."
Absolute Contraindications
The list of absolute contraindications for transplantation reflect the policies of the British Columbia
Transplant Society and the University
256
of California-San Francisco and include both medical and psychosocial conditions. Absolute oncologic
medical contraindications include extrahepatic malignancies (with, perhaps, a few exceptions, as

previously discussed), large hepatocellular carcinomas, and cholangiocarcinomas. Cholangiocarcinomas

deserve special mention because they were once thought to be ideal malignancies for liver
transplantation. The long-term post-transplant results proved to be disappointing, with the 2- to
three-year survival around 30%. Some patients with cholangiocarcinoma, however, may enjoy a good
long-term, recurrence-free survival.99 It has been suggested that the Klatskin variety of
cholangiocarcinomas may represent a subgroup that has a more favorable outcome.97 Despite that,
because of the issue of limited resource utilization, our centers and most others do not consider
transplantation for that malignancy.
The other absolute medical contraindications to transplantation include acute ongoing infection, which
may be only a temporary contraindication pending adequate antimicrobial therapy, human
immunodeficiency virus (HIV) seropositivity, multisystem organ failure (MSOF), and advanced
cardiopulmonary disease. It may be argued that HIV infection is a fatal, yet chronic disease, especially in
patients with favorable CD4 counts. We note, however, that post-transplant outcome in patients with
acquired immunodeficiency syndrome (AIDS) is poor100 and the that combined effect of an antirejection
protocol is likely to increase the likelihood of opportunistic infection and malignancy. Moreover, the
Toronto Hospital group101 has demonstrated that, in the case of coinfection with HIV and HBV, death
from AIDS-related causes predominates. Although the exclusion of patients with advanced
cardiopulmonary disease would appear obvious and the same medical contraindications to any other
major surgery apply, we note that 7% (3 of 41 patients) of deaths in a recent European report were
secondary to congestive heart failure.102 MSOF is likewise an absolute contraindication, with a dismal
short-term survival that is inversely related to the number of organ systems that have failed--from 50%
survival with one-organ failure to nil survival with four-organ failure.103
Although some may view psychosocial contraindications as "soft," there are, in fact, compelling reasons
to refuse certain patients on those grounds. To be listed for a transplant, the patient must be agreeable and
give informed consent. That may appear to be an obvious situation but, in reality, transplantation is a
very emotional issue involving not just the patient, but the patient's family. It has to be clear that the
patient, not just the patient's family, wants a transplant, no matter how vocal the family may be. The
issue of compliance, for medications, diagnostic and therapeutic tests, and clinic visits, is also of
importance. Documented, habitual noncompliance is an absolute contraindication for transplantation
because pretransplant noncompliance has been demonstrated to predict post-transplant graft loss.104
Active alcohol or drug abuse and unwillingness to abstain is similarly an absolute contraindication at
most centers.
257
There are many conditions that, of themselves, do not disqualify patients from transplant candidacy yet
may negatively affect long-term post-transplant survival, increase the operative mortality, or make the
technical aspects of surgery difficult. Many such relative contraindications are center specific, depending
on the expertise of the surgeons and the resources and expertise of the intensive care and transplant units.
What may not be feasible at a small center, may be possible at a larger center with more resources and
experience. Examples of such considerations include previous abdominal surgery, malnutrition, and
significant diabetes mellitus with target organ damage, as well as psychosocial issues such as lack of

social support.
THE FUTURE
As the list of transplantable medical conditions expands, the number of patients seeking organs also
grows, translating to longer waiting times and increased morbidity and mortality for those on the wait
list. Although greater organ donor awareness may increase the number of available organs, liver
allografts are likely to remain a limited and rationed resource. Recently, tremendous resources and
attention have been directed toward xenotransplantation. In theory, the availability of xenotransplants
would take pressure off the waiting lists. Xenotransplantation, at the current time, however, is far from
becoming a realistic generalizable alternative and introduces a whole new set of immunologic, infectious,
and ethical problems. Perhaps, in the global sense, a partial answer to our problems lies not in the arena
of transplantation, but in the areas of public health and primary care. Measures to reduce alcohol
dependence and alcoholic liver disease, intravenous drug abuse, and viral hepatitis, and to introduce
hepatitis B vaccination programs and reduce vertical transmission of that disease, would, in the decades
to come, reduce the waiting lists and make donor organs available to those with unpreventable liver
disease.
References
1. Abu-Elmagd KM, Selby R, Iwatsuki S, et al: Cholangiocarcinoma and sclerosing cholangitis: Clinical characteristics and
effect on survival after liver transplantation. Transplant Proc 25:1124-1125, 1993
2. Adams PC, Ghent CN, Grant DR, et al: Transplantation of a donor liver with hemochromatosis: Evidence against an
inherited intrahepatic defect. Gut 32:1082-1103, 1991
3. Adams PC, Grant DR, Wall WJ: Employment after liver transplantation. Hepatology 21:140-144, 1995
4. American Joint Committee on Cancer: Manual for Staging of Cancer, ed 3. Philadelphia, Lippincott, 1987, pp 87-92
5. Belle SH, Beringer C, Detre KM: Liver transplantation in the United States: Results
258
from the National Pitt-UNOS Liver Transplant Registry. In Terasaki PI, Cecka JM (eds): Clinical Transplants. Los
Angeles, UCLA Tissue Typing Laboratory, 1994, pp 19-35
6. Bernuau J, Samuel D, Durand F, et al: Criteria for emergency liver transplantation in patients with acute viral hepatitis
and factor V below 50% of normal: A prospective study. Hepatology 14:49A, 1991
7. Bilheimer
DW, Goldstein JL, Grundy SM, et al: Liver transplantation to provide low-density lipoprotein receptors and
lower plasma cholesterol in a child with homozygous familial hypercholesterolemia. N Engl J Med 311:1658-1664, 1984
8. Bird DL, O'Grady JG, Harvey FA, et al: Liver transplantation in patients with alcoholic cirrhosis: Selection criteria and
rates of survival and relapse. British Medical Journal 301:15-17, 1990
9. BismuthH, Samuel D, Castaing D, et al: Orthotopic liver transplantation in fulminant and subfulminant hepatitis: The
Paul Drousse experience. Ann Surg 222:109-119, 1995

10. BonselGJ, Essink-Bot ML, Klompmaker IJ, et al: Assessment of the quality of life before and following liver
transplantation. Transplantation 53:796-800, 1992
11. BrentnallTA, Haggitt RC, Rabinovitch PS, et al: Risk and natural history of colonic neoplasm in patients with primary
sclerosing cholangitis and ulcerative colitis. Gastroenterology 110:331-338, 1996
12. Calne R: Contraindications to liver transplantation. Hepatology 20:3S-4S, 1994
13. Conn HO: Transjugular intrahepatic portal-systemic shunts: The state of the art. Hepatology 17:148-158, 1993
14. D'Amico G, Morabito A, Pagilaro L, et al: Survival and prognostic indicators in compensated and decompensated
cirrhosis. Dig Dis Sci 31:468-475, 1986
15. Demetris AJ, Todo S, Van Thiel DH, et al: Evolution of hepatitis B liver disease after hepatic replacement: Practical
and theoretical considerations. Am J Pathol 137:667-676, 1990
16. Dickson
ER, Grambsch PM, Fleminng TR, et al: Prognosis in primary biliary cirrhosis: Model for decision making.
Hepatology 10:1-7, 1989
17. Dienstag
JL, Perrillo RP, Shiff ER, et al: A preliminary trial of lamivudine for chronic hepatitis B infection. N Engl J
Med 333:1657-1661, 1995
18. Dietz
O, Vogel W, Braunsperger B, et al: Liver transplantation in idiopathic hemochromotosis. Transplant Proc
22:1512-1513, 1990
19. DoussetB, Houssin D, Soubrane O, et al: Metastatic endocrine tumors: Is there a place for liver transplantation? Liver
20. Dupeyron C, Mangeney N, Sedrati L, et al: Rapid emergence of quinolone resistance in cirrhotic patients treated with
norfloxacin to prevent spontaneous bacterial peritonitis. Antimicrob Agents Chemother 38:340-344, 1994
21. Edwards CQ, Kushner JR: Screening for hemochromotosis. N Engl J Med 328:1616-1620, 1993
22. Emre S, Mor E, Schwartz ME, et al: Liver transplantation in patients beyond age 60. Transplant Proc 25:1075-1076,
1993
23. European FK506 Multicentre Study Group: Randomized trial comparing tacrolimus (FK506) and cyclosporin in
prevention of liver allograft rejection. Lancet 344:423-428, 1994
24. Farci P, Alter HJ, Shimoda A, et al: Hepatitis C virus-associated fulminant hepatic failure. N Engl J Med 335:631-634,
1996
25. Farges
O, Malassagne B, Sebagh M, et al: Primary sclerosing cholangitis: Liver transplantation or biliary surgery?
Surgery 117:146-155, 1995
26. GaneEJ, Portmann BC, Naoumov NV, et al: Longterm outcome of hepatitis C infection after liver transplantation. N
Engl J Med 334:815-820, 1996
27. GinesP, Rimola A, Planas R, et al: Norfloxacin prevents spontaneous bacterial peritonitis recurrence in cirrhosis:
Results of a double-blind placebo-controlled trial. Hepatology 12:716-724, 1990

28. Gish
RG, Lee AH, Keefe EB, et al: Liver transplantation for patients with alcoholism and end stage liver disease. Am J
29. GordonFD, Anastopoulos H, Khettry V, et al: Hepatitis C infection: A rare case of fulminant hepatic failure. Am J
259
30. GuarinoM, Stracciari A, D'Alessandro R, et al: No neurological improvement after liver transplantation for Wilson's
disease. Acta Neurol Scand 92:405-408, 1995
31. HayJE, Dickson ER, Wiesner RH, et al: Long-term effect of orthotopic liver transplantation on the osteopenia of
primary biliary cirrhosis [abstract]. Hepatology 12:838, 1990
32. Heath
Resources and Services Administration, Bureau of Health Resources Development: 1991 Report of
Center-Specific Graft and Patient Survival Rates. Rockville, MD, US Dept of Health and Human Services, 1992, pp 8-10
33. HuntCM, Tart JS, Dowdy E, et al: Effect of orthotopic liver transplantation on employment and health status. Liver
34. KaufmanSS, Wood RP, Shaw BJ, et al: Orthotopic liver transplantation for type I Crigler-Najjar syndrome. Hepatology
6:1259-1262, 1986
35. Kilpe
VE, Krakauer H, Wren RE: An analysis of liver transplant experience from 37 transplant centers as reported to
Medicare. Transplantation 56:554-561, 1993
36. Kneteman NM, Bain VG, Shapiro AMJ, et al: Prevention of hepatitis B recurrence after liver transplantation with
lamivudine [abstract]. Clin Invest Med 19:S101, 1996
37. Kosinkinas
J, Portmann B, Lombard M, et al: Persistent iron overload 4 years after inadvertent transplantation of a
haemochromatotic in a liver patient with primary biliary cirrhosis. J Hepatol 16:351-354, 1992
38. KowdleyKV, Hassanein T, Kaur S, et al: Primary liver cancer and survival in patients undergoing liver transplantation
for hemochromatosis. Liver Transplantation and Surgery 1:237-241, 1995
39. Lake JR: Changing indications for liver transplantation. Gastroenterol Clin North Am 22:213-229, 1993
40. LakeJR: Should liver transplantation be performed for patients with chronic hepatitis B? Yes! Liver Transplantation
and Surgery 1:260-265, 1995
41. Lange PA, Stoller JK: The hepatopulmonary syndrome. Ann Intern Med 122:521-529, 1995
42. Lee WM: Acute liver failure. N Engl J Med 329:1862-1872, 1993
43. Lee YM, Kaaplan M: Primary sclerosing cholangitis. N Engl J Med 332:924-933, 1994
44. LuceyMR, Merion AM, Henley KS, et al: Selection for and outcome of liver transplantation in alcoholic liver disease.

45. Markus BH, Grambsch PM, Dickson ER, et al: Efficiency of liver transplantation in patients with primary biliary
cirrhosis. N Engl J Med 320:1709-1713, 1989
46. MasonAL, Marsh W, Alpers DH: Intractable neurological Wilson's disease treated with orthotopic liver transplantation.
Dig Dis Sci 38:1746-1750, 1993
47. Masuda CT, Shaw BW Jr, Zetterman RK, et al: Fulminant hepatic failure with massive necrosis as a result of hepatitis
A infection. J Clin Gastroenterol 17:158-162, 1993
48. Mazzaferro V, Regalia E, Doci R, et al: Liver transplantation for the treatment of small hepatocellular carcinomas in
patients with cirrhosis. N Engl J Med 334:693-699, 1996
49. McDonald JC, Landreneau MD, Rohr MS, et al: Reversal by liver transplantation of the complications of primary
hyperoxaluria as well as the metabolic defect. N Engl J Med 321:1100-1103, 1989
50. McPeakeJR, O'Grady JG, Zaman S, et al: Liver transplantation for primary hepatocellular cancer: Tumor size and
number determine outcome. J Hepatol 18:226-234, 1993
51. Michieletti
P, Wanless IR, Katz A, et al: Antimitochondrial-negative primary biliary cirrhosis: A distinct syndrome of
autoimmune cholangitis. Gut 35:260-265, 1994
52. Millis
JM, Martin P, Gomes A, et al: Transjugular intrahepatic porto-systemic shunts: Impact on liver transplantation.
Liver Transplantation and Surgery 1:229-233, 1995
53. Mitchell I, Wendon J, Fitt S, et al: Antituberculous therapy and acute liver failure. Lancet 345:555-556, 1995
54. Mullen
MA, Kohut N, Sam M, et al: Access to adult liver transplantation in Canada: A survey and ethical analysis. Can
Med Assoc J 154:337-342, 1996
55. Narumi S, Roberts JP, Emond J, et al: Liver transplantation for sclerosing cholangitis. Hepatology 22:451-457, 1995
56. Ochs A, Rossle M, Haag K, et al: The transjugular intrahepatic portosystemic stent-shunt procedure for refractory
ascites. N Engl J Med 332:1192-1197, 1995
57. O'GradyJG, Alexander GJM, Hayllar KM, et al: Early indications of prognosis in fulminant hepatic failure.
260
58. Olthoff
KM, Millis JM, Rosove MH, et al: Is liver transplantation justified for treatment of hepatic malignancies? Arch
Surg 125:1261-1266, 1990
59. OrsorioRA, Ascher NL, Avery M, et al: Predicting recidivism after orthotopic liver transplantation for alcoholic liver
disease. Hepatology 20:105-110, 1994
60. PichlmayerR, Weimann A, Oldhafer KJ, et al: Role of liver transplantation in the treatment of unresected liver cancer.
World J Surg 19:807-813, 1995
61. Pirsch JD, D'Alessandro AM, Voss BJ, et al: Orthotopic liver transplantation in patients 60 years of age and older.

62. Platz
KP, Sollinger HW, Hullet DA, et al: RS-61443, a new potent immunosuppressive agent. Transplantation
51:428-432, 1991
63. PolsonRJ, Rolles RY, Calne RY, et al: Reversal of severe neurological manifestations of Wilson's disease following
orthotopic liver transplantation. Q J Med 64:685-691, 1987
64. Porayko MK, Wiesner RH, Hay JE, et al: Bone disease in liver transplantation recipients L-incidence, timing, and risk
factors. Transplant Proc 23:1462-1465, 1991
65. Ray DC, Drummond GB: Halothane hepatitis. Br J Anaesth 67:84-99, 1991
66. Ringe
B, Wittekind C, Weimann A, et al: Results of hepatic resection and transplantation for fibrolamellar carcinoma.
Surg Gynecol Obstet 175:299-305, 1992
67. Rolachon A, Cordier L, Bacq Y, et al: Ciprofloxacin and long-term prevention of spontaneous bacterial peritonitis:
Results of a prospective controlled trial. Hepatology 22:1171-1174, 1995
68. RossleM, Haag K, Ochs A, et al: The transjugular intrahepatic portosystemic stent-shunt procedure for variceal
bleeding. N Engl J Med 330:165-171, 1994
69. RoutleyD, Ramage JK, McPeake J, et al: Orthotopic liver transplantation in the treatment of metastatic neuroendocrine
tumors of the liver. Liver Transplantation and Surgery 2:118-121, 1995
70. Runyon BA, McHutchinson JG, Antillon MR, et al: Short course vs long-course antibiotic treatment of spontaneous
bacterial peritonitis: A randomized controlled study of 100 patients. Gastroenterology 100:1737-1742, 1991
71. Sallie
R, Chiyende J, Tan KC, et al: Fulminant hepatic failure resulting from co-existent Wilson's disease and hepatitis
E. Gut 35:849-853, 1994
72. SamuelD, Miller R, Alexander G, et al: Liver transplantation in European patients with the hepatitis B surface antigen.
N Engl J Med 329:1842-1847, 1993
73. Schilsky
ML, Scheinberg IH, Sternlieb I: Liver transplantation for Wilson's disease: Indications and outcome.
Hepatology 19:583-537, 1994
74. Selby R, Kadry Z, Carr B, et al: Liver transplantation for hepatocellular cancer. World J Surg 19:53-58, 1995
75. Shevell
MI, Bernard B, Adelson JW, et al: Crigler-Najjar syndrome type I: Treatment by home phototherapy followed
by home orthotopic hepatic transplantation. J Pediatr 110:429-431, 1987
76. Singh N, Gazowski T, Yu VL, et al: Trimethoprim-sulphamethoxazole for prevention of spontaneous bacterial
peritonitis in cirrhosis: A randomized trial. Ann Intern Med 122:595-598, 1995
77. Somberg KA, Riegler JL, Doherty M, et al: Hepatic encephalopathy following transjugular intrahepatic portosystemic
shunts (TIPS): Incidence and risk factors. Hepatology 16:122A, 1992
78. Sanchez-Urdazpal L, Czaja AJ, van Hoek B, et al: Prognostic features and role of liver transplantation in severe
corticosteroid-treated autoimmune chronic active hepatitis. Hepatology 15:215-221, 1992
79. Starzl TE: History of liver and other splanchnic organ transplantation. In Busuttil RW, Klintmalm GB (eds):

Transplantation of the Liver. Philadelphia, WB Saunders Company, 1996, pp 3-22
80. Starzl TE, Demetris AJ, Van Thiel D: Liver transplantation. N Engl J Med 329:1014-1022, 1092-1099, 1989
81. Starzl
TE, Klintmalm GBG, Porter KA, et al: Liver transplantation with the use of cyclosporin A and prednisone. N
Engl J Med 305:266-269, 1981
82. Starzl TE, Van Thiel D, Tzakis AG, et al: Orthotopic liver transplantation for alcoholic cirrhosis. JAMA 260:2542,
1988
261
83. Sternlieb I: Wilson's disease: Indications for liver transplants. Hepatology 4:15S-17S, 1984
84. StremmelW, Meyernose KW, Niederau C, et al: Wilson disease: Clinical presentation, treatment and survival. Ann
Intern Med 115:720-726, 1991
85. Tarter
RE, Switala J, Arria A, et al: Quality of life before and after orthotopic liver transplantation. Arch Intern Med
151:1521-1526, 1991
86. Taylor SL, Dean PJ, Riely CA: Primary autoimmune cholangitis: An alternative to antimitochondrial-negative primary
biliary cirrhosis. Am J Surg Pathol 18:91-99, 1994
87. TitoL, Rimola A, Gines P, et al: Recurrence of spontaneous bacterial peritonitis in cirrhosis: Frequency and predictive
factors. Hepatology 8:27-31, 1988
88. UNOS: Reported deaths on the OPTN waiting list, 1988 to 1994. UNOS Update 12:25, 1996
89. The
U.S. Multicenter FK506 Study Group: A comparison of tacrolimus (FK506) and cyclosporine for
immunosuppression in liver transplantation. N Engl J Med 331:1110-1115, 1994
90. Vennarecci G, Gunson BK, Ismail T, et al: Transplantation for end-stage liver disease related to alpha1 -antitrypsin.
91. WareAJ, D'Agostinho A, Combes B: Cerebral edema: A major complication of massive hepatic necrosis.
92. WattsRWE, Rolles K, Morgan SH, et al: Successful treatment of primary hyperoxaluria Type I by combined hepatic
and renal transplantation. Lancet 2:474-475, 1987
93. WiesnerRH, Porayko MK, Dickson ER, et al: Selection and timing of liver transplantation in primary biliary cirrhosis
and primary sclerosing cholangitis. Hepatology 16:1290-1299, 1992
94. YoshidaEM, McLean CA, Cheng ES, et al: Chinese herbal medicine, fulminant hepatitis and liver transplantation. Am
J Gastroenterol, in press
95. Zimmerman HJ, Maddrey WC: Acetoaminophen (paracetamol) hepatotoxicity with regular intake of alcohol: Analysis
of instances of therapeutic misadventure. Hepatology 22:767-773, 1995

96. ZuckerS, Gollan J: Copper metabolism and Wilson's disease An "ion" recent advances. In Thomson ABR, Shaffer E
(eds): Modern Concepts in Gastroenterology, vol 3. New York, Plenum Medical Book Company, 1992, pp 223-262


April 22, 2000
Conde Petra
PREFACE
Additional Article
Consult's full-text literature database. Volume 1 Number 2 August 1997
Full Text
PREFACE
Frontmatter
Figure . EMMET B. KEEFFE, MD

Guest Editor
The topics included in this issue of Clinics in Liver Disease

broadly address the science, politics, and economics of liver
transplantation. Since the first human liver transplantation in
1963 and endorsement of its clinical applicability by the
National Institutes of Health Consensus Development
Conference in 1983, the field of liver transplantation has
advanced at a rapid pace and undergone continuous medical and
surgical refinement. The initial focus of liver transplantation
was on technical aspects and patient and graft survival, but
more recently the focus has changed to broader issues such as
allocation and distribution of donor organs, selection of patients
for liver transplantation, determination of the ideal timing of
transplantation, new strategies in the management of specific
subsets of patients undergoing transplantation, refinement in the
use of established immunosuppressive agents and the addition
of newer agents, better treatment and prevention of common
infections associated with liver transplantation, and the
economics of liver transplantation. All of these topics are
addressed in this issue of Clinics in Liver Disease by renowned
experts.

This issue begins with an overview of the selection of patients

for liver transplantation which takes into account a limited
supply of donor organs and our inability to transplant all
patients who might benefit from the procedure. This overview is
complemented by a discussion of the role of prognostic models
in the timing of liver transplantation. The growing discrepancy
between supply and demand for donor livers has sparked
ongoing debates on allocation and distribution of donor livers,
and the changing economics in health care delivery are
affecting the way we care for liver transplant recipients. These
topics are discussed in two provocative contributions by
recognized experts and leaders in these areas. Several articles
address current thought and medical approaches to the
application of liver transplantation for specific diseases,
including alcoholic liver disease, chronic hepatitis B and C,
hepatocellular carcinoma, metabolic liver diseases, and
pediatric conditions. In an excellent literature review
supplemented by practical advice, cyclosporine-based and
tacrolimus-based immunosuppressive regimens are compared.
The current approach to the diagnosis, treatment and prevention
of cytomegalovirus and Epstein-Barr virus infections after liver
transplantation are reviewed in detail. Finally, the current
literature
14A
on reproductive function and results of pregnancy after liver

transplantation are presented.
I would like to thank all of the authors for their rigorous review
of the literature and sharing of their knowledge and expertise in
their outstanding contributions to this issue of Clinics in Liver
Disease.
EMMET B. KEEFFE
Guest Editor Division of Gastroenterology

Stanford University School of Medicine 750 Welch Road, Suite
210


April 22, 2000
Conde Petra
Additional Article
Full Text
Frontmatter 839
ALCOHOL USE IN
CONTEMPORARY AMERICAN ALCOHOLIC LIVER DISEASE
SOCIETY
SELECTING ALCOHOLIC LIVER

DISEASE PATIENTS FOR ORGAN LIVER TRANSPLANTATION FOR ALCOHOLIC LIVER DISEASE
LIVER TRANSPLANT
LISTING THE PATIENT
MEDICAL EVALUATION FOR LIVER

TRANSPLANTATION
Anne Burke 1 MB, BCh, BAO, MRCPI
PSYCHOSOCIAL EVALUATION Michael R. Lucey 1 2 MD, FRCPI
AND PREDICTING ALCOHOLIC
RELAPSE 1 Division of Gastroenterology (AB, MRL)
THE POSTTRANSPLANT PERIOD
2 Liver Transplant Program (MRL), Hospital of the University of Pennsylvania Philadelphia, Pennsylvania
Outcome
Monitoring the Posttransplant Address reprint requests to

Patient Michael R. Lucey, MD, RCPI
Division of Gastroenterology
Mortality Hospital of the University of Pennsylvania
Ravdin 3, 3400 Spruce Street
Alcohol Use After Liver Transplant Philadephia, PA 19104
Morbidity and Quality of Life
Healing is a
FUTURE DIRECTIONS matter of time, but it is sometimes also a matter of opportunity.
The Role of Naltrexone HIPPOCRATES, 460-400 BC
CONCLUSION In February 1998, 9823 patients in the United States were awaiting an orthotopic liver transplant (OLT). [1A] Among candidates with a
recorded diagnosis, 1212 (16.7%) had alcoholic liver disease (ALD). Only hepatitis C was more common among the diagnoses on the
References
waiting list (1639 or 22.8%). In addition, 524 (7.3%) patients carried both diagnoses. During 1996, 4062 liver transplants were carried out
http://home.mdconsult.com/das/journal/view/N/10888407?ja=129191&source=&stty=cow&nid=39988&tlt=Z&pos=&tag=relinfo&anc=ja (1 of 12) [22/04/2000 0:57:54]

in the United States. In the same year, 960 patients died on the OLT waiting list. Although there has been a four-fold increase in the
retrieval of organs considered suitable for use during the past 8 years, there has been a 10-fold growth both in the number of people
waiting for OLT and in the number of people dying on the OLT waiting list during the same period. The discrepancy between the number
of available donor organs and the potential recipients has led to a continuing debate about the appropriateness of using a scarce donor
liver to treat what is, in the estimation of some, a self-inflicted malady. [19] [27]
This article discusses the role of OLT in the treatment of the patient with alcoholic liver disease (ALD), the selection of patients with ALD
for OLT, and their outcome after liver transplantation. We set our discussion in the context of the ethical issues involved in this
controversial area.
ALCOHOL USE IN CONTEMPORARY AMERICAN SOCIETY

Alcohol use is very widespread in our society. Sixty-eight percent of adult Americans drink at least one beer a month. [16] Alcoholism, which is a
pathologic
840
loss of control over alcohol use, affects some 7% to 10% of Americans at some point in their lives. [5] Genetic susceptibility and diverse social and
cultural pressures contribute to the risk of alcoholism. There is considerable debate about the underlying mechanisms that cause alcoholism. At
one extreme, there is the personal responsibility school of thought that interprets alcoholism as a behavioral fault within an individual. On the
other hand, there is a considerable body of evidence to show that addiction is a chronic relapsing brain disease, in part mediated by a genetic
propensity, wherein drug use leads to changes in brain structure and function that persist long after the individual stops taking the drug. [23] The
fact that the majority of alcohol abusers do not develop ALD suggests that there are other environmental or genetic factors involved in the
pathogenesis of alcohol-induced liver damage. Putative cofactors include the gender of the individual, hereditary disorders, such as
hemachromatosis, and acquired factors, such as viral hepatitis. The emerging evidence that ALD is the result of a combination of pathogenic
influences, of which alcohol use is a required but not the sole factor, challenges the more facile formulation that places the responsibility for
developing ALD on the alcoholic.
Similarly, among persons who drink, there is a wide variation in susceptibility to ALD. The threshold of ethanol consumption that facilitates
development of ALD seems to be as little as 30 g/d (three drinks) [4] or 14 drinks per week. [2] The risk of developing liver disease increases with
increasing consumption. Nevertheless, only a minority (13.5%) of patients drinking in excess of 120 g/d (12 drinks) manifest features of ALD.
Despite the fact that fewer than 15% of alcoholics develop ALD, it is estimated that ALD affects between 1 and 2 million Americans at any time.
[5] Even though many of these patients are unsuitable for liver transplantation, either because they have not progressed to end-stage liver failure, or
because of concomitant confounding ailments (see The section on selecting ALD patients for OLT), it is easy to understand how ALD could
overwhelm the available donor resources.
SELECTING ALCOHOLIC LIVER DISEASE PATIENTS FOR ORGAN LIVER TRANSPLANT

The process of evaluating a patient with end-stage liver disease of any cause for liver transplantation can be reduced to three questions:
Does the patient have severe liver disease that warrants an OLT?
Does the patient have other medical, surgical, or psychosocial circumstances that would confound a successful OLT?

What are the patient's wishes regarding OLT?
In the program at the Hospital of the University of Pennsylvania, the authors have adopted a multidisciplinary team approach to answer these
questions. Each patientr receives an individualized assessment. Among alcoholic patients, we therefore take special care to assess the patient's
psychosocial background and support structures, which are crucial to maintaining sobriety and the health of the graft after transplant.
LISTING THE PATIENT

The immediate end point of the transplant evaluation process is determining whether a prospective candidate should be placed on the liver
transplant waiting
841
list, often called listing the patient. At the present time, in the United States, donor livers are allocated to patients on the liver transplant waiting
list according to a points priority scheme. Points are given for blood-type (ABO) compatibility, time on the waiting list, and severity of liver
failure. In recent years, as intervals spent on the waiting list have extended inexorably, estimating prognosis has become more difficult than ever
before. Transplant programs are faced with the dilemma of estimating a candidate's prognosis while factoring in an uncertain period of
pretransplant management on the waiting list.
Because time on the waiting list is recognized as a priority factor and the duration on the waiting list is uncertain, there is a tendency for placement
of stable patients on the list, a practice referred to as list inflation. To establish a degree of fairness and uniformity for entry onto the waiting list,
guidelines for minimal listing criteria were published in 1997. [24] The basis of these criteria is an expected 1-year survival of less than 90%. The
survival rate of patients with cirrhosis is related to the Child-Pugh score (Table 1) . The estimated 1- and 5-year survival rates for Child's C
patients are 95% and 75%, respectively. Predicted survival deteriorates considerably when cirrhosis becomes decompensated, as indicated by the
onset of any of the following clinical phenomena--ascites, variceal hemorrhage, jaundice, encephalopathy, or hepatorenal syndrome. A Child's
score of 7 or above or evidence of decompensation are indications for listing for OLT. [24] Patients on the list need to be reviewed regularly to
follow their course and modify their status on the waiting list as necessary.
ALD poses particular problems with regard to estimating prognosis. Many patients with alcoholic cirrhosis who deteriorate do so because of the
acute effects of alcohol on liver function. Furthermore, it is well documented that liver function can improve in acute alcoholic hepatitis with
abstinence. [1] This occurs even when the patient has already established cirrhosis. This is the basis of discriminating against alcoholics who have
been drinking in the previous 6 months, as is discussed further in the section on psychosocial assessment.
MEDICAL EVALUATION FOR LIVER TRANSPLANTATION

As liver transplantation has become more sophisticated, the list of absolute contraindications to placement on the waiting list has shortened.
Significant impediments to successful OLT include extrahepatic malignant disease, cholangiocarcinoma, or advanced hepatocellular carcinoma;
ongoing drug or alcohol use; active infection; significant organic brain injury or major intractable psychoses; or significant cardiopulmonary
disease. Infection by human immunodeficiency virus (HIV), with or without acquired immune deficiency syndrome, used
TABLE 1 -- CHILD-PUGH SCORE
Points
Criteria 1 2 3

Encephalopathy (grade) None 1-2 3-4

Ascites None Slight Moderate-severe
Bilirubin (mg/dL) 1-2 2-3 >3
Albumin (g/dL) >3.5 2.8-3.5 <2.8
Prothrombin time (seconds prolonged) 1-4 4-10 >10
Grading: Score 6 points = grade A; 7-9 points = grade B; 10 points = grade C.
842
to be a contraindication but this status has become less certain with the advent of more effective antiviral agents. Similarly, extensive thrombosis
of the portal and mesenteric venous system may no longer preclude liver transplantation. The impact of relative contraindications therefore change
with advances in medical practice.
Medical screening for alcoholic liver transplant candidates begins with history and physical examination, routine hematologic and biochemical
screening, electrocardiogram, and chest radiograph. HIV serology is checked in all candidates irrespective of underlying diagnosis. In view of the
increased risk of hepatoma in cirrhotic patients, all OLT candidates have abdominal cross-sectional imaging (sonography, MR imaging or CT
scanning as appropriate) and serum alpha-fetoprotein determination. Comorbid liver diseases should also be checked for by testing serology for
hepatitis B and C infection, iron studies (ferritin and iron saturation) for hemachromatosis, and alpha1 -antitrysin levels. Patients with ALD are a
fairly young group compared with those with other causes of liver disease. In those fewer than 35 years of age, it is prudent to check
ceruloplasmin or urinary copper to rule out Wilson's disease. The vast majority of patients will have had a liver biopsy to diagnose cause and
severity of their liver injury during the course of their disease. There are also a few conditions that are more likely to occur in the alcohol-abusing
patient--cardiomyopathy, organic brain syndromes, peripheral and autonomic neuropathy, chronic pancreatitis, tuberculosis, and osteopenia.
History and physical examination should be directed toward these areas. The following investigations should be done on case-by-case basis: echo
cardiography, arterial blood gases, pulmonary function tests, head CT image, nerve conduction studies, endoscopic retrograde
cholangiopancreatography, and bone densitometry. All alcoholic candidates should receive purified protein derivative (PPD) or Mantoux testing.
Malnutrition is common among ALD patients. Formal nutritional assessment and counseling by a dietician can be helpful.
PSYCHOSOCIAL EVALUATION AND PREDICTING ALCOHOLIC RELAPSE

Medical screening is not an exact science and places patients at high, moderate, or low risk for postoperative complications. Similarly,
psychosocial evaluation stratifies patients by their risk of relapse for alcoholism. A recent survey indicated that most (66 of 68 surveyed)
transplant programs in the United States give preference to alcoholic patients whom they assess to be more likely to remain abstinent from alcohol
after transplantation. It is much less certain how best to achieve the goal of predicting future drinking.
Most transplant programs in the United States require alcoholic patients to undergo formal psychiatric evaluation. [10] Furthermore, several payers
require psychiatric evaluation of an alcoholic patient before they will approve payment. Eighty-five percent of OLT centres surveyed make use of
the "6-month rule," which requires that an alcoholic patient must have demonstrated 6 months of abstinence before being considered for OLT. [10]
Almost half have had third-party payers refuse to pay for an OLT because of failure to complete a set period of abstinence. The accuracy of
predicting post-OLT sobriety based on short duration of pretransplant sobriety is questionable.
Most OLT centers in the United States demand a 6-month abstinence period before they will transplant a patient for ALD. [10] This 6-month period
is based on a 1990 report [22] suggesting more frequent rates of relapse among alcoholic OLT recipients who had been abstinent for fewer than 6

months before OLT.
843
The numbers of subjects involved, however, were too small for the statistical method used. Other follow-up data with greater numbers have found
no significant difference in relapse based on a 6-month threshold of pre-OLT abstinence. [7] [8] [11] [13] [20] [30] [35] [38]
A 6-month period of abstinence was proposed for minimal listing criteria for ALD, although the suggestion was rejected by the United Network
for Organ Sharing (UNOS). [20] The participants at the conference on minimal listing criteria noted that adherence to the 6-month abstinence
allows the inflammatory effects of recent alcohol consumption to resolve and exclude alcoholic hepatitis patients from OLT. Many authors also
agree that exceptional patients with fewer than 6 months abstinence could be considered for OLT. Many programs (75% of OLT centers surveyed)
ask the ALD patient to sign a contract to undergo alcohol rehabilitation. [10] Success with this approach is mixed and the authors do not use it at
their center.
At present, alcoholic hepatitis is a contraindication to liver transplantation. In a survey of 69 OLT centers in the United States, [10] active alcohol
use was considered an absolute contraindication to transplantation by 81% and a relative contraindication by the other 19%. When asked for their
protocol to manage a patient who resumed drinking while on the waiting list, 15% of centers would remove him or her from the list permanently,
whereas 47% would re-evaluate the patient after a further period of monitored abstinence lasting 6 months.
A more nuanced approach to the estimation of prognosis in alcoholism is drawn from the alcohol addiction literature. Vaillant analyzed a 12-year
prospective study of 100 alcoholic subjects and developed an alternative instrument to assess future alcohol use among alcoholics. He defined four
factors elicited from the patient's history that appeared to predict a favourable chance for establishing long-term sobriety [34] (Table 2) (Table Not
Available) . Social stability is important in any condition
TABLE 2 -- VAILLANT'S PROGNOSTIC FACTORS FOR LONG-TERM SOBRIETY
Modified from Van Thiel DH, Bonet H, Gavalar V, et al: Effect of alcohol on allograft rejection
rates after liver transplantation for alcoholic liver disease. Alcohol Clin Exp Res 19:1151-1155,
1995; with permission.
(Not Available)
844
requiring long-term compliance with a complex medical regimen. Strauss and Bacon [6] demonstrated that lack of social stability has been linked
with relapse from sobriety. Based on the work of Vaillent, Strauss and Bacon, and his own observations, Beresford [26] developed the Michigan
Alcoholism Prognosis Scale (MAPS) for Major Organ Transplant Candidates (Table 3) . [26] He did not advocate an absolute threshold value to
support placement on the transplant waiting list, but a higher score is more favourable. The MAPS is a useful instrument to guide the psychiatrist
or addiction specialist in making a judgment of a patient's prognosis.
The role of the psychiatrist, however, is not to confined to attempting to predict alcohol relapse. He or she can identify and manage premorbid
psychiatric disease, a risk factor for postoperative morbidity. [32] He or she can assist in the treatment of the alcohol abuse before and after OLT.
The psychiatrist can help distinguish between hepatic encephalopathy (an indication for OLT) and fixed organic brain syndromes that are a
contraindication to OLT. [14] [18]


Outcome
Part of the resistance to providing transplants to patients with ALD is a perception that they do less well post-OLT than patients receiving
transplants for other reasons. Review of the available data does not support this perception.
Monitoring the Posttransplant Patient
Several weeks after surgery, routine monitoring of the posttransplant patient consists of frequent--typically monthly--clinic visits for history and
physical examination and biochemical assays of liver enzyme, renal function, and immunosuppression
TABLE 3 -- MICHIGAN ALCOHOLISM PROGNOSIS SCALE FOR MAJOR ORGAN
TRANSPLANT CANDIDATES
From Lucey MR, Merion RM, Henley KS, et al: Selection for and outcome of liver
transplantation in alcoholic liver disease. Gastroenterology 102:1736-1741, 1992; with
permission.
Characteristic Points
Acceptance of Alcoholism
Patient and family 4
Patient only 3
Family only 2
Neither 1
Prognostic indices Yes No
Substitute activities 3 1
Behavioral 3 1
Consequences
Hope/self-esteem 3 1
Social relationship 3 1
Social stability
Steady job 1
Stable residence 1
Does not live alone 1
Stable marriage 1
845
levels. The use of a flow chart to follow trends in apparently minor changes in liver chemistries and to assist in immunosuppression dosing is
invaluable. Many centers perform "protocol" liver biopsies at specified times post-OLT. At the authors' centre liver biopsies are done based on

clinical indications, usually, rising liver enzymes.
Mortality
The Scientific Liver Transplant Registry of UNOS has information on more than 16,000 adult OLTs performed between 1988 and 1995. [1A] [3]
According to its data, the patient and graft survival rates for liver transplants performed for ALD are comparable to those for non-ALD transplant
recipients, with 1- and 5-year alcoholic patient survivals of approximately 80% and 70% respectively (Table 4) . More detailed information from
the National Institute of Diabetes and Digestive and Kidney Diseases Liver Transplantation Database [38] on 1346 patients followed over 3 years
indicated similar graft survival but slightly decreased patient survival in the ALD compared with non-ALD patients. The ALD patients tended to
be more ill than the non-ALD patients preoperatively (Child's C cirrhosis was present in 44%, compared with 32% of the nonalcoholic recipients).
This was particularly true for ALD patients with fewer than 6 months abstinence prior to OLT, 62% of whom had Child's C cirrhosis. The
retransplant rate was significantly less for the ALD group than the non-ALD group, (3% versus 9%; P=0.04). These latter two facts suggest that
the decrease in ALD survival may have been attributable to a reluctance to retransplant alcoholic patients whose first graft failed. Several other
small or short-term studies [11] [20] [29] [30] 31a show survival rates in patients receiving transplants for ALD comparable to those with transplants for
other indications.
Alcohol Use After Liver Transplant
There is no fail-safe method to detect alcohol relapse postoperatively. Relapse is often detected by simple questioning of the patient and family,
investigating
TABLE 4 -- ONE- AND FIVE-YEAR POST-LIVER TRANSPLANT PATIENT SURVIVAL
BY DIAGNOSIS (UNOS TRANSPLANTS 1988-1996)
From the 1997 Annual Report of the US Scientific Registry for Transplant Recipients and the
Organ Procurement and Transplantation Network--Transplant Data: 1988-1996. UNOS,
Richmond, VA; and the Division of Transplantation, Office of Special Programs, Health
Resources and Services Administration, US Department of Health and Human Services,
Rockville, MD.
1-Year Number 5-Year Number
Diagnosis Survival Patients Survival Patients
Acute hepatic necrosis 72.0 1403 65.4 7105
Alcoholic liver disease 81.9 3063 67.6 1561
alpha1 -Anti-trypsin 87.3 444 82.6 208
deficiency
Autoimmune hepatitis 83.5 907 77.2 386
Budd-Chiari syndrome 84.4 183 80.2 89
Chronic viral hepatitis 80.3 4267 65.3 2102
Hemochromatosis 73.8 170 58.3 89
Hepatocellular carcinoma 68.1 521 34.5 355
Primary biliary cirrhosis 85.8 1726 79.4 860
Primary sclerosing 87.0 1601 76.2 776
cholangitis
All diagnoses 81.3 14771 68.4 7141

846
elevated liver enzymes, or following up on missed appointments. Carbohydrate-deficient transferrin levels (CDT) are reported to be more sensitive
and specific than liver enzymes as markers of sustained alcohol consumption. [15] The value of CDT levels after liver transplant remains to be
demonstrated. Despite intensive monitoring efforts, patients may still relapse unbeknown to the transplant team until they present with
life-threatening complications of their alcohol use. [9] Most studies of alcohol use after liver transplantation define alcohol relapse as "any
drinking." This definition is in contrast to studies on alcoholism treatment, which tend to distinguish between "slips" and "pathological" drinking,
and regard a reduction in alcohol use as a legitimate goal. Furthermore, there is little consistency in nomenclature regarding alcohol use after
transplantation. Terms such as recidivism, active alcoholism, and harmful or pathologic drinking abound without careful definition.
Reports of alcohol use by alcoholics after liver transplantation are shown in (Table 5) , [25] which summarizes eight reports that follow 288 of 389
patients transplanted for ALD and followed for up to 11 years. All reports defined relapse as the recognition of any alcohol use. Furthermore, most
relied on retrospective methodology and patient self report to capture the frequency of return to drinking behavior. Other than one report of 95%
relapse, the relapse rate (any alcohol use) was 12% to 34%, with the higher figure in the studies of greatest duration. These numbers compare very
favorably with 60% to 95% relapse rates of conventional treatment programs for alcohol abuse. [35]
Despite the overall good prognosis following transplantation in alcoholic patients, prognosis is more guarded in the individuals who return to
pathologic drinking. UNOS data documented graft loss caused by recurrence of ALD to be 0.4/1000 patient years compared with 1.2/1000 patient
years for non-ALD patients. [3] The significance of these data is difficult to judge because of the manner in which the data are collected. Other
evidence of the injurious effects of a relapse to pathologic drinking derives from anecdotal accounts. Four of 6 post-OLT alcoholic patients at the
University of Colorado for example, died after relapse to pathologic drinking. [11] The remaining two patients have managed to become abstinent
again. In the Michigan cohort, [25] 6 of 17 relapsers (any alcohol use) had medical complications consequent to their drinking, including
pancreatitis, pneumonia, cellulitis, delerium tremens, and graft loss. One patient who was noncompliant with her immunosuppression while
drinking excessively developed rejection and died.
Wiesner [38] compared 139 ALD and 486 non-ALD patients in the National Institutes of Health liver transplant database. He found that ALD
patients were more likely to have prolonged intensive care unit stays and increased blood product requirements. They were also more ill at the
time of OLT, however, with 44% having Child's C cirrhosis at the time of transplant, as opposed to 32% in the non-ALD group. At least two
groups have reported a decreased incidence of acute cellular rejection in the alcoholic liver transplant recipients compared with nonalcoholic
patients. [36] [38] It is unlikely that this effect is caused by recent or concurrent alcohol consumption, especially because more than 80% of acute
cellular rejection episodes occur in the first 8 weeks after liver transplantation. This time interval is considerably sooner than the interval from
surgery to resumption of alcohol among the minority of alcoholics who are found to return to alcohol use. More likely, the effect on immune
tolerance is the result of
847
TABLE 5 -- SUMMARY OF PUBLISHED DATA ON ALCOHOL USE BY ALCOHOLIC PATIENTS AFTER LIVER
TRANSPLANTATION
From Lucey MR, Carr K, Beresford TP, et al: Alcohol use after transplantation in alcoholics: A clinical cohort follow-up study. Hepatology
25:1223-1227, 1997; with permission.

No. of No. of
Alcoholics Alcoholics
Receiving Receiving No. of 1-Yr
Transplants Transplants Alcoholics Survival 5-Yr Duration
Study Alcoholism (Study (Not Study Not Rate (% Survival of Relapse Relapse
Author Years Prognosis Cohort) Cohort) Selected ) Rate Follow-up Defined Frequency
Kumar [5] 1982-1988 Abstinent 52 21 Not stated 74 Not Not stated Any use 12
period not stated (phone
required survery)
Bird [22] 1980-1989 Abstinent 18 6 Not stated 66 Not 4 Any use 17
period not stated months-7 (laboratory
required years values,
biopsy)
Knechtle [6] 1984-1990 Abstinent 32 9 5 83 71% Not Any use 13
period not stated (psychiatric
required interview)
Gish [7] 1988-1991 Abstinent 29 0 Not stated 93 Not 24 Any use 21
period not stated months (prospective
required follow-up)
Osorio [8] 1988-1991 Six months 43 0 65 100 Not 21 Any use 19
abstinence stated months (mail
survery)
Berlakovich 1982-1993 Abstinent 44 36 Not stated 71 63% 78 Any use 32
[9] period not months (clinic
required follow-up)
Lucey* 1987-1991 Abstinent 50 9 75 80 77% 63 Any use 34
period not months (see text)
required
Howard [23] 1987-1992 Abstinent 20 20 Not stated 79 Not 34 Any use 95
period not stated months (psychiatric
required interview)
848
drinking prior to transplantation. Employment rates also are similar post-OLT in ALD and non-ALD groups. [20] [21] [38] Interestingly, the
perception of quality of life was not the same in alcoholic and nonalcoholic graft recipients in all studies. In two studies (one of 139 patients
followed for 3 years, [13] and another of 20 patients followed for 1-6 years [17] ), post-OLT alcoholic patients had a perception of worse health or
quality of life compared with post-OLT nonalcoholic patients. One study (30 patients followed for up to 6 years) [21] showed increased perception
of health compared with nonALD patients. In all of these studies, the vast majority of patients describe their postransplant quality of life as much
better than that pretransplant. [11]

FUTURE DIRECTIONS
The Role of Naltrexone
Two medications are approved for the treatment of alcohol dependence-- disulfiram, which was approved in 1951 and is limited by a 20%
compliance rate, and naltrexone, approved in the mid 1990s. Naltrexone is a pure opioid antagonist. Although it did not prevent subjects from
"sampling" alcohol, it decreased craving and, more important, the rate of relapse to problem drinking in those who did sample alcohol--81% (of 21
placebo samplers) versus 48% (of 21 naltrexone samplers). [37] Although various psychosocial therapies have not been shown to be particularly
useful alone, they do help in conjunction with naltrexone. [28] [37] It is unclear whether the efficacy of naltrexone is different between groups at high
or low risk of relapse. A study to assess the role of naltrexone in the post-OLT patient is being undertaken at the authors' center.
CONCLUSION
The outcome for liver transplantation in selected alcoholic patients is comparable to that of nonalcoholic patients. There is no justification for
denying these patients an OLT on the basis of their diagnosis.
Evaluation of candidates for liver transplantation remains an inexact process. The status of candidates being evaluated needs to be updated
regularly and the selection processes themselves, as used by various centers, need to be evaluated.
In addition to the actual OLT itself, the transplant team and setting provide structure and support to the alcoholic and may assist the transplanted
alcoholic maintain long-term sobriety.
References
1. Alexander JF, Lischner MW, Galambos JT: Natural history of alcoholic hepatitis: The long-term prognosis. Am J Gastroenterol 515-525, 1971
1A. Anonymous. 1997 Annual Report of the US Scientific Registry for Transplant Recipients and the Organ Procurement and Transplantation Network--Transplant Data:
1988-1996. UNOS, Richmond, VA; and the Division of Transplantation, Office of Special Programs, Health Resources and Services Administration, US Department of
Health and Human Services, Rockville, MD
849
2. Becker U, Deis A, Sorensen TIA, et al: Prediction of risk of liver disease by alcohol intake, sex and age: A prospective population study. Hepatology 23:1025-1029,
1996
3. Belle SH, Beringer KC, Detre KM: Liver transplantation for alcoholic liver disease in the United States: 1988 to 1995. Liver Transplantation and Surgery 3:212-219,
1997
4. Bellentani S, Saccoccio G, Costa G, et al: Drinking habits as cofactors of risk for alcohol induced liver damage. Gut 41:845-850, 1997
5. Beresford TP: Overt and covert alcoholism. In Lucey MR, Merion RM, Beresford TP (eds): Liver Transplantation and the Alcoholic Patient: Medical, Surgical and
Psychological Issues. Cambridge, England, Cambridge University Press, 1994, pp 6-28
6. Beresford TP: Psychiatric assessment of alcoholic candidates for liver transplantation. In Lucey MR, Merion RM, Beresford TP (eds): Liver transplantation and the
alcoholic patient. Cambridge, England, Cambridge University Press, 1994, pp 29-49

7. Beresford TP: Predictive factors for alcoholic relapse in the selection of alcohol-dependent persons for hepatic transplant. Liver Transplantation and Surgery 3:280-291,
1997
8. Berkalovich GA, Steininger R, Herbst, F, et al: Efficacy of liver transplantation for alcoholic cirrhosis with respect to recidivism and compliance. Transplantation 58:
560-565, 1994
9. Campbell DA, Punch JD: Monitoring for alcohol use relapse after liver transplantation for alcoholic liver disease. Liver Transplantation and Surgery 3:300-303, 1997
9A. Bird GLA, O'Grady JG, Harvey FAK, et al: Liver transplantation in patients with alcoholic cirrhosis: Selection criteria and rates of survival and relapse. BMJ
301:15-17, 1990
10. Everhart JE, Beresford TP: Liver transplantation for alcoholic liver disease: A survey of transplantation programs in the United States. Liver Transplantation and
Surgery 3:220-226, 1997
11. Everson GT, Bharadhwaj G, House R, et al: Long-term follow-up of patients with alcoholic liver disease who underwent hepatic transplantation. Liver Transplantation
and Surgery 3:263-274, 1997
12. Fattovich G, Giustina G, Degos F, et al: Morbidity and mortality in compensated cirrhosis type C: A retrospective follow-up study of 384 patients. Gastroenterology
112:463-472, 1997
13. Gerhardt TC, Goldstein RM, Urschel HC, et al: Alcohol use following liver transplantation for alcoholic cirrhosis. Transplantation 62:1060-1063, 1996
13A. Gish RL, Lee AN, Keeffe EB, et al: Liver transplantation for patients with alcoholism and end-stage liver disease. American Journal of Gastroenterology
88:1337-1342, 1995
14. Hans P, Gutheil TG: Liver transplantation and the disliked patient: A clinical and ethical dilemma. Clin Transplant 5:277-281, 1991
15. Helander A, Tabakoff B, WHO/ISBRA Study Centres: Biochemical markers of alcohol use and abuse: Experiences from the pilot study of the WHO/ISBRA
collaborative project on state and trait markers of alcohol. Alcohol Alcohol 32:133-144, 1997
16. Hoofnagle JH, Kresima T, Fuller RK, et al: Liver transplantation for alcoholic liver disease: Executive statement and recommendations. Liver Transplantation and
Surgery 3:347-350, 1997
17. Howard LM, Fahy TA, Wong P, et al: Psychiatric outcome in alcoholic liver transplant patients. QJM 87:731-736, 1994
18. Howard LM, Williams R, Fahy TA: The psychiatric assessment of liver transplant patients with alcoholic liver disease: A review. J Psychosom Res 38:643-653, 1994
19. Kluge E: Drawing the ethical line between organ transplantation and lifestyle abuse. Canadian Medical Association Journal 150:745-746, 1994
20. Knechtle SJ, Flemming MF, Barry KL, et al: Liver transplantation for alcoholic liver disease. Surgery 112:694-703, 1992
21. Knechtle SJ, Flemming MF, Barry KL, et al: Liver transplantation in alcoholics: Assessment of psychological health and work activity. Transplant Proc 25:1916-1918,
1993
22. Kumar S, Stauber RE, Gavaler JS, et al: Orthotopic liver transplantation for alcoholic liver disease. Hepatology 11:159-164, 1990
23. Leshner AI: Addiction is a brain disease, and it matters. Science 278:45-47, 1997
850
24. Lucey
MR, Brown KA, Everson GT, et al: Minimal criteria for placement of adults on the liver transplant waiting list: A report of a national conference organized by
the American Society of Transplant Physicians and the American Association for the Study of Liver Diseases. Liver Transplantation and Surgery 3:628-637, 1997

25. Lucey MR, Carr K, Beresford TP, et al: Alcohol use after transplantation in alcoholics: A clinical cohort follow-up study. Hepatology 25:1223-1227, 1997
26. Lucey MR, Merion RM, Henely KS, et al: Selection for and outcome of liver transplantation in alcoholic liver disease. Gastroenterology 102:1736-1741, 1992
27. Moss AH, Siegler M: Should alcoholics compete equally for liver transplantation? JAMA 265:1295-1298, 1992
28. O'Malley SS, Jaffe AJ, Chang G, et al: Naltrexone and coping skills therapy for alcohol dependence. Arch Gen Psychiatr 49:881-887, 1992
29. Osorio RW, Ascher NL, Avery M, et al: Predicting recidivism after orthoptic liver transplantation for alcoholic liver disease. Hepatology 20:105-110, 1996
30. Pereira SP, Williams R: Liver transplantation for alcoholic liver disease at King's College Hospital: Survival and quality of life. Liver Transplantation and Surgery
3:245-250, 1997
31. Starzl TE, Van Thiel DH, Tzakis AG, et al: Orthotopic liver transplantation for alcoholic cirrhosis. JAMA 260:2542-2544, 1988
32. Surman OS: Psychiatric aspects of organ transplantation. Am J Psychiatr 146:972-982, 1989
33. United Network for Organ Sharing Web site: UNOS Web site. http/www.UNOS.org
34. Vaillant GE: What can long-term follow-up teach us about relapse and prevention of relapse in addiction? British Journal of Addiction 83:1147-1157, 1983
35. Vaillant GE: The natural history of alcoholism and its relationship to liver transplantation. Liver Transplantation and Surgery 3:304-310, 1997
36. Van Thiel DH, Bonet H, Gavalar J, et al: Effect of alcohol on allograft rejection rates after liver transplantation for alcoholic liver disease. Alcohol Clin Exp
19:1151-1155, 1995
37. Volpicelli JR, Volpicelli LA, O'Brien CP: Medical management of alcohol dependence: Clinical use and limitations of naltrexone treatment. Alcohol Alcohol
30:789-798, 1995
38. Wiesner RH, Lombardero M, Lake JR, et al: Liver transplantation for end-stage alcoholic liver disease: An assessment of outcomes. Liver Transplantation and Surgery
3:231-239, 1997


839
Anne Burke 1 MB, BCh, BAO, MRCPI

Michael R. Lucey 1 2 MD, FRCPI
1 Division of Gastroenterology (AB, MRL)

2 Liver Transplant Program (MRL), Hospital of the University of Pennsylvania Philadelphia, Pennsylvania

Michael R. Lucey, MD, RCPI
Division of Gastroenterology
Hospital of the University of Pennsylvania
Ravdin 3, 3400 Spruce Street
Philadephia, PA 19104
Healing is a
matter of time, but it is sometimes also a matter of opportunity.
HIPPOCRATES, 460-400 BC
In February 1998, 9823 patients in the United States were awaiting an orthotopic liver transplant (OLT). [1A] Among candidates
with a recorded diagnosis, 1212 (16.7%) had alcoholic liver disease (ALD). Only hepatitis C was more common among the
diagnoses on the waiting list (1639 or 22.8%). In addition, 524 (7.3%) patients carried both diagnoses. During 1996, 4062 liver
transplants were carried out in the United States. In the same year, 960 patients died on the OLT waiting list. Although there has
been a four-fold increase in the retrieval of organs considered suitable for use during the past 8 years, there has been a 10-fold
growth both in the number of people waiting for OLT and in the number of people dying on the OLT waiting list during the same
period. The discrepancy between the number of available donor organs and the potential recipients has led to a continuing debate
about the appropriateness of using a scarce donor liver to treat what is, in the estimation of some, a self-inflicted malady. [19] [27]
This article discusses the role of OLT in the treatment of the patient with alcoholic liver disease (ALD), the selection of patients with
ALD for OLT, and their outcome after liver transplantation. We set our discussion in the context of the ethical issues involved in
this controversial area.
ALCOHOL USE IN CONTEMPORARY AMERICAN SOCIETY

Alcohol use is very widespread in our society. Sixty-eight percent of adult Americans drink at least one beer a month. [16] Alcoholism,
which is a pathologic
840
loss of control over alcohol use, affects some 7% to 10% of Americans at some point in their lives. [5] Genetic susceptibility and diverse
social and cultural pressures contribute to the risk of alcoholism. There is considerable debate about the underlying mechanisms that cause
alcoholism. At one extreme, there is the personal responsibility school of thought that interprets alcoholism as a behavioral fault within an

individual. On the other hand, there is a considerable body of evidence to show that addiction is a chronic relapsing brain disease, in part
mediated by a genetic propensity, wherein drug use leads to changes in brain structure and function that persist long after the individual
stops taking the drug. [23] The fact that the majority of alcohol abusers do not develop ALD suggests that there are other environmental or
genetic factors involved in the pathogenesis of alcohol-induced liver damage. Putative cofactors include the gender of the individual,
hereditary disorders, such as hemachromatosis, and acquired factors, such as viral hepatitis. The emerging evidence that ALD is the result
of a combination of pathogenic influences, of which alcohol use is a required but not the sole factor, challenges the more facile formulation
that places the responsibility for developing ALD on the alcoholic.
Similarly, among persons who drink, there is a wide variation in susceptibility to ALD. The threshold of ethanol consumption that
facilitates development of ALD seems to be as little as 30 g/d (three drinks) [4] or 14 drinks per week. [2] The risk of developing liver disease
increases with increasing consumption. Nevertheless, only a minority (13.5%) of patients drinking in excess of 120 g/d (12 drinks) manifest
features of ALD. Despite the fact that fewer than 15% of alcoholics develop ALD, it is estimated that ALD affects between 1 and 2 million
Americans at any time. [5] Even though many of these patients are unsuitable for liver transplantation, either because they have not
progressed to end-stage liver failure, or because of concomitant confounding ailments (see The section on selecting ALD patients for OLT),
it is easy to understand how ALD could overwhelm the available donor resources.
SELECTING ALCOHOLIC LIVER DISEASE PATIENTS FOR ORGAN LIVER TRANSPLANT

The process of evaluating a patient with end-stage liver disease of any cause for liver transplantation can be reduced to three questions:
Does the patient have severe liver disease that warrants an OLT?
Does the patient have other medical, surgical, or psychosocial circumstances that would confound a successful OLT?
What are the patient's wishes regarding OLT?
In the program at the Hospital of the University of Pennsylvania, the authors have adopted a multidisciplinary team approach to answer
these questions. Each patientr receives an individualized assessment. Among alcoholic patients, we therefore take special care to assess the
patient's psychosocial background and support structures, which are crucial to maintaining sobriety and the health of the graft after
transplant.
LISTING THE PATIENT

The immediate end point of the transplant evaluation process is determining whether a prospective candidate should be placed on the liver
transplant waiting
841
list, often called listing the patient. At the present time, in the United States, donor livers are allocated to patients on the liver transplant
waiting list according to a points priority scheme. Points are given for blood-type (ABO) compatibility, time on the waiting list, and
severity of liver failure. In recent years, as intervals spent on the waiting list have extended inexorably, estimating prognosis has become
more difficult than ever before. Transplant programs are faced with the dilemma of estimating a candidate's prognosis while factoring in an
uncertain period of pretransplant management on the waiting list.
Because time on the waiting list is recognized as a priority factor and the duration on the waiting list is uncertain, there is a tendency for
placement of stable patients on the list, a practice referred to as list inflation. To establish a degree of fairness and uniformity for entry onto
the waiting list, guidelines for minimal listing criteria were published in 1997. [24] The basis of these criteria is an expected 1-year survival
of less than 90%. The survival rate of patients with cirrhosis is related to the Child-Pugh score (Table 1) . The estimated 1- and 5-year
survival rates for Child's C patients are 95% and 75%, respectively. Predicted survival deteriorates considerably when cirrhosis becomes
decompensated, as indicated by the onset of any of the following clinical phenomena--ascites, variceal hemorrhage, jaundice,
encephalopathy, or hepatorenal syndrome. A Child's score of 7 or above or evidence of decompensation are indications for listing for OLT.
[24] Patients on the list need to be reviewed regularly to follow their course and modify their status on the waiting list as necessary.
ALD poses particular problems with regard to estimating prognosis. Many patients with alcoholic cirrhosis who deteriorate do so because
of the acute effects of alcohol on liver function. Furthermore, it is well documented that liver function can improve in acute alcoholic
hepatitis with abstinence. [1] This occurs even when the patient has already established cirrhosis. This is the basis of discriminating against
alcoholics who have been drinking in the previous 6 months, as is discussed further in the section on psychosocial assessment.

MEDICAL EVALUATION FOR LIVER TRANSPLANTATION

As liver transplantation has become more sophisticated, the list of absolute contraindications to placement on the waiting list has shortened.
Significant impediments to successful OLT include extrahepatic malignant disease, cholangiocarcinoma, or advanced hepatocellular
carcinoma; ongoing drug or alcohol use; active infection; significant organic brain injury or major intractable psychoses; or significant
cardiopulmonary disease. Infection by human immunodeficiency virus (HIV), with or without acquired immune deficiency syndrome, used
TABLE 1 -- CHILD-PUGH SCORE
Points
Criteria 1 2 3
Encephalopathy (grade) None 1-2 3-4
Ascites None Slight Moderate-severe
Bilirubin (mg/dL) 1-2 2-3 >3
Albumin (g/dL) >3.5 2.8-3.5 <2.8
Prothrombin time (seconds prolonged) 1-4 4-10 >10
Grading: Score 6 points = grade A; 7-9 points = grade B; 10 points = grade C.
842
to be a contraindication but this status has become less certain with the advent of more effective antiviral agents. Similarly, extensive
thrombosis of the portal and mesenteric venous system may no longer preclude liver transplantation. The impact of relative
contraindications therefore change with advances in medical practice.
Medical screening for alcoholic liver transplant candidates begins with history and physical examination, routine hematologic and
biochemical screening, electrocardiogram, and chest radiograph. HIV serology is checked in all candidates irrespective of underlying
diagnosis. In view of the increased risk of hepatoma in cirrhotic patients, all OLT candidates have abdominal cross-sectional imaging
(sonography, MR imaging or CT scanning as appropriate) and serum alpha-fetoprotein determination. Comorbid liver diseases should also
be checked for by testing serology for hepatitis B and C infection, iron studies (ferritin and iron saturation) for hemachromatosis, and alpha1
-antitrysin levels. Patients with ALD are a fairly young group compared with those with other causes of liver disease. In those fewer than 35
years of age, it is prudent to check ceruloplasmin or urinary copper to rule out Wilson's disease. The vast majority of patients will have had
a liver biopsy to diagnose cause and severity of their liver injury during the course of their disease. There are also a few conditions that are
more likely to occur in the alcohol-abusing patient--cardiomyopathy, organic brain syndromes, peripheral and autonomic neuropathy,
chronic pancreatitis, tuberculosis, and osteopenia. History and physical examination should be directed toward these areas. The following
investigations should be done on case-by-case basis: echo cardiography, arterial blood gases, pulmonary function tests, head CT image,
nerve conduction studies, endoscopic retrograde cholangiopancreatography, and bone densitometry. All alcoholic candidates should receive
purified protein derivative (PPD) or Mantoux testing. Malnutrition is common among ALD patients. Formal nutritional assessment and
counseling by a dietician can be helpful.
PSYCHOSOCIAL EVALUATION AND PREDICTING ALCOHOLIC RELAPSE

Medical screening is not an exact science and places patients at high, moderate, or low risk for postoperative complications. Similarly,
psychosocial evaluation stratifies patients by their risk of relapse for alcoholism. A recent survey indicated that most (66 of 68 surveyed)
transplant programs in the United States give preference to alcoholic patients whom they assess to be more likely to remain abstinent from
alcohol after transplantation. It is much less certain how best to achieve the goal of predicting future drinking.
Most transplant programs in the United States require alcoholic patients to undergo formal psychiatric evaluation. [10] Furthermore, several
payers require psychiatric evaluation of an alcoholic patient before they will approve payment. Eighty-five percent of OLT centres surveyed
make use of the "6-month rule," which requires that an alcoholic patient must have demonstrated 6 months of abstinence before being
considered for OLT. [10] Almost half have had third-party payers refuse to pay for an OLT because of failure to complete a set period of
abstinence. The accuracy of predicting post-OLT sobriety based on short duration of pretransplant sobriety is questionable.

Most OLT centers in the United States demand a 6-month abstinence period before they will transplant a patient for ALD. [10] This 6-month
period is based on a 1990 report [22] suggesting more frequent rates of relapse among alcoholic OLT recipients who had been abstinent for
fewer than 6 months before OLT.
843
The numbers of subjects involved, however, were too small for the statistical method used. Other follow-up data with greater numbers have
found no significant difference in relapse based on a 6-month threshold of pre-OLT abstinence. [7] [8] [11] [13] [20] [30] [35] [38]
A 6-month period of abstinence was proposed for minimal listing criteria for ALD, although the suggestion was rejected by the United
Network for Organ Sharing (UNOS). [20] The participants at the conference on minimal listing criteria noted that adherence to the 6-month
abstinence allows the inflammatory effects of recent alcohol consumption to resolve and exclude alcoholic hepatitis patients from OLT.
Many authors also agree that exceptional patients with fewer than 6 months abstinence could be considered for OLT. Many programs (75%
of OLT centers surveyed) ask the ALD patient to sign a contract to undergo alcohol rehabilitation. [10] Success with this approach is mixed
and the authors do not use it at their center.
At present, alcoholic hepatitis is a contraindication to liver transplantation. In a survey of 69 OLT centers in the United States, [10] active
alcohol use was considered an absolute contraindication to transplantation by 81% and a relative contraindication by the other 19%. When
asked for their protocol to manage a patient who resumed drinking while on the waiting list, 15% of centers would remove him or her from
the list permanently, whereas 47% would re-evaluate the patient after a further period of monitored abstinence lasting 6 months.
A more nuanced approach to the estimation of prognosis in alcoholism is drawn from the alcohol addiction literature. Vaillant analyzed a
12-year prospective study of 100 alcoholic subjects and developed an alternative instrument to assess future alcohol use among alcoholics.
He defined four factors elicited from the patient's history that appeared to predict a favourable chance for establishing long-term sobriety [34]
(Table 2) (Table Not Available) . Social stability is important in any condition
TABLE 2 -- VAILLANT'S PROGNOSTIC FACTORS FOR LONG-TERM SOBRIETY
Modified from Van Thiel DH, Bonet H, Gavalar V, et al: Effect of alcohol on allograft rejection rates
after liver transplantation for alcoholic liver disease. Alcohol Clin Exp Res 19:1151-1155, 1995; with
permission.
(Not Available)
844
requiring long-term compliance with a complex medical regimen. Strauss and Bacon [6] demonstrated that lack of social stability has been
linked with relapse from sobriety. Based on the work of Vaillent, Strauss and Bacon, and his own observations, Beresford [26] developed the
Michigan Alcoholism Prognosis Scale (MAPS) for Major Organ Transplant Candidates (Table 3) . [26] He did not advocate an absolute
threshold value to support placement on the transplant waiting list, but a higher score is more favourable. The MAPS is a useful instrument
to guide the psychiatrist or addiction specialist in making a judgment of a patient's prognosis.
The role of the psychiatrist, however, is not to confined to attempting to predict alcohol relapse. He or she can identify and manage
premorbid psychiatric disease, a risk factor for postoperative morbidity. [32] He or she can assist in the treatment of the alcohol abuse before
and after OLT. The psychiatrist can help distinguish between hepatic encephalopathy (an indication for OLT) and fixed organic brain
syndromes that are a contraindication to OLT. [14] [18]

Outcome
Part of the resistance to providing transplants to patients with ALD is a perception that they do less well post-OLT than patients receiving
transplants for other reasons. Review of the available data does not support this perception.
Monitoring the Posttransplant Patient
Several weeks after surgery, routine monitoring of the posttransplant patient consists of frequent--typically monthly--clinic visits for history
and physical examination and biochemical assays of liver enzyme, renal function, and immunosuppression
TABLE 3 -- MICHIGAN ALCOHOLISM PROGNOSIS SCALE FOR MAJOR ORGAN

TRANSPLANT CANDIDATES
From Lucey MR, Merion RM, Henley KS, et al: Selection for and outcome of liver transplantation in
alcoholic liver disease. Gastroenterology 102:1736-1741, 1992; with permission.
Characteristic Points
Acceptance of Alcoholism
Patient and family 4
Patient only 3
Family only 2
Neither 1
Prognostic indices Yes No
Substitute activities 3 1
Behavioral 3 1
Consequences
Hope/self-esteem 3 1
Social relationship 3 1
Social stability
Steady job 1
Stable residence 1
Does not live alone 1
Stable marriage 1
845
levels. The use of a flow chart to follow trends in apparently minor changes in liver chemistries and to assist in immunosuppression dosing
is invaluable. Many centers perform "protocol" liver biopsies at specified times post-OLT. At the authors' centre liver biopsies are done
based on clinical indications, usually, rising liver enzymes.
Mortality
The Scientific Liver Transplant Registry of UNOS has information on more than 16,000 adult OLTs performed between 1988 and 1995. [1A]
[3] According to its data, the patient and graft survival rates for liver transplants performed for ALD are comparable to those for non-ALD
transplant recipients, with 1- and 5-year alcoholic patient survivals of approximately 80% and 70% respectively (Table 4) . More detailed
information from the National Institute of Diabetes and Digestive and Kidney Diseases Liver Transplantation Database [38] on 1346 patients
followed over 3 years indicated similar graft survival but slightly decreased patient survival in the ALD compared with non-ALD patients.
The ALD patients tended to be more ill than the non-ALD patients preoperatively (Child's C cirrhosis was present in 44%, compared with
32% of the nonalcoholic recipients). This was particularly true for ALD patients with fewer than 6 months abstinence prior to OLT, 62% of
whom had Child's C cirrhosis. The retransplant rate was significantly less for the ALD group than the non-ALD group, (3% versus 9%;
P=0.04). These latter two facts suggest that the decrease in ALD survival may have been attributable to a reluctance to retransplant
alcoholic patients whose first graft failed. Several other small or short-term studies [11] [20] [29] [30] 31a show survival rates in patients receiving
transplants for ALD comparable to those with transplants for other indications.
Alcohol Use After Liver Transplant
There is no fail-safe method to detect alcohol relapse postoperatively. Relapse is often detected by simple questioning of the patient and
family, investigating
TABLE 4 -- ONE- AND FIVE-YEAR POST-LIVER TRANSPLANT PATIENT SURVIVAL BY
DIAGNOSIS (UNOS TRANSPLANTS 1988-1996)
From the 1997 Annual Report of the US Scientific Registry for Transplant Recipients and the Organ
Procurement and Transplantation Network--Transplant Data: 1988-1996. UNOS, Richmond, VA; and
the Division of Transplantation, Office of Special Programs, Health Resources and Services

Administration, US Department of Health and Human Services, Rockville, MD.
1-Year Number 5-Year Number
Diagnosis Survival Patients Survival Patients
Acute hepatic necrosis 72.0 1403 65.4 7105
Alcoholic liver disease 81.9 3063 67.6 1561
alpha1 -Anti-trypsin deficiency 87.3 444 82.6 208
Autoimmune hepatitis 83.5 907 77.2 386
Budd-Chiari syndrome 84.4 183 80.2 89
Chronic viral hepatitis 80.3 4267 65.3 2102
Hemochromatosis 73.8 170 58.3 89
Hepatocellular carcinoma 68.1 521 34.5 355
Primary biliary cirrhosis 85.8 1726 79.4 860
Primary sclerosing cholangitis 87.0 1601 76.2 776
All diagnoses 81.3 14771 68.4 7141
846
elevated liver enzymes, or following up on missed appointments. Carbohydrate-deficient transferrin levels (CDT) are reported to be more
sensitive and specific than liver enzymes as markers of sustained alcohol consumption. [15] The value of CDT levels after liver transplant
remains to be demonstrated. Despite intensive monitoring efforts, patients may still relapse unbeknown to the transplant team until they
present with life-threatening complications of their alcohol use. [9] Most studies of alcohol use after liver transplantation define alcohol
relapse as "any drinking." This definition is in contrast to studies on alcoholism treatment, which tend to distinguish between "slips" and
"pathological" drinking, and regard a reduction in alcohol use as a legitimate goal. Furthermore, there is little consistency in nomenclature
regarding alcohol use after transplantation. Terms such as recidivism, active alcoholism, and harmful or pathologic drinking abound without
careful definition.
Reports of alcohol use by alcoholics after liver transplantation are shown in (Table 5) , [25] which summarizes eight reports that follow 288
of 389 patients transplanted for ALD and followed for up to 11 years. All reports defined relapse as the recognition of any alcohol use.
Furthermore, most relied on retrospective methodology and patient self report to capture the frequency of return to drinking behavior. Other
than one report of 95% relapse, the relapse rate (any alcohol use) was 12% to 34%, with the higher figure in the studies of greatest duration.
These numbers compare very favorably with 60% to 95% relapse rates of conventional treatment programs for alcohol abuse. [35]
Despite the overall good prognosis following transplantation in alcoholic patients, prognosis is more guarded in the individuals who return
to pathologic drinking. UNOS data documented graft loss caused by recurrence of ALD to be 0.4/1000 patient years compared with
1.2/1000 patient years for non-ALD patients. [3] The significance of these data is difficult to judge because of the manner in which the data
are collected. Other evidence of the injurious effects of a relapse to pathologic drinking derives from anecdotal accounts. Four of 6
post-OLT alcoholic patients at the University of Colorado for example, died after relapse to pathologic drinking. [11] The remaining two
patients have managed to become abstinent again. In the Michigan cohort, [25] 6 of 17 relapsers (any alcohol use) had medical complications
consequent to their drinking, including pancreatitis, pneumonia, cellulitis, delerium tremens, and graft loss. One patient who was
noncompliant with her immunosuppression while drinking excessively developed rejection and died.
Wiesner [38] compared 139 ALD and 486 non-ALD patients in the National Institutes of Health liver transplant database. He found that
ALD patients were more likely to have prolonged intensive care unit stays and increased blood product requirements. They were also more
ill at the time of OLT, however, with 44% having Child's C cirrhosis at the time of transplant, as opposed to 32% in the non-ALD group. At
least two groups have reported a decreased incidence of acute cellular rejection in the alcoholic liver transplant recipients compared with
nonalcoholic patients. [36] [38] It is unlikely that this effect is caused by recent or concurrent alcohol consumption, especially because more
than 80% of acute cellular rejection episodes occur in the first 8 weeks after liver transplantation. This time interval is considerably sooner
than the interval from surgery to resumption of alcohol among the minority of alcoholics who are found to return to alcohol use. More
likely, the effect on immune tolerance is the result of
847

TABLE 5 -- SUMMARY OF PUBLISHED DATA ON ALCOHOL USE BY ALCOHOLIC PATIENTS AFTER LIVER
TRANSPLANTATION
From Lucey MR, Carr K, Beresford TP, et al: Alcohol use after transplantation in alcoholics: A clinical cohort follow-up study. Hepatology
25:1223-1227, 1997; with permission.
No. of No. of
Alcoholics Alcoholics
Receiving Receiving No. of 1-Yr
Transplants Transplants Alcoholics Survival 5-Yr Duration
Study Alcoholism (Study (Not Study Not Rate (% Survival of Relapse Relapse
Author Years Prognosis Cohort) Cohort) Selected ) Rate Follow-up Defined Frequency
Kumar [5] 1982-1988 Abstinent 52 21 Not stated 74 Not Not stated Any use 12
period not stated (phone
required survery)
Bird [22] 1980-1989 Abstinent 18 6 Not stated 66 Not 4 Any use 17
period not stated months-7 (laboratory
required years values,
biopsy)
Knechtle [6] 1984-1990 Abstinent 32 9 5 83 71% Not Any use 13
period not stated (psychiatric
required interview)
Gish [7] 1988-1991 Abstinent 29 0 Not stated 93 Not 24 Any use 21
period not stated months (prospective
required follow-up)
Osorio [8] 1988-1991 Six months 43 0 65 100 Not 21 Any use 19
abstinence stated months (mail
survery)
Berlakovich 1982-1993 Abstinent 44 36 Not stated 71 63% 78 Any use 32
[9] period not months (clinic
required follow-up)
Lucey* 1987-1991 Abstinent 50 9 75 80 77% 63 Any use 34
period not months (see text)
required
Howard [23] 1987-1992 Abstinent 20 20 Not stated 79 Not 34 Any use 95
period not stated months (psychiatric
required interview)
848
drinking prior to transplantation. Employment rates also are similar post-OLT in ALD and non-ALD groups. [20] [21] [38] Interestingly, the
perception of quality of life was not the same in alcoholic and nonalcoholic graft recipients in all studies. In two studies (one of 139 patients
followed for 3 years, [13] and another of 20 patients followed for 1-6 years [17] ), post-OLT alcoholic patients had a perception of worse
health or quality of life compared with post-OLT nonalcoholic patients. One study (30 patients followed for up to 6 years) [21] showed
increased perception of health compared with nonALD patients. In all of these studies, the vast majority of patients describe their
postransplant quality of life as much better than that pretransplant. [11]
FUTURE DIRECTIONS
The Role of Naltrexone
Two medications are approved for the treatment of alcohol dependence-- disulfiram, which was approved in 1951 and is limited by a 20%
compliance rate, and naltrexone, approved in the mid 1990s. Naltrexone is a pure opioid antagonist. Although it did not prevent subjects
from "sampling" alcohol, it decreased craving and, more important, the rate of relapse to problem drinking in those who did sample

alcohol--81% (of 21 placebo samplers) versus 48% (of 21 naltrexone samplers). [37] Although various psychosocial therapies have not been
shown to be particularly useful alone, they do help in conjunction with naltrexone. [28] [37] It is unclear whether the efficacy of naltrexone is
different between groups at high or low risk of relapse. A study to assess the role of naltrexone in the post-OLT patient is being undertaken
at the authors' center.
CONCLUSION
The outcome for liver transplantation in selected alcoholic patients is comparable to that of nonalcoholic patients. There is no justification
for denying these patients an OLT on the basis of their diagnosis.
Evaluation of candidates for liver transplantation remains an inexact process. The status of candidates being evaluated needs to be updated
regularly and the selection processes themselves, as used by various centers, need to be evaluated.
In addition to the actual OLT itself, the transplant team and setting provide structure and support to the alcoholic and may assist the
transplanted alcoholic maintain long-term sobriety.
References
1. Alexander JF, Lischner MW, Galambos JT: Natural history of alcoholic hepatitis: The long-term prognosis. Am J Gastroenterol 515-525, 1971
1A. Anonymous. 1997 Annual Report of the US Scientific Registry for Transplant Recipients and the Organ Procurement and Transplantation Network--Transplant
Data: 1988-1996. UNOS, Richmond, VA; and the Division of Transplantation, Office of Special Programs, Health Resources and Services Administration, US
Department of Health and Human Services, Rockville, MD
849
2. Becker U, Deis A, Sorensen TIA, et al: Prediction of risk of liver disease by alcohol intake, sex and age: A prospective population study. Hepatology
23:1025-1029, 1996
3. Belle SH, Beringer KC, Detre KM: Liver transplantation for alcoholic liver disease in the United States: 1988 to 1995. Liver Transplantation and Surgery
3:212-219, 1997
4. Bellentani S, Saccoccio G, Costa G, et al: Drinking habits as cofactors of risk for alcohol induced liver damage. Gut 41:845-850, 1997
5. Beresford TP: Overt and covert alcoholism. In Lucey MR, Merion RM, Beresford TP (eds): Liver Transplantation and the Alcoholic Patient: Medical, Surgical
and Psychological Issues. Cambridge, England, Cambridge University Press, 1994, pp 6-28
6. Beresford TP: Psychiatric assessment of alcoholic candidates for liver transplantation. In Lucey MR, Merion RM, Beresford TP (eds): Liver transplantation and
the alcoholic patient. Cambridge, England, Cambridge University Press, 1994, pp 29-49
7. Beresford TP: Predictive factors for alcoholic relapse in the selection of alcohol-dependent persons for hepatic transplant. Liver Transplantation and Surgery
3:280-291, 1997
8. Berkalovich GA, Steininger R, Herbst, F, et al: Efficacy of liver transplantation for alcoholic cirrhosis with respect to recidivism and compliance. Transplantation
58: 560-565, 1994
9. Campbell DA, Punch JD: Monitoring for alcohol use relapse after liver transplantation for alcoholic liver disease. Liver Transplantation and Surgery 3:300-303,
1997
9A. Bird GLA, O'Grady JG, Harvey FAK, et al: Liver transplantation in patients with alcoholic cirrhosis: Selection criteria and rates of survival and relapse. BMJ
301:15-17, 1990
10. Everhart JE, Beresford TP: Liver transplantation for alcoholic liver disease: A survey of transplantation programs in the United States. Liver Transplantation
and Surgery 3:220-226, 1997
11. Everson GT, Bharadhwaj G, House R, et al: Long-term follow-up of patients with alcoholic liver disease who underwent hepatic transplantation. Liver
12. Fattovich G, Giustina G, Degos F, et al: Morbidity and mortality in compensated cirrhosis type C: A retrospective follow-up study of 384 patients.

13. Gerhardt TC, Goldstein RM, Urschel HC, et al: Alcohol use following liver transplantation for alcoholic cirrhosis. Transplantation 62:1060-1063, 1996
13A. Gish RL, Lee AN, Keeffe EB, et al: Liver transplantation for patients with alcoholism and end-stage liver disease. American Journal of Gastroenterology
88:1337-1342, 1995
14. Hans P, Gutheil TG: Liver transplantation and the disliked patient: A clinical and ethical dilemma. Clin Transplant 5:277-281, 1991
15. Helander A, Tabakoff B, WHO/ISBRA Study Centres: Biochemical markers of alcohol use and abuse: Experiences from the pilot study of the WHO/ISBRA
collaborative project on state and trait markers of alcohol. Alcohol Alcohol 32:133-144, 1997
16. Hoofnagle JH, Kresima T, Fuller RK, et al: Liver transplantation for alcoholic liver disease: Executive statement and recommendations. Liver Transplantation
and Surgery 3:347-350, 1997
17. Howard LM, Fahy TA, Wong P, et al: Psychiatric outcome in alcoholic liver transplant patients. QJM 87:731-736, 1994
18. Howard LM, Williams R, Fahy TA: The psychiatric assessment of liver transplant patients with alcoholic liver disease: A review. J Psychosom Res 38:643-653,
1994
19. Kluge E: Drawing the ethical line between organ transplantation and lifestyle abuse. Canadian Medical Association Journal 150:745-746, 1994
20. Knechtle SJ, Flemming MF, Barry KL, et al: Liver transplantation for alcoholic liver disease. Surgery 112:694-703, 1992
21. Knechtle SJ, Flemming MF, Barry KL, et al: Liver transplantation in alcoholics: Assessment of psychological health and work activity. Transplant Proc
25:1916-1918, 1993
22. Kumar S, Stauber RE, Gavaler JS, et al: Orthotopic liver transplantation for alcoholic liver disease. Hepatology 11:159-164, 1990
23. Leshner AI: Addiction is a brain disease, and it matters. Science 278:45-47, 1997
850
24. Lucey MR, Brown KA, Everson GT, et al: Minimal criteria for placement of adults on the liver transplant waiting list: A report of a national conference
organized by the American Society of Transplant Physicians and the American Association for the Study of Liver Diseases. Liver Transplantation and Surgery
3:628-637, 1997
25. Lucey MR, Carr K, Beresford TP, et al: Alcohol use after transplantation in alcoholics: A clinical cohort follow-up study. Hepatology 25:1223-1227, 1997
26. Lucey MR, Merion RM, Henely KS, et al: Selection for and outcome of liver transplantation in alcoholic liver disease. Gastroenterology 102:1736-1741, 1992
27. Moss AH, Siegler M: Should alcoholics compete equally for liver transplantation? JAMA 265:1295-1298, 1992
28. O'Malley SS, Jaffe AJ, Chang G, et al: Naltrexone and coping skills therapy for alcohol dependence. Arch Gen Psychiatr 49:881-887, 1992
29. Osorio RW, Ascher NL, Avery M, et al: Predicting recidivism after orthoptic liver transplantation for alcoholic liver disease. Hepatology 20:105-110, 1996
30. Pereira SP, Williams R: Liver transplantation for alcoholic liver disease at King's College Hospital: Survival and quality of life. Liver Transplantation and
Surgery 3:245-250, 1997
31. Starzl TE, Van Thiel DH, Tzakis AG, et al: Orthotopic liver transplantation for alcoholic cirrhosis. JAMA 260:2542-2544, 1988
32. Surman OS: Psychiatric aspects of organ transplantation. Am J Psychiatr 146:972-982, 1989
33. United Network for Organ Sharing Web site: UNOS Web site. http/www.UNOS.org
34. Vaillant GE: What can long-term follow-up teach us about relapse and prevention of relapse in addiction? British Journal of Addiction 83:1147-1157, 1983
35. Vaillant GE: The natural history of alcoholism and its relationship to liver transplantation. Liver Transplantation and Surgery 3:304-310, 1997
36. Van Thiel DH, Bonet H, Gavalar J, et al: Effect of alcohol on allograft rejection rates after liver transplantation for alcoholic liver disease. Alcohol Clin Exp
19:1151-1155, 1995
37. Volpicelli JR, Volpicelli LA, O'Brien CP: Medical management of alcohol dependence: Clinical use and limitations of naltrexone treatment. Alcohol Alcohol
30:789-798, 1995
38. Wiesner RH, Lombardero M, Lake JR, et al: Liver transplantation for end-stage alcoholic liver disease: An assessment of outcomes. Liver Transplantation and

Surgery 3:231-239, 1997


April 22, 2000
Conde Petra
The changing spectrum of treatment for variceal

bleeding.
Citation
Bibliographic Data
Abstract
Indexing Data
The changing spectrum of treatment for
Copyright Notice and Disclaimer variceal bleeding.
NLM Citation ID:
99005003
Full Text
Full Source Title:
Frontmatter
Annals of Surgery
Methods
Publication Type:
Preoperative Data Language:
English
Operative Mortality
Author Affiliation:
Department of Surgery, University of Wisconsin, Madison,
Long-term Survival USA.
Discussion Authors:
Rikkers LF
Acknowledgments
Abstract:
References
OBJECTIVE: The objective of this study was to assess the
Discussion impact of endoscopic therapy, liver transplantation, and
transjugular intrahepatic portosystemic shunt (TIPS) on patient
selection and outcome of surgical treatment for this
complication of portal hypertension, as reflected in a single
surgeon's 18-year experience with operations for variceal
hemorrhage. SUMMARY BACKGROUND DATA: Definitive
treatment of patients who bleed from portal hypertension has
been progressively altered during the past 2 decades during
which endoscopic therapy, liver transplantation, and TIPS have
successively become available as alternative treatment options
to operative portosystemic shunts and devascularization

procedures. METHODS: Two hundred sixty-three consecutive

patients who were surgically treated for portal hypertensive
bleeding between 1978 and 1996 were reviewed retrospectively.
Four Eras separated by the dates when endoscopic therapy
(January 1981), liver transplantation (July 1985), and TIPS
(January 1993) became available in our institution were
analyzed. Throughout all four Eras, a selective operative
approach, using the distal splenorenal shunt (DSRS),
nonselective shunts, and esophagogastric devascularization, was
taken. The most common indications for nonselective shunts
and esophagogastric devascularization were medically
intractable ascites and splanchnic venous thrombosis,
respectively. Most other patients received a DSRS. RESULTS:
The risk status (Child's class) of patients undergoing surgery
progressively improved (p = 0.001) throughout the 4 Eras,
whereas the need for emergency surgery declined (p = 0.002).
The percentage of nonselective shunts performed decreased
because better options to manage acute bleeding episodes
(sclerotherapy, TIPS) and advanced liver disease complicated
by ascites (liver transplantation, TIPS) became available (p =
0.009). In all Eras, the operative mortality rate was directly
related to Child's class (A, 2.7%; B, 7.5%; and C, 26.1 %) (p =
0.001). As more good-risk patients underwent operations for
variceal bleeding, the incidence of postoperative
encephalopathy decreased (p = 0.015), and long-term survival
improved (p = 0.012), especially since liver transplantation
became available to salvage patients who developed hepatic
failure after a prior surgical procedure. There were no
differences between Eras with respect to rebleeding or shunt
occlusion. Distal splenorenal shunts (p = 0.004) and
nonselective shunts (p = 0.001) were more protective against
rebleeding than was esophagogastric devascularization.
CONCLUSIONS: The sequential introduction of endoscopic
therapy, liver transplantation, and TIPS has resulted in better
selection and improved results with respect to quality and
length of survival for patients treated surgically for variceal
bleeding. Despite these innovations, portosystemic shunts and
esophagogastric devascularization remain important and
effective options for selected patients with bleeding secondary
to portal hypertension.
Major Subjects:
Esophageal and Gastric Varices / Mortality / * Surgery
Gastrointestinal Hemorrhage / Mortality / * Surgery

Adult
Aged
Female
Human
Male
Middle Age
Postoperative Complications / Epidemiology
Survival Rate
Time Factors

April 22, 2000
Conde Petra
Guidelines for surgical procedures after liver

transplantation.
Citation
Bibliographic Data
Abstract
Indexing Data
Guidelines for surgical procedures after
Copyright Notice and Disclaimer liver transplantation.
Testa G - Ann Surg - 1998 Apr; 227(4): 590-9
Find More Articles Like This From NIH/NLM MEDLINE, HealthSTAR; NCI CANCERLIT
NLM Citation ID:
98223015
Full Text
Full Source Title:
Frontmatter
Annals of Surgery
Publication Type:
RESULTS Journal Article
Minor Procedures Language:
English
General Surgery
Author Affiliation:
Orthopedic Surgery
Baylor University Medical Center, Transplant Department,
Obstetrics and Gynecology, Dallas, Texas 75246, USA.
Thoracic and Plastic Surgery and
Neurosurgery Authors:
Testa G; Goldstein RM; Toughanipour A; Abbasoglu O;
Vascular Surgery Jeyarajah R; Levy MF; Husberg BS; Gonwa TA; Klintmalm
Head and Neck
GB
Urology Abstract:
OBJECTIVE: The first purpose of this study is to identify the
DISCUSSION types and incidences of surgical procedures in patients who
have previously undergone liver transplantation, with particular
Guidelines
focus on the complication rates and the lengths of hospital stay.
References The second purpose is to present the management guidelines for
patients with liver transplants at the preoperative,
intraoperative, and postoperative stages of surgical procedure.
SUMMARY BACKGROUND DATA: The surgical literature
on this issue is scant, and with the growing liver transplant
patient population it is not unlikey for any surgery specialist to

have to operate on a patient who has undergone liver

transplantation. METHODS: A sample of 409 patients with
available hospital records, with a minimum of a 2-year
follow-up, and with telephone access for interviews was chosen.
Type of surgery, time from the liver transplant, hospital stay,
immunosuppressive regimen, and complications were recorded.
RESULTS: A large proportion of patients (24.2%) underwent
some type of surgical procedure 2 to 10 years after liver
transplantation. The authors demonstrate that most of the
elective procedures can be safely carried out without an
increased incidence of complication and without longer hospital
stay than the general population. Conversely, emergent
procedures are plagued by a greater incidence of complications
that not only affect the function of the liver graft but may risk
the life of the patient.
Major Subjects:
* Liver Transplantation
* Surgical Procedures, Elective
Anastomosis, Roux-en-Y
Arthroplasty, Replacement, Hip
Digestive System Diseases / Surgery
Human
Immunosuppressive Agents / Therapeutic Use
Length of Stay
Postoperative Complications
Postoperative Period
Practice Guidelines
Surgical Mesh

(Immunosuppressive Agents)
Bookmark URL: /das/journal/view/N/10143154?source=CL,HS,MI

Annals of Surgery
Volume 227 Number 4 April 1998
Copyright 1998 Lippincott-Raven Publishers
590
ARTICLES
Guidelines for Surgical Procedures After Liver Transplantation
Giuliano Testa MD
Robert M. Goldstein MD
Abbas Toughanipour MD
Osman Abbasoglu MD
D. Rohan Jeyarajah MD
Marlon F. Levy MD
Bo S. Husberg MD, PhD
Thomas A. Gonwa MD
Goran B. Klintmalm MD, PhD
From the Baylor University Medical Center, Transplant Department, Dallas, Texas
Objective
The first purpose of this study is to identify the types and incidences of surgical procedures in
patients who have previously undergone liver transplantation, with particular focus on the
complication rates and the lengths of hospital stay. The second purpose is to present the
management guidelines for patients with liver transplants at the preoperative, intraoperative, and
postoperative stages of surgical procedure.
Summary Background Data
The surgical literature on this issue is scant, and with the growing liver transplant patient
population it is not unlikely for any surgery specialist to have to operate on a patient who has
undergone liver transplantation.
Methods

A sample of 409 patients with available hospital records, with a minimum of a 2-year follow-up,
and with telephone access for interviews was chosen. Type of surgery, time from the liver
transplant, hospital stay, immunosuppressive regimen, and complications were recorded.
Results
A large proportion of patients (24.2%) underwent some type of surgical procedure 2 to 10 years
after liver transplantation. The authors demonstrate that most of the elective procedures can be
safely carried out without an increased incidence of complication and without longer hospital stay
than the general population. Conversely, emergent procedures are plagued by a greater incidence
of complications that not only affectthe function of the liver graft but may risk the life of the
patient.
Address correspondence to: Robert M. Goldstein, MD, Transplant Services, Baylor University Medical Center, 3500
Gaston Avenue, Dallas, TX 75246.
Accepted for publication April 1998.
Patients increasingly undergo liver transplantation, and their survival rates are expected to exceed 5
years. Little has been written about the incidences and the types of other surgical procedures this patient
population may have after transplantation. Additionally, more surgeons without experience in
transplantation are faced with the care of the surgical diseases of these patients. Among the 1085 patients
who underwent a liver transplantation from 1985 to 1995, the records of 409 patients with a minimum of
a 2-year survival rate were reviewed. Two hundred sixty-three patients (24.2%) had at least one surgical
procedure, equaling 364 surgeries: 252 major procedures and 112 minor procedures. The surgeries were
divided by type, complications, hospital stay, and immunosuppressive regimen. Steroid dose
administration was recorded.
This study shows that elective surgery on patients with liver transplants can be safely performed and that
the complication rate and the hospital stay are similar to those of healthy patients. Moreover, once a few
simple guidelines are followed, the risk of graft dysfunction is practically nil. However, emergent
surgical procedures and procedures on patients with poor functioning grafts are affected by a greater
complication rate, which may cause a prolonged postoperative course.

From January 1985 to December 1995, liver transplantation was performed in 1085 patients (total
surgeries 1218). The 409 study patients were selected because their records were available. These
patients were interviewed, and the information collected was added to the information in the Baylor
University Medical Center Transplant Research database. The only procedures that were part of the study
were performed after discharge from the initial liver transplant
591
TABLE 1 -- Abdominal Surgery: Colectomies

Time Length of
From OLT Stay Steroids
(mo) Diagnosis (days) (mg) * Complications
28 Ulcerative colitis, 11 60
emergent
48 Carcinoma 10 10
28 GI bleed, CMV 60 10 Enterocutaneous fistula
22 Carcinoma, obstructed 14 40 Ascites
17 Diverticulitis 10 15
0.5 Perforation 16 40
2 Diverticulitis, emergent 60 20 Abdominal fluid collection,
percutaneous drainage
5 Lymphoma 9 10
20 Ulcerative colitis 8 10
2 Perforation 10 20
24 Diverticulitis 7 5
66 Carcinoma 7 10
24 Diverticulitis, emergent 15 7.5 Sepsis
CMV = cytomegalovirus.
* Average daily dose.
and those not immediately related to the transplant itself (hepatic artery reconstruction, bile duct surgery
because of leakage after T-tube removal, conversion to Roux-en-Y during the same admission of the
transplant, reexploration for bleeding or sepsis). Procedures were divided arbitrarily into major and
minor groups with regard to the use of general anesthesia and hospital or ambulatory setting. The major
procedures were divided into general surgery, which included abdominal and endocrine, orthopedic,
obstetrics and gynecology, cardiothoracic, vascular, head and neck, urology, plastic, and neurosurgical
procedures. The minor procedures included cataract surgery and removal of skin lesions. The time from
the transplantations, steroid dose, and immunosuppressive regimen during surgery, length of hospital
stay, complications, need for intravenous cyclosporine or tacrolimus, and postoperative liver
malfunctions were recorded. When calculating the number of minor procedures performed, only one

procedure was counted per patient because the data in our possession were not accurate, and the patients
themselves did not have a precise record. This held true especially for the removal of skin lesions
because many patients had more than one removed, and it was difficult to obtain accurate information.
RESULTS
Of the 409 patients who were contacted and whose records were reviewed, 263 underwent surgery after
their liver transplantation. One hundred forty-six did not. A total of 364 surgical procedures were
recorded, 252 major and 112 minor. Because 1085 patients underwent liver transplantation from 1985 to
1995, the incidence of patients requiring any kind of surgical procedure from 2 to 10 years after was
24.2%.
Minor Procedures
Fifty-six patients had at least one basal or squamous cell carcinoma skin lesion removed. The only
complication (1.7%) was an episode of local infection requiring drainage and wound care. Thirty-five
patients reported an episode of bleeding as the only complication after unilateral or bilateral cataract
surgery (2.8%). Twelve patients had some other type of ambulatory surgery requiring less than a 24-hour
hospital stay. No complications were recorded. Of the 10 patients who underwent placement of a
long-term central venous access, 2 (20%) had an infectious complication that required the removal of the
line.
General Surgery
One hundred sixteen patients had at least one general surgery procedure, which totaled 155 operations.
Eighty-one intraabdominal surgical procedures were performed on 63 patients. The breakdown was as
follows: 16 colectomies, (Table 1) , 6 ileostomy or colostomy take-down, (Table 2) , 14 Roux-en-y,
(Table 3) , 9 small bowel resections, 6 appendectomies, 7 exploratory laparotomies (Table 4) , 5
splenectomies, 4 lysis of adhesions, and 14 gastric, adrenal, duodenal, or liver surgery (Table 5) .
Forty-two patients underwent elective surgery, and 21 patients underwent emergent surgery. Infection,
localized or manifesting as generalized sepsis, was the most common complication. Infections were
observed in six patients undergoing emergent
592
TABLE 2 -- Abdominal Surgery Take Down

Time From OLT Length of Stay
(mo) (days) Steroids (mg) * Complications Reexploration
32 10 10 Bleeding
5 13 15 Bleeding,
reexploration
5 10 5 Yes

63 4 2.5
28 4 15
30 7 7.5
surgery and in four patients undergoing elective surgery. Postoperative bleeding was present in three
patients, renal failure in two patients, and ascites, enterocutaneous fistula, gastric fistula, anastomotic
leak, and pulmonary edema in one patient each (Table 6) . The overall morbidity rate for the
intraabdominal procedures was 31.7% (20 of 63 patients). The morbidity rate for the emergent surgeries
was 47.6% (10 of 21), and the rate for the elective surgeries was 23.8% (10 of 42). Five of the 10 patients
who suffered a complication after elective surgery went to surgery in suboptimal conditions as
demonstrated by their average albumin level of 2.5 mg/dL (2.2 to 3 mg/dL). Of the 37 patients who
underwent surgery in overall good general condition with an albumin level > 3.5 mg/dL, only five had
complications (13.5%). The hospital stay was greatly prolonged in the patients who underwent emergent
procedures. This is clearly demonstrated in the patients who underwent emergent colon surgery and had
an average stay of 28.6 days (10 to 60 days) compared with the average stay of 8.4 days (5 to 14 days)
with elective surgery.
The average amount of maintenance steroids administered to the patients who underwent emergent
surgery was 12.5 mg/day (2.5 mg-40 mg) compared with 11.2 mg/day (2.5 mg-60 mg) in the patients
undergoing elective surgery. Only four patients received a perioperative steroid bolus. Thirty-two
patients (50.7%) received intravenous immunosuppressive medications (30 cyclosporine and 2
tacrolimus), for a period of 2 to 5 days after surgery. The two patients who underwent intravenous
tacrolimus had complicated gastric surgery and were given nothing by mouth (NPO) for 5 days after the
procedure. A drastic change in the immunosuppressive regimen--more than 50% reduction in the main
immunosuppressive drug or discontinuation of azathioprine--was necessary in six patients (9.5%): five
for bacterial sepsis and one for systemic cytomegalovirus (CMV) infection.
Forty-two patients underwent 53 hernia repair procedures. Thirty-four incisional hernia repairs were
performed on 28 patients with a complication seen in 29.4% of the patients. The most common
complication was hernia recurrence, which was observed in 5 of the 34 patients (14.7%). Infection was
observed in 2 of the 34 patients (5.8%), prolonged ileus in 2 of the 34 patients (5.8%), and seroma in 1
patient (2.9%). Of the seven patients who had a repair with Marlex mesh (Davol, Cranston, RI), one
experienced an episode of infection requiring the removal of the mesh, and one patient had a recurrence.
TABLE 3 -- Roux-en-y
Time From Length of Stay
OLT (mo) (days) Steroids (mg) * Complications Revision
7 15 20 Yes
54 5 5

9 21 12.5 Bleeding/hematoma
22 6 10
12 5 10
42 5 10 Yes
39 5 5 Yes
52 11 10
48 9 15
46 7 7.5
20 7 10
70 24 10 Renal failure
3 8 15 Cholangitis
10 11 12.5 Yes
593
TABLE 4 -- Exploratory Laparotomy and Small Bowel Resection (SBR)

Time
From Length of Steroids
OLT (mo) Diagnosis Stay (days) (mg) * Complications
4 Strangulation 11 5
5 Lysis 7 15
7 Lysis 5 15
20 Appendectomy 5 7.5
34 Appendectomy (perforated) 8 10
2 Small bowel fistula 12 10
2 Exploratory 9 10
laparotomy/carcinomatosis/SBR
21 Lymphoma/SBR 17 20 Sepsis/death

Leak (4 more exploratory

laparotomies)
11 SBR/strangulation 70 12.5
22 Appendectomy 8 10
36 Lysis 6 20 Wound infection
14 Lysis 30 5 Renal failure
15 SBR 50 10 Gastric fistula,
reexploration
52 SBR/enterocutaneous fistula 24 5 Wound dehiscence
resection
46 Appendectomy 7 7.5
48 Carcinomatosis/SBR 7 7.5
72 SBR 5 10
The average steroid dose in the patients who experienced recurrence was 9 mg/day (7.5 mg to 10 mg)
compared with 9.7 mg/day (2.5 mg to 20 mg) in the patients without recurrence. For the patients who
experienced incisional hernia recurrence, the average time from transplant to repair was 26.4 months (12
to 40). The average time from transplant to repair for the patients who did not experience incisional
hernia recurrence, was 36.4 months (2 to 120). Only one episode of recurrence in
TABLE 5 -- Other Abdominal Surgeries
Time
From Length of
OLT Stay Steroids
(mo) Diagnosis Procedure (days) (mg) * Complications
6 Lymphomas Debulking hilar 21 15
lymphoma
4 G tube placement 60 12.5 Dehiscence, OR
reexploration
24 Pseudocyst Cystogastrostomy 11 5 Pulmonary edema
18 Gastric carcinoma Gastrectomy 48 2.5 Sepsis/pericarditis

18 Gastric ulcer Resection 6 10

36 Cystoadenoma Distal 6 10
pancreatectomy and
splenectomy
45 Perforated gastric Repair of ulcer and 44 7.5 Abdominal fluid
ulcer, hernia collection, sepsis,
paraesophageal pleural effusion
hernia
96 Metastatic Resection 7 7.5
retroperitoneal
hemangioma
12 Liver granuloma Resection 5 5
21 Duodenal mass Resection 15 5
2 G tube placement 7 20
17 Metastatic Adrenalectomy 7 10
hepatocarcinoma
10 Metastatic Adrenalectomy 7 12.5
hepatocarcinoma
84 Perforated duodenal Repair 80 5 CMV and ICU 27
ulcer days
CMV = cytomegalovirus.
594
TABLE 6 -- Abdominal Surgeries: Complications

Emergent (n = 21) Elective (n = 42)
Infection 6 4
Localized 2 3
Sepsis 3 1
Viral 1
Postoperative bleeding 0 3

Ascites 1
Renal failure 1 1
Enterocutaneous fistula 1
Gastric fistula 1
Anastomotic leak 1
Pulmonary edema 1
Total 10 (47.6%) 10 (23.8%)
11 procedures (9%) was recorded among the patients who underwent inguinal hernia repair. No
complications were recorded in the patients who underwent umbilical hernia repair.
With the twenty-two patients who underwent other types of general surgery, two minor complications
were observed. One patient experienced a recurrent abscess after incision and drainage of a wound
infection, and another patient had a urine leak after fenestration of a perirenal lymphocele (Table 7) .
Orthopedic Surgery
Orthopedic surgery procedures were the second most performed procedures (Table 8) . There were 56
orthopedic procedures performed on 47 patients. Thirty-one patients underwent hip replacement, five
patients had both hips replaced in a two-stage procedure, and three patients had simultaneous bilateral
hip replacement. Three patients (9.6%) suffered a complication after the surgery, which included one
footdrop, one prolonged recovery, and one pulmonary edema. The total complication rate for the
orthopedic surgery procedures was 10.7% (6 of 56). The average steroid dose at the time of the
procedure was 8.5 mg/day (2.5 mg to 20 mg), and the average time from the transplant was 31.5 months
(1 to 120 months). Thirty-nine patients were administered cyclosporine and eight were administered
tacrolimus.
Obstetrics and Gynecology, Thoracic and Plastic Surgery and Neurosurgery
Twenty patients underwent 21 obstetrics and gynecology procedures with no complications reported
(Table 9) . Additionally, no complications were reported among the 10 patients who underwent thoracic
surgery (Table 10) , the 5 patients who underwent 6 plastic surgery (Table 11) , and the 2 who underwent
neurosurgery.
Vascular Surgery
Twelve patients underwent 14 vascular operations. The complication rate was 25% (3 of 12). One patient
died during the repair of a mycotic aneurysm of the aorta, one patient had renal failure requiring
permanent hemodialysis
TABLE 7 -- Other General Surgeries

Time Length of
From Stay Steroids
OLT (mo) Diagnosis Procedure (days) (mg) * Complications
60 Mass Lumpectomy 1 5
6 Wound abscess Incision drainage 7 15 Infection,
reoperation
48 Lymphocele Fenestration 10 12.5 Urine leak
43 Infection Parotidectomy 10 10
50 Mass Lumpectomy 1 7.5
2 Mass, infiltrating Lumpectomy 1 15
ductal carcinoma
93 Pilonidal cyst Resection 1 7.5
31 Leg abscess Incision/drainage 3 10
4 Rectal Prolapse Repair 2 10
108 Hemorrhoids Hemorrhoidectomy 2 10
82 Wound infection Incision/drainage 2 5
3 Thyroid nodule Thyroidectomy 3 15
17 Breast cancer Mastectomy 5 10
44 Hurtle cell carcinoma Thyroidectomy 3 10
10 Leg lymphocele Drainage 2 10
14 Hyperparathyroidism Parathyroidectomy 3 10
22 Pilonidal cyst Resection 5 20
595

TABLE 8 -- Orthopedic Procedures

OLT (mo) Procedure (days) Steroids (mg) * Complications
42 Bilateral hip replacement 8 5
1 Hip surgery, incision, and drainage 30 10
8 Ankle surgery 9 15
24 Hip replacement 5 7.5
40 Knee replacement 5 2.5
17 Hip replacement 8 10
51 Arthroscopy 2 2.5
2 Hip replacement 3 15 Footdrop
2 Hip replacement 9 10 Pulmonary
edema

60 Back surgery 4 10
55 Laminectomy 4 10
32 Right shoulder cuff 2 2.5
34 Left shoulder cuff 1 2.5
24 Humerous surgery 4 10
24 Hand surgery debridement 10 10 Infection,
redebridement
48 Knee surgery 2 10 Technical
failure
51 Knee surgery 2 10
62 Knee replacement 5 10
70 Knee replacement 5 7.5
39 Knee replacement 5 5
4 Femur surgery 7 10
29 Knee surgery 7 7.5 Technical
failure
26 Knee surgery 7 7.5
1 Hip replacement 15 20 Prolonged
recovery
48 Knee surgery 4 7.5
12 Elbow surgery 2 10

12 Lumbar fusion 3 10
596
TABLE 9 -- Obstetrics and Gynecology Procedures

Time From OLT Length of Stay
(mo) Procedure (days) Steroids (mg) *
14 Hysterectomy 3 10
24 D&C 2 7.5
84 Resection vulvar lesion 1 10
1 Hysterectomy 3 20
57 D&C 1 5
22 Conization 1 10
58 Hysterectomy 4 5
14 Hysterectomy 7 5
20 Hysterectomy 2 5
97 Laparoscopy resection 7 5
17 C section 7 10
48 Ovariectomy/laparotomy 2 5
60 Laparoscopy 1 5
60 Conization 2 7.5
17 Tubal ligation 2 7.5
16 Ovariectomy 4 10
12 Ovariectomy 6 12.5

12 Hysterectomy and ovariectomy 5 12.5

D & C = dilation and curettage.
after the repair of a dissecting aneurysm of the thoracic aorta, and one patient had postoperative infection
of an arteriovenous fistula (Table 12) .
Head and Neck
Of the 11 patients who required head and neck surgery only one complication of postoperative bleeding
after tracheostomy (Table 13) . was recorded.
Urology
Five patients had a urology procedure. The only complication was the infection of a penile prosthesis.
The prosthesis was removed 10 days after placement (Table 14) .
DISCUSSION
The improvements in surgical technique and immunosuppressive treatment have made liver
transplantation standard therapy for end-stage liver disease. United Network of Organ Sharing (UNOS)
data [1] reported an overall 5-year survival rate of 68.8%, which, considering the almost 20,000 liver
transplants performed from 1988 to 1995 and the more than 4000 procedures performed in the past 3
years, indicates a constantly growing patient population with specific medical and surgical needs.
Although the treatment of the posttransplant drug-induced diseases (e.g., diabetes, hypertension, and
renal failure) has been addressed largely by the medical literature, little has been published regarding the
incidence of surgically treatable diseases after liver transplantation and the management of these
``special patients.'' The aim of this study is two fold: to record the incidences and the types of surgical
procedures that our patients have underwent once discharged from our institution after the liver
transplantation, and to address some of the important issues involved with the care of these patients --
issues that represent the basic questions that every surgeon not directly involved in transplantation would
ask himself or herself before, during, and after a surgical procedure on a patient with a transplanted liver.
The focus was mainly directed toward preoperative assessment, intraoperative, and postoperative
management of the most commonly encountered problems such as pain control, hypertension, renal
function, and immunosuppressive drug regimens.
Among the 1085 patients who underwent liver transplantation from 1985 to 1995, the incidence of
patients requiring any kind of surgical procedure 2 to 10 years after transplantation was 24.2%. This is
extremely high compared with the
TABLE 10 -- Thoractomy

Time From Length of Steroids

OLT (mo) Diagnosis Procedure Stay (days) (mg) *
24 Hepatocarcinoma Right nodule lobectomy 7 7.5
metastases
1 Hemothorax Right thoracotomy 10 20
96 Cryptococcus Wedge resection 7 10
6 Adenocarcinoma Thoracoscopy, 6 10
pleuroadhesis
30 Empyema Thoracotomy 15 15
51 Hepatocarcinoma Mediastinoscopy 4 5
metastases
24 Cholangiocarcinoma Wedge resection 5 10
metastases
24 Cystoadenoma carcinoma Wedge resection 4 10
metastases
12 Hemothorax Thoracotomy 15 5
14 Hemothorax Thoracotomy 7 10
597
TABLE 11 -- Plastic Surgery

Time From Length of Steroids (mg)
OLT (mo) Diagnosis Procedure Stay (days) *
2 Decubitus ulcer FLAP 10 10

33 Removal silicon implants 3 5
40 Cancer Right ear partial resection 2 10
24 Cancer Skin graft 3 5
36 Breast implants 6 10
105 Trauma Skin graft-leg 21 10

22 Perianal fistula FLAP (muscle) to anal area 3 10

nonhealing
incidence of surgical procedures in the general population. Considering that some of the diseases
requiring surgical treatment are drug-induced and that the life expectancy of these patients is increasing,
it is safe to predict that this percentage will increase because more patients will require at least one
surgical procedure.
This article shows that minor ambulatory surgery can be carried out without extra care. We recorded only
1 episode of bleeding in 35 patients undergoing cataract surgery (2.8%) and one episode of wound
infection in 56 patients undergoing skin lesion removal (1.7%). A greater rate of local infection was
documented in the patients undergoing Hickman line placement--2 of 10 (20%)--which stresses the need
for careful handling of these lines in the immunocompromised patient. Our recommendation is to
perform these procedures when the steroid dose is no greater than 15 mg/day to assure the best results in
wound healing. An extensive and expensive preoperative work-up is unnecessary, which limits
coagulation tests and platelet count to the patients with documented graft malfunction or renal failure.
Our data showed that the complication rate for major elective surgery is reasonably small, and any
procedure can be carried out safely without compromising the graft function. On the other hand, extreme
care must be accorded patients undergoing emergent surgery. Besides the risks and potential
complications that other patients have, they have the extra burden of the steroidal and
immunosuppressive therapy. This is clearly shown when the data regarding the complication rate in the
patients who underwent emergent intraabdominal surgery--47.6% versus 23.8%--in the patients who had
elective surgery are analyzed. Equally of interest, 50% of the complications related to elective surgeries
were sustained by patients with poor nutritional status as demonstrated by the low albumin levels.
Among the patients in overall good clinical condition, only 5 of 37 had complications, 3 had episodes of
bleeding, 1 had a localized infectious process, and 1 had temporary renal failure. Whether the surgery
was emergent or elective also had an impact on length of hospital stay. In fact, patients who underwent
emergent colon surgery had longer hospital stays
TABLE 12 -- Vascular Procedures
Time
From
OLT Length of Steroids
(mo) Diagnosis Procedure Stay (days) (mg) * Complications
48 Iatrogenic injury Arterial patch 3 5
52 Renal failure Arterial-venous fistula 2 10
12 Renal failure Aortohepatic conduit 11 20
120 Renal failure Arterial-venous fistula 4 10 Infection


11 Renal failure Arterial-venous fistula 15 12.5
24 Vascular Above knee amputation 7 15
insufficiency
27 Dissecting thoracic Repair 90 10 Renal failure
aneurysm
17 Myotic aneurysm Abdominal aortic 1 5 DIC/death
aneurysm
27 Iatrogenic injury Repair/patch 3 10
48 Graft infection Removal 2 10
4 Iatrogenic injury Patch 25 20
DIC = disseminated intravascular coagulopathy.
598
TABLE 13 -- Head and Neck

OLT (mo) Procedure (days) Steroids (mg) * Complications
48 Radical neck dissection 7 7.5
2 Tracheostomy 30 20
60 80% tongue resection 5 10
21 Removal 1 10
14 Drainage and packing 2 10
2 Tracheostomy 30 20
36 Partial tongue resection 9 5
104 Tracheostomy 60 10 Bleeding
52 Sinus surgery 1 10

(15 to 60 days, mean 45 days) compared with the patients who had elective colon surgery (5 to 14 days,
mean 8.4 days). [2] In elective general surgery, the greatest complication rate occurred in the patients who
underwent incisional hernia repair (29.4%; 10 in 34 procedures), the recurrence rate was 14.7%. None of
the patients who experienced a complication had a steroid dose administration greater than 12.5 mg/day.
The mean time from transplantation was 19.8 months (8 to 40 months), thus allowing a good recovery
period. It is possible that muscle depletion caused by liver dysfunction before transplantation, cumulative
steroid doses, and size of the incision do play a role in the high rate recurrence. Animal studies have
demonstrated no impact of cyclosporine or tacrolimus on wound healing. [3] [4] [5] [6] [7] The use of the
Marlex mesh is often dictated by the impossibility of obtaining a sound primary repair. Despite two
complications, one infection, and one recurrence, among seven patients in our early series, we have not
seen any complication in the patients who underwent repair after 1995 (data not published). We do not
discourage the use of Marlex mesh in the repair of large incisional hernias, especially when a poor
primary repair would put a patient at a greater risk for recurrence.
Forty-seven of 263 patients (17.8%) underwent 56 orthopedic surgical procedures. It is reasonable to
connect most of the orthopedic procedures to iatrogenic disease because of the use of steroids in the
immunosuppressive regimen. A similar increased incidence of skeletal complications compared with the
general population has been documented in patients who undergo kidney and pancreas transplantation
and is attributed by some investigators to the high dose of steroids used in these patient populations. In
our study, we found that the average dose of steroids at the time of the orthopedic procedure was only 8.5
mg/day (2.5-20 mg/day) and that the mean time from liver transplantation to skeletal injury was 31.5
months (1-120 months). The total amount of steroids administered to patients may be decreasing because
of the increasing use of tacrolimus as the main immunosuppressant drug. Therefore, in the near future we
may witness a decline in the number of orthopedic procedures in the patients with transplants. [8] [9] [10] Of
the patients in this study, 39 (82.9%) were administered cyclosporine and 8 were administered
tacrolimus. The complication rate was 10.7% (6 of 56 procedures). Three complications were results of
technical failures, and we could count only the three other complications--one prolonged recovery, one
pulmonary edema, and one infection--as directly related to the clinical status of the patients. Therefore,
we can conclude that orthopedic procedures can be carried out safely in patients after liver
transplantation, limiting the morbidity rate to approximately 5%.
None of the patients in this series had a documented decrease in graft function after the surgical
procedure. Postoperative adrenocortical insufficiency did not occur in any patient despite the fact that
only four patients received perioperative steroid boluses. This finding is supported by the study of
Bromberg et al. [11] on the kidney transplant population. The use of intravenous immunosuppression has
become rare since the introduction of tacrolimus and the micro-emulsion form of cyclosporine (Neoral,
Sandoz Pharmaceuticals, Basel, Switzerland), which have had excellent absorption even in the presence
of postoperative ileus.

TABLE 14 -- Urology
OLT (mo) Diagnosis Procedure (days) Steroids (mg) * Complications
22 Prostate cancer TURP 3 5
50 Impotence Penile 3 5 Infection
prosthesis removal
48 Prostate cancer TURP 2 7.5
60 Cancer Nephrectomy 10 5
85 Hydrocele Repair 1 5
83 Vasectomy 1 10
TURP = transurethral resection of the prostate.
599
Guidelines
The preoperative work-up should be carried out with the same criteria used for that particular procedure,
giving special attention to the coagulation parameters when the patient has documented graft
dysfunction. [12] [13] The immunosuppressive medication should be given at the scheduled times and no
dose should be withheld. In a case of sepsis, a decrease in the immunosuppressive regimen is
recommended, ie, stopping the azathioprine and/or lowering the dose of tacrolimus and cyclosporine.
However, this needs to be in consultation with transplant expertise.
Anesthesia in patients with normally functioning grafts does not require special precautions or drugs, but
it must be kept in mind that reduced acetylation is a feature of the poorly functioning graft and drug
doses requiring acetylation should be adjusted.14,15 Halothane is not only hepatotoxic, it decreases
hepatic blood flow and should not be used. Atracurium and cis-atracurium are safe paralyzing agents
because they do not require hepatic metabolism. In the postoperative period, special attention should be
given to the medications that affect the P450 microsomal enzyme pathway because they may interfere
with the metabolism of cyclosporine and create either toxic levels or decreased immunosuppression.
Moreover, because almost all patients have some impairment in renal function after liver transplantation,
any drug that requires renal metabolism should be adjusted for a clearance of 40 to 50 mL/minutes which
can be considered average in this population. This is especially important in the control of postoperative
pain when medications such as nonsteroidal antiinflammatory drugs must be avoided because they may
have an additive nephrotoxic effect with tacrolimus or cyclosporine. In fact, known steroidal analgesics
inhibit prostaglandin synthesis, may cause reduction in renal blood flow, and may precipitate renal
decompensation in patients who already take other nephrotoxic drugs such as cyclosporine and

tacrolimus. Postoperative pain control can be readily achieved with morphine or meperidine for
parenteral use and propoxyphene for enteral use without severe side effects. Postoperative hypertension
can be managed with central acting adrenergic drugs such as clonidine or some of the calcium-channel
blockers such as nifedipine. Because of its effects on cytochrome P450 and the consequent increase in
cyclosporine and tacrolimus levels, the use of doxazosin is not recommended. In patients who are stable
after surgery, daily levels of cyclosporine and tacrolimus and liver function tests are unnecessary. It is
sufficient to control the immunosuppressive medication level after surgery and to repeat it in patients
with variation greater than 25% of the baseline. When prolonged ileus or impaired oral intake is present,
our suggestion is to check the immunosuppressive medication level daily after the surgery until--with the
appropriate dose adjustments--the preoperative levels are obtained. If pharmacologic expertise with the
transplant-related medications is unavailable, our recommendation is to consult the transplant service that
supplied the donor organ.
In conclusion, any surgical procedure can be carried out safely in patients who undergo liver
transplantation once these guidelines are followed. Procedures that are more directly related to the
transplanted graft such as hepatic artery reconstruction or biliary tract surgery should be performed by
transplant surgeons more accustomed to the peculiarities of these types of surgery.
References
1. UNOS 1996 Annual Report, Richmond, Virgina.
2. Akamine S, Kawahara K, Takahashi T, et al. Effect of FK506 on bronchial healing of canine lung allografts. Journal of
the Japanese Association for Thoracic Surgery 1993;41:619-624.
3. Ahonen J, Nemlander A, Wiktorowicz, K, et al. Effect of Cyclosporine on wound healing. Transplant Proc
1983;15:3092-3093.
4. Guatelli R, Koh IHJ, Neto AB, et al. Effect of cyclosporine A on the healing process of ileal anastomosis in rats.
Transplant Proc 1996;28:2589.
5. Hassell TM, Romberg E, Sobhani S, et al. Lymphocyte-mediated effects of cyclosporine on human fibroblasts. In:
Kahan BD, ed. Cyclosporine. Grune & Stratton; 1988:993-1001.
6. Okubo T. Influences of cyclosporine A on wound healing. Hokkaido Journal of Medical Science 1993;68:665-682.
7. ChiuM, Coward M, Bruce D, et al. Long term incidence of bone fracture in kidney-pancreas transplant recipients is
high. Presented at the ASTP 15th Annual Meeting; Dallas, Texas.
8. Davidson JK, Tsakiris D, Briggs JD, Junor BJ. Osteonecrosis and fractures following renal transplantation. Clin Radiol
1985;36:27-35.
9. SchulzeS, Andersen J, Overgaard H, et al. Effect of prednisolone on the systemic response and wound healing after
colonic surgery. Arch Surg 1997;132:129-135.
10. BrombergJS, Baliga P, Cofer JB, et al. Stress steroids are not required for patients receiving a renal allograft and
undergoing operation. J Am Coll Surg 1995;180:532-536.
11. Baliga P, Merion RM, Turcotte JG, et al. Preoperative risk factor assessment in liver transplantation. Surgery

1992;112:704-710.
12. Johnston
TD, Katz SM. Special considerations in the transplant patient requiring other surgery. Surg Clin North Am
1994;74:1211-1221.
13. Jorm C, Stamford JA. Effects of liver disease on drug metabolism. Bailliere's Clinical Anesthesiology 1992;6:768-777.
14. Baubillier
E, Duvaldestin P. Anaesthetic management of patients with alcoholic liver disease. Bailliere's Clinical
Anaesthesiology 1992;6:847-859.

Bookmark URL: /das/journal/view/N/10143154?ja=104672&PAGE=1.html&ANCHOR=top&source=CL,HS,MI

April 22, 2000
Conde Petra
Liver transplantation. American Association for

the Study of Liver Diseases.
Citation
Bibliographic Data
Abstract
Indexing Data
Liver transplantation. American
Copyright Notice and Disclaimer Association for the Study of Liver
Diseases.
Carithers RL Jr - Liver Transpl - 2000 Jan; 6(1): 122-35
Full Source Title:
Liver transplantation : official publication of the American
Association for the Study of Liver Diseases and the
International Liver Transplantation Society.
Publication Type:
Review, Tutorial
Language:
English
Author Affiliation:
Division of Gastroenterology, University of Washington School
of Medicine, Seattle, WA.
Authors:
Carithers RL Jr
141
Abstract:
Liver transplantation has revolutionized the care of patients
with end-stage liver disease. Liver transplantation is indicated
for acute or chronic liver failure from any cause. Because there
are no randomized controlled trials of liver transplantation
versus no therapy, the efficacy of this surgery is best assessed

by carefully comparing postoperative survival with the known

natural history of the disease in question. The best examples of
this are in primary biliary cirrhosis and primary sclerosing
cholangitis, for which well-validated disease-specific models of
natural history are available. There are currently relatively few
absolute contraindications to liver transplantation. These
include severe cardiopulmonary disease, uncontrolled systemic
infection, extrahepatic malignancy, severe psychiatric or
neurological disorders, and absence of a viable splanchnic
venous inflow system. One of the most frequently encountered
contraindications to transplantation is ongoing destructive
behavior caused by drug and alcohol addiction. The timing of
the surgery can have a profound impact on the mortality and
morbidity of patients undergoing liver transplantation. Because
of the long waiting lists for donor organs, the need to project far
in advance when transplantation might be required has proven
to be one of the greatest challenges to those treating patients
with end-stage liver disease. Three important questions must be
addressed in a patient being considered for liver transplantation:
(1) when should the patient be referred for possible
transplantation? (2) when should the patient be listed for
transplantation? and (3) when is the patient too sick to have a
reasonable chance of surviving the perioperative period?
Major Subjects:
Human
Liver Diseases / Surgery
Liver Failure / Surgery
Patient Selection
Prognosis
Societies, Medical
Bookmark URL: /das/guideline/view/N/11264910?source=MI

April 22, 2000
Conde Petra
Indications for referral and assessment in adult

liver transplantation: a clinical guideline. British
Society of Gastroenterology.
Citation
Bibliographic Data
Indexing Data
Indications for referral and assessment in
adult liver transplantation: a clinical
guideline. British Society of
Gastroenterology.
Order a Full Text Copy of the

Devlin J - Gut - 1999 Dec; 45 Suppl 6: VI1-VI22
Original Journal Article From NIH/NLM MEDLINE, HealthSTAR
NLM Citation ID:
20029757
Full Source Title:
Gut
Publication Type:
Guideline; Journal Article; Practice Guideline
Language:
English
Author Affiliation:
Institute of Liver Studies, King's College School of Medicine
and Dentistry, Bessemer Road, London SE5 9PJ, UK.
Authors:
Devlin J; O'Grady J
Major Subjects:
Liver Diseases / Epidemiology / * Surgery
* Patient Selection
Referral and Consultation / * Organization &

Administration
Acute Disease

Adult
Chronic Disease
Great Britain / Epidemiology
Human

April 22, 2000
Conde Petra
Alcoholic liver disease: proposed

recommendations for the American College of
Gastroenterology [see comments]
Citation
Bibliographic Data
Abstract
Indexing Data
Alcoholic liver disease: proposed
Copyright Notice and Disclaimer recommendations for the American
College of Gastroenterology [see
Journal Articles On This Topic comments]
McCullough AJ - Am J Gastroenterol - 1998 Nov; 93(11):
Order a Full Text Copy of the 2022-36
Original Journal Article From NIH/NLM MEDLINE, HealthSTAR
NLM Citation ID:
99036255
Comment:
Am J Gastroenterol 1999 Oct;94(10):3065-6
Am J Gastroenterol 1999 Oct;94(10):3066-8
Full Source Title:
Publication Type:
Language:
English
Author Affiliation:
Center of Nutrition and Metabolism, Metro Health Medical
Center, Department of Medicine, Case Western Reserve
University, Cleveland, Ohio 44109-1998, USA.
Authors:
McCullough AJ; O'Connor JF
Abstract:
OBJECTIVE: The objective of this work was to develop
practice guidelines for the management of alcoholic liver
disease. METHOD: A computerized search using the Medline
Data Base from 1966-July 1997 was performed with the search

headings; alcohol, alcoholic hepatitis, alcoholic liver disease,

liver transplant, diagnosis, epidemiology, human, and English
only. All randomized controlled trials, case-control studies, and
meta-analyses were read in depth. A manual search was also
done using references from each retrieved report, review
articles, editorials, postgraduate course syllabi, and textbooks.
In the subsequent review, evidence was evaluated using a
hierarchical scale with randomized, controlled trials given the
most importance. Abstracts presented at national meetings were
included only when unique data were obtained from those
studies.
Major Subjects:
Liver Diseases, Alcoholic / * Diagnosis / * Therapy
Human

April 22, 2000
Conde Petra
Clinical practice guidelines for the management

of cirrhotic patients with ascites. Committee on
Ascites of the Italian Association for the Study
Citation of the Liver.
Bibliographic Data
Abstract
Indexing Data
Clinical practice guidelines for the
Copyright Notice and Disclaimer management of cirrhotic patients with
ascites. Committee on Ascites of the
Journal Articles On This Topic Italian Association for the Study of the
Liver.
Order a Full Text Copy of the Salerno F - Ital J Gastroenterol Hepatol - 1999 Oct; 31(7):
Original Journal Article 626-34
NLM Citation ID:
20071774
Full Source Title:
Italian Journal of Gastroenterology and Hepatology
Publication Type:
Review, Tutorial
Language:
English
Author Affiliation:
Department of Internal Medicine, A. Migliavacca Centre,
IRCCS Policlinico, University of Milan, Italy.
francesco.salerno@unimi.it
Authors:
Salerno F; Angeli P; Bernardi M; Laffi G; Riggio O; Salvagnini
M
67
Abstract:
An ad hoc Committee appointed by the Italian Association for
the Study of the Liver prepared these Clinical Practice
Guidelines for the Management of Cirrhotic Patients with

Ascites. The initial evaluation of a patient with ascites should
include a history, physical evaluation, paracentesis with ascitic
fluid analysis, abdominal ultrasonography and biochemistry to
assess the severity of liver disease and renal functionality. To
improve the efficiency of the choice between the different
opportunities available in the treatment of ascites, patients can
be classified into two subgroups: patients with uncomplicated
ascites and patients with complicated ascites, including
refractory ascites, bacterial peritonitis, hyponatraemia and renal
failure. Based upon evidence emerging from controlled clinical
trials or case-control studies, satisfactory treatment for
uncomplicated ascites is represented by paracentesis,
sodium-restricted diet and diuretics, whereas the treatment of
patients with complicated ascites requires other specific
approaches. As the prognosis for most patients with ascites is
poor, the last part of the paper offers simple criteria in the
selection of patients candidates for liver transplantation. The
aim of these guidelines is to reduce inappropriate practice and
to improve efficiency in the management of patients with
ascites. The Committee holds that a periodic update will be
necessary to conform to future scientific developments.
Major Subjects:
Ascites / Etiology / Mortality / * Therapy
Liver Cirrhosis / Complications / Diagnosis / Mortality /

* Therapy
Patient Care / * Standards
Female
Human
Italy
Liver Transplantation / Standards
Male
Prognosis
Societies, Medical
Survival Rate

April 22, 2000
Conde Petra
Minimal criteria for placement of adults on the

liver transplant waiting list: a report of a national
conference organized by the American Society
Citation of Transplant Physicians and the American
Bibliographic Data Association for the Study of Liver Diseases.
Indexing Data
Copyright Notice and Disclaimer Minimal criteria for placement of adults
on the liver transplant waiting list: a
report of a national conference organized
by the American Society of Transplant
Physicians and the American Association
Order a Full Text Copy of the for the Study of Liver Diseases.
Lucey MR - Transplantation - 1998 Oct 15; 66(7): 956-62
NLM Citation ID:
99013061
Full Source Title:
Transplantation
Publication Type:
Language:
English
Author Affiliation:
Division of Gastroenterology, University of Pennsylvania
Medical Center, Philadelphia 19104, USA.
lucey@mail.med.upenn.edu
Authors:
Lucey MR; Brown KA; Everson GT; Fung JJ; Gish R; Keefe
EB; Kneteman NM; Lake JR; Martin P; Rakela J; Shiffman
ML; So S; Wiesner RH
Major Subjects:
* Patient Selection
* Waiting Lists

Adult
Congresses
Gastroenterology
Human

April 22, 2000
Conde Petra
Ultrasound examination in the patient with

ascites.
Citation
Bibliographic Data
Indexing Data
Ultrasound examination in the patient
with ascites.
Find More Articles Like This Black M - Ann Intern Med - 1989 Feb 15; 110(4): 253-5

Full Source Title:
Annals of Internal Medicine
Publication Type:
Journal Article
Language:
English
Author Affiliation:
Temple University School of Medicine, Philadelphia,
Pennsylvania.
Authors:
Black M; Friedman AC
Major Subjects:
Ascites / * Etiology
Hepatic Vein Thrombosis / * Complications / Diagnosis
Thrombosis / * Complications / Diagnosis
* Ultrasonography
Adult
Case Report
Diagnostic Errors
Female
http://home.mdconsult.com/das/journal/view/O/8481091?source=HS,MI (1 of 2) [22/04/2000 1:05:58]

Hepatic Veins
Human
Male
Portal Vein
Time Factors
Bookmark URL: /das/journal/view/O/8481091?source=HS,MI

April 22, 2000
Conde Petra
Norfloxacin prevents spontaneous bacterial

peritonitis recurrence in cirrhosis: results of a
double-blind, placebo-controlled trial.
Citation
Bibliographic Data
Abstract
Indexing Data
Norfloxacin prevents spontaneous
Copyright Notice and Disclaimer bacterial peritonitis recurrence in
cirrhosis: results of a double-blind,
Find More Articles Like This placebo-controlled trial.
Gines P - Hepatology - 1990 Oct; 12(4 Pt 1): 716-24
91007687
Full Source Title:
Hepatology
Publication Type:
Clinical Trial; Controlled Clinical Trial; Journal Article;
Multicenter Study
Language:
English
Author Affiliation:
Liver Unit, Hospital Clinic i Provincial of Barcelona, Spain.
Authors:
Gines P; Rimola A; Planas R; Vargas V; Marco F; Almela M;
Forne M; Miranda ML; Llach J; Salmeron JM; et al
Abstract:
Eighty cirrhotic patients who had recovered from an episode of
spontaneous bacterial peritonitis were included in a multicenter,
double-blind trial aimed at comparing long-term norfloxacin
administration (400 mg/day; 40 patients) vs. placebo (40
patients) in the prevention of spontaneous bacterial peritonitis
recurrence. At entry, both groups were similar with respect to
clinical and laboratory data, ascitic fluid protein and
polymorphonuclear concentrations, number of previous
episodes of spontaneous bacterial peritonitis and causative
organisms of the index spontaneous bacterial peritonitis.

Norfloxacin administration produced a selective intestinal
decontamination (elimination of aerobic gram-negative bacilli
from the fecal flora without significant changes in other
microorganisms) throughout the study in six patients in whom
the effect of norfloxacin on the fecal flora was periodically
assessed. Fourteen patients from the placebo group (35%) and
five from the norfloxacin group (12%) developed spontaneous
bacterial peritonitis recurrence during follow-up (chi 2 = 5.97; p
= 0.014) (mean follow-up period = 6.4 +/- 0.6 mo; range = 1 to
19 mo). Ten of the 14 spontaneous bacterial peritonitis
recurrences in the placebo group and only one of the five
spontaneous bacterial peritonitis recurrences in the norfloxacin
group were caused by aerobic gram-negative bacilli (chi 2 =
8.87; p = 0.0029). The overall probability of spontaneous
bacterial peritonitis recurrence at 1 yr of follow-up was 20% in
the norfloxacin group and 68% in the placebo group (p =
0.0063) and the probability of spontaneous bacterial peritonitis
recurrence caused by aerobic gram-negative bacilli at 1 yr of
follow-up was 3% and 60%, respectively (p =
0.0013).(ABSTRACT TRUNCATED AT 250 WORDS)
Major Subjects:
Bacterial Infections / Mortality / * Prevention & Control
Norfloxacin / Administration & Dosage / * Therapeutic

Use
Peritonitis / Mortality / * Prevention & Control
Double-Blind Method
Drug Administration Schedule
Feces / Microbiology
Female
Human
Male
Middle Age
Patient Compliance
Recurrence

70458-96-7 (Norfloxacin)

April 22, 2000
Conde Petra
Paracentesis with intravenous infusion of

albumin as compared with peritoneovenous
shunting in cirrhosis with refractory ascites.
Citation
Bibliographic Data
Abstract
Indexing Data
Paracentesis with intravenous infusion of
Copyright Notice and Disclaimer albumin as compared with
peritoneovenous shunting in cirrhosis
Find More Articles Like This with refractory ascites.
Gines P - N Engl J Med - 1991 Sep 19; 325(12): 829-35
91342712
Full Source Title:
New England Journal of Medicine
Publication Type:
Clinical Trial; Journal Article; Multicenter Study; Randomized
Controlled Trial
Language:
English
Author Affiliation:
Liver Unit, Hospital Clinic i Provincial, Barcelona, Spain.
Authors:
Gines P; Arroyo V; Vargas V; Planas R; Casafont F; Panes J;
Hoyos M; Viladomiu L; Rimola A; Morillas R; et al
Abstract:
BACKGROUND. There is no satisfactory treatment for
refractory ascites in patients with cirrhosis. Both
peritoneovenous shunts and paracentesis have been used, but
there is uncertainty about their relative merits. METHODS. We
studied 89 patients with cirrhosis and refractory ascites who
were randomly assigned to receive either repeated large-volume
paracentesis plus intravenous albumin or a LeVeen
peritoneovenous shunt. Patients in the paracentesis group in
whom recurrent tense ascites developed during follow-up were
treated with paracentesis, and those in the

peritoneovenous-shunt group with diuretic agents or by the
insertion of a new shunt if there was shunt obstruction.
RESULTS. During the first hospitalization, ascites was
removed in all 41 patients in the paracentesis group and in 44 of
the 48 patients in the peritoneovenous-shunt group. The mean
(+/- SD) duration of hospitalization in the two groups was 11
+/- 5 and 19 +/- 9 days, respectively (P less than 0.01). There
were no significant differences in the number of patients who
had complications or died. During follow-up, 37 patients in
each group were hospitalized again. In the paracentesis group,
the number of rehospitalizations for any reason (174 vs. 97 in
the peritoneovenous-shunt group) or for ascites (125 vs. 38) was
significantly higher, and the median time to a first readmission
for any reason (1 +/- 1 vs. 2 +/- 2 months) or for ascites (2 +/- 2
vs. 8 +/- 17 months) was significantly shorter than in the
peritoneovenous-shunt group. The total times in the hospital
during follow-up, however, were similar in the two groups (48
+/- 49 and 44 +/- 39 days, respectively). Three patients had
obstructions of their peritoneovenous shunts during their first
hospitalizations, and 15 patients had a total of 20 obstructions
during follow-up. Survival was similar in both groups.
CONCLUSIONS. The LeVeen shunt and paracentesis are
equally effective in relieving refractory ascites. The former may
provide better long-term control of ascites, but shunt occlusion
is common and survival is not improved.
Major Subjects:
Albumins / * Administration & Dosage
Ascites / Mortality / Surgery / * Therapy
* Peritoneovenous Shunt / Adverse Effects
* Suction / Methods
Comparative Study
Female
Follow-Up Studies
Human
Infusions, Intravenous
Length of Stay
Male
Middle Age

Patient Readmission
Punctures
Recurrence
(Albumins)

April 22, 2000
Conde Petra
Randomized comparative study of therapeutic

paracentesis with and without intravenous
albumin in cirrhosis.
Citation
Bibliographic Data
Abstract
Indexing Data
Randomized comparative study of
Copyright Notice and Disclaimer therapeutic paracentesis with and without
intravenous albumin in cirrhosis.
Gines P - Gastroenterology - 1988 Jun; 94(6): 1493-502
Full Source Title:
Gastroenterology
Publication Type:
Language:
English
Author Affiliation:
Liver Unit, Hospital Clinic i Provincial of Barcelona, Spain.
Authors:
Gines P; Tito L; Arroyo V; Planas R; Panes J; Viver J; Torres
M; Humbert P; Rimola A; Llach J; et al
Abstract:
It has recently been shown that repeated large-volume
paracentesis associated with intravenous albumin infusion is a
rapid, effective, and safe therapy of ascites in cirrhosis. To
investigate whether intravenous albumin infusion is necessary
in the treatment of cirrhotics with large-volume paracentesis,
105 patients with tense ascites were randomly allocated into
two groups. Fifty-two patients (group 1) were treated with
paracentesis (4-6 L/day until disappearance of ascites) plus
intravenous albumin infusion (40 g after each tap), and 53
(group 2) with paracentesis without albumin infusion. After
disappearance of ascites, patients were discharged from the
hospital with diuretics. Patients developing tense ascites during

follow-up were treated according to their initial schedule.
Paracentesis was effective in eliminating the ascites in 50
patients from group 1 and in 48 from group 2, with the duration
of the hospital stay being approximately 11 days in both groups.
Paracentesis plus intravenous albumin did not induce significant
changes in standard renal function tests, plasma renin activity,
and plasma aldosterone. In contrast, paracentesis without
albumin was associated with a significant increase in blood urea
nitrogen, a marked elevation in plasma renin activity and
plasma aldosterone concentration, and a significant reduction in
serum sodium concentration. One patient from group 1 and 11
from group 2 developed renal impairment or severe
hyponatremia after treatment, or both (chi 2 = 9.19; p less than
0.01). The development of these complications could not be
predicted by clinical and laboratory data before treatment.
Although the probability of survival after entry into the study
was similar in patients from both groups, a multivariate analysis
identified the development of hyponatremia or renal
impairment, or both, following the first paracentesis treatment
and the occurrence of other complications during the first
hospitalization (encephalopathy, gastrointestinal bleeding, and
severe infection) as being the only independent predictors of
mortality. These results indicate that intravenous albumin
infusion is important in avoiding renal and electrolyte
complications and activation of endogenous vasoactive systems
in cirrhotics with ascites who are treated with repeated
large-volume paracentesis. The development of such
complications may impair survival in these patients.
Major Subjects:
Albumins / * Administration & Dosage
* Drainage
Liver Cirrhosis / * Therapy
Ascites / Therapy
Combined Modality Therapy
Comparative Study
Female
Human
Infusions, Intravenous
Length of Stay
Male

Middle Age
Random Allocation
(Albumins)

Alcoholismo y Cirrosis

Transféré par

Informations du document

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

Alcoholismo y Cirrosis

Transféré par

Droits d'auteur :

Formats disponibles

MD Consult - Case Of The Week

April 21, 2000

April 14, 2000

A 47-Year-Old Women With Alcoholic Cirrhosis and

How does ethanol damage the liver?

On physical examination her vital signs reveal a low-grade temperature and a

What are the causes and clinical features of cirrhosis?

http://home.mdconsult.com/das/stat/view/cow?nid=39981 (1 of 4) [21/04/2000 23:04:22]

The initial physical examination revealed a protuberant abdomen with a fluid

What is the likely diagnosis leading to such a mental status change?

The large-volume paracentesis was completed without complications, and the

http://home.mdconsult.com/das/stat/view/cow?nid=39981 (2 of 4) [21/04/2000 23:04:22]

Gines P, et al: Norfloxacin prevents spontaneous bacterial peritonitis recurrence in cirrhosis:

Gines P, et al: Paracentesis with intravenous infusion of albumin as compared with

Hoefs JC: Diagnostic paracentesis, a potent clinical tool,Gastroenterology 98:230-236, 1990

Rikkers LF: Variceal hemorrhage, Gastroenterol Clin North Am 17(2):289-301, 1988

Wilcox CM, Dismukes WE: Spontaneous bacterial peritonitis, a review of pathogenesis,

http://home.mdconsult.com/das/stat/view/cow?nid=39981 (3 of 4) [21/04/2000 23:04:22]

Copyright 2000 MD Consult L.L.C.

Bookmark URL: /das/stat/view/cow?nid=39981&sid=0

http://home.mdconsult.com/das/stat/view/cow?nid=39981 (4 of 4) [21/04/2000 23:04:22]

April 21, 2000

April 14, 2000

Question 1 - Alcohol Damage of the Liver

Here are some citations about alcohol damage of the liver in

http://home.mdconsult.com/das/stat/view/cow?nid=39982 (1 of 2) [21/04/2000 23:04:52]

Alcoholism And Alcohol Abuse

Alcohols and Glycols

ALCOHOL AND LIVER REGENERATION

Effects of prostaglandins on ethanol damage in primary cultured rat hepatocytes.

Ethanol-induced alterations of the microtubule cytoskeleton in hepatocytes.

Ethanol-induced retention of nascent proteins in rat hepatocytes is accompanied

Screening for Problem Drinking

Copyright 2000 MD Consult L.L.C.

Bookmark URL: /das/stat/view/cow?nid=39982&sid=0

http://home.mdconsult.com/das/stat/view/cow?nid=39982 (2 of 2) [21/04/2000 23:04:52]

April 21, 2000

Goldman: Cecil Textbook of Medicine, 21st Ed.,

Part IV - PREVENTIVE HEALTH CARE

ETHANOL METABOLISM. DEFINITIONS

ALCOHOL WITHDRAWAL ETIOLOGY

http://home.mdconsult.com/das/book/view/882/21.h...stty=cow&nid=39982&tlt=Z&pos=&tag=relinfo&anc=bo (1 of 17) [21/04/2000 23:05:55]

http://home.mdconsult.com/das/book/view/882/21.h...stty=cow&nid=39982&tlt=Z&pos=&tag=relinfo&anc=bo (2 of 17) [21/04/2000 23:05:55]

Ethanol is absorbed completely from the gastrointestinal tract

Ethanol oxidation to acetaldehyde by alcohol dehydrogenase in

Acetaldehyde is converted to acetate by aldehyde

http://home.mdconsult.com/das/book/view/882/21.h...stty=cow&nid=39982&tlt=Z&pos=&tag=relinfo&anc=bo (4 of 17) [21/04/2000 23:05:55]

The blood-brain barrier to ethanol is virtually non-existent, and

Figure 16-1 Ethanol metabolism. Alcohol

TABLE 16-1 -- ALCOHOL-RELATED MEDICAL

http://home.mdconsult.com/das/book/view/882/21.h...stty=cow&nid=39982&tlt=Z&pos=&tag=relinfo&anc=bo (5 of 17) [21/04/2000 23:05:55]

Nervous system Brain: Hepatic

of alcohol in the brain is nearly the same as in the blood. In a

http://home.mdconsult.com/das/book/view/882/21.h...stty=cow&nid=39982&tlt=Z&pos=&tag=relinfo&anc=bo (6 of 17) [21/04/2000 23:05:55]

and skin flushing. Cerebellar and vestibular function

depression with respiratory acidosis and hypotension. The

http://home.mdconsult.com/das/book/view/882/21.h...stty=cow&nid=39982&tlt=Z&pos=&tag=relinfo&anc=bo (7 of 17) [21/04/2000 23:05:55]

resemble the syndrome of transient global amnesia (see

A reduced response to ethanol, or tolerance, develops both

NATURAL HISTORY OF ALCOHOLISM.

Most people who consume alcohol begin drinking in

ALCOHOL WITHDRAWAL SYNDROME.

Ethanol is a central nervous system depressant. In alcoholics,

the nervous system appears to adapt to chronic exposure to