Vous êtes sur la page 1sur 9


Hepatopulmonary Syndrome and Liver

(1) Martinez G, Barbera JA, Navasa M, Roca J,
Visa J, Rodriguez-Roisin R. Hepatopulmonary
syndrome associated with cardiorespiratory
disease. J Hepatol 1999;30:882-889.
(Reprinted with permission. r1999 EASL, the
European Association for the Study of the Liver.)
Background/Aims: Hepatopulmonary syndrome is defined as a clinical triad including chronic liver disease,
abnormal pulmonary gas exchange resulting ultimately in
profound arterial hypoxaemia, and evidence of intrapulmonary vascular dilatation. We report five patients with
liver cirrhosis diagnosed with hepatopulmonary syndrome who had associated chronic obstructive or restrictive respiratory diseases. Methods: Clinical, radiographic
and contrast-enhanced echocardiographic findings, and
systemic and pulmonary haemodynamic and gas exchange, including ventilation-perfusion distribution, measurements were assessed in all five patients. Results:
Echocardiography was consistent with the presence of
intrapulmonary vasodilation without intracardiac abnormalities, and high resolution computed tomographic scan
features were compatible with clinical (3 cases) or histopathological diagnoses (2 cases) of associated respiratory
disorders. The most common prominent functional findings were moderate to severe arterial hypoxaemia, caused
by moderately to severely increased intrapulmonary shunting and/or mild to moderate low ventilation-perfusion
areas, and hypocarbia along with increased cardiac output
and a low pulmonary artery pressure and vascular resistance. Conclusions: These functional characteristics, classically reported in the setting of clinically stable, uncomplicated hepatopulmonary syndrome, conform to a
distinctively unique, chronic gas exchange pattern. Equally
important, these pulmonary haemodynamic-gas exchange hallmarks are not influenced by the coexistence of
chronic cardiorespiratory disease states. These data may
have clinical relevance for elective indication of hepatic
transplantation in patients with life-threatening hepatopulmonary syndrome.

(2) Egawa H, Kasahara M, Inomata Y, Uemoto S,

Asonuma K, Fujita S, Kiuchi T, Hayashi M,
Yonemura T, Yoshibayashi M, Adachi Y, Shapiro
JAM, Tanaka K. Long-term outcome of
living-related liver transplantation for patients
with intrapulmonary shunting and strategy
for complications. Transplantation
1999;67:712-717. (Reprinted with permission.)
Background: Of 320 living related liver transplantations
performed between June 1990 and September 1997, there

were 21 living related transplantations for patients with

intrapulmonary shunting, manifested by digital clubbing, cyanosis, and dyspnea. We report the long-term
outcome for more than 6 months and our strategy to
overcome complications in these recipients. Patients: A
total of 21 patients (age range 2-33 years; 19 children and
2 adults; 6 males and 15 females) were classified into three
grades according to shunt ratio calculated by TcMAA
pulmonary scintigraphy; 5 in mild group (shunt ratio:
less than 20%), 6 in moderated group (20%-40%), and
10 in severe group (more than 40%). The original
underlying liver disease was biliary atresia in all patients.
Results: Spearmens correlation coefficient rank test revealed that shunt ratio correlated significantly with PaO2
in room air (P 0.0001), PaO2 in 100% oxygen (P
0.0004), hematocrit (P 0.0276), and period of dyspnea
before transplantation (P 0.023). Complications: Wound
infection occurred in 80, 66, and 80%, and bile leakage in
20, 0, and 40% in mild, moderate, and severe groups,
respectively. Patients who had portal vein thrombosis and
intracranial complication were classified as severe group
and the incidence was 20 and 20%, respectively. The
patient actuarial one year survival was 80, 66.7, and 48%
in mild, moderate, and severe groups, respectively, although there was no significant difference. All patients
who survived improved hepatopulmonary syndrome and
the length of period required for the resolution was
significantly correlated to the preoperative shunt ratio
(P 0.023). Comments: Patients with severe shunting are
susceptible to wound infection and bile leak. The trend of
higher incidence of portal thrombosis and intracranial
complications in the severe group was closely related high
hematocrit. Secure surgical technique to reduce bile leak
and delayed primary wound closure to reduce wound
infection were found to be effective. Anticoagulant therapy
by infusing heparin through the portal vein followed by
Coumadin could prevent fatal portal vein thrombosis
without counter risk of fatal cerebral hemorrhage.

Hepatopulmonary syndrome (HPS) is characterized by
arterial hypoxemia caused by intrapulmonary vascular
dilatation that develops as a consequence of hepatic
dysfunction. This pulmonary vascular insult appears to
take at least two forms pathologically; diffuse precapillary/capillary dilatations and intrapulmonary anatomic
shunting caused by direct arteriovenous communication. The syndrome most commonly develops in
association with any cause of portal hypertension or
cirrhosis and thus is not infrequently recognized in
pediatric and adult liver transplant centers.
Remarkably, each component of this syndrome can
be eliminated with successful liver transplantation
techniques. Transplantation of a cadaveric or living
related liver has resulted in resolution of the vascular
dilatations and normalization of arterial oxygen levels.
These changes, however, may take months to occur,

Liver Transplantation, Vol 6, No 1 ( January), 2000: pp 113-121



Liver Transplantation Worldwide

suggesting that a significant remodeling occurs within

the pulmonary vasculature once the normal venous
blood constituents from the transplanted liver and
portal system perfuse the pulmonary arterial bed.
Thus, it appears that the pulmonary vessels, collectively as an innocent bystander, appear to have a
significant vulnerability to factors affected by hepatic
Two recent articles address pulmonary comorbidity
and postliver transplantation mortality in patients
with HPS. Regarding comorbidity, Martinez et al
described additional pulmonary pathophysiology in 5
of 16 adult patients (31%; age, 43 to 72 years) with
varying severity of HPS (PaO2 42 to 83 mm Hg,
depending on patient position). Using the sophisticated multiple inert gas elimination technique, routine
pulmonary function testing, and high-resolution chest
computed tomographic scanning, these investigators
reported that patients with well-defined HPS can have
other intrinsic pulmonary reasons for arterial hypoxemia that may not be a consequence of the liver disease.
More importantly, these other pulmonary insults (reported as expiratory airflow obstruction, transbronchial biopsy-proven usual interstitial pneumonitis with
traction bronchiectasis, and necropsy-proven interstitial lung disease associated with sarcoidosis) may not
improve after liver transplantation. In such patients,
the expectation of a normal PaO2 value after liver
transplantation may never be fulfilled. In addition, the
existence of other intrinsic and nonvascular pulmonary
problems may increase the risk for posttransplantation
morbidity. Recent experience at the Mayo Clinic
suggests a 20% incidence (5 of 25 patients) of
additional clinically significant pulmonary problems in
patients with HPS; 4 patients were considered inappropriate for liver transplantation because of those problems (emphysema, idiopathic pulmonary fibrosis, and
bilateral bronchiectasis).
Concerning mortality, Egawa et al reported eight
deaths (days 19 to 260) after living related liver
transplantation in 21 (38%) patients with HPS. A
mortality of 16% (13 of 81 patients) was reported in
the largest literature review (24 studies) published to
date.2 The conclusion from these series should include
the fact that although complete resolution of HPS
(even the most severe cases) can be realized as a
dramatic benefit of liver transplantation, such a result
is by no means attainable without significant risk.
The report by Egawa et al also stressed the importance of postoperative wound and biliary tract problems and alluded to the possible importance of increased hematocrit (secondary to chronic hypoxemia)

in the development of life-threatening portal vein

thrombosis. These investigators also described the
potential prognostic importance of pretransplantation
quantitative radionuclide lung scanning with brain
uptake (greatest mortality risk for uptake 30%;
normal 5%).
Regardless of the liver transplantation technique,
profound hypoxemia associated with HPS (PaO2 50
mm Hg) should be additionally evaluated by complete
pulmonary function testing, high-resolution chest computed tomographic scanning, radionuclide lung scanning, and pulmonary angiography in highly selected
cases (poor response to 100% inspired oxygen). Preliminary data suggest that the use of technetium-labeled
macroaggregated albumin (99mTcMAA) lung scanning
with brain uptake imaging can specifically quantify the
degree of hypoxemia associated with pulmonary vascular dilatation and thereby separate various vascular
from nonvascular contributions to hypoxemia in patients with liver disease.3,4
Although there is not a large experience that
characterizes poor outcome in patients with HPS after
liver transplantation, portal vein thrombosis, intracranial events, and multiorgan failure (associated with
refractory hypoxemia) appear to be associated with
mortality. Specific complications associated with living
related transplantation procedures and the importance
of hematologic response to chronic hypoxemia deserve
close study.
The spectrum of outcome (mortality versus complete syndrome resolution) after liver transplantation
emphasizes the importance of identifying practical,
diagnostic strategies that address prognosis. HPS has
evolved from an absolute contraindication5 to relative contraindication6 to indication for liver transplantation.2,7 This progress should not lull clinicians
into thinking these cases are without irreversible
comorbidities or increased risk.
Michael Krowka, MD
Mayo Clinic
200 First Street, SW
Rochester, MN 55905

1. Krowka MJ. Hepatopulmonary syndrome and portopulmonary
hypertension: Distinctions and dilemmas. Hepatology 1997:25:
2. Krowka MJ, Porayko MK, Plevak DJ, Pappas SC, Steers JL,
Krom RAF, et al. Hepatopulmonary syndrome with progressive
hypoxemia as an indication for liver transplantation: Case reports
and literature review. Mayo Clin Proc 1997;72:44-53.

Liver Transplantation Worldwide

3. Abrams GA, Nanda NC, Dubovsky EV, Krowka MJ, Fallon MB.
Use of macroaggregated albumin lung perfusion scan to diagnose
hepatopulmonary syndrome: A new approach. Gastroenterology
4. Whyte MKB, Hughes JMB, Peters AM, Ussov W, Patel S,
Burroughs AK. Analysis of intrapulmonary right-to-left shunt in
hepatopulmonary syndrome. J Hepatol 1998;29:85-93.
5. Van Thiel DH, Schade RR, Gavaler JS, Shaw BW Jr, Iwatsuki S,
Starzl TE. Medical aspects of liver transplantation. Hepatology
6. Maddrey WC, Van Thiel DH. Liver transplantation: An overview. Hepatology 1988;8:948-959.
7. Internet: UNOS Pediatric Policy 3.6. 1998. http://www.unos.org

Cross-Species Transmission of PERV Appears

Paradis K, Langjord G, Long Z, Heneine W,
Sandstrom P, Switzer W, et al. Search for
cross-species transmission of porcine endogenous
retrovirus in patients treated with living pig
tissue. Science 1999;285:1236-1241.
(Reprinted with permission. Copyright 1999
American Association for the Advancement of

Abstract not available

Pig organs have been proposed to address the shortage
of human donor tissue for solid-organ transplantation,
cellular transplantation, and a handful of extracorporeal therapies. Human case reports already exist regarding the use of porcine tissue in extracorporeal liver
perfusion2-4 and liver transplantation as a temporary
support in fulminant hepatic failure.5 At least one
extracorporeal bioartificial liver containing porcine
hepatocytes has completed preliminary (phase I) testing.6 However, reports of human cells infected by a
C-type porcine endogenous retrovirus (PERV)7-9 raise


concerns regarding the possibility of infectious transmission (pig to human) during human exposure to porcine
tissues. The recent report by Paradis et al1 provides
some reassurance on the safety of disease transmission
during cross-species exposure between pig tissue and
humans. Specifically, this report suggested that the
potential risk for infection by PERV is small and did
not occur in the 160 patients tested.
C-type porcine endogenous retroviruses such as
PERV have been recognized for nearly three decades.10
Early reports using porcine kidney cells as the source of
endogenous retrovirus suggested that these retroviruses
did not infect human cells.11,12 Only recently has
PERV infection been shown in vitro using human cell
lines and sensitive detection assays.7 At least three types
of PERV are now recognized based on envelope gene
sequencing.13 The term PERV is used generally in
reference to all three porcine endogenous retroviruses;
each share well-conserved sequences in the polreversetranscriptase (RT) region.14 A substantial number of
cultured porcine cells have been shown to produce
PERV based on the presence of PERV RNA8 or
porcine RT9 in their supernatant. These other porcine
cells include hepatocytes,8,15 lung,8 skin,8 endothelium,8 and peripheral-blood mononuclear cells
(PBMCs).9 Under in vitro coculture conditions, both
porcine endothelial cells8 and porcine PBMCs9 have
been shown to produce PERV capable of infecting
human embryonic kidney (HEK293) cells. In contrast,
PERV obtained from cultured porcine hepatocytes did
not produce sustained infection in HEK293 cells.15
The presence or absence of human fulminant hepatic
failure sera did not alter these findings.15
The possibility of PERV infection is significant for
at least two reasons. First, PERV is an endogenous
C-type retrovirus identified in nearly all pig tissues,16
although apparently not in fetal brain cells.17 Therefore, the possibility of breeding PERV-free pigs would
be quite complex, if at all feasible.14 Second, PERV is a
retrovirus and shares certain reproductive characteristics with human immunodeficiency virus type 1 (HIV1). As a result, concerns of an acquired immunodeficiency syndromelike epidemic caused by PERV are
inevitable, although no evidence exists regarding PERV
as a causative agent of human disease.
The possibility of PERV infection into human cells
in vivo was the major focus of the retrospective study
recently appearing in Science.1 These 160 patients were
studied after exposure to a variety of porcine therapies,
including extracorporeal liver perfusion (n 1),
extracorporeal kidney perfusion (n 2), bioartificial
liver perfusion (n 28), pancreatic islet transplanta-


Liver Transplantation Worldwide

tion (n 14), skin grafting (n 15), and extracorporeal splenic perfusion (n 100). All therapies used
living porcine cells at the time of human exposure.
Patients ranged in age from 2 to 77 years at the time of
exposure. The duration of exposure to porcine tissue
ranged from 15 minutes (extracorporeal pig kidney
perfusion) to 460 days (porcine pancreatic islet transplantation). The follow-up interval from treatment
ranged from less than 1 day to 12 years (mean, 38
At least one set of blood samples was obtained from
each patient to obtain PBMCs and serum for analysis.
Samples were analyzed in four laboratories using
multiple recently developed assays specific for PERV:
(1) polymerase chain reaction (PCR) for detection of
PERV DNA and pig DNA in PBMCs, (2) RT-PCR for
detection of PERV RNA in serum, and (3) protein
immunoblot analysis for detection of antibodies to
PERV in serum. In some cases, samples of saliva were
also collected for detection of PERV RNA by RT-PCR.
PCR testing showed that the PBMCs of 125
patients (81%) were negative for PERV DNA. This
group included all 29 patients who had undergone
either extracorporeal perfusion with a pig liver or
porcine bioartificial liver. Of the 30 patients found to
contain PERV DNA in their PBMCs, 23 also contained other sequences of pig DNA and were therefore
interpreted as microchimeric. The 23 patients found to
contain microchimerism had undergone extracorporeal splenic perfusion. The status of the remaining
patients was not interpretable because of insufficient
DNA for analysis. These PCR results were supported
by screening assays for PERV antibodies and RT-PCR
testing for PERV RNA. None of the serum and saliva
samples from these 160 patients, along with 5 close
contacts, contained PERV RNA. Four of 160 patients
were seroreactive to PERV Gag proteins, although
none of the 4 patients had other molecular evidence of
PERV infection.
A small number of unexplained symptoms were
noted by patients exposed to porcine tissue. Three
patients reported skin rashes after treatment with the
porcine bioartificial liver. One of these patients also
reported a 4-day febrile episode that was otherwise
unexplained. An ongoing skin rash was noted in a
patient after porcine skin grafting. All PERV testing
was negative in these 4 patients.
A few weaknesses of the study are apparent, primarily because of its retrospective nature. For example,
four separate laboratory sites were involved in sample
analysis. As a result, similar tests were conducted with
different primers, antigens, and methods. To provide a

most conservative approach, positive results from any

laboratory resulted in an overall positive interpretation.
Even with this conservative approach, no examples of
PERV infection (in the absence of microchimerism)
were detected. Of note, the study was unable to detect
PERV infection in the setting of microchimerism.
Another weakness of this study is the number of
samples found to contain insufficient DNA to complete analysis (i.e., 5 of 160 for PERV DNA, 7 of 30 for
pig DNA). In some cases, samples were redrawn and
analyzed at a later date. Nonetheless, data from
multiple patients were not interpretable because of
insufficient DNA to rule out or establish microchimerism. The study is also limited in that PBMCs were the
only cell type examined. PBMCs were chosen partly
because of the ease of obtaining these cells and because
of their potential importance in disease transmission.
However, infection of cells other than human PBMCs
is a possibility not addressed in this study. A final,
unavoidable weakness of the study relates to the lower
limits of detection of each assay. Namely, it is possible
that samples reported as negative may have contained
sequences of PERV in numbers less than the sensitivity
of the assay.
In summary, the possibility of PERV infection in
humans remains unknown. The recent retrospective
analysis by Paradis et al1 provides further reassurance
for safety in proceeding with clinical testing of pig
tissues as a treatment for human disease. Two smaller
retrospective studies previously did not detect evidence
of PERV infection in patents after extracorporeal pig
kidney perfusion18 and porcine islet transplantation.19
Well-designed prospective clinical trials are needed.
Based on recent recommendations from the Food and
Drug Administration Advisory Panel on Xenotransplantation,20 these trials should include the analysis of
blood and serum from all patients exposed to live
porcine cells. State-of-the-art assays of three general
classes should be used to screen for PERV transmission
in these patients: (1) assays to detect PERV DNA in
recipient cells (preferably PBMCs) using PCR, (2)
assays to detect evidence of viral infection by recipient
antibody responses, and (3) assays to detect viral
expression in plasma by enhanced RT-PCR techniques.
This level of rigor is needed to ensure safety when
porcine cells, tissues, or organs are used therapeutically
in humans.
Scott Nyberg, MD, PhD
Mayo Clinic
200 First Street, SW
Rochester, MN 55905


Liver Transplantation Worldwide

1. Paradis K, Langjord G, Long Z, Heneine W, Sandstrom P,
Switzer W, et al. Search for cross-species transmission of porcine
endogenous retrovirus in patients treated with living pig tissue.
Science 1999;285:1236-1241.
2. Abouna GM, Garry R, Hull C, Kirkley J, Walder DN. Pig-liver
perfusion in hepatic coma. Lancet 1968;2:508-509.
3. Fox IJ, Langnas AN, Fristoe LW, Shaefer MS, Vogel JE,
Antonson DL, et al. Successful application of extracorporeal
liver perfusion: A technology whose time has come. Am J
Gastroenterol 1993;88:1876-1881.
4. Chari RS, Collins BH, Magee JC, DiMaio JM, Kirk AD,
Harland RC, et al. Treatment of hepatic failure with ex vivo
pig-liver perfusion followed by liver transplantation. N Engl J
Med 1994;331:234-236.
5. Makowa L, Cramer D, Hoffman A, Breda M, Sher L, EirasHreha G, et al. The use of a pig liver xenograft for temporary
support of a patient with fulminant hepatic failure. Transplantation 1995;49:1654-1659.
6. Watanabe F, Mullon CJ-P, Hewitt W, Arkadopoulos N, Kahaku
E, Eguchi S, et al. Clinical experience with a bioartifical liver
(BAL) in the treatment of severe liver failure: A phase I clinical
trial. Ann Surg 1997;225:484-494.
7. Patience C, Takeuchi Y, Weiss R. Infection of human cells by an
endogenous retrovirus of pigs. Nat Med 1997;3:282-286.
8. Martin U, Kiessig V, Blusch J, Haverich A, von der Helm K,
Herden T, et al. Expression of pig endogenous retrovirus by
primary porcine endothelial cells and infection of human cells.
Lancet 1998;352:692-694.
9. Wilson C, Wong S, Muller J, Davidson C, Rose T, Burd P. Type
C retrovirus released from porcine primary peripheral-blood
mononuclear cells infects human cells. J Virol 1998;72:30823087.
10. Armstrong J, Porterfield J, De Madrid A. C-type virus particles
in pig kidney cell lines. J Gen Virol 1971;10:195-198.
11. Todaro G, Benveniste R, Lieber M, Sherr C. Characterization of
a type C virus released from the porcine cell line PK(15).
Virology 1974;58:65-74.
12. Lieber M, Sherr C, Benveniste R, Todaro G. Biologic and
immunologic properties of porcine type C viruses. Virology
13. Takeuchi Y, Patience C, Magre S, Weiss R, Banerjee P, LeTissier
P, et al. Host range and interference studies on three classes of
pig endogenous retrovirus. J Virol 1998;72:9986-9991.
14. Le Tissier C, Staye J, Takeuchi Y, Patience O, Weiss R. Two sets
of human-tropic pig retrovirus. Nature 1997;389:681-682.
15. Nyberg S, Hibbs J, Hardin J, Germer J, Platt J, Paya C, et al.
Influence of human fulminant hepatic failure sera on endogenous retroviral expression in pig hepatocytes. Liver Transpl
16. Stoye J. No clear answers on safety of pigs as tissue donor source.
Lancet 1998;352:666-667.
17. Weiss R. Xenografts and retroviruses. Science 1999;285:12211222.
18. Patience C, Patton G, Takeuchi Y, Weiss R, McClure M,
Rydberg L, et al. No evidence of pig DNA or retroviral infection
in patients with short-term extracorporeal connection to pig
kidneys. Lancet 1998;352:699-701.
19. Heneine W, Tibell A, Switzer W, Sandstrom P, Rosales GV,
Mathews A, et al. No evidence of infection with porcine

endogenous retrovirus in recipients of porcine islet-cell xenografts. Lancet 1998;352:695-698.

20. Auchincloss HJ. Chairmans report from the Food and Drug
Administration Advisory Panel on Xenotransplantation, June
3-4, 1999, Washington, DC. 14.

Transforming Power of Immunosuppression:

Experimental Mirage or Clinical Mirror
to the Future
Hojo M, Morimoto T, Maluccio M, Asano T,
Morimoto K, Lagman M, Shimbo T, Suthanthiran
M. Cyclosporine induces cancer progression by a
cell-autonomous mechanism. Nature
1999;397:530-534. (Reprinted by permission
from Nature Vol. 397, pp. 530. Copyright r1999
Macmillan Magazines Ltd.)

Abstract not available

In the practice of transplantation, most clinicians and
surgeons have had the unfortunate and disconcerting
experience of having a patient develop a malignancy
posttransplantation, with rapid progression of the
disease. One clinical question emanating from these
observations is always whether immunosuppressive
drugs themselves directly affect cell transformation.


Liver Transplantation Worldwide

These issues are especially germane to liver transplantation because this procedure is often performed on
people with malignant diseases. If specific immunosuppressive agents directly alter tumor biological characteristics, they likely will also impact on outcome after liver
transplantation for malignant diseases. This information is important for selecting the best immunosuppressive regiment for these specific patient populations.
The report by Hojo et al1 attempts to answer the
question, does cyclosporine directly affect tumor biology? First, they directly evaluated the effect of cyclosporine on the malignant phenotype in vitro by using a
nontransformed human pulmonary adenocarcinoma
(A-549) cell line. Cells cultured in the presence of
cyclosporine (1 g/mL) showed anchorage-independent proliferation (an assay correlating with invasive
tumor growth) and enhanced cell motility (an assay for
metastases). Second, experiments were performed to
determine the effect of cyclosporine on the metastatic
potential of a cancer in vivo. Murine renal adenocarcinoma cells showed increased metastases in immunodeficient (SCID) mice treated with cyclosporine (20
mg/kg orally every other day). The effects were associated with the ability of cyclosporine to increase
transforming growth factor- (TGF-) secretion by
the tumor cells and were blocked by monoclonal
antibodies to TGF-. The data were interpreted to
suggest that cyclosporine can promote cancer progression by direct effects on tumor cells independent of
their potential inhibition of immune surveillance. This
provocative study, published in a high-impact journal,
raises several key questions germane to the practice of
liver transplantation. These issues are discussed in the
following paragraphs.
Do the results of this study pertain to other
calcineurin inhibitors? The answer to this question is
likely. Tacrolimus, the other calcineurin inhibitor used
clinically, also increases TGF- generation at clinically
used concentrations.2 If the enhancement of the malignant phenotype is TGF- mediated, then either cyclosporine or tacrolimus should cause similar phenotypic
Can these data be extrapolated to transplant populations? The answer to this question is difficult to
address. First, the concentration dependence of the
data was not provided, and the concentrations of
cyclosporine used in these studies are greater than those
used in clinical transplantation. Second, TGF- does
not promote invasive behavior in all malignancies. For
example, TGF- induces cell-cycle arrest and apoptosis
in many hepatocellular carcinoma (HCC) cell lines.3
TGF- has even been suggested to be a tumor-

suppressor ligand.4 Despite these observations, cyclosporine has been shown to enhance the growth of
HCC cells and promote early recurrence in an animal
model of transplantation for HCC, perhaps by an
effect independent of TGF-.5 These observations
make it likely that cyclosporine directly enhances the
malignant phenotype in many solid tumors, including
Should this basic study impact on how we practice?
The answer is perhaps. First, this study should not alter
the way patients are counseled regarding the risks of
immunosuppression. Only skin cancer, lymphoma,
cervical cancer, and Kaposis sarcoma appear to be
increased in liver transplant recipients.6 These cancers,
aside from skin cancers, are related to viruses, and their
oncogenesis is uniquely different from other solidtissue malignancies. The most common solid-organ
tumors occurring in humans, breast, lung, and colon
cancers, are not increased in this patient population,
aside from patients with ulcerative colitis who undergo
transplantation for primary sclerosing cholangitis. These
patients are known to be at high risk for colon cancer
with or without transplantation. In this respect, it is
important to note that cyclosporine was not suggested
to promote tumor development in the study by Hojo
et al,1 but only the invasiveness of established cancers.
However, this study may alter our choice of immunosuppressants when liver transplantation is performed
for malignant disease.
The importance of this issue is highlighted by the
recent availability of rapamycin as an immunosuppressive agent. Rapamycin is growth inhibitory for transformed cell lines and has recently been used successfully as monotherapy in liver transplant recipients.7
Pilot studies comparing a classic calcineurin inhibitor
versus rapamycin after transplantation for malignant
disease are needed to answer this question in the era of
evidence-based medicine. However, it is possible that
rapamycin will become the agent of choice for these
patients in the future, based on studies like the one by
Hojo et al.1
In summary, I believe this study is a clinical mirror
to the future and not an experimental mirage. We are
now approaching the era of selected immunosuppression for different patient populations.
Gregory J. Gores, MD
Professor of Medicine
Mayo Clinic
200 First Street, SW
Rochester, MN 55905

Liver Transplantation Worldwide


1. Hojo M, Morimoto T, Maluccio M, Asano T, Morimoto K,
Lagman M, et al. Cyclosporine induces cancer progression by a
cell-autonomous mechanism. Nature 1999;397:530-534.
2. Khanna A, Cairns V, Hosenpud JD. Tacrolimus induces increased
expression of transforming growth factor-beta1 in mammalian
lymphoid as well as nonlymphoid cells. Transplantation 1999;67:
3. Gressner AM, Lahme B, Mannherz HG, Polzar B. TGF-beta
mediated hepatocellular apoptosis by rat and human hepatoma
cells and primary rat hepatocytes. J Hepatol 1997;26:1079-1092.
4. Tang B, Bottinger EP, Jakowlew SB, Bagnall KM, Mariano J,
Anver MR, et al. Transforming growth factor-beta1 is a new form
of tumor suppressor with true haploid insufficiency. Nat Med
5. Freise CE, Ferrell L, Liu T, Ascher NL, Roberts JP. Effect of
systemic cyclosporine on tumor recurrence after liver transplantation in a model of hepatocellular carcinoma. Transplantation
6. Penn I. Posttransplantation de novo tumors in liver allograft
recipients. Liver Transpl Surg 1996;2:52-59.
7. Watson CJ, Friend PJ, Jamieson NV, Frick TW, Alexander G,
Gimson AE, et al. Sirolimus: A potent new immunosuppressant
for liver transplantation. Transplantation 1999;67:505-509.

Discount Pricing and the Cost of Liver

Showstack J, Katz PP, Lake JR, Brown RS Jr,
Dudley RA, Belle S, Wiesner RH, Zetterman RK,
Everhart J. Resource utilization in liver
transplantation: Effects of patient characteristics
and clinical practice. JAMA
1999;281:1381-1386. (Reprinted with
permission. Copyright 1999, American Medical

Abstract not available

Liver transplantation remains one of the most expensive surgical procedures performed today.1,2 As a result,
third-party payers, providers, and public policymakers
have an acute interest in containing, if not reducing,
costs. Expense management is hardly a minor issue; it
is a major concern.
To address these issues, managed care organizations
have resorted to centers of excellence contracting and
case management.3-7 Based on the available evidence,
their efforts have apparently been successful. In 1993,
the actuarial firm, Milliman & Robertson, estimated
that the total first-year charges associated with liver
transplantation were $302,900.8 By 1996, this figure
increased to $314,500.9 However, in their most recent
report, Milliman & Robertson indicated that the total
first-year charges for liver transplantation have decreased significantly to $244,600.10
Unfortunately, most studies concerning the cost of
liver transplantation have actually focused on nonstandardized billed charges, which often bear little relationship to actual accounting costs. This issue has been
addressed at length in the literature.11,12
In an exceptional effort to overcome many of the
problems associated with previous analyses of liver
transplantation costs, Showstack et al,13 in a threecenter analysis, used a standard price list to value
resource utilization. Although their analysis excluded
all professional fees, they were able to examine in
exquisite detail the initial transplantation hospitalization for 711 patients who were aged 16 years or older,
had nonfulminant liver disease, and for whom the liver
was the only organ transplanted. Multivariate models
were tested in an effort to assess the independent
associations of resource utilization with patient demographic and clinical characteristics.


Liver Transplantation Worldwide

The results of their analyses both confirm and

elaborate on results previously reported.14-18 Resource
utilization was typically greater for patients who were
sickest at the time of transplantation, had alcoholic
liver disease as the primary diagnosis, received a donor
liver from a deceased person aged 60 years or older,
died in the hospital, or received multiple transplants.
They also found significant differences in resource
utilization across the three participating centers.13
Based on their results, Showstack et al13 expressed
concern about recommended changes in the allocation
of donor livers within the United States. These changes,
proposed by the Department of Health and Human
Services (DHHS) and recently endorsed by the Institute of Medicine, are intended to favor transplantation
of the sickest patients first.19 Showstack et al13
correctly concluded that if these policies were implemented, there will be a . . .substantial increase in the
resources used for liver transplantation in the United
The study by Showstack et al13 is eloquent in
methods compared with previous studies, although
they incorrectly characterized the results of the National Cooperative Transplantation Study (NCTS).14,15
In the NCTS, which included surgical and other
professional fees, Evans et al15 clearly showed substantial differences in transplantation procedurerelated
charges and hospital length of stay according to a
variety of demographic, prognostic, outcome, and
transplant center characteristics. However, unlike Showstack et al,13 Evans et al15 did not perform multivariate
Showstack et al13 also indicate that cost-effectiveness
has not been a serious consideration in the recent
national debate about organ allocation policies. This is
true in part, but in public testimony presented to the
DHHS at the National Institutes of Health on December 10, 1996, Evans and Kitzmann20 expressed concern
that the proposed DHHS policies would result in the
poorest transplant outcomes being achieved at the
greatest cost. In various contexts, officials representing
the United Network for Organ Sharing also have
expressed similar concerns.
Ultimately, analyses such as that conducted by
Showstack et al13 are of great significance to transplant
centers in their efforts to manage costs. The reason
should be obvious to most transplant center administrators. Managed care contracting based on global or
package prices requires a complete knowledge of actual
costs, resource utilization, and patient case-mix. To
their credit, Showstack et al13 help us better understand
each of these. Unfortunately, at this time, most trans-

plant centers are unable to quantify these financial

indicators in a meaningful way.
Increasingly, transplant centers are being forced to
aggressively compete on the basis of price because
outcomes have become similar across centers.21-23 In
effect, providers are assuming risk relative to their
patient-selection policies. As a result, in hopes of
remaining financially viable, it is likely transplant
centers will become more conservative in the patients
selected for transplantation. They will eventually recognize the financial penalty associated with adverse
patient selection. Thus, the highly controversial policies proposed by the DHHS may in principle gain
acceptance, but in clinical practice, their goal will not
be achieved.
Finally, note that the pricing policies of transplant
centers are highly variable based on the market penetration of managed care. This is readily apparent in Table
1. Clearly, transplant centers associated with hospitals
in highly managed care areas have a policy of aggressively marking up their hospital charges relative to
actual costs, whereas the policies of transplant hospitals
in markets with low-to-moderate managed care penetration are far more conservative in their pricing policies.
Consequently, based on a fee-for-service mentality,
transplant hospitals in markets heavily penetrated by
managed care can offer deep discounts for the services
they provide, which is apparent from the gross deductions from patient revenues listed in Table 1.

Table 1. Overall Inpatient Hospital Experience for All

Patient Admissions, 1997

Gross charges ($)
Adjusted charges
Cost ($)
Margin ($)
Margin as a percentage of cost
Gross deductions
from revenue

Managed Care Market



















*Definition of managed care penetration: high, 59%; moderate, 35%; low, 11%.
Case-mix and wage-index adjusted.

Liver Transplantation Worldwide

Of course, in a market now based on contracted

prices, actual accounting costs and markups become
meaningless considerations for payers intent on minimizing the prices they pay for the services their
beneficiaries require. Payers are not particularly concerned about the alleged unprofitability of academic
medical centers.24-27 They are much more interested in
maximizing the size of the discounts they negotiate,
even when hospitals with lower markups may offer
better prices. Thus, the solution should be obvious. In
the current managed care marketplace, hospitals should
aggressively mark up their prices, offer reasonably large
discounts, and thereby reap the benefits associated with
high profitability. Meanwhile, payers can remain disillusioned with their discount mentality.
Roger Evans, PhD
Mayo Clinic
200 First Street, SW,
Rochester, MN 55905

1. Evans RW. Organ transplantation and the inevitable debate as to
what constitutes a basic health care benefit. In: Terasaki P, Cecka
M (eds). Clinical transplants 1993. Los Angeles: UCLA Tissue
Typing Laboratory, 1994:359-391.
2. Evans RW. Effect of liver transplantation on local, regional, and
national health care. In: Busuttil RW, Klintmalm GB (eds).
Transplantation of the liver. Philadelphia: Saunders, 1996:869879.
3. Ascher NL, Evans RW. Designation of liver transplant centers in
the United States. Transplant Proc 1987;19:2405.
4. Evans RW. Public and private insurer designation of transplantation programs. Transplantation 1992;53:1041-1046.
5. Dahlberg R. Centers of excellence. Managed Care Q 1997;5(3):
6. Albrecht GL, Fitzpatrick R (eds). Advances in medical sociology.
Case and care management. Stanford: JAI Press, 1995.
7. Mayer GG. Case management as a mind set. Qual Manage
Health Care 1996;5(1):7-16.
8. Hauboldt RH. Cost implications of human organ transplantations, an update: 1993. Brookfield: Milliman & Robertson,
9. Hauboldt RH. Cost implications of human organ and tissue
transplantations, an update: 1996. Brookfield: Milliman &
Robertson, 1996.


10. Hauboldt RH, Courtney TD. Cost implications of human

organ and tissue transplantation, an update: 1999. Brookfield:
Milliman & Robertson, 1999.
11. Finkler SA. Issues in cost accounting for health care organizations. Gaithersburg: Aspen, 1994.
12. Finkler SA. The distinction between cost and charges. Ann
Intern Med 1982;96:102-109.
13. Showstack J, Katz PP, Lake JR, Brown RS Jr, Dudley RA, Belle
S, et al. Resource utilization in liver transplantation: Effects of
patient characteristics and clinical practice. JAMA 1999;281:
14. Evans RW, Manninen DL, Dong FB. The National Cooperative
Transplantation Study: Final report. Seattle: Battelle-Seattle
Research Center, 1991.
15. Evans RW, Manninen DL, Dong FB. An economic analysis of
liver transplantation: Costs, insurance coverage, and reimbursement. Gastroenterol Clin North Am 1993;22:451-473.
16. Kim WR, Therneau TM, Dickson ER, Evans RW. Preoperative
predictors of resource utilization in liver transplantation. In:
Cecka JM, Terasaki PI (eds). Clinical transplants, 1995. Los
Angeles: UCLA Tissue Typing Laboratory, 1996:315-322.
17. Brown RS Jr, Ascher NL, Lake JR, Emond JC, Bacchetti P,
Randall HB, Roberts JP. The impact of surgical complications
after liver transplantation on resource utilization. Arch Surg
18. Brown RS Jr, Lake JR, Ascher NL, Emond JC, Roberts JP.
Predictors of the cost of liver transplantation. Liver Transpl Surg
19. Institute of Medicine. Organ procurement and transplantation.
Washington, DC: National Academy of Sciences, 1999.
20. Evans RW, Kitzmann D. The arithmetic of donor liver
allocation. In: Terasaki PI, Cecka JM (eds). Clinical transplants,
1996. Los Angeles: UCLA Tissue Typing Laboratory, 1997:338342.
21. Evans RW. Organ transplantation in an era of economic
constraint: Liver transplantation as a case study. Semin Anesthesiol 1995;14:127-135.
22. Evans RW. Liver transplantation in a managed care environment. Liver Transpl Surg 1995;1:61-75.
23. Evans RW, Kitzmann DJ. Contracting for services: Liver
transplantation in the era of mismanaged care. Clin Liver Dis
24. Mechanic R, Coleman K, Dobson A. Teaching hospital costs:
Implications for academic missions in a competitive market.
JAMA 1998;280:1015-1019.
25. Goldsmith J. UCSF/Stanford: Building a prestige cartel.
Health Aff 1999;18(2):149-151.
26. Anderson GF, Greenberg G, Lisk CK. Academic health centers:
Exploring a financial paradox. Health Aff 1999;18(2):156-167.
27. Hallam K. Managed care linked to financial strain. Mod
Healthcare 1999;29(40):3.