Journal of Medical Virology 82:1318–1326 (2010)

Atypical Manifestations of Hepatitis A Infection:

A Prospective, Multicenter Study in Korea
Youn Mu Jung,1 Sang Jong Park,2 Jeong Sim Kim,1 Je-Hyuck Jang,1 Sang Hyub Lee,1 Jin-Wook Kim,1
Young Min Park,2 Seong Gyu Hwang,3 Kyu Sung Rim,3 Sook-Kyoung Kang,4 Hyun Seok Lee,4
Hae Sun Yun,5 Young Mee Jee,5 and Sook-Hyang Jeong1*
1
Department of Internal Medicine, Seoul National University Bundang Hospital, Seongnam-si, Gyeonggi-do, Korea
2
Department of Internal Medicine, Bundang Jesaeng General Hospital, Daejin Medical Center, Seongnam-si,
Gyeonggi-do, Korea
3
Department of Internal Medicine, Pochon CHA University Bundang CHA Hospital, Seongnam-si, Gyeonggi-do, Korea
4
Department of Internal Medicine, Seongnam Central Hospital, Seongnam-si, Gyeonggi-do, Korea
5
Division of Enteric and Hepatitis Viruses, Center for Infectious Diseases, Korea National Institute of Health,
Seoul, Korea

The clinical outcome of symptomatic hepatitis A
and the incidence and clinical characteristics of
atypical presentation of hepatitis A were studied
using prospective, multicenter design. The atyp-
ical presentation included delayed anti-hepatitis
A virus (HAV) immunoglobulin M (IgM) serocon-
version defined as positive anti-HAV IgM on the
repeated test within 7 days of hospital admission
after the initially negative result, prolonged
cholestasis, and acute kidney injury (AKI). A total
of 595 patients with symptomatic hepatitis A
requiring hospital admission were enrolled pro-
spectively from September 2006 to August 2008
in four major hospitals in a Korean city with a
population of
1 million. Clinical outcomes of
symptomatic hepatitis A showed a case fatality
rate of 0.2%, and fulminant hepatitis in 0.5%.
Delayed anti-HAV IgM seroconversion was found
in 6.4%, and was significantly associated with
shorter intervals from symptom onset to hospital
admission, higher body mass index, and lower
alanine aminotransferase (ALT) level at admis-
sion. Prolonged cholestasis was found in 4.7% of
patients, and could be predicted by preexisting
chronic hepatitis B viral infection, prolonged
prothrombin time, and higher total bilirubin level.
AKI was complicated in 1.5%, which could be
predicted by lower albumin level, higher ALT
level, and higher white blood cell (WBC) count.
More than half of the patients required hemo-
dialysis. Substantial occurrence of delayed anti-
HAV IgM seroconversion, prolonged cholestasis,
and AKI was confirmed with various predictable
factors, which could be helpful for accurate
diagnosis and management of hepatitis A
patients. J. Med. Virol. 82:1318–1326,
2010.
ß 2010 Wiley-Liss, Inc.
KEY WORDS: anti-HAV antibody; cholestasis;
hepatitis A virus; kidney injury;
prognosis
INTRODUCTION
Hepatitis virus A (HAV) is the fecal-orally trans-
mitted etiologic agent of acute viral hepatitis A identi-
fied in 1973. Following initial studies on the clinical
features of the disease and subsequent development of
safe and effective vaccines in the early 1990s, research
on acute hepatitis A faded away. However, according to
improvements in health sanitation and living condi-
tions, changes in the epidemiology of HAV have para-
doxically increased the disease burden in many regions
of the world [Martin and Lemon, 2006]. HAV remains an
important cause of hepatitis outbreaks and of fulminant
hepatitis in countries like Korea that are experiencing
an epidemiological shift [Lee et al., 2008]. During a
period of rapid economic development, Korea has
recently experienced a large community-wide outbreak
of hepatitis A, which provided us the opportunity to
None of the authors have any financial or other interest with
regard to the submitted manuscript that might be construed as a
conflict of interest.
Youn Mu Jung and Sang Jong Park contributed equally to this
work.
Grant sponsor: Korean Association for the Study of the Liver
(partially support).
*Correspondence to: Sook-Hyang Jeong, MD, Department of
Internal Medicine, Seoul National University Bundang Hospital,
Gumi-dong 300, Bundang-gu, Seongnam-si, Gyeonggi-do, Korea.
E-mail: jsh@snubh.org
Accepted 17 March 2010
DOI 10.1002/jmv.21822
Published online in Wiley InterScience
(www.interscience.wiley.com)

ß 2010 WILEY-LISS, INC.

Hepatitis A Infection With Atypical Course
study the clinical manifestations of hepatitis A, which
occurs most often in young adults who have not acquired
natural immunity.
Clinical spectrums of HAV infection range from
asymptomatic infection to fulminant hepatitis [Zimmer-
man et al., 1947]. Clinical presentation of hepatitis A
depends on the age of the patient; clinical presentations
are more severe in adults than in children [Nainan et al.,
2006]. Moreover, the clinical outcome in adults with
hepatitis A who also have preexisting liver disease has
been reported as poor; urgent liver transplantation is
required in some cases [Taylor et al., 2006]. However,
the clinical features and outcome of hepatitis A have
rarely been studied prospectively in this area, which has
been an obstacle to investigation of the cost-effective-
ness of a vaccination program. There have been reports
on atypical presentation of hepatitis A, including case
series with prolonged cholestasis, acute kidney injury
(AKI), recurrent hepatitis, hemolytic anemia, or other
extrahepatic manifestations [Cuthbert, 2001]. How-
ever, the clinical manifestations of hepatitis A that
include these atypical features were not studied in a
prospective manner, and detailed analysis of such
atypical presentation was limited.
Using a prospective, multicenter design, the aim
of this study was to investigate the clinical character-
istics and outcome of symptomatic hepatitis A requiring
hospital admission, with emphasis on the incidence
and clinical features of atypical presentation of
hepatitis A, which included delayed anti-HAV immuno-
globulin M (IgM) seroconversion, prolonged cholestasis,
and AKI.
METHODS
Patients and Diagnostic Criteria
of Hepatitis A
Serologically confirmed adult patients with acute
hepatitis A were prospectively enrolled from September
2006 to August 2008 in the four general hospitals located
in Seongnam city in Korea (Seoul National University
Bundang Hospital (SNUBH), Bundang Jesaeng Gen-
eral Hospital (BJGH), Bundang CHA University Hos-
pital (BCUH), and Seongnam Central Hospital (SCH)).
These four hospitals have a combined total of more than
3,000 beds and provide medical services for
1 million
residents of Seongnam city. Written informed consent
was obtained from all subjects, and this study was
approved by the institutional review board of the above
mentioned four hospitals, which was a part of a
comprehensive investigation on acute viral hepatitis
from A to E in this area. We included all of the
symptomatic patients requiring hospitalization because
of severe symptoms who occupied
60–70% of all the
patients with hepatitis A who visited hospitals. A total
23 patients excluded from this study: 6 patients refused
to participated in, and 17 patient showed negative result
on anti-HAV IgM test at first, but they did not undergo
the second test (incompletely studied group), because
1319
our diagnostic criteria for hepatitis A included obliga-
tory retesting of anti-HAV IgM after initially negative
result as described below. In addition, patients pre-
sented as acute liver injury other than viral etiology
such as toxic hepatitis, alcoholic hepatitis, autoimmune
hepatitis, genetic disorders including Wilson’s disease,
and biliary diseases were excluded.
The diagnosis of acute hepatitis A was made by
a positive result for anti-HAV IgM antibodies, typical
symptom manifestations, and elevation of alanine
aminotransferase (ALT). As a part of predefined
diagnostic criteria, anti-HAV IgM testing should be
repeated within 7 days of hospital admission if the initial
result was negative and other etiology was excluded.
The anti-HAV IgM test is repeated because some
hepatitis A cases turned out to be anti-HAV IgM positive
after only a short interval between the initial negative
result and the second test. In this study we defined these
cases as delayed anti-HAV IgM seroconversion, and two
negative results excluded acute hepatitis A. Commer-
cially available enzyme immunoassay kits were used to
detect serum anti-HAV IgM (anti-HAV IgM, Roche
Diagnostics, Indianapolis, IN in BJGH and SCH, and
aHAVM, Bayer Healthcare LLC, NY in BCUH, AxSYM
HAVAB-M, Abbott, Wiesbaden, Germany in SNUBH).
Prospective enrollment of the cases and fill-up of case
report forms were performed during the 2-year study
period. A total of 595 patients with confirmed acute
hepatitis A were enrolled in this study. Most of the
patients were followed until complete recovery, which
was defined as resolution of clinical symptoms and
normalization of liver tests.
Definition of Atypical Features of
Acute Hepatitis A
We studied atypical features of acute hepatitis A
in cases with delayed anti-HAV IgM seroconverison,
prolonged cholestasis, and AKI. Delayed anti-HAV IgM
seroconversion was defined as positive result on anti-
HAV IgM on the repeated test after initially negative
result described as above. Prolonged cholestasis was
defined as a total bilirubin level >5 mg dl1 lasting for
more than 4 weeks after hospital admission. AKI was
defined as an absolute level of serum creatinine (SCr)
higher than 2.0 mg dl1 or more in patients with no
history of renal dysfunction, or a percentage increase of
50% or more (1.5-fold) from baseline. Proteinuria was
defined as a urine protein concentration of more than
300 mg/day, or more than 1þ on a dipstick test.
Hematuria was defined as more than three red blood
cells per high power field. A comparative analysis was
made between the above three categories of atypical
hepatitis A and typical hepatitis A. Three patients
showed fulminant hepatitis defined as presence of
coagulopathy and hepatic encephalopathy within
4 weeks of symptom onset in the absence of preexisting
liver disease, and were therefore excluded from the
comparison of clinical characteristics between typical
and atypical features of hepatitis A.
J. Med. Virol. DOI 10.1002/jmv
 1320 Jung et al.

Statistical Analysis September 2006 and August 2008. Clinical character-

istics of the 595 hepatitis A cases were summarized in
Median, range, mean, and standard deviation were
Table I. Mean age was 31 years, and 61% were males.
used to describe statistics as appropriate. Continuous
The most prevalent age ranges were 20s (48%) and 30s
variables were compared using the independent sam-
(39%), and only 8% of the enrolled patients were older
ples’ t-test and are presented as mean  standard
than 40 years. We found 49 cases (8.6%) having the
deviation. Nominal data were compared using Fisher’s
contact history with hepatitis A cases within 2 months
exact test or Pearson’s chi-squared test as appropriate.
before symptom onset among the 572 patients who
Logistic regression and multiple regression were
responded to the elicited question on the contact history.
used for multivariate analysis. A P-value of <0.05 was
The mean interval from symptom onset to hospital
considered statistically significant. Data were analyzed
admission was 6 days and the mean hospital admission
using the SPSS statistical software package (SPSS
period was 8.7 days. Major symptoms included fever,
12.0K for Windows; SPSS, Seoul, Korea).
nausea, and jaundice. The peak total bilirubin level was
6.9 mg dl1, peak aspartate aminotransferase (AST) and
RESULTS ALT levels were 2,837 IU L1 and 3,298 IU L1, respec-
tively. Laboratory data were displayed as both initial
Clinical Characteristics and Outcomes of
levels at hospital admission and peak levels during the
Acute Hepatitis A in Korea
course of hepatitis.
A total of 595 patients with symptomatic acute The clinical outcome of acute hepatitis A showed
hepatitis A were admitted to the four hospitals between that 592 patients (99.5%) recovered without fulminant

TABLE I. Clinical Characteristics and Outcomes of the 595 Patients With Hepatitis A

Clinical features of hepatitis A Mean  SD or number (%) (n ¼ 595)

Age (years) 30.7  7.4
<20 30 (5%)
20–30 283 (48%)
30–40 234 (39%)
40–50 40 (7%)
>50 8 (1%)
Male (%) 365 (61%)
Body mass index (kg/m2) 22.9  3.5
Interval from symptom onset to hospital admission (days) 6.0  4.0
Hepatitis B virus carrier 28 (4.7%)
Seasonal case number
Spring (March to May) 196 (33%)
Summer (June to August) 288 (48%)
Fall (September to November) 51 (9%)
Winter (December to February) 60 (10%)
Presenting symptoms (%)
Fever 56%
General weakness 52%
Nausea/vomiting 61%
Dark urine 28%
Diarrhea 4%
Bilirubin (mg/dl), initial/peak level (normal range) 4.7  3.9/6.9  4.4 (0.2–1.2)
Alkaline phosphatase (IU/L), initial/peak level (normal range) 327  211/430  301 (30–115)
AST (IU/L), initial/peak level (normal range) 2,320  2,434/2,837  3,984 (<40)
ALT (IU/L), initial/peak level (normal range) 2,780  2,179/3,298  3,654 (<40)
Total cholesterol (mg/dl), initial/peak level (normal range) 124  36/162  62 (<200)
Albumin (g/dl), initial/peak level (normal range) 4.0  0.4/3.7  0.4 (3.3–5.2)
Prothrombin time (INR), initial/peak level (normal range) 1.3  0.3/1.0  0.2 (0.8–1.2)
WBC (103/mm3), initial/peak level(normal range) 5.5  3.1/6.9  3.1 (4–10)
Hemoglobin (g/dl), initial/peak level (normal range) 14.9  1.7/15.0  1.6 (12–16)
Platelet (103/mm3), initial/peak level (normal range) 179  72/167  70 (130–400)
Creatinine (mg/dl), initial/peak level 1.0  0.9/1.1  1.1
Clinical outcomes
Recovery without fulminant hepatitis 592 (99.5%)
Fulminant hepatitis 3 (0.5%)
Spontaneous recovery 2 (0.3%)
Mortality due to hepatic failure 1 (0.2%)
Atypical manifestations
Delayed anti-HAV IgM seroconverison 38 (6.4%)
Prolonged cholestasis 28 (4.7%)
Acute kidney injury 9 (1.5%)
WBC, white blood cell count; AST, aspartate aminotransferase; ALT, alanine aminotransferase; INR, international normalized ratio.
The laboratory results were presented as initial results at hospital admission and peak levels during the disease course.

J. Med. Virol. DOI 10.1002/jmv

Hepatitis A Infection With Atypical Course 1321

hepatitis, and 3 patients (0.5%) were complicated by
fulminant hepatitis; among them, 1 patient (24-year-old
female) died due to hepatic failure, and the 2 remaining
patients (24-year-old male and 31-year-old male) recov-
ered without liver transplantation. Therefore, overall
case fatality was 0.2% (1/595), and none of our study
subjects received liver transplantation. The comparison
of the laboratory results and the clinical outcome
between the patients younger than 40 years and the
patients over 40 years was displayed in Table II.
There were 28 patients with underlying chronic
hepatitis B virus (HBV) infection, and were either
inactive carriers or had chronic hepatitis, however,
none of them had liver cirrhosis. Among the 28 HBV
carriers superimposed with hepatitis A, 1 patient
experienced fulminant hepatitis A with spontaneous
recovery, and the overall mortality of hepatitis A in the
hepatitis B carriers was not different from that of non-
HBV carriers. However, six patients with HBV infection
(21%) had complications from prolonged cholestasis,
which showed a significantly higher incidence than that
of non-HBV carriers (4%). Therefore, overall morbidity
of hepatitis A increased in patients with chronic HBV
infection. Only one patient was positive for anti-HCV,
who was recovered without any complication.
Atypical Presentation: Delayed Anti-HAV IgM
Seroconversion
Among the 595 patients, 38 cases (6.4%) showed
delayed anti-HAV IgM seroconversion after hospital
admission. Comparison of clinical characteristics
between delayed and early anti-HAV IgM responders
was displayed in Table III. Clinical outcomes for delayed
responders were not different from those of early
responders. Delayed responders showed significantly
shorter intervals from symptom onset to hospital
admission and a higher frequency of fever, lower
frequency of dark urine and vomiting, higher body mass
index (BMI), lower levels of initial and peak platelet
counts, bilirubin, alkaline phosphatases, and ALT, and
higher peak level of total cholesterol than that of early
responders. However, independent variables associated
with delayed IgM response after logistic regression
analysis included the short interval from symptom onset
to hospital admission, high BMI, and low initial ALT
level (Table IV). This result showed that all delayed
responders had severe symptoms requiring hospital
admission, despite being within the serological window
period. Therefore, repeating an anti-HAV IgM test after
a short interval under clinical suspicion of hepatitis A is
important for accurate diagnosis of acute hepatitis A.
Atypical Presentation: Prolonged Cholestasis
Among the total of 592 patients, 28 patients (4.7%)
had complicated by prolonged cholestasis lasting for
more than 4 weeks after hospital admission. The
comparative analysis of clinical characteristics between
cases complicated with and without prolonged choles-
tasis was shown in Table V. Patients with prolonged
cholestasis were significantly older, and were more
frequently carriers of hepatitis B virus, had higher
frequency of fever and jaundice, and significantly low
albumin levels at hospital admission. Despite prolonged
cholestasis with problematic pruritus, all patients
recovered. The mean duration from hospital admission
to liver function recovery was 70 days in the cholestatic
hepatitis group, which was significantly longer than
that of non-cholestatic cases (36 days, P < 0.01). After
logistic regression, preexisting chronic hepatitis B
infection, prolonged prothrombin time, and higher total
bilirubin level at hospital admission were significantly
associated with prolonged cholestasis (Table IV).
Atypical Presentation: Acute Kidney Injury
Among the total 595 patients there were nine
complicated by AKI without fulminant hepatitis.
None of them had past history of renal dysfunction.
The clinical features of the nine patients were displayed

TABLE II. Comparison of Clinical Characteristics of Acute Hepatitis A According to Age

Age <40 (n ¼ 547) Age 40 (n ¼ 48) P-value

Male (%) 334 (61%) 31 (65%) 0.631
Interval from symptom onset to hospital admission (days) 5.8  3.6 7.8  6.3 0.038
Hepatitis B virus carrier (%) 26 (4.8%) 3 (6.3%) 0.722
Bilirubin (mg/dl), peak level 6.7  4.2 8.5  6.4 0.076
AST (IU/L), peak level 2,860  4,050 2,575  3,168 0.634
ALT (IU/L), peak level 3,299  3,655 3,289  3,683 0.985
Albumin (g/dl), peak level 3.7  0.4 3.5  0.4 0.026
Prothrombin time (INR), peak level 1.0  0.2 1.1  0.4 0.193
Clinical outcomes
Recovery without fulminant hepatitis 544 (99.5%) 48 (100%)
Fulminant hepatitis 3 (0.5%) 0
Spontaneous recovery 2 (0.3%) 0
Mortality due to hepatic failure 1 (0.2%) 0
Atypical features
Delayed IgM 35 (6.4%) 3 (6.3%) 1.000
Acute kidney injury 7 (1.3%) 2 (4.2%) 1.000
Prolonged cholestasis 26 (4.8%) 5 (10.4%) 0.095
WBC, white blood cell count; AST, aspartate aminotransferase; ALT, alanine aminotransferase; INR, international normalized ratio.

J. Med. Virol. DOI 10.1002/jmv

1322 Jung et al.

TABLE III. Clinical Features of Patients With Hepatitis A According to the Time of Anti-HAV IgM Seroconversion
After Hospital Admission

Delayed IgM seroconversion Early IgM seroconversion

(n ¼ 38) (n ¼ 554) P-value
Age (years) 31  8 31  7 0.837
Male (%) 25 (66%) 338 (61%) 0.558
Body mass index (kg/m2) 24.1  3.4 22.8  3.5 0.038
Interval from symptom onset to hospital 3.3  2.0 6.2  4.0 <0.001
admission (days)
Hepatitis B virus carrier (%) 2 (5.3%) 25 (4.5%) 0.689
Presenting symptoms (%)
Fever 79% 54% 0.003
Dark urine 8% 30% 0.004
Fatigue 8% 15% 0.240
Nausea 47% 62% 0.072
Vomiting 21% 40% 0.019
Diarrhea 3% 4% 1.000
Bilirubin (mg/dl), initial/peak level 1.5  1.0/5.0  3.4 4.9  3.9/7.0  4.4 <0.001/0.006
Alkaline phosphatase (IU/L), initial/peak level 191  147/309  229 337  211/448  304 <0.001/0.007
AST (IU/L), initial/peak level 1,695  1,812/3,395  2,933 2,362  2,465/2,797  4,046 0.102/0.372
ALT (IU/L), initial/peak level 1,559  1,539/3,226  2,136 2,963  2,192/3,300  3,740 <0.001/0.900
Platelet (103/mm3), initial/peak level 157  42/140  38 178  74/169  72 0.006/<0.001
WBC (103/mm3), initial/peak level 4.755  3.933/7.883  4.322 5.526  3.007/6.768  2.798 0.135/0.125
Hemoglobin (g/dl), initial/peak level 15.0  1.5/15.2  1.4 14.9  1.7/15.0  1.6 0.694/0.331
Prothrombin time (INR), initial/peak level 1.2  0.2/1.0  0.1 1.3  0.4/1.0  0.2 0.126/0.929
Total cholesterol (mg/dl), initial/peak level 133  29/184  69 123  37/160  61 0.182/0.030
Albumin (g/dl), initial/peak level 4.2  0.3/3.8  0.4 4.0  0.4/3.7  0.4 0.001/0.075
Creatinine (mg/dl) 1.1  0.5 1.1  1.2 0.855
WBC, white blood cell count; AST, aspartate aminotransferase; ALT, alanine aminotransferase; INR, international normalized ratio.
The laboratory results were presented as initial results at hospital admission and peak levels during the disease course.

in Table VI. Seven patients showed both proteinuria and
hematuria. Fractional excretion of sodium lower than
1.0 suggesting prerenal azotemia was found in three
patients. Five patients (56%) underwent hemodialysis,
and the eight patients eventually recovered normal
renal function, while one patient with underlying
hypertension did not recovered his renal function up
to 20 months after hospital admission (SCr level of
2.2 mg dl1). Clinical features of the five patients
receiving hemodialysis were not significantly different
from those of the four patients who recovered sponta-
neously, except for peak creatinine levels.
Comparative analysis of clinical characteristics
between cases complicated and not complicated with
AKI was shown in Table VII. Patients with AKI were
predominantly male, older in age, with higher BMI,
lower level of albumin, and higher level of white blood
cell (WBC) than those of patients with normal renal
function. Independent variables associated with AKI
after logistic regression included lower initial albumin
level, higher ALT level, and higher WBC count
(Table IV).
DISCUSSION
In this prospective, multicenter study, clinical out-
comes of symptomatic hepatitis A requiring hospital
admission showed a case fatality of 0.2% and an

TABLE IV. Results of Logistic Regression Analysis of Significant Variables Related to Atypical Features of Hepatitis A

P-value RC 95% CI
Variable related to delayed IgM seroconversion
Interval from symptom onset to hospital 0.003 0.705 0.561–0.887
admission (days)
Body mass index (>25 kg/m2) 0.041 2.628 1.039–6.647
ALT > 2,000 IU/L 0.001 0.190 0.069–0.524
Variable related to prolonged cholestasis
Preexisting chronic HBV infection 0.030 3.785 1.142–12.551
Prothrombin time (INR) 0.000 4.937 2.018–12.076
Total bilirubin (mg/dl) 0.004 1.113 1.035–1.198
Variable related to acute kidney injury
Albumin (g/dl) 0.016 0.008 0.000–0.401
ALT > 4,000 IU/L 0.015 78.514 2.328–2,648
WBC > 12,000/mm3 0.031 13.870 1.275–150.849
RC, regression coefficient; CI, confidence interval; WBC, white blood cell count; AST, aspartate aminotransferase; ALT, alanine aminotransferase;
INR, international normalized ratio.
The laboratory results were presented as initial results at hospital admission.

J. Med. Virol. DOI 10.1002/jmv

Hepatitis A Infection With Atypical Course 1323

TABLE V. Clinical Features of Hepatitis A According to the Presence of Prolonged Cholestasis

Prolonged cholestasis Non-cholestasis

(n ¼ 28) (n ¼ 564) P-value
Age (years) 35.0  8.0 30.5  7.3 0.002
Male (%) 61% 61% 0.946
Body mass index (kg/m2) 23.7  4.3 22.9  3.4 0.249
Interval from symptom onset to hospital 7.0  6.2 5.9  3.8 0.386
admission (days)
Hepatitis B virus carrier 18% 4% 0.006
Hepatitis C virus carrier 0 0.2% 1.000
Presenting symptoms (%)
Fever 75% 55% 0.037
Dark urine 46% 27% 0.028
Fatigue 18% 14% 0.580
Nausea/vomiting 61% 61% 0.961
Diarrhea 0 4% 0.616
WBC (103/mm3), initial/peak level 6.799  6.618/7.734  5.098 5.405  2.757/6.796  2.777 0.260/0.342
Hemoglobin (g/dl), initial/peak level 15.0  1.6/15.1  1.6 14.9  1.7/15.0  1.6 0.764/0.794
Platelet (103/mm3), initial/peak level 186  99/157  83 176  71/168  70 0.467/0.433
Total cholesterol (mg/dl), initial/peak level 113  30/189  117 124  37/160  57 0.147/0.231
Albumin (g/dl), initial/peak level 4.0  0.3/3.5  0.5 4.0  0.4/3.7  0.4 0.080/0.017
Bilirubin (mg/dl), initial/peak level 7.1  7.4/15.7  7.3 4.6  3.6/6.4  3.7 0.080/<0.001
Alkaline phosphatase (IU/L), initial/peak level 236  130/352  193 333  213/443  305 0.001/0.132
AST (IU/L), initial/peak level 2,888  3,442/4,061  3,968 2,289  2,368/2,775  3,980 0.354/0.102
ALT (IU/L), initial/peak level 3,074  2,361/3,824  2,454 2,764  2,169/3,269  3,707 0.449/0.433
Prothrombin time (INR), initial/peak level 1.6  0.8/1.1  0.5 1.2  0.3/1.0  0.2 0.020/0.224
Creatinine (mg/dl), initial/peak level 1.1  0.7/1.3  1.1 1.0  0.9/1.0  1.1 0.600/0.308
WBC, white blood cell count; AST, aspartate aminotransferase; ALT, alanine aminotransferase; INR, international normalized ratio.
The laboratory results were presented as initial results at hospital admission and peak levels during the disease course.

incidence of fulminant hepatitis of 0.5%. We found that
6.4% of acute hepatitis A patients were still within the
serological window period, although they experienced
severe symptoms with very high levels of AST/ALT. This
delayed anti-HAV IgM seroconversion has not been well
described in the literature; which highlights the impor-
tance of repeating the test after a short interval for
accurate diagnosis of hepatitis A. We found that 4.7% of
hepatitis A patients showed prolonged cholestasis,
which was defined as total bilirubin level >5 mg dl1
lasting for more than 4 weeks after hospital admission,
and was related significantly to the preexisting HBV
infection, suggesting higher morbidity from hepatitis A
among HBV carriers. There were 1.5% of hepatitis A
patients complicated with AKI. More than half of the
patients required hemodialysis treatment, and almost
all eventually recovered renal function.
Acute hepatitis A is a socioeconomic disease, and
improvements in living standards lead to a shift of peak
age of exposure to the virus from childhood to early
adulthood [Barzaga, 2000]. However, large scale, pro-
spective studies on the incidence, clinical character-
istics, and outcomes of hepatitis A have been limited.
Clinical features of hepatitis A in young adults are
generally severe, with abrupt and marked elevation of
aminotransferases, and about two-thirds of the sympto-
matic patients required hospital admission in our
previous study [Lee et al., 2008]. Our results on the
outcomes of hepatitis A showed that case fatality was
0.2%, which was compatible to previous results showing

TABLE VI. Clinical Features of Nine Cases of Hepatitis A Complicated With Acute Kidney Injury

Case no. 1 2 3 4 5 6 7 8 9
Age (years) 40 41 30 38 41 34 31 35 38
Sex Male Male Male Male Male Male Male Male Male
HBsAg         
Serum creatinine (mg/dl), initial 6.1 1.9 4.2 9.2 8.1 10.4 2.0 4.5 9.2
Serum creatinine (mg/dl), peak 6.1 5.6 11 9.2 12 10.4 2.0 4.5 11.3
Urine protein þþ þþ NA þþ þþþ þ þþþ Trace þþþ
Urine RBC þ þþ NA þþþ þþþ þþ þþ Trace þþþ
FENa 2.27 3.14 1.63 1.03 4.45 NA 0.11 0.14 0.75
Hemodialysis  þ þ þ þ    þ
ALT (IU/L), peak 3,467 56 5,921 7,655 5,359 1,843 10,286 4,082 4,471
Bilirubin (mg/dl), peak 9.2 37.6 5.6 19.1 9.5 11.9 4.9 7.0 10.5
Admission period 7 55 20 5 14 10 5 6 21
Duration of ALT recovery, days 14 62 20 16 17 10 NA 21 36
Duration of renal recovery, days 13 No recovery 20 89 56 27 5 21 132
ALT, alanine aminotransferase; RBC, red blood cell; FENa, fractional excretion of sodium; NA, Not available.
Case number 2 did not show the recovery of renal function till 20 months after hospital discharge with creatinine level of 2.2 mg/dl.

J. Med. Virol. DOI 10.1002/jmv

 1324 Jung et al.

TABLE VII. Clinical Features of Hepatitis A According to the Presence of Acute Kidney Injury

Acute kidney injury (þ) Acute kidney injury ()

(n ¼ 9) (n ¼ 583) P-value
Age (years) 36  4 31  7 0.019
Male (%) 9 (100%) 354 (60.7%) 0.015
Body mass index (kg/m2) 25.7  2.1 22.9  3.5 0.024
Interval from symptom onset to hospital 9.2  12.8 5.9  3.7 0.463
admission (days)
Hepatitis B virus carrier 0 4.6% 1.000
Hepatitis C virus carrier 0 0.2% 1.000
Presenting symptoms (%)
Fever 78% 56% 0.311
Dark urine 22% 28% 1.000
Fatigue 67% 15% 0.371
vomiting 11% 39% 0.322
WBC (/mm3), initial/peak level 11.570  7.611/ 5.382  2.869/6.753  2.729 0.041/0.049
12.473  7.417
Hemoglobin (g/dl), initial/peak level 15.2  1.9/15.2  1.9 14.9  1.7/15.0  1.6 0.643/0.727
Platelet (103/mm3), initial/peak level 185  145/175  144 176  71/167  69 0.868/0.877
Total cholesterol (mg/dl), initial/peak level 110  29/166  68 124  37/161  61 0.269/0.810
Albumin (g/dl), initial/peak level 3.7  0.3/3.3  0.4 4.0  0.4/3.7  0.4 0.008/0.003
Bilirubin (mg/dl), initial/peak level 9.5  9.8/13.2  9.9 4.6  3.7/6.8  4.2 0.176/0.087
Alkaline phosphatase (IU/L), initial/peak level 371  282/437  232 327  211/439  302 0.528/0.988
AST (IU/L), initial/peak level 4,355  4,480/4,359  4,476 2,288  2,381/2,814  3,976 0.204/0.249
ALT (IU/L), initial/peak level 5,007  3,083/5,010  3,078 2,745  2,147/3,268  3,662 0.002/0.157
Prothrombin time (INR), initial/peak level 1.47  0.32/1.03  0.13 1.26  0.34/1.04  0.18 0.093/0.903
BUN (mg/dl), initial/peak level 42.7  29.1/60.6  25.5 10.7  6.4/11.5  6.0 0.017/<0.001
Creatinine (mg/dl), initial/peak level 5.2  3.7/7.9  3.6 0.9  0.5/1.0  0.5 0.013/<0.001
WBC, white blood cell count; AST, aspartate aminotransferase; ALT, alanine aminotransferase; INR, international normalized ratio.
The laboratory results were presented as initial results at hospital admission and peak levels during the disease course.

an estimated case fatality rate of 0.015% in the
Shanghai epidemic of hepatitis A, which affected
310,746 people, primarily in the 20–40 years age range,
with 47 deaths in 1988 [Keeffe, 1995; Cooksley, 2000].
The fatality rate of hepatitis A among patients with
chronic HBV infection (0.05%) was 5.6 times greater
than among those without HBV infection (0.009%) in the
Shanghi epidemic. In our study, the morbidity of
hepatitis A in patients with chronic HBV infection
increased due to increased development of prolonged
cholestasis. Superinfection of HAV in underlying
chronic HBV infected liver with subsequent production
of cytokines such as interferon-gamma [Maier et al.,
1988] may exaggerate liver cell damage by activating
HBV-specific and HBV-non-specific cellular immune
responses [Locarnini, 2000] and resulted in a significant
decrease in HBV replication [Guidotti et al., 1996]. On
the other hand, the vigorous clinical presentation with
remarkably elevated ALT levels during acute hepatitis
A could result in severe outcomes in patients with
preexisting chronic liver disease and limited hepatic
functional reserve, regardless of the etiology of chronic
liver disease.
The diagnosis of hepatitis A by serological confirma-
tion of positive anti-HAV IgM is straightforward
[Nainan et al., 2006]. However, our study showed that
delayed anti-HAV IgM seroconversion after hospital
admission was found in 6.4% of symptomatic adult
patients. Independent factors related to delayed IgM
seroconversion included short interval from symptom
onset to hospital admission, high BMI, and low initial
ALT level. It revealed that these patients visited
J. Med. Virol. DOI 10.1002/jmv
hospital in the earlier phase of hepatitis A, after
symptom onset, with lower ALT levels; however, their
peak ALT levels were not different from those of early
IgM seroconverters. High BMI may indicate poor
tolerability of acute hepatitis symptoms in obese or
overweight people. Although we defined this as delayed
anti-HAV IgM response based on the view point of
caring physicians, the delayed period actually reflects
the seronegative window period of hepatitis A, which
has rarely been described in the literature [Cuthbert,
2001]. Two studies reported the presence of a sero-
negative viremic phase, emphasizing the importance of
HAV RNA testing for early diagnosis [Bower et al., 2000;
de Paula et al., 2004]. According to our results, repeating
the anti-HAV IgM test, rather than complicated testing
for HAV RNA, is not only an easy way to accurately
diagnose hepatitis A, but also to elucidate HAV-related
etiology in some cases of acute hepatitis of unknown
etiology. This phenomenon may be a reflection of a
vigorous cellular immune response to hepatitis A in
young adults, while neutralizing antibodies have not yet
fully developed in the early symptomatic phase of
hepatitis A.
Our study showed that 4.7% of hepatitis A patients
had complications associated with prolonged cholesta-
sis. Mean peak bilirubin level was 16 mg dl1 with mean
duration of 70 days from symptom onset to ALT
recovery, while the non-cholestatic group showed a
mean peak bilirubin level of 6 mg dl1 and a recovery
duration of 36 days (P < 0.001). These findings are
compatible with the report on cholestatic hepatitis A by
Tong et al. [1995], although their retrospective study
Hepatitis A Infection With Atypical Course
included only four patients with cholestatic hepatitis
among a total of 59 hepatitis A patients with different
diagnostic criteria. The optimal definition of prolonged
cholestasis complicating hepatitis A should be searched
further with exclusion of aggravating factors for
cholestasis, such as hemolytic anemia, renal dysfunc-
tion, and potentially hepatotoxic drug use. Patients
experienced a prolonged course with pruritus, but
reassurance of a good prognosis and symptomatic
management to relieve pruritus is required, while
corticosteroids should generally be avoided due to a
potential for immunosuppressive effects on viral clear-
ance [Cuthbert, 2001].
It is unclear what virologic or host factors may be
implicated in the pathogenesis of prolonged cholestasis.
In our study, the independent factors related to
prolonged cholestasis by logistic regression analysis
included preexisting chronic HBV infection, prolonged
initial prothrombin time, and higher initial bilirubin
level suggesting that host factors or severe damage to
liver cells in the early phase of hepatitis, rather than
viral factors, may be correlated to prolonged cholestasis
of hepatitis A. The cholestasis caused by viral infection is
probably induced by inhibition of bile salt transporter
function by proinflammatory cytokines, although the
detailed mechanisms of cholestasis are not exactly
known [Trauner et al., 1999]. On the other
hand, secretion of HAV across the apical canalicular
membrane of hepatocytes into the biliary system may
involve vectorial cellular vesicular transport mecha-
nisms [Martin and Lemon, 2006]. The duration of HAV
viremia was longer in patients with prolonged choles-
tasis than in those with a typical self-limiting course
[Sagnelli et al., 2003]. Therefore, direct viral effect on
the transporters related to cholestatic hepatitis could
not be excluded. One case report showed hepatitis A
virus (HAV) Ia and Ib coinfection in a prolonged and
severe cholestatic hepatitis A lasting 7 months, which
may have impaired antibody production of neutralizing
activity on the basis of simultaneous antigenic stimula-
tion of the two HAV isolates [Coppola et al., 2007].
However, there is no experimental data on the impaired
production of neutralizing antibody by infection of
multiple HAV strains.
Several case series have reported on AKI complicating
non-fulminant hepatitis A [Shroff et al., 2004; Song
et al., 2007; Kim et al., 2008], however, the incidence
rate of AKI in hepatitis A patients has not been reported.
Our results showed that 1.5% (9/595) of hepatitis A
patients had complications from AKI. Among the nine
patients, five (56%) underwent hemodialysis, however,
eight of the nine patients eventually recovered normal
renal function. Proposed mechanisms of renal damage
include prerenal factors associated with anorexia,
nausea, vomiting, diarrhea, and fever, as well as the
nephrotoxic effects of hyperbilirubinemia, immune
complex-mediated nephritis, interstitial nephritis, and
rarely massive intravascular hemolysis. A putative
receptor for HAV has been identified in African green
monkey kidney cells [Martin and Lemon, 2006], and
1325
viral antigen can be detected along the glomerular
basement membrane in some primates, as well as within
the cytoplasm of hepatocytes, germinal centers of the
spleen and lymph nodes [Nainan et al., 2006]. Therefore,
the direct role of HAV in the pathogenesis of AKI should
be studied further.
Our results showed that patients with AKI were
predominantly male, older in age, higher in BMI, and
had more delayed hospital admission, lower level of
albumin, and higher level of WBC than those of patients
with normal renal function. However, independent
variables associated with AKI after logistic regression
included lower albumin level, higher initial ALT level,
and higher WBC count, reflecting the suggestion that
more severe liver cell injury or inflammation may be
related to development of AKI in hepatitis A. Kim et al.
[2008] reported on independent predictors for AKI
development in hepatitis A, including lower hematocrit,
presence of coagulopathy, high CRP concentration, and
higher peak bilirubin level.
The limitations of our study were that we included
only symptomatic hepatitis A requiring hospital admis-
sion for the complete enrollment of the subjects and
that we did not study other atypical presentations of
hepatitis A [Tong et al., 1995], such as relapsing
hepatitis and hemolytic anemia. Therefore, our results
reflect the natural history of severely symptomatic
hepatitis A requiring hospital admission rather than
the overall symptomatic hepatitis A cases. Our previous
hepatitis A cohort study showed that 83% of the patients
needed hospital admission and 53% brought to the
emergency room, due to severe symptoms [Lee et al.,
2008].
In conclusion, we found that outcomes of symptomatic
hepatitis A in the young adult population showed a case
fatality rate of 0.2% and atypical presentations of
hepatitis A, such as delayed anti-HAV IgM seroconver-
sion, prolonged cholestasis, and AKI were not uncom-
mon with various predictive factors for development of
those atypical features. Therefore, further studies on
the mechanisms underlying these atypical features of
acute hepatitis A should be warranted.
REFERENCES
Barzaga BN. 2000. Hepatitis A shifting epidemiology in South-East
Asia and China. Vaccine 18:S61–S64.
Bower WA, Nainan OV, Han X, Margolis HS. 2000. Duration of viremia
in hepatitis A virus infection. J Infect Dis 182:12–17.
Cooksley WG. 2000. Consensus statement on the role of hepatitis A
vaccination in patients with chronic liver disease. J Viral Hepat
7:29–30.
Coppola N, Genovese D, Pisaturo M, Taffon S, Argentini C, Pasquale G,
Sagnelli C, Piccinino F, Rapicetta M, Sagnelli E. 2007. Acute
hepatitis with severe cholestasis and prolonged clinical course due
to hepatitis A virus Ia and Ib coinfection. Clin Infect Dis 44:e73–
e77.
Cuthbert JA. 2001. Hepatitis A: Old and new. Clin Microbiol Rev
14:38–58.
de Paula VS, Villar LM, Morais LM, Lewis-Ximenez LL, Niel C, Gaspar
AM. 2004. Detection of hepatitis A virus RNA in serum during the
window period of infection. J Clin Virol 29:254–259.
Guidotti LG, Borrow P, Hobbs MV, Matzke B, Gresser I, Oldstone MB,
Chisari FV. 1996. Viral cross talk: Intracellular inactivation of the
J. Med. Virol. DOI 10.1002/jmv
1326
hepatitis B virus during an unrelated viral infection of the liver.
Proc Natl Acad Sci USA 93:4589–4594.
Keeffe EB. 1995. Is hepatitis A more severe in patients with chronic
hepatitis B and other chronic liver diseases? Am J Gastroenterol
90:201–205.
Kim HW, Yu MH, Lee JH, Chang JW, Yang WS, Kim SB, Lee SK, Park
JS, Park SK. 2008. Experiences with acute kidney injury complicat-
ing non-fulminant hepatitis A. Nephrology (Carlton) 13:451–458.
Lee D, Cho YA, Park Y, Hwang JH, Kim JW, Kim NY, Lee DH, Lee W,
Jeong SH. 2008. Hepatitis a in Korea: Epidemiological shift and call
for vaccine strategy. Intervirology 51:70–74.
Locarnini S. 2000. A virological perspective on the need for vaccination.
J Viral Hepat 7:5–6.
Maier K, Gabriel P, Koscielniak E, Stierhof YD, Wiedmann KH,
Flehmig B, Vallbracht A. 1988. Human gamma interferon produc-
tion by cytotoxic T lymphocytes sensitized during hepatitis A virus
infection. J Virol 62:3756–3763.
Martin A, Lemon SM. 2006. Hepatitis A virus: From discovery to
vaccines. Hepatology 43:S164–S172.
Nainan OV, Xia G, Vaughan G, Margolis HS. 2006. Diagnosis of
hepatitis a virus infection: A molecular approach. Clin Microbiol
Rev 19:63–79.
J. Med. Virol. DOI 10.1002/jmv
Jung et al.
Sagnelli E, Coppola N, Marrocco C, Onofrio M, Scarano F, Marotta A,
Scolastico C, Catuogno A, Salzillo A, Sagnelli C, Piccinino F,
Filippini P. 2003. HAV replication in acute hepatitis with typical
and atypical clinical course. J Med Virol 71:1–6.
Shroff GR, Badve SV, Joshi AG, Desai DC, Abraham P, Sirsat RA. 2004.
Acute renal tubular injury with acute hepatitis A infection: Is it just
a coincidence? Nephrology (Carlton) 9:44–46.
Song KS, Kim MJ, Jang CS, Jung HS, Lee HH, Kwon OS, Kim YS, Choi
DJ, Kim JH, Ha SY. 2007. Clinical features of acute viral hepatitis A
complicated with acute renal failure. Korean J Hepatol 13:166–173.
Taylor RM, Davern T, Munoz S, Han SH, McGuire B, Larson AM,
Hynan L, Lee WM, Fontana RJ. 2006. Fulminant hepatitis A virus
infection in the United States: Incidence, prognosis, and outcomes.
Hepatology 44:1589–1597.
Tong MJ, el-Farra NS, Grew MI. 1995. Clinical manifestations of
hepatitis A: Recent experience in a community teaching hospital. J
Infect Dis 171:S15–S18.
Trauner M, Fickert P, Stauber RE. 1999. Inflammation-induced
cholestasis. J Gastroenterol Hepatol 14:946–959.
Zimmerman HJ, Lowry CF, Uyeyama K. 1947. Infectious hepatitis:
Clinical and laboratory features of 295 cases. Am J Med Sci
213:395–409.