Digestive Diseases and Sciences

, Volume 58, Issue 1, pp 275–281

Clinical and Virological Characteristics Post HBsAg Seroclearance in Hepatitis B Virus Carriers With Hepatic Steatosis Versus Those Without

Authors

    • Liver Research Unit, Department of Gastroenterology and Hepatology, Chang Gung Memorial HospitalChang Gung University College of Medicine
  • Deng-Yn Lin
    • Liver Research Unit, Department of Gastroenterology and Hepatology, Chang Gung Memorial HospitalChang Gung University College of Medicine
  • Yun-Fan Liaw
    • Liver Research Unit, Department of Gastroenterology and Hepatology, Chang Gung Memorial HospitalChang Gung University College of Medicine
Original Article

DOI: 10.1007/s10620-012-2343-9

Cite this article as:
Chu, C., Lin, D. & Liaw, Y. Dig Dis Sci (2013) 58: 275. doi:10.1007/s10620-012-2343-9

Abstract

Background

It has been suggested hepatic steatosis contributes to seroclearance of hepatitis B surface antigen (HBsAg) in chronic hepatitis B virus (HBV) infection. Although HBsAg seroclearance generally confers favorable outcome, hepatic steatosis may account for progressive liver fibrosis and cirrhosis. Further studies are needed to compare clinical and virological characteristics post HBsAg seroclearance between subjects with hepatic steatosis and those without.

Methods

One-hundred and fifty-five HBsAg carriers with HBsAg seroclearance were enrolled. Subjects with moderate–severe hepatic steatosis as diagnosed by ultrasonography were designated as having hepatic steatosis.

Results

There were 69 subjects with hepatic steatosis and 86 without. Subjects with hepatic steatosis had significantly higher body mass index (BMI; 27.8 ± 3.5 vs. 23.0 ± 3.1, P < 0.001), were more likely to be male (78.3 vs. 63.9 %, P = 0.05), and were significantly younger at HBsAg seroclearance (48.7 ± 8.9 years vs. 53.4 ± 8.9 years, P = 0.001), than those without. The frequency of anti-HBsAg seroconversion (56.5 vs. 59.3 %, P = 0.72) and HBV viremia (20.3 vs. 15.1 %, P = 0.40) was not significantly different between subjects with and without hepatic steatosis, but the incidence of abnormal AST and ALT was significantly higher in the former (23.2 vs. 0 %, P < 0.0001; and 30.4 vs. 0 %, P < 0.0001, respectively), and progression to liver cirrhosis tended to be more likely in the former than in the latter (10.1 vs. 3.5 %, P = 0.09).

Conclusions

In HBsAg carriers with increased body mass index, hepatic steatosis can accelerate HBsAg seroclearance by approximately 5 years. However, the beneficial effects of HBsAg seroclearance should be balanced against the harmful effects of hepatic steatosis.

Keywords

Body mass index, BMIHBsAg carriersHBsAg seroclearanceHBV viremiaHepatic steatosisLiver cirrhosis

Introduction

The natural course of chronic hepatitis B virus (HBV) infection constitutes three chronological phases: an initial immune tolerance phase during which the patients are positive for hepatitis B e antigen (HBeAg) and have high levels of HBV DNA but normal levels of alanine aminotransferase (ALT), followed by an immune clearance phase when the HBeAg-positive patients have decreasing levels of HBV DNA associated with raised levels of ALT, and finally the low replicative phase when HBeAg seroconverts to its antibody (anti-HBe) and ALT levels normalize [1]. Patients in the third phase are termed “inactive carriers”—they usually have serum levels of HBV DNA less than 2,000 IU/ml [1]. Although up to 25 % of these so called inactive carriers may encounter HBV reactivation and develop HBeAg-negative chronic hepatitis, most inactive carriers remain stably inactive in their lifetime [2, 3], and some eventually clear HBsAg from serum [1, 4].

Spontaneous HBsAg seroclearance in the natural history of chronic HBV infection has been increasingly recognized recently [5]. The annual incidence of HBsAg seroclearance ranges from 0.45 to 2.38 % in cohorts from Asian countries and from 0.54 to 1.98 % in cohorts from Western countries [6]. Several host and viral factors have been shown to be associated with HBsAg seroclearance in chronic HBV infection [6]. Recently, in a case–control study, we first proposed that increased body mass index (BMI) with moderate or severe hepatic steatosis may contribute to seroclearance of HBsAg [7]. Subsequently, in a longitudinal follow-up study, obesity (body mass index > 30 kg/m2) was shown to be an independent predictor for seroclearance of HBsAg [8]. The underlying mechanisms of HBsAg seroclearance associated with hepatic steatosis remain poorly understood. It is of note that a meta-analysis of seven studies involving 3,067 HBsAg-positive subjects found a strong negative association between the levels of HBV replication and hepatic steatosis [9].

Although hepatic steatosis may contribute to HBsAg seroclearance, which generally confers a favorable outcome in chronic HBV infection [10], approximately 10–20 % of patients with hepatic steatosis meet current criteria for nonalcoholic steatohepatitis, and progressive fibrosis and cirrhosis develop in up to 10 % of those with nonalcoholic steatohepatitis [11]. Further studies are required to compare clinical and virological characteristics between subjects with hepatic steatosis and those without.

Materials and Methods

This study was approved by the Institutional Review Board of the Chang Gung Memorial Hospital at Taipei, Taiwan.

Patients

Adult HBsAg carriers who visited the hepatitis B carrier outpatient clinic of Chang Gung Memorial Hospital at Taipei were registered from 1980. Patients were followed up periodically for liver biochemistry (aspartate aminotransferase (AST), ALT, total bilirubin), HBV serological markers (HBsAg, anti-HBs, HBeAg, and anti-HBe), α-fetoprotein, and liver ultrasonography every 6 months or more often if clinically indicated. The clinical and laboratory data at baseline and during follow-up were recorded, as described elsewhere [3, 5].

A consecutive series of 155 HBsAg carriers with documented HBsAg seroclearance who received regular follow-up evaluation during a 1.5-year period from January 2010 to June 2011 were enrolled in this study if they fulfilled the following criteria:

  1. 1.

    HBsAg-positive for at least 12 months before HBsAg seroclearance;

     
  2. 2.

    loss of serum HBsAg on two occasions at least 6 months apart during follow-up and remaining HBsAg negative up to the last visit [4];

     
  3. 3.

    no concomitant hepatitis C virus (HCV) or hepatitis D virus (HDV) infection;

     
  4. 4.

    no antiviral or immunomodulatory therapy before or during follow-up; and

     
  5. 5.

    no other causes of liver disease including alcoholic, drug-induced, or autoimmune liver diseases.

     

Because degrees of hepatic steatosis might vary with time, only patients who had changes in degree of hepatic steatosis ≤ one score within three years before HBsAg seroclearance and during the follow-up periods post HBsAg seroclearance were included for study. No patient in this cohort admitted intravenous drug abuse or homosexuality. Serum specimens were collected and stored at −80 °C for HBV DNA assay.

Methods

HBsAg, anti-HBs, HBeAg, ant-HBe, and antibody against HDV were assayed by radioimmunoassay or enzyme immunoassay (Abbott Diagnostics, North Chicago, IL, USA). Antibodies against HCV were tested by use of a second or third-generation enzyme immunoassay (Abbott Diagnostics). Serum specimens were assayed for HBV DNA, within six weeks of collection, by use of a commercially available, fully automated, real-time PCR assay (Cobas AmpliPrep/Cobas TaqMan HBV test; Roche Diagnostics). According to the manufacture’s instruction, serum specimens may be frozen at −20 to −80 °C for at least 6 weeks. This assay has a specificity of 100 % and sensitivity (95 % hit-rate) of 4–12 IU/ml (23–69 copies/ml) [12]. One IU is equivalent to 5.82 HBV DNA copies.

The degrees of hepatic steatosis were graded on a scale of 0–3 (0 = none, 1 = mild, 2 = moderate, 3 = severe) according to the hepatic echogenicity by liver ultrasonography, as described elsewhere [13]. Ultrasonography identified hepatic steatosis with a sensitivity of 94 % and a specificity of 84 % [14]. Diagnosis of cirrhosis was confirmed histologically or was based on clinical data and liver ultrasonography findings [15].

Statistical Analysis

Data were given as number (%) or mean ± SD. To compare characteristics between groups, either the chi-squared test or Fisher’s exact test was used for analysis of categorical variables and Student’s t test or the Wilcoxon nonparametric test was used for analysis of continuous variables. The incidence of cirrhosis post HBsAg seroclearance was estimated by use of Kaplan–Meier analysis, and the difference was determined by use of the log-rank test. Statistical procedures were performed with SAS statistical software (version 8.1; SAS Institute, Cary, NC, USA). P values of <0.05 were considered significant.

Results

Clinical and Demographic Features of HBsAg Carriers with HBsAg Seroclearance

The demographic and clinical data of the 155 study patients are summarized in Table 1. There were 109 men and 46 women. None had a previous history of acute hepatitis. The mean age (±SD) at HBsAg seroclearance was 51.3 ± 9.2 years and the mean age at enrollment in this study was 58.2 ± 8.5 years. The interval between HBsAg seroclearance and enrollment was 6.9 ± 6.1 years. At enrollment, 69 patients (44.5 %) had ultrasonographic features of moderate or severe hepatic steatosis and 86 patients (55.5 %) had no or mild hepatic steatosis. Because in our previous study the prevalence of moderate or severe hepatic steatosis was significantly higher in patients with HBsAg seroclearance than in those with HBsAg persistence whereas prevalence of mild hepatic steatosis was not significantly different between them [7], in this study patients with moderate or severe hepatic steatosis were designated as having hepatic steatosis whereas those with no or mild hepatic steatosis were designated as having no hepatic steatosis.
Table 1

Clinical and demographic data of study patients

Characteristic

Data

Number of cases

155

Male sex

109 (70.3)

Age at initial follow-up (years)

39.7 ± 8.2

Age at HBsAg seroclearance (years)

51.3 ± 9.2

Age at enrollment in this study (years)

58.2 ± 8.5

Intervals between HBsAg seroclearance and enrollment (years)

6.9 ± 6.1

Body mass index (kg/m2)

24.7 ± 3.4

Hepatic steatosis

69 (44.5)

Data are given as number (%) or mean ± SD

HBsAg, hepatitis B surface antigen

The comparison of clinical and virological characteristics post HBsAg seroclearance between subjects with hepatic steatosis and those without is summarized in Table 2.
Table 2

Comparison of clinical and laboratory data between patients with hepatic steatosis and those without

Characteristic

Without hepatic steatosis (n = 86)

With hepatic steatosis (n = 69)

P value

Age at HBsAg seroclearance (years)

53.4 ± 8.9

48.7 ± 8.9

0.001

Age at enrollment in this study (years)

59.6 ± 10.4

55.8 ± 11.7

0.001

Intervals between HBsAg seroclearance and enrollment (years)

6.8 ± 6.2

7.1 ± 6.1

0.75

Body mass index (kg/m2)

22.5 ± 3.1

27.5 ± 3.5

<0.0001

Male sex

55 (63.9)

54 (78.3)

0.05

Anti-HBsAg seroconversion

51 (69.3)

39 (56.5)

0.72

HBV viremia

13 (15.1)

14 (20.3)

0.40

AST (U/L)

19.8 ± 4.3

27.1 ± 14.51

<0.0001

Abnormal AST (>34 U/L)

0 (0)

16 (23.2)

<0.0001

ALT (U/L)

19.2 ± 5.9

36.1 ± 25.4

<0.0001

Abnormal ALT (>36 U/L)

0 (0)

21 (30.4)

<0.0001

Cirrhosis

3 (3.5)

7 (10.1)

0.09

New cirrhosisa

0 (0)

3 (4.6)

0.048

Hepatocellular carcinoma

0 (0)

0 (0)

1.0

Data are given as number (%) or mean ± SD

HBsAg, hepatitis B surface antigen; HBV, hepatitis B virus; AST, aspartate aminotransferase; ALT, alanine aminotransferase; anti-HBsAg, antibody against hepatitis B surface antigen

aExcluding seven patients (three without hepatic steatosis and four with hepatic steatosis) who had already had cirrhosis before HBsAg seroclearance

Body Mass Index, Gender, and Age at HBsAg Seroclearance

Body mass index was significantly higher in subjects with hepatic steatosis than in those without (P < 0.0001). Male gender was more frequent in the former than in the latter (P = 0.05). The mean age at HBsAg seroclearance and at enrollment in this study was significantly lower for subjects with hepatic steatosis than for those without (both P = 0.0001), but the interval between HBsAg seroclearance and enrollment was similar. Further analysis of the age at HBsAg seroclearance showed that the 10th, 25th, 50th (median), 75th, and 90th percentiles were 34.6, 42, 49, 55.8, and 60.4 years, respectively, for those with hepatic steatosis, and 44, 47, 52, 59, and 65 years, respectively, for those without hepatic steatosis (Fig. 1). Among patients with hepatic steatosis, the age at HBsAg seroclearance tended to be lower for patients with severe hepatic steatosis (n = 21; 45.9 ± 8.6 years) than for those with moderate hepatic steatosis (n = 48; 49.9 ± 9.1 years), but the difference did not reach statistical significance (P = 0.09). Both were significantly younger than those without hepatic steatosis (P = 0.005 and P = 0.03, respectively).
https://static-content.springer.com/image/art%3A10.1007%2Fs10620-012-2343-9/MediaObjects/10620_2012_2343_Fig1_HTML.gif
Fig. 1

Box plot of age at HBsAg seroclearance for subjects with hepatic steatosis (n = 69) and those without (n = 86). The 10th, 25th, 50th (median), 75th, and 90th percentiles are 34.6, 42, 49, 55.8, and 60.4 years, respectively, for those with hepatic steatosis, and 44, 47, 52, 59 and 65 years, respectively, for those without hepatic steatosis. Values above the 90th and below the 10th percentiles are plotted as points

Prevalence of Anti-HBsAg Seroconversion and HBV Viremia

The overall frequency of anti-HBs seroconversion post HBsAg seroclearance was 58.1 % (90/155), with no significant difference between subjects with hepatic steatosis and those without (P = 0.72), as shown in Table 2. The frequency of anti-HBsAg seroconversion increased progressively with increasing intervals post HBsAg seroclearance (P < 0.0001 by chi-squared test for trend). Therefore, the frequency of anti-HBs seroconversion at different times post HBsAg seroclearance was further analyzed. As shown in Table 3, the frequency of HBsAg seroconversion after different intervals from <1 year to >10 years post HBsAg seroclearance all showed no significant difference between subjects with hepatic steatosis and those without (all P > 0.2).
Table 3

Rates of anti-HBsAg seroconversion according to the intervals after HBsAg seroclearance: comparison between patients with hepatic steatosis and those without

Intervals post HBsAg seroclearance (years)

Rates of anti-HBsAg seroconversion

All (n = 155)

With hepatic steatosis (n = 69)

Without hepatic steatosis (n = 86)

P#

<1

18.5 % (5/27)*

8.3 % (1/12)**

26.7 % (4/15)***

>0.2

1–3

48.4 % (15/31)*

40 % (6/15)**

56.3 % (9/16)***

>0.2

4–6

55.8 % (21/32)*

61.5 % (8/13)**

68.4 % (13/19)***

>0.2

7–10

72.4 % (21/29)*

81.8 % (9/11)**

66.7 % (12/18)***

>0.2

>10

77.7 % (28/36)*

83.3 % (15/18)**

72.2 % (13/18)***

>0.2

HBsAg, hepatitis B surface antigen

P < 0.0001; ** P < 0.0001; *** P < 0.0001 by chi-squared test for trend; subjects with hepatic steatosis versus those without

Serum HBV DNA was detected in 27 (17.4 %) of 155 subjects post HBsAg seroclearance, with no significant difference between subjects with hepatic steatosis and those without (P = 0.40), as shown in Table 2. Of these 27, 22 (81.5 %) had levels of HBV DNA < 12 IU/ml, and only five (18.5 %) had levels of HBV DNA within the dynamic range of this assay (23, 28, 121, 2,430 and 2,770 IU/ml, respectively). The frequency of HBV viremia decreased progressively with increasing intervals post HBsAg seroclearance, but the difference was not statistically, as shown in Table 4. The frequency of HBV viremia at the different intervals post HBsAg seroclearance all showed no significant difference between subjects with hepatic steatosis and those without (Table 4).
Table 4

Rates of HBV viremia according to the intervals after HBsAg seroclearance: comparison between patients with hepatic steatosis and those without

Intervals post HBsAg seroclearance (years)

Rates of HBV viremia

All (n = 155)

With hepatic steatosis (n = 69)

Without hepatic steatosis (n = 86)

P#

<1

29.6 % (8/27)*

33.3 % (4/12)**

26.7 % (4/15)***

>0.2

1–3

19.4 % (6/31)*

20 % (3/15)**

18.8 % (3/16)***

>0.2

4–6

15.6 % (5/32)*

15.4 % (2/13)**

15.8 % (3/19)***

>0.2

7–10

13.8 % (4/29)*

18.2 % (2/11)**

11.1 % (2/18)***

>0.2

>10

11.1 % (4/36)*

11.1 % (2/18)**

11.1 % (2/18)***

>0.2

HBV, hepatitis B virus; HBsAg, hepatitis B surface antigen

P = 0.08; ** P > 0.2; *** P > 0.2 by chi-squared test for trend; subjects with hepatic steatosis versus those without

AST/ALT Levels and Risk of Liver Cirrhosis

Mean levels of AST and ALT were significantly higher in subjects with hepatic steatosis than in those without (both P < 0.0001). Whereas all subjects without hepatic steatosis had normal levels of AST (normal < 34 U/L) and ALT (normal < 36 U/L), 23.2 and 30.4 %, respectively, of those with hepatic steatosis had abnormal levels of AST and ALT (both P < 0.0001).

Liver cirrhosis was identified in seven (10.1 %) subjects with hepatic steatosis and in only three (3.5 %) of those without, but the difference was only marginally significant (0.05 < P < 0.1). Excluding seven subjects (three without hepatic steatosis and four with hepatic steatosis) who had already had cirrhosis before HBsAg seroclearance, none of the 83 without hepatic steatosis developed cirrhosis after HBsAg cirrhosis, compared with three (4.6 %) of 65 with hepatic steatosis (P = 0.048). Kaplan–Meier analysis showed that the cumulative probabilities of cirrhosis at 5, 10, and 15 years post HBsAg seroclearance were 2.3, 6.2 and 13.9 %, respectively, for patients with hepatic steatosis, and 0, 0, and 0 %, respectively, for those without hepatic steatosis (P = 0.067 by log-rank test).None of the subjects with or without hepatic steatosis had evidence of hepatocellular carcinoma based on α-fetoprotein and liver ultrasonographic examination.

Discussion

This study compared the clinical and serological characteristics post HBsAg seroclearance between HBsAg carriers with hepatic steatosis and those without. The major limitation of our study is that the diagnosis and grading of hepatic steatosis were based on liver ultrasonography. Liver biopsy is the recommended method for diagnosis of nonalcoholic fatty liver disease because it can distinguish those with hepatic steatosis alone from those with steatohepatitis and can be used to assess the stage of hepatic fibrosis.

The clinical and demographical data in this cohort, for example the male predominance, the mean age of 51.3 years at HBsAg seroclearance, and the high prevalence of hepatic steatosis, as shown in Table 1, were similar to those reported previously by us [5, 7].

There is no obvious sex dominancy in asymptomatic HBsAg carriers in Taiwan (male to female ratio is approximately 55:45), because most HBV infections are acquired in the perinatal period or early childhood with high incidence of persistent infection irrespective of gender [16, 17]. However, in this series, the male to female ratio of the HBsAg carriers with HBsAg seroclearance was approximately 70:30 (Table 1). These findings are in keeping with our previous observations in Taiwan that male HBsAg carriers were more likely to clear HBsAg from serum than female HBsAg carriers [5]. Furthermore, as shown in Table 2, among HBsAg carriers with HBsAg seroclearance, male gender is relatively more predominant in subjects with hepatic steatosis (78 %) than in those without (64 %), though the difference was only borderline significant (P = 0.05). These finding also are consistent with previous observations in Taiwan that the prevalence of overweight and/or obesity and hepatic steatosis were significantly higher in male subjects than female subjects [1820].

Perhaps the most important finding of this study is that HBsAg carriers with hepatic steatosis cleared HBsAg from serum significantly younger than those without, by a mean of 4.7 years (Table 2). None of these patients had previous history of acute hepatitis. Because most HBV infections in Taiwan are acquired in the perinatal period or early childhood [1], the age of HBsAg seroclearance is almost equivalent to the duration of HBV infection. These results provided strong evidence to support our previous postulation that increased body mass index with hepatic steatosis may contribute to HBsAg seroclearance in chronic HBV infection [7]. The underlying mechanisms by which hepatic steatosis can accelerate HBsAg seroclearance remains poorly understood. Interestingly, the HBV virus has adopted a regulatory system that is unique to the major hepatic metabolic genes. This unique virus–host interaction, mediated by metabolic events in the liver, designated the “metabolovirus model” by Shlomai et al. [21], probably explains the mechanisms of hepatic steatosis-related HBsAg seroclearance. For example, hepatic peroxisome proliferators-activated receptor α can activate fatty acid catabolism on the one hand and can also enhance HBV replication on the other hand [22]; in obese subjects with hepatic steatosis, expression of this transcriptional factor in the liver is reduced [23]; this, in turn, might result in suppression of HBV replication, as shown in a recent meta-analysis [9]. Further studies to explore the metabolic factors involved in HBsAg seroclearance in subjects with hepatic steatosis are warranted.

The overall frequency of anti-HBsAg serocoversion was 58.1 % in this series (Table 2). This figure is in the range 37–85.9 % in published reports [4, 2427]. As shown in Table 3, the frequency of anti-HBsAg seroconversion increased with the increasing intervals post HBsAg seroclearance. Further analysis revealed that the frequency of anti-HBsAg seroconversion at the different intervals post HBsAg seroclearance was not significantly different between subjects with hepatic steatosis and those without (Table 3).

The frequency of HBV viremia post HBsAg seroclearance varied substantially, from 0 to >70 %, in published reports [4, 2430]. In most previous studies, the use of in-house PCR for HBV DNA assays and study of stored serum specimens may account for such discrepancy. Our study overcame these drawbacks by using a commercially available, highly sensitive and specific, real-time PCR assay for detection of HBV DNA [12] in serum that was assayed within six weeks of collection, in accordance with the manufacture’s instruction, to avoid any potential effect of stored serum on assay accuracy, especially when the level of HBV DNA tended to be low. This study revealed that the overall frequency of HBV viremia post HBsAg seroclearance was 17.4 % (27/155) (Table 2). This figure is similar to that reported recently by us [31]. Among subjects with HBV viremia, the great majority (81.5 %) had extremely low levels of HBV DNA (<12 IU/ml) and none had levels of HBV DNA in the range for diagnosis of active hepatitis (>2,000 IU/ml) [1]. These results also showed that the frequency of HBV viremia at the various intervals post HBsAg seroclearance were also not significantly different between subjects with hepatic steatosis and those without (Table 4).

AST and ALT levels were significantly higher in subjects with hepatic steatosis than in those without (Table 2). None of the subjects without hepatic steatosis had abnormal AST or ALT levels, though 15.1 % had persistent HBV viremia (Table 2). These data suggested that the presence of low-level viremia post HBsAg seroclearance did not contribute to significant liver cell damage, if any. On the contrary, abnormal AST and ALT levels were found in 23.2 and 30.4 %, respectively, of the subjects with hepatic steatosis. Previous studies have suggested that HBV and body mass index may have synergistic effects on the progression of liver disease [32, 33]. Our results revealed that increased body mass index associated with hepatic steatosis can accelerate HBsAg seroclearance in subjects with chronic HBV infection, thus ameliorating the potential harmful effects of HBV on the liver. However, a substantial proportion had abnormal levels of AST and ALT post HBsAg seroclearance. Not unexpectedly, subjects with hepatic steatosis tended to be more likely to progress to cirrhosis than those without (Table 2). Chronic HBV infection can result in different degrees of fibrosis in the liver [34]. Pre-existing liver fibrosis because of previous HBV infection might account for the relatively high occurrence of cirrhosis in our patients with hepatic steatosis post HBsAg seroclearance. This study thus suggested that, in HBsAg carriers with overweight or obesity, the benefits of HBsAg seroclearance should be balanced against the harmful effects of hepatic steatosis.

In conclusion, HBsAg carriers with hepatic steatosis cleared HBsAg from serum significantly younger than those without, by a mean of approximately 5 years. These findings provide strong evidence to support the association between hepatic steatosis and HBsAg seroclearance in chronic HBV infection. The frequency of anti-HBs serconversion and HBV viremia was not significantly different between subjects with hepatic steatosis and those without. However, 20–30 % of the former had abnormal levels of AST and ALT, compared with none of the latter. Not unexpectedly, the former tended to be more likely to progress to cirrhosis than the latter.

Acknowledgments

The authors thank Lue LH and Shyu WC for their technical assistance. This study was supported by grants from National Science of Council of Taiwan (NSC 97-2314-B-182-019-MY2 and NSC 99-2314-B-182-029-MY3).

Copyright information

© Springer Science+Business Media, LLC 2012