Natural history of chronic hepatitis B virus infection: What determines prognosis after cirrhotic decompensation?

Chronic hepatitis B virus (HBV) infection affects more than 400 million people globally, of whom 75% are Asians. The natural history is complex and includes four phases: (i) immune tolerance; (ii) immune clearance; (iii) low or non-replication; and (iv) reactivation.1 The worst clinical outcomes include liver cirrhosis, hepatic decompensation and hepatocellular carcinoma (HCC). However, the occurrence of liver disease and rates of disease progression in persons with chronic HBV infection are influenced by factors that include the host, the virus and the environment. Further, outcomes and progress rates differ according to the different phases of infection/host response.2–4 Clinical studies have improved our understanding of the transition of chronic HBV infection to either cirrhosis or HCC. Disease progression in the immunotolerant has also been evaluated,2 and guidelines for diagnosis and treatment of chronic hepatitis B have been developed and updated based on the outcome of these studies. Accordingly, surveillance and intervention methods for treatment and management of HBV infection for the Asia–Pacific region and elsewhere have been recommended.1,5 However, the natural history of HBV-related cirrhosis and prognostic factors after the first episode of hepatic decompensation is less well known. Cirrhosis, the most important sequelae of chronic HBV infection, eventually progresses from the compensated to the decompensated phase, the latter being characterized by ascites, variceal bleeding, encephalopathy and/or jaundice.6,7 Every year, 5–7% of cases of HBV-related cirrhosis transition from the compensated to decompensated stage.5 Once the decompensated stage begins, the outcome is poor. In one study, 30 of 93 patients died after developing esophageal varices (nine cases of bleeding, nine liver failure, one sepsis, five HCC and six other) during 131 months (median survival, 20 months).8 Several other studies have reported median survival times after the appearance of ascites of 2–3 years.6,7 Overall, the probability of survival has ranged 55–70% at 1 year, and 14–28% at 5 years once decompensation occurred, with the highest survival rate occurring among patients presenting with ascites (50% at 3 years, 38% at 5 years).9,10 Lowest survival rates are observed in patients with more than one complication (35% at 3 years and 23% at 5 years).9 Guidelines for clinical practice are based on natural history, but details of the natural history and the prognostic factors associated with disease progression after the onset of ascites remain to be elucidated. One problem is that the majority of studies to date have included patients with a variety of diseases, such as alcohol liver cirrhosis or hepatitis C cirrhosis; hepatitis B cirrhosis patients represented a small fraction of the total.6,11 Therefore, more specific additional study of the natural history of HBV infection is needed. In this issue of The Journal of Gastroenterology and Hepatology, Dr Kim and colleagues highlight the natural history and prognosis of patients with HBV-related cirrhotic ascites among patients hospitalized to control ascites.12 Two hundred and three consecutive patients with HBV-related cirrhotic ascites were followed for a median of 81 months after the first significant episode of ascites requiring hospitalization. Thirty-one (19%) of these developed gastrointestinal variceal bleeding after a median of 8.0 months, 53 (32%) developed hepatic encephalopathy at 9.0 months, 24 (15%) developed spontaneous bacterial peritonitis (SBP) at 13 months, five (3.0%) developed hepatorenal syndrome (HRS) at 8.1 months and 10 (6.1%) developed HCC at 21 months. The median survival was 12 months and the 1- and 3-year survival rates were 51% and 19%, respectively. Child–Pugh class, serum white blood cell (WBC), serum creatinine (Cr), serum sodium (Na) and antiviral therapy were the independent factors that influenced survival. The marked difference between HBV-related cirrhosis and cirrhosis due to other causes is the fact of HBV infection itself. Long-term follow-up studies showed high HBV DNA viral load at the baseline and this persistence of high viral load during follow up is associated with increased risks of cirrhosis and HCC. In their article, Dr Kim and colleagues show clearly that lamivudine therapy can influence the outcome in patients after the appearance of ascites, and this is in agreement with earlier research by Villeneuve, Schiff and others.13,14 In Schiff's large open-label, multicenter study, 226 wait-listed and 241 post-liver transplant patients with lamivudine-resistant HBV were treated with adefovir dipivoxil for a median duration of 39 and 99 weeks, respectively. Treatment with adefovir dipivoxil markedly inhibited HBV replication, the mean decrease from baseline serum HBV DNA at week 48 being 3.5 log10 copies/mL. After 48 and 96 weeks of adefovir dipivoxil, serum HBV DNA became undetectable in 59% (45/76) and 65% (13/20) of patients. What is more, concomitant with inhibition of HBV replication, hepatic function improved with normalization of serum chemistries and improvement in functional indices, such as albumin, bilirubin and prothrombin time, as well as Model for End-Stage Liver Disease (MELD) and Child–Pugh scores. In light of the results from Dr Kim's study and the previous studies by Villeneuve and Schiff, it is reasonable to speculate that continued HBV replication promotes decompensation of cirrhosis. Therefore, HBV plays an important role across HBV infection from the very beginning to the end stage of liver disease, including onset of decompensated cirrhosis. It should be noted, however, that 35% of the wait-listed patients in Schiff's study still had detectable serum HBV DNA after 96 and 144 weeks of adefovir dipivoxil treatment. This implies that early antiviral therapy may be necessary to prevent disease progression;for cases treated too late, it may be impossible to obtain efficacy. Future randomized control trials could provide more information on when to start the therapy and what nucleotide analogs to use. The reason to raise these questions is that in some countries not all cirrhosis patients are allowed to be treated by antiviral nucleotide analogs, as pointed out in Dr Kim's paper. Ideally, however, all HBV DNA-positive patients (2000 IU/mL for compensated, and detectable for decompensated) with HBV-related cirrhosis should be recommended for antiviral therapy, regardless of alanine aminotransferase level.1 Nevertheless, despite the influence of antiviral therapy on survival, Dr Kim and colleagues also found that detectable HBV DNA did not influence survival statistically. This result is different from several previous studies on chronic hepatitis B.15 The reason for this difference might be the uncontrolled nature of the study. The authors suggest that hospitalized patients with ascites indicating very poor liver function, could contribute to this apparent discrepancy from other studies. One important finding in the current article by Dr Kim is that most cirrhotic complications, such as gastrointestinal variceal bleeding, hepatic encephalopathy, SBP, and HRS are likely to develop around 1 year after ascitic decompensation. This indicates that patients with ascites should be carefully screened to determine the optimal time for liver transplantation. Additionally, the lack of a statistically significant correlation between detectable HBV DNA and survival should be further explored through a control study. Patients enrolled in Dr Kim's study had ascitic decompensation of liver function, as opposed to previous studies that included patients with chronic hepatitis, and this difference could have affected the results. Thus, the findings of the present research, that Child–Pugh class, WBC, Cr and Na were the factors additional to antiviral therapy that primarily influenced survival at the time of ascites onset is consistent with the importance of liver function in determining survival. A large multicenter trial in patients with hyponatremia supports the result for serum sodium as a prognostic factor, and there is an inverse relationship between serum sodium concentration and frequency of encephalopathy, SBP and HRS.16 In summary, Dr Kim and colleagues' report provides a report of consecutive studies of the natural history of hepatitis B cirrhosis following the first episode of ascites. The results confirm that most HBV-related cirrhotic patients with ascitic decompensation have a very short survival. The report also highlights the effectiveness of antiviral therapy in treating many patients with HBV-related decompensated cirrhosis; it is found that antiviral therapy improves survival. These results can be used to design and carry out meaningful case-control or cohort studies that will help further describe the role of ascites volume and sensitivity to diuretics for prognosis in chronic HBV infection. Randomized controlled trials can also be done to develop treatment paradigms that achieve early and continued suppression of serum HBV DNA, in renewed attempts to either prevent the occurrence of decompensated cirrhosis or impede its progression.