Progression to cirrhosis and mortality of chronic hepatiita type B: a cohort study with 25 years follow-up in Italy.

The long-term liver-related morbidity and mortality of untreated Caucasian adult patients with chronic hepatitis (CH) B deserve further evaluation, particularly in the light of the possible use of antiviral agents. We conducted a longitudinal study to elucidate the long-term survival and prognostic factors in a cohort of Italian patients with hepatitis B surface antigen (HBsAg) positive CH. Methods: We re-evaluated, a cohort of 105 untreated patients (76% men; mean age 30±9 years; 64% hepatitis B e antigen (HBeAg) positive) with CHB without histological and clinical evidence of cirrhosis at presentation (1). We retrieved clinical, histological and ultrasound examinations, biochemical and virologic tests, and cause of death. Results: During a median period of 25.5 years 45 (43%) patients became inactive carriers, whereas the remaining 60 (57%) showed persistent alanine aminotransferase (ALT) elevation: 13 (12%) associated with HBeAg persistence, 32 (30.5%) with detectable serum HBV-DNA but HBeAg negative (HBeAg negative CH), 11 (10.5%) with concurrent hepatitis C virus and/or hepatitis delta virus infection and 4 (4%) with concurrent non alcoholic fatty liver disease. Twenty one (47%) inactive carriers cleared HBsAg. Cirrhosis developed in 32 (30.5%) patients (histologically documented in 30 and clinical in 2), 1 to 28 years after entry into the study with an incidence rate of 13.26 per 1000 person-years. Older age, male sex, absence of sustained remission and sustained HBV replication predicted cirrhosis occurrence independently. Liver-related death occurred in 7 (7%) patients, caused by HCC in 6 and liver failure in 1. None of the inactive carriers died of liver related causes. The 25-year survival probability was 87% in the 60 patients with active hepatitis and 80% in the subgroup of 45 patients with sustained HBV replication (HBeAg positive or HBeAg negative CH). Survival probability at 25 years was lower in patients who developed cirrhosis (79%) as compared to patients without cirrhosis (99%) (p = 0.0004). Cox proportional regression model showed that cirrhosis occurrence during follow-up was significantly associated with an increased risk of liver related death. The adjusted hazard ratios (95% Confidence Interval) for liver related death was 12.86 (1.48-111.7) for patients who developed cirrhosis relative to those without cirrhosis. Conclusions: In Caucasian patients with CH without cirrhosis at presentation progression to cirrhosis is relatively slow, but cirrhosis occurrence strongly predict disease-related mortality.