Background & aims: Predictors of clinical outcomes and histologic progression among patients with chronic hepatitis C and advanced fibrosis are poorly defined. We developed statistical models to predict clinical and histologic outcomes in such patients.
Methods: Baseline demographic, clinical, and histologic data from Hepatitis C Antiviral Long-Term Treatment Against Cirrhosis (HALT-C) Trial participants were subjected to multivariate analyses to determine their ability to predict clinical outcomes (ascites, spontaneous bacterial peritonitis, Child-Turcotte-Pugh score >or=7 on 2 consecutive visits, variceal bleeding, hepatic encephalopathy, and liver-related death) and histologic outcome (>or=2-point increase in Ishak fibrosis stage) during the 3.5 years of the trial.
Results: Of 1050 randomized patients, 135 had 1 or more clinical outcomes a median of 23 (range, 1-45) months after randomization. Factors associated with a clinical outcome in multivariate analyses were higher aspartate aminotransferase/alanine aminotransferase ratio, lower albumin, lower platelet count, higher total bilirubin, and more advanced Ishak fibrosis score (P < .0001). The cumulative 3.5-year incidence of a clinical outcome was 2% in the lowest and 65% in the highest risk group. Of 547 patients without cirrhosis at baseline and at least 1 follow-up biopsy, 152 had a histologic outcome. Independent variables associated with a histologic outcome were higher body mass index, lower platelet count, and greater hepatic steatosis (P < .0001).
Conclusions: In patients with chronic hepatitis C and advanced fibrosis, risk of clinical complications and fibrosis progression during 3.5 years can be predicted using baseline laboratory tests and histologic data. Our models may be useful in counseling patients and determining the frequency of monitoring.
Trial registration: ClinicalTrials.gov NCT00006164.
Copyright 2010 AGA Institute. Published by Elsevier Inc. All rights reserved.