The natural history of hepatitis C infection acquired through injection drug use: Meta-analysis and meta-regression

Background & Aims: Our aim was to estimate the rate of progression to cirrhosis for those infected with hepatitis C virus (HCV) through injection drug use. Methods: We searched the published literature for articles assessing cirrhosis in this population and abstracted data on cirrhosis prevalence, mean duration of infection, mean age, mean alanine aminotransferase (ALT) enzyme levels, proportion of males, proportion HIV co-infected, proportion consuming excessive alcohol, and study setting. Summary progression rates were estimated using weighted averages and random effects Poisson meta-regression. The impact of co-variates was assessed by estimating the posterior probability that the relative risk (RR) of progression exceeded 1.0. Results: A total of 47 published articles were identified. After adjusting for covariates in 44 studies representing 6457 patients, the estimated rate of progression to cirrhosis, was 8.1 per 1000 person-years (95% credible region (CR), 3.9-14.7). This corresponds to a 20-year cirrhosis prevalence of 14.8% (95% CR, 7.5-25.5). A 5% increase in the proportion of male participants and a 5% increase in the proportion consuming excessive alcohol were associated with faster progression (probability RR >1 = 0.97 and 0.92, respectively). A 5% increase in the proportion of HIV co-infected, an increase in ALT of 5 IU/L and studies in settings with a high risk of referral bias were not associated with faster progression (probability RR >1 = 0.42, 0.65, and 0.43, respectively). Conclusions: Analysis of aggregate level data suggests that for patients who contracted HCV through injection drug use prognosis is poor in populations with many male patients and high levels of alcohol consumption. (C) 2010 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved.