Five RCTs and 17 cohort studies were included in the review. One RCT scored 3 and one RCT scored 5 on quality, the remaining two RCTs scored 10 . The 17 cohorts received a quality score between eight and 10. Where reported, the median follow-up duration was 60 months (range 32 to 128.4 months) in studies of patients with hepatitis C and 74.5 months (range 48 to 86 months) in patients with hepatitis B.
Patients with Hepatitis C (four RCTs, 13 cohort studies; 12,459 patients):
The median cumulative incidence of hepatocellular carcinoma was statistically significantly reduced in patients receiving antiviral therapy at three years (-2.6%, 95% CI -1.2% to -4.0%) and five years (-7.8%, 95% CI -4.6% to -11.0%; 17 studies; random-effects model). Meta-regression identified the following factors as important predictors of efficacy at five years: RCT study design, higher quality, screening at least once every six months, and presence of histologically proven cirrhosis. Other results were discussed in the review.
Patients with Hepatitis B (one RCT, seven cohort studies; 3,438 patients):
Antiviral therapy statistically significantly reduced the median cumulative incidence of hepatocellular carcinoma at three years (-2.8%, 95% CI -0.5% to -5.1%) and five years (-7.1%, 95% CI -4.1% to -10.2%). Meta-regression identified the type of antiviral agent as a predictor of efficacy.
No details on heterogeneity were reported. There was evidence of publication bias according to funnel plot in patients with hepatitis C but not hepatitis B (funnel plots not presented).