The comparative efficacy of prednisolone followed by interferon alfa (IFN-alpha) versus IFN-alpha alone in enhancing the rate of antibody to hepatitis B e antigen (anti-HBe) seroconversion has not been evaluated in a large cohort of white children. To determine this, a multicenter-controlled trial was conducted in 95 hepatitis B virus (HBV)-DNA/hepatitis B e antigen (HBeAg)-positive children (median age, 9 years [range, 2-16 years]; 56 boys; 84 [89 percent] white), all having inflammatory changes on liver biopsy. Patients were randomized to receive either prednisolone followed by IFN-alpha (n = 34); placebo followed by IFN-alpha (n = 30); or no treatment (n = 31). The prednisolone/placebo was given on a double-blind basis. Lymphoblastoid IFN-alpha was given at 5 MU/m(2) three times a week for 12 weeks. Baseline clinical, biochemical, and histological features were similar for the three groups. The majority (85 percent) had a baseline aspartate aminotransferase (AST) level < or = 100 IU/L. On follow-up between 12 and 18 months (median, 15 months) after treatment, the loss of HBeAg with anti-HBe seroconversion was more common in patients pretreated with steroids (12 of 34 [35 percent]) or placebo [12 of 30 (40 percent)] as against controls (4 of 31 [13 percent], P< .05). Factors predictive of anti-HBe seroconversion were baseline HBV-DNA concentration of < or = 1,000 pg/mL and a greater degree of portal tract inflammation on pretrial biopsy. Our results show that in white children treatment with IFN-alpha, at the dose and duration used in this study, improves the rate of anti-HBe seroconversion. Steroid priming does not potentiate the effect of IFN-alpha.