To evaluate the effect of interferon and the benefit of prednisolone pretreatment in Oriental patients with chronic active hepatitis B, 120 male Chinese patients were randomly allocated to receive: 1) group A: a 4-week course of prednisolone followed by 2 weeks of no treatment and then a 12-week course of human lymphoblastoid interferon, 4 to 6 MU/m2 intramuscularly; 2) group B: as group A, but with placebo given instead of prednisolone; 3) group C: an 18-week course of placebo. Clearance of serum hepatitis B virus-DNA and HBeAg (complete response) was achieved in 21% of group A, 5% of group B and none of group C at the end of therapy (A vs B: p = 0.054; A vs C: p < 0.01). When assessed 12 months after the end of therapy, the complete response rate was 46% in group A, 24% in group B and 25% in group C (p < 0.05). Those with baseline alanine transaminase < or = 200 U/l showed a better response to interferon following prednisolone withdrawal (48%) than with interferon therapy alone (20%, p = 0.056) and no treatment (9%, p < 0.01). Those with a baseline serum hepatitis B virus-DNA < or = 1000 pg/ml also showed a higher complete response rate when pretreated with prednisolone (59%) than when treated with interferon alone (29%, p = 0.084) or untreated (22%, p < 0.03). The strongest independent predictor of a response to treatment was prednisolone withdrawal (p < 0.05). None of the responders lost hepatitis B surface antigen.(ABSTRACT TRUNCATED AT 250 WORDS)