Seventy-seven liver transplant candidates were enrolled in a multicenter study in which patients were treated with lamivudine (100 mg daily) without the adjunctive use of hepatitis B immune globulin. Treatment was begun while patients awaited liver transplantation and continued after transplantation. All were hepatitis B surface antigen (HBsAg) positive, and 61% had detectable hepatitis B e antigen (HBeAg) and HBV DNA when treatment was begun. Forty-seven underwent liver transplantation and 30 did not. Median study participation was 38 months (range, 2.7-48.5) in the transplanted patients and 26 months (range, 0.1-37) in the nontransplanted group. Twenty-five of 42 (60%) transplanted patients with 12 or more weeks of posttransplantation follow-up were HBsAg negative at the last study visit. At treatment week 156, 13 of 22 (59%) remained HBsAg negative, and all 9 reinfected patients were HBV-DNA positive before treatment. In the nontransplanted patients, HBeAg was initially detectable in 20 of 27 (74%) but this decreased to 3 of 17 (18%) after 104 weeks of treatment, and significant improvement in biochemical parameters was observed. HBV-DNA polymerase mutants were detected in 15 (21%) and 6 (20%) of the transplanted and nontransplanted patients, respectively. When compared with historical cohorts, lamivudine-treated patients appeared to have improved survival, and transplanted patients had a decrease in the rate of recurrent HBV infection. Lamivudine therapy was partially effective in preventing recurrent HBV infection when given before and after transplantation. Thus, future trials using a combination of HBIg and lamivudine are needed to assess the optimal prophylactic therapy.