An open, randomised, multicentre trial was conducted to evaluate the efficacy of thrice-weekly versus daily therapy with sulfadiazine-pyrimethamine in the prevention of relapses of toxoplasmic encephalitis in HIV-infected patients. Between February 1994 and July 1997, 124 patients with HIV infection were enrolled after resolution of the first acute episode of toxoplasmic encephalitis treated with sulfadiazine-pyrimethamine. Patients were randomly assigned to receive either a daily regimen consisting of sulfadiazine (1 g) twice a day plus 25 mg pyrimethamine and 15 mg folinic acid daily (n = 58), or a thrice-weekly regimen consisting of the same doses of sulfadiazine and folinic acid plus 50 mg pyrimethamine (n = 66). After a median follow-up period of 11 months (range 1-39 months), no differences were found in the incidence of toxoplasmic encephalitis relapses between the groups, there being 14.9 episodes per 100 patient-years (95% CI: 2.8-20.2) in the daily-regimen group versus 14.1 episodes (95% CI: 2.3-17.2) in the intermittent-regimen group. The estimated cumulative percentages of relapse at 12 months were 17% and 19%, respectively (P = 0.91). In a Cox multivariate analysis, not taking antiretroviral therapy was the only variable independently associated with relapse (adjusted risk ratio: 4.08; 95%CI: 1.32-12.66). Baseline CD4+ cell counts, prior AIDS, mental status, sequelae and allocated maintenance therapy regimen were not independent predictors of relapse. No differences were observed in the survival rate (P = 0.42), or in the incidence of severe adverse effects (P = 0.79). The efficacy of the thrice-weekly regimen was similar to that of the daily regimen in the prevention of relapses of toxoplasmic encephalitis. Administration of antiretroviral therapy was the only factor associated with a lower incidence of relapse.