This European multicenter study compares the efficacy and tolerance of the combination of pyrimethamine-clindamycin (Pyr-Cm) with the standard therapy pyrimethamine-sulfadiazine (Pyr-Sdz) for the treatment of toxoplasmic encephalitis (TE) in patients with AIDS. Two hundred ninety-nine human immunodeficiency virus-infected patients with TE were randomly assigned to receive 50 mg of pyrimethamine daily combined with either 2,400 mg of clindamycin or 4 g of sulfadiazine for 6 weeks followed by maintenance therapy with 25 mg of pyrimethamine daily with either 1,200 mg of clindamycin or 2 g of sulfadiazine. An intent-to-treat analysis showed that Pyr-Cm was less effective than Pyr-Sdz; the overall risk of progression of TE was 1.84 times higher for patients receiving Pyr-Cm therapy. There was no statistically significant difference in efficacy during acute therapy, although the rate of crossover motivated by a lack of response was higher among Pyr-Cm recipients. The difference in efficacy was evidenced during the maintenance phase of treatment; the relapse rate was twice as high among patients in the Pyr-Cm group (P = .02). The rate of side effects due to both regimens was similar, although the toxic effects of Pyr-Cm led to fewer discontinuations of therapy than did those of Pyr-Sdz (11% vs. 30%, respectively; P = .001). Pyr-Sdz appears to be the most effective treatment of TE. Pyr-Cm is a valuable alternative but is less effective for long-term prevention of relapses.