A randomized trial was conducted including 287 pregnant women seeking elective abortion to compare the efficacy of misoprostol given 3, 4, or 5 days after methotrexate for abortion at ≤63 days' gestation. Subjects received 50 mg/m 2 methotrexate intramuscularly and were randomly allocated to self-administer vaginally 800 μg of misoprostol 3, 4, or 5 days after the methotrexate. The misoprostol dose was repeated 48 and 96 h later if the abortion did not occur. Outcome measures included successful abortion (complete abortion without requiring a surgical procedure), and side effects. Eighty-six cases (93%; 95% confidence interval [CI] 85%-97%) aborted in Group I; 90 cases (92%; 95% CI 84%-96%) aborted in Group II (relative risk [RR] = 1.09; RR 95% CI 0.38-3.14); and 89 (93%; 95% CI 86%-97%) cases aborted in Group III (RR = 0.97; RR 95% CI 0.33-2.87). No significant statistical differences were obtained for the success rates when misoprostol was given days 3, 4, or 5 after the administration of methotrexate (p =0.97) nor with any of the characteristics of the subjects. Complete abortion occurred in 265287 (92%; 95% CI 89%-95%) patients. Twenty-two cases (8%; 95% CI 5%-11%) resulted in failure. Side effects for methotrexate were minimal while for misoprostol they were moderate. This combination could be an alternative to surgical abortion or the use of antiprogestins and prostaglandins for medical abortion.