The purpose of this study was to compare the outcome and side effects of using the drugs methotrexate and misoprostol, alone or in combination, to induce abortion. A total of 108 subjects who had requested elective termination elf pregnancy and medical abortion at 9 weeks gestation or less were randomized into three groups. The first group received 50 mg/m(2) intramuscular (IM) methotrexate on day 1 and, if the hCG level had risen by >50% of the initial level on day 4, a second dose was given. They were then followed-up at weekly intervals up to day 21. Group 2 received 800 mu g vaginal misoprostol on day 2 and, if ultrasound showed a gestational sac on day 4, they received a repeat dose and were re-examined on day 7. Group 3 received 50 mg/mg(2) methotrexate intramuscularly followed 3 days later by 800 mu g vaginal misoprostol and were re-examined on day 7. Complete abortion occurred in 25 (69%) of the 36 subjects in group 2, 22 (58%) of the 36 subjects in group 2, and 32 (89%) of the 36 subjects in group 3. The complete abortion rate in group 3 was significantly higher than that of both group I and group 2 (p < 0.05). The incomplete abortion rate was significantly higher in group 2 as compared with both of the other groups (p < 0.05). There were significant differences between the mean gestational age of the successful abortions and the failures in group 2 (no abortion occurred at more than 49 days gestation), but not in groups 2 or 3. Vaginal bleeding in subjects who successfully aborted began within 16 +/- 4 days in group 1 after the first dose, and within 24 h in 18 (86%) of the 21 subjects in group 2 and 27 (84%) of the 32 subjects in group 3 after the misoprostol dose. The drugs caused no serious or prolonged side effects. The combination of methotrexate and misoprostol is a more effective abortifacient regimen than when either drug is used alone. (C) 1999 Elsevier Science Inc. All rights reserved.