Does methotrexate confer a significant advantage over misoprostol alone for early medical abortion? A retrospective analysis of 8678 abortions


B Winikoff, Gynuity Health Projects, 15 East 26th Street, Suite 1617, New York, NY 10010, USA.


Objective  The objective of this study was to compare efficacy for four medical abortion regimens used in one clinic setting: (1) misoprostol alone, (2) oral methotrexate + buccal misoprostol, (3) oral methotrexate + vaginal misoprostol, and (4) intramuscular methotrexate + vaginal misoprostol.

Design  Retrospective analysis of data from clinical records.

Setting  An anonymous women’s health centre in Latin America, providing medical abortion services since 2001 in a highly restrictive setting.

Population  A total of 8678 women with gestations <56 days, who sought a medical abortion between April 2002 and December 2004.

Methods  Chi-square test was performed to compare patient characteristics by abortion outcome (success/failure). The impact of selected variables on method success was explored through logistic regression. A second regression analysis was conducted with a subsample (n= 4022), for which data on parity and previous abortion(s) were available.

Main outcome measure  Abortion outcome (success/failure) at 2-week follow up.

Results  Success rates for the three methotrexate regimens ranged from 81.7 to 83.5% and did not differ significantly; misoprostol-alone regimen had a success rate of 76.8%. Efficacy was significantly higher for the three combined methotrexate regimens compared with misoprostol alone and remained so in the multivariate model (OR = 1.35). In the final regression, lower gestational age, being nulliparous, and having no previous abortions were positively correlated with method success.

Conclusions  In this real-use setting, methotrexate appears to confer a significant advantage over misoprostol alone for early medical abortion. This finding is important for settings where mifepristone remains unavailable. Additional factors such as gestational age limits and patient preference should be considered in regimen selection.