Medical regimens for abortion at 12 weeks and above: A systematic review and meta-analysis

Abstract Background Mifepristone and misoprostol are recommended for second trimester medical abortion but consensus is unclear on the ideal regimen. Objectives To systematically review randomized controlled trials (RCTs) investigating efficacy, safety, and satisfaction of medical abortion ≥ 12 weeks' gestation. Data sources We searched PubMed, Popline, Embase, Global Index Medicus, Cochrane Controlled Register of Trials, and International Clinical Trials Registry Platform from January 2008 to May 2017. Study eligibility, participants, and interventions We included RCTs on medical abortion ≥ 12 weeks' gestation using mifepristone and/or misoprostol. We excluded studies with spontaneous abortion, fetal demise, mechanical cervical ripening, and those not reporting ongoing pregnancy (OP). Study appraisal and synthesis methods After extracting pre-specified data, and assessing risk of bias in accordance with the Cochrane handbook, we used Revman5 software to combine data and GRADE to assess certainty of evidence. Results We included 43 of the 1894 references identified. Combination mifepristone-misoprostol had lower rates of OP (RR 0.12, 95% CI 0.04–0.35) versus misoprostol-only. A 24-hour interval between mifepristone and misoprostol had lower OP rate at 24 hours than simultaneous dosing (RR 3.13, 95% CI 1.23–7.94). Every 3-hour dosing had lower OP rate at 48 hours (RR 0.39, 95% CI 0.17 to 0.88). Limitations Direct comparisons of buccal misoprostol to sublingual or vaginal routes after mifepristone were limited. Evidence from clinical trials on how to best manage women with prior uterine incisions was lacking. Conclusion Our analysis supports the use of mifepristone 200 mg one to two days before misoprostol 400 mcg vaginally every 3 hours at ≥ 12 weeks' gestation. Implications Where available, providers should use mifepristone plus misoprostol for second trimester medical abortion. Vaginal misoprostol appears to be most efficacious with fewest side effects, but sublingual and buccal routes are also acceptable.