Eric Schaff


  • close

    Monitoring the Abortion Process: Home Use Of Misoprostol in the United States

    Eric Schaff, MD

    Professor of Family Medicine and of Obstetrics and Gynecology

    University of Rochester School of Medicine

    New York, USA



    The United States medical abortion experience has been unique. Frustrated with not having mifepristone, U.S. researchers explored the combination of methotrexate and misoprostol for medical abortion in the mid-1990s.  The experience was mixed because the regimen often took several weeks to be successful.  But, it provided abundant opportunity to explore different approaches to medical abortion.


    Lessons from Methotrexate Medical Abortions


    We learned early on with methotrexate abortions, that vaginal misoprostol was more effective than oral misoprostol. We also learned that repeating the misoprostol dose was necessary, safe and increased the effectiveness rates.  We had to monitor the medical abortion with ultrasound because at two weeks, despite a history of bleeding, as many as 15-20% of pregnancies had not been expelled and the decision to perform an aspiration completion depended on the results of the viability of the pregnancy by ultrasound. And lastly, we learned that women could have a medical abortion safely at home as long as they had access to emergency medical care, if needed.  Women also found home use of misoprostol highly acceptable.


    Abortion Rights Mobilization Trials


    In 1996, the Abortion Rights Mobilization (ARM), a non-profit advocacy group in New York City, contracted with the University of Rochester medical abortion researchers to coordinate trials with mifepristone.  We applied the information learned from the U.S. experience with methotrexate, i.e., use of vaginal misoprostol, home administration of misoprostol, a repeat dose of misoprostol if needed, and routine ultrasound monitoring.


    Over the following four-year period, the ARM trials have enrolled 7000 women in seven multicenter trials. These trials have recently come to a halt as the commercial product has come to market.


    Summary of ARM Trials


    In 1966, the first ARM trial used mifepristone 600 mg and vaginal misoprostol. It was an open label, non‑randomized, prospective trial at one site, up to 56 days gestation. The first follow up occurred within the first 7 days. There were 166 subjects with a 97% effectiveness up to 8 weeks.  Four percent of subjects used a second misoprostol dose, 34% bled prior to misoprostol, and 11% did not require misoprostol. There were no transfusions, but there was 1 hospitalization related to bleeding.1


    The second trial in 1997 used  mifepristone 200 mg and vaginal misoprostol and was an open label, non‑randomized, prospective trial at 7 geographically diverse sites. There were 933 subjects with a 97% effectiveness up to 8 weeks pregnant. 2% used a second misoprostol dose and there were no transfusions and no hospitalizations.2


    The third trial in 1998 extended the gestational age to 9 weeks pregnant using mifepristone 200 mg and vaginal misoprostol at 15 geographically diverse sites. Women used home misoprostol. There were 1137 subjects and  308 between 56‑63 days gestation. The regimen was 97% effective up to 8 weeks, 96% after 8 weeks, and 2% uses a second dose of misoprostol. There were one transfusion and two hospitalizations.3


    The fourth ARM trial randomized misoprostol at 1, 2, or 3 days after mifepristone 200 mg.  Again, it was an open label, prospective trial at 15 geographically diverse sites.  There were 2255 women enrolled up to 8 weeks pregnant. The regimen was 98% effective at 1 and 2 days, and 97% effective at 3 days (2%  used a second dose). There were no transfusions and 2 hospitalizations.4


    We have had another trial completed comparing mifepristone 200 mg and Oral vs. Vaginal misoprostol at 24 hours. We used two doses of oral misoprostol 400 mcg two hours apart on day 2 up to 63 days pregnant.  There were 1168 women enrolled. The effectiveness at first follow up was 89% in the oral group and 97% in vaginal group, ( p<0.001).  In oral group, the regimen was 90% effective up to 7 weeks and 88% effective between 8‑9 weeks gestation. After a second vaginal dose of misoprostol was used, the overall effectiveness, 94% in oral group, 99% in vaginal group. There was more diarrhea with oral 33% v 18% vaginal misoprostol,( p<.001 ).  There were no transfusions and no hospitalizations.


    A sixth completed trial of 1000 women is still being analyzed to determine whether experienced US clinicians report that they need routine ultrasound.. Preliminary results show that about 57% of the time, clinicians report that ultrasound is not indicated. Ultrasound was reported to be indicated in about 22% of the time and “not indicated but desired” in the remaining 21%.


    We have just finished our seventh trial which was a randomization of oral v vaginal misoprostol after mifepristone 200 mg at 48 hours up to 9 weeks pregnant. We used two doses of oral misoprostol 2 hours apart. There will be about 750 subjects enrolled. The results are pending. Low dose mifepristone and one dose of oral misoprostol 400 mcg after 49 days was not effective and the study was amended.  There was 1 transfusion. 




    In summary,


    1. A 200 mg dose of mifepristone with vaginal misoprostol is as effective as the 600 mg standard mifepristone dose.  Effectiveness rates with vaginal misoprostol have ranged from 96-98%.


    2. With the use of vaginal misoprostol, we have demonstrated high effectiveness rates above 95% up to 9 weeks gestation.


    3. A repeat dose of misoprostol in women with a persistent gestational sac at first follow up in the first week, was both safe and might improve effectiveness.  A randomized trial of a repeat dose of misoprostol in initial non-responders would help to answer this question.  Since there are few (5%) women who do not respond initially, it would require a large sample.

    4. Vaginal misoprostol provided increased flexibility in the regimen and could be given 1, 2 or 3 days after mifepristone with effectiveness rates above 95%.


    5. Home use of misoprostol is safe. Only 2 women out of 7,000 have needed an emergent aspiration curettage within 4 hours of using misoprostol, i.e., the time period when medical observation is required in France.  Four transfusions occurred in these 7000 women (<0.1%).  Of note, they all were delayed ranging from 1-3 weeks after using mifepristone. 


    6. Ultrasound dating of the pregnancy and confirming the abortion is highly accurate. It allows women to return earlier for evaluation, as soon as one day after using misoprostol, at a time when the serum hCG may still be elevated and the uterus has not fully involuted. It can also allow an early ectopic pregnancy to be detected. The French experience has demonstrated that ultrasound need not be routine.


    From these studies, we conclude that the best regimen is mifepristone 200 mg in women up to 9 weeks pregnant (by ultrasound) followed 24-72 hours later by vaginal misoprostol at home, with a repeat dose of misoprostol in initial non-responders.





    1. Schaff EA, Stadalius L, Eisinger SH, Franks P. Vaginal misoprostol administered at home

    after mifepristone for abortion. J Fam Pract 1997;44:353‑360.2.


    2. Schaff EA, Eisinger SH, Stadalius LS, Franks P, Gore BZ, Poppema S. Low‑dose mifepristone

    200 mg and vaginal misoprostol for induced abortion. Contraception 1999; 59:1‑6.


    3. Schaff EA, Fielding, SL, Eisinger SH, Stadalius LS, Fuller L. Low‑dose mifepristone

     Followed by vaginal misoprostol at 48 hours for abortion up to 63 days. Contraception

    2000; 61:41‑ 46.


    4. Schaff EA, Fielding SL, Westhoff C, Ellerston C, Eisinger SH, Stadalius, LS, Fuller L.

    Vaginal misoprostol administered 1, 2, or 3 days after mifepristone for early medical abortion (< 56 days gestation): a randomized trial. JAMA  2000;284(15):1948‑1953.