Article Text

Download PDFPDF

A prospective study of mifepristone and unlimited dosing of sublingual misoprostol for termination of second-trimester pregnancy in Uzbekistan and Ukraine
  1. Ingrida Platais1,
  2. Tamar Tsereteli1,
  3. Galyna Maystruk2,
  4. Dilfuza Kurbanbekova3,
  5. Beverly Winikoff1
  1. 1 Gynuity Health Projects, New York City, New York, USA
  2. 2 Woman Health & Family Planning Charity Foundation, Kiev, Ukraine
  3. 3 Women’s Wellness Center, Tashkent, Uzbekistan
  1. Correspondence to Ingrida Platais, Gynuity Health Projects, New York City, NY 10017, USA; iplatais{at}


Introduction We aimed to assess the feasibility and acceptability of a second-trimester medical abortion regimen using mifepristone and sublingual misoprostol with no maximum number of misoprostol doses in Uzbekistan and Ukraine.

Methods This prospective open-label study enrolled 306 women with pregnancies of 13–22 weeks in Uzbekistan (n=134) and Ukraine (n=172). Women took 200 mg mifepristone at a place of their choosing, and 24–48 hours later received 400 μg sublingual misoprostol every 3 hours until the expulsion of both fetus and placenta. Study staff interviewed women about the intensity of pain, side effects, and satisfaction with the procedure. The primary outcome was the rate of complete uterine evacuation without surgical intervention or oxytocin at 15 hours after the first dose of misoprostol.

Results The median gestational age was 18.1 weeks, and half of the women (n=149, 48.9%) chose to take mifepristone at home. The majority of women (n=266, 86.9%, 95% CI 82.6% to 90.5%) expelled by 15 hours after the start of misoprostol treatment, and by 48 hours complete uterine evacuation was achieved in 296 (96.7%) women. Most women found pain (76.1%) and side effects (89.0%) to be acceptable or very acceptable, and reported high satisfaction (89.8%) with the procedure.

Conclusion Medical abortion in pregnancies of 13–22 weeks with 200 mg mifepristone followed 24–48 hours later by 400 μg sublingual misoprostol administered every 3 hours until complete expulsion is effective, safe and acceptable to women. Women can be given the option to take mifepristone at home and return to the hospital.

Clinical trials registration number,, NCT02415894, NCT02235155.

  • mifepristone
  • misoprostol
  • second trimester
  • abortion
  • home use

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Key messages

  • Medical abortion with mifepristone and an unlimited number of misoprostol doses in pregnancies of 13–22 weeks' gestation is safe, effective and acceptable to women.

  • Some women prefer to take mifepristone at home and return to the hospital.

  • For some women, an overnight hospitalisation can be avoided entirely if mifepristone is taken at home.


In Uzbekistan and Ukraine, mifepristone was registered in the early 2000s, and in 2010, the Ukrainian national abortion guidelines were revised to include a second-trimester medical abortion regimen of 200 mg mifepristone followed in 36–48 hours by 400 μg misoprostol every 3 hours up to a maximum of five doses.1 The guidelines do not permit home administration of mifepristone or provide a recommendation on actions to be taken if the expulsion does not occur after administration of the fifth misoprostol dose. In Uzbekistan, in the absence of protocols, procedures and regimens for second-trimester medical abortion varied by practitioner,; these included misoprostol only treatments and the use of other prostaglandins for induction (Dr D Kurbanbekova, personal communication, 2018). While in many settings women are hospitalised at the start of treatment with misoprostol,2 3 women in Uzbekistan and Ukraine are routinely hospitalised before taking mifepristone. Once the fetus and the placenta are expelled, providers often perform a sharp curettage to ensure all products of conception are removed, and hospital stays last several days after completion of the procedure.

Similar to the guidelines in Ukraine, some leading international medical abortion guidelines recommended 200 mg mifepristone followed in 24–48 hours by repeated ingestion of misoprostol up to a maximum of five doses.4 5 Recent research demonstrates that more than five doses of misoprostol can be used, increasing efficacy without increasing complications or adverse events.6 7 In 2016, the National Abortion Federation (NAF) and in 2017, the International Federation of Gynecology and Obstetrics (FIGO) issued guidelines that do not specify the maximum number of misoprostol doses.8 9 Studies have also found that home administration of mifepristone, which is safe and acceptable to women and providers,10 can shorten the time spent at the hospital by allowing women to be hospitalised only during treatment with misoprostol.7

We planned a prospective, open-label clinical study in Uzbekistan and Ukraine to evaluate the efficacy and feasibility of a medical abortion regimen for terminating pregnancies of 13–22 weeks using a protocol that allowed home administration of mifepristone and that did not limit the number of misoprostol doses. We hoped to provide local data to allow revision of the current national abortion guidelines in Ukraine and inform the national guidelines process in Uzbekistan.


The study was conducted from September 2014 to April 2016 at three tertiary-level hospitals in Uzbekistan (Clinic No. 2 at the Tashkent Medical Academy, Bukhara Maternity Hospital No. 1 and Samarkand Perinatal Centre) and three tertiary-level hospitals in Ukraine (Odessa Regional Hospital, Poltava City Maternity Hospital and Vinnitsa Regional Hospital). The ethical review board at the Ministry of Health of Uzbekistan approved the study for all facilities in Uzbekistan, and the institutional review board at each hospital in Ukraine approved the study protocol. Physicians and nurses at all sites were trained in the study procedures.

Women presenting at the study hospitals for termination of pregnancy with a gestational age between 13 and 22 weeks (based on menstrual history, clinical examination and/or ultrasound) were given the option to participate in the study. Eligibility criteria included willingness to follow study procedures and undergo a surgical completion if necessary. Exclusion criteria were known allergy to mifepristone, misoprostol or other prostaglandins, or any contraindication to vaginal delivery. In Uzbekistan, women with a previous transmural uterine incision were excluded from enrollment. Based on the latest recommendations,4 providers in Ukraine included women with a previous transmural uterine incision, without a limit on the number of incisions, to gain experience with this population.

After receiving counselling and signing a consent form to participate in the study, women took 200 mg mifepristone orally at the hospital or at home. Treatment with misoprostol took place at the hospital 24–48 hours after taking mifepristone. Women who chose to take mifepristone at home returned to the hospital 24–44 hours after taking mifepristone to ensure that treatment with misoprostol was started within the 24–48-hour time window. After an examination by a study physician or a nurse, women took 400 μg misoprostol sublingually. Doses of sublingual misoprostol were repeated every 3 hours until the complete expulsion of fetus and placenta unless the woman requested surgical intervention or the provider decided to stop treatment for medical reasons. If the placenta was not expelled within 30 minutes after fetal expulsion, study physicians had the option to administer an additional dose of misoprostol to aid expulsion. After each misoprostol dose, a study physician or a nurse monitored and documented women’s blood pressure, temperature and bleeding, collected women’s self-reported pain scores using a seven-level visual analogue scale, and offered pain medication (non-steroidal anti-inflammatory drugs or non-narcotic analgesics). Women were discharged from the hospital the same day as their expulsion unless they requested to prolong their hospitalisation or there was a medical need to extend it.

Medical abortion success was defined as complete evacuation of the uterus (both fetus and placenta) without surgical intervention or oxytocin. A complete uterine evacuation was determined by examination of the fetal and placental tissue or an ultrasound scan. At the time of discharge, study physicians interviewed women to assess their experience with the procedure and its acceptability. Study staff also documented unscheduled visits to the hospital after discharge.

The sample size was calculated for each country separately to provide sufficient data for the revision of the local medical abortion protocols. Based on previous research,11 12 we calculated a sample of 135 women which would allow us to detect a success rate of 91% at 15 hours with a ±5% precision (α=0.05) while assuming a 5% loss to follow-up/dropout. In Ukraine, the sample size was increased to 170 women to allow study staff to have more experience with the method. Staff at Gynuity Health Projects entered data into SPSS, version 19.13 Groups were compared using Fisher’s exact test for proportions and Mann–Whitney U-test for medians. Two-tailed p<0.05 was considered statistically significant.

Patient and public involvement

Participants were not involved in the study design or the recruitment process. Study planning included site visits to gather feedback from providers on ways to improve second-trimester medical abortion services. Study results were disseminated to study staff, other providers, and representatives from local ministries of health.


We enrolled 306 women in the study: 134 (43.8%) in Uzbekistan and 172 (56.2%) in Ukraine. Women in Ukraine had a higher median gestational age than women in Uzbekistan, and women in Uzbekistan were more likely to have been pregnant previously and to have given birth than women in Ukraine (table 1).

Table 1

Participant characteristics by country

Half of the women (48.9%) chose to take mifepristone outside of the hospital, 20.0% decided to take mifepristone at the hospital and went home before the start of the misoprostol treatment; 31.1% were hospitalised at the time of mifepristone ingestion. The median time between mifepristone ingestion and the first dose of misoprostol was 24.3 (range 18.0–60.0) hours. No woman expelled before the first dose of misoprostol. In four cases, providers mistakenly did not wait 24–48 hours after mifepristone ingestion before administering misoprostol, and in two cases women chose to take mifepristone at home and returned to the hospital to start misoprostol treatment at 56 and 60 hours, respectively. Two women requested to discontinue treatment after ingesting three misoprostol doses but did not want a surgical completion. Both women had successful abortions with complete expulsion of the fetus and placenta at 7.5 and 4.4 hours, respectively, after their third dose of misoprostol. All women were included in the analysis.

Almost all women (96.7%) had a complete abortion without an intervention. However, 10 (3.3%) women underwent an additional intervention (table 2). Two women (0.7%) received sharp curettage prior to fetal expulsion. Eight women (2.6%) received additional intervention to aid placental expulsion or to stop heavy bleeding: seven women received manual vacuum aspiration (MVA) or sharp curettage and one woman received oxytocin. Although not clinically recommended, sharp curettage was used in settings where MVA was not available.

Table 2

Complete uterine evacuation and time to expulsion from the start of treatment with misoprostol

The median time between the start of misoprostol treatment to complete evacuation of the uterus was 8.3 (range 1.3–32.3) hours, and 86.9% of women (95% CI 82.6% to 90.5%) expelled by 15 hours after the start of misoprostol treatment (table 2). Women with gestational ages of 13–18 weeks had a significantly shorter misoprostol treatment-to-abortion interval than those with gestational ages of 19–22 weeks (7.0 vs 9.7 hours, p<0.001) and required fewer misoprostol doses to complete the abortion (GA 13–18 weeks: three doses; GA 19–22 weeks: four doses, p=0.004; data not shown). The median time from fetal to placental expulsion was 10 (range 0–75) minutes; six women received an additional dose of misoprostol to facilitate placental expulsion. There was no statistical difference by gestational age group in the number of women who required additional misoprostol for placental expulsion (gestational age 13–18 weeks: four women; gestational age 19–22 weeks: two women, p=0.684). Almost all (90.2%) women required five or fewer doses of misoprostol; 9.8% needed six or more misoprostol doses to complete their abortions.

During the treatment period, the median maximum pain score that women reported was 3 (on a scale of 1–7, table 3). Women with gestational ages of 19–22 weeks reported a higher maximum pain score than women with gestational ages of 13–18 weeks (4 vs 3, p=0.004). Almost half of the women (n=141, 46.1%) took at least one dose of analgesia. Nearly all (95.1%) women reported at least one side effect associated with the abortion. Most women found the level of pain (76.1%) and side effects (89.0%) to be acceptable or very acceptable, and most (89.8%) reported a high level of satisfaction with the abortion procedure.

Table 3

Pain, side effects and satisfaction with the abortion procedure

Almost half of the women (n=139, 45.4%) were discharged from the hospital on the day of their abortion completion. Fewer women in Uzbekistan (n=56, 41.8%) had prolonged hospitalisation than women in Ukraine (n=111, 64.5%). Reasons for extended hospitalisation were: woman’s request for reasons such as long journey home and completion of the abortion late in the evening (Uzbekistan: n=47, 83.9%; Ukraine: n=93, 84.5%); provider recommended additional observation (Uzbekistan: n=4, 7.1%; Ukraine: n=15, 13.5%); provider request for additional ultrasound and/or blood test (Uzbekistan: n=2, 3.6%; Ukraine: n=4, 3.6%); and administrative reasons such as processing discharge paperwork (Uzbekistan: n=4, 7.1%; Ukraine: n=15, 13.6%). Nearly a third (27.1%, table 4) avoided an overnight stay at the hospital entirely because they were not hospitalised at the time of mifepristone ingestion and were discharged the same day as the start of their misoprostol treatment. Most women (n=58, 69.9%) who avoided the overnight stay were in the 13–18-week gestational age category.

Table 4

Length of hospitalisation

Two women returned to the hospital after discharge. One returned 1 day after discharge, and after an ultrasound examination was given additional uterotonics and antibiotics to treat lochia discharge. The other returned 5 days after discharge; she underwent an ultrasound exam,was diagnosed with haematometra and given an additional misoprostol dose. She returned a second time 8 days later with elevated body temperature and abdominal pain, and was then diagnosed with endometritis, hospitalised, and treated with intravenous antibiotics and oxytocin.


This prospective study provides further evidence that second-trimester medical abortion at 13–22 weeks gestation with 200 mg mifepristone and repeat doses of 400 μg sublingual misoprostol is safe, has a high success rate (97%), and is acceptable to women. Our findings support previous research that suggests that providers may use more than five misoprostol doses without a resting period or surgical intervention to complete the abortion.6 7 While buccal administration of misoprostol is reported in the literature, our study shows that the sublingual route can also be used in second-trimester medical abortion.7

In comparison to a similar study in Armenia,7 women in this study reported lower pain scores and fewer used analgesics. Similar to other second-trimester studies, we found that women with gestations of 19–22 weeks reported a higher level of pain, experienced longer expulsion times, and required more misoprostol doses to complete the abortion than women with gestations of 13–18 weeks.2 7 14 15 These results are particularly relevant to Ukraine and Uzbekistan, where the national guidelines require that abortion in the second trimester of pregnancy be provided only for certain medical and socioeconomic indications, such as fetal impairment or pregnancy resulting from rape. Any administrative delays in obtaining an approval for the procedure can contribute to abortions at later gestational ages, higher levels of pain, and longer procedure time.

Women in our study were given a choice to be hospitalised at the time of mifepristone ingestion or to take mifepristone at home and return to the hospital for misoprostol treatment. Having the option to take mifepristone at home gives women and providers greater flexibility when scheduling the start of misoprostol treatment. Almost a third of the women were able to avoid an overnight stay at the hospital entirely. This option provides the opportunity to decrease hospitalisation time, thus potentially making the procedure more comfortable for women and decreasing its cost. In Uzbekistan, it is common for patients from remote regions to stay at a nearby hotel before the start of a medical procedure to reduce the cost of overnight hospitalisation associated with the procedure (Dr D Kurbanbekova, personal communication, 2018). At one of the sites in Ukraine, a physician gave women the first dose (400 μg) of misoprostol to be self-administered by the woman at the hospital while she was waiting to be admitted (Dr I Kukuruza, personal communication, 2018). This approach ensured that any delays at the hospital, such as processing admissions paperwork, would not impede the start of the misoprostol treatment.

The WHO supports expanding the roles of nurses in safe abortion to increase access to abortion care.16 In Ukraine and Uzbekistan, only trained obstetricians/gynecologists are authorised to provide abortion, and usually the physician who examines and counsels women is present during the misoprostol treatment. In our study, nurses were trained in the study protocol. At some sites, nurses assessed women at the start of the treatment and actively monitored and administered misoprostol to women throughout the treatment period. By expanding the roles of all healthcare workers, this approach has the potential to increase flexibility in scheduling the start of misoprostol treatment independent of the physician’s schedule.

One limitation of this study is that study physicians interviewed women about their abortion experience, side effects, and level of pain, thus potentially introducing social desirability bias. In addition, our findings are specific to hospitals in Ukraine and Uzbekistan, where providers already have experience with provision of medical abortion in the second trimester; it may not be generalisble to other settings.

Our study supports the conclusion that second-trimester medical abortion with mifepristone administered at home, and repeat doses of sublingual misoprostol until the expulsion of the fetus and placenta is effective, safe and acceptable to women. This study’s findings were incorporated in the second-trimester medical abortion guidelines in Uzbekistan issued in November 2017. We encourage health experts in Ukraine in tandem with other organisations to revise their current abortion recommendations to make second-trimester medical abortion care more woman-centred by offering the option to take mifepristone at home, reducing the procedure time, and thus time spent at the hospital, and therefore increasing the acceptability of the method.


The authors thank all the women who participated in this study and their local partners and investigators who provided care: Drs Baxtiyor Djumanov, Gulchekhra Ikhtiyarova, Lyudmila Ivanova, Inna Kukuruza, Dilbar Najmutdinova, Svitlana Posohova and Dilfuza Sadikova. They are grateful to Ms Tatyana Lotarevich for assistance with data cleaning.



  • Contributors IP, TT and BW contributed to study design. IP, TT, GM and DK implemented the study and oversaw data collection. IP and TT led the analysis and writing of the manuscript with input from all authors.

  • Funding Financial support for this study was provided by an anonymous donor without financial interests in the outcome of this study.

  • Competing interests None declared.

  • Patient consent for publication Not required.

  • Ethics approval The study was approved by the institutional review board at each study facility in Ukraine and by the ethical review board at the Ministry of Health of the Republic of Uzbekistan.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data availability statement Data are available upon reasonable request.