Acceptability of misoprostol-only medical termination of pregnancy compared with vacuum aspiration: an international, multicentre trial


  • The original trial is registered as an International Standard Randomised Controlled Trial, number ISRCTN10531821.

MI Rodriguez, Department of Reproductive Health and Research, World Health Organization, 20 Avenue Appia, 1211 Geneva, Switzerland. Email


Please cite this paper as: Rodriguez M, Seuc A, Kapp N, von Hertzen H, Huong N, Wojdyla D, Mittal S, Arustamyan K, Shah R, the World Health Organization Research Group on Post-Ovulatory Methods of Fertility Regulation. Acceptability of misoprostol-only medical termination of pregnancy compared with vacuum aspiration: an international, multicentre trial. BJOG 2012;119:817–823.

Objective  To compare women’s acceptance of misoprostol-only medical termination of pregnancy (TOP) with surgical TOP.

Design  Prospective cohort study.

Setting  Termination of pregnancy clinics in New Delhi, Mumbai, Hanoi, Tbilisi, Trivandrum and Yerevan.

Population  Women requesting TOP, at 63 days of gestation or less, at study sites where both medical and surgical methods were available.

Methods  Serial surveys eliciting measures of women’s satisfaction and acceptance of TOP method were administered. Data were analysed using cross-tabulation and logistic regression to determine if TOP method was predictive of acceptability.

Main outcome measures  Patient acceptance.

Results  High acceptability of both surgical and misoprostol-only TOP.

Conclusions  Where medical TOP with mifepristone is not available, misoprostol-only medical TOP is acceptable to women who have the choice between medical or surgical techniques.


Access and patient preference are integral components of quality comprehensive care for termination of pregnancy (TOP).1,2 Both surgical and medical methods of TOP are effective and safe, however, their access is limited in many areas. Medical TOP with a regimen of mifepristone–misoprostol is widely recognised as a noninvasive, effective means of terminating early pregnancies.3 Unfortunately, drug registration, cost and supply issues restrict availability of mifepristone.4,5 In areas where mifepristone is unavailable, misoprostol-only regimens for medical TOP are recommended and widely used.4–10

We previously conducted a multicentre, randomised controlled equivalence trial to determine the most effective route of administration and dosing interval for medical TOP using misoprostol alone.4 The primary outcome was complete TOP rates at 2-week follow up. Complete TOP rates reported from the study ranged between 78% and 85% depending on the route and dosing interval.5 In addition to determining the efficacy of misoprostol-only medical TOP, we sought to assess the acceptability of the procedure. Data were collected on acceptability among participants at time of enrolment, after the TOP, and at 2-week follow up.

Women often have strong preferences in their TOP care.1,2,11 In studies comparing acceptability of medical TOP using combined mifepristone and misoprostol with surgical TOP, a range of factors influencing preferences has been elicited.12 Of women who prefer surgical TOP, the cited factors include: predictable time to completion, requirement of fewer visits, association with fewer adverse effects, shorter duration of bleeding, and that the procedure can be performed without witnessing blood or products of conception. Women who elect to have a medical TOP when a choice exists cite fear of surgery or anaesthesia, desire for a natural process and a greater degree of perceived privacy with the procedure.2

No published studies have assessed the acceptability of and patient preferences for first-trimester, misoprostol-only medical TOP, among women with a choice between surgical and medical TOP. The ability to adequately study acceptability of misoprostol-only medical TOP has previously been limited by uncertainty regarding the optimal route and dosage.5,7,8,13,14 Such evidence could help to guide protocol development and improve patient counselling. The aim of this report is to evaluate the acceptability of surgical versus medical TOP using misoprostol alone.


We conducted a multi-country study to compare efficacy, adverse effects and women’s perceptions of two intervals and two routes of administration of misoprostol-only medical TOP. A previously published report details the efficacy and adverse effects of the various misoprostol regimens.4 In this report, we evaluate the acceptability of misoprostol-only medical TOP compared with vacuum aspiration. Data from six international centres where women had the choice between misoprostol-only medical TOP and vacuum aspiration were used for this analysis. Standardised questions were used to assess acceptability at multiple time-points.

All participants provided written informed consent before enrolment. Institutional review boards at all participating hospitals and the World Health Organization (WHO) Secretariat Committee on Research Involving Human Subjects gave ethics approval.

Clinics included in the study were in New Delhi, India; Mumbai, India; Hanoi, Vietnam; Tbilisi, Georgia; Trivandrum, India; and Yerevan; Armenia—sites where both medical and surgical options were available, TOP was legal and more than 500 women were seen annually for first-trimester TOP. We sought to obtain regional balance and to study the issue in developing countries where access to mifepristone was not available. Women requesting TOP services were screened for study eligibility (Figure 1).

Figure 1.

 Study enrolment.

Women were eligible for inclusion if they were generally healthy, had attained the legal age of consent, had a single intrauterine pregnancy of 63 days or less, were willing to return for follow up, and accepted surgical completion if necessary.

Women were excluded from study participation if they had evidence of any past or present major medical illness such as anaemia with haemoglobin <95 g/l, uncontrolled asthma, liver disease, haemolytic disorders, thromboembolism, mitral stenosis or hypertension. Women with an intrauterine device, a uterine scar or anomaly, or who smoked more than 20 cigarettes per day were also excluded from the trial.

Approximately one-third of women who were found to be eligible for the study elected to undergo a surgical TOP procedure. Surgical TOP was performed using electric or manual aspiration. Anaesthesia used during the procedure ranged from none to conscious sedation. The remaining women underwent medical TOP with misoprostol. Enrolment in the study was the women’s only access to medical TOP at that site. Participants who chose medical TOP in the study were randomised to three doses of 800 μg misoprostol in one of the following four administrative route–interval combinations: sublingual repeated every 3 hours, vaginal every 3 hours, sublingual every 12 hours, vaginal every 12 hours. The first dose of misoprostol was administered to all women in the clinic, after which they were observed for 3 hours. Women had the choice of self-administering at home, the second dose in the 12-hour group and the third dose in the 3-hour group. All women electing to self-administer were seen at the hospital the following morning for evaluation of outcome and for administration of the third dose in the 12-hour group. Baseline characteristics including age, ethnic background, marital status, education, and gynaecological and obstetric history were recorded.

A standardised questionnaire was administered to all women immediately after the procedure and at 2-week follow up. The completion of a medical TOP was assessed with a pelvic examination and ultrasound (if indicated after pelvic examination) 3 hours after the final dose of misoprostol. Information collected at baseline included women’s previous knowledge of medical TOP and the source of their information about medical abortion. Data on women’s satisfaction and TOP experience were recorded after the procedure, and at 2-week follow up. Specifically, women were asked if they were satisfied with their TOP experience, if expectations were met, how this TOP compared with any previous TOPs (if applicable), which method they would choose in the future and which procedure they would recommend to a friend. Responses were tabulated by TOP method and compared.

A sample size of 2100 was calculated for the initial efficacy phase of the study (Figure 1). With this sample size, in the acceptability analysis, we would be able to detect as significant a difference in the acceptability percentages of (at least) 50% and 60% with a confidence level of 0.95 and 80% power. Comparative analysis of baseline demographics was conducted using a chi-square test for categorical variables and t test for quantitative variables. Acceptability was assessed both immediately following the procedure and at the 2-week follow up. Logistic regression was used to evaluate whether TOP technique was predictive of patient acceptability immediately after the procedure.

We fitted logistic regression models for the following measures of acceptability dichotomised to yes or no answers: if the woman was satisfied; if she would recommend the method to a friend; if she needed a future TOP, whether she would choose the same method, and satisfaction with the current method as compared with previous TOP experiences.

All four of the main outcome variables were assessed following the procedure. Women were considered satisfied if they would use the same method of TOP in the future, or would recommend it to a friend. In all cases, the logistic regression allowed for the estimation of the effect of the TOP technique on acceptability to be adjusted by age (four age groups), previous pregnancy (no/yes), previous TOP (no/yes) and centre. Data were analysed centrally at the World Health Organization in Geneva, Switzerland. Statistical analyses were conducted with spss v18.0, (SPSS Inc., Chicago, Illinois, USA) and logistic regressions accounted for centre as a random effect.


For this study of acceptability, a total of 2739 women consented to be included in the study. Of these, 400 were excluded from the analysis because there were no data for the comparison group. In the final analysis, 2339 were included with 1282 in the medical TOP group and 1057 in the surgical TOP group (Figure 1). Baseline characteristics were similar between the two groups with respect to ethnicity and education level (Table 1). Women choosing surgical TOP were more likely to be married, 88% as compared with 77% (P < 0.001), and to be slightly older (28.4 versus 26.6 years, P < 0.001). With respect to obstetric history (Table 1), women requesting medical TOP were more likely to have had no previous pregnancy (29% versus 19%, P < 0.01) or live birth (13% versus 4%, P < 0.001) and were less likely to have reported using contraception at time of conception (28% versus 37%, P < 0.001).

Table 1.   Characteristics of study participants
 Medical abortionSurgical abortion P
n Mean (SD) or % n Mean (SD) or %
  1. Chi-square tests performed on all categorical variables; Student’s t test for continuous variables.

Age (years) 128226.6 (5.21)104428.4 (6.03)<0.001
Education (years) 128211.8 (4.48)105312.0 (4.82)0.312
Ethnic group
Asian or Black90970.971868.20.169
Marital status
Using contraception at time of conception 35928.039037.0<0.001
No previous pregnancy 36828.720019.0<0.001
Prior pregnancy but no live birth 11412.5364.3<0.001
Prior delivery 80087.580995.7
History of induced medical abortion 495.4121.6<0.001
History of induced surgical abortion 35639.142852.1<0.001
Gestational age (days)

Acceptability was measured at two time points, immediately following the TOP and at 2-week follow up. With respect to measures of acceptability, the majority of women were satisfied with their TOP, both immediately following it and at 2-week follow up. Women choosing medical TOP reported being ‘highly satisfied’ or ‘satisfied’ 93% of the time, compared with 75% of women requesting surgical TOP (Table 2). Acceptability persisted over time. Among women receiving medical TOP, 84% were as satisfied, or more satisfied, at the 2-week follow up. In the surgical group, 96% of women were as satisfied, or more satisfied, at 2-week follow up.

Table 2.   Acceptability of abortion procedure and adverse effects at 2-week follow up
 Medical abortionSurgical abortion
n % n %
Satisfaction with method
Highly satisfied30724.414013.5
Not satisfied473.712412.0
No opinion423.313613.2
More than expected11215.628459.0
More than expected12133.74115.7
Adverse effects
More than expected7415.24013.5
Less than expected237.34927.8
Takes too long26821.0343.2

Regarding adverse effects, 16% of women receiving a medical TOP reported experiencing more pain than they had expected, contrasting with 59% in the surgical TOP group. Bleeding was reported as ‘more than expected’ by 34% and 16% of the medical and surgical TOP groups, respectively. In the medical TOP group, 21% of women felt that the procedure took too long, compared with 3% in the surgical TOP group. Only 7% of women in the medical TOP group described experiencing less privacy than they had expected compared with nearly one-third of women (28%) in the surgical group.

Subanalyses of acceptability assessments were performed among women who had a previous experience with either method in comparison with their most recent procedure (Table 3). The only significant differences noted were that women who had a prior medical TOP and chose surgical TOP during this study were significantly more likely to be ‘more satisfied’ with the surgical procedure (36% versus 10%, P < 0.001). The corollary was found among women who had previous experience with surgical TOP but chose a medical TOP during this pregnancy. This cohort was significantly more likely to report being more satisfied with their current TOP experience than those who chose a repeat surgical procedure (51% versus 24%, P < 0.001).

Table 3.   Acceptability of current abortion compared with previous experience at 2-week follow up
Acceptability measureMedical abortionSurgical abortion P value*
n % n %
  1. Chi-square tests performed on all categorical variables; Student’s t test for continuous variables.

  2. *‘other’, ‘missing’ and ‘no opinion’ were excluded from computed P values.

Prior medical abortion
More satisfied with current49.5836.4<0.001
As satisfied with current3378.629.1
Less satisfied with current24.8731.8
No opinion37.1522.7
Prior surgical abortion
More satisfied with current17450.910823.9<0.001
As satisfied with current11433.329665.5
Less satisfied with current185.3122.7
No opinion3610.5368.0

Logistic regression models were run for four indicators of acceptability: the TOP method a woman would choose in the future; what she would recommend to a friend; how satisfied she was; and satisfaction in comparison with previous experience (if applicable). As potential confounders we used in all cases age (four groups), previous pregnancy (no/yes), previous induced TOP (no/yes), and study centre as a random effect (Table 4). We also analysed results with centre as a fixed effect. No difference was noted in the treatment group results, when centre was considered as a random or fixed effect.

Table 4.   Acceptability of procedure as estimated by what a woman would choose in the future
Treatment groupAcceptability measures*
OR (95% CI)
  1. *Acceptability assessed immediately following procedure.

  2. p1: statistical significance of the predictor determined by comparing the full model with the model excluding the corresponding predictor. Results adjusted for centre as a random effect.

  3. Acceptability measures: 1. ‘If you needed another abortion, which method would you choose?’; 2. ‘Which method of abortion would you recommend to a friend?’; 3. ‘How satisfied are you?’; and 4. ‘How satisfied are you in comparison with previous experience?’ (if applicable). Sample size of 323.

  4. **No previous pregnancies are considered as no previous induced abortions. Includes medical and/or surgical.

Medical abortion43.2 (30.1–61.9)122.1 (81.2–183.5)8.5 (5.6–3.2)2.6 (0.5–13.2)
Surgical abortion    
Age group (years)
15–19 (reference category)    
20–241.266 (0.584–2.742)1.507 (0.685–3.316)1.401 (0.555–3.533)2.871 (0.304–27.13)
25–291.116 (0.494–2.520)1.469 (0.642–3.362)2.046 (0.766–5.462)2.852 (0.258–31.539)
30+1.535 (0.665–3.542)1.782 (0.768–4.134)2.724 (2.724–0.982)6.43 (0.672–61.527)
Pregnant before
No (reference category)    
Yes0.966 (0.610–1.528)0.835 (0.515–1.355)0.119 (0.077–0.184) 
Previous induced abortion**
No (reference category)    
Yes4.577 (3.135–6.682)4.479 (3.049–6.581)1.464 (0.585–3.663) 

Overall, women with the choice between misoprostol-only medical and surgical TOP were highly satisfied with misoprostol-only medical TOP. When acceptability was assessed by which method a woman would choose in the future, medical TOP was predictive of satisfaction, with an odds ratio (OR) of 43.2 (Table 4; 95% CI 30.1–61.9); the observed percentages were 96% and 53% in the medical and surgical TOP groups, respectively. Examining acceptability by whether a woman would recommend the method she used to others was also predictive of acceptability among the misoprostol-only medical TOP group. Women able to choose between the two methods were significantly more likely to recommend the medical method to their friend (Table 4; OR 122.1, 95% CI 81.2–183.5). The observed percentages were 97% and 38% in the medical and surgical TOP groups, respectively. High satisfaction persisted among the misoprostol-only medical TOP group through our third measure of acceptability. When asked ‘How satisfied are you?’ having had a medical TOP was predictive of satisfaction compared with surgical TOP (Table 4; OR 8.5 95% CI 5.5–13.2).

Finally, we examined the important question of whether the type of TOP a woman chose was predictive of acceptability among a subset of women who had experienced both methods of TOP. No significant difference existed in satisfaction between surgical TOP and misoprostol-only medical TOP (Table 4; OR 2.6, 95% CI 0.5–13.2).


This study examines the acceptability of misoprostol-only medical TOP as compared with surgical TOP. Misoprostol-only TOP, although less effective than a combined medicine regimen, is widely accessible and used for early TOP where mifepristone is not available. Women who had a previous suboptimal experience with TOP valued having the opportunity to choose another method. Those women who experienced both methods of TOP and had a choice of methods were equally satisfied with either medical TOP with misoprostol or a surgical procedure. These results demonstrate that misoprostol-only medical TOP is acceptable and preferred by many women when compared with surgical TOP. As women value choice in their TOP care, both methods should be offered to women who meet medical eligibility criteria.

We assessed acceptability at two different time points to investigate how acceptability might change for each TOP method over 2 weeks. At the 2-week follow up, high acceptability persisted for both methods. As the process of completing a misoprostol-only medical TOP takes longer than a surgical TOP, demonstrating that acceptability continues throughout the process, over time, is important. In addition to demonstrating the high acceptability of misoprostol-only medical TOP for women, these results highlight the importance of providing information, counselling and adequate pain control for surgical TOP. Anaesthesia for surgical TOP was not standardised among the study centres, and ranged from non-use to conscious sedation. Nearly 60% of women undergoing a surgical TOP found the procedure to be more painful than anticipated, emphasising the need for preoperative counselling and adequate procedural analgesia. Additionally, 28% of women undergoing surgery reported less privacy than expected. This highlights an important area for improvement in centres offering surgical TOP. Thorough counselling to prepare women for the amount of bleeding following a misoprostol-only medical TOP is also important to improve their experience and ameliorate anxiety, as indicated by our data (Table 3).

Our study is not without limitations. Women were not randomised to TOP method, which may introduce selection bias. A low follow-up rate of 48% at the 2-week follow up among the surgical TOP group may introduce bias into our data. It is possible that acceptability varies based on cultural factors for which our study could not account, for example, one centre in India reported 100% rates of satisfaction. Additionally, as women only had the option of medical TOP at the health centre by enrolling in the study, our results may inflate the participants’ acceptance of the misoprostol-only regimen from the inherent bias in the selection procedures. Importantly, we were not able to separately evaluate satisfaction in women who had a failed medical TOP and required subsequent surgery. It is not known how complications impact acceptability in our study.


Our research demonstrates that misoprostol-only medical TOP is a valued choice and highly acceptable for women in areas where mifepristone is not available. The ability to choose between medical TOP and surgery was valued by women for a range of individual reasons. Health systems without access to mifepristone should offer women the option of medical TOP using misoprostol alone.

Disclosure of interests

We declare that we have no conflict of interest.

Contribution to authorship

MIR, AS and NK developed the data analysis plan and AS analysed the data. HH designed the randomised controlled trial with the WHO Post-Ovulatory Fertility Regulation Research Advisory group. NTMH, DW, SM, KA and RS all provided valuable comments on the manuscript.

Details of ethics approval

Institutional review boards at all participating research sites and the WHO Secretariat Committee on Research Involving Human Subjects gave ethics approval.


The study was funded by the UNDP/UNFPA/WHO/World Bank Special Programme of Research, Development and Research Training in Human Reproduction.


We thank the primary investigators of the research sites for their contribution to this research: Drs Rajaeskharan Nair, Radnaabazar Erdenetungalad, Nguyen Duc Vy, Archil Khomassuridze, Nguyen Thi Ngoc Phuong and To Minh Huong. Dr Vinita Goyal’s assistance with preliminary reports on this project is gratefully acknowledged.