International Journal of Gynecology and Obstetrics (2007) 98, 222226

CLINICAL ARTICLE

Comparison of misoprostol and manual vacuum aspiration for the treatment of incomplete abortion
C. Biquea, M. Usta, B. Deboraa, E. Chongb,, E. Westheimerb, B. Winikoffb
a Jos b Gynuity

available at

Macamo Hospital, Maputo, Mozambique Health Projects, New York, NY, USA

Received 19 March 2007; received in revised form 30 April 2007; accepted 2 May 2007

KEYWORDS

Abstract Objective: To compare the safety, efficacy, and acceptability of misoprostol versus manual vacuum aspiration (MVA) for treatment of incomplete abortion in Maputo, Mozambique. Methods: A total of 270 women with clinically diagnosed incomplete abortions of up to 12 weeks of gestation were randomized to either 600 g oral misoprostol or MVA. Women were followed-up seven days later to evaluate whether the abortion was complete. Results: Success was high for both MVA and misoprostol groups (100% vs 91%, P=0.002). Women in the MVA arm reported fewer side effects but higher pain scores. Women who received misoprostol were significantly more likely to be very satisfied with the treatment and willing to choose the method again. Conclusion: Although oral misoprostol was less effective than MVA in this study, it was more acceptable to women. Misoprostol is well-suited for use in low-resource settings, and should be promoted as an option for the treatment of incomplete abortion. 2007 International Federation of Gynecology and Obstetrics. Published by Elsevier Ireland Ltd. All rights reserved.

Misoprostol; Incomplete abortion; Mozambique; Miscarriage

1. Introduction
Early pregnancy failure causes millions of women to seek emergency medical services. Estimates based on figures for the year 2000 indicate that of 210 million pregnancies worldwide, approximately 32 million ended in miscarriage and stillbirth, and another 46 million ended in induced
Corresponding author. 15 East26th St., #1617, NY NY 10010, USA. Tel.: +1 212 448 1230; fax: +1 212 448 1260. E-mail address: echong@gynuity.org (E. Chong).abortion [1].

Nearly half of induced abortions (19 million) were

considered to be unsafe (performed by unskilled providers or under unhygienic conditions, or both) and resulted in 67,900 deaths [2]. In Mozambique, abortion is available only under restricted conditions, thus many women attempt to terminate their pregnancies clandestinely with traditional practitioners, and later present to health care facilities with an incomplete abortion. Under a 1981 decree from the Ministry of Health, abortion is permitted in hospitals in cases of endangerment of health and contraceptive failure, and requires a committee to confirm whether legal grounds exist [3]. Although the decree has

0020-7292/$ see front matter 2007 International Federation of Gynecology and Obstetrics. Published by Elsevier Ireland Ltd. All rights reserved. doi: 10.1016/j.ijgo.2007.05.003

Comparison of misoprostol and manual vacuum aspiration for the treatment of incomplete abortion

been interpreted liberally in some hospitals (where abortion services are provided to women who sign a written statement), access is mostly limited to older, urban, higher income women [4,5]. A 20022003 assessment of abortion-related services in 45 public health facilities across Mozambique found that nearly all providers of uterine evacuation services used sharp curettage (98%), and smaller percentages reported the use of manual vacuum aspiration (51 %) or electric vacuum aspiration (20%) in their facilities [6]. The use of sharp curettage is discouraged by the World Health Organization (WHO), as it is considerably more painful for women than vacuum aspiration, and less safe (rates of major complications are two to three times higher than those of vacuum aspiration) [7,8]. A growing body of research has shown that the prostaglandin E1 analogue, misoprostol, offers a simple and noninvasive alternative to surgical treatment for early pregnancy failure. In addition, studies have found that treatment with misoprostol is considerably more effective than expectant management after one week [9,10]. Initially registered for the prevention of gastric ulcers during long-term use of nonsteroidal anti-inflammatory drugs, misoprostol is now used worldwide for numerous obstetric or gynecologic indications including induction of labor, preparation of the cervix for surgical procedures, prevention or treatment of postpartum hemorrhage, and pregnancy termination [11]. Previous studies of the effectiveness of misoprostol in evacuating the uterus in cases of early pregnancy failure have reported a wide range of success rates, in part due to the type of pregnancy failure being studied, the duration of the waiting period before the outcome is assessed, and the criteria of success. Among studies conducted since 2000 with sample sizes of 100 or greater, misoprostol completed the abortion process in 46 96% of participants [9,10,1219]. The best regimen of medical management is still under investigation, however the highest success rates have been achieved using 400 g vaginal misoprostol [20], 600 g vaginal misoprostol with repeat dose if needed [21], and 600 g oral misoprostol [16,17]. Few studies testing the use of misoprostol for incomplete abortion have been conducted in developing countries, and even fewer have been carried out in subSaharan Africa [9,17,21,22]. This study was designed to fill a gap in the literature by testing oral administration of misoprostol for incomplete abortion, in a hospital setting in a low-income country with restrictive abortion policies. 2. Materials and methods
The study was conducted at Jos Macamo Hospital, a mediumsized tertiary hospital located on the outskirts of Maputo, Mozambique. A total of 270 women who presented with an incomplete abortion, and gave their consent, were enrolled in the study and randomized to receive either 600 g oral misoprostol or manual vacuum aspiration (MVA) (the current surgical standard of care at the hospital). Since ultrasound is not routinely used in Mozambique and pregnancy tests are not widely available in government facilities, incomplete abortion was diagnosed based on past or present history of vaginal bleeding during pregnancy and an open cervical os. Women were eligible for the study if they had a confirmed incomplete abortion with uterine size equivalent to no more than 12 weeks

of gestation, were 18 years of age or older, lived or worked within the hospital's geographic area of coverage, had no known allergy to misoprostol, no signs of severe infection, and no hemodynamic disturbance. Ethical approval for this study was obtained from the Western Institutional Review Board and the Maputo Health Directorate. The informed consent form was provided in Portuguese; for those women who were illiterate, the form was carefully read aloud and explained to the participant by the nurse investigator. In a few cases the consent form was explained in the participant's local language for better understanding. The randomization scheme was generated by computer and the treatment allocations printed on cards and inserted into sequentially numbered opaque envelopes. Once a woman's eligibility was confirmed and the consent form signed, a member of the study staff opened the next envelope in the sequence and assigned the woman to the indicated treatment group. Women randomized to the misoprostol group swallowed the pills in the presence of the study investigator and remained in the clinic for approximately one hour of observation. For women randomized to the MVA group, a nurse-midwife with more than 5 years' experience with the method performed the MVA approximately 10 min after randomization. In contrast to similar studies of incomplete abortion in Uganda and South Africa (which used intramuscular pethidine and general anesthesia, respectively), women undergoing surgical evacuation received only verbal anesthesia (as well as kanamycin) [17,21]. Verbal anesthesia is the current standard practice at the hospital, as providers have found they are able to reduce women's anxiety by explaining the procedure thoroughly and by providing continuous reassurance during the surgery. Paracetamol was provided as needed for pain management to women in both groups. Follow-up visits were scheduled for seven days after the initial visit, and women were given a telephone number at which they could reach the study investigators with questions or in case any complications should arise. Women were asked to return to the hospital after one week to confirm the status of the abortion. If the abortion was still incomplete, women were given the option of waiting an additional week to see if the abortion would complete spontaneously, or immediately undergo a surgical evacuation. For those who elected to wait, they were evaluated again seven days later and, if there was still clinical evidence of substantial retained products of conception in the uterus, they underwent a surgical evacuation at that time. Prior to being discharged from the study, all women answered a series of questions about the acceptability of the treatment and its side effects. No monetary compensation was provided to the women, but those who attended the follow-up visit received a kapulana, a multipurpose cloth (worth approximately US $4) that is commonly used by women in Mozambique. The primary outcome of interest was the percentage of women in each treatment arm who experienced a complete abortion without recourse to additional surgical intervention. Secondary outcomes of interest included experience of side effects and acceptability of the treatment. This study was designed to be a feasibility study as it employed a treatment for incomplete abortion that was new to Mozambique. The sample size of 270 was chosen to allow staff to gain expertise in the new method and to allow comparisons to be drawn between the two treatments. However, in the process of monitoring the first 20 cases, it was noted that the randomization scheme was not being

Comparison of misoprostol and manual vacuum aspiration for the treatment of incomplete abortion

Misoprostol (n=123) M V A (n=124) ge, year completions in the misoprostol appropriately followed. Therefore, the process of randomization Table 2 Clinical outcome group, 5 were at the women's 25.8? 6.0 and how women were to be assigned to study arms was request, 4 were deemed medically necessary (due to Table 3 Side effects reported M VA 25.4?6.0 infection), and 1 was Misoprostolfor =123) conducted (n the provider's value Tablethoroughly reviewed and the study a was re-started. The 23 1 Participant characteristics P conveducation, year (n = 124) women Misoprostol (n=111)=101)P value enrolled prior toMthe A(n V discovery of the randomization error nience. Of the 5 women who requested surgical completion, 8.0?2.4 were excluded from the study analysis, although a review of Lost decided not to wait an additional 18.5 (23) requested 9.8 (12) 3 to follow-up week and 7.6?2.7 these data Side effects reported:yielded success rates consistent with that of the full Returned for at their scheduled follow-up (101) once the completion follow-up 90.2(111) 81.5 visit 0.067 arity study. Heavy bleeding 35.1 (39) 2.0 (2) <0.001 Success determined that their abortion was still (101) 0.002 91.0 (101) 100.0 incomplete. provider 1.5?1.5 bleeding Normal The sample95.5 (106) 35.6 (36)<0.001 who were not Failure (after excluding the 23 women 9.0 to the 0 . 0 The other two women returned (10) clinic 12 days after 1.5?1.5 Spotting randomized) is large enough to detect(94) or greater decrease in 82.9 (92) 93.1 a 6% 0.035 3.6 their 0.0 arried misoprostol administrationinitial treatment (before(4) scheduled follow-up medically necessary after efficacy with misoprostol, assuming a 99% success rate for MVA Nausea 11.7 (13) 2.0(2) 0.006 E E 21.5 (26) visit) presenting with heavy bleeding and cramping and Vomiting procedures [23] (one-tailed, = 0.05, = 0.2). 0.061 4.5 (5) 0.0 4.5 (5) 0.0 19.7 (24) E E requested afor patient's preference surgical surgical completion. Pain/cramps 99.1 (110) 100.0(101)1.000 The study forms were completed by staff at the study site and ad previous induced abortion At the follow-up visit, the provider completed with each Fever 25.2 (28) 0.0 <0.001 reviewed by the principal investigator for completeness and data 0.9 (1) 0.0 19.5 (24) Chills 84.7 (94) 0.0 < and data E E woman a table provider's preference she experienced and surgical for recording the side effects quality. Forms were sent to New York, USA0.001 were entered 21.0 (26) what days she experienced them. Women in the misoprostol into SPSS 14.084.4 (92) 98.0 (99) < and all ad previous spontaneous (SPSS, Chicago, IL, USA), 0.001discrepancies were abortion resolved by referral to the original study forms. Chi-square tests Abbreviations: MVA, manual vacuum aspiration. additional drugs following procedure 13.0 (16) Values were used to analyze categorical data, and t tests were used for are given as percentage (number) unless otherwise indicated. 9.7 (12) continuous current abortion was induced (also s reported byscore woman, data 2.63using SPSS). Mean pain 4.21 <0.001 50.4 (62)vacuum aspiration. <0.001 manual de effects characterized as: 49.2 (61) mean ? SD or 3. Results (number) unless otherwise indicated. percentage stically significant differences between(38) provider's opinion, woman's (94) Easily tolerable 85.5 abortion was induced 38.8 participants in the two groups. 50.4 (62) Tolerable Between December 2004 and January 2006, 270 women were 10.0 (11) 61.2 (60) enrolled in the study. As discussed above, 23(66) 53.7 women were not Bad 3.6 (4) 0.0

randomized and were not included in the analysis. A total of

tions: MVA, manual vacuum aspiration. to the misoprostol group, and 123 women were randomized e given as percentage (number) unless otherwise indicated. 124 to the MVA group. Participant characteristics are

shown in Table 1. The two groups did not differ significantly on background characteristics; participants had a mean age of 25 years, were predominantly unmarried, and about one out of five reported previously having an induced abortion. Approximately half of participants told the provider that their current abortion was induced, which coincided with providers' estimates of the proportion of pregnancies that were interfered with. Thirty-five women were lost to follow-up after the initial visit, 12 from the misoprostol group and 23 from the MVA group. Success rates for the misoprostol and MVA groups were 91.0% and 100%, respectively (Table 2). Of the 10 surgical
Misoprostol (n=111) M V A (n=101)

value Satisfaction with treatment:

<0.00 1 Very satisfactory 86.5 (96) 36.6 (37) Satisfactory 9.9 (11) 63.4 (64) Unsatisfactory 3.6 (4) 0.0 Would choose method again

<0.00 1 Yes

Comparison of misoprostol and manual vacuum aspiration for the treatment of incomplete abortion
Not sure 4.5 (5) 24.0 (24) No Table 4 Satisfaction with treatment method

68.0 (68)

e (number) unless otherwise indicated. a These women had a surgical completion after taking misoprostol. Worst features (asked for up to 2)

group1.8(2)significantly more likely than women in the MVA were group to experience the majority of side effects reported, 8.0(8) especially heavy bleeding (35.1% vs 2.0%, P<0.001), fever (25.2% vs 0%, P<0.001), method to a friend Would recommend and chills (84.7% vs 0%, P<0.001) (Table 3). These side effects, in addition to cramping, nausea, and vomiting, are all known and expected effects <0.00 1 of misoprostol and women are counseled about them. InterestYes ingly, although women in the misoprostol group experienced 93.7 (104) a greater number of side effects than the MVA group, they were more likely to characterize these side 69.3 (70) easily effects as tolerable (85.5% vs 38.8%, P<0.001 ).While they experienced Not sure fewer side effects, women in the MVA group experienced 4.5 (5) a greater degree of pain: their reported mean pain score 26.7 (27) was significantly higher (4.2 vs 2.6, P<0.001), and a higher percentage of women were given additional pain medication No following the procedure (98.0% vs 84.4%, P<0.001). Pain 1.8(2) level was measured with a seven-point Likert scale; a visual 4.0(4) analog scale was developed by the study team containing seven circles of increasing size, with the smallest circle Best features (asked for up to 2) representing no pain at all and the largest representing the worst pain imaginable. Women were asked to select the circle that best represented the level of pain they felt from the treatment. surgery Avoided Women (91) more than twice as likely to express being 82.0 were very0.0 satisfied with misoprostol than with MVA (86.5% vs 36.6%, P<0.001) (Table 4). More than 90% of women receiving misoprostol would choose the method again or Little pain recommend the method to a friend, whereas approximately 12.6 (14) one-fourth of women in the MVA group were not sure they 0.0 would take either of those actions. When asked to name up to two best and worstbleeding of the treatment, most Small amount of features women in(41)misoprostol group were happy to have avoided 36.9 the surgery (82.0%); common worst features 93.1 (94) fever/ included chills (71.2%) and cramps (64.0%). Women who underwent MVA appreciated the small amount of bleeding involved postQuick treatment procedure (93.1%) and the rapidity of the treatment (98.0%), 5.4 (6) but disliked most the pain (99.0%) and cramps (64.4%) they 98.0 (99) experienced. al vacuum aspiration.

commonly mentioned worst features by women receiving misoprostol, so this may not be a spurious finding. One particularly notable aspect of this study is the extremely limited use of ultrasound. As is the standard practice in Mozambique, clinical examination is the primary method of diagnosis, and ultrasound is reserved for cases where there is doubt on the success of the treatment or complications that need to be confirmed. While perhaps some women were incorrectly diagnosed as having an incomplete abortion that required treatment or had a surgical completion that was not necessary, the high success rate in the misoprostol group and the lack of adverse outcomes suggests that the use of misoprostol for treatment of incomplete abortion can be safe and effective without use of ultrasound, which is often less available in low-resource settings and requires skilled providers to conduct the examination. The main weakness of the study is the 14% of women who did not return for their follow-up visit. This is somewhat to be

4. Discussion

This study demonstrates that 600 g oral misoprostol, while Pain (related to the slightly less effective than MVA, is nonetheless a highly 4.5 treatment for incomplete abortion in a loweffective (5) a 99.0 (100) resource setting. The high success rate of the misoprostol group is consistent with other studies in low-resource countries using a 600 g dose [9,16,17,21]. Despite experiMVA procedure) encing more side effects, women receiving misoprostol reported feeling less pain and expressed a higher satisfaction with the procedure than those in the MVA group. This sizeable h e r p a i in satisfaction between the two groups is O t difference n (abdominal, stomach, pelvic) not typically found in the literature and it is possible that the 16.2 (18) providers in the study, pleased with the success of (2) new 2.0 this technology, may have over-reported women's satisfaction with misoprostol. Cramps Women (71)the misoprostol group reported experiencing in 64.0 fever and chills at a higher rate than that found (65) in similar 64.4 studies. This may reflect provider bias since providers filled out the table of side effects with the women at the follow-up visit,Fever/chills were more likely to prompt women in the perhaps they 71.2 (79) misoprostol group to recall these symptoms than those in the MVA0.0 group. On the other hand, fever and chills were the most
Bleeding 27.0 (30) 0.0

Nausea/vomiting 4.5 (5) 0.0

C. Bique et al.

expected as the study site was a tertiary-level hospital, and cultural norms dictate that one visits the hospital only in an emergency. We can surmise that women did not return because their treatment was successful, although it is also possible that they may have sought care elsewhere. There was greater attrition from the MVA group than the misoprostol group; this may be attributable to women's assumptions regarding the effectiveness of surgery, or may reflect women's discomfort during the treatment and their subsequent reluctance to return for follow-up care. From this study it is clear that oral misoprostol offers a safe and simple alternative to surgical treatment of incomplete abortion, and is well-suited for use in a country such as Mozambique. The health system is comprised of many small primary care facilities dispersed throughout the rural areas, where more than 70% of the country's population lives. These small units do not have the capacity or the equipment to offer manual vacuum aspiration, but with proper training, could potentially treat incomplete abortions with misoprostol. Making this simple and effective treatment widely available could avert numerous complications and adverse outcomes experienced by women in these settings who are often left to seek medical care outside the formal healthcare system. As an added incentive, recent analyses indicate that misoprostol is a less costly option than surgical evacuation for women presenting with early pregnancy failure [24,25]. Future work should include lobbying the government to include misoprostol in its essential medicines list, and to ensure an adequate supply is available across the country.

References
1] 2] Alan Guttmacher Institute (AGI). Sharing responsibility: women, society and abortion worldwide. New York: AGI; 1999. p. 7. Ahman E, Shah I. Unsafe abortion: global and regional estimates of the incidence of unsafe abortion and associated mortality in 2000. 4th ed. Geneva: World Health Organization; 2004. p. 13. United Nations Population Division. Abortion Policies: A Global Review, vol. II. New York: United Nations; 2002. p. 149. Hardy E, Bugalho A, Faundes A, Duarte GA, Bique C. Comparison of women having clandestine and hospital abortions: Maputo, Mozambique. Reprod Health Matters 1 997;9: 10815. Dgedge M, Gebreselassie H, Bique C, Victorino MT, Gallo MF, Mitchell EMH, et al. Confronting maternal mortality: the status of abortion care in public health facilities in Mozambique. Chapel Hill, NC: Ipas; 2005. p. 3. Ibid, 11. Grimes D, Schulz KF, Cates W, Tyler CW. The Joint Program for the Study of Abortion/CDC a preliminary report. In: Hem W, Andrikopolous B, editors. Abortion in the seventies. New York: National Abortion Federation; 1977. p. 416. Grimes DA, Cates W. Complications from legally-induced abortion: a review. Obstet Gynecol Surv 1979;34:17791. BagrateeJS, KhullarV, Regan L, MoodleyJ, Kagoro H. A randomized

3] 4] 0]

5] 1]

controlled trial comparing medical and expectant management of first trimester miscarriage. Hum Reprod 2004;19(2):26671. 2] Blohm F, Friden BE, Milsom I, Platz-Christensen JJ, Nielsen S. A randomized double-blind trial comparing misoprostol or placebo in the management of early miscarriage. BJOG Aug 2005;1 12(8):10905. 8] Winikoff B. Pregnancy failure and misoprostol time for a change. PAC in Action, vol. 9; 2006. p. 13. 9] Pandian Z, Ashok P, Templeton A. The treatment of incomplete miscarriage with oral misoprostol. BJOG Feb 2001 ;1 08(2):21 34. 10] PangMW, LeeTS, Chung TK. Incomplete miscarriage: a randomized controlled trial comparing oral with vaginal misoprostol for medical evacuation. Hum Reprod Nov 2001 ;16(1 1 ):22837. 11] Blanchard K, Taneepanichskul S, Kiriwat O, Sirimai K, Svirirojana N, Mavimbela N, et al. Two regimens of misoprostol for treatment of incomplete abortion. Obstet Gynecol 2004;103:8605. 3] Zhang J, Gilles JM, Barhart K, Creinin MD, Westhoff C, Frederick MM. National Institute of Child Health Human Development (NICHD) management of early pregnancy failure trial. A comparison of medical management with misoprostol and surgical management for early pregnancy failure. New Engl J Med 2005;353 (8):76 19. 12] Ngoc NTN, Blum J, Westheimer E, Quan TTV, Winikoff B. Medical management of incomplete abortion using 600 versus 1200 mcg of misoprostol. Contraception Dec 2005;72(6):43842. 4] Weeks A, Alia G, Blum J, Ekwaru P, Durocher J, Winikoff B, et al. A randomised trial of oral misoprostol versus manual vacuum aspiration for the treatment of incomplete abortion in Kampala, Uganda. Obstet Gynecol Sep 2005;106(3):5407. 5] Agostini A, Ronda I, Capelle M, Romain F, Bretelle F, Blanc B. Influence of clinical and ultrasound factors on the efficacy of misoprostol in first trimester pregnancy failure. Fertil Steril 2005;84(4):10302. 13] Kovavisarach E, Jamnansiri C. Intravaginal misoprostol 600 microg and 800 microg for the treatment of early pregnancy failure. Int J Gynecol Obstet Sep 2005;90(3):20812. 6] Gronlund L, Gronlund AL, Clevin L, Anderson B, Palmgren N, Lidegaard O. Spontaneous abortion: expectant management, medical treatment or surgical evacuation. Acta Obstet Gynecol Scand Aug 2002;81(8):7812. 14] Moodliar S, Bagratee JS, Moodley J. Medical vs. surgical evacuation of first-trimester spontaneous abortion. Int J Gynecol Obstet 2005;91 :216. 7] de Jonge ET, Makin JD, Manefeldt E, De Wet GH, Pattinson RC. Randomised clinical trial of medical evacuation and surgical curettage for incomplete miscarriage. Br Med J Sep 9 1995;31 1 (7006):662. 15] Greenslade FC, Leonard AH, Benson J, WinklerJ, Henderson VL. Manual vacuum aspiration: a summary of clinical and programmatic experience worldwide. Carrboro (NC): IPAS; 1993. 16] You JHS, Chung TKH. Expectant, medical or surgical treatment for spontaneous abortion in first trimester of pregnancy: a cost analysis. Hum Reprod 2005;20(10):28738. [25] Graziosi GC, van der Steeg JW, Reuwer PH, Drogtrop AP, Bruinse HW, Mol BW. Economic evaluation of misoprostol in the treatment of early pregnancy failure compared to curettage after an expectant management. Hum Reprod Apr 2005;20(4): 106771.

6] 7]