A growing body of research evidence indicates that medical treatment of incomplete abortion with misoprostol is an effective alternative to surgical intervention.1–6 Misoprostol could be an important alternative to dilatation and curettage or manual vacuum aspiration for treatment of incomplete abortion, allowing women to avoid surgical intervention and the attendant risks. Misoprostol is inexpensive and widely available and may also be more acceptable to women than the current standard of care.
A recent review of misoprostol for women’s health indications highlighted the variety of regimens that have been studied for treatment of incomplete abortion.1 In most cases, more than 50% of the women in these studies avoided surgical intervention by using misoprostol. However, the regimens evaluated involved multiple doses of misoprostol, sometimes over a number of days.2–4 Two recent studies support the role of medical management for incomplete abortion but again include regimens that stretch over a number of days (leading to higher efficacy)5 or evaluated higher doses (800 versus 400 μg).2–4,6
To add to this growing literature and potentially help identify a regimen of misoprostol that can be recommended for treatment of incomplete abortion, we evaluated 2 simple regimens of oral misoprostol among women with incomplete abortion attending 2 hospitals in Bangkok. We tested regimens using 600 μg for a number of reasons. First, 800-μg doses have been shown to be effective for induced abortion,7 and we hypothesized that lower doses would be needed for treatment of incomplete abortion. In addition, research on 400 μg has shown success rates of about 50–60% with repeat doses, and we hoped to achieve higher success with a simpler regimen. Finally, 600-μg doses were well tolerated in a study of early abortion using misoprostol alone (K.B., unpublished data, 2003). We chose oral regimens because data indicate that women find oral administration more acceptable than vaginal administration8 and because the drug is currently formulated for oral use.
MATERIALS AND METHODS
We enrolled women presenting with first-trimester incomplete abortion at 2 teaching hospitals in Bangkok, Chulalongkorn Hospital and Siriraj Hospital. Women presenting with incomplete abortion routinely receive surgical evacuation at both hospitals. Women with signs of incomplete abortion and ultrasound findings consistent with first-trimester abortion, who would have been advised to have a surgical evacuation of the uterus, were enrolled in the study. In addition, eligible women were in good general health, had no known allergy to misoprostol, agreed to return for follow-up and fill in a diary of side effects, and had good access to emergency care facilities.
Eligible women who signed an informed consent form were randomized to receive either a single oral dose of 600 μg of misoprostol (single-dose group) or 2 oral doses of 600 μg misoprostol, with 4 hours between doses (double-dose group). The randomization scheme was created by using the pseudo-random number generator in SPSS 9.0 (SPSS Inc, Chicago, IL) at the Population Council in New York. When a woman was deemed eligible, the study staff selected the next sequentially numbered opaque envelope; the number on the envelope became her study identification number. The envelopes contained a piece of paper indicating which regimen she was assigned to. Neither the provider nor the woman was blinded to the treatment regimen.
Whether or not to admit the woman to hospital was left to the discretion of the local investigator. Investigators were encouraged to admit the initial cases if they were concerned about the tolerability of the regimens. At one site (Site A), hospital admission was the standard of care for treatment of incomplete abortion, and a larger proportion of women was admitted.
We asked women to return to the clinic 2 days after misoprostol administration for their initial follow-up visit. Abortion outcome was determined by ultrasound examination. If the abortion was not complete, the woman was given the option of waiting an additional 5 days (1 week from misoprostol administration) to see if the abortion would become complete without further intervention. If the abortion was not complete 1 week after misoprostol administration or if the woman refused a study extension at her first follow-up visit, she underwent a surgical evacuation according to the standard practice at the hospital.
We aimed to enroll 100 women in each group (single and double dose), which would give 80% power (á = .05) to detect a difference of 20% between the 2 groups, based on the hypothesis that the single dose would result in complete abortion in 60% of women. We compared percentages of categorical variables using χ2 or Fisher exact tests. We compared means of continuous variables by using independent sample t tests and differences in ordinal variables by using the Mann-Whitney U test. We considered P < .05 to indicate a statistically significant result.
The Population Council Institutional Review Board and the local ethical review committees at Chulalongkorn and Siriraj Hospitals reviewed and approved the study protocol.
We enrolled 68 women at 1 hospital (Site A) and 101 women at the second hospital (Site B) between September 2000 and April 2002, for a total sample of 169 women (Figure 1). One woman at Site A and 2 women at Site B were lost to follow-up, 1.8% of the total sample. Table 1 summarizes the demographic characteristics of women enrolled at both sites and by study regimen. Women at both sites were in their late twenties, and the vast majority of enrolled women were married. Women enrolled at Site B had significantly more years of education (9.3 versus 7.3 years, P < .05). Approximately one third of the women at Site A and one half of the women at Site B had not given birth (P < .05), and 19.1% and 12.9%, respectively, reported a previous induced abortion (difference not significant). Twenty-two percent of the women at Site A and 75% of the women at Site B reported a previous spontaneous abortion (P < .05). Demographic characteristics were similar across the 2 study regimens in each site.
Table 2 shows women’s experience with the 2 regimens. Ninety-six percent of the women randomly assigned to the double-dose regimen elected to take their second dose of misoprostol at home at Site B, compared with 59.4% at Site A (women in the single-dose group at both sites took the misoprostol at the hospital). At both sites, the vast majority of women took their misoprostol as scheduled. More women at Site A reported heavy bleeding (defined as heavier than that of a normal period), normal bleeding, spotting, and nausea than at Site B. Heavy bleeding, nausea, and pain were reported more frequently in the double-dose group, although these differences were not statistically significant in this small sample. Women in the double-dose group also reported more mean days of fever/chills (although this difference was not statistically significant).
Outcome of misoprostol treatment is summarized in Table 3. Fewer women at Site A (42.6%) had complete abortions than at Site B (85.1%), although complete abortion rates were similar in the single-dose group (66.3%) and the double-dose group (69.9%). Site A had higher rates of medically necessary interventions and a slightly higher number of interventions at the request of the patient than Site B.
Women at both sites and in both treatment groups found treatment with misoprostol acceptable, although the regimen seemed overall to have been better accepted in Site B. More than 65% of the women at Site A and almost 90% of the women at Site B said they were satisfied or very satisfied with the treatment, and more than 80% of women from both sites would choose this method again. Results were similarly high when comparing the 2 treatment groups. Approximately 90% of women at both sites would recommend the method to a friend; 92.9% of women in the single-dose group and 87.7% of women in the double-dose group said they would recommend the method to a friend (Table 4).
The pain of the treatment was measured by asking women to indicate on a visual analogue scale, with the smallest circle indicating no pain and the largest circle indicating the worst pain they had ever experienced, which of the seven circles best approximated the pain they felt with misoprostol. Women at Site A reported significantly more pain than women at Site B; differences between the 2 treatment groups were small. Finally, the vast majority of women at both sites reported that side effects were tolerable or easily tolerable, as did more than 90% of women in both treatment groups. More women at Site A felt they did not receive adequate pain medication (each woman received 500-mg tablets of paracetamol and was instructed to take 2 tablets every 6 hours when necessary to manage pain).
Misoprostol is a useful treatment for incomplete abortion and allows a significant proportion of women to avoid surgical intervention. In this study, almost 70% of women in both the single- and double-dose groups had complete abortions with misoprostol treatment. In addition, a single oral 600-μg dose was as effective as a double dose in this relatively small sample of women. Women found this treatment acceptable, and the side effects were tolerable.
These results show higher success rates than previously published studies1 and are similar to the results reported by Pang et al9 in their comparison of 800 μg given vaginally or orally for treatment of incomplete abortion. They reported 61.1% success for vaginal administration and 64.4% success for oral administration. In their protocol, the 800-μg dose was repeated after 4 hours if women had not had a complete expulsion. Unfortunately, they do not report what percentage of women had the repeat dose. We achieved a similar rate of success with a lower, single-dose protocol of oral misoprostol administration, although we cannot be sure that the population of treated women in our study was identical to that in the earlier studies. Muffley et al6 also found similar rates of success with an 800-μg vaginal dose (60%); in that study 64% of women received 2 doses of 800 μg misoprostol. Again, we achieved similar success in this population with a lower dose and simpler regimen.
Sahin et al5 reported 93.3% success with a regimen of 200 μg vaginal misoprostol followed by 5 days of 200 μg oral misoprostol 4 times per day. Their protocol evaluated final study outcome after 10 days, compared with after 1 day and after 2 days in the Pang et al9 and Muffley et al6 studies. Waiting a longer time to determine study outcome may lead to higher success rates. In our study, if the treatment had not been successful 2 days after misoprostol administration, women were offered the opportunity to wait to see if their abortion would become complete. Fifty-nine percent of the women offered this option chose to wait. Many women in the study cited the longer waiting time to complete abortion as a negative factor in their experience of misoprostol treatment, and protocols with longer times to final outcome may not be acceptable. Further research evaluating the acceptability of longer times to final evaluation could shed light on whether a potential improvement in success by extending the treatment time would be acceptable and whether the increase in success would be significant.
Nielsen and Hahlin10 report that 71% of cases of spontaneous first-trimester abortion resolved in 3 days with no intervention. Blohm et al11 reported 83% complete abortion with expectant management. Misoprostol treatment should be compared directly with expectant management to evaluate whether there is an additional benefit over expectant management alone, either in higher rates of complete abortion or in shorter times to complete abortion among women presenting with incomplete abortion. Further research should also address the acceptability of expectant management among patients and providers in different contexts. Our results indicate that where surgical evacuation is the standard of care for treatment of incomplete abortion, switching to treatment with misoprostol can significantly reduce the need for surgical intervention and its attendant risks and may be more acceptable to women.
We found very different overall success rates at the 2 sites that participated in this study. Indeed, the 85% success rate at Site B is closer to the 93.3% in Sahin et al’s5 study than to the other published rates. We believe the difference in sites could be due to a number of factors. First, we did not standardize criteria for interventions but left it up to the clinical judgment of the local investigator. There may be significant differences in how clinicians judge when it is necessary to intervene. A difference in clinical opinion may underscore the difference in interventions classified as medically necessary. Studies of medical abortion have shown that providers who are more experienced or confident in the method may be less likely to intervene and have higher success rates.12 This could be true of medical treatment of incomplete abortion as well. In addition, differences in standard of care at each of the institutions may have an impact on clinical judgment about necessary interventions.
Women at Site A experienced more pain, were more likely to report heavy bleeding, normal bleeding, spotting, and nausea and found the treatment less acceptable. There were also more interventions at patient request at Site A. Differences in acceptability could be tied to differences in success rates; women who had a complete abortion are likely to find the treatment more acceptable. Differences in reported pain and side effects may be due to differences in counseling procedures. Or, there may be some underlying difference in the populations that attend the 2 hospitals that we were unable to measure. At baseline, years of education, parity, and reported history of spontaneous abortion were significantly different between the 2 sites, suggesting that the populations may indeed be different. These differences may also affect how women interpret the pain scale we used. Unfortunately, because of the small size of the study, we were not able to control for a variety of factors that may have influenced success rates, side effects, and acceptability. A larger study would be needed to investigate these issues in depth.
Misoprostol is an effective nonsurgical treatment for incomplete abortion. This study showed that a simple, single-dose regimen is as effective as more complicated regimens and is acceptable to women. We recommend further research on this simple regimen to determine whether it can be promoted widely for clinical use.
1. Blanchard K, Clark S, Winikoff B, Gaines G, Kabani G, Shannon C. Misoprostol for women’s health: a review. Obstet Gynecol 2002;99:316–32.
2. Chung TKH, Cheung LP, Leung TY, Haines CJ, Chang AMZ. Misoprostol in the management of spontaneous abortion. Br J Obstet Gynaecol 1995;102:832–5.
3. Chung T, Leung P, Cheung LP, Haines C, Chang AM. A medical approach to management of spontaneous abortion using misoprostol: extending misoprostol treatment to a maximum of 48 hours can further improve evacuation of retained products of conception in spontaneous abortion. Acta Obstet Gynecol Scand 1997;76:248–51.
4. Chung TKH, Lee DTS, Cheung LP, Haines CJ, Chang AM. Spontaneous abortion: a randomized, controlled trial comparing surgical evacuation with conservative management using misoprostol. Fertil Steril 1999;71:1054–9.
5. Sahin HG, Sahin HA, Kocer M. Randomized outpatient clinical trial of medical evacuation and surgical curettage in incomplete miscarriage [published erratum appears in Eur J Contracept Reprod Health Care 2002;7:iv]. Eur J Contracept Reprod Health Care 2001;6:141–4.
6. Muffley PE, Stitely ML, Gherman RB. Early intrauterine pregnancy failure: a randomized trial of medical versus surgical treatment. Am J Obstet Gynecol 2002;187:321–6.
7. Jain JK, Meckstroth KR, Mishell DR Jr. Early pregnancy termination with intravaginally administered sodium chloride solution-moistened misoprostol tablets: historical comparison with mifepristone and oral misoprostol. Am J Obstet Gynecol 1999;181:1386–91.
8. Ho PC, Ngai SW, Liv KL, Wong GCY, Lee SW. Vaginal misoprostol compared to oral misoprostol in termination of second trimester pregnancy. Obstet Gynecol 1997;90:735–8.
9. Pang MW, Lee TS, Chung TKH. Incomplete miscarriage: a randomized controlled trial comparing oral with vaginal misoprostol for medical evacuation. Hum Reprod 2001;16:2283–7.
10. Nielsen S, Hahlin M. Expectant management of first-trimester spontaneous abortion. Lancet 1995;345:84–6.
11. Blohm F, Fridén B, Platz-Christensen J, Milsom I, Neilsen S. Expectant management of first-trimester miscarriage in clinical practice. Acta Obstet Gynecol Scand 2003;82:654–8.
© 2004 The American College of Obstetricians and Gynecologists
12. Hedley A, Trussell J, Turner AN, Coyaji K, Ngoc NT, Winikoff B, et al. Differences in efficacy, differences in providers: results from a hazard analysis of medical abortion. Contraception 2004;69:157–63.