A prospective randomized, double-blinded, placebo-controlled trial comparing mifepristone and vaginal misoprostol to vaginal misoprostol alone for elective termination of early pregnancy

John K. Jain1,4, Caryn Dutton1, Bryna Harwood2, Karen R. Meckstroth3 and Daniel R. Mishell, Jr1

1 Department of Obstetrics and Gynecology, University of Southern California Keck School of Medicine, Women's and Children's Hospital, Los Angeles, CA 90033, 2 Department of Obstetrics, Gynecology, and Reproductive Sciences, University of Pittsburgh School of Medicine, Magee–Women's Hospital, Pittsburgh, PA 15213 and 3 Department of Obstetrics, Gynecology and Reproductive Sciences, University of California, San Francisco, San Francisco General Hospital, San Francisco, CA 94110, USA


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
BACKGROUND: Vaginal misoprostol has been shown to be an effective single agent for medical abortion. This randomized, double-blinded, placebo-controlled trial compared a regimen of mifepristone and misoprostol with misoprostol alone for termination of early pregnancy. METHODS: 250 women with gestations <=56 days were randomized by a random number table to receive either 200 mg mifepristone orally or placebo followed 48 h later by 800 µg vaginal misoprostol. Administration of misoprostol was repeated every 24 h up to three doses if abortion failed to occur. Abortion success was defined as complete abortion without the use of surgical aspiration. RESULTS: Successful medical abortions occurred in 114 out of 119 subjects (95.7%) after mifepristone followed by vaginal misoprostol. In all, 110 out of 125 subjects (88.0%) successfully aborted after placebo and vaginal misoprostol. The higher success rate of complete abortion with the mifepristone and misoprostol regimen was statistically significant compared with the placebo and misoprostol regimen (P < 0.05). CONCLUSIONS: A regimen of mifepristone and misoprostol was significantly more effective for termination of pregnancies <=56 days than misoprostol alone. The 88% efficacy obtained with vaginal misoprostol alone may be clinically acceptable when mifepristone is not available.

Key words: abortion/medical abortion/mifepristone/misoprostol/pregnancy


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
A regimen of 600 mg mifepristone followed by 400 µg oral misoprostol, a prostaglandin estrone analogue, was approved for use in the USA in September 2000 for elective termination of pregnancy up to 49 days gestation. Several large clinical trials have documented the safety and efficacy of this combination of agents for medical abortion (Ashok et al., 1998Go; Spitz et al., 1998Go; Schaff et al., 1999Go, 2000Go). Worldwide, millions of women have now received mifepristone and a prostaglandin agent for early pregnancy termination (Christin-Maitre et al., 2000Go). Studies have been ongoing to develop alternative regimens with these agents that are easier to administer, are cost-effective, have a low incidence of side-effects, and improve acceptability of the method.

Misoprostol has also been studied as a single agent for elective termination of pregnancy. This regimen has the potential to be easier to use, provide more rapid pregnancy termination, and is less expensive compared with the mifepristone and misoprostol regimen. A single agent regimen with misoprostol is also likely to be more accessible in areas where mifepristone is unavailable or difficult to obtain. Most published trials have evaluated a dose of 800 µg vaginal misoprostol, administered every 24 h, in gestations up to 70 days duration with reported success rates of 85–93% (Carbonell et al., 1997aGo,bGo; Esteve et al., 1999Go; Jain et al., 1999Go; Bugalho et al., 2000Go; Ngai et al., 2000Go; Jain et al., 2001Go). In a study comparing women who received 800 µg of vaginally administered, saline-moistened misoprostol to a historical control group of women who received 600 mg mifepristone followed by 400 µg misoprostol orally as part of a large multicentre US trial, there was no statistically significant difference in efficacy between the two groups (Jain et al., 1999Go). Abortion success occurred in 88% of subjects receiving vaginal misoprostol alone and 94% of the subjects receiving mifepristone and oral misoprostol in the non-concurrent control group. Therefore, this randomized, placebo-controlled, double-blinded study was designed to compare directly and prospectively the efficacy of mifepristone and vaginal misoprostol to vaginal misoprostol alone for elective termination of pregnancies <=56 days gestation.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Protocol
A total of 250 healthy women desiring termination of pregnancies <=56 days gestation were enrolled (190 at the Women's and Children's Hospital and affiliated clinics, Los Angeles County–University of Southern California Medical Center, Los Angeles and 60 at San Francisco General Hospital, University of California, San Francisco) after a review of the subject's history, a complete physical examination, pelvic ultrasound to confirm gestational age, and documentation of informed consent. Haemoglobin levels and Rh type were also determined at the enrolment visit (study day 1). Exclusion criteria included any evidence of threatened or spontaneous abortion as defined by prior episodes of uterine bleeding or the presence of cervical dilation. Other exclusion criteria included evidence of uterine infection, anaemia (haemoglobin <10 g/dl), bleeding disorders, cardiovascular or cerebrovascular disease, presence of uterine leiomyomata or any known allergy or contraindication to the study medications.

Immediately following confirmation of eligibility and consent, subjects were randomized to one of two groups. Women assigned to group 1 received mifepristone on study day 1, and women in group 2 received placebo medication. The assigned study medication was dispensed and ingested in the study clinic. The placebo medication used was a 500 mg tablet of vitamin C, similar in size, shape, and colour to the mifepristone tablets, but with a single score. Mifepristone was administered as a single 200 mg tablet, and was obtained from the Abortion Rights Mobilization Group. At the time subjects were enrolled, mifepristone was an investigational drug. Therefore, this study was performed and mifepristone studied under US Food and Drug Administration IND no. 58,204. At this enrolment visit, each subject was also given a diary to record uterine bleeding episodes, side-effects, and any medications used on each day until their final visit. Consents, diaries, and questionnaires were made available in both English and Spanish.

On day 3, the subjects returned and a repeat haemoglobin measurement was performed. For women reporting the presence of uterine bleeding, a pelvic ultrasound was performed. If a gestational sac was still present, or if no bleeding had occurred after the mifepristone or placebo tablets, four 200 µg tablets (800 µg) of misoprostol (Cytotec; Searle, Skokie, IL, USA) were placed vaginally under direct visualization with a speculum, followed by placement of 2 ml of normal saline to moisten the tablets. Prophylactic doses of loperamide (4 mg) and acetaminophen (1000 mg) were administered to each subject, as described in a previous protocol (Jain et al., 2001Go). Subjects then remained supine for at least 30 min before discharge to home.

On day 4, a repeat haemoglobin measurement and pelvic ultrasound examination were performed. If the gestational sac was absent, the subject was given a return appointment for study day 15, 2 weeks after the enrolment visit. If the gestational sac was still present, subjects were offered a repeat dose of 800 µg misoprostol, and women who received a second dose also received the same dose of prophylactic loperamide and acetaminophen. Only those subjects who required a second dose of vaginal misoprostol and had no subsequent bleeding returned for evaluation on day 5. On study day 5, subjects again were offered a repeat dose of 800 µg vaginal misoprostol if the gestational sac was visualized by pelvic ultrasound. Women were given the option to have a surgical evacuation to terminate the pregnancy at any time during the study, usually as an alternative to a repeat dose of misoprostol.

Abortion failure was defined as a need for evacuation of the uterus by a surgical technique for any reason, including the presence of a persistent gestational sac seen sonographically, excessive or prolonged uterine bleeding, incomplete abortion, or subject request. Surgical evacuation was performed by electric or manual vacuum aspiration after informed consent.

All subjects were contacted on study day 8 by telephone to evaluate symptoms and all subjects were scheduled to return on study day 15 (or for women who required surgical evacuation, a week or more following the procedure) for a follow-up visit. At that time, subjects underwent a repeat pelvic ultrasound and haemoglobin measurement. The subjects also returned their symptom diaries detailing uterine bleeding patterns (none, spotting, moderate or heavy), and incidence of pain, nausea, vomiting or diarrhoea for each day during the study protocol. All subjects returning for the final visit completed a questionnaire concerning acceptability of the method.

Assignment
Randomization was based on a computer-generated random number table; from this, a study coordinator who was not involved in patient care created the randomization list assigning treatment group. Subjects assigned to group 1 received mifepristone orally on day 1 of the protocol, and subjects assigned to group 2 received placebo.

Masking
An investigator placed single 200 mg mifepristone tablets into 125 unlabelled opaque vials, and single tablets of placebo into another identical, unlabelled 125 opaque vials. A study coordinator not involved with this protocol labelled the vials with subject numbers (1–250) using the randomization list, and placed the vials in sequential order. Each vial was then distributed in sequential numerical order to assign randomization to treatment. At the time of the administration of study medication on day 1, an investigator or a study coordinator witnessed the subjects swallowing the study medication, but did not see the individual tablets. In this manner both the providers and subjects remained blinded to group assignment. For each abortion failure, two separate investigators independently verified the outcome and reason for surgical evacuation: one at the time of the subject's clinic visit and another at the time of data entry and chart review.

Statistical analysis
Based on published rates of abortion success at the time of protocol development of 95% with mifepristone and misoprostol and of 85% with misoprostol alone, a sample size of 125 subjects in each group was needed to determine significance at an alpha of 0.05 with a power of 80%. Statistical analyses were performed using Pearson's {chi}2-test for comparison of proportions, and Student's t-test for comparison of means.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Participant flow and follow-up
A total of 125 women in each group received mifepristone or placebo medication on study day 1, followed by vaginal misoprostol on study day 3, and five women in the mifepristone group were subsequently classified as `lost to follow-up' (see Figure 1Go). The outcomes for four of these subjects are unknown, despite multiple attempts to contact the women. One of these subjects returned for follow-up on day 21 and was no longer pregnant, but was excluded from the analysis because of uncertainty as to when the abortion had occurred. Because of her poor compliance with the study protocol, the day of abortion relative to study medications was unknown, and therefore the chance that she would have needed repeat doses of misoprostol was unknown. An additional subject in the mifepristone group was excluded from further analysis due to inadequate documentation of her misoprostol administration on study day 3. She had a successful procedure, without surgical evacuation, but, due to lack of information about this subject's status on study day 3 and study day 4 (in marked contrast to the remainder of the subjects), her success was not included in the final tabulations.



View larger version (27K):
[in this window]
[in a new window]
 
Figure 1. Trial profile.

 
Analysis
Among all women randomized, there were no statistically significant differences in baseline characteristics by treatment group, as detailed in Table IGo. On study day 3, four subjects reported bleeding with passage of clots, and no gestational sac was visible sonographically. These subjects did not receive misoprostol, but were included in the overall success rates for abortion with <=1 dose of misoprostol, as listed in Table IIGo. All four of these subjects were determined to have been in the mifepristone group after the randomization code was broken at the end of the study. One subject in the placebo and misoprostol group noted urticaria a few minutes after the first dose of misoprostol had been administered, but she had no other complications or significant sequelae. The urticaria resolved within 20 min and symptomatic relief was obtained with an oral dose of diphenhydramine. No other significant adverse reaction to study medications was reported to the study investigators.


View this table:
[in this window]
[in a new window]
 
Table I. Demographic characteristics of the study population
 

View this table:
[in this window]
[in a new window]
 
Table II. Successful abortion rate by regimen, by number of doses of misoprostol, and by gestational age
 
Table IIGo details the successful abortion rates in each group, with success defined as complete expulsion of the pregnancy with no need for surgical intervention. For each dose of misoprostol (one, two or three), there was a statistically significant difference in efficacy between the two treatment regimens. Success rates for each regimen were then compared by number of doses of misoprostol and by gestational age groups. Within the gestational age groups of <=49 days and 50–56 days there was no statistically significant difference in efficacy when comparing mifepristone and misoprostol to misoprostol alone. Within each treatment group, there was also no statistically significant difference in treatment success between the earlier (<=49 days) and later gestations (50–56 days).

In the mifepristone and misoprostol treatment group, there were five surgical evacuations performed only for subject request, without a medical indication. The reasons for the surgical procedure were mild persistent bleeding (n = 3), or refusal to take a third dose of misoprostol with presence of a gestational sac (n = 2). Reasons for surgical evacuation in the 15 women who received placebo and misoprostol were presence of an ongoing pregnancy (n = 6), incomplete uterine evacuation of gestational tissue with no visible gestational sac (n = 5), clinical indication due to an allergic reaction to misoprostol (n = 1), or due to subject request (n = 3). The reasons for requesting surgical evacuation were mild, irregular, but persistent uterine bleeding without evidence of a gestational sac (range of bleeding 3 weeks to 2 months) (n = 2), or refusal to take a third dose of misoprostol with presence of a gestational sac (n = 1).

Side-effects were recorded by the subjects on each day during the study protocol, and included yes or no marks in columns designated `nausea', `vomiting', `diarrhoea', and `fever/chills'. For most subjective symptoms, side-effects between the two groups were similar (Table IIIGo). There was a significantly higher frequency of nausea and vomiting after administration of mifepristone compared with placebo (P = 0.009, P < 0.001). A total of 24% (34/140) of subjects in the combined groups reported a recorded maximum temperature >=100.5°F (combined median 99.8, range 96.5–104). Pain requiring use of opiate analgesics (acetaminophen with codeine) was reported by 27 women in group 1, and 29 women in group 2 (P = not significant). No severe complications occurred in any subjects, including transfusion, emergent surgery for haemorrhage, or sepsis.


View this table:
[in this window]
[in a new window]
 
Table III. Incidence of self-reported symptoms by treatment group
 
To evaluate the frequency of excessive uterine bleeding associated with the two regimens, the average change in haemoglobin over the course of the study was compared between the two groups of subjects. In the mifepristone and misoprostol group, the mean change in haemoglobin was –0.225 g/dl, and in the misoprostol alone group, the mean haemoglobin change was –0.198 g/dl. There was no statistically significant difference in mean haemoglobin change between the two regimens studied. A total of only three subjects in group 1 and six subjects in group 2 had a decrease of >2 g/dl in haemoglobin over the course of the study (P = not significant).

Bleeding patterns among the subjects having a successful medical abortion by either regimen were analysed, and there were no statistically significant differences between the two groups. By the final visit, women in group 1 reported a mean of 8.3 days of moderate or heavy bleeding, compared with 9.8 days in group 2 (P = not significant). The total number of bleeding days including days of spotting in group 1 was a mean of 12.4 days, and in group 2 was a mean of 13.0 days (P = not significant).

Acceptability of the regimens was estimated by analysis of the exit questionnaires. There were no statistically significant differences between the two groups for any of the parameters analysed. For the combined groups, the overall satisfaction with both regimens was high, with 91% (200/220) of the women rating their medical abortion experience as satisfactory or very satisfactory. Seventy-nine per cent of the respondents would choose to have a medical abortion again if they needed a subsequent abortion, and 89.6% would recommend the method to a friend if she needed an early abortion. Of those who had a prior surgical abortion, 64% thought that the medical abortion was `better'. Subjects also rated the overall side-effects as `tolerable' or `very tolerable' in 83% of the cases. Women commonly wrote down the benefit of `no surgery' as one of the best features of the method, while `cramping' or `pain,' `bleeding,' and `nausea' or `vomiting' were some of the worst features of the method.


    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Prior to the approval of mifepristone in the USA, a large multicentre trial confirmed its efficacy for termination of pregnancies up to 63 days gestation, and particularly for earlier gestations (Spitz et al., 1998Go). The regimen of 600mg mifepristone, followed 48 h later by 400 µg oral misoprostol, terminated 92% of pregnancies of <=49 days gestation. During the interval between this report and final regulatory approval of mifepristone, a variety of other clinical trials compared the efficacy of this standard regimen with regimens with lower doses of mifepristone (200 mg), regimens with vaginal administration of misoprostol, and in gestational ages up to 63 days (El-Refaey et al., 1995Go; Schaff et al., 1999Go; Schaff et al., 2000Go).

El-Rafaey et al. published one of the first reports demonstrating that vaginal administration of 800 µg misoprostol 48 h after mifepristone was more effective and better tolerated than 800 µg oral misoprostol (el-Refaey et al., 1995Go). This clinical finding was supported by pharmacokinetic studies that demonstrated prolonged serum levels of misoprostol and a greater area under the curve after vaginal administration compared with oral administration of the same dose (Zieman et al., 1997Go). Another large trial in the USA using a regimen of 200 mg mifepristone and 800 µg of vaginal misoprostol was effective in terminating 95% of pregnancies up to 63 days gestation (Schaff et al., 1999Go). Currently, many providers of medical abortion have adapted this specific regimen for its lower cost, increased effectiveness at later gestational ages, and more tolerable side-effect profile compared with use of higher doses of mifepristone and oral administration of misoprostol.

However, due to concerns that mifepristone would not be approved in the USA, investigators also developed medical abortion regimens without mifepristone. One of these regimens uses methotrexate, an anti-metabolite, in combination with vaginal misoprostol for early pregnancy termination. Trials evaluating methotrexate and misoprostol reported successful abortion rates of 92–96% for pregnancies up to 63 days gestation, but with an interval of as long as 2 or 3 weeks between administration of methotrexate and occurrence of abortion (Hausknecht, 1995Go; Creinin et al., 1996Go).

Another alternative to the mifepristone and misoprostol regimen, based on the successful experience of investigators outside the USA (Carbonell et al., 1997aGo,bGo) is the use of misoprostol as a single agent for termination of pregnancies of 56 days gestation or less. A non-concurrent cohort trial compared 800 µg of vaginally administered misoprostol, given every 24 h up to two doses, to the standard regimen of 600 mg mifepristone and 400 µg oral misoprostol, and found no statistically significant difference in efficacy (Jain et al., 1999Go). A subsequent trial reported the efficacy of 800 µg vaginal misoprostol given with oral loperamide and acetaminophen to be 93% overall (Jain et al., 2001Go). Comparing these efficacy rates with the mifepristone and misoprostol regimen is problematic, however, due to risk of bias in selection of subjects, and the non-blinded administration of medications. Therefore, the current study was designed to compare the two regimens directly using a more rigorous randomized clinical trial. This study is the first randomized, placebo-controlled trial directly comparing mifepristone and misoprostol with misoprostol alone for medical termination of early pregnancy.

Our results confirm that a regimen of mifepristone and vaginal misoprostol is more effective compared with a regimen of misoprostol alone for termination of pregnancies of <=56 days gestation. These findings also support the conclusion that mifepristone provides a small but significant contribution to the overall efficacy of a combined mifepristone and misoprostol regimen. The abortifacient action of mifepristone is supported by the documentation of four successful abortions in this trial that occurred after only mifepristone was administered. A recent report (Creinin et al., 2001Go) demonstrated an abortion rate of 95% using mifepristone 100 mg and a single dose of misoprostol 800 µg vaginally. They concluded that even a very low dose of mifepristone is clinically active, and that the combined regimen still had a higher efficacy compared with most trials using multiple doses of vaginal misoprostol alone.

Compared with the two earlier trials published by Jain et al., the efficacy rate in this placebo and misoprostol group was similar (88 versus 88 and 93%) (Jain et al., 1999Go, 2001Go). All studies allowed subjects either to receive multiple doses of misoprostol (two or three), or to request a surgical evacuation in lieu of a repeat dose if desired. Comparing successful abortion rates after only two doses of vaginal misoprostol again reveals consistent findings: 85.6% in the current study, compared with 88 and 89% in the previous trials.

The analysis of self-reported side-effects demonstrates that subjects receiving mifepristone had a significantly higher incidence of nausea (P = 0.009) and vomiting (P < 0.001) on day 1 compared with subjects receiving placebo. There were no other significantly different rates of side-effects reported between the two groups. Forty-four to 52% of subjects reported nausea, and 22–33% of subjects reported vomiting after the vaginal misoprostol insertion (on or after day 3). Despite the administration of prophylactic acetaminophen and loperamide before misoprostol, the rates of reported diarrhoea ranged from 11 to 16%, and subjective fever or chills was reported by 61–72% of subjects. These rates of side-effects are all similar to rates of previous studies of both the mifepristone and misoprostol alone regimens. Despite these relatively high incidences of side-effects, an overwhelming majority of subjects still reported the method to be tolerable or very tolerable. The use of additional medication to treat any of these side-effects was rare among the study population (3–5%).

One limitation of the study is that five subjects were lost to follow-up in the mifepristone and misoprostol group. Subjects may have failed to return because they perceived the method to be successful, and did not want further examinations. Alternatively, subjects with heavy bleeding or severe side-effects requiring treatment may have chosen to seek care with another provider, rather than return to our clinics for further treatment. Assuming that these five cases had successful outcomes strengthens the statistical difference in efficacy between the two groups. If all five subjects with no or inadequate outcome data are assumed to have had abortion failures (an intention-to-treat analysis), then the statistically significant difference remains only between the groups receiving one dose of misoprostol (P = 0.005), but not between the groups when subjects receiving <=2 or <=3 doses of misoprostol were included (P = 0.109, P = 0.292 respectively).

Despite the lower efficacy of misoprostol as a single agent compared with mifepristone and misoprostol to effect abortion up to 8 weeks gestation, further investigation is needed to evaluate the efficacy of misoprostol alone for earlier gestations (<=49 or even <=42 days gestation). Bugalho et al. have reported an abortion rate of 92.1% 1 week after vaginal administration of a single dose of 800 µg misoprostol in pregnancies of <6 weeks gestation (Bugalho et al., 2000Go). It is possible that the regimen of up to three doses of misoprostol every 24 h used in this trial and others is unnecessary to achieve high efficacy rates, and that expectant management would allow a higher proportion of pregnancies to abort much later than 24 h after misoprostol administration.

Current research in medical abortifacient regimens is focused on altering the dose of both mifepristone and misoprostol in order to reduce side-effects and optimize efficacy, altering the timing of doses to reduce the necessary clinic appointments and follow-up, and addressing issues such as need for ultrasound examination or administration of Rh immune globulin in early pregnancy. This study and others (Winikoff, 1995Go; Beckman and Harvey, 1997Go) confirm that subjects in medical abortion trials find the procedure highly acceptable. However, surgical termination of early pregnancy is also an effective and safe option for most women, and may be a preferred method when distance or resources limit the use of medical methods. Additional research is needed to identify the optimal regimens for specific populations of women.

In conclusion, mifepristone appears to have a small but clinically and statistically significant effect on the abortion success rate of the mifepristone and vaginal misoprostol regimen analysed in this randomized, double-blinded, placebo-controlled trial. However, use of misoprostol as a single agent for pregnancy termination remains of value in clinical situations where mifepristone is unavailable or contraindicated, especially due to the much lower cost of misoprostol compared with mifepristone. It is important to note that the expense of additional provider visits to determine the need for repeat doses of misoprostol may outweigh the expense saved by medication costs. The efficacy and convenience of medical abortion regimens, and the cost of each regimen, are important priorities for future research.


    Notes
 
4 To whom correspondence should be addressed at: 1240 N. Mission Rd, room 8K9, Los Angeles, CA 90033, USA.E-mail: jjain{at}hsc.usc.edu Back


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Ashok, P.W., Penney, G.C., Flett, G.M. and Templeton, A. (1998) An effective regimen for early medical abortion: a report of 2000 consecutive cases. Hum. Reprod., 13, 2962–2965.[Abstract/Free Full Text]

Beckman, L.J. and Harvey, S.M. (1997) Experience and acceptability of medical abortion with mifepristone and misoprostol among U.S. women. Women's Health Issues, 7, 253–262.[Medline]

Bugalho, A., Mocumbi, S., Faundes, A. and David, E. (2000) Termination of pregnancies of <=6 weeks gestation with a single dose of 800 microg of vaginal misoprostol. Contraception, 61, 47–50.[ISI][Medline]

Carbonell, J.L., Varela, L., Velazco, A. and Fernandez, C. (1997a) The use of misoprostol for termination of early pregnancy. Contraception, 55, 165–168.[ISI][Medline]

Carbonell, J.L., Varela, L., Velazco, A., Fernandez, C. and Sanchez, C. (1997b) The use of misoprostol for abortion at <=9 weeks' gestation. Eur. J. Contracept. Reprod. Health Care, 2, 181–5.[Medline]

Christin-Maitre, S., Bouchard, P. and Spitz, I.M. (2000) Drug therapy: medical termination of pregnancy. N. Engl. J. Med., 342, 946–956.[Free Full Text]

Creinin, M.D., Vittinghoff, E., Keder, L., Darney, P.D. and Tiller, G. (1996) Methotrexate and misoprostol for early abortion: a multicenter trial. I. Safety and efficacy. Contraception, 53, 321–327.[ISI][Medline]

Creinin, M.D., Pymar, H.C. and Schwartz, J.L. (2001) Mifepristone 100 mg in abortion regimens. Obstet. Gynecol., 98, 434–439.[Abstract/Free Full Text]

El-Refaey, H., Rajasekar, D., Abdalla, M., Calder, L. and Templeton, A. (1995) Induction of abortion with mifepristone (RU 486) and oral or vaginal misoprostol. N. Engl. J. Med., 332, 983–987.[Abstract/Free Full Text]

Esteve, J.L., Varela, L., Velazco, A., Tanda, R., Cabezas, E. and Sanchez, C. (1999) Early abortion with 800 micrograms of misoprostol by the vaginal route. Contraception, 59, 219–225.[ISI][Medline]

Hausknecht, R.U. (1995) Methotrexate and misoprostol to terminate early pregnancy. N. Engl. J. Med., 333, 537–540.[Abstract/Free Full Text]

Jain, J.K., Meckstroth, K.R. and Mishell, D.R. Jr (1999) Early pregnancy termination with intravaginally administered sodium chloride solution-moistened misoprostol tablets: historical comparison with mifepristone and oral misoprostol. Am. J. Obstet. Gynecol., 181, 1386–1391.[ISI][Medline]

Jain, J.K., Harwood, B., Meckstroth, K.R. and Mishell, D.R. Jr (2001) Early pregnancy termination with vaginal misoprostol combined with loperamide and acetaminophen prophylaxis. Contraception, 63, 217–221.[ISI][Medline]

Ngai, S.W., Tang, O.S., Chan, Y.M. and Ho, P.C. (2000) Vaginal misoprostol alone for medical abortion up to 9 weeks of gestation: efficacy and acceptability. Hum. Reprod., 15, 1159–1162.[Abstract/Free Full Text]

Schaff, E.A., Eisinger, S.H., Stadalius, L.S., Franks, P., Gore, B.Z. and Poppema, S. (1999) Low-dose mifepristone 200 mg and vaginal misoprostol for abortion. Contraception, 59, 1–6.[ISI][Medline]

Schaff, E.A., Fielding, S.L., Eisinger, S.H., Stadalius, L. and Fuller, L. (2000) Low-dose mifepristone followed by vaginal misoprostol at 48 hours for abortion up to 63 days. Contraception, 61, 41–46.[ISI][Medline]

Spitz, I.M., Bardin, C.W., Benton, L. and Robbins, A. (1998) Early pregnancy termination with mifepristone and misoprostol in the United States. N. Engl. J. Med., 338, 1241–1247.[Abstract/Free Full Text]

Winikoff, B. (1995) Acceptability of medical abortion in early pregnancy. Fam. Plann. Perspect., 27, 142–148, 185.[ISI][Medline]

Zieman, M., Fong, S.K., Benowitz, N.L., Bankster, D. and Darney, P.D. (1997) Absorption kinetics of misoprostol with oral or vaginal administration. Obstet. Gynecol., 90, 88–92.[Abstract/Free Full Text]

Submitted on November 19, 2001; accepted on January 25, 2002.