|
|
| |||||||||||||||||||||||||||||||
Background and Methods The combination of mifepristone (RU 486) and a prostaglandin analogue given either intramuscularly or intravaginally is effective in terminating early pregnancy, but the prostaglandin component of the regimen is cumbersome to administer and has side effects. We conducted two studies to determine the efficacy of 600 mg of mifepristone followed by a small dose of misoprostol, an orally active prostaglandin E1 analogue, for the same purpose. In the first study, 505 women who had had amenorrhea for less than 50 days received 400 µg of misoprostol 48 hours after receiving mifepristone, if the pregnancy was not terminated within that period. In the second study, 390 women initially received the same treatment, but if the pregnancy was not terminated within four hours after the administration of misoprostol, they were offered an additional 200-µg dose of misoprostol.
Results In study 1, the rate of success (termination of pregnancy and complete expulsion of the conceptus) was 96.9 percent (95 percent confidence interval, 94.1 to 97.7 percent) -- similar to the success rate of approximately 95 percent for mifepristone followed by the intramuscular or intravaginal administration of prostaglandin. Abortion occurred in 2.9 percent of the women within 48 hours after the administration of mifepristone, in 60.9 percent within 4 hours after the administration of misoprostol, and in 33.2 percent thereafter. The failures included ongoing pregnancies in four women (0.8 percent) and incomplete abortions in nine (1.8 percent); two other women (0.4 percent) required vacuum aspiration for prolonged uterine bleeding. In study 2, pregnancy was terminated in 5.5 percent of the women before the administration of misoprostol and within four hours after the first dose of misoprostol in 69.1 percent. Among the 71 women who received a second dose of misoprostol, 67 had complete abortions, 2 had partial retention of the conceptus, 1 had synechia with ongoing pregnancy, and 1 had an ectopic pregnancy. One ongoing pregnancy, which was terminated by vacuum aspiration, was recorded among the 27 women who declined to take the second dose of misoprostol. The overall rate of success of the regimen with the optional second dose of misoprostol was 98.7 percent (95 percent confidence interval, 96.8 to 99.5 percent). No woman had any serious adverse event.
Conclusions The combination of mifepristone and misoprostol is effective for the termination of early pregnancy in terms of success, tolerance, safety, and practicality.
In France, sulprostone, a prostaglandin E2 derivative (250 µg administered intramuscularly 36 to 48 hours after the administration of 600 mg of mifepristone), has been used for the voluntary termination of pregnancy in women who have had amenorrhea for less than 50 days8,9,14. Less often, gemeprost, a prostaglandin E1 derivative, has been given intravaginally8,9; in Great Britain gemeprost is used to terminate pregnancy in women who have had amenorrhea for less than 63 days6,15. Myocardial infarctions attributable to coronary spasm, one of them fatal, have occurred in 3 of the more than 60,000 women given sulprostone9,16. As a result, its use has been abandoned. Currently, termination of pregnancy with mifepristone and any prostaglandin should not be undertaken in any woman who is older than 35 years or is a heavy smoker.
The oral administration of a prostaglandin should improve the mifepristone-prostaglandin method by increasing safety, convenience, and potentially, privacy. Unlike prostaglandin E2,17 9-methylene prostaglandin E2 complements the action of mifepristone in the termination of early pregnancy,18 but it is not commercially available. The oral administration of 400 µg of misoprostol, a prostaglandin E1 derivative,19 48 hours after the administration of 600 mg of mifepristone is also effective for this purpose20. Among 100 pregnant women who had had amenorrhea for less than 50 days, the success rate was 95 percent, indicating that misoprostol, although not an efficient abortifacient when taken alone,21 is very active when given after mifepristone.
We describe here the results of two studies of the termination of pregnancy with mifepristone and misoprostol. In the first study, conducted at the request of the French Ministry of Health, we tested mifepristone and a single dose of misoprostol. In the second study, an additional dose of misoprostol was offered to women who had not aborted four hours after the first dose of misoprostol.
Methods
The study protocols were approved by the Delegation a la Recherche, Assistance Publique-Hopitaux de Paris, and the Comite d'Ethique, Hopital Broussais, and written informed consent (including acceptance of surgical evacuation in case of treatment failure) was given by all study subjects. Voluntary termination of pregnancy is legal in France during the first 84 days of amenorrhea, but the medical method is currently allowed only up to day 49. Pregnancy must be confirmed by measurement of the serum concentration of the beta subunit of human chorionic gonadotropin (beta-hCG) or ultrasonography.
The exclusion criteria for the two studies were as follows: age greater than 35 years, amenorrhea for more than 49 days, a possibility of ectopic pregnancy, ongoing spontaneous miscarriage, a concomitant blood-clotting abnormality, smoking more than 10 cigarettes per day, asthma, and cardiovascular or other serious disorders.
Study 1, conducted from June through October 1991, involved 505 women at 25 centers. Their mean (±SD) age was 26 ±5 years. On the first day, mifepristone was given orally in a single dose of 600 mg. Each woman was asked to return on day 3, when she received a single oral dose of 400 µg of misoprostol (two 200-µg tablets) and, when necessary, an injection of anti-D
immunoglobulin. She remained at the center for four hours for measurements of blood pressure and assessment of pelvic pain on a 100-mm visual-analogue scale (from 0 mm, indicating no pain, to 100 mm, indicating severe pain)22. The women were asked to return 8 to 15 days later for clinical evaluation and measurement of the serum 
-hCG concentration, ultrasonography, or both. Hemoglobin concentrations were measured before the administration of mifepristone and at the final visit. Success was defined as the complete expulsion of the conceptus, without the need for an additional surgical procedure. Continuation of pregnancy, incomplete expulsion, and hemorrhage requiring a surgical procedure were considered failures. Tolerance was evaluated, and any prolonged bleeding was assessed during the final visit.
Study 2, from March 1991 through March 1992, involved 390 women at one center. Their mean age was 26 ±5 years. These women initially received mifepristone and misoprostol as in study 1. If expulsion had not occurred four hours after the administration of misoprostol, the women were offered another 200-µg dose of misoprostol and asked to stay at the center for another two hours. Blood pressure and hemoglobin concentrations were not measured routinely in this study.
Statistical Analysis
Since the study was noncomparative, the statistical analysis was mainly descriptive. The results of study 1 were analyzed with use of Statistical Analysis System (SAS) software (SAS Institute, Cary, N.C.), and those of study 2 with Epistat software (T.L. Gustafson, Round Rock, Tex.). Confidence intervals were determined by solving the quadratic equation23. P values (two-tailed) of less than 0.05 were considered to indicate statistical significance.
Results
Study 1
Efficacy was analyzed in 488 women. The mean (±SD) duration of amenorrhea in the 488 women was 45 ±6 days. When verified by ultrasonography (in 325 women), the duration of pregnancy was estimated to be 42 ±5 days. Seventeen of the original 505 women were excluded: 1 because of an ectopic pregnancy, 1 who had had amenorrhea for more than 49 days, 1 who was mistakenly given sulprostone instead of misoprostol, and 14 who did not return for follow-up. Among these 14 women, pregnancy was terminated within four hours after the administration of misoprostol in 12, but they were not included in the analysis since the possibility of incomplete expulsion of the conceptus could not be completely excluded.
Among the 488 women, 14 (2.9 percent) did not receive misoprostol because the pregnancy was terminated within the first 48 hours, and 5 (1.0 percent) received a second dose of misoprostol because they vomited soon after receiving the first dose. The overall rate of success was 96.9 percent (95 percent confidence interval, 94.1 to 97.7 percent) (Table 1). Among the women in whom treatment was successful, the mean length of time between the administration of misoprostol and the expulsion of the conceptus was 12 ±36 hours. The median value was 3 hours, because in the majority of women expulsion took place soon after the administration of misoprostol (60.9 percent in the first 4 hours and 87.2 percent within 24 hours); the longest delay was 12 days.
|
All the women had uterine bleeding, whatever the outcome of drug administration. The mean duration of bleeding was 9 ±4 days (range, 1 to 32); bleeding lasted 12 or fewer days in 85 percent of the women. The mean hemoglobin concentration decreased from 12.8 ±0.9 g per deciliter (7.9 ±0.6 mmol per liter) to 12.1 ±1.0 g per deciliter (7.5 ±0.6 mmol per liter) at the final visit (P<0.001). There was no correlation between this decrease and the duration of the pregnancy as assessed by ultrasonography (P = 0.10). The mean decrease in the hemoglobin concentration was similar to that in women who received mifepristone and gemeprost9. One woman needed a blood transfusion and hemostatic curettage because the hemoglobin concentration fell from 13.0 g per deciliter (8.1 mmol per liter) to 6.1 g per deciliter (3.8 mmol per liter) nine days after the administration of misoprostol, and another woman underwent curettage for heavy bleeding four hours after the administration of misoprostol.
One woman, in whom an ovarian cyst was diagnosed at the time of entry into the study, had torsion of a uterine appendage 10 days after the administration of mifepristone and underwent cystectomy at that time.
Most of the women (80.5 percent) had uterine cramps, for which 16.0 percent received a nonopiate analgesic drug. The pain usually began within one hour after the administration of misoprostol, and it lasted one hour or less. The median intensity of pain recorded (for 484 women) on the visual-analogue scale was 31 mm. During the four-hour monitoring period, nausea, vomiting, and diarrhea occurred in approximately 43 percent, 17 percent, and 14 percent of the women, respectively. The mean systolic and diastolic blood pressure had decreased slightly four hours after the administration of misoprostol. Six women had a substantial but transient decrease in blood pressure (more than 30 mm Hg for the systolic pressure and 15 mm Hg for the diastolic pressure) attributable to a vagal reaction secondary to painful uterine cramps.
Study 2
The mean duration of amenorrhea in the 390 women in study 2 was 45 ±5 days. Most women had had amenorrhea for less than 50 days, and the duration of amenorrhea was 50 to 59 days in approximately 9 percent. The latter women were included because the date of fertilization was known to be 35 or fewer days earlier. Five women who mistakenly received an additional 400 µg of misoprostol, instead of 200 µg, were not considered further in the analyses reported here (all had complete abortions).
Among the 385 women included in the analysis of efficacy (Table 2), pregnancy was terminated before the administration of misoprostol in 21 (5.5 percent). Pregnancy was terminated within four hours after the administration of 400 µg of misoprostol in 266 women (69.1 percent), and in approximately 90 percent of these women pregnancy was terminated in the first three hours.
|
The mean duration of uterine bleeding was 10 ±4 days (range, 1 to 30). Bleeding lasted less than 15 days in approximately 90 percent of the women and less than 10 days in approximately 65 percent. No woman in the group received a transfusion. Not taking into account two women who had had amenorrhea for 68 and 80 days, who were enrolled in the study by mistake, and one woman who had minute vaginal spotting for 30 days, there was a positive correlation between the duration of amenorrhea preceding the administration of mifepristone and the duration of bleeding (r = 0.103, P<0.05). After the exclusion of women whose conceptuses were evacuated instrumentally, the mean duration of bleeding in the women who took 400 µg of misoprostol was 10 ±4 days, whereas the women who took 600 µg of misoprostol had bleeding for 8 ±4 days (P<0.002). The mean duration of bleeding in the 5 women who took 800 µg of misoprostol was 8 ±2 days, and it was 10 ±4 days in the 21 women who had abortions without the administration of prostaglandin.
Uterine cramps were of the same intensity as in study 1, but approximately 50 percent of the women had recurrent cramps after the additional dose of misoprostol. Only 12.5 percent of the women received a nonopiate analgesic drug, and none received an opiate. As in study 1, nausea, vomiting, and diarrhea occurred in approximately 40 percent, 15 percent, and 10 percent of the women, respectively. No woman had headache, fatigue, skin rash, or any cardiovascular side effects.
Discussion
The combination of mifepristone and a prostaglandin results in the complete and safe termination of early pregnancy, as demonstrated previously4,6,7,8,9,20 and as confirmed by the results presented here. The oral administration of prostaglandin appears to be an improvement over intramuscular or intravaginal prostaglandin administration for the termination of pregnancy in women who have had amenorrhea for less than 50 days. However, studies of women who have had amenorrhea for longer periods are needed, because mifepristone is approved for use up to day 63 in Great Britain6,15. Misoprostol is more convenient than any other prostaglandin, because it is given orally, and compliance is good. The drug is inexpensive, and it can be stored at room temperature.
The misoprostol regimen used in the second study appears to result in success rates that are slightly higher than the rate of approximately 95 percent currently obtained with sulprostone or gemeprost in women who have had amenorrhea for up to 49 days, but this regimen should be studied further before a definitive claim can be made. In our study the administration of the second, smaller dose of misoprostol was often followed by additional uterine cramps, which necessitated the administration of an analgesic drug in approximately 15 percent of the women, but no excessive bleeding or cardiovascular incidents were recorded.
It is not clear whether the second dose of misoprostol significantly shortened the waiting period before abortion, always a difficult time for women to endure. Of the 67 women who took the second dose in study 2 in whom an abortion occurred, over 50 percent of the women aborted within two hours, thus increasing significantly the number of women in whom pregnancy was terminated while they remained in the center. The reported amount of bleeding appeared to be essentially the same as with mifepristone alone,1,3 with other prostaglandins,4,6,7,8,9,24 or with only one dose of misoprostol20,21.
The apparent decrease in the duration of bleeding in the women who received the second dose of misoprostol is encouraging. The correlation between the duration of amenorrhea (and thus the duration of pregnancy) and the duration of bleeding argues for the earliest possible intervention after the decision to have an abortion has been made.
The side effects in the women who received misoprostol were neither more frequent nor more severe than those reported after any other prostaglandin. No woman in either study had a cardiovascular accident. Misoprostol is generally considered to be a safe drug,25,26,27 and larger doses (often 800 µg daily) have been given for long periods for other, mostly gastrointestinal indications in patients who are more likely to be smokers or to be at risk for cardiovascular accidents than healthy pregnant women. Until large-scale post-marketing surveillance studies are completed, however, caution should be exercised in using this agent in women at risk for cardiovascular accidents. In any case, the administration of misoprostol alone (not preceded by mifepristone) in order to induce abortion (a practice strongly disapproved of by the distributing company) is particularly unsuitable, since the woman would be exposed to the double risk of insufficient abortive action and embryonic abnormalities28,29,30.
Pregnancies continue in approximately 1 percent of women who receive mifepristone and a prostaglandin7,8,9,15. No teratogenic effects of mifepristone have been reported in nonhuman primates or in humans31,32. The malformations reported with the use of misoprostol in illegal or non-medically supervised attempts to induce abortion have occurred after the inappropriate administration of this prostaglandin and always in women who did not receive mifepristone28,29,30. There was only one ongoing pregnancy among the women in study 2, and the regimen may therefore result in fewer ongoing pregnancies than do other regimens of mifepristone plus a prostaglandin8,9,33,34. If such a trend is confirmed, the risk of fetal abnormality would be even lower than with other regimens, since in spite of strong advice, some women may not come back for a mechanical abortion if the administration of mifepristone and prostaglandin fails to terminate the pregnancy.
In spite of the satisfactory results reported here, we wish to emphasize that the abortion procedure should continue to be medically supervised14. Pregnancy constitutes an empirical risk for women, whether they choose to continue or to terminate it, and medical assistance is a woman's right in either case. In particular, the woman should be examined initially to determine the duration of pregnancy and to exclude ectopic pregnancy and any other medical or surgical contraindication to any method of abortion. Ectopic pregnancy is difficult to detect very early, and its possible occurrence makes a follow-up visit 8 to 15 days after the treatment mandatory, whether pregnancy is terminated by pharmacologic or surgical methods. It is too early to determine whether it will become possible to use the mifepristone-misoprostol method, under medical control, outside specialized centers.
In conclusion, we found in the studies reported here that, for the termination of pregnancy in women who had had amenorrhea for less than 50 days, orally administered misoprostol, in conjunction with mifepristone, is at least as successful and as well tolerated as other prostaglandins given parenterally or vaginally. This new regimen is simpler and potentially allows greater privacy than any other abortion method, and it has recently been approved in France. We suggest that mifepristone followed 48 hours later by misoprostol is a convenient and safe regimen for the early termination of pregnancy, but the possibility of rare accidents cannot be excluded, and caution should be exercised.
We are indebted to Mr. M. Arjomandi (of the Medical Student Scholars Program, Stanford University), Ms. F. Boussac, Dr. R. Fiddes (a postdoctoral fellow from Melbourne University), Mr. A. Gold (a summer medical student at Harvard University), Ms. C. Hogan (a summer medical student at the University of California at San Francisco), Mr. J.C. Lambert, Ms. C. Legris, and Ms. C. Meynieu for assistance in the preparation of the manuscript; to the physicians who participated in study 1 -- R. Boghossian (Avignon), J. Champion (Marseille), F. Charles (Toulon), R. Dreyfus (Paris), P. Fournie (Melun), R. Frydman (Clamart), B. Geffroy (Saint-Nazaire), M.C. Jourdan (Toulouse), M.C. Landeau (Paris), C. Levade Putois (Toulouse), B. Maria (Villeneuve Saint Georges), J. Milliez (Creteil), H. Missey Kolb (Saint Germain en Laye), M. Neny (Tours), Y. Pidoux (Morlaix), D. Plateaux (Le Kremlin-Bicetre), R. Renaud (Strasbourg), M. Rettel (Metz), W. Scharfman (Roubaix), C. Van Geem (Valenciennes), P. Vige (Saint Cloud), C. Vitani (Lyon), and Y.Y. Wang (Montrouge); and to Drs. A. Bureau, L. Meng, and M. Clipet, Ms. C. Der Adreassian, and Ms. G. Grimal for their help in study 2.
Source Information
From the Medical Department, Laboratoires Roussel, Paris (R.P., V.T., L.S., M.R., F.E., A.U.); the Centre d'Orthogenie de l'Hopital Broussais, Service du Pr. P. Poitout, Paris (E.A.); and the Institut National de la Sante et de la Recherche Medicale (INSERM) U33 and the Laboratoire d'Hormonologie de l'Hopital de Bicetre, Bicetre (P.L., E.-E.B.) -- all in France.
Address reprint requests to Dr. Baulieu at INSERM U33 and Laboratoire des Hormones, 80 rue du General Leclerc, 94276 Le Kremlin-Bicetre, France.
References
| |||||||||||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
