10th Postgraduate Course for Training in Reproductive Medicine and Reproductive Biology
Research on regimens for early medical abortion
Helena von Hertzen, MD,
DDS
Medical Officer
UNDP/UNFPA/WHO/World Bank Special Programme of
Research,
Development and Research Training in Human Reproduction
Geneva, Switzerland
Tel: 41-22-7913373
Fax: 41-22-7914171
E-mail: vonhertzenh@who.ch
The views expressed are those of the author and do not necessarily reflect the views of the Special Programme of Research, Development and Research Training in Human Reproduction or of its sponsors.
Abstract
Until recently the medical abortion regimen of mifepristone followed by a suitable prostaglandin analogue two days later has been available in only a few countries. Limited access to the drug has hampered research and forced investigators without access to seek alternatives. Where mifepristone has not been available, investigators have tested repeated doses of the prostaglandin misoprostol alone or in combination with methotrexate. However, these regimens cannot compete in efficacy, safety, and convenience with the mifepristone-prostaglandin regimen. The development of the mifepristone-prostaglandin regimen has continued since approval, and a more effective and better tolerated medical method than was available ten years ago can now be offered to women. Several randomized double-blind studies have demonstrated that the dose of mifepristone can be lowered to 200 mg, one-third of the original dose, without compromising efficacy. Misoprostol has now largely replaced other prostaglandin analogues in the regimen, and this has made medical abortion simpler to use and more cost effective. Although women prefer oral administration of drugs, recent evidence suggests that to achieve sufficient efficacy, misoprostol has to be administered vaginally in women beyond 49 days LMP. This medical abortion regimen has proved to be acceptable to women and safe when provided under proper conditions.
Introduction
Abortions have been induced throughout human history with a wide variety of compounds, but a systematic search for effective regimens to induce abortion medically only started in the 1970s following the synthesis of prostaglandin analogues. This search was boosted by the discovery of the first antiprogestogen, mifepristone. Neither prostaglandins nor mifepristone were sufficiently effective alone, however, and it was eventually the sequential regimen of mifepristone followed by a suitable prostaglandin analogue that offered women a viable alternative to surgical abortion. Mifepristone is the only antiprogestogen that has been used clinically to any extent, and its use is likely to expand significantly in the near future as it has been registered recently in several countries (Austria, Belgium, Denmark, Finland, Germany, Netherlands, Spain, Israel, Switzerland, and Russia) in addition to China, France, Sweden and the United Kingdom. The gestational age limit for first-trimester medical abortion is usually up to 49 days except in Russia, Sweden, and the United Kingdom where it is up to 63 days. This review will focus on mifepristone and a prostaglandin analogue for pregnancies up to 63 days (nine weeks) since last menstrual period (LMP).
Early Development
Research on the use of prostaglandins alone did not lead to a practical method for several reasons. In order to reach acceptable efficacy rates (> 90% complete abortion), repeated high doses of prostaglandins (sulprostone, gemeprost, or meteneprost were available at that time) were necessary.3 But these high doses had troublesome side effects such as vomiting and diarrhea (30%-40%), and about half the women in these studies required morphine analgesics for alleviation of uterine pain.
Since the prostaglandins alone did not fulfil expectations, the discovery of the antiprogestogen mifepristone in 1980 was welcomed with great enthusiasm. The importance of progesterone for the maintenance of pregnancy was well known, and it was expected that blocking its effects at the receptor level with this new antihormone would inevitably lead to abortion. Researchers were, therefore, disappointed to learn of the results of the first studies suggesting that this new drug could induce complete abortion in only 60%-70% of women treated.4,5 This finding was confirmed by a number of investigators who tried, in vain, to improve efficacy by changing the daily dose and the duration of treatment.6 For pregnancies up to eight weeks the complete abortion rate was generally less than 70%, although somewhat better rates could be achieved if treatment was given within the first 10 to14 days LMP. The frequency of complete abortion decreased with advancing pregnancy, and there did not seem to be any relationship between the success rate and the mifepristone regimen employed for women at the same stage of gestation.
An important step forward was made in 1985 with the discovery that mifepristone increased contractility and sensitized the myometrium to prostaglandins.10 The maximum effect was achieved when prostaglandins were administered 36 to 48 hours after mifepristone, an observation later confirmed in practice.8 This finding was tested in clinical studies using a sequential regimen of mifepristone and various prostaglandin analogues. Most prostaglandin analogues were found to induce complete abortion in about 95% of mifepristone-pretreated women,9 and the dose of prostaglandin could be reduced fivefold, thus reducing less side effects and increasing acceptability. This research led finally to a feasible medical alternative to surgical abortion and the regimen of 600 mg of mifepristone used with a prostaglandin analogue. It was approved in 1988 in France, in the United Kingdom in 1991,and in Sweden in 1992. Twenty-five mg tablets of mifepristone have been available in China since 1992.
Since approval of the regimen, research has focused mainly on the following issues: the optimal dose of mifepristone, the most appropriate type and dose of prostaglandin, the duration of pregnancy for which the treatment remains effective and safe, and the acceptability to users.
Optimal Dose
As mentioned above, earlier studies on mifepristone alone could not demonstrate a relationship between dose and efficacy. Also, studies in both pregnant and non pregnant women suggested that the drug's pharmacokinetics are non linear and that oral administration in single doses greater than 100 mg resulted in serum concentrations that differed only minimally or not at all.10 From these pharmacokinetic data it seemed likely that smaller doses than the registered 600 mg could render comparable efficacy rates. Multicenter trials conducted by the World Health Organization confirmed this assumption and demonstrated that the effectiveness of a 600 mg dose could be equalled with either repeated small doses (5 doses of 25 mg given at 12-hour intervals)11 or a single dose of 200 mg followed by a suitable prostaglandin.12,13 These findings have had an impact on clinical practice in China, where the repeated regimen of small doses is used. Other researchers have also confirmed the WHO findings.14,15
A randomized multicenter trial was carried out to determine if the dose of mifepristone could be reduced even further. Doses of 200 mg and 50 mg were followed by vaginal gemeprost in pregnant women up to 56 days LMP.16 The complete abortion rate in women receiving the 50 mg dose of mifepristone was significantly (p<0.01) lower and the rate of continuing pregnancy higher than in women who received 200 mg. Thus, 200 mg seemed to be approaching the minimum amount needed as a single dose in order for the sequential regimen to be effective.
The Most Appropriate Type and Dose of Prostaglandin
In China and France, where the medical abortion regimen is used in pregnancies of up to 49 days LMP, 0.6 mg (China) or 0.4 mg (France) of misoprostol is administered orally two days after mifepristone, with complete abortion rates of more than 95%.17 In Sweden and the United Kingdom, 1 mg vaginal gemeprost has been shown to be effective in pregnancies up to 63 days LMP.6 Gemeprost has several disadvantages, however, especially for use in developing countries, as it is both expensive and unstable at room temperature. Therefore, misoprostol, which is registered for the prevention and treatment of gastric ulcer in more than 70 countries, is a more attractive choice.
Although oral misoprostol was shown to be effective up to 49 days LMP, the results from a study including women with pregnancies of up to 63 days LMP suggested that it might be less effective than vaginal gemeprost in more advanced pregnancies.14 Recently, two large multicenter studies clearly demonstrated a statistically significant decline in efficacy with increasing length of gestation when misoprostol is administered orally.13,18 In both studies, the complete abortion rate was 83%13 to 87%18 in women at 50 to56 days LMP and only 78%18 to 80%13 at gestations of 57 to 63 days. More importantly, the continuing live pregnancy rate increased with the length of gestation in both studies (1% at 49 days and 9% at 57 to 63 days). These findings underscore the importance of accurately estimating gestation before treatment and the necessity of choosing an effective regimen. Recent articles referring to potential risks of uncontrolled use of misoprostol 19 also highlight the importance of followup after treatment.
The results of the first, and until now the only, randomized study comparing the efficacy of oral and vaginal administration of misoprostol after mifepristone in the first trimester suggested the superiority of the vaginal route.20 Ninety-five percent of women (n=270) within 63 days LMP in the vaginal misoprostol group (0.8 mg administered two days after 600 mg of mifepristone) had complete abortions compared to 87% of women who took the same dose of misoprostol orally. In addition, pregnancy continued in 9 of the 130 women (7%) in the oral group, compared to only one in the vaginal group. Rates of nausea, vomiting, and diarrhea were also somewhat lower in the vaginal group. The high efficacy after vaginal administration has since been replicated in case-series reports,15 and the improved side effect profile was demonstrated in a randomized second-trimester trial.21
These findings were somewhat unexpected, as the drug was developed and formulated for oral use, and gynecological application had not been envisaged. A possible explanation for the greater efficacy of vaginal misoprostol is provided by the results of a recent study,22 in which intrauterine pressure was measured for up to four hours in 30 pregnant women after vaginal or oral administration of misoprostol. Uterine tonus started to increase after 8 (7.8± 3.0) minutes and reached its maximum 25 (25.5± 5.0) minutes later in the oral group while the corresponding figures after vaginal administration were 21 (20.9± 5.3) minutes and 46 (46.3± 20.7) minutes. Although the initial increase in uterine tonus was more pronounced after oral administration, the contractions were not regular, and uterine activity increased continuously during the whole observation period after vaginal treatment. The effect of vaginal administration on contractility appeared to be longer lasting and more regular, which explains its greater efficacy.
Despite the good clinical results, vaginal absorption of misoprostol appears to be irregular, with large individual variations.23 A maximum plasma concentration is found at about 30 minutes after oral administration while it takes some 80 minutes to reach the peak after vaginal administration.22,23 The half-life of misoprostol is about 30 minutes, and as there is no reason to believe that this would depend on the route of administration, the difference observed in plasma levels appears to represent a difference in the rate of absorption.22 Because local factors may influence vaginal absorption, studies have tested adding acetic acid or water to improve efficacy even further. Acetic acid did not have any effect on clinical efficacy,24 and, although adding water improved the results somewhat, the difference was not statistically significant.25
The wide availability and low cost of misoprostol compared to other prostaglandin analogues have contributed to a renaissance in research for a prostaglandin-only method of abortion. Several studies have tested repeated vaginal doses during the last few years,26 and have found that repeated high doses are needed to reach acceptable rates. Also, the induction-to-abortion interval tends to be quite long, in some cases up to several days, making the procedure rather inconvenient. Further, one study has suggested that if pregnancy continues despite the treatment, strong uterine contractions induced by the drug may cause vascular disruption and ischemia, leading to fetal malformations.19 The efficacy data to date consist of case series reports, in which the definition for successful treatment sometimes included incomplete abortions. Randomized comparative trials are needed to evaluate the real value of the misoprostol-only regimen in the termination of first-trimester pregnancies.
Length of Gestation
Softening and dilation of the cervix becomes progressively more important as the length of gestation increases; after seven weeks, a more significant dilation is necessary. Also, the more advanced the pregnancy, the stronger the myometrial contractility required, thus, the role of prostaglandin is crucial. Several studies12 have demonstrated marked softening and dilation of the cervix after treatment with mifepristone, and prostaglandins are likely to enhance this effect.
Acceptability studies suggest that a large majority of women prefer oral administration of drugs,21 but clinical research13,18 has shown clearly that misoprostol should not be used orally beyond 49 days LMP. The complete abortion rate was 97% in a recent report of 2000 consecutive cases of gestations up to 63 days using 200 mg of mifepristone followed 36 to 48 hours later by 0.8 mg of vaginal misoprostol.15 The vaginal dose of 0.8 mg may be preferable to 0.4 mg in more advanced pregnancies, as the need for surgical intervention tends to increase with increasing gestational age. However, no studies have compared vaginal doses.
It is evident from previous research that efficacy cannot be improved by increasing the dose of mifepristone or modifying its administration. An effective regimen for pregnancies of more than 63 days' gestation may be possible by modifying the dose and administration of the prostaglandin component. The longer the gestation, the more the regimen becomes a prostaglandin regimen, perhaps even with repeated administration, as in second trimester abortion. The complete abortion rate is likely to decrease, and medical abortion in late first trimester may be less acceptable to women and not competitive with surgical interventions.
Acceptability
Comparing the acceptability of medical to surgical abortion is not straightforward, as women are generally relieved afterward, regardless of method.27 Preference may also be influenced by the quality of existing surgical services; for example, in settings where vacuum aspiration is performed without any analgesia, most women prefer the nonsurgical approach. However, even where surgical services are adequate, many women like the medical method because it appears more "natural" and discreet and allows them to avoid anaesthesia. The percentage of women choosing medical abortion is regularly increasing in countries where it has been in routine use, and is now about 34% in France, 40% in Scotland, and 35% in Sweden, although only 8% in the United Kingdom.
Three visits are required for medical abortion in countries where it has been approved so far, or one more visit than vacuum aspiration, which may influence a woman's choice. Some clinicians in the United States of America have investigated administering misoprostol at home,25 and it may be feasible in settings with good communication and transportation so that women can be transferred quickly to hospital in case of emergency. An ongoing 15-center study (7 in developed countries, 8 in the developing world) coordinated by WHO asked women whether they would prefer to abort at home or in the clinic. Interim results indicate that most women preferred the clinic, as they felt more secure there in case they needed help.
In China, which has extensive experience with medical abortion, the long duration of bleeding is regarded as the worst feature of medical abortion. Chinese investigators recently completed a double-blind randomized study to see if taking oral contraceptive pills immediately after medical abortion would affect the duration of bleeding or the abortion rate.29 Oral contraceptives did not influence the median duration of bleeding but there was a significant and unexpected drop in haemoglobin concentration in the oral contraceptive group. A further randomized double-blind study is ongoing to measure the actual blood loss. Also, the WHO research group has undertaken a randomized, placebo-controlled multinational study to investigate the effect of repeated administration of misoprostol on the amount of blood loss and the duration of bleeding.
Safety of medical abortion
It is important to note that, despite advances in abortion technology, procedure-related morbidity and mortality increase with gestational age. Vacuum aspiration is one of the safest surgical procedures, and the experience with medical abortion to date suggests that this method is also safe when provided at clinics with adequate back-up services. Both antiprogestogens and prostaglandins have contraindications that have to be respected. Women who suffer from adrenal insufficiency or severe asthma, or are allergic to mifepristone (absolute contraindications), patients with a history of cardiovascular disease (angina, Raynaud's disease, cardiac arrhythmias, severe hypotension) or known allergy to prostaglandins or porphyria should not be offered the method. As a special precaution, the medical method is not recommended for use in women who are older than 35 years and who smoke more than ten cigarettes a day.
Bleeding starts in about 50% of women after administration of mifepristone11, 12 and some 2% to 3% may abort before the prostaglandin is administered.18 In general, bleeding is most abundant around the time of expulsion, and some women require oxytocics at that time. Bleeding occurs in almost all cases and is not in anyway proof of expulsion. The need for uterine evacuation for hamostatic purposes varies between studies from 0.35%15 to 2%12. About one18 to three12 per thousand women treated have required blood transfusions. Some studies have found a decrease in hemoglobin concentrations, which is often significant12 one to two weeks after treatment, but returns to normal.
In the largest series reported to date8 the complete abortion rate was 95% among 15,709 women at 49 days LMP or less using the regimen of 600 mg of mifepristone followed by either gemeprost or sulprostone. The failures included continuing pregnancies in 1% of women and incomplete expulsions in 3%. Less than 1% of women required hemostatic surgical procedures (0.8%) or blood transfusions (0.1%). Other complications included skin rash (0.2%), fever (0.3%), endometritis (0.2%), and salpingitis (0.03%). One myocardial infarction was reported in a 38-year-old smoker after sulprostone due to coronary spasm, and three women suffered from severe hypotension after prostaglandin administration necessitating the infusion of macromolecular solutes.
The complete abortion rate was 97% in a more recent report on 2000 consecutive cases15 treated up to 63 days LMP with a dose of 200 mg of mifepristone followed 36 to 48 hours later by 0.8 mg of vaginal misoprostol the complete abortion rate was 97.5%. It should be noted, however, that 678 women were not seen by the research team at followup. Two percent of women had incomplete expulsions and 0.6 % required vacuum aspiration for continuing pregnancy. In addition, 2% of the women needed oxytocics at the time of abortion because of heavy bleeding, and less than 1% required haemostatic uterine evacuation or blood transfusions. The most common problems reported at follow-up were continued pain, vaginal bleeding, and offensive discharge. Antibiotics were prescribed for 5% of the 1322 women for presumed genital infection. Although not directly comparable with the previous report, the results with vaginal misoprostol appeared to be better than with gemeprost and sulprostone even though 54% of the women treated with misoprostol had gestational ages of 50 to 63 days.
Continuing pregnancies must be terminated surgically. However, if the woman changes her mind or the clinician fails to follow up or make a diagnosis, some pregnancies will continue to term. It is estimated that 405,000 early medical abortions were done in France, Sweden, and the United Kingdom between 1987 -1998.30 Eight malformations were reported among the 71 cases of continuing pregnancy; five were discovered during pregnancy and led to therapeutic abortions and three (fingernail defect, bilateral talipes equinovares, heart malformation) were found after delivery.30 These data show both the safety of medical abortion and the importance of rigorous adherence to the recommended procedure and proper counselling of women.
Conclusion
Research carried out during the last 20 years on mifepristone and prostaglandins has led to the development of a medical abortion regimen that has proven to be effective and safe when properly provided. Several recent studies have demonstrated that it is possible to lower the dose of 600 mg of mifepristone to 200 mg without affecting efficacy. These results have still to be put into practice as, with the exception of China, the only approved dose is 600 mg. Vaginal misoprostol makes the regimen effective up to nine weeks gestation, reduces the cost, and is more convenient to use than gemeprost. No doubt, the mifepristone-misoprostol approach is a valuable option to women seeking early termination of pregnancy. Medical abortion has also meant major changes in services; in several settings it has moved abortion out of operating theatres and made services more cost effective. The future will show whether the regimens used today can be improved even further.
References:
1. Creinin MD, Darney PD. Methotrexate and misoprostol for early abortion. Contraception. 1993;48:339-348.
2. Hausknecht RV. Methotrexate and misoprostol to terminate early pregnancy. N Engl J Med. 1995;333:537-540.
3. Swahn ML, Bygdeman M. Medical methods to terminate pregnancy. In: Bygdeman M, ed. Medical Induction of Abortion. Volume 4. Bailliere's Clinical Obstetrics and Gynaecology. London: Bailliere Tindall; 1990:293-306.
4. Herrmann W, Wyss R, Riondel A, et al. Effet d'un stéroide anti-progesterone chez la femme: interruption du cycle menstruel et de la grosesse au début. C R Acad Sci Paris. 1982;294:933-938.
5. Kovacs L, Sas M, Resch BA, et al. Termination of pregnancy by RU486 - an antiprogestational compound. Contraception. 1984;29:399-410.
6. Bygdeman M. Termination of pregnancy up to 8 or 9 weeks. In: eds. Baird D, Grimes D, Van Look P. Modern Methods of Inducing Abortion. Cornwall, U.K. Blackwell Science. 1995:39-53.
7. Bygdeman M, Swahn ML. Progesterone receptor blockage - effect on uterine contractility and early pregnancy. Contraception. 1985;32:45-51.
8. Ulmann A, Silvestre L, Chemama L, et al. Medical termination of early pregnancy with mifepristone (RU 486) followed by a prostaglandin analogue. Acta Obstet Gynecol Scand. 1992;71:278-283.
9. Van Look P, von Hertzen H. Clinical uses of antiprogestogens. Hum Reprod Update. 1995;1:19-34.
10. Heikinheimo O, Kekkonen R. Dose-response relationship of RU 486. Ann Med. 1993;25:71-76.
11. World Health Organization. Pregnancy termination with mifepristone and gemeprost: A multicentre comparison between repeated doses and a single dose of mifepristone. Fertil Steril. 1991;56:32-40.
12. World Health Organization Task Force on Post-ovulatory Methods of Fertility Regulation. Termination of pregnancy with reduced doses of mifepristone. Br Med J. 1993;307:532-537.
13. World Health Organization Task Force on Post-ovulatory Methods of Fertility Regulation. Comparison of two doses of mifepristone in combination with misoprostol for early medical abortion: a randomized trial. Br J Obstet Gynaecol. In press.
14. McKinley C, Thong KJ, Baird DT. The effect of dose of mifepristone and gestation on the efficacy of medical abortion with mifepristone and misoprostol. Hum Reprod. 1993;8:1502-1505.
15. Ashok PW, Penney GC, Flett GMM, Templeton A. An effective regimen for early medical abortion: A report of 2000 consecutive cases. Hum Reprod. 1998;13:2962-2965.
16. von Hertzen H, Van Look PFA. Post-ovulatory methods of fertility regulation. In: Annual Technical Report 1995. Geneva: World Health Organization, 1996;49-71.
17. Peyron R, Aubeny E, Targosz V, et al. Early termination of pregnancy with mifepristone (RU 486) and the orally active prostaglandin misoprostol. N Engl J Med. 1993;328:1509-1513.
18. Spitz I, Bardin W, Benton L, Robbins A. Early pregnancy termination with mifepristone and misoprostol in the United States. N Engl J Med. 1998;338:1241-1247.
19. Gonzales CH, Marques-Diaz MJ, Kim CA, et al. Congenital abnormalities in Brazilian children associated with misoprostol misuse in first trimester pregnancy. Lancet. 1998;351:1624-1627.
20. El-Refaey H, Rajasekar D, Abdalla M, Calder L, Templeton A. Induction of abortion with mifepristone (RU 486) and oral or vaginal misoprostol. N Engl J Med. 1995; 332:983-987.
21. Ho PC, Ngai SW, Liu KI, Wong GCY, Lee SWH. Vaginal misoprostol compared to oral misoprostol in termination of second trimester pregnancy. Obstet Gynaecol. 1997;90:735-738.
22. Gemzell-Danielsson K, Marions L, Rodriguez A, Spur BW, Wong K, Bygdeman M. Comparison between oral and vaginal administration of misoprostol on uterine contractility. Obstet Gynecol. 1999;93:275-280.
23. Zieman M, Fong SK, Benowitz NL, Banskter D, Darney PD. Absorption kinetics of misoprostol with oral or vaginal administration. Obstet Gynecol. 1997;90:88-92.
24. Singh K, Fong YF, Prasad RN, Dong F. Does an acidic medium enhance the efficacy of vaginal misoprostol for pre-abortion cervical priming? Hum Reprod. 1999;14:1635-1637.
25. Creinin MD, Carbonell JL, Schwartz JL, Varela L, Tanda R. A randomized trial of the effect of moistening misoprostol before vaginal administration when used with methotrexate for abortion. Contraception. 1999;59:11-16.
26. Blanchard K, Winikoff B, Ellertson C. Misoprostol used alone for the termination of early pregnancy. Contraception. 1999;59:209-217.
27. Winikoff B. Acceptability of medical abortion in early pregnancy. Fam Plann Perspect. 1995;27:142-148.
28. Schaff EA, Eisinger SH, Stadius LS Franks P, Gore BZ, Poppema S. Low-dose mifepristone 200 mg and vaginal misoprostol for abortion. Contraception. 1999;59:1-6.
29 Tang OS, Gao PP, Cheng L, Lee SWH, Ho PC. A randomized double-blind placebo-controlled study to assess the effect of oral contraceptive pills on the outcome of medical abortion with mifepristone and misoprostol. Hum Reprod. 1999;14:722-725.
30 Sitruk-Ware R, Davey A, Sakiz E. Fetal malformation and failed medical termination of pregnancy. Lancet. 1998;352:323.