Complications of first-trimester abortion by vacuum aspiration after cervical preparation with and without misoprostol: a multicentre randomised trial.

BACKGROUND: Little information is available about the incidence of complications from vacuum aspiration for first-trimester abortion after cervical preparation with prostaglandin analogues. We compared incidence of complications from vacuum aspiration in women who had had cervical preparation with misoprostol and those who had not. METHODS: We did a randomised parallel-group trial at 14 centres in nine countries between Oct 22, 2002, and Sept 24, 2005. Healthy women seeking first-trimester abortion were randomly assigned via a computer-generated randomisation sequence stratified by centre, to receive vaginal administration of either two 200 µg tablets of misoprostol or two placebo tablets 3 h before abortion by vacuum aspiration. Participants and health-care personnel other than staff administering the treatment were masked to group assignment. Follow-up was up to 2 weeks. The primary outcome was one or more complications of vacuum aspiration (cervical tear, uterine perforation, incomplete abortion, uterine re-evacuation, pelvic inflammatory disease, or any other serious adverse event). We included women undergoing treatment and vacuum aspiration in the analysis of immediate complications; whereas, in the analysis of delayed complications, we included only those followed-up. In the analysis of any immediate or delayed complication, we excluded women lost to follow-up. This trial is registered, number ISRCTN85366519. FINDINGS: We randomly assigned 2485 women to the misoprostol group and 2487 to the placebo group. Two women in the misoprostol group did not have vacuum aspiration. 56 women in each group were lost to follow-up. 50 (2%) of 2427 women in the misoprostol group and 74 (3%) of 2431 in the placebo group had one or more complication of vacuum aspiration (relative risk [RR] 0·68, 95% CI 0·47-0·96). No women in the misoprostol group had cervical tears and three had uterine perforations compared with two women in the placebo group who had cervical tears and one who had perforation. 19 (<1%) women given misoprostol and 55 (2%) on placebo had incomplete abortions (0·35, 0·21-0·58), of whom 14 (<1%) versus 48 (2%) needed uterine re-evacuation (0·29, 0·16-0·53). We noted no difference between groups in incidence of pelvic inflammatory disease (30 [1%] vs 25 [1%]; RR 1·20, 0·71-2·04) or other serious adverse events. The main side-effects of misoprostol during the 3 h treatment were abdominal pain (1355 [55%] of 2484 women vs 545 [22%] of 2487 women in the placebo group) and vaginal bleeding (909 [37%] vs 167 [7%]). INTERPRETATION: Cervical preparation with 400 µg of vaginal misoprostol can reduce incidence of complications from vacuum aspiration for first trimester abortion. FUNDING: UN Development Programme/UN Population Fund/WHO/World Bank Special Programme of Research, Development and Research Training in Human Reproduction, Department of Reproductive Health and Research, WHO, and the Packard Foundation.

Effect on pregnancy rates of the delay in the administration of levonorgestrel for emergency contraception: a combined analysis of four WHO trials.

BACKGROUND: Levonorgestrel is an effective method for emergency contraception (EC) and is used worldwide. Consistent with its mechanism of action in delaying ovulation, the earlier it is administered within 72 h of an unprotected act of intercourse, the more effective it is. There is uncertainty, however, about its effectiveness after 72 h. This analysis explores the effect of 24-h intervals of delay in levonorgestrel administration on pregnancy rates when used until 120 h of an unprotected act of intercourse. STUDY DESIGN: Data were analyzed from 6794 women participating in four World Health Organization randomized trials and receiving 1.5 mg of levonorgestrel for EC in a single dose or split into two doses 12 h apart, within 48, 72 or 120 h of an act of unprotected intercourse. The pregnancy rates among women in successive days after an unprotected act of intercourse and odds ratios of pregnancy were calculated using logistic regression with the first day as the reference. RESULTS: For the four trials combined, odds ratios for pregnancy in the second, third and fourth day with respect to the first day were not significantly different from 1 at the 5% level of significance. On the fifth day, the odds ratio of pregnancy compared to the first day was almost 6. CONCLUSIONS: Levonorgestrel for EC should be administered as soon as possible after unprotected intercourse. Delaying levonorgestrel administration until the fifth day after unprotected intercourse increases the risk of pregnancy over five times compared with administration within 24 h. It is uncertain whether levonorgestrel administration on the fifth day still offers some protection against unwanted pregnancy.

A randomized, double-blind, noninferiority study to compare two regimens of levonorgestrel for emergency contraception in Nigeria.

BACKGROUND: Unplanned pregnancies are common in Nigeria. Much of the unplanned pregnancies is due to low contraceptive prevalence and high contraceptive user failure rates. High user failure rates suggest the important role of emergency contraception to prevent unplanned pregnancy. STUDY DESIGN: Randomized, controlled, double-blind, multicenter, noninferiority trial comparing efficacy and side effects of two emergency contraceptive regimens up to 5 days after unprotected intercourse among 3022 Nigerian women: levonorgestrel administered in two doses of 0.75 mg given 12 h apart and levonorgestrel administered in a single dose of 1.5 mg. RESULTS: Efficacy was similar between the treatment groups; post-treatment pregnancy proportions were 0.57% in the two-dose regimen vs. 0.64% in the single-dose regimen (risk difference 0.07% (95% CI -0.50 to 0.64). The majority of women menstruated the first day of expected menses and the groups did not differ regarding reported side effects. CONCLUSIONS: This study shows the simplified emergency contraceptive regimen of single-dose levonorgestrel is not inferior in efficacy to the two-dose regimen among Nigerian women.

Emergency contraception: the state of the art.

Emergency contraception, otherwise known as post-coital contraception, is part of the continuum of contraceptive methods that women and couples can use for pregnancy prevention. Although emergency contraception should not be used as a regular, plan-ahead contraceptive method, it gives a woman one last-ditch effort to prevent unwanted pregnancy. This paper reviews the history of emergency contraception, the need for further studies, and results of studies conducted at the World Health Organization. Various methods used for emergency contraception are discussed, as well as their efficacies and side effects.

Comparison of vaginal and sublingual misoprostol for second trimester abortion: randomized controlled equivalence trial.

BACKGROUND: To identify an effective misoprostol-only regimen for the termination of second trimester pregnancy, we compared sublingual and vaginal administration of multiple doses of misoprostol in a randomized, placebo-controlled equivalence trial. METHODS: Six hundred and eighty-one healthy pregnant women requesting medical abortion at 13-20 weeks' gestation were randomly assigned within 11 gynaecological centres in seven countries into two treatment groups: 400 microg of misoprostol administered either sublingually or vaginally every 3 h up to five doses, followed by sublingual administration of 400 microg misoprostol every 3 h up to five doses if abortion had not occurred at 24 h after the start of treatment. We chose 10% as the margin of equivalence. The primary end-point was the efficacy of the treatments to terminate pregnancy in 24 h. Successful abortion within 48 h was also considered as an outcome along with the induction-to-abortion-interval, side effects and women's perceptions on these treatments. RESULTS: At 24 h, the success (complete or incomplete abortion) rate was 85.9% in the vaginal administration group and 79.8% in the sublingual group (difference: 6.1%, 95% CI: 0.5 to 11.8). Thus, equivalence could not be concluded overall; the difference, however, was driven by the nulliparous women, among whom vaginal administration was clearly superior to sublingual administration (87.3% versus 68.5%), whereas no significant difference was observed between vaginal and sublingual treatments among parous women (84.7% versus 88.5%). The rates of side effects were similar in both groups except for fever, which was more common in the vaginal group. About 70% of women in both groups preferred sublingual administration. CONCLUSIONS: Equivalence between vaginal and sublingual administration could not be demonstrated overall. Vaginal administration showed a higher effectiveness than sublingual administration in terminating second trimester pregnancies, but this result was mainly driven by nulliparous women. Fever was more prevalent with vaginal administration. Registered with International Standard Randomized Controlled Trial number ISRCTN72965671.

Comparison of two doses and two routes of administration of misoprostol after pre-treatment with mifepristone for early pregnancy termination.

BACKGROUND: It is not known whether a 400 microg dose of misoprostol has a similar efficacy as an 800 microg dose when administered sublingually or vaginally 24 hours after 200 mg mifepristone. METHODS: It is proposed to undertake a placebo-controlled, randomized, non-inferiority trial (3% margin of equivalence) of the two misoprostol doses when administered sublingually or vaginally using factorial design. A total of 3008 pregnant women (< 63 days of gestational age) who request legal termination of pregnancy will be recruited for the trial at 16 clinics in ten countries providing abortion services. Eligible women willing to join the study will be allocated randomly to one of the four treatment groups within each centre. Women in all treatment groups will first receive 200 mg mifepristone, followed 24 hours later by either 400 microg or 800 microg misoprostol, administered either sublingually or vaginally. The dose and route of administration of misoprostol will be blinded to women, each woman receiving four tablets vaginally and four tablets sublingually, two or four of which are 200 microg tablets of misoprostol and the rest are placebo tablets.The four treatment regimens will be compared in terms of: (i) their efficacy to induce complete abortion; (ii) induction-to-abortion interval when possible; (iii) the frequency of side effects; and (iv) women's perceptions. The initial judgment of the outcome of treatment is made at the follow-up visit on day 15 of the study and the final assessment four weeks later. It is estimated that the clinical phase will require 12-14 months for data collection.To compare the two routes and two doses, relative risks (RR) of failure to achieve a complete abortion and failure to terminate pregnancy and the two-sided 95% CIs will be calculated by standard methods, as well as risk differences and two-sided 95% CIs. The latter will be used to test the non-inferiority hypotheses (at 2.5% level of significance) for achieving complete abortion. The factorial structure will be taken into account in the analysis after testing the interaction. TRIAL REGISTRATION: ISRCTN87811512.

Efficacy of two intervals and two routes of administration of misoprostol for termination of early pregnancy: a randomised controlled equivalence trial.

BACKGROUND: The most effective route and best interval between several doses of misoprostol to induce abortion have not been defined. Our aim was to assess the effects of the interval between multiple doses of misoprostol and the route of administration to terminate pregnancy. METHODS: 2066 healthy pregnant women requesting medical abortion with 63 days or less of gestation were randomly assigned within 11 gynaecological centres in six countries to the four treatment groups (three doses of 0.8 mg misoprostol given sublingually at 3-h intervals, vaginally 3 h, sublingually 12 h, and vaginally 12 h), stratifying by gestational age. This was an equivalence trial with a 5% margin of equivalence. The primary endpoints were efficacy of treatment to achieve complete abortion and to terminate pregnancy. The main efficacy analysis excluded women lost to follow-up. This trial is registered as an International Standard Randomised Controlled Trial, number ISRCTN10531821. FINDINGS: Efficacy outcomes were analysed for 2046 women (99%), excluding 20 lost to follow-up. Complete abortion rates at 2-week follow-up were recorded for 431 (84%) in the sublingual and for 434 (85%) women in the vaginal group when misoprostol was given at 3-h intervals (difference 0.4%, 95% CI -4.0 to 4.9, p=0.85 equivalence shown), and for 399 (78%) in the sublingual and for 425 (83%) in the vaginal 12-h groups (4.6%, -0.2 to 9.5, p=0.06, equivalence not shown). In the 3-h groups, pregnancy continued in 29 (6%) women after sublingual and in 20 (4%) women after vaginal administration (difference 1.8%, 95% CI -0.8 to 4.4, p=0.19, equivalence shown); in the 12-h groups it continued in 47 (9%) after sublingual and in 25 (5%) after vaginal administration (4.4%, 1.2-7.5, p=0.01, vaginal better than sublingual). Differences for complete abortion between intervals for sublingual and vaginal routes were 6% (95% CI 1.0-10.6, p=0.02, 3 h better than 12 h) and 2% (-2.9 to 6.1, p=0.49, equivalence not shown), respectively; for continuing pregnancies they were 4% (0.4-6.8, p=0.03, 3 h better than 12 h) and 1% (-1.5 to 3.5, p=0.44, equivalence shown), respectively. INTERPRETATION: Administration interval can be chosen between 3 h and 12 h when misoprostol is given vaginally. If administration is sublingual, the intervals between misoprostol doses need to be short, but side-effects are then increased. With 12-h intervals, vaginal route should be used, whereas with 3-h intervals either route could be chosen.

Pharmacokinetics of a single oral dose of 1.5-mg levonorgestrel when administered as 750-mug tablets or as 30-microg minipills.

We examined the pharmacokinetics of a single dose of 1.5 mg of levonorgestrel when administered orally in two different formulations: two tablets of 0.75 mg or 50 minipills of 30 microg of levonorgestrel. Bioavailability of levonorgestrel with minipills was comparable to that with levonorgestrel tablets. These findings suggest that levonorgestrel-containing minipills can be considered as an alternative to standard levonorgestrel tablets for use in emergency contraception.

Frequently asked questions about medical abortion.

The development of methods of inducing abortion medically (nonsurgically) has created alternative options to make abortion available to women in a variety of health-care settings. Medical abortion is induced with a regimen of mifepristone followed by a prostaglandin analogue. Since its first introduction in the late 1980s, the regimen has undergone some modifications based on research evidence, and, in many countries, clinicians are using regimens that may differ from the one that has been licensed. This causes confusion among providers, also because only a few countries have developed national guidelines for the provision of medical abortion. We approached health care personnel providing abortion services in various countries and asked them to send us questions that they, or their colleagues, might have concerning the clinical practice of medical abortion in the early first trimester (up to 63 days since the first day of the last menstrual period). These questions were sent to experts representing the fields of biomedical and clinical research, clinical practice and family planning, who conducted literature reviews so that, whenever possible, the answers could be based on existing evidence. A consensus meeting was held in Bellagio, Italy, between November 1 and 5, 2004, to review the questions and to discuss the answers. The aim of this article is to provide a brief overview of some of the questions asked and the answers discussed.

Pharmacokinetics and endometrial tissue levels of levonorgestrel after administration of a single 1.5-mg dose by the oral and vaginal route.

OBJECTIVE: To determine the pharmacokinetics and endometrial tissue levels of levonorgestrel when taken as a single dose of 1.5 mg either orally or vaginally by healthy women in the periovulatory phase of their menstrual cycle. DESIGN: Prospective randomized study. SETTING: Academic research institution. PATIENT(S): Thirty women with regular cycles allocated to control (n = 5), oral (n = 13), and vaginal (n = 12) groups. INTERVENTION(S): Blood samples were drawn before (0 time) and at 0.5, 1, 2, 4, 6, 8, 24, 48, and 168 hours after levonorgestrel administration. Endometrial samples were collected 24 and 168 hours after levonorgestrel administration. MAIN OUTCOME MEASURE(S): Plasma and endometrial tissue levels of levonorgestrel. RESULT(S): Plasma concentrations of levonorgestrel were significantly greater during the first 48 hours after oral administration. However, 7 days after levonorgestrel administration, the plasma levels were similar for both treatments (3-5 nmol/L). Compared with vaginal administration, oral administration resulted in higher peak plasma concentrations (Cmax 64 vs. 10.7 nmol/L), with a shorter time to reach the maximal concentrations (Tmax 1.4 vs. 6.6 hours) and with a greater AUC (509 vs. 175 nmol/L). Interestingly, the half-life of levonorgestrel was shorter after oral administration (25 hours vs. 32.6 hours). Levonorgestrel tissue concentrations were not related to the plasma levels. Levonorgestrel values tended to be higher in endometrial tissue after vaginal administration. The ratio between plasma and endometrial concentrations of levonorgestrel differed significantly between the groups. CONCLUSION(S): These data indicate that orally administered levonorgestrel achieves higher plasma levels sooner than vaginally administered levonorgestrel. However, plasma levels after vaginal administration are more sustained and were likely to be sufficient for ovarian suppression. Therefore, the vaginally administered levonorgestrel could be considered as an alternative option for emergency contraception.

Late follicular phase administration of mifepristone suppresses circulating leptin and FSH - mechanism(s) of action in emergency contraception?

OBJECTIVE: Low dose mifepristone (RU486) is highly effective in emergency post-coital contraception (EC), although the mechanism(s) of action remains unclear. We studied the endocrine actions of 10 mg mifepristone administered orally as a single dose to eight healthy volunteers (aged 20-45 years) during the late follicular phase. METHODS: Serum levels of LH, FSH, oestradiol, progesterone, leptin, mifepristone, cortisol, and gluco-corticoid bioactivity (GBA) were measured before and 1, 2, 4 and 8 h after ingestion of mifepristone on cycle day 10 or 11 (study day 1), and follow-up was continued for 10 days. Ovarian ultrasonography was performed on study days 1 and 7. Similar measurements were carried out during a control cycle. RESULTS: Mifepristone postponed ovulation, as evidenced by a 3.4+/-1.1 day (means+/-s.d.) delay (P < 0.005) in the LH surge and 3.6+/-4.0 day prolongation of the treatment cycle (P = 0.08). During the mifepristone cycle, an LH surge was displayed by five subjects when serum mifepristone levels had declined to 9.5+/-7.1 nmol/l. During the day of mifepristone administration, circulating GBA (P < 0.001) and leptin (P < 0.001) levels declined. On the day after mifepristone administration, mean serum FSH and leptin levels were lower than pretreatment values (3.8+/-1.8 IU/l vs 5.2+/-1.1 IU/l, n = 7, P < 0.05; 28.9+/-6.7 microg/l vs 33.2+/-9.0 microg/l, n = 7, P < 0.05 respectively), and the corresponding difference in the mean serum oestradiol concentration was borderline (452+/-252 pmol/l vs 647+/-406 pmol/l, n = 7, P = 0.056). In contrast to the control cycle, individual leptin levels declined during the follow-up after ingestion of mifepristone (n = 8, P < 0.01). CONCLUSIONS: These data showed that the commonly employed dose of mifepristone for EC delays ovulation and prolongs the menstrual cycle, when given during the late follicular phase. The mechanism of action of mifepristone may include a reduction of FSH secretion via a decrease in circulating leptin.

Antiprogestins for contraception?

Antiprogestins could be a very promising group of compounds for contraception because they counteract the effects of progesterone, a key hormone for normal reproductive functioning. Great efforts have therefore been devoted to testing various ways of using these compounds for family planning. Most of this work has involved mifepristone; other antiprogestins have not been available for trials. With a few exceptions, the outcome of the studies performed to date has not met expectations. The most promising approaches seem to be the use of antiprogestins for emergency contraception, perhaps also as a daily pill when the dose is high enough to block ovulation, and in sequential regimens followed by a progestin. Given that antiprogestins differ in their affinity for target organs, better results could possibly be achieved by using more specific compounds than mifepristone. It is hoped that a wider choice of antiprogestational compounds will soon become available for research.

Tolerability of levonorgestrel emergency contraception in adolescents.

OBJECTIVE: We evaluated the tolerability of emergency contraception in adolescents. Study design In this descriptive study, 1 0.75 mg levonorgestrel tablet was administered to 52 females aged 13-16 with instructions to take the second tablet 12 hours later (unprotected intercourse was not an entry requirement). Participants kept diaries of side effects and menstrual patterns. We assessed correct use, side effects caused by treatment, and impact on menstrual cycle. RESULTS: Virtually all participants used the drug correctly, with no serious adverse events. Minor expected side effects occurred, including nausea, fatigue, and vomiting. There was no difference in reporting of side effects by age. Adolescents' mean duration of menses was comparable pre- and post-treatment (5.3 vs 5.0 days; P=.146), and onset of menses was within the expected range. Ninety percent of participants reported they would recommend emergency contraception to a friend or relative if needed. CONCLUSION: Adolescents tolerated the medication well, experiencing transient side effects.

Pharmacokinetics of 10 mg of mifepristone.

The results of several randomized studies have verified the efficacy of 10 mg mifepristone in emergency contraception. In the present study we characterized the pharmacokinetics of 10 mg mifepristone. Eight healthy female volunteers received a single oral dose of mifepristone on the day 10 or 11 of their menstrual cycle. Blood samples were collected at 0, 1, 2, 4 and 8 h, daily for the next 6 days and on day 10 after mifepristone. Mifepristone concentrations were determined by radioimmunoassay preceded by column chromatography. A peak level of 1.41 +/- 0.31 micromol/L (mean +/- SD) was measured at 1 h. Individual elimination phase half-lives varied from 15.3 to 26.8 h, the mean (+/- SD) value being 19.6 +/- 4.50 h. Serum mifepristone concentrations exceeded 10 nmol/L in all volunteers for an average of 4.9 days. The pharmacokinetic data on 10 mg mifepristone are in line with previous pharmacokinetic and clinical data, and encourage further development of the 10-mg dose in emergency contraception.

Expanded clinical trial of emergency contraception with 10 mg mifepristone.

We conducted a clinical single-arm trial to evaluate the effectiveness of 10 mg mifepristone for emergency contraception (EC) in a large population in China. The participating centers were 31 family-planning clinics and hospitals in the following 19 provinces or municipalities in China: Beijing, Shanghai, Tianjin, Harbin, Changchun, Shengyang, Shijiazhuang, Zhengzhou, Taiyuan, Nanjing, Jinan, Hangzhou, Guangzhou, Wuhan, Changsha, Chongqing, Guiyang, Chengdu, Kunming. A total of 4945 women requesting EC within 120 h after a single act of unprotected intercourse were recruited and treated with 10 mg mifepristone. A total of 28 women were lost to follow-up, and 4917 women were included in the analysis, of whom, 69 became pregnant. The combined pregnancy rate was 1.4 [95% confidence interval (CI): 1.0-1.9] and the percentage of pregnancies prevented was 82.2% (95% CI: 77.5-86.2%). There was a significant inverse trend in pregnancy rate with body mass index that disappeared when adjusted for other variables. The pregnancy risk was double among nulliparous women compared to parous women (2.3% compared to 1.0%), and it increased by a factor of 1.5 when the treatment was administered at 25-48 h and at 49-72 h compared to administration within 24 h, although this association was not significant. The risk of pregnancy was higher if intercourse took place during the follicular or preovulatory phase of the cycle. Women having repeated intercourse after treatment without any contraceptive methods had a dramatic increase in the risk of pregnancy, while those who used contraceptives had similar risk to those without acts. Side effects were mild and present in only small proportions of women: nausea and vomiting in 9% and other side effects in 2-3%. Delay of menstruation of 7 days or more occurred in 6.5% of women. The expanded study confirmed the high efficacy and safety of 10 mg mifepristone for EC.

Combined estimates of effectiveness of mifepristone 10 mg in emergency contraception.

The present paper combines the estimates of efficacy and side effects of 10 mg mifepristone for emergency contraception obtained from randomized trials. A total of 6083 women participating in 12 randomized trials and receiving 10 mg mifepristone for emergency contraception up to 120 h after intercourse, were analyzed for efficacy. Between 4188 and 5833 women were analyzed for side effects and 3601 for delay of menses of more than 7 days. Prevented fractions, the effect of delay and of further acts of intercourse after treatment administration were analyzed in 3440 women, using individual data. The combined pregnancy rate from all the 12 trials was 1.7% [101/6083, 95% confidence interval (CI): 1.3-2.2]. From the three trials providing individual data, the combined pregnancy rate was 1.3% (45/3440, 95% CI: 0.9-1.7) and the estimate of pregnancies prevented was 83.4% (95% CI: 77.4-87.8). There was a sharp decline in efficacy when treatment was administered during the 5th day after intercourse compared to administration during the 1st day, the odds of pregnancy increasing by a factor of 5.3 (95% CI: 1.9-14.9). The relative risk of pregnancy was about 28 times higher among women with unprotected acts of coitus between treatment administration and the onset of next menses, compared with women reporting none [odds ratio (OR) = 27.6, 95% CI: 12.7-60.2]. The increase in risk for women reporting protected acts of intercourse during this interval was not statistically significant (OR = 1.8, 95% CI: 0.9-3.8). There was a large heterogeneity among trials in all side effects and delay of menses of more than 7 days (all had p < 0.0001 for the test of homogeneity). The percentage of women with nausea ranged from 0.0-19.4% (highest upper 95% confidence limit: 23.0%), that of vomiting from 0.0-4.3% (highest upper 95% confidence limit: 6.1%), that of lower abdominal pain from 4.3-19.1% (highest upper 95% confidence limit: 22.7%). The percentage of women with delay of menses of more than 7 days ranged from 4.3-25.8% (highest upper 95% confidence limit: 34.1%). We conclude that 10 mg mifepristone is an effective emergency contraception regimen, with an acceptable side-effects profile. Postponing treatment until the 5th day seriously decreases efficacy. The risk of pregnancy is dramatically increased among women having unprotected acts of intercourse between treatment administration and the onset of next menses. This risk may be enhanced for women whose ovulation is postponed by treatment.

Meta-analyses of randomized trials comparing different doses of mifepristone in emergency contraception.

There is some evidence from randomized trials that different doses of mifepristone for emergency contraception do not differ in efficacy in the range from 10 mg to 600 mg. Lower doses have a better side effect profile and are cheaper and therefore they would be preferable in the absence of a dose effect. However, the lack of significance is not evidence of absence of an effect. More evidence can be obtained by combining results of trials. We present meta-analyses of randomized trials comparing doses of mifepristone for emergency contraception from 5 mg to 600 mg, with regard to the efficacy to prevent unwanted pregnancies. We use two approaches for analysis, one using only within-trial information and another one combining within-trial with between-trial information. We discuss the results in terms of equivalence. There is some evidence of a small dose effect on efficacy in the lower range of doses (<50 mg). The pregnancy rate increases by a factor of 1.6 when the dose of 10 mg is used instead of 25 mg (95% confidence interval: 1.1-2.4). In terms of the number of women needed to treat, however, using 10 mg in the place of 25 mg implies having one extra pregnancy every 146 women requesting emergency contraception, which might be a low cost compared to the benefit of more women having access to treatment.