Medical methods for first trimester abortion

Regina Kulier; Nathalie Kapp; A Metin Gülmezoglu; G Justus Hofmeyr; Linan Cheng; Aldo Campana

doi:10.1002/14651858.CD002855.pub4

. 2011 Nov 9;2011(11):CD002855. doi: 10.1002/14651858.CD002855.pub4

Medical methods for first trimester abortion

Regina Kulier ^1,^✉, Nathalie Kapp ², A Metin Gülmezoglu ³, G Justus Hofmeyr ⁴, Linan Cheng ⁵, Aldo Campana ⁶

Editor: Cochrane Fertility Regulation Group

PMCID: PMC7144729 PMID: 22071804

Abstract

Background

Surgical abortion by vacuum aspiration or dilatation and curettage has been the method of choice for early pregnancy termination since the 1960s. Medical abortion became an alternative method of first trimester pregnancy termination with the availability of prostaglandins in the early 1970s and anti‐progesterones in the 1980s. The most widely researched drugs are prostaglandins (PGs) alone, mifepristone alone, methotrexate alone, mifepristone with prostaglandins and methotrexate with prostaglandins.

Objectives

To compare different medical methods for first trimester abortion.

Search methods

The Cochrane Controlled Trials Register, MEDLINE and Popline were systematically searched. Reference lists of retrieved papers were also searched. Experts in WHO/HRP were contacted.

Selection criteria

Types of studies   Randomised controlled trials comparing different medical methods for abortion during first trimester (e.g. single drug, combination) were considered. Trials were assessed and included if they had adequate concealment of allocation, randomisation procedure and follow‐up. Women, pregnant during the first trimester, undergoing medical abortion were the participants. The outcomes were mortality, failure to achieve complete abortion, surgical evacuation, ongoing pregnancy at follow‐up, time until passing of conceptus, blood transfusion, side effects and women's dissatisfaction with the procedure.

Data collection and analysis

Two reviewers independently selected trials for inclusion from the results of the search strategy described previously.The selection of trials for inclusion in the review was performed independently by two reviewers after employing the search strategy described previously. Trials under consideration were evaluated for appropriateness for inclusion and methodological quality without consideration of their results. Data were processed using Revman software.

Main results

Fifty‐eight trials were included in the review. The effectiveness outcomes below refer to 'failure to achieve complete abortion' with the intended method unless otherwise stated. 1) Combined regimen mifepristone/prostaglandin: Mifepristone 600 mg compared to 200 mg shows similar effectiveness in achieving complete abortion (4 trials, RR 1.07, 95% CI 0.87 to 1.32). Misoprostol administered orally is less effective (more failures) than the vaginal route (RR 3.00, 95% CI 1.44 to 6.24) and may be associated with more frequent side effects such as nausea and diarrhoea. Sublingual and buccal routes were similarly effective compared to the vaginal route, but had higher rates of side effects. 2) Mifepristone alone is less effective when compared to the combined regimen mifepristone/prostaglandin (RR 3.76 95% CI 2.30 to 6.15). 3) Five trials compared prostaglandin alone to the combined regimen (mifepristone/prostaglandin). All but one reported higher effectiveness with the combined regimen. The results of these studies could not be combined but the RR of failure with prostaglandin alone is reportedly between 1.4 to 3.75 with the 95% confidence intervals indicating statistical significance. 4) In one trial comparing gemeprost 0.5 mg with misoprostol 800 mcg, misoprostol was more effective (failure with gemeprost: RR 2.86, 95% CI 1.14 to 7.18). 5) There was no difference in effectiveness with use of a divided dose compared to a single dose of prostaglandin. 6) Combined regimen methotrexate/prostaglandin demonstrates similar rates of failure to complete abortion when comparing intramuscular to oral methotrexate administration (RR 2.04, 95% CI 0.51 to 8.07). Similarly, day 3 vs. day 5 administration of prostaglandin following methotrexate administration showed no significant differences (RR 0.72, 95% CI 0.36 to 1.43). One trial compared the effect of tamoxifen vs. methotrexate and no statistically significant differences were observed in effectiveness between the groups.

Authors' conclusions

Safe and effective medical abortion methods are available. Combined regimens are more effective than single agents. In the combined regimen, the dose of mifepristone can be lowered to 200 mg without significantly decreasing the method effectiveness. Vaginal misoprostol is more effective than oral administration, and has less side effects than sublingual or buccal. Some results are limited by the small numbers of participants on which they are based. Almost all trials were conducted in settings with good access to emergency services, which may limit the generalizability of these results.

Plain language summary

Medical methods for early termination of pregnancy can be safe and effective

There are several different surgical techniques for abortion during the first three months. Several drugs can also be prescribed alone or in combination to terminate early pregnancy. This is called medical abortion, and uses the hormones prostaglandins and/or mifepristone (an antiprogesterone often called RU486), and/or methotrexate. This review of trials found that medical methods for abortion in early pregnancy can be safe and effective, with the most evidence of effectiveness for a combination of mifepristone and misoprostol (a prostaglandin). Almost all of the trials were done in well‐resourced settings where women returned for a check‐up.

Background

Up to 42 million abortions are performed each year (Sedgh 2007). Medical abortion has the potential to expand abortion services, where surgical services are limited, and to expand women's choice of abortion method and experience.

Surgical abortion by vacuum aspiration or dilatation and curettage has been the method of choice for early pregnancy termination since the 1960s. Medical abortion became an alternative method of first trimester pregnancy termination with the availability of prostaglandins in the early 1970s followed by the development of an antiprogesterone in the 1980s. Large uncontrolled studies suggested that early medical abortion with mifepristone and a prostaglandin would be an effective method for pregnancy termination (Urquhart 1997).

Various drugs have been used for first trimester medical abortion. The most widely researched are prostaglandins (PGs) alone, mifepristone alone, methotrexate alone, mifepristone with prostaglandins and methotrexate with prostaglandins. Prostaglandins soften the cervix, cause uterine contractions and are used orally or vaginally for ripening of the cervix before surgical or for medical abortion. The most commonly used prostaglandins are gemeprost given vaginally and misoprostol administered either orally (including buccal and sublingual) or vaginally. Misoprostol is a prostaglandin analogue registered for use in nonsteroidal anti‐inflammatory drug (NSAID) induced gastric ulcer prevention and treatment. It has a strong uterotonic effect and is used to induce pregnancy terminations illegally in some parts of the world (Blanchard 1999, Costa 1998) as well as legally, in areas where mifepristone is not available. The reported complete abortion rate for misoprostol alone varies between 61% for single and 93% for repeat doses (Bugalho 1996, Carbonell 1997b). Gemeprost used alone appears to be less effective in inducing complete abortion than when used in combination with mifepristone (Norman 1992).

Mifepristone, an antiprogestogen, blocks the receptors for progesterone and glucocorticosteroid and increases the sensitivity of the uterus to prostaglandins (Bygdeman 1985). This blockage results in the breakdown of maternal capillaries in the decidua, the synthesis of prostaglandins by the epithelium of decidual glands and inhibition of prostaglandin dehydrogenase (WHO 1997).

Mifepristone has been licensed in France and China since 1988, in Great Britain in 1991 and, in the USA and India in 2000 and 2002, respectively. Mifepristone given alone has been shown to result in abortion only in 60‐80% of cases, depending on the gestational age and the dose given (WHO 1997). However, in combination with a prostaglandin at up to 63 days of amenorrhoea, it leads to complete abortion in about 95% of pregnancies (United 1990) or more. The effect of mifepristone develops over a time period of 24‐48 hours; therefore, prostaglandins have usually been administered after 36‐48 hours. Currently, different regimens are in use. The recommended regimen by the manufacturer is mifepristone 600 mg followed by misoprostol (between 400 ‐ 800 mcg) or gemeprost (0.5 ‐ 1 mg vaginally) and is registered for abortion in pregnancies up to 49 days in France and up to 63 days of amenorrhoea in Great Britain. However, a reduced dose of mifepristone combined with a prostaglandin has similar effectiveness and has the advantage of being much less expensive (WHO 1997).

Methotrexate has been used successfully for the treatment of unruptured tubal pregnancy. It is a folic acid antagonist which inhibits purine and pyrimidine synthesis and is cytotoxic to the trophoblast. The use of methotrexate with misoprostol for first trimester abortion was first introduced in 1993 (Creinin 1993, Grimes 1997). This combination was more effective when misoprostol was administered 7 days after methotrexate as compared to 3 days, leading to a complete abortion rate of 98% (Creinin 1995 M800pv).

Side‐effects of medical methods are heavy bleeding, pain, nausea, vomiting and diarrhoea, varying in severity according to the protocols and gestational age (Henshaw 1994). In two randomised controlled trials included in the Cochrane review of the subject, compared to surgical procedures, medical methods are associated with a longer duration of bleeding (Say 2002, updated 2010).

Failed abortion is an infrequent but important complication of medical abortion. Both methotrexate and misoprostol may lead to fetal anomalies if the pregnancy persists, as described by some (Grimes 1997). However, other reports state that none of the malformations reported could be conclusively related to medications used for medical abortion (Wiebe 2006).

Some women prefer medical to surgical abortion. 'More natural', 'being easier', more private', and 'can be done earlier in pregnancy' were reasons to opt for a medical method by some women (Creinin 1996b). Characteristics such as the method being more new, less invasive and the possibility of verifying the expulsion were reported by others (Bachelot 1992).

Medical methods for first trimester abortion are already widely available in some countries and increasingly available throughout the world. It is therefore important to identify the best available agents and regimen for use. Comparison of medical methods with surgical evacuation in the first trimester is the subject of another review: Say 2002, updated 2010.

Objectives

To compare different medical methods for first trimester abortion.

Methods

Criteria for considering studies for this review

Types of studies

Randomised controlled trials comparing different medical methods (e.g. single drug, combination), ways of application, or different dose regimens, single or combined, for medical abortion, were considered. Trials were not excluded based on an arbitrary cut‐off limit regarding losses to follow‐up. Trials were excluded if there were unexplained imbalances in different groups at follow‐up and from available outcome data.Trials were assessed and included if they had adequate concealment of allocation, randomisation and follow‐up.

Types of participants

Women, pregnant in the first trimester, undergoing medical abortion.

Types of interventions

Different medical methods used for first trimester abortion, compared with each other or placebo. See 'Search methods for identification of studies' for a list of pharmaceutical preparations.

Types of outcome measures

The main outcome measure was failure to achieve complete abortion. Surgical evacuation (as emergency procedure, non‐emergency procedure, or undefined), ongoing pregnancy at follow‐up, time until passing of conceptus (> 3‐6 hours), blood transfusion, blood loss (measured or clinically relevant drop in haemoglobin), days of bleeding, pain resulting from the procedure (reported by the women or measured by use of analgesics), additional uterotonics used, women's dissatisfaction with the procedure, nausea, vomiting, and diarrhoea were also assessed. Although mortality is considered an important outcome we did not anticipate analyzing abortion‐related mortality within the context of these trials.

Search methods for identification of studies

The Cochrane Controlled Trials Register, MEDLINE and Popline were systematically searched. Reference lists of retrieved papers were also searched. Electronic literature search of MEDLINE (with the Cochrane 3‐stage search strategy)(1966‐2003) and POPLINE (1970‐2003) databases with the following key words: (abortion OR pregnancy termination OR termination of pregnancy) AND (first trimester OR early) AND (mifepristone OR misoprostol OR methotrexate OR dinoprost* OR carboprost OR sulprostone OR gemeprost OR meteneprost OR lilopristone OR onapristone OR epostane OR oxytocin OR RU 486 OR mifegyne). There were no language preferences in the application of the search.

Data collection and analysis

The selection of trials for inclusion in the review was performed independently by two reviewers after employing the search strategy described previously. Trials under consideration were evaluated for appropriateness for inclusion and methodological quality without consideration of their results. A quality score for concealment of allocation has been assigned to each trial, using the criteria described in the Cochrane Handbook:

(A) adequate concealment of the allocation   (B) unclear whether adequate concealment of the allocation   (C) inadequate concealment of allocation (includes quasi‐randomised studies)   (D) allocation concealment not used   Only trials scoring A or B were included in the review.

Failure to achieve complete abortion is defined as an abortion which is not completed by the described intended method. Other outcomes are failure of expulsion after 4 ‐ 6 hours, side effects (nausea, vomiting, diarrhoea, abdominal pain), and mean duration of days of bleeding. A further division into early (≤ 49 days of amenorrhoea) and late (> 49 days) gestational age at the time of abortion was made for subgroup analysis. Complications are defined as any serious complication described by the authors and which was not a failure or side effect.

A form was designed to facilitate the process of data extraction which has been performed by two of the reviewers independently. In case of discrepancies between reviewers in either the decision of inclusion/exclusion of studies or in data extraction, this was resolved by consensus. Attempts were made to obtain additional information from authors if required.   Whether or not an "intention‐to‐treat" analysis was done in the primary study was examined.

Data were processed using RevMan software. For reasons of clarification some coding was added to some trials included in the meta‐analysis: GP ‐gemeprost, the number next to it ‐ refers to the dose of gemeprost in grams, M ‐ misoprostol, the number next to it ‐ refers to the dose in mcg, MP ‐ minprostin, the number next to it refers to the dose in mg, PGF2 ‐ Prostaglandin F2alpha; PGE1‐ prostaglandin E1 analogue; MI ‐ mifepristone ‐ the number next to it refers to the dose in mg; MT ‐ methotrexate, T ‐ testosterone propionate, TM ‐ tamoxifen; po ‐ oral and pv ‐ vaginal administration.

Results are presented as relative risk and 95% confidence interval (RR; 95%CI) using the fixed effects model. If a large I² was found in the pooled analysis, a random effects model was applied and the tau ² value was evaluated for possible heterogeneity and reported if present.

Subgroup analyses were performed where possible for early and late first trimester abortions as the performance of some methods may differ with gestational age: 1) abortion up to 49 days, 2) abortion > 49 days of amenorrhoea.   The studies in this field use various combinations of agents, doses, intervals between the antiprogesterone and prostaglandin, and route of administration for prostaglandin. Since all of these variables may affect the outcomes, it was not considered appropriate to combine similar trials into meta‐analysis in many cases. However, it was possible to identify an experimental intervention and a constant (fixed ) intervention which enabled us to group the trials as follows:

Combined regimen mifepristone/prostaglandin:

Intervention: dose of mifepristone (comparison 1)
Intervention: dose of prostaglandin (comparison 2)
Intervention: type of prostaglandin (comparison 3)
Intervention: timing of prostaglandin (comparison 4)
Intervention: misoprostol oral versus vaginal (comparison 5)
Intervention: misoprostol buccal versus vaginal (comparison 6)
Intervention: misoprostol buccal versus oral (comparison 7)
Intervention: misoprostol sublingual versus vaginal (comparison 8)
Intervention: misoprostol sublingual versus oral (comparison 9)
Intervention:single versus split dose prostaglandin (comparison 10)
Intervention: single versus repeated prostaglandin (comparison 11)
Mifepristone alone versus combined regimen mifepristone/prostaglandin (comparison 12)
Prostaglandin alone versus a combined regimen (all) (comparison 13)

Single regimen:

Prostaglandin alone: route of administration (comparison 14)

Mifepristone single regimen ‐ high versus low dose (comparison 15)

Combined regimen methotrexate/prostaglandin:

Intervention: timing of prostaglandin (comparison 16)
Intervention: route of methotrexate: intramuscular versus oral (comparison 17)
Intervention: dose of methotrexate (comparison 18)
Intervention: route of prostaglandin (comparison 19)

Tamoxifen versus methotrexate (combined with prostaglandin):

Intervention: low dose tamoxifen (40 mg)(comparison 20)
Intervention: high dose tamoxifen (160 mg) (comparison 21)

Combined regimen mifepristone/prostaglandin versus mifepristone/prostaglandin plus tamoxifen (comparison 22)

Results

Description of studies

see table: Characteristics of included studies

Risk of bias in included studies

Thirty‐five trials scored adequate allocation concealment (A) and in 23 trials allocation concealment was unclear (B).Two trials used an open‐label design (Schaff 2000 MI200M800, Schaff 2001 M800MI200).   Two of the trials mentioned performing an 'intention ‐to ‐treat analysis' (WHO 2000 M400po, WHO 2001 GP1pv).

Effects of interventions

Fifty‐eight trials are included in this review. Due to the many different interventions, trials were grouped into comparisons, as listed below. The main outcome for which the meta‐analyses were performed was failure to achieve complete abortion with the method intended. Data on side‐effects could be combined for some comparisons. Major complications with any of the methods were rarely reported and if so, they are listed in the tables of Characteristics of included studies. Data are presented for different gestational ages where possible (≤ 49 days, > 49 days). One trial presented its data in two different publications (Honkanen 2004, von Hertzen 2003). One trial used 2 different comparisons, and is therefore listed as 2 different trials (Wiebe 1999 and Wiebe 1999 A). Six studies used different regimens/doses/timing of the drugs that could not be combined with any of the other regimens in the compraisons and are therefore listed separately inTable 23 (Wang 2000, Arvidsson 2005, Wiebe 2006, Liao 2004, WHO 1989, WHO 1991).

1. other studies included in the review.

Study	Intervention	Outcomes
Wang 2000	Group 1: Day 1: mifepristone 50mg/po 12hourly /2 doses),day 2‐7: mifepristone 25mg po/day Day 3: misoprostol 600mcg/po, day 4‐6: misoporstol 200mcg/day Group 2: Day 1: mifepristone 50mg, followed by 25mg/12hourly/4 times Day 3: misoprostol 600mcg/po	failure to achieve complete abortion: group 1: 18/1118 group 2: 59/494 ongoing pregnancy: group1:2/1118 group2: 6/494
Arvidsson 2005	Day 1: both groups receive mifepristone 600mg Day 3: group 1: misprostol 400mcg/po group 2: 800mcg/pv	nausea: group1: 23/48   group2: 17/49 vomiting: group1: 11/48   group2: 5/49 diarrhoea: group1: 3/48 group2: 1/49 women dissatisfied with procedure: group1: 2/48   group2: 1/49
Wiebe 2006	Group 1: methotrexate 50 mg/ m2 followed >/ 72 hours by 400mcg misoprostol vaginal Group 2: misoprostol 400mcg sublingual AND 400mcg misoprostol vaginal	failure to achieve complete abortion: group1: 62/149   group2: 57/149 nausea: group1: 53/49   group2: 54/149 vomiting: group1: 17/149   group2: 21/149 diarrhoea: group1: 16/149   group2: 41/149 surgical abortion: group1: 9/149   group2: 18/149
Liao 2004	Group 1: mifepristone given: 50 mg, then 12 hrs later 25 mg, then 12 hrs later 50 mg, and finally, 12 hrs later, 25 mg mifepristone. 24 hrs after, 600 mcg misoprostol oral (total: 150mg) Group 2: mifepristone given 30 mg, then 15 mg every 12 hours for 3 doses. 24 hrs after last dose, 600 mcg misoprostol given oral. (total: 75 mg)	failure to achieve complete abortion: group1: 11/240   group2: 9/240 ongoing pregnancy: group1: 2/240   group2: 1/240
WHO 1989	Group 1: mifepristone 25mg/twice daily for 3 days (total 150 mg) and sulprostone0.25 mg /intramuscular/ on third day a.m.   Group 2: mifepristone 25mg /twice daily for 4 days (total 200mg) and sulprostone0.25 mg /intramuscular/ on fourth day a.m.	failure to achieve complete abortion:   group1: 15/125   group2: 13/126 ongoing pregnancy: group1: 3/125   group2: 3/126
WHO 1991	Group 1: mifepristone 25mg/12 hourly/ 5 doses (total 125mg) and gemeprost 1mg/vaginally 60 hours after the start of the treatment   Group 2: mifepristone 600mg/single dose and gemeprost 1mg/vaginally 60 hours after the start of the treatment	failure to achieve complete abortion: group1: 12/181 group2: 15/187

Date	Event	Description
15 April 2008	Amended	Converted to new review format.
17 October 2003	New citation required and conclusions have changed	Substantive amendment

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 failure to achieve complete abortion	6		Risk Ratio (M‐H, Fixed, 95% CI)	Subtotals only
1.1 all	6	6841	Risk Ratio (M‐H, Fixed, 95% CI)	0.90 [0.77, 1.05]
1.2 600mg vs 200mg	4	3494	Risk Ratio (M‐H, Fixed, 95% CI)	1.07 [0.87, 1.32]
1.3 200mg vs 100 mg (>49 days)	1	1182	Risk Ratio (M‐H, Fixed, 95% CI)	0.89 [0.61, 1.29]
1.4 600mg vs 200mg with gemeprost 1mg/pv	2	1685	Risk Ratio (M‐H, Fixed, 95% CI)	1.02 [0.72, 1.45]
1.5 200mg vs 100 mg (</=49 days)	1	941	Risk Ratio (M‐H, Fixed, 95% CI)	0.79 [0.47, 1.33]
2 side effects	2		Risk Ratio (M‐H, Fixed, 95% CI)	Totals not selected
2.1 nausea	2		Risk Ratio (M‐H, Fixed, 95% CI)	0.0 [0.0, 0.0]
2.2 vomiting	1		Risk Ratio (M‐H, Fixed, 95% CI)	0.0 [0.0, 0.0]
2.3 diarrhoea	1		Risk Ratio (M‐H, Fixed, 95% CI)	0.0 [0.0, 0.0]
3 time until passing of conceptus > 3‐6 hours	2	1116	Risk Ratio (M‐H, Fixed, 95% CI)	0.89 [0.74, 1.07]
4 ongoing pregnancy	5		Risk Ratio (M‐H, Fixed, 95% CI)	Subtotals only
5 surgical evacuation	1		Risk Ratio (M‐H, Fixed, 95% CI)	Subtotals only

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 failure to achieve complete abortion	2	2814	Risk Ratio (M‐H, Fixed, 95% CI)	3.05 [2.24, 4.14]
2 side effects	2		Risk Ratio (M‐H, Fixed, 95% CI)	Subtotals only
2.1 nausea	2	1380	Risk Ratio (M‐H, Fixed, 95% CI)	1.13 [1.02, 1.25]
2.2 vomiting	2	1219	Risk Ratio (M‐H, Fixed, 95% CI)	0.83 [0.71, 0.98]
2.3 diarrhoea	2	1379	Risk Ratio (M‐H, Fixed, 95% CI)	1.80 [1.49, 2.18]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 failure to achieve complete abortion	1	429	Risk Ratio (M‐H, Fixed, 95% CI)	0.77 [0.36, 1.67]
2 side effects	1	1287	Risk Ratio (M‐H, Fixed, 95% CI)	1.21 [1.07, 1.36]
2.1 nausea	1	429	Risk Ratio (M‐H, Fixed, 95% CI)	1.12 [0.98, 1.29]
2.2 vomiting	1	429	Risk Ratio (M‐H, Fixed, 95% CI)	1.16 [0.89, 1.51]
2.3 diarrhoea	1	429	Risk Ratio (M‐H, Fixed, 95% CI)	1.51 [1.12, 2.03]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 failure to achieve complete abortion	1	224	Risk Ratio (M‐H, Fixed, 95% CI)	0.29 [0.06, 1.35]
2 surgical evacuation	1	327	Risk Ratio (M‐H, Fixed, 95% CI)	0.80 [0.18, 3.53]
3 ongoing pregnancy at follow‐up	1	224	Risk Ratio (M‐H, Fixed, 95% CI)	0.14 [0.01, 2.73]
4 side effects	2		Risk Ratio (M‐H, Fixed, 95% CI)	Totals not selected
4.1 nausea	2		Risk Ratio (M‐H, Fixed, 95% CI)	0.0 [0.0, 0.0]
4.2 vomiting	2		Risk Ratio (M‐H, Fixed, 95% CI)	0.0 [0.0, 0.0]
4.3 diarrhoea	2		Risk Ratio (M‐H, Fixed, 95% CI)	0.0 [0.0, 0.0]
5 women dissatisfied with the procedure	1	298	Risk Ratio (M‐H, Fixed, 95% CI)	2.81 [1.15, 6.87]
6 side effects	1	2490	Risk Ratio (M‐H, Fixed, 95% CI)	1.10 [1.02, 1.18]
6.1 nausea	1	830	Risk Ratio (M‐H, Fixed, 95% CI)	1.10 [1.01, 1.19]
6.2 vomiting	1	830	Risk Ratio (M‐H, Fixed, 95% CI)	1.09 [0.94, 1.27]
6.3 diarrhoea	1	830	Risk Ratio (M‐H, Fixed, 95% CI)	1.11 [0.94, 1.31]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 failure to achieve complete abortion	1	154	Risk Ratio (M‐H, Fixed, 95% CI)	0.70 [0.21, 2.39]
2 side effects	1		Risk Ratio (M‐H, Fixed, 95% CI)	Totals not selected
2.1 nausea	1		Risk Ratio (M‐H, Fixed, 95% CI)	0.0 [0.0, 0.0]
2.2 vomiting	1		Risk Ratio (M‐H, Fixed, 95% CI)	0.0 [0.0, 0.0]
2.3 diarrhoea	1		Risk Ratio (M‐H, Fixed, 95% CI)	0.0 [0.0, 0.0]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 failure to achieve complete abortion	1	2046	Risk Ratio (M‐H, Fixed, 95% CI)	1.16 [0.96, 1.40]
2 side effects	1		Risk Ratio (M‐H, Fixed, 95% CI)	Subtotals only
2.1 nausea	1	2066	Risk Ratio (M‐H, Fixed, 95% CI)	1.04 [0.90, 1.20]
2.2 vomiting	1	2066	Risk Ratio (M‐H, Fixed, 95% CI)	1.54 [1.14, 2.08]
2.3 diarrhoea	1	2066	Risk Ratio (M‐H, Fixed, 95% CI)	1.53 [1.33, 1.76]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 failure to achieve complete abortion	1	101	Risk Ratio (M‐H, Fixed, 95% CI)	1.32 [0.74, 2.38]
2 side effects	1		Risk Ratio (M‐H, Fixed, 95% CI)	Subtotals only
2.1 nausea	1	101	Risk Ratio (M‐H, Fixed, 95% CI)	0.40 [0.19, 0.84]
2.2 vomiting	1	101	Risk Ratio (M‐H, Fixed, 95% CI)	0.36 [0.07, 1.78]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 failure to achieve complete abortion	1	100	Risk Ratio (M‐H, Fixed, 95% CI)	2.04 [0.51, 8.07]
2 Side effects	1		Risk Ratio (M‐H, Fixed, 95% CI)	Subtotals only
2.1 nausea	1	100	Risk Ratio (M‐H, Fixed, 95% CI)	0.52 [0.22, 1.25]
2.2 vomiting	1	100	Risk Ratio (M‐H, Fixed, 95% CI)	4.89 [0.57, 42.21]
2.3 diarrhoea	1	100	Risk Ratio (M‐H, Fixed, 95% CI)	1.22 [0.18, 8.34]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 failure to achieve complete abortion	1	198	Risk Ratio (M‐H, Fixed, 95% CI)	2.04 [0.86, 4.84]
2 side effects	1		Risk Ratio (M‐H, Fixed, 95% CI)	Subtotals only
2.1 nausea	1	198	Risk Ratio (M‐H, Fixed, 95% CI)	0.56 [0.33, 0.97]
2.2 vomiting	1	198	Risk Ratio (M‐H, Fixed, 95% CI)	1.70 [0.42, 6.92]
2.3 diarrhoea	1	198	Risk Ratio (M‐H, Fixed, 95% CI)	1.53 [0.26, 8.96]

Methods	computer generated random numbers for the first 300 women, envelopes were shuffled in batches of 20 and numbered consecutively for the reminders   no blinding for clinical staff
Participants	800 pregnant women </= 63 days of amenorrhoea in Edinburgh/Scotland
Interventions	mifepristone 200mg (all) followed by:   group 1: gemeprost 0.5mg vaginal and 3 tabs placebo after 48 hours   group 2: misoprostol 600mcg oral and vaginal examination after 48 hours
Outcomes	complete, incomplete and missed abortion   ongoing pregnancy   side effects
Notes	power calculation (80% to detect 5% difference)   placebos were not identical to misoprostol   1 woman needed blood transfusion (group 2)
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment (selection bias)	Unclear risk	B ‐ Unclear

Methods	computer generated random numbers
Participants	999 pregnant women, < 63 days of gestation, confirmed by ultrasound if necessary, at the Royal Infirmary Hospital, Edinburgh   Inclusion criteria: aged =/> 16 years, available for follow‐up within 2 weeks   Exclusion criteria: ectopic pregnancy, active asthma, liver or renal disease, adrenal insufficiency, anaemia, haemolytic disease, treatment with anticoagulants, smoking > 20 cigarettes/day
Interventions	mifepristone 200mg (all) followed by:   group 1: gemeprost 0.5mg/vaginal   group 2: misoprostol 800mcg/vaginal
Outcomes	complete, incomplete abortion, ongoing pregnancy, duration of bleeding, side effects
Notes	single blinded   2 women required blood transfusions (1 in each group)
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment (selection bias)	Unclear risk	B ‐ Unclear

Methods	random allocation, not specified
Participants	153 women, ≤ 49 days of amenorrhoea, confirmed by positive pregnancy test and pelvic examination, Uppsala, Sweden
Interventions	group 1: mifepristone 10mg / twice daily for 7 days   group 2: mifepristone 25mg / twice daily for 7 days   group 3: mifepristone 50mg / twice daily for 7 days   (group 1 vs group 3)
Outcomes	complete, incomplete abortion   ongoing pregnancy   bleeding pattern   side effects
Notes	no mentioning of major complications
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment (selection bias)	Unclear risk	B ‐ Unclear

Methods	random numbers generated in SPSS; numbered opaque envelopes
Participants	>18 years old in good general health and willing   to return for follow‐up and living <1 hr from clinic; ≤ 56 days of gestation, exclusion: less than 18 years   old, suspected ectopic pregnancy. Study conducted in India and Vietnam
Interventions	misoprostol only: g1) 4X400mcg/3h/oral; g2) 2X800mcg/oral; g3) 1X600mcg/vaginal; g4) 2X800mcg/3h/oral; g5) 1X800mcg/vaginal
Outcomes	complete/incomplete abortion, ongoing pregnancy
Notes	Initially, women were randomised between the first three regimens. Subsequent review of their low efficacy resulted changing the regimens, and from that point on, women were randomized to treatment group 4 or 5.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment (selection bias)	Low risk	A ‐ Adequate

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 failure to achieve complete abortion	1	200	Risk Ratio (M‐H, Fixed, 95% CI)	1.96 [0.93, 4.15]
2 side effects	1		Risk Ratio (M‐H, Fixed, 95% CI)	Subtotals only
2.1 nausea	1	200	Risk Ratio (M‐H, Fixed, 95% CI)	0.78 [0.54, 1.10]
2.2 vomiting	1	200	Risk Ratio (M‐H, Fixed, 95% CI)	0.65 [0.28, 1.53]
2.3 diarrhoea	1	200	Risk Ratio (M‐H, Fixed, 95% CI)	1.23 [0.34, 4.43]

Methods	computer randomisation
Participants	100 women randomised; age and gestational ages averages not given; included gestational age up to 49 days confirmed by ultrasound; exclusion criteria: contraindications for medical abortion.Setting: Karolinska Hospital, Sweden
Interventions	mifepristone 600mg (all) followed 36‐48 hrs later by: group1) misoprostol 400mcg oral group 2) misoprostol 800mcg vaginal
Outcomes	experience of pain, occurrence of side‐effects, duration of bleeding
Notes	10 women could not be reached by phone 3‐7 weeks after abortion; 30 women did not agree or weren't asked to be called during this time frame

Methods	computer randomisation; sealed, opaque envelopes were numbered by a by a person unrelated to the study
Participants	300 pregnant women, ≤ 63 days of amenorrhoea confirmed by ultrasound   Exclusion criteria: previous use of vitamins/folates, white blood cell count <3000/uL, platelet count <100 000/uL, haemoglobin <10.0 mg/dL, aspartate aminotransferase >2 times normal or active liver disease, serum creatinine >1.5 mg/dL or active renal disease, inflammatory bowel disease, intolerance to the medication
Interventions	methotrexate 50mg/m2 intramuscular on recruitment day and misoprostol800 mcg vaginal (self administered) on:   group 1: day 3   group 2: day 4   group 3: day 5   additional 800mcg misoprostol in 48 hours interval (up to 4 doses)
Outcomes	complete, incomplete abortion (complete expulsion with additional doses of misoprostol), treatment failure, bleeding pattern, blood parameters, side effects
Notes	power calculation (85% power, significance level of 0.05)   no major complications occurred
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment (selection bias)	Low risk	A ‐ Adequate

Methods	double blind, randomisation generated centrally; sealed, opaque envelopes
Participants	151 women, ≤ 49 days of amenorrhoea confirmed by ultrasound at Shanghai Medical University without medical disorders, contraindication for the study medication or IUD in situ
Interventions	group 1:   day 1‐3: testosterone propionate 100mg/imi/day   day 4: PGE1 ester (ONO 802) 1mg/pv/6 hourly for a maximum of 4 doses   group2:   day 1‐3: placebo injections   day 4: PGE1 ester (ONO 802) 1mg/pv/6 hourly for a maximum of 4 doses
Outcomes	complete, incomplete abortion, ongoing pregnancy, blood transfusion, duration of bleeding
Notes	no major complications were reported
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment (selection bias)	Low risk	A ‐ Adequate

Methods	computer generated random sequence; consecutive numbered opaque envelopes
Participants	18 years or older; 300 women randomised; gestational age less than 8 weeks; study conducted between january 2004 ‐ june 2005; no contraindications to study medication   lived or worked within 1 hour of the study site, agreed to provide an address and telephone number and return for a follow‐up visit study site: India (Pune and Mumbai)
Interventions	mifepristone 200mg followed after 48 hours by: group 1) 400mcg oral misoprostol and placebo 3h later group 2) 400mg misoprostol repeated once 3 hours later
Outcomes	complete abortion, side effcets
Notes	ITT done
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment (selection bias)	Low risk	A ‐ Adequate

Methods	randomisation according to computer‐generated random number table numbered sealed, opaque envelopes
Participants	63 pregnant women, ≤ 56 days of amenorrhoea, confirmed by ultrasound, San Francisco General Hospital   Exclusion criteria:   Exclusion criteria: previous use of vitamins/folates, hematocrit ≤ 0.30, white blood cell count <3000/uL, platelet count <100 000/uL, haemoglobin <10.0 mg/dL, aspartate aminotransferase >2 times normal or active liver disease, serum creatinine >1.5 mg/dL or active renal disease, inflammatory bowel disease, asthma, intolerance to the medication
Interventions	group 1: methotrexate 50mg/m2 intramuscular and misoprostol 800mcg/vaginal after 3 days   group 2: misoprostol 800mcg/vaginal
Outcomes	complete abortion, duration of vaginal bleeding, side effects, change in beta‐HCG levels
Notes	power calculation (80% power, significance level of 0.05) based on 95% success with methotrexate and 75% success with misoprostol alone. The required sample size was 98.   no mentioning of major complications
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment (selection bias)	Low risk	A ‐ Adequate

Methods	randomisation according to random number tables   sealed, opaque envelopes were numbered by a by a person unrelated to the study   no blinding
Participants	20 pregnant women, ≤49 days, confirmed by ultrasound, Magee‐Women's Hospital, Pennsylvania, USA   Exclusion criteria: previous use of vitamins/folates, haemoglobin <10.0 mg/dL, aspartate aminotransferase >2 times normal or active liver disease, serum creatinine >1.5 mg/dL or active renal disease, inflammatory bowel disease, intolerance to the medication
Interventions	group 1: methotrexate 25mg/orally followed by misoprostol 800mcg/vaginal after 7 days   group 2: methotrexate 50mg/orally followed by misoprostol 800mcg/vaginal after 7 days
Outcomes	complete abortion, duration of vaginal bleeding, side effects, change in haemoglobin/aspartate transferase
Notes	no major complications occurred
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment (selection bias)	Low risk	A ‐ Adequate

Methods	random number tables in blocs of ten, sealed opaque envelopes prepared by person not involved in the trial
Participants	80 pregnant women, ≤ 49 days pregnant, single pregnancy, confirmed by ultrasound, at the University hospital Pittsburgh, USA;   exclusion criteria: contraindication to mifepristone/misoprostol administration, haemoglobin < 10 gm/dL, cardiovascular disease, coagulopathies, IUCD in situ, breast feeding
Interventions	mifepristone 100mg (all)   after 2 days, home administration:   group 1: misoprostol 400mcg oral   group 2: misoprostol 800mcg vaginal
Outcomes	complete abortion, onset of bleeding &cramping, duration of bleeding, side effects
Notes	power calculation power calculation (80% power, significance level of 0.05)   no major complications were reported
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment (selection bias)	Low risk	A ‐ Adequate

Methods	random number tables, sealed opaque envelopes
Participants	86 pregnant women, =/> 18 years, ≤ 49 days pregnant, single pregnancy, at the University hospital Pittsburgh, USA   exclusion criteria: contraindication to mifepristone/misoprostol administration, haemoglobin < 10 gm/dL, cardiovascular disease, coagulopathies, IUCD in situ, breastfeeding
Interventions	mifepristone 600mg (all)   group 1: misoprostol 400mcg after 6‐8 hours/oral   group 2: misoprostol 400mcg after 48 hours/oral
Outcomes	complete abortion, onset and duration of bleeding, side effects
Notes	no blinding   no major complications were reported
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment (selection bias)	Low risk	A ‐ Adequate

Methods	computer generated randomisation, permuted blocs, stratified by centre; sequentially numbered opaque envelopes
Participants	26 years old; 1080 women randomized no more than 63 days gestation confirmed by ultrasound; average gestational age of 51 days; willing to have surgical procedure and had a telephone; conducted in 2002‐2003 in USA MAgee‐Women's Hospital in Pittsburgh, Pennsylvania, columbia University, NY, Boston University,Massachusetts University of Rochester, NY
Interventions	mifepristone 200mg followed by misoprostol 800mcg vaginal: group1: administered 6‐8 hours after mifepristone group 2: administered 23‐25 hours after mifepristone
Outcomes	Complete abortion, side‐effects, bleeding, acceptability
Notes
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment (selection bias)	Low risk	A ‐ Adequate

PERMALINK

Medical methods for first trimester abortion

Regina Kulier

Nathalie Kapp

A Metin Gülmezoglu

G Justus Hofmeyr

Linan Cheng

Aldo Campana

Abstract

Background

Objectives

Search methods

Selection criteria

Data collection and analysis

Main results

Authors' conclusions

Plain language summary

Background

Objectives

Methods

Criteria for considering studies for this review

Types of studies

Types of participants

Types of interventions

Types of outcome measures

Search methods for identification of studies

Data collection and analysis

Results

Description of studies

Risk of bias in included studies

Effects of interventions

1. other studies included in the review.

1.

1.1. Analysis.

2.1. Analysis.

3.1. Analysis.

4.1. Analysis.

5.1. Analysis.

6.1. Analysis.

7.1. Analysis.

8.1. Analysis.

9.1. Analysis.

10.1. Analysis.

11.1. Analysis.

12.1. Analysis.

13.1. Analysis.

14.1. Analysis.

15.1. Analysis.

16.1. Analysis.

17.1. Analysis.

18.1. Analysis.

19.1. Analysis.

20.1. Analysis.

21.1. Analysis.

22.1. Analysis.

Discussion

Authors' conclusions

Implications for practice.

Implications for research.

What's new

History

Acknowledgements

Data and analyses

Comparison 1. combined regimen mifepristone/prostaglandin: dose of mifepristone.

1.2. Analysis.

1.3. Analysis.

1.4. Analysis.

1.5. Analysis.

Comparison 2. combined regimen mifepristone/prostaglandin: dose of prostaglandin.

2.2. Analysis.

2.3. Analysis.

2.4. Analysis.

2.5. Analysis.

Comparison 3. combined regimen mifepristone/prostaglandin: type of prostaglandin.

3.2. Analysis.

3.3. Analysis.

3.4. Analysis.

Comparison 4. combined regimen mifepristone/prostaglandin: time of prostaglandin.

4.2. Analysis.

4.3. Analysis.

Methods	computer generated random numbers, randomisation centrally, permuted block design with varying block sizes; randomisation after taking mifepristone ; no blinding; opaque envelopes
Participants	1128 women enrolled;mean age 27 years, women with no more than 63 days gestation (mean gestational age 51‐52 days gestation) and willing to follow‐up and with access to a telephone. Exclusion criteria: contraindications to mifepristone or misoprostol, Hbg< 10, IUD in place, on anticoagulants or with coagulopathy, active cervicits or currently breastfeeding. Gestational age confirmed by US. 4 academic centers in the USA; University of Pittsburgh, Oregon Health and Science University, Northwestern University, University of Southern California
Interventions	mifepristone 200mg followed by: group 1: within 15 minutes, 800mcg misoprostol vaginal group 2: 23‐25 hours later, 800mcg misoprostol vaginal
Outcomes	complete abortion; side‐effect; bleeding; acceptibility
Notes
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment (selection bias)	Low risk	A ‐ Adequate

Methods	sealed, opaque envelopes   random assignment before misoprostol administration
Participants	150 pregnant women </= 56 days of amenorrhoea, confirmed by ultrasound
Interventions	group 1: mifepristone 200mg and misoprostol 800mcg/oral after 48 hours   group 2: mifepristone 200mg and misoprostol 400mcg after 48 hours plus 400mcg 2 hours later/oral
Outcomes	changes in blood pressure, pulse rate and temperature   complete and incomplete abortion   ongoing pregnancy   side effects   bleeding pattern
Notes	power calculation (5% significance level to detect a 20% reduction in incidence of side effects)   no mentioning of major complications
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment (selection bias)	Low risk	A ‐ Adequate

Methods	computer generated random assignment before misoprostol administration, sealed opaque envelopes
Participants	270 women ≤ 63 days of amenorrhoea, confirmed by ultrasound   Exclusion criteria: contraindication for the use of mifepristone and/or misoprostol
Interventions	group 1: mifepristone 600mg and misoprostol 800mcg/orally after 48 hours   group 2: mifepristone 600mg and misoprostol 800mcg/vaginally (self‐administration) after 48 hours
Outcomes	complete, incomplete and missed abortion   ongoing pregnancy   expulsion within 4 hours   expulsion without need for surgery   side effects
Notes	power calculation (5% significance level to detect difference of 10% in the incidence of women aborting within 4 hours vaginal misoprostol by self administration   1 woman received a blood transfusion (group 2)
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment (selection bias)	Low risk	A ‐ Adequate

Methods	computer generated fixed blocks of 20; 1:1 randomisation; sealed opaque envelopes
Participants	450 women aged 24‐26 years; no more than 63 days gestation confirmed by US (average 51 days of gestation); exclusion criteria: contraindications for study medication, breastfeeding, Hbg<10, coagulopathy or treatment with anticoagulants, IUD in situ, presence of cardiovascular disease, ectopic pregnancy; study conducted between September 2003 ‐ March 2005
Interventions	mifepristone 200mg followed by: group 1: 800mcg vaginal misoprostol after 6 hours group 2: 800mcg of vaginal misoprostol after 36‐48 hrs later
Outcomes	complete abortion, side effcets, acceptability
Notes	ITT analysis
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment (selection bias)	Low risk	A ‐ Adequate

Methods	random number tables; sealed opaque envelopes
Participants	340 women, average age 24 years randomised; average gestational age 65‐68 days. Exclusion criteria: <16 years, severe asthma,haemorrhagic disorders and treatment with anticoagulants, known allergy to prostaglandins, history of cardiac disease, smoking over the age of 35 years with ECG abnormalities, breastfeeding. study conducted at Aberdeen Royal Infermary, UK, from July 2002 ‐ October 2003
Interventions	mifepristone 200 mg, followed 36‐48 hours after by: group 1: 600mcg sublingual misoprostol, followed 3 hours later by 400mcg (if 9‐13 weeks gestation, a third dose of 400mcg was administered) group 2: 800mcg vaginal misoprostol, followed 3 hours later by 400mcg (if 9‐13 weeks gestation, a third dose of 400mcg was administered)
Outcomes	complete/incomplete abortion, missed abortion, continuing pregnancy
Notes	No ITT; LTFU identical (13) in each group (total 26)
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment (selection bias)	Low risk	A ‐ adequate

Methods	computer generated random sequence; packing company prepared bags containing the medication according to the randomisation sequence
Participants	2219 women; mean age 27 years; </= 63 days of amenorrhoea; Inclusion criteria:single intrauterine pregnancies, haemoglobin > 100 g/L . Exclusion criteria: medical contraindications or allergy for either mifepristone or misoprostol; past or present thromboembolism; liver disease, pruritus of pregnancy; previous surgery of uterine cervix; presence of an intrauterine device; suspected or proven ectopic pregnancy; smoking > 10 cigarettes/day; risk factor for cardiovascular disease; breastfeeding; Study was conducted from October 1998 ‐ Decembre 2000 in 15 cities in 11 countries, including developed and developing countries: Beijing, Hong Kong and Shanghai ‐ China; Chandigarh, Mumbai and New Delhi ‐ India; Helsinki ‐ Finland;   Ho Chi Minh City ‐ Viet Nam; Ljubljana ‐Slovenia; Oslo ‐ Norway; Singapore ‐ Singapore; Stockholm ‐ Sweden;   Szeged ‐ Hungary; Targu Mures ‐ Romania; and Ulaanbaatar ‐ Mongolia.
Interventions	mifepristone 200mg followed 36‐48 hours later by: group 1: misoprostol 800mcg orally followed by misoprostol 400mcg twice/day for 6 days oral group 2: misoprostol 800mcg vaginally followed by misoprostol 400mcg twice/day for 6 days oral group 3: misoprostol 800mcg vaginally followed by placebo tablets twice/day for 6 days oral
Outcomes	side‐effects and acceptability
Notes
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment (selection bias)	Low risk	A ‐ adequate

Methods	computer generated random table, opaque vials
Participants	250 healthy women, </=56 days of amenorrhoea,   confirmed by ultrasound,   Exclusion criteria: evidence of threatened spontaneous abortion, uterine infection, anaemia, bleeding disorders, cardiovascular or cerebrovascular disease, uterine leiomyomata, allergy against the study medication.
Interventions	group 1: mifepristone 200mg, misoprostol 800mcg/pv on day 3, repeated on day 4 if gestational sac present   group 2: Placebo, misoprostol 800mcg/pv on day 3, repeated on day 4 if gestational sac present
Outcomes	successful abortion, side effects
Notes	Placebos were vitamin C tablets (not identical); opaque vials   were used to blind the investigator   power calculation (5% significance level to detect a 5% difference in success rates between the 2 study groups)   no major complications were reported
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment (selection bias)	Low risk	A ‐ Adequate

Methods	randomisation by using random number tables
Participants	150 women pregnant ≤ 56 days confirmed by ultrasound   exclusion criteria: cervical dilatation, anaemia, pelvic inflammatory disease, uterine bleeding, uterine leiomyomata, serious medical problems, allergy or contraindications to the study medication
Interventions	group 1:   tamoxifen 20mg/twice daily and misoprostol 800mcg/pv after 48 hours   group 2:   placebo twice daily and misoprostol 800mcg/pv after 48 hours
Outcomes	complete/incomplete abortion, ongoing pregnancy, complications, side effects
Notes	treatment and placebo were placed in identical capsules   no major complications were reported
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment (selection bias)	Low risk	A‐ Adequate

Methods	randomisation into 3 groups   randomisation procedure not stated
Participants	58 women, pregnant ≤ 10 weeks, confirmed by ultrasound, University Hospital Los Angeles, USA   Exclusion criteria: uterine infection, prior uterine bleeding, cervical dilatation, anaemia, cardiovascular or cerebral disease, allergy to misoprostol
Interventions	group A: misoprostol 100mcg/vaginally/ 8 hourly to a maximum of 6 doses   group B: misoprostol 100mcg/vaginally/ 8 hourly to a maximum of 6 doses and tamoxifen 10mg/orally after the first dose of misoprostol   group C: misoprostol 100mcg/vaginally/ 8 hourly to a maximum of 6 doses and laminaria/intracervical immediately before the first dose of misoprostol   the dose of misoprostol was increased after the success rate was unsatisfactory after the first 26 women
Outcomes	complete abortion, failure rate, side effects, mean number of doses of misoprostol used, time until passing of conceptus
Notes	no mentioning of major complications
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment (selection bias)	Unclear risk	B ‐ Unclear

Methods	computer random table; use of identical appearing packages and capsules/ tablets from pharmacy; identical placebo tablets
Participants	480 women,average age 26 years; </= 49 days gestation confirmed by ultrsound; Exclusion criteria: abnormal menses, IUD in situ, contraindications for use of study medication; study conducted between November 2001 to June 2002 in 3 hospitals affiliated to University of Beijing, China
Interventions	group 1: mifepristone: 50mg, then 12 hrs later 25mg, then 12 hrs later 50mg, and finally, 12 hrs later, 25mg (total: 150mg). 24 hrs after last dose 600mcg misoprostol orally group 2: mifepristone 30mg, then 15mg every 12 hours for 3 doses (total: 75mg). 24 hrs after last dose, 600mcg misoprostol orally.
Outcomes	complete abortion
Notes
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment (selection bias)	Low risk	A ‐ adequate