Doi:10.1016/j.ijgo.2005.10.00

International Journal of Gynecology and Obstetrics (2006) 92, 10 — 18 Misoprostol in preventing postpartum hemorrhage:A meta-analysis Epidemiology and Biostatistics, School of Public Health, University of California at Berkeley,Berkeley, CA, USA Received 25 May 2005; received in revised form 3 October 2005; accepted 4 October 2005 Objective: To assess misoprostol’s ability to prevent postpartum hemorrhage (PPH) where no alternatives exist. Comparison to oxytocics demonstrates how similarly misoprostol achieves a level of effectiveness—obtainable only in hospitals—in remote locations around the world. Method: Using the Mantel— Haenszel fixed-effects model and the DerSirmonian and Laird random-effects model, summary statistics indicated that misoprostol’s excess risk of PPH was only4% when compared to oxytocics. Result: This risk difference was well within therange of expected results for all uterotonic agents and does not warrant brandingmisoprostol as an inferior drug. Conclusion: Conventional uterotonic drugs shouldnot be used to set the lowest-accepted level of effectiveness in settings wherethey are entirely unsuitable. Continuing to weigh the benefits of one effectivedrug against another only delays the distribution of misoprostol in countries whereit is the only feasible choice and must be measured against no treatment at all.
D 2005 International Federation of Gynecology and Obstetrics. Published by ElsevierIreland Ltd. All rights reserved.
quarter of these deaths . Most of these deathsoccur in the resource-poor countries of Africa and The World Health Organization (WHO) estimated that Asia, particularly in rural areas. Not surprisingly, 529,000 women died from obstetric causes in 2000 records of maternal mortality are poor or non- . Postpartum hemorrhage (PPH), which afflicts existent where it most often occurs, which implies approximately 14 million women annually, caused a that even these sobering estimates are greatlyunderestimated. Indeed, data in remote regions areso scarce and the methods of collection so varied thatthe WHO warns against the formulation of confidence E-mail address: [email protected].
intervals around the available estimates.
0020-7292/$ - see front matter D 2005 International Federation of Gynecology and Obstetrics. Published by Elsevier Ireland Ltd.
All rights reserved.
doi:10.1016/j.ijgo.2005.10.001 Misoprostol in preventing postpartum hemorrhage: A meta-analysis PPH is defined as blood loss of 500 mL or more developing countries, where it is most needed.
within 24 h of delivery, but, this quantity is Misoprostol’s value as a prophylactic uterotonic extremely difficult to identify outside of a con- drug lies in its ability to prevent PPH and reduce trolled trial setting. Even trained physicians are maternal mortality where no alternatives exist.
reported to typically underestimate blood loss by Comparison to oxytocics should serve only to about half While there are a few known risk demonstrate how similarly misoprostol achieves a factors, PPH occurrence is random, making it level of effectiveness—obtainable only in hospi- impossible to predict in both low and high risk tals—in remote locations around the world. To populations. Furthermore, blood loss can be rapid.
date, this is the largest meta-analysis ever con- In developing countries, where nearly half the ducted on the efficacy of misoprostol for the women deliver without the aid of a skilled birth attendant , there is simply not enough time toseek treatment for PPH, and in most cases none isto be had. The only way to help women without access to trained attendants is through preventa-tive measures.
A literature search was conducted for all random- The most successful method for reducing PPH, ized control trials (RCT) which tested misoprostol’s Active Management of the Third Stage of Labor efficacy in preventing PPH. The electronic data- (AMTSL), requires prophylactic uterotonic drugs base PubMed (National Library of Medicine, which are unsuitable for use in the remote loca- Bethesda, MD) was searched for published articles, tions where prevention is most needed. Nonethe- along with the Cochrane CENTRAL database and the less, this nearly universal method has set the Population Council’s bibliographic website precedent for a standard of care unavailable in developing countries. The uterotonic drugs used in ing terms used were: misoprostol and postpartum AMTSL trials include oxytocics: oxytocin (Syn- hemorrhage, coupled with: prevention and active tocinonR, Alliance Pharmaceuticals, Chippenham, management. References from published articles UK or PitocinR, King Pharmaceuticals, Bristol, TN), were pursued and primary authors contacted in ergometrine malate (MethergineR, Novartis Phar- order to uncover any unpublished RCTs. The search maceuticals, East Hanover, NJ) and combinations of was conducted irrespective of language of publica- the two (SyntometrineR, Alliance Pharmaceuticals, tion or geographic region. All studies matching the Chippenham, UK) all of which must be administered inclusion criteria and published before May 2005 by injection, which not only requires a sterile needle, syringe and accurate dosing, but someoneto administer it. In addition, oxytocics are light-sensitive and require refrigeration to remain phar-macologically active, which limits their use to areas with refrigeration and reliable sources of energyand increases their cost.
All RCTs which assessed misoprostol efficacy in Misoprostol (CytotecR, Pfizer, New York, NY), a preventing PPH during third trimester vaginal births prostaglandin E1 analog, registered for the pre- were reviewed for inclusion in this analysis. All vention and treatment of gastric ulcers, is well- studies, irrespective of dose, route of administra- known for its off-label use as a uterotonic agent. It tion (with the exception of vaginal administration is inexpensive (one 200 g tablet is approximately due to its infeasibility after a vaginal birth) or type US$1 comes in tablets which can be admin- of control substance, were included. Three out- istered orally, rectally or sublingually, and does not comes were selected before analysis began: blood require refrigeration, dark storage or administra- loss z500 mL, blood loss z1000 mL and the need tion by an attendant. However, many studies have for additional uterotonic agents. Because the side found it to be slightly less effective than oxytocics effects of all uterotonic drugs have been well in controlled clinical settings. This circumstance has had the result of branding misoprostol as an ison to the life threatening alternative of PPH, side inferior drug , despite repeated praise for effects were not considered a relevant outcome for the feasibility of its use in resource-poor settings A total of 31 relevant studies were identified, The objective of this review is to analyze all but only 22 were selected for inclusion in the meta- existing trial data in order to reframe the current analysis. Six studies were excluded due to miso- debate surrounding the use of misoprostol in prostol being administered vaginally or after .5 mg or Oxytocin 10 units orErgometrine 500 mg a New studies since last meta-analysis.
Misoprostol in preventing postpartum hemorrhage: A meta-analysis All doses and routes of administration were pooled for three reasons. First, patients within studies were already included in other trials individual studies did not always receive the same and could not be used independently. A final trial was eliminated because it reported a pilot study it impossible to distinguish between methods.
Although the pooling of patients given oxytocics The studies which were included consisted of of varying administrations has been heavily criti- five placebo-controlled trials and 26 drug cized, especially in the WHO multi-center trial where intravenous and intramuscular administra- but in two instances , where three different heterogeneity only strengthened this meta-analy- oxytocics were tested, two arms were excluded sis. As oral, rectal and sublingual routes are known from the meta-analysis since they could not be to have slower up-take than intravenous or intra- pooled and analyzed against the misoprostol arm muscular injections, testing the three slowest simultaneously. For these two trials, the oxytocin methods against the two fastest biased the results arm was selected for the analysis since it is the against finding a similar relative risk. Therefore, most common prophylactic uterotonic drug. Seven the difference between misoprostol and oxytocics found in this analysis was actually greater than if published subsequent to the only other meta- misoprostol had only been administered by its analysis on the efficacy of misoprostol, published fastest method (sublingual and oxytocics by their slowest (intramuscularly). Second, numerousstudies on the efficacy of different oxytocics foundno significantly statistical difference in blood lossbetween them And third, comparison of misoprostol to each of these individual drugs hasalready been well documented ; given these Data were extracted from each study by the author, caveats, the purpose of this analysis was to who was not blinded. A validity analysis was compare misoprostol to the de facto standard of conducted to assess the methodological aspects of care (i.e. the collective efficacy of any drug each trial. Studies were scored from 0 to 10, using a approved for PPH prevention), rather than any For each outcome, a pooled risk ratio (RR) was 1) A research objective appropriate for this calculated comparing misoprostol to oxytocics or placebo. The Mantel—Haenszel fixed-effects model 2) Explicit inclusion and exclusion criteria was used instead of the inverse-variance method 3) Exclusion of patients with labor augmentation due to sparse outcome data. The test for hetero- geneity was based on weights provided by the inverse-variance method. When heterogeneity was 5) Reported baseline similarity between groups detected, the DerSirmonian and Laird random- effects model was used. Sensitivity analyses were conducted to investigate the influence of individual studies on the summary statistic by omitting each 9) No losses to follow-up before each outcome study in turn. Egger’s weighted regression and Begg’s rank correlation, where odds ratios were 10) Criteria for administration of additional ute- plotted against study size, were used for the detection of publication bias for each outcome.
Analysis of the data was performed using STATA 8.0 statistical software package (STATA Corpora- tion, College Station, TX). The raw data publishedin each article were compiled in two-by-two tables.
A total of 30,017 participants were included in the For only one study was it necessary to calculate cell 22 studies in the analysis, approximately half of counts from the reported percentages How- whom received misoprostol, with the remainder ever, not every study reported data for each specified outcome. When no data were provided, However, in several instances when the studies the study was dropped from the model.
reported a loss to follow-up for a specific outcome a One included study only reported outcome measurements for 500 mL.
misoprostol to placebo (n = 1706) produced a RR of the available data, not the original sample size, to 0.69 (95% CI: 0.53, 0.90). Thus, at this level avoid the inherent presumption that those lost to misoprostol demonstrates clear and statistically follow-up were not cases. Due to the infrequent significant reduction in the need for therapeutic incidence of PPH, little difference was found between the odds ratios (OR) and RR for any The 15 studies (n = 26,870) comparing misopros- tol and oxytocics for blood loss N500 mL produced a Whereas five studies compared misoprostol and RR of 1.4, which represents an excess risk (or risk placebo use, only two reported blood loss N500 mL; difference) of 5% greater incidence of blood loss thus, no analysis was conducted for this outcome level. Pooling the four studies reporting blood loss studies, a comparison of the fixed effects estimate N 1000 mL and one which only reported blood loss of and the random effects estimate showed little N 500 mL (n = 2112), misoprostol’s risk of difference between the two models. Although it PPH incidence over the risk of incidence with initially appeared that the largest trial dom- placebo was 0.85 (RR). Although misoprostol inated the pooled estimates, removing this study appeared to decrease the risk of PPH, this finding did not significantly alter the results (RR: 1.39; 95% was not statistically significant (95% CI: 0.63, 1.14).
CI: 1.219, 1.588). Further sensitivity analyses One included study specifically stated that its demonstrated that the heterogeneity present in objective was to measure side effects and was these models was the effect of the 15 varied not intended to be an efficacy trial due to its lack outcomes, not the effect of any single study.
of statistical power , which could help explain The RR of 1.36 for the 11 studies reporting blood this lack of significance. Of interest, including that loss N1000 mL for misoprostol versus oxytocics study with the three other trials reporting a need (n = 25,448) only demonstrated a 1% excess risk of for additional uterotonic agents when comparing All studies evaluating misoprostol vs. placebo with outcome blood loss N1000 mL. Mantel—Haenszel fixed- effects model. Heterogeneity chi-squared = 4.09 (df = 4), p = 0.394. I-squared (variation in RR attributable toheterogeneity) = 2.1%. Test of RR = 1: z = 1.08, p = 0.280. **Estimate reported for 500 mL or greater.
Misoprostol in preventing postpartum hemorrhage: A meta-analysis All studies evaluating misoprostol vs. placebo with outcome of additional uterotonic agents. Mantel— Haenszel fixed-effects model. Heterogeneity chi-squared = 1.62 (df = 3), p = 0.654. I-squared (variation in RRattributable to heterogeneity) = 0.0%. Test of RR = 1: z = 2.71, p = 0.007.
from this model produced a statistically insignifi- many milliliters lost warranted further interven- cant summary estimate (RR: 1.25; 95% CI: 0.94, tion. Because rectal administration of misoprostol 0.167), however this outcome was found to be requires a longer time to reach peak concentration consistent with the other studies’ findings and was levels than oral misoprostol it is possible that many studies intervened with additional uterotonic The random effects model of the 17 studies agents before there was adequate time for the comparing the need for additional uterotonic drugs misoprostol to take effect. A sub-analysis of oral in patients who received prophylactic misoprostol and sublingual misoprostol revealed no statistical to patients receiving oxytocics (n = 27,566) pro- difference between misoprostol and oxytocics (RR: duced a summary RR of 1.23 ). The hetero- 1.13, 95% CI: 0.81, 1.56). This observation strongly geneity present in this model can, in large part, be suggests that the rectal misoprostol had, in fact, explained by the subjective point at which addi- not yet peaked when additional uterotonic agents tional uterotonic drugs were administered. Not were administered. Therefore, inclusion of rectally only did many studies estimate (as opposed to administered misoprostol only masked the equiva- measure) blood loss, but not one specified how lency of oral and sublingual misoprostol to oxy- All studies evaluating misoprostol vs. oxytocics with outcome blood loss N500 mL 95% CI = 95% Confidence intervals.
M—H pooled RR = Mantel—Haenszel pooled risk ratios.
All studies evaluating misoprostol vs. oxytocics with outcome blood loss N1000 mL. Mantel—Haenszel fixed- effects model. Heterogeneity chi-squared = 3.64 (df = 10), p = 0.962. I-squared (variation in RR attributable toheterogeneity) = 0.0%. Test of RR = 1: z = 4.41, p = 0.000.
tocics. Despite many misoprostol patients receiving occurs in 5—18% of live births many trial doses which had not fully taken effect, only 4% findings were not statistically significant. The wide more received therapeutic drugs than patients in confidence intervals reflected the high probability of chance in each of these trials. Therefore, the RRof each trial were not definitive estimates, as itwas impossible to distinguish between individual RR with overlapping confidence intervals. However,the fact that the RR were consistent with individual Due to the small sample sizes and PPH’s high studies’ findings, demonstrated that the overall variability in general (even with oxytocics PPH risk of PPH when misoprostol was used was only 4% All studies evaluating misoprostol vs. oxytocics with outcome of additional uterotonic agents. DerSirmonian and Laird random-effects model. Heterogeneity chi-squared = 343.00 (df = 16), p = 0.000. I-squared (variation in RRattributable to heterogeneity) = 95.3%. Estimate of between-study variance Tau-squared = 0.2171. Test of RR = 1:z = 1.47, p = 0.141.
Misoprostol in preventing postpartum hemorrhage: A meta-analysis greater than when oxytocics were used. This risk [5] Roman AS, Andrei R. Seven ways to control postpartum difference was well within the range of expected hemorrhage. Contemp OB/GYN 2003;3:34 – 53.
[6] Chong YS, Su LL, Arulkumaran S. Current strategies for the results for any uterotonic agent and was not prevention of postpartum haemorrhage in the third stage of unique to misoprostol. Not surprisingly, the meta- labour. Curr Opin Obstet Gynecol 2004;16(2):143 – 50.
analysis published in 2003 had similar findings.
[7] Villar J, Gulmezoglu AM, Hofmeyr GJ, Forna F. Systematic In that review, Joy et al. reported OR for misopros- review of randomized controlled trials of misoprostol to tol versus oxytocin of 1.51 at blood loss prevent postpartum hemorrhage. Obstet Gynecol 2002; and 2.14 for additional oxytocics needed. Stated [8] Mousa HA, Walkinshaw S. Major postpartum haemorrhage.
another way, these OR only represented an excess Curr Opin Obstet Gynecol 2001;13(6):595 – 603.
risk of 5.8% and 4.5%, respectively. Although the [9] Darney PD. Misoprostol: a boon to safe motherhood or not? report stated that misoprostol was binferiorQ to other uterotonic drugs, its value in developing [10] El-Refaey H, Rodeck C. Post-partum haemorrhage: defini- tions, medical and surgical management. A time for countries was not dismissed due to its efficacy over change. Br Med Bull 2003;67:205 – 17.
[11] Joy SD, Sanchez-Ramos L, Kaunitz AM. Misoprostol use The primary limitation of this meta-analysis was during the third stage of labor. Int J Gynaecol Obstet the inability to conduct sub-analyses due to the small number of studies using each dose, adminis- [12] Miller S, Lester F, Hensleigh P. Prevention and treatment of postpartum hemorrhage: new advances for low-resource tration and uterotonic drug. Potential systematic settings. J Midwifery Women’s Health 2004;49(4):283 – 92.
biases, mentioned previously, were the inconsisten- [13] Potts M, Campbell M. Three meetings and fewer funerals— cy in blood loss measurement and lack of double- misoprostol in postpartum haemorrhage. Lancet 2004; masking. Because all outcomes were dependent on the amount of blood lost, it was critical to have [14] McCormick ML, Sanghvi HC, Kinzie B, McIntosh N. Prevent- ing postpartum hemorrhage in low-resource settings. Int J uniform measurement across, as well as within, Gynaecol Obstet 2002;77(3):267 – 75.
studies. Furthermore, because determining the [15] Hofmeyr GJ, Walraven G, Gulmezoglu AM, Maholwana B, severity of blood loss requires subjective judgment, Alfirevic Z, Villar J. Misoprostol to treat postpartum having all trial investigators and attending physi- haemorrhage: a systematic review. BJOG 2005;112(5): cians masked to drug allocation was exceedingly [16] Diaz QR, Arechiga EM, Resendiz BV, Lopez CO, Ayup HS, important. For these reasons, both double-masking Cuervo MJ. Vaginal misoprostol in the prevention of post- and blood measurement were criteria for the partum hemorrhage. Ginecol Obstet Mex 2002;70:572 – 5.
validity test. The tests for publication bias showed [17] Acharya G, Al-Sammarai MT, Patel N, Al-Habib A, Kiserud T.
no strong evidence of bias for any outcome level.
A randomized, controlled trial comparing effect of oral Nonetheless, there was a surprising lack of placebo misoprostol and intravenous syntocinon on intra-operativeblood loss during cesarean section. Acta Obstet Gynecol trials, despite nearly every study concluding with a request for more trials of this nature.
[18] Lokugamage AU, Paine M, Bassaw-Balroop K, Sullivan KR, Conventional uterotonic drugs, limited in their Refaey HE, Rodeck CH. Active management of the third use in remote areas, should not be used to set the stage at caesarean section: a randomised controlled trial of lowest-accepted standard for situations where they misoprostol versus syntocinon. Aust N Z J Obstet Gynaecol2001;41(4):411 – 4.
are entirely unsuitable. Continuing to weigh the [19] Chong YS, Chua S, El-Refaey H, Choo WL, Chanrachakul B, benefits of one effective drug against another, Tai BC, et al. Postpartum intrauterine pressure studies of when only misoprostol is currently feasible in the uterotonic effect of oral misoprostol and intramuscular developing countries, only delays its necessary syntometrine. BJOG 2001;108(1):41 – 7.
[20] Diab KM, Ramy AR, Yehia MA. The use of rectal misoprostol as active pharmacological management of the third stage oflabor. J Obstet Gynaecol Res 1999;25(5):327 – 32.
[21] El-Refaey H, O’Brien P, Morafa W, Walder J, Rodeck C. Use of oral misoprostol in the prevention of postpartumhaemorrhage. Br J Obstet Gynaecol 1997;104(3):336 – 9.
[1] Maternal Mortality in 2000: estimates developed by WHO, [22] Lumbiganon P, Villar J, Piaggio G, Gulmezoglu AM, Adetoro L, UNICEF and UNFPA. Geneva7 The World Health Organiza- Carroli G. Side effects of oral misoprostol during the first 24 hours after administration in the third stage of labour. BJOG [2] Mother—baby package: implementing safe motherhood in countries, vol. 94. Geneva7 World Health Organization, [23] Wright JM, Newton W. Is rectal misoprostol as effective as oxytocin in preventing postpartum hemorrhage? J Fam [3] Tsu VD, Langer A, Aldrich T. Postpartum hemorrhage in developing countries: is the public health community using [24] Lumbiganon P, Hofmeyr J, Gulmezoglu AM, Pinol A, Villar J.
the right tools? Int J Gynaecol Obstet 2004;85(Suppl 1): Misoprostol dose-related shivering and pyrexia in the third stage of labour. WHO Collaborative Trial of Misoprostol in [4] Coverage of Maternity Care: a listing of available informa- the Management of the Third Stage of Labour. Br J Obstet tion. Fourth edn. Geneva7 World Health Organization; 1997.
[25] Bamigboye AA, Hofmeyr GJ, Merrell DA. Rectal misopros- rhage after vaginal delivery. Am J Obstet Gynecol tol in the prevention of postpartum hemorrhage: a placebo-controlled trial. Am J Obstet Gynecol 1998; [38] Gulmezoglu AM, Villar J, Ngoc NT, Piaggio G, Carroli G, Adetoro L, et al. WHO multicentre randomised trial of [26] Benchimol M, Gondry J, Mention JE, Gagneur O, Boulanger misoprostol in the management of the third stage of labour.
JC. Role of misoprostol in the delivery outcome. J Gynecol Obstet Biol Reprod (Paris) 2001;30(6):576 – 83.
[39] Karkanis SG, Caloia D, Salenieks ME, Kingdom J, Walker M, [27] Hofmeyr GJ, Nikodem VC, de Jager M, Drakely A. Side- Meffe F, et al. Randomized controlled trial of rectal effects of oral misoprostol in the third stage of labour—a misoprostol versus oxytocin in third stage management. J randomised placebo-controlled trial. S Afr Med J 2001; Obstet Gynaecol Can 2002;24(2):149 – 54.
[40] Kundodyiwa TW, Majoko F, Rusakaniko S. Misoprostol versus [28] Hofmeyr GJ, Nikodem VC, de Jager M, Gelbart BR. A oxytocin in the third stage of labor. Int J Gynaecol Obstet randomised placebo controlled trial of oral misoprostol in the third stage of labour. Br J Obstet Gynaecol 1998; [41] Lam H, Tang OS, Lee CP, Ho PC. A pilot-randomized comparison of sublingual misoprostol with syntometrine [29] Surbek DV, Fehr PM, Hosli I, Holzgreve W. Oral misoprostol on the blood loss in third stage of labor. Acta Obstet for third stage of labor: a randomized placebo-controlled Gynecol Scand 2004;83(7):647 – 50.
trial. Obstet Gynecol 1999;94(2):255 – 8.
[42] Oboro VO, Tabowei TO. A randomised controlled trial of [30] Amant F, Spitz B, Timmerman D, Corremans A, Van Assche misoprostol versus oxytocin in the active management of FA. Misoprostol compared with methylergometrine for the the third stage of labour. J Obstet Gynaecol 2003;23(1): prevention of postpartum haemorrhage: a double-blind randomised trial. Br J Obstet Gynaecol 1999;106(10): [43] Ray A, Mukherjee P, Basu G, Chatterjee A. Misoprostol and third stage of labour. J Obstet Gynecol Ind 2001 (Novem- [31] Bamigboye AA, Merrell DA, Hofmeyr GJ, Mitchell R.
Randomized comparison of rectal misoprostol with Synto- [44] Vimala N, Mittal S, Kumar S, Dadhwal V, Mehta S. Sublingual metrine for management of third stage of labor. Acta misoprostol versus methylergometrine for active manage- Obstet Gynecol Scand 1998;77(2):178 – 81.
ment of the third stage of labor. Int J Gynaecol Obstet [32] Bugalho A, Daniel A, Faundes A, Cunha M. Misoprostol for prevention of postpartum hemorrhage. Int J Gynaecol [45] Walley RL, Wilson JB, Crane JM, Matthews K, Sawyer E, Hutchens D. A double-blind placebo controlled randomised [33] Caliskan E, Dilbaz B, Meydanli MM, Ozturk N, Narin MA, trial of misoprostol and oxytocin in the management of the Haberal A. Oral misoprostol for the third stage of labor: a third stage of labour. BJOG 2000;107(9):1111 – 5.
randomized controlled trial. Obstet Gynecol 2003;101(5 Pt [46] Jadad A. Randomized control trials. London7 BMJ Books; [34] Caliskan E, Meydanli MM, Dilbaz B, Aykan B, Sonmezer M, [47] El-ReFaey H. Use of misoprostol in third stage of labour.
Haberal A. Is rectal misoprostol really effective in the Lancet 2002;359(9307):707 – 8 [discussion 709—710].
treatment of third stage of labor? A randomized controlled [48] El-Refaey H, O’Brien P, Morafa W, Walder J, Rodeck C.
trial. Am J Obstet Gynecol 2002;187(4):1038 – 45.
Misoprostol for third stage of labour. Lancet 1996; [35] Cook CM, Spurrett B, Murray H. A randomized clinical trial comparing oral misoprostol with synthetic oxytocin or [49] Khan R, Sharma S. Use of misoprostol in third stage of labour.
syntometrine in the third stage of labour. Aust N Z J Obstet Lancet 2002;359(9307):708 – 9 [author reply 709—710].
[50] O’Brien P, Lokugamage AU, Guillebaud J, Rodeck CH. Use of [36] El-Refaey H, Nooh R, O’Brien P, Abdalla M, Geary M, Walder misoprostol in third stage of labour. Lancet 2002; J, et al. The misoprostol third stage of labour study: a 359(9307):708 [author reply 709—710].
randomised controlled comparison between orally admin- [51] Shannon C, Winikoff B. Use of misoprostol in third stage of istered misoprostol and standard management. BJOG labour. Lancet 2002;359(9307):709 [author reply 709—710].
[52] McDonald S, WJP, DE. Prophylactic syntometrine vs. oxyto- [37] Gerstenfeld TS, Wing DA. Rectal misoprostol versus intra- cin in the third stage of labour. The Cochrane Library; 1998.
venous oxytocin for the prevention of postpartum hemor-

Source: http://www.maternofetal.net/PDF/MisoprostolPostParto2.pdf