Description of the condition
The primary objective of postpartum haemorrhage (PPH; bleeding after childbirth) prevention and treatment is to reduce maternal deaths. Because mortality is generally too infrequent to be measured in randomised trials, blood loss is used as a surrogate outcome.
The postpartum period, also known as the puerperium, begins with the birth of the baby and placenta and ends six weeks after birth. Maternal mortality includes the death of a woman while pregnant or within 42 days of termination of pregnancy, irrespective of the duration and site of the pregnancy, from any cause related to, or aggravated by the pregnancy or its management, but not from accidental or incidental causes (AbouZahar 2003). Maternal morbidity, which is investigated in this review, includes major surgery (laparotomy, uterine artery ligation, internal iliac artery ligation, B-Lynch suture, hysterectomy, extensive vaginal repair), admission to the intensive care unit, vital organ failure (transient or permanent), and severe pyrexia (body temperature above 40ºC).
This review will focus only on maternal mortality and severe morbidity associated with the use of misoprostol for prevention and treatment of PPH. Other outcomes such as PPH and side-effects have been reviewed previously (Hofmeyr 2009), and are covered in The Cochrane Library by the reviews of prostaglandins for preventing PPH (Tuncalp 2012), and treatments for PPH (Mousa 2007). For this reason, PPH will not be reported on in this review.
The majority of trials on misoprostol were performed in low-income countries where PPH is more common in comparison to high-income countries. Maternal mortality is also much higher in low-income countries. Although the impact of this review will be more significant for low-income countries with a high incidence of PPH and maternal mortality, it is also important for high-income countries where misoprostol is commonly used for treatment of PPH.
Description of the intervention
Misoprostol is one of several uterotonic agents used for the prevention and treatment of PPH. It is an inexpensive and stable prostaglandin E1 analogue, and has been shown to stimulate uterine contractility in pregnancy (Norman 1991). Administered orally or vaginally, it is effective for induction of abortion and of labour, though it poses certain risks (Hofmeyr 1998).
Misoprostol has the major public health advantage over less stable, injectable medication such as oxytocin that it can more easily be distributed at community level (Rajbhandari 2010). A recent Monte Carlo simulation depicting mortality and anaemia-related morbidity attributable to PPH estimated that community-based distribution of misoprostol for prevention of PPH would lower mortality by 81% and for treatment of PPH by 70% (Sutherland 2010). A systematic review of misoprostol distribution for use at home births found evidence of a reduction in PPH, but concluded that more robust evidence was needed (Hundley 2013).
Misoprostol gained popularity as a treatment for PPH in both low- and high-income countries on the basis of several uncontrolled observational studies showing dramatic effects.
While common sense suggests that reduced blood loss should translate to reduced mortality, this intuitive link is not necessarily straightforward. It would be possible for a treatment to reduce blood loss while having other adverse effects which increase mortality. A good example of such a counter-intuitive effect is class I antiarrhythmics, which reduce post-myocardial infarction arrhythmias, but double mortality (Echt 1991). Misoprostol has ubiquitous effects on many organ systems. It is, therefore, important to monitor the effects on overall mortality as for any new intervention promoted on the basis of surrogate outcomes.
This review includes studies of misoprostol by any route or in any dose used to prevent or treat PPH compared with either placebo or another uterotonic. Other uterotonics may include oxytocin, ergometrine, sulprostone or combinations.
How the intervention might work
Misoprostol has well-established uterotonic effects. Given after birth, it decreases PPH in certain circumstances, though less effectively than oxytocin (Tuncalp 2012). However, being a prostaglandin analogue, misoprostol has ubiquitous effects on many organ systems. Pyrexia and shivering are common, dose-related side-effects (Lumbiganon 1999). A systematic review found an overall five-fold increase in pyrexia with misoprostol, which was dose-related and greatest with the sublingual route and least with the rectal route (Elati 2012). Hyperpyrexia (body temperature of 40°C or more) is a serious adverse event which may be life-threatening (Chong 1997). Adverse cardiovascular events reported following the use of sulprostone (which was subsequently withdrawn from the market) for treatment of PPH may have been related to pulmonary artery vasoconstriction, an adverse effect also described with misoprostol (Qian 1994). A statistically significant decrease in heart rate with misoprostol has been reported, which might impair the cardiovascular compensation for blood loss postpartum (Brecht 1987). Thus, while misoprostol might reduce PPH, it is possible that adverse effects might compromise homoeostasis in the postpartum period. To date, there is no direct evidence from randomised trials that postpartum misoprostol reduces maternal mortality.
Why it is important to do this review
Misoprostol is being advocated for wide use in prevention and treatment of PPH (Smith 2013), and is recommended for such in certain settings by the World Health Orgnization (Smith 2013; Tang 2013), the Clinical Practice Obstetrics Committee; Society of Obstetricians and Gynaecologists of Canada (Leduc 2009) and the American College of Obstetricians and Gynecologists (ACOG 2006). The Royal College of Obstetricians and Gynaecologists Green-Top Guideline no 52 of 2009 (revised 2011), recommends misoprostol 1000 µg rectally for treatment of PPH when other interventions fail (RCOG 2009) (http://www.rcog.org.uk/files/rcog-corp/GT52PostpartumHaemorrhage0411.pdf). Guidance for the use of misoprostol for the prevention (FIGO 2012a) and treatment (FIGO 2012b) of PPH has been issued by the International Federation Of Gynecology And Obstetrics. Given the fact that misoprostol has been introduced into practice through opportunistic use by clinicians, a number of investigator-initiated randomised trials and recommendations from global organisations, rather than by a manufacturing company with responsibility for post-marketing surveillance, it is most important to prospectively monitor maternal deaths and severe morbidity in all randomised trials of postpartum misoprostol to determine the net impact of the beneficial effects, known adverse effects, as well as any unknown adverse effects. The Cochrane review on treatments for primary PPH recommends: "A system of "adverse event registration" should be used to identify serious maternal morbidity and mortality associated with the use of misoprostol in clinical practice." (Mousa 2007) This is important both for low-resource settings where misoprostol is being introduced on a large scale for routine use after childbirth, and for well-resourced settings where misoprostol is commonly used to treat PPH when other uterotonics fail. Currently, no Cochrane review can monitor all adverse outcomes in trials of misoprostol, because trials of prevention (Tuncalp 2012) and treatment (Mousa 2007) are considered in separate reviews.
To review maternal deaths and severe morbidity in all randomised trials of misoprostol for prevention or treatment of postpartum haemorrhage.
Criteria for considering studies for this review
Types of studies
We included randomised controlled trials. We also planned to include cluster- and quasi-randomised trials in the analysis, as a very large number of women will be needed to obtain robust estimates of maternal mortality but we did not identify any for this version of the review. We will include trials reported only as abstracts if sufficient information is available from the abstract or from the authors.
Types of participants
Pregnant women ≥ 24 weeks' gestation who received misoprostol during the third stage of labour or in the postpartum period, versus placebo/no treatment or other uterotonics for prevention or treatment of postpartum haemorrhage (PPH). We included studies conducted in women who delivered by caesarean section.
Types of interventions
A: Different doses and different routes (sublingual, oral, vaginal, rectal) of misoprostol used for prevention or treatment of PPH compared versus B: other uterotonics, other doses or routes, or placebo/no treatment. We included comparisons of misoprostol versus placebo/no treatment or other routes or dosages in women who received routine uterotonics (e.g. we included misoprostol plus routine uterotonics versus placebo plus routine uterotonics as misoprostol versus placebo).
We considered it important to include in a single analysis comparisons of misoprostol versus both placebo/no treatment and versus other uterotonics for the following reasons:
- The main objective of the review was to monitor unexpected adverse effects related to misoprostol use in the third stage of labour, which, if existent, would be expected to occur irrespective of the comparator.
- Because severe adverse effects and death are likely to be rare, all relevant randomised trials needed to be considered in a single analysis in order to have the power to detect such effects as early as possible.
Types of outcome measures
- Maternal death
- Death or severe morbidity defined as any of the outcomes below.
- 1.1 Major surgery (laparotomy, uterine artery ligation, internal iliac artery ligation, B-Lynch suture, hysterectomy, extensive vaginal repair)
- 1.2 Admission to the intensive care unit
- 1.3 Vital organ failure (temporary or permanent)
- 1.4 Hyperpyrexia (body temperature ≥ 40ºC)
- 2. For dose comparisons, we also considered the outcome 'pyrexia ≥ 38°C'.
Because of the large number of women in the misoprostol groups with hyperpyrexia, we conducted a non-pre-specified analysis of maternal death or severe morbidity, excluding hyperpyrexia, to determine whether there were differences in morbidity from causes other than hyperpyrexia.
In a future update of the review, we will consider including another non-pre-specified analysis including blood transfusion as a component of severe morbidity.
Search methods for identification of studies
We contacted the Trials Search Co-ordinator to search the Cochrane Pregnancy and Childbirth Group’s Trials Register (11 January 2013).
The Cochrane Pregnancy and Childbirth Group’s Trials Register is maintained by the Trials Search Co-ordinator and contains trials identified from:
- monthly searches of the Cochrane Central Register of Controlled Trials (CENTRAL);
- weekly searches of MEDLINE;
- weekly searches of Embase;
- handsearches of 30 journals and the proceedings of major conferences;
- weekly current awareness alerts for a further 44 journals plus monthly BioMed Central email alerts.
Details of the search strategies for CENTRAL, MEDLINE and Embase, the list of handsearched journals and conference proceedings, and the list of journals reviewed via the current awareness service can be found in the ‘Specialized Register’ section within the editorial information about the Cochrane Pregnancy and Childbirth Group. Trials identified through the searching activities described above are each assigned to a review topic (or topics). The Trials Search Co-ordinator searches the register for each review using the topic list rather than keywords.
We did not apply any language restrictions.
Data collection and analysis
Selection of studies
Review author Tess Lawrie (TL) assessed for inclusion all the potential studies we identified as a result of the search strategy. We checked the selection against the selection made for a previous published version of the review (Hofmeyr 2009). For a subset of newer studies not assessed for the previous review, Justus Hofmeyr (GJH) repeated the assessment. We resolved any disagreement through discussion or, if required, we consulted Metin Gulmezoglu (AMG).
To the included studies, we allocated study identifiers which indicated the country (or countries) and date: misoprostol dosage in µg; route of misoprostol administration (PO = orally; SL = sublingually or buccally; PR = rectally); whether compared with a uterotonic ('U') or placebo ('P'), and whether used at caesarean section ('CS'); for example: 'Egypt 2009:800PR vs U'.
Data extraction and management
We designed a form to extract data. For eligible studies, TL extracted the data using the agreed form and compared these with the data extraction performed for the previous version of this review of (Hofmeyr 2009). For a subset of newer studies not assessed for the previous review, GJH performed a second data extraction. We resolved disagreements through discussion or, if required, we consulted AMG. We entered data into Review Manager software (RevMan 2011) and checked for accuracy. When information regarding any of the above was unclear or there were missing data, we attempted to contact authors of the original reports to provide further details.
Assessment of risk of bias in included studies
Measures of treatment effect
All review outcomes required dichotomous data only. We have presented most of the results of these data as summary risk ratios (RR) with 95% confidence intervals (CI). For the outcome maternal mortality, we used both RRs and Peto odds ratios (ORs) to exclude discrepancies between the two summary statistics (results not shown).
Unit of analysis issues
The unit of analysis for all outcomes was the individual participant. We did not identify any cluster-randomised trials for this version of the review. For trials with more than two treatment groups, we extracted only the relevant pair-wise comparisons that included misoprostol. If misoprostol was compared at different doses with other uterotonics or placebo, we extracted the data for each misoprostol dose comparison separately and entered them separately into the meta-analyses. If misoprostol was compared with two or more other types of standard uterotonics, we combined all relevant control intervention groups into a single control group, such that, for dichotomous outcomes, both the sample sizes and the number of people with events were summed across groups.
Dealing with missing data
For included studies, we noted levels of attrition. For all outcomes, we carried out analyses, as far as possible, on an intention-to-treat basis, i.e. we attempted to include all participants randomised to each group in the analyses, and all participants were analysed in the group to which they were allocated, regardless of whether or not they received the allocated intervention. The denominator for each outcome in each trial was the number randomised minus any participants whose outcomes were known to be missing.
Assessment of heterogeneity
We assessed statistical heterogeneity in each meta-analysis using the T², I² and Chi² statistics. We regarded heterogeneity as substantial if the I² was greater than 30% and either the T² was greater than zero, or there was a low P value (less than 0.10) in the Chi² test for heterogeneity.
Assessment of reporting biases
If there were 10 or more studies in the meta-analysis, we investigated reporting biases (such as publication bias) using funnel plots. We assessed funnel plot asymmetry visually. If asymmetry was suggested by a visual assessment, we performed exploratory analyses to investigate it.
We carried out statistical analysis using the Review Manager software (RevMan 2011). We used fixed-effect meta-analysis for combining data where it was reasonable to assume that studies were estimating the same underlying treatment effect: i.e. where trials were examining the same intervention, and the trials’ populations and methods were judged sufficiently similar. If there was clinical heterogeneity sufficient to expect that the underlying treatment effects differ between trials, or if substantial statistical heterogeneity was detected, we used random-effects meta-analysis to produce an overall summary if an average treatment effect across trials was considered clinically meaningful. The random-effects summary was treated as the average range of possible treatment effects and we discussed the clinical implications of treatment effects differing between trials. If the average treatment effect was not clinically meaningful, we did not combine trials. Where we used random-effects analyses, the results are presented as the average treatment effect with 95% CIs, and the estimates of T² and I².
Subgroup analysis and investigation of heterogeneity
If we identified substantial heterogeneity, we investigated it using subgroup analyses and sensitivity analyses. We considered whether an overall summary was meaningful, and if it was, used random-effects analysis to produce it.
We carried out the following subgroup analyses.
- By type of comparator: misoprostol versus placebo/no treatment; misoprostol versus other uterotonics; comparator mixed or unclear.
- By indication for misoprostol use: misoprostol to treat PPH; misoprostol to prevent PPH; indication mixed or unclear.
- By dose of misoprostol: 600 µg or more; 400 µg or less; more than 400 µg to less than 600 µg; dose mixed or unclear.
- By dose and route of misoprostol administration: orally; sublingually; rectally; vaginally; route mixed or unclear. We subdivided each route by dosages as in (3) above.
We used the following outcomes in subgroup analysis: 'maternal mortality' and 'maternal mortality or severe morbidity'. In addition, we performed an exploratory subgroup analysis for the outcome 'pyrexia ≥ 38°C'. For random-effects and fixed-effect meta-analyses, we assessed differences between subgroups by inspection of the subgroups’ confidence intervals, non-overlapping confidence intervals indicated a statistically significant difference in treatment effect between the subgroups, and by formal tests of subgroup differences.
In future versions of this review, we plan to perform subgroup analyses by co-treatment with routine uterotonics in both groups as follows: routine uterotonics used; routine uterotonics not used; use of routine uterotonics mixed or unclear.
We performed sensitivity analysis to determine the effects of trial quality on results, by excluding trials of lower quality for the following outcomes: 'maternal mortality' and 'maternal mortality or severe morbidity'. No effects of trial quality on the results were found (data not shown).
Description of studies
Results of the search
We located 117 studies. We excluded 34, and four are awaiting classification due to awaiting translation (1), full text publication (2) or further information (1), see Studies awaiting classification.
We included 78 studies (59,216 women); 71 postpartum haemorrhage (PPH) prevention studies and seven PPH treatment studies; 68 studies were conducted in women who underwent vaginal birth and 10 were conducted in women who underwent caesarean section. Misoprostol was compared to placebo (23), no additional treatment (2), other uterotonic agents (51) or uterotonic and placebo (2), at doses ranging from 50 µg to 1000 µg, via oral (35), sublingual (22), buccal (2), rectal (19), vaginal (1) and intrauterine routes (1).
Most of these studies reported side effects including fever, however several did not specifically report maternal morbidity and mortality data. For the initial WHO Bulletin version of the review, we obtained additional data from the authors of ten studies (Gambia 2004:600PO/SL vs P; Gambia 2005:600PO vs U; Guinea-B 2005:600SL vs P; India 2004a:400SL vs U; India 2005:600PO vs U; India 2005:600PO vs U; Nigeria 2003:600PO vs U; Turkey 2002:400PR vs P/U; WHO 1999:400/600PO vs U; Zim 2001:400PO vs U;). For the current review, we requested these additional data from contact authors of 24 studies and obtained them from the following (13): India 2012a:400SL vs U; India 2012b:400SL vs U; Spain 2009:400SL200PRvsN; Nigeria 2007:400PO vs U; Nigeria 2011:400SL vs P; Nigeria 2011:600PO vs U; Pakistan 2008:600SL vs P; Egypt 2009:800PR vs U; India 2009:400SL vs U; EEV 2010:800SL vs U; Egypt 2012b:CS400SL vs P; India 2010:CS800PR vs U; and India 2012:CS400SL vs P. No additional maternal deaths were identified through communication with the authors of these included trials. Several studies only contributed data relating to the incidence of fever, and three studies contributed no data (Bangla 2007:400PO vs U; China 2003:400PO vs N; Iran 2009:CS400SL vs U).
For more details see Characteristics of included studies.
We excluded 34 studies. For details see Characteristics of excluded studies.
Risk of bias in included studies
See Figure 1.
|Figure 1. 'Risk of bias' graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.|
The risk of allocation bias was considered low in 51 studies, unclear in 26 and high in one.
The risk of performance bias and detection bias was considered low in 42 studies, unclear in 13 and high in 23.
Incomplete outcome data
The risk of attrition bias was considered low in 62 studies and unclear in 16.
The risk of reporting bias was considered low in 65 studies, unclear in 12 and high in one.
Other potential sources of bias
The risk of other potential sources of bias was considered low in 62 studies and unclear in 16. Publication bias was assessed in funnel plots (Figure 2; Figure 3; Figure 4; Figure 5). Only for the outcome 'Pyrexia ≥ 38°C' was there some visual asymmetry (Figure 5).
|Figure 2. Funnel plot of comparison: 1 Misoprostol versus control (placebo or uterotonic) subgrouped by comparator, outcome: 1.1 Maternal deaths.|
|Figure 3. Funnel plot of comparison: 1 Misoprostol versus control (placebo or uterotonic) subgrouped by comparator, outcome: 1.2 Maternal deaths or severe morbidity.|
|Figure 4. Funnel plot of comparison: 1 Misoprostol versus control (placebo or uterotonic) subgrouped by comparator, outcome: 1.4 Maternal deaths or severe morbidity excluding hyperpyrexia.|
|Figure 5. Funnel plot of comparison: 1 Misoprostol versus control (placebo or uterotonic) subgrouped by comparator, outcome: 1.5 Pyrexia ≥ 38°C.|
Effects of interventions
Primary outcome: Maternal mortality
There was no statistically significant difference in maternal mortality with misoprostol compared to all control groups (31 studies; 11/19,715 (56/100,000) versus 4/20,076 deaths (20/100,000); risk ratio (RR) 2.08, 95% confidence interval (CI) 0.82 to 5.28; Analysis 1.1), or for any of the comparison subgroups; however, point estimates favoured the comparison groups:
- Misoprostol versus other uterotonics: 21 studies, 5/15,089 (3/100,000) versus 3/15,369 (19/100,000); RR 1.54; 95% CI 0.40 to 5.92; Analysis 1.1.
All maternal deaths occurred in studies evaluating misoprostol doses ≥ 600 µg versus controls ( Analysis 3.1). There was no statistically significant difference between these groups overall: 18 studies: 11/16,202 (68/100,000) versus 4/16,226 (25/100,000); RR 2.08; 95% CI 0.82 to 5.28; Analysis 3.1. Maternal deaths occurred in studies using the oral, sublingual and mixed routes of administration. No deaths occurred in studies using misoprostol 400 µg or less (12 studies; 0/3483 versus 0/3820). This may be because, in general, 400 µg was used in prevention rather than treatment studies, with a lower associated risk of mortality.
Maternal death or severe morbidity
'Maternal death or severe morbidity' was significantly higher with misoprostol compared with placebo: 12 studies; 43/5003 (0.86%) versus 24/5082 (0.47%); average RR 1.70, 95% CI 1.02 to 2.81; Tau² = 0, I² = 0%; Analysis 1.2. One study (SATAEV 2010:600SL vs P), contributed most of the morbidity events (41/67 events). When we excluded this study in a sensitivity analysis, there was no longer a statistically significant difference between the misoprostol and placebo groups (11 studies; average RR 1.15, 95% CI 0.51 to 2.57; I² = 0%; Analysis 1.3).
There was no statistically significant difference between misoprostol and other uterotonics: 17 studies; average RR 1.50, 95% CI 0.50, 4.52; Tau² = 1.81, I² = 69%, Analysis 1.2; however, there was significant heterogeneity in this subgroup. This was due to the very large effect in one study (EEV 2010:800SL vs U), contributed to by an unusually high rate of hyperpyrexia in one site (58/66 cases occurred in Ecuador). We performed sensitivity analysis by excluding this study. Following this exclusion, the point estimate of the average RR reduced to 1.16, results remained not statistically significantly different, and the heterogeneity disappeared (16 studies; RR 1.09, 95% CI 0.67 to 1.76; I² = 0%; Analysis 1.3).
When we excluded both (EEV 2010:800SL vs U and SATAEV 2010:600SL vs P), the overall relative effect of misoprostol versus control was not significantly different between the groups (27 studies; average RR 1.16, 95% CI 0.78 to 1.72; I² = 0%, Analysis 1.3).
Considering all the studies of misoprostol versus placebo or other uterotonics, most cases of 'maternal death or severe morbidity' occurred in studies using 600 µg misoprostol or more (19 trials; average RR 1.67; 95% CI 0.80 to 3.45; Analysis 3.2). When we excluded the two outlier studies as above in a sensitivity analysis (EEV 2010:800SL vs U and SATAEV 2010:600SL vs P), the results for 'maternal death or severe morbidity' remained statistically non-significant, Analysis 3.3).
Maternal death or severe morbidity, excluding hyperpyrexia
To determine whether differences in the outcome 'maternal death or severe morbidity' were mainly due to cases of hyperpyrexia, we conducted a non-prespecified analysis of the outcome excluding hyperpyrexia. There was no difference between misoprostol and the comparison groups overall (29 studies; RR 0.97; 95% CI 0.67 to 1.41; I² = 0%, Analysis 1.4), or when subgrouped by misoprostol dosage ( Analysis 3.4). This suggests that the increased morbidity with misoprostol was mainly due to hyperpyrexia, with most events occurring in two large multicentre studies of misoprostol for the treatment of PPH (EEV 2010:800SL vs U; SATAEV 2010:600SL vs P). In these latter studies, hyperpyrexia occurred mainly in Ecuador.
Pyrexia ≥ 38°C
There was considerable heterogeneity in the effect on pyrexia, which was increased with misoprostol (56 studies, 2776/25,647 (10.8%) versus 614/26,800 (2.3%); average RR 3.97, 95% CI 3.13 to 5.04, Tau² = 0.47, I² = 80% random-effects model; Analysis 1.5). This increase occurred both in trials of misoprostol versus placebo (20 studies; 1059/6212 (17%) versus 265/6228 (4.3%); average RR 3.25; 95% CI 2.30 to 4.59; Tau² = 0.40, I² = 79%) and versus other uterotonics (39 studies; 1717/19,435 (8.8%) versus 349/20,572 (1.7%); average RR 4.70; 95% CI 3.36 to 6.57; Tau² = 0.59, I² = 81%; Analysis 1.5). The effect was greater for trials using misoprostol 600 µg or more (27 studies; 2197/17,864 (12.3%) versus 422/18,161 (2.3%); average RR 4.64; 95% CI 3.33 to 6.46; Tau² = 0.51, I² = 86%; Analysis 3.5) than for those using misoprostol 400 µg or less (31 studies; 525/6751 (7.8%) versus 185/7668 (2.4%); average RR 3.07; 95% CI 2.25 to 4.18; Tau² = 0.29, I² = 58%; Analysis 3.5).
Summary of main results
The number of maternal deaths is too small for meaningful statistical analysis. The range of plausible effects lies between a small (18%) reduction and a large (5.28 times) increase with misoprostol. The outcome 'death or severe morbidity' was increased with misoprostol, due to a large increase in hyperpyrexia in dosages of 600 µg or more. When hyperpyrexia was excluded from the definition, there was no difference between groups. As most of these hyperpyrexia events occurred in Ecuador, this may indicate a genetic predisposition. Pyrexia was, as expected, increased with misoprostol and this effect was dose-related.
Overall completeness and applicability of evidence
This review only includes studies of women who received misoprostol after giving birth for prevention or treatment of PPH. It does not include studies of women receiving misoprostol for abortion or induction of labour, for which different dosages of misoprostol are used. Therefore, these results are not applicable to these other uses of misoprostol, which require a separate review.
Although not statistically significantly different to the control group data, all maternal deaths and most maternal morbidity with misoprostol occurred with dosages of ≥ 600 µg. We consider the current evidence to be incomplete, particularly with regard to the optimal dosage of misoprostol in relation to maternal mortality and morbidity, which requires further investigation in large randomised trials.
Quality of the evidence
The methodological quality of studies was variable, but most of the studies that contributed data on maternal death were at a low risk of bias. We consider the quality of evidence with regard to maternal mortality or morbidity to be moderate due to the small numbers of events and further evidence is likely to change our confidence in the estimates of effect.
A feature of the review is the considerable heterogeneity of results which may be related to the variety of study designs, populations studied and co-interventions. Sensitivity analysis performed by removing studies at higher risk of bias did not remove the heterogeneity. However, two studies were responsible for much of the heterogeneity related to maternal morbidity, especially the exceptionally high incidence of hyperpyrexia in Ecuador. When we excluded these studies in sensitivity analysis, heterogeneity related to the composite outcome 'death or severe morbidity' was eliminated. Hyperpyrexia data from these studies seem incompatible with those of other studies, however, they may indicate that genetic differences contribute to the pyrexial events with misoprostol use. Therefore, for the composite outcome, we consider the results of the non-prespecified analysis (that excludes hyperpyrexia per se, and not the two trials) to be of a better quality and more widely applicable.
Potential biases in the review process
We pooled the relative effects of comparator groups (placebo and uterotonics) in our meta-analyses in order to improve the power of our meta-analyses, and on the basis that severe adverse effects of misoprostol, if existent, would be expected to occur in the misoprostol groups irrespective of the comparator. We have also presented the results for all outcomes by individual comparator groups, as well as the pooled results, and therefore do not consider this to be a substantial source of bias in the review. In addition, we have used random-effects methods for most meta-analyses, except for maternal mortality which is a relatively rare outcome, or where there was no evidence of heterogeneity in the data.
It is possible that our composite outcome may be improved by including other outcomes, e,g, blood transfusions, which may be better indicators of severe morbidity. We plan to reconsider this composite outcome in future versions of this review. However, severe morbidity including hyperpyrexia may have been underestimated as some studies did not perform a quantitative assessment of pyrexia but included other severe morbidity, and so we included these in this meta-analysis (e.g. Nigeria 2011:600PO vs U). On the other hand, hyperpyrexia may be less important as an indicator of severe morbidity as it does not appear to have long-term effects on the woman or lead to near-misses or prolonged hospitalisation.
Many authors find it redundant to report that there were no maternal deaths in their studies, especially those small studies conducted in low-risk women. If no maternal death was mentioned in the paper, we asked the authors to confirm that no death had occurred. If we were unable to contact them, we excluded the study from the outcome analysis. By excluding studies where the reporting of maternal deaths was omitted, we may have over-estimated the risk of maternal death slightly; however, the studies excluded on this basis were numerically small and unlikely to have had a large effect.
Agreements and disagreements with other studies or reviews
The findings are consistent with our previous review (Hofmeyr 2009). We are not aware of other reviews that have focused on maternal deaths in randomised postpartum misoprostol trials.
Implications for practice
We have not found evidence that misoprostol significantly increases or decreases the risk of maternal death or serious morbidity (excluding hyperpyrexia) when used in the prevention and treatment of PPH. This review focusses on safety and has not addressed the effectiveness of misoprostol in terms of blood loss, which is covered in separate Cochrane reviews (Mousa 2007; Tuncalp 2012). Tuncalp 2012 found that misoprostol was less effective than injectable uterotonics with respect to blood loss, and more effective than placebo in settings in which injectable uterotonics were not used.
The increased risk of hyperpyrexia with doses of 600 µg or above is reason for concern. Hyperpyrexia is occasionally life-threatening, and at least (as is shivering and pyrexia) an unpleasant experience for women at a time that interactions with her newborn baby are of great importance. In one study of misoprostol 800 µg sublingually for the treatment of postpartum haemorrhage included in this review (EEV 2010:800SL vs U), 229 (47%) of women experienced shivering and 55 found it intolerable, while 66 women (14%) experienced hyperpyrexia > 40°C, of whom 22 found the pyrexia intolerable. These hyperpyrexia results are incompatible with the other included studies and are unlikely to be representative of most women using misoprostol. However, they do indicate that certain women may have a genetic predisposition to a hyperpyrexia response. Pyrexia > 38°C, however, was reported across all trials using misoprostol 600 µg or more.
Our previous review found no difference in effectiveness in terms of blood loss between misoprostol 600 µg or more and 400 µg, using both direct and adjusted indirect comparisons (Hofmeyr 2009). The current review has found no evidence of benefit in terms of 'maternal mortality or severe morbidity' with misoprostol in any dosage versus placebo, and increased adverse effects (pyrexia) in dosages of 600 µg or more. Given the fact that misoprostol is used prophylactically in very large numbers of healthy women, in settings without access to more effective uterotonics, the greatest emphasis should be placed on limiting adverse effects. In this context, the findings of this review support use of the lowest effective dose.
Implications for research
As for any new medication being used on a very large scale, continued vigilance for adverse effects is essential. In the context of ongoing programs to introduce misoprostol for routine management of the third stage of labour, there is a need for very large (cluster) randomised trials to further elucidate both the relative effectiveness and the risks of adverse effects of various dosages of misoprostol. There is also a need for more studies of the effects of various dosages of postpartum misoprostol on physiologic parameters such as pulmonary artery vasoconstriction (Qian 1994) and heart rate (Brecht 1987).
Pregnancy and Childbirth Group editorial base for technical support; Verena Linder, Sandra Ferreira and Gilda Piaggio for contributions to previous, non-Cochrane versions of this review.
As part of the pre-publication editorial process, this review has been commented on by four peers (an editor and three referees who are external to the editorial team), a member of the Pregnancy and Childbirth Group's international panel of consumers and the Group's Statistical Adviser.
The World Health Organization (and G Justus Hofmeyr, Natalia Novikova and Theresa A Lawrie) retain copyright and all other rights in their respective contributions to the manuscript of this Review as submitted for publication.
The National Institute for Health Research (NIHR) is the largest single funder of the Cochrane Pregnancy and Childbirth Group. The views and opinions expressed therein are those of the authors and do not necessarily reflect those of the NIHR, NHS or the Department of Health.
Data and analyses
- Top of page
- Authors' conclusions
- Data and analyses
- Contributions of authors
- Declarations of interest
- Sources of support
- Differences between protocol and review
- Index terms
Appendix 1. Risk of bias tool
(1) Random sequence generation (checking for possible selection bias)
We described for each included study the method used to generate the allocation sequence in sufficient detail to allow an assessment of whether it should produce comparable groups.
We will assess the method as:
- low risk of bias (any truly random process, e.g. random number table; computer random number generator);
- high risk of bias (any non-random process, e.g. odd or even date of birth; hospital or clinic record number); or
- unclear risk of bias.
(2) Allocation concealment (checking for possible selection bias)
We described for each included study the method used to conceal allocation to interventions prior to assignment and assessed whether intervention allocation could have been foreseen in advance of, or during recruitment, or changed after assignment.
We assessed the methods as:
- low risk of bias (e.g. telephone or central randomisation; consecutively numbered sealed opaque envelopes);
- high risk of bias (open random allocation; unsealed or non-opaque envelopes, alternation; date of birth);
- unclear risk of bias.
(3.1) Blinding of participants and personnel (checking for possible performance bias)
We described for each included study the methods used, if any, to blind study participants and personnel from knowledge of which intervention a participant received. We considered that studies were at low risk of bias if they were blinded, or if we judge that the lack of blinding would be unlikely to affect results. We assessed blinding separately for different outcomes or classes of outcomes.
We assessed the methods as:
- low, high or unclear risk of bias for participants;
- low, high or unclear risk of bias for personnel.
(3.2) Blinding of outcome assessment (checking for possible detection bias)
We described for each included study the methods used, if any, to blind outcome assessors from knowledge of which intervention a participant received. We assessed methods used to blind outcome assessment as:
- low, high or unclear risk of bias.
(4) Incomplete outcome data (checking for possible attrition bias due to the amount, nature and handling of incomplete outcome data)
We described for each included study, and for each outcome or class of outcomes, the completeness of data including attrition and exclusions from the analysis. We stated whether attrition and exclusions were reported and the numbers included in the analysis at each stage (compared with the total randomised participants), reasons for attrition or exclusion where reported, and whether missing data were balanced across groups or were related to outcomes. Where sufficient information was reported, or could be supplied by the trial authors, we re-included missing data in the analyses which we undertook.
We assessed methods as:
- low risk of bias (e.g. no missing outcome data; missing outcome data balanced across groups);
- high risk of bias (e.g. numbers or reasons for missing data imbalanced across groups; ‘as treated’ analysis done with substantial departure of intervention received from that assigned at randomisation);
- unclear risk of bias.
(5) Selective reporting (checking for reporting bias)
We assessed the methods as:
- low risk of bias (where it is clear that all of the study’s pre-specified outcomes and all expected outcomes of interest to the review have been reported);
- high risk of bias (where not all the study’s pre-specified outcomes have been reported; one or more reported primary outcomes were not pre-specified; outcomes of interest are reported incompletely and so cannot be used; study fails to include results of a key outcome that would have been expected to have been reported);
- unclear risk of bias.
(6) Other bias (checking for bias due to problems not covered by (1) to (5) above)
We assessed whether each study was free of other problems that could have put it at risk of bias:
- low risk of other bias;
- high risk of other bias;
- unclear whether there is risk of other bias.
(7) Overall risk of bias
We made explicit judgements about whether studies were at high risk of bias, according to the criteria given in the Handbook (Higgins 2011). With reference to (1) to (6) above, we assessed the likely magnitude and direction of the bias and whether we consider it likely to impact on the findings. We explored the impact of the level of bias through undertaking sensitivity analyses - see 'Sensitivity analysis'.
Contributions of authors
GJH conceived the review and developed the fist draft of the protocol. NN, AMG and GJH contributed to the development of the protocol. TL and GJH contributed to selection of studies and data extraction. GJH prepared the first draft of the review. All authors contributed to the final version of the review.
Declarations of interest
GJH and AMG have participated in trials eligible for consideration of inclusion in the review. They did not participate in decisions regarding such trials. GJH has received research funding from Gynuity, a non-profit organisation, for conducting a trial of misoprostol for preventing postpartum haemorrhage.
Sources of support
- No sources of support supplied
- National Institute for Health Research, UK.Cochrane Review Incentive Scheme Award 12/183/03
Differences between protocol and review
An exploratory, non-prespecified, analysis of maternal death or severe morbidity excluding hyperpyrexia was conducted to determine whether the increase in maternal death nor severe morbidity with misoprostol was due entirely to increased hyperpyrexia.
For future versions of this review, we may reformulate the composite outcome 'maternal death or severe morbidity', possibly to include 'blood transfusions'. Furthermore, it may be of value to adapt the protocol to include studies of misoprostol use for retained placenta.
Medical Subject Headings (MeSH)
*Maternal Mortality; Fever [chemically induced]; Misoprostol [*administration & dosage; adverse effects]; Oxytocics [*administration & dosage; adverse effects]; Postpartum Hemorrhage [mortality; *prevention & control]; Randomized Controlled Trials as Topic
MeSH check words
Female; Humans; Pregnancy
* Indicates the major publication for the study