Uterotonic agents for preventing postpartum haemorrhage: a network meta-analysis


Abstract

Background: Postpartum haemorrhage (PPH) is the leading cause of maternal mortality worldwide. Prophylactic uterotonic drugs can prevent PPH, and are routinely recommended. There are several uterotonic drugs for preventing PPH but it is still debatable which drug is best.

Objectives: To identify the most effective uterotonic drug(s) to prevent PPH, and generate a ranking according to their effectiveness and side-effect profile.

Search methods: We searched Cochrane Pregnancy and Childbirth''s Trials Register (1 June 2015), ClinicalTrials.gov and the World Health Organization (WHO) International Clinical Trials Registry Platform (ICTRP) for unpublished trial reports (30 June 2015) and reference lists of retrieved studies.

Selection criteria: All randomised controlled comparisons or cluster trials of effectiveness or side-effects of uterotonic drugs for preventing PPH.Quasi-randomised trials and cross-over trials are not eligible for inclusion in this review.

Data collection and analysis: At least three review authors independently assessed trials for inclusion and risk of bias, extracted data and checked them for accuracy. We estimated the relative effects and rankings for preventing PPH ≥ 500 mL and PPH ≥ 1000 mL as primary outcomes. We performed pairwise meta-analyses and network meta-analysis to determine the relative effects and rankings of all available drugs. We stratified our primary outcomes according to mode of birth, prior risk of PPH, healthcare setting, dosage, regimen and route of drug administration, to detect subgroup effects.The absolute risks in the oxytocin are based on meta-analyses of proportions from the studies included in this review and the risks in the intervention groups were based on the assumed risk in the oxytocin group and the relative effects of the interventions.

Main results: This network meta-analysis included 140 randomised trials with data from 88,947 women. There are two large ongoing studies. The trials were mostly carried out in hospital settings and recruited women who were predominantly more than 37 weeks of gestation having a vaginal birth. The majority of trials were assessed to have uncertain risk of bias due to poor reporting of study design. This primarily impacted on our confidence in comparisons involving carbetocin trials more than other uterotonics.The three most effective drugs for prevention of PPH ≥ 500 mL were ergometrine plus oxytocin combination, carbetocin, and misoprostol plus oxytocin combination. These three options were more effective at preventing PPH ≥ 500 mL compared with oxytocin, the drug currently recommended by the WHO (ergometrine plus oxytocin risk ratio (RR) 0.69 (95% confidence interval (CI) 0.57 to 0.83), moderate-quality evidence; carbetocin RR 0.72 (95% CI 0.52 to 1.00), very low-quality evidence; misoprostol plus oxytocin RR 0.73 (95% CI 0.60 to 0.90), moderate-quality evidence). Based on these results, about 10.5% women given oxytocin would experience a PPH of ≥ 500 mL compared with 7.2% given ergometrine plus oxytocin combination, 7.6% given carbetocin, and 7.7% given misoprostol plus oxytocin. Oxytocin was ranked fourth with close to 0% cumulative probability of being ranked in the top three for PPH ≥ 500 mL.The outcomes and rankings for the outcome of PPH ≥ 1000 mL were similar to those of PPH ≥ 500 mL. with the evidence for ergometrine plus oxytocin combination being more effective than oxytocin (RR 0.77 (95% CI 0.61 to 0.95), high-quality evidence) being more certain than that for carbetocin (RR 0.70 (95% CI 0.38 to 1.28), low-quality evidence), or misoprostol plus oxytocin combination (RR 0.90 (95% CI 0.72 to 1.14), moderate-quality evidence)There were no meaningful differences between all drugs for maternal deaths or severe morbidity as these outcomes were so rare in the included randomised trials.Two combination regimens had the poorest rankings for side-effects. Specifically, the ergometrine plus oxytocin combination had the higher risk for vomiting (RR 3.10 (95% CI 2.11 to 4.56), high-quality evidence; 1.9% versus 0.6%) and hypertension [RR 1.77 (95% CI 0.55 to 5.66), low-quality evidence; 1.2% versus 0.7%), while the misoprostol plus oxytocin combination had the higher risk for fever (RR 3.18 (95% CI 2.22 to 4.55), moderate-quality evidence; 11.4% versus 3.6%) when compared with oxytocin. Carbetocin had similar risk for side-effects compared with oxytocin although the quality evidence was very low for vomiting and for fever, and was low for hypertension.

Authors' conclusions: Ergometrine plus oxytocin combination, carbetocin, and misoprostol plus oxytocin combination were more effective for preventing PPH ≥ 500 mL than the current standard oxytocin. Ergometrine plus oxytocin combination was more effective for preventing PPH ≥ 1000 mL than oxytocin. Misoprostol plus oxytocin combination evidence is less consistent and may relate to different routes and doses of misoprostol used in the studies. Carbetocin had the most favourable side-effect profile amongst the top three options; however, most carbetocin trials were small and at high risk of bias.Amongst the 11 ongoing studies listed in this review there are two key studies that will inform a future update of this review. The first is a WHO-led multi-centre study comparing the effectiveness of a room temperature stable carbetocin versus oxytocin (administered intramuscularly) for preventing PPH in women having a vaginal birth. The trial includes around 30,000 women from 10 countries. The other is a UK-based trial recruiting more than 6000 women to a three-arm trial comparing carbetocin, oxytocin and ergometrine plus oxytocin combination. Both trials are expected to report in 2018.Consultation with our consumer group demonstrated the need for more research into PPH outcomes identified as priorities for women and their families, such as women''s views regarding the drugs used, clinical signs of excessive blood loss, neonatal unit admissions and breastfeeding at discharge. To date, trials have rarely investigated these outcomes. Consumers also considered the side-effects of uterotonic drugs to be important but these were often not reported. A forthcoming set of core outcomes relating to PPH will identify outcomes to prioritise in trial reporting and will inform futures updates of this review. We urge all trialists to consider measuring these outcomes for each drug in all future randomised trials. Lastly, future evidence synthesis research could compare the effects of different dosages and routes of administration for the most effective drugs.

Conflict of interest statement

Ioannis D Gallos (IDG): is a co‐applicant to the UK National Institute for Health Research HTA Project Award 14/139/17 entitled “Uterotonic drugs for preventing postpartum haemorrhage: a network meta‐analysis and cost‐effectiveness analysis”. He has been involved in one or more previous or ongoing trials related to the use of uterotonics for the prevention of PPH that could be eligible for inclusion in this review. He will not participate in any decisions regarding these trials (i.e. assessment for inclusion/exclusion, trial quality, data extraction) for the purposes of this review or future updates – these tasks will be carried out by other members of the team who are not directly involved in the trials. Ferring Pharmaceuticals and Novartis have supplied carbetocin and oxytocin for studies and an ongoing study is supported by WHO/Merck for Mothers. IDG has been supported by the MSD for mothers initiative for travel to a meeting for the study.

Helen M Williams (HMW): is part‐funded by the Birmingham Women’s NHS Foundation Trust, and a co‐applicant to the UK National Institute for Health Research HTA Project Award 14/139/17 entitled “Uterotonic drugs for preventing postpartum haemorrhage: a network meta‐analysis and cost‐effectiveness analysis”. Her salary is part‐funded by Tommy''s. She is a member of the Executive Board of Ammalife (UK registered charity 1120236). She has also assisted the administration of activities at a single study site in contribution to a multinational randomised controlled trial of carbetocin versus oxytocin,that could potentially be eligible for inclusion in this review. The trial is sponsored by the World Health Organization and supported by Merck for Mothers. She will not participate in decisions regarding the inclusion of this trial in the review or any tasks related to it such as data extraction or quality assessment.

Malcolm J Price (MP) is funded as a research fellow by the UK Medical Research Council (MRC) Project Award MR/J013595/1, and a co‐applicant to the UK National Institute for Health Research HTA Project Award 14/139/17 entitled “Uterotonic drugs for preventing postpartum haemorrhage: a network meta‐analysis and cost‐effectiveness analysis”.

Aurelio Tobias: none known.

Abi Merriel (AM): was part‐funded by Ammalife (UK Registered Charity 1120236) and the Birmingham Women’s NHS Foundation Trust.

Harold Gee (HG): is a Trustee of Ammalife (UK Registered Charity 1120236).

David Lissauer (DL): was previously a Trustee of Ammalife (UK Registered Charity 1120236).

Vidhya Moorthy: none known.

Mariana Widmer (MW): is involved in an ongoing trial related to the use of uterotonics for the prevention of PPH that could be eligible for inclusion in this review. Ferring Pharmaceuticals and Novartis have supplied carbetocin and oxytocin for the trial and the study is supported by WHO/Merck for Mothers. MW will not participate in any decisions regarding this trial (i.e. assessment for inclusion/exclusion, trial quality, data extraction)for the purposes of this review or future updates – these tasks will be carried out by other members of the team who are not directly involved in the trial.

?zge Tun?alp (OT): is a co‐applicant to the UK National Institute for Health Research HTA Project Award 14/139/17 entitled “Uterotonic drugs for preventing postpartum haemorrhage: a network meta‐analysis and cost‐effectiveness analysis”.

A Metin Gulmezoglu (AMG): is a co‐applicant to the UK National Institute for Health Research HTA Project Award 14/139/17 entitled “Uterotonic drugs for preventing postpartum haemorrhage: a network meta‐analysis and cost‐effectiveness analysis”. AMG was involved in the large multicentre trial (as part of the central coordination unit) which may be included in the review. AMG is involved in an ongoing trial related to the use of uterotonics for the prevention of PPH that could be eligible for inclusion in this review. Ferring Pharmaceuticals and Novartis have supplied carbetocin and oxytocin for the trial and the study is supported by WHO/Merck for Mothers. AMG will not participate in any decisions regarding this or previous trials (i.e. assessment for inclusion/exclusion, trial quality, data extraction)for the purposes of this review or future updates – these tasks will be carried out by other members of the team who are not directly involved in the trial.

Jonathan J Deeks (JJD): is a co‐applicant to the UK National Institute for Health Research HTA Project Award 14/139/17 entitled “Uterotonic drugs for preventing postpartum haemorrhage: a network meta‐analysis and cost‐effectiveness analysis”.

G Justus Hofmeyr (GJH) has been and continues to be involved in a number of studies that may be eligible for inclusion in this review, but will not participate in data extraction or quality assessment of the studies in which he was involved. He is a co‐investigator on the UK National Institute for Health Research HTA Project Award 14/139/17 entitled "Uterotonic drugs for preventing postpartum haemorrhage: a network meta‐analysis and cost‐effectiveness analysis". Neither he nor his institution receives funding from this grant.

Arri Coomarasamy (AC): is the Chief Investigator of UK National Institute for Health Research HTA Project Award 14/139/17 entitled "Uterotonic drugs for preventing postpartum haemorrhage: a network meta‐analysis and cost‐effectiveness analysis". He has been involved in one or more previous or ongoing trials related to the use of uterotonics for the prevention of PPH that could be eligible for inclusion in this review. Ferring Pharmaceuticals and Novartis have supplied carbetocin and oxytocin for studies and an ongoing study is supported by WHO/Merck for Mothers. AC will not participate in any decisions regarding these trials (i.e. assessment for inclusion/exclusion, trial quality, data extraction)for the purposes of this review or future updates – these tasks will be carried out by other members of the team who are not directly involved in the trials. AC is a member of the Executive Board of Ammalife (UK registered charity 1120236). He does not receive any payment for this relationship.

Figures

Figure 1
Figure 1
Study flow diagram.
Figure 2
Figure 2
''Risk of bias'' graph: review authors'' judgements about each risk of bias item presented as percentages across all included studies.
Figure 3
Figure 3
''Risk of bias'' summary: review authors'' judgements about each risk of bias item for each included study.
Figure 4
Figure 4
Network diagram for PPH ≥ 500 mL. The nodes represent an intervention and their size is proportional to the number of trials comparing this intervention to any other in the network. The lines connecting each pair of interventions represent a direct comparison and are drawn proportional to the number of trials making each direct comparison. Numbers on the lines represent the number of trials and participants for each comparison. The colour of the line is green when more than 50% of the trials involved in the specific direct comparison are judged to be at “low risk of bias” if they were double‐blinded, and had allocation concealment with little loss to follow‐up (less than 10%). The colour is red when less than 50% of the trials are at “low risk of bias”. Multi‐arm trials contribute to more than one comparison.
Figure 5
Figure 5
Forest plot with relative risk ratios and 95% CIs from network meta‐analysis and pairwise analyses for prevention of PPH ≥ 500 mL.
Figure 6
Figure 6
Cumulative rankograms comparing each of the uterotonic drugs for prevention of PPH ≥ 500 mL. Ranking indicates the cumulative probability of being the best drug, the second best, the third best, etc. The x‐axis shows the relative ranking and the y‐axis the cumulative probability of each ranking. We estimate the SUrface underneath this Cumulative RAnking line (SUCRA); the larger the SUCRA the higher its rank among all available drug options.
Figure 7
Figure 7
Network diagram for PPH ≥ 1000 mL.
Figure 8
Figure 8
Forest plot with relative risk ratios and 95% CIs from network meta‐analysis and pairwise analyses for prevention of PPH ≥ 1000 mL.
Figure 9
Figure 9
Cumulative rankograms comparing each of the uterotonic drugs for prevention of PPH ≥ 1000 mL.
Figure 10
Figure 10
Forest plot with relative risk ratios and 95% CIs from network meta‐analysis and pairwise analyses for maternal death.
Figure 11
Figure 11
Cumulative rankograms comparing each of the uterotonic drugs for prevention of maternal death.
Figure 12
Figure 12
Forest plot with relative risk ratios and 95% CIs from network meta‐analysis and pairwise analyses for maternal death or severe morbidity.
Figure 13
Figure 13
Cumulative rankograms comparing each of the uterotonic drugs for prevention of maternal deaths or severe morbidity events.
Figure 14
Figure 14
Forest plot with relative risk ratios and 95% CIs from network meta‐analysis and pairwise analyses for the requirement of additional uterotonics.
Figure 15
Figure 15
Cumulative rankograms comparing each of the uterotonic drugs for the requirement of additional uterotonics.
Figure 16
Figure 16
Forest plot with relative risk ratios and 95% CIs from network meta‐analysis and pairwise analyses for the requirement of blood transfusion.
Figure 17
Figure 17
Cumulative rankograms comparing each of the uterotonic drugs for the requirement of blood transfusion.
Figure 18
Figure 18
Forest plot with relative risk ratios and 95% CIs from network meta‐analysis and pairwise analyses for the requirement of manual removal of placenta.
Figure 19
Figure 19
Cumulative rankograms comparing each of the uterotonic drugs for the requirement of manual removal of placenta.
Figure 20
Figure 20
Forest plot with relative risk ratios and 95% CIs from network meta‐analysis and pairwise analyses for blood loss (mL).
Figure 21
Figure 21
Cumulative rankograms comparing each of the uterotonic drugs for blood loss (mL).
Figure 22
Figure 22
Forest plot with relative risk ratios and 95% CIs from network meta‐analysis and pairwise analyses for duration of third stage (minutes).
Figure 23
Figure 23
Cumulative rankograms comparing each of the uterotonic drugs for duration of third stage (minutes).
Figure 24
Figure 24
Forest plot with relative risk ratios and 95% CIs from network meta‐analysis and pairwise analyses for change in haemoglobin measurements before and after birth (g/L).
Figure 25
Figure 25
Cumulative rankograms comparing each of the uterotonic drugs for change in haemoglobin measurements before and after birth (g/L).
Figure 26
Figure 26
Forest plot with relative risk ratios and 95% CIs from network meta‐analysis and pairwise analyses for neonatal unit admissions.
Figure 27
Figure 27
Cumulative rankograms comparing each of the uterotonic drugs for neonatal unit admissions.
Figure 28
Figure 28
Forest plot with relative risk ratios and 95% CIs from network meta‐analysis and pairwise analyses for breastfeeding at discharge.
Figure 29
Figure 29
Cumulative rankograms comparing each of the uterotonic drugs for breastfeeding at discharge.
Figure 30
Figure 30
Forest plot with relative risk ratios and 95% CIs from network meta‐analysis and pairwise analyses for nausea.
Figure 31
Figure 31
Cumulative rankograms comparing each of the uterotonic drugs for nausea.
Figure 32
Figure 32
Forest plot with relative risk ratios and 95% CIs from network meta‐analysis and pairwise analyses for vomiting.
Figure 33
Figure 33
Cumulative rankograms comparing each of the uterotonic drugs for vomiting.
Figure 34
Figure 34
Forest plot with relative risk ratios and 95% CIs from network meta‐analysis and pairwise analyses for hypertension.
Figure 35
Figure 35
Cumulative rankograms comparing each of the uterotonic drugs for hypertension.
Figure 36
Figure 36
Forest plot with relative risk ratios and 95% CIs from network meta‐analysis and pairwise analyses for headache.
Figure 37
Figure 37
Cumulative rankograms comparing each of the uterotonic drugs for headache.
Figure 38
Figure 38
Forest plot with relative risk ratios and 95% CIs from network meta‐analysis and pairwise analyses for fever.
Figure 39
Figure 39
Cumulative rankograms comparing each of the uterotonic drugs for fever.
Figure 40
Figure 40
Forest plot with relative risk ratios and 95% CIs from network meta‐analysis and pairwise analyses for shivering.
Figure 41
Figure 41
Cumulative rankograms comparing each of the uterotonic drugs for shivering.
Figure 42
Figure 42
Forest plot with relative risk ratios and 95% CIs from network meta‐analysis and pairwise analyses for tachycardia.
Figure 43
Figure 43
Cumulative rankograms comparing each of the uterotonic drugs for tachycardia.
Figure 44
Figure 44
Forest plot with relative risk ratios and 95% CIs from network meta‐analysis and pairwise analyses for hypotension.
Figure 45
Figure 45
Cumulative rankograms comparing each of the uterotonic drugs for hypotension.
Figure 46
Figure 46
Forest plot with relative risk ratios and 95% CIs from network meta‐analysis and pairwise analyses for abdominal pain.
Figure 47
Figure 47
Cumulative rankograms comparing each of the uterotonic drugs for abdominal pain.
Figure 48
Figure 48
Forest plot with relative risk ratios and 95% CIs from network meta‐analysis and pairwise analyses for prevention of PPH ≥ 500 mL by mode of birth (vaginal birth).
Figure 49
Figure 49
Cumulative rankograms comparing each of the uterotonic drugs for prevention of PPH ≥ 500 mL by mode of birth (vaginal birth).
Figure 50
Figure 50
Forest plot with relative risk ratios and 95% CIs from network meta‐analysis and pairwise analyses for prevention of PPH ≥ 500 mL by mode of birth (caesarean).
Figure 51
Figure 51
Cumulative rankograms comparing each of the uterotonic drugs for prevention of PPH ≥ 500 mL by mode of birth (caesarean).
Figure 52
Figure 52
Forest plot with relative risk ratios and 95% CIs from network meta‐analysis and pairwise analyses for prevention of PPH ≥ 500 mL by prior risk for PPH (low risk).
Figure 53
Figure 53
Cumulative rankograms comparing each of the uterotonic drugs for prevention of PPH ≥ 500 mL by prior risk for PPH (low risk).
Figure 54
Figure 54
Forest plot with relative risk ratios and 95% CIs from network meta‐analysis and pairwise analyses for prevention of PPH ≥ 500 mL by prior risk for PPH (high risk).
Figure 55
Figure 55
Cumulative rankograms comparing each of the uterotonic drugs for prevention of PPH ≥ 500 mL by prior risk for PPH (high risk).
Figure 56
Figure 56
Forest plot with relative risk ratios and 95% CIs from network meta‐analysis and pairwise analyses for prevention of PPH ≥ 500 mL by healthcare setting (hospital setting).
Figure 57
Figure 57
Cumulative rankograms comparing each of the uterotonic drugs for prevention of PPH ≥ 500 mL by healthcare setting (hospital setting).
Figure 58
Figure 58
Forest plot with relative risk ratios and 95% CIs from network meta‐analysis and pairwise analyses for prevention of PPH ≥ 500 mL by healthcare setting (community setting).
Figure 59
Figure 59
Cumulative rankograms comparing each of the uterotonic drugs for prevention of PPH ≥ 500 mL by healthcare setting (community setting).
Figure 60
Figure 60
Forest plot with relative risk ratios and 95% CIs from network meta‐analysis and pairwise analyses for prevention of PPH ≥ 500 mL restricted to misoprostol studies that use a low dose (less or equal to 500 mcg).
Figure 61
Figure 61
Cumulative rankograms comparing each of the uterotonic drugs for prevention of PPH ≥ 500 mL restricted to misoprostol studies that use a low dose (less or equal to 500 mcg).
Figure 62
Figure 62
Forest plot with relative risk ratios and 95% CIs from network meta‐analysis and pairwise analyses for prevention of PPH ≥ 500 mL restricted to misoprostol studies that use a high dose (600 mcg or more).
Figure 63
Figure 63
Cumulative rankograms comparing each of the uterotonic drugs for prevention of PPH ≥ 1000 mL restricted to misoprostol studies that use a high dose (600 mcg or more).
Figure 64
Figure 64
Forest plot with relative risk ratios and 95% CIs from network meta‐analysis and pairwise analyses for prevention of PPH ≥ 500 mL restricted to oxytocin studies that used an intramuscular or intravenous bolus of any dose.
Figure 65
Figure 65
Cumulative rankograms comparing each of the uterotonic drugs for prevention of PPH ≥ 500 mL restricted to oxytocin studies that used an intramuscular or intravenous bolus of any dose.
Figure 66
Figure 66
Forest plot with relative risk ratios and 95% CIs from network meta‐analysis and pairwise analyses for prevention of PPH ≥ 500 mL restricted to oxytocin studies that used an intravenous bolus plus an infusion of any dose.
Figure 67
Figure 67
Cumulative rankograms comparing each of the uterotonic drugs for prevention of PPH ≥ 500 mL restricted to oxytocin studies that used an intravenous bolus plus an infusion of any dose.
Figure 68
Figure 68
Forest plot with relative risk ratios and 95% CIs from network meta‐analysis and pairwise analyses for prevention of PPH ≥ 500 mL restricted to oxytocin studies that used an intravenous infusion only of any dose.
Figure 69
Figure 69
Cumulative rankograms comparing each of the uterotonic drugs for prevention of PPH ≥ 500 mL restricted to oxytocin studies that used an intravenous infusion only of any dose.
Figure 70
Figure 70
Forest plot with relative risk ratios and 95% CIs from network meta‐analysis and pairwise analyses for prevention of PPH ≥ 500 mL restricted to low risk of bias studies only.
Figure 71
Figure 71
Cumulative rankograms comparing each of the uterotonic drugs for prevention of PPH ≥ 500 mL restricted to low risk of bias studies only.
Figure 72
Figure 72
Forest plot with relative risk ratios and 95% CIs from network meta‐analysis and pairwise analyses for prevention of PPH ≥ 500 mL restricted to studies with funding source at low risk of bias (public or no funding).
Figure 73
Figure 73
Cumulative rankograms comparing each of the uterotonic drugs for prevention of PPH ≥ 500 mL restricted to studies with funding source at low risk of bias (public or no funding).
Figure 74
Figure 74
Forest plot with relative risk ratios and 95% CIs from network meta‐analysis and pairwise analyses for prevention of PPH ≥ 500 mL restricted to studies with an objective method of measuring blood loss.
Figure 75
Figure 75
Cumulative rankograms comparing each of the uterotonic drugs for prevention of PPH ≥ 500 mL restricted to studies with an objective method of measuring blood loss.
Figure 76
Figure 76
Forest plot with relative risk ratios and 95% CIs from network meta‐analysis and pairwise analyses for prevention of PPH ≥ 500 mL restricted to large studies (> 400 participants).
Figure 77
Figure 77
Cumulative rankograms comparing each of the uterotonic drugs for prevention of PPH ≥ 500 mL restricted to large studies (> 400 participants).
See this image and copyright information in PMC

Similar articles