Source: Parry Smith WR, Papadopoulou A, Thomas E, et al. Uterotonic agents for first-line treatment of postpartum haemorrhage: a network meta-analysis. Cochrane Database Syst Rev. 2020;11:CD012754.
Study Population: 6 RCTs including women giving birth vaginally (n = 3674), and 1 RCT where women gave birth either vaginally or by caesarean (n = 64) in low resource settings
Efficacy Endpoints: Additional blood loss of at least 500 mL, death
Harm Endpoints: Transfusion or maternal mortality. Adverse effects including fever, hypothermia, nausea, vomiting, hypertension, headache, shivering, tachycardia, arrhythmia, diarrhea, abdominal pain
Narrative: Postpartum hemorrhage (PPH) may occur in 15% of women giving birth and is the leading cause of peripartum maternal death, with most cases occurring in low income countries.1, 2 PPH is commonly defined as greater than 500 mL of blood loss after birth. Uterine atony is the most common cause of PPH,3 and oxytocin is recommended as first-line medical therapy by the American College of Obstetricians and Gynecologists and the World Health Organization.3, 4 However, there are many effective (i.e. better than placebo) ‘uterotonic’ agents for the treatment of PPH,1 including misoprostol which may be used alone or in combination with oxytocin.5
The Cochrane Review summarized here aimed to compare the efficacy of different agents for PPH, and included randomized controlled trials (RCTs) or cluster randomized trials evaluating the benefits and harms of uterotonic agents in women with PPH after vaginal or caesarean birth.6 Trials were eligible if they compared systemically administered uterotonic agents of any dosage, route, or regimen.
The primary outcomes of the Cochrane review included blood loss of >500 mL after enrollment and a composite outcome of maternal death or severe morbidity (hysterectomy, organ dysfunction, transfer to higher level of care, coagulopathy, or shock). From this composite outcome, we report only maternal mortality, a patient-centered outcome reported consistently in the original trials.
The systematic review identified 7 RCTs (n = 3738) that met inclusion criteria. One trial included women giving birth vaginally or by cesarean section, while the others included only vaginal births. Agents evaluated included oxytocin (6 trial arms), misoprostol plus oxytocin (4 trials arms), misoprostol (3 trial arms), and fixed-dose oxytocin/ergometrine plus oxytocin infusion (1 trial arm). Data using this last regimen were limited, of low certainty, and showed unclear effects, therefore we have not summarized this comparison.
Two trials (n = 1787) found no difference in maternal mortality for misoprostol compared to oxytocin. These trials did suggest, however, misoprostol may increase blood transfusions (relative risk [RR]: 1.5; 95% confidence interval [CI]: 1.02-2.1; absolute risk increase: 2.3%; number needed to harm [NNH]: 43). Misoprostol also increased vomiting (RR: 2.5; 95% CI: 1.4-4.5; absolute risk increase: 2.9%; NNH: 34) and shivering (RR: 2.7; 95% CI: 2.3-3.2; absolute risk increase: 26.8%; NNH: 3).
Four trials (n = 1873) of misoprostol plus oxytocin versus oxytocin alone found no primary outcome benefit with the addition of misoprostol for maternal mortality, but did find an increase in adverse effects. These included fever (RR: 3.0; 95% CI: 2.6-3.6; absolute risk increase: 32.1%; NNH: 3) and vomiting (RR: 1.9; 95% CI: 1.2-3.0; absolute risk increase: 2.9%; NNH: 34).
Caveats: The quality of evidence for these analyses ranged from very low to high, with most data rated low or moderate certainty. No studies including injectable prostaglandins, ergometrine, or oxytocin/ergometrine as first line agents were available. Most subjects were women with a singleton term vaginal birth in a low-resource setting. Women with significant comorbidities were excluded from trials, limiting generalizability. There were also differences in dosing and route across interventions. Finally, blood loss can be challenging to quantify based on visual assessment, a measurement method used in some studies, which may have influenced the accuracy of this outcome.
Based on the existing evidence, misoprostol alone or in combination with oxytocin did not improve outcomes and is associated with more adverse effects, a finding supporting current recommendations from ACOG and WHO.3, 4 Therefore, we have assigned a color recommendation of Black (harms > benefits) for misoprostol. Further study in other settings on the benefits and harms of uterotonic agents is needed.
Author: Brit Long, MD; Michael Gottlieb, MD
Supervising Editors: Shahriar Zehtabchi, MD
Published/Updated: April 30, 2021
The title bar is color-coded with our overall recommendation.