Pharmacologic intervention for the management of retained placenta: a systematic review and meta-analysis of randomized trials

Background: Retained placenta affects 2% to 3.3% of all vaginal deliveries and is one of the leading causes of postpartum hemorrhage worldwide. Despite the prevalence of this condition, there is limited guidance on its management.

Objective: A systematic review and meta-analysis were performed to evaluate the efficacy of pharmacologic interventions for the management of retained placenta.

Study design: PubMed, ClinicalTrials.gov, Cochrane Library, Web of Science, and Scopus were searched for full-text publications in English. Search terms included "retained placenta" AND "treatment" OR "therapy" OR "disease management" OR "Pitocin" OR "misoprostol" OR "Cytotec" OR "dinoprostone" OR "nitroglycerin" OR "carbetocin" OR "ergotamine," with no restriction on publication dates. Only randomized controlled trials were included. The primary outcome was the need for manual extraction of the placenta or dilation and curettage. Reviewers evaluated the quality of included articles using the Cochrane Collaboration's tool for assessing the risk of bias. Pooled risk ratios were estimated based on random- and fixed-effects analyses. Interstudy heterogeneity was considered when I²≥50%.

Results: The literature search identified 29 randomized controlled trials that met the inclusion criteria (2682 subjects). The most commonly used agent across the studies was oxytocin administered via umbilical vein injection; there was high heterogeneity among these studies (I²=62%). Oxytocin was inferior to carbetocin (risk ratio, 1.61; 95% confidence interval, 1.03-2.52) and prostaglandins (risk ratio, 2.63; 95% confidence interval, 1.18-5.86) for the primary outcome. For oxytocin, prostaglandin agents, and nitroglycerin, there was a trend toward favoring the study drug for the primary outcome compared with control or placebo. Compared with placebo or control, estimated blood loss was lower if pharmacologic interventions were administered, with a mean difference of 121.5 mL (95% confidence interval, -185.7 to -52.3). There was no difference in postpartum hemorrhage or the need for blood transfusion between pharmacologic interventions and placebo or control.

Conclusion: Pooled estimates for oxytocin via umbilical vein injection, prostaglandin agents, and nitroglycerin performed favorably compared with placebo or control for the management of retained placenta. Carbetocin and prostaglandin agents were superior to oxytocin in reducing the need for manual extraction or dilation and curettage.