Vaginal progesterone compared with intramuscular 17-alpha-hydroxyprogesterone caproate for prevention of recurrent preterm birth in singleton gestations: a systematic review and meta-analysis

OBJECTIVE: Randomized trials have found benefits of both vaginal progesterone and 17-alphahydroxyprogesterone caproate in the prevention of recurrent preterm birth. A previous meta-analysis directly comparing the two was limited by low-quality evidence, and national and international society guidelines remain conflicting regarding progestin formulation recommended for prevention of recurrent preterm birth. The aim of this updated systematic review with meta-analysis was to evaluate the efficacy of vaginal progesterone compared with 17-alpha-hydroxyprogesterone caproate in the prevention of spontaneous preterm birth in patients with singleton gestations and previous spontaneous preterm birth. DATA SOURCES: Searches were performed in MEDLINE, Ovid, Scopus, ClinicalTrials.gov, the International Prospective Register of Systematic Reviews (PROSPERO), SciELO, Embase, and the Cochrane Central Register of Controlled Trials (CENTRAL) with the use of a combination of keywords and text words related to preterm birth, preterm delivery, singleton, cervical length, progesterone, progesto-gens, vaginal, 17-alpha-hydroxy-progesterone caproate, and intramuscular from inception of each database to September 2021. No restrictions for language or geographic location were applied. STUDY ELIGIBILITY CRITERIA: We included all randomized controlled trials of asymptomatic singleton gestations with previous spontaneous preterm birth that were randomized to prophylactic treatment with either vaginal progesterone (ie, intervention group) or intramuscular 17-alphahydroxyprogesterone caproate (ie, comparison group). Post hoc sensitivity analysis was performed for studies with low risk of bias and studies with protocol registration. METHODS: The primary outcome was preterm birth <34 weeks' gestation. The summary measures were reported as relative risks with 95% confidence intervals. RESULTS: Seven randomized controlled trials including 1910 patients were included in the meta analysis. Patients who received vaginal progesterone had a significantly lower rate of preterm birth at <34 weeks (14.7% vs 19.9%; relative risk, 0.74; 95% confidence interval, 0.57-0.96), preterm birth at <37 weeks (36.0% vs 46.6%; relative risk, 0.76; 95% confidence interval, 0.69-0.85), and preterm birth at <32 weeks of gestation (7.9% vs 13.6%; relative risk, 0.58; 95% confidence interval, 0.39-0.86), compared with women who received intramuscular 17-alpha-hydroxyprogesterone caproate. There were no significant differences in the rate of preterm birth at <28 weeks' gestation. Adverse drug reactions were significantly lower in the vaginal progesterone group than in the 17-alpha-hydroxyprogesterone caproate group (15.6% vs 22.2%; relative risk, 0.71; 95% confi-dence interval, 0.54-0.92). Perinatal mortality was lower in the vaginal progesterone group than in the 17-alpha-hydroxyprogesterone caproate group (2.2% vs 4.4%; relative risk, 0.51; 95% confi-dence interval, 0.25-1.01). In sensitivity analysis including trials rated with at least 4 Cochrane tools as of low risk of bias, 4 trials were included (N=575), and there was no longer a significant difference in preterm birth at <34 weeks' gestation between vaginal progesterone and 17-alpha-hydroxyprogesterone caproate (12.2% vs 13.9%; relative risk, 0.87; 95% confidence interval, 0.57-1.32). CONCLUSION: Overall, vaginal progesterone was superior to 17-alpha-hydroxyprogesterone caproate in the prevention of preterm birth at <34 weeks' gestation in singleton pregnancies with previous spontaneous preterm birth. Although sensitivity analysis of high-fidelity studies showed the same trend, findings were no longer statistically significant.

Automated summary

Vaginal progesterone is more effective than 17-alpha-hydroxyprogesterone caproate in preventing preterm birth at <34 weeks in singleton pregnancies with previous spontaneous preterm birth, but the evidence supporting this conclusion is no longer statistically significant in high-quality studies.