A Comparison of Vaginal and Intramuscular Progesterone for the Prevention of Recurrent Preterm Birth

Objective To examine whether vaginal progesterone is noninferior to 17-α hydroxyprogesterone caproate (17OHP-C) in the prevention of recurrent preterm birth (PTB). Study Design This retrospective cohort study included singleton pregnancies among women with a history of spontaneous PTB who received prenatal care at a single tertiary center from 2011 to 2016. Pregnancies were excluded if progesterone was not initiated prior to 24 weeks or the fetus had a major congenital anomaly. The primary outcome was PTB <37 weeks. A priori, noninferiority was to be established if the upper bound of the adjusted two-sided 90% confidence interval (CI) for the difference in PTB fell below 9%. Inverse probability of treatment weighting (IPTW) was used to carefully control for confounding associated with choice of treatment and PTB. Adjusted differences in PTB proportions were estimated via IPTW regression, with standard errors adjustment for multiple pregnancies per woman. Secondary outcomes included PTB <34 and <28 weeks, spontaneous PTB, neonatal intensive care unit admission, and gestational age at delivery. Results Among 858 pregnancies, 41% (n = 353) received vaginal progesterone and 59% (n = 505) were given 17OHP-C. Vaginal progesterone use was more common later in the study period, and among women who established prenatal care later, had prior PTBs at later gestational ages, and whose race/ethnicity was neither non-Hispanic white nor non-Hispanic black. Vaginal progesterone did not meet noninferiority criteria compared with 17-OHPC in examining PTB <37 weeks, with an IPTW adjusted difference of 3.4% (90% CI: −3.5, 10.3). For secondary outcomes, IPTW adjusted differences between treatment groups were generally small and CIs were wide. Conclusion We could not conclude noninferiority of vaginal progesterone to 17OHP-C; however, women and providers may be willing to accept a larger difference (>9%) when considering the cost and availability of vaginal progesterone versus 17OHP-C. A well-designed randomized trial is needed. Key Points