Association of Vaginal Progesterone Treatment With Prevention of Recurrent Preterm Birth

IMPORTANCE: Preterm birth (PTB) is the leading cause of infant morbidity and mortality worldwide. It has been suggested that vaginal progesterone (VP) treatment may reduce the recurrence of PTB. OBJECTIVE: To evaluate the association of VP treatment with prevention of recurrent PTB among patients with a singleton pregnancy. DESIGN, SETTING, AND PARTICIPANTS: This prospective, observational cohort study, set in a public health care system for inner-city pregnant patients, enrolled patients with prior spontaneous PTB (gestational age, ≤35 weeks) receiving VP from May 15, 2017, to May 7, 2019. Patients who delivered between 1998 and 2011 served as a referent cohort matched 3:1 for obesity, race and ethnicity, and individual specific preterm birth history. Statistical analysis was performed from August 19, 2021, to September 2, 2022. EXPOSURE: Patients received 90 mg of vaginal progesterone, 8%, nightly, initiated between 16 weeks and 0 days and 20 weeks and 6 days of pregnancy until 36 weeks and 6 days of pregnancy or delivery. MAIN OUTCOMES AND MEASURES: The primary outcome was overall rate of recurrent PTB at 35 weeks or less of patients given VP compared with the 3:1 matched untreated historical controls. Secondary outcomes included assessment of PTB according to adherence (≥80% completing scheduled doses), duration of pregnancy relative to index gestational age, progesterone blood levels, and outcomes for those who declined VP. RESULTS: A total of 417 patients (mean [SD] age, 30.4 [5.9] years; 64 Black patients [15.3%]; 272 [65.2%] with a body mass index of ≥30) received VP and were matched with 1251 controls (mean [SD] age, 28.8 [5.7] years; 192 Black patients [15.3%]; 816 [65.2%] with a body mass index of ≥30). The overall rate of recurrent PTB was 24.0% (100 of 417; 95% CI, 20.0%-28.4%) for the VP cohort compared with 16.8% (1394 of 8278) expected in the matched historical controls. Adherence was not associated with lower rates of recurrent PTB compared with nonadherence (odds ratio, 0.87 [95% CI, 0.51-1.41]). The mean difference between historical matched controls and those using VP was 0.2 weeks (95% CI, −1.4 to 1.0 weeks) without improvement in the interval of recurrent PTB after the implementation of VP (P = .73). Progesterone blood levels for patients who were adherent compared with those who were nonadherent were not significantly different at either 24 or 32 weeks (24 weeks: 99 ng/mL [95% CI, 85-121 ng/mL] vs 104 ng/mL [95% CI, 89-125 ng/mL]; P = .16; 32 weeks: 200 ng/mL [95% CI, 171-242 ng/mL] vs 196 ng/mL [95% CI, 155-271 ng/mL]; P = .69). CONCLUSIONS AND RELEVANCE: This cohort study of patients with a current singleton pregnancy suggests that VP was not associated with a reduction in recurrent PTB.