Does progesterone prophylaxis to prevent preterm labour improve outcome? A randomised double-blind placebo-controlled trial (OPPTIMUM)
Background Progesterone prophylaxis is widely used to prevent preterm birth but is not licensed and there is little information on long-term outcome. Objective To determine the effect of progesterone prophylaxis in women at high risk of preterm birth on obstetric, neonatal and childhood outcomes. Design Double-blind, randomised placebo-controlled trial. Setting Obstetric units in the UK and Europe between February 2009 and April 2013. Participants Women with a singleton pregnancy who are at high risk of preterm birth because of either a positive fibronectin test or a negative fibronectin test, and either previous spontaneous birth at ≤ 34 weeks+0 of gestation or a cervical length of ≤ 25 mm. Interventions Fibronectin test at 18+0 to 23+0 weeks of pregnancy to determine risk of preterm birth. Eligible women were allocated (using a web-based randomisation portal) to 200 mg of progesterone or placebo, taken vaginally daily from 22+0 to 24+0 until 34+0 weeks’ gestation. Participants, caregivers and those assessing the outcomes were blinded to group assignment until data collection was complete. Main outcome measures There were three primary outcomes, as follows: (1) obstetric – fetal death or delivery before 34+0 weeks’ gestation; (2) neonatal – a composite of death, brain injury on ultrasound scan (according to specific criteria in the protocol) and bronchopulmonary dysplasia; and (3) childhood – the Bayley-III cognitive composite score at 22–26 months of age. Results In total, 96 out of 600 (16%) women in the progesterone group and 108 out of 597 (18%) women in the placebo group had the primary obstetric outcome [odds ratio (OR) 0.86, 95% confidence interval (CI) 0.61 to 1.22]. Forty-six out of 589 (8%) babies of women in the progesterone group and 62 out of 587 (11%) babies of women in the placebo group experienced the primary neonatal outcome [OR 0.72, 95% CI 0.44 to 1.17]. The mean Bayley-III cognitive composite score of the children at 2 years of age was 97.3 points [standard deviation (SD) 17.9 points; n = 430] in the progesterone group and 97.7 points (SD 17.5 points; n = 439) in the placebo group (difference in means –0.48, 95% CI –2.77 to 1.81). Limitations Overall compliance with the intervention was 69%. Harms There were no major harms, although there was a trend of more deaths from trial entry to 2 years in the progesterone group (20/600) than in the placebo group (16/598) (OR 1.26, 95% CI 0.65 to 2.42). Conclusions In this study, progesterone had no significant beneficial or harmful effects on the primary obstetric, neonatal or childhood outcomes.The OPPTIMUM trial is now complete. We intend to participate in a comprehensive individual patient-level data meta-analysis examining women with a singleton pregnancy with a variety of risk factors for preterm birth. Trial registration Current Controlled Trials ISRCTN14568373. Funding This trial was funded by the Medical Research Council (MRC) and managed by the National Institute for Health Research (NIHR) on behalf of the MRC–NIHR partnership.