Effectiveness of progesterone, cerclage and pessary for preventing preterm birth in singleton pregnancies: a systematic review and network meta‐analysis
Journal Articles
Overview
Research
Identity
Additional Document Info
View All
Overview
abstract
BackgroundPreterm birth (PTB) is the leading cause of infant death, but it is unclear which intervention is best to prevent it.ObjectivesTo compare progesterone, cerclage and pessary, determine their relative effects and rank them.Search strategyWe searched Medline, EMBASE, CINAHL, Cochrane CENTRAL and Web of Science (to April 2016), without restrictions, and screened references of previous reviews.Selection criteriaWe included randomised trials of progesterone, cerclage or pessary for preventing PTB in women with singleton pregnancies at risk as defined by each study.Data collection and analysisWe extracted data by duplicate using a piloted form and performed Bayesian random‐effects network meta‐analyses and pairwise meta‐analyses. We rated evidence quality using GRADE, ranked interventions using SUCRA and calculated numbers needed to treat (NNT).Main resultsWe included 36 trials (9425 women; 25 low risk of bias trials). Progesterone ranked first or second for most outcomes, reducing PTB < 34 weeks [odds ratio (OR) 0.44; 95% credible interval (CrI) 0.22–0.79; NNT 9; low quality], <37 weeks (OR 0.58; 95% CrI 0.41–0.79; NNT 9; moderate quality), and neonatal death (OR 0.50; 95% CrI 0.28–0.85; NNT 35; high quality), compared with control, in women overall at risk. We found similar results in the subgroup with previous PTB, but only a reduction of PTB < 34 weeks in women with a short cervix. Pessary showed inconsistent benefit and cerclage did not reduce PTB < 37 or <34 weeks.ConclusionsProgesterone was the best intervention for preventing PTB in singleton pregnancies at risk, reducing PTB < 34 weeks, <37 weeks, neonatal demise and other sequelae.Tweetable abstractProgesterone was better than cerclage and pessary to prevent preterm birth, neonatal death and more in network meta‐analysis.
