Cervical pessaries to prevent preterm birth in women with a multiple pregnancy: a per-protocol analysis of a randomized clinical trial.

INTRODUCTION We recently showed that a cervical pessary prevents preterm birth and reduces poor neonatal outcomes in women with a twin pregnancy and a short cervix (<38 mm). The objective of this study was to evaluate the full potential treatment effect of the pessary in the whole group and in women with a short cervix. MATERIAL AND METHODS We performed a per-protocol analysis of a multicenter randomized controlled trial (ProTWIN trial, NTR1858) where we excluded women who were allocated to the pessary but never had it placed. Women who had the pessary removed before 36 gestational weeks and did not deliver within 7 days after removal, were excluded. Analyses were performed on all women and in those with a cervical length <38 mm. RESULTS In 23 (6%) women the pessary was not placed. In women with a cervical length <38 mm (25th percentile) the pessary reduced poor perinatal outcome (relative risk 0.32, 95% confidence interval 0.13-0.78) and birth at <32 weeks (relative risk 0.41, 95% confidence interval 0.20-0.87). After excluding 47 (12%) women, the time to delivery was longer in the pessary group than in the control group (whole group: hazard ratio 0.68, 95% confidence interval 0.55-0.82, cervical length <38 mm: hazard ratio 0.35, 95% confidence interval 0.22-0.57). CONCLUSIONS The analysis confirms the principal findings of the intention-to-treat analysis. Time to delivery was longer in the pessary group than in the control group when censored data were used. This implies the pessary should not be removed until labor is evident.