394: Timing of delivery in women with vasa previa: A decision analysis


      Prior studies have suggested that the optimal timing for delivery in pregnancies with vasa previa is in the late preterm period. However, these recommendations are based upon evidence from small studies. The purpose of the current study is to determine the optimal gestational age (GA) at which to deliver pregnancies complicated by vasa previa and examine how varying baseline assumptions would affect this decision.

      Study Design

      A decision-analytic model was built using TreeAge software to compare the outcomes of strategies of planned delivery at 32-37 weeks’ gestation in a theoretical cohort of 10,000 women diagnosed with vasa previa. Strategies involving expectant management until a later GA accounted for the risks of premature rupture of membranes (PROM), bleeding following PROM, spontaneous labor, and stillbirth during each successive week. GA- associated risks of neonatal complications included neonatal death and cerebral palsy and considered the reduction in risk afforded by antenatal corticosteroids. The risk of intrapartum death according to time from PROM or onset of labor to delivery was also incorporated. Probabilities and utilities were derived from the literature and total quality-adjusted life years (QALYs) were calculated. Univariate sensitivity analyses were used to vary model inputs to investigate the robustness of our baseline assumptions. We assumed a 28.0% baseline rate of bleeding following PROM.


      In our theoretical cohort of 10,000 women, delivery at 36 weeks maximized maternal and neonatal QALYs. Compared to 34 weeks’ gestation, delivery at 36 weeks would result in 47.3 fewer children with cerebral palsy, but 18.9 more stillbirths (Table 1). Univariate sensitivity analysis found that delivery at 36 weeks remained the optimal strategy until the probability of bleeding with PROM fell below 0.07, at which point 37 weeks became optimal (Figure 1). Monte Carlo Analysis demonstrated that when variation was incorporated into the model, delivery at 36 weeks was the dominant strategy 79% of the time.


      When weighing the risks of intrapartum mortality against GA-associated neonatal complications, the optimal time for delivery in women with vasa previa is at 36 weeks.
      Figure thumbnail fx1
      Figure thumbnail fx2