529: The effects of antenatal corticosteroids in twin pregnancies complicated by preterm birth

Abstract

Data regarding the effects of antenatal corticosteroids (ACS) in twins are limited due to the insufficient number of twin pregnancies enrolled in randomized controlled trials on ACS. Furthermore, the interpretation of available data is limited by the fact that the administration of ACS-to-delivery interval is greater than 7 days in a large proportion of twins, a factor that been shown to affect the efficacy of ACS and has not been controlled for in some of the previous studies. The objective of the current study was to assess the effect of ACS on neonatal outcomes in women with twin pregnancies who completed a course of ACS 1-7 days prior to delivery. In this retrospective cohort study, data on twin infants born between 24+0 and 33+6 weeks and admitted to tertiary neonatal units in Canada during 2010-2013 were obtained from the Canadian Neonatal Network. Neonatal outcomes were compared between twins who received ACS between 24 hours and 7 days prior to delivery (twins-ACS group) and those who did not receive ACS (twins-control group). 1) Out of 3,024 twin neonates who met the inclusion criteria, 1382 (69.5%) received a complete course of ACS 1-7 days prior to delivery and were compared to 607 (30.5%) who did not receive ACS. 2) Neonates in the twins-ACS group had a lower rate of mortality (2% vs. 7%, p<0.001) and grade 3-4 intraventricular hemorrhage (IVH) (7% vs. 12%, p=0.01). 3) Following adjustment for confounding factors including gestational age at delivery, hypertension, outborn status, parity, small for gestational age, and mode of delivery, administration of ACS in twins was associated with a lower odds of neonatal mortality, but was not associated with any of the neonatal morbidity outcomes (Table). 4) Findings were similar for the subgroup of twins delivered at less than 30+0 weeks of gestation. Administration of ACS 1-7 days prior to delivery in twin pregnancies is important as it is associated with a lower odds of neonatal mortality, and the magnitude of effect is similar to that previously observed in singletons.