Objectives To examine the effects of antenatal steroids, tocolysis, and delivery mode on extremely preterm delivery; to identify other potential antecedents through a hypothesis-generating analysis.
Design National cohort study.
Participants Singleton pregnancies delivering before 27 weeks gestation during 2006.
Main outcome measures Survival at birth (heart rate > 100 at 5 minutes) and at hospital discharge.
Results 1723 births were identified for inclusion. Antenatal steroids were associated with improvements in live birth (adjusted odds ratio (aOR) for any steroids: 1.65, 95% CI: 1.20–2.27) and survival to discharge (partial course of steroids: aOR 1.80, (1.25–2.59); complete course: aOR 2.38, (1.69–3.34)). Tocolysis was positively associated with both: aORs 1.38 (1.04-1.83) and 1.32 (1.04-1.68) for survival at birth and discharge, respectively. Caesarean delivery did not affect either outcome after adjustment.
In the hypothesis-generating analysis, gestational age had the strongest influence (aORs per week: 2.08 (1.86–2.34) for live birth; aOR 2.89 (2.56–3.28) for survival to discharge). Male sex reduced survival to discharge (aOR 0.64 (0.50–0.82)), as did non-cephalic presentation (aOR 0.59 (0.46–0.75)), in contrast with an aOR of 1.99 (1.17–3.43) for cervical cerclage and of 1.76 (1.27–2.45) for a spontaneous onset of labour.
Discussion Our results confirm the efficacy of antenatal steroids and tocolysis for promoting survival of extremely premature babies. Fetal sex, presentation, presence of a cervical suture, and a spontaneous onset of labour potentially provide extra information concerning the odds of survival over and above gestational age; these factors warrant further investigation and may influence management decisions and counselling.