Objective: To assess the impact of antenatal corticosteroids (ACS) on neonatal mortality, cerebral lesions and 5-year neurodevelopmental outcome of infants born at 24-27 and 28-32 weeks of gestational age (GA).
Design: Observational population-based study including all births at GAs between 22 and 32 weeks in 1997 in nine regions of France. Survivors were assessed at the age of 5 years.
Sample and methods: The population enrolled in the follow up comprised 2323 infants; there were 23 deaths before age 5 years and outcome at 5 years was available for up to 1781 subjects. Two GA subgroups (24-27 and 28-32 weeks of GA) were analysed separately. Propensity scores were used to reduce bias in the estimation of the association between ACS treatment and outcomes.
Main outcome measures: Neonatal death, neonatal white matter injury, cerebral palsy, mental processing composite (MPC) of the Kaufman Assessment Battery for Children test and behavioural difficulties at 5 years.
Results: In the 28- to 32-week GA subgroup, there was a significant association between ACS and a decreased risk of both neonatal death (OR = 0.61 [0.41-0.91]) and white matter injury (OR = 0.60 [0.46-0.79]) but only a nonsignificant trend for improved 5-year outcome (cerebral palsy, MPC < 70). In the 24- to 27-week GA subgroup, ACS was associated with a significant decrease risk of neonatal death (OR = 0.43 [0.27-0.68]) but there was only a trend for a lower risk of white matter injury and no beneficial impact on outcome at 5 years. Limiting the analysis to only those who received complete courses of ACS did not modify the results.
Conclusion: The study shows that ACS therapy greatly increases the survival of very preterm infants, including the most immature, but there is little evidence that ACS affects long-term neurodevelopmental and behavioural outcome in 28- to 32-week survivors, and none in <28-week survivors.