Abstract
Objective: To determine the association between postnatal age (PNA) at first administration of systemic postnatal steroids (sPNS) for bronchopulmonary dysplasia (BPD) and mortality or significant neurodevelopmental impairment (sNDI) at 18−24 months corrected age (CA) in infants < 29 weeks’ gestation. Methods: Data from the Canadian Neonatal Network and Canadian Neonatal Follow-up Network databases were used to conduct this retrospective cohort study. Infants exposed to sPNS for BPD after the 1st week of age were included and categorized into 8 groups based on the postnatal week of the exposure. The primary outcome was a composite of mortality or sNDI. A multivariable logistic regression model adjusting for potential confounders was used to determine the association between the sPNS and ND outcomes. Results: Of the 10,448 eligible infants, follow-up data were available for 6200 (59.3%) infants. The proportion of infants at first sPNS administration was: 8%, 17.5%, 23.1%, 18.7%, 12.6%, 8.3%, 5.8%, and 6% in the 2nd, 3rd, 4th, 5th, 6th, 7th, 8−9th, and ≥10th week of PNA respectively. No significant association between the timing of sPNS administration and the composite outcome of mortality or sNDI was observed. The odds of sNDI and Bayley-III motor composite < 70 increased by 1.5% (95% CI 0.4, 2.9%) and 2.6% (95% CI 0.9, 4.4%), respectively, with each one-week delay in the age of initiation of sPNS. Conclusions: No significant association was observed between the composite outcome of mortality or sNDI and PNA of sPNS. Among survivors, each week’s delay in initiation of sPNS may increase the odds of sNDI and motor delay.