Timing of postnatal corticosteroid treatment for bronchopulmonary dysplasia and its effect on outcomes

Objective To determine the association of timing of steroid therapy for bronchopulmonary dysplasia (BPD) and outcomes. Methods Retrospective cohort study of preterm infants treated with low-dose dexamethasone for BPD. Infants treated with steroids at day of life (DOL) 14-28 (moderately late group) were compared to infants treated at DOL 29-42 (delayed group). Inverse probability of treatment weighting (IPTW) adjusted propensity scores were used to correct for potential confounders. The primary outcome of interest was postmenstrual age (PMA) at discharge. Results Fifty-five infants (25 with moderately late treatment; 30 with delayed treatment) were identified. The mean age at treatment was 23 days in the moderately late group and 35 days in the delayed group. At time of treatment, infants treated moderately late were more likely to be on high frequency ventilation (96% vs 47%, P < 0.0001) and had higher fraction of inspired oxygen (70.7 +/- 17.9% vs 56.2 +/- 18.4%, P = 0.005) compared to infants treated later. Despite being the sicker group, moderately late treated infants were discharged at an earlier corrected age compared to infants with delayed treatment (PMA 42.9 +/- 4.5 vs 47.5 +/- 8.3 weeks, IPTW adjusted P = 0.03). Moderately late treatment was also associated with fewer days on mechanical ventilation (46.0 +/- 19.0 days vs 77.4 +/- 67.0 days, IPTW adjusted P = 0.02) and fewer days on supplemental oxygen (114.3 +/- 40.8 days vs 149.8 +/- 57.0 days, IPTW adjusted P = 0.005). Conclusions Among preterm infants at high risk of BPD, delaying treatment with postnatal steroids is associated with comparatively worse short-term outcomes as compared to earlier treatment.