23 Assessing the effects of introducing prophylactic hydrocortisone in extremely low gestational age infants in a tertiary Canadian NICU

Abstract

Abstract Background Bronchopulmonary dysplasia (BPD) is a common complication of preterm birth. A large trial and a meta-analysis support the early use of physiologic doses of hydrocortisone to reduce the incidence of death or moderate/severe BPD. In 2019, this practice was introduced for all preterm infants <28 weeks gestational age (GA) at our university-affiliated tertiary care NICU. Objectives To determine the effect of prophylactic hydrocortisone on the incidence of survival with mild or no BPD in neonates born <28 weeks GA. Secondary objectives were to compare safety outcomes related to prophylactic hydrocortisone use: late onset sepsis, gastrointestinal perforation, severe intraventricular hemorrhage (IVH) and use of postnatal dexamethasone. Design/Methods We performed a retrospective cohort study of infants born <28 weeks GA admitted to our NICU during 2 epochs, before and after the implementation of prophylactic hydrocortisone. The primary objective was evaluated using logistic regression adjusted for GA. Safety outcomes were compared using chi-square or Fisher’s exact test. Results Epoch 1 included 76 infants and epoch 2 included 71 infants, of whom 63 (88.7%) received prophylactic hydrocortisone. Infants in the two epochs were similar in terms of median (IQR) GA 26.5 (25.0, 27.2) vs 26.7 (25.1, 27.1) weeks GA, mean (SD) weight at admission (882g +/- 212 vs 841g +/- 176), APGAR at 5 minutes, use of antenatal steroids and/or surfactant, and need for intubation at delivery (all p>0.5). Three infants were transferred to another hospital and their BPD status could not be determined. Adjusting for GA, there was a statistically non-significant signal for increased survival with mild or no BPD in epoch 2 (adjusted odds ratio 1.68, 95% CI 0.80, 3.59, p= 0.17) compared to epoch 1. The two epochs had similar incidence of late onset sepsis, gastrointestinal perforation and need for postnatal dexamethasone, however, in epoch 2, fewer infants developed grade 3/4 IVH on head ultrasound within the first two weeks of life (7/71, 9.9%) compared to epoch 1 (21/76, 27.6%, p= 0.01). Conclusion Prophylactic hydrocortisone may be associated with increased survival with mild or no BPD. Safety outcomes in both epochs were similar, but with a significantly decreased incidence of grade 3/4 IVH in the prophylactic hydrocortisone group. Larger cohort studies are required before a clear recommendation for prophylactic hydrocortisone can be made.