293 - Race/Ethnicity and the Trajectory of Non-Invasive Respiratory Support Among Infants with Evolving Bronchopulmonary Dysplasia

Non-invasive ventilation (NIV) has become the predominant form of respiratory support in premature infants. Preterm infants with evolving bronchopulmonary dysplasia (BPD) often require long NIV courses that may be influenced by clinician variation, parental preference, and different risk for BPD. We hypothesize the management of infants on NIV are affected by social determinants.

Objective: To estimate the effect of maternal race/ethnicity on the weaning of non-invasive respiratory support for infants with evolving BPD born < 32 weeks gestation.

Design/Methods: We identified preterm infants < 32 weeks gestation using our center’s data in the Children's Hospital Neonatal Consortium from 2017-2022. Infants receiving respiratory support at 32 weeks post-menstrual age were included. The main outcome was the time from 28 days until infants were weaned to room-air.  Those who died, were transferred to other institutions, received tracheostomy, or reached 180 days of age were censored. Maternal race/ethnicity was categorized as black, white, Hispanic, and other. Differences in the main outcome were assessed via Cox regression by racial/ethnic and patient characteristics.

Results: There were 329 patients included (median gestation = 28 weeks [25^th-75^th %ile; 25, 30]; birthweight = 1.03 kg [0.73, 1.43]; age at referral = 35 days [1, 75]). Mortality was significant (12%), and SGA < 10^th centile was prevalent (16%). Each infant underwent a mean of 5 [2,7] changes in respiratory support between mechanical ventilation (mean duration 20.4 days), non-invasive positive pressure (12.7 days), nasal cannula (10.2 days) prior to reaching room-air. Regression analyses demonstrated that SGA was related to a decreased time to room-air in unadjusted (SGA: 84 days; non-SGA 113 days, p< 0.001) and multivariable analyses (adjusted hazard ratio (aHR) = 1.71, 95% CI=1.07.2.25, p=0.025) after adjusting for mode of support at 28 days of life, gestational age, and maternal race/ethnicity. Black infants who were SGA demonstrated a trend toward the shortest time-to-room air (aHR 2.25, p=0.1).

Conclusion(s):

SGA < 10^th centile was an independent risk factor for time-to-achieve room-air in infants with evolving BPD. We did not find evidence of different weaning outcomes by maternal racial/ethnic backgrounds. However, this framework may be useful in identifying variants in future diagnostic and management strategies.