350 - Prolonged Hyperoxia is an Independent Predictor for Severe Neurodevelopmental Impairment and Mortality in Extremely Premature Neonates

Extremely premature neonates born < 28 weeks’ gestation are susceptible to oxidative stress injury due to free oxygen radical damage coupled with immature antioxidant defenses. These factors contribute to impaired lung development and bronchopulmonary dysplasia (BPD). Experimental studies have demonstrated that hyperoxia may also disrupt brain development in premature infants. Few human studies have investigated the effects of prolonged hyperoxia on neurodevelopmental outcomes and survival in this population.

Objective:

To investigate the hypothesis that prolonged exposure to hyperoxia is an independent risk factor for severe neurodevelopmental impairment (NDI) or death in extremely premature neonates.

Design/Methods:

Retrospective cohort study of infants born <27 gestational age (GA) at a single, academic quaternary hospital from January 2008 to December 2018. Prolonged hyperoxia was defined as requiring fraction of inspired oxygen (FiO₂) >0.4 delivered via positive pressure ventilation for >72 hours aggregated or continuous. Neurodevelopmental assessment was conducted at 22-26 months corrected age. Severe NDI was defined as Bayley III cognitive or motor composite scores < 70, Gross Motor Function Classification System (GMFCS) level 4-5, bilateral blindness, or severe hearing impairment. Multiple logistic regression model was used to assess the association of prolonged hyperoxia and the primary composite outcome of severe NDI or death adjusting for co-morbidities: intraventricular hemorrhage, necrotizing enterocolitis, and BPD.

Results:

Of the 635 patients eligible for the study, follow-up outcomes were available for 546 (86.0%). Of these, 327 (59.9%) had exposure to prolonged hyperoxia, 100 (18.3%) had severe NDI, and 107 (16.8%) died. Prolonged hyperoxia in the extremely premature population was associated with increased odds of the composite outcomes of severe NDI or death (adjusted OR 2.29, 95% CI 1.39 to 3.81). When the components of the primary outcome were analyzed separately, prolonged hyperoxia was not associated with severe NDI among survivors (adjusted OR 1.61, 95% CI 0.89 to 2.95) but was associated with death (adjusted OR 3.34, 95% CI 1.66 to 6.92).

Conclusion(s):

Oxygen is a critical therapy in respiratory management for extremely premature infants. While oxidative stress has been shown to harm brain development in experimental studies, its long-term effects on neurodevelopment have not been extensively studied in children. In our study population, prolonged exposure to hyperoxia was not associated with severe NDI among survivors but was associated with increased odds of death.