Abstract 4344150: Perinatal PM2.5 Exposure Varies by Race and Ethnicity but Does Not Mediate Disparities in Infant Cardiac Surgery Outcomes: A Causal Mediation Analysis

Introduction: Racial and ethnic disparities in congenital heart surgery outcomes are well documented and attributed to social determinants, but the full range of contributing factors remains unclear. Chronic exposure to particulate matter ≤2.5 µm diameter (PM2.5) is linked to adverse outcomes in adults. Given that minority communities are exposed to higher PM2.5 levels, we investigated how perinatal PM2.5 exposure mediates the relationship between race/ethnicity and infant congenital heart surgery outcomes. Hypothesis: Perinatal exposure to PM2.5 may be in the causal pathway between race/ethnicity and the days alive and out of healthcare (DAoH) among infants undergoing cardiac surgery (Image 1). Methods: We linked clinical registry data from the New York State (NYS) Congenital Heart Surgery Collaborative for Longitudinal Outcomes and Utilization of Resources with Medicaid claims and census tract-level daily PM2.5 concentrations from 2006-2017. Race/ethnicity was categorized as Hispanic, Non-Hispanic (NH) Asian, NH-Black, and NH-White. Our outcome was DAoH during the first two years of life (as % of 730 days). Mediator candidates included average exposure to PM2.5 during gestational weeks 0-8, 0-35, and postnatal period ages 0-1 and 0-2 years. We applied fractional logit and linear regression for each mediator, adjusting for clinical factors and surgical center, within a causal mediation framework. Analyses were then stratified by urbanicity. Results: Our cohort included 2,505 infants (33% Hispanic, 11% NH-Asian, 22% NH-Black, and 35% NH-White). Adjusted mean DAoH was 79.6% (95% CI 68.3-86.3) for Hispanic, 84.5% (70.3-87.3) for NH-Asian, 77.4% (66.6-86.4) for NH-Black, and 81.7% (70.0-86.3) for NH-White. In urban settings, NH-Black children experienced higher levels of PM2.5 (µg/m3) compared to NH-White children (0.44 µg/m3 [CI 0.25-0.64] at 0-35 gestational weeks, 0.44µg/m3[CI 0.27-0.61] age 0-1 year, and 0.32 µg/m3 [CI 0.16-0.48] age 0-2 years). PM2.5 exposure was not significantly associated with DAoH. Mediation analyses revealed no significant indirect effect. Conclusion: While differences in perinatal exposure to PM2.5 were observed by race/ethnicity among infants living in urban areas, our findings suggest that this environmental factor does not mediate racial and ethnic disparities in outcomes after congenital heart surgery in NYS. This highlights the need to explore other determinants of postoperative disparities in pediatric cardiac care.