P1.023: Association of Sociodemographic Factors with Mortality in Pediatric ECMO Patients

Background: Extracorporeal membrane oxygenation (ECMO) has been established as a life-sustaining therapy for pediatric patients experiencing severe respiratory and/or cardiac failure. Although ECMO can significantly improve survival in critically ill children, the impact of social determinants on ECMO outcomes remains understudied. Recent reports suggest sociodemographic factors such as race, insurance status, and median household income may contribute to differences in mortality among pediatric ECMO recipients. This study aims to evaluate the association between sociodemographic factors and mortality in children treated with ECMO at a tertiary care academic center.

Methods: All neonatal and pediatric patients (0–17 years) at a tertiary care academic center who underwent ECMO support for respiratory or cardiopulmonary failure between January 1, 2014, and January 1, 2024, were identified. Data were obtained through retrospective review, capturing sociodemographic variables including age, sex, race, insurance payer, and zip codes. Clinical parameters were collected, including ECMO indication, length of ECMO support, and hospital length of stay. Zip code, household population, and median income data were obtained from American Community Survey 2023 estimates. In‐hospital death before discharge/transfer was designated as mortality. Univariate and multivariate logistic regression models assessed associations between sociodemographic variables and mortality. All analyses were performed using Stata 14.2.

Results: The univariate analysis noted significantly lower odds of in-hospital mortality in patients transferred from outside the hospital (OR 0.3; 95% confidence interval [CI] 0.16-0.55; p< 0.001) compared to patients transferred within the hospital. Odds of mortality were significantly lower among patients with an ECMO indication of meconium aspiration (OR 0.18; CI 0.06-0.60; p=0.005) compared to infection. Mortality odds approached but did not reach significance for respiratory distress syndrome/persistent pulmonary hypertension (OR 0.38; CI 0.13-1.08; p=0.07) compared to patients with infection as ECMO indication. Infants ages 1 to 12 months showed higher odds of mortality compared to those younger than 1 month (OR 4.31; CI 1.70-10.92; p=0.002). Additionally, patients with congenital heart disease (OR 2.57; CI 1.06-6.24; p=0.037) had increased odds of mortality compared to those with infection as the primary indication for ECMO, while patients with congenital diaphragmatic hernia approached significance (OR 2.27; CI 0.90-5.75; p=0.08). Each additional day on ECMO was associated with an 11% increase in the odds of mortality. In the multivariate analysis, after adjusting for ECMO duration and ECMO indication, no significant association was seen between sociodemographic factors and odds of in-hospital mortality. Each 1-day change was significantly associated with 9% higher odds of mortality.

Conclusion: This analysis showed no sociodemographic disparity in odds of mortality after adjusting for other sociodemographic characteristics. After adjusting for covariates, higher odds of mortality were noted with increased length of ECMO therapy. These results can guide clinical decision making and improve resource allocation.