Back

Pre-extracorporeal support impaired oxygen delivery as a predictor of neurologic outcomes in infants; insights into decision support

Rabinowitz, E.; Guerriero, R.; Bauer Huang, S.; Guilliams, K.; Payne, P.; Said, A. S.

2025-04-28 health informatics
10.1101/2025.04.26.25320858 medRxiv
Show abstract

ObjectivesThis study aims to evaluate the association between pre-Extracorporeal Membrane Oxygenation (ECMO) markers of impaired oxygen delivery, as quantified by the iDO2 predictive analytics platform, and neurologic outcomes in infants supported on ECMO. The goal is to determine whether these markers can inform decision-support systems for optimizing ECMO initiation timing. Materials & MethodsWe performed a single-center retrospective cohort study of infants <1 year supported on ECMO from 2013-2017, excluding cases with congenital diaphragmatic hernia or post-cardiac surgery ECMO. Data included demographics, clinical variables, and iDO2 estimates retroactively calculated in 120-minute intervals prior to ECMO initiation. Primary outcomes included mortality, EEG abnormalities, and head imaging findings; secondary outcomes included MRI abnormalities and Functional Status Scores (FSS). ResultsOf 219 patients, 47 met inclusion criteria. Median age and weight at ECMO initiation were 16 days [IQR 6-112] and 3.3 kg [IQR 2.8-4.8], with an overall mortality rate of 55%. Non-survivors had higher rates of congenital heart disease (77% vs. 42%, p=0.03) and pre-ECMO cardiac arrest (53% vs. 14%, p=0.006). Time spent above iDO2 thresholds of 25%, 50%, and 75% increased closer to ECMO initiation. Higher iDO2 dose correlated with adverse neurologic outcomes, including EEG abnormalities and abnormal imaging, and predicted poor composite functional outcomes (p<0.05). Discussion & ConclusionMarkers of impaired oxygen delivery, such as iDO2, may inform the development of decision-support systems to optimize ECMO timing, potentially improving neurologic outcomes. Further research is needed to validate these findings and develop decision-support systems for clinical practice.

Matching journals

The top 3 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.