Back

Identification of Early Risk Factors for Mortality in Pediatric Veno-Arterial Extra Corporeal Membrane Oxygenation: The Patient Matters

Weinerman, B.; Kwon, S. b.; Alalqum, T.; Nametz, D.; Megjhani, M.; Clark, E.; Verner, C.; Cheung, E. W.; Park, S.

2024-10-18 pediatrics
10.1101/2024.10.17.24315712 medRxiv
Show abstract

ObjectivePediatric Veno-Arterial Extra Corporeal Membrane Oxygenation (VA ECMO) is a life saving technology associated with high mortality. A successful VA ECMO course requires attention to multiple aspects of patient care, including ECMO and patient parameters. Early, potentially modifiable, risk factors associated with patient mortality should be analyzed and adjusted for when assessing VA ECMO risk profiles. MethodRetrospective single center experience of pediatric patients requiring VA ECMO from January 2021 to October 2023. Laboratory and ECMO flow parameters were extracted from the patients record and analyzed. Risk factors were analyzed using a Cox proportion hazard regression Main ResultsThere were 45 patients studied. Overall survival was 51%. Upon uncorrected analysis there were no significant differences between the patients who survived and those who died. Utilizing a Cox proportion hazard regression, platelet count, fibrinogen level and creatine level were significant risk factors within the first twenty-four hours of a patients ECMO course. SignificanceAlthough we did not find a significant difference among ECMO flow parameters in this study, this work highlights that granular ECMO flow data can be incorporated to risk analysis profiles and potential modeling in pediatric VA ECMO. This study demonstrated, that when controlling for ECMO flow parameters, kidney dysfunction and clotting regulation remain key risk factors for pediatric VA ECMO mortality.

Matching journals

The top 4 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.