WOFAPS 2025 8th World Congress of Pediatric Surgery

Mingwei Sun, Ru Lin

Children’s Hospital, Zhejiang University School of Medicine, National Clinical Research Center for Child Health, National Children’s Regional Medical Center

Objectives

ECMO support for pediatric refractory cardiogenic shock is associated with high mortality and poses significant clinical challenges. Leveraging data from the Chinese Pediatric Extracorporeal Life Support (CP-ECLS) Registry, this study established a predictive model to identify early warning factors for in-hospital mortality and assist clinical decision-making.

Methods

This multicenter retrospective study utilized data from the CP-ECLS Registry to analyze children receiving VA-ECMO for RCS between January 2017 and December 2023. Multivariate Cox regression identified early warning factors for in-hospital mortality, while a prognostic model equation by integrating the weighted contributions of risk factors. A time-dependent risk stratification assessment tool: Pediatric ECMO Cardiogenic Shock Outcome Score（PECSOS）was developed to visualize the prognostic model for clinical application.

Results

A total of 1,113 eligible children receiving ECMO between 2017-2023 were identified from the CP-ECLS registry, with 666 (59%) surviving to discharge.Through integration of Cox regression analysis and variable importance ranking to develop a model equation. A risk score cutoff of 1.74 was selected in the model equation to stratify the dataset into high-risk and low-risk groups for in-hospital mortality. Total dataset underwent stratified random sampling (6:4 ratio) yielding into training set and test set. The prognostic model developed on the training set demonstrated satisfactory performance upon external validation in the test cohort (Training set: days 3, 7, and 14 were 0.816, 0.807, and 0.800; Test set: days 3, 7, and 14 were 0.797, 0.766, and 0.759). To enhance clinical utility, we developed the PECSOS (Pediatric ECMO-Cardiogenic Shock Outcome Stratification) risk stratification schema. The high-risk cohort necessitates intensified hemodynamic monitoring and aggressive therapeutic escalation to mitigate adverse outcomes.

Conclusion

The risk stratification model and PECSOS in this study demonstrate potential to optimize clinical decision-making for pediatric RCS requiring ECMO support, with implications for improving prognosis and enhancing healthcare resource allocation efficiency.