Interpretable mortality prediction at VA-ECMO establishment in acute myocardial infarction: A multicenter study from the CSECLS registry.
CatBoost model predicted in-hospital mortality after venoarterial extracorporeal membrane oxygenation for acute myocardial infarction accurately
Interpretable mortality prediction at VA-ECMO establishment in acute myocardial infarction: A multicenter study from the CSECLS registry.
Despite the growing use of venoarterial extracorporeal membrane oxygenation (venoarterial extracorporeal membrane oxygenation) as a rescue therapy for acute myocardial infarction (acute myocardial infarction), in-hospital mortality remains high.
To develop and evaluate interpretable in-hospital mortality prediction models for venoarterial extracorporeal membrane oxygenation-supported acute myocardial infarction patients using multicenter data from the CSECLS registry.
We retrospectively analyzed 1,833 patients from 84 centers between 2016 and 2023.
higher scores mean better accuracy in predicting death before discharge
CatBoost was recommended for the primary W2 window (pooled internal-external cross-validation AUROC, 0.650).
In the secondary W1 analysis, random forest was selected and achieved an AUROC of 0.855 (95% CI, 0.794-0.912) in the same held-out center.
Calibration analysis revealed systematic underestimation of baseline risk and conservative probability estimates, highlighting the need for context-specific recalibration.
SHAP analysis identified invasive mechanical ventilation, norepinephrine dose, ECMO indication category, arterial H, mean arterial pressure, and pre-ECMO lactate as the leading predictors.
The CatBoost model for the designated primary W2 window provided an accurate and interpretable tool to support early risk assessment, shared decision-making, and precision management in high-risk cardiogenic shock populations.
Local recalibration and further multicenter validation are required before clinical use.