Key result
XGBoost machine learning model predicts postoperative cardiac arrest in elderly patients with ~0.97 AUC.
Why the study?
In-hospital cardiac arrest is a rare but life-threatening complication in elderly surgical patients, and predictive models for this outcome are needed.
Does a machine learning predictive model integrating perioperative clinical and intraoperative variables accurately predict 30-day postoperative in-hospital cardiac arrest in elderly patients undergoing non-cardiac surgery?
Cohort (n=7,327)
Yes
Does a machine learning predictive model integrating perioperative clinical and intraoperative variables accurately predict 30-day postoperative in-hospital cardiac arrest in elderly patients undergoing non-cardiac surgery?
An interpretable XGBoost machine learning model incorporating intraoperative and early postoperative data successfully predicted 30-day in-hospital cardiac arrest in elderly non-cardiac surgery patients with excellent discrimination.
May support perioperative risk stratification; requires prospective validation before clinical adoption.
Background In-hospital cardiac arrest (IHCA) is a rare but life-threatening complication in elderly surgical patients. This study aimed to develop and validate a machine learning model for predicting IHCA and to identify key perioperative risk factors. Methods A multicenter retrospective cohort study was conducted among patients aged ≥65 years who underwent non-cardiac surgery at six tertiary hospitals between January 2020 and January 2025. Clinical, laboratory, and perioperative variables were collected. Five machine learning algorithms, including Extreme Gradient Boosting (XGBoost), Random Forest (RF), Support Vector Machine (SVM), Multilayer Perceptron (MLP), and K-Nearest Neighbor (KNN), were developed and compared. Model performance was evaluated using receiver operating characteristic (ROC) analysis, calibration, decision curve analysis, cross-validation, and external validation. SHAP analysis was applied to interpret the optimal model. Results Among the five models, XGBoost demonstrated the best predictive performance, with AUCs of 0.967 and 0.960 in the training and validation cohorts, respectively. It also achieved the lowest Brier score and the highest clinical net benefit. External validation confirmed excellent discrimination (AUC=0.971) and calibration. SHAP analysis identified advanced age, intraoperative tachycardia, intraoperative hypotension, intraoperative blood transfusion, intraoperative hypoxemia, elevated postoperative neutrophil-to-lymphocyte ratio, B-type natriuretic peptide, lactate, and D-dimer levels as the most important predictors of IHCA. Conclusion An interpretable XGBoost-based model demonstrated excellent performance for predicting postoperative cardiac arrest in elderly patients. The identified risk factors reflected perioperative vulnerability, intraoperative hemodynamic instability, and postoperative physiological deterioration.
No takes yet. Share an insight, caveat, or question.
Chen et al. (2026) conducted a cohort in In-hospital postoperative cardiac arrest (n=7,327). XGBoost machine learning model was evaluated on Prediction of 30-day postoperative in-hospital cardiac arrest. An interpretable XGBoost machine learning model accurately predicted 30-day postoperative in-hospital cardiac arrest in elderly surgical patients, achieving an AUC of 0.971 in external validation.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: