Development and Validation of a Machine Learning-based Explainable Predictive Model for Long-term net Adverse Clinical Events in Patients With High Bleeding Risk Undergoing Percutaneous Coronary Intervention: Results From a Prospective Cohort Study

Junyan Zhang; Yuting Lei; Ran Liu; Hongsen Zhao; Yuxiao Li; Minggang Zhou; Chen Li; Li Rao; Dapeng Jiang; Zhongxiu Chen; Yong He

doi:10.1097/JS9.0000000000002744

Development and Validation of a Machine Learning-based Explainable Predictive Model for Long-term net Adverse Clinical Events in Patients With High Bleeding Risk Undergoing Percutaneous Coronary Intervention: Results From a Prospective Cohort Study

Int J Surg. 2025 Jun 23. doi: 10.1097/JS9.0000000000002744. Online ahead of print.

Authors

Junyan Zhang¹, Yuting Lei², Ran Liu³, Hongsen Zhao³, Yuxiao Li^{1

4}, Minggang Zhou¹, Chen Li¹, Li Rao¹, Dapeng Jiang³, Zhongxiu Chen¹, Yong He¹

Affiliations

¹ Department of Cardiology, West China Hospital of Sichuan University, Chengdu, China.
² Integrated Care Management Center, West China Hospital, Sichuan University, Chengdu, China.
³ Information Center of West China Hospital, Sichuan University, Chengdu, China.
⁴ West China School of Nursing, Sichuan University, Chengdu, China.

PMID: 40549441
DOI: 10.1097/JS9.0000000000002744

Abstract

Background: Patients classified as having a high bleeding risk (HBR) and undergoing percutaneous coronary intervention (PCI) face a significantly greater incidence of net adverse clinical events (NACEs) than non-HBR patients do. Existing risk assessment models, such as the CRUSADE and TIMI scores, do not adequately address the unique risks faced by the HBR population. There is an urgent need for a precise and comprehensive predictive model tailored to PCI-HBR patients to guide clinical decision-making and improve patient outcomes.

Methods: This study aimed to develop a machine learning-based predictive model for long-term NACE in PCI-HBR patients. We utilized data from the Prognostic Analysis and an Appropriate Antiplatelet Strategy for Patients with Percutaneous Coronary Intervention and High Bleeding Risk (PPP-PCI) registry database. Feature selection and interpretation were performed via a SHapley Additive exPlanations (SHAP) model based on recursive feature elimination (RFE). Model construction and evaluation were conducted via four algorithms: logistic regression, random forest, gradient boosting, and XGBoost.

Results: A total of 1512 PCI-HBR patients were included in the study. The XGBoost model demonstrated the highest predictive performance, achieving an area under the receiver operating characteristic curve (AUC) of 0.85. The SHAP model identified 24 significant variables contributing to the prediction of NACE, including clinical parameters, laboratory findings, and echocardiographic data.

Conclusions: Our machine learning-based model offers a promising tool for predicting long-term NACE in PCI-HBR patients. The model's high predictive accuracy and interpretability have the potential to enhance clinical decision-making and improve patient care. Further validation in larger, diverse populations is warranted to confirm these findings.

Keywords: high bleeding risk; machine learning; percutaneous coronary intervention; recursive feature elimination (RFE); shapley additive explanations (SHAP).