Establishment of a Prediction Model for Adverse Cardiovascular Events after PCI in Acute Myocardial Infarction Based on the Decision Tree Algorithm

YOU Zhuozhi; WU Shanshan; ZENG Xiaoyan; YANG Nan

doi:10.12102/j.issn.1672-1349.2025.12.012

Abstract：Objective: To construct a model for predicting the risk of adverse cardiovascular events after percutaneous coronary intervention (PCI) in patients with acute myocardial infarction based on decision tree algorithm. Methods: A total of 230 patients with acute myocardial infarction who underwent PCI at the 910 Hospital of the Joint Logistics Support Force from January 2018 to June 2023 were selected as the study subjects. Clinical data were collected, and the patients were divided into the occurrence group (64 cases) and non-occurrence group (166 cases) according to whether they developed adverse cardiovascular events. Univariate and multivariate logistic regression analyses were used to identify the influencing factors of adverse cardiovascular events after PCI in patients with acute myocardial infarction, and a predictive model was established based on the decision tree algorithm. Results: The incidence of adverse cardiovascular events after PCI in patients with acute myocardial infarction was 27.8% (64/230). Univariate analysis showed statistically significant differences between the two groups in the time from onset to PCI, left ventricular ejection fraction, number of lesion vessels, low-density lipoprotein cholesterol (LDL-C), and troponin I (TnI) (P < 0.05). Multivariate logistic regression analysis showed that longer time from onset to PCI, lower left ventricular ejection fraction, multi-vessel lesions, higher LDL-C, and higher TnI were independent risk factors for adverse cardiovascular events after PCI in patients with acute myocardial infarction (P < 0.05). A decision tree model was established based on these factors. The model calculated the predictive importance of TnI, left ventricular ejection fraction, number of lesion vessels, time from onset to PCI, and LDL-C as 0.55, 0.24, 0.13, 0.06, and 0.03, respectively. Model validation results showed an area under the curve (AUC) of 0.811 [95% CI (0.779, 0.843)]. Conclusion: Risk factors for adverse cardiovascular events after PCI in patients with acute myocardial infarction include longer time from onset to PCI, lower left ventricular ejection fraction, multi-vessel lesions, higher LDL-C, and higher TnI. The decision tree model constructed based on these factors has good predictive ability for the risk of adverse cardiovascular events after PCI in patients with acute myocardial infarction.