Early Predictive nomogram of Ventricular Fibrillation after Percutaneous Coronary Intervention in Acute Myocardial Infarction using LASSO-Logistic regression

Objective Ventricular fibrillation (VF) is a serious complication of acute myocardial infarction (AMI) that can occur after percutaneous coronary intervention (PCI). This study aimed to identify early predictors of VF and develop a nomogram to predict VF risk in this patient population. Methods This single-center retrospective cohort study included 162 AMI patients who underwent PCI at Beijing Friendship Hospital between 2018 and present. Patients were divided into VF and non-VF groups (1:3 matched) based on the occurrence of VF after PCI. Baseline demographics, medical history, laboratory results, medications, and in-hospital outcomes were compared between groups. LASSO regression analysis, followed by univariable and multivariable logistic regression, identified predictors of VF. A nomogram incorporating significant predictors was constructed and validated using ROC curve analysis, calibration curves, and the Hosmer-Lemeshow goodness-of-fit test. Results Significant differences in baseline characteristics (Table 1A), clinical data including coagulation parameters and complications (Table 1B), and in-hospital prognoses (Table 1C) were observed between VF and non-VF groups. LASSO regression analysis identified key predictors, and multivariable analysis confirmed diabetes, metabolic acidosis, anemia, hypokalemia, and ventricular premature beats as independent predictors of VF (Table 2). The nomogram incorporating these factors showed excellent discrimination (AUC = 0.942, 95% CI: 0.882-1.000) and good calibration (Hosmer-Lemeshow test p = 0.898). Conclusion This study identified several readily available clinical parameters as independent predictors of VF in AMI patients undergoing PCI. The developed nomogram based on these predictors showed promising predictive accuracy and calibration, potentially assisting clinicians in early risk stratification and guiding proactive management strategies to improve outcomes in this high-risk population.