Early Prediction Model for In-Hospital all-cause Death in Patients with Acute ST-Elevation Myocardial Infarction After Primary Coronary Artery Stenting

Abstract Background To establish an early prediction model for in-hospital all-cause death in patients with acute ST-segment elevation myocardial infarction (STEMI) after primary coronary artery stenting. Methods This retrospective study analyzed 3,916 STEMI patients undergoing primary coronary stenting within 24 hours of symptom onset at TEDA International Cardiovascular Hospital (2014–2022). We collected demographic, clinical, and procedural data, along with 48-hour laboratory results (including echocardiography and Holter monitoring). The primary outcome was in-hospital all-cause mortality. Eighty clinical parameters were compared between survivors and non-survivors to identify risk factors and develop an early prediction model. Result In a cohort of 3916 patients, 54 experienced in-hospital all-cause death. Comparison of 80 clinical variables between groups, followed by univariate logistic regression and least absolute shrinkage and selection operator (LASSO) regression, identified nine risk factors. Multicollinearity analysis confirmed no significant interactions. Multivariate logistic regression revealed six independent predictors: B-type natriuretic peptide (BNP) (per 200 pg/mL), creatine kinase-MB (CK-MB) (per 100 ng/mL), blood urea nitrogen (BUN) (per 1 mmol/L), lactic acid (LAC) (per 1 mmol/L), Holter mean heart rate (MHR) (per 10 bpm), and Holter total atrial beats (TAB) (per 1000 beats). Receiver operating characteristic (ROC) curve and decision curve analysis (DCA) demonstrated superior net benefit of the combined model over individual predictors. Conclusion The combination of BNP, CK-MB, BUN, LAC, Holter MHR, and Holter TAB can effectively predicts in-hospital all-cause death in STEMI patients undergoing primary coronary artery stenting, offering potential clinical utility.