Construction and validation of a personalized risk prediction model for in-hospital mortality in patients with acute myocardial infarction undergoing percutaneous coronary intervention

Background: Although emergency Percutaneous Coronary Intervention (PCI) has been shown to reduce mortality in patients with Acute Myocardial Infarction (AMI), the risk of in-hospital death remains high. In this study, the authors aimed to identify risk factors associated with in-hospital mortality in AMI patients who underwent PCI, develop a nomogram prediction model, and evaluate its effectiveness. Methods: The authors retrospectively analyzed data from 1260 patients who underwent emergency PCI at Dongyang People's Hospital between June 1, 2013, and December 31, 2021. Patients were divided into two groups based on in-hospital mortality: the death group (n = 61) and the survival group (n = 1199). Clinical data between the two groups were compared. The Least Absolute Shrinkage and Selection Operator (LASSO) regression was used to select non-zero coefficients of predictive factors. Multivariable logistic regression analysis was then performed to identify independent risk factors for in-hospital mortality in AMI patients after emergency PCI. A nomogram model for predicting the risk of in-hospital mortality in AMI patients after PCI was constructed, and its predictive performance was evaluated using the c-index. Internal validation was performed using the bootstrap method with 1000 resamples. The Hosmer-Lemeshow test was used to assess the goodness of fit, and a calibration curve was plotted to evaluate the model's calibration. Results: LASSO regression identified d-dimer, B-type natriuretic peptide, white blood cell count, heart rate, aspartate aminotransferase, systolic blood pressure, and the presence of postoperative respiratory failure as important predictive factors for in-hospital mortality in AMI patients after PCI. Multivariable logistic regression analysis showed that d-dimer, B-type natriuretic peptide, white blood cell count, systolic blood pressure, and the presence of postoperative respiratory failure were independent factors for in-hospital mortality. A nomogram model for predicting the risk of in-hospital mortality in AMI patients after PCI was constructed using these independent predictive factors. The Hosmer-Lemeshow test yielded a Chi-Square value of 9.43 (p = 0.331), indicating a good fit for the model, and the calibration curve closely approximated the ideal model. The c-index for internal validation was 0.700 (0.560-0.834), further confirming the predictive performance of the model. Clinical decision analysis demonstrated that the nomogram model had good clinical utility, with an area under the ROC curve of 0.944 (95 % CI 0.903-0.963), indicating excellent discriminative ability. Conclusion: This study identified B-type natriuretic peptide, white blood cell count, systolic blood pressure, ddimer, and the presence of respiratory failure as independent factors for in-hospital mortality in AMI patients undergoing emergency PCI. The nomogram model based on these factors showed high predictive accuracy and feasibility.