Nomogram for predicting 30-day mortality in patients with infective endocarditis: a retrospective study

The prognosis for patients with infective endocarditis (IE) remains suboptimal. This study aimed to identify risk factors for 30-day mortality after hospital admission in IE patients and to construct a prognostic prediction model. We conducted a retrospective study of 391 patients diagnosed with IE between January 2017 and September 2024. Enrolled patients were randomly allocated to a training cohort and a validation cohort in a 7:3 ratio. Within the training cohort, patients were stratified into survival and non-survival groups based on 30-day outcomes. Least Absolute Shrinkage and Selection Operator (LASSO) regression and logistic regression were employed for feature selection and prediction model construction. Model performance was evaluated using the area under the curve (AUC), calibration curves, and decision curve analysis (DCA). Surgical, lower lymphocyte count (LY), prolonged activated partial thromboplastin time (APTT), higher lactate dehydrogenase (LDH), higher creatine kinase (CK), and higher UREA were identified as independent risk factors. The nomogram model demonstrated an AUC of 0.889 in the training cohort and 0.783 in the validation cohort. Calibration curves showed excellent alignment between predicted and actual outcomes, and DCA confirmed the model’s positive net clinical benefit across reasonable thresholds. Surgical, LY, APTT, LDH, CK, and UREA are significant predictors of prognosis in IE patients. The risk prediction model constructed from these factors shows high clinical utility for assessing the 30-day mortality risk in hospitalized IE patients.