Development and validation of a clinical prediction model for sepsis-induced cardiomyopathy.

Objective: The objective of this study was to develop and validate a clinically applicable risk-prediction model for sepsis-induced cardiomyopathy (SICM) and to evaluate its predictive performance comprehensively. Methods: A retrospective cohort study was conducted using clinical data from patients with sepsis, obtained from the Medical Information Mart for Intensive Care IV database. Propensity score matching was applied to minimize confounding and achieve balance in baseline characteristics between groups. Candidate predictors were initially screened using univariate analysis, and feature selection was performed using the least absolute shrinkage and selection operator regression method. A multivariate logistic regression model was subsequently developed and externally validated with an independent dataset. Results: The prediction model was derived from 956 patients and externally validated in a cohort of 104 patients. Five independent predictors were retained in the final model: serum phosphate concentration, neutrophil percentage, troponin concentration, heart rate, and Charlson Comorbidity Index. The model demonstrated strong discriminatory ability, with C-statistic values of 0.80 in the derivation cohort, 0.79 in the internal validation cohort, and 0.76 in the external validation cohort. Conclusion: The validated prediction model provides accurate estimation of SICM risk. Based on routinely available clinical variables, the model has potential utility for early risk stratification and individualized management of patients with sepsis, supporting improved clinical outcomes in those at elevated risk of SICM.