Objective:

To develop and validate a clinically applicable risk-prediction model for sepsis-induced cardiomyopathy (SICM) and to evaluate its predictive performance comprehensively.

Approach:

• Study Design: A retrospective cohort study using clinical data from the MIMIC-IV database.
• Data Analysis: Propensity score matching, univariate analysis, least absolute shrinkage and selection operator regression, and multivariate logistic regression were used.
• Validation: The model was externally validated with an independent dataset of 104 patients.

Key Findings:

• Five independent predictors were identified: serum phosphate concentration, neutrophil percentage, troponin concentration, heart rate, and Charlson Comorbidity Index from a cohort of 956 patients.
• The model showed strong discriminatory ability with C-statistic values of 0.80 for the derivation cohort, 0.79 for the internal validation cohort, and 0.76 for the external validation cohort.

Interpretation:

The validated prediction model provides accurate estimation of SICM risk based on routinely available clinical variables.

Limitations:

• The study is retrospective and may be subject to biases inherent in such designs.
• The definition of SICM may not encompass all phenotypes of the condition.

Conclusion:

The model has potential utility for early risk stratification and individualized management of patients with sepsis.