Development and Validation of a Risk Prediction Model for New-Onset Atrial Fibrillation in Sepsis.

Background: Sepsis patients face a high risk of new-onset atrial fibrillation (NOAF), which increases mortality. Thus, it is significant to construct a risk prediction model for early risk stratification. Objective: To construct and validate a risk prediction model for NOAF in sepsis. Methods: A total of 423 sepsis patients were randomly divided into training (n=299) and validation (n=124) cohorts. Predictors were selected using least absolute shrinkage and selection operator (LASSO) regression, and independent risk factors were identified by multivariate logistic regression to construct a nomogram. Model performance was assessed by the area under the receiver operating characteristic curve (AUC), Hosmer-Lemeshow test, and calibration curves. Clinical utility was evaluated using decision curve analysis (DCA) and clinical impact curves (CIC). Results: Log interleukin-6 (Log IL-6), blood urea nitrogen (BUN), and heart rate (HR) were identified as independent risk factors for NOAF. The nomogram demonstrated strong discriminative ability, with AUCs of 0.925 in the training cohort and 0.866 in the validation cohort. Calibration was good in both cohorts, and DCA and CIC indicated favorable clinical utility across a range of threshold probabilities. Conclusion: A risk prediction model incorporating Log IL-6, BUN, and HR effectively could predict NOAF in sepsis patients, with good discrimination, calibration, and potential clinical applicability for early risk identification. However, prior to further clinical application, additional multicenter, prospective studies are required for external validation.