Construction of a Risk Prediction Model for Stroke Occurrence in Septic Shock Patients: A Combined Analysis Using LASSO and Multivariate Logistic Regression.

Background: This study aimed to develop and validate a risk prediction model for stroke within 90 days in patients with septic shock and to identify independent risk factors. Methods: A retrospective, single-center study was performed, including 2127 septic shock patients admitted to Dongyang City People's Hospital from June 2016 to December 2024. Clinical variables were selected using LASSO regression, and the prediction model was established by multivariate logistic regression. Internal validation was conducted using nomogram, calibration curves, ROC curves, and decision curve analysis (DCA). Results: Seven variables (age, hypertension, ALB, TC, Cr, TBIL, and WBC) were screened by LASSO regression. Multivariate analysis confirmed that age, hypertension, ALB, TC, and TBIL were independent risk factors. These factors may reflect age-related vascular vulnerability, chronic hypertension-related cerebrovascular damage, and metabolic or hepatic dysfunction associated with stroke risk in septic shock. The model showed good predictive performance, with AUCs of 0.754 and 0.76 in the modeling and testing cohorts, respectively. Calibration and DCA curves confirmed satisfactory discrimination, calibration, and clinical utility. Conclusion: This prediction model demonstrates favorable performance for stroke risk stratification in septic shock patients. It may allow early identification of high-risk individuals, enabling timely intervention to reduce stroke-related mortality and disability.