Development and Validation of a Nomogram Prediction Model for Sepsis-Induced Coagulopathy: A Multicenter Retrospective Study.

Objective: This study aimed to develop a prediction model to assess the risk of sepsis-induced coagulopathy (SIC) in sepsis patients.

Methods: We conducted a retrospective study of septic patients admitted to the Intensive Care Units of Shandong Provincial Hospital (Central Campus and East Campus), and Shenxian People's Hospital from January 2019 to September 2024. We used Kaplan-Meier analysis to assess survival outcomes. LASSO regression identified predictive variables, and logistic regression was employed to analyze risk factors for pre-SIC. A nomogram prediction model was developed via R software and evaluated via receiver operating characteristic (ROC) curves, calibration curves, and decision curve analysis (DCA).

Results: Among 309 patients, 236 were in the training set, and 73 were in the test set. The pre-SIC group had higher mortality (44.8% vs. 21.3%) and disseminated intravascular coagulation (DIC) incidence (56.3% vs. 29.1%) than the non-SIC group. LASSO regression identified lactate, coagulation index, creatinine, and SIC scores as predictors of pre-SIC. The nomogram model demonstrated good calibration, with an AUC of 0.766 in the development cohort and 0.776 in the validation cohort. DCA confirmed the model's clinical utility.

Conclusion: SIC is associated with increased mortality, with pre-SIC further increasing the risk of death. The nomogram-based prediction model provides a reliable tool for early SIC identification, potentially improving sepsis management and outcomes.