Development and validation of a nomogram for predicting sepsis-associated encephalopathy in ICU patients

Abstract

Abstract Background: Sepsis-associated encephalopathy (SAE) is associated with systemic inflammation caused by sepsis. It is estimated that a majority of sepsis patients develop severe acute effects (SAE) during their stay in the intensive care unit (ICU), and a significant number of survivors have persistent cognitive impairment even after they have recovered from the illness. The aim of this study was to develop a useful predictive nomogram for patients with ICU sepsis and screen for SAE risk factors. Methods: We conducted a retrospective cohort study using the Medical Information Mart for Intensive Care IV (MIMIC-IV) database, defining SAE as a Glasgow Coma Scale (GCS) score of ≤15 or delirium. We randomly divided patients into training and validation cohorts, and used least absolute shrinkage and selection operator (LASSO) regression modeling to optimize feature selection. The independent risk factors were determined through a multivariable logistic regression analysis, and a prediction model was built. Nomogram performance was evaluated using the area under the receiver operating characteristic curve (AUC), calibration plots, Hosmer-Lemeshow test, decision curve analysis (DCA), net reclassification improvement (NRI) and integrated discrimination improvement (IDI). Results: Among 4476 sepsis patients screened, 2781 (62.1%) developed SAE. In-hospital mortality was higher in the SAE group than in the non-SAE group (9.5% vs 3.7% p<0.001). A number of variables were screened, such as the patient's age, gender, BMI on the first day of admission, the mean arterial pressure, the body temperature, the platelet count, the sodium level, and the use of midazolam. The variables that were assessed encompassed the patient's age, gender, BMI upon admission, initial mean arterial pressure, body temperature, platelet count, sodium level, utilization of midazolam, and SOFA score. These were used to construct and validate a nomogram. Comparisons between the nomogram's AUC, NRI, IDI, and DCA with those of the conventional SOFA score in conjunction with delirium revealed superior performance. The nomogram's calibration plots and the results of the Hosmer-Lemeshow test indicated accurate calibration. Enhanced NRI and IDI values demonstrated that our scoring system surpassed traditional diagnostic approaches. Furthermore, the DCA curve indicated favorable clinical applicability of the nomogram. Conclusion: This study identified independent risk factors for the development of SAE in sepsis patients and used them to construct a predictive model. The findings of this study can provide a clinical reference for the early diagnosis of SAE in patients.