The following is a summary of “Improved prediction of sepsis-associated encephalopathy in intensive care unit sepsis patients with an innovative nomogram tool,” published in the February 2024 issue of Neurology by Jin et al.
Researchers conducted a retrospective study to unlock risk factors for sepsis-associated encephalopathy (SAE) in ICU patients, aiming to create a predictive nomogram for early identification.
They utilized the Medical Information Mart for Intensive Care IV (MIMIC-IV) database. SAE was defined as a Glasgow Coma Scale (GCS) score of 15 or lower, or delirium. The patients were divided randomly into two groups, namely, training and validation groups. Feature selection was optimized using least absolute shrinkage and selection operator (LASSO) regression modeling. Multivariable logistic regression analysis identified independent risk factors and constructed a prediction model. The nomogram’s performance was assessed using various metrics, such as 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).
The results showed 4,476 sepsis patients, 2,781 (62.1%) developed SAE. In-hospital mortality was higher in the SAE cohort compared to the non-SAE group (9.5% vs. 3.7%, P<0.001). Various variables, such as age, gender, BMI on admission, mean arterial pressure, body temperature, platelet count, sodium level, and use of midazolam, were analyzed. Variables were utilized to develop and validate a nomogram. The nomogram’s performance, evaluated by AUC, NRI, IDI, and DCA, surpassed that of the conventional SOFA score combined with delirium. Calibration plots and the Hosmer-Lemeshow test confirmed the nomogram’s accuracy. Enhanced NRI and IDI values show our system’s superiority. The DCA curve highlights its practicality.
They concluded that the study built a predictive model for early detection by pinpointing SAE risk factors in sepsis, potentially aiding clinical management.
Source: frontiersin.org/journals/neurology/articles/10.3389/fneur.2024.1344004/full