This retrospective cohort study developed and validated a nomogram—a graphic calculating tool—to predict ICU mortality in patients with Sepsis-Associated Encephalopathy (SAE), a serious complication where sepsis affects the brain.

Goal: To create a simple, effective model to identify high-risk SAE patients early, enabling faster and more targeted interventions.

Study Data: The model was trained using data from 5,242 patients in the MIMIC-IV database and then externally validated on 3,103 patients from the eICU-CRD database.

Predictive Variables: Using LASSO regression, the researchers identified eight key variables for inclusion in the final predictive nomogram.

Superior Performance: The resulting nomogram significantly outperformed several standard ICU scoring systems in predicting mortality. Its predictive power, measured by the Area Under the Receiver Operating Characteristic (AUROC) curve, was 0.832, which was notably higher than the AUROCs for SAPS II (0.769), SOFA (0.607), and GCS/GOS scores (0.575) in the training set.

The new nomogram offers a more accurate and reliable method for predicting mortality in patients with Sepsis-Associated Encephalopathy than current standard scores, providing a valuable tool for clinicians to guide early intervention strategies.

Read More