Related Experiment Video
Updated: Jan 14, 2026
A Data-Driven Approach to Quantifying Immune States in Sepsis
Published on: February 7, 2025
A nomogram for predicting ICU mortality of sepsis associated encephalopathy: a retrospective cohort study based on
Xuemei Hu1,2, Jianbao Wang1,2, Shaowei Wang1,2,3
1The Second Department of Intensive Care Unit, The Second Affiliated Hospital of Anhui Medical University, 678 Furong Road, Hefei, Anhui Province, 230601, China.
Background:
Sepsis-associated encephalopathy (SAE) is a fatal complication of sepsis, with a high mortality rate worldwide. This study aimed to reduce mortality and improve the quality of life of patients with SAEs by developing a practical nomogram to predict the risk factors associated with ICU mortality.
Method:
The MIMIC database was used as the training set to build the model, and the eICU-CRD served as the validation set for external verification. LASSO regression analysis was conducted to identify predictive variables and develop the nomogram model. Receiver Operating Characteristic (ROC) curves were generated to assess the model's discriminative ability. Model calibration was assessed using calibration curves and the Hosmer-Lemeshow goodness-of-fit tests. Clinical decision curves were plotted to assess the model's net benefit and evaluate its clinical applicability.
Results:
A total of 5,242 patients from the MIMIC database and 3,103 from the eICU-CRD were included in the study. LASSO regression, identified eight predictive variables for inclusion in the final model. The nomogram was evaluated against standard ICU scoring systems, including SAPS II, SOFA and GOS scores, with AUROC values of 0.832, 0.769, 0.607, and 0.575, respectively, in the training set. Conversely, the validation set demonstrated AUROC values of 0.825, 0.715, 0.714, and 0.587. P-values from the Hosmer-Lemeshow goodness-of-fit test for both the training and validation sets were 0.129 and 0.583, respectively, indicating a good fit quality. DCA revealed that the nomogram consistently provides greater net benefits compared to SAPS II, SOFA, and GCS scores.
Conclusion:
Developing mortality prediction models for SAE patients in the ICU can facilitate early intervention strategies and potentially reduce mortality rates in this high-risk population.
Clinical Trial Number:
Not applicable.
