Related Experiment Video
Updated: Jun 29, 2026
Identification and Quantification of Deranged Metabolites in Critically Ill Patients Using NMR-Based Metabolomics
Published on: November 29, 2024
Risk Factor Analysis and Nomogram for Predicting 28-day in-hospital Mortality in ICU Patients with Liver Necrosis
Fanbing Wang1,2, Haohong Huang2, Nuoqi Liu2
1Department of Hepatobiliary Surgery, The First Affiliated Hospital, Jinan University, Guangzhou, Guangdong Province, 524000, China.
Introduction:
The mortality risk factors in intensive care unit (ICU) patients with liver necrosis (LN) remain unclear. This study aimed to develop a prediction model for mortality risk in LN patients to assist clinical practice.
Methods:
We enrolled 1153 acute and subacute LN patients from the MIMIC-III database, randomly divided into training and validation cohorts. LASSO regression was used to screen risk factors. A multiple logistic regression model and nomogram were established for 28-day mortality prediction. The model was assessed by AUC, calibration curves, and DCA.
Results:
LASSO regression identified age, gender, ethnicity, marital status, bicarbonate, bilirubin, hemoglobin, electrolytes, heart rate, temperature, and SpO₂as independent risk factors.
Discussion:
Although there is currently no suitable model for predicting patients with sepsis, the model we have constructed includes some factors that were previously considered independent but were not as significant. This time, we have highlighted the more important positions. Our model will provide ideas for predicting liver necrosis insepsis.
Conclusion:
Our model integrates previously underappreciated predictors and emphasizes their clinical significance, providing novel insights for predicting LN in patients with sepsis.