Related Experiment Videos
Development and validation of a multidimensional nomogram for predicting 28-day mortality in sepsis patients with
Piao Zhang1, Chengcheng Sun2, Yingjing Wu3
1Department of Anaesthesiology, West China Hospital, Sichuan University, Chengdu, Sichuan, 610041, China.
Respiratory Medicine
|June 24, 2026
Summary
A new nomogram accurately predicts 28-day mortality in sepsis patients with acute lung injury (ALI). This tool offers reliable clinical insights for precise interventions, improving patient outcomes.
Area of Science:
- Critical Care Medicine
- Pulmonology
- Medical Informatics
Background:
- Sepsis-associated acute lung injury (ALI) presents a significant challenge in critical care.
- Accurate prediction of mortality is crucial for timely and effective clinical management.
Purpose of the Study:
- To develop and validate a multidimensional nomogram for predicting 28-day mortality in sepsis patients with ALI.
- To establish a reliable tool for precise clinical intervention in this patient population.
Main Methods:
- Retrospective analysis of 4620 sepsis patients with ALI from the MIMIC-IV database.
- Development of a nomogram using LASSO and Boruta algorithms for variable selection.
- Performance evaluation via AUROC, calibration curves, decision curve analysis (DCA), and SHAP analysis.
Main Results:
- A 12-variable nomogram demonstrated superior predictive ability (AUROC=0.869 training, 0.884 validation) over existing scores.
- Lactate, creatinine, and SOFA score identified as core risk factors; platelet count and albumin as protective.
- Good agreement between predicted and actual mortality risks confirmed by calibration curves and DCA.
Conclusions:
- The developed nomogram exhibits excellent predictive efficacy and clinical interpretability for 28-day mortality in sepsis with ALI.
- This tool provides a reliable basis for precise clinical interventions, potentially improving patient outcomes.