Development and validation of a nomogram to predict mortality of patients with DIC in ICU
View abstract on PubMed
Summary
This summary is machine-generated.This study developed a predictive nomogram for short-term mortality in disseminated intravascular coagulation (DIC) patients. The Lasso-Cox model identifies key risk factors, offering reliable, individualized risk predictions for critically ill patients.
Area Of Science
- Critical Care Medicine
- Hematology
- Biostatistics
Background
- Disseminated intravascular coagulation (DIC) is a severe condition associated with high mortality in critically ill patients.
- Predictive models for short-term mortality in DIC patients are scarce.
- Identifying risk factors is crucial for improving patient outcomes.
Purpose Of The Study
- To identify risk factors associated with short-term mortality in overt DIC.
- To construct a predictive nomogram for DIC mortality risk.
- To provide individualized and reliable mortality risk assessments.
Main Methods
- A total of 676 overt DIC patients were included.
- A Cox proportional hazards regression model was developed using variables selected by LASSO regression.
- Model performance was validated across MIMIC-III, MIMIC-IV, and the 908th Hospital databases.
Main Results
- The predictive model incorporated heart failure, sepsis, height, SBP, lactate, HCT, PLT, INR, AST, and norepinephrine use.
- The model achieved a C-index >0.65 across validation datasets, demonstrating good predictive performance at 7 and 28 days.
- A nomogram was developed, showing superior net benefits in decision curve analysis.
Conclusions
- A Lasso-Cox regression-based nomogram provides a tool for predicting short-term mortality in overt DIC.
- The nomogram offers individualized and reliable risk stratification for critically ill DIC patients.
- This tool can aid in clinical decision-making and patient management.

