A Predictive Model for Acute Respiratory Distress Syndrome Mortality Using Red Cell Distribution Width

Area of Science:

• Critical Care Medicine
• Pulmonary Medicine
• Biomarkers

Background:

• Acute Respiratory Distress Syndrome (ARDS) is a severe lung injury with high ICU mortality.
• Predictive models for ARDS mortality often rely on clinical factors.
• The utility of novel biomarkers for improving these predictions is under investigation.

Purpose of the Study:

• To evaluate if Red Cell Distribution Width (RDW) improves the predictive accuracy of ICU mortality in ARDS patients.
• To assess the impact of RDW on a base predictive model using established clinical factors.

Main Methods:

• Retrospective cohort study of 318 ARDS patients from an ICU database (2001-2008).
• Clinical factors (age, gender, comorbidity score, SOFA score, PaO2/FiO2 ratio) formed the base model.
• RDW was added to the base model to assess its impact on predicting ICU mortality.

Main Results:

• The base model (without RDW) had an AUC of 0.76.
• Adding RDW to the model increased the AUC to 0.78 (p=0.048).
• The Net Reclassification Improvement (NRI) was 0.46 (p=0.001), indicating improved predictive probability for 46% of patients.

Conclusions:

• RDW measurement at the time of ARDS diagnosis significantly improves the predictive discrimination of ICU mortality.
• RDW serves as a valuable adjunct biomarker for risk stratification in ARDS patients.
• Incorporating RDW into clinical models enhances the ability to predict outcomes in critically ill ARDS patients.