A Practical Nomogram Based on RDW-CV for Predicting Clinical Outcome in Elderly Septic Patients

Area of Science:

• Critical care medicine
• Geriatrics
• Hematology

Background:

• Sepsis is a leading cause of mortality in elderly patients.
• Predictive markers for sepsis outcomes are crucial for timely intervention.
• Red blood cell distribution width-coefficient of variation (RDW-CV) has emerged as a potential prognostic indicator.

Purpose of the Study:

• To establish a prognostic nomogram for predicting mortality in elderly septic patients.
• To evaluate the predictive value of RDW-CV for mortality in this population.
• To assess the clinical applicability of the developed nomogram.

Main Methods:

• Retrospective analysis of 986 elderly septic patients.
• Machine learning-based feature importance and multivariate logistic regression.
• Development and validation of a prognostic nomogram incorporating RDW-CV and other clinical indicators.

Main Results:

• Mortality rate in the study cohort was 27.48%.
• RDW-CV was significantly higher in non-survivors (17.22 ±3.98%) than survivors (15.30 ±2.81%).
• The nomogram achieved an AUC of 0.755, demonstrating good predictive accuracy and clinical utility.

Conclusions:

• RDW-CV is a significant independent predictor of mortality in elderly sepsis.
• The developed nomogram, integrating RDW-CV, offers a reliable tool for predicting clinical outcomes.
• The nomogram's accuracy and clinical applicability were confirmed through validation analyses.