Predictors of 30-Day Mortality in Trauma: A Single-Center Retrospective Exploratory Study

Area of Science:

• Trauma research
• Medical outcome prediction

Background:

• Trauma is a leading cause of preventable death globally.
• Accurate outcome prediction is vital for trauma patient triage and resource management.
• Existing trauma scores may lack precision for specific patient cohorts.

Purpose of the Study:

• To develop a more precise logistic regression model for predicting 30-day mortality in trauma patients.
• To tailor a predictive model to a specific patient cohort for future research.

Main Methods:

• Retrospective analysis of 91 trauma patients.
• Development of a logistic regression model using clinical and paraclinical factors.
• Inclusion of systolic blood pressure, glycemia, urea serum levels, and number of fractures.

Main Results:

• The final model demonstrated excellent predictive performance (McFadden R² = 0.682; AUC = 0.94).
• Systolic blood pressure was a significant predictor of mortality (OR = 0.944).
• The model's performance surpassed that of The Trauma and Injury Severity Score in this cohort.

Conclusions:

• Commonly available clinical parameters can effectively stratify trauma patient risk.
• Hemodynamic instability and metabolic response are key prognostic indicators.
• Further validation in larger, multicentric cohorts is necessary.