Related Experiment Video
Updated: Aug 7, 2026
06:11
Uncontrolled Hemorrhagic Shock Modeled via Liver Laceration in Mice with Real Time Hemodynamic Monitoring
Published on: May 21, 2017
A predictive model for mortality in massively transfused trauma patients
Ronald R Barbosa1, Susan E Rowell, Chitra N Sambasivan
1Trauma Services, Legacy Emanuel Hospital and Health Center, Portland 97227, USA. rbarbosa@lhs.org
The Journal of Trauma
|August 5, 2011
Summary
Predicting mortality in massively transfused trauma patients is possible using early post-injury data. Key predictors include Glasgow Coma Scale score, pH, heart rate, age, Injury Severity Score, and red blood cell transfusion needs.
Area of Science:
- Trauma Surgery
- Critical Care Medicine
- Emergency Medicine
Background:
- Trauma resuscitation and systems have improved survival rates.
- Massive transfusion is increasingly utilized in civilian trauma care.
- Objective mortality predictors in these patients remain underexplored.
Purpose of the Study:
- To identify early post-injury variables predicting 24-hour and 30-day mortality.
- To develop predictive models for massively transfused trauma patients.
Main Methods:
- Analysis of data from 704 massively transfused patients across 23 Level I trauma centers.
- Logistic regression models were used to assess predictors at patient arrival and 6 hours post-injury.
- Receiver operating characteristic curves evaluated model performance.
Main Results:
- pH, Glasgow Coma Scale score, and heart rate predicted 24-hour mortality (AUROC 0.747).
- Adding 6-hour red blood cell requirement improved 24-hour prediction (AUROC 0.769).
- Age and Injury Severity Score further enhanced 30-day mortality prediction (AUROC 0.828).
Conclusions:
- Glasgow Coma Scale score, pH, heart rate, age, Injury Severity Score, and 6-hour RBC transfusion requirement are independent mortality predictors.
- Current predictive models offer modest accuracy.
- These models should not guide decisions to withhold massive transfusion.
