Development of a prehospital clinical prediction score for predicting massive blood transfusion in patients with trauma.

Background: Most prehospital trauma deaths result from exsanguination. Developing a scoring system to predict massive blood transfusion (MBT) is essential in urban areas of middle- to lower-income developing countries where traffic accidents are frequent. Aim: To develop and validate a clinical score for predicting MBT in prehospital trauma patients. Methods: Data were obtained from emergency medical service (EMS) patient care reports between January 2018 and October 2023. Multiple logistic regression was used to develop the score. Discrimination was assessed using receiver operating characteristics (ROC) curve analysis, and the area under the ROC curve (AuROC) with 95 % confidence interval was reported. Risk groups were defined by positive likelihood ratio (LR + ).Results: Six hundred patients were included. Significant predictors were injury severity score, wound type, serious injury body part, scene distance ≥ 4 km, response time < 8 min, hospital transfer time < 8 min, airway management, and fluid access. The score classified patients into low- (≤3), moderate- (4-9), and high-risk (≥10) groups, with MBT probabilities of 1.6 %, 20.7 %, and 74.2 %, respectively. LR + values were 0.07, 1.08, and 11.87. The AuROC was 0.883. Conclusion: This study derived a prehospital clinical score that demonstrated good discrimination for predicting MBT. Internal validation using bootstrap resampling confirmed the robustness of the model. Nevertheless, external and prospective validation in diverse trauma populations and EMS settings are required before the score can be considered for clinical application.