Prediction of massive transfusion and mortality in early trauma care: A retrospective analysis of scoring systems

Hemorrhagic shock is a leading cause of preventable trauma deaths, particularly within the first hours following injury. Early identification of patients requiring massive transfusion or with high mortality risk is critical to optimizing trauma management. Early identification of massive transfusion needs supports timely blood product preparation. Likewise, predicting mortality risk early can influence therapeutic planning and clinical decisions. Numerous trauma and transfusion scoring systems have been developed to guide such early decisions; however, their comparative predictive performance remains unclear. This study aimed to evaluate the effectiveness of trauma and transfusion scoring systems in predicting massive transfusion requirements and in-hospital mortality within the first four hours of trauma.

This retrospective study included 117 trauma patients who received at least one unit of red blood cell transfusion within the first four hours of admission to a tertiary care center between 2018 and 2022. Data on demographics, trauma mechanism, clinical and laboratory findings were collected. Each patient was evaluated using 16 trauma and transfusion scoring systems. Patients were categorized based on the need for massive transfusion, defined as receiving ≥5 units of blood products within four hours. Receiver Operating Characteristic (ROC) analysis was used to assess the performance of each scoring system, and optimal cut-off values were determined using the Youden Index.

Massive transfusion was required in 23 patients (19.7%), with firearm injuries being the most common mechanism among these cases. All 16 scoring systems significantly differentiated patients with and without massive transfusion. The Shock Index demonstrated the highest predictive accuracy for massive transfusion (area under the curve [AUC]=0.911). For in-hospital mortality, all scoring systems except the Schreiber Score showed significant predictive ability. The Trauma Related Injury Severity Score (TRISS) achieved the highest predictive value for mortality (AUC=0.975). Several scoring systems required revised threshold values for optimal performance in this cohort, highlighting the need for population-specific calibration.

Early-phase application of trauma and transfusion scoring systems provides valuable insights for predicting clinical outcomes in trauma patients. Among the systems analyzed, the Shock Index was the most reliable predictor of massive transfusion. Separately, TRISS demonstrated superior accuracy in forecasting in-hospital mortality. These findings emphasize the importance of rapid, score-based assessment in early trauma care and support further validation of scoring systems across diverse patient populations.