Pediatric resuscitation: Weight-based packed red blood cell volume is a reliable predictor of mortality

The definition of massive transfusion (MT) in civilian pediatric trauma patients is not established. In combat-injured pediatric patients, the definition of MT is based on the volume of total blood products transfused. The aim of this study is to define MT in civilian pediatric trauma patients based on a packed red blood cell (PRBC) volume threshold and compare its predictive power to a total blood products volume threshold. METHODS An analysis of the pediatric American College of Surgeons Trauma Quality Improvement Program database was performed (2014-2016) including pediatric trauma patients (4-18 years) who received blood products within 24 hours. Receiver operator characteristic curves for predicting mortality determined the optimal PRBC MT threshold. Area under receiver operating characteristic curve (AUROC) curve analysis was performed to compare the predictive power of a PRBC threshold to a total blood product threshold. RESULTS A total of 1,495 patients were included. Sensitivity and specificity for 24-hour and in-hospital mortality were optimal at a PRBC threshold of 20 mL/kg. As compared with total blood products threshold, 20 mL/kg PRBCs volume achieved higher discriminatory power for predicting 24-hour (AUROC, 0.803 vs. 0.672; p < 0.001) and in-hospital mortality (AUROC, 0.815 vs. 0.686, p < 0.001). Patients who received an MT had higher Injury Severity Score (p < 0.001) and were more likely to receive mechanical ventilation (p < 0.001) and intensive care unit admission (p < 0.001). Overall 24-hour mortality (23.1% vs. 7.6%, p < 0.001) and in-hospital mortality (44.9% vs. 15.8%, p < 0.001) were higher in the MT group. On regression analysis, MT significantly predicted in-hospital mortality (odds ratio, 3.8 [2.9-4.9, 95% CI]) and 24-hour mortality (odds ratio, 3.3 [2.4-4.7, 95% CI]). CONCLUSION The use of a PRBCs MT definition in civilian pediatric patients is a better predictor of mortality compared with total blood products threshold. These results provide a framework for MT protocol development. LEVEL OF EVIDENCE Prognostic study, level III.