BACKGROUND: The purpose of this study was to identify an optimal definition of massive transfusion in civilian pediatric trauma with severe traumatic brain injury (TBI) METHODS: Severely injured children (age ≤18 y) with severe TBI in the Trauma Quality Improvement Program research data sets 2015-2016 that received blood products were identified. Data were analyzed using descriptive statistics, Wilcoxon rank-sum, chi-square, and logistic regression. Continuous variables are presented as median (interquartile range). Massive transfusion thresholds were determined based on receiver operating curves and optimization of sensitivity and specificity RESULTS: Of the 460 included children, the mortality rate was 43%. There were no differences in demographics, heart rate at presentation, or injury severity score between children that lived or died. However, those who died had lower Glasgow coma scores (3 [3, 8] versus 3 [3, 3]; P < 0.01), were more likely to have had a penetrating injury (20% versus 11%; P < 0.01) and were more likely to be hypotensive for age (62% versus 34%; P < 0.01). Total blood products infused were greater in those who died (34 mL/kg/4-h [17, 65] versus 22 [12, 44]; P < 0.01). Sensitivity and specificity for delayed mortality was optimized at 40 mL/kg/4 h, and for the need for a hemorrhage control procedure at 50 mL/kg/4 h. These thresholds predicted delayed mortality (OR 2.12; 95% CI 1.28-3.50; P < 0.01) and the need for hemorrhage control procedures (5.47; 95% CI 2.82-10.61; P < 0.01) CONCLUSIONS: For children with TBI, a massive transfusion threshold of 40 mL/kg/4-h of total administered blood products may be used to identify at-risk patients, improve resource utilization, and guide future research methodology.