Severe traumatic brain injury in the very young: data from the Trauma Quality Improvement Program

OBJECTIVE Mechanisms and patterns of severe traumatic brain injury (TBI) in infants and toddlers are different from adults and even from older children, but there is no taxonomy of injuries that can provide a basis for individualized care. The current project was undertaken as a proof of concept to demonstrate that a taxonomy can be constructed with clinical correlations. METHODS Data were taken from the American College of Surgeons Trauma Quality Improvement Program for the years 2014 through 2022. Inclusion criteria were an age < 2 years and Abbreviated Injury Scale head score of 4, 5, or 6. Latent class analysis (LCA) was performed based on covariates reflecting socioeconomic status, access to definitive care, mechanism of injury, severity of injury, and pathology. RESULTS There were 3735 cases meeting study inclusion criteria. Overall mortality among cases with a known outcome was 19%, and 70% of discharges were routine. Intracranial monitoring was coded for 28% of cases. An LCA model with 4 classes was studied. The mortality rates of the 4 classes were highly distinct at 2%, 13%, 37%, and 65%, respectively. The routine discharge rates were similarly distinct at 92%, 66%, 41%, and 12%, respectively. The rates of intracranial monitoring varied much more narrowly between 26% and 39%. CONCLUSIONS A scheme for classifying injuries based on LCA had powerful associations with outcomes and proves that a clinically meaningful taxonomy of severe TBI in the very young is possible. The fact that 25% of cases with the best prognosis underwent intracranial monitoring and only 37% of cases with the worst prognosis did so demonstrates the need for an evidence-based taxonomy to guide individualized care. Large-scale prospective study may define categories that are conceptually accessible rather than latent and that guide therapies as well as predict outcomes.