Prediction of mortality in pediatric traumatic brain injury: Implementations from a tertiary pediatric intensive care facility

Abstract

BACKGROUND: To explore the mortality risk factors of traumatic brain injury in pediatric intensive care unit admissions. METHODS: Eighty-eight children (categorized using the Glasgow Coma Scale) between September 2014 and December 2016 were analyzed. Emergency department and intensive care course, treatment strategies, axonal injury, intubation and tracheostomy rates, length of intensive care and hospitalization, Rotterdam-CT scores, injury severity scores, and PRISM-III scores were recorded. RESULTS: Older age was associated with trauma severity (p=0.010). Target serum osmolality was reached at 8.5 (3.5–40) hours in patients undergoing anti-edema therapy. ICP-monitoring rates was 8%; in absence of ICP-monitorization clinical follow-up was performed through repeated brain tomographies. Axonal injury was associated with prolonged intubation, intensive care and hospital stay (p<0.001, p<0.001, p=0.030). Six children required tracheostomy at 14.33±1.03 days; decannulations were performed within 6 months in five children. CONCLUSION: Mortality rate was 12.5%; six patients progressed to brain death with organ donor approvals in five. Initial hypotension, lung contusion, injury severity scores and Rotterdam-CT scores were related with mortality. Rotterdam-CT score was determined as the independent risk factor for mortality; one increment in the score increased the odd of recovery by 20.334 times (%95 CI 1.999–206.879). ISS score was also borderline significant (p=0.052; OR:1.195 %95 CI 0.999–1.430).