PP06.12 – 2814: Laboratory predictors of outcome following paediatric traumatic brain injury

Abstract

Objective Paediatric traumatic brain injury (TBI) has a mortality of around 10% and there is poor functional outcome in a substantial proportion of survivors. Limited understanding of pathophysiology means that there are few data on prediction of outcome. This audit aimed to review whether admission laboratory data could predict outcome after TBI in a regional centre. Methods A retrospective review was conducted between 1998 and 2013 of consecutive children Results Data from 243 children (median age 11.07; range 0.18–17.45 years; 71.8% boys) were available. Median minimum Glasgow coma score before intubation (GCS) was 7; range 3–15. 23 children died and 96 had poor outcome. White cell count was higher in those who died than in those who survived with poor (p=0.007) or good (p=0.002) outcome. Haemoglobin was lower in those who died than in survivors with poor (p=0.037) and good (p=0.057) outcomes but between the two groups of survivors there was no difference. In multiple logistic regression, death was predicted by high white cell count (odds ratio 1.159; 95% confidence intervals [95%CI] 1.032–1.301) and there was a trend for an effect of low haemoglobin (odds ratio 0.63, 95%CI 0.38, 1.03; p=0.066) independent of age on admission and GCS. Conclusion Mortality following severe paediatric TBI is higher in children with high white cell count and low haemoglobin, perhaps because head and neck infections and anaemia predispose to acutely abnormal intracranial hydro- and haemodynamics.