Background: Traumatic brain injury (TBI) is a major source of morbidity and mortality in children. The Glasgow Coma Scale (GCS) can be challenging to calculate in pediatric patients. Our objective was to determine its reproducibility between prehospital providers and pediatric trauma hospital personnel.
Materials and methods: The institutional trauma database for a level 1 pediatric trauma center was queried for patients aged ≤18 y who presented with a TBI. Demographics, mechanism, prehospital GCS, and trauma center GCS were collected. Agreement was evaluated with weighted kappa (κ) coefficients (0 = agreement no better than that expected by chance alone, 1 = perfect agreement).
Results: The inclusion criteria were met by 1711 patients, 263 of whom were aged <3 y. Prehospital GCS and trauma center GCS differed in 766 patients (44.8%). Agreement between prehospital GCS and trauma center GCS was moderate for all patients (κ = 0.61, 95% confidence interval [CI] 0.57-0.64). Agreement was slightly better than chance alone in patients with trauma center GCS between 9 and 12 y (κ = 0.09, 95% CI 0.03-0.15) and was lower for children aged 0-2 y (κ = 0.51, 95% CI 0.42-0.61) than for those aged between 3 and 18 y (κ = 0.63, 95% CI 0.59-0.66). Younger children were more likely to have score differences of at least 3 points (21.3% versus 13.6% of 3- to 18-y-olds, P < 0.001).
Conclusions: Prehospital and trauma center GCS scores frequently disagree in children, particularly in TBI patients aged <3 y and those with moderate TBI. Centers should consider the inconsistency of the pediatric GCS when triaging TBI patients.
Keywords: Agreement; Glasgow coma scale; Pediatric; Traumatic brain injury.
Copyright © 2019 Elsevier Inc. All rights reserved.