Detection of traumatic brain injury with magnetic resonance imaging and S-100B protein in children, despite normal computed tomography of the brain.

OBJECTIVE: The objective of this study was to obtain data to further define the extent of traumatic brain injury by using S-100B protein and standard noncontrast magnetic resonance imaging with added fluid-attenuated inversion recovery (FLAIR) and gradient echo sequence in children with normal head computed tomography. DESIGN: Pilot, single cohort, prospective, clinical diagnostic study. SETTING: Pediatric intensive care and intermediate care unit in a tertiary care children's hospital. PATIENTS: Children ages 5-18 yrs who sustained traumatic brain injury, had a negative computed tomography of the brain, and were admitted to hospital were eligible for enrollment. INTERVENTIONS: Two blood samples were drawn for S-100B protein analysis: the first (t-1) as soon as possible or close to 6 hrs of injury and the second (t-2) close to 12 hrs from the time of injury. A magnetic resonance image of the brain was obtained within 96 hrs of injury. MEASUREMENTS AND MAIN RESULTS: Seven of 17 patients (41%) had positive magnetic resonance image. Of the seven patients with positive magnetic resonance image, 100% (seven of seven) had a positive magnetic resonance image with FLAIR sequence, 85% (six of seven) with axial T2 sequence and 50% (three of six) with gradient echo sequence. There was no statistically significant difference in S-100B protein concentrations in patients with a positive magnetic resonance image (n = 7) and those with a negative magnetic resonance image (n = 10; p =.40 at t-1 and p =.13 at t-2). The concentration of S-100B protein was statistically significantly higher in patients with head and other bodily injury (n = 9) compared with isolated head injury (n = 6; p =.018 at t-1 and p =.025 at t-2). Patients with a positive magnetic resonance image had a lower Glasgow Coma Scale score and longer duration of hospital stay. CONCLUSIONS: Magnetic resonance imaging seems to be a useful modality to better define the spectrum of brain injury in children with mild head trauma. The addition of S-100B protein measurement does not seem to be useful in this setting.