J. D. Kauffman1, C. N. Litz1, S. A. Thiel1, A. Carey1, P. D. Danielson1, N. M. Chandler1 1Johns Hopkins All Children’s Hospital,Division Of Pediatric Surgery,St. Petersburg, FLORIDA, USA
Introduction: Traumatic brain injury (TBI) results in nearly half a million pediatric emergency department (ED) visits annually in the U.S. Computed Tomography (CT) is often used to evaluate for TBI but increases a child’s risk of malignancy. The PECARN Pediatric Head Injury/Trauma Algorithm was developed to provide guidance as to when CT is indicated for children presenting after head trauma. One criterion in the algorithm, history of loss of consciousness (LOC), may be indeterminate if the event is unwitnessed. The purpose of this study is to determine whether children presenting with unknown history of LOC are at greater risk for TBI than those with no history of LOC.
Methods: Following IRB approval, the institutional trauma registry was reviewed to identify children 0–17 years of age presenting within 24 hours of minor head injury with score of 14 or 15 on the Glasgow Coma Scale (GCS) from January, 2010 to April, 2017. Those who underwent CT prior to arrival, those with penetrating injuries, and those suspected of non-accidental trauma were excluded. Age-specific predictor variables for clinically important TBI (ciTBI), defined as TBI that results in hospital admission for two or more nights, intubation for greater than 24 hours, neurosurgical intervention, or death, were extracted. Ordinal data was analyzed using a chi-square test or Fisher’s exact test as indicated.
Results: Among 1852 patients reviewed, 741 met inclusion criteria. Median age was 7.6 years; 66% were male. The majority (56.4%) reported no LOC, 260 (35.1%) reported LOC, and in 63 (8.5%) LOC history was indeterminate. Those in the indeterminate LOC group were significantly more likely than those with no reported LOC to undergo CT, but significantly less likely to have evidence of TBI on CT (Table 1). There was no difference in rate of ciTBI or neurosurgical intervention between groups. Each of the three children in the indeterminate LOC group who developed ciTBI met criteria (altered mental status and/or severe mechanism of injury) that would have resulted in CT being recommended even if LOC history had not been considered. Overall, 89% of those in the indeterminate LOC group exhibited findings apart from LOC status that justified CT. Of the remaining 11% (those with no clinical criteria for CT apart from indeterminate LOC status), none developed TBI.
Conclusion: Children presenting to the ED within 24 hours of minor head injury for whom history of LOC is unknown and who otherwise meet PECARN criteria for observation may not be at greater risk than those with no history of LOC for findings of TBI on CT, ciTBI, or need for neurosurgical intervention. Additional positive findings on PECARN algorithm may be used to direct need for CT or observation.
