RESUMO
OBJECTIVES: Minor head trauma accounts for a significant proportion of pediatric emergency department (ED) visits. In children younger than 24 months, scalp hematomas are thought to be associated with the presence of intracranial injury (ICI). We investigated which scalp hematoma characteristics were associated with increased odds of ICI in children less than 17 years who presented to the ED following minor head injury and whether an underlying linear skull fracture may explain this relationship. METHODS: This was a secondary analysis of 3,866 patients enrolled in the Canadian Assessment of Tomography of Childhood Head Injury (CATCH) study. Information about scalp hematoma presence (yes/no), location (frontal, temporal/parietal, occipital), and size (small and localized, large and boggy) was collected by emergency physicians using a structured data collection form. ICI was defined as the presence of an acute brain lesion on computed tomography. Logistic regression analyses were adjusted for age, sex, dangerous injury mechanism, irritability on examination, suspected open or depressed skull fracture, and clinical signs of basal skull fracture. RESULTS: ICI was present in 159 (4.1%) patients. The presence of a scalp hematoma (n = 1,189) in any location was associated with significantly greater odds of ICI (odds ratio [OR] = 4.4, 95% confidence interval [CI] = 3.06 to 6.02), particularly for those located in temporal/parietal (OR = 6.0, 95% CI = 3.9 to 9.3) and occipital regions (OR = 5.6, 95% CI = 3.5 to 8.9). Both small and localized and large and boggy hematomas were significantly associated with ICI, although larger hematomas conferred larger odds (OR = 9.9, 95% CI = 6.3 to 15.5). Although the presence of a scalp hematoma was associated with greater odds of ICI in all age groups, odds were greatest in children aged 0 to 6 months (OR = 13.5, 95% CI = 1.5 to 119.3). Linear skull fractures were present in 156 (4.0%) patients. Of the 111 patients with scalp hematoma and ICI, 57 (51%) patients had a linear skull fracture and 54 (49%) did not. The association between scalp hematoma and ICI attenuated but remained significant after excluding patients with linear skull fracture (OR = 3.3, 95% CI = 2.1 to 5.1). CONCLUSIONS: Large and boggy and nonfrontal scalp hematomas had the strongest association with the presence of ICI in this large pediatric cohort. Although children 0 to 6 months of age were at highest odds, the presence of a scalp hematoma also independently increased the odds of ICI in older children and adolescents. The presence of a linear skull fracture only partially explained this relation, indicating that ruling out a skull fracture beneath a hematoma does not obviate the risk of intracranial pathology.