Background/objective: Conventional neuroimaging is not recommended for the routine diagnosis of concussion, but some patients with concussion undergo computed tomography (CT) or magnetic resonance imaging (MRI). The objective of this study was to explore the clinical factors that predict neuroimaging utilization in concussion.
Methods: Concussion-related CT and MRI data were analysed from 1953 patients, aged 10-19 years, who presented to a sports concussion clinic within 30 days of injury.
Results: The majority of CT scans (n = 193) were obtained during the acute concussion period (mean = 2.7 days post-concussion), whereas MRI scans (n = 134) were ordered later during recovery (mean = 39.4 days post-concussion). Predictors of CT utilization included loss of consciousness, amnesia and vomiting (all p < 0.001). Prior concussion (p = 0.002) and continued participation in activity after injury (p = 0.03) predicted greater MRI utilization. Neuroimaging with either CT (p = 0.024, hazard ratio = 1.2) or MRI (p < 0.001, hazard ratio = 2.75) was associated with prolonged symptoms. Only 3.1% of CTs and 1.5% of MRIs demonstrated signs of traumatic brain injury.
Conclusion: Several clinical factors predict neuroimaging utilization in patients with concussion. CT is generally used acutely, while MRI is used in the sub-acute and chronic post-concussion periods. In a sports concussion clinic, delayed neuroimaging has limited clinical yield.
Keywords: CT; MRI; Mild traumatic brain injury; paediatric; post-concussion syndrome.