Purpose: Evidence guiding the use of CT and MRI scans in blunt trauma patients who are obtunded remains controversial. This study aims to determine and predict if computed tomography (CT) scans alone can be performed without risking oversight of substantial injuries found on follow-up magnetic resonance imaging (MRI).
Methods: This is a retrospective cohort study of 63 blunt trauma patients with a Glasgow Coma Scale of < 8. Data were collated from electronic medical records and included patient demographics, premorbid mobility, mechanism of injury, suspected level of injury and neurological examination findings. Patients were urgently evaluated using CT scans, followed by non-contrast MRI scans within 48 h of admission. The accuracy of CT scan was evaluated using MRI as a reference. Adjusted multivariable analysis was also performed to identify predictors for findings detected on MRI but not on CT.
Results: The mean age of patients was 42.3 years and 90.5% were males. CT scans had a high specificity of 100% and sensitivity of 87.2%. Predictors of MRI abnormalities include females, patients with relatively milder mechanisms of injury, patients with suspected thoracic spine injury, and CT scan findings of facet dislocation and intracranial haemorrhage. There was no predictor for spinal cord oedema.
Conclusions: MRI should be performed in the presence of the aforementioned predictive factors and in the presence of neurological deficits. Otherwise, patients can be treated medically without the fear of missing a substantial cervical injury.
Keywords: Cervical spine; Computed tomography; Magnetic resonance imaging; Obtunded patients; Predictors.