Variability in computed tomography and magnetic resonance imaging in patients with cervical spine injuries

The purpose of this study was to describe the performance of adjunctive radiologic imaging in patients with cervical spine injury.

All patients undergoing cervical spine radiography were prospectively enrolled at 16 diverse emergency departments. We recorded the imaging modalities and radiographic interpretations rendered by unblinded faculty radiologists at each center. Only patients with cervical spine injury were included in this analysis. Findings revealed by individual modalities were compared with the final diagnosis (after all evaluations) in each patient.

Six hundred eighty-eight patients with 1,302 separate cervical spine injuries were enrolled. Four hundred seventy-six (69%) patients had magnetic resonance imaging (MRI) and/or computed tomography (CT) of the cervical spine. MRI identified the following injuries among 124 imaged patients: osseous fractures, 85 of 154 (55%); spinal cord injury, 69 of 69 (100%); vertebral subluxation/dislocation, 37 of 43 (86%); ligamentous injury, 38 of 38 (100%); and unilateral/bilateral locked facets, 14 of 18 (78%). Among 418 patients undergoing CT, the following injuries were identified: osseous fractures, 721 of 740 (97%); spinal cord injury, 0 of 30 (0%); vertebral subluxation/dislocation, 76 of 88 (86%); ligamentous injury, 9 of 36 (25%); and unilateral/bilateral locked facets, 34 of 35 (97%). CT identified 29 patients with fractures who had normal plain radiographs. Cervical myelograms were obtained in two patients and cervical tomograms in seven patients.

The majority of patients with cervical spine injury undergo MRI and/or CT imaging. In clinical practice, MRI is superior at identifying soft tissue injuries, whereas CT performs better in identifying bony injuries. Cervical myelograms and tomograms are rarely obtained.