The utility of magnetic resonance imaging in pediatric trauma patients suspected of having cervical spine injuries

BACKGROUND Pediatric cervical spine injuries (CSI) are rare but potentially devastating sequelae of blunt trauma. Existing protocols to evaluate children at risk for CSI frequently incorporate computed topography (CT) and magnetic resonance imaging (MRI); however, the clinical value of performing both remains unclear. METHODS Single-center retrospective review of pediatric trauma patients who underwent both CT and MRI of the cervical spine between 2001 and 2015. Based on radiographic findings, CT and MRI results were grouped into one of three categories: no injury, stable injury, or unstable injury. Radiographic instability was defined by disruption of two or more contiguous spinal columns while radiographic stability was defined by any other acute cervical spine abnormality on imaging. Clinical instability was defined by the need for surgical intervention (halo or spinal fusion), with the remaining patients, including children discharged in a cervical collar, considered clinically stable. RESULTS In total, 221 children met inclusion criteria, with a median age of 9 (interquartile range, 3-14). The Glasgow Coma Scale (GCS) score of the cohort was 9 (interquartile range, 4-15). Thirty-three (14.9%) children had clinically unstable injuries, requiring surgical intervention. Among the 160 (72.4%) children with no injury on CT, MRI identified no injury in 84 (52.5%) cases, a stable injury in 76 (47.5%) cases, and an unstable injury in none. Among the 21 children with stable injuries on CT, MRI findings were concordant in 17 (81.0%) cases. In four (19.0%) cases, a spinal column injury was identified on CT and appeared to be stable, but later deemed unstable on MRI. Forty (18.1%) patients had an unstable injury on CT with 100% MRI concordance. CONCLUSION In pediatric trauma patients suspected of having a CSI, a normal cervical spine CT is sufficient to rule out a clinically significant CSI as no child with a normal cervical CT was found to be radiographically or clinically unstable. LEVEL OF EVIDENCE Diagnostic Test, level III.