Clearance of the cervical spine in patients who have sustained trauma remains a contentious issue. Clinical examination alone is sufficient in neurologically intact patients without neck pain. Patients with neck pain or those with altered mental status or a depressed level of consciousness require further radiographic evaluation. However, no consensus exists as to the appropriate imaging modality. Some advocate multidetector CT (MDCT) scanning alone, but this has been criticized because MDCT is not sensitive in detecting ligamentous injuries that can often only be identified on MRI.
Patients were identified retrospectively from a prospectively maintained database at a Level I trauma center. All patients admitted between January 2004 and June 2011 who had a cervical MDCT scan interpreted by a board-certified radiologist as being without evidence of acute traumatic injury and who also had a cervical MRI study obtained during the same hospital admission were included. Data collected included patient demographics, mechanism of injury, Glasgow Coma Scale score at the time of MRI, the indication for and findings on MRI, and the number, type, and indication for cervical spine procedures.
A total of 1004 patients were reviewed, of whom 614 were male, with an overall mean age of 47 years. The indication for MRI was neck pain in 662 patients, altered mental status in 467, and neurological signs or symptoms in 157. The MRI studies were interpreted as normal in 645 patients, evidencing ligamentous injury alone in 125, and showing nonspecific degenerative changes in the remaining patients. Of the 125 patients with ligamentous injuries, 66 (52.8%) had documentation of clearance (29 clinical, 37 with flexion-extension radiographs). Another 32 patients were presumed to be self-cleared, bringing the follow-up rate to 82% (98 of 119). Five patients died prior to clearance, and 1 patient was transferred to another facility prior to clearance. Based on these data, the 95% confidence interval for the assertion that clinically irrelevant ligamentous injury in the face of normal MDCT is 97%–100%. No patient with ligamentous injury on MRI was documented to require a surgical procedure or halo orthosis for instability. Thirty-nine patients ultimately underwent cervical surgical procedures (29 anterior and 10 posterior; 5 delayed) for central cord syndrome (21), quadriparesis (9), or discogenic radicular pain (9). None had an unstable spine.
In this study population, MRI did not add any additional information beyond MDCT in identifying unstable cervical spine injuries. Magnetic resonance imaging frequently detected ligamentous injuries, none of which were found to be unstable at the time of detection, during the course of admission, or on follow-up. Magnetic resonance imaging provided beneficial clinical information and guided surgical procedures in patients with neurological deficits or radicular pain. An MDCT study with sagittal and coronal reconstructions negative for acute injury in patients without an abnormal motor examination may be sufficient alone for clearance.