It is estimated that approximately 800,000 cervical spine radiographs are obtained annually in the United States in patients who sustain blunt trauma. Although the vast majority of these radiographs are ultimately discovered to be normal, an unrecognized injury to the cervical spine can result in significant neurologic disability. Several study results have revealed that five specific criteria (no focal neurologic deficit, normal level of alertness, no evidence of intoxication, absence of posterior midline tenderness and no clinically apparent painful injury) had a sensitivity of 100 percent for ruling out cervical spine injury. However, the lower confidence limit for sensitivity of the criteria was 89 percent, which has been considered too low to be acceptable for general use of the five criteria. Hoffman and colleagues organized the National Emergency X-Radiography Utilization Study (NEXUS) in an attempt to validate these pre-determined criteria (the decision instrument).

This prospective, observational study involved the participation of 21 different centers in the United States, including university and community hospitals, and teaching and nonteaching hospitals. Persons were enrolled who presented to an emergency department with a history of blunt trauma. Persons who presented with penetrating trauma or those who had cervical imaging for any other reason unrelated to trauma were excluded. A standard three-view cervical spine series was obtained for all eligible patients. At the discretion of the examining physician, additional imaging studies could be ordered. The radiographs were formally read by designated radiologists at each of the study sites. A list of potential cervical spine injuries was created before the study began. A diagnosis of cervical spinal injury and the type of fracture was made according to the final interpretation of all imaging studies. All injuries were classified as clinically significant or insignificant. A clinically insignificant injury was defined as one that required no specific treatment and that, if not identified, would not result in harm to the patient. Radiographically documented cervical spine injuries were considered clinically insignificant only if they were isolated and no evidence of other bone injury or ligamentous or spinal cord injury existed. All other cervical spine injuries were considered clinically significant.

Of the 34,069 patients between one and 101 years of age who underwent cervical spine imaging following blunt trauma, 818 patients (2.4 percent) were found to have a radiographically documented injury. Among these patients, eight injuries were not detected by application of the five screening criteria (false negatives). However, two of these eight patients met the predetermined criteria for a clinically significant injury. Sensitivity approached 100 percent for clinically important injuries. Overall, it was retrospectively determined that by applying the five criteria, 4,309 patients (approximately 13 percent) would not have needed cervical spine radiographs.

The authors conclude that by applying these pre-determined criteria, physicians can accurately identify patients with blunt trauma who require cervical spine radiography. The overall rate of “missed” cervical spine injuries in this study was less than one in 4,000. Considering that the average full-time emergency department physician orders approximately 32 cervical spine films annually, a physician might encounter a single case of occult cervical spine injury once every 125 years of clinical practice. The authors believe that the application of this now-validated clinical instrument would result in significant clinical and economic benefit.