STUDY OBJECTIVE: We sought to describe and measure 3 radiographic variables in normal male volunteers and determine whether these variables could be useful in establishing more objective radiographic criteria for evaluation of flexion-extension studies of the cervical spine. In addition, we hypothesized that patients with a normal cervical spine should not have greater than 2 mm of subluxation present with flexion or extension. METHODS: A prospective, observational study of normal male volunteers between the ages of 18 and 40 years was performed. We obtained radiographs of all participants in neutral, flexion, and extension positions and measured the amount of subluxation and interspinous distance, as well as the degree of vertebral angulation between C3 and C7. RESULTS: One hundred male volunteers were included in the study. Subluxation during flexion (compared with neutral) was greater than 2 mm in none of the participants at each level from C3 to C7 (95% confidence interval [CI] 0 to 3.6); subluxation in extension was greater than 2 mm in one participant at one level from C3 to C4 (95% CI 0 to 5.5) and none of the participants at each of the remaining 3 levels from C4 to C7 (95% CI 0 to 3.6). Comparing flexion with the neutral position, the mean vertebral angulation and SD were 24.2 degrees and 13.3 degrees, and the 95% certainty interval (CEI) was -1.9 to 50.2. Comparing extension with the neutral position, the mean vertebral angulation was 14.1 degrees, with an SD of 13.3 degrees, and the 95% CEI was -4.6 to 32.8. The mean change in interspinous distance between the neutral position and flexion varied from 1.2 to 4.6 mm (SD 1.7 to 2.4 mm), depending on the level of cervical spine studied. When comparing the neutral position and extension, the mean change in interspinous distance varied from 1.2 to 2.2 mm (SD 1.7 to 1.9). CONCLUSION: Currently, there are no clinically validated criteria for evaluating flexion-extension studies of the cervical spine. Our study suggests that subluxation greater than 2 mm in men 18 to 40 years of age may be a useful variable for further study as an indicator of ligamentous injury. Interspinous distance and vertebral angulation appear less likely to have useful clinical application.