Are five-view plain films of the cervical spine unreliable? A prospective evaluation in blunt trauma patients with altered mental status

Objectives: Clearing the cervical spine in a time-sensitive fashion is difficult. We hypothesized that admission computed tomographic scan of the occiput to T1 (CTS) with multiplanar reformatted images will replace five-view (odontoid, anteroposterior, lateral, and oblique) plain films of the cervical spine (CSX) in the initial evaluation of blunt trauma patients with altered mental status. Methods: Between January and July 2001, all patients aged 16 years or older with altered mental status undergoing both CTS and CSX were prospectively entered into the study group. Attending physician interpretation defined the presence of cervical spine injury. Unstable fractures were defined as requiring surgical or halo stabilization. Results: One thousand six patients met study criteria. One hundred sixteen patients had 172 cervical spine injuries (CSIs) (fracture and subluxation). CSX missed 90 of 172 (52.3%) CSIs in 65 of 172 (56.0%) patients. Anatomically, CSX failed to identify 14 of 15 occipital fractures (93.3%), 17 of 36 (47.2%) C1-3 fractures, and 59 of 121 (48.8%) C4-T1 CSIs. CSX failed to identify 5 of 29 (17.2%) patients with unstable CSIs. CTS failed to diagnose 3 of 172 (1.7%) CSIs that were stable (spinous process fractures at C6-7). Two patients exhibited spinal cord injury without radiologic abnormality missed by both modalities. CTS had a sensitivity of 97.4%, a specificity of 100%, a prevalence of 11.5%, a positive predictive value of 100%, and a negative predictive value of 99.7%. CSX had a sensitivity of 44.0%, a specificity of 100%, a prevalence of 11.5%, a positive predictive value of 100%, and a negative predictive value of 93.2%. Conclusion: CTS outperformed five-view CSX in a group of patients with altered mental status or distracting injuries. Five-view CSX failed to diagnose 52.3% of cervical spine fractures identified by CTS. Five-view CSX failed to diagnose five patients with unstable cervical fractures and failed to identify 93.3% of patients with occipital condyle fractures.