The effect of spinal deformity on cervical spine (C-spine) alignment has been a topic of recent interest. Several authors have proposed standardized radiographic sagittal alignment parameters for the C-spine. These papers recommend the use of full-length 36-inch spine radiographs. However, full-length spine radiographs often produce poor images of the cervical region, making evaluation difficult. Routine use of full-length spine radiographs also increases cost and patient exposure to radiation. Casey L. Smith, MD, Grand Rapids Medical Education Partners, Grand Rapids, Michigan, USA, presented the Correlation of Cervical Sagittal Alignment Parameters on Full-Length Spine Radiographs Compared with Dedicated Cervical Radiographs study [Smith CL et al. Spine. 2014].

In this retrospective cross-sectional study from a single tertiary spine practice, a billing database was used to identify patients who had both dedicated C- and full-length spine radiographs within a period of 4 months. Sets of radiographs from single patients were measured by one reviewer. Each radiograph set consisted of a dedicated lateral C-spine radiograph and a lateral full-length spine radiograph. Radiograph sets were excluded if the sagittal parameters on 1 or both images could not be accurately measured.

The outcome measures were the following sagittal parameters: C0-C2, C0-C1, C0-C7, C1-C2, and C2-C7 Cobb angles; C1-C7, C2-C7, and CGH-C7 sagittal vertical axis (SVA) differences; T1 tilt; and chin-brow vertical angle. Paired t tests and correlation analyses were performed for the corresponding paired radiographs from each patient. A marked correlation was defined as correlation coefficients between 0.60 and 0.80. A robust correlation was defined as correlation coefficients between 0.8 and 1.0.

Radiograph sets were collected from 40 patients. There were 33 women and 7 men. The mean patient age was 48.9 ± 14.5 years. All of the cervical sagittal alignment parameters measured were significantly correlated between the full-length and dedicated C-spine radiographs (P < .001). The sagittal Cobb angle correlation coefficients ranged from 0.62 to 0.81. The SVA difference correlation coefficients ranged from 0.42 to 0.65. Paired t tests showed that only the C0-C7 angle and T1 tilt angles were significantly different between the full-length and C-spine radiographs (P = .000, both).

There were no statistically significant differences between the full-length and C-spine radiographs in C1-C7 SVA and CGH-C7 SVA. The full-length and C-spine radiograph measurement differences were statistically significant for the C2-C7 SVA (P = .049).

The results of this study are limited by the small sample size. Although > 100 pairs of radiographs were identified, many were excluded due to poor C-spine visualization on the full-length spine radiograph.

Strong correlations were observed between most cervical sagittal measurements taken from dedicated C-spine radiographs and full-length spine radiographs. Most of the measurements were similar between the pairs of radiographs. Dr Smith concluded that a 36-inch full-length spine radiograph may not be necessary for evaluation of C-spine sagittal alignment when a C-spine radiograph has already been obtained.