Analysis of cervical and global spine alignment under Roussouly sagittal classification in Chinese cervical spondylotic patients and asymptomatic subjects

To explore the relationship between cervical spine and the global spine alignment and to postulate the hypotheses that a lordotic alignment of cervical spine is not the only standard to identify asymptomatic subjects, and the degenerative modification of cervical curves depends primarily on their spinal-pelvic alignment. A cohort of 120 cases of Chinese asymptomatic subjects and a cohort of 121 cases of Chinese cervical spondylotic patients were recruited prospectively from 2011 to 2012. Roussouly Classification was utilized to categorize all subjects and patients according to their thoracic spine, lumbar spine and pelvic alignment. The cervical alignments were evaluated as lordosis, straight, sigmoid or kyphosis. Through the lateral X-ray images of neutral cervical and global spine, a number of parameters were measured and analyzed, including pelvic incidence, pelvic tilt, sacral slope, thoracic kyphosis (TK), lumbar lordosis, global cervical angles (angles between two lines parallel with posterior walls of C2 and C7), practical cervical angles (the addition of different cervical end plate angles from C3 to C7, and inter-vertebral angles from C23 to C67), T1 slope, spinal sacral angles (SSA), Hip to C7/Hip to Sacrum and C0-C2 angle. The percentages of cervical lordosis were 28.3 % and 36.4 % in asymptomatic and spondylotic group, respectively. The cervical spine alignments correlated with Roussouly types of global spine alignment in both asymptomatic and cervical spondylotic group (P < 0.001). And there were significant differences between Roussouly Type 2 and 4, Type 3 and 4, Type 1 and 3 in cervical angles in spondylotic group (P < 0.05). In the comparison of the two cohorts, significant differences were found in both general and practical cervical angles in Roussouly Type 4 (P = 0.00 and 0.01, respectively), and there were significant differences in inter-vertebral angle in Roussouly Type 2 at C4-5 and C5-6 levels (P = 0.04 and 0.04, respectively), and in Roussouly Type 3 at C6-7 level (P = 0.01). The SSA showed significant difference between Roussouly Type 2 and 4 in asymptomatic subjects (P = 0.00), and between Type 1 and 3, 1 and 4, 2 and 3, 2 and 4 in cervical spondylotic patients (P = 0.01, 0.02, 0.00 and 0.01, respectively). The T1 slope was significantly different among Roussouly types (P = 0.04) with its largest value in Type 1 in cervical spondylotic group. There are significant differences in C0-C2 angles in all Roussouly types (P = 0.01, 0.02, 0.00 and 0.01, respectively), as well as in the ratio of Hip to C7/hip to sacrum in Type 2 (P = 0.01), and Type 3 (P = 0.00) in the comparison of the two cohorts. The multiple linear regression of all parameters showed both general and practical cervical angles were significantly related to TK, C0-C2 and T1 slope (P = 0.01, 0.00 and 0.00, respectively). The cervical alignment correlates with their global spine and pelvic curves. And lordosis is not the only presentation in asymptomatic subjects. The degenerative modification of cervical disc angles was the compensation of global spine degeneration for horizontal gaze. Cervical angles are influenced by their TK angles, occipital-C2 joint and the tilt of T1 vertebral body. The occipital-C2 joint has a compensating mechanism in all Roussouly types in cervical spondylosis.