Spino-cranial angle as an alternative to T1 slope for predicting loss of cervical lordosis after laminoplasty: A retrospective cohort study

The T1 slope (T1s) has been extensively studied and is recognized as a predictor of loss of cervical lordosis (LCL). However, the visibility of the T1 vertebral body is often obscured by rib shielding, complicating the application of T1s. This study evaluates the spino-cranial angle as a potential alternative predictor instead of T1s for predicting LCL post-laminoplasty. Adults undergoing C3 to C6 laminoplasty for degenerative myelopathy were eligible. Exclusions included prior cervical surgery, combined anterior/posterior procedures, trauma, infection, tumor, inflammatory disease, or inadequate standing lateral radiographs. Cervical lordosis was measured to assess cervical alignment, defining LCL as any postoperative change in alignment exceeding 0°. Candidate variables entered multivariable modeling based on univariable screening (P < .05) and clinical relevance; SCA and T1s were modeled simultaneously after multicollinearity testing. This analysis was complemented by comparing linear regression models of preoperative SCA and T1s in relation to LCL. The ability of preoperative SCA and T1s to predict LCL was further evaluated using receiver operating characteristic curves. Based on the optimal thresholds of preoperative SCA, patients were categorized into 2 groups to compare postoperative cervical alignment changes at the 2-year follow-up. One hundred eight consecutive patients with cervical myelopathy who underwent laminoplasty met the criteria and were included in the study. Multivariate logistic regression analysis identified both lower SCA and higher T1s as independent predictors of LCL. A negative correlation was observed between preoperative SCA and LCL (r = -0.597, P < .001), and a positive correlation existed between preoperative T1s and LCL (R = 0.493, P < .001). Although both SCA and T1s predicted LCL effectively (SCA ≤ 80.5°, P = .001; T1s > 26.1°, P = .007), no significant difference in diagnostic value was found between them (P = .644). Based on the optimal preoperative SCA cutoff value (80.5°), patients were categorized into 2 groups. Those with lower SCA demonstrated more lordotic preoperative cervical alignment and greater LCL at 2 years post-laminoplasty. In cases where the T1 vertebral body is obscured, SCA proves to be an effective predictor of LCL, with diagnostic accuracy similar to that of T1s. The findings further indicated that patients with a preoperative SCA ≤ 80.5° showed more pronounced preoperative cervical lordosis and underwent greater LCL post-laminoplasty after 2 years.