Related Experiment Video
Updated: Aug 6, 2026
Reliability of Artificial Intelligence-Based Cone Beam Computed Tomography Integration with Digital Dental Images
Published on: February 23, 2024
Comparison of Cervical Sagittal Alignment Between EOS and Conventional Standing Radiographs: Agreement, Variability,
Sereen Halayqeh1, Tarek Harhash1, Tomoyuki Asada1
1Hospital for Special Surgery.
Study Design:
Retrospective cohort study.
Objective:
This study aims to assess the agreement and differences between EOS imaging and conventional cervical radiographs (cXR) in measuring cervical alignment, and to identify factors associated with their differences.
Summary Of Background Data:
Accurate assessment of cervical spine alignment is essential for effective surgical planning. While cXR has been widely used for this purpose, the EOS imaging system provides a full-body, low-radiation alternative that captures global spinal alignment in a standing position. Despite its advantages, the agreement between EOS and cXR in evaluating cervical alignment parameters remains unclear.
Methods:
Patients who had both preoperative EOS and cXR images within a 6-month interval were included. The primary outcome measures included cervical alignment parameters in EOS and standing cXR, including C0-C2 Cobb angle, C2-C7 Cobb angle, C2-C7 sagittal vertical axis (SVA), and T1 slope. The humerus angle was measured on EOS imaging as the angle between the humerus and a plumb line to account for arm positioning during imaging. Bland-Altman analysis was used to evaluate the agreement between EOS and cXR. Multiple linear regression identified predictors of measurement differences of cervical alignment.
Results:
A total of 171 patients were included, with a mean age of 58 years, of which 56% were male. Significant differences were observed in C0-C2 Cobb (EOS, 27.5° versus cXR, 26.6°; P=0.020), C2-C7 SVA (20.9 mm versus 26.3 mm, P<0.001), and T1 slope (26.7° versus 25.2°, P=0.016), while no significant difference was observed for C2-C7 Cobb (EOS, 11.3° versus cXR, 10.4°; P=0.065). The Bland-Altman analysis suggested significant variability for C2-C7 SVA. Multiple linear regression identified humerus angle (B=-0.159, P=0.026), height (B=-0.153, P=0.036), and C2-C7 SVA on cXR (B=-0.515, P<0.001) as significant predictors of C2-C7 SVA measurement differences.
Conclusions:
EOS and cXR are generally interchangeable for cervical alignment assessment. However, C2-C7 SVA measurements require cautious interpretation due to variability stemming from the humerus angle and the patient's height.