Sagittal balance measures are more reproducible when measured in 3D vs in 2D using full-body EOS ^registered images

An innovative low-dose X-ray biplanar imager (EOS ^registered) allows measurement of the whole-body in standing-position which is necessary for the evaluation of spinal deformities. A total of 60 asymptomatic subjects (ages 20-81 years) were evaluated using the 3D workflow called postural assessment and 2D measures. Subjects were measured twice each by two new observers following training, including: lordosis/kyphosis, pelvic parameters, sagittal-vertical axis, and spinal-sacral angle. Intra- and inter-observer reproducibility and similarity were compared between 2D and 3D measures. The intraclass correlation coefficient (ICC) was very high for the 3D measures (>0.9) and excellent for the 2D measures (>0.75). In all cases, the overall mean absolute difference between repeated 3D measures was less than 2 , or 2 mm. For all parameters, the inter- and intra-observer reproducibility in 3D measures were significantly superior to 2D measures (p < 0.03). This study demonstrated that 3D measures have better reproducibility than 2D for sagittal balance. (orig.)