New sagittal classification of AIS: validation by 3D characterization.

INTRODUCTION AND AIM: In order to improve surgical planning of sagittal correction in AIS, we proposed a new sagittal classification-Abelin-Genevois et al. Eur Spine J (27(9):2192-2202, 2018. https://doi.org/10.1007/s00586-018-5613-1 ). The main criticism is related to the fact that 2D lateral view results from the projection of the 3D deformity. The aim of this study is to show that the new sagittal classification system is a reliable system to describe the different sagittal scenarios that AIS could create both in 2D and 3D.

METHODS

We performed retrospective radiograph analysis of prospectively collected data from 93 consecutive AIS patients who underwent an examination of the whole spine using the EOS® imaging system. 2D (Keops®) and 3D analyses (sterEOS®) provided frontal and sagittal spinal and spinopelvic parameters. In addition, 3D analysis provided apical vertebra rotation (AVR).

RESULTS

Comparing 2D and 3D measurements for the general cohort, excellent correlation can be found for all parameters, but only fairly good for T10L2 and L1S1 angles. The highest variability was observed for T10L2, differences between 2D and 3D measurements being greater when the Cobb angle increased. AVR did not influence concordance between 2D and 3D measurements. Eighty-two percent were similarly classified in 2D and 3D according to the new classification. Misclassified patients were all AIS sagittal type 3 in 3D analysis, thoracolumbar junction (TLJ) lordosis being underestimated on 2D view.

DISCUSSION: In conclusion, for the majority of cases (82%), 2D analysis may provide enough information for decision making when using a semi-automated 2D measurement system. However, in severe cases, especially when Cobb angle exceeds 55°, 3D analysis should be used to get a more accurate view on the thoracolumbar junction behavior. These slides can be retrieved under Electronic Supplementary Material.