Angle measurement reproducibility using EOS three-dimensional reconstructions in adolescent idiopathic scoliosis treated by posterior instrumentation.

STUDY DESIGN: A reproducibility study was conducted in preoperative and postoperative three-dimensional (3D) measurements for patients operated for adolescent idiopathic scoliosis (AIS).

OBJECTIVE: To assess the reliability of preoperative and postoperative 3D reconstructions using EOS in patients operated for AIS.

SUMMARY OF BACKGROUND DATA: No prior reliability study of 3D measurements has been performed in the literature for severe scoliosis and for operated patients.

METHODS: This series included 24 patients (62° ± 11) operated for Lenke 1 or 2 AIS, using either all-pedicle screw constructs (group 1) or hybrid constructs, with universal clamps at thoracic levels (group 2). All patients underwent low-dose standing biplanar radiographs, pre- and postoperatively. Three operators performed the 3D reconstruction process two times preoperatively and two times postoperatively (total 288 reconstructions). Intraoperator repeatability and interoperator reproducibility were calculated and compared between groups.

RESULTS: The preoperative reproducibility was between 4° and 6.5° for parameters dedicated to scoliosis (Cobb and apical vertebral rotation), between 4° and 7° for kyphosis and lordosis values, and between 1° and 5° for pelvic measurements. The postoperative reproducibility was between 5° and 8° for values of kyphosis and lordosis, between 1° and 5.5° for pelvic parameters, and between 6.5° and 10.5° for the scoliotic parameters. The reproducibility of the scoliotic parameters was slightly better in the hybrid construct group, but the difference was not significant (P = 0.8). No difference was found between groups for the other parameters.

CONCLUSION: 3D postoperative reconstructions are as reproducible as preoperative ones. The reproducibility is not influenced by the type of implant used for correction. Mean difference between operator was higher than previously reported for the apical rotation measurement, but this difference can be explained by the severity of the curves and the lower visibility of the anatomical landmarks due to the implants.