The Sagittal Thoracic Modifier Component of the Lenke Classification System in Moderate and Severe Adolescent Idiopathic Scoliosis: Reliability and Reproducibility of Measurements Among 5 Observers.

Study Design: A prospective radiographic study.
Summary of Background Data: As the importance of the spinal sagittal profile becomes increasingly evident, there is a need to ensure that the measuring methods used to evaluate thoracic kyphosis (TK) are both accurate and reproducible.
Objective: The purpose of the following study was to determine the intraobserver and interobserver variability of measurements of the sagittal profile in moderate and severe thoracic scoliosis.
Methods: Five experienced Faculty Spine surgeons independently reviewed thirty standing long 30-inch cassette lateral radiographs of preoperative moderate and severe curves ≥50 degrees of adolescent idiopathic scoliosis (AIS) patients on 2 different occasions. The parameters measured were the vertebral endplate clarity and measurability of the sagittal angle from D5 to D12 and categories of thoracic sagittal modifier. κ statistics and Intraclass Correlation Coefficient (ICC) were used for analysis.
Results: The interobserver percentage of agreement for the Sagittal modifier was 58% in both trials. The mean κ coefficient value was only moderate 0.43 (range, 0.14-0.66) for both trials. The number of the vertebral endplates that were difficult to identify was 201 of 300 measurements (67%). There was a predominance of difficulty to identify vertebral endplate clarity in all curve types.
Conclusions: The results of this study yielded poor to moderate interobserver reliability of the thoracic sagittal profile component of the Lenke classification system in moderate and severe AIS. This was attributed to the difficulty in identification of the vertebral endplates. The current standard lateral radiographs routinely used in AIS patients have inherent difficulties and limitations to visualize, identify, and analyze the thoracic endplates in moderate and severe curves.