Intraobserver Reliability And Interobserver Reproducibility Of The King, Lenke And PUMC Classification Systems Of Adolescent Idiopathic Scoliosis

Objective: To determine the reliability and reproducibility of the King, Lenke and PUMC classification systems for idiopathic scoliosis using radiographs that had not been premeasured. Methods: Preoperative radiographs (standing full-length posteroanterior, lateral and two supine side-bending radiographs) of 56 patients were evaluated by four orthopedic surgeons independently on two separate occasions. The results were determined by calculating the average percentage of intraobserver and interobserver agreement. Reliability and reproducibility was quantified using kappa statistics. Results: The King classification demonstrated good to excellent intraobserver reproducibility and fair interobserver reliability. The mean interobserver reliability was 65.8% (mean Kappa coefficient, 0.542), while intraobserver reproducibility was 82.6% (mean Kappa coefficient, 0.767). The main reason of disagreement was distinguishment of the King type II and type III. Another reason is assessment of King type V. All three parameters of the overall Lenke curve classification demonstrated poor reliability. The mean interobserver reliability was 50.0% (mean Kappa coefficient, 0.438), while intraobserver reproducibility was 47.0% (mean Kappa coefficient, 0.402). The three parameters had fair interobserver reliabilities and intraobserver reproducibilities when it was examined separately. The main reasons for disagreement arose from judging a structural upper thoracic curve and from assigning of sagittal thoracic modifier. The PUMC type demonstrated good to excellent intraobserver reproducibility and interobserver reliability, and the subtype demonstrated fair. The mean interobserver reliability was 70.2% (mean Kappa coefficient, 0.629), while intraobserver reproducibility was 74.1% (mean Kappa coefficient, 0.674). The main reasons for disagreement were definition of a curvature and Cobb angle measurements. Conclusion: This three classifications do not appear to eliminate the dispute for the definition of the upper thoracic curve. The main reason of lower reliability and reproducibility of the Lenke and PUMC classification systems can be attribute to the variable of Cobb angle measurements. Orthopedic surgeons should be think over the effect of the factors mentioned above, when make the surgical decisions.