Comparison analysis between different predictive methods of height loss in adolescent idiopathic scoliosis patients

English Keywords:Adolescent idiopathic scoliosis  Height loss  Bjure method  Kono method  Ylikoski method  Stokes method

English Abstract:

  【Abstract】 Objectives: To compare the Bjure method, Kono method, Ylikoski method, and Stokes method in the prediction of height loss in patients with adolescent idiopathic scoliosis(AIS), and to evaluate their influencing factors. Methods: A total of 250 female AIS patients surgically treated in the Department of Pediatric Orthopedics, Gansu Hospital of Traditional Chinese Medicine from January 2016 to May 2020 were retrospectively reviewed. The age was 15.0±2.6 years(12-18 years) of the cohort. The patients were divided into single curve group and double curves group based on the curve type; And Cobb angle < 60° group and Cobb angle ≥ 60° group. The radiographic parameters including Cobb angle, spinal length(SL) and spinal height(SH) were measured, and the height loss(ΔH) was defined as SL minus SH. Using the fomulas of Bjure method, Kono method, Ylikoski method, and Stokes method to calculate height losses respectively, and comparing between the predicitive values of the 4 methods and real measurement with paired t test while applying Pearson coefficients for correlation analysis. Results: The average Cobb angle was 53.9°±15.8° and the value of ΔH was 4.4±2.8cm for the cohort. The height loss calculated from Bjure method, Kono method, Ylikoski method and Stokes method were 2.9±2.2cm, 1.9±1.3cm, 1.4±1.4cm and 2.5±1.8cm, respectively. The paired t test demonstrated significant differences between ΔH and the calculated height losses(P<0.001). Significant correlations were found between ΔH and the calculated height losses(P<0.001). The subgroup analysis showed significant differences between ΔH and the calculated height losses in single curve group, double curves group, Cobb angle of main curve ≥60° group and Cobb angle of main curve <60° group(P<0.05), and the calculated height loss by Bjure method was the closest to the real measurement value of ΔH in the Cobb angle≥60° group(5.2cm vs 5.9cm). Conclusions: None of the Bjure method, Kono method, Ylikoski method and Stokes method is able to predict the height loss in female AIS patients accurately, and the Bjure method seems to be relatively reliable in female AIS patients with Cobb angle of main curve ≥60° and single curve comparing with the other 3 methods.