BACKGROUND Flexibility evaluation methods were only used to assess the changes of coronal Cobb angle in patients with adolescent idiopathic scoliosis (AIS). Little attention was attached to the vertebral rotation in these methods. MATERIAL AND METHODS 21 patients with severe adolescent idiopathic scoliosis were enrolled in this study. Coronal flexibility and rotation correction were compared on the supine bending, traction and fulcrum bending radiographs. The apical vertebral body rib ratio (AVB-R), and Perdriolle rotation angles were used to measure the rotation of the main thoracic curve. Statistical analysis was performed with one-way Analysis of Variance(ANOVA). Correlations between postoperative AVB-R and AVB-R in supine bending, traction and fulcrum bending radiographs were assessed utilizing the Linear Regression. RESULTS There were trends towards increased coronal flexibility in fulcrum bending versus traction versus supine bending, but there were no significant differences due to the limited sample size. And all were significantly lower than postoperative correction. The correction of AVB-R at traction and supine bending radiographs were significantly better than fulcrum bending, however, all were significantly lower than postoperative correction. Correction of Perdriolle rotation angle at traction radiograph was best among these methods. A univariant linear regression analysis showed a strong linear correlation between the postoperative AVB-R and the AVB-R in the traction radiograph. CONCLUSIONS As to patients with severe AIS, the coronal plane flexibility evaluated at the fulcrum bending radiograph is superior to that at the traction radiograph. This may be explained by the measurement errors induced by the better derotation capacity at the traction radiograph. Rotation correction evaluated at the traction radiograph proves better than the fulcrum bending radiographs, showing a linear correlation with the postoperative correction.