Objective: To compare the effects in assessing the curve flexibility of the adolescent idiopathic scoliosis (AIS) and predicting the outcomes of operation among different radiological techniques: supine lateral bending (SB), traction (Tr), and fulcrum bending radiographs.
Methods: 68 consecutive AIS patients, all with the single-curve types Ia/Ib/Ic according to the PUMC classification, divided into 4 groups according to the magnitude of Cobb's angle: moderate thoracic curve (n = 19, 40 degrees < Cobb's angle < or = 60 degrees ), severe thoracic curve (n = 13, Cobb's angle > 60 degrees ), moderate lumbar curve (n = 28, 35 degrees < Cobb's angle < or = 60 degrees ), and severe lumbar curve(n = 8, Cobb's angle > 60 degrees ) who were treated surgically underwent preoperative radiological evaluation including standing anteroposterior and lateral Tr, SB, and fulcrum bending radiographs. COBB angle was measured and the flexibility ratio was determined on each radiograph. The amounts of correction obtained by all radiographic methods were compared with the amount of surgical correction.
Results: The post-operative Cobb's angle of the moderate thoracic curve group was 9 degrees , not significantly different from that by fulcrum bending radiograph (P = 0.076), but significantly different from those by the other methods (both P < 0.01). The post-operative COBB angle of the severe thoracic curve group was 40 degrees , significantly different from all the radiographs before operation (all P < 0.01). The post-operative Cobb's angle of the moderate lumbar curve group was 4 degrees , significantly different from those by fulcrum bending and Tr radiographs (both P < 0.01) and that by SB (P = 0.013). The post-operative Cobb's angle of the severe lumbar curve group was 24 degrees , significantly different from those of anteroposterior and Tr radiograph (both P < 0.01) and those of fulcrum-bending and SB radiographs (P = 0.021 and P = 0.011). In the moderate thoracic curve group the operation correction rate was not significantly different from the flexibility rate by fulcrum-bending radiograph (P = 0.111), and was significantly different from the flexibility rates by SB and Tr radiographs (P = 0.011 and P = 0.000). In the severe thoracic curve group the operation correction rate was significantly different from the flexibility rates by different kinds of radiograph (all P = 0.111). In the moderate lumbar curve group the operation correction rate was significantly different from the flexibility rates by different kinds of radiograph (P < 0.111 or P = 0.019). In the severe lumbar curve group the operation correction rate was significantly different from the flexibility rates by different kinds of radiograph (P < 0.01 or P = 0.017).
Conclusion: Fulcrum-bending radiography can better assess the flexibility and correction rate of thoracic curves in AIS, however, it can only predict those in moderate thoracic curves. Fulcrum-bending radiograph and SB radiograph are similar in predicting the flexibility in lumbar curves.