The validity of Lenke criteria for defining structural proximal thoracic curves in patients with adolescent idiopathic scoliosis

Design. Retrospective series. Objective. To delineate the efficiency of using Lenke criteria during the decision of whether to include the proximal thoracic curve into instrumented fusion or not in patients with adolescent idiopathic scoliosis (AIS) treated with posterior translational instrumentation. Summary of Background Data. Lenke and colleagues classified the curves of patients with AIS and assigned the term structural or nonstructural to each curve. However, there is still not much consensus on the definition of structural proximal thoracic curve, and structurality criteria for proximal thoracic curve have not been validated yet. Methods. Inclusion criteria were: (1) patients with AIS treated with a posterior translational instrumentation system and older than 10 years; (2) nonstructural upper thoracic curves (side bending Cobb < 25 degrees and T2-T5 kyphosis <+20 degrees); and (3) a minimum of 2 years of follow-up. A total of 37 consecutive patients, including 6 males and 31 females, with AIS that was treated, with an average age of 15 years (range 11-24) and average follow-up of 55 months (range 24-90) were studied. For radiologic evaluation of the patients, preoperative, postoperative, and latest follow-up radiographs were used. There were 2 groups constructed according to the involvement of proximal thoracic curve into instrumented fusion. Group 1 included those patients who underwent uppermost extent of the instrumentation, either T2 or T3, indicating inclusion of proximal thoracic curve into instrumentation. Group 2 included individuals who underwent uppermost extent of the instrumentation at T4 or lower, indicating partial or no inclusion of the proximal thoracic curve into instrumented fusion. Radiographic evaluation included the measurement of proximal thoracic, main thoracic, and thoracolumbar-lumbar curves. On the lateral radiographs, sagittal Cobb angles of T2-T5, T5-T12, and T10-L2 were measured. Difference between right and left coracoid process heights, clavicle angle, and T1 tilt were determined for assessment of shoulder balance. Results. The 2 groups were statistically equivalent in terms of age at surgery, follow-up time, preoperative proximal thoracic and main thoracic, and their corresponding side bending curve magnitudes, as well as the parameters related to shoulder balance before surgery (P > 0.05). The 2 groups were also statistically equivalent regarding immediate postoperative and latest follow-up proximal thoracic and main thoracic, and parameters related to shoulder balance (P > 0.05). Conclusion. It was observed that the Lenke description for structurality of proximal thoracic curves can effectively determine which curves need fusion and which curves do not. Because there was no difference among inclusion of a nonstructural proximal thoracic curve into fusion or solely fusing the main thoracic curve in terms of outcomes, extension of fusion to T2 or T3 is unnecessary. In this patient population, the question of what the upper extent of instrumentation should be could not be answered.