Purpose: The Lenke classification has established criteria which designate the proximal thoracic (PT) curve as structural (Lenke 2). However, this classification may overestimate the necessity to include the proximal thoracic curve in the fusion construct. The objective of this study was to compare the incidence of fusing the PT curve for true Lenke 2 curve patterns, comparing a time period when the classification was not utilized and when it was first used.

Method: A retrospective review of a consecutive series of patients with adolescent idiopathic scoliosis (AIS) at a single institution from 1996–2000 (early group) and from 2002–2004 (late group) were reviewed. The curves were all classified by the Lenke classification retrospectively. Patients were also grouped into those who have had inclusion of the PT curve (+PT fusion) and those who did not (−PT fusion).

Results: There were 44 in the early group and 33 in the late group. There were no differences in the early and late groups with respect to age (14.3 vs. 14.4yrs), gender (79.5% vs. 69.7% female), BMI (21.7 vs. 22.4kg/cm2), the preoperative PT magnitude (40.0° vs. 38.6°), curve flexibility (16.0% vs. 14.5%), the main thoracic (MT) magnitude (63.4° vs. 62.7°), T1 tilt (7.3° vs. 5.2°), pre-operative clavicle angle (1.0° vs. −0.2°), and preoperative shoulder height (1.2 vs. −0.8mm. The early group had fusion of the PT less often (36% vs. 57%)(p< 0.05) which resulted in a greater residual PT curve (26.5 vs. 22.2°), MT curve (33.8 vs. 27.8°), and a greater clavicle angle (4.6 vs. 2.5°)(p< 0.05). At two years the PT continued to be significantly greater in the early group (28.6 vs. 22.8°)(p< 0.05), however T1 tilt (8.8 vs. 8.1°), clavicle angle (1.3 vs. 1.0°), and shoulder height (3.5 vs. 4.7mm) were the same.

Conclusion: The application of the Lenke classification system for AIS increases the likelihood of instrumenting a PT curve resulting in improved postoperative PT curve magnitude. However, similar shoulder balance, T1 tilt and clavicle angle were seen compared between groups. The classification system should be used to guide decision making, however, other parameters should be evaluated when deciding when to instrument a PT curve.