Breaking the Rules in Three Dimensions: What to Expect After a Thoracic-only Fusion With Structural Thoracic and Thoracolumbar Curves

Abstract

Objective: Despite guidelines to fuse both thoracic and thoracolumbar/lumbar (TH/L) curves in patients with structural curves in both regions, a thoracic-only fusion allows preservation of lumbar motion segments. The purpose of this study was to assess the 2-year postoperative three-dimensional (3D) radiographic and clinical outcomes of patients with double or triple major (thoracic curves >TH/L curves) structural curves who underwent a thoracic-only fusion. Methods: A prospective adolescent idiopathic scoliosis registry was queried for double or triple major curves undergoing thoracic-only posterior fusion and a minimum 2-year follow-up. 3D reconstructions were generated from bi-planar radiographs. Paired sample t tests were used to assess differences in the coronal, sagittal, and axial planes pre and postoperatively, as well as Scoliosis Research Society Questionnaire–22 scores. Pearson correlations were utilized to identify variables related to spontaneous lumbar derotation. Results: Twenty-two patients met the inclusion criteria. Both thoracic [61 ± 10 degrees to 20 ± 9 degrees (P < 0.001)] and lumbar curves [41 ± 7 degrees to 22±7 degrees (P < 0.001)] had significant coronal improvement and T5 to T12 kyphosis improved from 7 ± 14 degrees to 23 ± 8 degrees (P < 0.001). The thoracic apical translation was significantly improved postoperatively (4.7 ± 1.5 to 0.5 ± 1 cm, P < 0.001), but the lumbar apical translation was unchanged (−1.7 ± 0.6 to −1.7±0.8 cm, P = 0.94). Scoliosis Research Society Questionnaire–22 scores significantly improved by 2 years postoperative. Conclusions: Unlike the 3D correction observed in nonstructural TH/L curves after thoracic-only fusion, patients with double or triple major curves demonstrated only spontaneous coronal correction of the lumbar curve, whereas the sagittal and axial planes were not significantly improved. These radiographic parameters did not negatively affect subjective or clinical outcomes at minimum 2-year follow-up. Level of Evidence: Level IV—therapeutic.