Antagonistic role of vertebral translation against vertebral rotation in the spontaneous postoperative modulation of the anterior chest wall contour in thoracic idiopathic scoliosis

STUDY DESIGN A computed tomography study. OBJECTIVE To identify the best scoliotic deformity components that show impact upon the spontaneous postoperative modulation of the deformed anterior chest wall contour in right convex thoracic adolescent idiopathic scoliosis. SUMMARY OF BACKGROUND DATA Spontaneous postoperative aggravation of the anterior concave costal projection was a common occurrence in adolescent idiopathic scoliosis, yet the risk factors that effectively bridged the gap between what the surgeons did in the interior and how the rib cages reacted on the exterior were still open to debate. METHODS Pre- and postoperative computed tomographic scans of 77 patients with right convex thoracic adolescent idiopathic scoliosis were retrieved and analyzed. According to the postoperative variation of anterior chest wall angle (CWA), the patients were divided into 2 groups with either aggravated or improved CWA. Multiple scoliotic deformity parameters and their surgical correction rates were evaluated, correlated, and then compared between the 2 groups. Moreover, patients with apex located at T9 were isolated and evaluated independently. A logistic regression analysis was used to determine the independent predictors of the spontaneous postoperative modulation of the anterior chest wall contour. RESULTS The surgical correction rate of Cobb angle (supine), the rotational angle with respect to the sagittal plane (RAsag angle), the rotational angle with respect to the anterior midline of the body (RAml angle), the angle of lateral deviation of the apical vertebrae from the midline (MLdev angle), the posterior hemithorax ratio, the vertebral translation (VT), and the thoracic rotation averaged 64.6%, 19.5%, 30.8%, 39.2%, 15.0%, 41.2%, and 28.7%, respectively. Ratio of aggravated anterior chest wall contour was the highest at the T7 apex group (84.6%) as compared with T8 apex group (47.1%), T9 apex group (19.5%), and T10 apex group (0.0%). The preoperative CWA was significantly lower in the aggravated CWA group when compared with the improved group (2.1 ± 1.8°vs. 6.6 ± 2.4°, P < 0.001). Besides, in the aggravated CWA group, significantly greater surgical correction of VT and lesser correction of RAsag angle were demonstrated when compared with the improved CWA group (VT: 53.0% vs. 34.8%, P = 0.001; RAsag: 2.5% vs. 28.7%, P = 0.000). In the T9 subgroup, remarkably different correction rate of VT and RAsag were similarly observed (VT: 54.9% vs. 35.3%, P = 0.046; RAsag: 4.9% vs. 23.5%, P = 0.034). In terms of other deformity parameters, no significantly different correction rate was consistently detected. In the logistic regression analysis, apex location, CWA, and correction rate of RAsag were demonstrated to be independent factors predictive of the alteration of chest wall contour. CONCLUSION In addition to the smaller preoperative CWA and higher apex location, lesser correction of vertebral rotation, if accompanied by great surgical correction of apical VT, could also largely result in a poor postoperative anterior chest wall contour.