Does levelling of T1 tilt intra-operatively affect post-operative shoulder balance in adolescent idiopathic scoliosis patients?

Abstract

Previous studies have shown that T1 tilt is positively correlated with post-operative shoulder balance (SB). The aim of this study was to explore the role of intra-operative T1 tilt, among other shoulder parameters as a potential parameter to predict post-operative SB in adolescent idiopathic scoliosis (AIS) patients. A retrospective review of AIS patients with structural thoracic curves with minimum 2 year follow up was conducted from a single tertiary center. Standing pre-operative, 1st erect, 1 year and 2-year follow-up; and intra-operative final prone radiographs were reviewed along with clinical data. Patients were stratified into 2 cohorts: Group A-Final intra-operative T1 tilt ≤5° and Group B-Final intra-operative T1 tilt >5°. These groups were compared for post-operative SB as a whole and separately for patients with baseline right or left shoulder high and if UIV was T2 or T3/T4. Patients with optimal SB (Radiographic shoulder height (RSH) <2 cm) at 2 years were compared to sub-optimal SB (RSH ≥ 2 cm) with respect to multiple SB variables. 55 patients (mean age 15.1 years-old, 43 F, mean BMI 22, mean thoracic Cobb-49.8°) were included. Based on Lenke curve types, there were 13 patients with type 1A, 10 patients with 1B, 12 patients with 1C, 7 patients with 2A, 4 patients with 2B and 9 patients with type 3C. T1 tilt was significantly correlated with RSH, Clavicle angle difference (CAD), First Rib Angle (FRA), and UIV tilt at first erect, 1-year, and 2-year post-op radiographs (p < 0.05 for all). When comparing groups, A and B, Group A patients showed significantly better restoration of their 2-year SB parameters; RSH (6.8 vs 11.8 mm, p = 0.01), CAD (3.9 vs 9.1 p < 0.001) and T1 tilt (4.7 vs 7.8° p = 0.01). Similar results were found for patients with baseline right shoulder high; RSH (p = 0.04), CAD (p < 0.001) and T1 tilt (p < 0.001) and whether UIV was T2 or T3/T4. Eight patients with sub-optimal SB had worse intra-operative T1 tilt (p = 0.03) compared to 47 patients with optimal SB despite no difference in MT Cobb correction (83.1 vs 79.8%, p = 0.57). Post-operative T1 tilt correlates with lateral shoulder parameters at first erect, 1 year, and 2-year radiographs. Therefore, T1 tilt can potentially be used as a surrogate to predict post-operative SB. Leveling intra-operative T1 tilt ≤5° is associated with better 2-year post-operative shoulder balance parameters irrespective of whether the UIV was T2 or T3/T4. Patients with sub-optimal SB at 2 years had worse final intra-operative T1 tilt despite similar percent correction of main thoracic curve for all patients.