ABSTRACT
BACKGROUND: Postoperative shoulder imbalance is a common complication in adolescent idiopathic scoliosis and may impair patient satisfaction. The little data in the literature on the prediction and treatment of postoperative shoulder imbalance is incongruous. According to some studies, postoperative shoulder imbalance depends on the superior instrumented vertebral bodies, skeletal maturity, extent of correction and flexibility of the major curve. QUESTION: Can preoperative radiological parameters from X-ray and traction films as well as correction of the curves be used to identify a factor impacting on postoperative shoulder imbalance? MATERIAL AND METHODS: Prospective data in adolescent idiopathic scoliosis with thoracic curve (Lenke type 1), operated on between 2015 and 2018 at a scoliosis centre, were analysed retrospectively based on full-length X-rays of the spine (pre-/postOP and follow-up (FU)) and preoperative-traction films: age; correction of proximal, main and lumbar curve; shoulder height [mm]; clavicle angle [°]; T1 tilt [°]; coronal plumb line deviation [mm]. The findings were expressed as means with standard deviation. Changes in parameters over time (postOP-FU) were compared by t test (significance level α = 0.05). The correlation between preOP parameters and extent of correction with postoperative shoulder imbalance (≥ 15 mm) was determined by Pearson correlation and regression classification analysis. OUTCOMES: 55 patients with adolescent idiopathic scoliosis, mean age of 15 ± 1.4 years. The FU-rate after a mean of 15 months was 80% (n = 44). Correction of proximal, main and lumbar curve: 47.0%, 75.8% and 68.8%, respectively, without statistically significant change (Δ) in FU (p > 0.05). Shoulder height was - 11.0 ± 12.7 mm (preOP), 15.5 ± 13.4 mm (postOP), 10.1 ± 10.6 mm (FU) (p < 0.05). 38% of those with adolescent idiopathic scoliosis had preOP right shoulder elevation; 55% (postOP) and 32% (FU) respectively had postoperative shoulder imbalance (left shoulder elevation). Strong statistical correlation was found for Δshoulder position (FU-preOP) with pre-OP shoulder position (r = - 0.7), and Δshoulder position (pre-OP traction films) (r = 0.5) with pre-OP clavicle angle (r = - 0.5). On regression classification analysis, 81.8% of cases did display postoperative shoulder imbalance if proximal curve correction was ≤ 64.4%; main and lumbar curve correction, Δshoulder elevation (preOP traction films) played a secondary role. DISCUSSION: One common complication even in Lenke type 1 adolescent idiopathic scoliosis is postoperative shoulder imbalance. Preoperative planning should include traction films, preoperative shoulder position and clavicle angle to avoid postoperative shoulder imbalance. Moderate correction of proximal curve is critical for postoperative shoulder balance.
