Preoperative left shoulder elevation > 1 cm is predictive of severe postoperative shoulder imbalance in early onset idiopathic scoliosis patients treated with growth-friendly instrumentation

Purpose: Prior research has demonstrated the influence of preoperative shoulder elevation (SE), proximal thoracic curve magnitude, and upper instrumented vertebra (UIV) on shoulder balance after PSF for AIS. Our purpose was to evaluate the impact of these factors on shoulder balance in early onset idiopathic scoliosis (EOIS) patients treated with growth-friendly instrumentation. Methods: This was a multicenter retrospective review. Children with EOIS treated with dual TGR, MCGR, or VEPTR and minimum 2-year follow-up were identified. Demographics and radiographic/surgical data were collected. Results: 145 patients met inclusion criteria: 74 had right SE (RSE), 49 left SE (LSE), and 22 even shoulders (EVEN) preoperatively. Mean follow-up was 5.3 years (range, 2.0–13.1 years). The LSE group had a larger pre-index mean main thoracic curve (p = 0.021) but no difference was observed between groups at the post-index or most recent timepoints. RSE patients with UIV of T2 were more likely to have balanced shoulders post-index than patients with UIV of T3 or T4 (p = 0.011). Pre-index radiographic shoulder height (RSH) was predictive of post-index shoulder imbalance ≥ 2 cm in the LSE group (p = 0.007). A ROC curve showed a cut-off of 1.0 cm for RSH. 0/16 LSE patients with pre-index RSH < 1.0 cm had post-index shoulder imbalance ≥ 2 cm compared to 8/28 (29%) patients with pre-index RSH > 1.0 cm (p = 0.006). Conclusion: Preoperative LSE > 1.0 cm is predictive of shoulder imbalance ≥ 2 cm after insertion of TGR, MCGR, or VEPTR in children with EOIS. In patients with preoperative RSE, UIV of T2 resulted in a higher likelihood of balanced shoulders postoperatively.