Does Obesity Affect Preoperative Shoulder Balance in Adolescent Idiopathic Scoliosis?

Abstract

<b><i>Background:</i></b> Obese patients with adolescent idiopathic scoliosis (AIS) have been shown to present with larger curve magnitudes preoperatively. However, the effect of obesity on shoulder balance in AIS remains unknown. The purpose of our study was to determine if overweight and obese patients with AIS have worse radiographic shoulder balance on initial presentation when compared with normal weight patients. <b><i>Methods:</i></b> AIS patients &#x3c;18 years old, with Lenke 1 or 2 curves, who underwent a posterior spinal fusion between March 2013 and December 2018 were retrospectively evaluated. BMI-for-age percentiles as defined by the Center for Disease Control and Prevention were used: obese (≥95th percentile), overweight (85th to &#x3c;95th percentile), and normal weight (5th to &#x3c;85th percentile). Shoulder height was measured via the radiographic shoulder height (RSH) method, with an RSH ≤ 1 cm considered balanced. The primary outcome was preoperative shoulder balance. Secondary outcomes included postoperative shoulder balance, major curve correction, and UIV selection. <b><i>Results:</i></b> One hundred eighty-four patients (116 [63%] normal weight and 68 [37%] overweight/obese) were included. The mean age at surgery was 13.1 ± 2 years, and mean follow-up was 17.4 ± 13 months. Preoperative shoulder imbalance was significantly greater in the overweight/obese group compared to the normal weight group (1.9 ± 1 cm vs. 1.5 ± 1 cm, <i>p</i> = 0.04). The odds ratio of presenting with unbalanced shoulders was 2.0 (95% CI: 1.02–3.83, <i>p</i> = 0.04) for the overweight/obese group. No significant differences were found for postoperative shoulder balance, UIV selection, or major curve correction. <b><i>Conclusions:</i></b> Overweight and obese patients with AIS are twice as likely to present with unbalanced shoulders preoperatively; however, this difference is not clinically relevant with a mean difference of 0.4 cm between cohorts. Finally, the preoperative BMI percentile did not show a significant effect on the chosen UIV or curve magnitude correction. <b><i>Level of Evidence:</i></b> Level III: this is a retrospective case-control study.