High Preoperative Body Mass Index Is Associated With Implant Breakage in Patients Treated With Magnetically Controlled Growing Rods for Early-onset Scoliosis.

Abstract

Introduction: Magnetically controlled growing rods (MCGRs) have become the current standard in the growth-friendly treatment of patients with early-onset scoliosis (EOS). MCGRs allow noninvasive lengthenings with external lengthening device and reduce the need for surgical procedures. The association of preoperative body mass index (BMI) and the outcomes of the MCGR treatment is not well known.

Methods: Prospectively collected international database was reviewed for EOS patients treated with MCGR. Patients without preoperative BMI data or follow-up <2 years were excluded. Patients were classified as healthy weight, overweight, and underweight using Centers for Disease Control and Prevention (CDC) growth charts. Quality of life was assessed using EOSQ-24. Results were analyzed from the 2-year follow-up.

Results: A total of 663 patients were categorized into underweight (n=91), healthy weight (n=417), and overweight (n=155) groups. There were no significant differences in major curve correction or thoracic height increase among the BMI groups, irrespective of etiology. Distribution of BMI categories differed significantly by etiology ( P =0.009), with lower healthy weight proportions in the syndromic group (92/167, 55%) compared with idiopathic (131/177, 74%) (adjusted P =0.004), and a higher underweight proportion in neuromuscular (36/244, 15%) compared with idiopathic (15/177, 8.5%) (adjusted P =0.044). Higher BMI z-scores were associated with an increased incidence of complications, including implant-related complications (RR 1.1, 95% CI 1.0-1.3) and implant breakage (RR 1.3, 95% CI 1.1-1.7). Healthy weight and underweight patients experienced lower overall complication rates compared with overweight patients. Implant-related complications were less common in underweight patients compared with overweight patients (RR 0.45, 95% CI 0.20-0.90). Higher BMI z-score was a significant predictor of implant breakage, whereas preoperative major curve, kyphosis, and etiology were not. EOSQ-24 scores did not differ significantly among BMI groups, and changes in scores were comparable across groups during follow-up.

Conclusion: BMI status did not influence curve correction, thoracic height increase, or EOSQ-24 outcomes in early-onset scoliosis patients. However, the higher incidence of implant breakage in overweight patients suggests that elevated BMI should be carefully considered when planning treatment.

Level of evidence: Level III.