Systemic Metabolic and Skeletal Muscle-Related Profiles Associated with Annual Axial Elongation Across Optical Myopia-Control Strategies in Adolescents: A Retrospective Cohort Study.

INTRODUCTION

Axial elongation varies among adolescents receiving myopia-control treatment, but the contribution of systemic metabolic and skeletal muscle-related characteristics remains uncertain. We examined these associations, treating recorded glucagon-like peptide 1 receptor agonist (GLP-1RA) status as a clinical marker rather than an independent drug exposure.

METHODS

This retrospective study included 3270 adolescents aged 7-18 years receiving single-vision spectacles, defocus incorporated multiple segments (DIMS) lenses, or orthokeratology. The primary outcome was annual axial elongation; composite treatment failure was exploratory. Multivariable regression, spline models, treatment-stratified analyses, and stabilized inverse probability of treatment weighting (IPTW) were performed. Primary adjusted analyses included 2267 complete cases. The revised exploratory muscle-vulnerability score excluded GLP-1RA status and duration.

RESULTS

Among 3270 participants, 484 had recorded GLP-1RA use and showed greater obesity and insulin-resistance burden and less favorable muscle and activity profiles. In the comprehensive model, BMI z-score (β = 0.020 mm/year per unit; 95% CI 0.014-0.026) and HOMA-IR (β = 0.007 mm/year per unit; 95% CI 0.004-0.009) were associated with greater annual axial elongation. Skeletal muscle index was not independently associated, whereas handgrip strength showed a nonlinear association (P for nonlinearity = 0.002). Recorded GLP-1RA status was associated with 0.040 mm/year greater elongation (95% CI 0.031-0.049), without heterogeneity across myopia-control strategies (P for interaction = 0.646). After IPTW and comprehensive adjustment, the estimate attenuated to 0.032 mm/year (95% CI 0.021-0.043), with residual imbalance in four of 17 covariates. Associations with exploratory composite treatment failure were statistically uncertain in the comprehensive model (OR 1.23; 95% CI 0.93-1.63) and after IPTW (OR 1.17; 95% CI 0.80-1.70).

CONCLUSIONS

Metabolic vulnerability was associated with modestly greater annual axial elongation, although the clinical importance of the observed difference remains uncertain. Recorded GLP-1RA status identified a metabolically higher-risk subgroup but did not establish a drug effect. Muscle-related findings were not uniform, and residual confounding remained.