Real World Axial Elongation in Children Wearing Diversified Segmental Defocus Optimization Spectacle Lenses: A...

To describe longitudinal axial elongation in children wearing spectacle lenses incorporating Diversified Segmental Defocus Optimization (D.S.D.O.) technology in routine clinical practice and to place the observed change in context using two historical external benchmarks.

This retrospective cohort included first time D.S.D.O. spectacle lens users with no documented prior myopia control lens wear and bilateral follow up of at least 12 months. Repeated axial length measurements from March 2022 through July 2026 underwent prespecified longitudinal quality control. The primary mixed effects model adjusted for age, sex, baseline axial length, baseline spherical equivalent, and any recorded atropine use. Sensitivity analyses addressed cycloplegic refraction, quality control exclusions, delay between lens initiation and baseline measurement, missing baseline refraction, and longitudinal covariance. External benchmarks used a Chinese pediatric normative model and age updated piecewise axial elongation rates from two Hong Kong single vision spectacle cohorts.

The cohort comprised 2396 children and 4792 eyes. Mean age was 9.23 ± 2.18 years and median follow up was 1.92 years. The primary adjusted axial elongation rate was 0.173 mm/year (95% CI, 0.171 to 0.174), with similar point estimates in cycloplegic, alternative quality control, baseline timing, missing refraction, random slope, and robust covariance analyses. Among 1665 children eligible for the Chinese model, observed change was 0.323 mm versus 0.625 mm expected, an absolute difference of 0.302 mm (95% CI, 0.290 to 0.314). Among 1515 children eligible for the Hong Kong benchmark, age updated expected change was 0.425 mm versus 0.319 mm observed, an absolute difference of 0.106 mm (95% CI, 0.096 to 0.117). Findings were similar after excluding children with any atropine record.

Axial elongation during D.S.D.O. spectacle lens wear averaged approximately 0.17 mm/year in this large routine care cohort. Observed changes were lower than both historical external expectations, but these benchmark contrasts are descriptive associations and should not be interpreted as causal efficacy estimates. Prospective studies with concurrent controls and standardized behavioral and treatment exposure data are required.