Axial length changes and baseline predictors over five years in children treated with orthokeratology lenses: a retrospective cohort study.

Background: Orthokeratology lenses (Ortho-K) are widely used in pediatric myopia management. However, axial length (AL) changes over 5 years, and the baseline predictors in treated cohorts remain insufficiently characterized. Methods: This retrospective cohort study included 656 right eyes of children who were treated with Ortho-K lenses and were followed for 5 years. Baseline variables included sex, age, refractive parameters (Cycloplegic Spherical Equivalent (SE), cycloplegic spherical power (DS), cycloplegic cylindrical power (DC)), keratometric indices, and AL. All refractive measurements were recorded in minus-cylinder notation. AL was assessed at baseline and at follow-up visits at 1, 3, and 5 years. Linear regression was used to analyze continuous 5-year changes in AL. Logistic regression was additionally performed using a prespecified axial elongation criterion (≤0.2 mm/year; ≤1.0 mm over 5 years) for within-cohort stratification. Results: Over 5 years, the mean AL elongation was 0.75 ± 0.49 mm in this cohort of children treated with Ortho-K lenses. In univariable analyses, baseline age, sex, SE, DS, and baseline AL were associated with the five-year change in AL. In the multivariable models, younger age, sex, and refractive status remained associated with five-year AL change, whereas baseline AL was not independently associated after adjustment. In the stratified logistic analyses, DS was associated mainly with the 1-year outcome grouping, whereas baseline age remained the most consistent factor over longer follow-up. Conclusion: This retrospective cohort study describes AL changes over 5 years and their baseline predictors in myopic children treated with Ortho-K lenses. Among the baseline variables, younger age was the most consistent factor associated with greater AL elongation during follow up, while sex and baseline refractive status showed less consistent associations. Due to the absence of an untreated or alternative treatment control group, this study cannot directly quantify the treatment effect of Ortho-K relative to natural myopia progression. These findings should therefore be interpreted as within cohort observational data and support closer AL monitoring and individualized management for younger children receiving Ortho-K treatment.