Abstract
Purpose:
To describe changes in corneal power and axial length (AL) in the COMET cohort followed for 14 years, and explore the relationship between AL and corneal radius (CR) over this time period.
Methods:
469 ethnically diverse, 6-11 year old children with -1.25 to -4.50 D of myopia were enrolled in COMET. Children wore either single vision (SVLs) or progressive addition lenses (PALs) for 5 years and were followed for an additional 9 years wearing PALs, SVLs, or contact lenses. Additionally, 206 non-myopic young adults matched by gender, ethnicity, and age with COMET myopes were recruited at the 12-year visit. Refractive error (cycloplegic autorefraction), corneal curvature (CC, auto-keratometry), and ocular component dimensions (A-scan ultrasound) were measured annually for COMET children, and once for the non-myopic, matched young-adult subjects. Linear mixed model was used to evaluate longitudinal changes based on all available records adjusting for covariates (gender, ethnicity, lens type, baseline age and baseline refraction). Unpaired t-test was used to compare myopes and non-myopes at the 12-year visit. The Pearson correlation coefficient (PCC) between AL and CC was computed at each visit. The comparison of PCCs between myopes and non-myopes at the 12-year visit was conducted using Fisher’s transformation.
Results:
Longitudinally, COMET girls had significantly steeper CC than boys (p<0.0001). Caucasians had the steepest CC, and Hispanics the flattest (p=0.001). The correlation between AL and CC was -0.70 (p<0.0001) at baseline and decreased to -0.53 (p<0.0001) at the 14-year visit. The average AL to CR ratio (AL/CR) was 3.15 at baseline and increased to 3.31 at the 14-year visit. In the cross-sectional analysis, the correlation between AL and CC for the COMET myopes at the 12-year visit was lower than for the matched non-myopes (r =-0.57 vs.-0.76; p<0.0001). In addition, the AL/CR in myopes was significantly higher than in non-myopes (3.30 vs. 3.00; p<0.0001).
Conclusions:
These data demonstrate significant gender and ethnicity differences in average corneal curvature as myopia progresses. In addition, our findings suggest that as axial length continues to increase in myopic children the cornea may be incapable of further flattening, contributing to myopia progression.