Abstract
Purpose :
To investigate the correlative factors of myopia and to establish the model for predicting myopia onset.
Methods :
This was a 1.5-year longitudinal study by collecting ocular biometry (OB) data from 3 schools’ students (age 6-14) in China. Refractive error was assessed with cycloplegic autorefraction. Partial correlation analysis was used to assess the factors associated with spherical equivalent (SE) progression. Generalized Estimated Equation was applied to evaluate the correlation between astigmatism and myopia onset over time. Logistic regression analysis was performed to build the model of axial length to corneal radius (AL/CR) ratio predicting myopia onset and ROCs were used to measure its performance.
Results :
SE progressions showed significant correlation with anterior chamber depth (OD:r=-0[DH1] .181, P<0.001; OS:r=-0.193, P<0.001), lens thickness (OD:r=0.102, P<0.001; OS:r=0.138, P<0.001), AL/CR(OD:r=-0.314, P<0.001; OS:r=-0.348, P<0.001). The odd ratio (OR) of myopia onset in astigmatism group was 1.797(95%CI:1.696-1.903) times compared to non-astigmatism group, while the OR of interaction astigmatism group with time was 0.853(95%CI:0.837-0.868) times compared to non-astigmatism group with time. The area under the ROC curve (AUC) of the AL/CR ratio predictive model was OD: 0.823(95%CI: 0.803-0.842), OS: 0.822(95%CI:0.802-0.841) in training group, and OD: 0.823(95%CI:0.794-0.851), OS: 0.792(95% CI:0.763-0.821) in validation group as the approximate results after adjusted by age, gender, school and grade.
Conclusions :
Anterior chamber depth, lens thickness, AR/CR ratio, had a significant correlation with progression rate of myopia. Besides, astigmatism played a positive stimulating role in developing myopia. Finally, a relatively feasible model using AL/CR ratio for predicting myopia onset was proposed.
This is a 2020 ARVO Annual Meeting abstract.