The Nepal Longitudinal Study: biometric characteristics of developing eyes.

Purpose. To identify differences in potential biometric markers for predicting refractive error in school children. Methods. Biometric data on 895 Tibetan children, aged 6 to 18 years, residing in Katmandu, Nepal, were collected biennially from 1992 to 2000. Measurements included cycloplegic autorefraction, A-scan ultrasonography, and video phacometry. Only those children who had been studied at least once at age 12 years or more were included in the analysis. Subjects were divided into two groups: a myopia group if the refractive error was myopic by more than -0.50 D and a nonmyopia group if the refractive error was maximally myopic by -0.50 D, expressed as a spherical equivalent error in the left eye. Results. Biometric measures that differed significantly with increasing age between the two refractive groups included: anterior chamber depth + 0.012 mm/year (p = 0.014), anterior lens radius of curvature + 0.073 mm/year (p = 0.001), lens power -0.059 D/year (p = 0.082), lens thickness -0.005 mm/year (p = 0.02), and vitreous chamber depth + 0.084 mm/year (p < 0.001). Corneal radii of curvature of the myopic group were steeper at all ages by 0.09 mm (p < 0.001), but the rate of change with age was equivalent across the refractive groups. Conclusions. Compared with those who remained nonmyopic, children who developed myopia had a crystalline lens that was initially thicker and steeper, and a vitreous chamber that was initially shorter. With age, children who became myopic developed greater lens thinning, greater flattening of the anterior lens surface radius, and a greater increase in vitreous chamber depth than their nonmyopic counterparts