Relative Peripheral Refractive Error and the Risk of Juvenile-Onset Myopia

Abstract

Purpose: : Peripheral hyperopic defocus may accelerate the overall growth of the eye, often resulting in a myopic foveal refractive state (Smith et al., ARVO 2007). We investigated whether relative peripheral refractive error is related to the risk of myopia onset in children.

Methods: : Subjects were 1,981 participants in the Collaborative Longitudinal Evaluation of Ethnicity and Refractive Error (CLEERE) Study who were not myopic in the third grade and who returned for at least one subsequent annual visit. Examinations included autorefraction (Canon R-1, Grand Seiko WR-5100K) under cycloplegia (tropicamide or tropicamide plus cyclopentolate) in primary gaze and in 30 degrees temporal gaze. Relative peripheral refractive error (RPR) was defined as peripheral spherical equivalent minus central spherical equivalent refractive error. Central and RPR refractions were dichotomized into risk factors (cut points <+0.75 D or ≥+0.75 D centrally; any relative peripheral hyperopia or not for RPR). Risk of myopia onset (more than -0.75 D of myopia in each meridian) was analyzed using a discrete time survival analysis model.

Results: : Children with a central refractive error ≥+0.75 D had a low risk of myopia onset regardless of whether RPR was hyperopic or non-hyperopic (probability of no myopia by grade 8 = 0.98 and 0.99, respectively). Children with a central refractive error <+0.75 D had a higher risk of myopia onset that depended on whether RPR was hyperopic or non-hyperopic (probability of no myopia by grade 8 = 0.60 and 0.70, respectively). In a multivariate model, the hazard probability odds ratio for central refractive error <+0.75 D was 28.7 (95% CI = 14.1-58.6) and 1.46 for hyperopic RPR (95% CI = 1.14-1.87).

Conclusions: : Relative peripheral hyperopia is a risk factor for myopia onset independent of central refractive error. Central refractive error <+0.75 D seems to be the much stronger predictive factor, however. Possible mechanisms for RPR’s influence include defocus-driven eye growth or mechanical or physiologic restriction to equatorial expansion. Clinical trials of correction of peripheral defocus in children at risk for myopia onset using corneal reshaping or bifocal contact lenses may be justifiable but may also require a large sample considering the estimated size of RPR’s effect.

Keywords: myopia • clinical (human) or epidemiologic studies: risk factor assessment • refractive error development