High Heritability of Refractive Error and Axial Length

Abstract

Purpose: : A classical twin study was performed to investigate the heritability of axial length and refractive error.

Methods: : Monozygotic (MZ) and dizygotic (DZ) twin pairs were recruited from the Twin Eye Study in Tasmania and the Brisbane Adolescent Twin Study. In each subject refractive error was measured using a Humphrey 598 autorefractor, after cycloplegia and axial length was measured using an Alcon Ocuscan. Subjects were excluded if they had undergone cataract surgery, had anterior segment disease or a spherical equivalent (SE) > ±5D. The mean axial length (AL) and SE of each subject was used in analysis. Structural equation modeling was performed with the Mx program.

Results: : 152 MZ and 318 DZ twin pairs were recruited. The mean age of subjects was 20.2 ± 13.3 years (range 5–68 years). The mean SE for the cohort was –0.033 and was more highly correlated in MZ versus DZ twins (intraclass correlation coefficients being 0.82 and 0.45 respectively). A significant difference in SE variation was found between sexes. An age–adjusted model incorporating 81.7% additive genetic (95%CI:66.3–85.7%) and 18.3% unique environment (95%CI:14.3–23.5%) components provided the best fit for SE. Mean AL of all eyes was 23.00 mm (range 20.39–26.41mm). SE correlated with AL (R²=0.15, p<0.0001). The correlation coefficients for AL were 0.85 and 0.43 for MZ and DZ twin pairs respectively. Modeling revealed that AL was 84.9% (95%CI:79.3–88.8%) heritable with unique environmental factors accounting for the remaining variance (15.1%: 95%CI:11.2–20.7%).

Conclusions: : This study of Australian twins supports earlier work that SE and AL has a strongly genetic component

Keywords: refraction • genetics