Purchase this article with an account.
ENEH JONES-ODEH, Ekaterina Hristova Yonova, Edward Bloch, Katie Williams, Claire Steves, Christopher J Hammond; Is the relationship between retinal nerve fibre layer thickness and cognitive performance explained by genetic or environmental factors? A twin study.. Invest. Ophthalmol. Vis. Sci. 2016;57(12):2542.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Retinal Nerve Fibre Layer Thickness (RNFLT) has been associated with cognitive function but it is unclear whether RNFLT thinning is secondary to cortical loss, or if the same disease process affects both. Twin studies can explore whether phenotypic sharing is due to shared genetic or environmental factors.
823 twins from 471 families (mean age 61years; range 19-89) from the TwinsUK cohort underwent detailed eye and cognitive examination including: spectral domain optical coherence tomography, autorefraction, intraocular pressure (IOP) measurements, mini mental state examination (MMSE), verbal fluency (VF), reaction time (RT) and recall (RC). Associations between average retinal nerve fibre layer (RNFL) or RNFL quadrants and age, sex, spherical equivalent (SphE), IOP and RT were assessed using uni- and multivariable linear regression models. Associations between RNFLT and MMSE, VF and RC was assessed using one-way ANOVA and MANOVA. Heritability analyses of RNFLT were performed using univariate and bivariate Cholesky decomposition heritability models on 686 twins (201 monozygotic and 142 dizygotic twin pairs).
Average RNFLT was thinner with increasing age (β=-0.18; p<0.001) and with increase in myopia (β=1.41; p<0.001). A thicker RNFLT was associated with better MMSE performance ( F(5,450)=2.8, p=0.01), and with faster RT in the inferior quadrant only (β=-0.02, p=0.04); these stayed significant after adjusting for age and SphE (p<0.05). There was no statistically significant association between RNFLT and IOP, VF or RC. RNFLT was highly heritable with additive genetic effects accounting for 82% of variance in the average RNFLT, while MMSE score and RT were moderately heritable (40% and 53% respectively). RNFLT shared 30% (95%CI: 18%-33%) of phenotypic variance with SphE, 12% (95%CI: 4%-19%) with RT (inferior quadrant) and 14% (95%CI: 6%-21%) with MMSE. Shared additive genetic effects accounted for 73%, 3% and 94% of the shared variance respectively. The last two estimates were not statistically significant, however.
MMSE performance was associated with RNFLT independent of age and SphE but we did not find conclusive evidence for shared genetic factors driving this association. RNFLT was independently associated with SphE; 73% of the shared variance was due to shared genetic factors.
This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.
This PDF is available to Subscribers Only