Decreased visual acuity is related to thinner cortex in cognitively normal adults: cross-sectional, single-center cohort study

Abstract

Background: Decreased visual acuity (VA) is reported to be a risk factor for dementia. However, the association between VA and cortical thickness has not been established. We investigated the association between VA and cortical thickness in cognitively normal adults.

Method: We conducted a cross-sectional, single-center cohort study with cognitively normal adults (aged ≥ 45) who received medical screening examinations at the Health Promotion Center at Samsung Medical Center. Subjects were categorized as bad (VA ≤ 20/40), fair (20/40 < VA ≤ 20/25), and good (VA > 20/25) VA group by using corrected VA in the Snellen system. Using 3D volumetric brain MRI, cortical thickness was calculated using the Euclidean distance between the linked vertices of the inner and outer surfaces. We analyzed the association between VA and cortical thickness after controlling for age, sex, hypertension, diabetes, dyslipidemia, intracranial volume, and education level.

Results: A total of 2756 subjects were analyzed in this study. Compared to the good VA group, the bad VA group showed overall thinner cortex (p = 0.015), especially in the parietal (p = 0.018) and occipital (p = 0.011) lobes. Topographical color maps of vertex-wise analysis also showed that the bad VA group showed a thinner cortex in the parieto-temporo-occipital area. These results were more robust in younger adults (aged 45 to 65) as decreased VA was associated with thinner cortex in more widespread regions in the parieto-temporo-occipital area.

Conclusion: Our results suggest that a thinner cortex in the visual processing area of the brain is related to decreased visual stimuli.

Keywords: Cortical thickness; Dementia; Visual acuity.

Fig. 1

Statistical topography of the relationship between cortical thickness and visual acuity (VA). Compared to the good VA group, the bad VA group showed a thinner cortex in the bilateral temporo-parieto-occipital lobes (A). In the analysis of the subjects who were equal or under 65 years old, a similar result was found with a larger extent and greater significance (B). False discovery rate corrected (q < 0.05). This result was adjusted for age, sex, hypertension, diabetes mellitus, hyperlipidemia, education, and intracranial volume. The MATLAB-based toolbox 2014b (https://kr.mathworks.com/products/matlab.html) was used for the drawing and the configuration of this figure