Purchase this article with an account.
Hui Wang, Tobias Elze, Tobias Luck, Jana Kynast, Francisca S. Rodriguez, A. Veronica Witte, Christoph Engel, Johanna Girbardt, Mengyu Wang, Neda Baniasadi, Dian Li, Markus Loeffler, Matthias L. Schroeter, Steffi G. Riedel-Heller, Arno Villringer, Franziska G Rauscher; Association between retinal nerve fiber layer thickness and human cognitive performance. Invest. Ophthalmol. Vis. Sci. 2018;59(9):1128.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To study the association between pointwise interocular circumpapillary retinal nerve fiber layer thickness (RNFLT) and different dimensions of cognitive performance.
From the age and gender stratified, population-based LIFE-Adult-Study, spectral domain optical coherence tomography (OCT) scans of circumpapillary RNFLT (768 A-scans, diameter: 12°) were selected from participants with reliable measurements (≥50 B-scan repetitions, quality ≥20 dB, ≤5% missing A-scans) and without clinically significant findings on fundus and/or OCT images in either eye. One eye per subject was randomly chosen. Linear regression was applied to model the outcome of the following cognitive tests (see Figures for descriptions/examples): Stroop (executive functions/cognitive speed), trail-making test (TMT; Part A: cognitive speed/B: executive functions), and the German version of the Hill Mill vocabulary scale (WST; verbal IQ). Apart from pointwise RNFLT, age and an estimation of real scanning diameter based on scanning focus (to adjust for eye anatomy and lens effects) were used as regressors.
5,646 eyes of 5,646 subjects were selected (54.8% female; age range:20-79 years). For Stroop, no significant (p<0.05) locations were detected after adjustment for multiple comparisons (Fig.1). For cognitive speed (TMT-A), worse performance (longer time) was related to lower RNFLT at large parts of the temporal/supero-temporal area (11.3% of the 768 locations significant; Fig.2 top). The same pattern, but less pronounced (3.6% significant locations) was found for task switching (TMT-B). The strongest effect was observed for verbal IQ (WST, 26.4% significant locations; Fig.2 bottom), where higher performance was related to thicker RNFLT at the same temporal/supero-temporal retinal locations.
Lower RNFLT was previously associated with cognitive decline in the context of dementia. Based on data from a large population-based study, we demonstrate linear relationships between RNFLT at specific retinal locations and the whole range of cognitive performance in selected domains. While no or at most weak relationships were found for executive functions (Stroop, TMT-B), lower verbal IQ (WST) and cognitive speed (TMT-A) were significantly related to thinner RNFLT exclusively at temporal/supero-temporal retinal areas. Our results might help to detect early cognitive decline and to better monitor its progression.
This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.
This PDF is available to Subscribers Only