Reading cognition from the eyes: association of retinal nerve fibre layer thickness with cognitive performance in a population-based study

With the eye as a window to the brain, non-invasive fast screening of retinal nerve fibre layer thickness poses the opportunity for early detection of cognitive decline leading to dementia. Our objective is to determine whether performance in various neurocognitive tests has an association with itemized retinal nerve fibre layer thickness. Detailed investigation of associations factored in sex and eye-side. The large population-based LIFE-Adult study (Leipzig Research Centre for Civilization Diseases) was conducted at Leipzig University, Germany from 2011 to 2014. Randomly selected participants (N = 10 000) were drawn from population registry in an age- and gender-stratified manner, focusing on 40-80 years. Cognitive function was examined with the CERAD-NP Plus test-battery (Consortium to Establish a Registry for Alzheimer's Disease), Stroop-Test, Reading the Mind in the Eyes-Test and Multiple-Choice Vocabulary Intelligence Test. Circumpapillary retinal nerve fibre layer thickness was measured with Optical Coherence Tomography. Subjects with reliable measurements (≥50 B-scan repetitions, signal-to-noise-ratio ≥20 dB, ≤5% missing A-scans) and without clinical eye pathology (sample A) and additional exclusion due to conditions of the central nervous system (sample B) were evaluated. The relationship between cognitive function and retinal nerve fibre layer thickness was investigated for six segments: temporal, temporal-superior, temporal-inferior, nasal, nasal-superior and nasal-inferior. For comparison with other studies, global mean is given. Brain-side projection analysis links results to the corresponding brain hemisphere. We analysed 11 124 eyes of 6471 subjects [55.5 years of age (19.1-79.8 years), 46.9% male]. Low cognitive performance was predominantly associated with thinner retinal nerve fibre layer thickness. Correlation analysis indicated emphasis on global and temporally located effects. Multivariable regression analysis with adjustments (age, sex and scan radius) presented individual results for each test, differentiating between sex and eye-side. For instance, verbal fluency tests and Trail Making Test-B show stronger association in females; Trail Making Test-A shows right-eye dominance. Findings in Trail-Making-Test-A projected to left brain hemisphere, and the ratio incongruent to neutral in the Stroop test projected to right brain-hemisphere. Separate assessment for sex and eye-side is presented for the first time in a population-based study. Location-specific sectorial retinal nerve fibre layer thickness was found to be an indicator for cognitive performance, giving an option for early detection of cognitive decline and the potential of early treatment. The eye as a window to the brain was studied with optical coherence tomography and connected to cognition. Girbardt et al. report that thinner retinal nerve fibre layer thickness was found to be a meaningful index for poorer cognitive performance which presents the potential for prediction of future cognitive decline.

Test-retest measurements of dopamine D1-type receptors using simultaneous PET/MRI imaging

PURPOSE

The role of dopamine D₁-type receptor (D₁R)-expressing neurons in the regulation of motivated behavior and reward prediction has not yet been fully established. As a prerequisite for future research assessing D₁-mediated neuronal network regulation using simultaneous PET/MRI and D₁R-selective [¹¹C]SCH23390, this study investigated the stability of central D₁R measurements between two independent PET/MRI sessions under baseline conditions.

METHODS

Thirteen healthy volunteers (7 female, age 33 ± 13 yrs) underwent 90-min emission scans, each after 90-s bolus injection of 486 ± 16 MBq [¹¹C]SCH23390, on two separate days within 2-4 weeks using a PET/MRI system. Parametric images of D₁R distribution volume ratio (DVR) and binding potential (BP_ND) were generated by a multi-linear reference tissue model with two parameters and the cerebellar cortex as receptor-free reference region. Volume-of-interest (VOI) analysis was performed with manual VOIs drawn on consecutive transverse MRI slices for brain regions with high and low D₁R density.

RESULTS

The DVR varied from 2.5 ± 0.3 to 2.9 ± 0.5 in regions with high D₁R density (e.g. the head of the caudate) and from 1.2 ± 0.1 to 1.6 ± 0.2 in regions with low D₁R density (e.g. the prefrontal cortex). The absolute variability of the DVR ranged from 2.4% ± 1.3% to 5.1% ± 5.3%, while Bland-Altman analyses revealed very low differences in mean DVR (e.g. 0.013 ± 0.17 for the nucleus accumbens). Intraclass correlation (one-way, random) indicated very high agreement (0.93 in average) for both DVR and BP_ND values. Accordingly, the absolute variability of BP_ND ranged from 7.0% ± 4.7% to 12.5% ± 10.6%; however, there were regions with very low D₁R content, such as the occipital cortex, with higher mean variability.

CONCLUSION

The test-retest reliability of D₁R measurements in this study was very high. This was the case not only for D₁R-rich brain areas, but also for regions with low D₁R density. These results will provide a solid base for future joint PET/MRI data analyses in stimulation-dependent mapping of D₁R-containing neurons and their effects on projections in neuronal circuits that determine behavior.