Retina geography not associated with cognitive decline

December 21, 2016

Retinal vascular changes do not show non-pathological age-related cognitive decline, researchers say. The study does not rule out the use of retinal information in diseases such as Alzheimer’s.

Retinal vascular changes do not show non-pathological age-related cognitive decline, researchers say. The study, however, does not rule out the use of retinal information in diseases such as Alzheimer’s.

Sarah McGrory and colleagues at the University of Edinburgh in Edinburgh, United Kingdom, published the finding in the British Journal of Ophthalmology.

The retina offers a potential non-invasive window into changes in the brain. Previous research has shown associations between retinopathy and retinal branching parameters, poorer verbal memory, information processing speed, and executive function. However, these associations appear stronger in younger people than in older ones.

Some research suggests that about half of the variance in cognitive status in later life is associated with childhood intelligence quotient (IQ), but it is challenging to find data with which to study that relationship.

A study of retinal vascular abnormalities found differences between people with Alzheimer’s disease and healthy people of the same age. In particular, those with high neocortical plaque burden had increased arteriolar length-to-diameter ratio (LDR) (a measure of vessel width) and venular branching asymmetry factor values.

The authors could not find any studies examining the relationship between LDR and non-pathological cognitive ageing.

They examined the relationship of this and other retinal parameters in the Lothian Birth Cohort 1936, a large population of healthy older adults whose IQ scores were measured when they were 11 years old and again when they were older adults.

Most of the participants took part in the Scottish Mental Survey of 1947, which tested the intelligence of almost all Scottish schoolchildren born in 1936.

McGrory and her colleagues cross-referenced this data with digital retinal photographs and cognitive tests obtained in tests of 570 subjects completed from 2008 to 2010 when the participants were about 73 years old. The sample was 51.5% male.

In bivariate analyses, the researchers found few retinal vascular parameters that have an association with cognitive ability either when the subjects were children or when they were older adults, or with changes in cognitive ability over that time.

Negative associations


They did find a slight negative association between venular tortuosity and memory (r = -0.08, P = 0.04), and crystallised ability (r - -0.09, P = 0.02) and IQ at age 11 (r = -0.10, P = 0.02).

They found a negative association between arteriolar fractal dimension and memory (r = -0.09, P = 0.03), and between venular fractal dimension IQ at age 11 (r = -0.08, P = 0.04).

They found a negative association between lifetime change in IQ and arteriolar junctional exponent, but a positive association between lifetime change in IQ and arteriolar branching coefficient.

Multivariate analysis, which took into consideration such factors as age, sex, socioeconomic status, education, IQ at age 11, and vascular risk factors, produced a different set of a few nominally significant associations.

However, the retinal parameters appeared to contribute less than 1% of the variance in most of the associations. One of the associations remained statistically insignificant when the researchers adjusted them according to the false discovery rate method to eliminate type 1 errors.

To see if the lack of statistical significance could be attributed to a lack of statistical power, the researchers calculated that the sample size of 570 subjects had 80% power to detect a change of 0.118 standard deviation units. The sample excluded 6 people with a Mini-Mental State Examination score of less than 24, a commonly used indication of dementia.

The findings agree with what other studies have found. The researchers speculate that “associations between vessel width and dementia suggest that alterations may not manifest until late in the disease process, or that alternative measurements such as LDR may be more sensitive.” This study provides tentative support for this hypothesis, even though the associations it uncovered did not survive the false discovery rate analysis, they wrote.

A previous study using the same database found little evidence of an association between fractal dimension and cognitive ability. It did show an association between higher fractal dimension and poorer memory performance, but the low number of significant associations raised the possibility of type 1 errors, McGrory and colleagues reported.

At the same time, other studies have reported lower fractal dimensions in people with cognitive dysfunction and dementia. The researchers called for further study to elucidate the inconsistency between findings in people with this type of pathology and people with more typical age-related cognitive decline.

The researchers noted limitation to their study, including the potential for intergrader and intragrader variability, the variation of retinal vessel calibres within individuals, and the assumption that measurements of retinal parameters from one eye can adequately represent the same parameters in the other eye.

They note that stronger associations might have surfaced if retinal measurements had been available from participants in middle age.

“Though mostly null, we judge the present study’s results to be valuable,” they concluded.