Prospective Evaluation of Macular and Parapapillary Morphology and Their Association With Cognition in Childhood and Old Age.

Purpose We investigated the potential association between cognitive function and retinal morphology based on shared age-related changes in a cohort of older people. Methods Participants (n=250) from the Scottish Mental Survey 1947 undertook the Moray House Test (MHT) of general cognitive ability (IQ) at age 11 and 70. All were born in 1936. Fundus photographs were assessed by an ophthalmologist masked to the results of these tests. Results Univariate analyses of variance were computed to compare the effects of retinal morphological features on cognitive test scores. Presence of retinal drusen was associated with lower age 70 IQ scores when compared to those without this abnormality. This difference approached statistical significance (F (1,166)=3.77, p=0.054, η2=0.22). Scores on the same cognitive test administered at age 11 were significantly lower in those with retinal drusen when compared to IQ scores from a group without this abnormality. This difference was statistically significant (F (1,158)=4.03, p=0.046, η2=0.025), though the effect size was small. But when age 11 IQ was included as a covariate in the earlier model, age 70 IQ scores were no longer significantly different by retinal drusen grouping (F (1,155)=1.47, p=0.228, η2=0.009). Similar analyses using parapapillary atrophy yielded null findings. Conclusion Initial analysis of the age 70 IQ data implied that macular drusen could serve as a potential bio-marker of cognitive decline. But the added information from the age 11 IQ data raises the possibility of reverse causation: rather than late-life retinal damage being a marker of cognitive decline, early-life cognitive ability might account for this retinal change in late-life.