Abstract
The retina has been proposed as a potential biomarker for Alzheimer’s disease, despite a thin but nonetheless growing body of evidence to support its utility during preclinical stages of the disease. This thesis evaluated whether quantitative retinal measurements of the neural layers (retinal nerve fibre layer [RNFL] and ganglion cell layer [GCL]) and the microvasculature (vessel density [VD] and perfusion density [PD]) were associated with cognitive abilities and their decline, in a birth cohort followed to middle-age. Further, a lifecourse approach was adopted to investigate potential early psychosocial determinants of these retinal measures.
Participants were members of the Dunedin Multidisciplinary Health and Development Study, a representative birth cohort (n = 1037). Study members were assessed at birth and at ages 3, 5, 7, 9, 11, 13, 15, 18, 21, 26, 32, 38, and most recently at age 45 years, with 94% still participating. Importantly, while the participants were relatively young at the latest assessment, some were already showing evidence of cognitive decline.
The first study examined the association between cognitive performance in childhood and at age 45, and RNFL and GCL measured via optical coherence tomography (OCT) at age 45. Full scale IQ was associated with RNFL and GCL thickness cross-sectionally in middle age, and prospectively from childhood. Although neither RNFL nor GCL were associated with global cognitive decline across the same period, thinner RNFL was associated with a greater decline in processing speed.
The second study investigated whether a psychosocial risk factor in childhood, specifically social isolation, was associated with RNFL and GCL thickness. Social isolation in adulthood is a known risk factor for poorer physical health, but the effects of social isolation during childhood on adult physical health has been less explored. Childhood social isolation may become biologically embedded, such that effects on the body remain apparent across the lifecourse. Childhood social isolation was associated with average RNFL thickness at age 45, as well as in the nasal and inferior quadrants. The association with the nasal quadrant was not attenuated by co-occurring risk factors from childhood (family socioeconomic status, maltreatment, perinatal complications, IQ, or physical health) or adulthood (socioeconomic status, blood pressure, tobacco use, or cannabis use). Further, this association was not mediated by adult loneliness or social support, suggesting that childhood social isolation is an independent predictor of adult RNFL thickness.
The third study examined whether there was an association between cognitive decline and retinal microvasculature using a novel extension of OCT technology, OCT angiography. This study presents evidence of a weak association between vessel density and cognitive performance in childhood and adulthood, but no association with cognitive decline over this period.
Considered together, these findings suggest that RNFL, but not other retinal measures, may contribute to identifying those in the earliest stages of cognitive decline. Psychosocial factors may ‘get under the skin’ to influence RNFL thickness, specifically childhood social isolation. This suggests that the retina is subject to the mechanisms of chronic stress which affect other bodily systems, and that there might be a shared pathway between neurodegeneration in the retina and the brain.