Retinal parameters, cortical cerebral microinfarcts, and their interaction with cognitive impairment

Saima Hilal,Carol Y Cheung,Tien Yin Wong,Leopold Schmetterer,Christopher Chen
DOI: https://doi.org/10.1177/17474930221097737
Abstract:Background: Quantitative changes in retinal vessels and thinning of optic nerves have been associated with subclinical (atherosclerosis, inflammation) and clinical age-related brain pathologies (stroke and neurodegeneration). However, data on the association between both retinal vascular and neuronal parameters with cortical cerebral microinfarcts (CMIs) and how these factors jointly influence cognition are lacking. Aim: We investigated the association of retinal vascular and neuronal changes with CMIs on 3 T MRI and explored their interaction with cognitive impairment in a memory-clinic population. Methods: A total of 538 participants were included. Retinal vascular parameters (caliber, tortuosity, and fractal dimension) were measured from retinal fundus photographs using a semi-automated computer-assisted program. Retinal nerve fiber layer (RNFL) and ganglion cell-inner plexiform layer (GC-IPL) thicknesses were obtained from optical coherence tomography. Cortical CMIs were defined as hypointense on T1-weighted MRI, <5 mm in diameter and restricted to the cortex. Cognition was assessed using Clinical Dementia Rating Sum-of-Boxes (CDR-SoB) score and detailed neuropsychological test. Multivariable regression analysis was conducted adjusting for age, sex, hypertension, hyperlipidemia, diabetes mellitus, smoking, diagnosis, white matter hyperintensities volume, lacunes, and cerebral microbleeds. Results: Larger venular caliber (Rate ratios (RR): 1.15, 95% CI: 1.01-1.38, p = 0.014), increased venular fractal dimension (RR: 1.58, 95% CI: 1.31-1.91, p ⩽ 0.001), increased venular tortuosity (RR: 1.54, 95% CI: 1.35-1.75, p ⩽ 0.001), and thinner GC-IPL (RR: 1.24, 95% CI: 1.13-1.36, p ⩽ 0.001) were associated with CMI counts. Among individuals in highest tertile of retinal parameters, a significant interaction was observed between venular tortuosity (RR: 1.12, 95% CI: 1.02-1.22, p-interaction = 0.014) and GC-IPL (RR: 1.05, 95% CI: 1.01-1.11, p-interaction < 0.001) with CMIs on CDR-SoB. Conclusion: Retinal vascular and neuronal parameters are associated with cortical CMIs, and persons with both pathologies are likely to have cognitive impairment. Further studies may be warranted to evaluate the clinical utility of retinal parameters and CMI in risk prediction for cognitive dysfunction.
What problem does this paper attempt to address?