Previous studies have explored the associations of retinal vessel calibre, measured from retinal photographs or fundus images using semi-automated computer programs, with cognitive impairment and dementia, supporting the concept that retinal blood vessels reflect microvascular changes in the brain. Recently, artificial intelligence deep-learning algorithms have been developed for the fully automated assessment of retinal vessel calibres. Therefore, we aimed to determine whether deep-learning based retinal vessel calibre measurements are predictive of risk of cognitive decline and dementia. We conducted a prospective study recruiting participants from memory clinics at the National University Hospital and St. Luke’s Hospital in Singapore; all participants had comprehensive clinical and neuropsychological examinations at baseline and annually for up to 5 years. Fully automated measurements of retinal arteriolar and venular calibres from retinal fundus images were estimated using a deep-learning system. Cox regression models were then used to assess the relationship between baseline retinal vessel calibre and the risk of cognitive decline, and developing dementia, adjusting for age, gender, ethnicity, education, cerebrovascular disease status, hypertension, hyperlipidemia, diabetes, and smoking. A total of 491 participants were included in this study, of whom 254 developed cognitive decline over 5 years. In multivariable models, narrower retinal arteriolar calibre (hazard ratio per standard deviation decrease = 1·258, p = 0·008) and wider retinal venular calibre (hazard ratio per standard deviation increase = 1·204, p = 0·037) were associated with increased risk of cognitive decline. Among participants with cognitive impairment but no dementia at baseline (n = 212), 44 progressed to have incident dementia; narrower retinal arteriolar calibre was also associated with incident dementia (hazard ratio per standard deviation decrease= 1·624, p = 0·021). In summary, deep-learning based measurement of retinal vessel calibre was associated with risk of cognitive decline and dementia.

Background: Cerebral small vessel disease (SVD) and neuropsychiatric symptoms (NPS) independently increase the risk of cognitive decline. While their co-existence has been reported in the preclinical stage of dementia, longitudinal data establishing the prognosis of their associations, especially in an Asian context remains limited. Objective: This study investigated the role of SVD and NPS progressions on cognitive outcomes over 2 years in a dementiafree elderly cohort. Methods: 170 dementia-free elderly with baseline and 2-year neuropsychological assessments and MRI scans were included in this study. White matter hyperintensities (WMH), lacunes, and microbleeds (CMBs) were graded as markers of SVD. The Neuropsychiatric Inventory (NPI) was used to measure NPS. Generalized estimating equations modelling evaluated the relationship between NPI change and SVD progression. Logistic regression evaluated the risk of incident cognitive decline with both SVD and NPS. All models were adjusted for demographics, baseline cerebrovascular diease, and medial temporal lobe atrophy. Results: Higher NPI scores were associated with higher SVD burden at baseline. Subjects with WMH progression had greater increase in total NPI (␤[SE] = 0.46[0.19], p = 0.016), driven by hyperactivity subsyndrome (␤[SE] = 0.88[0.34], p = 0.007). Subjects with incident CMBs had greater increase in psychosis subsyndrome (␤[SE] = 0.89[0.30], p < 0.001). Subjects with progressions in both SVD and NPS were more likely to develop cognitive decline over 2 years (OR[95% CI] = 4.17[1.06-16.40], p < 0.05). Conclusion:Our findings support worsening of NPS as a clinical indicator of SVD progression and are associated with cognitive decline over 2 years. Early detection of NPS and targeted interventions on SVD burden may improve NPS outcomes.

No abstract