AbstractBackgroundMicrovascular injury is a significant contributor to cognitive impairment. Conventional neuroimaging is inadequate for assessing microvascular changes in the brain. Intensity‐based Retinal Arteriolar Visualization in OCT (I‐bRAVO) is a novel method that enables rapid, direct and non‐invasive visualization of retinal arterioles using standard optical coherence tomography images. We aimed to assess retinal microvascular changes using I‐bRAVO and compare it to measures of arteriolosclerosis, white and grey matter damage, and cognitive function.MethodWe included 500 eyes from 250 subjects with sporadic (n=91) and genetic (i.e., Cerebral Autosomal Dominant Arteriopathy with Subcortical Infarcts and Leukoencephalopathy, n=21) cerebral small vessel diseases (CSVD), non‐vascular neurodegenerative diseases (NVND, n=113), and healthy controls (HC, n=25). Mean Wall Thickness (MWT) and Wall to Lumen Ratio (WLR) were analyzed in association with imaging and cognitive markers of vascular pathology, including novel imaging classifier of brain arteriolosclerosis (ARTS), MRI markers of white matter injury (Fazekas scale), and grey and white matter atrophy, as well as the Uniform Data Set (v3.0) executive function (UDS3‐EF) score. Linear mixed‐effect models with subject‐specific intercepts were fitted using restricted maximum likelihood estimation.ResultARTS score correlated with MWT (ρ =0.44, P<0.001) and WLR (ρ=0.48, P<0.001). MWT and WLR were higher in CSVD (MWT: 28.18µm, WLR: 0.30) compared with NVND (MWT: 26.72µm, WLR: 0.29) and HC (MWT: 25.17µm, WLR: 0.25) (P<0.001). MWT and WLR were positively associated with superficial white matter disease burden (B= 0.07, 95% CI [0.05 – 0.1], and 5.63 [4.03 – 7.22], respectively, P<0.0001). A similar association was observed between MWT, WLR and deep white matter lesion load. WLR correlated inversely with grey (ρ =0.40, P<0.05) and white matter volume (ρ =0.48, P<0.01) in subjects with vascular cognitive impairment. WLR differentiated cognitively healthy individuals with vascular white matter lesions from those without (AUC= 0.81, P<0.001) with 94% specificity (cutoff value: 0.275). In asymptomatic CSVD, higher WLR was associated with a lower age‐, sex‐ and education‐adjusted UDS3‐EF (β=‐3.1 [‐6.3 – ‐0.01], P<0.05).ConclusionWe describe a rapid, inexpensive novel technique for assessing retinal arterioles as a surrogate marker for intracranial vascular pathology and as a tool for early recognition of presymptomatic CSVD.