According to a multicenter study published in Investigative Ophthalmology & Visual Science, optical coherence tomography and angiography imaging can flag silent brain injury and cognitive risk in people with large artery stenosis, even before an overt stroke occurs. Researchers found that specific eye measurements correlate with cerebral atrophy and lower cognitive scores.
How Retinal Imaging Detects Vascular Brain Injury
Large artery stenosis can drive cognitive impairment and neurodegeneration without causing obvious strokes. Because scalable biomarkers for capturing this silent vascular injury are limited, investigators examined whether retinal and choroidal microvascular and neurostructural measures reflect these changes, according to the study. Investigators enrolled 1,239 participants across three tertiary hospitals in China. This cohort included 1,035 patients diagnosed with noninfarcted large artery stenosis and 204 age- and sex-matched community controls.
Did you know?
The study analyzed 1,941 eyes from patients with large artery stenosis alongside 392 eyes from the control group.
Patients with large artery stenosis exhibited significantly thinner ganglion cell-inner plexiform layer thickness, measuring 63.17μm compared to 66.58μm in controls. They also showed lower superficial vascular complex vessel length density at 18.64mm-1 versus 20.69mm-1. Furthermore, deep vascular complex vessel length density, choriocapillaris vessel length density, and the choroidal vascularity index were all markedly reduced in the stenosis group, according to the published data.
Linking Ocular Microvascular Metrics to Brain Atrophy
Brain volume reductions accompanied these ocular changes in patients with large artery stenosis. Gray matter to total intracranial volume dropped to 35.24% compared to 35.66% in controls, while white matter ratios fell to 30.75% from 31.37%. Median Montreal Cognitive Assessment scores also dipped, registering at 22 out of 30 for patients versus 25 for controls.
Retinal and choroidal measures showed positive associations with both gray and white matter volumes, as well as cognitive scores. These correlations appeared strongest in subcortical regions, particularly the thalamus, hippocampus, and basal ganglia, and maintained a generally bilateral pattern. According to the authors, ocular microvascular metrics—especially choriocapillaris and superficial vascular complex density—displayed stronger associations with brain and cognitive outcomes in stenosis patients than in controls. This suggests that ocular microvascular biomarkers link closely to neurodegenerative burden under vascular stress.
Machine Learning Models Improve Risk Stratification
To evaluate the predictive value of these ocular features, researchers trained machine learning models internally on 891 participants and validated them externally on 348 participants, as detailed in the PubMed publication data. Integrating optical coherence tomography and angiography metrics into these models substantially improved predictions for cerebral atrophy and cognitive impairment beyond standard clinical baselines.
For gray matter volume, a random forest model achieved an external test R2 of 0.456, outperforming the enhanced clinical baseline model score of 0.254. For white matter volume, the random forest model reached an R2 of 0.430 compared to 0.251 for the baseline. Meanwhile, a support vector machine delivered the top classification performance for cognitive impairment, securing an area under the curve of 0.794 against 0.668 for the baseline model.
While cross-sectional data limits causal determinations, combining ocular imaging with machine learning offers substantial incremental value for individual risk stratification in vascular contributions to cognitive impairment and dementia, according to the study authors.
Future Directions in Dementia Research
The study data supports retinal and choroidal optical coherence tomography and angiography biomarkers as complementary indicators of vascular-related neurodegeneration. Investigators concluded that these findings advance the understanding of retina-brain relationships in vascular disease, highlighting a path toward early detection and precision approaches in dementia research. However, the team clarified that future interventional studies remain necessary to determine whether improving vascular function modifies ocular biomarkers and downstream brain outcomes.
Frequently Asked Questions
What is large artery stenosis?
Large artery stenosis involves the narrowing of major blood vessels, which can drive cognitive impairment and neurodegeneration even without causing an overt stroke.
How do eye scans measure brain health?
Optical coherence tomography and angiography capture retinal and choroidal microvascular changes that track closely with cerebral atrophy and lower cognitive scores in subcortical and medial temporal brain regions.
Can these biomarkers replace standard cognitive tests?
No. Researchers found that integrating ocular imaging into machine learning models improves risk prediction beyond clinical baselines, serving as a complementary indicator rather than a standalone replacement.
Stay Updated on Vascular Research
Explore more articles on neurology, advanced imaging technologies, and dementia research. Subscribe to our newsletter to receive the latest clinical findings directly in your inbox.