What retinal imaging reveals about the brain-eye connection with Sharon Fekrat, MD, FACS, FASRS

Sharon Fekrat, MD, FACS, FASRS, is a professor of ophthalmology, neurology, and surgery at Duke University School of Medicine as well as the founder and director of the iMIND Study Group, a multidisciplinary research program evaluating retinal imaging as a biomarker for neurologic health. Her work spans Alzheimer disease, Lewy body dementia, traumatic brain injury (TBI), and other neurodegenerative and neurovascular conditions, with more than 2000 participants enrolled to date.

In a recent conversation with the Eye Care Network and Optometry Times, she discussed the current and potential role of retinal imaging in cerebral visual impairment (CVI), how the eye-brain connection could inform future clinical training and referral patterns, and what she hopes the field will discover next.

What role can retinal imaging play in the diagnosis and understanding of cerebral visual impairment (CVI)?

Sharon Fekrat, MD, FACS, FASRS: Retinal imaging cannot directly diagnose CVI, since the dysfunction lies in the posterior visual pathways and cortical visual networks in the brain, beyond what any retinal imaging can visualize. A patient with CVI may have a completely normal eye exam and normal OCT. That said, retinal imaging can help in several ways.

First, it can help rule out an ocular or ophthalmic cause of visual impairment, which is especially useful in patients who cannot complete visual acuity testing, visual field testing, or electrophysiologic testing. Second, damage behind the eye can sometimes cause retrograde degeneration visible on optic nerve imaging, such as thinning of the retinal nerve fiber layer, functioning as a kind of downstream structural signal of injury elsewhere in the visual pathway. Third, in children who are premature or have developmental brain abnormalities, retinal and optic nerve measurements may eventually show abnormalities that help characterize the broader brain injury when combined with neuroimaging and functional visual assessments. OCT is a complementary test for CVI, but it is not diagnostic for it.

Can you tell us about the origin and goals of the iMIND research program?

Fekrat: iMIND began in 2017 after I saw a pair of 96-year-old identical twins in clinic. One was cognitively normal; the other had advanced Alzheimer disease. Since they had the same DNA, I wanted to image their retinas to see if there were differences, and there were: the twin with Alzheimer disease had decreased vessel density and perfusion density in the macula. That finding is what led to iMIND.

We study multiple diseases because it lets us ask 2 important questions: are there retinal findings shared across neurodegenerative, neurovascular, and neurocognitive disorders, and are there disease-specific patterns that could help distinguish, for example, Alzheimer disease from Lewy body dementia or frontotemporal dementia.

I don't think retinal imaging will be a standalone diagnostic tool. It's more likely to be one part of a larger framework that includes a neurologic exam, plasma biomarkers, neurocognitive testing, genetics, and brain imaging, with AI and machine learning helping to extract information from these images that we may not be able to see ourselves.

How does traumatic brain injury fit into this research, and what have you found?

Fekrat: TBI is an important model of the brain-eye relationship, in some ways similar to CVI, since people can walk around with TBI without it being obvious. TBI involves both immediate and more chronic changes in the brain, including changes in blood flow and inflammation, and many patients have persistent, difficult-to-define visual symptoms rather than simple blurry vision.

In our study, we imaged 36 patients with TBI, mostly mild, and 36 age- and sex-matched controls, representing about 60 eyes in the TBI group and about 70 in the control group. Most patients were imaged years after their injury rather than acutely. We found decreased vessel and perfusion density in the macula, a pattern similar to what we've observed across other neurodegenerative conditions we've studied. This work, like most of our studies to date, was cross-sectional rather than longitudinal, and I think longitudinal work will tell us much more about the timing and sequence of these changes.

As someone who trains fellows and residents, how do you think ophthalmology and optometry training could evolve to better prepare clinicians to recognize brain-based visual conditions rather than only structural or refractive eye disease?

Fekrat: Trainees first have to understand normal anatomy and function before they can understand disease states, and then they need to learn the full range of eye diseases, both isolated ocular conditions and those related to systemic health.

For ophthalmology residents especially, the demands of learning surgery and laser technique mean that understanding how the brain affects the retina and optic nerve tends to become a secondary focus until there's something more tangible to offer patients. I think this interest is more likely to develop during fellowship, particularly a neuro-ophthalmology fellowship, which draws trainees from both ophthalmology and neurology residencies.

When should a general optometrist or ophthalmologist recognize that a patient's visual complaint might have a neurological origin and warrant referral?

Fekrat: If a patient has visual perception complaints despite normal visual acuity, for example, 20/20 on the eye chart, and a normal complete eye exam, that's when clinicians should start considering a cause related to the brain rather than dismissing the complaint. That's the point at which further workup, such as neuroimaging, becomes appropriate rather than optional.

What is the one message you want eye care professionals to take away about the brain-eye connection?

Fekrat: The future is bright. There's still a great deal we don't understand about the eye-brain connection, but there's growing recognition that changes happening in the retina and optic nerve mirror what's going on in the brain. With the explosion of high-resolution retinal imaging technology and the application of AI, I think we're going to discover things we never thought existed.