Are eye tests a window to the brain?

Tim Shakespeare, Research Communications Officer, takes a look at the science behind the headlines regarding using eye tests to diagnose dementia.

Image by Keith Williams, used under the Creative Commons license.

From the above newspaper ‘Alzheimer’s eye test’ headlines you’d be forgiven for thinking that new eye tests to diagnose dementia are about to hit the high street. There’s certainly something in the idea that the eyes can give us clues to what’s going on in the brain, but what do we know about the eyes and dementia and what would it take to make a useful diagnostic test?

Interest in the eye stems from the strong relationship it has with the brain. The part of the eye that is sensitive to light (called the retina) is closely connected to the brain by a bundle of nerve fibres. This bundle relays visual information from the retina for the brain to process. As the retina and brain are strongly connected parts of the same system, the nerve fibres in the retina are likely to reflect changes seen in the brain.

Unlike the brain it is easy and inexpensive to take detailed images of the retina and examine them for signs of disease. For example, people who have diabetes are invited to have annual retinal photographs to detect early changes in blood supply of the retina. This enables early treatment if a problem is detected. The strong relationship to the brain and the ability to examine the eye in detail makes eye testing an interesting prospect for diagnosis of dementia.

What do we know?

A number of studies have investigated changes in the retina in people with Alzheimer’s disease. Most research has focused on a layer of nerves in the retina that sends visual signals to the brain (called the retinal nerve fibre layer). This layer tends to be thinner in people with Alzheimer’s disease. It’s thought that thinning of the retinal nerve fibre layer might reflect the shrinking of the brain seen in Alzheimer’s disease. However not all of the studies have found this difference. These conflicting results could be down to things like differences in the number of people tested and the particular techniques used to collect images.

The fact that people with Alzheimer’s disease seem to have thinner retinal layers on average doesn’t tell us whether this information can actually help doctors to make an accurate diagnosis. Whether it is useful for diagnosis depends on how big that difference is, and how much variation there is. For example we know that on average men are taller than women, but there’s so much variation that you couldn’t use someone’s height to reliably say if they were male or female.

In order to be effective, a good diagnostic test needs to be both sensitive and specific. Sensitive enough to make sure it doesn’t miss people who have the disease, and specific enough to make sure it doesn’t wrongly identify people as having the disease. These aspects aren’t well established for eye tests, so working them out is the next step to investigate their use for diagnosis.

The future of eye tests for dementia

We’ve described tests that look at the thickness of a certain layer of the retina, but there are other ways of imaging the eye that might also reveal changes in Alzheimer’s disease. These include attempts to use a special form of photography to see one of the hallmarks of the disease (called amyloid protein) in the retina. However, for the moment eye tests for dementia are at a very early stage. There are some interesting signs that they could be helpful, but there’s a lot of work to do to find out just how helpful they are, and whether they can add useful information to the methods that doctors already use for diagnosis.

We can expect to hear more about the eye and Alzheimer’s disease; two very large studies in the UK called the UK Biobank and the Deep and Frequent Phenotyping study are both measuring changes in the retina of people in later life, and have the potential to shed further light on the issue.



Add your own

I need to get my father diagnosed as I'm sure from his behaviour that he is in the early stages of dimensa. The problem I have is that he refuses to goto his GP. Is it at all possible for the doctor to perform a diagnosis from dad's home?

Hi Nicola, please contact our Helpline, where you'll be able to chat in more detail to one of our advisers about your father's symptoms and the next steps to take. Call 0300 222 1122 or take a look at our Helpline page here:…

I'm sure there is much to be said for this. I am convinced that following a visit to the optician by my wife, where they noticed something of enough concern to refer her to the eye clinic, was an early, albeit unrecognised sign of what unfolded over the next couple of years. I can't be specific about dates, but I remember after 3 appointments with the eye specialist, a referral for an MRI scan and neurosurgeon, all inconclusive and all missing what was actually happening, left me thinking that my wife's increasingly strange and uncharacteristic behaviour was just a figment of my imagination, rather than symptoms of early onset Alzheimer's. I fully understand nothing would have changed the outcome, and hindsight is a wonderful thing, but had such a test been available, it could have fore shortened much anguish and soul searching for the whole family. This was over 6 years ago now, I hope things have since advanced, and that serious work will proceed in this key area.

I am very interested in any ways of 'finding/seeing' signs indicating early onset Alzheimer's, I'm a firm believer in the earlier the detection of it the more chance of a reasonable quality of life.

I am convinced loss of the sense of smell is a precursor to dementia. My wife lost hers about 35 years before onset. As I worked for the NHS once I was able to search for confirmation of this and quite a bit was available, now available on the internet of course. When I mentioned it to my wife's psychiatrist she said 'Nonsense I have never heard of such a thing'. So I gave her a reading list of references to follow up and proved I was right.

Leave a comment

Your email address will not be published. Required fields are marked with *

The content of this field is kept private and will not be shown publicly.