More than 2,000 years ago, the Greek mathematician Euclid began his treatise Optics with an assumption that now sounds peculiar: visual rays travelled out from the eye towards the objects we see.
Author
- Barbara Pierscionek
Professor of Eye Health and New Technologies for Sight Improvement, Anglia Ruskin University
We now know that vision works the other way around. But modern imaging has given the old idea an unexpected twist: light directed into the eye and scattered back from the retina can reveal information about health elsewhere in the body.
One of the technologies making this possible is optical coherence tomography (OCT) . OCT uses light to produce detailed cross-sectional images of structures at the back of the eye. It has transformed the diagnosis and monitoring of retinal diseases, allowing clinicians to examine individual layers of the retina in remarkable detail.
Researchers are now investigating whether these scans can tell us about diseases beyond the eye.
The retina, the light-sensitive tissue at the back of the eye, contains several specialised layers of cells. Some detect light and convert it into electrical signals, while others help transmit those signals through the optic nerve towards the brain.
This process requires a great deal of energy. The retina has an exceptionally high demand for oxygen and an extensive blood supply. Its blood vessels are also part of the wider circulation, which means changes visible in the eye may reflect processes happening elsewhere in the body.
This has led to growing interest in retinal imaging as a way of investigating diseases elsewhere in the body, including cardiovascular and neurological conditions. A recent study suggests this could include atrial fibrillation, a common heart rhythm disorder.
Researchers analysed retinal images and health records from two large datasets: AlzEye, which links eye scans from Moorfields Eye Hospital with hospital records, and UK Biobank. In the AlzEye group of 50,651 people, those with atrial fibrillation had differences in several retinal layers and blood-vessel measurements compared with those without the condition.
One finding in particular was also seen in UK Biobank: people with atrial fibrillation had a thinner macular ganglion cell-inner plexiform layer. This is part of the inner retina containing cells involved in transmitting visual information towards the brain.
The researchers then looked at 39,013 UK Biobank participants who did not have a recorded diagnosis of atrial fibrillation when their retinal images were taken. Over an average follow-up of 10.3 years, 838 were subsequently diagnosed with the condition. Among these people, diagnosis occurred an average of about four years after retinal imaging.
Those with a thinner ganglion cell-inner plexiform layer were more likely to be diagnosed with atrial fibrillation later. After researchers adjusted for factors including age, sex, ethnicity, high blood pressure, diabetes, smoking and alcohol consumption, each standard-deviation decrease in its thickness was associated with a 27% higher risk of atrial fibrillation.
What can the eye really tell us?
Atrial fibrillation occurs when electrical activity in the heart's upper chambers becomes disorganised, causing an irregular heart rhythm. Some people experience palpitations, breathlessness, tiredness or dizziness, while others have no noticeable symptoms.
The condition is associated with serious complications, particularly stroke and heart failure. Detecting atrial fibrillation can also be difficult because abnormal heart rhythms may occur intermittently and therefore may not be present when somebody is examined.
This makes the retinal findings intriguing. A scan already being carried out for eye health might potentially provide additional information that prompts investigation of somebody's cardiovascular health.
But there is an important gap between finding an association and having a useful screening test.
The study was observational, so it cannot establish why retinal thinning and atrial fibrillation are associated. Atrial fibrillation might contribute to changes in the retina, or both could be influenced by underlying cardiovascular processes.
Nor can the study show that an eye scan can reliably predict who will develop atrial fibrillation.
There is another complication. Atrial fibrillation can come and go, particularly during its early stages, and can cause no noticeable symptoms. The researchers did not continuously monitor participants' heart rhythms. In the longitudinal analysis, a new case was identified when atrial fibrillation appeared in hospital records.
Some people classified as developing atrial fibrillation after their retinal scan may therefore already have had an undiagnosed form of the condition. Around one in five of those who went on to receive a diagnosis did so within two years of retinal imaging.
The researchers also acknowledge that other factors that were not measured or accounted for could influence both retinal structure and the likelihood of developing atrial fibrillation.
An eye on future heart health
The findings are part of a growing field known as "oculomics" , which investigates whether information captured in images of the eye can reveal signs of health and disease elsewhere in the body.
Research has already linked characteristics of the retinal blood vessels and other structures with cardiovascular disease and its risk factors . Advances in artificial intelligence are also allowing researchers to extract increasingly complex information about systemic health from retinal images.
The attraction is clear. Retinal imaging is non-invasive and is already widely used in eye-care settings. If future research establishes that particular retinal features can reliably identify people at higher cardiovascular risk, information contained in scans taken for other reasons could potentially help doctors decide who needs further investigation.
But retinal imaging cannot currently diagnose atrial fibrillation. Diagnosis requires evidence of the abnormal heart rhythm, generally using an electrocardiogram (ECG) or another form of heart-rhythm monitoring .
There are also broader questions about whether retinal biomarkers will prove accurate and specific enough for clinical use. Reviews of retinal imaging and cardiovascular risk have concluded that the technology is promising, but its role in screening and predicting cardiovascular disease remains uncertain.
The next step is therefore to establish whether retinal imaging adds useful information to methods already available, and whether acting on that information improves patients' health.
For now, the study provides another example of how examining the eye may reveal more than the health of our vision. What begins as an image of the retina could eventually help researchers understand what is happening in the heart as well.
Barbara Pierscionek does not work for, consult, own shares in or receive funding from any company or organisation that would benefit from this article, and has disclosed no relevant affiliations beyond their academic appointment.
