Eye tests could be a simple way to assess stroke risk. A new study identified 29 blood vessel 'fingerprints' in the retina – the light-sensitive layer of tissue in the back of our eyes – that are significantly associated with increased risk of stroke.
The international team of researchers says their AI-powered technique could help flag problems earlier and save lives – as around 90 percent of strokes are attributed to modifiable risk factors, including blood pressure and poor diet.
"When combined with age and sex, the newly identified retinal parameters had comparable predictive power for stroke risk when compared with established traditional risk factors," writes the team in their published paper.
The work builds on previous research into how the eyes and brain are connected – and how eye blood vessels can reflect characteristics of brain blood vessels. Strokes are caused by disruptions or blockages to normal blood flow to the brain, depriving it of oxygen and nutrients.
Retina images taken via fundus photography – with a microscope-like camera – were collected from 45,161 individuals enrolled in a wider research database. Of those participants, 749 people had a stroke during the period covered by the database, which averaged 12.5 years.
A machine learning AI approach was used to spot consistent eye blood vessel patterns in volunteers who had suffered strokes – including the density and shape of the blood vessels. A total of 29 retinal vascular traits were spotted that were associated with stroke risk.
The retinal parameters identified by AI were linked to a 9.8–19.5 percent increase in the likelihood of a stroke, the researchers found – and they were able to tie in earlier research on eye blood vessels and stroke risk.
"This finding was consistent with previous studies that found associations with stroke risk factors, including age, hypertension and atherosclerosis," write the researchers.
"Our findings indicate that this association is mainly due to arterial density parameters. Pathologically, this could result from compromised oxygen and nutrient supply."
In other words, some of the underlying problems that can trigger strokes could also be affecting the blood vessels in the eyes – though the study isn't comprehensive enough to prove direct cause and effect.
Predicting the likelihood of a stroke is complex, as many different factors play into it, from what we eat to how we sleep. Not all of those will show up in eye tests, but these could certainly help with early identification of increased risk.
As with any kind of reliable diagnostic scan that assesses health, if more people are able to get warnings earlier, that means more opportunity to do something about it – and potentially prevent progression to a more serious condition.
"This model presents a practical and easily implementable approach for incident stroke risk assessment, particularly for primary healthcare and low-resource settings," write the researchers.
The research has been published in Heart.