0
0
A new study published in Heart reveals that a vascular “fingerprint” on the retina—the light-sensitive tissue layer at the back of the eye—could predict a person’s risk of stroke with the same accuracy as traditional risk factors, but without the need for invasive lab tests. This breakthrough could make stroke risk assessment more accessible, particularly in primary health care settings or regions with limited resources.
The retina’s intricate vascular network shares common features with the vasculature of the brain, making it a valuable tool for assessing systemic health issues, such as diabetes. While the retina’s potential for predicting stroke risk has been explored before, inconsistent findings and the use of specialized imaging techniques have hindered its widespread adoption. However, with advancements in machine learning, researchers have now developed the Retina-based Microvascular Health Assessment System (RMHAS), which can accurately identify biological markers of stroke risk from retinal images.
In this study, researchers analyzed fundus images from 68,753 participants of the UK Biobank, measuring 30 retinal indicators across five categories: caliber (length, diameter, ratio), density, twistedness, branching angle, and complexity of blood vessels. They also considered traditional stroke risk factors, including age, sex, smoking, diabetes, blood pressure, cholesterol levels, and body mass index (BMI).
The study followed 45,161 participants, with an average age of 55, for an average of 12.5 years. During this period, 749 participants suffered a stroke. The results revealed that 29 specific retinal vascular indicators were significantly associated with the risk of first-time stroke, after adjusting for traditional risk factors. Notably, changes in vascular density, complexity, and twistedness were linked to an increased stroke risk, with a 10–19% higher likelihood for each change in certain indicators.
When combined with just basic demographic factors like age and sex, the retinal vascular fingerprint proved as effective as traditional stroke risk factors in predicting future strokes.
Although this study’s observational nature prevents drawing definitive conclusions about cause and effect, the researchers believe their findings are promising. Given that retinal parameters can be easily measured through routine fundus photography, this approach offers a practical and non-invasive way to assess stroke risk, particularly in primary care and low-resource settings.
Despite the encouraging results, the study acknowledges some limitations. Most of the UK Biobank participants were white, so the findings may not be applicable to all ethnic groups. Additionally, the study did not differentiate between types of strokes, meaning further research is needed to explore these distinctions.
In conclusion, this research suggests that retinal vascular fingerprints could be a game-changer in stroke risk assessment, offering a low-cost and non-invasive alternative to traditional diagnostic methods.
For more information:
Retinal vascular fingerprints predict incident stroke: findings from the UK Biobank cohort study, Heart (2025). DOI: 10.1136/heartjnl-2024-324705
