A recent study explored a comprehensive set of retinal vascular parameters for associations with stroke to understand whether these parameters could be used to predict stroke risk.

Led by Mayinuer Yusufu of the Centre for Eye Research Australia at The Royal Victorian Eye and Ear Hospital in East Melbourne, Victoria, Australia, the researchers analyzed 45,161 participants from the UK Biobank, with a median follow-up of 12.5 years. Retinal vascular parameters were extracted using the Retina-based Microvascular Health Assessment System, and Cox regression analysis was performed to assess associations between these parameters and incident stroke.

To control for multiple comparisons, false discovery rate adjustments were applied. The investigators also evaluated the predictive value of retinal parameters using receiver operating characteristic (ROC) curves to compare their performance against traditional stroke risk factors.

The findings, recently published in Heart, showed a significant association between retinal vascular parameters and stroke risk. The researchers identified 29 retinal vascular parameters as significantly associated with stroke risk. Density parameters accounted for more than half of these indicators. Each standard deviation (SD) change in density parameters increased stroke risk by 9.8% to 19.0%. 

Each SD change in caliber parameters was linked to an increased stroke risk of 10.1% to 14.1%, while complexity parameters and arterial inflection count tortuosity showed a different trend—each SD decrease was linked to an increased stroke risk of 10.4% to 19.5%.

When retinal vascular parameters were incorporated into the stroke prediction model, the area under the ROC curve (AUC) improved from 0.739 to 0.752, which suggests that retinal imaging can enhance traditional risk assessment models.

The investigators concluded: “The 29 novel retinal indicators identified offer new avenues for stroke pathophysiology research.”

A full list of author disclosures can be found in the published research.