EURETINA 2024: Subretinal drusenoid deposit load and correlation with an increased risk of stroke
Research findings from Moorfields Eye Hospital also indicate SDD and drusen differ in many factors related to stroke risk
The study included optical coherence tomography and colour fundus photographs for 67,687 participants. Image credit: ©David L/peopleimages.com – stock.adobe.com
Roy Schwartz, MD, MRes, and colleagues reported that a higher load of subretinal drusenoid deposits (SDD) may be a marker for an elevated risk of stroke. Dr Schwartz, from Moorfields Eye Hospital, London, reported this result at the 24th European Society of Retina Specialists (EURETINA) Congress in Barcelona.
The investigators explained that there is increasing evidence that drusen and SDD differ in a number of factors including theanatomic location, risk of late age-related macular degeneration and biochemical composition. These differences, they commented, suggest that drusen and SDD are different disease entities with different biologic pathways.
In the study under discussion, they investigated the associations between the SDD number and myocardial infarction and stroke.
Dr Schwartz and colleagues used data from the large, multi-centre, community-cohort UK Biobank study that was comprised of 500,000 volunteer participants (aged 40 to 69 years) who had been recruited between 2006 and 2010 throughout the UK.
The study included optical coherence tomography and colour fundus photographs for 67,687 participants and primary and tertiary care clinical data.
Methods
Patients with SDD and drusen were identified, and the number of drusen and SDD were quantified in each patient. The study also included controls.
The patients were included if they were older 60 years or older, had a minimum of 5 drusen or SDD identified by the framework that were included in this analysis and the drusen exceeded 63 microns in at least 1 eye, the investigators explained.
A history of stroke that included ischaemic stroke, intracerebral haemorrhage and subarachnoid haemorrhage or myocardial infarction (MI) that included both ST elevation MI and non-ST elevation MI was recorded. The investigators performed univariable and multivariable logistic regression for stroke with adjustment for age, sex and SDD count in stroke model 1; the ratio of high-density lipoprotein-cholesterol (HDL-C) to low-density lipoprotein-cholesterol (LDL-C) in stroke model 2; and smoking history in stroke model 3. The logistic regression models for MI were adjusted sequentially for the same variables as in the 3 stroke models.
Effects of SDD
The study included 768 subjects with drusen, of whom 401 had drusen without SDD ("pure drusen"); 439 with SDD (of whom 71 had "pure SDD"); and 1,170 controls. The patients were a median age of 64 years (interquartile range, 61-66 years).
The results showed that the univariable analysis showed that “the odds ratio [OR] for experiencing a stroke increased by 1.02 (or 2% increase) for every 20 reticular pseudodrusenlesions (p=0.028).”
Dr Schwartz and colleagues reported, “SDD remained associated with the increased odds of stroke even after adjusting for age and sex (stroke model 1) and HDL-C/LDL-C (stroke model 2). The OR for stroke (1.02, p = 0.063) was no longer significant after adjusting for smoking history (stroke model 3).”
The mean number of SDDs was 197 (range, 5-2,834). “Considering the estimate of multivariable stroke model 2 (OR 1.02 per 20 SDD), the average patient with SDD would have a 19% increase in the odds of stroke compared with patients without SDD,” the investigators said.
The authors found no significant associations for MI with SDD, or for stroke or MI with drusen.
Dr Schwartz and colleagues concluded, “A higher SDD load could be a marker for elevated stroke risk and should possibly warrant targeted cardiovascular investigations in patients with high SDD numbers. A similar association was not found for drusen, supporting the hypothesis that the two phenotypes represent different disease entities, with different systemic implications. Studies with larger sample sizes are needed to confirm these findings.”
