Retinal arteriolar tortuosity and fractal dimension are associated with long-term cardiovascular outcomes in people with type 2 diabetes

Abstract

Aims/hypothesis Our aim was to determine whether quantitative retinal traits in people with type 2 diabetes are independently associated with incident major cardiovascular events including CHD and stroke. Methods A total of 1066 men and women with type 2 diabetes, aged 65–74 years, were followed up over 8 years in the population-based Edinburgh Type 2 Diabetes Study. Using retinal photographs taken at baseline and specialist software, a number of quantitative retinal traits were measured, including arteriolar and venular widths and tortuosity as well as fractal dimension (a measure of the branching pattern complexity of the retinal vasculature network). Incident CHD events occurring during follow-up included fatal and non-fatal myocardial infarction, first episodes of angina and coronary interventions for CHD. Incident cerebrovascular events included fatal and non-fatal stroke or transient ischaemic attack. Cox proportional hazard regression analyses were performed to identify the association of the retinal traits with cardiovascular events in the population with retinal data available (n = 1028). Results A total of 200 participants had an incident cardiovascular event (139 CHD and 61 cerebrovascular events). Following adjustment for age and sex, arteriolar tortuosity and fractal dimension were associated with cerebrovascular events (HR 1.27 [95% CI 1.02, 1.58] and HR 0.74 [95% CI 0.57, 0.95], respectively), including with stroke alone (HR 1.30 [95% CI 1.01, 1.66] and HR 0.73 [95% CI 0.56, 0.97], respectively). These associations persisted after further adjustment for established cardiovascular risk factors (HR 1.26 [95% CI 1.01, 1.58] and HR 0.73 [95% CI 0.56, 0.94], respectively). Associations generally reduced in strength after a final adjustment for the presence of diabetic retinopathy, but the association of fractal dimension with incident cerebrovascular events and stroke retained statistical significance (HR 0.73 [95% CI 0.57, 0.95] and HR 0.72 [95% CI 0.54, 0.97], respectively). Associations of retinal traits with CHD were generally weak and showed no evidence of statistical significance. Conclusions/interpretation Arteriolar tortuosity and fractal dimension were associated with incident cerebrovascular events, independent of a wide range of traditional cardiovascular risk factors including diabetic retinopathy. These findings suggest potential for measurements of early retinal vasculature change to aid in the identification of people with type 2 diabetes who are at increased risk from stroke. Graphical abstract

Risk Assessment of CHD Using Retinal Images with Machine Learning Approaches for People with Cardiometabolic Disorders

“…For retinal fractal dimension characteristics, we found a negative association of MAangle, MAasymmetry, and MVasymmetry with CHD. Similar associations have been found with cerebrovascular events and stroke previously [ 74 ]. We have also found larger occlusion and more exudates in the CHD group than the control group in this study, consistent with previous studies [ 75 , 76 , 77 ].…”

“…Several studies have reported positive associations between retinal vessel tortuosity and cyanotic CHD, intracranial artery atherosclerosis, diabetes, and cerebrovascular events. However, one research study reported a negative association with ischemic heart disease death [ 68 , 71 , 72 , 73 , 74 ]. Our study result demonstrated a positive association between retinal vessel tortuosity and CHD.…”

Diabetic retinopathy is also an important marker of cardiovascular risk in type 2 diabetes, with practical implications

“…3 Subtle alterations of the retinal status, as arterial tortuosity and fractal dimension, can even predict such events in the absence of diabetic retinopathy. 4 Due to the lower effective of our population, our analysis does not question the relevance of Khunti's findings: we agree that CKD and conventional risk factors can be considered as interesting predictors of cardiovascular events in type 2 diabetes. A problem may, however, rise if the categorization to high cardiovascular risk leads to prescribe new antidiabetic drugs like GLP-1 receptor agonists, as suggested by Khunti.…”

Evolution of Quantitative Optical Coherence Tomography Angiography Markers with Glycemic Control: A Pilot Study

“…FAZ area is a metric that correlates to DR stage, predicts risk of progression to a more severe DR stage, and has been very recently associated with long-term cardiovascular outcomes in people with T2D [ 16 ]. A recent meta-analysis by Zhang et al (2021) [ 17 ] compiled and analyzed 2241 diabetic eyes and showed that FAZ area increased with worsening DR and in DR-free patients compared to healthy controls.…”