We read with great interest the recent study by Lee et al. exploring associations between retinal vascular disease (RVD) and the risk of subsequent ischaemic and haemorrhagic stroke.1 The article offers robust evidence that retinal vascular occlusion, diabetic retinopathy and retinal haemorrhage are each associated with elevated stroke risk, even after adjustment for traditional vascular risk factors. However, within the context of vascular prevention, the article also demonstrates the limitations of employing static, cross-sectional risk factor assessment. In the diabetic retinopathy cohort, the greater stroke risk in individuals with ‘normal’ cholesterol levels, compared with those with elevated cholesterol, is particularly illustrative. The authors list several possible explanations, including sarcopenia and malnutrition from severe diabetes. Yet, another possible explanation is that a significant proportion of these individuals achieved lipid normalisation through statin therapy after prolonged periods of elevated lifetime vascular risk.2 This is especially likely, given diabetic retinopathy patients are likely to have accumulated sufficient cardiometabolic comorbidity to prompt statin initiation even before their retinal manifestations had become clinically apparent.3 Conditioning on retinal vascular disease effectively selects individuals in whom cumulative vascular exposure has manifested structurally. Subsequent associations with contemporaneous risk factor levels may therefore be distorted by index-event bias, treatment-modified risk factors and single time-point measurements, creating apparent paradoxes such as higher risk among those with ‘normal’ lipid levels. In such instances, current lipid profiles may reflect secondary prevention efforts rather than favourable underlying cardiovascular biology. Thus, retinal microvascular pathology may function less as a conventional risk factor and more as an anatomical record of lifetime vascular injury. Unlike biochemical parameters, which may fluctuate greatly with treatment, behaviour or acute illness, structural microvascular changes can reflect years of exposure to hypertension, dysglycaemia, lipid derangement and endothelial dysfunction.4 This framework offers a biologically coherent explanation for why adjustment for current risk factor levels does not eliminate the observed associations between RVD and stroke. This distinction has important implications for prevention. Reliance on static, normalised risk factor values may lead to false reassurance in individuals with substantial lifetime vascular exposure, while underestimating risk in those whose disease burden is already anatomically encoded. Therefore, dynamic or integrative markers, such as retinal vascular pathology, may better identify patients who would benefit from more aggressive prevention strategies, even if conventional parameters (such as lipids, blood pressure or body mass index) appear well controlled. More broadly, these findings advocate for a shift in preventive paradigms from snapshot-based risk assessment toward approaches that acknowledge the nature of cumulative vascular injury. Retinal vascular assessment, which is performed routinely and non-invasively for ophthalmic indications, may thus represent an underutilised opportunity to detect advanced microvascular disease and prompt timely escalation of stroke prevention efforts. This perspective does not negate the value of risk factor modification, but instead emphasises the requirement to interpret normalised values within the context of lifetime exposure and end-organ effects. Therefore, the authors strengthen the case for integrating structural and dynamic markers into ongoing vascular risk assessment and position the retina as an accessible window into the long-term consequences of systemic vascular stress. Data sharing not applicable to this article as no datasets were generated or analysed during the current study.
Jesudason et al. (Sat,) studied this question.