Retinal artery occlusion as indicator of stroke in diabetes

Abstract

Abstract Background Diabetes is associated with multiple complications in organs throughout the body, including an increased risk cardiovascular disease (1). Furthermore, an association between retinal artery occlusion and diabetes have been described as well (1,2). Early predictors of complications associated with diabetes would be useful for a better prognosis of the disease. Purpose Based on data from the Danish registries, the aim of this study was to investigate whether RAO is a predictor of stroke in diabetic patients. Methods We included 992 diabetic patients with incident RAO from the Danish National Patient Register with incident RAO between 1st of January 2000 and 31st of December 2018. Each of these cases were matched on age at index, sex, and time since diabetes diagnosis with 5 random diabetes patients with no subsequent diagnosis of RAO. Survival analyses were used to estimate and contrast the risk of stroke in diabetic RAO patients compared to diabetic patients with no prior diagnosis of RAO. The included analyses were all performed at 1 and 5-year follow-up. Results At 1-year follow-up, the incidence rate of stroke in diabetic RAO patients was 10.0 per 100 person years (PY) (95% CI: 8.13–12.10). The incidence for the controls at this time point was much lower at 1.82 per 100 PY (95% CI: 1.47–2.25). After 5-years of follow-up a large difference persisted, where the cumulative incidence was 19.3% (95% CI: 16.5–22.2%) for the RAO patients compared to 7.4% (95% CI: 6.5–8.2%) for the controls. The hazard rate ratio for stroke was 5.48 (95% CI: 4.06–7.39) after 1-year follow-up and 3.17 (95% CI: 2.60–3.86) after 5-year follow-up. The results support the use of RAO as a predictor of stroke in diabetic patients. Cardiovascular diseases contribute substantially to the morbidity and mortality in diabetic patients (3,4), underlining the need for especially a predictor of cardiovascular complications in these patients. To our knowledge, the predictive usage of RAO for cardiovascular complications in diabetic patients have not been investigated previously. Small occlusions, which could indicate systemic atherosclerosis, may not always be symptomatic. Therefore, the condition may not be diagnosed before more severe events occur. Conversely, RAO usually cause visual impairments, making RAO suitable as a potential predictor of subsequent cardiovascular events. The incidence rates of stroke suggest that diabetic RAO patients should be especially monitored the first year after their RAO diagnosis, but the monitoring would be beneficial even 5 years after the RAO diagnosis to continuously evaluate risk factors and instigate early intensified treatment. Conclusion The data of this study identified a substantial larger risk of stroke in diabetic RAO patients than diabetic patients with no RAO diagnosis, indicating RAO could be a predictor of stroke in diabetic patients. Funding Acknowledgement Type of funding sources: Private grant(s) and/or Sponsorship. Main funding source(s): The Obel Family Foundation