P2783Diabetic retinopathy is associated with echocardiographic structural abnormalities and both heart failure with reduced and preserved ejection fraction

Abstract

Background: Type 2 diabetes mellitus (DM) and heart failure (HF) frequently co-exist. There is major interest in the role of microvascular dysfunction in the development and progression of HF in diabetic individuals. However assessing microvascular health is difficult in clinical practice. Diabetic retinopathy screening provides a unique opportunity to assess microvascular health. In this study, we have determined the relationship between retinopathy, echocardiographic abnormalities and HF in DM patients. Methods: We evaluated 8,576 patients in the Genetics of Diabetes Audit and Research in Tayside Scotland (GoDARTS) study with available retinal screening data and echocardiography. Using anonymised record linkage we evaluated the association between diabetic retinopathy, relevant clinical covariates, echocardiographic parameters and cardiovascular outcomes. Logistic regression was used to examine the association between retinopathy and echo parameters and Cox regression used to determine the association between retinopathy and CV outcomes. The combined primary outcome was CV death or HF hospitalisation. Results: Median follow-up from first retinal photograph was 6.6 years. Mean age at time of first retinal photograph was 71±12 years and 57% of the cohort were male. Patients with moderate or worse DR were significantly older, more likely to be male, smokers, use insulin therapy and have a longer duration of diabetes. After adjustment for clinical variables including age, gender, duration of diabetes, mean HbA1c and mean systolic blood pressure prior to study entry, prior CV event and insulin use, the presence of moderate or worse DR was associated with both left ventricular systolic dysfunction (LVSD) and left ventricular hypertrophy (LVH) on echo (OR 1.13, p=0.00012 and 1.14, p=0.00038 respectively), however referable maculopathy was only significantly associated with LVSD (OR 1.05, p=0.036). After adjustment for these same variables both mild and moderate/severe retinopathy were significantly associated with the primary outcome (mild: HR 1.59, 95% CI 1.38-1.83, p=1.18 x10-10, moderate/severe: HR 2.29, 95% CI 1.42-3.68, p=0.0006) (Figure). These remained the same when only patients with preserved LV function were evaluated (mild: HR 1.65, p<0.000001, moderate/ severe: HR 1.97, p=0.047). There was no significant interaction between either the presence of LVSD or LVH on the primary outcome. Conclusion: DR is associated with adverse cardiovascular outcome including HF hospitalisation, including in patients without echocardiographic evidence of reduced EF. The absence of any significant interaction suggests that DR and thus microvascular disease may be associated with both HF with reduced EF and HF with preserved EF. Critically this seems to be independent of important risk factors including severity and duration of diabetes and prior CV events. (Figure Presented) .