Chronic Kidney Disease and Diabetic Retinopathy in an Asian population

Abstract

Purpose: Diabetic nephropathy (DN) is commonly associated with diabetic retinopathy (DR). Few studies have demonstrated that chronic kidney disease (CKD) defined as an estimated glomerular filtration rate (eGFR) <60 mL/min/1.73m2 is associated with DR. We hypothesized that the association between CKD and DR would be stronger in the presence of microalbuminuria, an early marker of DN.

Methods: 301 participants with diabetes and of Chinese, Malay and Indian ethnicity aged ≥40 years who participated in the Singapore Prospective Study Program were included. Diabetes was defined according to the recent American Diabetes Association guidelines. Retinal photographs taken from both eyes were graded for DR using the modified Airlie House Classification. Based on the presence of CKD and micro/macroalbuminuria (urinary albumin-to-creatinine ratio ≥30) we defined 4 categories: 1) no CKD and normoalbuminuria (referent) 2) no CKD and micro/macroalbuminuria 3) CKD and normoalbuminuria 4) CKD and micro/macroalbuminuria. We calculated the odds ratios (OR) and 95% confidence interval (CI) of any-DR associated with CKD (n= 54) and micro/macroalbuminuria (n=116) separately and in combination adjusting for potential confounders. We validated our findings, using an independent cohort of Malay adults (n=265) with similar methodology in Singapore.

Results: The prevalence of any-DR was 32.9% in the study population. In separate models, micro/macroalbuminuria was significantly associated with any-DR (OR [95% CI] = 2.27 [1.32-3.91]). However, CKD was not significantly associated with any-DR (1.39 [0.70-2.77]). In the combination model, compared to the referent group, the OR (95% CI) of any-DR were: 2.30 (1.26-4.19) for no CKD and micro/macroalbuminuria, 1.38 (0.49-3.87) for CKD and normoalbuminuria, 2.73 [1.10-6.77] for CKD and micro/macroalbuminuria. Similar findings for any-DR were observed in the replication cohort (3.86 [1.34-9.09] for no CKD and micro/macroalbuminuria, 1.89 (0.56-6.06) for CKD and normoalbuminuria, 4.86 [1.91-12.36] for CKD and micro/macroalbuminuria.

Conclusions: We found that in a multi-ethnic sample of Asian adults, CKD is strongly associated with DR in the presence of micro/macroalbuminuria. Our findings suggest that in the presence of micro/macroalbuminuria, CKD is more likely to be due to DN and therefore more closely associated with DR suggesting a common microvascular pathogenesis.

Keywords: 499 diabetic retinopathy • 498 diabetes • 464 clinical (human) or epidemiologic studies: risk factor assessment