Abstract: Objective Currently mechanisms of diabetic nephropathy(DN) and diabetic retinopathy(DR) remains unclear, but the hypothesis that persistent hyperglycemia and its abnormal metabolites play a vital role has been controversial. It is necessary to analyze the relationship of type 2 diabetic retinopathy and diabetic nephropathy in the aspect of onset and development of them, and whether we can use fundus examination to predict the renal injury level. This study aimed to find the relationship of diabetic retinopathy and diabetic nephropathy. Methods Patients with non-diabetic retinopathy(NDR) or proliferative diabetic retinopathy(PDR) were enrolled. Based on 8-hour urinary albumin excretion rate(UAER), the patients were divided into 3 groups: group A: UAER<20 μg/min, group B: UAER≥20~199 μg/min; group C: UAER>200 μg/min. Group A had 169 patients, including 53 NDR patients, PDR 116 patients; group B had a total of 76 patients, including 37 patients with NDR, 39 with PDR; group C had a total of 66 patients, including 40 patients with NDR, 26 with PDR. The glomerular filtration rate was compared between the subgroups. Results Glomerular filtration rate(GFR) was calculated according to the currently recommended MDRD simplified formula: eGFR=186×(SCr)^-1.154×(Age)^-0.203×(0.742 women). We found that among the three groups, PDR subgroup eGFR was lower than those in NDR. Significant difference in eGFR was found in groups B and C in PDR sub-group compared with the NDR sub-group(P<0.01). Conclusion Proliferative diabetic retinopathy appears to predict the degree of renal injury, but urinary albumin excretion rate had poor correlation with renal injury. The cause may be poor specificity of microalbuminuria as a biomarker of DN in diagnosis.

Key words: type 2 diabetes mellitus, diabetic retinopathy, glomerular filtration