Background/Aims: Diabetic kidney disease (DKD) is a leading cause of end-stage renal disease. However, no effective treatments for this disease are available. Calcium dobesilate (CaD) is widely used to treat diabetic retinopathy. DKD and retinopathy often co-exist and have similar mechanisms of pathogenesis. The aim of the present study was to elucidate the safety and efficacy of CaD in the treatment of DKD. Methods: In the prospective randomised controlled study, 100 DKD from Type 2 diabetes mellitus (DM) patients with a urinary albumin/ creatinine ratio (ACR) ≥30 mg/g and urinary protein level between 150 mg/24 h and 2 g/24 h with GFR> 90ml/min were enrolled. The patients were randomly divided into the treatment group (500 mg of CaD, administered orally, 3 times per day) and the control group. DKD patients were treated for 3 months. In the case control study, DM patients without proteinuria and healthy individuals were also enrolled. Clinical data and related biochemical parameters were collected. Endothelial function markers (VEGF, ET-1, eNOS, NO) and inflammatory markers (MCP-1, ICAM, PTX3) were detected by ELISA. Results: In the prospective randomised controlled study, the 24 h urinary albumin and 24 h urinary protein levels significantly decreased after three months of treatment with CaD in the patients with DKD, but the cystatin C-based glomerular filtration rate (GFR) remained unchanged. In addition, the levels of inflammatory markers (PTX3, MCP-1, hsCRP, ICAM) and endothelial dysfunction markers (VEGF, ET-1) were significantly reduced compared to pre-treatment levels, NO was signifcantly increased post treatment. In the case control study, we found that PTX3, MCP-1, ICAM, VEGF and ET-1 levels were positively correlated with urinary albumin in DKD patients, while the NO level was negatively correlated. Logistic regression analysis showed that PTX3, NO and HbAlc were influential factors in DKD. After patients with DKD were treated with CaD for three months, the 24 h urinary albumin and 24 h urinary protein levels significantly decreased, but the cystatin C-based glomerular filtration rate (GFR) remained unchanged. In addition, the levels of inflammatory markers (PTX3, MCP-1, hsCRP, ICAM) and endothelial dysfunction markers (VEGF, NO, ET-1) were significantly reduced compared to pre-treatment levels. Conclusion: CaD can be safely and effectively used to treatdiabetic nephropathy.