The aims of the present study were to investigate the effects of fenofibrate and bezafibrate on the risk of development of diabetic retinopathy (DR) in patients with type 2 diabetes and dyslipidemia. Japanese working age patients with type 2 diabetes and dyslipidemia were extracted from the Nihon University School of Medicine Clinical Data Warehouse. These patients were divided into three groups: control group (n=2549), fenofibrate group (n=40), and bezafibrate group (n=135). Multivariate logistic regression analysis was performed to assess the association between fibrates and the development of DR. After adjustment for covariates, fenofibrate showed no association with the risk of DR [adjusted odds ratio (OR), 0.160; 95% CI, 0.021-1.209; p=0.0758]. Bezafibrate also showed no association with the risk of DR (adjusted OR, 0.731; 95% CI, 0.411-1.299; p=0.2855). However, poor control of hemoglobin A1c (HbA1c ≥8.0%; adjusted OR, 3.623; 95% CI, 2.649-4.956; p<0.0001) and high low-density lipoprotein cholesterol (LDL-C ≥140 mg/dL; adjusted OR, 1.399; 95% CI, 1.013-1.932; p=0.0415) within the follow-up period of type 2 diabetes and dyslipidemia increased the risk of DR. Our results suggested that to prevent development of DR in patients with type 2 diabetes and dyslipidemia, controlling LDL-C levels as well as HbA1c levels under coexistence type 2 diabetes and dyslipidemia is more important than the selection of fibrate.