Objective To observe the effect of resveratrol on retinal vasculopathy in diabetic rats. Methods Forty-five Sprague-Dawley male rats were randomly divided into the resveratrol group, treatment control group and the normal control group, 15 rats in each group. Diabetic rat models were induced with streptozotocin injection in resveratrol group and treatment control group. The same volume of sterile saline solution was injected into the rats of the normal control group. The rats of resveratrol group and treatment control group were feed with highfat diet. The rats of resveratrol group received oral gavage of resveratrol (75 mg/kg) twice a day for four months. The same volume of sterile saline solution was given by gavage in rats of treatment control group twice a day for four months. 2 ml femoral vein blood and 50 mu;l aqueous fluid of anterior chamber of the eye from rats of three groups were collected to detect fasting blood glucose, aqueous fluid glucose, cholesterol and triglyceride. The retinal vascular permeability was test by labeling with evans blue. Whole retina was isolated to detect the pericyte number. Total protein was extracted from retina to test the level of vascular endothelial growth factor (VEGF). Results The fasting blood glucose, aqueous fluid glucose, cholesterol and triglyceride in treatment control group were higher than those in normal control group, also higher than those in resveratrol group except cholesterol. The differences among the three groups were statistically significant (F=152.809, 65.230, 3.861, 15.059; P＜0.05). The retinal vascular permeability in treatment control group was higher than that in normal control group, while it in resveratrol group was lower than that in treatment control group. The differences among the three groups was statistically significant (F=11.626,P＜0.05). The pericyte number in treatment control group decreased as compared to normal control group, while it in resveratrol group increased as compared to treatment control group. The differences among the three groups was statistically significant (F=43.284, P＜0.05). The VEGF expression in treatment control group increased as compared to normal control group, while it in resveratrol group decreased as compared to treatment control group. The differences among the three groups was statistically significant (F=14.017, P＜0.05). Conclusion Resveratrol can improve abnormal retinal vasculopathy structure and function, down-regulated level of fasting blood glucose, aqueous fluid glucose, triglyceride and VEGF may be the mechanism.