Although good glycaemic control can delay the development and progression of diabetic retinopathy, new therapies are needed to obtain a better control of this diabetic complication. Oxidative stress seems to be a contributing factor in diabetic retinal alterations, therefore, it has been suggested that antioxidants may be beneficial in reducing diabetic retinal changes. However, many questions are still open. In fact, it remains to be ascertained which antioxidants are the most active when they are chronically administered in vivo and their effective dosages. Therefore, we compared the effect of chronic taurine supplementations versus a mixture of vitamin E + selenium on biochemical retinal changes induced by diabetes at different stages of the disease. Briefly, streptozotocin (STZ) diabetic rats were administered for 4 months following the dietary supplements: (a) 2% (w/w) taurine; (b) 5% (w/w) taurine; (c) 200 IU vitamin E + 8 mg selenium/kg diet (d) 500 IU vitamin E + 8 mg selenium/kg diet. In STZ diabetic rat in poor metabolic control (i.e. serum glucose >16.5 mmol/l), at 2, 4, 8, 16 weeks following the onset of diabetes, retinal conjugated dienes (CD) and lipid hydroperoxides (LP) were significantly and progressively increased, while sodium pump activity was gradually and significantly reduced. In taurine and vitamin E + selenium supplemented diabetic rats, glycaemia and body weight were not significantly different from those of non-supplemented diabetic animals. In diabetic rats, 2 and 5% taurine significantly decreased CD. This reduction is long lasting. Regarding CD, both vitamin E + selenium supplementations reduced CD only during the first 4 weeks of diabetes. Two percent taurine supplementation significantly lowered LP for the first 8 weeks of the disease while 5% taurine-induced-reduction lasted for the whole experimental time. A 200 IU vitamin E + 8 mg selenium supplementation did not significantly modify LP, while 500 IU vitamin E + 8 mg selenium significantly lowered them for the whole studied period. Finally, taurine preserved ATPase activity being more effective at 5% than 2%. Two hundred IU vitamin E + 8 mg selenium did not generally modify pump activity, while 500 IU vitamin E + 8 mg selenium partially prevented the decrease in pump activity. We conclude that taurine and vitamin E + selenium supplementations ameliorate biochemical retinal abnormalities caused by diabetes. These effects are dose- and time-dependent Moreover, the effect of taurine on CD is longer lasting than that of vitamin E + selenium. In addition, taurine seems to better preserve ATPase activity in comparison with vitamin E + selenium. Finally, in diabetic animals a negative correlation is found between CD and LP on one side and Na+K+ATPase activity on the other; thus, lipid peroxidation and pump activity seem to be associated. The same inverse correlations are present in vitamin E + selenium supplemented diabetic rats, but are lost in taurine supplemented animals. Therefore, taurine effects may not be simply mediated by its antioxidant activity. Thus, chronical (4 months) taurine and vitamin E + selenium supplementations reduce biochemical retinal alterations in diabetic rat in poor metabolic control.