The aim of this study was to investigate the effect of thiamine pyrophosphate (TPP), administered via sugar water, on retinal neovascularisation in rats. Animals were assigned to three groups, namely the TPP sugar-water group (TPSWG, n = 12), the control group (CG, n = 12) and the healthy group (HG, n = 12). The TPSWG was injected intraperitoneally with TPP once a day for 6 months. CG and HG rats were given distilled water in the same way. TPSWG and CG rats were left free to access an additional 0.292 mmol/ml of sugar water for 6 months. The fasting blood glucose (FBG) levels of the animals were measured monthly. After 6 months, biochemical, gene expression and histopathologic analyses were carried out in the retinal tissues removed from the animals after they were killed. The measured FBG levels were 6.96 +/- 0.09 mmol/ml (p < 0.0001 vs. HG), 6.95 +/- 0.06 mmol/ml (p < 0.0001 vs. HG) and 3.94 +/- 0.10 mmol/ml in the CG, TPSWG and HG groups, respectively. The malondialdehyde (MDA) levels were found to be 2.82 +/- 0.23 (p < 0.0001 vs. HG), 1.40 +/- 0.32 (p < 0.0001 vs. HG) and 1.66 +/- 0.17 in the CG, TPSWG and HG, respectively. Interleukin 1 beta (IL-1 beta) gene expression was increased (3.78 +/- 0.29, p < 0.0001) and total glutathione (tGSH) was decreased (1.32 +/- 0.25, p < 0.0001) in the retinal tissue of CG compared with TPSWG (1.92 +/- 0.29 and 3.18 +/- 0.46, respectively). Increased vascularisation and oedema were observed in the retinal tissue of CG, while the retinal tissues of TPSWG and HG rats had a normal histopathological appearance. A carbohydrate-rich diet may lead to pathological changes in the retina even in nondiabetics, but this may be overcome by TPP administration.