There is overwhelming evidence in favour of a significant role of reactive oxygen metabolites (ROM) in the pathophysiology of inflammatory bowel disease (IBD) in man and in experimental animal models. This study was undertaken to investigate the possible protective effects of pretreatment with trimetazidine (TMZ) on the oxidant-anti-oxidant balance in ethanol- and acetic acid-induced colonic damage in rats. TMZ was chosen because of its various cytoprotective features (preserving cellular ATP levels, limiting intracellular acidosis and limiting inorganic phosphate, Na(+) and Ca(2+) accumulation) and anti-oxy characteristics which were previously reported. A total of 80 rats were randomized into eight major groups each consisting of 10 animals. Animals in groups 1, 2 and 3 served as models of ethanol-induced colitis (0.25 ml of 30% (v/v) ethanol), while group 4 served as their control. Animals in groups 5, 6 and 7 served as models of acetic acid-induced colitis (1 ml of 4% (v/v) acetic acid), while group 8 served as their control. TMZ was administered 5 mg/kg by intrarectal (i.r.) and intraperitoneal (i.p.) routes to groups 1, 2, 5 and 6. Intraperitoneal administration of TMZ was used in order to evaluate its systemic effect while i.r. administration was used to determine its local effect. After decapitation, colon mucosa samples were obtained and evaluated macroscopically and microscopically. Myeloperoxidase (MPO) activities as markers for inflammation, malondialdehyde (MDA) levels as markers for oxidant stress and reduced glutathione (GSH) and oxidized glutathione (GSSG) levels as markers for anti-oxidant status were determined. Acute colitis was observed in macroscopic and microscopic evaluation in ethanol- and acetic acid-administered groups compared with controls (P = 0.000). The macroscopic and microscopic scores in colitis groups were correlated with MPO activities (r = 0.5365, P = 0.000 and r = 0.5499, P = 0.000, respectively). MDA and GSSG levels in the acetic acid-induced colitis group were higher compared with ethanol-induced colitis group (P < 0.008 and P < 0.005, respectively), while GSH levels were significantly lower (P < 0.05). While TMZ pretreatment did not improve the oxidant state, it preserved the GSH levels significantly (P < 0.05). In conclusion, ethanol- and acetic acid-induced colitis models are appropriate experimental colitis models which in many ways manifest the characteristics seen in tissue injury related to colitis in humans. Of these two, the acetic acid-induced colitis model proved more suitable than the ethanol model for investigating the alterations in long-term and in more severe tissue injury. While TMZ pretreatment via i.p. or i.r. route did not improve the oxidative-inflammative state in either of these models, it did contribute significantly to the preservation of the anti-oxidant pool via the conservation of intracellular GSH levels. This conserving effect of TMZ was substantially more pronounced in the i.p. route compared with the i.r. route. Based on our results, we conclude that the 'GSH-preservation' role of TMZ can be the mode of action it manifests as an anti-oxy compound.