Exercise is known to increase reactive oxygen species and alter glutathione peroxidase activity (GPx), a selenoenzyme responsible for neutralising hydrogen peroxide. This study evaluated the effect of selenium (Se) status on the response of unfit horses to mild exercise. 25 mature horses received one of four dietary treatments for 29 weeks: low Se (LS, n=6), adequate Se (AS, sodium selenite, n=6), high organic Se (SP; Sel-Plex®, n=7) or high inorganic Se (SS, sodium selenite, n=6). Total dietary Se concentration for LS, AS, SP and SS was 0.06, 0.12, 0.3 and 0.3 mg/kg respectively. Blood samples were collected before and at 0, 4 and 24 h after a 36 min exercise test covering 4.41 km. Blood samples were evaluated for blood GPx, pro-inflammatory cytokine expression, serum malondialdehyde (MDA), creatine kinase (CK), aspartate aminotransferase (AST) and blood pro-inflammatory cytokine expression. Data were analysed as ANOVA with repeated measures. Prior to the exercise test blood Se was higher (P<0.01) for SP and SS than AS and LS. An effect of treatment and treatment×time (P<0.05) existed for GPx. Post-exercise GPx decreased in LS and did not recover by 24 h while remaining unchanged in AS. GPx increased in SP and decreased in SS post-exercise, but both returned to pre-values at 24 h. Serum MDA was elevated at 24 h (P<0.05), but unaffected by treatment. Serum CK and AST were unaffected by Se status (P≯0.05). Pro-inflammatory cytokine expression was unaffected by Se status, although tumour necrosis factor alpha decreased immediately after exercise followed by an increase post-exercise while granzyme B was elevated immediately post-exercise (time, P<0.05). Overall Se status had minimal effects on exercise response in unfit horses. The differences in post-exercise GPx response may warrant further investigation.