Aims/hypothesis Patients with type 1 diabetes mellitus are more susceptible than healthy individuals to exercise-induced oxidative stress and vascular endothelial dysfunction, which has important implications for the progression of disease. Thus, in the present study, we designed a randomised doubleblind, placebo-controlled trial to test the original hypothesis that oral prophylaxis with vitamin C attenuates rest and exercise-induced free radical-mediated lipid peroxidation in type 1 diabetes mellitus. Methods All data were collected from hospitalised diabetic patients. The electron paramagnetic resonance spectroscopic detection of spin-trapped α-phenyl-tert-butylnitrone (PBN) adducts was combined with the use of supporting markers of lipid peroxidation and non-enzymatic antioxidants to assess exercise-induced oxidative stress in male patients with type 1 diabetes (HbA 1c 7.9±1%, n=12) and healthy controls (HbA 1c 4.6±0.5%, n=14). Following participant randomisation using numbers in a sealed envelope, venous blood samples were obtained at rest, after a maximal exercise challenge and before and 2 h after oral ingestion of 1 g ascorbate or placebo. Participants and lead investigators were blinded to the administration of either placebo or ascorbate treatments. Primary outcome was the difference in changes in free radicals following ascorbate ingestion. Results Six diabetic patients and seven healthy control participants were randomised to each of the placebo and ascorbate groups. Diabetic patients (n=12) exhibited an elevated concentration of PBN adducts (p<0.05 vs healthy, n=14), which were confirmed as secondary, lipid-derived oxygen-centred alkoxyl (RO·) radicals (a nitrogen =1.37 mT and aβ hydrogen =0.18 mT). Lipid hydroperoxides were also selectively elevated and associated with a depression of retinol and lycopene (p<0.05 vs healthy). Vitamin C supplementation increased plasma vitamin C concentration to a similar degree in both groups (p<0.05 vs presupplementation) and attenuated the exercise-induced oxidative stress response (p<0.05 vs healthy).There were no selective treatment differences between groups in the primary outcome variable. Conclusions/interpretation These findings are the first to suggest that oral vitamin C supplementation provides an Diabetologia