1. The role of oxidative stress in the pathogenesis of the diabetic state is being investigated extensively. Although oxidative stress has been reported in terms of glycoxidation, protein oxidation and DNA oxidation in diabetes mellitus, oxidation parameters have not been determined in parallel on the same study population.2. We studied 24 patients with diabetes mellitus (14 patients with Type I diabetes with a mean age of 62.3+/-6.3 years and 10 patients with Type II diabetes aged 67.3+/-5.9 years) and compared them with age-matched non-diabetic controls. Urinary o-tyrosine, 8-hydroxy-2'-deoxyguanosine and pentosidine measurements by HPLC were made on two occasions (t1 and t2).3.A clear statistical difference was found between diabetic patients and controls at t1 or t2 for 8-hydroxy-2'-deoxyguanosine and pentosidine, but not for o-tyrosine. No significant correlations were found between clinical and other laboratory parameters except high-density lipoprotein and uric acid. We revealed significantly increased glycoxidation and DNA oxidation in patients with Type I and Type II diabetes, but protein oxidation was not different from controls.4. The finding of increased glycoxidation reflects increased oxidation of the carbohydrate moiety, whereas the increased levels of oxidized DNA may also be interpreted as due to increased DNA repair. The increased 8-hydroxy-2'-deoxyguanosine does not indicate the generation of an individual active oxygen species, but DNA could have been oxidized simply by alkenals from lipid peroxidation, as e.g. malondialdehyde. As no difference in protein oxidation (i.e. o-tyrosine) between diabetics and controls could be revealed, the oxidation of DNA by hydroxyl radical attack is unlikely, as o-tyrosine was proposed as a marker for hydroxyl radical attack. Therefore, the message is that increased glycoxidation can be confirmed, protein oxidation does not appear to take place and increased DNA oxidation is still not proven, as increased 8-hydroxy-2'-deoxyguanosine may simply reflect repair.