Accelerated atherosclerotic vascular disease is the leading cause of mortality in patients with diabetes mellitus. Endothelium-derived nitric oxide (NO) is a potent endogenous nitrovasodilator and plays a major role in modulation of vascular tone. Selective impairment of endothelium-dependent relaxation has been demonstrated in aortas of both nondiabetic animals exposed to elevated concentrations of glucose in vitro and insulin-dependent diabetic animals. The impaired NO release in experimentally induced diabetes may be prevented by a number of antioxidants. It has been hypothesized that oxygen-derived free radicals (OFR) generated during both glucose autoxidation and formation of advanced glycosylation end products may interfere with NO action and attenuate its vasodilatory activity. The oxidative injury may also be increased in diabetes mellitus because of a weakened defense due to reduced endogenous antioxidants (vitamin E, reduced glutathione [GSH]). A defective endothelium-dependent vascular relaxation has been found in animal models of hypertension and in hypertensive patients. An imbalance due to reduced production of NO or increased production of free radicals, mainly superoxide anion, may facilitate the development of an arterial functional spasm. Treatment with different antioxidants increases blood flow in the forearm and decreases blood pressure and viscosity in normal humans; vitamin E inhibits nonenzymatic glycosylation, oxidative stress, and red blood cell microviscosity in diabetic patients. Long-term randomized clinical trials of adequate size in secondary and primary prevention could support the free-radical hypothesis for diabetic diabetic vascular complications and the use of antioxidants to reduce the risk of coronary heart disease.