The suggestion that insulin is associated with atherosclerosis is based on clinical, epidemiologic, and experimental evidence. In general, atherosclerosis of the coronary, cerebral, and peripheral arteries is associated with abnormally high insulin responses to oral glucose. This hyperinsulinemia is not related to acute injury or to tissue necrosis, does not occur in response to intravenous glucose or tolbutamide, and is independent of other cardiovascular risk factors. Populations who are at high risk for cardiovascular disease have higher insulin responses to oral glucose than those at lower risk. Prospective studies carried out in Australia, Finland, and France have shown that elevated insulin levels, either fasting or in response to oral glucose, have a predictive role in the development of cardiovascular disease. This association is independent of the effects of other cardiovascular risk factors. In experimental animals, insulin deficiency retards the development of diet-induced arterial disease, whereas administration of insulin promotes lesion development and prevents lesion regression. Insulin stimulates lipid synthesis in isolated arteries and stimulates proliferation and lipid accumulation in cultured arterial smooth muscle cells. The evidence linking insulin with atherosclerosis has been gathered from nondiabetic subjects; this evidence is unavailable in diabetics. As it is clear that hyperinsulinemia is often present in diabetes, either in relation to mild glucose intolerance or obesity (in noninsulin-dependent diabetes mellitus) or because of insulin therapy (in insulin-dependent diabetes mellitus), it is essential that further consideration should be given to the possibility that hyperinsulinemia may have harmful effects on the arterial wall.