Atherosclerosis is the leading cause of mortality in the developed world. In the United States alone there are approximately 700 000 deaths per year attributable to coronary artery disease, stroke and peripheral vascular diseases (National Center for Health Statistics, 1994). Although national data demonstrate an increase in coronary artery disease prevalence since 1970 (Morano, 1996), there has been a significant decline in mortality from atherosclerosis over the past three decades. Much of this decline may be attributable to risk factor modification. Intriguingly, many of the risk factors for atherosclerosis (hypercholesterolaemia, hypertension, homocysteinaemia, diabetes mellitus and exposure to tobacco) are associated with a reduced elaboration of nitric oxide by the endothelium (endothelium-derived nitric oxide (EDNO)) (Lüscher, Raij & Vanhoutte, 1987; Celermajer et al. 1993; Wu & Meininger, 1995). The result is a perturbation in the normal functions of the endothelium to maintain appropriate vessel calibre and resist the development of atherosclerosis. This endothelial impairment occurs well before any structural changes of atherogenesis are detected. Indeed when otherwise normal vessels are exposed in vitro to oxidized lipoproteins, an impairment in EDNO-dependent vasodilatation can be detected within minutes.It is well documented that risk factor modification has reduced the incidence of morbidity and mortality due to cardiovascular disease (Pasternak, Grundy, Levy & Thompson, 1996). More recently, modification of risk factors has also been shown to restore endothelial function and it may be by this mechanism that morbidity and mortality are curtailed. For example, cigarette smoking, which has been significantly linked to the occurrence of coronary artery disease, is associated with endothelial vasodilator dysfunction (Celermajer et al. 1993). Cessation of smoking appears to improve endothelial vasodilator dysfunction and reverses this increased risk within a few years (Rosenberg, Kaufman, Helmrich & Shapiro, 1985; Ockene, Kuller, Svendsen & Meilahn, 1990). Likewise, lowering cholesterol has been shown to improve endothelial function (Stroes, Koomans, de Bruin & Rabelink, 1995; Treasure et al. 1995) and reduce the incidence of initial and recurrent coronary artery disease events (Sacks et al. 1996). The effect of these therapies on endothelial function may explain their beneficial impact upon clinical events. Indeed, the effect of lipid-lowering therapy on mortality greatly exceeds the rather modest effect upon angiographic measures of lesion progression (Fuster, Gotto, Libby, Loscalzo & McGill, 1996). The success of risk factor modification in reversing endothelial dysfunction and clinical events serves to illustrate the importance of restoring endothelial function by other means. A greater understanding of the involvement of the nitric oxide system in the pathophysiology of vascular disease may provide the basis for new therapeutic strategies.