Risk assessment and clinical management of patients with coronary artery disease (CAD) has traditionally concentrated on the anatomical features of atherosclerosis. Over the past decade,-however, the importance of the dynamic aspects of vascular physiology, particularly the role of the endothelium, have become clear. In 1980, Furchgott demonstrated the production of a relaxant factor from healthy endothelium (EDRF),'which has subsequently been characterised as nitric oxide or a related molecule. 2 Nitric oxide is not only an important physiological vasodila-tor'counteracting the effect of vasoconstrictors but it also inhibits platelet activation, monocyte-endothelium cell interactions, and smooth muscle cell proliferation and migration. 4 Studies in patients with established CAD have used pharmacological and physiological stimuli to examine coronary vascular tone and flow responses which are dependent on endothelial production of nitric oxide. 5 6 These studies have shown marked abnormalitiesin coronary vasoreactivity in atherosclerotic vessels, which may be the basis for disturbances of coronary supply involved in the genesis of many episodes of transient myocardial ischaemia. 7 This represents a target for therapy as reduction of risk factors such as hypercholesterolaemia, which may result in dysfunctional endothelium, have recently been shown to reduce the activity of transient myocardial ischaemia. 8 Furthermore, dysfunctional vascular endothe-lium may play a role in plaque destabilisation and the risk of acute coronary events. 9 Improvement in endothelial dys-function at the site of atherosclerotic plaques may therefore contribute to the impressive decreasein clinical events seen in large prospective clinical trials of cholesterol lowering in both primary and secondary prevention.'01'While CAD normally presents from middle age onwards, morphological evidence of atherosclerosis has been detected as early as the first two decades oflife.'2 This early atherosclerosis is associated with a similar risk factor profile to that seen in patients with CAD, and epidemiological evidence suggests that the damaging effects of risk factors in early lifeare related to coronary morbidity and mortality decades later." 3 Strategies which aim to retard the progression of atherogenesis from an earlier stage may, therefore, have a substantial impact on the subsequent incidence of clinical disease. This approach is supported by the continu-ing high incidence of cardiovascular events in patients with CAD despite active treatment. Endothelial dysfunction is now recognised as a key early event in atherogenesis.'4 The failure of the endothelium dependent vasodilatory responses of large conduit arteries to physiological stimuli is likely to represent either decreased production or increased inactivation of nitric oxide. Experimental evidence has demonstrated that endothelial derived nitric oxide may be an important endogenous antiatherogenic molecule and that reduction in nitric oxide activity may facilitate the progression of athero-sclerosis.'5 In addition to loss of the protective actions of nitric oxide, endothelial dysfunction may result in a number of abnormalities which may enhance progression of atherosclerosis including: increased release ofvasoconstrictor sub-stances; expression of surface adhesion molecules which facilitate monocyte recruitment and egress into the intima; production of growth factors which promote vascular smooth muscle cell proliferation and migration; and enhanced thrombogenicity mediated by increased platelet activation, plasminogen activator inhibitor-1, and expression of tissue factor. 4

Research on the benefit of early intervention has been limited …