Endothelial dysfunction in coronary heart disease is more than a systemic process.

Since its initial description by our laboratory, now 27 years ago, much has been learned about the causes and consequences of human endothelial dysfunction. Impairment of endothelial function assessed by nitric oxide bioavailability (referred to as endothelial dysfunction) has been observed in atherosclerosis, is associated with traditional risk factors, and has been described in the setting of various systemic inflammatory disorders linked to high rates of coronary heart disease. Endothelial dysfunction exerts a key role in precipitating myocardial ischaemia during daily activities. Endothelial dysfunction confers adverse cardiovascular prognosis independently of known risk factors and also by becoming a pathophysiological target for both knownand unknowncardiovascular risk factors. Fortunately, endothelial function can be restored with treatments that also improve long-term outcomes. Notably, failure of treatments to re-establish normal endothelial function among individuals with coronary heart disease or in individuals with hypertension identifies a subset of individuals who remain at high risk for adverse clinical events. Thus, endothelial function testing may ultimately be used to identify the non-responders and to individualize treatment with the purpose of preventing future clinical events. The pharmaceutical industry has recognized the importance of endothelial function testing as a tool for development of antiatherosclerotic drugs as endothelial function can change rapidly in response to interventions. Lastly, ‘endothelial biopsies’ which allow for the isolation of human endothelial cells are enabling investigators to characterize the biology of the endothelium on a molecular level. During this procedure, endothelial cells are removed from blood vessels using guidewires and analysed, for example, for protein expression, protein modifications (e.g. serine phosphorylation or tyrosine nitration), and for the function of subcellular organelles, such as mitochondria. Endothelial biopsies have deciphered the molecular signatures of endothelial dysfunction in several human cardiovascular diseases, providing novel biomarkers of disease along with therapeutic targets that can be used for rational development of new treatments. Despite these notable advances, many unanswered questions regarding human endothelial dysfunction remain. For example, although endothelial dysfunction is generally viewed as a ‘systemic’ process, it has long been recognized that endothelial dysfunction does not affect all arteries to the same extent and that, even within a single coronary artery, there are regional differences in endothelial function. Early in the course of human atherosclerosis, endothelial dysfunction is confined to sites of abnormal haemodynamic stresses, particularly the coronary branch points, which are areas of predilection for the development of atherosclerosis. Thus, in concert with systemic risk factors, local anatomy plays a role in the formation of human atherosclerotic lesions, presumably by altering shear stresses and thereby disturbing local coronary endothelial function. At the other end of the spectrum in human atherosclerosis, endothelial dysfunction is more pronounced at sites of culprit coronary lesions of patients presenting with unstable angina compared with non-culprit coronary segments in the same arteries or compared with stenotic segments of patients presenting with stable angina. Similarly, following myocardial infarction, endothelial dysfunction is more impaired in stenotic segments of infarct-related coronary arteries compared with non-infarct-related stenotic segments. While these studies have convincingly demonstrated regional heterogeneity in coronary endothelial function after an acute coronary syndrome (ACS), whether or not the marked endothelial dysfunction at sites of culprit plaques pre-dated the clinical event or was simply a consequence of the event remained unknown. After all, ACS is typically caused by disruption of an atherosclerotic plaque, resulting in the physical denudation and consequently functional loss of the overlying endothelium. Thus, to implicate endothelial dysfunction as a potential cause as opposed to a consequence of ACS, it is important to assess coronary endothelial function in the types of plaques that are