Thomas F. Luscher, M.D.
N Engl J Med 1994; 330:1081-1083April 14, 1994DOI: 10.1056/NEJM199404143301511
Cardiovascular disease accounts for considerable mortality and morbidity in Western countries. Most of the common forms of cardiovascular disease, such as atherosclerosis, are caused by functional and structural changes in the blood-vessel wall. These changes include abnormal vasoconstriction, enhanced interaction of blood cells with the vessel wall, activation of coagulation mechanisms, and migration and proliferation of vascular smooth-muscle cells1-3. Depending on the stage and location of the disease, one or more of these factors predominate. These vascular abnormalities have an important role in the pathogenesis of angina pectoris, myocardial infarction, stroke, and vascular forms of renal failure.
A crucial vascular structure is the endothelium,1 not only because it is strategically located between the circulating blood and vascular smooth muscle, but also because it is a source of a variety of mediators regulating vascular tone1,4 and growth3 as well as platelet function and coagulation1,5. One important endothelium-derived mediator is nitric oxide, which is formed from L-arginine6 by the action of a constitutive form of the enzyme nitric oxide synthase7. The fact that nitric oxide is released both into the lumen (to inactivate platelets) and away from the lumen (to relax vascular smooth muscle) suggests that it protects against thrombosis and vasoconstriction1. Indeed, the L-arginine-nitric oxide pathway is stimulated not only by mechanical forces and neurohumoral mediators, but also by platelet-derived products and coagulation factors5,8. The latter effects may represent a negative feedback mechanism that prevents vasoconstriction and thrombus formation at sites of normal endothelium. The capacity of nitric oxide to inhibit the migration and proliferation of vascular smooth-muscle cells is a further protective property. For example, it is known that endothelial denudation in vivo invariably leads to vascular proliferation. Nitric oxide is also involved in blood-pressure regulation. Indeed, intravenous or oral administration of nitric oxide inhibitors induces sustained hypertension9.
The endothelium is both a target for and a mediator of cardiovascular disease. Changes in endothelial function occur early in the course of vascular disease. Hence, the development of diagnostic tests able to detect early vascular changes could make possible the identification of patients at risk for the progression of vascular disease and provide a means of monitoring therapy. Tests to assess the function of the L-arginine pathway are still invasive and involve the infusion of nitric oxide inhibitors (such as NG-monomethyl-L-arginine10) or acetylcholine (which stimulates the release of nitric oxide)11. Although vasoconstriction in response to NG-monomethyl-L-arginine indicates the presence of basal nitric oxide formation, an infusion of acetylcholine allows the assessment of receptor-mediated release of nitric oxide. Vasodilatation in response to acetylcholine in human arteries is endothelium-dependent8 and is inhibited by NG-monomethyl-L-arginine, but not by inhibitors of other mediators11. Flow-dependent vasodilatation of large arteries is another nitric oxide-dependent response.
Tests of endothelial function may detect early vascular abnormalities in patients with cardiovascular risk factors but without clinically evident disease. For example, in patients with hypercholesterolemia and diabetes, endothelium-dependent vasodilatation in response to acetylcholine is reduced in both the peripheral and coronary circulation long before structural vascular changes or clinical symptoms occur12-15. Studies of the coronary circulation suggest that during the natural history of atherosclerosis, impairment of receptor-mediated activation of the endothelium is the first event, followed by flow-induced vasodilatation and later by a reduced response of smooth muscle to nitric oxide12.
Hyperterier nsion is an important cardiovascular risk factor associated with myocardial infarction, heart failure, and stroke. Morphologic and functional alterations of endothelial cells occur in experimental hypertension1. In patients with hypertension, basal formation of nitric oxide and the effect of an acetylcholine infusion on the release of nitric oxide were found to be reduced,11,16-18whereas flow-induced vasodilatation was maintained19. Such changes in endothelial function could be a cause or a consequence of hypertension. Most experimental data suggest that endothelial dysfunction develops as blood pressure increases and that the degree of dysfunction is related to the blood-pressure level. Hence, dysfunction of the L-arginine pathway is likely to be a secondary event involved in the maintenance rather than the initiation of hypertension. Most important, however, endothelial dysfunction may contribute to the vascular complications of hypertension, such as myocardial infarction and stroke. If so, endothelial dysfunction may be an indicator of the risk of such complications, particularly since they do not actually develop in all patients with hypertension or other risk factors.
In this issue of the Journal, Cockcroft et al.20 report that a large number of patients with essential hypertension had normal endothelium-dependent vasodilatation. They used acetylcholine and carbachol to investigate the L-arginine-nitric oxide pathway in the forearm circulation. Using both muscarinic agonists and the direct vasodilator sodium nitroprusside, they obtained similar responses in normotensive and hypertensive subjects regardless of whether the subjects had been treated with antihypertensive drugs. In view of previous studies that did find abnormal responses to acetylcholine in patients with hypertension, the authors conclude that selective impairment of the vasodilator effects of acetylcholine in the forearm circulation is not a universal finding in hypertension.
How can these different findings be reconciled? One possibility, which would be in line with experimental data, is that abnormal endothelium-dependent vasodilatation is a consequence of hypertension and that the occurrence of such changes does not affect all hypertensive patients to the same degree. Furthermore, early defects may involve the basal formation of nitric oxide rather than acetylcholine-stimulated release of nitric oxide. Factors that appear to be important for the development of endothelial dysfunction are the duration and severity of hypertension. It is possible that certain groups of hypertensive patients are more likely to have endothelial dysfunction than others because of different pathogenetic mechanisms or dietary and genetic factors. Also, other variables, such as insulin resistance, may be involved.
It is obvious that the relation between the endothelium and cardiovascular disease is complex. In the early stages of vascular disease, alterations in endothelial function are subtle and selective and can affect both basal formation or receptor-mediated release of endothelial mediators. In contrast, the release of mediators in response to vascular shear stresses is affected only later in the course of disease. Not all endothelial mediators are similarly affected unless severe atherosclerotic changes have occurred. In addition, the coronary and cerebral arteries are probably more prone to endothelial dysfunction than other vascular beds. Endothelial dysfunction may occur more readily in patients with hyperlipidemia or diabetes than in those with hypertension. Finally, even in patients with a particular risk factor, the degree of endothelial dysfunction may differ, as does the incidence of cardiovascular complications at later stages. Hence, we need further studies using sophisticated, noninvasive tests of endothelial function to define more clearly the occurrence, natural history, anatomical distribution, and type of endothelial dysfunction in different populations and individual patients. This may allow more precise risk stratification and eventually more effective therapy.