Endothelial Dysfunction and Coronary Risk Reduction

Coronary artery disease (CAD) is the leading cause of death in the United States.1 In this century, tremendous progress has been made to elucidate the roles of chronic risk factors such as hypercholesterolemia, hypertension, diabetes mellitus, and cigarette smoking in the development of atherosclerotic disease.2 In addition, studies in recent years have clearly shown that risk factor reduction, particularly lipid-lowering therapy, reduces the incidence of cardiovascular events and overall mortality.3-5 Because hypercholesterolemia and other risk factors contribute to the development of atherosclerosis, it is frequently assumed that regression of atherosclerotic lesions explains the beneficial effects of risk factor reduction. However, there is growing appreciation that lesion regression is an inadequate explanation for the observed benefits.6 Atherosclerotic lesions may be present and clinically silent for decades before a clinical event such as an acute myocardial infarction.7 In addition, there is evidence that lipid-lowering therapy may reduce cardiovascular risk without necessarily altering the presence or severity of coronary artery lesions.6 We now understand that many acute coronary events are the result of a fundamental alteration in vascular homeostasis that leads to plaque rupture, intramural hemorrhage, vasospasm, and formation of an occlusive thrombus. In this review, we will focus on the role played by the vascular endothelium in the development of acute and chronic coronary syndromes and the impact of risk reduction on endothelial function. Normal Endothelial Function Although previously viewed as a passive conduit for the flow of blood, we now understand that the vascular system actively regulates blood flow and fluidity. Located at the interface between blood and the vessel wall, the vascular endothelium plays a particularly important role in the control of vasomotor tone, thrombosis, and intimal growth by releasing a number of important regulatory factors. Nitric oxide (NO) is synthesized in endothelial cells from the amino acid L-arginine through the action of endothelial nitric oxide synthase (eNOS).8 Endothelial nitric oxide synthase activity depends on the cytosolic calcium concentration. A number of agonists including acetylcholine, catecholamines, bradykinin, and products of aggregating platelets such as adenosine diphosphate and serotonin may act on endothelial membrane receptors to increase cytosolic calcium concentration activating eNOS and stimulating NO production.8 Increased blood flow in a vessel may also activate eNOS by increasing shear stress at the endothelial surface.9 Nitric oxide is a potent vasodilator and activates soluble guanylyl cyclase, increasing intracellular cyclic 3′,5′-guanosine monophosphate in vascular smooth-muscle cells.10 It is known that agonists for NO synthesis produce vasodilation of normal human arteries,11 and it is now firmly established that endothelium-derived NO is an important regulator of vasomotor tone in humans because eNOS inhibition blocks agonist-induced and flow-mediated dilation.12,13 Endothelium-derived NO also inhibits platelet adhesion and activation through a guanylyl cyclase-dependent mechanism.14 In addition to these vasodilator and antiplatelet effects, NO also has anti-inflammatory properties and inhibits adhesion of leukocytes to the endothelial surface.15 The endothelium further regulates vasomotor tone through the release of other vasodilators such as prostacyclin16 and endothelium-derived hyperpolarizing factor.17 Conversely, under certain conditions, the endothelium may produce vasoconstrictors such as endothelin,18 angiotensin II,19 and thromboxane A2.20 Thus, physiologic regulation of vasomotor tone is determined by a balance between endothelium-derived vasodilators and vasoconstrictors. In addition to regulating vascular tone, the endothelium also acts to control fibrinolysis and thrombosis. Tissue plasminogen activator (t-PA) and urokinase-type plasminogen activator are produced by endothelial cells and induce clot lysis by activating plasmin and promoting fibrin degradation.21 The primary inhibitor of t-PA in plasma, plasminogen activator inhibitor-1 (PAI-1), is also primarily produced by endothelial cells.22 The relative balance between t-PA and PAI-1 in plasma plays a critical role in determining the response to thrombus formation, and under normal conditions this balance favors fibrinolysis. The endothelium has a number of other antithrombotic functions including production of thrombomodulin, a cell surface protein that inhibits thrombin-induced platelet aggregation and promotes activation of the anticoagulant factor protein,23 and production of the heparin-like molecules heparan and dermatan sulfate.24 Recently, it was discovered that endothelial cells express a membrane associated ecto-ADPase that limits platelet aggregation by inactivating platelet-derived ADP, which otherwise would promote further platelet aggregation.25 Finally, under pathologic conditions, the endothelium may also produce prothrombotic factors including tissue factor, which activates the extrinsic coagulation pathway.26 The endothelium is also involved in the inflammatory response. Under normal conditions, the endothelium limits the entry of inflammatory cells into tissues.15 However, exposure of endothelial cells to cytokines such as tumor necrosis factor may "activate" endothelial cells, leading expression of endothelial-leukocyte adhesion molecules (intercellular adhesion molecule-1, vascular cell adhesion molecule-1, and E-selectin) and leukocyte chemoattractants (interleukin-8 and monocyte chemotactic protein-1).27 These products regulate leukocyte adhesion to the endothelial surface and their retention in the subendothelial space. Endothelial activation is also associated with other changes in endothelial phenotype leading to increased permeability,28 and increased production of PAI-1, endothelin, and tissue factor. Although these responses may be an adaptive part of the normal immune response, inappropriate vascular inflammation also may contribute to the atherosclerotic process and the pathophysiology of acute coronary events.29 Finally, the presence of an intact, normally functioning endothelium acts to inhibit the growth of underlying vascular smooth muscle. Endothelial denudation is well recognized as a stimulus for intimal growth.30 Endothelial products that inhibit the growth of intimal cells include NO and heparan sulfate.31,32 Further, by inhibiting platelet adhesion and activation, the normal endothelium prevents release of growth promoting factors from activated platelets, including platelet-derived growth factor.33 In summary, the healthy endothelium maintains normal vasomotor tone, promotes clot lysis, and inhibits vasoconstriction, platelet aggregation, inflammation, and intimal growth. As will be discussed in the following section, coronary risk factors and fully developed atherosclerosis are associated with impairment of these essential functions of the endothelium, and there is strong evidence that endothelial dysfunction contributes to the clinical expression of atherosclerosis. Coronary Risk Factors and Endothelial Dysfunction In light of the multiple regulatory functions of the endothelium, the term "endothelial dysfunction" is vague and merits definition. A large number of studies in humans have shown loss of NO-dependent vasodilation in subjects with coronary risk factors, and there is a tendency to equate endothelial dysfunction with impaired vasomotor tone. However, we use the term endothelial dysfunction to denote a more generalized abnormality of endothelial control of vasomotor tone, thrombosis, platelet activity, inflammation, and intimal growth that are likely to be relevant to ischemic syndromes in atherosclerosis.6 It is clear that hypercholesterolemia is associated with loss of NO-dependent vasodilation in patients. In the coronary and peripheral circulations, the normal vasodilator responses to acetylcholine34,35 or increased flow36 are impaired in subjects with hypercholesterolemia. In fact, such impairment can be shown in children with familial hypercholesterolemia (age 7 to 17 years).36 Subjects with angiographically proven coronary atherosclerosis show similar impairment in NO-dependent vasodilation.11 Experimental studies suggest that impaired NO-dependent vasodilation in hypercholesterolemia is a consequence of increased oxidative stress in the vasculature.37 Specifically, increased production of superoxide anion may lead to inactivation of NO and formation of oxidized low-density lipoprotein may interfere with receptor-dependent NO release from endothelial cells and have a number of other proinflammatory effects that lead to loss of NO action.37 Support for the importance of these mechanisms in humans is provided by recent studies that show improved NO-dependent vasodilation after treatment with antioxidants.38,39 Subjects with hypercholesterolemia and early or advanced atherosclerosis also show increased circulating levels of the endothelin,40 decreased fibrinolytic activity,41 and increased PAI-1 activity.42 In experimental models, oxidized low-density lipoprotein and related compounds may stimulate intercellular adhesion molecule-1, tissue factor, and PAI-1 expression and decrease t-PA production.43 Thus, hypercholesterolemia and atherosclerosis are associated with loss of the vasodilator and antiplatelet effects of NO as well as a number of the other regulatory functions of the endothelium. Other coronary risk factors also are associated with endothelial dysfunction. For example, NO-dependent vasodilation is impaired in the coronary and brachial circulations of subjects with diabetes mellitus,44 hypertension,45,46 and history of cigarette smoking.47,48 Passive exposure to cigarette smoke is also associated with loss of endothelium-dependent vasodilation.49 Compared with premenopausal women, postmenopausal women show impaired endothelial vasomotor function.50 Although less well-characterized, there is growing evidence that other endothelial functions are impaired in these disease states or appropriate experimental models. For example, diabetes mellitus is associated with increased production of vascular cell adhesion molecule-151 and PAI-1,52 whereas PAI-1 activity also is increased in subjects with hypertension.53 Thus, hypercholesterolemia, diabetes mellitus, hypertension, cigarette smoking, post-menopausal status, and established atherosclerosis are all associated with the loss of the normal vasodilator, antithrombotic, and anti-inflammatory properties of the endothelium. Endothelial Dysfunction and Clinical Ischemia How does endothelial dysfunction relate to clinical ischemia in patients with CAD? Although a direct link between endothelial dysfunction and increased risk for cardiovascular events remains to be established, there is growing evidence that loss of the normal functions of the endothelium may contribute to the pathophysiology of clinical ischemia in atherosclerosis. In the past decade, studies have emphasized that the anatomic severity of an atherosclerotic lesion bears little relationship to the risk for development of acute coronary syndromes such as myocardial infarction and unstable angina. Instead, we now understand that these events occur after plaque rupture, mural hemorrhage, vasospasm, and formation of an obstructive thrombus.54 Studies indicate that relatively mild lesions, rather than severely stenosed lesions, are most prone to rupture.55 In regard to stable coronary syndromes, lesion severity also fails to predict the level of effort required to provoke angina.56 Such observations have emphasized the importance of vascular function in the pathophysiology of ischemic coronary syndromes. In patients with stable angina and obstructive lesions, impaired endothelial vasomotor function may aggravate the supply-demand mismatch that occurs during exercise and other stimuli for angina. Exercise is associated with increased coronary blood flow, activation of the sympathetic nervous system leading to increased blood pressure, and increased circulating catecholamines. In patients with atherosclerosis, loss of endothelium-derived NO leads to increased sensitivity to the constrictor effects of catecholamines57 and loss of normal flow-mediated dilation.58 These effects tend to worsen the severity of stenoses and further limit coronary blood flow in the face of increased myocardial oxygen demand. Thus, an intervention that improves endothelial NO action has the potential to lessen ischemia. There also is evidence to support the idea that endothelial dysfunction contributes to acute coronary syndromes. Patients with recent myocardial infarction or unstable angina show more severe impairment of endothelial vasodilator function compared with stable control subjects.59,60 Loss of flow-mediated NO release and increased catecholamine-induced coronary vasoconstriction may lead to increased shear stress at the site of an atherosclerotic lesion making it more vulnerable to rupture.61,62 Loss of NO and expression of adhesion molecules may promote leukocyte accumulation in plaque where expression of metalloproteinases may lead to collagen degradation and further increase the risk for rupture.29 After plaque rupture, platelet activation and the clotting cascade may generate increased local concentrations of serotonin and thrombin. These compounds are potent vasoconstrictors, but also stimulate NO production in healthy endothelial cells. In atherosclerosis, loss of the counter-regulatory effects of NO may lead to a more severe constrictor response to these products.63 Loss of the antiplatelet effects of NO may further worsen the thrombotic response after plaque rupture. In atherosclerosis, increased expression of endothelin, PAI-1, and tissue factor would promote vasospasm and thrombosis in this setting. Thus, endothelial dysfunction may increase both the risk for plaque rupture and worsen the severity of ischemia after this event. Thus, improvement of endothelial function has the potential to reduce cardiovascular risk.6 Interventions to Improve Endothelial Dysfunction Given the association between coronary risk factors and endothelial dysfunction, it is reasonable to hypothesize that risk factor reduction will improve endothelial function. Several studies have examined this possibility in patients, and the results of these studies provide insight into the mechanisms of benefit of therapies known to reduce cardiovascular events. Further, studies of drug therapies to designed to improve endothelial function may suggest new approaches to the management of patients with CAD. Recently, three large, randomized, placebo-controlled trials showed that lipid-lowering therapy with an 3-hydroxyl-3-methylglutaryl-coenzyme A reductase inhibitor reduces the risk for cardiovascular events3,5 and reduces mortality3 in subjects with CAD, and reduces cardiovascular events in hyper-cholesterolemic men without known CAD.4 Pooling the results of the large number of earlier studies that used less potent lipid-lowering agents provides similar results.64 Despite this strong evidence for a beneficial effect of lipid-lowering therapy, the mechanism of this effect remains uncertain. The idea that reducing plasma cholesterol induces lesion regression or dramatically slows lesion progression has been difficult to confirm in nearly a dozen angiographic studies.6 These findings are consistent with the idea that plaque activation may play a more important role than lesion severity in the pathogenesis of acute coronary events. To date, five human studies have examined the effect of lipid-lowering therapy on endothelial vasomotor function in the coronary circulation.38,65-68 These studies suggest that more prolonged treatment and treatment of subjects with higher baseline total cholesterol levels is associated with the greatest improvement in endothelial function. However, studies in the brachial circulation have emphasized that the improvement in endothelial vasomotor function can occur over relatively short periods of treatment (1 to 3 months),69,70 a period that is likely too short for any meaningful regression of atherosclerosis. In fact, two recent studies indicate that rapid, drastic reduction of LDL cholesterol by plasma apheresis can produce an improvement in endothelial vasomotor function almost immediately.71,72 Lipid-lowering therapy also seems to improve some other functions of the endothelium. For example, pravastatin has been reported to reduce circulating levels of PAI-173 and E-selectin, but not vascular cell adhesion molecule-1 or intercellular adhesion molecule-1.74 Two recent studies suggest that lipid-lowering therapy may have a relatively rapid effect on clinical ischemia in patients with a coronary disease. Three months of intensive dietary lipid-lowering therapy was associated with a reduction in dobutamine-induced ischemia.75 Similarly, 3 months of lovastatin treatment markedly reduced episodes of ischemia as assessed by ambulatory electrocardiogram monitoring.76 The finding that endothelial function and clinical ischemia may improve over a similar time period supports the contention that improved endothelial function is a mechanism of benefit for lipid-lowering therapy in patients with CAD. It should be acknowledged that lipid-lowering therapy likely has other beneficial effects. For example, investigators have proposed that a reduction in lesion lipid content and increased lesion fibrosis in the setting of lipid-lowering therapy may reduce the vulnerability of a plaque to rupture, and may explain, in part, the beneficial effects of lipid-lowering therapy.6 Further, some individuals do show impressive changes in lesion severity with treatment and there is evidence to suggest that preventing lesion progression has a beneficial effect on prognosis.77 However, it is likely that in addition, improvement of the vasodilator, antithrombotic, and anti-inflammatory functions of the endothelium play an important role in this regard. The effects of several other risk-reducing interventions on endothelial function have been examined in patients with CAD, although in general, the results remain less conclusive than the results following lipid-lowering therapy. For example, treatment of postmenopausal women with hormone replacement therapy is associated with improved brachial artery flow-mediated dilation.78 Although estrogens also have favorable effects on plasma lipids and other potential contributing factors, investigators have proposed that the reduction in cardiovascular risk associated with postmenopausal estrogen use79 may result, in part, from a direct effect to improve endothelial vasomotor function. A similar story is emerging to explain the beneficial effects of exercise. It is known that regular exercise reduces cardiovascular risk,80,81 and controversy exists regarding the mechanism of benefit. Investigators have proposed that weight loss, decreased body fat, increased high-density lipoprotein cholesterol, reduced glucose-intolerance, and favorable alterations in the sympathetic nervous system and renin-aldosterone system explain this benefit.82 There is now evidence that exercise also improves endothelial function. In animal studies, chronic and repetitive increases in coronary blood flow by treadmill exercise improves endothelium-dependent vasodilation83 and increases eNOS gene expression.84 In patients, 4 weeks of handgrip exercise in subjects with heart failure improved radial artery flow-mediated dilation.85 A preliminary study involving young army recruits showed improved NO-dependent vasodilation of the brachial artery after 10 weeks of intensive training.86 To date, however, no study has examined the effect of exercise training on coronary artery endothelial function in humans. In addition to NO action, exercise training also improves other endothelial functions in patients. For example, exercise training is associated with increased t-PA activity and decreased PAI-1 activity in healthy adults.87 In addition, chronic exercise seems to promote vessel growth and improve non-endothelium-dependent vasodilator capacity.82,88 Thus, it seems likely that exercise training has a number of important effects that contribute to improved outcome, including improved endothelial function. Although improvement of hypercholesterolemia and sedentary lifestyle improve endothelial function, it remains unknown whether other forms of risk reduction will have a similar benefit. Although the angiotensin-converting enzyme inhibitor quinapril improved endothelium-dependent coronary vasodilation in patients with CAD,89 in general, antihypertensive therapy is unsuccessful in having a consistent beneficial effect in hypertensive patients.90,91 As a group, former smokers have better flow-mediated dilation than current smokers;48 however, there is no prospective data suggesting that smoking cessation improves endothelial function. Similarly, there is no evidence to date that rigorous control of diabetes mellitus has such a beneficial effect. Finally, there are emerging therapies that may reduce cardiovascular risk through an effect on endothelial function (Table 1). For example a recent randomized, placebo-controlled trial showed that vitamin E, a lipid-soluble antioxidant, dramatically reduces non-fatal myocardial infarction in subjects with coronary artery disease.92 Treatment with the lipid-soluble antioxidant probucol is associated with improved coronary endothelial function.38 The water-soluble antioxidant vitamin C also improves brachial artery endothelial vasomotor function in patients with CAD.39 In addition, recent studies suggest that treatment with L-arginine, the precursor for NO synthesis, also improves endothelium-dependent vasodilation.93Table 1: Interventions Known to Improve Endothelial Vasomotor FunctionConclusion We have reviewed the normal functions of the endothelium and the impact of coronary risk factors and risk factor reduction on endothelial function. It is apparent that loss of endothelial control of vasomotion, thrombosis, and vessel inflammation contributes to the pathophysiology of acute and chronic coronary syndromes; and improvement of endothelial function has the potential to reduce cardiovascular risk. Lipid-lowering therapy has been proven to reduce cardiovascular risk, and it is highly likely that improved endothelial function contributes to this beneficial effect. Regarding the beneficial effects of exercise, emerging studies strongly support a similar mechanism. Development of other strategies to improve endothelial function are likely to provide additional tools that will benefit patients with CAD.