sign in sign up
<<
Home , Endothelium

CORE Metadata, citation and similar papers at core.ac.uk

Provided by Elsevier - Publisher Connector JACC Vol. 30, No. 2 325 August 1997:325–33 REVIEW ARTICLES

Endothelial Dysfunction: Does It Matter? Is It Reversible?

DAVID S. CELERMAJER, MBBS, PHD, FRACP Sydney, Australia

Until recently, the endothelium was regarded as a relatively The recent demonstration that may be inert cell layer. However, over the past 20 years, research has reversible raises the possibility of slowing the progression of revealed an extraordinary array of endothelial functions, includ- or modifying arterial function, or both, to decrease ing control over , fibrinolysis, arterial tone and vascu- the risk of acute cardiovascular events. lar growth. Importantly, endothelial dysfunction has been impli- (J Am Coll Cardiol 1997;30:325–33) cated as a key event in the pathogenesis of atherosclerosis, ©1997 by the American College of coronary and, probably, myocardial ischemia.

For many decades the endothelium, the cell layer that lines the children and young adults with risk factors for atherosclerosis, vessels, was viewed simply as a semipermeable barrier such as and cigarette smoking (8). between blood and interstitium, facilitating the exchange of Therefore, abnormal endothelial physiology has been impli- water and small molecules. Recently, however, a series of cated both in early atherogenesis and later in the disease experiments have demonstrated that the endothelium has an process, in the control of dynamic plaque behavior. The enormous range of vital homeostatic functions (1). Instead of “biologic link” between endothelial damage and atherosclero- serving as an inert barrier, the endothelium is an antithrom- sis may be related to decreased arterial bioavailability of NO, bogenic vascular lining that also participates in metabolic, which may predispose to leucocyte and adhesion, synthetic and regulatory pathways (2). vasoconstriction and smooth muscle cell proliferation (9). Normal endothelial functions include control over throm- bosis and thrombolysis, platelet and leucocyte interactions with the vessel wall and regulation of vascular tone and growth (Fig. Normal Endothelial Function 1). Of particular interest to cardiologists, the endothelium The endothelium lies between the and the vascular secretes both powerful vasorelaxing (e.g., [NO]) smooth muscle. Although only one cell layer thick, it is able to and vasoconstricting substances (e.g., -1 [ET-1]) “sense” changes in hemodynamic forces, or blood-borne sig- (3,4). nals, by membrane receptor mechanisms and respond to Given that normal endothelial function plays a central role physical and chemical stimuli by synthesis or release of a in vascular , it follows that endothelial dysfunction variety of vasoactive and thromboregulatory molecules and probably contributes to disease states characterized by vaso- growth factors. Substances released by the endothelium in- spasm, vasoconstriction, excessive or abnormal clude , NO (the endothelium-derived relaxing fac- vascular proliferation, alone or in combination, including tor [EDRF]), , endothelial cell (s), atherosclerosis and (Fig. 2). , plasminogen inhibitors and Indeed in 1986, endothelial dysfunction was found in the (Fig. 1). In addition to these “universal” functions, the endo- coronary of humans with advanced atherosclerosis (5). thelium may have organ-specific roles that are differentiated Subsequent work (6,7) has demonstrated that such endothelial for various parts of the body, such as in the , dysfunction predisposes to vasoconstriction at the site of control of myocardial function in the or phagocytosis in arterial plaque, instead of the more normal dilator response, the liver and spleen. with potentially adverse clinical consequences. In 1992, endo- Studies of endothelial structure and function have been thelial dysfunction was also demonstrated in asymptomatic accomplished by a variety of techniques, including ultrastruc- tural studies (10), in vitro experiments for endothelial cell isolation and culture (11,12), physiologic studies in animals From the Department of Cardiology, Royal Prince Alfred Hospital and The (13) and, most recently, clinical studies in humans (5,8). This Heart Research Institute, Sydney, Australia. Dr. Celermajer is supported by The Medical Foundation, University of Sydney. knowledge has facilitated the development of certain treat- Manuscript received December 12, 1996; revised manuscript received April ment strategies based on administration of endothelial prod- 14, 1997, accepted April 24, 1997. ucts, such as prostacyclin and NO, or their antagonists. Address for correspondence: Dr. David S. Celermajer, Department of Cardiology, Royal Prince Alfred Hospital, Missenden Road, Camperdown 2050, Thromboresistance. The endothelium has anticoagulant, Sydney, Australia. E-mail: [email protected]. antiplatelet and fibrinolytic properties. Endothelial cells are

©1997 by the American College of Cardiology 0735-1097/97/$17.00 Published by Elsevier Science Inc. PII S0735-1097(97)00189-7 326 CELERMAJER JACC Vol. 30, No. 2 ENDOTHELIAL DYSFUNCTION August 1997:325–33

Abbreviations and Acronyms ACE ϭ -converting enzyme ADP ϭ diphosphate ATP ϭ adenosine triphosphate EDRF ϭ endothelium-derived relaxing factor ET-1 ϭ endothelin-1 LDL ϭ low density L-NMMA ϭ NG-monomethyl-L- NO ϭ nitric oxide NOS ϭ t-PA ϭ -type plasminogen activator

Figure 2. Some consequences of endothelial dysfunction. In the presence of certain risk factors, endothelial cells may produce less NO the major site for anticoagulant reactions involving Ϫ or more oxygen-derived free radicals (such as O2 ), or both. These (14,15). Platelet adhesion to endothelial cells is markedly changes may in turn result in certain proischemic or proatherogenic inhibited by the endothelium-derived arachidonic acid metab- effects, as indicated, including increased transcription of certain redox- olite prostacyclin (16,17). The same stimuli that activate plate- sensitive proinflammatory genes (Marui N, Offermann MK, Swerlick lets, such as thrombin and adenosine diphosphate (ADP) and R, et al. Vascular molecule-1 (VCAM-1) gene transcrip- tion and expression are regulated through an antioxidant-sensitive adenosine triphosphate (ATP), also act to release prostacyclin mechanism in human vascular endothelial cells. J Clin Invest 1993;92: from the endothelium, which allows the endothelium to limit 1886–94). Agents that can lead to increased NO bioavailability in the Ϫ the extent of platelet plug formation (18). The interactions vessel wall, or decreased O2 production, may be useful in reversing between and endothelium regulate platelet function, endothelial dysfunction (see text). In addition, decreased endothelial coagulation cascades and local vascular tone (19). production of the anticoagulant vasodilator prostacyclin or increased production of the vasoconstrictor ET-1 may also be pathogenetically In addition, endothelial cells may secrete tissue-type plas- important consequences of endothelial dysfunction. minogen activator (t-PA) (20), the powerful thrombolytic agent in frequent clinical use for treatment of coronary throm- botic occlusion. t-PA release is stimulated in vivo by norepi- nephrine, or stasis within the vessel lumen. Throm- even inactivate, circulating vasodilators and constrictors, such bin may also stimulate t-PA release, providing a further as thrombin, , ADP and ATP (23). endothelium-mediated safeguard against uncontrolled coagu- NO, an EDRF. The existence of an endothelial relaxing lation. factor was first postulated by Furchgott and Zawadzki (21) in Regulation of vascular tone. The central role of the endo- 1980, when they noticed that rabbit aortic rings relaxed in thelium in controlling vascular tone has only been appreciated response to only in the presence of an intact since the discovery of the potent vasodilators prostacyclin and endothelium. The biologic effects of EDRF are mediated by NO (16,21). The endothelium controls underlying smooth NO. NO is synthesized from L-arginine by an enzyme, nitric muscle tone in response to certain pharmacologic and physio- oxide synthase (NOS) (24). The reaction is stereospecific, and logic stimuli (22). This involves a number of lumen membrane L-arginine is converted to NO and L-citrulline. The generation receptors and complex intracellular pathways and the synthesis of NO from L-arginine can also be specifically blocked by G and release of a variety of relaxing and constricting substances, arginine analogues, such as N -monomethyl-L-arginine (L- described below. In addition to making their own vasoactive NMMA), which has recently proved to be a useful tool in mediators, endothelial cells may transduce signals from, or clinical research, allowing investigation of the biologic distri-

Figure 1. Some functions of normal endothelium. Fac- tors secreted into the lumen (upward arrows) include prostacyclin and t-PA, which influence coagulation. Cell surface adhesion molecules (such as intercellular adhe- sion molecule-1 [ICAM-1] and vascular -1 [VCAM-1]) regulate leucocyte adhesion. Factors secreted abluminally (toward the smooth muscle [downward arrows]) may powerfully influence vessel tone and growth. Coronary and endocardial en- dothelium may also influence myocardial contractility (41). JACC Vol. 30, No. 2 CELERMAJER 327 August 1997:325–33 ENDOTHELIAL DYSFUNCTION

Figure 3. In 1986, Ludmer et al. (5) showed paradoxic coronary vasoconstriction at plaque sites, suggesting endothelial dysfunction (*p Ͻ 0.001). These data first suggested an important role for endothelial physiology in determining dynamic plaque behavior at the sites of coronary artery stenosis. ACh ϭ acetylcholine; TNG ϭ trinitroglycerin. Adapted, with permission, from Ludmer et al. (5). Copyright 1986, Massachu- setts Medical Society. All rights reserved.

bution and role of NO. NOS has at least three isoforms, Endothelins (endothelium-derived contracting factors). including constitutively expressed and inducible enzymes. Endothelin is an extremely potent vasoconstrictor that was NO maintains low arterial tone at rest, in both the systemic isolated, purified and sequenced in 1988 (4). This 21-amino and pulmonary circulations (25–27). In addition, NO release is acid peptide is closely related to the snake venom sarafatoxin stimulated by increased flow (leading to increased shear stress and is one of the most potent vasoconstrictors yet identified. on the endothelium) (28) and by bradykinin, thrombin, acetyl- There are three isoforms of endothelin, but only one (ET-1) choline and a variety of other circulating agents that increase has been shown (35) to be released from human endothelial EDRF release through activation of specific endothelial cell cells. Like NO, ET-1 is synthesized in the endothelial cells and membrane receptors (22). released in response to a variety of stimuli, such as adrenaline The vasodilator activity of NO is due to interaction with the and hypoxia (36). ET-1 has a short half-life (37), suggesting iron atom of heme in guanylate cyclase, causing its activation that it, too, is mainly a locally active vasoregulator. Its physi- and thereby increasing intracellular cyclic guanosine mono- ologic role includes maintenance of basal , phosphate levels (29,30). In smooth muscle cells, this results in and it is present in healthy subjects in low concentrations (38). a reduction of intracellular calcium and thereby relaxation Elevated endothelin levels have been found in systemic and (31). The same pathway is involved in the mechanism of action pulmonary hypertension, and heart of exogenous nitrovasodilators, such as sodium nitroprusside failure; in these conditions, a pathologic role has been postu- and nitroglycerin. lated but not proved (39,40). Endothelium-derived hyperpolarizing factor. Stimulation In addition, normal endothelium also plays an important of the normal endothelium by acetylcholine also produces role in vascular growth, monocyte adhesion, immunologic regulation, metabolism of circulating amines, lipoprotein me- hyperpolarization of the underlying smooth muscle and tabolism and integration and transduction of blood-borne thereby vasorelaxation. This is not mediated by NO, but by signals. Endocardial and coronary microvascular endothelium another endothelium-derived factor, which acts by increasing may also regulate myocardial contractility (41). Kϩ conductance. The resulting is not inhibited by L-NMMA, the specific antagonist of NO (32,33). In contrast ouabain, a Naϩ/Kϩ-ATPase inhibitor, does not affect NO- induced relaxation, but prevents the action of endothelium- Clinical Assessment of Endothelial Function derived hyperpolarizing factor. The precise identity and phys- Over the past decade, a large number of studies have iologic role of this factor are uncertain. assessed arterial endothelial function in health and disease. Prostacyclin. Another major endothelium-derived vasodi- Most of these have tested the ability of normal endothelium to lator is prostacyclin, which is produced from arachidonic acid release the vasorelaxing factor NO in response to pharmaco- through the enzyme cyclooxygenase (16). Prostacyclin also has logic or physiologic stimuli, or both. Although this is only one antithrombotic and antiplatelet activity. Its release may be of many endothelial functions, NO release may be particularly stimulated by bradykinin and adenine nucleotides. Like NO, it important because of its actions on platelets, monocytes and is chemically unstable, with a short half-life (34). However, smooth muscle cells (9). unlike NO, prostacyclin acts through stimulation of adenylate Coronary artery testing. In vivo assessment of coronary cyclase and an increase in intracellular levels of cyclic AMP. It endothelial function in humans was first reported in the is a potent vasodilator and is active in both the pulmonary and mid-1980s (5) (Fig. 3). Coronary artery diameter was measured systemic circulations. by quantitative angiography before and after intracoronary 328 CELERMAJER JACC Vol. 30, No. 2 ENDOTHELIAL DYSFUNCTION August 1997:325–33

shown to be reproducible (49), due mainly to endothelial release of NO (50), and to correlate well with invasive testing of coronary endothelial function (51). Endothelial function has also been extensively investigated in the forearm microcircu- lation by intraarterial infusion of endothelium-dependent and -independent vasodilator substances, followed by measure- ment of forearm flow using plethysmographic techniques (52,53). Although these peripheral artery techniques provide only “surrogate” measures of coronary endothelial function, they have provided important insights into the risk factors for atherogenesis in childhood and young adult life and are also being used in studies of reversibility of endothelial dysfunction in asymptomatic subjects at high risk of arterial disease (54,55).

Figure 4. Endothelial dysfunction is also important in the early stages of atherogenesis; for example, significantly impaired endothelium- dependent flow-mediated dilation (EDD) has been found in asymp- Endothelial Dysfunction in tomatic children and young adults with risk factors for atherosclerosis, such as hypercholesterolemia and cigarette smoking. Adapted, with Early Atherogenesis permission, from Celermajer et al. (8). Endothelial injury, either physical trauma or more subtle cellular damage, is now regarded as an important initial event in atherogenesis (56,57). Even in nomocholesterolemic ani- infusion of acetylcholine. In normal arteries, acetylcholine mals, physical damage to the endothelium can lead to athero- stimulated the endothelial release of NO, resulting in vasodi- sclerotic lesion formation (58). Hypertension has been shown lation, whereas in subjects with endothelial dysfunction, vaso- experimentally to disrupt endothelial integrity (59). Hyperho- constriction was observed (due to a direct smooth muscle mocysteinemia, which causes chemical endothelial injury, is constrictor effect of acetylcholine). This response was con- associated with premature atherosclerosis and thrombosis trasted with the response to nitroglycerin, an exogenous source (60). The finding that many of these insults, which are associ- of NO and therefore an endothelium-independent vasodilator. ated with clinical progression of , are clearly Soon after, invasive testing of coronary microvascular endo- related to endothelial injury has added weight to the “response thelium was described (42) by measuring the response of to injury” hypothesis of Ross and Glomset (61) and its later coronary flow to administration of endothelium-dependent modifications (56). Other hypotheses have also been proposed and -independent small-vessel dilator substances, using Dopp- to explain the initiating events in atherosclerosis (62). ler wires or catheters. The consequences of endothelial damage that promote These techniques have given valuable insights into the risk and plaque formation include increased adherence factors for coronary endothelial dysfunction in angiographi- of monocytes (63), increased permeability to monocyte/ cally smooth arteries (43,44) and into the role of endothelial and , which then accumulate in the dysfunction in predisposing to dynamic plaque activation and vessel wall, increased platelet adherence and increased smooth constriction (5,6,45). More recently, the potential for revers- muscle cell migration and proliferation (64). The latter may be ibility of endothelial dysfunction in the coronary arteries has mediated by loss of endothelium-derived -like oligosac- been assessed using this methodology to assess novel thera- charides, which normally inhibit smooth muscle cells (65), or peutic strategies (e.g., cholesterol-lowering therapy and by allowing platelet adhesion and thereby increased local angiotensin-converting enzyme [ACE] inhibition) (46–48) concentration of platelet derived growth factor, which is also (see later). an important smooth muscle mitogen (66). Endothelial dys- Peripheral artery testing. The major disadvantage of intra- function may also be accompanied by decreased local avail- coronary testing is its invasive nature, and it is therefore ability of NO. This may be due to decreased endothelial generally unsuitable for use in children or adults who are at production of NO or to excess production of superoxide high risk of atherosclerosis but who have no clinical symptoms anions, or both, with consequent degradation of NO before it or signs of disease. For this reason, noninvasive or minimally can reach its target tissues. For example, superoxide produc- invasive clinical tests of endothelial function have been devel- tion is enhanced in the presence of hypercholesterolemia, with oped. Noninvasive detection of endothelial dysfunction in the consequently decreased bioavailability of NO (67). Because brachial and femoral arteries was first described in 1992 (8) NO is a local vasodilator and also inhibits platelet adherence (Fig. 4). In this test, arterial diameter is measured in response and aggregation, smooth muscle proliferation and endothelial- to an increase in shear stress, which causes endothelium- cell leucocyte interactions, reduced NO activity may also dependent dilation, and in response to sublingual nitroglyc- contribute to the initiation and progression of atherogenesis erin, an endothelium-independent dilator. (9). Indeed, supplementation with oral L-arginine, the physio- The brachial artery dilator response to shear stress has been logic substrate for NO production, has profound antiathero- JACC Vol. 30, No. 2 CELERMAJER 329 August 1997:325–33 ENDOTHELIAL DYSFUNCTION genic effects in cholesterol-fed animals (68) and has recently endothelial impairment and later advanced atherosclerotic been associated with decreased platelet aggregation (69) and disease. monocyte/endothelial cell adhesion (70) in humans. Endothelial dysfunction has been demonstrated in asymp- tomatic children and young adults with risk factors for athero- Endothelial Dysfunction in sclerosis (8). Hypercholesterolemia is associated with arterial Advanced Atherosclerosis endothelial dysfunction in children as young as 7 years old, with significant correlations between the degree of endothelial Obstructive coronary stenoses are usually thought to con- impairment and the levels of both low density lipoprotein tribute to angina pectoris by providing a fixed limitation to (LDL) cholesterol and lipoprotein(a) (71). Both active ciga- coronary flow during periods of increased myocardial oxygen rette smoking (72) and even prolonged exposure to environ- demand (e.g., during exertion). However, a series of studies mental tobacco smoke (73) have been shown to be associated over the past 10 years (5,6,45,86) has shown that coronary with impaired endothelium-dependent dilation, in a dose- stenoses are dynamic rather than fixed and that impaired dependent manner, in otherwise healthy teenagers and young endothelium-dependent dilation at the site of coronary adults. Aging, too, has been associated with progressive endo- plaques may result in paradoxic vasoconstriction during exer- thelial impairment (74,75), and this age-related dysfunction cise or mental stress, just when coronary vasodilation is most appears to occur earlier in men than in women (76). Endothe- required. lial dysfunction has also been implicated in the pathogenesis of In 1986, Ludmer et al. (5) first demonstrated this phenom- hypertension. Many studies have found impaired endothelial enon in human coronary arteries. Whereas normal arteries release of NO in hypertensive subjects (77–79), although one dilated in response to intracoronary acetylcholine, which stim- large study has failed to confirm this finding (80). Overproduc- ulates endothelial NO production, stenotic arteries showed paradoxic vasoconstriction. Subsequent studies demonstrated tion of ET-1 may also produce hypertension (81). However, it that coronary endothelial dysfunction (assessed as an abnor- is unclear whether abnormal endothelial production of vaso- mal response to acetylcholine) also resulted in impaired flow active factors can be a primary cause of high mediated coronary dilation in response to intracoronary pa- rather than simply an effect of the hypertensive state. paverine (45), cardiac pacing (87) and physiologic levels of It has also been shown (44,82) that the traditional vascular exercise (6). Abnormal coronary responses to in the risk factors may interact to damage the endothelium in asymp- presence of endothelial dysfunction may be particularly impor- tomatic subjects, in the same way as they are known to interact tant in unstable coronary syndromes in vivo, given the involve- in determining the risk of clinical cardiovascular end points. In ment of (serotonin containing) platelets in the pathophysiology the coronary circulation, endothelial dysfunction is not only of these processes (88). observed at the sites of obstructive stenoses, but has also been Similar endothelial dysfunction also occurs in the coronary documented in angiographically smooth arteries of subjects , particularly in response to hypercholesterol- with risk factors for atherosclerosis (43,44). Taken together, emia (89). Because endothelial vasodilator function of the these data from in vivo human studies indicate the importance coronary microvessels may be an important determinant of of impaired endothelium-dependent NO-mediated dilation in myocardial perfusion during periods of increased demand, the early stages of the atherosclerotic process. microvascular endothelial dysfunction may play a particularly Some limitations apply to these studies. To date, large- significant role in the pathogenesis of myocardial ischemia vessel endothelial function has only been investigated in (90,91). asymptomatic children and young adults using peripheral Some studies have demonstrated heterogeneous endothe- arteries (such as the brachial and femoral), which can be lial responses within the same coronary artery or within the studied noninvasively. Invasive coronary studies are confined same patient (92), and this finding is consistent with in vitro to symptomatic patients and therefore may suffer from a experiments demonstrating nonuniform endothelial cell be- selection bias. Nevertheless, there appears to be good corre- havior when exposed to apparently similar conditions of shear lation between coronary and peripheral artery endothelial stress (93). However, most studies overall have consistently function in the same subjects (51). Furthermore, no longitudi- reported the presence of impaired endothelium-dependent nal studies in humans have yet proved that those young dilation in atherosclerotic coronary vessels. Although serial subjects with endothelial dysfunction will go on to develop coronary physiologic studies are only infrequently performed, advanced atherosclerosis; such studies would take decades to recent intervention trials (46–48) have reported coronary complete. Despite this limitation, endothelial dysfunction is endothelial responses before and after 6 months of placebo spatially and temporally linked to atherosclerosis; endothelial treatment; in these studies, the endothelial responses to ace- dysfunction occurs first at coronary branch points (83), as do tylcholine appeared highly reproducible. advanced plaques; and endothelial dysfunction also precedes The clinical correlate of impaired endothelial function in occlusive arterial disease in both the experimental primate the coronary arteries may be episodic myocardial ischemia, model (84) and in human heart transplant recipients (85). either with or without chest pain. It has recently been ob- These data serve to support an important link between arterial served, for example, that cholesterol-lowering therapy may 330 CELERMAJER JACC Vol. 30, No. 2 ENDOTHELIAL DYSFUNCTION August 1997:325–33 result in significantly decreased myocardial ischemia (94,95) and improved cardiovascular survival in patients with coronary atherosclerosis (96,97). These benefits have occurred even though angiographic regression trials of cholesterol lowering have demonstrated only very minor (if any) regression of actual coronary plaque size (98,99). The mechanism for clinical improvement in the absence of coronary anatomic changes might relate to beneficial endothelial effects; that is, coronary function rather than structure might improve, resulting in fewer clinical symptoms and events. Alternate hypotheses for this observation include plaque remodeling or stabilization, or both, and a decreased tendency to thrombosis as the mecha- nisms for clinical benefit. Coronary endothelial injury may also play an important role Figure 5. Endothelial dysfunction may be reversible under certain in postangioplasty restenosis (100). Balloon angioplasty always circumstances. In this example (54), impaired endothelium-dependent results in some degree of endothelial injury, and the response dilation (EDD) in young adults with hypercholesterolemia was im- to this injury includes growth factor secretion and consequent proved significantly by 4 weeks of treatment with L-arginine, the myointimal proliferation. Again, the role of endothelium- precursor of NO. Reprinted, with permission of the American Society for Clinical Investigation, from Clarkson et al. (54). derived NO may be particularly important, with the recent demonstration in experimental studies (101) that local trans- fection of endothelial cells with NO synthase after angioplasty may attenuate this proliferative response, with a consequent weeks (Fig. 5). However, animal data suggest that the benefi- decrease in the incidence of restenosis. cial effects of L-arginine on endothelial function and athero- sclerosis may not be sustained over the long term (105), although the experimental data on this point are not consistent Reversibility of Endothelial Dysfunction (106). Current research is examining strategies that might improve ACE inhibition with quinapril has also been shown (48) to arterial endothelial function. This is an attractive possibility; if improve coronary endothelial function in patients with athero- such strategies could be implemented early in the disease sclerosis. This effect may contribute to the potential anti- process, it might be possible to prevent or retard atherogene- ischemic effect of this class of drugs and may be mediated in sis. Even in advanced disease, modification of coronary artery part by suppression of vascular wall superoxide production and endothelial dysfunction might decrease the propensity to va- consequently by enhanced bioavailability of NO (107). Such soconstriction or thrombosis, or both, and thereby decrease the findings suggest important links between the –angiotensin risk of acute cardiovascular events. and the L-arginine–NO systems in the vessel wall. Some animal Coronary endothelial function can be improved in hyper- data have indicated that calcium channel blocking agents, as cholesterolemic subjects by lowering (46,47), either short well as angiotensin blockade, might improve endothelial dys- term, for example, by LDL apheresis (102), or in the longer function in experimental models of hypertension (108), and term, with either diet and resin therapy (103) or HMG CoA short-term calcium channel antagonism has recently been reductase inhibitors such as lovastatin. In these latter studies, associated with improved coronary vasomotion in hypertensive coronary vasoconstriction observed at baseline was signifi- humans (109); however, long-term human studies with calcium cantly attenuated after cholesterol-lowering treatment for 6 channel blockers have not yet been reported. months. Improved dilation has been particularly observed in therapy has also been shown (100–112) to have a patients treated with a combination of cholesterol-lowering beneficial effect on endothelial function in postmenopausal and antioxidant therapy (46), highlighting the importance of women. This effect may mediate some of the apparent cardio- in the pathogenesis of endothelial dysfunction. protective benefit of estrogen replacement in older women. Oral administration of high dose vitamin C has also been Recent data (113,114) indicate that the addition of progester- shown (55) to improve brachial artery endothelial function in one to estrogen, in combined hormone replacement therapy, patients with established coronary disease, although it is not may also be associated with beneficial effects on endothelial known whether this effect is sustained with prolonged therapy. function. Improvement in endothelial function has also been ob- Early data suggest that the same therapeutic strategies that tained by therapeutic manipulation of the L-arginine–NO improve arterial endothelial function may also be associated pathway. Short-term parenteral administration of L-arginine with a decrease in myocardial ischemia and with ischemic has been associated with improved coronary and peripheral events rates. For example, cholesterol-lowering therapy not microvascular function (89,104), and a recent study (54) has only improves coronary endothelial physiology in hypercholes- demonstrated improved endothelium-dependent dilation in terolemic subjects, but may also significantly reduce the num- young hypercholesterolemic subjects given oral L-arginine for 4 ber and duration of transient myocardial ischemic episodes JACC Vol. 30, No. 2 CELERMAJER 331 August 1997:325–33 ENDOTHELIAL DYSFUNCTION

(94,95). Furthermore, ACE inhibition may improve coronary 10. Palade GE. Fine structure of blood . J Appl Physiol 1953;24: endothelial physiology and may also reduce the incidence of 1424–30. 11. Gimbrone MA. Culture of vascular endothelium. Prog Hemost Thromb clinical cardiovascular events (115). Although there may be 1976;3:1–28. other non–endothelium-related benefits of these therapeutic 12. Jaffe EA, editor. Biology of Endothelial Cells. Boston: Martinus Nijhoff, manipulations, it is possible that at least some of the observed 1984. benefit of such interventions may relate to reversibility of 13. Renkin EM. Multiple pathways of permeability. Circ Res 1977; 41:735–43. endothelial dysfunction. 14. Machovich R. Choices among the possible reaction routes catalysed by thrombin. Ann N Y Acad Sci 1986;485:170–83. 15. Esmon NL. Thrombomodulin. Semin Thromb Hemost 1987;13:454–63. Future Directions 16. Moncada S, Herman AG, Higgs EA, Vane JR. Differential formation of prostacyclin (PGX or PGI ) by layers of the arterial wall: an explanation for Diagnosis. 2 A simple noninvasive test for endothelial dys- the anti-thrombotic properties of vascular endothelium. Thromb Res function would potentially have immediate clinical applicabil- 1977;11:323–44. ity. Such a test might identify presymptomatic subjects at high 17. Moncada S. Biological importance of prostacyclin. Br J Pharmacol 1982; risk of atherosclerotic complications and therefore allow spe- 76:3–31. cific “targeting” of primary preventive strategies. Although 18. Jaffe EA. Cell biology of endothelial cells. Hum Pathol 1987;18:234–9. 19. Pearson JD, Gordon JL. Nucleotide metabolism by endothelium. Annu endothelial function testing is available in the research setting, Rev Physiol 1985;47:617–27. no technique yet exists that is simple, safe, reproducible and 20. Hekman CM, Loskutoff DJ. Fibrinolytic pathways and the endothelium. easily performed as a screening method. Semin Thromb Hemost 1987;13:514–27. Treatment. Current research has raised the exciting possi- 21. Furchgott RF, Zawadzki JV. The obligatory role of endothelial cells in the relaxation of arterial smooth muscle by acetylcholine. Nature 1980;288: bility that endothelial dysfunction might be reversible with 373–6. certain interventional strategies (46–48,54,55). In the case of 22. Vanhoutte PM. Endothelium and control of vascular function. Hyperten- naturally occurring substances, such as L-arginine and vitamin sion 1989;13:658–67. C, future studies in humans will be directed toward defining 23. Ryan JW, Ryan US. Endothelial surface enzymes and the dynamic process- ing of plasma substrates. Int Rev Exp Pathol 1984;26:1–43. the sustainability of any beneficial arterial effects with pro- 24. Palmer RMJ, Ashton DS, Moncada S. Vascular endothelial cells synthesise longed therapy. Certain pharmacologic agents have been iden- nitric oxide from L-arginine. Nature 1988;333:664–6. tified that result in improved endothelial function over at least 25. Vallance P, Collier J, Moncada S, et al. Effects of endothelium-derived a 6-month period (46–48). It is now important to investigate nitric oxide on peripheral arteriolar tone in man. Lancet 1989;2:103–10. 26. Celermajer DS, Dollery C, Burch M, Deanfield JE. Role of endothelium in whether such improvements in arterial physiology will be the maintenance of low pulmonary vascular tone in normal children. correlated with evidence of clinical benefit. Circulation 1994;89:2041–4. Potentially, the most exciting future direction relates to 27. Stamler JS, Loh E, Roddy MA, Currie KE, Creager MA. Nitric oxide improvement of arterial endothelial function in children and regulates basal systemic and pulmonary vascular resistance in healthy humans. Circulation 1994;89:2035–40. young adults, early in the atherogenic process. Intuitively, it is 28. Pohl U, Holtz J, Busse R, Bassenge E. Crucial role of endothelium in the at this early stage that arterial wall damage is most likely to be vasodilator response to increased flow in vivo. Hypertension 1986;8:37–44. easily reversible (116). If strategies to protect the endothelium 29. Arnold WP, Mittal CK, Katsuki S, Murad F. Nitric oxide activates early in life can be defined and applied successfully, this may guanylate cyclase and increases guanosine 3:5Ј-cyclic monophosphate levels have important public health benefits into the next millenium. in various tissue preparations. Proc Natl Acad Sci 1977;74:3203–7. 30. Griffith TM, Edwards DH, Lewis MJ, Newby AC, Henderson AH. The nature of endothelium-derived relaxant factor. Nature 1984;308:645–7. 31. Collins P, Griffith TM, Henderson AH, Lewis MJ. Endothelium-derived References relaxing factor alters calcium fluxes in rabbit : a cyclic guanosine monophosphate-mediated effect. J Physiol (Lond) 1986;381:427–37. 1. Petty RG, Pearson JD. Endothelium—the axis of vascular health and 32. Feletou M, Vanhoutte PM. Endothelium-dependent hyperpolarization of disease. J R Coll Physicians 1989;23:92–101. canine smooth muscle. Br J Pharmacol 1988;93:515–24. 2. Rubanyi GM. The role of endothelium in cardiovascular homeostasis and 33. Taylor SG, Weston AH. Endothelium-derived hyperpolarizing factor: a diseases. J Cardiovasc Pharmacol 1993;22 Suppl 4:S1–14. new endogenous inhibitor from the vascular endothelium. Trends Pharma- 3. Moncada S, Palmer RMJ, Higgs EA. Nitric oxide: physiology, pathophys- col Sci 1988;9:272–4. iology, and pharmacology. Pharmacol Rev 1991;43:109–42. 34. FitzGerald GA, Brash AR, Falardeau P, Oates JA. Estimated rate of 4. Yanagisawa M, Kurihara H, Kimura S, et al. A novel potent vasoconstrictor prostacyclin secretion into the circulation of normal man. J Clin Invest peptide produced by vascular endothelial cells. Nature 1988;332:411–5. 1981;68:1272–5. 5. Ludmer PL, Selwyn AP, Shook TL, et al. Paradoxical vasoconstriction 35. Masaki T. The discovery, the present state, and the future prospects of induced by acetylcholine in atherosclerotic coronary arteries. N Engl J Med endothelin. J Cardiovasc Pharmacol 1989;13 Suppl 5:S1–4. 1986;315:1046–51. 36. Boulanger C, Luscher TF. Release of endothelin from the porcine aorta: 6. Gordon JB, Ganz P, Nabel EG, et al. Atherosclerosis and endothelial inhibition by endothelium-derived nitric oxide. J Clin Invest 1990;85:587– function influence the coronary vasomotor response to exercise. J Clin 90. Invest 1989;83:1946–52. 37. Anggard E, Galton S, Rae G, et al. The fate of radio-iodinated endothelin-1 7. Healy B. Endothelial cell dysfunction: an emerging endocrinopathy linked and endothelin-3 in the rat. J Cardiovasc Pharmacol 1989;13 Suppl 5:S46–9. to coronary disease. J Am Coll Cardiol 1990;16:357–8. 38. Davenport AP, Ashley MJ, Easton P, et al. A sensitive radioimmunoassay 8. Celermajer DS, Sorensen KE, Gooch VM, et al. Non-invasive detection of measuring endothelin-like immunoreactivity in human plasma: comparison endothelial dysfunction in children and adults at risk of atherosclerosis. of levels in patients with essential hypertension and normotensive control Lancet 1992;340:1111–5. subjects. Clin Sci 1990;78:261–4. 9. Cooke JP, Tsao PS. Is NO an endogenous antiatherogenic molecule? 39. Warrens AN, Cassidy MJD, Takahashi K, Ghatei MA, Bloom SR. Endo- Arterioscler Thromb 1994;14:653–5. thelin in renal failure. Nephrol Dial Transplant 1990;5:418–22. 332 CELERMAJER JACC Vol. 30, No. 2 ENDOTHELIAL DYSFUNCTION August 1997:325–33

40. Cody RJ. The potential role of endothelin as a vasoconstrictor substrate in 68. Cooke JP, Singer AH, Tsao P, Zera P, Rowan RA, Billingham ME. congestive heart failure. Eur Heart J 1992;13:1573–8. Antiatherogenic effects of L-arginine in the hypercholesterolemic rabbit. 41. Paulus WJ, Vantrimpont PJ, Shah AM. Paracrine coronary endothelial J Clin Invest 1992;90:1168–72. control of left ventricular function in humans. Circulation 1995;92:2119–26. 69. Adams MR, Forsyth CJ, Jessup W, Robinson J, Celermajer DS. Oral 42. Drexler H, Zeiher AM. Endothelial function in human coronary arteries in L-arginine inhibits platelet aggregation but does not enhance endothelium- vivo. Hypertension 1991;18:II-90–9. dependent dilation in healthy young men. J Am Coll Cardiol 1995;26:1054– 43. Zeiher AM, Drexler H, Wollschlager H, Just H. Modulation of coronary 61. vasomotor tone in humans: progressive endothelial dysfunction with differ- 70. Adams MR, Jessup W, Celermajer DS. Cigarette smoking is associated with ent early stages of coronary atherosclerosis. Circulation 1991;83:391–401. increased monocyte adhesion to endothelial cells in humans: reversibility 44. Vita JA, Treasure CB, Nabel EG, et al. Coronary vasomotor response to with L-arginine but not vitamin C. J Am Coll Cardiol 1997;29:491–7. acetylcholine relates to risk factors for coronary artery disease. Circulation 71. Sorensen KE, Celermajer DS, Georgakopoulos D, Hatcher G, Betteridge 1990;81:491–7. DJ, Deanfield JE. Impairment of endothelium-dependent dilation is an 45. Nabel EG, Selwyn AP, Ganz P. Large coronary arteries in humans are early event in children with familial hypercholesterolemia and is related to the lipoprotein(a) level. J Clin Invest 1994;93:50–5. responsive to changing blood flow: an endothelium-dependent mechanism that fails in patients with atherosclerosis. J Am Coll Cardiol 1990;16:349– 72. Celermajer DS, Sorensen KE, Georgakopoulos D, et al. Cigarette smoking 56. is associated with dose-related and potentially reversible impairment of endothelium-dependent dilation in healthy young adults. Circulation 1993; 46. Anderson TJ, Meredith IT, Yeung AC, Frei B, Selwyn AP, Ganz P. The 88:2149–55. effect of cholesterol-lowering and antioxidant therapy on endothelium- 73. Celermajer DS, Adams MR, Clarkson P, et al. and dependent coronary vasomotion. N Engl J Med 1995;332:488–93. impaired endothelium-dependent arterial dilatation in healthy young 47. Treasure CB, Klein JL, Weintraub WS, et al. Beneficial effects of adults. N Engl J Med 1996;334:150–4. cholesterol-lowering therapy on the coronary endothelium in patients with 74. Zeiher AM, Drexler H, Saurbier B, Just H. Endothelium-mediated coro- coronary artery disease. N Engl J Med 1995;332:481–7. nary blood flow modulation in humans: effects of age, atherosclerosis, 48. Mancini GBJ, Henry GC, Macaya C, et al. Angiotensin-converting enzyme: hypercholesterolemia, and hypertension. J Clin Invest 1993;92:652–62. inhibition with quinapril improves endothelial vasomotor dysfunction in 75. Egashira K, Inou T, Hirooka Y, et al. Effects of age on endothelium- patients with coronary artery disease. Circulation 1996;94:258–65. dependent vasodilation of resistance coronary artery by acetylcholine in 49. Sorensen KE, Celermajer DS, Spiegelhalter DJ, et al. Non-invasive mea- humans. Circulation 1993;88:77–81. surement of human endothelium dependent arterial responses: accuracy 76. Celermajer DS, Sorensen KE, Spiegelhalter DJ, Georgakopoulos D, Rob- and reproducibility. Br Heart J 1995;74:247–53. inson J, Deanfield JE. Aging is associated with endothelial dysfunction in 50. Joannides R, Haefeli WE, Linder L, et al. Nitric oxide is responsible for healthy men years before the age-related decline in women. J Am Coll flow-dependent dilatation of human peripheral conduit arteries in vivo. Cardiol 1994;24:471–6. Circulation 1995;91:1314–9. 77. Linder L, Kiowski W, Buhler FR, Luscher TF. Indirect evidence for release 51. Anderson TJ, Uehata A, Gerhard MD, et al. Close relationship of of endothelium-derived relaxing factor in human forearm circulation in endothelial function in the human coronary and peripheral circulations. vivo: blunted response in essential hypertension. Circulation 1990;81: J Am Coll Cardiol 1995;26:1235–41. 1762–7. 52. Creager MA, Cooke JP, Mendelsohn ME, et al. Impaired vasodilation of 78. Panza JA, Quyyumi AA, Brush JE, Jr., Epstein SE. Abnormal endothelium- forearm resistance vessels in hypercholesterolemic humans. J Clin Invest dependent vascular relaxation in patients with essential hypertension. 1990;86:228–34. N Engl J Med 1990;323:22–7. 53. Panza JA, Quyyumi AA, Brush JE Jr, Epstein SE. Abnormal endothelium- 79. Calver A, Collier J, Moncada S, Vallance P. Effect of local intra-arterial dependent vascular relaxation in patients with essential hypertension. NG-monomethyl-L-arginine in patients with hypertension: the nitric oxide N Engl J Med 1990;323:22–7. dilator mechanism appears abnormal. J Hypertens 1992;10:1025–31. 54. Clarkson P, Adams MR, Powe AJ, et al. Oral L-arginine improves 80. Cockcroft JR, Chowienczyk PJ, Benjamin N, Ritter JM. Preserved endothelium-dependent dilation in hypercholesterolemic young adults. endothelium-dependent vasodilatation in patients with essential hyperten- J Clin Invest 1996;97:1989–94. sion. N Engl J Med 1994;330:1036–40. 55. Levine GN, Frei B, Kouloukis SN, Gerhard MO, Keaney JFJ, Vita JA. 81. Yokokawa K, Tahara H, Kohno M, et al. Hypertension associated with Ascorbic acid reverses endothelial vasomotor dysfunction in patients with endothelin-secreting malignant . Ann Intern Med coronary artery disease. Circulation 1996;93:1107–13. 1991;114:213–5. 56. Ross R. The pathogenesis of atherosclerosis: a perspective for the 1990s. 82. Celermajer DS, Sorensen KE, Bull C, Robinson J, Deanfield JE. Nature 1993;362:801–9. Endothelium-dependent dilation in the systemic arteries of asymptomatic 57. Clarkson TB, Weingard KW, Kaplan JR, Adams MR. Mechanisms of subjects relates to coronary risk factors and their interaction. J Am Coll atherogenesis. Circulation 1987;76 Suppl I:I-20–8. Cardiol 1994;24:1468–74. 58. Moore S. Thromboatherosclerosis in normolipemic rabbits: a result of 83. McLenachan JM, Vita J, Fish RD, et al. Early evidence of endothelial continued endothelial damage. Lab Invest 1973;29:478–87. vasodilator dysfunction at coronary branch points. Circulation 1990;82: 59. Reidy MA, Schwartz SM. A technique to investigate surface morphology 1169–73. and endothelial cell replication of small arteries: a study in acute 84. Harrison DG, Armstrong ML, Freiman PC, Heistad DD. Restoration of angiotensin-induced hypertension. Microvasc Res 1982;24:158–67. endothelium-dependent relaxation by dietary treatment of atherosclerosis. 60. DeGroot PG, Willems C, Boers GHJ, et al. Endothelial cell dysfunction in J Clin Invest 1987;80:1808–11. homocystinuria. Eur J Clin Invest 1983;13:405–10. 85. Fish RD, Nabel EG, Selwyn AP, et al. Responses of coronary arteries of 61. Ross R, Glomset JA. Atherosclerosis and the arterial smooth muscle cell. cardiac transplant patients to acetylcholine. J Clin Invest 1988;81:21–31. Science 1973;180:1332–39. 86. Yeung AC, Vekshtein VI, Krantz DS, et al. The effects of atherosclerosis on 62. Williams KJ, Tabas I. The response-to-retention hypothesis of early the vasomotor response of coronary arteries to mental stress. N Engl J Med atherogenesis. Arterioscler Thromb Vasc Biol 1995;15:551–61. 1991;325:1551–6. 63. Steinberg DS. Lipoproteins and the pathogenesis of atherosclerosis. Circu- 87. Nabel EG, Selwyn AP, Ganz P. Paradoxical narrowing of atherosclerotic lation 1987;76:508–14. coronary arteries induced by increases in heart rate. Circulation 1990;81: 64. Henderson AH. Endothelium in control. Br Heart J 1991;65:116–25. 850–9. 65. Castellot J, Addonizo ML, Rosenberg R, et al. Cultured endothelial cells 88. McFadden EP, Clarke JG, Davies GJ, Kaski JC, Haider AW, Maseri A. produce a heparin-like inhibitor of smooth muscle cell growth. J Cell Biol Effect of intracoronary serotonin on coronary vessels in patients with stable 1981;90:372. angina and patients with variant angina. N Engl J Med 1991;324:648–54. 66. Stiles CD. The molecular biology of platelet-derived growth factor. Cell 89. Sellke FW, Armstrong ML, Harrison DG. Endothelium-dependent vascu- 1983;33:653–5. lar relaxation is abnormal in the coronary microcirculation of atheroscle- 67. Ohara Y, Peterson TE, Harrison DG. Hypercholesterolemia increases rotic primates. Circulation 1990;81:1586–93. endothelial superoxide anion production. J Clin Invest 1993;91:2546–51. 90. Drexler H, Zeiher AM, Meinzer K, Just H. Correction of endothelial JACC Vol. 30, No. 2 CELERMAJER 333 August 1997:325–33 ENDOTHELIAL DYSFUNCTION

dysfunction in coronary microcirculation of hypercholesterolaemic patients therapy on coronary endothelium-dependent relaxation in hypercholester- by L-arginine. Lancet 1991;338:1546–50. olaemic patients. Lancet 1993;341:1496–500. 91. Zeiher AM, Krause T, Schachinger V, Minners J, Moser E. Impaired 104. Creager MA, Gallagher SH, Girerd XJ, Coleman S, Dzau VJ, Cooke JP. endothelium-dependent vasodilation of coronary resistance vessels is asso- L-arginine improves endothelium-dependent vasodilation in hypercholes- ciated with exercise-induced myocardial ischemia. Circulation 1995;91: terolemic humans. J Clin Invest 1992;90:1248–53. 2345–52. 105. Jeremy RW, McCarron H, Sullivan D. Effects of dietary L-arginine on 92. El Tamimi H, Mansour M, Wargovich TJ, et al. Constrictor and dilator atherosclerosis and endothelium-dependent vasodilation in the hypercho- responses to intracoronary acetylcholine in adjacent segments of the same lesterolemic rabbit. Circulation 1996;94:498–506. coronary artery in patients with coronary artery disease. Circulation 1994; 89:45–51. 106. Aji W, Ravalli S, Szabolcs M, et al. L-Arginine prevents xanthoma development and inhibits atherosclerosis in LDL receptor knockout mice. 93. Barbee KA, Mundel T, Lal R, Davies PF. Subcellular distribution of shear stress at the surface of flow-aligned and nonaligned endothelial monolayers. Circulation 1997;95:430–7. Am J Physiol 1995;268:H1765–72. 107. Rajagopalan S, Harrison DG. Reversing endothelial dysfunction. Circula- 94. Van Boven AJ, Jukema JW, Zwinderman AH, Crijins HJGM, Lie KI, tion 1996;94:240–3. Bruschke AVG. Reduction of transient myocardial ischemia with pravasta- 108. Dohi Y, Criscione L, Pfeiffer K, Luscher TF. Angiotensin blockade or tin in addition to the conventional treatment in patients with angina calcium antagonists improve endothelial dysfunction in hypertension: stud- pectoris. Circulation 1996;94:1503–5. ies in perfused mesenteric resistance arteries. J Cardiovasc Pharmacol 95. Andrews TC, Raby K, Barry J, et al. Effect of cholesterol reduction on 1994;24:372–9. myocardial ischemia in patients with coronary disease. Circulation 1997;95: 109. Frielingsdorf J, Seiler C, Kaufmann P, Vassalli G, Suter T, Hess OM. 324–8. Normalization of abnormal coronary vasomotion by calcium antagonists in 96. Randomised trial of cholesterol lowering in 4444 patients with coronary patients with hypertension. Circulation 1996;93:1380–7. heart disease: the Scandinavian Simvastatin Survival Study (4S). Lancet 110. Gilligan DM, Quyyumi AA, Cannon RO. Effects of physiological levels of 1994;344:1383–9. estrogen on coronary vasomotor function in postmenopausal women. 97. Sacks FM, Pfeffer MA, Moye LA, et al. The effects of pravastatin on Circulation 1994;89:2545–51. coronary events after myocardial infarction in patients with average cho- 111. Collins P, Rosano GM, Sarrel PM, et al. 17-Beta-estradiol attenuates lesterol levels. N Engl J Med 1996;335:1001–9. acetylcholine-induced coronary arterial constriction in women but not men 98. Blankenhorn DH, Azen SP, Kramsch DM, et al. Coronary angiographic with coronary heart disease. Circulation 1995;92:24–30. changes with lovastatin therapy: the Monitored Atherosclerosis Regression Study (MARS). The MARS Research Group. Ann Intern Med 1993;119: 112. Lieberman EH, Gerhard MD, Uehata A, et al. Estrogen improves 969–76. endothelium-dependent, flow-mediated vasodilation in postmenopausal 99. Brown G, Albers JJ, Fisher LD, et al. Regression of coronary artery disease women. Ann Intern Med 1994;121:936–41. as a result of intensive lipid-lowering therapy in men with high levels of 113. McCrohon JA, Adams MR, McCredie RJ, et al. Hormone replacement . N Engl J Med 1990;323:1289–98. therapy is associated with improved arterial physiology in healthy post- 100. Chesebro JH, Lam JY, Badimon L, Fuster V. Restenosis after arterial menopausal women. Clin Endocrinol (Oxf) 1996;45:435–41. angioplasty: a hemorrheologic response to injury. Am J Cardiol 1987;60: 114. Gerhard MD, Tawakol A, Halley EA. Long-term estradiol therapy with or 10B–6B. without progesterone improves endothelium dependent vasodilatation in 101. Von der Leyen HE, Gibbons GH, Morishita R, et al. Gene therapy post-menopausal women [abstract]. Circulation 1996;94 Suppl I:I-279. inhibiting neointimal vascular lesion: in vivo transfer of endothelial cell 115. Garg R, Yusuf S. Overview of randomized trials of angiotensin-converting nitric oxide synthase gene. Proc Natl Acad Sci USA 1995;92:1137–41. enzyme inhibitors on mortality and morbidity in patients with heart failure: 102. Tamai O, Matsuoka H, Itabe H, Wada Y, Kohno K, Imaizumi T. Single Collaborative Group on ACE Inhibitor Trials. JAMA 1995;273:1450–6. LDL apheresis improves endothelium-dependent vasodilatation in hyper- 116. Kwiterovich PO Jr. Diagnosis and management of familial dyslipoprotein- cholesterolemic humans. Circulation 1997;95:76–82. emia in children and adolescents: current issues in pediatric and adolescent 103. Leung WH, Lau CP, Wong CK. Beneficial effect of cholesterol-lowering endocrinology. Pediatr Clin North Am 1990;37:1489–523.