Vascular Compliance and Cardiovascular Disease

Home , Vascular resistance, Windkessel effect

AJH 1997;10:1175–1189 REVIEW

A Risk Factor or a Marker? Downloaded from https://academic.oup.com/ajh/article/10/10/1175/156506 by guest on 28 September 2021 Stephen P. Glasser, Donna K. Arnett, Gary E. McVeigh, Stanley M. Finkelstein, Alan J. Bank, Dennis J. Morgan, and Jay N. Cohn

athophysiologic changes in the blood vessels aorta of the aortic trunk in Chinese has a larger diam- are associated with a wide variety of cardio- eter and thinner media than that in Australians and vascular events, but our ability to assess vas- population differences such as these may be geneti- cular structure and function are limited. Al- cally determined.2 Studies have suggested that the Pthough arteriography provides some information angiotensin II type 1 receptor (AT1) gene is involved in regarding intimal pathology, it provides little informa- the development of aortic stiffness.3 A conceptual ex- tion about the structure of the arterial wall or its ample where abnormalities in vascular compliance physiology. A reduction in arterial compliance has might be both a risk factor and a marker is hyperten- long been regarded as a potentially useful indicator of sion. Hypertension may alter arterial wall tone and 1 the presence of arterial disease. Changes in the arte- structure increasing blood pressure, which results in a rial wall leading to reductions in arterial compliance decrease in compliance (ie, the decrease in compliance may precede the onset of clinically apparent disease, is a marker for hypertension). Alternatively, when and may identify individuals at risk before disease sclerotic changes occur in vessels arising from diseases onset (symptoms due to disease are, in general, late that may or may not increase blood pressure, de- manifestations of alterations in organ function). The creased compliance becomes a risk factor for the de- ability to predict alterations in vascular structure and velopment of hypertension. In the following discus- function before the onset of clinical diseases such as sions, it should be kept in mind that there is both atherosclerosis, hypertension, and diabetes mellitus morphologic (structural) and functional heterogeneity has potential advantages. Whether reduced vascular in the different vascular beds. Also, there is no ac- compliance precedes the development of cardiovascu- cepted ‘‘gold standard’’ methodology for estimating lar disease (ie, is a risk factor) or is the consequence of vascular compliance, so comparison of results ob- established cardiovascular disease (ie, a marker) is a tained with differing methodologies is difﬁcult if not matter of debate. To qualify as a risk factor the pres- impossible. ence of a condition must increase the probability of disease compared to those without the condition (im- DEFINITION OF COMPLIANCE, plying stronger causality). DISTENSIBILITY, ELASTICITY, Recent studies have suggested that the ascending AND STIFFNESS There exist a number of terms characterizing vascular wall dynamics, so that some studies report results in Received January 27, 1997. Accepted June 10, 1997. terms of stiffness or elastic modulus (Ep), whereas From the Department of Internal Medicine, Division of Cardiol- others report compliance or distensibility. See the Ap- ogy (SPG, AJB, DJM, JNC), and the Division of General Medicine (GEM), the Department of Laboratory Medicine and Pathology, pendix for a further deﬁnition of those terms. In this Division of Health Computer Sciences (SMF), School of Medicine, article, cross-sectional area is used as the reference and the Division of Epidemiology, School of Public Health (DKA), vascular dimension. Vessel diameter or volume can University of Minnesota, Minneapolis, Minnesota. Address correspondence and reprint requests to Stephen P. also be used, although it should be cautioned that Glasser, 3500 E. Fletcher Ave., Suite 218, Tampa, FL 33613. during growth and aging, vessel diameter and volume

are not necessarily concordant. Compliance will be ties of the arterial tree, they have limitations. For in- defined as the change in area for a given change in stance, many of these models assume that the arterial pressure, whereas distensibility is the fractional wall is uniform in terms of compliance, and thereby change in area for a given change in pressure (by assume that the arterial tree is nontapering and non- using the percent change in area rather than absolute branching. The models that do attempt to take these change, blood vessels of different size can be more aspects into account are usually too complex for clin- readily compared). Elastic modulus is a term that ical use. Vascular tone has traditionally been assessed describes the stiffness of the blood vessel wall. Unlike by determining systemic vascular resistance (SVR), compliance, it is independent of size or geometry; and which is a calculation based on steady state flow that is defined as the change in wall stress for a given does not exist in the pulsatile arterial system. As a change in strain (see the discussion of wall stress and relatively crude measure of arterial homeostasis, SVR strain below). Since the relationship between stress is determined by the arterioles and not the large con- Downloaded from https://academic.oup.com/ajh/article/10/10/1175/156506 by guest on 28 September 2021 and strain in blood vessels is nonlinear, the term in- duit vessels, and, therefore is not influenced by large cremental elastic modulus is used and is defined as the vessel compliance. slope of a tangent to the stress-strain curve. The non- Characteristics of Blood Pressure Somewhat sim- linear relationship between pressure and area or stress plistically, blood pressure can be characterized by con- and strain requires that a single value of compliance, sidering two variables: mean arterial pressure and distensibility, or elastic modulus cannot be reported pulse pressure. Mean arterial pressure is dependent for a blood vessel. Rather, these values must be spec- on cardiac output and peripheral vascular resistance, ified at a given pressure. To more fully understand and represents the steady state component of blood compliance, a brief discussion of the Windkessel con- pressure. Pulse pressure is more complex and is influ- cept follows. enced by arterial stiffness, which is a term used to Windkessel Concept The arterial system represents describe a vessel’s ability to undergo deformation, a network of vessels designed to convert intermittent stroke volume, and left ventricular ejection rate (since flow from the heart to a continuous and steady flow as much as 66% of the stroke volume remains in the across the capillaries (the Windkessel effect). The level aorta and large arteries at the end of systole, the of vascular tone, wave reflection, compliance, and in- compliance of this ‘‘compartment’’ is a major determi- ertance are important contributors to this process. The nant of pulsatile flow). Pulsatile pressure is influenced Windkessel concept can be portrayed in mechanical or by the summation of forward (ie, the pressure wave electrical terms. Stephen Hale likened the arterial sys- established when the stroke volume is ejected into the tem to a contemporary fire engine device that con- aorta) and reflected (the wave generated when the verted intermittent spurts of water from a pump to forward propagated wave meets a change in imped- smooth flow by the use of a cushioning device that ance pressure waves) (ie, resistance/compliance mis- was an inverted air filled dome (called a Windkessel).1 match). These forward and reflecting waves primarily The fire engine consisted of a pump (the electronic depend on the system’s compliance. For a more com- plete discussion of this area the reader is referred to equivalent being a voltage source that generates cur- 4 rent, ie, flow). The Windkessel served as the compli- McDonald. ance component (the electronic equivalent being a Wall Stress and Strain If one accepts the postulate capacitor), the fire hose represented the conduit func- that the consequences of vascular damage are related tion of the arterial system, the wall of the hose pro- not only to the level of mean arterial pressure but to viding the recoil effect (contributing to inertance), and pulsatile pressure as well, then an understanding of the nozzle introduced resistance (electronically por- the different kinds of wall stress that the arterial vas- trayed by a resistor). A number of models have been culature deals with is important. Stress is the intensity created in order to approximate the properties of the of force over a unit of area, and may be expansile arterial tree; in many engineering problems, complex (consisting of distending pressures that are both cir- systems are often reduced to more elementary electri- cumferential and longitudinal, this is called tensile cal models. The simplest electrical model of the Wind- stress), compressive, longitudinal, and shear (stress kessel consists of a resistor and capacitor connected in that is tangential to the axis of flow). Shear stress also parallel to a voltage source, representing the left ven- has longitudinal dynamics, but reflects the fact that tricle. Although this basic Windkessel and more com- fluids such as blood move in layers, with flow in the plex models are useful, they fail to represent the phe- middle of the stream moving more rapidly than at the nomenon of wave transmission and reflection within vessel-blood interface. Shear stress may be an impor- the arterial tree, so that other models have also been tant determinant in stimulating the release of endo- used to account for that phenomenon. Although all thelium-derived relaxing factor and prostaglandins these models are useful in representing many proper- that affect smooth muscle tone.5 Strain (the ratio of AJH–OCTOBER 1997–VOL. 10, NO. 10, PART 1 VASCULAR COMPLIANCE AND CARDIOVASCULAR RESEARCH 1177

change in area to the initial area) is the result of stress, and the stress/strain ratio, or elastic modulus ex- presses the magnitude of stress required to produce a given strain. STRUCTURAL AND FUNCTIONAL DETERMINANTS OF COMPLIANCE Compliance is not only inﬂuenced by changes in both the large and small artery size, but may have structural and functional components.

Functional Determinants of Compliance The func- Downloaded from https://academic.oup.com/ajh/article/10/10/1175/156506 by guest on 28 September 2021 tional determinants relate to neurohumoral influences, such as the renin-angiotensin system, the adrenergic nervous system, and endothelium-derived factors, FIGURE 1. This conceptual model depicts the relationship be- such as endothelium derived relaxing factor (EDRF) tween stress and strain, which is initially linear, then becomes nonlinear to the point of fracture. Adapted from Lee and Kamm.10 and endothelin. Because abnormalities in EDRF effect have been observed in hypertension and atherosclerosis and with aging and cigarette smoking, the same conditions associated with abnormalities in vascular changes in arterial compliance, can have an important compliance, the relationship between EDRF and com- effect on organ function. That is, if the buffering ca- pliance is intriguing.6,7 pability of the vasculature is decreased, there would Structural Determinants of Compliance The struc- be insufficient damping of the pulsatile pressure, and tural elements of the vessel wall that are particularly thus an alteration of the smooth continuous flow at the important in determining the vessel’s stiffness are col- capillary level. Although little is known about the lagen and elastin. With aging (accentuated if there is influence of pulsatile versus continuous flow into the the concurrent presence of hypertension and athero- precapillary and capillary vasculature, it has been pos- sclerosis), there is a thinning and fracturing of elastin tulated that pulsatile pressure in the small vessels and increased collagen deposition resulting in in- could contribute to small vessel damage and thereby creased wall thickness; these changes adversely affect organ dysfunction. 8 compliance. Elastin is very stretchable and is impor- Large Versus Small Artery Compliance When con- tant to pulsatile behavior, but probably not of great sidering compliance there is no distinct separation importance in determining vessel wall strength. Col- anatomically between large and small arteries. How- lagen fibers, in contrast, are stiffer and can resist ever, it may prove useful to separate them conceptu- stresses Ͼ 100 times the fracture stress of elastin fi- 9 ally. This separation may be important, as some dis- bers. Thus, the blood vessel wall is considered to act eases predominantly affect the systolic wave form in a biphasic manner with the elastin fibers important (mostly as a result of alterations in the larger arteries), in determining stiffness at low distending pressures, whereas other diseases affect the diastolic waveform and collagen in determining stiffness at high distend- (mostly a result of alterations in the smaller arteries), ing pressures. This biphasic behavior is one reason resulting from pulse wave transmission and reflected that wall stiffness has a nonlinear relationship with waves. pressure (Figure 1).10 Furthermore, increased smooth muscle tone or smooth muscle cell hypertrophy influ- Pulse Wave Transmission The arterial pressure ence arterial stiffness, as do cell-cell and cell-matrix waveform is derived from the complex interaction of attachments, such as integrins. the LV stroke volume, the physical properties of the To summarize, in normal vasculature, the large ar- arterial tree, and the characteristics of the fluid in the teries act as a ‘‘buffering’’ system that is dependent on system. At the time of LV ejection, a pressure wave is vessel compliance. During systole, the stroke volume initiated that is propagated forward in the blood as is rapidly infused into the arterial tree, with only 20% well as by the aortic and arterial walls. The flow and to 30% resulting in forward flow, whereas the rest is pressure waves are propagated out of phase, with the stored in the large arteries and subsequently released pressure wave in the arterial wall propagated more into the periphery during diastole. This buffering ac- rapidly than the flow (ie, blood) wave. This wave tion essentially converts the pulsatile flow at the level propagation is faster in stiffer and smaller vessels. It is of the aorta to continuous flow in the capillaries (ie, also well known that there is a distortion of the pres- the Windkessel effect). Alterations in this buffering sure waveform as it travels forward (although the capability of the large arteries, mediated through mean pressure is not affected), and that this distortion 1178 GLASSER ET AL AJH–OCTOBER 1997–VOL. 10, NO. 10, PART 1

is also a function of vessel compliance. The pressure wave distortion is also affected by reflected waves returning from the periphery and interacting with the forward propagating waves, and is a function of peripheral vascular resistance, pulse wave reflection at bifurcations along the arterial tree, and small vessel compliance. This reflected wave augmentation of the systolic portion of the pressure waveform has recently been popularized as the augmentation index, an index incorporated into the ‘‘vascular overload concept’’ (see below).11 In summary, compliance is a measure of the storage Downloaded from https://academic.oup.com/ajh/article/10/10/1175/156506 by guest on 28 September 2021 capacity of the arteries. Although it is recognized that the proximal aorta and its major branches are the most compliant portion of the arterial circulation, the distal vessels contribute to circulatory regulation. The distal vessels have a small storage capacity, but are the site FIGURE 2. An electrical schematic of the modified Windkessel where the forward propagating waves are reflected model of the vasculature. back (reverberate or oscillate) centrally. Clinical application of compliance measurement is hampered by the fact that the volume in a vessel and the intraarterial artery segments only, whereas transcutaneous ultra- pressure are not easily measured, and when mea- sonic techniques are limited by the ability of the ul- sured, compliance often is estimated in one arterial trasonic method to accurately image the anterior and segment and may not be representative of compliance posterior arterial wall, and may only be used with throughout the vascular tree. larger vessels. Another noninvasive imaging technique, magnetic resonance imaging (MRI), offers the METHODS TO ESTIMATE COMPLIANCE advantage of not being limited by an acoustic win- Currently, arterial compliance can be estimated by dow; unfortunately, it also shares many of the disad- indirect methods using pulse wave velocity (PWV), vantages of ultrasound. Forearm blood flow estimated pulse pressure, pulse contour analysis, and Fourier by plethysmography measures the volume changes in transformation, and can be measured as lumped com- limbs and can be adapted to estimating large vessel pliance (ie, an estimate of global compliance), com- compliance. A newly developed noninvasive instru- partmental compliance, or large or small artery com- ment that estimates compliance using pulse contour pliance (this latter term does not refer to any distinct analysis recorded by tonometry from the radial artery anatomical division, but rather a conceptual divi- is also undergoing clinical testing.13 This technique sion).12 Historically, arterial compliance was esti- uses a modified Windkessel model that includes two mated by in vitro methods using excised arteries. In compliance elements (generally termed C1 and C2) vivo (invasive) methods of determining compliance combined with inertance and resistance elements (Fig- frequently use pulse contour analysis and require ure 2).12 A computer algorithm calculates the decay in catheterization and the use of high fidelity pressure diastolic pressure along with an estimation of stroke transducers (thus limiting its applicability to the study volume. The entire compliance evaluation can usually of large numbers of patients). Pulse contour analysis be obtained in Ͻ 10 min. With this method, compli- uses a mathematical formula to match the shape of the ance can be assessed separately as a function of capac- pressure wave during diastole and an electrical analog itance of the arterial system (C1) or of reflectance or model of the circulation to calculate compliance. A oscillations in the arterial system (C2). The capacitance new invasive technique for estimating compliance is resides in the larger arteries, which are particularly intravascular ultrasound, where change in dimension sensitive to aging, and the reflectance resides in the is accurately measured at different transmural pres- smaller vessels, which have been shown to be partic- sures by changing the pressure of an overlying occlu- ularly sensitive to the vascular disease associated with sive blood pressure cuff. The greater the vessel com- hypertension, diabetes mellitus, and atherosclerosis. pliance, the slower the pulse wave transmission, this There currently is no agreement as to ‘‘the best’’ being the underlying assumption of pulse wave veloc- measure of compliance, and the literature is replete ity (PWV). Ultrasound imaging, along with other im- with examples using compliance interchangeably with aging techniques, measures change in diameter com- distensibility, elasticity, or stiffness, as well as measur- pared to measured changes in pressure. Each tech- ing compliance with a variety of techniques that may nique has inherent limitations; PWV measures large be assessing different functions of the blood vessel AJH–OCTOBER 1997–VOL. 10, NO. 10, PART 1 VASCULAR COMPLIANCE AND CARDIOVASCULAR RESEARCH 1179

wall. For example, compliance is pressure-dependent and many studies do not take this into account. Whether the technique is estimating overall compliance, large vessel compliance only, or large and small vessel compliance from compliance that is a result of changes at the arteriolar level may be important. Fur- thermore, there are as yet no well designed prospective longitudinal studies reported relating abnormal compliance to outcome. However, these facts should not obfuscate the commonality of pulse wave alteration in conditions of reduced compliance. What is evident, even from some of the earlier research in Downloaded from https://academic.oup.com/ajh/article/10/10/1175/156506 by guest on 28 September 2021 compliance, is that the pressure waveform is altered by a variety of diseases that reduce vascular compli- FIGURE 3. This demonstrates the relationship between trans- ance, and that alteration is a decrease in the magni- mural pressure and compliance at baseline (comp.@base.) demon- strating that as the transmural pressure increases, compliance tude of the diastolic wave (as altered by large vessel decreases. This same relationship holds after nitroglycerin admin- compliance and systemic vascular resistance) and a istration (comp.@NTG). However, a new pressure/compliance measure of the slope of the diastolic pressure or aug- curve is present. Adapted from Bank et al.16 mentation in the systolic wave (as estimates of small vessel compliance). With these above considerations in mind, cardiovascular disease morbidity and mor- Nitroprusside is a drug in which compliance may be tality that have been linked to alterations in vascular an important consideration. Nitroprusside is a smooth compliance, and drugs that have a significant effect on muscle relaxant that is used to lower blood pressure in vascular compliance will be discussed. hypertension and to reduce vascular impedance in congestive heart failure. Whereas nitroprusside’s vas- THE EFFECT OF SELECTED cular effect is usually determined by its resultant re- CARDIOVASCULAR DRUGS duction in systemic vascular resistance, its effect on ON COMPLIANCE compliance may actually be more important. Feske et Vasodilating drugs are widely used in clinical practice al demonstrated that with nitroprusside there was a and represent a group of agents that have effects on 19% decrease in mean arterial pressure and a 6% blood pressure, systemic vascular resistance, and vas- variable effect on systemic vascular resistance, but cular compliance.14 Drugs that are vasodilators will there was a 19% increase in large vessel compliance usually decrease blood pressure, which in turn will and a 326% increase in small vessel compliance, sug- affect compliance. Many studies cannot differentiate a gesting that compliance measures were more sensitive compliance change that is the result of a drug’s effect indicators of change in the vascular functional state on blood pressure from a drug’s direct effect on the than systemic vascular resistance.17 Similar concepts vessel wall; an important point because it would ap- might also apply to nitroglycerin. Smulyan et al mea- pear that a therapy that directly affects the vessel wall sured forearm arterial distensibility using a technique structurally could be of greater clinical benefit com- whereby the pressure effect on distensibility could be pared to a therapy that only lowers blood pressure separated from any direct effect on the artery.18 They secondary to functional vasodilation.15 In addition, concluded that, at every studied pressure, nitroglyc- the effect of these therapies on left ventricular contrac- erin infusion increased forearm arterial distensibility. tility and in relation to pulsatile flow must also be With the above caveats in mind, a number of car- considered. Drugs that relax smooth muscle can in- diovascular drugs have been studied regarding their crease compliance by a number of mechanisms. First, effect on compliance. Atenolol (a ␤-blocking agent), they can decrease arterial pressure, which results in a nitrendipine (a dihydropyridine calcium antagonist), shift to a more compliant portion of a given compli- and lisinopril (an angiotensin converting enzyme in- ance-pressure curve. Second, drugs can directly relax hibitor), were studied by Perret et al in an attempt to the smooth muscle in the arterial wall, and cause an separate the effects on compliance from effects on upward shift to a new compliance-pressure curve pressure in 32 normotensive volunteers.19 Compared (Figure 3). The decrease in wall tension following to placebo, only lisinopril appeared to have any effect smooth muscle relaxation can, however, be offset by on vascular compliance. Of course, the effect of these increased wall stiffness as a result of transferring ad- drugs in normotensives might differ from that of hy- ditional stress to the stiff collagen fibers.16 Thus, the pertensives (see the additional discussion of drug ef- effects of a drug on arterial compliance are complex fects in the section on hypertension). Other drugs and can involve a number of different mechanisms. studied under a variety of conditions have suggested 1180 GLASSER ET AL AJH–OCTOBER 1997–VOL. 10, NO. 10, PART 1

measure was corrected for age and sex, signiﬁcant correlations were present between compliance and cholesterol, LDL-cholesterol, HDL-cholesterol, and the HDL/LDL ratio.

Aging It has been repeatedly suggested that both systolic and diastolic BP increase while compliance decreases between the ages of 20 and 50 years (in women there is about a 10-year delay compared to men). After the age of 50 there is a more rapid increase in systolic pressure and a more rapid decrease in

compliance (Figure 4 and Table 1), whereas the diasto- Downloaded from https://academic.oup.com/ajh/article/10/10/1175/156506 by guest on 28 September 2021 lic pressure plateaus, and then declines in later FIGURE 4. Reproduction of the variation in compliance in the years.28–35 Some of these observed changes are almost aorta. Similar changes were observed in the right and left iliac certainly due to changes in arterial stiffness, the result arteries and left femoral artery. Compliance was assessed by Dopp- of structural changes in collagen and elastin, along ler signals from two detectors (PWV) and expressed as a percent- 38 with the potential for a noncausal association with age per 10 mm Hg. Adapted from Laogun and Gosling. atherosclerosis. The Baltimore Longitudinal Study of Aging reported on changes in systolic pressure, the carotid pulse augmentation index, and the aortic pulse that compliance is increased by a number of vasodi- wave velocity in 146 male and female volunteers aged lators but less so or not at all by ␤-blockers or thiazide 21 to 96 years. Systolic BP increased 14% between diuretics.20,21 these ages and the rise was similar in sedentary compared to endurance-trained individuals.36 Arterial CARDIOVASCULAR RISK FACTORS AND stiffness increased ﬁve-fold in the sedentary and two- VASCULAR COMPLIANCE fold in the endurance-trained groups. A number of studies have suggested that the widened Studies have suggested that, in the younger age pulse pressure by itself (an indirect measure of blood groups, ischemic heart disease is related to both sys- vessel stiffness) is a cardiovascular risk factor.19–25 In tolic and diastolic blood pressure, but that after the addition, Kupari et al attempted to identify factors age of 50 the relation to systolic pressure is much that would be associated with abnormalities in vessel stronger.37 Keely et al noninvasively recorded arterial compliance by studying cardiac risk factors and MRI pressure waveforms (from the carotid, femoral, or determined compliance in 55 healthy subjects born in radial arteries) in 1005 normal subjects ranging in age 1954 and studied in the 1990s.26 Reduced compliance from 2 to 91 years.29 Observed changes per decade was related to the magnitude of the mean blood pres- were increasing pulse amplitude, steepening of the sure, LDL and HDL level, and physical inactivity. diastolic decay, and decreased prominence of the di- Lehmann et al measured aortic compliance using astolic wave, all likely the result of arterial stiffening. Doppler ultrasound PWV in 20 young patients with One continuing controversy relates to whether the heterozygous familial hypercholesterolemia and com- decreasing compliance is a natural phenomenon of pared the results with 20 age and sex matched con- aging or is due to associated atherosclerosis. Two ep- trols.27 They concluded that when the compliance idemiologic studies in Chinese subjects demonstrated

TABLE 1. DECREASED VASCULAR ‘‘COMPLIANCE’’ WITH AGING Study(tech.,ref.) No. of Subjects Comments

KeelyD,29 1005 Normals age 2–91 years ShimojoC,30 22 Normals, 30 with HTN, 36 post-MI Presence of disease accelerated aging changes BenetosE,31 78 Normals and HTN Also found heterogeniety amongst vascular beds AvolioC,D,32 480 Normals AvolioC,D,33 524 Normals PaganiA,34 9 Adults, 7 newborns, 5 fetal sheep MohiaddinF,35 70 Normals

Compliance measured in a variety of patient populations by a variety of techniques: A, invasive pulse contour analysis; B, invasive pulse wave velocity; C, ultrasound; D, noninvasive pulse wave velocity; E, tonometry. tech, technique; ref., reference number; HTN, hypertensives; MI, myocardial infarction. AJH–OCTOBER 1997–VOL. 10, NO. 10, PART 1 VASCULAR COMPLIANCE AND CARDIOVASCULAR RESEARCH 1181

decreasing compliance with age, and these subjects demonstrate short-term changes in arterial distensibil- were free of overt cardiac disease and represent a ity, but not long-term effects, when smokers were population that manifests little atherosclerosis.33,36 compared to nonsmokers.48 However, Kool et al did However, it has also been observed that the decreas- postulate that because plaque rupture is the inciting ing vascular compliance associated with aging is ac- mechanism of acute cardiovascular events that the celerated and magnified with aging in patients with short-term effects of smoking might, in fact, be a more hypertension.38 important risk than the long-term effects. The association of smoking and hypertension with increased Gender and Menopause Changes in endothelial cardiovascular morbidity and mortality has also been function have been observed in association with shown; and, acutely, cigarette smoking raises the menopause, and this change has in part been linked to 39 blood pressure, although with chronic exposure to estrogens. Estrogen receptors are present in the vas- Downloaded from https://academic.oup.com/ajh/article/10/10/1175/156506 by guest on 28 September 2021 cigarette smoke the relationship to BP is less clear. cular endothelium and several studies have evaluated Levenson et al studied the effect of cigarette smoking the effects of postmenopausal hormone replacement on arterial stiffness in 33 normotensive and 81 hyper- on endothelial function.40,41 Fewer studies are avail- tensive men of whom 22 and 24, respectively, were able that pertain directly to vascular compliance. Ar- cigarette smokers.49 They found a decrease in compli- nett et al studied the relationship between arterial ance in normotensive smokers compared to normoten- stiffness, menopausal status, and left ventricular hy- sive nonsmokers. Moreover, compliance was reduced pertrophy in a cross-sectional study of the Atheroscle- in hypertensive smokers compared to hypertensive rosis Risk In Communities (ARIC) population.42 They nonsmokers (despite similar BP in the two groups). found that increasing stiffness was associated with increased left ventricular mass, and that both were Exercise There are only a few studies relating exer- higher in post- compared to premenopausal women. cise and vascular compliance. Cameron et al investi- Arterial stiffness changes in men versus women across gated the effect of a 4-week exercise training program the age spectrum were similar between the two sexes in 13 previously sedentary young men.50 They found up to age 12 or so and after age 50. Between the ages that arterial compliance, as measured by aortic arch of 12 (adolescence) and 50 (the time of menopause) echocardiography, increased with exercise training women had less vessel stiffness than men, again im- and the increase was associated with changes in max- plying the potential role of estrogens. To further ad- imal oxygen consumption. They suggested that dress this relationship, Arnett et al evaluated arterial change in compliance was not the result of changes in stiffness (as estimated from carotid B-mode ultra- blood pressure, but that there were structural as well sound) and hormone replacement therapy in surgi- as functional contributions to the change in compli- cally and naturally occurring postmenopausal wom- ance. Wijnen et al studied various vessels by ultra- 43 en. Hormone replacement therapy in that study was sound techniques in 15 trained male cyclists and 15 associated with significantly lower arterial stiffness healthy sedentary males.51 They found that compli- compared to nonuse. Also, naturally occurring meno- ance was higher in the brachial artery (a muscular pause was associated with a lower arterial stiffness artery) of the trained athletes; this suggesting a func- compared to surgically induced menopause. Sonesson tional but not structural improvement in compliance. et al studied ultrasound determined compliance of the In another study, Wijnen et al found that a 6 week abdominal aorta in women age 4 to 74 years and moderate exercise training program in previously sed- compared these changes to previously recorded entary normotensive subjects did not lead to compli- 44 changes in healthy men. In women, they found an ance changes despite clear training effects in respect to almost linear decrease in compliance with age, oxygen consumption and heart rate.52 whereas in men the compliance decrease with age was exponential in nature. DISEASES ASSOCIATED WITH ABNORMALITIES IN VASCULAR Cigarette Smoking Cigarette smoking is a known COMPLIANCE risk factor for atherosclerosis, and is known to adversely effect endothelial function.45 The mechanisms Hypertension Hypertension is a seemingly easily di- whereby smoking accelerates blood vessel damage agnosed condition for which there are multiple ther- remain poorly understood. Cigarette smoke has been apeutic modalities and treatment is of proven benefit shown to reduce basal (although not methacholine- for improving long-term mortality and morbidity. stimulated) nitric oxide production.46 McVeigh et al, However, within the spectrum of the condition called using intravascular pulse contour analysis, demon- hypertension, there are many patients unaffected by strated reduced small vessel compliance in smokers the elevation of blood pressure, particularly as it ap- compared to nonsmokers,47 but Kool et al, using non- plies to vascular and atherosclerotic complications, or invasive pulse contour analysis, were only able to to the development of left ventricular hypertrophy 1182 GLASSER ET AL AJH–OCTOBER 1997–VOL. 10, NO. 10, PART 1

TABLE 2. ‘‘COMPLIANCE’’ CHANGES IN CROSS-SECTIONAL STUDIES OF HYPERTENSION (HTN) Studytech,ref No. of Subjects Comments

McVeighA,60 38 with HTN, 32 controls Proximal compliance decreased 19%, distal compliance decreased 72% FinkelsteinA,61 7 with HTN, 7 normals FeskeA,62 6 with HTN Proximal and distal compliance increased with exercise LiuA,63 11 with HTN, 7 normals Compliance normalized with nitroprusside RomanC,64 43 with HTN, 43 normals Compliance decreased when nonlinearity of p/v and distending pressure were considered VenturaA,65 36 with HTN, 16 borderline HTN, 27 Compliance decreased in HTN and borderline Downloaded from https://academic.oup.com/ajh/article/10/10/1175/156506 by guest on 28 September 2021 normals HTN GribbinB,66 17 with HTN, 21 normals Compliance decreased but related to pressure Toto-MoulouoD,67 25 obese c/w 25 nonobese HTN Compliance greater in obese c/w nonobese MerillonA,68 12 with HTN, 13 normals Van MerodeC,69 16 with borderline HTN, 15 normals BenetosD,E,70 72 with HTN, 39 normals Effect of ACE inhibitors was hetergeneous PearsonC,38 25 with HTN, 53 normals ? HTN accelerates compliance changes of aging

Compliance measured in a variety of patient populations by a variety of techniques: A, invasive waveform analysis; B, invasive pulse wave velocity (PWV); C, ultrasound; D, noninvasive PWV; E, tonometry. tech, technique; ref, reference number; HTN, hypertensives; c/w, compared with; ACE, angiotensin converting enzyme; p/v, pressure/volume.

(LVH). Researchers have suggested that LVH may There is also increasing evidence that vessel stiff- indicate, or be the result of, an abnormality in arterial ness or compliance may be an important component compliance, separate and distinct from elevation in in determining risk in patients with hypertension and the BP, and the frequent observation of LVH in nor- atherosclerosis. The altered physiology in essential hy- motensive subjects and in patients with well-con- pertension is, at least in part, related to vasoactive trolled hypertension indirectly supports that view.53 substances such as norepinephrine and angiotensin Indeed, one of the major problems in dealing with (both of which have been experimentally shown to hypertension is its definition and thereby, its existence result in reversible artery stiffening) and endothelium- as a cardiovascular risk factor.1 The classical method derived substances such as EDRF and endothelin.56,57 of estimating intravascular pressure by manual sphyg- Recently, Franklin and Weber attempted to address momonometry, of course, determines only the highest the above concerns by suggesting that hypertensive and lowest pressure (and even the measurement of the cardiovascular risk is primarily related to the sum of lowest pressure by indirect means is controversial), three vascular abnormalities: increased arteriolar re- and it is now more and more apparent that analysis of sistance, increased large artery stiffness, and the effect the total blood pressure curve should be analyzed in of increased and early pulse wave reflection.11 Further detail. Certainly, both steady state and pulsatile phe- research has suggested that increased pulse pressure nomena need to be considered.1 and decreased diastolic pressure are superior to in- Currently, considerable controversy exists as to creased systolic pressure alone as predictors of risk, whether abnormalities in arterial compliance in hyper- particularly in isolated systolic hypertension.21–25 That tensive patients represents intrinsic change in the ar- is, in vascular systems with normal compliance, the terial wall or merely a reflection of pressure changes, reflected waves have an impact on the central arteries and whether the changes in compliance are located during diastole (there is coordinated ventricular-vas- primarily in the large vessels or the small ves- cular coupling), and this enhances coronary perfusion. sels.39,54–56 Several observations would suggest that However, with increasing vessel stiffness, there is decreased compliance is not solely a mechanical con- early wave reflection that, if early enough, could, dur- sequence of increased blood pressure.54 These obser- ing systole, collide with the central pulse during sys- vations include the finding that compliance is reduced tole. This increased systolic component of the pressure in patients with borderline hypertension where the wave can be measured as an ‘‘augmentation index.’’ slight elevation in pressure is unlikely to induce a Arnett et al recently summarized what is known epi- reduction in compliance, and that, in patients with demiologically regarding arterial stiffness as a cardio- established hypertension, compliance estimates are re- vascular risk factor by compiling the literature that duced to the same extent regardless of the degree of reported compliance measures recorded by a variety BP elevation.55,56 of techniques.58 For example, the Bogalusa Heart AJH–OCTOBER 1997–VOL. 10, NO. 10, PART 1 VASCULAR COMPLIANCE AND CARDIOVASCULAR RESEARCH 1183

Study, a cross-sectional analysis of 384 children (mean phenomenon. Sumimoto et al and Duchier et al found age 11 years), demonstrated that systolic blood pres- that nitroglycerin had a beneﬁcial effect in patients sure was related to arterial stiffness.59 Studies in with ISH and felt that this was because of the known adults that demonstrate a positive association of hy- greater effect of this substance on systolic compared to pertension and abnormal vessel compliance are listed diastolic pressure.75,76 in Table 2.54,60–70 However, all of these studies are Left Ventricular Hypertrophy A major complica- cross-sectional and when all studies on the association tion of established hypertension is LVH. The asso- of hypertension and compliance are reviewed in total- ciation between LVH and blood pressure elevation ity, the data is less consistent. Some of the inconsis- is unexpectedly low, suggesting that LVH is not tencies in the relationship of BP to compliance might totally a pressure-dependent phenomenon.77,78 The relate to recent evidence suggesting that in hyperten-

regression of LVH with treatment is also not neces- Downloaded from https://academic.oup.com/ajh/article/10/10/1175/156506 by guest on 28 September 2021 sion small vessel compliance is predominantly af- sarily pressure-dependent. Several studies have fected (except in isolated systolic hypertension, see evaluated the relationship of compliance to the pres- below) and there has been a general inability of most ence of LVH and have attempted to separate these techniques to assess both proximal and distal compli- effects from the level of BP. The studies are too few ance. For example, McVeigh et al compared the prox- 79,80 to draw any inferences. imal and distal compliance in 38 patients with established essential hypertension with 32 age-matched Antihypertensive Therapy The effects of antihyper- controls and found a 19% lower proximal (C1) and tensive drugs on the large arteries consist of two major 72% lower distal (C2) compliance.54 Heintz et al, using effects: the effect due to the blood pressure lowering intravascular ultrasound, found differences in some and the direct effect of the drug on the vascular wall. measures of compliance with increasing distance from In the past, all antihypertensive drugs have been the heart.71 In contrast, Roman et al64 and Gribbin et gauged by their effectiveness in reducing blood pres- al66 suggested that compliance abnormalities in hyper- sure and at times by their effect on systemic vascular tensive patients were the result of the increase in resistance. Overall, few antihypertensive agents have distending pressure rather than secondary to struc- been assessed relative to their effect on compliance tural changes in the arterial wall. In contrast, Heintz et and all are hampered by virtue of the fact that com- 81 al noted that even with normalization of the BP compliance is a pressure-related variable. Chau et al pliance abnormalities persisted.71 studied the effect of a variety of drugs on blood pressure and compliance and found that there was a dif- Isolated Systolic Hypertension The importance of ferential impact on those two variables.82 Safar re- vascular compliance in isolated systolic hypertension viewed the therapeutic trials in hypertension along (ISH, elevated systolic pressure with normal or de- with the effects on hemodynamic’s with various va- creased diastolic pressure) may be related to increas- sodilating drugs and concluded that it ‘‘is not sufﬁ- ing stiffness within the vascular system as aging oc- cient to reduce blood pressure in patients being curs). Beltran et al demonstrated reduced proximal treated for hypertension. It is also important to im- compliance (measured by intravascular pulse contour prove the status of the arterial wall.’’83 That is, vascu- analysis) in 19 patients with ISH compared to 29 pa- lar hypertrophy may occur in hypertension (some- tients with systolic/diastolic hypertension and 47 nor- what analogous to cardiac muscle hypertrophy) and motensives.72 In four other studies, reduced compli- vascular hypertrophy may not be entirely pressure- ance (studied by pulse contour analysis, ultrasound, dependent. Sano and Tarazi, for instance, found that, or noninvasive PWV) was observed in ISH compared in a rat model, despite equal blood pressure control, to normotensive subjects. Smulyan et al thought the wall thickness of resistance vessels regressed more reduced compliance was the result and not the cause with a captopril-hydrochlorothiazide treatment com- of ISH,73 whereas Dart et al observed that the reduced bination compared to treatment with hydralazine compliance correlated with an increase in LV mass.74 alone.84 This relationship between drug effects on Analysis of the relationship of blood pressure and pressure and on the vascular wall may be further arterial stiffness by Beltran et al, suggest that in ISH a complicated by race and gender differences. Arnett et 35% to 63% reduction in compliance produces a 12% al found, in a cross-sectional analysis from the ARIC to 18% rise in systolic and a 12% to 24% decrease in study, that, at the same level of BP and age, signiﬁcant diastolic pressure.72 That is, that although the compli- differences in arterial stiffness were present between ance of smaller arteries is similarly reduced in essen- users and nonusers of ␤-blocking agents and diuret- tial compared to isolated systolic hypertension, com- ics.85 These drug effects were seen with ␤-blockers in pliance in large vessels is reduced to a greater degree whites but not blacks; whereas diuretic use was asso- in ISH. Their data also suggest that the abnormality of ciated with lower stiffness in women compared to compliance in ISH may not merely be an age-related men. Using noninvasive pulse contour analysis, Mock 1184 GLASSER ET AL AJH–OCTOBER 1997–VOL. 10, NO. 10, PART 1

et al assessed the effects of felodipine separately from al, in experimental atherosclerosis in the rhesus mon- an isosorbide dintrate (ISDN)-hydralazine combina- key, demonstrated a decrease in aortic distensibility tion on large and small vessel compliance.86 They that occurred with the development of atherosclerosis found that the ISDN-hydralazine combination in- and a return towards normality with atherosclerosis creased both large and small vessel compliance regression.95 Dart et al studied ultrasound determined whereas felodipine increased large vessel but not aortic arch compliance in four groups: a normocholes- 86 small vessel compliance. Given the importance of terolemic symptom-free group, symptom-free patients adverse effects of arterial dysfunction on pulsatile with hypercholesterolemia, patients with coronary ar- flow, logical goals for the treatment of hypertension in tery disease, and post–heart-transplant patients.103 In the future might include the drug’s effect on large and all groups, distensibility fell with age, but the change small vessel compliance, pulse wave transmission and was greater in the group with known coronary disease Downloaded from https://academic.oup.com/ajh/article/10/10/1175/156506 by guest on 28 September 2021 reflection, and their effects on minimizing the cyclic and in the posttransplant group with atheroma. stress on the arterial wall. Another interesting aspect relating atherosclerosis Renal Hypertension LVH develops in a large pro- and abnormalities in compliance is provided by re- portion of patents with end-stage renal disease, but searchers who have demonstrated abnormalities in LVH is only loosely correlated with BP. London et al vessels free from atherosclerosis in myocardial infarc- found reduced aortic and large vessel compliance in tion patients,104,105 as well as in vessels not generally 92 hemodialysis patients when compared to 90 control associated with atherosclerosis, such as the digital ar- subjects.87 However, Marchais et al in 44 hemodialysis teries, or the smaller coronary arteries.106 This sug- patients, found that there was an alteration of wave gests that the atherogenic process somehow signals a reflection (ie, small vessel compliance) associated with widespread abnormality in the arterial circulation. the development of LVH.88 London et al have also However, other studies have observed that arterial demonstrated that angiotensin converting enzyme in- distensibility was less in areas with atherosclerosis hibition decreased left ventricular mass independently when compared to nonatherosclerotic areas.106 of its effect on BP.89 Barenbrock et al studied 20 normotensive renal transplant patients and compared Diabetes Mellitus Using a variety of techniques it them to age-, sex-, and BP-matched controls.90 They has been shown that compliance is reduced in diabetes 107–116 found a decrease in common carotid vessel compli- mellitus (Table 3). Woolam et al studied 52 dia- ance in the transplant patients that was related to the betics compared to 87 normal controls (ages 4 to 75 time that patients had been on dialysis. years) and found reduced compliance in the diabetics for all but the younger ages.110 These observations Atherosclerosis There are no prospective studies re- were supported by Lehmann et al,109 whereas others garding the longitudinal association between arterial have demonstrated that these changes in compliance stiffness and atherosclerosis. However, there are cross- are unrelated to wall thickness. McVeigh et al demon- sectional studies that show an association between strated that compliance was significantly reduced in atherosclerosis and abnormal compliance, although diabetic subjects, regardless of the presence or absence others have not shown this association.91–97 Using in- of physical complications of the disease.111 Although travascular ultrasound, alterations in vascular compliance within the coronary artery bed have been ob- the association of atherosclerosis is an attractive expla- served.98 Reduced compliance is related to the pres- nation for reduced compliance in diabetics, Oxlund et ence, size, and characteristics of the atherosclerotic al demonstrated that alterations in arterial compliance plaque. Invasive and noninvasive measures have were not correlated with the degree of atherosclerosis shown that the distensibility of peripheral vessels is and that diabetic patients develop alterations in arte- reduced in patients with coronary artery disease, al- rial connective tissue independent of the presence of 117 though the stimulus for structural or functional atherosclerosis. Others have observed that chronic changes in these vessels is unknown.99 Some studies elevations in plasma glucose have been shown to implicate endothelin in the process, and Heintz et al cause connective tissue damage. Airaksinen et al sug- did show an association of vessel stiffness with endo- gest that reduced compliance in diabetics might be thelin levels.100 It should be noted, however, that the related to nonenzymatic glycosylation of matrix pro- 118 effect of atherosclerosis on endothelial function is less teins (specifically collagen). On the basis of these controversial, and the effect of EDRF on pulsatile ar- observations, the ARIC Study performed a cross-sec- terial function is plausible.101 Separating the effects of tional analysis of the prediabetic state by assessing the aging and the effects of atherosclerosis on compliance relationship of arterial stiffness indices with glucose is problematic. However, Blankenhorn and Kramsch tolerance and serum insulin concentrations in 4701 demonstrated the coexistence of increased arterial men and women.119 Their findings were compatible stiffness and atherosis of the carotid artery.102 Farrar et with the view that the prediabetic state is associated AJH–OCTOBER 1997–VOL. 10, NO. 10, PART 1 VASCULAR COMPLIANCE AND CARDIOVASCULAR RESEARCH 1185

TABLE 3. STUDIES EXAMINING DIABETICS Studytech,ref No. of Subjects Comments

LoD,107 10 Diabetics, 10 normals Decreased compliance in diabetics ScarpelloD,108 43 Diabetics, 11 normals Decreased compliance in lower but not upper limbs LehmannD,109 25 Type 1, 25 type 2 diabetics, Compliance in normals was greater than than in type 1 60 normals which was greater than type 2 diabetics WoolamB,110 52 Diabetics, 87 normals Decreased compliance in diabetics (except for the young) McVeighA,111 28 Diabetics, 22 normals Decreased distal but not proximal compliance PillsburyB,112 39 Diabetics, 27 normals Decreased compliance in diabetics MegnienD,113 9 Diabetics, 9 HTN, 11 normals Decreased compliance in diabetics and HTN ChristensenC,114 19 Insulin-dependent diabetics Decreased compliance correlated with vessel wall thickness Downloaded from https://academic.oup.com/ajh/article/10/10/1175/156506 by guest on 28 September 2021 ChristensenC,115 47 Insulin-dependent diabetics, Decreased compliance correlated with diabetes duration 24 normals ThordarsonC,116 52 Diabetics, 26 normals Decreased aortic compliance correlated with diabetes duration only in men

Compliance measured in a variety of patient populations by a varity of techniques: A, invasive pulse contour analysis; B, invasive pulse wave velocity (PWV); C, ultrasound; D, noninvasive PWV; E, tonometry. tech, technique; ref, reference number; HTN, hypertensives. with increased arterial stiffness and that the stiffness tered compliance may help explain inconsistencies in was unrelated to vessel wall thickness. the natural history of disease processes. In addition, if abnormal compliance occurs before clinical manifesta- Congestive Heart Failure The characteristics of the tions of disease are apparent, it could be important in peripheral vasculature are important determinants of identifying patients at risk, thereby allowing for ear- arterial pressure and left ventricular performance in lier and more cost-effective preventative therapy. That patients with congestive heart failure (CHF). Finkel- is, early identification of vascular abnormalities may stein et al studied the effects of CHF on arterial com- offer a means to anticipate, and possibly avert, clinical pliance in 46 heart failure patients compared to 48 events. For most of the cardiovascular risk factors, healthy control subjects, all age 19 to 75 years.120 They particularly when the risk factor is only mildly abnor- demonstrated that large vessel compliance was un- mal, an important clinical problem presents itself, that changed by CHF, but small vessels were less compli- is, whether there is a cost/effect benefit in treating all ant than the controls even after age adjustment. Gian- nattasio et al demonstrated the beneficial effects of patients with a risk factor knowing that the benefit angiotensin converting enzyme inhibitor therapy on will be to only a few. If there were a way of further compliance in patients with CHF.121 Separating the stratifying increased risk in a subset of this larger effects of the underlying disease process from the group of patients with a given risk factor, the cost/ pathophysiologic changes that are a result of heart effect benefit would be improved. Abnormal vascular failure is problematic. compliance may represent a way of improving risk stratification. SUMMARY AND CONCLUSIONS REFERENCES Arterial compliance decreases with age and is also reduced in hypertension, diabetes mellitus, and ath- 1. Safar ME, O’Rourke MF: The Arterial System in Hy- pertension. Kluwer Academic Publishers, Dordecht– erosclerosis. Understanding the concepts that revolve Boston–London, 1993. around compliance is complex, particularly to clini- 2. Avolio A: Genetic and environmental factors in the cians who are not routinely exposed to the physics function and structure of the arterial wall. Hyperten- and mathematics necessary to fully appreciate its nu- sion 1995;26:34–37. ances. Limiting the study of compliance is the lack of 3. Benetos A, Topouchian J, Ricard S, et al: Influence of a gold standard for its measurement and the limita- angiotensin II type 1 receptor polymorphism on aortic tions of invasive and noninvasive techniques for its stiffness in never-treated hypertensive patients. Hy- estimation. Nonetheless, the association of abnormal- pertension 1995;26:44–47. ities in compliance with a number of disease processes 4. McDonald DA: Blood Flow in Arteries: Theoretic, Ex- that have blood vessel wall damage as their common perimental, and Clinical Principles, 3rd ed. Edward denominator have been studied. Whether abnormal Arnold Publishers, London, 1990. compliance is the result of a disease process or a 5. O’Rourke MF: Mechanical principles in arterial dis- primary abnormality of vascular wall integrity, al- ease. Hypertension 1995;26:2–9. 1186 GLASSER ET AL AJH–OCTOBER 1997–VOL. 10, NO. 10, PART 1

6. Glasser SP, Selwyn A, Ganz P: Atherosclerosis, risk intracranial hemmorhage. Circulation 1994;90:1657– factors, and the vascular endothelium. Am Heart J 1661. 1996;131:379–384. 24. Fang J, Madhaven S, Cohen H, Alderman MH: Mea- 7. Levy BI, Benessiano J, Poitevin P, Safer ME: Endothe- sures of blood pressure and myocardial infarction in lium dependent mechanical properties of the carotid treated hypertensive patients. J Hypertens 1995;13: artery in WKY and SHR. Circ Res 1990;66:321–328. 413–419. 8. O’Rourke MF, Blozek JV, Moreels CL: Pressure wave 25. Madhaven S, Ooi WL, Cohen H, Alderman MH: Re- transmission along the human aorta. Changes in age lation of pulse pressure and blood pressure reduction and in arterial degenerative disease. Circ Res 1968;23: to the incidence of myocardial infarction. Hyperten- 567–579. sion 1994;23:395–401. 9. Johnson WTM, Salanga G, Lee W, et al: Arterial inti- 26. Kupari M, Hekali P, Keto P, et al: Relation of aortic mal embrittlement: a possible factor in atherogenesis. stiffness to factors modifying the risk of atherosclero- Downloaded from https://academic.oup.com/ajh/article/10/10/1175/156506 by guest on 28 September 2021 Atherosclerosis 1986;59:161–171. sis in healthy people. Arterioscler Thromb 1994;14: 386–394. 10. Lee RT, Kamm RD: Vascular mechanics for the cardi- ologist. J Am Coll Cardiol 1994;23:1289–1295. 27. Lehmann ED, Watts GF, Fatemi-Langroudi B, Gosling RG: Aortic compliance in young patients with het- 11. Franklin SS, Weber MA: Measuring hypertensive car- erozygous familial hypercholesterolemia. Clin Sci diovascular risk: the vascular overload concept. Am 1992;83:717–721. Heart J 1994;128:793–801. 28. Laogun AA, Gosling RG: In vivo arterial compliance 12. Finkelstein SM, Collins VR, Cohn JN: Arterial vascular in man. Clin Phys Phsiol Meas 1982;3:201–212. compliance: response to vasodilators by Fourier and pulse contour analysis. Hypertension 1988;12:380– 29. Keely R, Hayward C, Avolio A, O’Rourke M: Nonin- 387. vasive determination of age-related changes in the human arterial pulse. Circulation 1989:1652–1659. 13. Cohn JN, Finkelstein SM, McVeigh GE, et al: Nonin- vasive pulse wave analysis for the early detection of 30. Shimojo M, Tsuda N, Iwasaka T, Inada M: Age-related vascular disease. Hypertension 1995;26:503–508. changes in aortic elasticity determined by gated radionuclide angiography in patients with systemic hyper- 14. Safar ME, Simon AC, Levenson JA, Cazor JL: Hemo- tension or healed myocardial infarcts in normal sub- dynamic effects of diltiazem in hypertension. Circ Res jects. Am J Cardiol 1991;68:950–953. 1983;52(suppl I):169–173. 31. Benetos A, Laurent S, Hoeks AP, et al: Arterial alter- 15. Zobel LR, Finkelstein SM, Carlysle PF, Cohn JN: Pres- ations with aging and high blood pressure: a nonin- sure pulse contour analysis in determining the effect vasive study of carotid and femoral arteries. Arterio- of vasodilator drugs on hemodynamic impedance scler Thromb 1993;13:90–97. characteristics in dogs. Am Heart J 1980;100:81–88. 32. Avolio AP, Chen SG, Wang RP, et al: Effects of aging 16. Bank AJ, Wilson RF, Kubo SH, et al: Direct effects of on changing arterial compliance and left ventricular smooth muscle contraction on in vivo human brachial load in a northern Chinese urban community. Circu- artery elastic properties. Circulation 1995;77:1008– lation 1983;68:50–58. 1016. 33. Avolio AP, Fa-Quan D, Wei-Quiang L, et al: Effects of 17. Feske W, Finkelstein SM, Francis G, et al: Vascular aging on arterial distensibility in populations with compliance after nitroprusside in hypertension. IEEE high and low prevalence of hypertension: comparison Proc 1988;14:277–280. between urban and rural communities in China. Cir- 18. Smulyan H, Mookherjee MD, Warner RA: The effect culation 1985;71:202–210. of nitroglycerin on forearm arterial distensibility. Cir- 34. Pagani M, Mirsky I, Baig H, et al: Effects of age on culation 1986;73:1264–1269. aortic pressure-diameter and elastic stiffness-stress re- 19. Perret F, Mooser V, Hayoz D, et al: Evaluation of lationships in unanesthetized sheep. Circ Res 1979;44: arterial compliance-pressure curves: effect of antihy- 420–429. pertensive drugs. Hypertension 1991;18(suppl II): 35. Mohiaddin RH, Underwood SR, Bogren HG, et al: II77–II83. Regional aortic compliance studied by magnetic reso- 20. Ting C-T, Chen C-H, Chang M-S, Yin FCP: Short- and nance imaging: the effects of age, training and coro- long-term effects of antihypertensive drugs on arterial nary disease. Br Heart J 1989;62:90–96. reﬂection, compliance, and impedance. Hypertension 36. Vaitkevicius PV, Fleg JL, Engel JH, et al: Effects of age 1995;26:524–530. and aerobic capacity on arterial stiffness in healthy 21. Van Bortel LM, Kool MJ, Boudier HAS: Effects of adults. Circulation 1993;88:1456–1462. antihypertensive agents on local arterial distensibility 37. London GM, Gruerin AP, Pannier B, et al: Inﬂuence of and compliance. Hypertension 1995;26:531–534. sex on arterial hemodynamics and blood pressure. 22. Millar JA, Lever AF: Pulse pressure predicts coronary Hypertension 1995;26:514–519. but not cerebrovascular events in placebo-treated 38. Pearson AC, Ruiqiang G, Orsinelli DA, et al: Trans- male subjects in the MRC trial. J Hypertens 1996; esophageal echocardiographic assessment of the ef- 14(suppl 1):P897. fects of age, gender, and hypertension on thoracic 23. Selker HP, Beshansky JR, Schmid CH, et al: Presenting aortic wall size, thickness and stiffness. Am Heart J pulse pressure predicts thrombolytic therapy-related 1994;128:344–351. AJH–OCTOBER 1997–VOL. 10, NO. 10, PART 1 VASCULAR COMPLIANCE AND CARDIOVASCULAR RESEARCH 1187

39. Simon A, Levenson J: Use of arterial compliance for cause or secondary consequence of hypertension? Hy- the evaluation of hypertension. Am J Hypertens 1991; pertension 1991;18(suppl I):I52–I57. 4:97–105. 56. Smith DHG, Vaziri ND, Winer RL, et al: The relation- 40. Losordo, DW, Kearney M, Kim EA, et al: Variable ship of arterial compliance with endothelial-derived expression of the estrogen receptor in normal and proteins of the hemostatic system. Life Sci 1993;52: atherosclerotic arteries of premenopausal women. Cir- 1005–1011. culation 1994;69:1501–1510. 57. Dzau VJ, Safar ME: Large conduit arteries in hyper- 41. Williams JK, Adams MR, Herrington DM, Clarckson tension: role of the vascular renin-angiotensin system. TB: Short-term administration of estrogen and vascu- Circulation 1988;77:947–954. lar responses of atherosclerotic arteries. J Am Coll 58. Arnett DK, Evans GW, Riley WA: Arterial stiffness: a Cardiol 1992;20:452–457. new cardiovascular risk factor? Am J Epidemiol 1994; 42. Arnett DK, Riley W, Rautaharju P, et al: Electrocar- 140(8):657–664. Downloaded from https://academic.oup.com/ajh/article/10/10/1175/156506 by guest on 28 September 2021 diographic let ventricular mass and its association 59. Riley WA, Freedman DS, Higgs NA, et al: Decreased with arterial stiffness and menopause: the ARIC arterial elasticity associated with cardiovascular dis- study. Am J Cardiol 1994;74:247–252. ease risk factors in the young. Bogalusa Heart study. 43. Arnett DK, Ephross SA, Riley WA, et al: Is hormone Arteriosclerosis 1986;6:378–386. replacement among surgically of naturally post- 60. McVeigh GE, Burns DE, Finkelstein SM, et al: Re- menopausal women associated with lower arterial duced vascular compliance as a marker for essential stiffness? The ARIC study. Circulation 1992;82(suppl hypertension. Am J Hypertens 1991;4:245–251. II):II675. 61. Finkelstein SM, Cohn JN, Shelley W, et al: Vascular 44. Sonesson B, Hansen F, Stale H, Lanne T: Compliance impedance in health, hypertension and heart failure. and diameter in the human abdominal aorta—the in- Engineering Med Biol 1983;25:202. fluence of age and sex. Eur J Vasc Surg 1993;7:690– 697. 62. Feske W, Finkelstein SM, Francis G, Cohn JN: Arterial vascular compliance response to exercise in hyperten- 45. Celemajer DS, Sorensen KE, Georgakopoulos D, et al: sion. Biomed Sci Instrum 1988;24:161–165. Cigarette smoking is associated with dose-related and potentially reversible impairment of endothelium-de- 63. Liu Z, Ting CT, Zhu S, Yin FCP: Aortic compliance in pendent dilation in healthy young adults. Circulation human hypertension. Hypertension 1989;14:129–136. 1993;88(Part 1):2149–2155. 64. Roman MJ, Pini R, Pickering TG, et al: Noninvasive 46. McVeigh GE, LeMay L, Morgan DJ, Cohn JN: Ciga- measurements of arterial compliance in hypertensive rette smoking inhibits basal but not stimulated release compared to normotensive adults. J Hypertens 1992; of nitric oxide from the forearm vasculature. Am J 10(suppl 6):S115–S118. Cardiol 1996;78:668–672. 65. Ventura H, Messerli FH, Oigman W, et al: Impaired 47. McVeigh GE, LeMay L, Morgan DJ, et al: Influence of systemic arterial compliance in borderline hyperten- cigarette smoking on arterial compliance. Am J Med sion. Am Heart J 1984;108:132–136. 1997;102:227–231. 66. Gribbin B, Pickering TG, Sleight P: Arterial distensi- 48. Kool MJF, Hoeks APG, Boudier HAJ, et al: Short and bility in normal and hypertensive man. Clin Sci 1979; long term effects of smoking on arterial wall proper- 56:413–417. ties in habitual smokers. J Am Coll Cardiol 1993;22: 67. Toto-Moulouo JJ, Achimastos A, Asmar RG: Pulse 1881–1886. wave velocity in patients with obesity and hyperten- 49. Levenson J, Simon AC, Cambien FA, et al: Cigarette sion. Am Heart J 1986;112:136–140. smoking and hypertension: factors independently as- 68. Merillon JP, Fontenier GJ, Lerallut JF, et al: Aortic sociated with blood hyperviscosity and arterial rigid- input impedance in normal man and arterial hyper- ity. Arteriosclerosis 1987;7:572–577. tension: its modification during changes in aortic pressure. Cardiovasc Res 1982;16:646–656. 50. Cameron JD, Dart AM: Exercise training increases total systemic arterial compliance in humans. Am J 69. Van Merode T, Hick PJ, Hoeks AP, et al: Carotid Physiol 1994;266:H693–H701. artery wall properties in normotensive and borderline hypertensive subjects of various ages. Ultrasound 51. Wijnen JAG, Kuipers H, Kool MJF, et al: Vessel wall Med Biol 1988;14:563–569. properties of large arteries in trained and sedentary subjects. Basic Res Cardiol 1991;86:25–29. 70. Benetos A, Laurent S, Hoeks A, et al: Arterial alterations with aging and high blood pressure: a non- 52. Wijnen JAG, Kuipers H, Kool MJF, et al: Vessel wall invasive study of carotid and femoral arteries. Arte- properties of large arteries and endurance training. rioscler Thromb 1993;13:90–97. J Hypertens 1993;11(suppl 5):S88–S89. 71. Heintz B, Gillessen T, Walkenhorst F, et al: Evaluation 53. Randall OS, VanDenBos GC, Westerhof AN: Systemic of segmental elastic properties of the aorta in normo- compliance: does it play a role in essential hyperten- tensive and medically treated hypertensive patients sion? Cardiovasc Res 1984;18:455–462. by intravascular ultrasound. J Hypertens 1994;11: 54. McVeigh GE, Burns DE, Finkelstein SM, et al: Re- 1253–1258. duced vascular compliance as a marker for essential 72. Beltran A, Morgan DJ, Weber M, et al: Arterial com- hypertension. Am J Hypertens 1991;4:245–257. pliance abnormality in isolated systolic hypertension 55. Mulvany MJ: Are vascular abnormalities a primary (abst). Circulation 1994;90:I506. 1188 GLASSER ET AL AJH–OCTOBER 1997–VOL. 10, NO. 10, PART 1

73. Smulyan H, Vardan S, Griffiths A, et al: Forearm 92. Newman DL, Gosling RG, Bowden NLR: Changes in arterial distensibility in systolic hypertension. J Am aortic distensibility and area ratio with the develop- Coll Cardiol 1984;3:387–393. ment of atherosclerosis. Atherosclerosis 1973;14:231– 234. 74. Dart A, Silagy C, Dewar E, et al: Aortic distensibility and left ventricular structure and function in isolated 93. Farrar DJ, Green HD, Wagner WD, Bond MG: Reduc- systolic hypertension. Eur Heart J 1993;14:1465–1470. tion in pulse wave velocity and improvement of aortic 75. Sumimoto T, Hamada M, Kawakami H, et al: Effects distensibility accompanying regression of atheroscle- of glyceryl trinitrate on blood pressure and arterial rosis in the Rhesus monkey. Circ Res 1980;47:425–432. compliance. Angiology 1993;44:951–955. 94. Shankar R, Bond MG. Correlation of noninvasive ar- 76. Duchier J, Iannascoli F, Safar M: Antihypertensive terial compliance with anatomic pathology of athero- effect of sustained release Isosorbide dinitrate for iso- sclerotic nonhuman primates. Atherosclerosis 1990;85: lated systolic systemic hypertension in the elderly. 37–46. Downloaded from https://academic.oup.com/ajh/article/10/10/1175/156506 by guest on 28 September 2021 Am J Cardiol 1987;60:99–102. 95. Farrar DJ, Bond MG, Riley WA, Sawyer JK: Anatomic 77. Kannel WB, Dannenberg AL, Levy D: Population im- correlates of aortic pulse wave velocity and carotid plications of electrocardiographic left ventricular hy- artery elasticity during atherosclerosis progression pertrophy. Am J Cardiol 1987;60:851–856. and regression in monkeys. Circulation 1991;83:1754– 1763. 78. Bouthier JD, DeLuca N, Safer ME, Simon AC: Cardiac hypertrophy and arterial distensibility in essential hy- 96. Kogure S, Hayashi R, Sakaibori K, et al: Aortic pulse pertension. Am Heart J 1985;109:1345–1352. wave velocity in atherosclerotic diseases. Am J Med 1988;85:739–745. 79. Pannier B, Brunel P, El Aroussy W, et al: Pulse pressure and echocardiographic findings in essential hy- 97. Haira T, Susayama S, Kawasaki J, Yagi SJ: Stiffness of pertension. J Hypertens 1989;7:127–132. systemic arteries in patients with myocardial infarction. Circulation 1989;80:78–86. 80. Oparil S: Pathogenesis of ventricular hypertrophy. J Am Coll Cardiol 1985;5:57B–65B. 98. El-Tamimi H, Mansour M, Wargovich TJ, et al: Con- strictor and dilator responses in intracoronary acetyl- 81. Albaladejo P, Bouaziz H, Duriez M, et al: Angiotensin choline in adjacent segments of the same coronary converting enzyme inhibition prevents the increase in artery in patients with coronary disease. Circulation aortic collagen in rats. Hypertension 1994;23:74–82. 1994;89:45–51. 82. Chau NP, Simon A, Vilar J, et al: Active and passive 99. Anderson TJ, Gerhard MD, Meredith IT, et al: Sys- effects of antihypertensive drugs on large artery di- temic nature of endothelial dysfunction in atheroscle- ameter and elasticity in human essential hypertension. rosis. Am J Cardiol 1995;75:1B–5B. J Cardiovasc Pharm 1992;19:78–85. 100. Heintz B, Dorr R, Gillessen T, et al: Do arterial endo- 83. Safar M: Therapeutic trials and large arteries in hy- thelin 1 levels affect local arterial stiffness? Am Heart J pertension. Am Heart J 1988;115:702–710. 1993;126:987–989. 84. Sano T, Tarazi RC: Differential structural responses of 101. Rubanyi GM, Romero JC, Vanhoutte PM: Flow-in- small resistance vessels to antihypertensive therapy. duced release of endothelium derived relaxing factor. Circulation 1987;75:618–626. Am J Physiol 1986;250:H1145–H1149. 85. Arnett DK, Evans GW, Riley WA, et al: Differences in 102. Blankenhorn DH, Kramsch DM: Reversal of atherosis arterial elasticity by antihypertensive medication, gen- and sclerosis. The two components of atherosclerosis. der, and ethnicity. Circulation 1993;87(suppl II):II13. Circulation 1989;79:1–7. 86. Mock JE, Morgan DJ, et al: Contrasting effects of va- 103. Dart AM, LaCombe F, Yeoh JK, et al: Aortic distensi- sodilators on arterial compliance in borderline hyper- bility in patients with isolated hypercholesterolemia, tension. Circulation 1995;92(suppl I):68. coronary artery disease, or cardiac transplant. Lancet 87. London GM, Marchais SJ, Safar ME, et al: Aortic and 1991;338:270–273. large artery compliance in end-stage renal failure. 104. Shimogo M, Tsuda N, Iwasaka T, Inada M: Age re- Kidney Int 1990;37:137–142. lated changes in aortic elasticity determined by gated 88. Marchais SJ, Guerin AP, Pannier BM, et al: Wave radionuclide angiography in patients with systemic reflections and cardiac hypertrophy in chronic ure- hypertension or healed myocardial infarcts and in mia: Influence of body size. Hypertension 1993;22: normal subjects. Am J Cardiol 1991;68:950–953. 876–883. 105. Nakatani S, Yamagishi M, Tamai J, et al: Assessment 89. London G, Marchais SJ, Safar ME, et al: Aortic and of coronary artery distensibility by intravascular ul- large artery compliance in end-stage renal failure. trasound. Application of simultaneous measurements Kidney Int 1990;37:137–142. of luminal area and pressure. Circulation 1995;91: 2904–2910. 90. Barenbrock M, Spieker C, Laske V, et al: Effect of long term hemodialysis on arterial compliance in end-stage 106. Alfonso F, Macaya C, Goicolea J, et al: Determinants of renal failure. Nephron 1993;65:249–253. coronary compliance in patients with coronary artery disease: an intravascular ultrasound study. J Am Coll 91. Farrar DJ, Green HD, Bond MG, et al: Aortic pulse Cardiol 1994;23:879–884. wave velocity, elasticity, and composition in a nonhuman primate model of atherosclerosis. Circ Res 1978; 107. Lo CS, Relf IR, Myers KA, Wahlquist ML: Doppler 43:52–62. ultrasound recognition of preclinical changes in arte- AJH–OCTOBER 1997–VOL. 10, NO. 10, PART 1 VASCULAR COMPLIANCE AND CARDIOVASCULAR RESEARCH 1189

rial wall in diabetic subjects: compliance and pulse- minished arterial elasticity in diabetes: association wave damping. Diabetes Care 1986;9:27–31. with ﬂuorescent advance glycosylation end products 108. Scarpello JHB, Martin TRP, Ward JD: Ultrasound mea- in collagen. Cardiovasc Res 1993;27:942–945. surements of pulse wave velocity in the peripheral 119. Salomaa V, Riley W, Kark JD, et al: Non–insulin- arteries of diabetic subjects. Clin Sci 1980;58:53–57. dependent diabetes mellitus and fasting glucose and 109. Lehmann ED, Gosling RG, Sonksen PH: Arterial wall insulin concentrations are associated with arterial compliance in diabetes. Diabetes Med 1992;9:114–119. stiffness indexes, the ARIC study. Circulation 1995;91: 1432–1443. 110. Woolam GL, Schnur PL, Vallbona C, Hoff HE: The pulse wave velocity as an early indicator of athero- 120. Finkelstein SM, Cohn JN, Collins VR, et al: Vascular sclerosis in diabetic subjects. Circulation 1962;25:533– hemodynamic impedance in congestive heart failure. 539. Am J Cardiol 1985;55:423–427.

111. McVeigh G, Brennan G, Hayes R, et al: Vascular ab- 121. Giannattasio C, Mangoni AA, Carugo S, et al: Arterial Downloaded from https://academic.oup.com/ajh/article/10/10/1175/156506 by guest on 28 September 2021 normalities in non-insulin dependent diabetes melli- compliance in mild and severe congestive heart fail- tus identified by arterial waveform analysis. Am J ure (abst). Eur Heart J 1993;14(suppl):168. Med 1993;95:424–430. APPENDIX 112. Pillsbury HC, Hung W, Kyle MC, Freis ED: Arterial pulse waves and velocity and systolic time intervals in Compliance: The change in geometry (volume, diameter, diabetic subjects. Am Heart J 1974;87:783–790. area, etc) per change in pressure. It is a measure of the 113. Megnien JL, Simon A, Valensi P, et al: Comparative storage capacity of the system. Operative compliance refers effects of diabetes mellitus and hypertension on phys- to both the structural and functional components, eg, intrin- ical properties of human large arteries. J Am Coll sic changes in the vessel wall and blood pressure. Cardiol 1992;20:1562–1568. Distensibility: Percent change in area per change in pressure. 114. Christensen T, Nubauer B: Increased arterial wall stiff- Stiffness: A general term describing resistance to deforma- ness and thickness in medium-sized arteries in pa- tion. tients with insulin-dependent diabetes mellitus. Acta Elastic modulus: Change in stress for a given change in Radiol 1988;29:299–302. strain. This is often used as a measure of vessel stiffness. 115. Christensen T, Neubauer B: Arterial wall stiffness in Stress: Force per unit area. insulin-dependent diabetes mellitus: an in vivo study. Circumferential stress: Comparable to wall tension (Law Acta Radiol 1987;28:207–208. of LaPlace) 116. Thordarson H, Thorgeirsson G, Helgason T: Aortic Strain: Ratio of change in area to the initial area. stiffness in insulin-dependent diabetics: an echocar- Impedance: The total opposition to flow offered by the diographic study. Diabetes Med 1986;3:449–454. arterial system. Wall tension: The circumferential force per unit vessel 117. Oxlund H, Rasmussen LM, Andreassen TT, Heicken- length. dorf L: Increased aortic stiffness in patients with type C1 and C2: Terms used by some investigators to refer to I (insulin decedent) diabetes mellitus. Diabetologia proximal or large vessel compliance (C1) or distal, oscilla- 1989;32:748–752. tory, or small vessel compliance (C2), as borrowed from the 118. Airaksinen KE, Salmela PI, Linnaluoto MK, et al: Di- electrical modified Windkessel model.