Endothelial Control of Arterial Distensibility Is Impaired in Chronic Heart Failure
Background Vascular tone is a determinant of conduit artery distensibility. The aim of this study was to establish whether endothelium-derived relaxing factor (EDRF) influences the distensibility of conduit arteries and whether endothelium-mediated increases in distensibility are impaired in chronic heart failure (CHF).
Methods and Results Conduit artery distensibility was measured by two methods in healthy subjects and in nine patients with CHF caused by dilated cardiomyopathy. In the first method, pulse-wave velocity (PWV) was measured in the right common iliac artery at rest and during local infusions of acetylcholine (10−7 to 10−5 mol/L) or adenosine (2×10−7 to 2×10−5 mol/L), with correction for systemic effects. Acetylcholine induced concentration-dependent local reductions of PWV in healthy subjects (−5%, −15%, and −26%) but not in CHF patients (3%, 1% , −4%, P<.01), whereas adenosine induced similar reductions of PWV in healthy subjects and CHF patients. In the second method, brachial artery diameter, blood flow, and blood pressure were measured noninvasively by high-resolution ultrasound, continuous-wave Doppler, and photoplethysmography during reactive hyperemia in the hand and after sublingual glyceryl trinitrate (GTN, 400 μg). Hyperemic flow, similar in healthy subjects and CHF patients, was associated with increases in diameter and distensibility in healthy subjects (8.8% and 18.4%, respectively) but not in CHF patients (0.3% and −4.5%), whereas GTN induced similar effects in healthy subjects and CHF patients.
Conclusions These data indicate that conduit artery distensibility is increased by acetylcholine and increased blood flow in healthy subjects but not in CHF patients, whereas the effects of adenosine and GTN on distensibility are preserved in CHF patients. This implies that EDRF-mediated increases in distensibility are impaired in CHF patients, thus adding to cardiac work.
The elastic behavior of conduit arteries converts pulsatile cardiac ejection into nearly continuous tissue perfusion and reduces systolic pressure relative to flow, resulting in reduced workload relative to perfusion (“cardiovascular efficiency”). It may be characterized by distensibility (ΔV/VΔP, where V is luminal volume and P is transmural pressure) and is determined by structural components of the arterial wall, smooth muscle tone, and transmural pressure. Elastic behavior may thus change acutely with changes in smooth muscle tone or transmural pressure1 or chronically with changes in structure such as in aging,2 3 4 5 hypertension,6 7 and atherosclerosis.8 Specifically, it may be influenced by the activity of EDRF in response to flow or agonist stimulation, and it could be reduced in those conditions in which EDRF activity is impaired in conduit arteries.
CHF is characterized by increased peripheral vascular resistance, which may be due in part to impaired EDRF activity.9 Endothelium-dependent vasodilatation has been reported to be abnormal in heart failure; EDRF activity in response to acetylcholine is reduced in heart failure models in the femoral artery of dogs and rats10 11 12 and in the peripheral and coronary microcirculation in humans.9 13 14 15 Changes in radial artery diameter induced by acetylcholine16 and in brachial artery diameter induced by methacholine17 in patients with CHF have been reported to be normal, suggesting that impairment of EDRF secretion from conduit artery endothelium is not a prominent feature of CHF in humans.
The present studies were designed to establish whether the distensibility of a muscular conduit artery is influenced by EDRF activity in healthy subjects and whether EDRF-induced increases in arterial diameter and distensibility are diminished in patients with CHF. We studied patients with CHF caused by DCM to avoid the potentially confounding influences of impaired EDRF activity resulting from hypercholesterolemia, atheroma, or hypertension, which are commonly associated with CHF from other causes. Using two different methods, we studied iliac and brachial arteries and tested the effects of both agonist- and flow-stimulated increases in EDRF activity relative to endothelium-independent vasodilators.
Chronic Heart Failure
Nine patients (mean age, 52 years [range, 23 to 63 years]; 6 men, 3 women) with idiopathic DCM and CHF of New York Heart Association grade I (n=1), II (n=6), or III (n=2) were studied. These patients included all patients with DCM who underwent diagnostic catheterization in the University Hospital of Wales (Cardiff, UK) and who were available for and gave informed consent for this study during its course. All patients had fractional shortening measured by transthoracic echocardiography of <25%, radiological evidence of cardiomegaly, and normal coronary angiograms. All were on stable medical treatment for ≥2 months with diuretics and ACE inhibitors; 5 patients were on digoxin, and 1 patient took nitrovasodilators. ACE inhibitors, digoxin, and nitrovasodilators were withdrawn 48 hours before and diuretics 12 hours before the study.
Iliac artery study. Nine healthy subjects (mean age, 49 years [range, 37 to 58 years]; 4 men, 5 women) were studied after diagnostic catheterization for chest pain. All had a negative treadmill exercise test, no other evidence of cardiovascular disease, and a normal coronary angiogram. All medications (β-blockers in 3 patients, nitrates in 3 patients, and calcium antagonists in 5 patients) were withdrawn for >24 hours before investigation.
Brachial artery study. Fifteen healthy subjects (mean age, 51 years [range, 22 to 63 years]; 8 men, 7 women) were studied. None of the healthy subjects in either control group and no patients with CHF had a history of hypertension (blood pressure >160/90 mm Hg), hypercholesterolemia (cholesterol >7 mmol/L), or diabetes mellitus. Patients with valvular heart disease, previous cardiac surgery, peripheral arterial disease, or evidence of hematologic, hepatic, or renal dysfunction were excluded. All subjects refrained from the use of alcohol, caffeine, and cigarettes for >12 hours before the study.
All subjects gave written informed consent, and the protocol was approved by the ethical committee of the South Glamorgan Health Authority.
Iliac Artery Study: PWV
Distensibility may be derived from PWV, which is inversely related to the square root of distensibility.18 For ethical reasons, we took advantage of the opportunity offered by diagnostic catheterization to measure directly the pressure wave in the RCIA in this study, performed 2 to 4 hours after the diagnostic catheterization. Concentration-related, directly mediated local effects of an endothelium-dependent dilator (acetylcholine) and an endothelium-independent dilator (adenosine) were derived by normalizing for hemodynamic, flow-related, and reflex neurohumoral effects. This was achieved by infusing the same agents proximally and distally to the length of the RCIA over which PWV was measured and subtracting the effects of distal infusion from the effects of proximal infusion.
After an overnight fast and routine oral premedication with diazepam (10 mg), a right femoral artery puncture was performed under local anesthesia (1% lignocaine). A 6F end-hole catheter with a 0.46-mm lumen and two high-fidelity pressure transducers mounted 10 and 60 mm from its distal end (Gaeltec) was inserted through a 2.3-mm arterial sheath into the RCIA under radiographic screening. Small hand injections of contrast media (iohexol 300) were delivered through the lumen of the catheter to ensure that the catheter was positioned away from major arterial bifurcations and that subsequent infusions through the catheter lumen would remain in the ipsilateral RCIA. All subjects were given heparin (5000 U bolus and then 1000 U/h IV) for anticoagulation, and saline (150 mmol/L) was infused continuously through the catheter and sheath to prevent local thrombus formation. Subjects were subsequently studied at supine rest on a bed in a temperature-controlled environment (22°C to 24°C).
All measurements were derived from the signals obtained from the pressure transducers mounted 50 mm apart on the catheter in the RCIA. PWV was determined from the delay in the foot of the pressure wave arriving at the two transducers. The delay was measured in real time by use of a novel parallel processing algorithm implemented on a T800 30-MHz transputer housed within a personal computer.19 In brief, the computer performs multiple time shifts of the distal waveform relative to the proximal waveform in the region of the systolic edge. The sum of the moduli of the differences between discrete sections of the waveforms is calculated repeatedly, and the instance when this sum is at a minimum represents the time delay between the two waveforms. The accuracy of this system is largely dependent on the signal-to-noise ratio; with physiological pressure waveforms, the accuracy is estimated to be ≈50 microseconds.19 Mean blood pressure in the RCIA was calculated as the time integral of the pressure signal during each cardiac cycle, and heart rate was determined from the time between consecutive systolic pressure wave fronts.
After baseline recordings were taken with saline (150 mmol/L) infused through the catheter lumen into the proximal RCIA, adenosine was infused at 5 mL/min in incremental doses (10, 100, and 1000 μg/mL each for 3 minutes to give approximate local RCIA concentrations of 2×10−7, 2×10−6, and 2×10−5 mol/L, respectively), followed by washout with physiological saline for 5 minutes. After a return to steady state conditions, the baseline recordings were repeated and acetylcholine was infused similarly in incremental concentrations (3.6, 36, and 360 μg/mL each for 3 minutes to give approximate local concentrations of 10−7, 10−6, and 10−5 mol/L, respectively). Adenosine solutions were freshly prepared by diluting a sterile 25 mg/5 mL solution (South Glamorgan Pharmaceutical Services) with saline (150 mmol/L). Acetylcholine solutions were freshly prepared by diluting the reconstituted sterile powder (Miochol, Johnson & Johnson) with saline (150 mmol/L). Changes in concentration resulting from changes in flow during infusion of active agents were negligible relative to the 10- and 100-fold differences in drug concentrations. Measurements were recorded when steady state responses of PWV, blood pressure, and heart rate had been reached during the final minute of each infusion.
To control for flow-related and neurohumorally mediated changes in conduit artery distensibility after downstream microvascular dilatation and for reflex responses to any change in blood pressure, the protocol was repeated with the vasodilators infused through the arterial sheath >3 cm distal to the two pressure transducers. Thus, the segment of conduit artery being studied was not exposed to the distally infused vasodilator, although indirect effects were common to both proximal and distal infusions. The half-lives of adenosine and acetylcholine in vivo are <9.3 seconds20 and <1 millisecond,21 respectively, so the concentration of these agents recirculating from distal infusion to the segment of artery under study may be estimated as <0.15 of the original concentration (assuming a circulation time of 60 seconds22 ) and may be regarded as negligible for the purposes of this study. In addition, the directly mediated local effect of the drug was derived by subtracting the effects of the distal infusions from those of the proximal infusions, so that any recirculation effect would be negated.
Brachial Artery Study: Ultrasound Wall Tracking
Distensibility was also derived noninvasively from high-resolution ultrasound measurement of brachial artery diameter in relation to blood pressure. This method provides a measure of distensibility from beat to beat and over the range of distension induced by the pulse. It also gives data for blood flow and brachial artery vasodilatation in response to the dilator interventions used.
Studies were performed on fasting subjects in the morning after 15 minutes of supine rest in a temperature-controlled room (21°C to 23°C ), with the outstretched arm supported by a pneumatic cushion.
The high-resolution ultrasonic wall tracking system used in this study consisted of a specially adapted duplex color flow ultrasound machine (Diasonics Spectra) with a 7.5-MHz linear phased-array transducer. The brachial artery was identified with the ultrasound transducer. Anatomic landmarks (usually arterial bifurcations) were identified to allow repeated studies on the same section of artery. A standoff device containing ultrasound coupling gel was placed beneath the transducer to prevent compression of the anterior wall of the artery. The transducer was held in a stereotactic clamp, and a two-dimensional B-mode image of the brachial artery was obtained. The M-mode cursor was then positioned perpendicular to the vessel, and the horizontal distance between the cursor and the anatomic landmark was noted. With the ultrasound machine in M mode, the radiofrequency signals from the M-mode output were digitized and relayed to the wall tracking system. The sampling frequency of the radiofrequency signals was 1 kHz, and the total recording time was 10 seconds. On completion of data acquisition, the first radiofrequency signal was displayed on the computer screen. The operator marked the positions of the anterior and posterior vessel walls using sample volume markers. The wall tracking system tracked the vessel wall movements using the stored radiofrequency signals (axial resolution, ≈3 μm) to produce displacement waveforms of the anterior and posterior artery walls (Fig 1⇓), which allowed end-diastolic and end-systolic intraluminal diameters and thus the maximum change in diameter or distension for each beat to be determined.
Reproducibility, interoperator error, and repeatability of the method were validated in preliminary studies. Consecutive baseline measurements of arterial diameter made 30 minutes apart by the same operator in 22 subjects showed good correlation (r=.997). The mean coefficient of variation for baseline arterial diameter measurements in 20 subjects made by two different operators was 1.6%. In studies repeated 2 to 16 weeks apart by different operators in 22 subjects, the variance in diameter at baseline was 0.02 mm; the increase in diameter was 1.6% during reactive hyperemia and 2.3% after GTN administration.
Blood pressure was recorded noninvasively with photoplethysmography (Finapres) with a cuff on the middle finger of the arm being studied. Diastolic and systolic pressures measured in this study have been shown to correlate well with intrabrachial arterial pressure23 and intra-aortic pressure (M.W.R., J.G., personal observations, 1994).
Arterial distensibility (ΔV/VΔP, where V is volume and P is pressure) can be represented as ΔD/D/ΔP (where D is diameter) if cylindrical arterial conformation is assumed. Distensibility was derived from the systolic distension (difference between peak systolic and end-diastolic intravascular diameter, ΔD) and simultaneously recorded pulse pressure (ΔP) for each beat. Data represent the average during a 10-second recording.
Blood flow was measured with an 8-MHz continuous-wave Doppler probe mounted 60° over the brachial artery just distal to the 7.5-MHz imaging transducer. Brachial artery blood flow was calculated as the product of the mean spatial blood flow velocity corrected for the Doppler angle (SciMed Dopstation) and internal brachial artery diameter.
Measurements were made at baseline, during increased blood flow secondary to reactive hyperemia in the hand, and after administration of sublingual GTN (400 μg). Hyperemia was induced after inflation of a pediatric sphygmomanometer cuff at the wrist for 5 minutes (preliminary studies had shown that hyperemic blood flow remained similar [±5%] after cuff inflation ranging from 4 to 15 minutes). Blood pressure was measured continuously (except during wrist-cuff inflation), brachial artery blood flow was measured from −15 to 90 seconds, and internal brachial artery diameter was measured at 60 to 70 seconds after cuff release (preliminary studies had shown that the increase in diameter was maximal [±5%] from 45 to 90 seconds under all conditions studied). Return of all hemodynamic measurements to baseline was confirmed before each intervention. Data are presented at initial baseline, 60 seconds after cuff release, and 3 minutes after GTN administration.
Age, blood cholesterol, and resting heart rate, blood pressure, and PWV were analyzed with unpaired t tests. Sex distribution between control subjects and CHF patients was compared by use of Fisher’s exact probability test. For the iliac artery study, PWV, blood pressure, and heart rate recorded for each beat during the final minute of each infusion were averaged, and statistical significance was assessed by repeated-measures ANOVA24 with patient group, drug type and concentration, PWV, blood pressure, and heart rate as covariates, using the sas statistical software package (SAS Institute). For the brachial artery study, effects were related to the preceding baseline, and groups compared by an unpaired t test. A value of P<.05 was considered significant.
There were no significant differences in the age or sex distribution, serum cholesterol levels, or smoking habits between the CHF patients and the control groups of healthy subjects (Table 1⇓). Resting heart rate also was similar, although pressure tended to be lower in CHF patients than in the control group for the iliac artery study. There also were no significant differences between CHF patients and their corresponding healthy control groups in resting RCIA PWV or in resting brachial artery blood flow, systolic or diastolic diameters, distension, or distensibility.
Iliac Artery Study: PWV
Adenosine and acetylcholine infused into the proximal RCIA of healthy subjects significantly decreased PWV (increased distensibility) in a concentration-dependent manner (Table 2⇓, Fig 2⇓). Both agents caused concentration-dependent reductions in blood pressure and increases in heart rate. When infused into the distal RCIA, adenosine and acetylcholine caused concentration-dependent increases in PWV, but their effects on blood pressure and heart rate were not significantly different from those caused by the proximal infusions. Subtracting the results induced by the distal infusions from those induced by the proximal infusions gave the direct effects of adenosine and acetylcholine on RCIA PWV independent of flow-mediated and other indirect effects (Fig 3⇓). Both adenosine and acetylcholine caused significant direct concentration-dependent reductions in PWV (increases in distensibility) in these healthy subjects.
In contrast to results in healthy subjects, proximal infusion of adenosine and of acetylcholine did not significantly affect PWV (Table 2⇑, Fig 2⇑), although changes in blood pressure and heart rate were similar to those seen in healthy subjects. Distally infused adenosine and acetylcholine increased PWV dose dependently, with effects on blood pressure and heart rate that were similar to those caused by the proximal infusions. The derived direct local effect of these agents (Fig 3⇑) differed from that in healthy subjects in that acetylcholine-induced reductions in PWV were no longer present (P<.01, CHF patients versus healthy subjects), whereas the significant adenosine-induced reductions in PWV were retained.
The results in smokers and nonsmokers appeared to be the same, and exclusion of smokers did not influence the difference between CHF patients and healthy subjects.
Brachial Artery Study: Ultrasound Wall Tracking
Increased blood flow in the brachial artery in response to reactive hyperemia in the hand and sublingual administration of GTN significantly increased arterial diastolic and systolic diameters, distension, and distensibility, with no effect on heart rate or blood pressure (Table 3⇑, Fig 4⇓).
Blood flow, both during reactive hyperemia and after GTN administration, increased to a similar extent in CHF patients and healthy subjects (Table 3⇑, Fig 4⇑). Compared with healthy subjects, increased blood flow in the brachial artery did not increase arterial diameters, distension, or distensibility in CHF patients, whereas the responses to GTN administration were preserved.
The principal findings of this study are that the distensibility of a muscular conduit artery is increased by endothelium-mediated vasodilator stimuli in healthy subjects but not in CHF patients and that the increase in distensibility induced by the endothelium-independent stimuli, adenosine and GTN administration, is preserved in CHF. Endothelium-mediated vasodilator stimuli used in this study included acetylcholine, an agonist of EDRF release, and increased flow, shown in such studies to act by increasing EDRF release.25 26 Increased brachial artery flow was induced by reactive hyperemia of the hand after a brief period of cuff inflation distal to the brachial artery, thereby avoiding possible direct consequences of ischemia on the artery being studied caused by proximal cuff inflation.
Impairment of EDRF activity is a feature of hypercholesterolemia,27 hypertension,28 atherosclerosis,29 and diabetes mellitus,30 all of which were excluded in our patients. We therefore studied only patients whose heart failure was a result of DCM; unlike several previous investigators,9 14 15 we excluded those in whom heart failure was due to coronary artery disease. Smokers were not excluded, in the interest of obtaining adequate numbers of patients to investigate. Smoking history, however, was matched between healthy subjects and CHF patients; individual patient responses did not appear to differ between smokers and nonsmokers, and exclusion of smokers did not influence the differences between CHF patients and healthy subjects.
Distensibility of a conduit artery is influenced by its wall tension. For a simple thin-walled elastic tube, wall tension increases with diameter (by the Laplace law), but the relation between these variables is more complex in a muscular conduit artery. The wall has finite thickness, its individual constituents have different elastic properties, and resting arterial diameters vary regionally. In an artery wall containing elastin and collagen, load bearing shifts from elastin to collagen as the artery is distended. Because collagen is stiffer than elastin, elastic properties change nonlinearly with changes in arterial pressure and diameter. The relative contribution of smooth muscle to arterial elasticity will depend directly on its tone1 and on the extent to which wall tension is borne by passive elements over the range of distension being considered.
One method of assessing distensibility is to measure PWV, which is inversely related to the square root of distensibility (from the Moens-Koeteweg equation18 ). Accordingly, we measured the velocity of the pressure wave from the foot of the waveform because wave reflections from the previous cardiac cycle have generally ceased by the time of arrival of a new systolic wave front.31 The foot of the waveform is difficult to define precisely either visually or mathematically. We therefore developed a novel algorithm that allows the time delay between the foot of two pressure waveforms of similar morphology to be measured with high precision and in real time.19 The RCIA was chosen because it was easily accessible from a right femoral artery puncture (used as the standard approach for diagnostic coronary angiography). This invasive protocol allowed high-fidelity pressure measurements and thus the precise determination of PWV. The possibility that the presence of an in-dwelling catheter might have influenced PWV was minimized by ensuring that the pressure waveforms were of consistently high quality, implying that the catheter was positioned centrally in the blood flow stream throughout the study. The cross-sectional area of the catheter (3 mm2) was small relative to that of the RCIA (≈50 to 100 mm2), minimizing catheter-induced flow disturbances. Moreover, it is unlikely that the presence of a small, relatively incompressible catheter within the lumen would influence the distensibility of the artery wall.
To determine the direct local effects of intra-arterial infusions of acetylcholine and adenosine on the elastic properties of the RCIA, it was necessary to correct for indirect consequences of downstream resistance vessel dilatation. This was achieved by infusing the agents both proximally and distally to the length of the artery over which PWV was measured and subtracting the effects of distal infusion from those of proximal infusion. The results demonstrated that the direct local action of acetylcholine and adenosine on the RCIA of healthy subjects was to decrease PWV, reflecting an increase in distensibility by up to ≈75% with each agent. In CHF patients, by contrast, the direct local acetylcholine-induced reduction in PWV was lost and the adenosine-mediated effect was preserved, implying loss of EDRF-mediated increase in distensibility in CHF.
An alternative method of measuring conduit artery distensibility is to measure the luminal diameter and its distension (peak systolic minus end-diastolic diameter) during systolic pulsation by ultrasonic wall tracking, with simultaneous recording of blood pressure. The system developed for this study provides high-resolution diameter measurements (±3 μm)32 validated as giving repeatable and reproducible results and with the practical and ethical advantages of being noninvasive. The brachial artery was chosen as being readily accessible, with the distal wrist cuff used to induce reactive hyperemia. Both reactive hyperemia and GTN administration were associated with increases in flow in CHF patients and healthy subjects (and thus similar local flow signals within each group, although the magnitude of the increase observed with reactive hyperemia was always greater than that seen with GTN). The increase in brachial artery distensibility associated with an increase in flow in healthy subjects, however, was lost in CHF patients. This may have been due to an increase in passive stiffness manifest at larger diameters, but this possibility is effectively precluded by showing that the GTN-induced increase in distensibility occurred at larger intravascular diameters and over a greater systolic distension than those induced by an increase in flow. Loss of the flow-related increase in distensibility in CHF therefore appears to reflect loss of vasodilatation, consistent with impairment of EDRF activity. GTN-induced increases tended to be less in CHF patients (albeit not significantly with these relatively small numbers), which may reflect loss of the additional flow-related stimulus in CHF.
The study provides evidence that EDRF contributes to systemic conduit artery distensibility in healthy subjects when stimulated by either an agonist or increased flow. It also provides evidence that both agonist- and flow-induced EDRF-mediated increases in distensibility are impaired in CHF.
CHF was due to DCM in this study to ensure that other conditions known to be associated with impairment of EDRF activity were excluded, therefore suggesting that the impairment is probably the consequence of CHF per se, although the possibility that it is related to the underlying DCM cannot be ruled out on the evidence of this study. The findings, using two different systemic conduit arteries, two different methods, and two different stimuli of EDRF activity, are entirely consistent and thus mutually supportive, despite potential limitations in methods used clinically to measure distensibility. The loss of the EDRF-mediated increase in distensibility was associated with a loss of EDRF-mediated dilatation, implying greater vasoconstrictor tone as the predicted basis for the loss of distensibility.
This study does not directly address the question of basal EDRF activity, eg, by investigating the effects of nonmetabolized l-arginine analogues. The similar findings in the basal state provide no evidence for a difference in basal EDRF activity, but it remains possible that differences exist and are counterbalanced by compensatory mechanisms.
For the purpose of this study, it was fortuitous that the hyperemic increase in blood flow was similar in CHF patients and healthy subjects, thus providing the same flow signal with which to compare the local brachial artery response. This cannot, however, be taken as evidence that EDRF activity in resistance vessels is normal, for it is known that metabolic signals involved in hyperemic flow can override impairment of EDRF activity within the limits of dilator capacity.33
Flow-related increases in conduit artery distensibility will contribute to overall cardiovascular efficiency (defined in principle as cardiac work relative to tissue perfusion) when, for instance, cardiac output is increased during exercise; it will reduce relative cardiac workload by reducing systolic wall tension for a given stroke output, allowing increased stroke output under otherwise constant conditions and thus reducing relative myocardial energy consumption. Reduced arterial distensibility may be particularly deleterious when cardiac function is depressed, as in CHF.34 Moreover, when PWV is increased as a consequence of reduced distensibility, more rapid reflections of pressure waves to the heart will oppose left ventricular ejection, reducing stroke volume and increasing left ventricular pressure and myocardial oxygen consumption.35 Correction of endothelial dysfunction and increasing conduit artery distensibility may provide a novel means of improving impaired cardiac performance.
Selected Abbreviations and Acronyms
|CHF||=||chronic heart failure|
|EDRF||=||endothelium-derived relaxing factor|
|RCIA||=||right common iliac artery|
This study was supported by the British Heart Foundation (BHF). Dr Ramsey was a BHF junior research fellow, C.J.H. Jones was a BHF lecturer, and A.H. Henderson holds the BHF Sir Thomas Lewis Chair of Cardiology. We are grateful to Dr F. Dunstan, Department of Medical Statistics, University of Wales College of Medicine, for his assistance in the statistical analysis of the data and to J. Davies for her secretarial support.
- Received August 30, 1994.
- Revision received May 10, 1995.
- Accepted July 20, 1995.
- Copyright © 1995 by American Heart Association
Cox RH. Mechanics of canine iliac artery smooth muscle in vitro. Am J Physiol. 1976;230:462-470.
Learoyd BM, Taylor MG. Alterations of age in the viscoelastic properties of human arterial walls. Circ Res. 1966;18:278-292.
Avolio AP, Chen S, Wang R, Zhang C, Li M, O’Rourke MF. Effects of aging on changing arterial compliance and left ventricular load in a northern Chinese urban community. Circulation. 1983;68:50-58.
Kelly R, Hayward C, Avolio AP, O’Rourke MF. Noninvasive determination of age-related changes in the human arterial pulse. Circulation. 1989;80:1652-1659.
Vaitkevicius PV, Fleg JL, Engel JH, O’Conner FC, Wright JG, Lakatta LE. Effects of age and aerobic capacity on arterial stiffness in healthy adults. Circulation. 1993;88:1456-1462.
Ting CT, Brin KP, Lin SJ, Wang SP, Chang MS, Chiang BN, Yin FCP. Arterial hemodynamics in human hypertension. J Clin Invest. 1986;78:1462-1471.
Hirai T, Sasayama S, Kawasaki T, Yagi S. Stiffness of systemic arteries in patients with myocardial infarction: a noninvasive method to predict severity of coronary atherosclerosis. Circulation. 1989;80:78-86.
Kaiser L, Spickard RC, Olivier NB. Heart failure depresses endothelium-dependent responses in canine femoral artery. Am J Physiol. 1989;25:H962-H967.
Ontkean M, Gay R, Greenberg B. Diminished endothelium-derived relaxing factor activity in an experimental model of chronic heart failure. Circulation. 1991;69:1088-1096.
Lindsay DC, Jiang C, Brunotte F, Adamopoulos S, Coats AJS, Rajagopalan B, Poole-Wilson PA, Collins P. Impairment of endothelium dependent responses in a rat model of chronic heart failure: effects of an exercise training protocol. Cardiovasc Res. 1992;26:694-697.
Treasure CB, Vita JA, Cox DA, Fish RD, Gordon JB, Mudge GH, Colucci WS, St John Sutton MG, Selwyn AP, Alexander RW, Ganz P. Endothelium-dependent dilation of the coronary microvasculature is impaired in dilated cardiomyopathy. Circulation. 1990;81:772-779.
Katz SD, Schwarz M, Yuen J, LeJemtel TH. Impaired acetylcholine-mediated vasodilation in patients with congestive heart failure. Circulation. 1993;88:55-61.
Kubo SH, Rector TS, Bank AJ, Williams RE, Heifetz SM. Endothelium-dependent vasodilation is attenuated in patients with heart failure. Circulation. 1991;84:1589-1596.
Bank AJ, Rector TS, Tschumperlin LK, Kraemer MD, Kubo SH. Endothelium-dependent vasodilatation of conduit arteries is not impaired in patients with heart failure. Circulation. 1992;86:I-50. Abstract.
Caro CG, Pedley JG, Schroter RC, Seed WA. The Mechanics of the Circulation. Oxford, UK: Oxford University Press, 1978:281.
Goodman Gilman A, Rall TW, Nies AS, Taylor P. The Pharmacological Basis of Therapeutics. New York, NY: Pergamon Press; 1990:96.
Tortora GJ, Anagnostakos NP. Principles of Anatomy and Physiology. 7th ed. New York, NY: Harper and Row; 1993:630.
Imholz BPM, Weiling W, Langewouters GJ, van Montfrans GA. Continuous finger arterial pressure: utility in the cardiovascular laboratory. Clin Auton Res. 1991;1:45-53.
Crowder MJ, Hand DJ. Analysis of Repeated Measures. London, UK: Chapman and Hall; 1990:25-59.
Tagawa T, Imaizumi T, Endo T, Shiramoto M, Harasawa Y, Takeshita A. Role of nitric oxide in reactive hyperemia in human forearm vessels. Circulation. 1994;90:2285-2290.
Lieberman EH, Knab ST, Creager MA. Nitric oxide mediates the vasodilator response to flow in humans. Circulation. 1994;90:I-138. Abstract.
Zeiher AM, Drexler H, Wollschläger H, Just H. Modulation of coronary vasomotor tone in humans: progressive endothelial dysfunction with different early stages of coronary atherosclerosis. Circulation. 1991;83:391-401.
McDonald DA. Blood Flow in Arteries. London, UK: Edward Arnold; 1974:393.
Hoeks APG, Brands PJ, Reneman RS. Assessment of the arterial distension waveform using Doppler signal processing. J Hypertens. 1992;10:S19-S22.
Kostic MM, Schrader J. Role of nitric oxide in reactive hyperemia of the guinea pig heart. Circ Res. 1992;70:208-212.
Maruyama Y, Nishioka O, Nozaki E, Kinoshita H, Kyono H, Koiwa Y, Takishima T. Effects of arterial distensibility on left ventricular ejection in the depressed contractile state. Cardiovasc Res. 1993;27:182-187.
Nichols WW, MF O’Rourke. McDonald’s Blood Flow in Arteries: Theoretical, Experimental and Clinical Principles. London, UK: Edward Arnold; 1990:chap 9.