| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Hypertension. 2007;49:1134.)
© 2007 American Heart Association, Inc.
Original Articles |
From the Institute of Child Health (J.P.J.H.), University College London, London, United Kingdom; the Division of Cardiovascular Medicine (K.R.A.N.), Henry Ford Hospital, Detroit, Mich; Cardiology Branch (G.Z.), National Heart Lung and Blood Institute, Bethesda, Md; and the Division of Cardiology (A.A.Q.), Emory University School of Medicine, Atlanta, Ga.
Correspondence to Julian P. J. Halcox, Vascular Physiology Unit, Institute of Child Health, 30 Guildford St, London WC1N 1EH, United Kingdom. E-mail j.halcox{at}ich.ucl.ac.uk
| Abstract |
|---|
|
|
|---|
Key Words: clinical science blood flow regulation endothelin endothelium atherosclerosis
| Introduction |
|---|
|
|
|---|
We hypothesized a differential contribution from endogenous ETA and ETB receptor activation in human coronary vasomotor regulation. Herein we report the first clinical study investigating the effect of selective ETB and combined ETA+B receptor antagonism on coronary vascular function in subjects with CAD and its risk factors.
| Methods |
|---|
|
|
|---|
Study Protocols
Study protocols were initiated after completion of diagnostic coronary angiography. A 7-French guide catheter was introduced into a coronary artery with insignificant stenosis (<20%). Blood flow velocity was measured using a 0.014-inch Doppler wire (FloWire, Volcano Corporation). Medications were infused via a 3-French infusion catheter, advanced over the FloWire. Infusion flow rates were similar in both protocols.
Protocol 1
In 25 patients, baseline coronary blood flow velocity and coronary angiography were performed after a 10-minute infusion of 5% dextrose at 2 mL/min. Endothelium-dependent coronary vasodilatation was estimated in 22 patients by administering incremental 2-minute infusions of acetylcholine (ACH) at 1.5, 15, and 50 µg/min (estimated intracoronary concentrations: 107, 106, and 3.3x106 mol/L, respectively). Subsequently, a 3-minute intracoronary infusion of sodium nitroprusside (SNP) at a dose of 20 µg/min, was administered to assess endothelium-independent coronary vasodilatation. BQ-788 (Bachem), a selective ETB receptor antagonist, was then infused for 1 hour at a rate of 100 nmol/min. To record the maximal response, measurements were made over 1 hour, based on previous observations in the forearm circulation.16 ACH and SNP infusions were then repeated during coadministration of BQ-788.
Systemic hemodynamics and coronary blood flow velocity were recorded, and coronary angiography was performed after each intervention and also after 30, 45, and 60 minutes of the BQ-788 infusion. In addition, pulmonary artery pressure, pulmonary capillary wedge pressure, and cardiac output were also measured using a Swan-Ganz catheter at baseline and during the 60-minute infusion of BQ-788.
In 12 patients in whom the native left anterior descending coronary artery was used as the study vessel, a 7-French gauge A2 catheter (Cordis Inc) was introduced percutaneously via the coronary sinus into the great cardiac vein. Arterial and coronary venous blood was drawn at baseline and after 60 minutes of BQ-788 infusion.
Protocol 2
This protocol was conducted in a similar fashion to protocol 1 in 14 patients with measurement of systemic, pulmonary, and coronary hemodynamics and endothelium-dependent and -independent coronary vasodilatation (with ACH at 1.5 and 15 µg/min for 2 minutes and SNP at 20 µg/min for 3 minutes, respectively) determined before and after a 1-hour combined intracoronary infusion of BQ-123 (Bachem) and BQ-788 (at doses of 200 nmol/min and 100 nmol/min, respectively).
Measurement of Coronary Blood Flow and Diameter
Coronary blood flow, vascular resistance, and epicardial coronary diameter were assessed during each intervention using Doppler flow velocity and quantitative coronary angiography as described previously.22,32
Measurement of Plasma Nitrogen Oxides, Big ET-1, and ET-1
Blood was drawn in EDTA tubes and immediately chilled on ice, centrifuged at 4°C at 2500 rpm for 10 minutes, and plasma was frozen at 70°C. Coronary venous plasma nitrogen oxide (NOx) levels were measured with the use of the Sievers Nitric Oxide Analyzer (model 280).33 Plasma levels of ET-1 and big ET-1 were measured using radioimmunoassay techniques.34,35
Statistical Analysis
Data are expressed as mean±SE. Differences between means were compared by paired or unpaired Students t test as appropriate. Dose response curves with ACH were compared by ANOVA using the SAS software 6.12 (SAS Institute). If the F value was significant, a Bonferroni multiple comparison test was performed. Univariate correlations were performed using the Pearsons correlation coefficient. Multiple stepwise regression analysis was performed to test whether the magnitude of change with BQ-788 or BQ-123+BQ-788 was related to the age, sex, presence of atherosclerosis, hypertension, diabetes, cigarette use, or the total, low-density lipoprotein, and high-density lipoprotein cholesterol levels (general linear models procedure). Subjects with a vasoconstrictor response in the epicardial vessels and/or in the lower 50% of the distribution for microvascular dilatation in response to ACH at a dose of 15 µg/min were considered as a subgroup with "worse endothelial function" for secondary analyses where appropriate (n=15 [63%] in protocol 1; n=8 [57%] in protocol 2). All of the P values are 2-tailed, and a value of <0.05 was considered of statistical significance.
| Results |
|---|
|
|
|---|
|
Protocol 2: Effects of Combined Administration of BQ-123+BQ-788
Systemic blood pressure and cardiac index were unchanged and heart rate increased by 7 bpm (P=0.003). Mild pulmonary vasodilatation was observed with BQ-123+BQ-788; after 1 hour, mean pulmonary artery pressure and pulmonary artery wedge pressure fell by 3.0 mm Hg (P<0.01), and 2.6 mm Hg (P<0.01), respectively (Table).
Resting Coronary Vascular Tone
Effect of BQ-788 on the Epicardial Arteries
No change in epicardial vascular tone was observed after BQ-788 (diameter [D] change: 0.6±1%; P=0.4; Figure 1). This lack of effect was noted in both mid- and distal coronary arterial segments and in subgroups with and without CAD.
|
Effect of BQ-788 on the Coronary Microcirculation
There was progressive coronary microvascular constriction with BQ-788; coronary blood flow fell by 13.1±3.8% (P=0.0003), and coronary vascular resistance (CVR) increased by 25.9±4.9% (P<0.0001) after 1 hour (Figure 1). We observed that the decrease in coronary blood flow (P=0.03) and increase in CVR (P=0.01) was apparent after 30 minutes of BQ-788 infusion, at which time blood pressure remained similar to baseline (P=0.2). There was no correlation between the presence of CAD or any of its risk factors with the magnitude of the microvascular response to BQ-788.
Effect of BQ-123+BQ-788 on the Epicardial Arteries
Significant vasodilatation of the epicardial coronary arteries was observed after 30 minutes and reached a maximum of 6.1±1.4% after 1 hour of BQ-123+BQ-788 (P<0.0001; Figure 2). There was no correlation between the magnitude of epicardial vasodilatation and the presence of atherosclerosis or its risk factors, and vasodilatation was similar in mid- and distal epicardial arteries.
|
Effect of BQ-123+BQ-788 on the Coronary Microcirculation
There was progressive coronary microvascular dilation with BQ-123+BQ-788; coronary blood flow increased by 24±8% (P=0.025), and CVR fell by 15±7% (P=0.022) after 1 hour (Figure 2). As with epicardial responses, the magnitude of microvascular vasodilatation did not correlate with the presence of atherosclerosis or its risk factors.
Endothelium-Dependent and -Independent Coronary Vascular Function
Effect of BQ-788 on the Epicardial Circulation
Epicardial coronary responses to ACH at baseline were heterogeneous with no net change in D (0.7±1.4%; P=0.36 at 15 µg/min of ACH). After BQ-788, a trend toward an improved epicardial vasodilator response to ACH (P=0.06 ANOVA) was observed (Figure 3). There was no difference in the response to SNP (P=0.19). To assess the effect of BQ-788 independent of both baseline changes in tone and any direct smooth muscle effects, we calculated the ratio of ACH:SNP responses in the epicardial coronary arteries before and after BQ-788. No significant difference in the epicardial ACH:SNP ratio was observed after BQ-788 in the whole group and in the subgroups with worse and better endothelial function. There was no correlation between the presence of CAD or its risk factors and the change in the response to ACH or SNP after BQ-788.
|
Effect of BQ-788 on Coronary Microvascular Function
At baseline, ACH infusion produced progressive microvascular dilation (see the data supplement). No difference in coronary blood flow with either ACH (93±13 versus 102±11 mL/min pre- versus post-BQ-788 at the 15-µg/min dose of ACH; P=0.1) or SNP (82±14 versus 77±11 mL/min pre- versus post-BQ-788; P=0.9) was observed after BQ-788. Also, absolute CVR after both ACH and SNP remained unchanged after BQ-788. In addition, to assess the effect of ETB receptor blockade independent of any direct smooth muscle effects of BQ-788, and to take into account the observed change in resting blood flow, we calculated the ratio of ACH:SNP responses in the coronary microcirculation before and after BQ-788. ACH:SNP resistance ratio was no different after BQ-788 in the whole group or in the subgroups defined by endothelial function.
Effect of BQ-123+BQ-788 on the Epicardial Circulation
Epicardial D was greater during ACH infusion after BQ-123+BQ-788 (1.83±0.09 versus 1.98±0.1 mm, with the 15-µg/min dose of ACH, pre- versus post-BQ-123+BQ-788; P<0.001). When expressed as the percentage of change in D, ACH-induced epicardial vasoconstriction at baseline (7.7±1.7%; P=0.008 with 15 µg/min ACH) was significantly attenuated (P=0.013 ANOVA) after BQ-123+BQ-788 (Figure 4). Epicardial responses to SNP were unchanged after BQ-123+BQ-788 (P=0.18). Furthermore, a significant improvement in the ratio of epicardial ACH:SNP was noted after BQ-123+BQ-788 (P=0.007 ANOVA), suggesting that combined ETA+B receptor blockade selectively improves epicardial endothelial function. The subgroup with worse endothelial dysfunction at baseline had significant improvement after ETA+B receptor blockade (13% to 5%; P=0.03), whereas in those with less severe dysfunction at baseline, the change did not reach significance (0.4% to 1%; P not significant).
|
Effect of BQ123+BQ-788 on Coronary Microvascular Function
After BQ-123+BQ-788, there was no difference in coronary blood flow with ACH (93±13 versus 102±11 mL/min pre- versus post-BQ-123+BQ-788 at the 15-µg/min dose of ACH; P=0.1) or the percentage increase in flow with ACH (85±23 versus 76±22% pre- versus post-BQ-123+BQ-788 at the 15-µg/min dose of ACH; P=0.7; please see the data supplement). Absolute flow and percentage increase in flow with SNP were also unchanged after BQ-123+BQ-788 (85±16 versus 88±16 mL/min [P=0.5] and 116±33 versus 84±23% [P=0.2], pre- versus post-BQ-123+BQ-788, respectively). Similarly, CVR, after both ACH and SNP, and the ACH:SNP resistance ratio remained unchanged after BQ-123+BQ-788.
Effect on Plasma Levels of Big ET-1, ET-1, and NOx in the Coronary Circulation
No correlation was observed between baseline levels of ET-1 or big ET-1 and the change in coronary blood flow in response to BQ-788, BQ123+BQ-788, or ACH. Big ET-1 levels were unaltered in arterial and coronary venous blood after either BQ-788 or BQ-123+BQ-788. Arterial ET-1 levels were unchanged (9.3±0.8 versus 9.1±1.0 pg/mL), but coronary venous ET-1 levels increased after BQ-788 (10.1±1.4 versus 12.4±1.0 pg/mL); although there was absence of an arteriovenous difference in ET-1 (0.8±1.9 pg/mL; P not significant) at baseline, after BQ-788, coronary venous ET-1 levels were higher (arteriovenous difference: 3.3±1.6 pg/mL; P=0.06). No changes in arterial or venous ET-1 levels were observed after BQ-123+BQ-788.
At baseline, coronary arterial and venous levels of NOx were similar (20.9±2.4 versus 21.2±2.4 µmol/L, arterial versus venous; P=0.6). After BQ-788, a significant trans-coronary arterial and venous difference in plasma NOx was observed (20.0±2.3 versus 18.8±1.2 µmol/L, arterial versus venous; P=0.01), predominantly because of the significant fall in coronary sinus NOx levels (P=0.002).
| Discussion |
|---|
|
|
|---|
Effect of ET Receptor Blockade on Basal Coronary Vascular Tone
Epicardial Coronary Circulation
Epicardial tone was unchanged after ETB receptor blockade, reflecting the sparse distribution of epicardial ETB receptors in conduit compared with resistance vessels.2,36,37 Notably, similar epicardial coronary dilation (6%) is seen with combined ETA+B and selective ETA receptor blockade using BQ-123 alone,22 indicating that basal epicardial coronary constriction is principally mediated via ETA receptors.
Coronary Microcirculation
Vasoconstriction is the predominant coronary microvascular effect of ETB receptor blockade. This response occurred before the increase in systemic blood pressure, implicating local rather than reflex systemic mechanisms. Possible explanations include reduced endothelial NO release and displacement of ET-1 from ETB receptors and/or ETB-mediated ET-1 clearance. The fall in coronary sinus NOx levels and the increase in coronary sinus levels of ET-1 after BQ-788 support both as potential mechanisms.11,13,38,39 In contrast, as with selective ETA receptor blockade,22 coronary microvascular dilatation is the predominant local physiological effect of combined ETA+B receptor blockade. In keeping with observations in the forearm microvessels of patients with CAD,15 the magnitude of coronary blood flow increase (mean 24%) with combined ETA+B receptor blockade in this study appears greater than that observed with selective ETA blockade (mean: 9% increase) in similar patients.22 This suggests a complementary effect of both ETB and ETA receptor blockades on microvascular smooth muscle. Although this suggests a lesser role for ETB-mediated endothelial NO release than direct ETA-mediated coronary microvascular constriction in our study, this may not be the case in healthy subjects with preserved endothelial NO bioavailability.16,17
Effect of ET Receptor Blockade on Coronary Vascular Endothelial Function
Selective ETB antagonism did not affect epicardial vessel endothelium-dependent or -independent function. Combined ETA+B receptor blockade improved epicardial endothelial dysfunction, consistent with our previous observations with selective ETA antagonism22 and in keeping with the low to absent epicardial coronary ETB receptor expression.2,37,39 These findings implicate ETA activation as an important mediator of epicardial coronary endothelial dysfunction. Reduced oxygen free-radical production with increased NO bioavailability may contribute to this effect.4042
In contrast, microvascular responses to ACH and SNP were unaffected by either BQ-788 alone or in combination with BQ-123. Because selective ETA receptor antagonism improves coronary and forearm microvascular endothelial dysfunction,16,22,26 it is likely that attenuated NO release with ETB blockade, as seen in animal studies,43 offsets the beneficial effect of ETA receptor antagonism on endothelial function. Alternatively, our cohort may be underpowered to demonstrate a smaller effect of dual antagonism on microvascular endothelial function.
Limitations
We cannot quantify the contribution of individual risk factors, and the impact of atherosclerosis burden (intravascular ultrasound) was not assessed. Although it is possible that the coronary vascular effects of ET receptor antagonists may differ somewhat according to these parameters, our cohort is representative of the general population with, or at increased risk, of developing CAD.
ET receptor antagonists may have different effects on coronary vascular function at different doses. Human coronary vascular physiology studies are technically challenging, time consuming, invasive, and, therefore, not without risk. Because a 60-minute infusion of each ET antagonist dose would be required, dose-ranging coronary studies cannot be justified. The concentrations of BQ-788 and BQ-123 were selected based on experience in the forearm circulation.16,17,44 At these doses, systemic effects on blood pressure and autonomic tone may have modulated coronary physiology and accounted, in part, for our findings.45,46 However, we observed an increase in coronary microvascular tone before the rise in systemic blood pressure during the BQ-788 infusion. Similarly, coronary vascular responses to ACH remained unchanged after adjustment for endothelium-independent vasomotor responses.
The fall in coronary NOx levels with associated increase in CVR after BQ-788 cannot be considered causal. Confirmation of this would require evaluation of responses to BQ-788 before and after coronary NO synthesis inhibition with NG-monomethyl-L-arginine, which is not practically or ethically feasible in this clinical setting.
Perspectives
This is the first clinical study investigating the contribution of ETB receptors to human coronary vascular tone and to endothelial function and both ETA+B receptors to coronary endothelial function in patients with CAD and/or risk factors. Our findings demonstrate that endogenous ET-1 contributes to the maintenance of basal coronary vasoconstrictor tone via ETA receptor activation in epicardial coronary vessels and by stimulation of both ETA and ETB receptors in the microcirculation. Of note, selective ETB receptor antagonism results in microvascular vasoconstriction because of increased ET-1 and decreased NO. Moreover, both selective ETA and combined ETA+B receptor blockade improve epicardial coronary endothelial dysfunction,22 but dysfunction of the microcirculation was only improved by selective ETA blockade. Figure 5 summarizes our observations of the coronary vascular effects of endogenous ET-1.22
|
Consistent with the conclusions of a recent comprehensive review of the literature, we found no incremental benefit with combined ETA+B antagonism in the coronary circulation, suggesting a potential therapeutic advantage for selective ETA receptor antagonism in atherosclerosis. In addition, selective ETB antagonism should be avoided.47 The long-term value of appropriate doses of selective oral ETA receptor antagonists in CAD warrants further exploration.
| Acknowledgments |
|---|
This work was funded by the National Heart Lung and Blood Institute Intramural Research Program.
Disclosures
A.A.Q. receives significant funding as a principal investigator, collaborator, or consultant from grants awarded by National Institutes of Health, Lilly, Amorcyte, Pfizer, Berlex, and the Woodruff Foundation. He is a modest recipient of speakers bureau appointments and honoraria from Pfizer, CV Therapeutics, and Wyeth and is a consultant for Amorcyte and Endothelix. The other authors report no conflicts.
Received November 9, 2006; first decision November 26, 2006; accepted February 20, 2007.
| References |
|---|
|
|
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Hypertension Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2007 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |