Author + information
- Received March 28, 2002
- Revision received June 13, 2002
- Accepted July 24, 2002
- Published online November 6, 2002.
- Laurent M Riou, PhD*,
- Mirta Ruiz, MD*,
- Jayson M Rieger, PhD*,
- Timothy L Macdonald, PhD*,
- Denny D Watson, PhD*,
- Joel Linden, PhD*,
- George A Beller, MD, FACC* and
- David K Glover, ME*,* ()
- ↵*Reprint requests and correspondence:
David K. Glover, ME, Cardiovascular Division, University of Virginia Health Sciences System, P.O. Box 800500, Charlottesville, Virginia 22908-0500, USA.
Objectives The study was done to determine the effects of propranolol, enalaprilat, verapamil, and caffeine on the vasodilatory properties of the adenosine A2A-receptor agonist ATL-146e (ATL).
Background ATL is a new adenosine A2A-receptor agonist proposed as a vasodilator for myocardial stress perfusion imaging. Beta-blockers, angiotensin-converting enzyme (ACE) inhibitors, and calcium blockers are commonly used for the treatment of coronary artery disease (CAD), and their effect on ATL-mediated vasodilation is unknown. Dietary intake of caffeine is also common.
Methods In 19 anesthetized, open-chest dogs, hemodynamic responses to bolus injections of ATL (1.0 μg/kg) and adenosine (60 μg/kg) were recorded before and after administration of propranolol (1.0 mg/kg, ATL only), enalaprilat (0.3 mg/kg, ATL only), caffeine (5.0 mg/kg, ATL only), and verapamil (0.2 mg/kg bolus, ATL and adenosine).
Results Neither propranolol nor enalaprilat attenuated the ATL-mediated vasodilation (225 ± 86% and 237 ± 67% increase, respectively, p = NS vs. control). Caffeine had an inhibitory effect (97 ± 28% increase, p < 0.05 vs. control). Verapamil blunted both ATL- and adenosine-induced vasodilation (63 ± 20% and 35 ± 7%, respectively, p < 0.05 vs. baseline), and also inhibited the vasodilation induced by the adenosine triphosphate-sensitive potassium (KATP) channel activator pinacidil.
Conclusions Beta-blockers and ACE inhibitors do not reduce the maximal coronary flow response to adenosine A2A-agonists, whereas verapamil attenuated this vasodilation through inhibition of KATPchannels. The inhibitory effect of verapamil and KATPchannel inhibitors like glybenclamide on pharmacologic stress using adenosine or adenosine A2A-receptor agonists should be evaluated in the clinical setting to determine their potential for reducing the sensitivity of CAD detection with perfusion imaging.
Pharmacologic stress using adenosine or dipyridamole is an alternative to exercise stress for the detection of coronary artery disease (CAD) with myocardial perfusion imaging. However, the frequent occurrence of undesirable side effects such as chest pain, headache, dyspnea, and atrioventricular conduction abnormalities (1–3)has stimulated the development of highly selective adenosine A2A-receptor agonists (4–6). These compounds induce an A2A-receptor–mediated coronary vasodilation without stimulation of the adenosine A1-, A2B-, and A3-receptors, thereby potentially eliminating the side effects. Recently, a new class of stable and highly potent and selective adenosine A2A-receptor agonists has been synthesized at the University of Virginia (7). Among them, the compound ATL-146e has shown promise as a vasodilator for myocardial perfusion imaging (8).
Beta-adrenergic receptor blockers, angiotensin-converting enzyme (ACE) inhibitors, and calcium blockers are beneficial in the treatment of CAD and are commonly prescribed (9–11). Additionally, dietary intake of caffeinated food or beverages is also common and has been shown to alter the pharmacological vasodilation induced by dipyridamole (12)due to the nonselective adenosine receptor antagonist property of caffeine (13). The kinetics of the ATL-146e–induced vasodilation in the absence of other pharmacologic agents have been previously characterized in detail (8). In the present study, we sought to investigate any pharmacological interactions between the hemodynamic effects of ATL-146e and a beta-adrenergic blocker (propranalol), an ACE inhibitor (enalaprilat), a calcium channel blocker (verapamil), and caffeine. Before ATL-146e can be introduced for clinical imaging, the influence of these commonly used drugs in patients with cardiovascular disease requires examination.
Nineteen fasted, adult mongrel dogs (20.9 ± 0.7 kg, range 16.8 to 28.2 kg) were anesthetized with sodium pentobarbital (30 mg/kg IV), tracheally intubated, and mechanically ventilated. Open-chest surgery and instrumentation were performed as previously described (8).
Protocol 1: effects of propranolol, enalaprilat, verapamil, and caffeine on hemodynamic responses to vasodilators
After baseline measurements, hemodynamic responses to boluses of adenosine (60 μg/kg), ATL-146e (1.0 μg/kg), and a second adenosine A2A-receptor agonist, CGS-21680 (2.0 μg/kg), were monitored. The effect of IV bolus injections of propranolol (1.0 mg/kg), enalaprilat (0.3 mg/kg), verapamil (0.2 mg/kg), and caffeine (5.0 mg/kg) on ATL-146e hemodynamic responses were monitored 10 min after each drug injection when hemodynamics had reached a steady state. Drug injection order was randomized, except for caffeine, which was always injected last because of its known nonselective inhibition of adenosine receptors (13). The effect of verapamil on hemodynamic responses to adenosine and CGS-21680 was also studied.
Protocol 2: effect of increasing verapamil doses on the hemodynamic responses to adenosine and ATL-146e
The hemodynamic responses to bolus injections of adenosine and ATL-146e were recorded in the absence or presence of varying infusion rates of verapamil (0.002, 0.004, and 0.02 mg/kg/min IV). In one dog, blood sampling was performed for determination of plasma verapamil concentration at each infusion rate. In two dogs, the effect of a 0.02 mg/kg/min infusion rate on the increase in regional myocardial blood flow mediated by an IV adenosine infusion (250 μg/kg/min) was determined with radioactive microspheres after placement of a critical stenosis on the left anterior descending coronary artery (LAD). A critical stenosis was defined as the point at which baseline flow was unchanged but the reactive hyperemic response to a brief LAD occlusion was completely abolished. The dose of adenosine used has been shown to produce maximal coronary vasodilation in our canine model (8).
protocol 3: mechanism for the verapamil-mediated inhibition of adenosine and ATL vasodilator action
(a) Effect of pressure drop reversal: after baseline recordings, hemodynamic responses to boluses of adenosine and ATL-146e were recorded. An IV infusion of verapamil (0.02 mg/kg/min) was started, and adenosine and ATL-146e responses were again recorded. An IV calcium infusion (3.0 mg/kg/min) was then begun and maintained until reversal of the verapamil-induced decrease in mean arterial pressure (MAP) was observed. Adenosine and ATL-146e hemodynamic responses were then recorded.
(b) Role of adenosine triphosphate-dependent potassium (KATP) channels: the hemodynamic response to an adenosine bolus was recorded before and after IV administration of the KATPchannel inhibitor glybenclamide (0.3 mg/kg), and the hemodynamic responses to intracoronary infusions of the KATPchannel activator pinacidil (1.0 μg/kg/min) were recorded in the absence or presence of verapamil (0.02 mg/kg/min IV).
Quantification of regional myocardial blood flow with radioactive microspheres
After euthanasia, hearts were removed and sliced into four rings from apex to base. Each slice was divided into six transmural sections, which were subdivided into epicardial, midwall, and endocardial segments. The resulting 72 myocardial tissue samples were counted for microspheres in a gamma-well scintillation counter (MINAXI 5500, PerkinElmer Life Sciences Inc., Downer’s Grove, Illinois). The window settings on the gamma counter were Sn-113, 340 to 440 keV; Sr-85, 450 to 580 keV; Nb-95, 640 to 840 keV; and Sc-46, 842 to 1300 keV.
In one dog in Protocol 2, 2 ml of arterial blood was withdrawn at baseline and during each verapamil infusion and centrifuged (1,900 rpm × 10 min at 4°C) for the assessment of plasma verapamil concentration (National Medical Service, Willow Grove, Pennsylvania).
When expressed in the text as a mean percent increase, coronary vasodilation was quantified as the ratio of peak ultrasonic LAD flow observed after vasodilator injection to baseline flow observed immediately prior to the vasodilator injection in each animal and the results averaged.
Results are presented as mean ± SEM. Computations were performed with SYSTAT software (SPSS Inc., Chicago, Illinois). Comparisons were performed using one-way analysis of variance and repeated-measures analysis of variance. When appropriate, post hoc testing was performed using the Bonferroni tests. The p values < 0.05 were considered statistically significant.
Protocol 1: effects of propranolol, enalaprilat, verapamil, and caffeine on hemodynamic responses to vasodilators
Results are shown in Table 1. At baseline, LAD flow increased by an average of 249 ± 44% following ATL-146e bolus injection (p < 0.05 vs. baseline). ATL-146e induced a slight decrease in MAP (p < 0.05 vs. baseline) and a significant increase in both heart rate (HR) and peak positive first derivative of left ventricular pressure with respect to time (dP/dt) (p < 0.05 vs. baseline). Left atrial pressure (LAP) was not affected by the adenosine A2A-receptor agonist.
Propranolol (1.0 mg/kg) significantly decreased resting LAD flow, MAP, HR, and dP/dt, and had no effect on LAP. In the presence of propranolol, ATL-146e induced a mean 225 ± 86% increase in LAD flow (p = NS vs. control) without decreasing MAP. The increases in HR and dP/dt observed with ATL-146e alone at baseline were abolished.
Enalaprilat (0.3 mg/kg) had no effect on resting LAD flow, dP/dt, and LAP. The MAP was significantly decreased and HR was slightly but significantly increased by the ACE inhibitor. In the presence of enalaprilat, ATL-146e induced a mean 237 ± 67% increase in LAD flow (p = NS vs. control) and significantly decreased MAP and increased dP/dt. However, enalaprilat abolished the increase in HR observed with ATL-146e alone at baseline.
Caffeine (5.0 mg/kg) had no effect on resting LAD flow, dP/dt, and LAP. A trend toward an increase in MAP, which did not reach statistical significance (p = 0.069), was observed, as was a small but significant increase in HR (p < 0.05). As expected, the increase in LAD flow with ATL-146e was attenuated in the presence of caffeine (97 ± 28%, p < 0.05 vs. control). ATL-146e produced no changes in MAP, HR, and LAP.
Verapamil (0.2 mg/kg) increased LAD resting flow by an average 46 ± 15% (p < 0.05), and significantly decreased MAP and HR. As expected, dP/dt decreased in response to the calcium blocker (p = 0.058). Unexpectedly, in the presence of verapamil, the increases in LAD flow produced by the adenosine A2A-receptor agonists ATL-146e and CGS-21680 were markedly blunted (63 ± 20% and 67 ± 8% increase, respectively; p < 0.05 vs. control) (Table 2). The hemodynamic responses to a bolus injection of adenosine in the absence and presence of verapamil are also presented in Table 2. As was observed with the adenosine A2A-receptor agonists, the mean 149 ± 34% increase in LAD flow produced by adenosine in the absence of verapamil was blunted at 35 ± 7% (p < 0.05 vs. control) in the presence of this calcium channel blocker.
Protocol 2: effect of increasing verapamil doses on the hemodynamic responses to vasodilators
Analysis of plasma verapamil concentration in one dog revealed a linear relationship between the verapamil dose and its plasma concentration (r2= 0.99, p < 0.001). Infusion rates of 0.002, 0.004, and 0.02 mg/kg/min resulted in verapamil plasma concentrations of 22, 46, and 230 ng/ml, respectively. Hemodynamic data are presented in Table 3and Figures 1 and 2. ⇓⇓As the verapamil infusion rate increased, baseline coronary flow tended to increase, and resting MAP progressively decreased (p < 0.01 vs. baseline at the highest verapamil infusion rate).
Coronary flow responses to adenosine and ATL-146e in the absence or presence of increasing doses of verapamil are depicted in Figures 1A and 1B. Note that with increasing verapamil infusion rates, a progressive inhibition of the coronary flow responses to adenosine and ATL-146e was observed. At the highest verapamil dose, peak coronary flow after ATL-146e or adenosine administration was significantly reduced (p < 0.05 vs. control), and coronary flow reserve (CFR) decreased from 4.4 ± 0.8 to 2.0 ± 0.1 with ATL-146e, and from 3.1 ± 0.1 to 1.8 ± 0.2 with adenosine (p < 0.05). As shown in Table 3, with increasing verapamil infusion rates there was also a progressive attenuation of the ATL-induced increases in dP/dt and HR that were completely abolished by the highest verapamil dose.
Absolute myocardial blood flow disparity
Shown in Figure 2are the absolute regional myocardial blood flows before and during IV administration of adenosine (250 μg/kg/min) in the absence and presence of a 0.02 mg/kg/min IV verapamil infusion. In the absence of verapamil, adenosine induced a fivefold increase in regional flow in the left circumflex coronary artery (LCx) zone, from 1.01 ± 0.01 ml/min/g to 5.05 ± 0.11 ml/min/g (p < 0.05), whereas flow in the LAD zone did not increase owing to the presence of a critical stenosis. During verapamil infusion, the increase in flow in the normal LCx zone was only 1.6-fold and did not reach statistical significance. Therefore, the marked fourfold LCx-to-LAD flow disparity with adenosine was severely attenuated in the presence of verapamil (1.8-fold, p < 0.05).
Protocol 3: mechanism for the verapamil-mediated inhibition of adenosine and ATL-146e vasodilator action
(a) Effect of pressure drop reversal: to determine whether the impaired adenosine- or ATL-146e–mediated vasodilation in the presence of verapamil was a nonspecific consequence of the calcium blocker-mediated decrease in MAP, we reversed the drop in MAP with an IV calcium infusion (3.0 mg/kg/min). As shown in Figures 3A and 3B, despite the recovery of MAP with calcium infusion, the vasodilator action of both adenosine (Fig. 3A) and ATL (Fig. 3B) remained blunted by the presence of verapamil (0.02 mg/kg/min).
(b) Role of KATPchannels: results are presented in Figure 4. The adenosine-mediated increase in flow (211 ± 49%) was totally abolished by the KATPchannel inhibitor glybenclamide. Moreover, the vasodilation induced by an intracoronary infusion of the KATPchannel activator pinacidil (134 ± 10%, p < 0.05 vs. baseline) was severely blunted in the presence of verapamil (22 ± 19%, p < 0.05 vs. pinacidil).
The new adenosine A2A-receptor agonist ATL-146e displays favorable properties for use as a vasodilator for myocardial stress perfusion imaging after an IV bolus injection (7,8). We sought to determine whether commonly used cardiac drugs such as a beta-adrenergic blocker, an ACE inhibitor, and a calcium blocker interfered with the pharmacologic effects of this novel A2A-agonist. We also evaluated the effect of caffeine, a nonselective adenosine receptor antagonist (13), on the hemodynamic response to a bolus injection of ATL-146e.
Effects of a beta-blocker, an ACE inhibitor, and caffeine on hemodynamic responses to ATL-146e
The control hemodynamic response to a bolus injection of ATL-146e observed in this study is identical to our previously published data (8). The 3.5-fold increase in coronary flow was accompanied by a slight decrease in MAP and by increases in both HR and dP/dt. These responses are direct consequences of adenosine A2A-receptor stimulation (14,15).
Effect of a beta-blocker
The dose of propranolol used in this study (1.0 mg/kg) was previously shown to produce beta-adrenergic receptor blockade in a canine model (16). We observed a significant decrease in MAP with propranolol administration, which has also been reported in patients after IV injection of a beta-blocker (17). The decreases in coronary flow, HR, and dP/dt observed in our study after propranolol injection have also been reported previously in anesthetized, open-chest dogs (18). In the present study, the ATL-146e–mediated increase in coronary flow was not affected by the presence of propranolol. An increase in maximal coronary flow after adenosine injection has been observed clinically after administration of metoprolol, a selective beta1-blocker (17,19), and has been attributed to a decrease in coronary vascular resistance during hyperemia. This in turn was attributed to a diminution of extravascular compressive forces in the presence of the beta1-blocker. However, Kern et al. (20)have shown that nonselective beta-blockers such as propranolol could potentiate coronary vasoconstriction after cold pressor testing in patients with CAD, leading to a reduced CFR. Such effects on coronary vascular resistance and CFR were not observed in healthy volunteers (20), which is consistent with our experimental data in normal dogs.
Effect of an ACE inhibitor
As shown by others in the same experimental model (21), the dose of enalaprilat (0.3 mg/kg) used in this study significantly reduced MAP. Baseline coronary flow was not affected, in accordance with previously published clinical data (22,23). We observed that the ATL-146e–mediated increase in coronary flow was not significantly affected by the presence of enalaprilat. Previous studies have shown that ACE inhibition may improve endothelium-dependent vasodilation both in healthy volunteers and in patients with atherosclerosis (22,23)through nitric oxide (NO)-mediated mechanisms. However, potentiation of vasodilation with ACE inhibition only occurs when the specific tissue affinity of the ACE inhibitor being used is favorable (24). Because of its low tissue affinity, enalaprilat has been shown previously to have no effect on radial artery vasodilation (24), in accordance with the results of the present study.
Effect of caffeine
The dose of caffeine used in this study (5.0 mg/kg) is equivalent to the serum level found in patients after consumption of two cups of brewed coffee 2 to 3 h before stress (12). As previously reported by Jain et al. (25)in anesthetized dogs, this dose of caffeine has little effect on baseline hemodynamic parameters. Caffeine significantly decreased ATL-induced vasodilation as a result of its nonselective antagonist effect on adenosine receptors (13). The reduction in CFR observed in our study in the presence of caffeine (from 3.5 ± 0.4 to 2.0 ± 0.3, p < 0.05) was similar to that observed clinically by Bottcher et al. (12)in 12 healthy volunteers after IV dipyridamole injection. These observations emphasize the need to screen patients for caffeine dietary intake prior to performing a stress test, because the drug could blunt vasodilator stress, thereby decreasing the sensitivity for detection of a significant coronary stenosis.
Effect of a calcium blocker
Verapamil bolus administration had a vasodilatory effect, as shown by an increase in coronary blood flow and a decrease in MAP, and a negative chronotropic effect. These effects have previously been observed in both experimental and clinical studies (26,27). Verapamil significantly decreased the maximal vasodilatory response to adenosine and ATL-146e. The increase in flow following another adenosine A2A-receptor agonist, CGS-21680, was also blunted by verapamil, indicating that the effect of the calcium blocker was not limited to a specific adenosine A2A-receptor agonist like ATL-146e. Although the attenuation in CFR may in part be due to an increase in baseline coronary flow after verapamil administration, we also observed a decrease in peak flow following vasodilation. As shown in Figures 1A and 1B, the decrease in peak coronary flow achieved after either adenosine or ATL-146e injection was dose-dependent. Moreover, assessment of absolute regional myocardial blood flow with radioactive microspheres showed that infusion of verapamil significantly reduced CFR. The three IV verapamil infusion rates (0.002, 0.004, and 0.02 mg/kg/min) evaluated in this study resulted in proportional drug plasma concentrations of 22, 46, and 230 ng/ml, respectively. Clinically, the mean verapamil plasma concentration is 120 ± 20 ng/ml in patients chronically treated with the drug (28). Therefore, the range of verapamil concentrations achieved in our study is likely to be encountered in patients.
Calcium infusion reversed the decrease in MAP induced by calcium channel inhibition with verapamil. However, the attenuation in ATL-146e– and adenosine-induced vasodilation remained. Therefore, the attenuation of the coronary flow response was not a secondary effect of calcium channel blockade on MAP by verapamil.
Adenosine and adenosine A2A-receptor agonists mediate vasodilation by stimulation of adenosine receptors on endothelial and smooth muscle cells (29), leading to activation of KATPchannels on these cell types. Subsequent activation of NO production in endothelial cells and hyperpolarization in smooth muscle cells lead to vasodilation (30). In our model, inhibition of KATPchannels with glybenclamide totally inhibited the adenosine-mediated vasodilation. In addition to inhibiting l-type voltage-dependent calcium channels, Haworth et al. (31)showed that verapamil also strongly inhibits KATPchannels in vitro. Our results confirm this finding in vivo, because verapamil inhibited the increase in flow produced by the KATPchannel activator pinacidil.
This study was performed in anesthetized animals. Sodium pentobarbital anesthesia is known to produce hemodynamic changes such as mild tachycardia and may also blunt reflex changes in HR and/or contractility. However, the effects of the drugs tested in our study on baseline hemodynamics were in accordance with previously published clinical and experimental data, suggesting that our model was appropriate for investigating the effects of these drugs on the vasodilatory properties of adenosine and adenosine A2A-receptor agonists.
The vasodilator properties of the adenosine A2A-receptor agonist ATL are not affected by the presence of the beta-adrenergic blocker propranolol, or the ACE inhibitor enalaprilat. As reported previously with dipyridamole, caffeine inhibits the increase in flow induced by an IV injection of ATL-146e due to its known nonselective inhibition of adenosine receptors. The important new finding in this study is that the calcium blocker verapamil inhibits the vasodilation produced by adenosine or adenosine A2A-receptor agonists through inhibition of KATPchannels in vivo. The inhibitory effect of verapamil and KATPchannel inhibitors like glybenclamide on pharmacologic stress using adenosine or adenosine A2A-receptor agonists should be evaluated in the clinical setting to determine their potential for reducing the sensitivity of CAD detection with perfusion imaging.
☆ Supported in part by a research grant from Adenosine Therapeutics, LLC. The adenosine A2Aagonist, ATL-146e, is owned by Adenosine Therapeutics, LLC. Drs. Macdonald, Linden, Beller, Glover, and the University of Virginia have a financial interest in the company.
- angiotensin-converting enzyme
- coronary artery disease
- coronary flow reserve
- peak positive first derivative of left ventricular pressure with respect to time
- heart rate
- adenosine triphosphate-sensitive potassium
- left anterior descending coronary artery
- left atrial pressure
- left circumflex coronary artery
- mean arterial pressure
- nitric oxide
- Received March 28, 2002.
- Revision received June 13, 2002.
- Accepted July 24, 2002.
- American College of Cardiology Foundation
- Abreu A.,
- Mahmarian J.J.,
- Nishimura S.,
- et al.
- Cerqueira M.D.,
- Verani M.S.,
- Schwaiger M.,
- et al.
- Webb R.L.,
- Sills M.A.,
- Chovan J.P.,
- et al.
- Glover D.K.,
- Ruiz M.,
- Takehana K.,
- et al.
- Rainwater J.,
- Steele P.,
- Kirch D.,
- LeFree M.,
- Jensen D.,
- Vogel R.
- Bottcher M.,
- Czernin J.,
- Sun K.T.,
- Phelps M.E.,
- Schelbert H.R.
- Monahan T.,
- Sawmiller D.,
- Fenton R.,
- Dobson J.
- Fujii A.M.,
- Vatner S.F.
- Billinger M.,
- Seiler C.,
- Fleisch M.,
- Eberli F.,
- Meier B.,
- Hess O.
- Bottcher M.,
- Czernin J.,
- Sun K.,
- Phelps M.,
- Schelbert H.
- Kern M.J.,
- Ganz P.,
- Horowitz J.D.,
- et al.
- Hornig B.,
- Kohler C.,
- Drexler H.
- Hornig B.,
- Arakawa N.,
- Haussmann D.,
- Drexler H.
- Hein T.W.,
- Belardinelli L.,
- Kuo L.
- Hein T.W.,
- Kuo L.
- Haworth R.A.,
- Goknur A.B.,
- Berkoff H.A.