Endogenous Adenosine Is an Antiarrhythmic Agent
Background Adenosine administered intravenously terminates supraventricular tachycardias (SVT) involving the AV node as part of the reentrant circuit. Dipyridamole increases interstitial myocardial levels of this nucleoside. This study was designed to determine whether intravenous dipyridamole increases coronary sinus plasma adenosine concentrations ([Ado]cs) in humans to levels sufficient to alter electrophysiological parameters and terminate SVT.
Methods and Results A custom-designed catheter and syringe for sampling blood for measurement of [Ado]cs was placed in the coronary sinuses of 7 patients. [Ado]cs and refractory periods and conduction characteristics of the atrium and AV node were determined after autonomic blockade and dipyridamole infusion (5 μg · kg−1 · min−1 after a loading dose of 0.56 mg/kg). The atrial effective and functional refractory periods remained unchanged after dipyridamole infusion. In contrast, the AV nodal functional refractory period increased from 350±32 to 381±41 milliseconds (P=.03); the Wenckebach cycle length also increased from 309±47 to 350±57 milliseconds (P<.0001). Coincident with these changes, [ADO]cs increased from 0.18±0.11 to 0.31±0.12 μmol/L (P=.02). In another 10 patients with AV or AV nodal reentrant tachycardia, SVT was induced, and coronary sinus blood samples were drawn. Dipyridamole was infused, and coronary sinus blood samples were obtained after 15 minutes or coincident with termination of SVT. Mean [ADO]cs increased from 0.17±0.06 μmol/L during SVT to 0.38±0.21 μmol/L after dipyridamole (P=.02). Mean tachycardia cycle length increased from 334±132 to 375±139 milliseconds (P=.02); this effect was confined to the AV node, as demonstrated by an increase in AH interval from 171±144 to 214±140 milliseconds (P=.003). SVT terminated with the infusion of dipyridamole in 4 of the 10 patients.
Conclusions Administration of dipyridamole is associated with elevation of [ADO]cs, with coincident prolongation of the mean Wenckebach cycle length and AV nodal functional refractory period. During SVT, dipyridamole leads to prolongation of the AH interval and tachycardia cycle length and to an increase in [ADO]cs, with termination of SVT in four patients. These results support the hypothesis that adenosine may function as an endogenous antiarrhythmic metabolite.
Adenosine administered intravenously terminates supraventricular tachycardias (SVT) involving the AV node as part of the reentrant circuit.1 2 3 Dipyridamole, an inhibitor of cellular nucleoside transport, can potentiate the effects of exogenously administered adenosine and increase plasma4 and interstitial myocardial5 levels of the nucleoside. Animal studies have revealed that in the presence of dipyridamole, endogenously released adenosine significantly increases during rapid pacing to simulate tachycardia.6 Dipyridamole has also been shown to slow AV conduction in laboratory animals and humans, an effect antagonized by theophylline and abolished by adenosine deaminase.6 7 However, although the latter findings support the hypothesis that the electrophysiological effects of dipyridamole are due to elevation of endogenous adenosine concentrations, no direct measurements of coronary sinus adenosine concentrations in humans have been made yet to confirm this. This investigation was designed to demonstrate that in the presence of dipyridamole, coronary sinus adenosine concentration increases and that during rapid atrial pacing or SVT, adenosine levels rise further, depressing AV nodal conduction and potentially terminating adenosine-sensitive SVT.
Specifically, the goals of this investigation were (1) to determine the effects of dipyridamole on conduction and refractoriness in the human atrium and AV node under conditions of total autonomic blockade; (2) to determine the correlation of the electrophysiological effects of dipyridamole with changes in coronary sinus adenosine levels; (3) to determine whether dipyridamole, administered in the setting of an adenosine-sensitive tachycardia, raises myocardial adenosine levels sufficiently to terminate the tachycardia; and (4) to demonstrate a causal relation between elevated coronary sinus adenosine levels and modulation of AV nodal conduction by use of the adenosine receptor antagonist theophylline.
Seventeen patients were enrolled in this study. The protocols were conducted in patients having electrophysiological studies for standard clinical indications. Patients were considered for entry into the study if they had documented SVT or syncope. Criteria for exclusion from the study included congestive heart failure, coronary artery disease, second-degree AV block, hemodynamically unstable rhythms requiring cardioversion, contraindication to β-blockade, and concomitant use of dipyridamole or theophylline preparations.
All subjects gave informed consent for the study. The Institutional Review Board of the University of Florida and the Gainesville Veterans Administration Hospital Subcommittee on Human Studies approved the study protocols.
All procedures were performed on patients in a fasting, postabsorptive state. Antiarrhythmic drugs were discontinued at least five half-lives before electrophysiological study. Three quadripolar electrode catheters were introduced through a femoral vein and positioned in the high right atrium, in the right ventricular apex, and across the tricuspid valve to measure the His bundle electrogram. An additional sheath was placed in the right subclavian vein to gain access to the coronary sinus. Intracardiac electrograms were filtered at 30 to 500 Hz and simultaneously displayed with three ECG leads on a multichannel electrophysiological recorder (PPG Industry). In addition, the obtained data were stored on optical disks (Biomedical Instrumentation, Inc). Systemic arterial pressure was monitored continually through a femoral arterial catheter connected to standard pressure transducers. Stimulation was performed with a programmable stimulator and isolated constant-current source (Bloom Associates). Patients underwent a routine electrophysiological study, including incremental atrial and ventricular pacing, determination of atrial and ventricular refractory periods with the extrastimulus technique, and induction of tachycardia.
After completion of the clinical electrophysiological study, a custom-designed, dual-lumen catheter (100 cm long, 8F; Cordis Corp) was advanced through the subclavian sheath into the coronary sinus. This catheter, coupled to a specially designed double-barrel syringe, was used to obtain sequential samples of coronary sinus blood for measurement of adenosine levels. This catheter-syringe system was previously shown to be capable of sampling blood from the coronary sinus and detecting increases in plasma adenosine concentration.8 9
Protocol A: Effects of Dipyridamole on Conduction and Refractoriness of the Atrium and AV Node
In seven patients, the custom-designed catheter was placed in the coronary sinus, and then total autonomic blockade was achieved with intravenous atropine (0.04 mg/kg) and esmolol (loading dose of 500 μg/kg followed by a 50 μg · kg−1 · min−1 infusion). Baseline coronary sinus blood samples for measurement of plasma adenosine concentrations were collected before and after autonomic blockade, as were measurements of refractory periods and conduction characteristics of the atrium and AV node. The heart was then paced at a cycle length 10 milliseconds longer than that at which Wenckebach periodicity occurred to simulate tachycardia, and coronary sinus samples were again drawn. Dipyridamole (Dupont Radiopharmaceutical Division) was infused at 5 μg · kg−1 · min−1 after a loading dose of 0.56 mg/kg, and all measurements were repeated.
Protocol B: Effects of Dipyridamole on Sustained SVT
A second group of 10 patients was studied. This group had inducible, hemodynamically stable SVT involving the AV node as part of the reentrant circuit. Tachycardia was considered sustained if it required either administration of adenosine or pacing for termination. The patients had either AV nodal reentrant tachycardia (AVNRT) or AV reentrant tachycardia (AVRT). Electrophysiological parameters measured at baseline included the cycle length of the tachycardia and AH, HV, and VA intervals. Coronary sinus blood samples for measurement of plasma adenosine concentration were obtained in 8 patients, 4 with AVNRT and 4 with AVRT, before and after each intervention. Baseline coronary sinus blood samples for measurement of adenosine concentration were drawn before the induction of tachycardia and after 30 seconds of sustained tachycardia, based on the assumption that plasma adenosine levels rise and reach steady state rapidly during tachycardia. Collection of each blood sample took approximately 15 seconds. To confirm that adenosine coronary sinus plasma levels did reach steady state by 30 seconds, coronary sinus blood samples were drawn at 30 seconds and 2.5 and 5 minutes after the initiation of the tachycardia in 4 patients. Tachycardia cycle length and AH, HV, and VA intervals were evaluated in all 10 patients. Dipyridamole was then given as a bolus injection of 0.56 mg/kg followed by a continuous infusion 5 μg · kg−1 · min−1. Electrophysiological measurements were repeated, and adenosine levels were drawn either 15 minutes after the bolus injection of dipyridamole or coincident with the termination of the tachycardia. In 5 patients, theophylline 3 to 5 mg/kg was administered 15 minutes after the bolus injection of dipyridamole, and all measured electrophysiological measurements were reexamined.
Measurement of Plasma Adenosine Concentrations
Because adenosine has an ultrashort half-life, its concentration in blood from the central circulation is difficult to determine accurately. We therefore used a catheter-syringe system developed specifically for sampling blood from the central circulation for determination of plasma adenosine concentrations.8 9 This system delivers a “stop solution” that stabilizes the adenosine content in blood close to the sampling site within the catheter. The composition of the stop solution, the processing of the blood samples, and the assay for adenosine were described in detail previously.8 9 With this catheter and the methods described in References 8 and 9, 75% to 85% of administered adenosine can be reliably recovered in coronary sinus blood samples.8
All measurements of AV nodal function (ie, atrial and AV nodal refractory periods, Wenckebach cycle lengths, AH intervals) and coronary sinus blood adenosine levels are reported as mean±SD. Student’s paired t test was used for comparisons between control and infusion with dipyridamole and/or theophylline.
Effects of Dipyridamole on Conduction and Refractoriness in the Atrium and AV Node
Seven patients were studied with this protocol. One patient had inducible AVNRT; another had inducible AVRT. A third patient manifested only inducible atrial fibrillation when studied. Four patients had a history of syncope but a normal electrophysiological study. All patients had normal ejection fractions and no evidence of coronary artery disease.
Table 1⇓ summarizes the electrophysiological parameters before and after dipyridamole. All measurements were made after total autonomic blockade. The atrial effective and functional refractory periods remained unchanged after dipyridamole infusion. In contrast, the AV nodal functional refractory period increased after dipyridamole infusion from 350±32 to 381±41 milliseconds (P=.03). In all seven patients, because atrial refractoriness was reached before the AV nodal effective refractory period, the latter could not be determined. The mean Wenckebach cycle length was significantly prolonged after dipyridamole infusion from 309±47 to 350±57 milliseconds (P<.0001) (Table 1⇓). After these baseline measurements, tachycardia was then simulated by pacing the right atrium at a cycle length 10 milliseconds longer than the Wenckebach cycle length. Coronary sinus adenosine concentration during rapid atrial pacing increased from a mean of 0.18±0.11 μmol/L before to 0.31±0.12 μmol/L after dipyridamole infusion (P=.02) (Fig 1⇓).
Effects of Dipyridamole on SVT Involving the AV Node
Ten patients with AVNRT or AVRT were given an infusion of dipyridamole during sustained tachycardia. In 4 of the 10 patients (40%), 2 with AVNRT and 2 with AVRT, the tachycardia terminated within 383±328 seconds of the initiation of dipyridamole infusion. In the remaining 6 patients, tachycardia did not terminate during dipyridamole infusion. However, the mean tachycardia cycle length in these 10 patients increased from 325±86 to 348±90 milliseconds (P=.02) (Fig 2⇓). The greatest increase in tachycardia cycle length was seen in those patients whose tachycardias terminated. The tachycardia cycle length increased from 334±132 to 375±139 milliseconds (P=.02) in those patients whose tachycardias terminated after dipyridamole infusion versus 319±54 to 331±47 milliseconds in those patients whose tachycardias did not terminate. Prolongation of tachycardia cycle length was primarily confined to the AV node, as reflected by prolongation of the mean AH interval from 198±98 to 221±92 milliseconds (P=.03) (Fig 3⇓) and termination of the tachycardia at the level of the AV node (Fig 4⇓). HV intervals were unchanged during dipyridamole infusion.
Coronary sinus plasma concentrations of adenosine increased from 0.17±0.06 to 0.38±0.21 μmol/L (P=.02) (Fig 5⇓), coinciding with termination of the tachycardia in 4 of 10 patients and prolongation of the tachycardia cycle length in 5 of the remaining 6 patients. Although coronary sinus plasma concentrations of adenosine during tachycardia tended to be higher in the group whose tachycardias terminated (0.27±21 versus 0.13±0.10 μmol/L), this difference was not statistically significant. However, the mean increase in coronary sinus plasma concentrations of adenosine was significantly greater in those patients whose tachycardias terminated compared with those in whom it did not. Adenosine concentrations increased from 0.11±0.07 to 0.48±0.21 μmol/L (P=.02) in the group of patients whose tachycardias terminated (n=4) versus 0.12±0.05 to 0.20±0.13 μmol/L (P=NS) in the group whose tachycardias did not (n=4) (Fig 6⇓). Fig 7⇓ shows the individual changes in adenosine concentrations.
Effect of Theophylline on Electrophysiological Parameters and Adenosine Concentration in Patients Receiving Dipyridamole
In all five patients with persistent tachycardia who received theophylline after the dipyridamole infusion was completed, AH intervals and tachycardia cycle length returned to baseline. HV and VA intervals did not change with either dipyridamole or theophylline infusion (Table 2⇓). Adenosine concentrations did not change significantly, as shown by a comparison of values obtained after dipyridamole infusion and after theophylline infusion (0.14±0.08 versus 0.33±0.13 μmol/L, P=NS).
Time Course of Change in Coronary Sinus Plasma of Adenosine Concentration During SVT
Before dipyridamole infusion, coronary sinus adenosine concentrations measured at 0.5, 2.5, and 5 minutes after initiation of SVT were not significantly different (Fig 8⇓). Adenosine concentration before tachycardia was 0.11±0.05 μmol/L and increased during tachycardia to 0.19±0.11, 0.16±0.04, and 0.22±0.09 μmol/L at 0.5, 2.5, and 5 minutes (P=NS), respectively.
The unique finding of this study is objective demonstration that coronary sinus adenosine concentrations are elevated during rapid atrial pacing and SVT when the metabolism and uptake of adenosine are inhibited by dipyridamole. In fact, in patients in whom the largest increases in coronary sinus adenosine concentration occurred, the tachycardia terminated. It had previously been hypothesized that endogenous adenosine could modulate AV nodal conduction in the presence of dipyridamole. This conclusion was based solely on the reversal of the electrophysiological effects of dipyridamole by theophylline, without direct measurement of coronary sinus adenosine concentrations. The data presented in the present study are the first in humans to provide a direct association between the electrophysiological effects of dipyridamole and coronary sinus adenosine concentrations. Specifically, dipyridamole-induced prolongation of AV nodal refractoriness was associated with significant elevation of coronary sinus (ie, endogenous) adenosine levels. Consistent with this finding, dipyridamole terminated or prolonged the cycle length of adenosine-sensitive tachycardias involving the AV node. Theophylline, an adenosine antagonist, reversed the electrophysiological effects of dipyridamole, demonstrating a cause and effect relation between the elevated coronary sinus adenosine levels and modulation of AV nodal conduction.
Adenosine is an endogenous nucleoside that has multiple cardiovascular effects, including depression of AV nodal conduction.10 Until recently, it was believed that endogenous adenosine did not have significant physiological effects on the AV node except under conditions of hypoxia or ischemia.11 12 13 In fact, under normoxic, nonischemic conditions in guinea pig isolated perfused hearts, endogenously released adenosine did not reach concentrations sufficient to prolong AV nodal conduction time, regardless of the atrial pacing rate.6 In contrast, in the presence of dipyridamole, a nucleoside uptake inhibitor, endogenous adenosine levels increased, and this was associated with prolongation of AH intervals during atrial pacing.6 Of note, this effect was most pronounced at fast rates of pacing and led to AV block. In humans, endogenously produced adenosine can exert electrophysiological effects on the AV node if dipyridamole inhibits uptake and degradation. In the study of Lerman et al,7 the AH interval increased at a constant pacing cycle length, as did the cycle length at which Wenckebach periodicity occurred. These findings supported the hypothesis that in the presence of a nucleoside uptake inhibitor, depression of AV nodal conduction is due to increased endogenous levels of adenosine. However, Lerman et al7 did not demonstrate that coronary sinus plasma adenosine levels were elevated coincident with increased AV nodal conduction delay. Thus, the present study is the first to demonstrate a significant rise in coronary sinus plasma adenosine coincident with modulation of AV nodal conduction by dipyridamole.
Lerman et al7 also studied six patients with SVT that involved the AV node as part of the reentrant circuit. Although coronary sinus plasma concentration of adenosine was not determined, an increase in endogenous adenosine was proposed as the mechanism underlying the electrophysiological effects of dipyridamole. Again, the evidence presented to support this interpretation was the reversal of the electrophysiological effects of dipyridamole by theophylline.10 14 Regardless, the findings raised the possibility of using nucleoside transport inhibitors like dipyridamole to reveal the antiarrhythmic properties of endogenous adenosine. Hence, the present study advances this hypothesis by demonstrating that endogenous adenosine concentrations in humans are indeed elevated in the presence of dipyridamole during SVT. This suggests that the concentration of adenosine in the AV nodal region rises sufficiently to modulate AV nodal conduction.
The antiarrhythmic effects of dipyridamole were evaluated in 17 patients, 10 of whom had adenosine-sensitive tachycardias. The remaining 7 patients had a clinical indication for electrophysiological study. During fast rates of atrial pacing, dipyridamole infusion affects the AV node by increasing the AV nodal functional refractory period. In contrast, the atrial functional and effective refractory periods remained unchanged. Because the atrial effective refractory period was reached before the AV nodal effective refractory period in most cases, changes in AV nodal effective refractory periods could not be evaluated.
We found no evidence that endogenous adenosine concentrations increase significantly during tachycardia when the nucleoside uptake inhibitor dipyridamole is not present. This is consistent with previous work in laboratory animals in which adenosine concentrations did not rise sufficiently to modulate AV conduction except in the presence of a nucleoside uptake inhibitor.6
The 10 patients studied with tachycardias involving the AV node as part of the reentrant circuit were affected by dipyridamole in one of two ways: either the tachycardia terminated during dipyridamole infusion (40% of patients) or the cycle length prolonged significantly (all patients). Neither the etiology of the tachycardia nor the cycle length of the tachycardia before dipyridamole predicted the response to this nucleoside uptake blocker. Of note, the mean increase in coronary sinus plasma concentrations of adenosine was significantly greater in the patient group whose tachycardias terminated than in the group whose tachycardias did not (0.36±0.15 versus 0.08±0.12 μmol/L, P<.05). This latter finding provides further evidence to support our hypothesis that the delay in AV nodal conduction caused by dipyridamole is mediated by endogenous adenosine.
Theophylline, a competitive antagonist of adenosine, has been shown in both laboratory animals and humans to antagonize the negative dromotropic effect of adenosine by a mechanism of competitive inhibition.10 15 The results of the present study are consistent with and confirm previous reports that theophylline reverses the effects of adenosine and dipyridamole on the AV node.10 15 This finding is consistent with adenosine being the causative agent for the dipyridamole-induced prolongation of tachycardia cycle length, specifically at the level of the AV node. This, together with the demonstration that coronary sinus plasma adenosine concentrations were elevated coincident with the electrophysiological changes, provides the strongest support yet for the hypothesis that endogenous adenosine is the most likely cause of the electrophysiological effects of dipyridamole.
Adenosine has an ultrashort half-life of less than 10 seconds.16 The catheter-syringe system described above allowed us to estimate the adenosine levels in the coronary sinus by delivering a solution at the tip of the catheter that inhibits production and metabolism of adenosine in the plasma during the blood sample withdrawal. Other investigators using this catheter-syringe system have shown reproducible recovery of about 80% of exogenously infused adenosine.8 9 Coronary sinus plasma adenosine levels parallel the changes in interstitial concentration.17 However, because of rapid metabolism of this nucleoside, the coronary sinus blood levels of adenosine measured are likely to be an underestimation of the true interstitial adenosine concentrations. Nevertheless, the coronary sinus plasma concentrations of adenosine should reflect changes in the myocardial interstitial concentration of adenosine.
Dipyridamole is a drug with multiple effects. For instance, in addition to inhibiting adenosine uptake, dipyridamole has been shown to inhibit the enzymes adenosine deaminase and phosphodiesterase. However, the concentration of dipyridamole necessary to inhibit adenosine deaminase and phosphodiesterase is higher than that required to inhibit the nucleoside transporter. The magnitude of the potentiation of the AV nodal effects of adenosine by various nucleoside uptake blockers correlates significantly with the degree of inhibition of adenosine uptake.10 In addition, the effects of nucleoside uptake blockers on AV nodal conduction are completely abolished by adenosine deaminase and antagonized by adenosine receptor blockers.6 10 Thus, the most likely mechanism by which dipyridamole modulates AV nodal conduction is the potentiation of the effects of endogenous adenosine.
Dipyridamole causes peripheral vasodilation and consequently hypotension, which in turn may trigger reflex tachycardia. This could have posed a problem in that rather than terminating a tachycardia, dipyridamole might have actually accelerated it. However, there were no appreciable differences in either blood pressure or heart rate after autonomic blockade in the presence of dipyridamole, and there was no acceleration of rate or measurable difference in blood pressure. If there was a reflex tachycardia, it was masked by the β-adrenergic blockade (esmolol) and by the effect of prolongation of the tachycardia cycle length by endogenous adenosine.
We are unable to explain the wide variation in coronary sinus levels of adenosine among patients. There was no difference in ejection fraction, coronary artery disease, sex, or blood pressure during the procedure among the patients studied. It is unlikely that differences in dipyridamole metabolism and uptake are significantly different between patients. More likely, variability in adenosine metabolism and uptake among individuals is responsible for the minimal rise in coronary sinus adenosine levels in some patients. In fact, demonstration that little or no electrophysiological effects of dipyridamole were detected in those patients with minimal or no increases in coronary sinus adenosine levels provides strong evidence for the hypothesis that increasing adenosine levels are responsible for the electrophysiological effects of dipyridamole.
The results of the present study demonstrate that coronary sinus plasma adenosine concentration is elevated by dipyridamole and has antiarrhythmic properties that are confined to the AV node. Thus, nucleoside uptake inhibitors may be the basis for a unique new treatment approach to adenosine-sensitive tachyarrhythmias. With minimal effects on sinus rhythm and significant effects seen only during fast rates of pacing or tachycardia, nucleoside uptake inhibitors or other agents that potentiate the actions of adenosine (such as allosteric enhancers of this nucleoside18 ) may prove to be ideal antiarrhythmic agents that are both site- and event-specific.
This work was supported by a grant from the American Heart Association, Florida Affiliate. We thank J.Y. Lu, MD, for his invaluable assistance with this investigation.
Reprint requests to Jamie Beth Conti, MD, Box 100277, Division of Cardiology, University of Florida, Gainesville, FL 32610.
- Received September 1, 1994.
- Accepted October 26, 1994.
- Copyright © 1995 by American Heart Association
DiMarco JP, Sellers D, Beren RM, West GA, Belardinelli L. Adenosine: electrophysiologic effects and therapeutic use for terminating paroxysmal supraventricular tachycardia. Circulation. 1983;6: 1254-1263.
Wang T, Mentzer RM, Van Wylen DGL. Interstitial adenosine with dipyridamole: effect of adenosine receptor blockade and adenosine deaminase. Am J Physiol. 1992;263:H552-H558.
Jenkins JR, Belardinelli L. Atrioventricular nodal accommodation in isolated guinea pig hearts: physiological significance and role of adenosine. Circ Res. 1988;63:97-116.
Lerman BB, Wesley RC, Belardinelli L. Electrophysiologic effects of dipyridamole on atrioventricular nodal conduction and supraventricular tachycardia: role of endogenous adenosine. Circulation. 1989;80:1536-1543.
Shryock JC, Boykin MT, Hill JA, Belardinelli L. A new method of sampling blood for measurement of plasma adenosine. Am J Physiol. 1990;258(Heart Circ Physiol 27):H1232-H1239.
Hill JA, Utterback DB, Keim SG, Dugger D, Mayfield WR, Shryock JC, Belardinelli L. A catheter-syringe sampling system for measurement of adenosine in coronary sinus blood in humans. Coron Artery Dis. 1992;3:963-971.
Belardinelli L, Fenton RA, West A, Joel L, Althaus JS, Berne RM. Extracellular action of adenosine and the antagonism by theophylline on the atrioventricular conduction of isolated perfused guinea pig and rat hearts. Circ Res. 1982;51:569-579.
DiMarco JP. Electrophysiology of adenosine. J Cardiovasc Electrophysiol. 1990;1:340-348.
Ely SW, Berne RM. Protective effects of adenosine in myocardial ischemia. Circulation. 1992;85:893-904.
Bardenheuer H, Schrader J. Supply-to-demand ratio for oxygen determines formation of adenosine by the heart. Am J Physiol. 1986;250(Heart Circ Physiol 19):H173-H180.
Ontyd J, Schrader J. Measurement of adenosine, inosine and hypoxanthine in human plasma. J Chromatogr Sci. 1984;307:404-409.
Lasley RD, Hegge JO, Zhou Z, Utterback DB, Belardinelli L, Mentzer RM. Effects of dobutamine and regional ischemia on interstitial fluid and coronary sinus plasma adenosine levels in in situ porcine myocardium. Drug Develop Res. 1994;31:288,1994. Abstract.
Amoah-Apraku B, Xu J, Lu JY, Pelleg A, Bruns RF, Belardinelli L. Selective potentiation by an A1 adenosine receptor enhancer of the negative dromotropic action of adenosine in the guinea pig heart. J Pharmacol Exp Ther. 1993;266:611-617.