Antiarrhythmic effects of adenosine on ischemia-induced ventricular fibrillation

Antiarrhythmic Effects of Adenosine on Ischemia-Induced Ventricular Fibrillation Ulrike Stark, M a r i a n n e B r o d m a n n , A n d r e a s Lueger, and G e r h a r d Stark

Purpose:The antiarrhythmic efficacy of adenosine during states of AV-nodal reentrant tachycardias is well known and clinically established. Adenosine is also able to reduce ventricular arrhythmias when applied before coronary ligation in rats. Hypoxia or ischemia leads to an increased production of adenosine by cardiac myocytes. The purpose of this study was to evaluate if adenosine also has a direct antiarrhythmic effect on ischemia-induced ventricular fibrillation. Materials and Methods: In this study, the antiarrhythmic effects of adenosine on ventricular fibrillation during global (low flow) ischemia were evaluated in isolated guinea pig hearts perfused by the method of Langendorff. Results: Adenosine showed a dose-dependent pro-

Iongation of the peak to peak interval of the ventricular ECG signal during ventricular fibrillation until ventricular flutter or tachycardia occurred at a concentration of 2 mmol/L. At a concentration of 20 mmol/L, adenosine converted ventricular fibrillation into ventricular tachycardia with intermittent periods of asystole.This conversion of ventricular fibrillation to asystole was antagonised by 200 ~mol/L theophylline. Conclusion: Adenosine appears to have an antiarrhythmogenic effect both in supraventricular and ventricular rhythm disturbances. During myocardial infarction, where huge amounts of adenosine are present in ischemic regions, asystole may respond to adenosine antagonists.

HE PRECISE ROLE of endogenous adenosine

cult. 5-7 Adenosine also inhibits the His-Purkinje pacemaker, s In clinical practice, this effect accounts for the short periods of asystole seen with bolus administration of adenosine clinically.9 Because adenosine exhibits a direct effect on the ventricular pacemaker, the influence of adenosine on ventricular arrhythmias was studied in coronary ligation experiments. Adenosine can reduce or overcome ischemia-induced arrhythmias when applied before coronary ligation. 1~ Furthermore, it was shown that the Al-receptor antagonist theophylline is able to abolish bradyasystole refractory to atropine and epinephrine. 11-13 Therefore, the purpose of this study was to evaluate the effects of adenosine on ventricular fibrillation induced by global ischemia in isolated guinea pig hearts to determine whether (1) adenosine is able to influence ventricular fibrillation; (2) the possible effect of adenosine on ventricular fibrillation is mediated by the Al-receptor; (3) endogenous adenosine (elevated by global ischemia of the heart) plays a role in influencing ventricular fibrillation; and (4) theophylline antagonizes adenosine induced effects on ventricular fibrillation.

in the heart remains hitherto uncovered in the T literature. Several clues point to adenosine as an "endogenous myocardial protective substance." Hypoxia or ischemia leads to an increased production of adenosine by cardiac myocytes.1 The cardiac effects of adenosine are blocked by methylxanthines and potentiated by dipyridamole. 2'3 Adenosine aids in the regulation of the myocardial oxygen supply-demand balance by exerting a negative chronotropic effect on the sinus node, by decreasing conduction velocity through the AV-node and prolonging its refractoriness, by counteracting catecholamine induced effects, by decreasing myocardial contractility, and by inducing myocardial vasodilatation. 4 Adenosine is an endogenous nucleoside, with special electrophysiologic properties on the cardiac pacemaker and conduction system. Adenosine effectively terminates supraventricular arrhythmias, where the AV-node is involved in the reentrant cir-

From the Department of Internal Medicine, Karl-FranzensUniversity, Graz, Austria. This work was supported by Grant No. P l O239from the Austrian Research Foundation and the Austrian Nationalbank (5913). Received March 27, 2000. Accepted October 16, 2000. Address reprint requests to Ulrike Stark, MD; Department of Internal Medicine, Karl-Franzens- University, Auenbruggerplatz 15, A-8036 Graz, Austria. Copyright 9 2001 by W.B. Saunders Company 0883-9441/01/1601-0002535.00/0 ' doi: lO.1053/jcrc.2001.21791 8

Copyright 9 2001 by W.B. Saunders Company

MATERIALS AND METHODS

Design Thirty-two guinea-pigs of either sex weighing 300 to 400 g and fed ad libimm were used. Complete results were obtained from 29 of 32 preparations studied because asystole occurred in three hearts after the application of 100 ~mol/L adenosine. In the first group (n = 7), the depressant action of adenosine on the ventricular escape rhythm, triggered by the His Purkinje system, was investigated. Journal o f Critical Care, Vol 16, No 1 (March), 2001: pp 8-16

ANTIFIBRILLATORY EFFECT OF ADENOSINE

9

In the second group, the effects of increasing concentrations of adenosine on ventricular fibrillation during low flow ischemia were studied (n = 7). To evaluate if the antifibrillatory effect of adenosine is mediated by the A~ receptor, the effect of the Al-receptor agonist 2-Chloro-N6-cyclopentyladenosine ~(CCPA) on ventricular fibrillation was studied in a further group of experiments (n = 6). To evaluate if endogenous adenosine plays any role during ventricular fibrillation, experiments were performed in the presence of the combination of the adenosine uptake blocker dipyridamole and adenosine (n = 6). The efficacy of theophylline to counteract the effects of adenosine during ventricular fibrillation was studied in another group of experiments (n = 6).

irregularities occurred during the equilibration period, the heart was discarded. Two FeC13 chloridized silver wire electrodes (wire 0.3 mm, 1.5-mm electrode tip) were placed on the epicardial surface of the heart free to move with the contractions. Both electrodes were positioned in the AV-valve plane, anteriorly near the origin of the interventricular artery and posteriorly between the two auricles. The unfiltered signals were amplified by a factor of 100 with an instrumentation amplifier (Anton Paar, Graz, Austria) with AC input (fc = 0.72 Hz). His-bundle activity was visible in the bipolar recorded ECG signals monitored on a digital storage oscilloscope and stored on a tape-recorder sampling at 5 kHz. Details of this high resolution ECG recording technique are described in earlier publications. 14 The ECG signals were further digitized by an analog to digital converter (TL-125, Axon Instruments, Foster City, USA), monitored, and stored on a personal computer (486/50 MHz) for further analysis.

Experimental Process Guinea pigs were injected intraperitoneally with 250 IU heparin 1 hour before being killed by dislocation of the neck. The chest was quickly opened, the heart removed and attached to a modified Langendorff perfusion system (Auton Paar, Graz, Austria). All procedures met the guidelines set by the Committee on Animal Care of our medical center. Tyrode's solution, saturated with a mixture of 95% oxygen and 5% carbon dioxide warmed to 36~ was used as perfnsate (in mmol/L: NaC1 132.1, KC12.7, CaCI2 2.5, MgC12 1.15, NaHCO3 24.0, NaH2PO4 0.42, e-glucose 5.6). Immediately after the heart had been attached to the Langendorff perfusion system, electrocardiographic (ECG) recordings were taken from the epicardial surface. The perfusion rate was individually adjusted for every heart (8 to 10 mL/minute), so that atrioventricular conduction, a sensitive parameter for acute ischemia in this preparation, was shorter than 65 ms and the spontaneous sinus rate was about 200 beats/rain. Each heart was allowed to equilibrate for 30 minutes. If rhythm

Ventricular Escape Rhythm We measured the spontaneous sinus rate, rate of ventricular escape rhythm, and intraventricular conduction time (QRS interval) to show the depressant action of adenosine on sinus rate and ventricnlar escape rhythm triggered by the I-Iis-Purkinje system, during control conditions, and after 10 minutes of perfusion with either 30 Ixmol/L or 100 txmol/L of adenosine.

Ventricular Fibrillation To induce ventricular fibrillation in isolated guinea pig hearts, global ischemia was induced by reduction of the flow rate to 1 mL/minute. During the low-flow period, ventricular stimulation at a pacing rate of 300 beats per minute was started. The pacing rate was doubled within 10 minutes if no ventricular fib-

V P

Control

100 ms I-----t

V Fig 1, Effects of 100 itmol/L adenosine on the cardiac conduction and pacemaker system. 100 ttmol/L adenosine caused a third-degree AV block with a consecutive sinus node and ventricular arrest. V, ventricuiar activity, P, atdal activity (P wave).

100 laM Adenosine

I

I~

Asystole >

250 ms I---t

10

STARK ET AL

rillation could be induced under these conditions. If ventricular fibrillatiofi occurred, ventricular pacing was continued for a furtiler 5 minutes. In earlier experiments (data not shown in dais article) a 98% stability of ventricular fibrillation for at least 1 hour v~as shown. After 5 minutes of stable ventricular fibrillation during global (low-flow) ischemia without pacing, the heart was perfused with adenosine at a concentration of 0.1 mmol/L, 2 mmol/L, and 20 mmol/L. Each concentration was acting on the heart for 5 minutes. The same protocol as described above to induce ventricular fibrillation was used in an additional group of experiments. After 5 minutes of stable ventricular fibrillation, 20 mmol/L adenosine was applied to the heart. To antagonize the effects of adenosine, 200 izmol/L theophylline was added to the perfus/tte after a perfnsion period of 5 minutes with adenosine alone. In another group of experiments, after 5 minutes of stable ventricular fibrillation, CCPA in increasing concentrations of 1 i.~mol/L, 10 ~mol/L, and 20 ~mol/L was added to the perfnsate. In the last group of experiments, ventricular fibrillation was again induced by the same way as described above, and afterwards the effects of dipyridamole (1 ~mol/L) and increasing dosages of adenosine (0.1 mmol/L and 0.5 mmol/L) were tested.

Compounds Used Adenosine, dipyridamole, theophylline (Sigma, Germany), and 2-Chloro-N6-cyclopentyladenosine (gift of Prof. Belardinelli, University of Florida, Gainesville, FL) were dissolved in Tyrode's solution and freshly prepared before each experiment. All measurements were performed after a perfusion period of 5 minutes. Control measurements were done in the presence of Tyrode's solution.

Data Analysis The peak-to-peak interval of the ventricular ECG signals (peaks higher than 0.3 mV were recognized) was continuously measured by a computer during ventricular fibrillation. The mean • standard error of the mean (SEM) of the peak-to-peak interval of the ventricular ECG signal during a period of 1 minute of ventricular fibrillation was calculated under baseline conditions (10 minutes after stable ventricular fibrillation) and 5 minutes after the addition of each drug concentration. During the perfusion period with each drug, the occurrence of asystole was monitored and classified as a short period of asystole with a duration of 1 to 10 seconds and a complete asystole with a duration of more than 60 seconds. All data are expressed as mean _+ SEM. The data were compared using a Student's t test or Wilcoxon test after a test of homogeneity of variance. Comparisons between baseline conditions and different drug concentrations were done by repeated measures of ANOVA followed by Bonferroni's method (Sigmastat, statistical software package, version 1.0; Jandel, Erkrath, Germany).

RESULTS

Effects of Adenosine on Ventricular Escape Rhythm The spontaneous sinus rate was significantly reduced by 30 and 100 p.mol/L adenosine. At the used

concentrations of adenosine, a third-degree atrioventricular block was present in all experiments. The ventricular escape rhythm was significantly reduced by 30 tzmol/L adenosine. At the concentration of 100 Ixmol/L adenosine, asystole occurred (Fig 1). The QRS interval was not affected by adenosine (Table 1). In our experiments, the exact location of the ventricular pacemaker during ventricular escape rhythm was not mapped. However, in all experiments a His bundle electrogram could be registered with the width of the QRS complex being comparable to sinus rhythm under control conditions, which makes a pacemaker site in the His-Purkinje system likely.

Effects of Adenosine on Ventricular Fibrillation Adenosine caused a dose-dependent, significant increase of the peak-to-peak interval during ventricular fibrillation resulting in ventricular flutter at a concentration of 2 mmol/L adenosine (Table 2). Adenosine 20 mmol/L converted ventricular fibrillation into ventricular tachycardia with short periods of asystole (duration of asystole between 1 and 10 seconds) (Fig 2). In 2 out of 7 experiments, 20 mmol/L of adenosine induced a complete asystole (duration of asystole of more than 60 seconds). In a separate group of experiments (n = 6), shortand long-lasting periods of asystole in the presence of 20 mmol/L adenosine were abolished completely in all experiments by theophylline (200 ixmol/L) (Fig 3). The At-receptor agonist CCPA also decreased the ventricular fibrillation rate in a concentration dependent manner (Table 3, Fig 4). In one out of six experiments a complete asystole occurred at concentrations of 20 ixmol/L. Short periods of asys-

Table 1. Effects of Increasing Adenosine Concentrations (30 i~mol/L, 100 i~mol/L) on Intraventricular Conduction (QRS Interval) and Ventricular Escape Rhythm (VR)

n QRS interval (ms) VR (beats/minute) Sinus rate (beats/minute)

Control

30 ixrnol/L

100 ixrnol/L

7 32.9 + 5.3 88 • 5 234 • 12

7 34.6 • 5.8 47 • 14" 183 • 9*

7 Asystole Asystole 82 • 14" (n = 3)

NOTE. At the highest concentration of adenosine (100 ixmol/L) in three out of six experiments a sinus node arrest occurred. Values are mean + SEM. * P < .01.

11

ANTIFIBRILLATORY EFFECT OF ADENOSINE

Table 2. Effects Of Adenosine on Ventricular Fibrillation During Global Ischemia

n V-V (ms) ~ Asystole (1 "/o 10 seconds) Asystole (> 1 minute)

Baseline

0.1 mmol/L Adenosine

2 mmol/L Adenosine

20 mmol/L Adenosine

7 44.9 + 7.82

7 63.8 -+ 17.81

7 69.2 +- 17.82"

5 179 -+ 52.3*

0 out of 7 0 out of 7

0 out of 7 0 out of 7

0 out of 7 0 out of 7

7 out of 7 2 out of 7

NOTE. Values are mean _+ SEM. *P < .01. Abbreviations: VV, ventricular fibrillation cycle length; baseline, global ischemia in the presence of Tyrode's solution alone.

tole were observed in two out of six experiments at a concentration of 10 p~mol/L and in all experiments at concentrations of 20 txmol/L. The adenosine uptake blocker dipyridamole (1 ixmol/L) had only a slight effect on the ventricular fibrillation rate. The addition of adenosine at concentrations of 0.5 mmol/L induced a complete asystole in two out of six experiments in the presence of dipyridamole (Table 4, Fig 5).

DISCUSSION

The results of this study demonstrate that at high doses, adenosine is able to induce asystole. In a state of global, low-flow ischemia, high doses of adenosine converted ventrictflar fibrillation to ventricular flutter, ventricular tachycardia, or asystole. This antifibrillatory effect of adenosine was antagonized by theophylline.

Baseline VF

0.1 m M Adenosine

2 m M Adenosine

Fig 2. Dose-dependent antifibrillatory effect of adenosine in isolated guinea pig hearts during ventricular fibrillation induced by global (low flow) ischemia. Baseline VF, ventricular fibrillation induced by global (low flow) ischemia.

20 m M Adenosine

2 0 0 ms I 1

12

STARK ET AL

Baseline VF

V

20 m M Adenosine

^

(~

Asystole

"IV-

V

20 mM Adenosine

+

200 pM Theophylline

u

L/L.,k.-.-J_Av

jv

~YJ":

1 ~.~ 200 ms ! t

Earlier studies had demonstrated that adenosine decreases ventricular pacemaker activity.S ~5~6 This effect is enhanced by the adenosine transport blocker dipyridamole. 17 Therefore, the negative cbronotropic effect of adenosine on the rate of the ventricular escape rhythm probably is mediated by adenosine cell surface receptors. Different pace-

Fig 3. Theophylline antagonises adenosine induced asysrole during ventricular fibrillation. Baseline VF, ventricular fibrillation induced by global (low flow) ischemia; V, ventricul a r activity.

makers of the heart exhibit different sensitivity to adenosine. Adenosine suppresses ventricutar pacemaker cells more than the sinus node pacemaker. 8 The most obvious explanation for the greater inhibiting effect of adenosine on ventricular automaticity compared with sinus node activity may be that the A1 receptor varies in number and affinity

Table 3. Effects of CCPA (2-Chloro-NS-cyclopentyladenosine) Alone on Ventricular Fibrillation During Global Ischemia Baseline n V-V (ms) Asystole (1 to 10 seconds) Asystole (> 1 minute)

6 40.5 • 7.6 0 out of 6 0 out of 6

1 ~mol/L CCPA 9 6 63.5 • 1.6" 0 out of 6 0 out of 6

10 ~mol/L CCPA

20 p,rnol/L CCPA

6 74,3 ~ 3.91" 2 out of 6 0 out of 6

6 99.6 -+ 1.61" 6 out of 6 1 out of 6

NOTE. Values are mean _+ SEM. Abbreviations: VV, ventricular fibrillation cycle length; baseline, global ischemia in the presence of Tyrode's solution alone. t P < .01. * P < .05.

ANTIFIBRILLATORY EFFECT OF ADENOSINE

13

Baseline VF

1 ~M CCPA

10 ~ M C C P A

Fig 4. Dose-dependent antifibrillatory effect of CCPA (2Chloro-NS-cyclopentyladeno sine) in isolated guinea pig hearts during ventricular fibrillation induced by global (low flow} ischemia, BaselineVF, ventricular fibrillation induced by global (low flow) ischemia.

20 ~ M C C P A

200 ms I I

with the tissue. 18Adenosine receptors may be more numerous or may have greater affinity to adenosine in the His-Purkinje system. In anesthetized dog models of coronary artery occlusions, it was shown that the exogenous administration of adenosine or an increase in myocardial levels of adenosine by nucleoside uptake inhibitors have a profound antiarrhythmic effect in the early stage of myocardial ischemia, w-2a A similar protective effect of adenosine was demon-

strated in isolated rat hearts. 22 The mechanism underlying this antiarrhythmic effect is unclear. It is not even certain whether this antiarrhythmic effect is mediated by adenosine itself or by an adenosine breakdown product, such as hypoxanthine, or by antagonism of the proarrhythmic effects of catecholamines. In the in vivo models, the reduction of afterload and heart rate as well as the increase of coronary blood flow by adenosine may contribute indirectly to the antiarrhythmic effect. In the iso-

Table 4. Effects of the Combination of Dipyridamole and Adenosine on Ventricular Fibrillation During Global Ischemia

n V-V (ms) Asystole (1 to 10 seconds) Asystole (> 1 minute)

Baseline

1 ~.rnol/L DPM

1 I~mol/L DPM + 0.1 mrnN/L Adenosine

1 i~mo[/L DPM + 0.5 mrnol/L Adenosine

6 36.6 -+ 1.9 0 out of 6 0 out of 6

6 59.4 -+ 6.7* 0 out of 6 0 out of 6

6 106.1 _+ 2.1t 6 out of 6 0 out of 6

6 219,1 _+ 6.7t 6 out of 6 2 out of 6

NOTE. Values are mean _+ SEM. Abbreviations: VV, ventricular fibrillation cycle length; baseline, global ischemia in the presence of Tyrode's solution alone. t P < .01. *P < .05.

14

STARK ET AL

Baseline VF

1~

Dipyridamole

1 ltM Dipyridamole +

0.1 mM Adenosine

1 ~tM Dipyridamole +

0.5 mM Adenosine

200 ms 1

lated guinea pig heart during global low flow, these indirect effects do not play any major role. Adenosine directly decreases ventricular pacemaker activity. 16 However, under normoxemic conditions, adenosine does not modify ionic currents in isolated single ventricular rnyocytes in concentrations between 100 and 1000 txmol/L. The only significant effect of adenosine on normal Purkinje fibers is a slight concentration-dependent abbreviation of action potential duration. 23 Under ischemic conditions, adenosine has quite different effects. Ischemia results in marked reductions in resting membrane potential, action potential amplitude, upstroke velocity, and action potential duration. The ischemia-induced reduction in upstroke velocity and action potential amplitude are significantly attenuated by adenosine. 23 This occurs without any concomitant change in resting membrane potential or action potential duration and is observed whether adenosine is given before, throughout the period of ischemia, or after the onset of the ischemia-induced changes. The precise underlying mechanism of

!

Fig 5. Dose-dependent antifibrillatory effect of adenosine in the presence of the adenosine uptake blocker dipyddamole in isolated guinea pig hearts during ventdcular fibrillation induced by global (low flow) ischemia. Baseline VF, ventricular fibrillation induced by global (low flow) ischemia.

these effects in unknown. However, these effects would be potentially antiarrhythmic because ischemia-induced reduction of upstroke velocity and action potential amplitude will be accompanied by an increased risk of unidirectional block, a factor that predisposes to re-entrant arrhythmias. 1~ Another contributing factor to the antiarrhythmic effect of adenosine may be the inhibition of noradrenaline release from cardiac sympathetic nerves by adenosine. This effect is mediated through presynaptic adenosine A1 receptors. 24 Enhanced release of adenosine during ventricular fibrillation has been shown to mediate postdefibrillatory bradycardia, AV block, hypotension, and myocardial depression, effects that are reversed by adenosine antagonists. 25'26 In this article, the depressant effect of adenosine on the His-Purkinje pacemaker was demonstrated in isolated guinea pig hearts. This direct inhibitory effect of adenosine on the His-Purkinje pacemaker as well as the direct effects of adenosine on the ac: tion potential during ischemia may be responsible

ANTIFIBRILLATORY EFFECT OF ADENOSINE

15

for the anti arrhythmic activity of adenosine during ventricular fibrillation induced by global ischemia. Short- and long-lasting periods of asystole in the presence, o f 20 mmol/L adenosine were antagonized by the adenosine antagonist theophylline. The Al-receptor agonist CCPA exhibit comparable antifibrillatory effects, such as those observed in the presence of adenosine alone. Therefore, it appears that the antifibrillatory effect of adenosine is directly mediated by the A1 receptor. In the presence of the adenosine uptake blocker dipyridamole an approximately 40 times lower concentration of adenosine was able to convert ventricular fibrillation to ventricular tachycardia. This finding suggests that endogenous adenosine levels, which are increased during ischemia, may act antifibrillatory. The major limitation of our study is that the experiments were performed in isolated and therefore denervated hearts. A high concentration of adenosine, which exceeds under in vivo conditions, was necessary to induce the antifibrillatory effect. This might be a limitation of our global ischemia model, which entails enormous damage of the myocardium, and this damage may affect the number

of A1 receptors or the affinity of adenosine to the receptor. However, this enormous damage of the heart was necessary to guarantee a stable ventricular fibrillation in this small heart. During local ischemia and in an early stage of ischemia, lower concentrations of adenosine may be able to show the same antifibrillatory effects. On the basis of our results, we suggest that adenosine is an endogenous antiarrhythmic substance not only for supraventricular but also for ventricular arrhythmias. It could be expected that during ischemia where a large amount of adenosine will be produced by the myocardium, adenosine will act as an endogenous antiarrhythmic substance against ventricular fibrillation, but may also be able to induce asystole. This theory is confirmed by the clinical finding that aminophylline is effective against bradyasystolic cardiac arrest refractory to atropine and epinephrine. 11-13 ACKNOWLEDGMENTS The authors would like to thank Luiz Belardinelli, MD, Hans Domanovits, MD, and Fritz Sterz, MD, for their helpful suggestions.

REFERENCES 1. Kitakaze M, Hori M, Kamada T: Role of adenosine and its interaction with adrenoceptor activity in ischaemic and reperfusion injury of the myocardium. Cardiovasc Res 27:18-27, 1993 2. Favale S, DiBiase M, Rizzo V, et al: Effect of adenosine and adenosine-5'-triphosphate on atrioventricular conduction in patients. J Am Coil Cardiol 5:1212-1219, 1985 3. Belhassen B, Pelleg A: Adenosine triphosphate and adenosine: Perspectives in the acute management of paroxysmal supraventricular tachycardia. Clin Cardiol 8:460-464, 1985 4. Conti M, Belardinelli L: Adenosine is an endogenous antiarrhythmic drug. Circulation 91:1961-1967, 1995 5. Clarke B, Rowland E, Burnes PJ, et al: Rapid and safe termination of supraventricular tachycardia in children by adenosine. Lancet 1:299-301, 1987 6. Di Marco JR Sellers TD, Berne RM, et al: Adenosine: electrophysiologic effects and therapeutic'rose for terminating paroxysmal supraventricular tachycardia. Circulation 68:12541263, 1993 7. Di Marco JP: Adenosine and supraventricular tachycardia, in Pelleg A, Michelson EL, Dreifus LS (eds): Cardiac Electrophysiology and Pharmacology of Adenosine and ATP: Basic and Clinical Aspects. New York, NY, Alan R. Liss, 1987, pp 271282 8. Stark G, Domanovits H, Sterz F, et al: Action of ATP on ventricular automaticity. J Cardiovasc Parmaco124:740-744, 1994 9. Reed R, Falk JL, O'Brian J: Untoward reaction to adenosine therapy for supraventricular tachycardia. Am J Emerg Med 9:566-570, 1991

10. Boachie-Ansah G, Kane KA, Parratt JR: Is adenosine an endogenous myocardial protective (antiarrhythmic) substance under conditions of ischaemia. Cardiovasc Res 27:77-83, 1993 l l . Viskin S, Belhassen B, Roth A, et al: Aminophylline for bradyasystolic cardiac arrest refractory to atropine and epinephrine. Ann Intern Med 118:279-281, 1993 12. Perouansky M, Shamir M, Hershkowitz E, et al: Successful resuscitation using aminophylline in refractory cardiac arrest with asystole. Resuscitation 38:39-41, 1998 13. Mader TJ, Gibson P: Adenosine receptor antagonism in refractory asystolic cardiac arrest: Results of a human pilot study. Resuscitation 35:3-7, 1997 14. Stark G, Stark U, Tritthart HA: Assessment of the conduction of the cardiac impulse by a new epicardiac surface and stimulation technique (SST-ECG) in Langendorff perfused mammalian hearts. J Pharmacol Methods 21:195-209, 1989 15. Heller IJ, Olsson RA: Inhibition of rat ventricular automaticity by adenosine. Am J Physiol 248:907-913, 1985 16. Stark G, Stark U, Bachernegg M, et al: A comparison of the effects of adenosine and verapamil on the conduction and pacemaker system of isolated guinea pig hearts. Clin Cardiol 16:859-862, 1993 17. Pelleg A, Mitamura H, Mitsuoka T, et al: Adenosine and ATP supress ventricular escape rhythms in the canine heart. J Am Coil Cardiol 8:1145-1151, 1986 18. Linden J, Hollen CE, Patel A: The mechanism by which adenosine and cholinergic agents reduce contractility in rat myocardium. Circ Res 56:728-735, 1985 19. Parratt JR, Wainwright CL: The effects of adenosine on

16

arrhythmias induced by myocardial ischaemia and reperfusion in dogs. Br J Pharmacol 85:228R 1985 20. Wainwright CL, Parratt JR: An antiarrhythmic effect of adenosine during myocardial ischaemia and reperfusion. Eur J Pharm'acol 145:183-194, 1988 21. Wainwright CL, Parratt JR, Van Belle H: The haemodynamic and antiarrhythmic effects of the nucleoside transport inhibitor R75231 in anaesthetised pigs. Br J Pharmaco199:14P, 1990 22. Lubbe WT, Guyen T, West EJ: Modulation of myocardial cyclic AMP and vulnerability to fibrillation in the rat heart. Fed Proc 42:2460-2464, 1983 23. Boachie-Ansah G, Kane KA, Parratt JR: Electrophysiological effects of adenosine and adenosine triphosphate on sheep

STARK ET AL

Purkinje fibres under normal and simulated ischaemic conditions. Br J Pharmacol 97:240-246, 1989 24. Richardt G, Waas W, Kranzhofer R, et al: Adenosine inhibits exocytic release of endogenous noradrenaline in rat heart: A protective mechanism in early myocardial ischemia. Circ Res 61:117-123, 1987 25. Wesley RC, Belardinelli L: Role of endogenous adenosine in postdefibrillation bradyarrhythmia and hemodynamic depression. Circulation 80:128-137, 1989 26. Wesley RC, Porzio D, Sadeghi M: Effect of selective A1 adenosine receptor antagonism on postdefibrillation cardiovascular depression: Evidence for an antiadrenergic role of endogenous adenosine. Cardiovasc Res 27:129-133, 1993