Background The single most important determinant of cardiac arrest outcome is the duration of ventricular fibrillation (VF) preceding delivery of a high-energy shock, because of the adverse effect of VF duration on defibrillation threshold (DFT). Although a metabolic mechanism has been proposed, hypoxia, metabolic acidosis, or alkalosis do not adversely affect DFT. However, since (1) catecholamines and adenosine levels are markedly increased during hypoxia, (2) exogenous catecholamines decrease DFT, and (3) adenosine is a potent antagonist of the electrophysiological effects of catecholamines on ventricular myocardium, we hypothesized that release of adenosine during prolonged VF adversely affects DFT and that this effect occurs through an antiadrenergic mechanism.
Methods and Results DFT was determined in dogs during infusion of adenosine (300 μg · kg−1 · min−1) and dipyridamole (0.25 mg/kg), an adenosine uptake blocker, a regimen that resulted in adenosine levels in the myocardial effluent equivalent to those achieved after 5 minutes of VF. Adenosine increased transthoracic DFT in each dog by 49±14% (n=21) (mean±SEM) and transmyocardial DFT in a separate group of 10 dogs by 103±16%, P=.0003. Pretreatment with the specific A1 adenosine receptor antagonist 8-cyclopentyltheophylline (CPT) 5 mg/kg completely abolished the effects of adenosine on DFT. The effects of adenosine on DFT were also examined in the denervated state (propranolol 0.2 mg/kg plus bilateral vagotomy). In contrast to its effect in the innervated condition, adenosine had no effect on DFT in the same dogs when denervated, 49±11 versus 53±10 J (P=NS).
Conclusions Adenosine significantly increases transthoracic and transmyocardial DFT, effects that are mediated by the A1 adenosine myocardial receptor through an antiadrenergic mechanism. These results suggest that enhanced release of adenosine during VF may have a deleterious effect on defibrillation and that intramyocardial delivery of a specific A1 adenosine antagonist during VF may facilitate defibrillation and significantly reduce defibrillation threshold.
Sudden death due to ventricular fibrillation (VF) is the major cause of death in the United States. The single most important factor determining the outcome of cardiac arrest is the duration of VF before a high-energy shock is delivered, because of the adverse effect of VF duration on defibrillation threshold (DFT).1 2 3 4 The mechanism governing this phenomenon is unknown. Although a metabolic cause has been hypothesized, it has not been proved, since metabolic acidosis or alkalosis does not elevate defibrillation requirements.5 6 The potential role for the endogenous metabolite adenosine in mediating elevation of DFT has not been systematically evaluated. We considered a possible role for adenosine in mediating this phenomenon on the basis of several lines of evidence: First, myocardial production and interstitial release of adenosine is triggered by an imbalance in O2 supply-demand,7 a condition that exists during VF. Second, the effects of adenosine in ventricular myocardium are antiadrenergic (inhibiting production of cAMP) and are mediated by the inhibitory GTP-binding protein Gi.8 Third, endogenous catecholamine levels are markedly increased during VF, and catecholamines are known to decrease DFT, whereas β-adrenergic blockade elevates DFT.9 10 11 In view of these data, we hypothesized that enhanced release of adenosine from the myocardium during VF may mediate an increase in energy requirements for defibrillation and that this occurs through an antiadrenergic action.
All studies were performed in accordance with the guidelines of the American Physiological Society in a protocol approved by the Institutional Animal Care and Use Committee at the Cornell University Medical College. Experiments were performed in adult mongrel dogs weighing 15 to 28 kg. All animals were premedicated with acepromethazine (2 mg/kg IM), anesthetized with sodium pentobarbital (30 mg/kg IV), and ventilated with a Harvard respirator adjusted to maintain arterial pH within the physiological range. Additional doses of pentobarbital were administered as necessary. ECG surface lead I and arterial blood pressure were continuously monitored on a physiological recorder.
The right or left internal jugular vein was cannulated percutaneously, and a 6F quadripolar catheter (Bard Inc) was advanced to the right ventricle. The position of the catheter was verified by the intracardiac recording and by pacing with an external pacemaker. In open-chest dogs, the catheter was sutured to the surface of the right ventricle. For protocols requiring a bilateral vagotomy, a paramedian incision was made on both sides of the trachea, and the vagus nerve was identified, isolated adjacent to the carotid artery, and then ligated.
Sampling of Coronary Sinus Effluent
A median sternotomy was performed, the pericardium was reflected, and the heart was suspended in a pericardial cradle. A 6F right coronary Judkins catheter was introduced via an internal jugular vein and advanced under manual guidance into the coronary sinus (CS). The position of the catheter was verified by manual palpation before each sampling of CS blood.
Measurement of Plasma Adenosine, Catecholamines, and Lactate
For adenosine measurements, blood samples from the CS and the femoral artery were withdrawn simultaneously and placed immediately into iced tubes containing a stop solution that inhibits adenosine metabolism. The solution contained 0.2 mmol/L dipyridamole, 4.2 mmol/L Na2EDTA, 5 μmol/L erythro-9(2-hydroxy-3-nonyl)adenine, 79 μmol/L α,β-methylene adenosine 5′-diphosphate, and 25 U/mL heparin in 0.9% NaCl.12 Samples were obtained immediately before induction of VF (0 minutes) and after 5 minutes of VF. The transit time of blood through the catheters was approximately 5 seconds. The samples were cold-centrifuged (4°C) for 20 minutes at 11 000g to separate blood cells from the diluted plasma. The supernatant fraction was then ultrafiltered during centrifugation at 900g to 950g for 60 minutes. Adenosine content was determined from aliquots of reconstituted ultrafiltrate by high-performance liquid chromatography.
Epinephrine, norepinephrine, and lactate serum levels were determined from blood samples withdrawn from the CS and femoral artery simultaneously. As for adenosine, samples were obtained just before induction of VF and 5 minutes after continuous VF. Serum epinephrine and norepinephrine levels were measured by radioenzymatic assay,13 and serum lactate was measured by an enzymatic method previously described.14
Determination of Transthoracic Defibrillation Threshold
Defibrillation shocks were delivered by a modified commercially available defibrillator (Lifepak 6, Physio-Control) designed to deliver a pulse calibrated in units of current through a constant-load current divider circuit.15 Shocks were calibrated to deliver 3000 V at 400 J across a 50-Ω load. Variable parallel and series resistors were adjusted to deliver the desired current and simultaneously maintain a 50-Ω load to the defibrillator. Electrode area (60 cm2, circular stainless steel) and force (50 N) were held constant with a precision control system as previously described.15 The electrodes were covered with a fresh, thin film of electrode paste (Redux Paste, Hewlett-Packard) and were positioned over the shaven right and left lateral chest walls at the transverse level of the heart.
At the beginning of the protocol, transthoracic resistance was determined in each dog by a 20-A shock delivered during sinus rhythm. Voltage delivered across the thorax was measured with a 1000:1 voltage divider in parallel with the defibrillator output, and delivered current was measured with a 0.10-Ω resistor in series with the defibrillator output. Voltage and current wave forms were displayed on a triggered-sweep storage oscilloscope (model 5113, Tektronix) with a frequency response from DC to 1 MHz.
DFT was determined by a method previously described.16 In brief, VF was induced by AC delivered through the right ventricular catheter. After induction of VF, the endotracheal tube was clamped at peak inspiration. Twenty seconds after induction of VF, a subthreshold shock, usually starting at 15 A, was delivered, and the current increased in 2- to 5-A steps every 10 seconds until defibrillation occurred. This sequence was repeated with an initial shock strength 2 A lower than the successful shock from the preceding episode. Data were used from second-shock conversions only (a successful shock that followed an initial failed shock), since by definition threshold cannot be determined from a first-shock conversion. Thresholds determined during two consecutive trials were averaged in each phase of the protocol. Thresholds for consecutive trials varied by approximately ±2 A. A symmetrical protocol was implemented, with control DFTs determined before and after an intervention.16
Determination of Transmyocardial Defibrillation Threshold
A median sternotomy was performed, and the heart was suspended in a pericardial cradle. A small and a large ventricular patch electrode (models 0040 and 0041, Cardiac Pacemakers Inc) were sutured onto the anterior and posterior surfaces of the heart, respectively, so as to maximize the amount of interventricular septum between the leads. The electrode surface areas including insulation were 30 and 53 cm2, respectively. The leads were connected to an external cardioverter defibrillator (Ventak ECD, model 2805, Cardiac Pacemakers Inc) that can be set to deliver energy-based shocks of 1, 2, 3, 4, 5, 8, 10, 15, 20, 25, 30, 35, or 40 J. DFT was determined by a subthreshold shock delivered 20 seconds after induction of VF, and two subsequent shocks of incremental energy levels were applied until defibrillation occurred. A rescue shock of 40 J was delivered if defibrillation did not occur with the first three shocks. As described above for transthoracic shocks, data from second-shock conversions from two consecutive successful trials were averaged to determine DFT during each phase of the protocol.
Drugs and Chemicals
The following drugs and preparations were used for the experiments. Adenosine (Sigma Chemical Co) was dissolved in normal saline and diluted to a concentration of 1 mg/mL. Adenosine was administered as a constant infusion at a rate of 300 μg · kg−1 · min−1 with an infusion pump (LifeCare Pump, model 4P, Abbott Laboratories). Atropine (Elkins-Sinn Inc) was administered intravenously at a dose of 0.04 mg/kg. Dipyridamole (Sigma) was dissolved in normal saline at a pH of 2 to 3 (after addition of HCl) and was diluted to a concentration of 1 mg/mL. It was administered as an infusion over 30 to 60 seconds at a dose of 0.25 mg/kg. 8-Cyclopentyltheophylline (CPT) (Research Biochemical Inc) was dissolved in a solution of normal saline with 1N NaOH in a 3:1 ratio. It was administered as an infusion over 1 minute at a dose of 5 mg/kg. Isoproterenol (Elkins-Sinn Inc) was diluted in normal saline or 5% dextrose to a concentration of 4 μg/mL and infused at a continuous rate of 6 to 12 μg/min.9 Propranolol (Inderal 1 mg/mL, Ayerst Laboratories Inc) was administered intravenously at a dose of 0.2 mg/kg and at a rate of 1 mg/min.
Coronary Sinus Levels of Adenosine, Catecholamines, and Lactate During Ventricular Fibrillation (Protocol 1)
After surgical instrumentation, the animals were allowed to rest for 15 to 30 minutes until a stable baseline blood pressure and heart rate were observed. Blood samples for control measurements of adenosine, catecholamines (epinephrine and norepinephrine), and lactate were obtained simultaneously from the femoral artery and CS. VF was then induced, and samples for adenosine, catecholamines, and lactate were obtained from the femoral artery and CS 5 minutes after induction of VF.
Effects of Adenosine on Transthoracic and Transmyocardial Defibrillation Threshold (Protocol 2)
In this protocol, DFT was determined either transthoracically (protocol 2A) or transmyocardially (protocol 2B). The effect of adenosine on DFT was evaluated during concurrent administration of adenosine and dipyridamole, a nucleoside transport blocker, and isoproterenol. Steady-state effects of isoproterenol were achieved by its infusion being started 15 minutes before administration of adenosine. Two minutes after the adenosine infusion was begun (300 μg · kg−1 · min−1), dipyridamole 0.25 mg/kg was given. One minute after the dipyridamole bolus, DFT was determined. A control DFT was determined before adenosine and dipyridamole, and a recontrol DFT was determined 1 hour after the adenosine and isoproterenol infusions were discontinued (Fig 1⇓). The two control DFTs were averaged and compared with DFT determined during adenosine and dipyridamole.
Pilot experiments in 3 dogs showed that the doses of adenosine and dipyridamole given in protocol 2 (during sinus rhythm) were sufficient for achieving CS levels of adenosine comparable to those obtained after 5 minutes of VF, 1098±476 (sinus rhythm) versus 817±257 nmol/L (VF, protocol 1), P=NS. This was an important objective because, although DFT was determined after approximately 30 seconds of VF in the protocol, myocardial effluent levels of adenosine were intended to correlate with those achieved after 5 minutes of VF. In the pilot experiments as well as in protocol 2, adenosine infusion was maintained for 15 minutes after the bolus dose of dipyridamole was given. A reduction in heart rate and blood pressure (30% to 50%) was maintained during this time frame; heart rate and blood pressure returned toward baseline within 5 minutes of discontinuation of the adenosine infusion (Fig 2⇓). Therefore, the effects of a single dose of dipyridamole, in combination with adenosine infusion, lasted approximately 15 minutes, which was the maximum time required to determine DFT. CPT, a selective A1 adenosine receptor antagonist (5 mg/kg), was shown to rapidly reverse the effects of adenosine and dipyridamole on heart rate (Fig 2B⇓). Although pretreatment with CPT alone had no effect on heart rate or blood pressure, it also markedly attenuated the effects of adenosine and dipyridamole on these parameters (Fig 2C⇓).
Effect of A1 Adenosine Antagonist on Transthoracic Defibrillation Threshold (Protocol 3)
In a subset of dogs from protocol 2A in which adenosine elevated DFT, the effects of CPT (5 mg/kg) were evaluated. The effects of adenosine on DFT were initially determined as described in protocol 2. The sequence was then repeated after pretreatment with CPT, which was infused 10 minutes before adenosine and dipyridamole were infused. Recontrol DFT was determined after a 90-minute washout period of the antagonist (Fig 1⇑). Preliminary experiments in 6 dogs showed that CPT alone (5 mg/kg) had no effect on DFT (<1.0%, P=NS).
Effects of Adenosine on Transthoracic Defibrillation Threshold in Denervated Dogs (Protocol 4)
To determine whether the effects of adenosine were catecholamine-independent or catecholamine-dependent, the effects of adenosine on DFT were determined after surgical (bilateral vagotomy in 1 dog) or pharmacological (atropine 0.04 mg/kg in 4 dogs) vagal denervation and intravenous propranolol, 0.2 mg/kg, in a subset of dogs from protocol 2A in which adenosine elevated DFT. DFT was first determined during control and then during dipyridamole and adenosine infusion as outlined in protocol 2. After autonomic denervation and steady-state heart rate and blood pressure were achieved, control DFT in the denervated dog was determined. Adenosine (300 μg · kg−1 · min−1) and dipyridamole (0.25 mg/kg) were then administered as described in protocol 2, and DFT was determined. Adenosine infusion was then discontinued, and a recontrol DFT was determined after a 30-minute washout period (Fig 1⇑).
Statistical analysis was performed with a two-tailed Student’s t test for paired or unpaired observations. Multiple comparisons within and between groups were analyzed by repeated-measures ANOVA. A value of P<.05 was considered significant. All data are expressed as mean±SEM.
Coronary Sinus and Systemic Plasma Levels of Adenosine, Catecholamines, and Lactate During Ventricular Fibrillation
VF resulted in marked elevation of myocardial adenosine, as reflected indirectly by CS plasma concentration. During sinus rhythm (control), the CS adenosine concentration measured in 10 dogs was 64±16 nmol/L. After 5 minutes of VF, the adenosine concentration increased to 816±257 nmol/L (P=.01) (Fig 3⇓). Plasma concentrations in the systemic circulation also increased in response to VF, but to a lesser degree. During control, the plasma concentration measured from the femoral artery was 58±28 nmol/L, and after 5 minutes of VF, it was 108±46 nmol/L, P=.04.
Lactate levels also increased in response to VF. The control CS level was 22±2 mg/dL, which increased to 38±4 mg/dL during VF, P=.002 (Fig 3⇑). Plasma lactate levels determined from femoral artery samples were not altered by VF.
VF resulted in a marked release of endogenous catecholamines from the myocardium. The CS plasma level of norepinephrine during control was 322±54 pg/mL, and it increased to 7935±1936 pg/mL during VF, P=.004 (Fig 3⇑). Similarly, there was a marked increase in myocardial release of epinephrine during VF. The control level was 290±63 pg/mL, which increased to 28 451±10 308 pg/mL during VF, P=.03. In contrast, systemic increments of endogenous catecholamines were not observed.
Effects of Adenosine on Transthoracic and Transmyocardial Defibrillation Threshold
The effects of adenosine on transthoracic DFT were determined in 21 dogs. Control DFT for the group of dogs receiving isoproterenol was 49±6 J, which increased to 67±7 J during adenosine and dipyridamole infusion, P=.0003. (The control DFT before adenosine and dipyridamole was 52±6 J, and recontrol was 47±6 J, P=NS.) The mean increment in DFT in response to adenosine for each dog was 49±14% (Fig 4A⇓). Nine of 21 dogs had an increment of 50% to 241% in DFT, whereas 7 of 21 dogs had <15% change in DFT. Since pilot studies in 5 dogs showed that after 30 seconds of VF, myocardial effluent concentrations of norepinephrine and epinephrine increased by approximately 4- and 15-fold, respectively, the effects of adenosine and dipyridamole were examined in 11 other dogs that did not receive concurrent isoproterenol infusion. In these dogs, control DFT was 58±11 J, which increased to 76±17 J during adenosine and dipyridamole, P<.0001. There was no statistical difference in the effect of adenosine on DFT between the two groups of dogs that were either infused or not infused with isoproterenol.
The effects of adenosine on transmyocardial DFT were determined in 10 other dogs. Control DFT was 10±3 J, which increased to 19±4 J during adenosine infusion, P=.0003. The mean increment for each dog was 103±16%; range, 24% to 204% (Fig 4B⇑). Adenosine and dipyridamole did not alter transmyocardial resistance. The control transmyocardial resistance during VF was 59±2 Ω, and the resistance during VF when adenosine and dipyridamole were administered was 59±2 Ω, P=NS.
Effects of Adenosine Antagonist (CPT) on Defibrillation Threshold
As indicated above, CPT alone had no effect on DFT; however, it did abolish the adverse effects of adenosine on transthoracic DFT. In a subset of 12 dogs from protocol 2A, adenosine increased DFT from 62±9 J (control) to 89±16 J (adenosine), P=.0001. When these dogs were pretreated with CPT, adenosine had no effect on DFT (Fig 5⇓).
Effects of Adenosine on DFT in Denervated Dogs
In contrast to adenosine’s effects on DFT in innervated dogs, adenosine had no effect on DFT when the same dogs were denervated. Five dogs from protocol 2A were denervated, and the effects of adenosine on DFT were reassessed. Adenosine increased DFT in the innervated condition from 38±8 J (control) to 61±15 J (adenosine), P=.03. However, when these dogs were denervated, adenosine had no effect on DFT, 49±11 J (control) versus 53±10 J (adenosine), P=NS (Fig 6⇓).
The major findings in this study are that (1) myocardial effluent levels of adenosine are significantly increased during VF; (2) adenosine markedly elevates transthoracic and transmyocardial DFT; (3) these adverse effects on threshold are mediated by the A1 adenosine myocardial receptor, since they are reversed by the specific A1 receptor antagonist CPT; and (4) the effects of adenosine on defibrillation are mediated through a catecholamine-dependent (antiadrenergic) mechanism. These results suggest that enhanced myocardial production of adenosine during VF has a deleterious effect on defibrillation and in part mediates the increase in threshold related to VF duration.
Although it is recognized that for a given shock strength, the duration of VF before countershock is the major determinant of defibrillation (that is, the longer the VF duration before shock, the more difficult it is to defibrillate), the mechanism governing elevated defibrillation requirements under these conditions is poorly understood. Respiratory acidosis and alkalosis do not elevate threshold requirements, nor does metabolic acidosis or alkalosis.5 On the contrary, neurohumoral substances, such as catecholamines, that are released during hypoxia have been shown to reduce DFT,9 10 11 whereas β-blockers increase it.9 10
One hypothesis of this study was that the effects of adenosine release during VF are in part responsible for mediating the increase in DFT related to VF duration. Related to this hypothesis, enhanced release of adenosine during VF has been shown to mediate postdefibrillation bradycardia, AV block, hypotension, and myocardial depression,17 18 effects that are reversed by adenosine antagonists.17 18 19 Moreover, aminophylline, a nonselective adenosine antagonist, has been shown to decrease DFT. Although this result was attributed to possible phosphodiesterase inhibition and elevation of cellular cAMP levels,20 this explanation is unlikely, since methylxanthines inhibit phosphodiesterase at levels at least 20-fold greater than those required to antagonize the A1 receptor.21
In this study, we have shown that adenosine has potent adverse effects on DFT, whether evaluated during transmyocardial or transthoracic defibrillation. Preliminary data from a previous study failed to demonstrate any effect of adenosine on DFT. In that study, adenosine (60 μg · kg−1 · min−1) was administered without concurrent infusion of dipyridamole in two dogs.5 Coronary sinus effluent levels of adenosine were not determined, but it is unlikely that adenosine levels were significantly elevated, because DFT was determined 15 seconds after induction of VF, before appreciable endog- enous adenosine is released from the myocardium.
To permit measurement of DFT 30 seconds after induction of VF and yet achieve coronary sinus effluent concentrations of adenosine similar to those observed during cardiac arrest, ie, after 5 minutes of VF, we determined the effluent concentrations of adenosine after 5 minutes of continuous VF. These concentrations were then reproduced during our study by administration of both dipyridamole (0.25 mg/kg IV) and adenosine (300 μg · kg−1 · min−1). These doses produced consistent electrophysiological (sinus slowing) and hemodynamic (decrease in blood pressure) effects in each animal. The CS effluent concentrations of adenosine measured during VF showed a >12-fold increase compared with control measurements during sinus rhythm. In contrast, the other ischemic metabolite measured in CS effluent, lactate, increased by less than twofold. Despite the large increment in effluent concentrations of adenosine achieved in the study, it is likely that our protocol underestimated the true adverse impact of adenosine on DFT. First, adenosine undergoes metabolism during its transit through the sampling catheter.12 Furthermore, during enhanced release of endogenous adenosine (such as occurs during VF, protocol 1), myocardial interstitial concentrations of adenosine are severalfold higher than those in the venous effluent because of the coronary endothelium, which serves as a metabolic and physical barrier to the exchange of adenosine between the interstitium and effluent.22 In contrast, when exogenous adenosine is administered (protocol 2), the concentration gradient of adenosine is inverted (effluent > interstitium)22 because of the uptake and metabolism of adenosine by endothelial cells. Therefore, since infusion of adenosine during defibrillation (protocol 2) was designed to approximate the CS effluent concentration measured after 5 minutes of VF, the interstitial myocardial concentrations of adenosine achieved during protocol 2 were probably less than those actually achieved at 5 minutes of VF, and this factor may account for the finding that adenosine did not adversely affect DFT in all dogs.
The rise in adenosine concentration in CS effluent during VF was paralleled by a rise in norepinephrine and epinephrine effluent levels. We initially considered it necessary to infuse exogenous catecholamines during assessment of the effects of adenosine on DFT (to reproduce conditions at 5 minutes of VF). In 21 dogs, isoproterenol was therefore infused with adenosine and dipyridamole (protocol 2A, Fig 1⇑). Eleven other dogs did not receive isoproterenol. No statistical difference was observed in the effects of adenosine on DFT between the two groups of dogs. We interpreted these results as indicating that sufficient amounts of endogenous catecholamines, necessary to exert its facilitatory effect on DFT, were released after 30 seconds of VF. Therefore, it is important to emphasize that the data for control DFT (protocols 2A and 2B) also reflect the facilitatory effects of endogenous catecholamines on DFT.
The unfavorable effects of adenosine on DFT were observed during both transthoracic and transmyocardial defibrillation. These effects, although unequivocal in both circumstances, were more pronounced during transmyocardial defibrillation. The explanation for this discrepancy is not readily apparent. Differences between the two protocols included a different waveform for defibrillation, damped sine wave (transthoracic) versus trapezoidal wave (transmyocardial), and current-based defibrillation (transthoracic) versus energy-based defibrillation (transmyocardial). In contrast to current-based defibrillation, the energy-based method depends on the resistance of the defibrillation load. However, transmyocardial resistance was equivalent in all phases of protocol 2B, eliminating this consideration as a potential explanation.
The cardiac effects of adenosine are mediated by the cell surface A1 and A2 adenosine receptors. Activation of the A1 receptor in ventricular tissue antagonizes the stimulatory actions of catecholamines on the slow inward calcium current (ICa) and on the transient inward current (ITI).23 This effect is mediated by an inhibitory G-protein that decreases intracellular cAMP. The hypothesis that the effects of adenosine on DFT observed in this study were also mediated by the A1 receptor was supported by the complete reversal of the effects of adenosine by the alkylxanthine CPT, a competitive adenosine antagonist with an A1/A2 affinity of 140.24 25
Although alkylxanthines also inhibit phosphodiesterase and the release of catecholamines from nerve terminals and the adrenal medulla,21 these actions are unlikely to account for the effects of CPT observed in this study, for several reasons. First, as indicated above, phosphodiesterase inhibition occurs at levels far greater than that required to antagonize the A1 receptor.21 Second, as shown in this study, CPT had little or no effect on heart rate (Fig 2C⇑), results consistent with a minimal effect on phosphodiesterase inhibition and catecholamine release. Third, although CPT has some effect on the A2 receptor, partially reversing adenosine- and dipyridamole-induced hypotension, its effects on the A1 receptor are more pronounced (ie, reversing heart rate effects, Fig 2B⇑).
Dipyridamole was used in this study to potentiate the effects of adenosine through its inhibition of cellular adenosine uptake.17 26 Although dipyridamole inhibits prostacyclin and also elevates cAMP levels by inhibiting phosphodiesterase,26 the evidence strongly suggests that the primary effects of dipyridamole in this study were related to its inhibition of adenosine transport. This is supported by the abolition of the effects of dipyridamole (and thus adenosine) on defibrillation threshold by CPT. If dipyridamole resulted in an elevation of cAMP and therefore lowered DFT, the effects of adenosine on threshold observed in this study would have been masked. Finally, possible reflex sympathetic activation due to dipyridamole- (and adenosine-) induced hypotension would have been expected to attenuate an elevation of DFT due to the myocardial effects of adenosine.
To determine the mechanism of action of adenosine on DFT, its effects were determined in the same dog during both the innervated and denervated states. In contrast to its adverse effect on DFT in the innervated condition, its lack of effect during the denervated state indicates that its effects on DFT are probably mediated through an antiadrenergic or cAMP-dependent mechanism. These actions are consistent with the known antiadrenergic effects of adenosine in ventricular myocardium, which include negative inotropy, termination of catecholamine-mediated ventricular tachycardia, and suppression of catecholamine-mediated His-Purkinje system automaticity.27 28
One may question whether adenosine elevates defibrillation threshold or merely abolishes the facilitatory effects of catecholamines on threshold. This is essentially a semantic argument, however. As shown in this study, there is an enhanced release of endogenous catecholamines at 30 seconds of VF. Therefore, a “control” defibrillation trial, which consists of several inductions of ventricular fibrillation and subsequent defibrillation, is performed under conditions consistent with elevated endogenous catecholamine levels. As a result, “control” DFT under these conditions is lower (because of the effects of catecholamines) than a theoretical control threshold determined in the absence of catecholamines. Since this latter condition never occurs under physiological experimental conditions, we believe it is conceptually more meaningful to interpret the effects of adenosine on DFT as those consistent with elevation.
The antiadrenergic effects of adenosine enumerated above are consistent with its role as a homeostatic metabolite that functions to maintain an optimal O2 supply-demand ratio within the myocardium. By functioning as a negative feedback inhibitor of β-adrenergic stimulation, mediated at both prejunctional and postjunctional levels,7 adenosine protects the myocardium against excessive O2 demand. The deleterious effects of adenosine on defibrillation shown in this study, however, are a “paradoxic” consequence of the otherwise protective role associated with the antiadrenergic mechanism of action of adenosine. Thus, it appears that the spectrum and consequences of the myocardial antiadrenergic effects of adenosine are more complex than previously appreciated. The results of this study suggest that release of adenosine during prolonged VF mediates a significant increase of DFT through an antiadrenergic mechanism. Reversal of the effects of adenosine with an A1 receptor antagonist should facilitate defibrillation by lowering threshold.
This work was supported in part by a grant from the National Institutes of Health (RO1-44747) and the Michael Wolk Heart Foundation. Dr Lerman is an Established Investigator of the American Heart Association.
Reprint requests to Bruce B. Lerman, MD, Division of Cardiology, New York Hospital–Cornell Medical Center, 525 E 68th St, Starr Pavilion, 4th Floor, New York, NY 10021.
- Received June 23, 1994.
- Accepted September 23, 1994.
- Copyright © 1995 by American Heart Association
Echt DS, Black JN, Barbey JT, Coxe DR, Cato E. Evaluation of antiarrhythmic drugs on defibrillation energy requirements in dogs. Circulation. 1988;79:1106-1117.
Winkle RA, Mead RH, Ruder MA, Smith NA, Buch WS, Gaudiani VA. Effect of duration of ventricular fibrillation on defibrillation efficacy in humans. Circulation. 1990;81:1477-1481.
Kerber RE, Pandian NG, Hoyt R, Jensen SR, Koyanagi S, Grayzel J, Kieso R. Effect of ischemia, hypertrophy, hypoxia, acidosis, and alkalosis on canine defibrillation. Am J Physiol. 1983;244: H825-H831.
Echt DS, Cato EL, Coxe DR. pH-Dependent effects of lidocaine on defibrillation energy requirements in dogs. Circulation. 1989;80: 1003-1009.
Belardinelli L, Isenberg G. Actions of adenosine and isoproterenol on isolated mammalian ventricular myocytes. Circ Res. 1983;53: 287-297.
Ruffy R, Schechtman K, Monje E. β-Adrenergic modulation of direct defibrillation energy in anesthetized dog heart. Am J Physiol. 1985;248:H674-H677.
Ruffy R, Schechtman K, Monje E. Adrenergically mediated variations in the energy required to defibrillate the heart: observations in closed-chest, nonanesthetized dogs. Circulation. 1986;73:374-380.
Rattes MF, Sharma AD, Klein GJ, Szabo T, Jones DL. Adrenergic effects on internal cardiac defibrillation threshold. Am J Physiol. 1987;253:H500-H506.
Shryock JC, Boykin MT, Hill JA, Belardinelli L. A new method of sampling blood for measurement of plasma adenosine. Am J Physiol. 1990;258:H1232-H1239.
Marbach EP, Weil MH. Rapid enzymatic measurement of blood lactate and pyruvate. Clin Chem. 1967;13:314-325.
Lerman BB, Halperin HR, Tsitlik JE, Brin K, Clark CW, Deale OC. Relationship between canine transthoracic impedance and defibrillation threshold. J Clin Invest. 1987;80:797-803.
Deale OC, Wesley RC Jr, Morgan D, Lerman BB. Nature of defibrillation: determinism versus probabilism. Am J Physiol. 1990;259:H1544-H1550.
Wesley RC Jr, Belardinelli L. Role of endogenous adenosine in postdefibrillation bradyarrhythmia and hemodynamic depression. Circulation. 1989;80:128-137.
Wesley RC Jr, 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. 1993;27:129-133.
Ruffy R, Monje E, Schechtman K. Facilitation of cardiac defibrillation by aminophylline in the conscious, closed-chest dog. J Electrophysiol. 1988;2:450-454.
Fredholm BB. Are methylxanthine effects due to antagonism of endogenous adenosine? Trends Pharmacol Sci. 1980;1:129-132.
Decking UKM, Juengling E, Kammermeier H. Interstitial transudate concentration of adenosine and inosine in rat and guinea pig hearts. Am J Physiol. 1988;254:H1125-H1132.
Song Y, Thedford S, Lerman BB, Belardinelli L. Adenosine-sensitive afterdepolarizations and triggered activity in guinea pig ventricular myocytes. Circ Res. 1992;70:743-753.
Dennis D, Boykin M, Jacobson KA, Belardinelli L. A novel irreversible adenosine antagonist attenuates the negative dromotropic effect of adenosine in guinea pig heart. FASEB J. 1990;4:A1008. Abstract.
Bruns RF, Lu GF, Pugskey TA. Adenosine receptor subtypes: binding studies. In: Becker B, Gerlach E, eds. Topics and Perspectives of Adenosine Research. Berlin, Germany: Springer-Verlag; 1987:59-73.
Harker LA, Kadatz RA. Mechanism of action of dipyridamole. Thromb Res Suppl. 1983;4:39-46.
Lerman BB, Belardinelli L, West GA, Berne RM, DiMarco JP. Adenosine-sensitive ventricular tachycardia: evidence suggesting cyclic AMP-mediated triggered activity. Circulation. 1986;74: 270-280.
Lerman BB, Wesley RC Jr, DiMarco JP, Haines DE, Belardinelli L. Antiadrenergic effects of adenosine on His-Purkinje automaticity: evidence for accentuated antagonism. J Clin Invest. 1988;82: 2127-2135.