| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 University Laboratory of Physiology, Oxford OX1 3PT; and 2 Department of Cardiovascular Medicine, John Radcliffe Hospital, Oxford OX3 9DU, United Kingdom; and 3 Institute for Mathematics and its Applications, University of Minnesota, Minneapolis, Minnesota 55455
 |
ABSTRACT |
|---|
 TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
We characterized the epicardial activation sequence during a norepinephrine (NE)-induced ventricular arrhythmia in anesthetized pigs and studied factors that modulated it. Subepicardial NE infusion caused the QRS complex to invert within a single beat (n = 35 animals, 101 observations), and the earliest epicardial activation consistently shifted to the randomly located infusion site (n = 14). This preceded right atrial activation, whereas the total ventricular epicardial activation time increased from 20 ± 4 to 50 ± 9 ms (P < 0.01). These events were accompanied by a ventricular tachycardia and a drop in left ventricular pressure, which were fully reversed after the infusion was stopped. Epicardial pacing at the infusion site mimicked all electrical and hemodynamic changes induced by NE. The arrhythmia was prevented by propranolol and abolished by cardiac sympathetic or vagal nerve stimulation. Focal automaticity was computationally reconstructed using a two-dimensional sheet of 256 × 256 resistively coupled ventricular cells, where calcium handling was abnormally high in the central region. We conclude that adrenergic stimulation to a small region of the ventricle elicits triggered automaticity and that computational reconstruction implicates calcium overload. Interventions that reduce spatial inhomogeneities of intracellular calcium may prevent this type of arrhythmia.
catecholamines; cardiac mapping; porcine heart; ventricular arrhythmia
| |
INTRODUCTION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
IT IS WELL ESTABLISHED
THAT cardiac sympathetic imbalance can lead to ventricular
arrhythmias, although the underlying cellular electrophysiological
mechanisms are not fully understood. Phase 4 depolarization (abnormal
automaticity) and afterdepolarizations (triggered automaticity)
(22) have been implicated because both can be provoked by
catecholamines (8, 27) or exercise (8) and
prevented or terminated by 
-adrenergic blockade (8) and acetylcholine (27), all of which modulate the levels of
intracellular calcium in ventricular myocytes. Calcium-channel
antagonists also prevent triggered automaticity (8, 27),
whereas they are ineffective in terminating abnormal automaticity
(27). Previous studies (18, 19) have shown
that, in the in vivo nonischemic pig heart, a localized subepicardial
infusion of norepinephrine (NE), the permeant cyclic-AMP analog
dibutyryl-cAMP, or inhibitors of cyclic nucleotide phosphodiesterase
were effective in causing a sustained ventricular arrhythmia, resulting
in depressed hemodynamics. In the clinical situation, high-adrenergic
tone is known to occur around regions of ischemia (11) or
during right ventricular (RV) outflow tract dysfunction
(16), both of which may facilitate spontaneous activity in
the ventricular myocardium (10).
Our aims were threefold: 1) to test whether the hemodynamic changes were a direct consequence of the formation of a new pacemaker at the NE infusion site; 2) to examine whether changing global cardiac autonomic input, through stimulation of the cardiac sympathetic or vagal nerves, modulated the electrical activation sequence during the arrhythmia and whether electrical pacing at the infusion site mimicked the pattern of activation caused by the infusion of NE; and 3) to elucidate the electrophysiological basis underlying this arrhythmia, we attempted to reconstruct it in two dimensions using a network model of 256 × 256 resistively coupled ventricular cells, for which the central region had high-adrenergic tone. Some of the results have been communicated in abstract form (12, 13).
| |
MATERIALS AND METHODS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
The investigation conforms to the "Guide for the Care and Use of Laboratory Animals" published by the National Institutes of Health [DHEW publication no. (NIH) 85-23, revised 1996, Office of Science and Health Reports, DRR/NIH, Bethesda, MD 20205] and under a British Home Office Project License (no. PPL 30/1133).
Anesthesia
Thirty-five domestic pigs of either sex [mean weight, 24 ± 5 (SD) kg] were premedicated with azaperone (5 mg/kg im; Stresnil, Janssen), and a surgical plane of anesthesia was induced with 2-3% halothane (Fluothane, ICI) in 100% oxygen via an Ayre's T piece with a mask and bag. After the surgery, halothane was replaced with a bolus infusion of
-chloralose (100 mg/kg iv; Sigma Chemical) and repeated approximately every 2 h as required. Stainless steel needle electrodes were then inserted subcutaneously into each limb and
into the chest wall to monitor the electrocardiograph (ECG).
Surgery and Instrumentation
A tracheotomy was performed, and a cuffed endotracheal tube (9 mm, Vygon) was introduced ~80 mm into the trachea. The cuff was inflated to provide an airtight seal. Fluid-filled catheters (Portex) were inserted into two femoral arteries and one femoral vein for sampling of arterial blood, monitoring of arterial blood pressure (ABP), and intravenous infusion of anesthetic and fluids, respectively. Artificial ventilation (Oxford Mark II ventilator, Penlon) was used to maintain arterial carbon dioxide tension between 35 and 45 Torr by adjusting tidal volume and frequency as required. Animals were ventilated with a hyperoxic gas mixture throughout the experiment. A midline thoracotomy was performed, and the chest was retracted laterally. After incision of the pericardium, the heart was suspended in a pericardial cradle. For all experiments, the cranial projections of the vagi were cut.In the first group of experiments (group A, n = 21), a catheter (Wallace Y-can, 19 gauge) was inserted into the left ventricle (LV) via the apical dimple and secured. The cranial projections of the cardiac sympathetic nerves were ligated and cut. Communicating branches to the thoracic sympathetic chain were also cut.
In the second group of animals (group B, n = 14), a Millar catheter (Mikro-Tip, TC-510 control unit; Millar Instruments, Houston, TX) pressure sensor was inserted into the LV via the apical dimple. An elasticized sock containing 127 unipolar stainless steel electrodes (Biomedical Instruments Designers, Montreal, Quebec) was slipped over the ventricles. In addition, one custom-built silver electrode was attached to the right atrium with the use of continuous suction.
Intensive Care
After completion of surgery, at least 1 h elapsed before experiments were begun. The thoracic cavity was covered by plastic film to reduce cooling. Core body temperature was measured by a rectal thermistor and maintained at 38 ± 1°C by heating lamps beneath the operating table. Fluid was replaced by an intravenous sterile saline drip (~100 ml/h). Arterial blood samples (0.3 ml) were regularly withdrawn into preheparinized syringes and analyzed for pH, blood gases, and electrolytes (Na+, K+, Ca2+) (Radiometer ABL505, Copenhagen, Denmark). For all animals (n = 35), mean values for arterial pH, arterial CO2 tension, and arterial O2 tension were 7.44 ± 0.07 (SD), 41 ± 6 Torr, and 319 ± 86 Torr, respectively. Respiratory acidosis was corrected by adjusting the ventilator. Metabolic acidosis was corrected by intravenous infusion of 4.2% sodium bicarbonate solution as required.Hemodynamic Measurements
ABP was measured by using a saline-filled pressure transducer (SensoNor 840) calibrated in the midaxillary line. For group A, left ventricular pressure (LVP) was also measured in this way, and, for group B, LVP was measured directly by using the Millar catheter. All signals were recorded onto a penwriter (MT8P, Lectromed). The analog inputs were passed to a real-time data-acquisition system (MP 100, Biopac Systems) employing Acqknowledge 3.0 software for the Macintosh (Macintosh Quadra 950). Heart rate (HR) and the rate of change of LVP (LVdP/dt) were computed using this software.Epicardial Electropotential Mapping
For the experiments in group B, 63 of the sock electrodes and the right atrial electrode were connected to a 64-channel Unemap cardiac mapping system (Uniservices, Auckland, New Zealand). The selected sock electrodes spanned the whole of the ventricular epicardium, with an interelectrode spacing of ~15 mm. The animal and all cable shields were earthed to the system ground (computer earth potential). Unipolar epicardial electropotentials (referred to the system ground) were continuously sampled at 900 Hz with a rolling buffer of 8 s and could be digitally stored for subsequent analysis. During three of these experiments, two-dimensional echocardiography (HP SONOS 5500) was used to estimate LV ejection fraction.Protocols
In all animals (n = 35), ventricular arrhythmias were induced by regional subepicardial infusions of NE bitartrate (10 µmol/l in 0.9% saline, 150 µl/min). The concentration and rate of infusion were chosen because they elicited reproducible arrhythmias within 1-5 min after the onset of the infusion. Infusion was through a 25-gauge hypodermic needle to which was attached a 3FG polyvinyl chloride catheter (~150 mm long) and a three-way tap. The needle was inserted 1-5 mm at randomly selected locations throughout the ventricular subepicardium. Solutions were infused by using a three-syringe microinjection pump (CMA/100). In cases in which subepicardial infusion of NE did not elicit a ventricular arrhythmia within 5 min (~20% of infusions), the needle was removed, and infusion was begun in another part of the ventricle. Susceptibility to the arrhythmia did not depend on the choice of infusion site.Group A (n = 21). The vehicle (saline) was infused (150 µl/min) subepicardially in all animals for ~10-15 min.
To test whether the arrhythmia was due to-adrenergic stimulation
alone, propranolol (0.1-0.25 mg/kg iv; Inderal, ICI) was administered during the NE-induced ventricular arrhythmia
(n = 8). The specific -agonist isoproterenol (1 µmol/l, 150 µl/min, n = 2) or the -agonist
phenylephrine (1 and 10 µmol/l, 150 µl/min, n = 2) was also infused subepicardially in the absence of
-blockade.
To test whether the arrhythmia could be modulated by global autonomic
input, cardiac sympathetic nerve stimulation (5 Hz, 5 V, 3-ms width,
30-s duration; caudal limb of the right ansa subclavia,
n = 6) or vagal nerve stimulation (10 Hz, 10 V, 3-ms width, 30-s duration; left vagus, n = 2; right vagus,
n = 4) was undertaken during the NE-induced ventricular
arrhythmia. Each period of stimulation lasted for 30 s and was
followed by a recovery period, while the infusion of NE was maintained.
Pre- and postarrhythmia stimulations were also performed to test the
stability of the nerves. In addition, NE was infused intravenously (1 µg · kg
1 · min1 for 1 min) during stable arrhythmias (n = 3). Control
intravenous infusions of NE were also performed before subepicardial NE infusion.
Group B (n = 14). Ventricular electropotential mapping was performed for control, NE-induced ventricular arrhythmia, and recovery states. In addition, a single electrode was used to detect right atrial activation (n = 10).
A custom-built roving bipolar electrode was used to pace hearts (0.5-1.5 V, 2-ms width, 30-s duration, n = 3) at the same site in which NE had previously been infused. The pacing needle was inserted ~2 mm subepicardially, and the pacing rate was matched to the HR observed during the previous NE-induced arrhythmia. Epicardial surface (n = 2) or transesophageal (n = 1) echocardiography was used to estimate LV ejection fraction during the ventricular arrhythmias.Electropotential Signal Analysis
With reference to Fig. 1, epicardial electropotential signal analysis was performed as follows. After epicardial signal acquisition using the elasticized electrode sock (Fig. 1A), the electropotentials for a single heart cycle were identified from the 8 s of recorded data, as shown in Fig. 1B. The most negative electropotential slope was computed for each signal (Fig. 1C), and this was chosen to mark the instant at which the associated sock electrode detected excitation of its adjacent portion of epicardium. An event marker was used to identify the time of excitation for each signal.
|
The set of event markers was then used to order the signals, such that the signal that detected earliest activation was listed first (Fig. 1C). For this earliest signal, the marker time was referred to as the datum (red dashed line in Fig. 1C). Activation times were then computed for all other signals by taking the difference between their marker times and the datum.
Two-dimensional Activation Maps
From the known topographical organization of the sock electrodes, estimated epicardial electrode locations were plotted by using the two-dimensional Hammer projection (20), for which the LV and RV free walls constitute the central and lateral portions, respectively (see Fig. 1D). The left anterior and posterior descending coronary arteries were superimposed for anatomic clarity. Nodal activation times were computed from the set of spatially arranged electrode activation times using a two-dimensional finite element-fitting technique, which incorporated linear least squares (6). Red portions of this smoothly continuous two-dimensional activation map highlight regions of earliest epicardial excitation, whereas blue denotes regions of latest epicardial excitation. The spatial resolution of the electrodes was considered sufficient during control and arrhythmic states, because doubling the electrode density (i.e., halving the interelectrode spacing) over one-half of the epicardial surface produced a similar two-dimensional activation pattern and range of times for that region (not shown).Epicardial Activation Sequence Rendered in Three Dimensions
A further technique was developed to produce epicardial activation maps with anatomic realism. The epicardial activation sequence was reconstructed by fitting a scalar field of activation times onto the epicardial surface of an anatomically accurate mathematical model of the canine ventricles (14) rendered in three dimensions, as shown in Fig. 1E. The estimated three-dimensional locations of the electrodes were projected onto the epicardial surface of the anatomic model, and the epicardial activation time field was approximated by using linear interpolation of a set of parameters defined at spatial locations that coincided with the finite element nodes used to define the geometry. As for the two-dimensional activation map, nodal epicardial activation times were computed by using a standard finite element-fitting technique.Computer Modeling
Abnormal intracellular calcium handling has been implicated in the onset of this arrhythmia (18). In an attempt to simulate this, we modified a recent model of the electrical activity in ventricular myocytes (15) to include effects of NE on cell membrane properties. The effect of NE was modeled by using a fivefold elevation in the conductances of the L- type calcium channel of the cell sarcolemmal membrane (from 0.25 to 1.25 µA · l · mmol1) and the sarcoplasmic
reticulum (SR) calcium-ATPase uptake pump (from 1 to 5 mmol · l1 · s1). This is
consistent with changes in calcium currents due to isoproterenol that
have been reported in the literature (1). Action potential
propagation was studied by using a two-dimensional sheet of 256 × 256 cells isotropically interconnected via linear resistors to
represent the intercellular gap junctions. To simulate the spatial
decrease in NE concentration due to ligand diffusion and metabolism,
the upregulation factor for the L-type and SR uptake calcium
conductances was chosen to decrease according to a two-dimensional
normal distribution with a variance of 60 cell lengths. In this way,
conductance changes were negligible at a distance of 90 cell lengths
from the center, consistent with the reported zone of increased cAMP
levels surrounding intramyocardial NE infusion sites (18).
Action potentials were initiated by stimulating one edge of the sheet
at a rate of 1 Hz with a twice-threshold stimulus for 2 ms. A control
simulation with normal intracellular calcium handling was undertaken
for the same number of cycles. The ionic currents of a single-model
myocyte were also investigated to elucidate a possible cellular
mechanism for the NE-induced arrhythmia.
Cellular simulations were initially performed by using a Silicon Graphics Power Challenge XL (12 symmetric MIPS 90-MHz R8000 processors; 1-GB physical DRAM) and subsequently by using a Silicon Graphics Cray Origin2000 (84 symmetric MIPS 195-MHz R10000 processors; 21-GB physical DRAM; Silicon Graphics, Mountain View, CA). The parallel computations incorporated a variable-step, fourth-order, Runge-Kutta numerical integration method, as described previously (26).
Statistical Analysis
All data are expressed as means ± SD. Mean arterial blood pressure (MAP), LVP, HR, LVdP/dt, and activation times for control, arrhythmic, and recovery states were compared by using a one-way analysis of variance with repeated measures. Post hoc comparisons were performed by using Dunnett's method, and statistical significance was accepted at P < 0.05.| |
RESULTS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
NE-induced Ventricular Arrhythmia
Infusion of NE into the ventricular subepicardium elicited a reproducible ventricular arrhythmia on 101 separate occasions in 35 animals. Within 1-5 min after the start of the infusion, there was a marked change in the amplitude and direction of the QRS complex, which masked the P wave (Fig. 2). The abnormal electrical activity was accompanied by a ventricular tachycardia (control, 126 ± 18 beats/min; arrhythmia, 153 ± 20 beats/min; P < 0.01, n = 35) and a profound hemodynamic depression (see Figs. 2 and 3). Restoration of normal sinus rhythm and hemodynamic performance occurred within 10 min after the subepicardial infusion of NE was stopped. Recovery was accompanied by a small, transient overshoot of all hemodynamic variables (Fig. 3), which was significant compared with control (P < 0.05; n = 35). A subsequent ventricular arrhythmia could be established by using the same NE infusion site.
|
|
Sustained ventricular arrhythmia was also elicited by subepicardial
infusion of the -agonist isoproterenol (n = 2), but
could not be provoked with subepicardial infusions of the -agonist phenylephrine (n = 2) or the saline vehicle
(n = 21), or by similar infusions of NE administered
intravenously. After a stable arrhythmia was established, intravenous
administration of a bolus of propranolol (n = 8)
abolished it within ~2 min. After this, the hemodynamic performance
quickly returned to its prearrhythmic state. It was not possible to
elicit the arrhythmia in the -blocked animal.
Sympathetic nerve stimulation.
Stimulation of the right cardiac sympathetic nerve abolished the
ventricular arrhythmia within 10.6 ± 7.5 s (17 observations in 6 animals). Sympathetic nerve stimulation caused the QRS complex to
revert to its control form and the HR to increase by 90 ± 16 beats/min above the arrhythmic rate. No recurrence of the arrhythmia was seen during any 30-s period of sympathetic stimulation. However, the arrhythmia and hemodynamic depression were reestablished 47 ± 16 s after stimulation was stopped. Infusion of NE (1 µg · kg1 · min1 iv) for 1 min reversed the arrhythmia within 35 ± 21 s (5 observations in 3 animals). As observed during sympathetic stimulation, the QRS
complex reverted, and all hemodynamic variables were elevated (HR
increased by 42 ± 34 beats/min above the arrhythmic rate). Recurrence of the arrhythmia took place 30 ± 28 s after the
infusion was stopped.
Vagal nerve stimulation.
Stimulation of the left (5 observations in 2 animals) or right (13 observations in 4 animals) vagus nerves reversed the ventricular arrhythmia after 7.4 ± 4.9 s. Although control vagal
stimulation reduced all hemodynamic variables, vagal stimulation during
a NE-induced arrhythmia caused small increases in ABP, LVP, and LVdP/dt (see Fig. 4), whereas
the HR decreased by 46 ± 11 beats/min after the termination of
the arrhythmia. After the stimulation was stopped, hemodynamics
continued to increase in the presence of normal electrical activity
(Fig. 4) for 31 ± 9 s until the arrhythmia was
reestablished.
|
Epicardial Electropotential Mapping
During the NE-induced arrhythmia for the group B experiments, right atrial activation followed the earliest ventricular epicardial activation (20 observations in 10 animals). Figure 5 illustrates the onset of one arrhythmia. The first two cycles show control conditions, in which right atrial excitation preceded ventricular activation. After the ventricular repolarization of the second cycle, a spontaneous ventricular depolarization was observed. Subsequently, earliest ventricular epicardial activation preceded right atrial activation on a one-to-one basis until after the NE infusion was stopped.
|
Figure 6 illustrates epicardial
activation maps during one experiment, for which a ventricular
arrhythmia was elicited after ~90 s of NE infusion. Under control
conditions, the earliest epicardial activation occurred near the
equatorial region of the RV and the latest epicardial activation
occurred in the posterobasal region. In this experiment, NE was infused
near the apex of the posterior LV wall, and, during the ensuing
arrhythmia, the location of earliest epicardial activation closely
matched the NE infusion site. This change in the cardiac activation
sequence caused the QRS complex to invert within a single beat and
resulted in a marked depression of the ABP. Within 2 min of the stop of
the NE infusion, the cardiac activation sequence, ECG, and hemodynamics
were restored to their control states.
|
The events described above were observed consistently for all
group B experiments (27 observations in 14 animals). The
group data (see Fig. 7) showed that the
total ventricular epicardial activation time (defined as the time taken
for the wave of excitation to spread from earliest to latest epicardial
activation sites) significantly increased from 20 ± 4 ms under
control conditions to 50 ± 9 ms (P < 0.01, n = 14) during the NE-induced ventricular arrhythmia.
During all 27 NE-induced arrhythmias, earliest epicardial activation
was detected within one electrode spacing (~15 mm) of the needle
site. The MAP significantly decreased from 62 ± 13 mmHg (control)
to 43 ± 10 mmHg during the arrhythmia (P < 0.01, n = 14). The total ventricular epicardial activation
time and MAP recovered to control values after the NE infusion was
stopped.
|
Epicardial electrical stimulation.
Ventricular epicardial pacing at the NE infusion site mimicked all
electromechanical changes observed during the NE-induced ventricular arrhythmia. Figure 8 shows
activation maps obtained during one experiment, for which NE was
infused into a LV equatorial site. The subsequent arrhythmia resulted
in an increase in the total ventricular epicardial activation time from
21 ms (control) to 49 ms. After a full recovery, the needle was
withdrawn, and the epicardium was electrically stimulated at the same
location with a similar rate to that observed during the NE-induced
arrhythmia. The total ventricular epicardial activation time for the
pacing-induced arrhythmia increased from 21 ms (control, not shown) to
54 ms, and the activation maps for the NE- and pacing-induced
arrhythmias were qualitatively similar (see Fig. 8). In two subsequent
experiments, epicardial activation times increased from 14 and 18 ms
(control) to 39 and 51 ms for the NE-induced arrhythmias and to 39 and
46 ms for the pacing-induced arrhythmias. The corresponding decreases in MAP were 15, 36, and 28 mmHg, respectively, for the NE-induced arrhythmias and 16, 31, and 27 mmHg, respectively, for the
pacing-induced arrhythmias. The morphologies of the lead II ECG QRS
complex were similar during corresponding NE- and pacing-induced
ventricular arrhythmias.
|
Echocardiographic studies. Pacing-induced arrhythmias resulted in estimated ejection fraction decreases from 64, 52, and 50% (controls) to 20, 38, and 33%, respectively. Echocardiographic recordings were taken during two NE-induced arrhythmias, for which the estimated ejection fractions decreased from 60 and 55% (controls) to 30 and 30%, respectively. In all cases, mitral and aortic valve function remained normal.
Computer Modeling
At the beginning of the two-dimensional cellular network model simulation, the stimulated wave of electrical excitation proceeded "normally" from left to right, as illustrated in Fig. 9A. Several cycles later, the central region of NE-affected cells spontaneously depolarized before the normal wave of activation (Fig. 9B). A further seven cycles later, triggered automaticity was fully established, and the activation sequence was completely dominated by the central zone, which acted as an abnormal pacemaker with a rate of ~2 Hz (Fig. 9C). Spontaneous depolarization did not occur during the control simulation (not shown), for which calcium handling was unchanged.
|
Alternative simulations with 1) just the L-type calcium conductance increased and not the SR uptake calcium conductance, 2) the L-type and SR uptake calcium conductances upregulated by a factor of two instead of five, and 3) the normal pacemaker reduced from 1 to 0.5 Hz also reproduced triggered automaticity, although the time to onset was greater in all three cases (simulation results not shown). The size of the NE-affected zone in the network model was varied to determine whether this altered arrhythmogenesis. NE-affected zones smaller than that described above did not result in arrhythmia, whereas larger zones did during the same simulated time period. Moreover, when the overall size of the network was increased (while the size of the central zone was kept constant), characteristics of the simulated arrhythmias remained unchanged, ruling out any mathematical edge effects on the required size of the NE-affected zone to elicit arrhythmia. To test whether other effects of NE could produce triggered automaticity, an alternative model simulation that incorporated modulation of the slow component of the delayed rectifier current and the sodium/potassium exchanger current instead of the calcium conductances was also performed (not shown). This model did not result in arrhythmia over the same simulated time period.
Analysis of the ionic behavior for a single-model cell with upregulated
calcium conductances revealed a significant increase in the peak
cytosolic calcium concentration and SR calcium release flux compared
with the control case, as illustrated in Fig.
10. This led to spontaneous calcium
release from the SR, which caused the sodium-calcium exchanger to admit
sodium in an attempt to remove the excess calcium. The effect of this
inward sodium current was to depolarize the cell beyond threshold.
Spontaneous beating was sustained as long as intracellular calcium
remained high.
|
| |
DISCUSSION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
The new findings of this work are as follows. 1) Characterization of the epicardial electrical events during a ventricular arrhythmia, induced by a unifocal infusion of NE directly into the subepicardium, revealed activation patterns consistent with a region of triggered automaticity near the infusion site, which could be comprehensively mimicked by epicardial pacing. 2) Stimulation of the right cardiac sympathetic or left or right cardiac vagal nerves abolished the NE-induced ventricular arrhythmia. 3) Two-dimensional computational reconstruction of the NE-induced arrhythmia replicated triggered automaticity due to calcium overload, which caused the sodium-calcium exchanger to act as an abnormal pacemaker.
Electromechanical Characterization of the NE-induced Ventricular Arrhythmia
We elicited a stable ventricular arrhythmia through subepicardial infusion of NE or by using epicardial electrical stimulation and compared the observed epicardial activation sequence with that obtained during control conditions. The location of NE infusion was randomly chosen throughout the exposed ventricles. During all sustained NE-induced arrhythmias, the zone of earliest ventricular epicardial activation consistently matched the NE infusion site, and there was a significant increase in the total ventricular epicardial activation time compared with the control state. This was associated with a drop in cardiac output and ABP. Epicardial electrical stimulation near NE infusion sites that had previously been used to elicit ventricular arrhythmias closely mimicked the electrical events and hemodynamic changes associated with the NE-induced arrhythmias. Epicardial activation sequences and times were similar for the NE- vs. pacing-induced arrhythmias, as were the decreases in ABP, LVP, LVdP/dt, and LV ejection fraction, consistent with the notion that the region of tissue adjacent to the infusion site became the dominant pacemaker.We chose a high concentration of NE (10 µmol/l in saline at 150 µl/min) because it elicited an arrhythmia within 1-5 min from the start of the infusion. Podzuweit et al. (18,
19) used the same concentration of NE in normal Tyrode infused
at 10-20 µl/min and established arrhythmias within 2 min. They
also reported arrhythmias with 1 µmol/l NE. We could not elicit
reproducible arrhythmias using the reduced infusion rate, and this was
most likely due to the absence of calcium in our infusate. To determine the effective region of influence of NE on the ventricular myocardium, Podzuweit (18) measured levels of cAMP in small
ventricular tissue cores (~4-mm diameter and 6-mm length) at the
onset of the arrhythmias and found that myocardial cAMP was ~45%
greater in the 4-mm core at the infusion site compared with the
adjacent control tissue cores. The localized inhomogeneous nature of
the NE infusion was critical in eliciting this arrhythmia because similar concentrations administered intravenously did not elicit the
sustained ventricular arrhythmia. Furthermore, reversal of the
arrhythmia by intravenous NE administration (1 µg · kg1 · min1) is most
likely due to the overdrive pacing effect of the sinoatrial node. This
illustrates the requirement for localization of the effects of NE to
the ventricular myocardium to elicit the sustained ventricular
arrhythmia. The subepicardial infusion of NE was unlikely to have
permanently damaged the affected cells, because the arrhythmia was
fully reversible and could be elicited by using the same needle site on
successive occasions with an interim period of full recovery of the
activation sequence and hemodynamics.
Modulation of the NE-induced Ventricular Arrhythmia
The arrhythmia was also reproduced by subepicardial infusion of the specific-agonist isoproterenol but not the -agonist phenylephrine, whereas intravenous infusion of propranolol abolished the ectopic activity, consistent with a role for modulation of cAMP and
calcium currents in arrhythmogenesis. Indeed, Podzuweit (18) elicited similar ventricular arrhythmias in the pig
using intramyocardial infusion of a membrane-permeable analog of cAMP, inhibitors of cyclic nucleotide phosphodiesterases, or NE. In that
study, the NE-induced ectopic activity was abolished by the coinfusion
of choline esters or calcium antagonists, but not by tetrodotoxin.
These observations strongly implicate a -adrenergic, cAMP modulation
of intracellular calcium that leads to spontaneous depolarization of
the ventricular myocardium. Catecholamines are known to enhance
automaticity in Purkinje fibers (17), although this is not
a unique property of specialized conducting fibers, because
automaticity can also occur in isolated, perfused ventricular tissue (7). Moreover, catecholamines can induce
afterdepolarizations in normal canine ventricular myocytes
(21) and reentry in the ventricle (9).
Therefore, we cannot exclude the possibility that spontaneous discharge
may have elicited a premature stimulus to initiate ventricular reentry,
although our results showed a regular cycle length and stationary site
of earliest epicardial activation during the NE-induced arrhythmias.
The results of the present study show that stimulation of the right cardiac sympathetic nerves and intravenous infusion of NE can abolish the adrenergic-induced ventricular arrhythmia. Both interventions increased HR and facilitated the reestablishment of the sinoatrial node as the dominant pacemaker. Left- and right-sided vagal stimulation also abolished the arrhythmia (see the ECG trace in Fig. 4); however, ABP did not fully recover because of the decrease in HR caused by vagal stimulation itself. Vagal stimulation took ~7 s to terminate the arrhythmia, which is more characteristic of responses involving the indirect modulation of ion channels by intracellular messengers rather than the rapid activation of the acetylcholine-stimulated outward potassium current that can decrease HR within one or two heartbeats (4). The indirect effect of cholinergic activation, which is known to cause a decrease in the L-type calcium current in ventricular cells via the inhibition of adenylate cyclase-cAMP system (5) and via the nitric oxide-cGMP pathway (3, 25), is implicated in the antiarrhythmic property of vagal stimulation. Therefore, the antiarrhythmic effects of sympathetic and vagal nerve stimulation may act via two different mechanisms: the former directly restoring the natural site of pacemaking, and the latter involving a reduction in calcium currents and diastolic intracellular calcium in the ventricle.
Computational Reconstruction of the NE-induced Arrhythmia
Abnormal modulation of intracellular calcium is thought to be central to the NE-induced arrhythmia. First, the arrhythmia is abolished by-adrenergic blockers (Ref. 18,
see also RESULTS), calcium-channel
antagonists, and choline esters (18). Second, it is
initiated by permeant cAMP analogs (18). Therefore, we attempted to mimic aspects of this arrhythmia by altering calcium handling in a two-dimensional computational network of excitable cardiac myocytes. These changes led to intracellular calcium overload, which in turn caused an abnormal influx of sodium through the sodium-calcium exchanger. This inward current depolarized the model
cells beyond threshold and acted as an abnormal pacemaker (8). This suggests a possible cellular mechanism for the
focal automaticity observed during subepicardial NE infusion.
Study Limitations
The preparation used for the present study was an open-chest, anesthetized pig. This altered the global ECG and could, therefore, have affected the magnitudes of the recorded epicardial electropotentials compared with the closed-chest animal. However, we recently found that the pattern of epicardial activation was not significantly altered after the chest had been reclosed (unpublished observation).Perhaps the biggest limitation of this study is that the activation sequence was deduced from just epicardial electropotential recordings. For this reason, we cannot identify which specific cell types are responsible for the abnormal pacemaking during the NE infusion. Indeed, intracellular iontophoresis of cAMP enhances phase 4 depolarization and increases the rate of contraction in spontaneously active Purkinje fibers (24). In addition, the pig heart is known to host Purkinje tissue throughout the ventricular wall (23). Therefore, it is possible that the NE infusion enhanced Purkinje cell automaticity to elicit the arrhythmia. However, this is unlikely in the present preparation, because Purkinje fiber abnormal automaticity is unaffected by calcium-channel antagonism (8), whereas triggered automaticity is terminated by verapamil (8, 18, 27). Endocardial-to-epicardial ventricular cell heterogeneity may also play a role in the genesis of this arrhythmia, although this is difficult to determine because the electrophysiology of the various ventricular cell types in the pig is not well characterized.
Whereas our computational reconstruction implicates abnormal calcium handling in the central zone of the cellular network model, it is possible that automaticity could have arisen from a time-dependent decay of the potassium current (17), which has been shown to be modulated by catecholamines (2), coupled with a progressive activation of the L-type calcium-channel current. However, model simulations that incorporated modulation of the slow component of delayed rectifier current and the sodium/potassium exchanger current instead of the calcium conductances showed comparatively smaller effects. Other limitations of the computational model that prevent a direct comparison between the simulation results and in vivo measurements include species differences (guinea pig computational model vs. in vivo pig), topography (2-dimensional cellular network model vs. 3-dimensional in vivo ventricles), and effects of anisotropy (isotropic intercellular resistances vs. anisotropic in vivo conductances). Nevertheless, the computational model demonstrates that triggered automaticity can be reproduced by considering a small region of calcium-overloaded ventricular tissue.
The clinical correlates of this arrhythmia remain to be determined, although pathological circumstances can lead to high-sympathetic drive to a localized region of myocardium (10, 11, 16). We conclude that the ventricular arrhythmia described in this study is a likely consequence of spatial inhomogeneities in the concentration of NE in the ventricular myocardium that lead to triggered automaticity and result in depressed hemodynamics. Interventions that reestablish the sinoatrial node as the dominant pacemaker or that reduce regional inhomogeneities in NE may prevent or terminate this type of arrhythmia.
| |
ACKNOWLEDGEMENTS |
|---|
We thank Chris Hirst and Vivienne Harris for technical assistance and Dr. A. Banning for performing the echocardiography.
| |
FOOTNOTES |
|---|
This study was supported by British Heart Foundation Programme Grant BG 95003. Computational simulations were made possible by support from the Oxford Supercomputing Centre of Oxford University and the Wellcome Trust, which provided access to two Silicon Graphics high-performance parallel computers.
Address for reprint requests and other correspondence: M. P. Nash, Univ. Laboratory of Physiology, Parks Road, Oxford OX1 3PT, UK (E-mail: martyn.nash{at}physiol.ox.ac.uk).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Received 20 June 2000; accepted in final form 11 August 2000.
| |
REFERENCES |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
1.
Bean, BB.
-Adrenergic modulation of cardiac calcium channel gating.
In: Ion Channels in the Cardiovascular System: Function and Dysfunction, edited by Spooner PM,
and Brown AM.. Armonk, NY: Futura, 1994, p. 237-252.
2.
Giles, W,
Nakajima T,
Ono K,
and
Shibata EF.
Modulation of the delayed rectifier K+ current by isoprenaline in bull-frog atrial myocytes.
J Physiol (Lond)
415:
233-249,
1989
3.
Han, X,
Kobzik L,
Zhao YY,
Opel DJ,
Liu WD,
Kelly RA,
and
Smith TW.
Nitric oxide regulation of atrioventricular node excitability.
Can J Cardiol
13:
1191-1201,
1997[ISI][Medline].
4.
Hartzell, HC.
Regulation of cardiac ion channels by catecholamines, acetylcholine and second messenger systems.
Prog Biophys Mol Biol
52:
165-247,
1988[ISI][Medline].
5.
Hescheler, J,
Kameyama M,
and
Trautwein W.
On the mechanism of muscarinic inhibition of the cardiac Ca2+ current.
Pflügers Arch
407:
182-189,
1986[ISI][Medline].
6.
Hunter, PJ,
Smaill BH,
Nielsen PMF,
and
Le Grice IJ.
A mathematical model of cardiac anatomy.
In: Computational Biology of the Heart, edited by Panfilov AV,
and Holden AV.. West Sussex, UK: Wiley, 1997, p. 171-215.
7.
Imanishi, S,
and
Surawicz B.
Automatic activity in depolarized guinea pig ventricular myocardium. Characteristics and mechanisms.
Circ Res
39:
751-759,
1976[Abstract].
8.
Lerman, BB.
Response of nonreentrant catecholamine-mediated ventricular tachycardia to endogenous adenosine and acetylcholine. Evidence for myocardial receptor-mediated effects.
Circulation
87:
382-390,
1993[ISI][Medline].
9.
Lerman, BB,
Stein KM,
Markowitz SM,
Mittal S,
and
Slotwiner D.
Catecholamine facilitated reentrant ventricular tachycardia: uncoupling of adenosine's antiadrenergic effects.
J Cardiovasc Electrophysiol
10:
17-26,
1999[ISI][Medline].
10.
Mandel, WJ.
Sustained monomorphic ventricular tachycardia.
In: Cardiac Arrhythmia: Mechanisms, Diagnosis, and Management, edited by Podrid PJ,
and Kowey PR.. London: Williams & Wilkins, 1995, p. 919-935.
11.
Mertes, PM,
el Abbassi K,
Jaboin Y,
Michel C,
Beck B,
Pinelli G,
Carteaux JP,
Villemot JP,
and
Burlet C.
Consequences of coronary occlusion on changes in regional interstitial myocardial neuropeptide Y and norepinephrine concentrations.
J Mol Cell Cardiol
28:
1995-2004,
1996[ISI][Medline].
12.
Nash, MP,
Thornton JM,
Banning AP,
and
Paterson DJ.
3D epicardial activation sequence during a ventricular arrhythmia in the anesthetized pig (Abstract).
J Physiol (Lond)
513.P:
162P,
1998.
13.
Nash, MP,
Thornton JM,
and
Paterson DJ.
Characterization of the epicardial electrical activation sequence during a ventricular arrhythmia in the anesthetized pig (Abstract).
J Physiol (Lond)
507.P:
60P,
1998.
14.
Nielsen, PM,
Le Grice IJ,
Smaill BH,
and
Hunter PJ.
Mathematical model of geometry and fibrous structure of the heart.
Am J Physiol Heart Circ Physiol
260:
H1365-H1378,
1991
15.
Noble, D,
Varghese A,
Kohl P,
and
Noble P.
Improved guinea-pig ventricular cell model incorporating a diadic space, IKr and IKs, and length- and tension-dependent processes.
Can J Cardiol
14:
123-134,
1998[ISI][Medline].
16.
Palileo, EV,
Ashley WW,
Swiryn S,
Bauernfeind RA,
Strasberg B,
Petropoulos AT,
and
Rosen KM.
Exercise provocable right ventricular outflow tract tachycardia.
Am Heart J
104:
185-193,
1982[ISI][Medline].
17.
Pappano, AJ,
and
Carmeliet EE.
Epinephrine and the pacemaking mechanism at plateau potentials in sheep cardiac Purkinje fibers.
Pflügers Arch
382:
17-26,
1979[ISI][Medline].
18.
Podzuweit, T.
Catecholamine-cyclic-AMP-Ca2+-induced ventricular tachycardia in the intact pig heart.
Basic Res Cardiol
75:
772-779,
1980[ISI][Medline].
19.
Podzuweit, T,
Binz KH,
Nennstiel P,
and
Flaig W.
The anti-arrhythmic effects of myocardial ischaemia. Relation to reperfusion arrhythmias?
Cardiovasc Res
23:
81-90,
1989[ISI][Medline].
20.
Raisz, E.
Principles of Cartography. New York: McGraw-Hill, 1962.
21.
Sosunov, EA,
Anyukhovsky EP,
Shvilkin A,
Hara M,
Steinberg SF,
Danilo P, Jr,
Rosen MR,
Moise NS,
Merot J,
Probst V,
Charpentier F,
Legeay Y,
and
Le Marec H.
Abnormal cardiac repolarization and impulse initiation in German shepherd dogs with inherited ventricular arrhythmias and sudden death.
Cardiovasc Res
42:
65-79,
1999
22.
Tomaselli, GF,
and
Marban E.
Electrophysiological remodelling in hypertrophy and heart failure.
Cardiovasc Res
42:
270-283,
1999
23.
Tranum-Jensen, J,
Wilde AAM,
Vermeulen JT,
and
Janse MJ.
Morphology of electrophysiologically identified junctions between Purkinje fibers and ventricular muscle in rabbit and pig hearts.
Circ Res
69:
429-437,
1991[Abstract].
24.
Tsien, RW.
Adrenaline-like effects of intracellular iontophoresis of cyclic AMP in cardiac Purkinje fibres.
Nat New Biol
245:
120-122,
1973[ISI][Medline].
25.
Wahler, GM,
and
Dollinger SJ.
Nitric oxide donor SIN-1 inhibits mammalian cardiac calcium current through cGMP-dependent protein kinase.
Am J Physiol Cell Physiol
268:
C45-C54,
1995
26.
Winslow, RL,
Varghese A,
Noble D,
Adlakha C,
and
Hoythya A.
Generation and propagation of ectopic beats induced by spatially localized Na-K pump inhibition in atrial network models.
Proc R Soc Lond B Biol Sci
254:
55-61,
1993[Medline].
27.
Wit, AL,
and
Cranefield PF.
Triggered and automatic activity in the canine coronary sinus.
Circ Res
41:
434-445,
1977[Medline].
This article has been cited by other articles:
|
|
 |
|
  M. P. Nash, C. P. Bradley, P. M. Sutton, R. H. Clayton, P. Kallis, M. P. Hayward, D. J. Paterson, and P. Taggart Whole heart action potential duration restitution properties in cardiac patients: a combined clinical and modelling study Exp Physiol, March 1, 2006; 91(2): 339 - 354. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. P. Nash, C. P. Bradley, and D. J. Paterson Imaging Electrocardiographic Dispersion of Depolarization and Repolarization During Ischemia: Simultaneous Body Surface and Epicardial Mapping Circulation, May 6, 2003; 107(17): 2257 - 2263. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Marret, O. Pruszkowski, A. Deleuze, and F. Bonnet Accelerated Idioventricular Rhythm Associated with Desflurane Administration Anesth. Analg., August 1, 2002; 95(2): 319 - 321. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
t, Time difference. D:
an epicardial activation map was fitted to the electrode activation
times and displayed by using the Hammer projection. This 2-dimensional
(2D) topography represents the ventricular epicardial surface, where
the left ventricle (LV) makes up the central portion of the projection,
the right ventricle (RV) comprises the border regions, but the apex is
retained as a single point. E: the activation time field was
also fitted on the epicardial surface of an anatomically accurate
3-dimensional (3D) mathematical model of the ventricles, for which
anterior (left) and posterior (right) views are
illustrated and contours of simultaneous epicardial activation have
been superimposed. LAD and PDA (thick lines), left anterior and
posterior descending coronary arteries, respectively.
Statistically significant (P < 0.01) changes in the hemodynamic variables compared with the control
state.
