| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Departments of Pharmacology, Pathology and Cell Biology, and Surgery, Université de Montréal, and Research Center, Hôpital du Sacré-Coeur de Montréal, Montréal, Québec, Canada H4J 1C5
Submitted 16 July 2003 ; accepted in final form 16 December 2003
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONREFERENCES |
|---|
coronary artery disease; mapping; ventricular tachycardia
Chronically increased myocardial metabolic demand, blood flow deficit, and oxidative stress are induced in experimental animals by long-term rapid ventricular pacing (27, 43). Prolongation of repolarization intervals, increased total epicardial activation time, and enhanced vulnerability to ventricular fibrillation have been reported in studies concerned with arrhythmia generation in this paradigm (1, 24, 30). We sought to amplify energetic stress regionally by limiting blood supply to the left circumflex coronary artery (LCx) territory, but in a progressive fashion to avoid segmental scar formation (19, 33, 34, 36, 42). This was achieved by implanting around the proximal LCx an Ameroid (AM) constrictor, which causes the artery to become slowly obliterated while collaterals develop as the material takes up tissue water and swells over 3–4 wk (34, 36). Canines with chronically implanted AM constrictors studied in the conscious state displayed normal subendocardial contractile function at rest that deteriorated during exercise as total flow increased but was redistributed away from subendocardial layers (6, 10, 14, 22). Accordingly, ST segment elevation developed in subendocardial unipolar recordings (10), whereas ST segment depression was detected at body surface during transient rapid pacing (26).
We used epicardial mapping to investigate long-term regional electrical alterations in canines with combined AM constrictor implantation and chronic rapid ventricular pacing compared with healthy canines and canines with either ameroid constrictor or chronic rapid ventricular pacing (CRP). Unipolar electrograms recorded from the LCx territory in situ displayed increased spatial inhomogeneity in variables related to excitability and repolarization, and conduction disturbances as well as ventricular tachycardias were induced in response to programmed stimulation. The possible contribution of reduced Na+ channel activity to depression of excitability that was identified in situ was further investigated with the use of whole cell patch clamp and analysis of Na+ channel protein expression.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The experiments were performed in accordance with guidelines of the Canadian Council for Animal Care. Initial surgery was conducted under aseptic conditions and anesthesia (25 mg/kg iv thiopental, followed by 1% isoflurane inhalation) in three groups of mongrel canines (22–32 kg, either sex): 1) CRP with AM constrictor (CRP-AM), 2) CRP alone (CRP), and 3) AM constrictor alone (AM). Comparisons were made with a fourth group consisting of healthy canines. In CRP-AM and CRP, a pacing electrode was positioned at the right ventricular (RV) apex and a pacemaker placed in a subcutaneous pocket (23). In CRP-AM and AM, the proximal LCx artery was exposed via limited left thoracotomy for constrictor implantation (2.25–2.75 mm ID; Research Instruments; Escondido, CA). The chest was closed and animals recovered (antibiotics, 0.3 mg/kg im buprenorphine three times per day). Pacemakers were activated at 240 beats/min 
2 wk after surgery. Animals were carefully observed to detect any pain reaction. The protocol called for immediate termination of rapid pacing and, if needed, euthanasia (thiopental overdose) if pain reactions occurred.
Electrophysiological Study
In CRP-AM and CRP, study was promptly scheduled once signs of heart failure became apparent (particularly, signs of respiratory distress) after 4 wk, amounting to 6 wk after constrictor implantation in CRP-AM. Under anesthesia (50 mg meperidine and 8 mg/kg iv thiopental, followed by 1% isoflurane), pulmonary capillary wedge pressure and cardiac output measurements were repeated for comparison with initial values. Coronary angiography was performed to assess constrictor closure. After sternotomy and pericardiotomy, a silicon plaque electrode carrying 191 unipolar recording contacts (2.6 mm spacing) was positioned onto the posterolateral left ventricular (LV) wall. Unipolar electrograms and ECG lead II were connected to a multichannel recording system (12). Eight needle electrodes were inserted into the LV wall in 6 CRP-AM, and 16 electrodes in 2 CRP-AM to record electrograms at depths of 2, 7, and 12 mm (4). Needle insertion caused slowly reversible ST segment changes; values at stabilization were considered. Signals were amplified by programmable-gain analog amplifiers (0.05–450 Hz), converted to digital format at 1,000 samples·channel–1·s–1, and stored on hard drive for analysis using custom-designed software (http://www.crhsc.umontreal/cardiomap). The findings are reported from additional cases, in which a sock electrode array was employed to record 127 electrograms from the entire biventricular epicardial surface (4, 39).
Protocol. After recording electrograms in sinus rhythm, atrioventricular block was induced by formaldehyde injection (37%, 0.1 ml) into the atrioventricular node and ventricular pacing was maintained at cycle length of 400–500 ms with bouts of transient rapid pacing at 250 ms (1 min). Refractory periods were determined using extrastimuli (S2) coupled to repetitive trains of basic stimuli (S1, S1-S1 = 400–500 ms) in 5-ms decrements, applied from bipolar electrodes in the center of the plaque electrode (11). Arrhythmias were induced using 1–4 closely coupled extrastimuli (S2-S5) applied via bipolar electrodes to the RV wall or from the center of the plaque electrode. Tachycardias terminated spontaneously or by direct current counter-shock. Recording and programmed stimulation were repeated under lidocaine (bolus: 2 mg/kg, infusion: 4 mg·kg–1·h–1) or procainamide in 7 CRP-AM (bolus: 8 mg/kg, infusion: 4 mg·kg–1·h–1) (11, 13). After death (thiopental overdose, KCl), the hearts were excised and examined for gross evidence of tissue scarring. In 50% of preparations, posterolateral wall was dissected and a descending branch of LCx was cannulated for perfusion with enzyme solutions to isolate cardiomyocytes from the full wall thickness (23); this method does not permit to selectively isolate cells from subepicardial, subendocardial, and intramural locations. Blocks of tissue were immersed in liquid nitrogen for later analysis of Na+ channel protein expression. Ventricles of 10 CRP-AM were sliced (1 cm) and incubated with buffered triphenyltetrazolium chloride solution for delineation of necrotic areas.
Electrophysiological Data Analysis
The following variables were extracted from recordings at each site: 1) ST segment potential depression or elevation with reference to isoelectric lines (T-Q) at 60 ms after activation, 2) maximum slope of negative (RS) deflections in unipolar activation complexes [the maximum negative slope of unipolar activation complexes (–dV/dtmax) and its absolute value |–dV/dtmax|] (12), 3) activation-recovery intervals (ARI) from –dV/dtmax to maximum positive slope of T wave (4), 4) activation time determined at the point of –dV/dtmax of unipolar deflections displaying an "RS" (or "rs") morphology and |–dV/dtmax| generally >0.5 mV/ms (5, 12); values were expressed with reference to the earliest activation determined under the plaque electrode. It was considered that excitation failed whenever |–dV/dtmax| was <0.5 mV/ms. Conduction velocities were determined from the analysis of ellipsoid isochronal patterns identified when pacing was performed from the center of the plaque electrode, as reported previously (11). Velocities in the longitudinal and transverse directions were measured as distance divided by conduction time along the long and short axes of the ellipsoid pattern, respectively. Ventricular tachycardias were considered as monomorphic or polymorphic depending on whether or not surface electrocardiogram and unipolar electrograms displayed a constant morphology in all beats (13, 39).
Na+ Channel Activity and Expression
Cardiomyocytes isolated as described above were transferred to a bath (17°C) on an inverted microscope (Nikon; Tokyo, Japan) and superfused with K+-free solution containing (in mM) 5 NaCl, 132.5 CsCl, 1 MgCl2, 1 CaCl2, 11 glucose, and 20 HEPES, pH 7.4 adjusted with CsOH (7). Na+ current (INa) was recorded in the whole cell configuration of patch clamp, with the use of an EPC-7 amplifier (List Medical; Darmstadt, Germany) and suction pipettes filled with solution containing (in mM) 5 NaCl, 135 CsF, 5 HEPES, 10 EGTA, and 5 Mg2-ATP. Currents were monitored on an oscilloscope and stored on a personal computer with pCLAMP 6 software (Axon Instruments; Foster City, CA) used for voltage-clamp protocols and data analysis. Voltage dependence of INa activation was determined by delivering depolarizing voltage-clamp pulses (30-ms duration) in 5-mV increments every 10 s from a holding potential of –140 mV (7).
Membrane preparations were extracted as reported previously (2) from 200 mg tissue. Proteins (100 µg) were fractionated on 8% SDS-polyacrylamide gels, transferred onto nitrocellulose strips, and exposed to Na+ channel (
-subunit) antibody (1:150, Alomone Labs; Jerusalem, Israel) detected using the Renaissance assay kit (Mandel Scientific; Guelph, Canada).
Data (expressed as means ± SD) were analyzed using ANOVA, in which the group (CRP-AM, CRP, AM, and healthy) was the "between" factor and individual preparations were "within" factors. Differences were considered as statistically significant when P < 0.05.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
At postmortem, small areas of necrosis were detected in five CRP-AM preparations: necrosis was limited to parts of the posterior papillary muscle in two (<1% of biventricular slice surfaces), extending to small areas in adjacent subendocardial muscle in two (<2% of slice surfaces), and to small areas in the posterior septum in the fifth (4%). Necrosis extending from subendocardial to middle third of the wall (16% of slice surfaces) was seen in the CRP-AM preparation that died at anesthesia induction. Postmortem was not performed in the five CRP-AM animals that died suddenly. There was no gross evidence of scarring in hearts from AM and CRP.
Electrophysiological Study
ST segment displacements and –dV/dtmax. Unipolar electrograms displaying slight ST segment depression (Fig. 1A, sites a and b) and isoelectric ST segments (site c) were recorded from the epicardial surface at the anterolateral border of the LCx territory. In contrast, ST segment potentials became elevated as recording was extended further into LCx territory (sites d–g). Electrograms recorded more posteriorly displayed slight depression (site h) or isoelectric ST segment potentials. Confluent regional distributions of ST segment alterations within the LCx territory were identified in both CRP-AM and AM (but not in healthy and CRP preparations: vide infra). When intramural recordings were obtained, similar ST segment alterations were identified at subendocardial and intramural sites (not shown). ST segment displacements on either side of the ±2 mV range were identified at >20% of recording sites during ventricular pacing at cycle lengths of 400–500 ms; this AM constrictor effect was typical of both CRP-AM and AM preparations (Fig. 2A). In healthy hearts, the majority of unipolar recordings displayed isoelectric ST potentials (±2 mV from isoelectric) and a minority (5%) displayed slight elevation (+2 to +4 mV) that was related to slurring in the mounting phase of T waves detected. ST segment displacements were identified at only 10% of recording sites in CRP preparations. The number of sites displaying ST segment displacements in CRP-AM (and AM preparations) increased to >50% in response to transient rapid ventricular pacing at 250 ms (Fig. 2B).
|
|
–dV/dtmax values were regionally depressed within the LCx territory of CRP-AM preparations. Depression ranging from moderate (|–dV/dtmax| of 2–10 mV/ms) to marked (|–dV/dtmax| < 2 mV/ms) was identified in association with ST segment elevation in some areas (Fig. 1: |–dV/dtmax| < 2 mV/ms at sites e–g) but without ST segment displacement in others (Fig. 1, B and C, bottom of map). Depressed |–dV/dtmax| values (2–10 mV/ms) were also detected in the LCx territory of CRP preparations, but, in contrast to CRP-AM, they displayed irregular spatial patterns (Fig. 3A). In fact, –dV/dtmax depression was detected diffusely throughout the ventricular epicardial surface when mapping was performed using a sock electrode array. In healthy hearts (Fig. 3B), ST segment displacements were minimal (left map) and |–dV/dtmax| values <10 mV/ms were detected at a minority of sites, particularly at sites neighboring the edges of the plaque electrode. –dV/dtmax depression was one of the two main features of the CRP effect (together with prolongation of repolarization intervals: vide infra).
|
We previously reported from studies in a different pathophysiological setting (12) that 2 mV/ms is the threshold below which conduction velocity locally decreases as |–dV/dtmax| is further reduced. Among all CRP-AM preparations, depression to values <2 mV/ms was identified at 12 ± 22 sites, whereas the incidence of such marked depression was smaller in CRP, and minimal in AM and healthy preparations (Table 1). The maximum number of sites displaying |–dV/dtmax| < 2 mV/ms in any one preparation was much greater among CRP-AM (91 sites) than in CRP and AM. Mean overall values were significantly depressed in CRP-AM and CRP compared with healthy preparations (Table 1); note, however, that the regional differences identified at a minority of sites in CRP-AM (see above) did not translate into a reduction of the overall values measured in CRP-AM versus CRP. –dV/dtmax values measured in AM preparations were also significantly reduced compared with healthy but were not as depressed as in CRP-AM and CRP.
|
Likewise, global conduction velocities measured in the longitudinal and transverse directions were significantly depressed in CRP-AM (longitudinal: 0.63 ± 0.11 m/s and transverse: 0.30 ± 0.04 m/s) and CRP (0.56 ± 0.11 and 0.28 ± 0.04 m/s) compared with healthy (0.76 ± 0.11 and 0.39 ± 0.04 m/s) and AM preparations (0.75 ± 0.14 and 0.38 ± 0.04 m/s).
Inexcitability. Several unipolar recordings from the center of the LCx territory displayed monophasic wave forms in successive beats (6 sites displayed inexcitability in the preparation illustrated in Fig. 1). Interestingly, monophasic waveforms could occur, albeit infrequently, in alternation with delayed activation (Fig. 1, site f). Sites displaying inexcitability in sinus rhythm or at cycle lengths of 400–500 ms were detected from epicardial recordings in 40% (7 of 18 preparations) of CRP-AM (4–34 sites); this was a characteristic of this group because inexcitability was not detected at a slow rate in any preparation of the other three groups (Table 1). Inexcitability occurred in areas in which both depression of –dV/dtmax and elevation of ST segment potential were identified (Fig. 1, site f). The preparation illustrated in Fig. 1 was from one among the majority of CRP-AM preparations in which subendocardial areas of necrosis were not detected at postmortem. In fact, epicardial electrograms displaying QS wave forms (transmural necrosis) or QS waves, followed by RS complexes (subepicardial muscle overlying necrosis) were never detected in any preparation.
ARI. Overall values were significantly prolonged in CRP-AM and CRP (Table 2), a previously reported feature of the CRP effect (17, 24, 30, 37). However, localized shortening was identified during sinus rhythm (or pacing at 400–500 ms) in the LCx territory of one-third of the CRP-AM preparations, usually at sites that also displayed ST segment elevation (Fig. 1, sites d, e, and g) but not always (site c). Sites displaying ARI shortening were identified in proximity to sites displaying prolonged ARI (sites a and b) thereby creating marked spatial inhomogeneities beyond the slight increases in temporal dispersion determined from differences between maximum and minimum ARI values (see Table 2). Thus CRP-AM displayed irregular isocontour patterns, in which central areas of the LCx displayed relatively shorter repolarization intervals, thereby causing local gradients >20 ms between neighboring sites (Fig. 4B). Whereas smooth isocontour patterns were seen in healthy preparations, the longest repolarization intervals being recorded near the site of stimulation (Fig. 4A). Normal shortening occurred homogeneously in response to transient rapid pacing in healthy (Fig. 4A, bottom map). However, repolarization intervals became markedly shorter in some areas and such areas were enlarged in CRP-AM (Fig. 4B, bottom map). Increased spatial inhomogeneities were identified in 80% of CRP-AM preparations in response to transient rapid pacing. Local shortening and spatial inhomogeneities of a similar magnitude were detected in 40% of AM and in 20% of CRP preparations in response to transient rapid pacing.
|
|
In general, localized shortening of repolarization intervals occurred in association with ST segment elevation >2 mV. Note that in Fig. 4C the spatial distribution of ST segment elevation (blue) corresponded to the pattern of ARI shortening (red), ST elevation of up to 12 mV developing in the areas in which ARI was shortened to 125 ms. Note also that ARI shortening was not detected at sites that displayed ST segment depression of as much as –8 mV in response to transient rapid pacing. The association of ST segment elevation and ARI shortening is consistent with a common relation to a locally compromised myocardial ischemic substrate. However, instances were noted in which ARI shortening occurred without ST segment displacement (see above, Fig. 1, site c) or, conversely, ST elevation was detected without being associated with concomitant ARI shortening.
Programmed stimulation. Refractory periods (S2 applied to the center of LCx territory) were significantly prolonged in CRP-AM and CRP (Table 2). Figure 5 illustrates that, in CRP-AM, conduction disturbances in the form of increasing delay and inexcitability (blank areas) occurred in response to closely coupled extrastimuli (Fig. 5, A–D, S1-S4 applied at RV site). The fact that all recording sites displayed activity under pacing from the center (Fig. 5E) illustrates the functional character of inexcitability. As noted above, inexcitability developed in areas in which both –dV/dtmax depression and elevation of ST segment potential (Fig. 5F) were identified. In response to closely coupled S3 and S4, 60% (11 of 18 preparations) of CRP-AM displayed an increasing number of inexcitable sites. Maximal delay increased and the number of inexcitable sites was doubled (from an average of 13 to 26 per CRP-AM preparation) under therapeutic plasma levels of lidocaine (15 ± 4.8 µM, 11 preparations) and procainamide (28 ± 5.8 µM, 7 preparations). In contrast, only two CRP and two AM displayed 
8 inexcitable sites in response to extrastimuli, even under lidocaine. In CRP-AM, the number of recording sites displaying |–dV/dtmax| < 2 mV/ms increased from 12 ± 22 (Table 1) to 46 ± 16 in response to S2 under lidocaine.
|
Ventricular arrhythmias. Polymorphic beats were induced in 9 of 18 CRP-AM with 3 to 4 extrastimuli (coupling interval of initiating extrastimuli of 120–180 ms). Monomorphic tachycardias were induced in two other CRP-AM preparations with two and four extrastimuli, respectively. Remarkably, the incidence of monomorphic tachycardias (lasting for 9 beats to >30 s) increased to seven preparations under lidocaine (5) or procainamide (2), in response to a single (Fig. 6, top trace) or up to three extrastimuli. Monomorphic ventricular tachycardias were never induced in healthy preparations, and their maximal responses to programmed stimulation consisted of runs of 7–20 polymorphic ventricular beats (6 of 8 preparations) or ventricular fibrillation (1 of 8 preparations) that required application of 3–4 extrastimuli at extremely short coupling intervals (S3-S4 intervals of 90–120 ms); similar responses were obtained under lidocaine. Runs of polymorphic ventricular beats were induced in 6 of 10 AM and in 3 of 8 CRP preparations by application of 3–4 extrastimuli (with coupling intervals of 130–220 ms). Runs of monomorphic beats were induced in only a single CRP and in a single AM preparation under lidocaine.
|
Conduction disturbances (inexcitability, delay) were detected from epicardial recordings during ventricular tachycardias in CRP-AM. Isochronal patterns consistent with complete reentrant activity in subepicardial muscle were obtained in two such preparations (Fig. 6). In this example, programmed stimulation was performed from the center of the plaque electrode; the activation map determined in response to S1 (Fig. 6A) displayed an elliptical isochronal pattern in which the direction of rapid conduction velocity is presumed to correspond to the longitudinal fiber direction, and the direction of slow conduction corresponds to the transverse fiber direction (11). The activation sequence mapped during the tachycardia (Fig. 6, B and C: successive beats) displayed a double loop reentrant pattern (8, 25), in which the lower loop (blue) circulated around an arc of dissociation (red line marking differences in activation time >120 ms between neighboring sites, or sites at which |–dV/dtmax| < 0.5 mV/ms) that was aligned along the direction of fast conduction during pacing from the center of the plaque electrode (Fig. 6A). This suggested that dissociation occurred in the transverse direction, as occurs in anisotropic reentry (44), and that accordingly, the return pathway of the reentrant pattern (160–235 ms) occurred in the longitudinal fiber direction. The upper loop (green) revolved around a shorter, irregular line of dissociation (also in red). Note that the areas of dissociation as well as the return pathways were identified in areas in which –dV/dtmax values were depressed (Fig. 6D; |–dV/dtmax| < 2 mV/ms) and that the upper loop occurred in association with localized ARI heterogeneity (Fig. 6E).
Na+ channel expression and current. Measurements were made to investigate Na+ channel alteration in the context of the overall depression of excitability identified in the experimental groups. Protein expression was significantly reduced in CRP-AM (–54 ± 20%) and CRP (–35 ± 16%) compared with healthy, but not in AM. Within CRP-AM, expression was significantly reduced in the LCx (–51 ± 17%) versus LAD territory. Peak INa values in cardiomyocytes isolated from the posterolateral LV wall in AM (36.9 ± 22.9 pA/pF) were similar to healthy (39.8 ± 20.5), whereas peak INa tended to be lower in CRP-AM (32.3 ± 12.4) and values were significantly lower in CRP (24.2 ± 14.3). Peak current density consistently occurred at –40 mV in all four groups, as reported by Gaspo et al. (7) in studies of canine atrial myocytes performed under similar conditions. Membrane capacitance measured in cells from CRP (237 ± 64 pF) was significantly greater than in healthy animals (204 ± 46 pF); capacitance measurements in CRP-AM cells also tended to be increased (227 ± 40 pF).
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Recordings from the LCx territory were typical of unipolar electrograms generated by populations of electrically coupled viable cells displaying depressed action potentials (15) of the type described in acutely ischemic myocardium (21, 44). Confluent patterns of ST segment displacement (presumably related to patterns of myocardial hypoperfusion after ameroid constrictor closure) were identified by epicardial mapping in the LCx territory of CRP-AM and AM preparations, but not in CRP and in healthy preparations. The patterned distribution of ST segment displacements was enhanced by transient rapid pacing. Epicardial ST segment elevation was classically interpreted as a sign of transmural ischemia, whereas ST depression indicated either ischemia of underlying myocardium or reciprocal changes from sideways or more distant ischemic borders (14). Although the main focus of this study was on epicardial recordings, needle electrode recordings obtained in several preparations indicated that similar alterations occurred at intramural and subendocardial sites.
Epicardial QS waves typical of transmural or extensive subendocardial necrosis (5, 44) were never detected. Subendocardial necrosis limited to the posterior papillary muscle and small adjacent areas was detected in a minority of CRP-AM preparations. In previous studies, no confluent necrosis was detected by gross examination in chronic rapid ventricular pacing from the right ventricle (43) and little or none in almost all AM preparations (19, 26, 33, 36, 42). Collateral vessel development is advanced 2 wk after ameroid constrictor implantation (36), the time at which rapid pacing was started in this study. Preexisting collaterals in canines (and their further development under energetic stress) probably was an important factor limiting myocardial ischemic damage in AM and CRP-AM. Other factors that possibly contributed to limiting anginal pain may have included impaired function of afferent neurons subserving cardiac nociception in myocardial ischemia (16) and loss of cardiac innervation in CRP (31).
In situ variables related to excitability (–dV/dtmax) were globally depressed in CRP-AM and CRP compared with healthy preparations. The major difference between CRP-AM and CRP preparations in this regard was that confluent patterns including areas of marked |–dV/dtmax| depression were identified in the LCx territory of CRP-AM. Depression to values <2 mV/ms was identified at a greater, albeit limited, number of sites in the LCx territory of CRP-AM preparations. A much greater incidence of inexcitability in the LCx territory of CRP-AM preparations was another major difference between CRP-AM and CRP. This suggested that interaction occurred between the chronic rapid pacing and the chronic blood flow restriction at a number of sites located in central areas of the LCx territory, as identified with the relatively high mapping resolution available in this study.
Our finding that whole cell INa was reduced by chronic rapid ventricular pacing is in contrast with Kaab et al.'s (17) finding that whole cell INa was similar to healthy in cardiomyocytes isolated from similar CRP canine preparations. However, the cells that they studied may have been less damaged than the ones isolated from CRP and CRP-AM herein because membrane capacitance was similar to healthy in their study, whereas we measured higher than normal capacitance, a sign of cell hypertrophy. We found that decreased whole cell INa was associated with a reduction in Na+ channel expression. We may speculate that chronic rapid ventricular pacing, which is known to induce a state of increased intracellular Ca2+ (35), might have in turn caused a reduction in Na+ channel expression, as shown in cultured neonatal rat cardiac myocytes (3).
Peak INa (as well as Na+ channel expression) measured in AM were similar to measurements made in healthy, whereas –dV/dtmax depression was identified in AM preparations (although to a lesser extent than in CRP-AM and CRP). The discrepancy could be due to 1) greater sensitivity of –dV/dtmax reflecting alterations in large cell populations, and 2) contributions of factors other than reduction of Na+ channel activity to –dV/dtmax depression, namely cell uncoupling.
Overall prolongation of repolarization intervals in CRP-AM and CRP is another feature of the CRP effect, in accordance with previously documented downregulation of K+ currents in CRP (37). However, areas of local shortening and attendant spatial inhomogeneities were detected in the LCx territory of CRP-AM preparations, an effect that was exaggerated by a transient tachycardic stress. Moreover, preliminary data indicate that the localized shortening of ARI can be inhibited by glibenclamide, an ATP-dependent K+ channel blocker. Localized shortening in some areas and prolongation in others contributed to spatial inhomogeneities that may have been involved, together with depression of excitability (|–dV/dtmax| < 2 mV/ms), in conduction disturbances.
The induction of monomorphic ventricular tachycardias, especially with facilitation under therapeutic concentrations of sodium channel blocking drugs, was a major difference between CRP-AM and all other groups. In the few cases of reentrant ventricular tachycardia that could be mapped at the epicardial level, a line of dissociation could be identified between adjacent fiber tracts parallel to the longitudinal fiber orientation, suggesting that conduction was impaired in the transverse direction. This suggests that anisotropic conduction properties may also have contributed to arrhythmogenesis in CRP-AM preparations (32, 44). Moreover, it is possible that cellular uncoupling might have contributed to both –dV/dtmax depression and increased dispersion of repolarization intervals. It will also be necessary to clarify the role of other transmembrane currents (ICa,L, Ito) that are known to be altered in CRP (17, 23, 35, 37) and may be involved in maintaining conduction in the face of reduced excitability and decreased cell coupling (40).
Perspective. Our interest in long-term electrical alterations occurring independently from segmental scarring arose from an atypical case of arrhythmia surgery. In this patient, LAD revascularization was the primary indication for surgery but he was also subjected to intraoperative electrophysiological study because of prior syncopal episodes and akinetic apical LV segments (suspected aneurysm). Unipolar epicardial recordings did not present QS complexes (Fig. 7A), in agreement with the fact that scar was found to be limited to a small, remote subendocardial location in the anterior LV wall. Yet, nonsustained monomorphic ventricular tachycardias were induced by programmed stimulation, which were stabilized into sustained monomorphic with 500 mg procainamide. A reentrant pattern was mapped epicardially, with bridging activity detected in the anteroapical LV wall. This is in contrast with a conventional tachycardia substrate (Fig. 7B) in which the return pathway occurred in muscle surviving next to scar (29, 39, 44). Unipolar recordings from the anterior LV wall (sites 3,4) displayed QS waves, followed by delayed activation (rs) in sinus rhythm and bridging activity from one beat to the next during tachycardia. Consideration of the data illustrated in Fig. 7A (obtained in the context of coronary revascularization therapy) suggested to us that this type of nonscar substrate might be investigated in the animal laboratory by combining blood flow restriction with an ameroid constrictor and chronic rapid ventricular pacing because the latter is a model of reversible ventricular dysfunction in which alterations of the flow-function relationship is involved (28).
|
Limitations. Although constrictor closure was assessed by angiography, local collateral-dependent blood flow was not determined (microspheres). The variable development of collateral vessels between animals would probably explain much of the variability of electrical alterations between CRP-AM preparations. Another limitation was that sequences of transmural activation during ventricular tachycardias were not determined and that, therefore, reentrant activity and the involvement of intramural and subendocardial muscle layers therein were not systematically documented. Finally, morphological data are limited to gross examination, and the histological features of muscle fibers in areas of long-term electrical alterations were not determined. In the present study, we emphasized the role of altered electrophysiological variables related to excitability and repolarization properties. However, we recognize that interstitial fibrosis that may develop with CRP (18, 41) could have played a role in the dissociation between adjacent fiber tracts leading to reentry. Although Kavanagh et al. (20) did not find evidence of structural alteration in gap junction distribution or change in cellular interconnections in similar canine CRP preparations, we cannot rule out the possibility that such changes might have contributed to arrhythmogenesis in CRP-AM preparations.
In conclusion, CRP has been proposed as a paradigm to study arrhythmogenic mechanisms underlying sudden death in heart failure (30). Combination with progressive coronary artery constriction provides an enhanced model displaying increased regional heterogeneity, responsiveness to transient tachycardic stress and enhanced arrhythmia formation. This model could also be useful to investigate the potential proarrhythmic effects of drugs because interaction with myocardial ischemia has been proposed as a mechanism for the increased mortality in the Cardiac Arrhythmia Suppression Trial I study (9).
|
ACKNOWLEDGMENTS |
|---|
This study was supported by the Canadian Institutes of Health Research.
|
FOOTNOTES |
|---|
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.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  P. Dorian Antiarrhythmic Action of{beta}-Blockers: Potential Mechanisms Journal of Cardiovascular Pharmacology and Therapeutics, October 1, 2005; 10(4_suppl): S15 - S22. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
