Electrophysiological Effects of Flecainide on Anisotropic Conduction and Reentry in Infarcted Canine Hearts
Background The class IC antiarrhythmic drug flecainide has been shown to be ineffective for the treatment of ventricular arrhythmias in some patients who have had a prior myocardial infarction and sometimes even provoke arrhythmias (proarrhythmic effect). Since some ventricular tachycardias may be caused by anisotropic reentry, we determined the effects of flecainide on this mechanism for reentry in infarcted canine hearts in order to determine possible causes for its clinical effects.
Methods and Results The effects of flecainide were determined on ventricular tachycardia induced by programmed electrical stimulation in dogs with healing myocardial infarction 4 days after coronary artery occlusion. Activation in the reentrant circuits causing tachycardia was mapped with a 196-channel computerized mapping system. We found that flecainide converted inducible unsustained ventricular tachycardia to inducible sustained ventricular tachycardia by modifying conduction in the reentrant circuit. In general, by slowing conduction, the reentrant wave front did not block after flecainide, leading to perpetuation of reentrant excitation. When sustained ventricular tachycardia could be induced before the drug, flecainide prolonged the coupling interval of premature impulses necessary to induce tachycardia by lengthening the line of block and slowing conduction around it. Flecainide also slowed the rate of the tachycardia but did not terminate it. The anisotropic reentrant circuits were modified so that the central common pathway of “figure-of-eight” circuits was narrowed and lengthened due to extension of the lines of block that bounded the pathways. Extension of the lines of block resulted from depression of conduction in the direction transverse to the long axis of the myocardial fiber bundles caused by flecainide. Flecainide also slowed conduction in the longitudinal direction in part of the circuits. The depressant effects of flecainide on both longitudinal and transverse anisotropic conduction were quantified by pacing from the center of the electrode array and it was found, contrary to predictions, that transverse conduction was depressed as much as longitudinal conduction.
Conclusions Flecainide slows conduction in both the longitudinal and transverse direction relative to the orientation of the myocardial fibers. This enables sustained reentry to occur more easily. Flecainide does not cause conduction block in crucial regions of reentrant circuits (central common pathway) and therefore does not prevent reentrant tachycardia in healing infarcts.
Class I antiarrhythmic drugs (drugs that block fast cardiac sodium channels in the Vaughan-Williams classification) are effective in only 10% to 30% of clinical cases of reentrant ventricular tachycardia associated with healing or healed myocardial infarction, with class IA drugs being more effective (approximately 30%) than class IB or IC drugs (approximately 10%).1 All the class I drugs share the property of sodium channel blockade, but the kinetics of the interactions between the drug and the sodium channel determine whether these drugs produce little (IB), moderate (IA), or marked (IC) slowing of conduction.1 2 It has been proposed that “a drug that depresses conduction can convert unidirectional block to bidirectional block and thus terminate or prevent reentry.”3 Class IA drugs also prolong the action potential duration and refractory period more than the other class I drugs.1 2 This property also may account for efficacy in preventing ventricular tachycardia.
In addition to being relatively ineffective in the treatment of ventricular tachycardia, the class IC drugs can be proarrhythmic, particularly in patients with ventricular arrhythmias who have left ventricular dysfunction.4 5 Although not yet proven, the increased incidence of sudden death in the flecainide limb of the Cardiac Arrhythmia Suppression Trial (CAST) might have been a consequence of the proarrhythmic action of flecainide.6 7
The reason why class IC drugs are often not effective in preventing ventricular tachycardia and the mechanism for their proarrhythmic actions have not been completely elucidated. One mechanism that has been proposed for the proarrhythmic effects is that these drugs can facilitate the occurrence of reentry because they depress conduction,8 9 the very property proposed to underlie the antiarrhythmic effects.3 However, the determination of the effects of drugs on reentrant circuits requires tracking impulse propagation in circuits during drug administration. This had not been done in electrophysiological investigations of class IC drugs on reentrant circuits in infarcts at the time we began our investigation.10
Activation in reentrant circuits can be mapped in a canine model of myocardial infarction.11 12 13 14 During the healing phase after ligation of the left anterior descending coronary artery (LAD), reentry often occurs in the narrow rim of parallel oriented fibers that survive on the epicardial surface of the infarct as the epicardial border zone. We have shown that much of the slow conduction that enables reentry to occur results from the anisotropic properties of the epicardial border zone, that is, slow conduction transverse to fiber orientation, and therefore have called this mechanism for reentry anisotropic reentry.14 15 The major aims of the present study were to determine how class IC drugs affect this mechanism for reentry and to determine if they prevent reentrant excitation or how they might facilitate reentry. For this purpose, we used flecainide, which is a prototypical class IC drug. Our results have been reported previously in abstract form.10
Creation of Myocardial Infarction
Myocardial infarction was produced in 18 mongrel dogs weighing 30 to 40 kg by a two-stage ligation of the LAD according to the procedure originally described by Harris.16 We have described our methods in detail previously.14 After surgery, the dogs recovered from anesthesia and were returned to the animal quarters for postoperative care until the electrophysiological study 4 days later. Studies were done at this time because reentrant circuits can be clearly defined, and anisotropic properties contribute importantly to their occurrence. Our use of animals conformed to the guidelines of the American Physiological Society and AAALAC.
Electrode Array and Recording Instrumentation
Four days after coronary occlusion, the dogs were anesthetized with intravenous pentobarbital sodium (20 to 30 mg/kg) and ventilated by a positive pressure respirator. The chest was opened by a median sternotomy, the pericardium was opened, and the heart was supported in a pericardial cradle. An electrode array, consisting of 292 bipolar electrodes embedded in a 0.3-mm-thick sheet of rubberlike material (Biomer), was then sutured on the epicardial surface of the left ventricle (Fig 1A⇓). The array covered virtually the entire left ventricular anterior and lateral surfaces. The electrodes were made from silver disks with a diameter of 1.0 mm; the distance between the centers of two disks forming a bipolar pair was 2.0 mm. We could record simultaneously from 196 of the 292 electrodes at any one time with a computerized mapping system that has been described previously.14 The electrode array was divided into two overlapping but separate configurations of 196 electrodes each. Each configuration could be selected with a switch box. One configuration consisted of 196 electrodes, which covered the entire left ventricle with the exception of part of the posterior wall adjacent to the right ventricle. These electrodes were located throughout the entire array shown in Fig 1A⇓. In this configuration, in the center of the grid (area within the central square in Fig 1A⇓), the center of each bipolar pair was 5 mm from the center of the closest bipolar pair in the vertical direction and 7.5 mm from the center of the closest bipolar pair in the horizontal direction. In the area outside the central grid, the center of each bipolar pair was 5 to 10 mm from the center of the closest bipolar pair in both vertical and horizontal directions. The second configuration of 196 electrodes was a higher-density array, with twice the resolution of the large field array. This was achieved by adding 96 electrodes to the central region of the grid (within the central square in Fig 1A⇓) so that each bipolar electrode was now in the center of a hexagon; each bipolar pair was 4.5 to 5 mm away from each of six surrounding bipolar pairs. While recording in this second configuration, the 96 electrodes outside the central square were not used.
Two ECG leads, arterial blood pressure, and a selected electrogram from the multiple electrode array were continuously monitored on an Electronics for Medicine DR 12 oscillographic recorder. The heart was stimulated through rows of bipolar electrodes located at different sites on the ventricles. Each pole had a 1.0-mm diameter, and the poles were 1.5 mm apart. One row of four bipoles was embedded in a narrow sheet of Biomer 10 mm wide and 3 cm long and sutured on the right ventricle with the long axis parallel and adjacent to the LAD (Fig 1A⇑, LAD stim). The other stimulating electrodes were embedded in the recording matrix: a row of four electrodes on the basal-lateral margin with the long axis perpendicular to the LAD (Fig 1A⇑, basal stim) and a row of four electrodes on the lateral margin parallel to the LAD (Fig 1A⇑, lateral stim). Two bipolar electrodes were located at the center of the central high-resolution portion of the grid (Fig 1A⇑, central stim).
For induction of ventricular tachycardia, programmed stimulation protocols with either single, double, or triple premature stimuli were used from each of the four stimulation sites (LAD, base, lateral, and center). The stimulus pulse was 2 milliseconds in duration and two to four times diastolic threshold. For the purpose of this study, sustained monomorphic ventricular tachycardia was defined as the occurrence of repetitive complexes of ventricular origin with a uniform QRS morphology lasting longer than 30 seconds. All sustained ventricular tachycardias had a stable cycle length. Unsustained ventricular tachycardia was defined as runs of three or more repetitive complexes that terminated spontaneously before 30 seconds. The stimulation protocol was continued to completion even if sustained ventricular tachycardia was initiated. If ventricular fibrillation was repeatedly induced from a single site, stimulation was discontinued from that site and refractoriness was not determined.
The magnetic tapes on which the digitized electrograms were stored were replayed afterward for off-line data analysis. We have described our methods in detail previously for determining local activation times, drawing isochrones, and designating regions of conduction block.14 Some of the uncertainties of determining conduction block from isochronic maps are discussed along with the results. We also determined the probable exit route of the reentrant impulse from the circuit causing ventricular tachycardia to the rest of the ventricles from the activation maps with methods previously described.17
Calculation of Conduction Velocity
Apparent conduction velocities in the epicardial border zone were calculated by computer from activation maps generated during stimulation at the center of the high-density electrode array (Fig 1A⇑, central stim) by a method that we have previously described in detail.18 19 As shown in Fig 1B⇑, in a map resulting from stimulation at the center, 16 equally spaced vectors were drawn by computer from the stimulation point to the edges in all directions (indicated by circled numbers in Fig 1B⇑). The average conduction velocity along each of the 16 vectors was calculated as previously described.18 19 The first isochrone where the electrograms were clearly distinguishable from the stimulus artifact served as the zero time point for the calculation of conduction velocity (for example, the 10-millisecond isochrone in Fig 1B⇑). In some situations it was necessary to truncate the actual length of a vector along which a velocity was measured or discard it entirely. We did this when there was possible conduction block in a region along the vector. Because there is no precise way in which to distinguish very slow conduction from conduction block, we arbitrarily designated three or more interpolated isochrones (at least 40 milliseconds) between any two adjacent electrode pairs in this experimental protocol as possible conduction block. Calculation of conduction velocity across these bunched isochrones would yield values of 0.12 m/s or less. While it is possible that conduction did occur with such low velocity, since we could not exclude actual block from occurring in this area, we did not include these values in conduction velocity calculations.18 19 This is illustrated by vectors 9 through 11 in Fig 1C⇑ (possible block is indicated by the thick lines). We do indicate in the results when a region through which conduction occurred was converted to a region of possible block after the administration of drug. The length of vectors along which there was possible epicardial breakthrough (activation from deeper surviving muscle layers) was also truncated for the calculation of conduction velocities. We arbitrarily designated regions where more than three electrodes were activated along a vector within one 10-millisecond isochrone as regions of possible transmural breakthrough. This corresponds to a distance of 11 to 15 mm being activated in less than 10 milliseconds, which would give very rapid conduction velocities of greater than 1.5 m/s. An example of such a region is shown at the distal end of vector 16 in Figure 1B⇑. While such rapid velocities might occur, we could not be certain whether or not the epicardial border zone in these areas was being activated transmurally.18 19
The fast axis of conduction was identified by the vectors with the two fastest calculated velocities (Fig 1B⇑, vectors 3 and 12). The slow axis of conduction was indicated by the two slowest calculated sector velocities. These vectors extended away from the stimulus site in nearly opposite directions (Fig 1B⇑, vectors 10 and 16). The fast axes were parallel to the orientation of myocardial fibers (longitudinal conduction).14 20 The slow axes were perpendicular to the long axis of fiber orientation (transverse conduction). Velocities were calculated for three to five consecutive stimulated impulses during pacing at a regular cycle length and averaged. The anisotropic ratio was calculated as the average conduction velocity of the two fastest vectors divided by the average conduction velocity of the two slowest vectors for each experiment.
Measurement of Effective Refractory Period
The effective refractory period was measured at each of the sites of stimulation during the protocol in which single premature stimuli were applied to initiate tachycardia. The LAD stimulus site was always on the noninfarcted right ventricle and therefore gave us a measurement of the effects of flecainide on normal myocardium. The central stimulus site was always in the middle of the epicardial border zone in or near the region of the reentrant circuit and therefore gave us a measurement of the effects of flecainide on the surviving muscle in the infarct. The basal and lateral sites were sometimes in normal myocardium and sometimes in the infarct, depending on the extent of the infarct in different experiments. Premature stimuli had a strength of twice diastolic threshold, which was the same as the basic drive stimuli. The effective refractory period was defined as the maximum S1S2 interval at which a conducted response was not elicited by S2. The effective refractory period was determined at the longest pacing cycle length in each experiment at which the ventricles could be reliably captured. This cycle length ranged from 280 to 400 milliseconds.
Flecainide acetate (Riker Laboratories) was administered intravenously starting with a dose of 2.5 mg/kg given over several minutes and followed by a maintenance infusion of 0.025 to 0.033 mg/kg per minute. These doses resulted in plasma levels of 1.15 to 2.4 ng/mL. A second dose of 2 mg/kg followed by an increase in the maintenance infusion to 0.067 mg/kg per minute was given to achieve plasma levels of 2 to 4 ng/mL. Plasma levels were determined commercially (Roche Pharmaceuticals).
For the experiments on conduction velocity, data were analyzed by repeated-measures ANOVA. Post hoc comparisons of means were performed using Tukey’s procedure. Comparisons between two means only were made by Student’s t test. All data are expressed as mean±SD.
Effects of Flecainide on Reentrant Circuits Causing Unsustained Ventricular Tachycardia
Unsustained ventricular tachycardia was induced by programmed stimulation in 12 of the 18 experiments. In 9 of these 12, ventricular fibrillation was also induced. Sustained tachycardia could not be induced in any of these 12 experiments before drug administration. Flecainide in either of the dose schedules used (see “Methods”) did not prevent the induction of unsustained tachycardia, but rather, after flecainide administration, sustained ventricular tachycardia could be induced in 7 of the 12 experiments. In 4 of these experiments, the plasma levels associated with sustained tachycardia were between 1 and 2 ng/mL while in 2 experiments, levels were between 2 and 4 ng/mL. Plasma levels were not available for 1 experiment.
For the description of the effects of flecainide on unsustained ventricular tachycardia, we divided these experiments into two groups based on the morphology of the tachycardia before drug. In the first group (7 experiments), polymorphic unsustained ventricular tachycardia was induced before flecainide administration, and in 4 of these 7, monomorphic sustained tachycardia was induced after flecainide administration (Fig 2⇓, A and B). In the second group (5 experiments), monomorphic unsustained ventricular tachycardia was induced before flecainide, and in 3 of these 5, monomorphic sustained tachycardia was induced after flecainide (Fig 2⇓, C and D).
Conversion of Polymorphic Unsustained Tachycardia to Monomorphic Sustained Tachycardia
Summary. In these experiments, the unsustained tachycardias were characterized by an unstable reentrant circuit; that is, the location of the circuit changed during the tachycardia. The circuit sometimes switched from the epicardial border zone to another location that might have been at least partly intramural. The change from one circuit to another occurred after conduction block of the reentrant impulse in the epicardial border zone. Conduction block occurred because excursion of the reentrant impulse around a circuit was too rapid compared with circuits in which sustained tachycardia occurred. Changes in the circuit were accompanied by changes in the exit route from circuits to the rest of the ventricles, resulting in polymorphic QRS. After flecainide, monomorphic sustained tachycardia was caused by reentry in a new stable circuit that was not present before drug administration, with a single exit route to the rest of the ventricles. The new circuit was either in the form of a single reentrant loop or a double loop, that is, a “figure-of-eight” configuration. The new circuit was established because flecainide caused new regions of functional conduction block that sometimes served as a fulcrum around which reentry occurred. Flecainide also significantly slowed conduction of the reentrant impulse, preventing it from propagating into a region that had not recovered from prior activation and blocking. The slowing of conduction was manifested on the ECG as a prolongation of the cycle length. We next describe the effects of flecainide for one representative experiment in this group.
Example of polymorphic unsustained ventricular tachycardia. The activation maps of the unsustained polymorphic ventricular tachycardia in Fig 2A⇑ (which was converted to the monomorphic sustained tachycardia in Fig 2B⇑) are shown in Fig 3⇓. Fig 3A⇓ is the map of the epicardial border zone activation during the last of 8 basic drive stimuli initiated from the LAD electrodes. Activation begins within the 10-millisecond isochrone at the LAD margin (at the pulse symbol) and moves in a direction parallel to fiber orientation, from the LAD to the lateral (LL) margin of the electrode array (isochrones 10 to 70, black arrows).
Fig 3B⇑ shows the activation after the premature stimulus (S1S2 coupling interval of 190 milliseconds) that induced the unsustained tachycardia. Activation again begins at the LAD margin (10-millisecond isochrone, pulse symbol) and moves from the LAD toward the lateral margin (orthodromic for LAD stimulation), but after 50 to 70 milliseconds, the wave front of activation blocks along a long line, indicated by the thick black line. The block is indicated by the large differences in activation times on either side of this line as well as movement of the wave front in opposite directions on either side of the line. The wave front splits into two, with one wave front going around the apical edge of the line of block and the second wave front going around the basal edge of the line of block. These two wave fronts coalesce at the lateral margin and move antidromically back to the line of block (black arrows). The area on the distal side of the line of block is excited at 160 milliseconds (asterisk), 96 milliseconds after the area proximal to the block was initially excited. This wave front from the distal side of the line of block reenters the area proximal to the line of block as shown in Fig 3C⇑.
The activation map in Fig 3C⇑ begins (10-millisecond isochrone, asterisk) where the one in panel B ends. The reentering wave front antidromically activates the area toward the LAD margin, leading to the first nonstimulated complex (first complex of the tachycardia, Fig 2A⇑). This wave front then splits into two wave fronts. One wave front moves to the left, toward the apex (40- to 60-millisecond isochrones) and then toward the lateral margin (60- to 110-millisecond isochrones). The other wave front moves to the right, toward the base (40- to 60-millisecond isochrones) and then toward the lateral margin (60- to 110-millisecond isochrones). The two wave fronts coalesce at the lateral margin and move back toward the LAD through a central common pathway (110- to 190-millisecond isochrones) bounded by two lines of functional block (thick black lines) oriented parallel to the direction of the myocardial fiber bundles.
Fig 3D⇑ continues where activation in Fig 3C⇑ ends. Activation begins in the 10-millisecond isochrone at the asterisk and moves toward the LAD (10- to 40-millisecond isochrones), where the wave front splits into two. The wave front that exits the epicardial border zone at the LAD margin produces the second tachycardia impulse on the ECG in Fig 2A⇑. One wave front again moves to the left, toward the apex (40- to 80-millisecond isochrones). The second wave front moves to the right, toward the base (40- to 80-millisecond isochrones), and then toward the lateral margin where the two wave fronts coalesce at 130 milliseconds. The merged wave front activates the central common pathway in the antidromic direction (isochrones 130 to 160). The impulse then blocks at the site indicated by the thick black transverse line (at 160 milliseconds). However, despite the block, there was one more tachycardia impulse.
In Fig 3E⇑, activation begins, after a 30-millisecond quiescent period, within isochrone 30 (asterisk). The origin of this wave front is uncertain but might have resulted from an intramural reentrant circuit. The wave fronts then follow an activation pattern similar to those during the previous reentrant tachycardia impulses, but block again occurs after 180 milliseconds at the thick black transverse line.
Fig 3F⇑ shows the activation pattern during the last QRS complex in Fig 2A⇑. This is probably a sinus beat, judging by the morphology of the QRS, which is identical to the QRS during sinus rhythm. The activation pattern shown in Fig 3F⇑, with activity arising around all margins of the epicardial border zone and moving toward the center, is also identical to the activation pattern during sinus rhythm.
Fig 4⇓ (top panels) shows the enlarged activation maps of Figs 3C⇑ and 3D⇑, corresponding to the first two beats of the tachycardia with the location of 5 bipolar electrode recording sites (circled) spanning the line of block of the second reentrant impulse. These 5 electrograms are shown below for the last basic stimulated impulse (S1), the prematurely stimulated impulse (S2), and the three tachycardia impulses (T1, T2, T3). During the last basic drive, S1, this region is activated orthodromically from the electrogram in the top trace, toward the bottom. These electrograms are distal to the line of transverse block that occurred with S2 (see Fig 3B⇑) and hence, during S2 they were activated antidromically after a long delay (the time required for the wave front to move around the edge of the line of block and come back antidromically). During T1, this region is again activated antidromically. Each of these sites is activated at a shorter interval (ie, S2T1<S1S2) and conduction is slower in this region during T1 than during S2. This is evidenced by the broader electrograms during T1 and the longer time interval required for activation of this region during T1 (59 milliseconds) compared with S2 (32 milliseconds). Activation of this region during T2 occurs at an even shorter coupling interval (T1T2<S2T1), and block occurs between the second and third electrogram traces. Block of T3 occurred at a different location (between the first and second electrogram traces) at a still shorter coupling interval. Thus, the progressive shortening of the cycle length during the unsustained tachycardia may cause the reentrant wave fronts to encounter refractory tissue with resultant block.
Effect of flecainide on reentrant circuit causing polymorphic unsustained ventricular tachycardia. After flecainide administration in the experiment described in Fig 2A⇑ and Fig 3⇑, programmed ventricular stimulation induced a monomorphic sustained ventricular tachycardia with a QRS morphology that was different from the polymorphic unsustained tachycardia (Fig 2B⇑). Fig 5⇓ shows the activation maps of the induction and final reentrant circuit causing this sustained tachycardia. Fig 5A⇓ shows the activation map of the last of 8 stimulated basic impulses. The activation pattern of the basic drive after flecainide is similar to that before the administration of the drug (compare with Fig 3A⇑), but conduction is slower.
Fig 5B⇑ shows the activation pattern after a premature stimulus with a 190-millisecond coupling interval. The location of the line of block (thick black line) is approximately the same after flecainide as it was with a comparable coupled premature stimulus before flecainide, but the line is somewhat longer, particularly in the transverse direction (compare Fig 5B⇑ with Fig 3B⇑). Activity spreads orthodromically around the ends of the line of block. The area distal to the line of block was excited after 160 to 200 milliseconds (see asterisks). The wave front reenters the area proximal to the line of block as shown by the asterisks within the 10-millisecond isochrone in Fig 5C⇑, which continues where activation in Fig 5B⇑ ends. From this point, excitation exits the epicardial border zone at the LAD margin to cause the first tachycardia impulse.
Fig 5C⇑ shows the activation pattern during this first complex of the tachycardia. The wave front divides into two wave fronts that activate the basal and apical areas from the LAD toward the lateral margin (isochrones 20 to 100). The wave front activating the apical area (to the left on the map) blocks after 90 milliseconds at the thick black line. Because of this block, the subsequent activation pattern suddenly changes from the pattern that occurred before flecainide.
Panel D in Fig 5⇑ begins where panel C ended. The wave front beginning at the 10-millisecond isochrone moves toward the LAD margin (isochrones 10 to 130). This wave front also moves toward the base around a line of block and activates the upper half of the base orthodromically. It then collides with a second wave front that arises near the lower end of the line of block at the 110-millisecond isochrone.
In Fig 5E⇑, the wave front during the third tachycardia impulse winds up moving in a circular pattern around a single long line of block. Finally in Fig 5F⇑, when the tachycardia was stable, there is a single reentrant circuit with a revolution time of 200 milliseconds around a central depressed area where no activity was recorded. The exit route from this circuit is at the lateral margin, accounting for the change in the QRS morphology of the tachycardia.
Conversion of Monomorphic Unsustained Tachycardia to Monomorphic Sustained Tachycardia
Summary. In these experiments, the unsustained tachycardias were characterized by a stable reentrant circuit composed of either a single or double loop (figure of eight) with the same size, shape, location, and exit route to the ventricles for every beat. The reentrant wave front blocked spontaneously in the circuit after 5 to 60 excursions around it in a direction parallel to the long axis of the myocardial fiber orientation without a progressive decrease in cycle length as occurred for the polymorphic tachycardias. After flecainide, reentry occurred in the same circuit with the same exit route to the ventricles as before. Conduction of the reentrant excitation wave around the circuit was slowed, prolonging the cycle length, and the reentrant impulse no longer blocked to terminate tachycardia.
Example of monomorphic unsustained ventricular tachycardia. The reentrant circuit during the last two impulses of the unsustained (nonsustained) tachycardia shown on the ECG in Fig 2C⇑ is displayed in Fig 6⇓ (left panel, NSVT). This reentrant circuit is in the basal area of the epicardial border zone, which is the only region of the electrode array shown in the figure. In the time window shown at the far left, activity during the next to the last beat of the unsustained tachycardia starts at the LAD-basal margin (10-millisecond isochrone and asterisks), and the wave front activates the basal region from LAD toward the lateral margin (isochrones 10 to 80), pivots around the lower edge of a line of functional block, and comes back up in the opposite direction toward the LAD (isochrones 110 to 170). This pattern was found for all unsustained tachycardia impulses. The exit route to the ventricles during each revolution is at the LAD-basal margin. Representative electrograms from around the reentrant circuit are shown below the maps. (The recording sites are circled on the maps.)
The unsustained tachycardia terminated when the circulating wave front moving toward the LAD margin blocked in a region of slow activation at the 150-millisecond isochrone (dark horizontal line) in the right map in the NSVT panel. Electrograms recorded from the region where the block occurred are shown below. Block on the electrogram traces is indicated by the arrow with the two horizontal lines. The reason for the block and the termination of the unsustained tachycardia in this experiment and other experiments in this group is not apparent from analysis of the activation maps or from the electrograms. There are no oscillations in the cycle length before termination, nor is there a progressive decrease in the cycle length or a sudden premature activation at the site of block, as was present in polymorphic unsustained tachycardias (Fig 4⇑).
Effects of flecainide on reentrant circuit causing monomorphic unsustained ventricular tachycardia. The activation sequence in Fig 6⇑ (right panel, SVT) during sustained ventricular tachycardia after flecainide administration in this experiment is similar to the sequence during unsustained tachycardia. However, conduction around the entire circuit during the sustained tachycardia is slower. The cycle length of the unsustained tachycardia is 178 milliseconds, and the cycle length of the sustained tachycardia is 214 milliseconds. Electrograms from the exact same sites shown for the unsustained tachycardia are shown for the sustained tachycardia below the activation map. The electrograms during the sustained tachycardia are broader due to the slower conduction after flecainide but have the same morphology. The exit route to the ventricles remains the same after flecainide, at the LAD-basal margin.
Effects of Flecainide on Reentrant Circuits Causing Sustained Ventricular Tachycardia
In four experiments, sustained monomorphic ventricular tachycardia with a stable cycle length was reproducibly induced by one of the stimulation protocols before flecainide administration. Sustained tachycardia could also be induced in each experiment after flecainide; the drug did not prevent tachycardia initiation with either of the dose schedules used. The electrocardiograms recorded in one of these experiments are shown in Fig 7⇓. The top two control panels show that before flecainide administration, tachycardia was not induced by a premature stimulated impulse (S2) with a coupling interval of 170 milliseconds but was induced by a prematurely stimulated impulse with a 150-millisecond coupling interval. After flecainide, tachycardia was induced by a prematurely stimulated impulse with a 170-millisecond coupling interval. The increase in the coupling interval at which tachycardia could be induced was found in all four experiments. There was a significant increase in the cycle length of tachycardia in all experiments after flecainide, from a mean of 171±20 milliseconds in control to 210±23 milliseconds (P<.05). This increase in cycle length is also apparent in the records shown in Fig 7⇓.
Changes in Activation Patterns and Conduction in Reentrant Circuits
Summary. All sustained tachycardias were associated with double loop (figure-of-eight) reentrant circuits. Accompanying the flecainide-induced changes in the ECG during the sustained tachycardia were changes in these reentrant circuits including (1) slowing of total activation time around the circuits, (2) elongation and narrowing of the central common pathway, and (3) acceleration of activation in the central common pathway. These changes occurred initially at flecainide plasma levels of 1.15 ng/mL and became more pronounced as plasma levels increased to 4 ng/mL.
Example of reentrant circuit causing sustained ventricular tachycardia. Fig 8B⇓ (top) shows the reentrant circuit during the sustained ventricular tachycardia (cycle length, 153 milliseconds) shown in Fig 7⇑ (control ECG), before flecainide administration. The activation pattern is a double loop configuration, with the functional lines of block (thick black lines in the figure) on either side of the central common pathway oriented parallel to the orientation of the long axis of the myocardial fibers. At time zero (within the 10-millisecond isochrone), the reentrant wave front leaves the LAD end of the central common pathway and splits into two wave fronts. One wave front moves toward the apex (to the left) and then turns to activate the apical region from LAD toward the lateral margin (isochrones 50 to 90 at the left). The second wave front moves toward the base (to the right) and then turns to activate the basal region from LAD toward the lateral margin (isochrones 50 to 90 at the right). The two wave fronts coalesce after 100 milliseconds near the lateral margin, and activity then enters the central common pathway and completes the circuit after 150 milliseconds.
Representative electrograms recorded from around each of the reentrant circuits are shown in Figs 8A⇑ and 8C⇑. The recording sites are indicated on the map by the circled activation times. The exit route from the circuit to the ventricles is at the LAD margin.
Effects of flecainide on reentrant circuit causing sustained ventricular tachycardia. After flecainide, the cycle length of tachycardia in the example shown in Fig 7⇑ increased to 212 milliseconds and the morphology of the tachycardia QRS was unchanged, but there was a marked increase in QRS duration (Fig 7⇑, ECG in bottom panel). The reentrant circuit after flecainide was located in the same area of the epicardial border zone as before the drug was administered, and is shown in Fig 8E⇑. The activation pattern is still a double loop; in the time window shown, activation begins in the 10-millisecond isochrone toward the LAD margin and splits into two wave fronts, one moving along the left and one along the right margins. The two wave fronts coalesce after 150 to 160 milliseconds and return toward the LAD margin in the central common pathway between the two lines of block (thick black lines), which is activated between 195 and 210 milliseconds. The exit route from the circuit is still at the LAD margin.
There are several alterations in the characteristics of the reentrant circuit described in Fig 8⇑ that were caused by flecainide and were found in the other experiments as well. First, activation time around the entire circuit is prolonged. Slowing of conduction occurs in the longitudinal direction (parallel to fiber orientation) outside the central common pathway, as is evidenced in the map (Fig 8E⇑) by the bunching of the isochrones on the apical and basal sides of the lines of block. Slowing in the transverse direction (perpendicular to fiber orientation) occurs around the ends of the lines of block, as evidenced by the bunching of the isochrones in these regions. Activation of the central common pathway (which is parallel to the long axis of the fiber bundles) after flecainide, however, is faster than control, decreasing from about 45 to about 17 milliseconds, and the central common pathway is narrower and longer (see below). Nevertheless, the time for activation of the complete circuit increased from 153 to 210 milliseconds, accounting for the increased cycle length of the tachycardia. Slowing of activation is also shown by the representative electrograms recorded around both circuits (Fig 8D⇑ and 8F⇑), which are more widely separated and broader than before drug administration. (The locations of the electrogram recording sites are indicated by the circles on the activation map in Fig 8E⇑ and do not correspond to the recording sites in Fig 8B⇑ because the reentrant pathways are slightly different after flecainide, as we describe below.)
Changes in the central common pathway. The decrease in the width of the central common pathway of the reentrant circuit described in Fig 8⇑ and the acceleration of activation after flecainide are shown in more detail in Fig 9⇓. Before flecainide, there are two rows of electrodes within the central common pathway, with activation along each row taking 49 to 50 milliseconds (Fig 9A⇓, circled activation times). After flecainide, there is one row of electrodes in the central common pathway, with activation taking 30 milliseconds (Fig 9D⇓). Only the line of block on the basal side (to the right) remains in the same position after flecainide. The left (apical) line of block is shifted to the right. This shift is further illustrated on the right side of Fig 9⇓. In the control, the two rows of electrodes within the central common pathway (Fig 9B⇓ and 9C⇓) are both activated in the same direction from the lateral to the LAD margin. After flecainide, the left row of electrodes (Fig 9E⇓) is no longer in the central common pathway because of the shift in position of the line of block. This row is now activated in the opposite direction (from LAD to LL margin) (Fig 9E⇓) than the electrodes that remained in the central common pathway (Fig 9F⇓) because of its location in the apical part of the reentrant circuit.
Flecainide markedly slowed conduction in the transverse direction around the ends of the lines of block that formed the central common pathway and in doing so, extended the length of these lines and the length of the central common pathway (Fig 10⇓). Before flecainide (Fig 10⇓, control), the reentrant wave front moving around the top of the left line of block activated the row of electrodes that are circled nearly simultaneously (activation times of 37, 32, 34, 34). These electrograms are shown at the left of the map. This is characteristic of a broad wave front moving transversely, from right to left. After flecainide (Fig 10⇓, flecainide), conduction in this right to left direction is slowed so that more than 3 isochrones are interpolated in this region, meeting our criterion for designating block, thereby extending the left line of block upward. In addition to this observation and further confirming the transformation from slow conduction to block in this region is the sequence of activation of the same recording sites (circled), which are now on the distal side of the line of block. These sites are no longer activated nearly simultaneously, which indicated a broad transverse conduction wave front in the control panel, but rather are activated with increasing delay (76, 80, 106, 123), indicating a wave front moving longitudinally (from top to bottom in the map) on the side of the line of block outside of the central common pathway. These electrograms are shown at the left of the map.
There was also an increased delay in activation around the ends of the right line of block and possible extension of this line. Before flecainide (control at the top), it takes 31 milliseconds for activation to move around the top of the right line of block (from 16 to 47 milliseconds) in the transverse direction and 40 milliseconds for activation to move around the bottom of the right line of block in the transverse direction (from 87 to 127 milliseconds). The electrograms at these pivoting points are shown at the right. After flecainide (bottom panel), activation time increases to 45 milliseconds around the top of the right line of block and to 68 milliseconds around the bottom. The increase is shown in the corresponding electrograms. The increase in activation time increases the number of interpolated isochrones, thereby meeting our criteria for conduction block, and extending the length of this line of block.
Effects of Flecainide on Induction of Sustained Ventricular Tachycardia
Summary. Flecainide facilitated the induction of sustained ventricular tachycardia by prolonging the coupling interval of single premature impulses (S1S2) required to induce tachycardia from 152±8 to 175±9 milliseconds in the four experiments. Prolongation of the coupling interval occurred because, after both doses of flecainide, conduction of premature impulses in the epicardial border zone blocked at longer coupling intervals. Figs 11⇓ and 12⇓ show the effects of flecainide on propagation of premature impulses in a representative experiment (the ECG is shown in Fig 7⇑). Similar results were found in our other experiments.
Example of control induction of sustained ventricular tachycardia. In control (Fig 11⇑), stimulation at a basic cycle length of 280 milliseconds from the LAD margin results in activation of the epicardial border zone from the LAD toward the lateral margin in 90 to 105 milliseconds (panel A). After a premature stimulus with a coupling interval of 170 milliseconds (panel B), activation again proceeds from the LAD to the lateral margin, but local delay or block occurs in two regions indicated by the thick black lines at the 80-millisecond isochrone at the left and the 60-millisecond isochrone at the right. The areas distal to these short lines of block (length of 5 to 8 mm) are activated 40 to 60 milliseconds later by wave fronts that move around the lines of block in a nearly transverse direction. Activation also occurs toward the lateral (LL) margin between the two lines of block (isochrones 70 to 110). When the coupling interval of the premature stimulus is shortened to 150 milliseconds (panel C), the two short lines of transverse block are extended and become confluent (thick black line at the 60-millisecond isochrone). The wave front of activation moves around this line of block and the area distal to the line is activated antidromically 148 milliseconds after the stimulus and 86 milliseconds after activation proximal to the site of block. The delayed activation, compared with that following the S2 at 170 milliseconds, occurs because of the longer route the wave front travels and the slower transverse conduction that occurs as the wave front moves around the ends of the line of block. The delay in activation of the myocardium distal to the line of block is long enough so that the regions proximal to the block recover excitability. As a result, the premature impulse reactivates the proximal side of the line 118 milliseconds after block initially occurred. This is shown in panel E (T1). Activation begins at the asterisk on this map (10-millisecond isochrone), which shows continuation of the wave front from the 160-millisecond isochrone in the previous map (bottom left). This wave front travels in the antidromic direction (toward the LAD margin), splits into two separate wave fronts that move toward the apical and basal margins (isochrones 10 to 70), and then toward the lateral margin establishing the figure-of-eight reentrant pattern that causes sustained ventricular tachycardia.
The electrograms that were recorded along the pathway of propagation of the last basic stimulus (S1), the premature impulse (S2), and the first beat of the tachycardia (T1) are shown in Fig 11D⇑. The sites from which the electrograms were recorded are indicated on the maps by a box or circle around the activation times. Boxes are used when conduction is orthodromic in relation to the stimulus, and circles are used when conduction is antidromic in relation to the stimulus. During the last basic drive (S1), conduction proceeds uniformly in the orthodromic direction (arrow, S1). After the premature stimulus, at 150 milliseconds (S2), block occurs, as indicated by the two horizontal lines at the end of the long arrow. Electrogram morphology becomes monophasic as the wave front blocks. Activation distal to the line of block is shown by the electrogram that is activated at 148 milliseconds on map S2. Activation then proceeds retrogradely as the first reentrant impulse (electrograms labeled T1).
Induction of sustained ventricular tachycardia after flecainide. After the administration of flecainide, a premature stimulus of 170 milliseconds induced tachycardia in the example shown in Figs 7⇑ and 11⇑, although this coupling interval did not induce tachycardia in control. The activation maps during the initiation of tachycardia after flecainide are shown in Fig 12⇑. During the basic drive at the 280-millisecond cycle length from the LAD margin, activation proceeds more slowly than in control (Fig 12A⇑). Electrograms recorded along the pathway of propagation are shown in Fig 12D⇑ (S1). The electrograms have a longer duration than before flecainide. Following a premature stimulus with a coupling interval of 170 milliseconds (Fig 12B⇑; S2, 170 milliseconds), block of the premature wave front occurs after 100 to 110 milliseconds in two areas indicated by the thick black lines formed by the bunched isochrones. The electrograms at the sites of block become monophasic (see electrograms recorded during S2 in Fig 12D⇑). Although the lines of block are in a similar location as during control at the same coupling interval, they extend more toward the lateral margin (compare with Fig 11B⇑). The wave front of activation that is moving around the ends of the two lines of block in a direction transverse to the myocardial fibers turns back to activate the area distal to the block. It takes 122 milliseconds from the time the proximal side of the right line of block is activated until the distal side of this line of block is activated, whereas in control, activation at the distal side of the block occurs after 40 to 60 milliseconds after the 170-millisecond coupled premature impulse. Activity then spreads back across this line of block and reenters the area proximal to it. The reentering wave front is shown in the lower left map (T1) at the 20-millisecond isochrone and also by electrograms labeled T1 in Fig 12D⇑. The reentering wave front splits and propagates toward the base and apex to form a double-loop reentrant circuit.
Effects of Flecainide on Anisotropic Conduction in the Epicardial Border Zone and the Ventricular Effective Refractory Period
The effects of flecainide on initiation and perpetuation of reentry were associated with alterations in conduction caused by the drug that are evident on the activation maps. We quantified these effects on conduction in the longitudinal and transverse directions (see “Methods”) in the anisotropic epicardial border zone in 6 experiments (12 longitudinal vectors and 12 transverse vectors) during stimulation through the central electrodes (see Fig 1⇑). Flecainide significantly decreased conduction velocity at fast and slow stimulation rates in the direction parallel to fiber orientation (P<.02, ANOVA) and in the direction transverse to fiber orientation (P<.002, ANOVA) (see Table 1⇓). There was no significant effect of cycle length in these experiments on conduction velocity either in control or with flecainide. At the longest stimulus cycle length (336±71 milliseconds), flecainide decreased conduction velocity by 28% (64.2 to 46.4 cm/s) in the direction parallel to fiber orientation and by 20% (32.9 to 26.4 cm/s) in the direction transverse to fiber orientation. At the shortest stimulus cycle length (220±53 milliseconds), flecainide decreased conduction velocity by 29% (59.1 to 42.0 cm/s) in the direction parallel to fiber orientation and by 23% (32.0 to 24.8 cm/s) in the direction transverse to fiber orientation. The decrease in conduction velocity in the longitudinal direction was not significantly different from the decrease in the transverse direction (P>.1). The anisotropic ratio decreased insignificantly at each stimulus cycle length (P>.1) (Table 1⇓). The decreases in conduction velocity in the longitudinal and transverse direction are illustrated for one of the experiments in Fig 1⇑, B and C. In Fig 1B⇑ in control, the fast axis of conduction is along vectors 3 and 12 (conduction velocity is 57.25 and 53.12 cm/s, respectively) and the slow axis of conduction is along vectors 10 and 16 (conduction velocity is 22.11 and 27.59 cm/s). After flecainide (Fig 1C⇑), fast axis velocity is decreased to 26.1 and 39.1 cm/s, which is evident by the increased time it took the stimulated wave front to reach the LAD and lateral-apical margins. Slow axis velocity along vector 16 decreased to 21.4 cm/s, as shown by the increased time required to reach the basal margin. Conduction block occurred along vector 10, so no velocity was measured in this direction.
We also quantified the incidence of conduction block in the longitudinal and transverse direction caused by flecainide during stimulation at the central electrodes. In the control, we did not observe conduction block in the longitudinal direction at the long cycle length in 6 experiments (12 vectors), nor did block occur in any of the experiments after flecainide. We also did not observe conduction block in the control in the transverse direction. After flecainide, however, block occurred along 5 of the 12 transverse vectors. In Fig 1⇑, block occurred along vector 10 in panel B after flecainide (compare with control vector 10 in panel A). At the short-stimulus cycle length, conduction block occurred in the longitudinal direction in control in 1 of the 12 vectors and in the transverse direction in 2 of the 12 vectors. After flecainide, conduction block occurred in the longitudinal direction in 3 of the 12 vectors and in the transverse direction in 9 of the 12 vectors. Therefore, the occurrence of block caused by flecainide was more prevalent in the transverse direction at both the long- and short-stimulus cycle lengths.
Ventricular effective refractory period was determined at each of the stimulation sites (see “Methods”). Flecainide did not significantly increase effective refractory period at 3 of the 4 stimulation sites, including the LAD site, which was always on normal myocardium, and the central site, which was always in the epicardial border zone (Table 2⇓). There was a small but significant increase at the lateral stimulation site.
New Experimental Results
Our experiments provide detailed data for the first time on the effects of flecainide on reentrant circuits that cause both unsustained and sustained ventricular tachycardias in the in situ infarcted heart. Previously published studies have investigated the effects of this drug on reentry in circuits composed of normal atrial myocardium in both isolated perfused preparations21 and in situ22 and in normal ventricular myocardium in isolated hearts,23 where the electrophysiological effects of antiarrhythmic drugs are expected to be different from their effects on infarcts. Our data describe the effects of flecainide on a model of anisotropic functional reentry in nonuniformly anisotropic myocardium,14 where slow transverse activation caused by sparse electrical coupling among fiber bundles is a major contributing cause for the occurrence of reentry.15 The effects of antiarrhythmic drugs on this mechanism have not been previously investigated and are expected to be different from their effects on other reentrant mechanisms such as anatomical reentry around a fixed obstacle21 23 24 or functional reentry caused by nonanisotropic mechanism22 that have been studied previously.
Our results show new and some unexpected effects of flecainide on nonuniform anisotropic reentrant circuits, which include (1) conversion of unsustained ventricular tachycardia to sustained ventricular tachycardia by two different mechanisms, depending on whether unsustained tachycardia is polymorphic or monomorphic, (2) prolongation of the coupling interval at which premature impulses initiate sustained tachycardia without prolonging the refractory period, implicating an important role for anisotropy in the initiation of reentry, (3) narrowing and lengthening of the central common pathway of reentrant circuits causing sustained tachycardia and speeding of activation in it despite a generalized effect to slow conduction in both the longitudinal and transverse direction in other regions of the circuit that results in a prolongation of tachycardia cycle length, and (4) failure of sodium channel blockade to cause conduction block in the central common pathway that would terminate tachycardia despite very high plasma levels of drug. We discuss each of these new findings in terms of possible mechanisms that are responsible for them. Insights into these mechanisms are provided both by the results of our mapping experiments and by other published data in the literature from experiments that have used techniques to study cellular electrophysiological effects of antiarrhythmic drugs.
Effects of Flecainide on Unsustained Ventricular Tachycardia
Our results show that flecainide converts unsustained ventricular tachycardia to sustained ventricular tachycardia by two different mechanisms, depending on whether unsustained tachycardia is polymorphic or monomorphic. The most common effect that we observed was the conversion of a changing, unstable, functional anisotropic reentrant circuit causing polymorphic unsustained tachycardia to a new and stable circuit causing monomorphic sustained tachycardia (Figs 3⇑ and 5⇑). Less common was the utilization of the same anisotropic reentrant circuit to cause unsustained tachycardia before drug and sustained tachycardia after drug (Fig 6⇑). Frame et al21 showed that flecainide increased the persistence of reentry in experiments on isolated, superfused tricuspid rings of canine atrial muscle, converting unsustained arrhythmias to sustained arrhythmias, while Brugada et al23 demonstrated similar results in anatomic rings of ventricular muscle. Because of the fixed, anatomic nature of the reentrant pathway in those experiments, the sustained arrhythmia after flecainide had to be constrained to the exact same circuit as the unsustained arrhythmia before flecainide. Drug-induced changes in the reentrant pathway were prevented from occurring; therefore, this important effect of the antiarrhythmic drug that we observed was not described. Brugada et al23 were able to show changes in the reentrant circuit associated with conversion of unsustained tachycardias to sustained tachycardias in another model of functional reentry in normal ventricular myocardium.
The effect of flecainide on reentrant circuits in the epicardial border zone of infarcts causing unsustained tachycardia is likely to be related to its depressant effect on conduction resulting from sodium channel blockade.2 25 26 We found that sodium channel blockade caused by the drug was manifested as depression of conduction equally in both the longitudinal and transverse direction (Table 1⇑), contrary to the findings of others in normal ventricle, where longitudinal conduction was depressed more than transverse.27 28 It had previously been predicted that sodium channel blockade should have a more marked effect on longitudinal conduction,27 although this has not always been found.23 29 The difference in the results of our experiments from others may be a consequence of the nonuniform anisotropic myocardium that we studied as compared with normal uniform anisotropic myocardium in the other studies. During polymorphic unsustained tachycardia in the nonuniform anisotropic myocardium of the epicardial border zone, the size, shape, and location of the reentrant circuits are not stable,30 possibly because there is no region with a sufficiently high anisotropic ratio (fast longitudinal conduction velocity/slow transverse conduction velocity) to provide an adequate region of slow conduction to anchor the circuit in one location. Concomitant with changes in the circuit, tachycardia cycle lengths vary and tachycardia terminates after a short cycle (Fig 4⇑). At the short cycle lengths, the reentrant wave front probably encounters refractory myocardium, causing block of the reentrant impulse; conduction of the reentrant impulse is too rapid with respect to the path length of the circuit because of the absence of a region of sufficiently slow transverse conduction. Flecainide, by slowing conduction, prevents the short cycle length and block caused by it. Depression of conduction by flecainide also causes a region of sufficiently slow conduction for stabilization of the circuit so that it occurs in a fixed location.
Although our stimulation studies showed that there was a tendency for flecainide to depress both longitudinal and transverse conduction a little more at faster rates than at slower rates (Table 1⇑), these effects were marginal. One would expect flecainide to produce more depression of conduction at more rapid rates of stimulation because of its well-described use-dependent effects.31 However, our range of stimulation rates were limited from a long cycle length of 336 milliseconds to a short cycle length of 220 milliseconds.
In several experiments, as exemplified by the one shown in Figs 3 through 5⇑⇑⇑, flecainide also caused block in other regions of the epicardial border zone at short cycle lengths where block did not exist before the drug. This resulted in the formation of a new reentrant circuit that was stable. The reentrant circuit causing sustained tachycardia in Fig 5⇑ no longer appears to be a typical, functional anisotropic circuit because of the large central region of inexcitability, although anisotropy may contribute to the slow activation around this region. Therefore, the drug may sometimes convert one mechanism for reentry into another mechanism.
Unsustained monomorphic ventricular tachycardia occurred in circuits with a stable location, size, and shape, as shown in Fig 6⇑. Therefore, the epicardial border zone had appropriate nonuniform anisotropic properties to anchor the circuit in one region without drug. Reentry terminated without prior changes in cycle lengths or oscillations because of conduction block. There appeared to be a “weak link” in the circuit where block always occurred after a period of rapid repetitive activation. This “weak link” was always in a longitudinally oriented segment of the circuit where the safety factor for conduction has been described as being lower than in the transverse direction.32 It was surprising to us that flecainide, by further depressing conduction, did not convert this area to complete block, which would have prevented tachycardia. Flecainide slowed conduction throughout the rest of the reentrant circuit in both longitudinal and transverse directions, as was evident in the activation maps, thereby increasing the cycle length at this critical site. The slower rate of repetitive activation at the site of the “weak link” may have prevented the occurrence of block and caused reentry to become sustained. The importance of slowing of conduction in causing sustained reentry can be seen by comparing these results with the results of our previous experiments on the effects of d-sotalol on anisotropic reentry.33 Sotalol did not slow conduction in the epicardial border zone and did not convert unsustained reentry to sustained reentry.
We never found in these experiments that flecainide prevented tachycardia, unlike the experiments of Frame et al21 on the tricuspid ring, in which concentrations of the drug that were higher than those facilitating reentry terminated reentry by causing conduction block. Flecainide has a much stronger effect to prolong the effective refractory period in the atria than the ventricle,22 34 35 perhaps accounting for a more consistent atrial antiarrhythmic action.36 37
Effects of Flecainide on Sustained Ventricular Tachycardia
Flecainide did not prevent sustained ventricular tachycardia in any of our experiments in which sustained tachycardia was present before drug administration. The failure of flecainide to prevent tachycardia can be attributed to the inability of the drug to cause conduction block in critical regions of the reentrant circuit either during initiation by the premature impulse or during the sustained period.
The initiation of reentry by a premature impulse requires block of that impulse and conduction around the area of block that is slow enough to allow sufficient time for the myocardium proximal to the block to recover excitability. This enables an impulse to reenter and reexcite the area proximal to the block (see Fig 11⇑). Thus, the area of block of the premature impulse is an area of unidirectional block. Theoretically, sodium channel–blocking drugs might exert an antiarrhythmic effect by converting such an area of unidirectional block to bidirectional block by preventing conduction of the retrograde impulse and thus preventing the initiation of reentry. Flecainide has been shown to prolong the action potential duration of premature impulses in some regions of normal ventricular myocardium38 39 and to prolong refractoriness at short cycle lengths,40 an effect that might be predicted to lead to an increase in the effective refractory period of the premature response and retrograde (bidirectional) block. However, in our experiments, even high doses of flecainide did not cause retrograde block of the reentering premature impulse during initiation of tachycardia.
Flecainide increased the propensity for premature impulses to block in the antegrade longitudinal direction, as evidenced by increased block at longer coupling intervals (Fig 12⇑).41 The block was not a result of prolongation in effective refractory period that we did not observe to occur when we determined the effects of the drug on this parameter. The effective refractory period was measured at the central pacing electrode, which, although not exactly in the region of block, was very close to it. The failure of flecainide to prolong refractoriness of epicardial border zone muscle is contrary to the results of experiments by Krishnan and Antzelevitch40 on normal epicardium and may be a consequence of the lack of the transient outward current in epicardial border zone cells.42 The increased propensity for block of premature impulses at longer coupling intervals may be a consequence of the effects of flecainide on anisotropic conduction properties. Spach et al32 43 have proposed that there is an increased propensity for premature impulses to block in the longitudinal direction rather than the transverse direction in nonuniformly anisotropic myocardium because of the lower safety factor for conduction longitudinally. Myocardium at the site of block is still excitable (not effectively refractory) but is not excited by the weak inward current of the propagating premature impulse. Flecainide decreases the upstroke velocity by blocking sodium channels.2 25 26 Therefore, it causes premature impulses to conduct more slowly with a reduced stimulating efficacy. Since the safety factor in the longitudinal direction is already low,32 a severe reduction in sodium current of premature impulses caused by the drug should further reduce the stimulating efficacy causing block to occur at longer coupling intervals even without a prolongation of the effective refractory period. Because flecainide also depresses transverse conduction, it also facilitated induction of tachycardia by extending the lines of block transversely while slowing conduction around them and allowing more time for myocardium proximal to the area of unidirectional block to recover excitability.
During sustained tachycardia in the epicardial border zone of healing infarcts, the reentrant circuit often has a figure-of-eight configuration.11 Termination of tachycardia requires that conduction blocks in the central common pathway,44 which in most instances is oriented in the direction of the long axis of the myocardial fibers.45 Block in other regions of the circuit outside the central common pathway will still permit propagation of one of the two reentrant excitation waves to continue. Flecainide never caused longitudinal conduction block in the central common pathway in circuits causing sustained tachycardia in our experiments and therefore did not terminate tachycardia once it was initiated. Failure to cause longitudinal block during the sustained period of tachycardia compared with the propensity to cause longitudinal block of premature impulses during the initiation of tachycardia is most likely a consequence of the longer cycle length during the sustained phase and a faster action potential upstroke. During sustained tachycardia, flecainide decreased activation time through the central common pathway. In contrast, it slowed conduction in all other regions of the circuit, including the transverse direction and the longitudinal direction outside the central common pathway (see Fig 9⇑), in agreement with the results of our experiments that quantified the effects of flecainide on anisotropic conduction. Flecainide did cause block in the transverse direction, which extended the lines of functional block on either side of the central common pathway, thereby increasing the path length of the reentrant circuit and contributing to the increase in the cycle length of tachycardia. Our stimulation studies showed that flecainide increases the propensity for transverse block.
The lines of block bounding the central common pathway shifted slightly after flecainide, narrowing the central common pathway. This shift may have contributed to the acceleration of activation in the central common pathway. A possible mechanism for the acceleration is that narrowing of the central common pathway altered the shape of the leading edge of the wave front. In a wide central common pathway before drug, the leading edge of the reentrant wave front is expected to be convex. Computer modeling and theoretical considerations indicate that the more convex the wave front, the slower the propagation because current from the wave front required to depolarize myocardium in front of the advancing wave front is dispersed over a larger area.46 Narrowing of the central common pathway would decrease the convexity of the wave front, making it “flatter” and accelerating conduction because of the reduced area of myocardium it needs to depolarize. Another possible mechanism for acceleration of activation in the central common pathway is that electrotonic influences from wave fronts on the opposite sides of the lines of block outside the central common pathway may increase excitability of the myocardium in the central common pathway. Since the region outside the central common pathway is in a depolarized state at the time when the central common pathway is repolarized, current flow from the depolarized regions into the central common pathway is expected to reduce the level of the resting potential toward threshold. This effect would be facilitated by narrowing of the central common pathway. If the resulting increase in excitability outweighs any effect of inactivation of sodium channels, activation would be expected to speed up.
Why did flecainide fail to cause conduction block in the central common pathway that would terminate sustained tachycardia? One explanation is the lack of prolongation of the effective refractory period, as we discussed for the failure of the drug to prevent unsustained tachycardia. In addition, sodium channel–blocking drugs might cause conduction block in a region of a reentrant circuit by severely depressing the action potential upstroke. For this effect to occur without significant impairment of conduction in normal regions of the ventricles, the resting potential of muscle fibers in the circuit need to be partially depolarized and the inward sodium current of the cells significantly reduced.47 This would render the upstroke of the cells in the circuit more sensitive to the depressant effect of flecainide because unblocking is reduced at decreased membrane potentials and depression of the sodium current is intensified.26 We do not think that the membrane potentials of the cells in the central common pathway of the reentrant circuits in healed infarcts are sufficiently depolarized or the upstrokes are sufficiently depressed to enable block to occur with drug concentrations that would not cause significant toxic effects on normal regions of the heart. This statement is made on the basis of our measurements that show that conduction velocity in the central common pathway is often nearly normal.14 45
In clinical electrophysiological studies, class IC drugs such as flecainide have often been shown to be ineffective in preventing inducible reentrant ventricular tachycardia in patients with a prior myocardial infarction.1 5 48 This may explain failures of flecainide in the chronic therapy of this arrhythmia. In addition, class IC drugs may result in inducibility of sustained ventricular tachycardia during electrophysiological testing in patients who have only unsustained tachycardia induced in the absence of the drug.48 49 50 While this may be a manifestation of the proarrhythmic effect of the drug, it is still uncertain how conversion of inducible unsustained to sustained tachycardia is related to clinical proarrhythmia responses that are usually not defined on the basis of electrophysiological testing.5 8 51
There are many similarities of the electrophysiological responses to flecainide in the canine model of infarction that we studied and in patients with scar-related ventricular tachycardia. However, at the present time, we do not know the relation between the effects of flecainide on reentrant circuits that we have described in the experimental animal model and the effects of the drug on reentrant circuits causing tachycardia in humans. Although there is some evidence that the anisotropic properties of the myocardial cells in healed human infarcts may be instrumental in the genesis of tachycardia,52 the importance of anisotropic reentry as a cause of clinical tachycardia is, at present, uncertain. In addition to the possibility of a role for anisotropic reentry, other data also suggest the presence of reentrant circuits with discrete anatomic pathways in healed human infarcts.53 Therefore, the clinical significance of our experiments await a more complete understanding of the electrophysiological mechanisms that cause clinical tachycardia.
This study was supported by grant R37-HL-31393 and by Program Project Grant HL-30557 from the National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Md.
- Received September 26, 1994.
- Accepted November 20, 1994.
- Copyright © 1995 by American Heart Association
Coromilas J. Classification of antiarrhythmic agents: electropharmacologic basis and clinical relevance. In: Greenspan A, Waxman H, eds. Contemporary Management of Ventricular Arrhythmias. New York, NY: Cardiovascular Clinics, FA Davis Co; 1992:97-116.
Zipes DG. Management of cardiac arrhythmias: pharmacological, electrical and surgical techniques. In: Braunwald E, ed. Heart Disease. Philadelphia, Pa: WB Saunders Co; 1992:633.
Herre JM, Titus C, Oeff M, Eldar M, Franz MR, Griffin JC, Scheinman MM. Inefficacy and proarrhythmic effects of flecainide and encainide for sustained ventricular tachycardia and ventricular fibrillation. Ann Intern Med. 1990;113:671-676.
Echt DS, Liebson PR, Mitchell LB, Peters RW, Obias-Manno D, Barker AH, Arensberg D, Baker A, Friedman L, Greene HL, Huther ML, Richardson DW, CAST Investigators. Mortality and morbidity in patients receiving encainide, flecainide, or placebo: the cardiac arrhythmia suppression trial. N Engl J Med. 1991;324:781-788.
Levine JH, Morganroth J, Kadish AH. Mechanisms and risk factors for proarrhythmia with type IA compared with IC antiarrhythmic drug therapy. Circulation. 1989;80:1063-1069.
Ranger S, Talajic M, Lemery R, Roy D, Nattel S. Amplification of flecainide-induced ventricular conduction slowing by exercise: a potentially significant clinical consequence of use-dependent sodium channel blockade. Circulation. 1989;79:1000-1006.
Coromilas J, Saltman AE, Waldecker B, Dillon SM, Wit AL. Effect of class IC antiarrhythmic drugs on anisotropic reentry in canine infarcts. Circulation. 1988;78(suppl II):II-460. Abstract.
El-Sherif N. The figure 8⇑ model of reentrant excitation in the canine postinfarction heart. In: Zipes DP, Jalife J, eds. Cardiac Electrophysiology and Arrhythmias. New York, NY: Grune & Stratton; 1985:363-378.
Cardinal R, Savard P, Carson DL, Perry JB, Page P. Mapping of ventricular tachycardia induced by programmed stimulation in canine preparations of myocardial infarction. Circulation. 1984;70:136-148.
Kramer JB, Saffitz JE, Witkowski FX, Corr PB. Intramural reentry as a mechanism of ventricular tachycardia during evolving canine myocardial infarction. Circ Res. 1985;56:736-754.
Dillon SM, Allessie MA, Ursell PC, Wit AL. Influence of anisotropic tissue structure on reentrant circuits in the epicardial border zone of subacute canine infarcts. Circ Res. 1988;63:182-206.
Harris AS. Delayed development of ventricular ectopic rhythms following experimental coronary occlusion. Circulation. 1950;1:1318-1328.
Saltman AE. Anisotropic Conduction in the Infarcted Canine Ventricle: Conduction Characteristics of Stimulated and Reentrant Beats and the Influence of the Antiarrhythmic Drug Flecainide. Graduate School of Arts and Sciences, Columbia University, New York, NY, 1990. Thesis.
Ursell PC, Gardner PI, Albala A, Fenoglio JJ Jr, Wit AL. Structural and electrophysiological changes in epicardial border zone of canine myocardial infarcts during infarct healing. Circ Res. 1985;56:436-451.
Frame LH, Rhee EK, Fei H, Luchetti W. Proarrhythmic and antiarrhythmic effects of flecainide on nonsustained reentry around the canine atrial tricuspid ring in vitro. PACE. 1991;14:1728-1734.
Wang Z, Page P, Nattel S. Mechanism of flecainide’s antiarrhythmic action in experimental atrial fibrillation. Circ Res. 1992;71:271-287.
Brugada J, Boersma L, Kirchhof C, Allessie M. Proarrhythmic effects of flecainide: experimental evidence for increased susceptibility to reentrant arrhythmias. Circulation. 1991;84:1808-1818.
Frame LH, Page RL, Boyden PA Fenoglio JJ Jr, Hoffman BF. Circus movement in the canine atrium around the tricuspid ring during experimental atrial flutter and during reentry in vitro. Circulation. 1987;76:1155-1175.
Anno T, Hondeghem LM. Interactions of flecainide with guinea pig cardiac sodium channels: importance of activation unblocking to the voltage dependence of recovery. Circ Res. 1990;66:789-803.
Spach MS, Dolber PC, Heidlage JF, Kootsey JM, Johnson EA. Propagating depolarization in anisotropic human and canine cardiac muscle: apparent directional differences in membrane capacitance: a simplified model for selective directional effects of modifying the sodium conductance on Vmax, tfoot and the propagation safety factor. Circ Res. 1987;60:206-219.
Kadish AH, Spear JF, Levine JH, Moore EN. The effects of procainamide on conduction in anisotropic canine ventricular myocardium. Circulation. 1986;74:616-625.
Turgeon J, Wisialowski TA, Wong W, Altemeier WA, Wikswo JP, Roden DM. Suppression of longitudinal versus transverse conduction by sodium channel block: effects of sodium bolus. Circulation. 1992;85:2221-2226.
Coromilas J, Waldecker B, Wit AL. Characteristics of inducible nonsustained ventricular tachycardia in the subacute phase of canine myocardial infarction. Circulation. 1992;86(suppl I):I-133. Abstract.
Spach MS, Miller WT, Geselowitz DB, Barr RC, Kootsey JM, Johnson EA. The discontinous nature of propagation in normal canine cardiac muscle: evidence for recurrent discontinuities on intracellular resistance that affect the membrane currents. Circ Res. 1981;48:39-54.
Coromilas J, Saltman AE, Wit AL. Effect of class III antiarrhythmic drugs on anisotropic reentry in canine infarcts. Circulation. 1990;82(suppl III):III-452. Abstract.
Kidwell GA, Gonzalez MD. Effects of flecainide and D-sotalol on myocardial conduction and refractoriness: relation to antiarrhythmic and proarrhythmic drug effects. J Cardiovasc Pharmacol. 1993:21:621-632.
Wang Z, Pelletier LC, Talajic M, Nattel S. Effects of flecainide and quinidine on human atrial action potentials: role of rate-dependence and comparison with guinea pig, rabbit and dog tissue. Circulation. 1990;82:274-283.
Krishnan SC, Antzelevitch C. Flecainide-induced arrhythmia in canine ventricular epicardium: phase 2 reentry? Circulation. 1993;87:562-572.
Olsson SB, Edvardsson N. Clinical electrophysiological study of antiarrhythmic properties of flecainide: acute intraventricular delayed conduction and prolonged repolarization in regular paced and premature beats using intracardiac monophasic action potentials with programmed stimulation. Am Heart J. 1981;102:864-871.
Krishnan SC, Antzelevitch C. Sodium channel block produces opposite electrophysiological effects in canine ventricular epicardium and endocardium. Circ Res. 1991;69:277-291.
Starmer CF, Lastra AA, Nesterenko VV, Grant AO. Proarrhythmic response to sodium channel blockade: theoretical model and numerical experiments. Circulation. 1991;84:1364-1377.
Lue W-M, Boyden PA. Electrophysiologic properties of myocytes from the chronically infarcted canine heart: alterations in Vmax and the transient outward current. Circulation. 1992;85:1175-1188.
Spach MS, Dolber PC. Relating extracellular potentials and their derivatives to anisotropic propagation at a microscopic level in human cardiac muscle: evidence for electrical uncoupling of side-to-side fiber connections with increasing age. Circ Res. 1986;58:356-371.
El-Sherif N, Mehra R, Gough WB, Zeiler RH. Reentrant ventricular arrhythmias in the late myocardial infarction period: interruption of reentrant circuits by cryothermal techniques. Circulation. 1983;68:644-656.
Josephson ME, Hanna MS, Coromilas J, Waldecker B, Dillon SM, Wit AL. Is there a critical zone of slow conduction in reentrant circuits causing ventricular tachycardia? Circulation. 1992;86(suppl I):I-132. Abstract.
Cabo C, Pertsov AM, Baxter WT, Davidenko JM, Gray RA, Jalife J. Wavefront curvature as a cause of slow conduction and block in isolated cardiac muscle. Circ Res. 1994;75:1014-1028.
Boyden PA, Wit AL. Pharmacology of the antiarrhythmic drugs. In: Rosen MR, Hoffman BF, eds. Cardiac Therapy. Boston, Mass: Martinus-Nijhoff Publishers; 1983:171-234.
Rinkenberger RL, Prystowsky EN, Jackman WM, Naccarelli GV, Heger JJ, Zipes DP. Drug conversion of nonsustained ventricular tachycardia to sustained ventricular tachycardia during serial electrophysiologic studies: identification of drugs that exacerbate tachycardia and potential mechanisms. Am Heart J. 1982;103:177-184.
Wit AL, Janse MJJ. The Ventricular Arrhythmia of Ischemia and Infarction: Electrophysiological Mechanism. Mt Kisco, NY: Futura Publishing Co; 1993.
DeBakker JMT, van Capelle FJL, Janse MJ, Wilde AAM, Cornel R, Becker AE, Dingemas KP, van Hemel NM, Hauer RNW. Reentry as a cause of ventricular tachycardia in patients with chronic ischemic heart disease: electrophysiologic and anatomic correlation. Circulation. 1988;77:589-606.