Reentrant Arrhythmias in the Subacute Infarction Period
The Proarrhythmic Effect of Flecainide Acetate on Functional Reentrant Circuits
Background The Cardiac Arrhythmia Suppression Trial has shown that flecainide was associated with an increased incidence of sudden cardiac death in postinfarction patients. The exact mechanism(s) of the proarrhythmic effects of flecainide remain unclear. We performed a detailed analysis of the electrophysiological and proarrhythmic effects of flecainide in a well-characterized model of reentrant arrhythmias in the subacute phase of myocardial infarction.
Methods and Results Sixteen dogs were studied 4 days after ligation of the left anterior descending coronary artery. Isochronal mapping of ventricular activation showed that flecainide facilitated both the induction and sustenance of ventricular tachycardia, especially at shorter basic cycle lengths. Flecainide had negligible effect on the length of the arc of functional conduction block but markedly depressed conduction of the common reentrant wave front that was usually oriented parallel to fiber axis. Whole heart mapping was analyzed in combination with basic measurements of the effects of flecainide on conduction and refractory properties of both normal and ischemic myocardia using a high-resolution cross electrode consisting of four orthogonal arms, each comprised of 16 poles with an interelectrode spacing of 500 μm. The electrode was especially designed to study the effects of the drug on anisotropic conduction as determined by a linear regression of activation time and distance in each direction. Flecainide resulted preferentially in more marked rate-dependent depression of conduction in ischemic compared with normal myocardium. On the other hand, the effect of flecainide on refractoriness in both normal and ischemic myocardia was negligible.
Conclusions Because flecainide caused no significant change in refractoriness in both normal and ischemic myocardia, there was no difference in the dimension of the potential reentrant pathway, that is, the continuous line of functional conduction block, around which the reentrant wave fronts circulate. Yet, flecainide resulted in significant rate-dependent slowing of conduction preferentially in ischemic myocardium. The additional slowing of conduction of the common reentrant wave front coupled with minimal changes in the length of the reentrant pathway allowed additional time for the wave front to reexcite normal myocardium on the proximal side of the arc of block. After flecainide, reentry could be induced in hearts in which reentry could not be induced during control. The same proarrhythmic mechanism explains the propensity of nonsustained figure-8 reentrant tachycardias to become sustained after flecainide.
Type Ic antiarrhythmic agents (flecainide and encainide) have been associated with an increased incidence of sudden cardiac death when used as therapy for ventricular arrhythmias in patients after myocardial infarction. These results were highlighted by the Cardiac Arrhythmia Suppression Trial (CAST).1 The exact mechanism of proarrhythmia remains unclear in both clinical2 3 and experimental studies.4 5 6 Changes in conduction properties of reentrant circuits by pharmacological agents involve a complex interplay of functional properties, including excitability and refractoriness as well as anisotropic properties, which are difficult to assess using standard electrophysiological techniques.7 Furthermore, there are no reports of the differential effects of flecainide on rate-dependent changes in conduction velocity and refractoriness in normal and ischemic ventricular myocardia.
In the present report, we performed a detailed analysis of the electrophysiological effects and mechanism(s) of proarrhythmia associated with flecainide, using a well-characterized functional model of circus movement reentry in the subacute myocardial infarction period in the dog. Using high-resolution mapping techniques and whole heart activation mapping in the same heart in which reentrant arrhythmias were induced, the mechanism of proarrhythmic action of flecainide was investigated. Preliminary results have been reported.8
Details of the ligation procedure have been described previously.9 Briefly, conditioned, heartworm-negative, mongrel male dogs were anesthetized with sodium pentobarbital (30 mg/kg IV) and received supplemental doses as needed. Dogs were ventilated by a cuffed endotracheal tube with a Harvard Apparatus positive-pressure ventilator. Under sterile conditions, a left-sided thoracotomy was performed at the fourth intercostal space to expose the heart. The left anterior descending coronary artery was dissected free and ligated (one stage) just distal to the septal artery. The chest cavity then was closed, the thorax evacuated, and the animal was allowed to recover. Postoperative pain was treated with buprenorphine hydrochloride (0.005 to 0.02 mg/kg IV). Four to 5 days after the initial ligation, the dog was reanesthetized with sodium pentobarbital in the same manner as described above, receiving supplemental doses and fluids through a catheter placed in the cephalic vein. The dogs were ventilated with room air through a cuffed endotracheal tube using a Harvard Apparatus positive-pressure respirator. ECG lead II and aortic blood pressure (Statham) were continuously monitored on an Electronics for Medicine DR12 monitor (PPG Industries). The heart was exposed through a left thoracotomy. Core temperature and intrathoracic temperature were monitored using two electronic thermometers (Yellow Spring Instruments). To slow the sinus rate, a Grass S88 stimulator was used to stimulate the right and left vagosympathetic trunks through two pairs of PTFE-insulated silver wire (.010-in. diameter) with square wave pulses, 0.1 to 0.5-ms duration, at a frequency of 20 Hz, at 1 to 10 V.10
Whole Ventricle Isochronal Mapping
A sock electrode array was placed on the ventricular surface for simultaneous recording at 126 epicardial sites. Each bipolar electrode consisted of a pair of silver wires (.005-in. diameter) sutured to the sock with an interpolar distance of 0.8 to 1.4 mm. The distance between electrodes ranged between 3 and 10 mm, with a higher concentration of electrodes covering the zone overlying the infarction. Additional details of the electrode arrangement and recording technique are reported elsewhere.11
Programmed electrical stimulation was provided by a Bloom DTU-101 digital stimulator through a bipolar plunge electrode consisting of two hooked stainless steel wires (enamel coated, .005-in. diameter) placed in a 23-gauge hypodermic needle. The control stimulation site was located either in (1) the right ventricle, adjacent to the septal border of the infarct, or (2) the left ventricular base, near the basal lateral border of the infarct.
After surgical preparation and sock electrode placement, the ribs were approximated and the chest cavity was closed. A circulating hot water heating blanket was positioned about the thorax. Once the core temperature had stabilized, programmed stimulation was applied to the control site in order to induce reentrant rhythms. The control stimulation sequence consisted of a train of 29 basic driven beats using two cycle lengths (S1S1) of 500 and 300 ms. Stimuli were applied at twice diastolic threshold, followed by a premature stimulus, S2. The premature stimulus was introduced at decreasing coupling intervals, beginning at 280 ms until refractoriness was reached.
Induced ventricular rhythms resulting from a single premature stimulus, S2, were classified into three categories: (1) no reentrant response (NR), (2) nonsustained reentrant activity (NS-RA, one or more unstimulated reentrant beats lasting less than 30 seconds), and (3) reentrant ventricular tachycardia (VT, a sustained monomorphic rhythm lasting more than 30 seconds).
Electrogram recordings of activation during programmed electrical stimulation were obtained from the data acquisition system and used to construct isochronal maps of epicardial activation. Isochrons were delineated by closed contours at 20-ms intervals beginning with the earliest detected time of activation. For the whole ventricle activation maps, sites of functional unidirectional conduction block were identified using previously defined criteria of an activation time difference between adjacent recording sites of more than 40 ms, electrotonic deflections representing distant activation, or evidence of retrograde conduction to that site.11 A continuous line was drawn through these regions and was defined as an arc of functional conduction block.
High-Density Electrode Recordings of Conduction Properties
To measure conduction velocities, we used a cross arrangement of closely spaced unipolar electrodes. The electrode arrangement is shown in Fig 1⇓. Each electrode pole consisted of a PTFE-coated silver wire (125 μm in diameter) imbedded in a flexible acrylic substrate (Flexacryl, Lang Industries). The cross-electrode plaque consisted of four orthogonal arms comprised of 16 poles each, with an interelectrode spacing of 500 μm. To minimize virtual cathode effects, the first electrode pole in each arm was situated 2 mm from a central unipolar stimulating wire (200-μm diameter). The cross electrodes were polished flush at the beginning of each experiment, permitting only the exposed cross section of the wire to contact the tissue. We find that this procedure is essential for high-fidelity recordings and low-threshold currents. All electrograms were acquired at a sampling rate of 1 to 8 kHz per channel using two DSC2000 128-channel data acquisition modules (INET Corp) connected to an IBM-compatible 486/33 EISA bus computer (Touche Corp). All data were filtered at a bandpass of 0.05 to 500 Hz.
Crosses were carefully situated under the sock so as not to disturb the position of the sock recording sites. The plaques were placed in at least two locations: (1) ischemic zone, with the distal pole of the cross 3 to 4 mm distal to the arc of block (as defined by the whole ventricle maps) at the septal border of the infarct, and (2) normal zone, with the distal pole of the cross 3 to 4 mm proximal to the arc either at the left ventricular base or the right ventricle. The cross was situated parallel and perpendicular to the longitudinal fiber axis of the ventricles. Correct positioning was verified by maximizing anisotropic conduction velocity ratio.12 Unipolar stimulation was applied at the center of the cross at 1.3 times diastolic threshold. Measurements were only performed for a diastolic threshold less than 0.6 mA. Once electrode positions and recordings proved satisfactory, the ribs were approximated and the chest cavity was closed. Electrode sites from the high-resolution plaque were used in conjunction with the remaining sock electrode site to construct whole heart maps before and after drug administration. Whole heart and high-resolution recordings were made once the intrathoracic temperature had returned to control level. Whole heart activation maps were validated after plaque placement and after each electrophysiological measurement protocol, during control, and after drug.
Conduction velocity was determined by a linear regression of activation time and distance along each row. Activation times were determined by either the maximum negative derivative of the unipolar electrogram or peak of uniphasic positive bipolar electrogram. A minimum of six activation points were used for each determination.
Flecainide was administered as a bolus dose of 1 mg/kg given over 8 minutes, followed by a maintenance infusion of 1 mg/kg per hour. For determination of plasma levels of flecainide, blood samples were obtained just before the electrophysiological study. Plasma levels were measured 30 to 40 minutes after start of infusion at a time after the rapid elimination phase.13 14 In four experiments, a second plasma level was obtained 1 hour later. Levels were measured by immunoassay technique (SmithKline Beecham Clinical Laboratories).
Refractory Period Determination
Effective refractory periods were determined at selected sites proximal and distal to the arc of functional conduction block during control and flecainide infusion. Refractory periods were determined at basic cycle lengths of 500 and 300 ms. As in our previous studies,11 refractory periods were determined at twice diastolic threshold. Threshold was determined during the basic driven rate (S1S1) of 500 ms, using a current resolution of 0.05 mA. Because of the frequency dependent binding of type I agents, the standard 8–driven beat protocol was modified to ensure steady-state binding.15 The procedure was as follows: premature stimuli (S2) were introduced at decreasing intervals (10 ms), starting at 280 ms, following a train of 9 basic driven beats; upon failure of a propagated response, the drive train length was increased to 29 beats. The S1S2 coupling interval was increased by 15 ms, then again introduced at decreasing coupling intervals of 5 ms. The effective refractory period of each site was defined as the longest interstimulus interval, S1S2, that failed to produce a locally propagated response. A maximum diastolic threshold of 3.0 mA was used throughout.
Strength-interval curves were measured at normal and ischemic sites on either side of the line of functional conduction block by unipolar application of cathodal current through a site on the high-density electrode array. After determination of threshold current at a 500-ms basic cycle length, current strengths were determined beginning at an S1S2 interval of 300 ms, then at decreasing increments of 2 to 15 ms. For each successive test interval, the current intensity was decreased at least 100 μA from the preceding test interval and varied in increasing 50-μA steps until local impulse propagation was detected. The procedure was terminated when the current exceeded either 10 times threshold or 6 mA.
Strength-interval data were fit to a hyperbolic function. The fitted equation is of the form
where I is the applied current strength at each premature interval, ΔS is the premature coupling interval, ARP is the absolute refractory period, Ithr is the diastolic threshold, and a and n are constants.
Data are reported as mean±SD. Conduction velocities were determined as described above by linear regression of activation times versus interelectrode distance. Effective refractory periods and diastolic thresholds for normal and ischemic zones before and after flecainide were tabulated. The data were compared using Student’s t test for paired data. A confidence level of 95% was considered statistically significant. Comparison of rate-dependent changes in longitudinal and transverse conduction properties for normal and ischemic zones before and after flecainide were made using ANOVA for repeated measures. Where ANOVA failed to reach statistical significance, Bonferroni’s correction for paired data was applied to test for effects of drug.16 Pearson’s χ2 test was applied to test for proarrhythmic versus nonproarrhythmic events for each cycle length. Statistical tests were performed using spss, version 4.1, for IBM VM/CMS (SPSS) or systat for Windows, version 5.03 (Systat). Curve fitting was performed with table curve for Windows, version 1.0 (Jandel Scientific). Computer-generated graphics were performed using origin for Windows, version 3.0 (MicroCal Software, Inc) or corel draw, version 4.0 (Corel Corporation).
Effects of Flecainide on Epicardial Activation Patterns
Sixteen dogs were included in the analysis. Flecainide had an insignificant effect on blood pressure (control, 114±12/81±14 mm Hg; flecainide, 112±18/81±14 mm Hg). The mean plasma level of flecainide was 0.88±0.07 mg/L for 15 dogs (in one experiment, blood samples were damaged during shipping). In 4 dogs, plasma levels were measured 1 hour after the beginning of the electrophysiological protocol. There was no significant difference in plasma level over this period (0.83±0.10 versus 0.78±0.10 mg/L).
In 5 of 7 dogs for which reentry was not induced during control, flecainide induced nonsustained reentrant activity (3) or sustained reentrant tachycardia (2). In the remaining 9 dogs, flecainide transformed nonsustained into sustained reentrant ventricular tachycardia in 5 dogs. In 2 dogs, two distinct tachycardia morphologies were induced. Seven of the 9 sustained monomorphic tachycardias induced by flecainide could be mapped in space and time as complete epicardial circuits. Sustained ventricular tachycardia (cycle length, 270±67 ms; range, 165 to 310 ms) was hemodynamically stable in 6 of 7 dogs. In 1 dog, ventricular tachycardia rapidly degenerated into ventricular fibrillation.
The proarrhythmic effect of flecainide at two different basic cycle lengths, 500 and 300 ms, is summarized in Fig 2⇓. Data shown are for 16 dogs in which induction of reentry was attempted with only a single premature beat (S2). Proarrhythmic events were defined as the drug-induced transformation of (1) no inducible reentry (NR) to either nonsustained reentrant activity (NS-RA) or sustained ventricular tachycardia (VT) or (2) nonsustained reentrant activity (NS-RA) to sustained ventricular tachycardia (VT). There were four proarrhythmic events in 16 dogs (25%) for a basic cycle length of 500 ms and 10 proarrhythmic events for a basic cycle length of 300 ms (62.5%). The proarrhythmic effect of flecainide was more prominent at a basic cycle length of 300 ms (P<.005, Pearson χ2). There was no significant difference in plasma levels between the two groups (0.88±0.07 mg/L, n=9, proarrhythmia; 0.88±0.08 mg/L, no proarrhythmia, n=6).
Fig 3⇓ shows a heart in which flecainide resulted in unstimulated reentrant responses and sustained ventricular tachycardia that were not observed in the control, predrug state. The maps shown in this and following figures are polar projections of the ventricles. The perimeter represents the atrioventricular annulus with the apex at the center. The dashed line in the left map represents the location of the left anterior descending artery. The ischemic zone is indicated by the dotted contour. Stimulation was applied from the right ventricle, near the outflow tract, and is indicated by the square pulse symbol. In this particular experiment, there was a region of necrosis extending through to the epicardial surface and is indicated in the maps by shading. The left map, a control isochronal map, shows premature stimulation after a drive of 29 S1 beats. After a premature stimulus (S1S2 of 170 ms), the premature activation wave front encountered a continuous line of functional unidirectional conduction block within 40 ms. Activation continued as two wave fronts that circulated clockwise and counterclockwise around the functional obstacle and coalesced within 80 ms. Conduction then proceeded retrogradely through the slow zone and blocked during the 140-ms isochron before reaching the distal border of the arc.
The proarrhythmic effect of flecainide is illustrated in the middle map of Fig 3⇑. After application of a premature stimulus at the same S1S2 interval, the activation wave front again encountered a continuous line of functional conduction block within 40 ms. The line of conduction block was essentially identical to the control predrug map. Similarly, the coalesced common reentrant impulse reached the entrance to the slow zone at the 100-ms isochron. However, conduction within the slow zone proceeded much slower after flecainide. Because of the additional delays incurred, the impulse was able to conduct slowly through the area of bidirectional block seen during control. Reactivation at the septal border near the apex occurred within 220 ms. In this experiment, a sustained monomorphic ventricular tachycardia (right map, Fig 3⇑) was induced. The reentrant circuit appeared as two impulses that circulated clockwise and counterclockwise around two continuous lines of functional conduction block. The cycle length of the tachycardia was 228 ms.
Fig 4⇓ shows lead II ECGs from an experiment during which, in the control predrug state, premature stimulation after a train of 29 basic driven beats did not induce unstimulated ventricular responses at either basic cycle length (500 or 300 ms). The ECGs for a premature coupling of 170 ms are shown. After flecainide infusion, no unstimulated responses were elicited at a cycle length of 500 ms. However, at a cycle length of 300 ms, a single extrastimulus induced a sustained monomorphic ventricular tachycardia.
The activation patterns during basic cycle lengths of 500 and 300 ms for the example shown in Fig 4⇑ are presented in Fig 5⇓. The upper left map is the isochronal activation pattern of the 29th basic driven beat at 500 ms. Both ventricles were activated within 80 ms. The isochronal maps indicate some slowing of conduction at the septal border of the infarct. There was no evidence of conduction block in this and the remaining three maps. The upper right map shows that after flecainide, the heart was again activated within 80 ms. Except for some additional crowding of isochrons at the septal infarct border, the activation pattern was quite similar to control. The lower row of S1 maps shows that reducing the basic cycle length to 300 ms had little effect on the control activation pattern. However, after flecainide, there was additional conduction delay that resulted in an increase in total epicardial activation time to 100 ms.
The activation maps of S2 for the experiment illustrated in Figs 4⇑ and 5⇑ are shown in Fig 6⇓. The upper left map shows the control predrug activation pattern for an S1S2 coupling of 170 ms after a 500-ms drive train. Similar to the description of Fig 3⇑, an arc of functional conduction block formed during the 40-ms isochron. Activation reached the distal border of the arc during the 160-ms isochron. Note that the direction of the common wave front is oriented perpendicular to the left anterior descending coronary artery, or parallel with the epicardial fiber axis. Because the maximum activation time difference across the arc of block was only 120 ms, the impulse was unable to reenter. The upper right map illustrates that after flecainide, reentry also could not be induced by premature stimulation at the same coupling interval. Except for a slight shift in the apical portion of the arc and subtle changes in the isochronal contours, the activation map is essentially the same as control. The lower left map shows that after a control drive train of 300 ms, reentry was not elicited by the same premature coupling. The lower right map shows that flecainide produced more dramatic changes in S2. Again, the premature wave front blocked within 40 ms. Flecainide depressed conduction in the slow zone and delayed activation at the distal border of the arc of block to the 200-ms isochron. The activation time difference across the arc of block was now 160 ms, and the impulse was able to reenter near the site of initial conduction block.
Effects of Flecainide on Refractoriness in Normal and Ischemic Myocardia
The role of flecainide-induced changes in refractoriness were examined. Fig 7⇓ shows arcs of functional conduction block before and after flecainide for the same experiment shown in the previous figure. The solid line represents the arc of block during flecainide infusion; the dotted line represents the control drug-free state. Effective refractory periods at selected sites are indicated; refractory periods after flecainide for the same site are indicated in parentheses. Our previous published reports established that arcs of functional conduction block are due to regional disparities in refractoriness along a continuous border.11 One can see that the refractory period differences across the arc were similar before and after flecainide administration. There was some shortening of the arc of block during flecainide that can be explained by the additional drug-induced slowing of conduction proximal to this region during S2.
Table 1⇓ illustrates the effect of flecainide on excitability and effective refractory periods proximal and distal to the arc of block. As expected, flecainide increased diastolic threshold at both normal and ischemic sites. However, there were no significant changes in refractory values by flecainide at either cycle length in both the normal and ischemic zones.
Effects of Flecainide on Strength-Interval Relation of Normal and Ischemic Myocardia
The effect of flecainide on strength-interval relations of normal and ischemic epicardia also was investigated. Fig 8⇓ shows the results from an experiment in which the normal test site was in the right ventricle, just proximal to the line of block along the septal border of the infarct. The ischemic site was approximately 1 cm lateral to the normal site, just distal to the line of block. Because of the biasymptotic nature of strength-interval relations about threshold and the absolute refractory period, data were fit with a hyperbolic function. The fitted equation is of the form
where I is the current strength, ΔS is the coupling interval, ARP is the absolute refractory period, and Ithr is the diastolic threshold. The constants a and n define the shape of the curve. As one would expect, the strength interval for normal myocardium in the predrug state (left curve) is sharply asymptotic. The absolute refractory period was 167 ms. In keeping with the expected type Ic behavior of flecainide, there was no change in the absolute refractory period. Because of the membrane anesthetic effect, there was an upward shift in the curve. The knee of the strength-interval curve after flecainide became blunted compared with control. For the ischemic site (right curve), the knee of the control curve was markedly blunted compared with the shape of the curve for the normal site. After flecainide, the curve shifted upward and to the right. The fitted data showed an increase in the absolute refractory period due to flecainide (198 versus 190 ms, control). Still, based on the definition of effective refractory period used in the present study (two times diastolic threshold), there was no significant change in effective refractory period by flecainide (see Table 1⇑). For this example, there was a difference in diastolic threshold between the normal and ischemic sites; however, it should be noted that no significant difference in threshold values was detected for the grouped analysis (see Table 1⇑). This finding is in agreement also with a previous report by Restivo et al,11 who used the same animal model.
Effects of Flecainide on Anisotropic Conduction Properties of Normal and Ischemic Myocardia
The role of flecainide-induced changes in conduction velocity in normal and ischemic tissues was examined using the high-resolution cross electrode. Fig 9⇓ shows selected bipolar electrograms (500-μm interelectrode spacing) from a cross electrode placed in the ischemic zone, distal to the defined arc of block. During control, conduction was more rapid for propagation along the longitudinal axis of the ischemic myocardial fibers (first electrogram) than for propagation across the fiber axis (third electrogram). Flecainide caused a reduction in conduction velocity for longitudinal (second electrogram) and transverse conduction (fourth electrogram). The slowing of conduction is demonstrated by the increased conduction time and widening of the bipolar electrograms. For accurate determination of conduction velocity, linear regressions of activation time versus distance from the central unipolar stimulating electrode were performed. An example of the regression analysis for the same experiment shown in Fig 9⇓ is presented in Fig 10⇓.
Grouped analysis of the rate-dependent effect of flecainide is illustrated in Fig 11⇓. To eliminate tonic effect of flecainide, drug values were normalized to the longest possible basic cycle length. In these infarcted hearts, it was difficult to achieve pacing rates at intervals much greater that 600 ms because of underlying idioventricular rhythms. Because of this fact and since a heart rate corresponding to a basic drive length of 600 ms is more physiologically relevant for dog heart, we present the effects of flecainide as the fractional conduction velocity normalized to the conduction velocity at the basic driven cycle length of 600 ms. Results shown are from 6 dogs. Repeated-measures ANOVA was used to test for differences between control and flecainide across different cycle lengths. There was a significant difference between groups for longitudinal propagation in normal and ischemic zones and for transverse propagation in the ischemic zone. Paired t tests then were used to localize significant differences between the control and flecainide responses (see Fig 11⇓). For transverse propagation in the normal zone, ANOVA failed to reach statistical significance. In this case, a Bonferroni correction was applied to the paired t test to localize effects between control and flecainide responses. Fig 11⇓ shows that the effect of the drug was more pronounced for propagation in the ischemic zone relative to the normal zone and was statistically significant at the shorter cycle lengths for propagation in all four groups. In the range of paced cycle lengths used in the present study, flecainide caused a rate-dependent reduction of conduction velocity of 14% (longitudinal) and 8% (transverse) for a change in basic drive from 600 to 250 ms in normal tissue and a reduction of 27% (longitudinal) and 25% (transverse) for a change in basic drive from 600 to 250 ms in ischemic tissue. However, there was no significant difference in the anisotropic conduction ratio before and after flecainide for all cycle lengths (see Table 2⇓).
The present study attempted to provide a comprehensive analysis of the proarrhythmic mechanism of flecainide in a documented model of reentrant ventricular arrhythmias in the subacute phase of myocardial infarction. Basic measurements of the effects of the drug on conduction, strength intervals, and effective refractory periods of both normal and ischemic myocardia were analyzed. These were combined with analysis of the drug effect on isochronal maps of epicardial activation from the same hearts from which basic measurements were obtained. The electrode array used for conduction studies on both normal and ischemic myocardia was specifically designed to study the effects of the drug on anisotropic conduction and used a large number of closely spaced (500 μm) recording sites. Measurements of conduction velocity were determined by a linear regression of activation time and distance along and across myocardial fibers. This technique was used to eliminate artifacts such as virtual cathode effect.17 Both basic measurements and epicardial activation patterns were analyzed for a plasma concentration of flecainide that is well within the therapeutic range.18
The study has shown that flecainide resulted in preferentially more marked depression of conduction in ischemic compared with normal myocardium. The reduction in conduction velocity was evident at the relatively long cycle length of 600 ms, and there was a significant rate-dependent augmentation of this effect at shorter cycle lengths. It should be noted that there was also a rate-dependent change in conduction velocity of normal and ischemic zones during control. This may be due in part to the membrane anesthetic effect of sodium pentobarbital, which has been reported by others.19 In contrast to the significant effect of flecainide on conduction, the effects on refractoriness at both normal and ischemic myocardia were negligible. Isochronal mapping of ventricular activation showed that flecainide facilitated both the induction and sustenance of reentrant tachycardia, especially at shorter basic cycle lengths.
Effects of Flecainide on Conduction and Refractoriness
The depressant effect of flecainide on conduction is consistent with its known effect as a sodium channel blocker. Takafumi and Hondeghem20 have shown that flecainide results in little tonic block of Vmax or INa, but block develops in a use-dependent fashion that is more pronounced at a shorter interpulse interval. The block was similar for long and short depolarizations, suggesting that the use-dependent block by flecainide is associated with the activated state of the sodium channel. The block was also more pronounced at less negative potentials. This may provide some explanation for the preferentially greater depression of conduction in ischemic versus normal myocardium observed in the present study. The use-dependent effects of flecainide are exemplified by amplification of flecainide-induced ventricular conduction slowing by exercise.3 Some studies suggested that flecainide-induced ventricular tachycardia has a propensity to occur during exercise.21
The lack of a differential effect of flecainide on longitudinal versus transverse conduction is somewhat in contrast to results with other type I antiarrhythmic drugs.22 23 24 A study by Turgeon et al12 has recently shown that an active metabolite of the type Ic drug encainide, O-demethyl encainide, caused disproportionate depression of longitudinal conduction in a concentration-dependent fashion. The smaller difference between the degree of reduction of conduction velocity in longitudinal versus transverse direction in the present study compared with Turgeon et al may be ascribed to the fact that our measurements were obtained with a moderate “therapeutic” dose of flecainide. It is possible that at a higher dose, the differential effect of the drug could be greater.
We were unable to evaluate the tonic depression of conduction block in the present study because the longest feasible interval for continuous pacing, without unmasking an underlying escape rhythm, was usually 600 ms. Because fewer arrhythmias occurred at a basic cycle length of 500 ms, the mechanism by which flecainide facilitated reentry at shorter cycle lengths was best described by the preferential rate-dependent decrease in conduction velocity in the ischemic zone. The drug-induced depression of conduction was similar for propagation along and across the ischemic fibers. While the primary direction of the reentrant wave front usually is oriented parallel to the longitudinal axis of the fibers, it is difficult to separate the possible combination of longitudinal and transverse conductions during complex propagation of impulses within the slow zone during the shortly coupled intervals of reentrant activation.
The negligible effect of flecainide on refractoriness in the present study is consistent with previous reports.25 26 27 However, flecainide was shown to markedly affect atrial refractoriness, particularly at high rates of stimulation,28 probably mediated through an effect on IK,29 in contrast to its minimal effect on ventricular refractoriness.27 On the other hand, in a study by Sakai et al,5 in an animal model similar to the one used in the present study, flecainide resulted in differential lengthening of refractoriness in ischemic versus normal myocardium and an increased degree of dispersion of local refractoriness in some experiments. This was consistent with development of new functional arcs of block and inducible sustained ventricular tachycardia after flecainide. In the present study, the negligible effects of flecainide on refractoriness in both normal and ischemic myocardia were consistent with our observation that the functional arcs of conduction block did not significantly change after flecainide. In this experimental model, the arcs of functional conduction block primarily are due to spatial dispersion of refractoriness at the border of the ischemic zone.11 The difference between our findings and those of Sakai et al5 are difficult to reconcile and may be related to differences in the drug dose or the technique of measuring the refractory periods, which in the present study was designed to ensure steady-state binding of the drug. Another consideration is the fact that in the present study, careful attention was directed to maintaining relatively constant core and epicardial temperatures before and after drug administration due to the significant effects of temperature on ventricular refractoriness.30
Electrophysiological Mechanism of the Proarrhythmic Effect of Flecainide
Flecainide, even at toxic doses, does not induce ventricular arrhythmias in normal dogs,31 but it does cause a dose-related proarrhythmic effect in dogs with prior myocardial infarction5 32 and when acute ischemia is induced at a site distant from a previous myocardial infarction.4 Flecainide also was shown to be proarrhythmic in an experimental model of functional reentry in which circus movement could be induced in a thin layer of “normal epicardium” obtained by endocardial cryotechnique in Langendorff-perfused rabbit hearts.6 However, this model of “anisotropic” reentry may have less relevance to the proarrhythmic effect of the drug in the presence of ischemic heart disease.
Early studies on arrhythmias caused by circus movement recognized that reentry was possible only if the length of the circus path exceeded the wavelength of the impulse.33 34 Lewis35 subsequently postulated that an antiarrhythmic drug would terminate circus movement only if it prolonged refractoriness more than it slowed conduction, so that the wavelength of refractoriness exceeded the path length. The wavelength of the cardiac impulse was mathematically formulated by Wiener and Rosenblueth36 as the distance the depolarization wave traveled during the duration of the refractory period (wavelength equals conduction velocity times refractory period). Since many antiarrhythmic agents affect both the refractory period and conduction velocity, it was suggested that measurement of the wavelength of the impulse may help to understand their mechanism of action.37 38 This approach has been criticized because of the lack of homogeneity in most preparations in which reentry is observed.39 In other words, conduction velocities and refractory periods are not uniformly distributed within the reentrant path. This is especially true in postinfarction models, as studied here, where the spatial distribution of electrophysiological properties has been documented.40
In the present model of reentry, the length of the pathway is related to the length of a “continuous” line of functional conduction block.11 The latter is determined by the spatial dispersion of refractoriness between normal and ischemic myocardium.11 An intervention that extends the area showing a critical degree of spatial dispersion of refractoriness would result in a longer arc of block. This can occur if a drug causes preferential lengthening of refractoriness in ischemic compared with normal myocardium. Flecainide caused negligible increases in refractoriness in both normal and ischemic myocardia and hence did not significantly change the length of a potential reentrant pathway. However, it resulted in a significant slowing of conduction, preferentially in ischemic myocardium, which was characteristically more pronounced at faster heart rates. The further slowing of conduction of the common reentrant wave front coupled with minimal changes in the length of reentrant pathway allowed enough time for the wave front to reexcite normal myocardium on the other side of the arc of functional block and to initiate reentry in hearts in which reentry could not be induced during control. The same mechanism also explains the propensity of reentrant activation to be sustained after flecainide. Of special interest in this regard is the observation by us41 and others42 43 that the direction of the slow common reentrant wave front during a sustained figure-8 reentrant circuit in this experimental model is usually oriented parallel to the myocardial fiber axis, that is, perpendicular to the left anterior descending coronary artery. While no directional differences in conduction velocity depression were found at the dosage used in the present study, isochronal mapping of functional reentrant circuits indicates that most of the slow conduction occurs in the common reentrant wave front for propagation oriented parallel to the fiber axis. Therefore, in circuits in which the potential reentrant pathway is preserved, drug-induced depression of longitudinal conduction would be of greater concern as a proarrhythmic factor.
In the range of cycle lengths used in the present study, the linear regression of conduction velocity was highly correlated (r>.98). However, one cannot be sure of the conduction path of short coupled beats in the ischemic zone. For this reason, we chose to measure strength-interval relations as an indication of conduction during premature stimulation. Excitability as a determinant of changes in conduction velocity has been used by others.44 During control, the strength-interval relation of normal muscle was asymptotic, as expected. The blunting of the strength interval of normal tissue by flecainide implied that there was an extended range of premature coupling intervals during which conduction could be depressed before block. However, the most dramatic changes in conduction of premature beats were evidenced in the ischemic zone during faster-paced basic rhythms, which can be attributed to the enhanced binding of the drug at faster rates. In the ischemic zone, there was a wider range of decreased excitability during control. The strength-interval relation showed that flecainide shifted the curve upward and to the right, allowing conduction to become further slowed at longer premature intervals.
The marked slowing of conduction in the common reentrant pathway at faster basic rhythms also can be attributed to the depression of conduction during S1. Recovery of excitability during S2 is a function of refractoriness plus the conduction delays during S1 activation. We have shown previously that the recovery time distribution influences the activation pattern of premature beats and the induction of reentry.45 Although there were negligible changes in the lines of block before and after flecainide, it is possible that the additional delays during propagation of S1 in the ischemic zone by flecainide shifted the recovery of excitability sufficiently to depress conduction in the common reentrant pathway during premature stimulation.
Limitations of the Study
Measurements of conduction velocities at very short cycle lengths, or during premature stimulation, were technically unfeasible in this infarction model using the cross electrode method for two reasons. First, we found that many sites exhibited variable or 2:1 conduction patterns at rates less than 250 ms, and premature extrastimuli could not be introduced at intervals similar to those delivered from control sites eliciting unstimulated responses.
The electrophysiological substrates of reentrant arrhythmias and their modulation by pharmacological agents involve a complex interplay of functional properties. The ultimate impact of experimental models of reentrant arrhythmias depends on how closely these resemble possible clinical situations.7 Care must be excercised when extrapolating the findings of animal studies to the human condition. For example, the metabolism of flecainide in dogs is quite different than in humans.13
The present study and others have demonstrated the propensity of flecainide to induce sustained reentrant tachycardia. However, those tachycardias, if well tolerated, do not directly explain the increased incidence of sudden death in the clinical setting in the presence of flecainide. It is possible that flecainide-induced sustained tachycardia in the presence of poor ventricular function, or even drug-induced negative inotropic effect,46 can result in a hemodynamically unstable situation. It should be pointed out that while flecainide did not result in any degradation in blood pressure, the hemodynamic state of these hearts after infarction was generally good. The possibility of additional deleterious effects of flecainide among the CAST patients with poor cardiac function may limit the extrapolation of some of the present findings to the clinical setting. On the other hand, the marked slowing of conduction by flecainide without lengthening of refractoriness allowed reentry to be maintained around shorter reentrant pathways.6 This fact alone implicates the probable proarrhythmic effects of even the purest type Ic agents. The subsequent induction of one or more smaller reentrant circuits may enhance the potential for the degeneration into ventricular fibrillation.7 The exact mechanism of flecainide-induced sudden death in humans remains to be established.
The proarrhythmic effect of flecainide acetate in the subacute infarction period in the dog is due to negligible changes in the dimension of the potential reentrant pathway, due to a lack of drug effect on refractoriness, combined with significant selective rate-dependent slowing of conduction in ischemic myocardium.
This work was supported by Veterans Administration Medical Research Funds (Merit Grants) to Dr Restivo and Dr El-Sherif. Dr Ndrepepa was supported by a Fulbright Scholar Fellowship. The authors appreciate the capable technical assistance of Antoinette Wells and her staff in the Animal Laboratory. The authors wish to thank Michael Yu and the Medical Media staff for the artwork. The authors also would like to thank Drs Alberico Sorgato and Jianhua Hu and Kathleen Stergiopolis for their assistance in some of the experiments. We thank David Lang of Lang Industries for his technical advice regarding the use of Flexacryl and Dr Dennis White for his suggestions for electrode construction. Most of all, we thank Richard Levin for his invaluable help in all phases of the project.
- Received July 21, 1994.
- Accepted October 5, 1994.
- Copyright © 1995 by American Heart Association
Levine JH, Morganroth J, Kadish AH. Mechanisms and risk factors for proarrhythmias 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.
Kou WH, Nelson SD, Lynch JJ, Montgomery DG, DiCarlo L, Lucchesi BR. Effect of flecainide acetate on prevention of electrical induction of ventricular tachycardia and occurrence of ischemic ventricular fibrillation during the early postmyocardial infarction period: evaluation in a conscious canine model of sudden death. J Am Coll Cardiol. 1987;9:359-365.
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.
El-Sherif N. Experimental models of reentry, antiarrhythmic, and proarrhythmic actions of drugs: complexities galore. Circulation. 1991;84:1871-1875.
Restivo M, Yin H, Caref EB, Patel A, Gjin N, Assadi M, Isber N, El-Sherif N. Mechanism of pro-arrhythmic action of flecainide in the post infarction canine heart: high resolution determination of conduction and refractory properties of normal and ischemic tissue. J Am Coll Cardiol. 1994;23:183A. Abstract.
El-Sherif N, Scherlag BJ, Lazzara R, Hope RR. Reentrant ventricular arrhythmias in the late myocardial infarction period, I: conduction characteristics in the infarction zone. Circulation. 1977;55:686-702.
Lazzara R, Scherlag BJ, Robinson MJ, Samet P. Selective in situ parasympathetic control of the canine sinoatrial and atrioventricular nodes. Circ Res. 1973;32:393-401.
Restivo M, Gough WB, El-Sherif N. Ventricular arrhythmias in subacute myocardial infarction period: high-resolution activation and refractory patterns of reentrant rhythms. Circ Res. 1990;66:1310-1327.
Turgeon J, Wisialowski TA, Wong W, Altemerei WA, Wikswo JP, Roden DM. Suppression of longitudinal versus transverse conduction by sodium channel block: effects of sodium bolus. Circulation. 1992;85:2221-2226.
Hondeghem LM, Katzung BG. Time- and voltage-dependent interactions of antiarrhythmic drugs with cardiac sodium channels. Biochim Biophys Acta. 1977;472:377-398.
Milliken GA, Johnson DE. Analysis of Messy Data. Vol I. Designed Experiments. Belmont, Calif: Lifetime Learning Publishers, Inc; 1984:31.
Nattel S, Wang Z, Matthews. Direct electrophysiological actions of pentobarbital at concentrations achieved during general anesthesia. Am J Physiol. 1990;259:H1743-H1751.
Takafumi A, 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.
Anastasiou-Nana MI, Anderson JL, Stewart JR, Crevey BJ, Yanowitz FG, Lutz JR, Johnson TA. Occurrence of exercise-induced and spontaneous wide complex tachycardia during therapy with flecainide for complex ventricular arrhythmias: a probable proarrhythmic effect. Am Heart J. 1987;113:1071-1077.
Kadish AH, Spear JF, Levine JH, Moore EN. The effects of procainamide on conduction in anisotropic canine ventricular myocardium. Circulation. 1986;74:616-625.
Bajaj AK, Kopelman HA, Wikswo JP Jr, Woosley RL, Roden DM. Frequency- and orientation-dependent effects of mexiletine and quinidine on conduction in the intact dog heart. Circulation. 1987;75:1065-1073.
Anderson KP, Walker R, Lux RL, Ershler PR, Menlove R, Williams MR, Krall R, Moddrelle D. Conduction velocity depression and drug-induced ventricular tachyarrhythmias: effects of lidocaine in the intact canine heart. Circulation. 1990;81:1024-1038.
Wang Z, Pelletier C, 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 tissues. Circulation. 1990;82:274-283.
Follmer CH, Colatsky TJ. Block of delayed rectifier potassium current Ik, by flecainide, and E-4031 in cat ventricular myocytes. Circulation. 1990;82:289-293.
Riker Investigator’s Brochure: R-818 (Flecainide Acetate, Tambocor) Oral and Parenteral Antiarrhythmic. St Paul, Minnesota: Riker Laboratories; 1987.
Zimmerman JM, Bigger JT, Coromilas J. Flecainide acetate is arrhythmogenic in a canine model of sustained ventricular tachycardia. J Am Coll Cardiol. 1985;5:391. Abstract.
Mines GR. On dynamic equilibrium in the heart. J Physiol (Lond). 1913;46:350-383.
Mines GR. On circulating excitations in heart muscles and their possible relation to tachycardia and fibrillation. Trans R Soc Can. 1914;8:43-52.
Lewis T. The Mechanism and Graphic Registration of the Heart Beat. 3rd ed. London: Shaw and Sons; 1925.
Smeets JLRM, Allessie MA, Lammers WJEP, Bonke FIM, Hollen J. The wavelength of cardiac impulse and reentrant arrhythmias in isolated rabbit atrium: the role of heart rate, autonomic transmitters, temperature, and potassium. Circ Res. 1986;58:96-108.
Rensma PL, Allessie MA, Lammers WJEP, Bonke FIM, Schalij MA. Length of excitation wave and susceptibility to reentrant atrial arrhythmias in normal conscious dogs. Circ Res. 1986;62:395-410.
Spinelli W, Hoffman BF. Mechanisms of termination of reentrant atrial arrhythmias by class I and class III antiarrhythmic agents. Circ Res. 1989;63:1563-1579.
Gough WB, Mehra R, Restivo M, Zeiler RH, El-Sherif N. Reentrant ventricular arrhythmias in the late myocardial infarction period, XIII: correlation of activation and refractory maps. Circ Res. 1985;57:432-442.
Dillon SM, Allessie MA, Ursell PC, Wit AL. Influences of anisotropic tissue structure in reentrant circuits in the epicardial border zone of subacute canine infarcts. Circ Res. 1988;63:182-206.
Cardinal R, Vermeulen M, Shenasa M, Roberge F, Page P, Helie F, Savard P. Anisotropic conduction and functional dissociation of ischemic tissue during reentrant ventricular tachycardia in canine myocardial infarction. Circulation. 1988;77:1162-1176.
Peon J, Ferrier GR, Moe GK. The relationship of excitability to conduction in canine Purkinje tissue. Circ Res. 1978;43:125-135.
Restivo M, Gough WB, El-Sherif N. Reentrant ventricular arrhythmias in the late myocardial infarction period: mechanism of dual stimulation in the prevention of reentry. Circulation. 1988;77:429-444.
Packer M. Hemodynamic consequences of antiarrhythmic therapy in patients with chronic heart failure. J Cardiovasc Electrophysiol. 1991;2:S240-S247.