Donate Help Contact The AHA Sign In Home
American Heart Association
Circulation
Search: search_blue_button Advanced Search
« Previous Article | Table of Contents | Next Article »
Circulation. 1995;92:1300-1311

This Article
Right arrow Abstract Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ranger, S.
Right arrow Articles by Nattel, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ranger, S.
Right arrow Articles by Nattel, S.

(Circulation. 1995;92:1300-1311.)
© 1995 American Heart Association, Inc.

Articles

Determinants and Mechanisms of Flecainide-Induced Promotion of Ventricular Tachycardia in Anesthetized Dogs

Suzanne Ranger, BSC; Stanley Nattel, MD

From the Department of Medicine (S.N.), Montreal Heart Institute and the University of Montreal, and Department of Pharmacology and Therapeutics (S.R., S.N.), McGill University, Montreal, Canada.


*    Abstract
up arrowTop
 *Abstract
down arrowIntroduction
 down arrowMethods
down arrowResults
down arrowDiscussion
down arrowReferences
 
Background Class IC antiarrhythmic agents such as flecainide are known to have potentially significant ventricular proarrhythmic actions, but the underlying mechanisms are incompletely understood. While some studies have reported proarrhythmia in both healthy dogs and dogs that previously have had a myocardial infarction (MI), there are no published, controlled studies comparing proarrhythmia in healthy dogs vs in dogs with MI. In addition, the concentration dependence of proarrhythmia is unknown and the electrophysiological changes associated with proarrhythmia are not well established.

Methods We administered successive loading and maintenance infusions of flecainide until ventricular tachyarrhythmia or death occurred in 13 healthy dogs and 19 dogs with 72-hour-old MIs (MI dogs). Ventricular proarrhythmia, defined as reproducible ventricular tachycardia absent under control conditions and occurring in the presence of flecainide, was observed in 4 of 13 healthy dogs (31%) and 15 of 19 MI dogs (79%, P=.02), and drug-induced spontaneous ventricular tachycardia occurred in 8 of 19 MI dogs but in no healthy dogs (P=.007). Activation data at the time of proarrhythmia were available for 11 MI dogs and provided evidence for reentry in 9, with a complete epicardial reentry circuit identified in 4 dogs and a partial circuit in 5. While flecainide slowed ventricular conduction in both the longitudinal and transverse directions, there were no significant differences between overall drug-induced conduction changes in MI dogs compared with healthy dogs. However, in 7 MI dogs for whom activation data were available during ventricular pacing at concentrations comparable to those causing proarrhythmia, flecainide induced a new arc of block in 6 of 7, whereas an arc of block was never observed in the absence of proarrhythmia. Conduction block was induced transverse to fiber orientation in a rate-dependent fashion and was caused by a regionally-specific effect of the drug. No differences were noted between refractory periods proximal and distal to the site of block.

Conclusions Prior MI strongly predisposes dogs to flecainide proarrhythmia, which occurs in the majority of such dogs in a concentration-related way. In most cases, activation data suggest that anisotropic reentry around a localized arc of rate-dependent transverse conduction block underlies proarrhythmia. These results provide insights into the conditions and mechanisms underlying the ability of flecainide to promote the occurrence of ventricular tachycardia.


Key Words: myocardial infarction • antiarrhythmia agents • sodium • arrhythmia • death, sudden


*    Introduction
up arrowTop
 up arrowAbstract
 *Introduction
down arrowMethods
down arrowResults
down arrowDiscussion
down arrowReferences
 
Ventricular proarrhythmia, consisting of the de novo induction or aggravation of preexisting ventricular tachyarrhythmias (VTs) by an antiarrhythmic agent, is probably the most important factor limiting the application of antiarrhythmic drug therapy.1 2 Proarrhythmic responses are believed to underlie the propensity of class I drugs, particularly the IC agents flecainide and encainide, to increase the mortality rate in patients with a recent myocardial infarction (MI).3 4 Indirect evidence suggests that similar phenomena may be operative in patients resuscitated from sudden cardiac death5 and those with atrial fibrillation.6 7 Improved understanding of the mechanisms of drug-induced proarrhythmia is needed to develop strategies to optimize the risk to benefit ratio of antiarrhythmic drug therapy.

Several studies have suggested that class IC agents can result in the induction of VT by programmed electrical stimulation in dogs with a previous MI even when VT was not inducible before drug administration.8 9 10 11 12 13 14 Results have also been presented that suggest that flecainide causes proarrhythmia (including a 50% prevalence of sustained VT) in a large percentage of normal dogs.15 These findings raise questions about the need for a pathological substrate to promote proarrhythmia. Limitations of previous studies have included a lack of consideration of the concentration dependence of drug action, the variable definition of proarrhythmia that often did not require reproducibility of VT induction, study protocols that did not include observations in the presence of stable drug concentrations, and the lack of a comparison between healthy dogs and those with prior infarction. In addition, activation mapping studies of mechanisms have been presented in a very limited form (48-channel recordings from two dogs) in only one published study,12 and preliminary data from more detailed studies have been presented but published only in abstract form.10 A recent study showed that flecainide permitted the induction of sustained VT in rabbit hearts through a thin layer of surviving epicardial tissue after extensive ventricular cryoablation.16 Proarrhythmia in this model was associated with a small decrease in the wavelength for reentry and the induction of arcs of conduction block of variable location.

The present experiments were designed to address several questions regarding the ability of flecainide to induce proarrhythmic responses in anesthetized, open-chest dogs: (1) To what extent does a previous myocardial infarction predispose to such proarrhythmic responses, and at what drug concentrations do they occur? (2) What electrophysiological changes permit the development of drug-induced proarrhythmia? (3) How frequently can epicardial mapping reveal a functional mechanism of proarrhythmia, and what mechanisms are involved? Preliminary findings have been presented in abstract form.17 18


*    Methods
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 *Methods
down arrowResults
down arrowDiscussion
down arrowReferences
 
Infarct Model
Mongrel dogs (17 to 27 kg) were anesthetized with sodium pentobarbital 30 mg/kg IV and ventilated through an endotracheal tube. A left thoracotomy was performed under sterile conditions, and a small opening was created in the pericardium. The left anterior descending coronary artery (LAD) was dissected proximal to the first diagonal branch and then occluded in two stages.19 Nadolol 0.5 mg/kg IV was given before occlusion and then daily (20 mg PO) for 2 days. Benzathine penicillin G 150 000 U and procaine penicillin G 150 000 U SC were also given daily. Levorphanol 0.056 mg/kg SC BID was used to control postoperative pain.

Study Procedures
The proarrhythmic effects of flecainide were studied in 13 healthy dogs and 31 dogs subjected to LAD occlusion 72 hours before study (MI dogs). On the day of study, dogs were anesthetized with morphine 2 mg/kg SC and {alpha}-chloralose 120 mg/kg IV, intubated, and mechanically ventilated. Catheters were inserted into one femoral artery and two femoral veins. Arterial blood gases were measured at 1-hour intervals and maintained in the physiological range (SAO2 >90%, pH 7.35 to 7.45) by adjusting the ventilator or using supplementary oxygen. A left thoracotomy was performed at the fifth intercostal space, and a pericardial cradle was created. The chest cavity was covered with plastic wrap to prevent cooling or dehydration, and body temperature was maintained at 37°C to 38°C with a heating blanket.

An array of 56 or 112 bipolar electrodes in a 4x6-cm plastic sheet was sewn to the surface of the left ventricle over the LAD territory. A bipolar platinum electrode was fixed to the left atrial appendage to record the left atrial electrogram. ECG leads II, III, aVL, and aVR were filtered at 0.05 to 40 Hz, and electrograms were filtered at 30 to 400 Hz (amplifiers, Bloom Associates Ltd) recorded, along with stimulus artifacts and arterial pressure, with a paper recorder (MT-95000 Host Control, Astromed Inc).

Activation Mapping
Previously-described techniques were used to create activation maps.20 21 The interpolar distance was 2 mm, and electrodes were arranged in a parallel fashion with an interelectrode distance of 4 mm for the 112-electrode array and 6 mm for the 56-electrode array. The 56-electrode array was used in the first 18 dogs, and the 112-electrode array was used in the remaining 26 dogs. A bipolar electrode at the center of the array was used for programmed ventricular stimulation with the use of 2-ms current pulses of twice diastolic threshold (model EP-2 Stimulator, Digital Cardiovascular Instruments). Electrograms were filtered at 40 to 300 Hz, digitized with 12-bit resolution and a 1-kHz sampling rate, and transmitted by means of duplex fiber-optic cables into a microcomputer. The activation time at each electrode site was defined as the time of maximal rate of voltage change as calculated by the computer. Each electrogram was reviewed manually to exclude low-amplitude recordings resulting from poor contact, electrical artifacts, and interference by electrical noise. The presence of block was defined by a conduction velocity <0.1 m/s between adjacent electrode sites, generally with an abrupt alteration in the direction of wave front propagation across the line of block. Isochrones were constructed with a computer-based interpolation algorithm.

Experimental Procedures
Successively increasing doses of flecainide (Table 1Down) were administered until ventricular proarrhythmia or death occurred in 19 MI dogs and 13 healthy dogs ("controlled series"). Under control conditions and at each drug concentration, ventricular activation, ventricular refractory period, and arrhythmia occurrence were evaluated over a range of cycle lengths from 180 to 500 ms. Two minutes of stimulation were allowed at each basic cycle length before the introduction of extrastimuli to determine the refractory period. Single extrastimuli were then applied after every 15 basic stimuli at the central electrode site to determine effective refractory period (ERP, the longest S1S2 interval failing to capture the ventricle) and to induce VT.


View this table:
[in this window]
[in a new window]
  		Table 1. Flecainide Doses and Plasma Concentrations

Proarrhythmia was defined as the reproducible occurrence of spontaneous or inducible VT (>=5 successive ventricular complexes) in the presence of flecainide but absent under control conditions. Sustained VT was defined by the occurrence of >30 successive ventricular complexes. Plasma flecainide concentrations were measured by high-performance liquid chromatography.22 Infarct size was determined with triphenyl tetrazolium chloride.23

An additional 12 MI dogs ("additional series") were studied to relate the occurrence of proarrhythmia to infarct histopathology and to differences in ERP on either side of lines of block. The same drug infusion protocol described above was used, but instead of acquiring detailed mapping data at each dose, we identified the dose causing proarrhythmia and then determined the ERP at multiple sites on either side of the line of block. Portions of myocardium were cut into 0.3x1x2.4-cm sections, which were stained with hematoxylin-phloxine-saffron or Gomori's one-step trichrome24 and subjected to microscopic examination.

Data Analysis
Conduction velocity was analyzed by linear regression of interelectrode distance on activation time. Group values are presented as mean±SD. Statistical comparisons were performed by ANOVA with a range test (Student's t test with the Bonferroni correction)25 or for contingency data by Fisher's exact test or {chi}2 test. All comparisons were two-tailed, and a value of P<.05 was taken to indicate statistical significance.


*    Results
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMethods
*Results
down arrowDiscussion
down arrowReferences
 
Occurrence and Type of Proarrhythmia in Controlled Series
Flecainide caused proarrhythmic responses in 15 of 19 controlled series MI dogs, and 4 of 13 healthy dogs (P=.02). MI dogs also presented more severe forms of proarrhythmia, such as sustained and spontaneous VT (Fig 1Down). Spontaneous VT did not occur in healthy dogs, but was observed in 8 of 19 MI dogs (P=.007). Drug concentrations at the time of VT averaged 5.3±1.9 mg/L in healthy dogs and 5.6±2.3 mg/L in MI dogs (P=NS). In dogs without proarrhythmia, death occurred because of progressive hypotension at plasma concentrations averaging 9.8±4.8 mg/L in healthy dogs (n=9) and 10.4±1.9 mg/L in MI dogs (n=4, P=NS versus healthy dogs). Flecainide significantly reduced blood pressure in MI dogs only at the highest doses, with blood pressure remaining unchanged before proarrhythmia in most cases.



View larger version (16K):
[in this window]
[in a new window]
  		Figure 1. Bar graph showing the incidence of various forms of proarrhythmia (PROA) in healthy (normal) dogs and dogs that had prior myocardial infarction (MI dogs). The percentage of dogs with any form of proarrhythmia (total PROA) was significantly greater (*P=.02) for MI dogs. NSVT indicates nonsustained ventricular tachycardia (VT); SVT, sustained VT; and spVT, spontaneous VT. Some dogs had more than one form of proarrhythmia.

General Determinants of Proarrhythmia
In the four MI dogs without proarrhythmia, infarct size averaged 3±2% of the left ventricle, compared with 32±5% in dogs with proarrhythmia (P<.05). The cumulative concentration-response curve for proarrhythmic responses in infarct dogs is shown in Fig 2Down. The EC50 for proarrhythmia was {approx}6 mg/L, and the maximum proarrhythmic incidence of 79% occurred at a plasma concentration of 9 mg/L. Mean plasma concentrations at the time of proarrhythmia were higher for spontaneous VT (6.8±1.4 mg/L, n=7) than for sustained (4.0±2.5 mg/L, n=7) and nonsustained (4.4±2.2 mg/L, n=5) VT. When multiple forms of VT occurred within a given dog, more severe forms occurred later in the protocol at higher drug concentrations. Because of the smaller number of normal dogs experiencing proarrhythmia, a concentration-response curve for proarrhythmia could not be constructed, but the mean drug concentration at the time of proarrhythmia was similar to that in MI dogs.



View larger version (13K):
[in this window]
[in a new window]
  		Figure 2. Line graph showing cumulative concentration-response curve for flecainide proarrhythmia (PROA) in dogs that had prior myocardial infarction. Flec. Conc. indicates flecainide concentration; dots, data points; solid line, nonlinear best-fit curve to the equation N=Nmax(1/[1+{FC/EC50}K]), where N is the cumulative number of dogs with proarrhythmia (PROA); Nmax, maximum predicted number of dogs with PROA; FC, flecainide concentration; EC50, concentration for 50% of maximum incidence of PROA; and K, a constant.

Drug-Induced Changes in Conduction and Refractoriness
Drug-induced conduction slowing that exceeds changes in refractoriness is believed to play an important role in the arrhythmogenic properties of class IC drugs.1 16 26 27 28 Fig 3Down, left, shows an analysis of the effects of flecainide on conduction velocity. Drug-induced conduction slowing was not significantly different for longitudinal versus transverse propagation and was not significantly different in infarcted versus healthy hearts. Flecainide did not significantly alter ventricular ERP in either infarcted or healthy hearts (Fig 3Down, right). The very similar changes produced by the drug in healthy versus infarcted hearts do not clarify the mechanism of the increased susceptibility to proarrhythmia of the latter group. Conduction slowing alone appears to be insufficient to cause proarrhythmic responses in a large percentage of hearts.



View larger version (20K):
[in this window]
[in a new window]
  		Figure 3. Plots: Left, Changes in conduction velocity (CV) produced by flecainide in healthy dogs (N, circles) and dogs that had prior myocardial infarction (MI, triangles), when conduction was assessed in the longitudinal (top) or transverse (bottom) direction. For a similar mean concentration ([F]), drug effects were not significantly dependent on direction of propagation or the presence or absence of MI. BCL indicates basic cycle length. Right, Ventricular effective refractory period (ERP) as a function of BCL in the absence and presence of flecainide. No significant changes were seen.

Activation During Ventricular Tachycardia
Fig 4Down shows activation data at the onset of VT in one dog (dog No. 34): recordings of selected electrograms and a surface ECG lead (top right) and maps of the first three activations of VT (cycles 1 through 3, corresponding to the cycles delimited by the vertical dashed lines on the analogue panel). All activation maps are oriented in the same fashion, with the LAD running parallel to the long axis of the array along its left border. Electrode locations indicated by letters on each map correspond to the electrograms shown in the analogue recordings. After the last basic stimulus (S1) at a cycle length of 250 ms, a single extrastimulus is applied (S1S2, 210 ms), initiating a run of sustained VT (200 per minute). Selected activation times relative to S2 are shown on the activation maps, along with 20-ms isochrons. The activation initiated by S2 (cycle 1) is shown at the upper left. The tissue underlying the top two thirds of the array (toward the base of the heart) was not captured by the extrastimulus, and a line of horizontal unidirectional block (solid line) resulted. Propagation occurred slowly in the inferior direction, and the counterclockwise limb propagated around the lateral border of the arc of block near site E. The impulse then continued to propagate counterclockwise around a transverse arc of functional block to produce cycles 2 and 3, and all subsequent beats of VT displayed a similar activation pattern.



View larger version (43K):
[in this window]
[in a new window]
  		Figure 4. Top left, bottom left, and bottom right, Maps of cycles 1, 2, and 3, respectively, showing a complete reentry circuit responsible for proarrhythmic ventricular tachycardia (VT) in dog No. 34. The activation patterns during these cycles are delimited by the vertical dashed lines in the analogue recordings (top right). The electrode array is shown schematically for each activation map, with dots representing the sites of bipolar recording electrodes. The long axis of the array is oriented parallel to the left anterior descending coronary artery, which runs along the left side of the array; the numbers on each map are activation times at selected sites; and 20-ms isochrons are shown. All activation times are relative to the time of the extrastimulus (S2) that initiated VT. The heavy lines represent arcs of block, and the dashed arrows indicate the sequence of activation. Top right, Analogue recordings from nine electrode sites (A through I), with locations indicated on each map, from the surface ECG and from a stimulus artifact channel (S). (For detailed discussion, see text).

Fig 5Down shows the initiation of ventricular tachycardia in another dog (dog No. 26). An extrastimulus (S1S2, 145 ms; basic cycle length, 200 ms) initiates activity at the center of the array (activation times in Fig 5ADown through 5C are referenced to S2), which propagates rapidly in the longitudinal direction (perpendicular to the LAD and horizontally in Fig 5ADown) and more slowly in the transverse direction. A broad arc of conduction block is encountered (solid line), forcing the impulse to travel around the lateral margin of the arc of block. The impulse travels toward the LAD, past a shorter arc of transverse conduction block, and then appears to propagate around the inferior border of the array in a counterclockwise direction (Fig 5BDown). The excitation of tissue at the inferolateral portion of the array then initiates a full counterclockwise reentry circuit around the arc of transverse block, as shown in Fig 5CDown, producing the first unstimulated beat of VT. A stable reentry pattern results, with the activation shown in Fig 5DDown recorded several seconds later during VT (vertical dashed lines in Fig 5FDown delimit cycle mapped in Fig 5DDown). The ECG and selected electrogram tracings corresponding to Fig 5ADown through 5C are shown in Fig 5EDown, and those corresponding to Fig 5DDown are shown in Fig 5FDown.



View larger version (49K):
[in this window]
[in a new window]
  		Figure 5. A through D, Activation maps during sustained ventricular tachycardia (VT) initiated by an extrastimulus at a coupling interval of 145 ms in dog 26. Format for activation maps as in Fig 4Up. A and B, Activation during the beat initiated by the extrastimulus; C, activation during the first reentrant beat of VT; and D, activation during a cycle of VT several seconds after VT onset. E, Analogue recordings of stimulus artifact (S), surface ECG, and electrograms from five locations in A through C. The beginning and end of each cycle is delimited by the vertical dashed lines, and activation times are given relative to the peak of the stimulus artifact. F, Analogue recordings of surface ECG and electrograms at selected sites (leads I through V) during VT. The cycle corresponding to D is delimited by the vertical dashed lines in F, and D shows the locations corresponding to electrogram leads I through V. (For detailed discussion, see text.)

Fig 6Down shows activation during a spontaneously occurring episode of VT in dog No. 42. An ECG and stimulus channel (showing lack of stimulation) are provided on the bottom, and activation maps of beats A through D on the ECG recording are shown in the corresponding sections of the figure (Fig 6ADown through 6D). Epicardial breakthrough of the last sinus beat (Fig 6ADown) occurs at the site marked by an asterisk, and a large region of the LAD territory is activated within 30 ms. Activation is delayed towards the apex, as expected in the presence of an almost transmural infarction that interferes with normal endocardial to epicardial impulse propagation.29 A spontaneous ventricular ectopic beat, with initial activation indicated by the asterisk in Fig 6BDown, initiates 14 beats of VT at a cycle length of 280 ms. The propagating impulse encounters an arc of conduction block (solid line in Fig 6BDown) and appears to turn around the arc of block at the medial border of the electrode array. Activation proceeds in a counterclockwise direction beyond the arc of block. Initial activation of the next cycle (C) occurs at the lateral border of the array, 106 ms after the last recorded activation during the preceding cycle. The impulse conducts medially, encountering an arc of block similar to that shown in B, and returns in a counterclockwise direction beyond the arc of block. A similar activation pattern was recorded for the next beat (D) and for subsequent cycles of VT. Unlike the data shown in Figs 4Up and 5Up, which show complete circuses of impulse propagation during VT, the activation data in Fig 6Down account for only part of a potential reentry cycle. In addition, VT occurs spontaneously in Fig 6Down, whereas it results from premature stimulation in the other figures. In both cases of spontaneous VT recorded with the mapping system, the ventricular beat initiating tachycardia had a pattern of activation different from that recorded during VT, consistent with the possibility that the mechanism of the premature beat initiating reentry is different from the mechanism of VT. Enhanced automaticity is known to cause ventricular ectopy in this model30 31 32 and could explain the ventricular premature beat that initiated reentry. Alternatively, the beat initiating reentry could be a result of reentry with a different epicardial breakthrough point compared with sustained VT.



View larger version (35K):
[in this window]
[in a new window]
  		Figure 6. Activation during ventricular tachycardia (VT) of spontaneous onset. A through D, Activation maps corresponding to complexes A through D in the ECG tracing (bottom). Format as in Fig 4Up. *Site (S) of earliest activation (time 0) for A through D. A, Activation during last sinus beat B, Activation of ventricular ectopic beat that initiated VT. C and D, Activation during second and third beat, respectively, of VT. Note that site of earliest activation has moved to lateral margin of array. Activation times for B through D are referenced to earliest activation during cycle B. (For detailed discussion, see text.)

Of 15 dogs with myocardial infarction and proarrhythmic VT in the controlled series, activation data during VT were available in 11. In 4 dogs, a complete reentry circuit was resolved (Table 2Down), as in Figs 4Up and 5Up. In 5 other dogs, the activation pattern was consistent with reentry, but recorded activation times did not account for all the reentry cycle. Among all dogs with a complete or partial reentry circuit visualized, an arc of transverse conduction block was noted between the central stimulation point and the apex of the left ventricle.


View this table:
[in this window]
[in a new window]
  		Table 2. Occurrence of Arrhythmia and Block AFter Flecainide Infusion in Controlled Series of Dogs

Activation During Ventricular Paced Rhythm and Relation to Proarrhythmia
Transverse conduction block appeared to play a central role in proarrhythmic responses. We therefore evaluated the hypothesis that while overall conduction changes caused by the drug were similar in infarcted and normal hearts, infarction may predispose localized regions to drug-induced conduction slowing. Figs 7 through 9DownDownDown show activation in the LAD territory before and after flecainide in a representative healthy dog, an MI dog without proarrhythmia, and an MI dog with proarrhythmia, respectively. In the healthy dog (Fig 7Down), flecainide slowed conduction by 16% at a cycle length of 400 ms, without altering the pattern of activation (Fig 7CDown). When the pacing cycle length was decreased to 200 ms (Fig 7DDown), the drug slowed conduction by 34% without changing the activation pattern. In MI dogs without proarrhythmia (Fig 8Down), similar drug concentrations produced conduction changes very similar to those in the healthy dogs. Fig 9Down shows corresponding results in a dog with proarrhythmia. Under control conditions, activation in the LAD territory was not perceptibly different from that in the healthy dog in the absence of drug at either cycle length. Flecainide, at a concentration similar to that in the healthy dog (Fig 7Down), slowed conduction by 18% at a cycle length of 400 ms (Fig 7CDown), without changing activation pattern. When the pacing cycle length was shortened to 200 ms (Fig 7DDown), conduction in the longitudinal and superior (basal) direction was uniformly slowed, as in the healthy dog in Fig 7DDown. In the apical direction, however, an arc of conduction block appeared (solid line).



View larger version (35K):
[in this window]
[in a new window]
  		Figure 7. Activation maps from a healthy (normal in the figure) dog with no proarrhythmia (PROA). A and B, Control results as obtained at basic cycle lengths (BCLs) of 400 and 200 ms, respectively. C and D, Results in the presence of 2.8 mg/L flecainide at the same cycle lengths. Format for maps as in Fig 4Up.



View larger version (31K):
[in this window]
[in a new window]
  		Figure 8. Activation maps from a dog that had a myocardial infarction (MI) but no proarrhythmia (PROA). A and B, Control results at basic cycle lengths (BCLs) of 300 and 150 ms, respectively. C and D, Results in the presence of flecainide (2.7 mg/L) at the same BCL values. Format for maps as in Fig 4Up.



View larger version (36K):
[in this window]
[in a new window]
  		Figure 9. Activation data from a dog that had a myocardial infarction (MI) with proarrhythmia (PROA) (dog No. 22). A and B, Activation maps at basic cycle lengths (BCLs) of 400 and 200 ms, respectively, under control conditions. C and D, Activation maps at BCLs of 400 and 200 ms, respectively, in the presence of 2.6 mg/L flecainide. Format for maps as in Fig 4Up.

In 7 MI dogs with proarrhythmia, activation data during ventricular pacing were obtained as part of the controlled series study protocol at a plasma concentration >=75% (Table 2Up) of the concentration at which proarrhythmia occurred. In 6 of these dogs, an arc of conduction block was observed during rapid pacing. In all 6, conduction block was absent at slower rates and the location of block during rapid ventricular pacing remained the same as during VT. Arcs of block were never observed in dogs without VT. In the additional 12 dogs studied to evaluate regional refractoriness, all 6 with proarrhythmia had arcs of block during rapid ventricular pacing and no block was seen in the 6 without proarrhythmia.

Studies of Conduction and Refractoriness at Sites of Block
The above data indicate a close association between the development of regional block and the occurrence of proarrhythmia, consistent with a potentially important role for regional block in the mechanism of VT induced by flecainide. Fig 10Down presents an analysis of results from the controlled series of dogs that was designed to establish whether the susceptibility of dogs to proarrhythmia was due to generally enhanced drug effects on conduction or to regional factors at the site of block. For equivalent drug concentrations, overall drug-induced conduction slowing during both longitudinal and transverse propagation was similar for dogs with (Fig 10Down, bottom) and without (Fig 10Down, top) block. During rapid pacing (bottom right), conduction slowing was much greater at the site of transverse conduction block (filled bar) than elsewhere over the infarct (bars with horizontal lines). In contrast also in Fig 10Down, during slower pacing (bottom left) transverse conduction slowing was the same at the site of block (filled bar) as in other regions (horizontal lines). Thus, block was due to an interaction between rate and the underlying substrate and not to a generalized susceptibility to drug action.



View larger version (20K):
[in this window]
[in a new window]
  		Figure 10. Bar graphs showing percentage decrease in conduction velocity (mean±SD) caused by flecainide in a controlled series of dogs that had a myocardial infarction (MI dogs) without block (top, 3.3±1.5 mg/L) and MI dogs with an arc of conduction block (bottom, 4.6±2.5 mg/L), during fast and slow pacing. Conduction velocity (CV) in the longitudinal and transverse directions were calculated by linear regression of distance on activation times, excluding zones of block. Drug-induced conduction velocity changes at the site of block were calculated from the interelectrode distance and difference in activation times across the site of block. CL indicates cycle length. *P<.05 for difference in drug-induced change in conduction velocity across the arc of transverse conduction block vs overall longitudinal or transverse conduction, **P<.01 for difference in conduction velocity during fast vs slow pacing.

We then determined whether the interaction is due to a progressive reduction in conduction velocity at the site of block with increased rate in the presence of flecainide. Fig 11Down shows the cycle length dependence of transverse conduction velocity at the site of block under control conditions and in the presence of flecainide in the 7 MI dogs with block. While conduction tended to slow with decreased cycle length in the presence of the drug, in all but 1 dog (dog No. 26) there was an abrupt decrease in conduction velocity (averaging 58±17%) associated with block at the shortest cycle length.



View larger version (33K):
[in this window]
[in a new window]
  		Figure 11. Bar graphs showing apparent conduction velocity (CV) between electrodes at the site of rate-dependent block, calculated from the interelectrode distance divided by difference in activation times. Conduction block was defined by a velocity <0.1 m/s. Results are shown under control conditions (hatched bars) and in the presence of flecainide (solid bars) during the dose that caused proarrhythmia. BCL indicates basic cycle length.

The final possibility that we evaluated was that block results from spatially determined drug effects on refractoriness, with ERP prolonged to a greater extent distal (versus proximal) to the site of block. In an additional series of 12 dogs studied to evaluate this possibility, block occurred in 6. For each of these dogs, ERP was determined by local stimulation on either side of the arc of block at a cycle length similar to the VT cycle length in that dog. As shown in Fig 12Down, no systematic differences in ERP were observed between sites proximal and distal to the site of block. Overall, ERP averaged 246±16 ms at 13 sites proximal to the arc of block and 246±21 ms at 13 sites distal to the arc of block.



View larger version (37K):
[in this window]
[in a new window]
  		Figure 12. Activation map of a representative dog (during block) showing refractory periods (ERP) measured selectively by stimulating at sites proximal and distal to an arc of transverse conduction block. ERP values were measured at six adjacent sites across the arc of block in six dogs. TableUp, Mean results from 13 pairs of sites in the six dogs studied.

Histological Analysis
Examination of histological sections confirmed that epicardial muscle fibers were oriented in the direction of rapid impulse propagation, ie, perpendicular to the LAD. Infarctions became progressively denser and more transmural toward the apex, and at the zones of block there were areas in which the infarction extended to the epicardial surface (ie, zones in which there were no surviving epicardial cells). However, similar zones were present in dogs without an arc of block, so that no qualitative histological differences could be identified with the occurrence of an arc of block.


*    Discussion
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMethods
up arrowResults
*Discussion
down arrowReferences
 
We have shown that the presence of a prior MI predisposes dogs to the ventricular proarrhythmic actions of flecainide. The occurrence of proarrhythmia in our dogs was related to plasma drug concentration, and the mechanism of proarrhythmia appeared to be reentry involving an arc of rate-related transverse conduction block in the presence of the drug.

Comparison With Previous Studies of Flecainide-Induced Proarrhythmia in Dogs
Several studies have shown that flecainide can facilitate the occurrence of VT in dogs with prior MI.8 9 10 11 12 13 In addition, one group reported a high incidence of flecainide proarrhythmia at high concentrations in dogs with healthy hearts.15 We found that under controlled conditions, dogs with healthy hearts are relatively resistant to flecainide proarrhythmia, whereas MI dogs that do not experience proarrhythmia have relatively small infarctions. While clinical experience suggests that diseased hearts are predisposed to flecainide proarrhythmia,27 the present study comprises, to our knowledge, the first direct experimental demonstration of this phenomenon.

Little information is available about changes in activation that underlie flecainide proarrhythmia. Only one published study12 presented activation data that were obtained in two dogs with a 48-electrode array. The limited data presented in that paper are consistent with the arrhythmia mechanism that we observed, reentry around a drug-induced arc of conduction block. Another study, published only in abstract form,10 showed that flecainide induced VT in four dogs by extending the line of block transverse to fiber orientation while slowing conduction equally in the longitudinal and transverse directions.

Mechanism of Proarrhythmia
In 9 of 11 MI dogs in the controlled series for which activation maps during proarrhythmic VT were available, activation compatible with a reentry circuit could be mapped around an arc of conduction block, which always occurred in the direction transverse to fiber orientation. The anisotropic properties of myocardium thus played an important role in proarrhythmia. According to the original descriptions of VT inducibility by programmed electrical stimulation in dogs with prior MI,33 34 35 36 epicardial mapping studies showed that VT in this model is due to reentry around an arc of conduction block within the infarct,37 38 with block occurring in the direction of impulse propagation transverse to myocardial fiber bundles.39 40 The intrinsically anisotropic properties of myocardial conduction41 appear to play an important role in producing this type of VT,39 40 which has been called "anisotropic reentry."40 Similar mechanisms underlay proarrhythmic VT in our dogs. While in normal tissues, the safety factor for impulse propagation is greater in the transverse direction,41 42 transverse conduction block is predicted to occur more readily than longitudinal when active membrane properties are impaired.43 A lower safety factor for transverse impulse propagation has been observed in the presence of hyperkalemia,43 and the intrinsically anisotropic properties of myocardial cell coupling44 are believed to play a central role in the genesis of reentrant clinical arrhythmias related to MI.45 46 47 We obtained evidence that suggested that flecainide caused proarrhythmia by inducing reentry around arcs of transverse conduction block, as also noted in infarcted hearts in the absence of drugs.39 40 Block was rate related and appeared abruptly at critical cycle lengths (see Figs 10Up and 11Up). These findings are compatible with computer simulations of the behavior of anisotropic tissues in the presence of impaired active properties and suggest that flecainide causes proarrhythmic VT by impairing active membrane properties and causing conduction to fail in the direction of weaker cell-to-cell coupling, ie, transverse-to-fiber orientation.

New Findings and Potential Significance
The present study is the first to study experimentally the dose-response relation for flecainide proarrhythmia in an organized fashion, to evaluate associated electrophysiological changes in the presence of stable plasma concentrations, and to compare directly proarrhythmia and electrophysiological changes produced by flecainide in infarcted hearts with those in control dogs studied in parallel. While limited mapping data previously have been reported during flecainide-induced VT,10 12 the present study provides a systematic analysis of activation during VT and relates this to ventricular activation at a series of basic cycle lengths under control and drug conditions in healthy and infarcted hearts.

The specific proarrhythmic risks associated with class IC antiarrhythmic drugs were first noted in the early 1980s48 49 50 51 and led to a distinction between proarrhythmic responses to drugs that block sodium channels compared with those that prolong action potential duration.27 28 The presence of structural heart disease, a substrate that can support VT, and rapid escalation of flecainide dose have been identified as factors that increase the likelihood of proarrhythmia.52 53 In addition, exercise has been identified to be especially likely to precipitate proarrhythmic reactions, particularly during flecainide therapy,54 55 possibly by causing a sinus tachycardia.

The present study sheds light on potential mechanisms underlying these clinical findings. Anisotropic reentry, strongly facilitated by the presence of previous infarction, was found to be the likely mechanism underlying VT in 9 of our 11 MI dogs for which activation data during VT were available. Abrupt conduction block at rapid rates appeared to underlie the rate dependence of drug-induced proarrhythmia. While use-dependent sodium channel blockade causes progressive conduction slowing in the presence of sodium channel blockers,26 56 sudden decreases in conduction velocity are not seen in healthy tissues.56 Sudden rate-dependent failure of transverse conduction occurs in infarcted hearts not exposed to antiarrhythmic drugs37 40 and is predicted to occur during transverse propagation when sodium current is depressed.43 Thus, in addition to enhancing drug-induced sodium channel block, tachycardia favors the occurrence of transverse block by exposing critical discrepancies in the source-sink relation. Our findings suggest that proarrhythmia is due to interaction among cycle length, the underlying substrate, and drug effects on conduction.

Model Limitations and Results
We found that, while proarrhythmia could be induced by flecainide in dogs with prior MI, relatively high drug concentrations were necessary in most dogs. These observations are consistent with clinical findings that therapeutic flecainide concentrations cause sustained VT to be inducible in a relatively small fraction (7% to 29%) of patients.57 58 59 It is difficult to know to what extent the mechanisms of flecainide-induced VT in our dogs apply to drug-induced VT in humans. Because of important interspecies differences in drug sensitivity, it is important to evaluate indexes of drug pharmacodynamics to relate the effects of drugs seen in animal models to those noted in man. At therapeutic doses, flecainide slows intraventricular conduction (as indicated by the QRS duration) by {approx}25% in humans.51 This degree of conduction slowing is in the range of the mean changes that we observed in our dogs (Fig 3Up). Furthermore, the risk of clinical proarrhythmia increases with increasing flecainide dose,52 as does the degree of intraventricular conduction slowing.51 In a well-documented series of patients with VT that was caused by a pharmacologically similar class IC agent, encainide, drug-induced QRS prolongation averaged 46±10%,48 suggesting a degree of conduction slowing in the range that we observed in dogs exposed to 5 to 6 mg/L plasma flecainide concentrations (Fig 3Up). Thus, while VT in some dogs required concentrations that are equivalent to toxic levels in man, the similarity in conduction slowing to values in patients experiencing proarrhythmia at therapeutic doses suggests pharmacodynamic equivalence and possible relevance to clinical phenomena. Proarrhythmia in our dogs was always associated with significant conduction slowing. Clinical drug-induced VT can occur in the absence of substantial ventricular conduction slowing,50 in which case it may be due to idiosyncratic predisposition to other drug-induced arrhythmia mechanisms. Potent sodium channel blockers such as flecainide strongly increase the risk of ventricular fibrillation during acute ischemia60 61 62 at concentrations much lower than those required for proarrhythmia in the present model.63 This mechanism may well play an important role in class I drug effects on mortality.

We were able to account for a complete reentry circuit in just under half the dogs (4 of 9) in which reentrant activity could be mapped. In the other 5 dogs, activation accounting for part of a potential reentry circuit was obtained. As for most previous studies in the literature in this arrhythmia model, mapping was limited to epicardial sites over the infarct zone. Intramural and endocardial activation were not recorded. Our inability to account fully for reentry in some dogs may have been due to the involvement of intramural and subendocardial tissues in the reentry circuit, as previously demonstrated in both dog models64 65 66 and human hearts with inducible VT.67 Alternatively, some of the reentry circuit may have been epicardial but outside the field covered by our electrode array. This condition may also account for the fact that the circuits we mapped had only one limb of reentry, whereas reentry in the postinfarction VT model is frequently associated with a figure eight pattern.39 40 66 Another possibility, particularly when activation data do not reveal evidence for reentry or account for only part of a reentry circuit, would be the participation of other arrhythmia mechanisms, such as abnormal and triggered automaticity, that can occur in infarcted preparations.33 68

In any case, it is important to be aware that the results presented in this manuscript, while pointing strongly to reentry mechanisms (particularly when a complete circuit is mapped), do not provide absolute proof for reentry. We were able to satisfy two of the Mines criteria for demonstrating reentry,69 the identification of unidirectional block and the delineation of a repetitive recirculating wave front during tachycardia. We did not attempt to satisfy the final criterion, the termination of tachycardia by anatomic disruption of the reentry circuit, because the demanding technical requirements of such a demonstration (requiring a stable and hemodynamically tolerated tachycardia, rapid on-line delineation of the circuit, and precise localized interruption of the circuit without disruption of function in a beating heart) place it beyond the scope of the present study.

The model we used simulates many conditions of clinical ventricular tachyarrhythmia occurring after recovery from acute MI.33 34 35 36 37 38 39 40 67 68 Nevertheless, the model cannot be considered to mimic directly any specific clinical condition, and many aspects of the procedures involved (general anesthesia, open-chest preparation, intravenous drug administration, and programmed electrical stimulation) create potentially important differences from the clinical setting in which proarrhythmia typically occurs. Caution is therefore warranted in extrapolating the results of the present studies to man.

Several studies have suggested that sodium channel blockers may have more profound effects on longitudinal versus transverse conduction.70 71 72 73 74 However, the relative magnitude of directional differences was highly variable, and at some concentrations no differences were seen.71 The apparent discrepancy may be due to differences in action between flecainide and previously studied drugs (procainamide,70 mexiletine,71 quinidine,71 amiodarone,72 lidocaine,73 and 0–desmethyl encainide74 ), to the drug concentrations studied, or to some other technical factor.

Conclusions
We have shown that flecainide promotes the occurrence of ventricular tachyarrhythmias in a concentration-dependent fashion in a large proportion of dogs with prior MI. The presence and extent of infarction are important in determining the likelihood of VT, which appears to be caused by reentry around an arc of functional transverse conduction block. The occurrence of block is rate dependent and appears to be caused by an interaction between sodium channel blockade and the underlying substrate. These experiments support the importance of the anisotropy of ventricular conduction in the genesis of cardiac arrhythmias, particularly when intercellular coupling and active membrane properties are disturbed by MI and potent sodium channel blocking drugs.


*    Acknowledgments
 
This work was supported by the Medical Research Council of Canada, the Quebec Heart Foundation, and the Fonds de Recherche de l'Institut de Cardiologie de Montréal. Suzanne Ranger was supported by a studentship award from the Fonds pour la Formation de Chercheurs et de l'Aide à la Recherche (FCAR). We thank Emma De Blasio, Carol Matthews, and Christine Villemaire for technical assistance, Mary Morello for typing the manuscript, Drs Jean-Gilles Latour and Tack Ki Leung for advice and technical support for infarct size measurement and histologic analysis, and 3M-Riker Pharmaceuticals for supplying the flecainide used in the present experiments.


*    Footnotes
 
Reprint requests to Dr Stanley Nattel, Research Center, Montreal Heart Institute, 5000 Belanger St E, Montreal, Quebec H1T 1C8, Canada.

Received January 24, 1995; accepted February 27, 1995.


*    References
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMethods
up arrowResults
up arrowDiscussion
*References
 

  1. Nattel S. Antiarrhythmic drug classifications: a critical appraisal of their history, present status and clinical relevance. Drugs. 1991;41:672-701. [Medline] [Order article via Infotrieve]
  2. Horowitz LN, Zipes DP, Bigger JT Jr, Campbell RWF, Morganroth J, Podrid PJ, Rosen MR, Woosley RL. Proarrhythmia, arrhythmogenesis or aggravation of arrhythmia: a status report, 1987. Am J Cardiol. 1987;59:54E-56E. [Medline] [Order article via Infotrieve]
  3. The Cardiac Arrhythmia Suppression Trial (CAST) Investigators. Preliminary report: effect of encainide and flecainide on mortality in a randomized trial of arrhythmia suppression after myocardial infarction. N Engl J Med. 1989;321:406-412. [Abstract]
  4. Greene HL, Roden DM, Katz RJ, Woosley RL, Salerno DM, Henthorn RW, for the CAST Investigators. The Cardiac Arrhythmia Suppression Trial: first CAST, then CAST-II. J Am Coll Cardiol. 1992;19:894-898. [Abstract]
  5. Moosvi AR, Goldstein S, Medendorp SV, Landis JR, Wolfe RA, Leighton R, Ritter G, Vasu CM, Acheson A. Effect of empiric antiarrhythmic therapy in resuscitated out-of-hospital cardiac arrest victims with coronary artery disease. Am J Cardiol. 1990;65:1192-1197. [Medline] [Order article via Infotrieve]
  6. Coplen SE, Antman EM, Berlin JA, Hewitt P, Chalmers TC. Efficacy and safety of quinidine therapy for maintenance of sinus rhythm after cardioversion. Circulation. 1990;82:1106-1116. [Abstract/Free Full Text]
  7. Flaker GC, Blackshear JL, McBride R, Kronmal RA, Halperin JL, Hart RG, on behalf of the Stroke Prevention in Atrial Fibrillation Investigators. Antiarrhythmic drug therapy and cardiac mortality in atrial fibrillation. J Am Coll Cardiol. 1992;20:527-532. [Abstract]
  8. Zimmerman JM, Bigger JT Jr, Coromilas J. Flecainide acetate is arrhythmogenic in a canine model of sustained ventricular tachycardia. J Am Coll Cardiol. 1985;5:391. Abstract.
  9. Lynch JJ, DiCarlo LA, Montgomery DG, Lucchesi BR. Effects of flecainide acetate on ventricular tachyarrhythmia and fibrillation in dogs with recent myocardial infarction. Pharmacology. 1987;35:181-193. [Medline] [Order article via Infotrieve]
  10. Coromilas J, Saltman AE, Waldecker B, Dillon SM, Wit AL. Effects of class IC antiarrhythmic drugs on anisotropic reentry in canine infarcts. Circulation. 1988;78(suppl II):II-460. Abstract.
  11. Miyazaki T, Ogawa S, Sakai T, Hosokawa M, Sakurai K, Nakamura Y. Effects of class I drugs on reentrant ventricular tachycardia in a canine model of myocardial infarction: on their antiarrhythmic and proarrhythmic mechanisms. Jpn Circ J. 1988;52:254-261.[Medline] [Order article via Infotrieve]
  12. Sakai T, Ogawa S, Hosokawa M, Miyazaki T, Sakurai K, Nakamura Y. Electrophysiological effects of flecainide in a canine 7 day old myocardial infarction model. Cardiovasc Res. 1989;23:177-183. [Medline] [Order article via Infotrieve]
  13. 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. [Medline] [Order article via Infotrieve]
  14. Wallace AA, Stupienski RF III, Kothstein T, Gehret JR, Lynch JJ Jr. Demonstration of proarrhythmic activity with the class IC antiarrhythmic agent encainide in a canine model of previous myocardial infarction. J Cardiovasc Pharmacol. 1993;21:397-404. [Medline] [Order article via Infotrieve]
  15. Salerno DM, Murakami MAM, Asinger DA, Johnston RB, Keyler DE, Pentel PE. Reversal of flecainide-induced ventricular arrhythmia by hypertonic sodium bicarbonate vs. normal saline in dogs: a blinded randomized comparison. J Am Coll Cardiol. 1991;17(suppl A):43A. Abstract.
  16. 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. [Abstract/Free Full Text]
  17. Ranger S, Villemaire C, Nattel S. Role of previous myocardial infarction in flecainide proarrhythmia. FASEB J. 1993;7:A107. Abstract.
  18. Ranger S, Nattel S. Rate-dependent transverse conduction block and flecainide proarrhythmia in infarcted hearts. Circulation. 1993;88(suppl I):I-623. Abstract.
  19. Harris AS. Delayed development of ventricular ectopic rhythms following experimental coronary occlusion. Circulation. 1950;1:1318-1328. [Medline] [Order article via Infotrieve]
  20. Nattel S, Jing W. Rate-dependent changes in intraventricular conduction produced by procainamide in anesthetized dogs: a quantitative analysis based on the relation between phase 0 inward current and conduction velocity. Circ Res. 1989;65:1485-1498. [Abstract/Free Full Text]
  21. Wang J, Bourne GW, Wang Z, Villemaire C, Talajic M, Nattel S. Comparative mechanisms of antiarrhythmic drug action in experimental atrial fibrillation: importance of use-dependent effects on refractoriness. Circulation. 1993;88:1030-1044. [Abstract/Free Full Text]
  22. O'Hara G, Villemaire C, Talajic M, Nattel S. Effects of flecainide on the rate dependence of atrial refractoriness, atrial repolarization and atrioventricular node conduction in anesthetized dogs. J Am Coll Cardiol. 1992;19:1335-1342. [Abstract]
  23. Nattel S, Beau S, McCarragher G. Effect of frequent ventricular ectopy on myocardial infarct size in dogs. Cardiovasc Res. 1987;21:286-292. [Medline] [Order article via Infotrieve]
  24. Rousseau G, Provost P, Latour JG. Sustained myocardial protection by clentiazem (TA-3090) after a 90-minute coronary occlusion and 72 hours of reperfusion in dogs with collateral flow. J Cardiovasc Pharmacol. 1993;22:264-272. [Medline] [Order article via Infotrieve]
  25. Sachs L. In: Applied Statistics. New York, NY: Springer-Verlag Inc; 1984.
  26. 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. [Abstract/Free Full Text]
  27. 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. [Abstract/Free Full Text]
  28. Bigger JT Jr, Sahar DI. Clinical types of proarrhythmic response to antiarrhythmic drugs. Am J Cardiol. 1987;59:2E-9E. [Medline] [Order article via Infotrieve]
  29. Arenal A, Villemaire C, Nattel S. Mechanism of selective epicardial activation delay during acute myocardial ischemia in dogs. Circulation. 1993;88(pt 1):2381-2388.
  30. Horowitz LN, Spear JF, Moore EN. Subendocardial origin of ventricular arrhythmias in 24-hour-old experimental myocardial infarction. Circulation. 1976;53:56-63. [Abstract/Free Full Text]
  31. El-Sherif N, Mehra R, Gough WB, Zeiler RH. Ventricular activation patterns of spontaneous and induced ventricular rhythms in canine one-day-old myocardial infarction: evidence for focal and reentrant mechanisms. Circ Res. 1982;51:152-166. [Abstract/Free Full Text]
  32. Le Marec H, Dangman KH, Danilo P Jr, Rosen MR. An evaluation of automaticity and triggered activity in the canine heart one to four days after myocardial infarction. Circulation. 1985;71:1224-1236. [Abstract/Free Full Text]
  33. El-Sherif N, Scherlag BJ, Lazzara R, Hope RR. Re-entrant ventricular arrhythmias in the late myocardial infarction period, I: conduction characteristics in the infarction zone. Circulation. 1977;55:686-702. [Abstract/Free Full Text]
  34. Garan H, Fallon JT, Ruskin JN. Sustained ventricular tachycardia in recent canine myocardial infarction. Circulation. 1980;62:980-987. [Abstract/Free Full Text]
  35. Karagueuzian HS, Fenoglio JJ Jr, Weiss MB, Wit AL. Protracted ventricular tachycardia induced by premature stimulation of the canine heart after coronary artery occlusion and reperfusion. Circ Res. 1979;44:833-846. [Abstract/Free Full Text]
  36. Michelson EL, Spear JF, Moore EN. Electrophysiologic and anatomic correlates of sustained ventricular tachyarrhythmias in a model of chronic myocardial infarction. Am J Cardiol. 1980;45:583-590. [Medline] [Order article via Infotrieve]
  37. El-Sherif N, Smith RA, Evans K. Canine ventricular arrhythmias in the late myocardial infarction period, VIII: Epicardial mapping of reentrant circuits. Circ Res. 1981;49:255-265. [Abstract/Free Full Text]
  38. Wit AL, Allessie MA, Bonke FIM, Lammers W, Smeets J, Fenoglio JJ Jr. Electrophysiologic mapping to determine the mechanism of experimental ventricular tachycardia initiated by premature impulses. Am J Cardiol. 1982;49:167-185.
  39. Cardinal R, Vermeulen M, Shenasa M, Roberge F, Page P, Hélie F, Savard P. Anisotropic conduction and functional dissociation of ischemic tissue during reentrant ventricular tachycardia in canine myocardial infarction. Circulation. 1988;77:1162-1176. [Abstract/Free Full Text]
  40. Dillon SM, Allessie MA, Ursell PC, Wit AL. Influences of anisotropic tissue structure on reentrant circuits in the epicardial border zone of subacute canine infarcts. Circ Res. 1988;63:182-206. [Abstract/Free Full Text]
  41. Spach MS, Kootsey JM. The nature of electrical propagation in cardiac muscle. Am J Physiol. 1983;244:H3-H22.
  42. Spach MS, Dolber PC, Heidlage JF. Influence of the passive anisotropic properties on directional differences in propagation following modification of the sodium conductance in human atrial muscle. Circ Res. 1988;62:811-832. [Abstract/Free Full Text]
  43. Delgado C, Steinhaus B, Delmar M, Chialvo DR, Jalife J. Directional differences in excitability and margin of safety for propagation in sheep ventricular epicardial muscle. Circ Res. 1990;67:97-110. [Abstract/Free Full Text]
  44. Hoyt RH, Cohen ML, Saffitz JE. Distribution and three-dimensional structure of intercellular junctions in canine myocardium. Circ Res. 1989;64:563-574. [Abstract/Free Full Text]
  45. Saffitz JE, Corr PB, Sobel BE. Arrhythmogenesis and ventricular dysfunction after myocardial infarction: is anomalous cellular coupling the elusive link? Circulation. 1993;87:1742-1745. [Free Full Text]
  46. Spear JF, Kieval RS, Moore EN. The role of myocardial anisotropy in arrhythmogenesis associated with myocardial ischemia and infarction. J Cardiovasc Electrophysiol. 1992;3:579-588.
  47. Saffitz JE, Hoyt RH, Luke RA, Kanter HL, Beyer EC. Cardiac myocyte interconnections at gap junctions: role in normal and abnormal electrical conduction. Trends Cardiovasc Med. 1992;2:56-60.
  48. Winkle RA, Mason JW, Griffin JC, Ross D. Malignant ventricular tachyarrhythmias associated with the use of encainide. Am Heart J. 1981;102:857-863. [Medline] [Order article via Infotrieve]
  49. 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. [Medline] [Order article via Infotrieve]
  50. Nathan AW, Hellestrand KJ, Bexton RS, Banim SO, Spurrell RAJ, Camm AJ. Proarrhythmic effects of the new antiarrhythmic agent flecainide acetate. Am Heart J. 1984;107:222-228. [Medline] [Order article via Infotrieve]
  51. Morganroth J, Horowitz LN. Flecainide: its proarrhythmic effect and expected changes on the surface electrocardiogram. Am J Cardiol. 1984;53:89B-94B. [Medline] [Order article via Infotrieve]
  52. Morganroth J. Risk factors for the development of proarrhythmic events. Am J Cardiol. 1987;59:32E-37E. [Medline] [Order article via Infotrieve]
  53. Morganroth J, Anderson JL, Gentzkow GD. Classification by type of ventricular arrhythmia predicts frequency of adverse cardiac