Search

Magnetic resonance imaging (MRI) is often considered the gold-standard test for characterizing cardiac as well as noncardiac structure and function. However, many patients with cardiac implantable electronic devices (CIEDs) and/or severe renal dysfunction are unable to undergo this test because of safety concerns. In the past 10 years, newer-generation CIEDs and gadolinium-based contrast agents (GBCAs) as well as coordinated care between imaging and heart rhythm device teams have mitigated risk to patients and improved access to MRI at many hospitals. The purpose of this statement is to review published data on safety of MRI in patients with conditional and nonconditional CIEDs in addition to patient risks from older and newer GBCAs. This statement was developed through multidisciplinary collaboration of pan-Canadian experts after a relevant and independent literature search by the Canadian Agency for Drugs and Technologies in Health. All recommendations align with the Grading of Recommendations Assessment, Development and Evaluation (GRADE) system. Key recommendations include: (1) the development of standardized protocols for patients with a CIED undergoing MRI; (2) patients with MRI nonconditional pacemakers and pacemaker dependency should be programmed to asynchronous mode and those with MRI nonconditional transvenous defibrillators should have tachycardia therapies turned off during the scan; and (3) macrocyclic or newer linear GBCAs should be used in preference to older GBCAs because of their better safety profile in patients at higher risk of nephrogenic systemic fibrosis.

Background Access to pacemakers and defibrillators is problematic in places with limited resources. Resterilization and reuse of implantable cardiac devices obtained post mortem from patie...

To our knowledge, heart rate responses have not previously been assessed in hockey fans. We quantified heart rate increases in spectators of the Montreal Canadiens, compared televised with live games, explored features associated with peak heart rates, and assessed whether increases correlate with a fan passion score.Healthy adults were enrolled, with half attending live games and half viewing televised games. All subjects completed questionnaires and had continuous Holter monitoring. Intensity of the physical stress response was defined according to previously published heart rate index thresholds as mild (1.33), moderate (1.33-1.83), or vigorous (1.83).In 20 participants, 35% women, age 46 ± 10 years, the heart rate increased by a median of 92% during the hockey game, from 60 (interquartile range, 54-65) beats per minute at rest to 114 (interquartile range, 103-129) beats per minute (P0.001). The heart rate increased by 110% vs 75% during live vs televised games (P0.001). Heart rate index (2.16 ± 0.27 vs 1.73 ± 0.15; P0.001) and percent maximum predicted heart rate attained (75% ± 8% vs 58% ± 7%; P 0.001) were significantly higher during live vs televised games. Number of premature beats was nonsignificantly higher during live games (5 vs 1; P = 0.181). The fan passion score was not predictive of the heart rate response (P = 0.753). Peak heart rates most commonly occurred during overtime (40%) and scoring opportunities for (25%) and against (15%).It is exciting to watch the Montreal Canadiens! Viewing a live hockey game is associated with a heart rate response equivalent to vigorous physical stress and a televised game to moderate physical stress.

Background: Clinical trials suggest that sotalol and dofetilide are much more effective in preventing atrial fibrillation (AF) than in terminating it. This study evaluated potential mechanisms of discordant sotalol and dofetilide effects on AF termination vs. prevention. Methods: We applied 240-electrode epicardial mapping and programmed stimulation in a vagotonic dog model of AF before and after dofetilide or sotalol. Results: Under control conditions, sustained AF could be induced by single S 2 extrastimuli that caused unidirectional block and macroreentry. Sotalol (2 mg/kg) and dofetilide (0.04 mg/kg) failed to terminate AF in any dog, but prevented AF induction by S 2 stimuli in 19/22 (86%) and 4/5 (80%) of animals, respectively. With sotalol and dofetilide, unidirectional block still occurred, but wavefront reentry failed. The prevention of S 2-induced reentry was related to large increases in the effective refractory period (ERP) at a BCL of 1000 ms, leading to ERPs that exceeded the conduction delay following S 2. Reverse use-dependent effects resulted in smaller ERP increases at BCLs closer to the AF cycle length. Although the number of zones of reactivation per cycle during sustained AF were decreased by sotalol and dofetilide, the changes were small and insufficient to terminate AF. Conclusions: Sotalol and dofetilide prevent AF initiation by premature depolarizations at doses that fail to terminate vagotonic AF, by increasing ERP at the basic cycle length beyond the associated conduction delay that leads to reentry.

BACKGROUND Tachycardia enhances the channel-blocking effects of antiarrhythmic drugs. In contrast to the extensive data regarding the rate-dependent effects of sodium channel blockers in humans, little is known about the frequency-dependent effects of calcium channel blockers on human atrioventricular (AV) nodal properties. Accordingly, the purpose of this study was to evaluate the importance of heart rate in modulating the electrophysiological effects of diltiazem in humans. METHODS AND RESULTS Electrophysiological studies were performed in 25 patients. Sinus node, atrial, and AV nodal function were evaluated at multiple atrial rates under control conditions and after administration of one of three intravenous doses of diltiazem designed to produce low, intermediate, and high stable plasma concentrations (designated doses 1, 2, and 3, respectively). Results were analyzed in terms of the longest and shortest cycle lengths obtainable in each patient under control and drug conditions. Plasma concentrations of diltiazem were stable and averaged 43 +/- 4, 73 +/- 6, and 136 +/- 11 ng/ml for doses 1, 2, and 3, respectively. Sinus node recovery time, intra-atrial conduction time, atrial effective refractory period, and HV interval were unaffected by diltiazem infusion. Effects of diltiazem were limited to changes in AV nodal parameters. Stable, dose-dependent increases in Wenckebach cycle length were observed after all three doses of diltiazem (increases of 54 +/- 13, 84 +/- 18, and 174 +/- 33 msec for doses 1, 2, and 3, respectively). Small nonsignificant increases in AH interval and atrioventricular effective refractory period (AVERP) were observed after dose 1 of diltiazem. At long cycle lengths, diltiazem caused modest increases in AH interval (3 +/- 4 and 25 +/- 8 msec for doses 2 and 3, respectively) and AVERP (36 +/- 12 and 70 +/- 25 msec). Drug effects were far greater at short cycle lengths (45 +/- 17 msec, 58 +/- 12 msec for AH interval and 80 +/- 24 msec, 163 +/- 41 msec for AVERP; p less than 0.05 versus values at long cycle lengths). At rapid rates, effects of diltiazem on AVERP substantially exceeded those on AV conduction, a result that could account for the beneficial effects of diltiazem during paroxysmal AV reentrant tachycardia by decreasing the excitable gap. CONCLUSIONS Depressant effects of diltiazem on human AV nodal function are highly dependent on atrial rate; the rate-dependent actions on AV nodal refractoriness probably contribute to beneficial effects of diltiazem in patients with supraventricular arrhythmias.

Autonomic blockade for in vivo electrophysiological studies generally involves atropine and beta blockers, ignoring the potential role of alpha-adrenergic activity. To evaluate the importance, if any, of alpha-adrenergic tone, the electrophysiological effects of incremental doses of phenylephrine were examined in eight chloralose-anesthetized dogs. In order to study direct effects, all dogs were both beta blocked (with nadolol) and vagally blocked (with the combination of vagotomy and atropine). Results were also obtained after normalization of blood pressure with nitroprusside or the alpha-blocker, prazosin. Phenylephrine caused dose-dependent increases in systolic and diastolic blood pressure. This was accompanied by consistent but modest decreases in sinus cycle length (control RR interval 492 +/- 34 msec vs 459 +/- 29 msec after dose 1 of phenylephrine, P less than 0.05, 516 +/- 41 vs 484 +/- 34 msec control versus dose 2, P less than 0.05). These increases in automaticity were not prevented after normalization of arterial pressure by nitroprusside, but were reversed when concomitant alpha-receptor blockade was achieved with prazosin, suggesting that sinus node acceleration resulted directly from alpha-receptor stimulation. No effects on atrial, AV nodal or His-Purkinje conduction were noted. In addition, phenylephrine did not affect atrial, AV nodal, or ventricular refractoriness. In conclusion, conduction and refractoriness of normal cardiac tissue (other than the sinus node) are unaffected by direct alpha-receptor stimulation. This justifies the use of combined beta and muscarinic blockers to achieve autonomic blockade under most circumstances.

While radiofrequency catheter ablation is very effective, it does not allow for prediction of success prior to full delivery of the energy. We investigated the use of cryoablation using a new catheter on the AV node to determine (1) if a successful site might be identified prior to the ablation itself, and (2) the parameters of cryoablation of the AV node using a new cryocatheter. In eight dogs, the cryoablation catheter was advanced to the AV node to produce transient high degree AV block by lowering the temperature to a minimum of -40 degrees C (ice mapping). Transient high degree AV node block was obtained in seven of eight animals at a mean temperature of -39.9 +/- 11.6 degrees C. No significant pathological modification was found in all animals but one and, in all cases, electrophysiological parameters of the AV node measured before, 20 minutes, 60 minutes, and up to 56 days after cryoapplication were not significantly different. In the 12 other dogs, after ice mapping, cryoablation of the AV node was attempted with a single freeze-thaw cycle in 6 dogs (group I) and a double freeze-thaw cycle in the other 6 dogs (group II). Chronic complete AV block was obtained in only one animal in group I compared to all animals in group II. Ablation of the AV node is effective with a double freeze-thaw cycle using a percutaneous catheter cryoablation system. Ice mapping of the area allows for identification of the targeted site.

Background —Rapid atrial activation causes electrical remodeling that promotes the occurrence and the maintenance of atrial fibrillation (AF). Although remodeling has been shown to alter electrophysiological variables, the spatial uniformity of these changes is unknown. Methods and Results —Dogs subjected to rapid atrial pacing (400 bpm) for 24 hours (n=12) were compared with sham-operated dogs (instrumented but not paced, n=12). Epicardial mapping (240 bipolar electrodes) and extrastimulation at a large number of sites (mean±SEM, 66±4 per dog) were used to evaluate atrial activation and the heterogeneity of the effective refractory period (ERP), respectively. Rapid pacing increased both the percentage of sites at which AF could be induced by single premature stimuli (from 2.6±0.9% to 11.8±2.8%, P =0.007) and AF duration (from 39±28 to 146±49 seconds, P =0.03). Atrial tachycardia decreased atrial ERP (from 120±4 to 103±2 ms, P =0.003), increased the coefficient of variation of ERP (from 14.9±0.9% to 20.7±0.9%, P P =0.0004), with no change in the wavelength. The increase in ERP heterogeneity was due both to interregional differences in the extent of ERP remodeling and to increased intersite variability within regions. Stepwise multilinear regression indicated that ERP heterogeneity was an independent determinant of the inducibility ( P P Conclusions —Atrial tachycardia causes nonuniform remodeling of atrial refractoriness that plays an important role in increasing atrial vulnerability to AF induction and the duration of induced AF.

BACKGROUND Antiarrhythmic drugs are known to have state-dependent interactions with cardiac sodium channels, and these have potentially important implications for drug effects on cardiac conduction, particularly in situations of changed resting potential and heart rate. Recent advances in theoretical approaches permit beat-to-beat changes in sodium channel block to be inferred from conduction changes in vivo and allow for an analysis of state-dependent drug action from conduction changes occurring on the onset of pacing at different rates. The purpose of the present study was to use this method to analyze the interaction between hyperkalemia and procainamide's sodium channel-blocking action in terms of resulting changes in left ventricular conduction. METHODS AND RESULTS Epicardial mapping with a 56-electrode array was used to assess ventricular conduction in open chest, anesthetized mongrel dogs with Formalin-induced atrioventricular block. Procainamide was infused as a series of loading and maintenance infusions until at least 20% conduction slowing was obtained at the shortest basic cycle length (300 milliseconds). Results in a control set of normokalemic dogs were compared with results in dogs with moderate hyperkalemia produced by a loading and maintenance infusion of potassium chloride. Plasma procainamide concentration was measured by high-performance liquid chromatography, and the constancy of serum potassium concentration was verified with ion-sensitive electrode measurement. Although hyperkalemia itself (mean +/- SEM potassium concentration, 6.64 +/- 0.66 mmol/L) did not alter conduction, it resulted in substantially increased conduction slowing by procainamide despite substantially lower plasma drug concentrations (102 +/- 10 mumol/L) compared with normokalemic dogs (potassium concentration, 3.87 +/- 0.24 mmol/L; procainamide concentration, 277 +/- 16 mumol/L). The onset of conduction slowing and block followed basic molecular theory, with an exponential time constant that was faster at longer cycle lengths and total block that increased as cycle length decreased. Piecewise exponential analysis of block during the rested and depolarized phases of the action potential showed that the enhancement of procainamide's action by hyperkalemia was due almost exclusively to increased rested-phase block. Hyperkalemia produced a bradycardia-dependent and slight reduction in action potential duration and antagonized the action potential-prolonging effect of procainamide, particularly at shorter cycle lengths. CONCLUSIONS Hyperkalemia strongly enhances procainamide-induced conduction slowing by increasing the interaction between the drug and sodium channels during the rested phase of the cardiac cycle. These results indicate the applicability of basic molecular theories of antiarrhythmic drug action to understanding drug-induced changes in conduction velocity in vivo and highlight the potential importance of heterogeneous magnification of sodium channel-blocking drug action by the spatially variable hyperkalemia that occurs with acute myocardial ischemia. The latter could play an important role in the known proarrhythmic potential of sodium channel-blocking drugs in patients with coronary artery disease.

BACKGROUND Use-dependent effects of antiarrhythmic drugs on phase 0 sodium current result in rate-dependent conduction slowing with important potential clinical consequences. The purpose of the present study was to determine whether state-dependent interactions of procainamide with sodium channels can be analyzed based on conduction changes in vivo. METHODS AND RESULTS Procainamide infusions were used to produce stable drug concentrations causing greater than or equal to 25% conduction slowing at a basic cycle length (BCL) of 300 msec in morphine/chloralose-anesthetized dogs with formalin-induced atrioventricular block. Computer-based epicardial activation mapping was applied to assess the time course and pattern of conduction over a wide range of BCLs before and after drug administration. Action potential duration was measured from recordings of monophasic action potentials. The onset and steady-state values of fractional sodium channel block estimated from conduction changes were fitted to equations obtained from a stepwise exponential analysis. The rate constant for the onset of block (lambda *) decreased, as predicted, with decreasing cycle length. The slope of the relation between lambda * and recovery time at each BCL averaged 0.29 +/- 0.03 sec-1, resulting in a calculated recovery time constant (3.4 seconds) similar to values previously obtained by direct measurement. Estimates of binding and unbinding rate constants for the sodium channel during the action potential plateau and after repolarization were of the same order as previous results obtained using microelectrode methods in vitro. CONCLUSIONS Use-dependent conduction changes produced by procainamide in vivo closely follow the predictions of mathematical models of drug-channel interactions, and underlying kinetic interactions with the sodium channel inferred from conduction changes agree with previous, more direct observations. These results support the relevance of basic concepts about antiarrhythmic drug actions on sodium channels for understanding drug effects on conduction in vivo and advance analytical tools that can be used to explore the latter in humans.

BACKGROUND Rate-dependent conduction slowing by class I antiarrhythmic agents has clinically important consequences. Class I drugs are known to produce use-dependent sodium channel blockade. If rate-dependent conduction slowing by class I agents is due to sodium channel blocking actions, the kinetics of conduction slowing should be similar to those of depression of sodium current indexes in vitro. The purpose of the present investigation was to study the onset time course of ventricular conduction slowing caused by a variety of class I agents in humans. METHODS AND RESULTS Twenty-seven patients undergoing electrophysiological evaluation for antiarrhythmic therapy were studied. Changes in QRS duration at initiation of ventricular pacing at cycle lengths of 400 and 500 msec were used to evaluate the kinetics of drug action. Mean time constants for each drug were similar to values for Vmax depression reported in vitro studies: flecainide, 24.9 +/- 11.6 beats in eight patients (versus 34.5 beats reported for Vmax block); propafenone, 17.8 +/- 6.9 beats in five patients (versus 8.4-20.8 beats); quinidine, 7.0 +/- 2.4 beats in six patients (versus 5.6-6.2 beats); and amiodarone, 3.6 +/- 2.0 beats for eight patients (versus 3.0 beats). Time constants were significantly different among the various drugs tested (p = 0.0002 at a cycle length of 400 msec; p = 0.002 at 500 msec), and there was a strong correlation (r = 0.89, p less than 0.0001) between values obtained at a cycle length of 400 msec and those at a cycle length of 500 msec. No rate-dependent changes in QRS duration were seen at onset of ventricular pacing among eight age- and disease-matched control patients not taking class I antiarrhythmic drugs, including three patients subsequently showing such changes during type I antiarrhythmic drug therapy. CONCLUSIONS We conclude that class I agents produce use-dependent QRS prolongation in humans with characteristic kinetics for each agent that are similar to the kinetics of Vmax depression in vitro. These results suggest that rate-dependent ventricular conduction slowing by antiarrhythmic drugs in humans is due to use-dependent sodium channel blockade.

Cycle length alternation (CLA) is commonly observed during supraventricular tachycardia (SVT) onset and termination. The present study was designed to gain insights into the mechanism and potential clinical relevance of CLA by comparing computer simulations of tachycardia to directly observed behavior in a canine model of AV reentrant tachycardia (AVRT). The computer model was based on the hypothesis that CLA is secondary to feedback between AV nodal output during SVT and subsequent AV nodal input, and used the measured anterograde AV nodal recovery curve (AV vs A1A2) to predict sequential AV and RR intervals during SVT. Orthodromic AVRT was created experimentally in 11 open-chested, autonomically-blocked (atropine plus nadolol) dogs using a sensing and pacing circuit that mimicked a retrograde-conducting accessory pathway. Steady-state cycle length and AV interval during experimental AVRT closely paralleled predictions made by the computer model. CLA appeared consistently at the onset of experimental AVRT at programmed VA intervals less than or equal to 100 msec (corresponding to VA less than or equal to 150 msec as measured clinically) in all dogs. The amplitude and duration of CLA increased as the VA interval decreased, and closely paralleled predictions based on the computer model. Abrupt accelerations in atrial pacing to the same rate as AVRT did not result in alternation of cycle length. In conclusion, alternation of cycle length results from feedback between AV nodal output and subsequent AV nodal input at the onset of reentrant supraventricular tachycardia, and does not require changes in autonomic tone or dual AV nodal pathways. CLA occurrence, amplitude, and duration are predictable based on AV node recovery properties, and depend on retrograde conduction properties of the reentrant circuit. The presence of CLA suggests that the AV node is an integral component of the SVT reentry circuit, and may be useful clinically to identify the mechanism of supraventricular tachycardias.

The purpose of this study was to determine if the known frequency-dependent effects of diltiazem on inward calcium current result in selective actions during supraventricular tachycardia. These effects were evaluated by use of an experimental model of orthodromic atrioventricular reentrant tachycardia (AVRT). AVRT was induced in 15 dogs over a wide range of retrograde conduction times before and after two doses of diltiazem. Diltiazem produced a tachycardia-related suppression of atrioventricular nodal conduction resulting in greater efficacy for faster than for slower AVRTs. The degree of slowing for tachycardias that remained inducible after diltiazem administration was greater for AVRTs with a rapid initial rate (dose 1, 29%; dose 2, 40%) than for slower AVRTs (dose 1, 11%, p less than 0.01; dose 2, 18%, p less than 0.001). Rate-dependent AVRT slowing occurred because of a time-dependent phase of AH interval prolongation after the onset of tachycardia, which was observed only after diltiazem administration. to further clarify the mechanism of diltiazem's selective actions against faster tachycardias, its effects on the minimum pathway for reentry, or wavelength, were examined in four dogs. The ratio of refractory period to revolution time (RP/RT), an index of wavelength, was measured for each AVRT before and after diltiazem administration. Diltiazem increased the positive slope of the relation between RP/RT and the AVRT rate threefold compared with control (p less than 0.05). This rate-dependent effect prevented AVRT when RP/RT became greater than unity. In conclusion, rate-dependent atrioventricular node depression by diltiazem results in greater tachycardia slowing and higher rates of termination during atrioventricular reentrant tachycardias with faster initial rates and shorter retrograde conduction intervals.

Flecainide is effective against certain supraventricular arrhythmias (atrial fibrillation and atrioventricular [AV] node reentrant tachycardia), but its mechanisms of action are unknown. Previous in vitro work suggests that flecainide attenuates rate-dependent action potential duration shortening, producing tachycardia-dependent prolongation of the refractory period. This study was designed to assess whether similar changes occur in vivo and whether the effects of flecainide on AV node conduction depend on heart rate and on direction of propagation (anterograde vs. retrograde). The effects of flecainide at three clinically relevant concentrations were assessed in open chest, morphine-chloralose-anesthetized dogs. Flecainide increased atrial refractory period in a concentration- and rate-related fashion (e.g., dose 3 increased the atrial effective refractory period by 9 +/- 4% at a cycle length of 1,000 ms but by 36 +/- 5% and 55 +/- 10% at a basic cycle length of 400 and 300 ms, respectively; p less than 0.001 for each). Flecainide attenuated the action potential duration accommodation (measured by monophasic action potentials) to heart rate, causing tachycardia-dependent action potential duration prolongation and accounting for most of the rate-dependent atrial effective refractory period changes. Flecainide increased Wenckebach cycle length, but the concentration-response curve was much steeper in the retrograde (slope 41 +/- 7 ms/mumol.liter-1) than in the anterograde direction (17 +/- 4 ms/mumol.liter-1; p less than 0.01), indicating more potent effects on retrograde conduction. The depressant action of the drug on the AV node was also rate dependent, with an effect on the AH interval at a basic cycle length of 400 ms that averaged 1.8, 1.5 and 2 times that at a basic cycle length of 1,000 ms for doses 1 (p less than 0.05), 2 (p less than 0.01) and 3 (p less than 0.001), respectively.1) Flecainide suppresses atrial action potential duration accommodation to heart rate changes in vivo, leading to rate-dependent atrial effective refractory period prolongation, which may be important in suppressing atrial fibrillation. 2) The drug has frequency- and direction-dependent effects on AV node conduction, which may lead to selective antiarrhythmic actions during AV node reentry.

Vagal effects on atrioventricular (AV) nodal conduction are accentuated by increases in heart rate. To establish the mechanism of these rate-dependent negative dromotropic actions, we studied the properties governing AV nodal adaptation to changes in heart rate in chloralose-anesthetized dogs in the absence and presence of bilateral cervical vagal nerve stimulation (20 Hz, 0.2 msec). Stimulation protocols were applied to evaluate the contributions of changes in AV nodal recovery, facilitation, and fatigue independently of each other. Vagal stimulation slowed AV nodal recovery in a voltage-dependent way, increasing the time constant of recovery (tau r) from 80 +/- 7 to 194 +/- 16 msec (mean +/- SEM, p less than 0.01) at the highest voltage studied. The facilitating effect of a premature (A2) beat was manifested by a leftward shift of the recovery curve (A3H3 versus H2A3) of a subsequent A3 beat. The magnitude of shift depended on the A1A2 coupling interval and was reduced by vagal stimulation at all A1A2 intervals (maximum shift: control, 63 +/- 12 msec; vagus, 24 +/- 11 msec; p less than 0.01). When recovery and facilitation were kept constant, abrupt increases in AV nodal activation rate caused a slow (tau = 75 beats) increase in AH interval (fatigue). Vagal stimulation increased the magnitude of this process (maximum: control, 11 +/- 2 msec; vagus, 27 +/- 3 msec; p less than 0.001), without altering its time course. At activation rates comparable to sinus rhythm in humans, vagal stimulation at an intermediate voltage increased the AH interval by 25 msec. As heart rate increased, vagally induced changes in dynamic processes amplified AH prolongation up to fivefold at maximum rate. The role of vagal changes in individual functional properties depended on heart rate, but slowing of recovery was the single most important factor, constituting over 50% of overall vagal action at rapid rates. We conclude that vagal stimulation alters the ways in which the AV node responds to changes in activation rate and that at rapid rates most of the negative dromotropic action of the vagus is due to changes in the AV nodal response to tachycardia. Alterations in rate-dependent AV nodal properties are a novel and potentially important mechanism through which interventions may affect AV nodal conduction.

BACKGROUND Antiarrhythmic drugs are considered to terminate atrial fibrillation by prolonging refractoriness, but direct experimental evaluation of this concept has been limited. The atria are activated rapidly during atrial fibrillation, and antiarrhythmic drugs are known to have important rate-dependent actions. The potential role of such properties in determining drug effects during atrial fibrillation has not been evaluated. METHODS AND RESULTS We evaluated the effects of representative class Ia (procainamide), Ic (propafenone), and III (sotalol) antiarrhythmic drugs on sustained cholinergic atrial fibrillation and atrial electrophysiological properties in anesthetized, open-chest dogs. Loading and maintenance doses were used to produce stable plasma concentrations, and computer-based 112-electrode epicardial mapping was used to study atrial conduction and activation during atrial fibrillation. Clinically used doses of procainamide and propafenone terminated atrial fibrillation in 13 of 13 (100%) and 7 of 10 (70%) dogs, respectively, but a dose of sotalol (2 mg/kg IV) in the clinical range terminated atrial fibrillation in only 2 of 8 (25%) dogs (P = .0005 vs procainamide, P = .08 vs propafenone). Procainamide and propafenone prevented atrial fibrillation induction in 13 of 13 (100%) and 7 of 10 (70%) dogs, respectively, compared with none of 8 dogs for 2 mg/kg sotalol (P < .0001 vs procainamide, P = .004 vs propafenone). A larger dose of sotalol (cumulative dose, 8 mg/kg) was uniformly effective in terminating atrial fibrillation and preventing its induction. All drugs significantly increased atrial refractory period, with effects that were use dependent for propafenone but reverse use dependent for sotalol. Effective doses of all drugs significantly increased the wavelength for reentry at rapid atrial rates in the presence of vagal stimulation into the range observed under drug-free conditions in the absence of vagal input. The inefficacy of clinical doses of sotalol was explained by the reverse use dependence of its effects on refractoriness, which resulted in reduced effects on wavelength at rapid rates. The effects of propafenone on refractoriness were significantly increased at rapid rates, contributing to its ability to increase wavelength and terminate atrial fibrillation. Activation mapping showed that drugs terminated atrial fibrillation by reducing the number and increasing the size of reentry circuits, leading to termination by mechanisms related to block in the remaining circuit(s). CONCLUSIONS We conclude that antiarrhythmic drugs terminate experimental atrial fibrillation by increasing the wavelength for reentry at rapid rates, leading to a reduction in the number of functional reentry circuits and, eventually, failure of reentrant excitation. Use-dependent effects on refractoriness can limit (in the case of the reverse use dependence of sotalol) or contribute (in the case of propafenone) to antiarrhythmic drug efficacy against atrial fibrillation by determining drug-induced changes in wavelength at rapid atrial rates.

Parameters describing the static and dynamic properties of AV nodal conduction were individually examined and quantitatively assessed using different pacing protocols on 7 autonomically blocked dogs. A computer model was developed to simulate AV nodal conduction on a beat to beat basis following any type of stimulation history. The model was used to simulate time-dependent changes observed in conduction patterns during second degree AV nodal block (Wenckebach rhythms). The model's predictions show excellent agreement with the experimental observations.