Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Arrhythmias are one of the most important causes of mortality in patients with heart failure (HF), although the mechanisms of ventricular arrhythmias (VA) during the development of the disease remain unclear (Luu et al., 1989; Kjekshus, 1990). To date, however, amiodarone might be the only effective antiarrhythmic drug demonstrating a reduction of arrhythmic death in patients with depressed left ventricular function and some beneficial effect on survival in selected patients with HF (Amiodarone Trials Meta-Analysis Investigators, 1997; Cairns et al., 1997; Julian et al., 1997). Unfortunately, the use of this drug in clinical practice is limited by its very long and unpredictable half-life and by some serious toxic side effects (Zipes et al., 1984). There is thus still a need to search for new antiarrhythmic agents.

Dronedarone, previously labeled SR 33589, is a noniodinated benzofurane derivative structurally related to amiodarone with proven effectiveness on ischemia and reperfusion-induced arrhythmias in animal models, but presumably without its deleterious effects (Chatelain et al., 1995; Finance et al., 1995; Manning et al., 1995a,b). The acute administration of dronedarone in dogs results in electrophysiological actions similar to those produced by amiodarone. Nevertheless, this compound has not been studied in models with chronic HF associated with VA, nor have its electrophysiological effects been assessed in untethered, awake animals.

The present study combines in vitro and in vivo models. Our purpose was to investigate the cellular electrophysiological effects of dronedarone on both ionic currents and action potential characteristics, as well as its effects on arrhythmias, in control rats and in postmyocardial infarcted (PMI) rats, a well documented model of ventricular remodeling with significant electrophysiological alterations (Aimond et al., 1999). The cellular electrophysiological study was focused on potassium currents because dronedarone is close to amiodarone, a known class III antiarrhythmic agent. The major effects of dronedarone are to markedly reduce the sustained outward current in both control and PMI rats without significantly modifying the action potential (AP) time course in the latter animals. Dronedarone also reduced spontaneous arrhythmias in untethered animals under telemetry ECG monitoring.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Experimental Myocardial Infarction. Male Wistar rats weighing 180 to 230 g underwent left anterior coronary artery ligation according to Pfeffer et al. (1979, 1991). Briefly, rats were anesthetized with a mixture of 150 mg/kg i.p. ketamine and 15 mg/kg i.p. chlorpromazine before being intubated and ventilated. After median-left thoracotomy and pericardium opening, the left main coronary artery was occluded at the postproximal point below the left atrial appendage. Successful coronary ligation was recognized by pallor of the anterior left ventricular free wall and by the occurrence of immediate regional dyskinesia. Sham-operated rats were submitted to the same protocol except for the coronary artery ligation. Rats were then allowed to recover in individual cages. Rats surviving the ligation (70% at 4 months, not including the initial death during the surgery and the first 2 weeks) and shams received similar housing conditions, including ad libitum food, water, and a 12-h light/dark cycle. Four months after operation, PMI (n = 6) and sham animals (n = 6) were sacrified for electrophysiological experiments. Scar size was 24 ± 3% (n = 6) of the left ventricle free wall area; necrosis was transmural as checked at the late stage of cell dissociation. Hemodynamic measurements in rats undergoing the same operation and showing similar range of large infarcted scar areas demonstrated a significant increase in end-diastolic pressure from 2.5 ± 0.5 to 17.1 ± 13.5 mm Hg (n = 9). Using M-mode echocardiography, PMI rats demonstrated a 45.5 ± 0.8% increase in end-diastolic left ventricular internal diameter and a 57.0 ± 5.9% decrease in in vivo shortening fraction (n = 8; P < .0001) compared with basal values of 7.25 ± 0.43 mm and 47.9 ± 4.2%, respectively, in normal rats of same age.

In Vitro Study of Cellular Action Potential. Normal and PMI male Wistar rats (450-500 g) were stunned and the heart quickly removed. A right papillary muscle was cut and set in a bath where it was superfused at 36°C by a modified Tyrode's solution (107 mM NaCl, 4 mM KCl, 1.8 mM CaCl₂, 1.05 mM MgCl₂, 1.04 mM NaH₂PO₄, 30 mM NaHCO₃, 2 mM Na pyruvate, 11.5 mM glucose, gassed with O₂/CO₂ 95%:5%; pH 7.4). After a 2-h stabilization period, transmembrane action potentials were recorded by a semifixed glass microelectrode (15-20 M) kept 1 mm away from the stimulating bipolar platinum electrode. The basic stimulation length was 400 ms. APs were acquired and processed with a DATAPAC computer program (Biologic, Grenoble, France). Mean ± S.E. and percentage of change compared with control were calculated for all parameters at the end of the 30-min superfusion period with each of the dronedarone concentrations. The comparison of control values or drug effects between normal and PMI rats and the dose-dependent relationship were analyzed by ANOVAREP and Duncan's Multiple Range Test (RS1 computer program; BBN Software Products, Cambridge, MA).

Cell Isolation. Ventricular myocytes were dissociated from the hearts of urethane-anesthetized (2 g/kg i.p.) young (6 weeks), sham, or PMI (24 weeks) Wistar rats as previously described (Pucéat et al., 1995). The heart was first perfused for 4 min at 35°C with a nominally Ca²⁺-free HEPES-buffered solution containing 117 mM NaCl, 5.7 mM KCl, 4.4 mM NaHCO₃, 1.5 mM KH₂PO₄, 1.7 mM MgCl₂, 21 mM HEPES, 11 mM glucose, and 20 mM taurine. This was followed by a 50-min perfusion with the same solution also containing 20 µM Ca²⁺ ions and 1.9 mg/ml collagenase (CLS4; Worthington Biochemical, Lakewood, NJ). At the end of the collagenase perfusion, the ventricles were cut off and stirred to isolate cells. The cells were suspended in HEPES buffer with 1 mM Ca²⁺ and 0.5% BSA (pH 7.4). Two decantations were performed to separate dead cells. The yield of well striated, elongated cells was 70% for sham and 50% for PMI animals.

Solutions and Drugs. For voltage-clamp experiments, a cell aliquot was put in a Petri dish containing the control solution (117 mM NaCl, 2.5 mM KCl, 1.7 mM MgCl₂, 1.8 mM CaCl₂, 10 mM HEPES, and 10 mM glucose; pH adjusted to 7.4 with NaOH). Experiments were performed at room temperature (22 ± 2°C). After a cell had been sealed to the electrode and the patch broken, it was exposed to different extracellular solutions by positioning it at the extremity of one of six capillaries (250 µm i.d.). Such a system allowed for rapid changes of solutions (<2 s). From these capillaries flowed the control solution to which was added 50 µM tetrodotoxin and 2 mM CoCl₂, respectively, to block Na⁺ and Ca²⁺ currents, in the presence or absence of the compound to be tested. At 2 mM, CoCl₂ also blocked the steady state K⁺ current (Scamps, 1996).

Cellular Electrophysiological Recording. The whole-cell patch-clamp technique (Hamill et al., 1981) was used. Recordings were obtained with a patch-clamp amplifier (model RK-400; Biologic) and filtered at 3 kHz. Current traces were digitized at 1 kHz (12-bit analog-to-digital converted) and stored on a computer disk; acquisition and analyses were perfomed with the pClamp6 software (Axon Instruments, Foster City, CA). The series resistance (R_s), membrane capacitance (C_m), and time constant of membrane capacitance (T_c) were determined on most of voltage-clamped cells according to the following equations:
in which I_o is the maximum membrane current, I is the current at the end of the 10-ms pulse, and E_m is the amplitude of the voltage step applied (2 mV from a holding potential of 70 mV).

"In Vivo" Holter Monitoring Study in Untethered Animals. Holter recording on PMI rats was designed to discriminate eventual anti- or proarrhythmic effects of dronedarone. Four weeks after infarction, survivors were anesthetized again and a hermetically sealed transmitter (DSI, St. Paul, MN) was implanted i.p. The positive electrode was placed in a V 5 position and the negative one near the right scapula. To increase the power of the analysis of the arrhythmic effects of the compound, PMI rats were included only if their original arrhythmic score was >50 ventricular beats per hour (VPB/h) on two different days. Continuous recordings (3 h from 7 to 10 PM) of the untethered rats were fed into a personal computer and a human Holter monitoring equipment (Model 2448; Ela Medical, Le Plessis-Robinson, France) at a recording speed of 4 mm/s. Holter recordings were analyzed with a recent version of Ela Medical software "Elatec" to evaluate the severity of arrhythmias, two scores were considered, a quantitative score (VPB/h) and a qualitative score. In the latter, only the most severe event (VT or doublets) on each rat was taken into account.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Effects of Dronedarone on AP Characteristics. The general effects of dronedarone were first compared on AP characteristics recorded in papillary muscles isolated from normal and PMI rat hearts under a basic stimulation cycle length of 400 ms. Three concentrations (3, 10, and 30 µM) were applied sequentially for 30 min each. Figure 1 shows that the main effect of dronedarone in normal hearts was to increase AP duration, particularly the late repolarizing phase. Thus, 10 µM dronedarone increased the action potential durations at 70 and 90% repolarization by 27 and 29%, respectively. This effect was associated with a 10% reduction in the maximal rate of the ascending phase (dV/dt_max). At the highest concentration (30 µM), dronedarone also induced a significant depolarization of the resting membrane potential that was associated with a reduced dV/dt_max and a marked AP prolongation (Table 1).

View larger version (16K): [in this window] [in a new window]   Fig. 1.   General effects of dronedarone on cellular APs. Original AP recordings on papillary muscles isolated from normal (A) and PMI (B) rats submitted to increasing concentrations (3, 10, and 30 µM) of dronedarone for a 30-min period.

General Effects of Dronedarone on Potassium Currents. Cardiac cell repolarization is mostly controlled by potassium currents. Figure 2A shows such currents recorded on a sham cell in the presence of Na⁺ and Ca²⁺ current inhibitors. Depolarizing pulses activated two types of outward currents: an early outward current (Ipeak) due to the activation of the transient outward potassium current, I_to superimposed with the slower activation of the delayed rectifier potassium current, I_K (also named Isus). Hyperpolarizing pulses activated an inward current, I_K1.

View larger version (27K): [in this window] [in a new window]   Fig. 2.   General effects of dronedarone on the K+ currents elicited in a sham ventricular rat myocyte. A, K+ currents were recorded in the presence of tetrodotoxin and Co2+ to inhibit, respectively, the Na+ and Ca2+ currents when applying membrane voltages from 130 to +50 mV for 400 ms every 4 s from a 70-mV holding potential. B, dronedarone was applied at 1 µM during 6 min under the same voltage-clamp conditions. C, dronedarone-sensitive current estimated as the difference of the currents recorded at each voltage in the presence or absence of dronedarone. D, dronedarone-sensitive current-voltage relations established at peak of the outward current and at the end of the 400-ms depolarizing pulses (n = 6).

View larger version (21K): [in this window] [in a new window]   Fig. 3.   Current-voltage relationships established in control conditions and after 6 min in the presence of 1 µM dronedarone at peak of the outward current (Ap, Bp, and Cp) and at the end of the 400-ms voltage pulses (As, Bs, and Cs) in young (6 week-old; Ap and As; n = 4), in sham (6-month-old; Bp and Bs; n = 5), and in PMI rats (Cp and Cs; n = 8).*P < .05. Note the decrease in peak outward current in sham compared with young animals and further decrease in PMI rats.

Time Course and Dose-Dependent Dronedarone Inhibition of I_K. The application of 1 µM dronedarone induced a rapid exponential decrease in I_K amplitude elicited during repetitive +50-mV depolarizing pulses every 10 s (Fig. 4). The reduction in I_K further progressed with drug application such that I_K reached a null amplitude while Ipeak and I_K1 could still be observed; however, this situation was soon associated with gigaseal disruption (n = 4; data not shown). The application of dronedarone during interruption of the repetitive depolarizing protocol had the same inhibitory effect on I_K (Fig. 4B). In both patch-clamp conditions, the time for half-inhibition of I_K in the presence of the antiarrhythmic agent was ~4 min (220.4 ± 0.4 s; n = 12). This indicates that dronedarone induces only a tonic-block with no use-dependent effects. Figure 4 also shows that I_K recovery was very slow with <50% recovery after a 12-min washout (667 ± 153 s; n = 4). I_K recovery was also independent of the stimulation protocol. It had been verified that under control conditions, I_K was stable with <10% amplitude variation within 1 h.

View larger version (13K): [in this window] [in a new window]   Fig. 4.   Comparison of the time courses of the reduction in the delayed outward current, IK, induced by 1 µM dronedarone and washout when the cell was (A) or was not (B) stimulated by a +50-mV depolarizing pulse every 4 s.

View larger version (14K): [in this window] [in a new window]   Fig. 5.   Dose-dependent inhibitory effects of dronedarone on the delayed outward current elicited by a +50-mV depolarizing pulse in sham (; n = 10) and PMI (; n = 10) rat cells. Dronedarone effects were estimated after a 6-min application; at most, two concentrations were tested on a given cell. Curves are fitted according to a sigmoid shape: y = ECmax [dronedarone]nH/([dronedarone] + EC50nH) and normalized to ECmax, the concentration giving the maximal value. Number of cells in brackets.

Dronedarone Effects on Transient Outward Potassium Current, I_to. In the following, the effects of dronedarone were compared on the fast transient outward current, I_to, estimated as the difference between Ipeak and I_K in both sham and PMI rat cells. It was verified that this method gave similar I_to than the 4-AP sensitive K⁺ current. Figure 6A shows typical outward currents elicited during a test pulse at +50 mV after a 6-min dronedarone application in a sham cell. Associated with the marked reduction in I_K after 6 min in the presence of dronadarone, I_to appeared to be slightly increased. At a +50-mV depolarization, I_to densities on sham cells (n = 6) were 12.5 ± 2.7 pA/pF and 13.8 ± 2.1 pA/pF (+10.3%; NS) in control conditions and after a 6-min dronedarone application, respectively. These effects were more marked in PMI animals (n = 8) in which I_to was reduced. Dronedarone (1 µM) increased I_to from 9.4 ± 1.9 pA/pF in control conditions to 11.3 ± 2 pA/pF (+20.2%; P < .05) after a 6-min dronedarone application. In the two types of animals, during prolonged dronedarone application I_to decreased back to its control level or below (data not shown).

View larger version (24K): [in this window] [in a new window]   Fig. 6.   Dronedarone-induced alterations in the kinetics of the transient outward current, Ito, elicited in sham cells. Ito was estimated as the difference between the peak and the late outward current elicited by a +50-mV depolarizing pulse in control conditions () or after 6 min in the presence of 1 µM dronedarone (). A, current recordings elicited by a depolarization to +50 mV applied on a sham cell in control conditions and after a 6-min dronedarone application. B, voltage-dependent activation of Ito. The relations are fitted by a Boltzmann equation with I = Imax/(1 + exp [(V  V1/2)/k]), in which V1/2 and k were 4.76 ± 0.69 and 9.28 ± 0.62 mV, respectively (n = 7). C, voltage-dependent inactivation of Ito. The relations are fitted by a Boltzmann equation in which V1/2 and k were, respectively 39.53 ± 0.25 and 5.51 ± 0.22 mV (n = 7). D, recovery from inactivation of Ito elicited by 200-ms, +50-mV depolarizing pulses applied at various, aleatory intervals after the same conditioning pulse. The curve is fitted by a single exponential in control conditions, whereas two exponentials are required after dronedarone application (sham; n = 6).

Holter Monitoring Study in Untethered Animals. The ability of the compound to demonstrate in vivo effects was assessed by Holter monitoring with telemetry. We first investigated whether dronedarone exhibited proarrhythmic effects on sham rats (n = 6). VPB/h were 0.6 ± 0.3, 0.3 ± 0.3, and 0.6 ± 0.3 before, during treatment with 30 mg/kg/day (days 4-7), and after wash-out phases (days 15-22). The arrhythmic score was statistically similar for the three phases that indicates dronedarone had no proarrhythmic effect. Similarly, no arrhythmic effect could be detected during treatment with 90 mg/kg/day (n = 6). However after 4 days of treatment, animals were clearly sick and dyspneic; these toxic effects led to stop this protocol. Complete recovery occurred soon after stopping drug administration.

View larger version (22K): [in this window] [in a new window]   Fig. 7.   Effects of dronedarone (40 mg/kg/day) on electrical parameters in eight PMI rats. Recordings were performed in untethered animals by Holter monitoring during three consecutive hours at baseline (two recordings), during, and after oral administration of the drug (days 4 and 7 during treatment, and then days 15, 22, and 30). A, heart rate (beats/min). B, VPB/h, counts per hour. C, severity score is expressed by the number of rats demonstrating complex forms (VT and doublets) of ventricular arrhyhtmias with only the most severe form in each animal being considered.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The purpose of this study was to evaluate the antiarrhythmic effects of dronedarone in control rats and in rats after postinfarct remodeling, a pathophysiological model known to develop anomalous electrical activities. The most important findings of this study were that this new amiodarone-like compound significantly decreased I_K by a use-independent, tonic-block and markedly prolonged recovery from inactivation of I_to. Both findings are consistent with the observed AP prolongation in normal hearts. However, there was no significant modification of the AP time course in PMI rats that already exhibited prolonged APs. These effects on ionic currents might be the basis for the increase in the effective refractory period and the antiarrhythmic effects demonstrated during the telemetry Holter monitoring.

The chronic infarcted rat heart as a model of left ventricular dysfunction is clinically relevant and has predicted results of pathophysiological and pharmacological studies in humans (Pfeffer et al., 1979; Hasenfuss, 1998). In this experimental model of HF and chronic arrhythmias, antiarrhythmic drug activity is now easily technically accessible due to the availability of in vivo monitoring of untethered animals (Carré et al., 1992; Chevalier et al., 1995; Opitz et al., 1995). Our study was performed in the same model several weeks after occlusion, during the chronic phase of ventricular remodeling. In spite of the relatively small size of the group involved in the Holter monitoring study, dronedarone is shown to have pharmacological effects on ventricular arrhythmias, particularly to decrease VPB/h and to a lesser extent, the severity of recorded arrhythmias. Recovery was completed only 2 weeks after the last oral administration that suggests, at first, a prolonged effect of the drug. A sustained direct effect of the drug is not consistent with other studies designed to evaluate the pharmacokinetics of the drug. Another view would be that preventing arrhythmias during a week or so has a beneficial effect on remodeling that prolonged over a few weeks.

The antiarrhythmic effects of dronedarone have already been compared with those of amiodarone and other class III antiarrhythmic agents such as d-sotalol in different species and models. In anesthetized dogs, dronedarone and amiodarone share similar electrophysiological actions (Manning et al., 1995b). Moreover, dronedarone, like amiodarone, reduces ischemia-induced ventricular arrhythmias in anesthetized pigs (Finance et al., 1995), and ischemia- and reperfusion-induced ventricular arrhythmias in anesthetized rats (Manning et al., 1995a). In the latter report, dronedarone administrated either i.v. or orally exhibits a profile similar to that of amiodarone but is 3 to 10 times more active to reduce ischemia- and reperfusion-induced arrhythmias. In these studies, significant decreases in induced ventricular fibrillation, VTs, and VPBs were noted. Increases in AP duration and effective refractory period were suggested by an increase in the QTc interval (time elapsed between waves Q and T of the ECG and corrected for frequency). The antiarrhythmic effect was attributed to the increase in refractoriness and the prevention of reentrant excitation, a well known mechanism for the induction of malignant ventricular arrhythmias occurring during myocardial ischemia. At the cellular level, dronedarone, like amiodarone, has a many-sided pharmacological profile. Amiodarone is known to alter Na⁺ (Follmer et al., 1987; Kolhardt and Fichtner, 1988; Honjo et al., 1991) and Ca²⁺ currents (Nishimura et al., 1989; Valenzuela and Bennett, 1991; Varro et al., 1996) and to partially inhibit the effects of sympathetic activation (Chatelain et al., 1995; Hodeige et al., 1995). Dronedarone demonstrates similar effects on both inward currents (Gautier et al., 1997) and on the sympathetic system although without triggering a hypothyroid-like state.

Besides these multiple effects, class III antiarrhythmic agents are most known to block outward potassium channels that leads to prolongation of repolarization phase and refractory period. In isolated adult rat ventricular myocytes, outward potassium currents have been well-studied (Apkon and Nerbonne, 1991; Barry and Nerbonne, 1996). In the present study, dronedarone application had specific effects on the different K⁺ currents. First, it seems that dronedarone induces no or a small and nonsignificant decrease in I_K1, whereas at 10 to 20 µM amiodarone was reported to induce a relatively small (14%) but statistically significant reduction in I_K1 (Sato et al., 1994). Such an effect was, however, not seen during acute application of amiodarone at 5 µM (Varro et al., 1996). Indeed, I_K1 is responsible for the cell resting potential; we did not obtain any variation in the resting potential of rat papillary muscle studied until applying 30 µM dronedarone. The most obvious effect of dronedarone occurred on I_K. The dronedarone-sensitive current had a relatively slow activation and a very slow inactivation phase. It activated about 40 mV and was inhibited by tetraethylammonium. These characteristics strongly suggest the dronedarone-sensitive current is the I_K previously described in rat cells by Apkon and Nerbonne (1991). The only report describing the effects of amiodarone on I_K in rat cells is by Guo et al. (1997) who showed a 30% inhibition of I_K after a 20-min application of 10 µM amiodarone in cultured newborn ventricular myocytes. Such an inhibition of I_K by amiodarone has been previously described in other species. Amiodarone at 10 µM reduces the La³⁺-resistant delayed rectifier, I_K by ~50% in isolated guinea pig ventricular myocytes (Balser et al., 1991) and exhibits a dose-dependent inhibition in the range between 1 and 10 µM in rabbit ventricular myocytes (Kamiya et al., 1995; Varro et al., 1996). Amiodarone was initially reported to have little effect on the La³⁺-sensitive current (Balser et al., 1991) that includes the noninactivating component I_Kr (Sanguinetti and Jurkiewicz, 1990); however, it also was recently demonstrated that amiodarone induces a decrease in I_Kr in rabbit ventricular cells (Carmeliet, 1993).

It has been suggested that dronedarone, like amiodarone, acts in the lipid bilayer as a membrane-active drug (Chatelain et al., 1985; Trumbore et al., 1988). An extensive study of dronedarone-induced I_K channel block has not been performed in the present work; however, recovery from block by washout of dronedarone was very slow as previously described for amiodarone (Herbette et al., 1988; Carmeliet, 1993) in line with the above-mentioned reports. The major difference between amiodarone and dronedarone results from the observation that dronedarone induced only a tonic block with no use-dependent effects on I_K. That suggests dronedarone acts by block of the channel in the rested state, or it exerts a very fast block of the open channel. These results differ from those obtained with amiodarone that showed mostly a use-dependent block on I_Kr (Carmeliet, 1993).

The fast transient outward current I_to was reported to be unaffected by amiodarone application (up to 10 µM) in single rabbit ventricular myocytes (Kamiya et al., 1995; Varro et al., 1996), as well as in cultured newborn rat cardiomyocytes (Guo et al., 1997). However, in the latter study, amiodarone at higher concentrations (10-30 µM) induced a dose-dependent inhibition of I_to that occurred after approximately a 20-min application. In our experimental conditions at a stimulation frequency of 0.25 s¹, only prolonged application of 1 µM dronedarone decreased I_to following a weak increase observed after 6 min of drug application. However, the most salient effect of dronedarone was on the I_to reactivation curve. After a 6-min application of 1 µM dronedarone, I_to reactivation was markedly slowed. The 7-mV rightward shift of I_to activation curve accounting for a delayed activation together with the slower recovery from inactivation are very important parameters to enhance AP duration as well as the refractory period. In addition, during arrhythmias, the dronedarone-induced slowing of I_to reactivation would further contribute to reduce this current, prolong the AP, and thus antagonize the anomalous activities. Indeed, I_to is responsible of the first phase of repolarization in rat AP (Barry and Nerbonne, 1996). A reduced I_to activation after dronedarone application is a reliable explanation for the increase in AP duration. This K⁺ current is mostly attributable to Kv4.2 and Kv4.3 -subunits that are found in all animal species examined to date (Nerbonne, 1998). A rather similar effect of dronedarone on AP duration could thus be expected in other species. However I_K, the delayed outward current that is also markedly altered by dronedarone, shows distinct kinetics and voltage-dependent properties according to species; a fact that should be born in mind before extrapolating our results to humans.

In conclusion, the inhibition by dronedarone of K⁺ currents that are all-important in the late phase of repolarization of APs is thus expected to play a major role in AP lenghtening and cardiac arrhythmia prevention. In this study, dronedarone effects are only seen on AP duration recorded in normal rats, whereas AP duration in PMI rats seems unaffected at the low experimental stimulation frequency. However, dronedarone exhibits significant in vivo antiarrhythmic effects as demonstrated by the decrease in VPB/h recorded in unthetered PMI rats. These effects are directly attributable to an increase in the effective refractory period that results from a tonic block of I_K and a marked slowing of I_to reactivation.