Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Delay of myocardial repolarization (class III electrophysiologic activity) via blockade of repolarizing potassium currents has been advanced as a potential antiarrhythmic mechanism. This is based on the premise that sufficient prolongation of myocardial refractoriness results in a wavelength of excitation that exceeds the path length of reentrant circuits, thereby preventing the initiation and/or maintenance of reentrant rhythms (Wellens et al., 1984). Myocardial repolarization in the majority of mammalian species studied, including humans, is controlled mainly by the interplay of the rapidly (I_Kr) and slowly (I_Ks) activating, delayed rectifier potassium currents (Sanguinetti and Jurkiewicz, 1990; Wang et al., 1994; Liu and Antzelevitch, 1995; Li et al., 1996; Salata et al., 1996a; Virag et al., 2001). The clinical assessment of selective blockers of I_Kr for the treatment of malignant ventricular arrhythmia has yielded disappointing results. d-Sotalol increased mortality in patients with previous myocardial infarction and left ventricular dysfunction (Waldo et al., 1996), and dofetilide displayed a neutral effect on mortality in patients with reduced left ventricular function and congestive heart failure (Torp-Pedersen et al., 1999). Characteristics of I_Kr blockers, which have been proposed to limit clinical antiarrhythmic efficacy, include reverse frequency dependence, whereby class III activity is diminished at faster heart rates and exaggerated at slower rates (Nattel and Zeng, 1984; Hondeghem and Snyders, 1990), and reduction of class III activity in the setting of sympathetic stimulation (Sanguinetti et al., 1991; Schreieck et al., 1997).

A number of structurally distinct selective I_Ks blockers recently have become available for preclinical assessment. Initial studies indicate that selective I_Ks block may provide a profile of class III action differing significantly from that of I_Kr block, particularly with regard to frequency dependence and activity during sympathetic stimulation, which may impart improved antiarrhythmic efficacy (Gerlach, 2001). To this point, the benzodiazepine I_Ks blocker L-768673 (Fig. 1) has been shown to prevent the development of malignant ventricular arrhythmia in anesthetized dogs with acute thrombotic coronary ischemia superimposed upon a recent myocardial infarction, and in conscious dogs with acute coronary ischemia and exercise superimposed upon a healed myocardial infarction (Lynch et al., 1999). Antiarrhythmic efficacy in the latter model was noteworthy because this conscious preparation possesses high sympathetic tone, and the I_Kr blocker d-sotalol was ineffective in this model (Vanoli et al., 1995).

View larger version (16K): [in this window] [in a new window]   Fig. 1.   Chemical structures of L-768673 and L-735821.

-Adrenergic receptor blockade also has demonstrated antiarrhythmic efficacy in preclinical animal models (e.g., Gang et al., 1984; Euler and Scanlon, 1988). In vitro studies have reported that -adrenergic stimulation increases the magnitude of I_Ks current (Sanguinetti et al., 1991; Han et al., 2001) and increases the effect of I_Ks blockers (Schreieck et al., 1997; Han et al., 2001), and that concomitant I_Ks block and -adrenergic stimulation results in exaggerated, inhomogeneous, and potentially proarrhythmic effects (Schreieck et al., 1997; Burashnikov and Antzelevitch, 2000; Shimizu and Antzelevitch, 2000). We therefore hypothesized that concomitant low-dose I_Ks and -adrenergic receptor blockade might provide enhanced and potentially safer antiarrhythmic activity than either mechanism alone. We tested this hypothesis by extending our previous antiarrhythmic assessment of the I_Ks blocker L-768673 in dogs with recent myocardial infarction (Lynch et al., 1999). The present study assesses the cardiac electrophysiologic and antiarrhythmic actions of a subeffective dose of L-768673 and a subeffective, minimally -adrenergic blocking dose of timolol, administered individually or in combination. The results of this study demonstrate significantly greater suppression of malignant arrhythmias with combination low-dose I_Ks and -adrenergic blockade.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

All procedures related to the use of animals in these studies were reviewed and approved by the Institutional Animal Care and Use Committee at Merck Research Laboratories at West Point and conform with the Guide for the Care and Use of Laboratory Animals (Institute of Laboratory Animal Resources, Commission on Life Sciences, National Research Council, 1996).

Determination of -Adrenergic Blocking Dose of Timolol in Dogs with Recent Anterior Myocardial Infarction. A separate group of dogs (7.5-11.0 kg, 9.6 ± 0.2 kg; total n = 20) was used to determine a minimally -adrenergic blocking dose of timolol for use in subsequent antiarrhythmic studies. These animals possessed anterior myocardial infarctions produced surgically as described below, 10.1 ± 0.8 days prior to study. The postinfarction animals were studied during anesthesia with -chloralose (80.0-100.0 mg/kg i.v.) and minimal sodium pentobarbital as needed (5 mg/kg i.v.), and ventilation using a cuffed endotracheal tube and a volume-cycled respirator. -Adrenergic activity was assessed by monitoring heart rate increases in response to bolus intravenous doses of 0.01 to 3.0 µg/kg isoproterenol at 15, 75, 135, and 195 min after the following treatments: no timolol (n = 5), 1.0 µg/kg i.v. timolol (n = 4), 10.0 µg/kg i.v. timolol (n = 6), and 100.0 µg/kg i.v. timolol (n = 5). The 15- to 195-min time frame for assessment of -adrenergic blocking activity of timolol encompassed the time frame for response of postinfarction dogs to the development of acute, thrombotically induced posterolateral myocardial ischemia in the subsequent antiarrhythmic studies described below.

Canine Model of Recent Anterior Myocardial Infarction. The anesthetized canine model of recent anterior myocardial infarction in which ventricular tachyarrhythmias may be induced by programmed ventricular stimulation (PVS) and in which malignant ventricular arrhythmias develop in response to acute coronary artery thrombosis was used to assess the antiarrhythmic efficacy of low-dose I_Ks and -adrenergic receptor blockade. This model was used previously to characterize the dose-dependent cardiac electrophysiologic and antiarrhythmic actions of the I_Ks blocker L-768673 (Lynch et al., 1999). The present study extends this initial evaluation through a comparison of the following four treatment groups: 1) microemulsion vehicle (n = 10), 2) a subeffective 0.3 µg/kg i.v. dose of L-768673 (n = 10), 3) a minimally -adrenergic blocking dose of 1.0 µg/kg timolol (n = 8) determined in studies described above, and 4) coadministration of the subeffective 0.3 µg/kg i.v. L-768673 + the 1.0 µg/kg timolol doses (n = 10). These four treatment groups were studied concurrently and in randomized fashion.

Surgical Preparation. Male or female purpose-bred mongrel dogs (7.2-11.4 kg; 9.1 ± 0.1 kg; total n = 38) were preanesthetized with sodium thiamylal (5.0 mg/kg i.v.), and general anesthesia was induced with isoflurane. A left thoracotomy was performed in the fourth intercostal space, the pericardium was incised, and the heart was suspended in a pericardial cradle. Anterior myocardial infarction was produced by a 2-h occlusion of the left anterior descending coronary artery followed by reperfusion. Surgical incisions were closed, and the animals were allowed to recover.

Electrophysiologic Testing, Programmed Ventricular Stimulation, and Acute Posterolateral Myocardial Ischemia. Animals were studied at 8.0 ± 0.4 days after anterior myocardial infarction. Postinfarction dogs were anesthetized with -chloralose (80.0-100.0 mg/kg i.v.) and minimal sodium pentobarbital as needed (5 mg/kg i.v.) and were ventilated by means of a cuffed endotracheal tube and a volume-cycled respirator. Systemic arterial pressure was monitored via the cannulated left common carotid artery, and the right femoral vein was isolated and cannulated for test compound administration. The heart was re-exposed via a left thoracotomy and suspended in a pericardial cradle. A surface bipolar electrode was sutured to the left atrial appendage for atrial pacing, and a bipolar plunge electrode was inserted into the interventricular septum near the right ventricular outflow tract (RVOT) adjacent to the site of left anterior descending coronary artery occlusion for the introduction of ventricular extrastimuli during PVS. One bipolar plunge electrode per zone was sutured into the infarcted anterior region of the left ventricle distal to the site of coronary artery occlusion and within the area of myocardial scarring as ascertained visually and by palpation (infarct zone, IZ) and into the noninfarcted posterolateral region of the left ventricle (noninfarct zone, NZ) for the measurement of ventricular excitation thresholds and refractory periods. Lead II electrocardiogram was monitored continuously.

1/2

Statistical Analysis. Data are expressed as mean ± S.E.M. Within a given treatment group, pre- versus post-treatment comparisons were made using a two-tailed paired Student's t test. Comparisons among treatment groups were made using a single-factor ANOVA followed by a Fisher's protected least significant difference post hoc test or a Fisher's exact test, as appropriate.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Determination of -Adrenergic Receptor Blocking Dose of Timolol in Dogs with Recent Anterior Myocardial Infarction. Figure 2, A to D, summarizes the heart rate responses of postinfarction dogs to bolus i.v. challenges of isoproterenol at 15, 75, 135, and 195 min after the i.v. administration of 1.0, 10.0, and 100.0 µg/kg timolol. The 10.0 and 100.0 µg/kg i.v. timolol doses produced clear rightward shifts in the isoproterenol dose-response curve throughout the 195-min study period. The 1.0 µg/kg i.v. timolol dose produced only a slight rightward shift at the lower doses of isoproterenol at the 15- and 75-min time points and, therefore, was considered a minimally -adrenergic blocking dose. Underlying anterior myocardial infarct size for the animals used in this assessment was 15.1 ± 2.2% of left ventricle and did not vary significantly among treatment groups.

View larger version (23K): [in this window] [in a new window]   Fig. 2.   Heart rate responses to exogenous bolus intravenous administration of 0.01 to 3.0 µg/kg isoproterenol at 15 min (A), 75 min (B), 135 min (C), and 195 min (D) after no timolol () or after 1.0 (), 10.0 (), or 100.0 () µg/kg i.v. timolol. Baseline heart rates ranged from 111 ± 3 to 133 ± 2 in these four treatment groups; change in heart rate with isoproterenol is expressed as percent change from each baseline heart rate. Data are mean ± S.E.M.

Cardiac Electrophysiologic Effects in Dogs with Recent Anterior Myocardial Infarction. Table 1 summarizes the effects of 0.3 µg/kg i.v. L-768673, 1.0 µg/kg i.v. timolol and the coadministration of 0.3 µg/kg L-768673 + 1.0 µg/kg i.v. timolol on heart rate, mean arterial pressure, electrocardiographic intervals, and cardiac electrophysiologic parameters in dogs with recent anterior myocardial infarction. L-768673 at 0.3 µg/kg i.v. elicited a slight decrease in sinus heart rate (5.8 ± 4.7%), modest but significant increases in QTc interval (3.6 ± 1.2%; p < 0.05) and in NZ VRRP and VERP (9.7 ± 2.0 and 10.2 ± 2.7%; both p < 0.01), and lesser increases in IZ VRRP and VERP (4.0 ± 2.3 and 7.4 ± 4.6%). Timolol at 1.0 µg/kg i.v. had no effect on sinus heart rate or QTc interval but modestly increased NZ VRRP and VERP (6.0 ± 2.2 and 7.3 ± 2.4%; both p < 0.05) and IZ VERP (7.0 ± 2.7%; p < 0.05). Coadministration of 0.3 µg/kg i.v. L-768673 + 1.0 µg/kg i.v. timolol elicited effects generally similar to those of L-768673 alone. Sinus heart rate was decreased slightly (4.5 ± 1.5%; p < 0.05), QTc and paced QT intervals were increased modestly but consistently (4.5 ± 1.2% and 5.5 ± 1.4%; both p < 0.01), NZ VRRP and VERP were increased (7.0 ± 1.7 and 12.3 ± 3.0%; both p < 0.01), and IZ VRRP and VERP also were increased to a lesser extent (5.3 ± 1.6 and 5.8 ± 1.4%; both p < 0.01).

Antiarrhythmic Activity in Dogs with Recent Anterior Myocardial Infarction. Table 2 compares the effects of the microemulsion vehicle versus 0.3 µg/kg i.v. L-768673, 1.0 µg/kg i.v. timolol, or the coadministration of 0.3 µg/kg i.v. L-768673 + 1.0 µg/kg i.v. timolol on the induction of VT by PVS, as well as on the incidence of lethal arrhythmias developing in response to acute, thrombotically induced posterolateral myocardial ischemia in dogs with recent anterior myocardial infarction. By entry criterion, all preparations in all treatment groups responded to pretreatment baseline PVS with sustained VT or VT/VF. Significant suppression of PVS-induced tachyarrhythmias was achieved only with the coadministration of 0.3 µg/kg i.v. L-768673 + 1.0 µg/kg i.v. timolol (8/10, 80% rendered noninducible versus 1/10, 10% noninducible in vehicle group; p < 0.01).

View larger version (26K): [in this window] [in a new window]   Fig. 3.   Survival of 0.3 µg/kg i.v. L-768673, 1.0 µg/kg i.v. timolol alone (n = 8), or concomitant 0.3 µg/kg i.v. L-768673 + 1.0 µg/kg i.v. timolol (n = 10) versus microemulsion vehicle-treated (n = 10) dogs with previous anterior myocardial infarction expressed as a function of time after onset of thrombotically induced acute posterolateral myocardial ischemia.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

To date, two structurally distinct classes of selective I_Ks blockers have been identified. These are the chromanols exemplified by compound 293b and HMR1556 (Busch et al., 1996; Gogelein et al., 2000), and the benzodiazepines L-735821 and L-768673 (Fig. 1) (Salata et al., 1996b; Selnick et al., 1997). Both classes have been used to characterize the physiologic consequences of cardiac I_Ks blockade. In guinea pig and human ventricular myocytes, chromanol 293b prolonged action potential duration (APD) in a frequency-independent manner (Bosch et al., 1998), whereas small but frequency-independent APD prolongations were elicited in guinea pig, rabbit, and dog papillary muscles (Schreieck et al., 1997; Varro et al., 2000; Lengyel et al., 2001) and in dog Purkinje fiber (Varro et al., 2000, Han et al., 2001). The magnitude of APD prolongation with chromanol 293b was increased in guinea pig papillary muscle and dog Purkinje fiber in the presence of isoproterenol (Schreieck et al., 1997; Han et al., 2001). HMR1556 prolonged APD in vitro in guinea pig papillary muscle in a frequency-independent manner, again with APD prolongation increased in the presence of isoproterenol (Gogelein et al., 2000). In dog left ventricular epicardial, mid-myocardial, and endocardial tissues, exposure to chromanol 293b produced a homogeneous frequency-independent prolongation in APD (Burashnikov and Antzelevitch, 2000). Similarly, chromanol 293b produced a homogeneous prolongation of APD in endo-, mid-, and epicardium in an arterially perfused dog left ventricular wedge preparation, with concomitant QT interval prolongation but no change in transmural dispersion of repolarization (Shimizu and Antzelevitch, 2000). However, concomitant -adrenergic stimulation with isoproterenol in the setting of I_Ks block with chromanol 293b produced nonhomogeneous APD prolongations in both isolated myocytes and the wedge preparation, resulting in an increased transmural dispersion of APD and a markedly prolonged QT interval (Burashnikov and Antzelevitch, 2000; Shimizu and Antzelevitch, 2000). In guinea pig papillary muscle, isolated canine ventricular myocytes and in the ventricular wedge preparation, I_Ks block with chromanol 293b produced moderate homogeneous effects with no provocation of arrhythmogenic afterdepolarizations, whereas -adrenergic stimulation with isoproterenol combined with I_Ks block with chromanol 293b evoked afterdepolarizations (Schreieck et al., 1997; Burashnikov and Antzelevitch, 2000; Shimizu and Antzelevitch, 2000).

The prototype benzodiazepine I_Ks blocker L-735821 produced frequency-independent increases in APD in guinea ventricular myocytes (Salata et al., 1996b) as well as a marked prolongation of APD in rabbit Purkinje cells (Cordeiro et al., 1998). Like chromanol 293b, L-735821 produced small prolongations of APD in rabbit and dog papillary muscle (Varro et al., 2000; Lengyel et al., 2001) and dog Purkinje fiber (Varro et al., 2000). L-768673, a benzodiazepine with optimized pharmacokinetic properties, produced a self-limiting prolongation of APD in vitro in guinea pig ventricular myocytes (Selnick et al., 1997). In rabbit ventricular myocytes obtained from normal control rabbits versus rabbits with renovascular hypertension and left ventricular hypertrophy characterized by abnormally prolonged repolarization, L-768673 produced modest prolongation of APD, with APD prolongation greater in normal control rabbit myocytes than in left ventricular hypertrophy rabbit myocytes (Xu et al., 2001).

Insufficient studies have been conducted to determine similarities or differences in pharmacologic profile between the chromanol and benzodiazepine classes of I_Ks blockers. Taken together, however, the studies summarized above suggest I_Ks blockade to produce homogeneous, frequency-independent or forward frequency-dependent increases in myocardial APD and refractoriness, with the potential for increased and possibly inhomogeneous and exaggerated activity in the setting of direct -adrenergic stimulation. This profile for I_Ks block differs from that of I_Kr block, which is characterized by reverse frequency-dependent prolongations of cardiac APD and refractoriness, whereby activity is reduced at a faster rate (Nattel and Zeng, 1984; Hondeghem and Snyders, 1990), and a diminution of activity in the setting of -adrenergic stimulation (Sanguinetti et al., 1991; Schreieck et al., 1997).

Fewer studies have assessed the cardiac electrophysiologic effects of selective I_Ks blockade in vivo. In anesthetized noninfarcted dogs, intravenous chromanol 293b produced a homogeneous forward frequency-dependent increase in ventricular refractory periods measured at the endo-, mid-, and epicardial levels (Bauer et al., 1999). In anesthetized dogs 3 to 5 days after anterior myocardial infarction, intravenous chromanol 293b produced forward frequency-dependent, transmurally homogeneous increases in ventricular refractoriness, with greater increases in refractoriness observed in the infarct zone compared with the noninfarct zone (Bauer et al., 2000). The antiarrhythmic potential of I_Ks blockade has been addressed through the study of L-768673 in two canine models of myocardial ischemic injury: 1) chloralose-anesthetized dogs in which malignant ventricular arrhythmias develop in response to acute thrombotic coronary ischemia superimposed upon a recent myocardial infarction, and 2) conscious dogs in which malignant ventricular arrhythmias develop in response to acute mechanical coronary ischemia and exercise superimposed upon a healed myocardial infarction (Lynch et al., 1999). Intravenous L-768673 produced modest increases in noninfarct zone and infarct zone refractory periods and ECG QTc interval, and prevented the development of malignant arrhythmia in both preparations. In the conscious healed myocardial infarction model, the increase in QTc with L-768673 was preserved but not increased during exercise. The efficacy of L-768673 in the latter preparation was noteworthy in that this preparation possesses high sympathetic tone induced by exercise and acute coronary artery occlusion, and the I_Kr blocker d-sotalol was ineffective in this preparation (Vanoli et al., 1995; Schwartz, 1998).

The present study extends the previous assessment of the antiarrhythmic potential of I_Ks blockade by demonstrating significant suppression of ventricular arrhythmias in dogs with recent myocardial infarction with the combined administration of a subeffective dose of L-768673 and a subeffective, minimally -blocking dose of timolol. A low-dose of L-768673 (0.3 µg/kg i.v.) administered alone failed to reduce the incidence of PVS- or thrombotically induced ischemic arrhythmias. Timolol (1.0 µg/kg i.v.), at a dose lower than the 100.0 µg/kg to 1.0 mg/kg i.v. dose range reported to be effective in canine arrhythmia models (Gang et al., 1984; Euler and Scanlon, 1988), produced only slight inhibition of isoproterenol-induced chronotropic effects and failed to prevent the development of ventricular arrhythmia when administered alone. Concomitant low-dose I_Ks and -blockade, however, afforded significant protection against the development of malignant ventricular arrhythmia in concert with modest increases in ventricular refractoriness and ECG QTc interval.

The mechanism underlying improved antiarrhythmic efficacy with combined I_Ks and -adrenergic receptor blockade may be multifactorial. -Adrenergic stimulation with isoproterenol has been reported to increase the magnitude of I_Ks current in guinea pig ventricular myocytes and in dog Purkinje cells (Sanguinetti et al., 1991; Han et al., 2001), and, as summarized above, concomitant -adrenergic stimulation with isoproterenol in the setting of I_Ks block is reported to produce inhomogeneous, exaggerated, and potentially arrhythmogenic effects (Schreieck et al., 1997; Burashnikov and Antzelevitch, 2000; Shimizu and Antzelevitch, 2000). Therefore, low-dose -blockade might produce a salutary modulation and prevent exaggeration of the effect of I_Ks block in the setting of sympathetic hyperactivity. Alternatively, elevation in sympathetic tone is an established and significant trigger for malignant arrhythmias (Meredith et al., 1991), and -adrenergic blocking agents are known to reduce the incidence of postinfarction mortality, presumably in part due to reduction in malignant arrhythmia and sudden death (Rehnqvist, 1990). Therefore, the addition of -adrenergic blockade to a given antiarrhythmic therapy might be expected to result in an enhanced efficacy through an independent mechanism, e.g., attenuation of augmented L-type calcium current during sympathetic stimulation. Finally, the possibility cannot be excluded that improved efficacy with combined low-dose I_Ks and -adrenergic receptor blockade might result from a combined class III effect, with a previous study reporting that some -adrenergic receptor blockers possess intrinsic class III electrophysiologic activity (Taggart et al., 1984).

Additional insight regarding interactions between I_Ks and sympathetic stimulation and modulation thereof by -adrenergic blocking agents may be derived from clinical studies on congenital long-QT (LQT) syndromes. LQT syndromes are characterized by delayed ventricular repolarization, long ECG QT interval, and the high risk of ventricular tachyarrhythmias. LQT-1 results from mutation of the KVLQT1 (KCNQ1) gene encoding the  subunit of the I_Ks channel (Priori et al., 1999). The triggering of ventricular arrhythmia in LQT-1 patients occurs frequently in settings of high sympathetic activity, such as emotion and exercise (Schwartz et al., 2001). -Adrenergic receptor blockers have been considered the treatment of choice in LQT syndrome patients; recent clinical studies have reported varying success rates with beta blockers among different LQT genotypes, with beta blockade particularly effective in the management of LQT-1 patients (Moss et al., 2000; Schwartz et al., 2001; Vincent et al., 2001). These clinical observations coupled with the preclinical findings summarized above are consistent with an arrhythmogenic risk associated with genetically or pharmacologically reduced I_Ks in the setting of sympathetic hyperactivity and support modulation of I_Ks blockade by concomitant -adrenergic blockade in the setting of increased sympathetic tone.

Presently, patients with malignant ventricular arrhythmias are managed with implantable cardioverter-defibrillators, a reflection of the ability of these devices to efficiently and repeatedly terminate ongoing ventricular tachyarrhythmias and fibrillation. This also reflects the inadequacy of existing pharmacologic therapies to safely and effectively prevent the development of malignant ventricular arrhythmias. In the current treatment paradigm, the potential therapeutic role of antiarrhythmic agents lies in the reduced need or frequency of implantable cardioverter-defibrillators discharges. The present findings suggest that the concomitant administration of low-dose I_Ks blockade and -adrenergic blockade may constitute a potential pharmacologic strategy to prevent the development of malignant ischemic ventricular arrhythmias.