Journal Logo


Salutary Antiarrhythmic Effect of Combining a K Channel Blocker and a β-Blocker in a Canine Model of 7-Day-Old Myocardial Infarction

Takatsuki, Seiji; Mitamura, Hideo; Kanki, Hideaki; Sueyoshi, Koichiro; Ogawa, Satoshi

Author Information
Journal of Cardiovascular Pharmacology: June 2000 - Volume 35 - Issue 6 - p 914-918
  • Free


Antiarrhythmic drugs with a K channel-blocking action are characterized by their effect to prolong repolarization. These drugs have been considered to be effective in preventing arrhythmias caused by reentry, especially with a short excitable gap. However, the SWORD trial (1) demonstrated that a K channel blocker may exert a proarrhythmic effect and decrease survival of patients after myocardial infarction. In contrast, K channel blockers having a β-blocking activity or when combined with a β-blocker can reduce arrhythmic death (2,3) and improve survival of patients after myocardial infarction (4,5).

In normal myocardium, the effect of a K channel blocker can be antagonized by β-adrenergic stimulation (6). However, how the electrophysiologic characteristics of infarcted myocardium with an inhomogeneous milieu would be modulated by the interaction between a K channel blocker and a patient's β-adrenergic status remains to be validated. Especially in myocardial infarction, higher adrenergic tone may develop in association with impaired cardiac function. In this study, we therefore examined the electrophysiologic characteristics of infarcted myocardium and its modulation by E4031, a pure IKr blocker, under β-adrenergic stimulation and blockade, and aimed to verify the significance and usefulness of the combination therapy of IKr blockade and β-blockade in this pathologic setting.


Surgical preparation

Ten adult mongrel dogs of either sex (weight, 9.8-14.2 kg) were anesthetized with sodium pentobarbital (30 mg/kg, i.v.), intubated with a cuffed endotracheal tube, and ventilated with room air by a constant-volume ventilator (model 80V; Harvard Apparatus, South Natick, MA, U.S.A.). With sterile technique, a thoracotomy was performed in the fifth left intercostal space, the ribs and lungs retracted, and the pericardium incised. The left anterior descending artery was isolated and tied just distal to the first diagonal branch with silk ligature. Five minutes before the ligation, methylprednisolone (30 mg/kg) was administered intravenously, because its use is reported to increase the likelihood of survival of the narrow rim of myocardium (7). Thirty minutes after the ligation, the chest was closed in layers, and negative pressure was reestablished in the pleural cavity. Antibiotics were administered for 5 days after surgery.

Seven days later, all dogs were subjected to open-chest electrophysiologic study. They were again anesthetized, intubated, and ventilated in the same manner as 7 days earlier. The right femoral artery was cannulated with a heparinized saline-filled polyethylene catheter to monitor the blood pressure continuously. The right femoral vein was cannulated with a polyethylene catheter to infuse saline solution (0.9%) and administer drugs during the study. A median sternotomy was performed and a pericardial cradle constructed. Body temperature was maintained at ∼37°C with a heating lamp, and lead II ECG also was monitored continuously.

Epicardial mapping

A polyethylene-coated stainless steel electrode was plunged into the epicardium of the right ventricular outflow tract (RVOT), serving for constant pacing during local electrogram recording and the induction of arrhythmias. A programmable stimulator (BC-02A; Fukuda Denshi Inc., Tokyo, Japan) was used to deliver rectangular cathodal current pulses of 2-ms duration at twice the diastolic threshold. A patch electrode (2.5 × 3 cm) equipped with 47 bipolar electrodes (distance between electrodes of 3 mm) was placed on the epicardial surface of the infarcted zone for recording the local bipolar and unipolar electrogram using a mapping system (HPM7100; Fukuda Denshi Inc., Tokyo, Japan). The bipolar electrogram was filtered from 30 to 400 Hz, and the unipolar electrogram was filtered from 0.05 to 100 Hz. A train of 30 beats of stimulation was applied to the RVOT electrode at a cycle length of 300 ms, and the 47 local electrograms, both bipolar and unipolar, were recorded at the last driven beat. The local QT interval was measured at each site, which was defined as the distance from the initial deflection in the bipolar electrogram to the end of the T wave in the unipolar electrogram (Fig. 1A). The coefficient of variation of QT intervals from 47 sites was calculated as standard deviation/mean × 100 and was adopted as an index of spatial dispersion of QT intervals.

FIG. 1
FIG. 1:
A: Measurement of the local QT interval on an epicardial mapping electrode. A large black triangle indicates the onset of local activation recorded on the bipolar electrogram (Bipolar Eg); a large white triangle indicates the end of QT on the unipolar electrogram (Unipolar Eg). A representative change of the local QT interval is measured between these triangles. B: QT interval change with drug interventions. The end of QT is indicated by a longitudinal transect.

Induction of ventricular arrhythmias

The protocol for the induction of ventricular tachyarrhythmias was as follows. After a train of eight beats of cathodal stimulation from RVOT at 300 ms, an extrastimulus was applied, with a coupling interval decreasing by 10 ms until the effective refractory period was reached. If sustained ventricular tachyarrhythmias were not induced, a second extrastimulus was added in a similar fashion. Ventricular tachyarrhythmias were defined as nonsustained (NSVT) if three or more repetitive ventricular responses were initiated but terminated spontaneously within 30 s, and sustained (SVT) if they persisted for 30 s or were accompanied by hemodynamic collapse. Ventricular responses of fewer than three nonstimulated complexes were classified as noninducible (NI).

Drug administration

These procedures described were repeated during the administration of E4031 (E), a pure IKr blocker, E with isoproterenol (E + Iso), and E with ONO1101 (E + βB), an ultrashort-acting β-blocker provided by ONO Pharmaceutical Co. Ltd. (Tokyo, Japan).

E4031 was provided by Eisai Inc. (Tokyo, Japan) and was administered in normal saline solution with a bolus injection of 50 μg/kg followed by the maintenance dose of 0.5 μg/kg/min. Iso and βB also were administered in normal saline solution with continuous injection of 35 ng/kg/min and 15 μg/kg/min, respectively, each with the maintenance dose of E.

Statistical analysis

Data are given as mean ± SEM, and the statistical comparisons were performed using analysis of variance (ANOVA). A value of p < 0.05 was considered statistically significant.


QT interval and QT dispersion

The representative QT changes in the unipolar electrogram with drug interventions are shown in Fig. 1B. The average QT interval obtained from values at 47 points on the mapping electrode in 10 dogs was 239.0 ± 4.6 ms at baseline and was significantly prolonged to 261.0 ± 6.1 ms after E administration (p < 0.05). E + Iso administration also showed a tendency to prolong the QT interval to 256 ± 10 ms from baseline (p = 0.07). E + βB administration prolonged the QT interval to 304 ± 6.3 ms, significantly more than either E or E + Iso administration (p = 0.0001 and <0.0001, respectively; Fig. 2).

FIG. 2
FIG. 2:
Average QT intervals during drug interventions. *p < 0.0001 vs. E + βB; †p < 0.05 vs. E.

QT dispersion, as defined by the coefficient of variation of 47 QT intervals, was 7.8 ± 0.6 at baseline (Fig. 3). It tended to decrease during E (6.9 ± 0.9), to increase during E + Iso (8.5 ± 0.7), and significantly decreased during E + βB administration (5.2 ± 0.6; p < 0.05). Importantly, this decrease of QT dispersion by E and E + βB was brought about by inhomogeneous prolongation of QT intervals in infarcted tissues. Local QT prolongation in the infarcted myocardium by E was greater at sites where the baseline QT interval was shorter than at sites where the baseline QT interval was longer. There was a negative correlation between the baseline QT interval and the degree of QT prolongation by E and E + βB (r = 0.611, p < 0.0001; and r = 0.584, p < 0.0001, respectively; Fig. 4) with the slope steeper in the absence of concomitant β-blockade (−0.776 in E vs. −0.425 in βB). Although E + Iso also showed a negative correlation between the baseline QT interval and the degree of QT prolongation (r = 0.463, p < 0.0001; Fig. 4), the degree of QT prolongation was more variable against baseline QT interval with E + Iso than with E or E + βB. Hence the correlation coefficient for E + Iso was smaller than that for E or E + βB, reflecting increased local QT dispersion by the addition of isoproterenol.

FIG. 3
FIG. 3:
QT dispersion during drug interventions. *p < 0.05 vs. E + βB.
FIG. 4
FIG. 4:
The relation between the baseline QT interval and the change of QT interval (ΔQT) by E, E + Iso, and E + βB at multiple epicardial electrodes in all the dogs. Negative correlations between the baseline QT interval and ΔQT for each intervention were all significant (p < 0.0001). With E, E + Iso, and E + βB, r was 0.611, 0.463, and 0.584, and the slope was −0.776, −0.946, and −0.425, respectively.

Induction of ventricular arrhythmias

Figure 5 shows the induction rate of ventricular arrhythmias in studied dogs. Although ventricular fibrillation (VF) was induced in 20% of the dogs at baseline, its induction rate decreased slightly to 10% with E administration, but increased to 50% when E was administered with Iso, none of which had this arrhythmia during E administration alone. On the contrary, the combination of E + βB consistently prevented VF induction.

FIG. 5
FIG. 5:
Inducibility of ventricular arrhythmias during drug interventions. The black, cross stripes and oblique stripes indicate inducibility of VF, sustained VT, and nonsustained VT, respectively. VF was induced in 50% of dogs during E + Iso administration. On the contrary, VF was never induced during E + βB administration.

Inducibility of VF in each dog is graphically displayed in Fig. 6 with regard to the average QT interval and QT dispersion. Black circles indicate dogs with inducible VF, and white circles represent those with noninducible VF. The average QT interval was not different between dogs with inducible VF and those without inducible VF. The probability of VF induction was 44% when the coefficient of variation of QT intervals (a marker for QT dispersion) was >9.0 (dashed line in Fig. 6) and 11% when it was <9.0 (p = 0.047; Fisher's Exact method). However, the difference of QT dispersion between dogs with VF and those with noninducible VF was not statistically significant.

FIG. 6
FIG. 6:
Prediction of ventricular fibrillation (VF) inducibility. Data under various drug were combined and plotted, showing a relation between the average QT interval and QT dispersion. The black circle indicates inducible VF and the blunt circle indicates noninducible VF. Sensitivity and specificity of QT dispersion for predicting VF induction are 44 and 89%, respectively, when a cut-off line of QT dispersion was set at 9.0 (p < 0.05, Fisher's Exact method).


In the canine 7-day-old infarcted myocardium, E4031 combined with a β-blocker prolonged the local QT interval more than E4031 alone, and significantly reduced local QT dispersion. This regimen also showed a salutary effect in preventing VF induction.

It has been reported that the efficacy of K channel blockers is counteracted by β-adrenergic stimulation. Isoproterenol antagonizes prolongation of the effective refractory period by E4031 (8), and prolongation of action potential duration by d-sotalol (9), both in isolated guinea pig myocytes. In human subjects, isoproterenol reversed the effect of prolonging the action potential duration by the K channel blocker, sematilide (10). These K channel blockers are known to suppress only the rapid component of IK (IKr), whereas β-adrenergic stimulation increases the slow component of IK (IKs), thus antagonizing the effect of the former (11).

Those studies were performed only in the normal myocardium and their clinical implications are limited. For the first time, we have shown that the effect of an IKr blocker also is reduced in the infarcted myocardium under β-adrenergic stimulation, an effect associated with decreased antiarrhythmic efficacy. Circulating catecholamines are elevated in patients with left ventricular dysfunction, and this could play a significant role in stimulating β-adrenergic receptors. Because these receptors may be upregulated in denervated infarcted myocardium, exaggerated β-adrenergic stimulation may occur as a result of the denervation supersensitivity (12). Therefore elevated circulating catecholamines associated with left ventricular dysfunction, and the temporal process of denervation and reinnervation as a function of the numbers of β receptors available, are both critical in determining the antiarrhythmic efficacy of a K channel blockers in the setting of subacute myocardial infarction.

In this study, E4031 reduced QT dispersion in the infarcted myocardium. Because each QT interval was considered to reflect repolarization of local areas of myocardium, QT dispersion would represent a dispersion of repolarization constituting an arrhythmogenic substrate forming reentrant tachyarrhythmias. In concordance with our result, Dhein et al. (13) reported that sotalol decreased the dispersion of the activation-recovery interval measured from unipolar electrodes in the perfused rabbit heart. Lubisky et al. (14), by measuring local VF intervals on epicardial mapping electrodes, which would represent local refractoriness, also showed that sotalol decreased dispersion of refractoriness in the canine myocardial infarction model.

K channel blockers have been shown to exert a differential effect on normal and infarcted myocardium (15,16). Interestingly, the effect of E4031 in prolonging the QT interval in the infarcted myocardium was not homogeneous. E4031 prolonged the QT interval more at sites with shorter baseline QT intervals. To our knowledge, this finding has not been reported previously, and could be the result of a decrease in the number or function of IKr channels in the central area of infarcted myocardium (17). The voltage dependence of K channels also may account for this finding. For example, Pinto and Boyden (18) showed that IK1 was decreased and E4031-sensitive currents were increased in subendocardial Purkinje myocytes of the canine 48-h infarcted heart.

QT dispersion of the epicardial border zone of infarcted myocardium can serve as a predictor of inducibility of VF, as shown in this study. Reduction of QT dispersion could be at least one of the antifibrillatory mechanisms of antiarrhythmic drugs. Because no fatal arrhythmias were inducible when E4031 was given with a β-blocker, this regimen could exert its most potent antiarrhythmic efficacy by synergism of both prolonging the local QT intervals and decreasing QT dispersion. It cannot exclude the possibility that the antiarrhythmic effect of E4031 with a β-blocker also can be reproduced by a β-blocker alone (19).


We adopted QT dispersion measured on the local unipolar electrogram as an arrhythmogenic index. These observations were recorded by a relatively small mapping electrode covering only the infarcted myocardium. Therefore QT dispersion in this study does not represent that in the whole heart. In addition, the unipolar electrogram can be sensitive to distant potentials or electrical noise. The resolution of QT intervals among adjacent electrodes may have been affected to some degree by these signals, although it has been demonstrated that there is a significant correlation between the local myocardial repolarization and the QT interval on a local unipolar electrogram (20,21). We should also be aware that repolarization does not always coincide with refractoriness in infarcted myocardium (postrepolarization refractoriness) (22,23). We did not measure the local refractory period directly, which should be more responsible for vulnerability to reentrant ventricular arrhythmias. It is thus possible that the QT interval or QT dispersion measured in the unipolar electrogram may not truly reflect local refractoriness or its dispersion.

Most important, the findings obtained in this animal model with 7-day-old myocardial infarction, open chest, and anesthesia may not be the same as those in the clinical state of the postinfarcted human heart, which continues to be subjected to ventricular remodeling, denervation and innervation, neurohumoral activation, as well as periods of regional myocardial ischemia.


Although K channel blockers at first were expected to improve survival of patients after myocardial infarction, the SWORD trial suggested the opposite. Although the mechanism is unknown, this drug abbreviated survival of patients with better left ventricular function more than those with poorer functions (24). It is possible that patients with better left ventricular function have more heterogeneously distributed surviving ventricular muscles with various degrees of autonomic innervation and denervation than those with worse function. Although this is speculative, those areas of heterogeneous tissue may be more benefited by β-blockade than the areas with homogeneous tissues. In addition, it is interesting to note that in two recent multicenter studies, EMIAT and CAMIAT, the improvement of survival by amiodarone was augmented by the addition of β-blockers (25). Although amiodarone has multiple in channel-blocking effects, including that of IK blockade, these findings are compatible with our results. Our study strongly suggests that to protect and even to enhance antiarrhythmic efficacy of K channel blockers, concomitant use of a β-blocker is crucial.

Acknowledgement: This research was partially funded by ONO pharmaceutical Co, Ltd.


1. Waldo AL, Camm AJ, deRuyter H, et al. For the SWORD investigators: effect of d-sotalol on mortality in patients with left ventricular dysfunction after recent and remote myocardial infarction. Lancet 1996;348:7-12.
2. Cairns JA, Connolly SJ, Roberts R, et al. Randomized trial of outcome after myocardial infarction in patients with frequent or repetitive ventricular premature depolarisations: CAMIAT: Canadian Amiodarone Myocardial Infarction Arrhythmia Trial Investigators. Lancet 1997;349:675-82.
3. Julian DG, Camm AJ, Frangin G, et al. Randomised trial of effect of amiodarone on mortality in patients with left-ventricular dysfunction after recent myocardial infarction: EMIAT, European Myocardial Infarct Amiodarone Trial Investigators. Lancet 1997;349:667-74.
4. Burkart F, Pfisterer M, Kiowski W, et al. Effect of antiarrhythmic therapy on mortality in survivors of myocardial infarction with asymptomatic complex arrhythmias: Basel Antiarrhythmic Study of Infarct Survival (BASIS). J Am Coll Cardiol 1990;16:1711-8.
5. Amiodarone Trials Meta-Analysis Investigators. Effect of prophylactic amiodarone on mortality after acute myocardial infarction and in congestive heart failure: meta-analysis of individual data from 6500 patients in randomised trials. Lancet 1997;350:1417-24.
6. Duchcatte-Gourdon I, Hartzell HC, Lagrutta AA. Modulation of the delayed rectifier potassium current in frog cardiomyocytes by beta-adrenergic agonists and magnesium. J Physiol 1989;415:251-74.
7. Scherlag BJ, Brachmann J, Kabell G, et al. Sustained ventricular tachycardia: common functional properties of different anatomic substrates. In: Zipes DP, Jalife J, eds. Cardiac electrophysiology and arrhythmias. Orlando: Grune & Stratton, 1985:379-87.
8. Sanguinetti MC, Jurkiewicz NK, Scott A, et al. Isoproterenol antagonizes prolongation of refractory period by the class III agent E-4031 in guinea pig myocytes: mechanism of action. Circ Res 1991;68:77-84.
9. Groh WJ, Gibson KJ, McAnulty JH, et al. Beta-adrenergic blocking property of dl-sotalol maintains class III efficacy in guinea pig ventricular muscle after isoproterenol. Circulation 1995;91:262-4.
10. Sager PT, Follmer C, Uppal P, et al. The effect of beta adrenergic stimulation on the frequency-dependent electrophysiologic actions of amiodarone and sematilide in humans. Circulation 1994;90:1811-9.
11. Salata JJ, Jurkiewicz NK, Jow B, et al. IK of rabbit ventricle is composed of two currents: evidence for IKs. Am J Physiol 1996;271:H2477-89.
12. Stanton MS, Zipes DP. Modulation of drug effects by regional sympathetic denervation and supersensitivity. Circulation 1991;84:1709-14.
13. Dhein S, Muller A, Gerwin MSc, et al. Comparative study on the proarrhythmic effects of some antiarrhythmic agents. Circulation 1993;87:617-30.
14. Lubisky A, Hindricks G, Shenasa M, et al. Dispersion of ventricular refractoriness in experimental myocardial infarction: effect of sotalol. PACE Pacing Clin Electrophysiol 1994;17:2090-4.
15. Freigang KD, Bauer A, Becker R, et al. Differential effects of d-sotalol on normal and infarcted myocardium: an experimental study using epicardial mapping. Cardiovasc Res 1997;35:52-9.
16. Marschang H, Schols W, Karolyi L, et al. Differential effects of d-sotalol on subendocardial Purkinje myocytes isolated from normal or 10 to 14 days postinfarction canine hearts: role of extracellular potassium concentration. J Cardiovasc Pharmacol 1998;32:274-83.
17. Gidh-Jain M, Huang B, Jain P, et al. Differential expression of voltage-gated K+ channel genes in left ventricular remodeled myocardium after experimental myocardial infarction. Circ Res 1996;79:669-75.
18. Pinto JM, Boyden PA. Reduced inward rectifying and increased E-4031-sensitive K+ current density in arrhythmogenic subendocardial Purkinje myocytes from the infarcted heart. J Cardiovasc Electrophysiol 1998;9:299-311.
19. Aupetit JF, Frassati D, Bui-Xuan B, et al. Efficacy of a beta-adrenergic receptor antagonist, propranolol, in preventing ischaemic ventricular fibrillation: dependence on heart rate and ischaemia duration. Cardiovasc Res 1998;37:646-55.
20. Haws CW, Lux RL. Correlation between in vivo transmembrane action potential durations and activation-recovery intervals from electrograms: effects of interventions that alter repolarization time. Circulation 1990;81:281-8.
21. El-Sherif N, Caref EB, Yin H, et al. The electrophysiological mechanism of ventricular arrhythmias in the long QT syndrome: tridimensional mapping of activation and recovery patterns. Circ Res 1996;79:474-92.
22. Takatsuki S, Mitamura H, Sueyoshi K, et al. Discordant prolongation of the refractory period and repolarization time by a class III agent, E4031, in the healing phase of myocardial infarction. Jpn Heart J 1998;39:687-97.
23. Downar E, Janse MJ, Durrer D. The effect of "ischemic" blood on transmembrane potentials of normal porcine ventricular myocardium. Circulation 1977;55:455-62.
24. Pratt CM, Camm AJ, Cooper W, et al. Mortality in the Survival With Oral D-sotalol (SWORD) trial: why did patients die? Am J Cardiol 1998;81:869-76.
25. Kennedy HL. Beta-blocker prevention of proarrhythmic and proischemia: clues from CAST, CAMIAT, EMIAT. Am J Cardiol 1997;80:1208-11.

IKr blocker; β-Blocker; Myocardial infarction; QT dispersion

© 2000 Lippincott Williams & Wilkins, Inc.