Author + information
- Received December 9, 1998
- Revision received March 23, 2000
- Accepted May 24, 2000
- Published online October 1, 2000.
- Adelqui O. Peralta, MDa,
- Roy M. John, MBBS, PhD, FRCP, FACCa,* (, )
- William H. Gaasch, MD, FACCa,
- Peter I. Taggart, MD, FRCP∗,
- David T. Martin, MD, FRCP, FACCa and
- Ferdinand J. Venditti, MD, FACCa
- ↵*Reprint requests and correspondence: Dr. Roy M. John, Section of Cardiovascular Medicine, Lahey Clinic Medical Center, 41 Mall Road, Burlington, Massachusetts 01805
We sought to study the rate related effects of sotalol on myocardial contractility and to test the hypothesis that the class III antiarrhythmic effect of sotalol has a reverse use-dependent positive inotropic effect in the intact heart.
Antiarrhythmic drugs exert significant negative inotropic effects. Sotalol, a beta-adrenergic blocking agent with class III antiarrhythmic properties, may augment contractility by virtue of its ability to prolong the action potential duration (APD).
In 10 anesthetized dogs, measurements of left ventricle (LV) peak (+)dP/dt and simultaneous endocardial action potentials were made during baseline conditions and after sequential administration of esmolol and sotalol. In addition, electrical and mechanical restitution curves were constructed at a basic pacing cycle length of 600 ms by introducing a test pulse of altered cycle length ranging from 200 ms to 2,000 ms.
In the steady state pacing experiments, sotalol prolonged the APD in a reverse use-dependent manner; such an effect was not seen with esmolol. At cycle lengths exceeding 400 ms, LV (+)dP/dt was significantly higher with sotalol than it was with esmolol. There was a direct relation between APD and LV (+)dP/dt with sotalol (r = 0.46, p < 0.001), but there was no significant relation between APD and LV (+)dP/dt with esmolol (r = 0.27, p = NS). Results in the single beat (restitution) studies were qualitatively similar to the steady state results; APD (at cycle length >400 ms) and LV (+)dP/dt (at cycle length >600 ms) were significantly higher with sotalol than they were with esmolol.
The reverse use-dependent prolongation of APD by sotalol is associated with a positive inotropic effect.
Cardioactive agents that act through their effects on ion channels exhibit rate dependent effects on myocardial contractile function and electrophysiologic properties. Use dependency whereby a progressive increase in sodium channel blockade is seen with faster heart rates has been documented for many of the class I antiarrhythmic agents (1,2). For example, the slowing of conduction produced by flecainide becomes more pronounced at faster rates. By contrast, class III antiarrhythmic agents, such as sotalol that selectively block the rapid component of the delayed rectifier potassium channel, produce a progressive prolongation of action potential duration (APD) at slower heart rates (3,4). Thus, these class III agents exhibit reverse use dependency on the action potential duration.
In addition to their electrophysiologic effects, many antiarrhythmic agents have significant influences on myocardial contractility. Most exert a negative inotropic effect that limits their use in patients with depressed left ventricular (LV) function (5,6). It has been suggested that class III agents, by their effect of prolonging APD and, hence, the time for calcium entry, may in fact negate or attenuate any potential effect on ventricular contractility. For example, sotalol, which is a beta-adrenergic receptor blocking agent with potassium channel blocking effects, has been reported to have less depressant effect on contractility compared with propranolol (7,8). The rate dependent effects of sotalol on contractility and the relationship between changes in contractility and changes in APD produced by sotalol have not been systematically explored in the intact heart.
We, therefore, designed this study to define the effects of sotalol on the rate related changes in APD and contractility in canine hearts and to examine the relationship between the changes produced in APD and contractility. These effects were investigated under steady state pacing conditions and with single test pulses at progressively longer coupling intervals. In order to identify the additional class III effect of sotalol and separate this from its beta-blocker effect, the sotalol data were compared with those obtained with esmolol. Esmolol is a beta-1 selective blocker without intrinsic sympathomimetic or membrane stabilizing activity and has no known electrophysiological effects other than via beta-blocking effects (9).
1.1 Experimental preparation
The study protocol was approved by the Lahey Clinic Animal Research Committee and conforms to the position of the American Heart Association on research animal use. Ten adult mongrel dogs of either sex with an approximate weight of 25 kg were premedicated with intravenous pentobarbital and intubated. Positive pressure ventilation was maintained on an artificial respirator. Halothane carried in a 1:1 oxygen nitrogen mixture was used for maintenance of anesthesia. Ventilation was controlled so as to maintain arterial pH between 7.35 and 7.4 and PCO2 between 35 and 40 mm Hg. Core temperature was maintained between 37°C and 38°C by a heating blanket on the operating table. The left external jugular vein and the left carotid artery were cannulated with 8F sheaths. The chest was then opened with a left thoracotomy, and the heart was suspended in a pericardial cradle. An 8F micromanometer tipped catheter (Millar Mikro-Tip, Millar Instruments, Inc., Houston, Texas) was inserted through an apical stab incision to record LV pressure and the time derivative of pressure (dP/dt); LV peak (+)dP/dt was used as an index of contractility. Pacing wires were sutured to the right ventricular outflow tract and left atrium appendage for bipolar pacing of these chambers.
Complete heart block was created using catheter based radiofrequency ablation of the atrioventricular (AV) node. Under fluoroscopy, a 7F deflectable tip catheter (EP Technologies, Inc., Sunnyvale, California) was positioned in the right AV junction where a His bundle electrogram was recorded (three experiments) or via a transaortic retrograde approach in the LV septum immediately below the noncoronary cusp (seven experiments). Radiofrequency current in the form of a 500 kHz continuous unmodulated wave form was delivered from a generator (Radionics, Inc., Burlington, Massachusetts) in a unipolar fashion to the catheter tip. The energy output was increased to levels between 25 and 40 W until permanent complete heart block was achieved. The hearts were paced from the right ventricular outflow tract.
1.2 Monophasic action potential (MAP) recordings and signal processing
Monophasic action potential recordings were obtained from the endocardial surface of the heart. A 7Fr endocardial MAP catheter mounted with silver/silver chloride electrodes (EP Technologies, Inc.) was introduced through the sheath in the carotid artery into the LV and positioned in a location where stable signals of amplitude greater than 10 mV were obtained. The MAP signals were processed using a DC-coupled isolated preamplifier (EP Technologies, Inc.) and the output processed and amplified in a Model 5900 Signal Conditioner (Gould Instruments, Inc., Valley View, Ohio). The MAP and LV pressure signals were displayed on a MAG monitor (Gould Instruments, Inc.) and printed real time on a TA5000 thermal array recorder (Gould Instruments, Inc.) at 200 mm/s paper speed.
Recognizing that electrical or mechanical restitution may be variably governed by diastolic interval rather than cycle length, we plotted data (in pilot studies) using diastolic intervals and cycle lengths; no clear differences were evident. Therefore, we chose to display the data using cycle lengths in order to maintain uniformity of data presentation.
1.3 Pacing protocol
The atria were paced at rapid rates (300 beats/min) to eliminate effective atrial contraction and minimize variations in LV preload that could have resulted from AV dissociation caused by complete heart block. Steady state curves were constructed by pacing the right ventricle using a constant current stimulator with 1.9 ms pulse duration and twice diastolic current threshold at cycle lengths of 300, 400, 600, 800, 1,000 and 1,500 ms. Each rate was maintained for a minimum of 3 min to allow attainment of steady states. Subsequently, electrical and mechanical restitution curves were constructed at a basic cycle length of 600 ms by introducing a single test pulse of altered cycle length after each eighth steady state beat such that the test pulse intervals spanned a range of 200 to 2,000 ms. During pacing protocols, a surface electrocardiogram lead, LV pressure, LV dP/dt and endocardial MAPs were recorded.
Measurements were made during the baseline drug-free state, after an infusion of esmolol at a rate of 150 to 250 μg/kg/min to achieve a 20% reduction of the sinus rate (determined by examining the unpaced atrial rates) and after sotalol (2 mg/kg) administration (10). A period of 45 to 60 min was allowed between cessation of esmolol infusion and commencement of sotalol to permit complete elimination of esmolol.
1.4 Data analysis
Monophasic action potential recordings with an amplitude greater than 10 mV were used for analysis. Measurements of MAP duration were made manually at 90% repolarization (APD90). Data obtained when heart rate control was lost (for example, during spontaneous junctional rhythms) were not employed for analysis.
Data are expressed as mean ± SEM. Left ventricular peak (+)dP/dt and APD at different cycle lengths were compared utilizing two-way repeated measures analysis of variance. Best fit linear regression analysis was performed for the force frequency relationship and cycle length dependence of APD. A Pearson product moment correlation coefficient was obtained between changes in LV peak (+)dP/dt and APD produced by sotalol and esmolol. A p value of <0.05 was considered statistically significant.
An example of a portion of a steady state pacing experiment is shown in Figure 1. In the baseline state, the APD increased from 185 to 260 ms when the pacing interval increased from 400 to 1,000 ms; this was accompanied by a decline in LV systolic pressure (78 to 64 mm Hg) and a reduction in peak (+)dP/dt (770 to 525 mm Hg). After sotalol, the action potential increased from 230 to 340 ms; this was accompanied by smaller changes in LV systolic pressure (94 to 88 mm Hg) and peak (+)dP/dt (945 to 805 mm Hg/s). This example illustrates the direct relation between cycle length and APD and the effect of sotalol. The negative inotropic effect of a long interval is attenuated by sotalol.
2.1 Steady state pacing experiments (n = 10)
The relation between steady state APD and cycle length is shown in the upper panel of Figure 2. In the baseline state and with esmolol, APD increased as a function of cycle length. After sotalol, the APD was significantly longer than that seen during the baseline state and after esmolol (p < 0.001); the absolute prolongation was most marked at the longest cycle lengths (17% at short cycle lengths vs. 23% at long cycle lengths). This greater prolongation of APD at slower heart rates indicates a reverse use-dependent effect of sotalol.
The relation between LV peak (+)dP/dt and cycle length is shown in the lower panel of Figure 2. During the baseline state, LV peak (+)dP/dt was maximal at a pacing cycle length of 400 ms, and it decreased as a function of increasing cycle length (r = −0.43, p < 0.01). Esmolol, while maintaining a similar inverse relation (r = −0.39, p < 0.01), displaced the curve downward, indicating a depressant effect on contractility. Left ventricular peak (+)dP/dt with sotalol was significantly higher than with esmolol at cycle lengths of exceeding 400 ms; it was higher than baseline only at the longest cycle lengths. Thus, sotalol nearly abolished the force frequency relation, and it appeared to exert a relatively positive inotropic effect at long cycle lengths.
As shown in Figure 3, there was a positive relation between changes in APD and peak (+)dP/dt with sotalol (r = 0.46, p < 0.001). By contrast, there was no significant relation between APD and peak (+)dP/dt with esmolol (r = 0.27, p = NS). These data indicate that the preserved/augmented contractility seen with sotalol (Fig. 2) is related to its class III reverse use-dependent effects.
2.2 Restitution experiments
The relation between a single test pulse (with progressive increase in coupling interval) and APD is shown in Figure 4. After sotalol, the APD of the premature beat (test pulse) was longer than that seen in the baseline and esmolol states (at coupling intervals exceeding 400 ms, p < 0.001). The peak (+)dP/dt was higher than that seen with esmolol (at coupling intervals exceeding 600 ms, p < 0.01); the peak (+)dP/dt was also higher than that seen in the baseline state, but this did not achieve statistical significance. These data are qualitatively similar to those in the steady state pacing studies in that APD, and (+)dP/dt with sotalol were significantly higher than those seen with esmolol.
The primary findings of this study are: 1) sotalol produced an increase in the APD in a reverse use-dependent manner. 2) During steady state pacing, esmolol had a negative effect on LV contractility, while sotalol exhibited a positive effect, particularly at longer cycle lengths. 3) The preservation of contractility by sotalol correlates with prolongation of the APD. 4) The single test pulse (restitution) results were qualitatively similar to the steady state pacing results; relative to esmolol, sotalol produced an increase in APD and contractility. It appears, therefore, that the depressant effect of a beta-blocker on the relation between electromechanical parameters and cycle length are largely reversed when the agent has class III reverse use-dependent effects. Indeed, our data indicate a positive inotropic effect of sotalol, which becomes apparent at long cycle lengths.
3.1 Sotalol and steady state pacing
Sotalol, a beta-blocker with potassium blocking properties, is known to produce action potential and QT interval prolongation at slow rates with little or no prolongation at rapid rates (1,3,4,11). This effect has been demonstrated in vitro and in vivo and is related to blockage of the rapid component of the delayed rectifier potassium current (12). This current plays an important role in physiologic repolarization, but its contribution is reduced at rapid rates or during sympathetic stimulation when the slow component of the current becomes more prominent. Thus, compounds that predominantly block the rapid component delayed rectifier (Ikr) potassium current manifest reverse use dependency, prolonging APD at longer cycle lengths.
A positive inotropic effect of APD prolongation has been suggested by a number of investigators (7,8,13,14). Aberg et al. (7) and Hoffman et al. (8) showed that sotalol exerted less depression of contractility compared with propranolol in intact dog hearts. Exploring the rate related effects of sotalol, Lathrop (12) utilized canine Purkinje strand fibers to demonstrate that force development by sotalol was enhanced when stimulated at slow rates. Although it was proposed that action potential prolongation and increased calcium availability were the underlying mechanisms, concomitant organ (i.e., intact heart) electrophysiological data are lacking. In this study, radiofrequency ablation of the AV junction was used to allow pacing at very long cycle lengths and permit evaluation over a wide range of heart rates, a feature usually reserved for in vitro studies. Our studies indicate that sotalol, by prolonging APD in a reverse use-dependent manner, is associated with an increase in ventricular contractility. Interestingly, this effect of the drug appears to abolish the positive force frequency relation by attenuating the negative inotropic effect of bradycardia.
The precise mechanism underlying a maintained contractility at slow rates is unknown. In pharmacological doses, sotalol does not have a direct effect on the calcium channels (15); therefore, an indirect effect must be considered. White et al. (13) studied compound II, a class III antiarrhythmic drug similar to sotalol but with no beta-blocking properties. They were able to demonstrate that the drug reduced calcium extrusion via the sodium-calcium exchange, allowing augmented cellular uptake of calcium and more calcium available for release in subsequent beats. A second potential effect of prolonging APD is that more time would be available for L-type calcium channels to conduct calcium; this could provide additional “trigger calcium” and thus augment release of calcium from the sarcoplasmic reticulum (16). It is known that the calcium influx through the L-type calcium channels is unable (by itself) to produce a significant muscle twitch. Rather, it acts as a trigger for calcium release from the sarcoplasmic reticulum, the so-called “calcium-induced calcium release” (17).
3.2 Sotalol and restitution
Mechanical restitution is usually characterized by a steep initial slope of the strength-interval relation at short coupling intervals followed by a plateau at long coupling intervals (18). This form of strength-interval relation has been interpreted in terms of calcium ion movements and the presence of a two compartment calcium model (19,20). During relaxation, calcium is actively taken up by the sarcoplasmic reticulum, stored in a temporary compartment and eventually transferred into a labile or release compartment. This transfer is time-dependent. Thus, the calcium content of the labile store, which determines the inotropic state of the myocardium, is heavily dependent on the preceding history of activity (21). During excitation-contraction, the amount of calcium released from the labile store is governed by the rate of rise and magnitude of the inward calcium current.
The electrical and mechanical restitution curves shown in Figure 4 illustrate a prolonged APD and a higher (+)dP/dt during the plateau phase in the sotalol versus esmolol states. Thus, a longer APD appears to provide for more calcium availability and an increased contractility. These restitution data are consonant with those in the steady state pacing studies; there was approximately a 70 to 80 ms difference in the APD and approximately a 300 to 400 mm Hg/s difference in (+)dP/dt in the sotalol versus esmolol states. When expressed as a percent difference, however, the difference in max (+)dP/dt (sotalol vs. esmolol) was greater in the steady state than it was in the single test restitution studies (approximately 30% vs. 90%). This finding is in keeping with the complex dependence of contractility on previous APDs and diastolic intervals.
In clinical practice, many antiarrhythmic drugs exert negative inotropic effects that limit their use in patients with arrhythmia and coexistent LV dysfunction. In such patients, prolongation of the APD may confer a protective effect on ventricular function. For example, amiodarone, a drug with predominant class III activity, has been shown to improve contractility in patients with severe ventricular dysfunction (22). Similarly, in the clinical setting, sotalol has minimal effects on LV function, despite its beta-blocking properties (23,24). In this study, we have demonstrated a reverse use-dependent inotropic benefit of sotalol on cardiac contractility, and it is possible that the clinical combination of the APD prolongation and beta-blockade will slow the heart rate to a range in which a protective effect on contractility is seen. Our results offer a mechanistic explanation for the potential beneficial effects of combining beta-blockade and a class III antiarrhythmic effect.
3.3 Study limitations
It is possible that the beta-blocking effects of esmolol and sotalol in doses used in this study were not equivalent. Comparing sinus rates would not have ensured equipotent beta-blockade because the class III activity of sotalol exerts direct dromotropic effects on the sinus node. Hence, we had to rely on standard doses of sotalol used in previously published experimental work. Sotalol has different concentration response curves for beta-blockade and class III activity; beta-blockade is achieved at lower concentrations before class III activity becomes apparent. In doses used in this study where significant class III activity was manifest, we expect to have achieved levels of beta-blockade comparable to that seen with clinical use of the drug. Study utilizing a pure class III agent such as d-sotalol would have been of interest. This drug is, however, not available for clinical use. Our data with d-l sotalol maintains a degree of clinical applicability in that slow heart rates at which inotropic effects were evident are more likely to be achieved with a beta-blocker–class III combination than with a pure class III agent. Secondly, unlike sotalol, esmolol has relative cardioselectivity, and a potential drawback is that the two drugs may not be strictly comparable in their beta-blocking properties. Esmolol was selected for these experiments because of its short duration of action and because, like sotalol, it has no intrinsic sympathomimetic activity. Finally, our recordings of action potentials were obtained from the endocardial surface. It is possible that recordings from the midmyocardial layers, where APD-cycle length relationships are most pronounced, may have yielded different results. However, APD changes follow the same direction in both layers, and it is likely that our results are representative of the global LV repolarization characteristics.
☆ Dr. Peralta was supported by the Eleanor-Dana, Siemens Corporation and Gordon Research Grant 1995 to 1996.
- action potential duration
- left ventricle or ventricular
- monophasic action potential
- time derivative of pressure
- Received December 9, 1998.
- Revision received March 23, 2000.
- Accepted May 24, 2000.
- American College of Cardiology
- Campbell T.J.
- Strauss H.C.,
- Bigger T.,
- Hoffmann B.F.
- Ravid S.,
- Podrid P.J.,
- Lampert S.,
- Lown B.
- Strauss H.C.,
- Bigger J.T.,
- Hoffman B.F.
- Holubarsch C.,
- Schneider R.,
- Pieske B.,
- et al.
- Komeichi K.,
- Tohse N.,
- Nakaya H.
- Beuckelmann D.J.,
- Weir W.G.
- Franz M.R.,
- Schaefer J.,
- Schottler M.,
- et al.
- Pidgeon J.,
- Lab M.,
- Seed A.,
- et al.
- Prabhu S.D.,
- Freeman G.L.
- Fabiato A.
- CHF-STAT Investigators,
- Massie B.M.,
- Fisher S.G.,
- Deedwania P.C.,
- et al.
- Mahmarian J.J.,
- Verani M.,
- Hohmann T.,
- et al.