Clinical and Experimental Pharmacology and Physiology (1994) 21, 179-188

COMPARATIVE HAEMODYNAMIC EFFECTS OF VERAPAMIL, FLECAINIDE, AMIODARONE AND SOTALOL

IN THE CONSCIOUS RABBIT

N. Twidale,* P. Roberts-Thomson, R. J. McRitchie and J. P. Chalmers

Department of Medicine, Flinders Medical Centre, Bedford Park, South Australia, Australia

SUMMARY

1. The effect of intravenous boluses of verapamil (0.15 mg/ kg), flecainide (2 mg/ kg), amio- darone (5 mg/ kg), and sotalol(l.5 mg/ kg) on mean arterial pressure, heart rate (HR), cardiac output (CO), total peripheral resistance (TPR), and peak rate of change of left ventricular pressure (LV dP/ dt) were assessed in the conscious rabbit.

2. All four drugs had negative inotropic effects: verapamil reduced peak LV dP/d t by 19 f 4 % (mean+s.e.m.; P<O.Ol), flecainide by 27+9% (P<O.OOl), amiodarone by 11 +2% (P<O.Ol) and sotalol by 133~3% (P<O.Ol).

3. The drugs had different effects on CO as a result of differences in their actions on peripheral blood vessels: verapamil and amiodarone produced, respectively, a 12 f 4% (P<0.03) and 16 f 6% (P<O.Ol) increase in CO associated with a substantial vasodilatory effect (TPR reduced 15 f 7 % [P<0.05] and 20+5% [P<O.Ol], respectively). Flecainide caused only a small (6+ 1%; P<O.Ol) increase in CO and sotalol had no effect on either CO or TPR.

4. Bolus intravenous injections of verapamil, flecainide and amiodarone produced an increase in HR, while sotalol reduced HR by 10 k 2% (P<O.Ol). The increase in HR and cardiac output seen with verapamil, flecainide and amiodarone was in part secondary to reflex increase in sympathetic tone and these changes were abolished after total cardiac autonomic blockade.

5. The modest reduction in cardiac performance associated with sotalol was abolished by cardiac autonomic blockade, suggesting that the predominant effect of sotalol on contractility was mediated through its P-adrenoceptor blocking effect.

Key words: amiodarone, antiarrhythmic drugs, flecainide, haemodynamic effects, sotalol, verapamil.

INTRODUCTION

Anti-arrhythmic drugs exert their effects on cardiac rhythm by complex interactions with sodium, calcium and potassium ionic channels (Hondeghem 1987). The resulting changes in ionic conductance may affect

Correspondence: Dr R. J . McRitchie, Department of Medicine, Flinders Medical Centre, Bedford Park, SA 5042, Australia.

*Present address: Dr N. Twidale, 615 E. Oklahoma, #207, Enid, OK 73701, USA.

Received 13 October 1993; revision 5 January 1994; accepted 5 January 1994.

the contractile response of cardiac myocytes and vas- cular smooth muscle, leading to complex and often profound haemodynamic changes (Schlepper 1989). The net circulatory effects of anti-arrhythmic drugs depend on the balance between their direct negative inotropic properties, the indirect effects resulting from changes in peripheral vascular resistance, and any reflex neurohumoral responses which may be evoked (Bourke et al. 1987).

Verapamil inhibits slow calcium ion conductance and reduces the release of cytosolic calcium, therefore producing a marked direct cardio-depressant action in isolated preparations (Nayler & Szeto 1972). How- ever, this effect is difficult to demonstrate in humans

180 N. Twidule et al.

because of a marked peripheral vasodilatory effect on vascular smooth muscle which offsets the reduction in contractility (Nayler & Horowitz 1983). Amiodarone, which produces significant prolongation of the car- diac action potential duration and is thus classified as possessing Class 3 anti-arrhythmic properties (Vaughan Williams 1970), is reported to have little direct negative inotropic action, and because periph- eral vascular resistance and hence cardiac afterload are reduced, cardiac output is well maintained even during intravenous bolus administration (Remme et al. 1991).

Flecainide has powerful and prolonged sodium channel blocking activity and thus, might be expected to produce at least moderate direct cardiodepressant effects (Cowan & Vaughan Williams 1981). Although early clinical experience suggested a more favourable haemodynamic profile (Anderson et al. 1981), more recent studies have substantiated concerns regarding the potential for adverse haemodynamic effects with flecainide (De Paola et al. 1987). Sotalol shares with amiodarone the effect of prolonging action potential duration and has been reported to be well tolerated, even in patients with impaired ventricular function (Singh et al. 1988). By the lengthening of the action potential duration, amiodarone and sotalol may delay inactivation of the slow inward calcium current and augment contractility in isolated preparations (Tande & Refsum 1988,1990). However, sotalol has additional /3-adrenoceptor blocking properties which have direct depressant actions and may impair reflex compen- satory mechanisms in some patients dependent on augmented sympathetic drive (Ruder et al. 1989).

Thus, the overall haemodynamic effects of anti- arrhythmic drugs will depend critically on the balance between any direct cardiodepressant action and in- direct effects on cardiac loading conditions and changes in autonomic reflexes. These responses will be absent in studies performed in vitro, or impaired in studies performed in animals under general anaesthesia (Vatner & Braunwald 1975).

The aim of this study, therefore, was to compare in the conscious rabbit, the changes in cardiac perfor- mance produced by verapamil (a known cardiode- pressant) and amiodarone (a drug with minor haerno- dynamic effects), with those of flecainide and sotalol, and assess the relative contribution of direct, peripheral and reflex actions of these drugs on the heart.

METHODS Operative procedures Animal care and instrumentation were performed in accordance with the guidelines proposed by the Aus-

tralian National Health and Medical Research Council and the study protocol was approved by this hospital's animal ethics committee.

Male New Zealand white rabbits (mean weight 2.5 kg) underwent preliminary operations under halothane anaesthesia following induction with sodium thiopen- tone (Pentothal, Abbott, Sydney, NSW, Australia: 5-15 mg/ kg i.v.). Intrathoracic procedures required assisted ventilation. In one group of rabbits, Doppler ultrasonic flow transducers were implanted around the ascending thoracic aorta at a right thoracotomy, with the leads positioned subcutaneously. In some animals, atrial pacing leads were sewn to the epicardial surface of the right atrium at the same operation. The animals were allowed to recover fully before experi- ments were performed 2 weeks later. In other rabbits, specially fabricated silastic catheters were inserted into the left ventricle via the left carotid artery to measure left ventricular (LV) pressure and rate of change of ventricular pressure (LV dP/dt). The tubing was tunnelled subcutaneously to exit at the back of the rabbit's neck. This catheter was flushed daily with heparinized saline to maintain patency, and experi- ments were performed 2 days later when the rabbit had fully recovered.

Measurement of circulatory variables On the day of the experiment, after local infiltration with 0.5 mL 1% lignocaine, the central ear artery and marginal vein were cannulated and following con- nection of the left ventricular catheter, pacing and Doppler transducer leads to the chart recorder, each rabbit was allowed to rest quietly in a covered box for 30 min prior to recording. Arterial blood pressure was measured with a Bell and Howell physiological trans- ducer (Bell and Howell, Pasadena, CA, USA) con- nected to the ear artery catheter. Left ventricular pressure was measured by connecting the indwelling left ventricular catheter to a high-fidelity transducer- tipped catheter (Millar MIKRO-TIP 7F, Millar In- struments Inc., Houston, TX, USA). With optimal damping the frequency response is linear to 60 Hz and is suitable for accurate measurement of LV pressure ant) LV dPjdt (Aylwatd 1983). LV dP/dt was differ- entbted.from%b hY pressure signal by a Tektronix A M 501 operational amplifier (Tektronix Inc., Bcaverton, 13R, USA) which had a fresuency response

ar wave signal of known slope pressure signd so as to give

direct calibration of the dP/dt sigqal (Tektronix Model FG 502; McRitchie et al. 1984). Peak LV dP/dt was used as the optimum isovolumCtric index of con- tractility.

Haemodynamic effects of anti-arrhythmic drugs 181

Cardiac output (CO) was measured by connecting the ultrasonic flow transducer leads to a Doppler ultrasonic flowmeter (Baker Institute, Melbourne, Vic., Australia) with the signal threshold adjusted to assume zero phasic flow at end-diastole (White et al. 1974). Total peripheral resistance (TPR) was calculated by dividing the mean arterial pressure (MAP) by the mean CO, while instantaneous heart rate (HR) was recorded on a beat-to-beat basis by a cardiotachometer triggered by the phasic LV pressure or CO signal.

Continuous recordings of phasic and MAP, LV pressure, high-gain LV pressure, LV dP/dt, HR and phasic and mean Doppler shift were displayed on an eight-channel Gould 2800 biomedical recording system (Gould Instruments Division, Cleveland, OH, USA).

Drug administration

Verapamil (Isoptin, Knoll, Alexandria, NSW, Aus- tralia; 0.15 mg/ kg i.v.), flecainide (Tambocor, Riker, Sydney, NSW, Australia; 2 mg/ kg i.v.), amiodarone (Cordarone, Labaz, Sydney, NSW, Australia; 5 mg/ kg i.v. with diluent polysorbate 80), and sotalol (Sotacor, Astra, Sydney, NSW, Australia; 1.5 mg/ kg i.v.) were administered by slow intravenous injection over 2 min. Plasma flecainide levels were measured using fluorescence polarization (Abbott Laboratories, Abbott Park, IL, USA), and verapamil, amiodarone and sotalol levels were measured by high performance liquid chromatography.

Cardiac autonomic blockade was produced by administration of propranolol (initial dose, 500 pg/ kg i.v., followed by 100 pg/ kg per rnin i.v.; West et al. 1975) and methylscopolamine (initial dose, 50 pg/ kg i.v., supplemented by 25 pg/kg i.v. every 30 min; Dorward et al. 1982).

As negative chronotropic effects of anti-arrhythmic drugs may alone reduce peak LV dP/d t and cardiac output (Mahler et al. 1975; Broughton & Korner 1980), anti-arrhythmic drugs were administered in some animals during constant atrial pacing.

Experimental protocol

After the rabbits had been rested quietly for 30 min, circulatory variables were measured five times over a further 30 rnin period prior to drug administration. The drugs in similar volume, or saline, were then given over 2 rnin and the lines flushed with saline. Circulatory variables were measured at 3, 5 and then every 5 min, up to 30 rnin after drug administration. In rabbits in which autonomic blockade was produced, propranolol and methylscopolamine were given after the initial five control readings, and a further four

control readings were taken during the 15 min prior to administration of the anti-arrhythmic drugs or saline. In rabbits studied with maintenance of constant HR by atrial pacing, the pacing leads were connected to a Tektronix FG 502 (Tektronix Inc.) function stimulator. The rabbits were paced at a rate sufficiently above their intrinsic rate to prevent breakthrough of their own rhythm (usually 10-15 beats/min >intrinsic rhythm).

Drug levels were assessed at rest and 3, 10, 15 and 30 rnin after drug administration. Each experimental group comprised at least four rabbits, some of which received all four drugs on separate days, while some received only one drug (total study population approximately 30 rabbits). For rabbits receiving multiple drug testing, amiodarone was always the final drug tested.

Statistical analysis The results are expressed as mean f s.e.m. The sig- nificance of changes in circulatory variables produced by administration of an anti-arrhythmic drug were assessed using paired two-tailed Student’s t-test. Com- parison between the drug-induced changes produced in control animals (without cardiac autonomic block- ade) and corresponding drug-induced changes pro- duced in animals pretreated with cardiac autonomic blockade were tested using unpaired Student’s t-test, after applying the Bonferroni adjustment for multiple comparisons (Wallenstein et al. 1980). For comparison of the drug-induced circulatory changes within experi- mental groups, Scheffe’s test and one factor analysis of variance for multiple comparisons was performed (Wallenstein et al. 1980).

RESULTS

Plasma drug concentrations

Verapamil The mean plasma concentration of verapamil meas- ured immediately (at 3 min) after i.v. bolus injection (0.15 mg/ kg) was 68 f 11 ng/ mL, and was within the range of plasma verapamil levels reported to produce therapeutic effect (60-70 ng/ mL; Dominic et al. 1979). Plasma verapamil concentration fell and by 30 rnin was 15 f 0.6 ng/mL as shown in Fig. la.

Flecainide The mean plasma concentration of flecainide measured immediately (at 3 min) after i.v. bolus injection (2

182 N. Twidale et al.

mg/kg) was 0.44f0.02 mg/L, and was within the appropriate human therapeutic range of 0.2-0.9 mg/ L (Duff et al. 1981). Plasma flecainide concentration fell but remained within the therapeutic range even after 30 rnin following drug administration as shown in Fig. lb.

Amiodarone The mean concentration of amiodarone measured at 3 rnin after a 5 mg/ kg i.v. bolus injection was 4.1 f 1.5 mg/ L which is above the recommended human thera- peutic range of 1-2.5 mg/ L (Counihan & McKenna 1990), but was similar to values obtained in man with the same weight adjusted i.v. dose (Andreasen et al. 1981). Plasma amiodarone concentration was within the therapeutic range 10 rnin after drug injection and continued to fall and by 30 rnin was 0.4 f 0.04 mg/ L as shown in Fig. lc.

So talo 1 The mean plasma concentration of total sotalol (enantiomeric ratio d/l-sotalol 1.1) measured 3 rnin after a 1.5 mg/kg i.v. bolus injection was 2.2f0.2

(a) Verapamil (0.15 mglkg. i.v.) (b) Flecainide (2 mglkg. i.v.)

0 ' , 0 3 10 15 10

Time (min)

( c ) Amiodarone (5 mg/kg, i.v.) (d)

1 $0 15 30

Time (min)

4

3 10 I 5 30

Time (min)

Sotalol (1.5 mglkg, i.v.)

3 10 ts Jo

Time (rnin)

Fig. 1. Plasma drug concentration (mean f s.e.m.) following i.v. administration of (a) verapamil; (b) flecainide; (c) amiodarone; (d) sotalol. The human therapeutic range for verapamil is 60-70 ng/mL; for flecainide, 0.2-0.9 mg/L; for amiodarone, 1.0-2.5 rng/L; and for sotalol, 0.5-9.0

mgl L.

mg/ L. The plasma concentrations of sotalol found in this study were similar to those measured in man after the same weight adjusted i.v. bolus (Nademanee et al. 1985) and were within the proposed therapeutic range of 0.5 to 5 mg/L (Kopelman et al. 1988). The mean plasma concentration of sotalol fell and by 30 rnin was 0.6 f 0.1 mg/ L as shown in Fig. Id.

(a) Control

MAP co TPR dP/d t HR

-60

(b) Cardiac autonomic blockade

4 0

20

g o

$ -20

c 0 c 0

- 4 0

-60

MAP co TPR

I I

dP/dt HR

s in MAP, CO, TPR, peak LV

kg), and sotdol (p4, 1.5 mg/kg). (a) s with intact cardiac autonomic

effectors. (b) Changes in animals with cardiac autonomic effectors blocked by propranolol and methylscopolamine. values are means.f:s.e.m. *P<O.OS refers to changes from baseline. 'P<0.05 refers to changes after CAB compared

to control.

Haemodynamic effects of anti-arrhythmic drugs 183

Haemodynamic effects

Verapamil No cardiac autonomic blockade The haemo-

dynamic changes produced by an i.v. bolus of ver- apamil are shown in Fig. 2a. Following i.v. bolus injection, MAP fell by 7 f 2 % (P<O.Ol) and was accompanied by a marked 1 5 f 7 % fall in TPR (P<0 .05) . This was associated with a significant increase in HR of l O f 4 % (P<O.Ol). Combined with a fall in impedance to left ventricular ejection, the increase in HR resulted in a 1 2 f 4 % increase in CO (P < 0.03), this occurring despite a significant decrease in peak LV dP/d t of 1 9 f 4 % (P<O.Ol). As the drug level fell with time (Fig. la), the haemodynamic changes returned toward control and by 30 min, only HR remained elevated compared to resting values.

Following cardiac autonomic blockade Combined p-adrenoceptor and vagal blockade with propranolol and methylscopolamine reduced the resting peak LV dP/dt by 3 3 f 7 % (P<O.Ol) and decreased HR by 7 f 3% (P<O.Ol) as shown in Table 1.

After pretreatment with propranolol and methyl- scopolamine, verapamil reduced MAP by 12 f 5% (P<0.05) which accompanied a significant fall in CO of 23 f 3% (P<O.Ol). As a result, the fall in TPR seen with verapamil was attenuated (8 f 1%; P<0.05). As sympathetic effectors were blocked, HR decreased significantly by l O f 2 % (P<O.Ol). Peak LV dP/d t decreased markedly by 4 2 f 3 % (P<O.Ol) , as shown in Fig. 2b.

Flecain ide No cardiac autonomic blockade The haemo-

dynamic effects of flecainide are shown in Fig. 2a. Following the i.v. bolus, MAP fell by lOf5% (P<O.Ol) accompanying a small decrease in TPR of 8 f 3% (P<O.Ol). This was associated with an increase in HR of 8 f 3 % (P<O.Ol). As a result, CO was

Table 1. Effect of cardiac autonomic blockade on resting haemodynamic parameters

Post-autonomic blockade (propanolol and

Control methylscopolamine)

n 17 17 MAP (mmHg) 77 f 4.0 75 * 5.0 CO (kHz) 2.4 k 0.2 2.5 f 0.2 TPR (units) 31 k 3.0 30 f 3.0 LV dP/dt (mmHg/ s) 5545 f 410 3703 f 310* HR Ibeatslmin) 259 f 8.0 241 f 8.0*

Values are means f s.e.m. *P<0.05 for difference from control.

relatively maintained (increase of 6 f 1%; P<O.Ol) despite a 2 7 f 9 % decrease in peak LV d P / d t (P<O.OOl). As the drug level fell with time (Fig. lb), the haemodynamic variables returned toward control until by 30 min, only HR remained increased compared to resting values.

After pre- treatment with propranolol and methylscopolamine, flecainide bolus injection reduced MAP by 11 f 6% (P<O.Ol). This was accompanied by a significant fall in CO of 20 f 3% (P<O.Ol). TPR increased 5 f 3% but this was not statistically significant. As sympathetic effectors were blocked, HR increased only 5 f 5% and this was not statistically significant. Peak LV dP/d t was reduced by 2 6 f 9 % (P<0.03), as shown in Fig. 2b.

Following cardiac autonomic blockade

Amiodarone No cardiac autonomic blockade The haemo-

dynamic effects of amiodarone are shown in Fig. 2a. Immediately following bolus injection, MAP was reduced 7 f 2% (P < 0.02), accompanied by a marked fall in TPR of 20f5% (P<O.Ol). HR increased by 5+2% (P<O.Ol) and, combined with the fall in impedance to left ventricular ejection due to the fall in afterload, CO increased 16+6% (P<O.Ol). Peak LV dP/dt was reduced 11 f 2 % (P<O.Ol). As drug levels fell (Fig. lc), haemodynamic variables returned toward control values.

Following cardiac autonomic blockade After pre- treatment with propranolol and methylscopolamine, bolus injection of amiodarone reduced MAP by 9 f 2% (P<O.Ol) accompanying a non-significant fall in CO of 5 * 3%. Due to blockade of sympathetic effectors, HR did not change significantly. Peak LV dP/dt was reduced significantly by 21 f 4 % (P<0.02) as shown in Fig. 2b.

Sot alo 1 N o cardiac autonomic blockade The haemo-

dynamic changes produced by an i.v. bolus of sotalol are shown in Fig. 2a. Following the boIus injection, MAP, CO and TPR were not significantly changed. HR decreased significantly by l O f 2 % (P<O.Ol) and peak LV dP/d t was reduced by 13*3% (P<O.Ol). As the drug level fell (Fig. ld), haemodynamic vari- ables returned toward control and at 30 min, variables were not significantly different from control values.

Following cardiac autonomic blockade Pretreat- ment with propranolol and methylscopolamine abol- ished the reduction in HR seen with sotalol, and

84 N. Twidale et al.

blocked the fall in peak LV dP/dt associated with sotalol injection when administered to animals without autonomic blockade (Fig. 2b).

Provision of constant atrial rate with atrial pacing LVdPldt Following administration of verapamil, flecainide and amiodarone in animals with HR maintained constant by fixed rate atrial pacing, the change in peak LV dP/dt associated with drug administration was not significantly different from that produced in control

(a) Control

MAP co TPR dPIdt HR I 1 I I

animals, as shown in Fig. 3a,b. By contrast, wheQ sotalol was administered to animals during constant atrial pacing, peak LV dP/dt fell by only 8 f 4% (P = NS) compared with the 1 3 f 3 % fall in LV dP/dt in animals without HR held constant.

Cardiac output Following administration of verapamil and amid- arone in animals with HR maintained constant, there was a rise in CO of 16 f 6% and 14 f 696, respectively. These did not reach statistical significance as the number of animals studied was small (n = 4). How-. ever, the change was similar in magnitude to that seen in animals without fixed rate atrial pacing (25 f 6% and 15 f 6%, respectively). Neither flecainide nor sotalol was associated with significant change in CO in any of the animals, paced or non-paced, as shown in Fig. 3a,b.

(b) Atrial pacing

MAP co TPR dPIdt HR

20

a 0 m E m c

-20 z

-.OI -60 I Fig. 3. Percentage c--anges in MAP, CO, TPR, pee- LV dP/dt and HR observed 3 min after intravenous adminis- tration of verapamil (0, 0.15 mg/kg), flecainide (m, 2 mg/kg), amiodarone (H, 5 mg/kg), and sotalol (W, 1.5 mg/kg). (a) Changes in animals without constant atrial pacing. (b) Changes in animals with heart rate held constant by right atrial pacing. Values are means f s.e.m. *P<0.05

for difference from baseline.

Experiments in sham animals Haemodynamic variables were measured following i.v. administration of 12 mL saline to assess the validity of the experimental protocol. No significant changes were observed when saline was administered alone or following cardiac autonomic blockade with propranolol and methylscopolamine.

DISCUSSION

In the present study, i.v. bolus administration of verapamil, flecainide, amiodarone and sotalol reduced peak LV dP/dt in the conscious rabbit. Since peak LV dP/dt is principally determined by the inotropic state (Mason 1969), the fall in peak LV dP/dt observed 3 min after drug administration probably reflects a decrease in myocardial contractility. As peak LV dP/dt can be influenced by the HR (Higgins et al. 1973), experiments were repeated in animals in which HR was held constant by atrial pacing. Following anti-arrhythmic drug administration, the haemo- dynamic changes were similar to those obtained in control animals (Fig. 3b).

This study confirms previous work showing that verapamil muses o reduction in inotropic state (Nayler & Szeto 1972; Nayler & Horowitz 1983). Total peripheral resistance was reduced confirming the findings of Greenberg and Wilson (1974) who demori- strated that verapamil causes marked vasodilatation of vascular beds. The fall in MAP leads to stimulation of arterial baroreceptors resulting in sympathetic neural stimulation and a rise in HR (Angus et al. 1976). Thus, although verapamil was associated with

Haemodynamic effects of anti-arrhythmic drugs 185

a significant cardiodepressant action, cardiac output increased due to indirect effects of verapamil in producing marked peripheral vasodilatation and sym- pathetic stimulation. After blockade of cardiac auto- nomic effectors, sympathetic stimulation was abolished and the direct negative inotropic and chronotropic effects of verapamil were manifest by a significant fall in cardiac output. Verapamil also produced a signifi- cantly greater fall in LV dP/dt in animals pretreated with propranolol compared to animals without auto- nomic blockade. These results support clinical data showing that combined P-adrenergic blockade and verapamil therapy may be hazardous, and produce adverse haemodynamic effects in some patients (Packer et al. 1982).

The results of this study corroborate other studies (Verdouw et al. 1979; Serruys et al. 1983) showing that flecainide is a negative inotrope as it reduces peak LV dP/dt by>25%. The change in peak LV dP/d t induced by flecainide appeared to be due to direct cardiodepression, as there was little change after cardiac autonomic blockade with propranolol and methylscopolamine. Flecainide had a slight vaso- dilatory action and reduced TPR by 8 * 3% resulting in a decrease in MAP of 10 f 5%, in agreement with the results of Hoffmeister et al. (1989). Thus, HR increased due to baroreceptor mediated sympathetic stimulation and cardiac output was generally main- tained as reported in man (Cohen et al. 1985; Jackson et al. 1985; Silke et al. 1986). However, after blockade of cardiac autonomic effectors, cardiac output was reduced significantly by almost 25%.

Amiodarone had a comparatively weak negative inotropic effect, as reported in other studies (Singh et al. 1976). The present study also confirmed previous studies showing that amiodarone has a profound vasodilatory action, similar in magnitude to that seen with verapamil (Cote et al. 1979). MAP was reduced 7*2% and there was a small but significant rise in HR (5 f 2%), similar to that reported by Remme et al. (199 I), resulting in a rise in cardiac output of 16 f 6%. The reflex increase in HR seen with amiodarone was less than that recorded for verapamil(10 * 4%), despite a greater fall in TPR ( 2 0 f 5 and 15*7%, respec- tively). The apparent blunting of the HR response seen with amiodarone may relate to its P-adrenoceptor blocking properties (Nokin et al. 1983). Another contributing factor responsible for the marked fall in TPR seen with amiodarone is the vasodilatory effect of the diluent used in this preparation of amiodarone, polysorbate 80 (Tween 80). This diluent causes a fall in vascular resistance and MAP which lasts for up to 4 min (Platou & Refsum 1986). Munoz et al. (1988) examined the effect of the solvent on myocardial

contractility and found that amiodarone with and without Tween 80 produced similar negative inotropic effects in humans when studied during routine coron- ary arteriography. Therefore, it is possible that Tween 80 contributes to the vasodilatory effect seen with amiodarone, but cannot solely explain the reduction in peak LV dP/d t seen in the present study. Because of the reduction in afterload after injection of amio- darone and the reflex increase in HR, cardiac output was increased despite the fall in contractility. When reflex sympathetic effectors were blocked, amiodarone produced a 21 *4% fall in LV dP/dt and cardiac output fell slightly by 5 5 3%. Like verapamil, the fall in contractility seen with amiodarone after pretreat- ment with propranolol was greater than in animals without autonomic blockade (1 1 * 2%) and may relate to the calcium blocking actions recognized with amiodarone (Yabec et al. 1986).

Surprisingly, sotalol produced a comparatively weak negative inotropic response, reducing peak LV dP/ dt by 13 f 3% in the present study, this being similar to findings reported in humans (Singh et al. 1988). The fall in contractility associated with sotalol injection appears to be predominantly due to its @-adrenoceptor blocking effects as after autonomic blockade, peak LV dP/d t did not change significantly after sotalol injection (Fig. 2b). It should be noted that control LV dP/dt was reduced by 33 f 7% following pretreatment with propranolol (0.5 mg/kg i.v.) in those animals studied after autonomic blockade. As such, the per- centage change in LV dP/d t after sotalol adminis- tration in these animals may be reduced because of the lower control value for LV dP/dt . Importantly however, all four anti-arrhythmic drugs were assessed in autonomically blocked animals and so assessment of comparative changes remains valid.

Allowing for this caveat, the changes seen after autonomic blockade were similar to those reported by Fitzgerald ef al. (1971) who found that sotalol (<20 mg/ kg) had no effect on contractility when adminis- tered in catecholamine depleted dogs. In contrast, Hoffmann and Grupp (1969) found that sotalol reduced peak LV dP/d t in reserpenized dogs suggesting that sotalol had intrinsic negative inotropic effects beyond its P-adrenoceptor blocking actions. The reason for this apparent discrepancy may relate to the different methods used to produce autonomic blockade in the different studies. In the present study, propranolol in a dose of 0.5 mg/kg was used, this being combined with methylscopolamine i.v. This dose of propranolol is higher than that used in other studies and may explain the more marked effect of cardiac autonomic blockade on peak LV dP /d t and HR seen in this study (Table 1). By comparison, Jose (1966) was able

186 N. Twidale et al.

to block totally the chronotropic response, and at least 90% of inotropic response to isoprenaline, using propranololO.2 mg/ kg i.v. (less than half that used in the present study).

Sotalol is known to cause bradycardia in humans (Brooks et al. 1970) even at doses of 0.2-0.6 mglkg and this was confirmed in the present study in which HR decreased by 10%. In support of the conclusion that the predominant effect on contractility results from its P-adrenoceptor blocking action was the finding that the fall in peak LV dP/dt was less (8 f 4% vs 13f3%) when HR was held constant by atrial pacing, but this difference was not statistically sig- nificant. Similar findings were reported in other animal studies (Puri & Bing 1969). By contrast, Hutton et al. (1972) studied patients during atrial pacing and con- cluded that sotalol reduced contractility even with constant pacing, but no comparative data were pre- sented.

Clinical implications

The results of the present study support the present trend towards use of anti-arrhythmic drugs with class 3 effect. While this in part relates to concerns of lethal pro-arrhythmia associated with flecainide and other class lc drugs reported by the Cardiac Arrhythmia Suppression Trial (CAST) investigators (Echt et al. 199 l), clinical experience with amiodarone and sotalol has been favourable, even in patients with severe left ventricular dysfunction (Amiodarone vs Sotalol Study Group 1989). While pro-arrhythmia may still occur with either of these two drugs, the incidence appears in general, to be less than that seen with class 1 anti- arrhythmic drugs (Counihan & McKenna 1990).

While sotalol had only a modest negative inotropic effect in this study, this was largely attributable to its 8-adrenergic blocking action. This capacity might potentially result in exacerbation of cardiac failure and cause hypotension in the patient with limited cardiac reserve whose sympathetic drive is stimulated maximally to compensate for poor cardiac perfor- mance (Mahmarian et al. 1987). By comparison, amiodarone has been used specifically in patients with severe left ventricular dysfunction and has been reasonably well tolerated when administered carefully by i.v. infusion (Remme et al. 1991) and with long- term oral therapy (Cleland et al. 1987).

The experimental model used in this study has obvious limitations. Most notably, the extrapolation of acute haemodynamic data obtained in animals with normal left ventricular function to predict results in patients with impaired cardiac reserve, may be inaccurate. The present series of experiments also do

not take into account the effects of periods of myo- cardial ischaemia which frequently occur in patients during episodes of cardiac arrhythmia. However, the conclusion that amiodarone and sotalol are less likely to be hazardous than either verapamil or flecainide in patients with compromised left ventricular function, is supported by clinical experience obtained over the past several years.

ACKNOWLEDGEMENTS

We are grateful for the expert technical assistance of Maria Kay and Benjamin Passehl, and to Shelly Baldwin and Lisa Bailey for preparing the manuscript. Dr N. Twidale held an Australian National Health and Medical Research Council Postgraduate Research Fellowship and Dr P. Roberts-Thomson was sup- ported by a National Health and Medical Research Council scholarship.

REFERENCES

Amiodarone vs Sotalol Study Group (1989) Multicentre randomized trial of sotalol vs amiodarone for chronic malignant ventricular tachyarrhythmias. European Heart Journal, 10,685-694.

Anderson, J. L., Stewart, J. R., Perry, B. A. et al. (1981) Oral flecainide acetate for the treatment of ventricular arrhythmias. New England Journal of Medicine, 305, 473-477.

Andreasen, F., Agerbaek, H., Bjerregaard, P. & Gotzshe, H. (1981) Pharmacokinetics of amiodarone after intra- venous and oral administration. European Journal of Clinical Pharmacology, 19,293-299.

Angus, J. A., Richmond, D., Dhumma-Upakorn, P., Cobbin, L. B. & Goodman, A. H. (1976) Cardiovascular action of verapamil in the dog with particular reference to myocardial contractility and atrioventricular con- duction. Cardiovascular Research, 10, 623-632.

Aylward, P. E. (1983) Cardiac performance in the hyper- tensive rabbit. PhD thesis, School of Medicine, Flinders University of South Australia (unpubl.).

Bourke, J, P., Cmmn, J. C., Tansuphaswadikul, S. & cunpbdl, B, w. FI (1987) Antiarrhythmic drug effects on&& veqt#cda.r Rerformance. Ewopenn Heart Journal,

B uced beta blockade in

Broughton, A. & Korner, P. I. (1980) Steady-state effects of preload and afterload on isovoIumetnc indices of con- tractility in autonomically blocked dogs. Cardiuvascular Research, 14,245-253.

Cleland, J. G., Dargie, H. J., FLedlr)., L N. & Wilson, J. T. (1987) Clinical, haemodynamic, and antiarrhythmic

Haemodynamic effects of anti-arrhythmic drugs 187

effects of long term treatment with amiodarone of patients in heart failure. British Heart Journal, 57, 436-445.

Cohen, A. A., Daru, V., Covelli, G., Gonzalez, M., Villa- mayor, R. & Tronge, J. E. (1985) Hemodynamic effects of intravenous flecainide in acute non-complicated myo- cardial infarction. American Heart Journal, 110, 1 193- 1196.

Cote, P., Bourassa, M. G., Delaye, J., Janin, A., Froment, R. & David, P. (1979) Effects of amiodarone on cardiac and coronary hemodynamics and on myocardial metab- olism in patients with coronary artery disease. Circu- lation, 59, 1165-1 172.

Counihan, P. J. & McKenna, W. J. (1990) Risk benefit assessment of amiodarone in the treatment of cardiac arrhythmias. Drug Safety, 5,286-303.

Cowan, J. C. & Vaughan Williams, E. M. (1981) Charac- terisation of a new oral antiarrhythmic drug flecainide R818. European Journal of Pharmacology, 73,333-342.

De Paola, A. A., Horowitz, L. N., Morganroth, J. et al. (1987) Influence of left ventricular dysfunction on fle- cainide therapy. Journal of the American College of Cardiology, 9, 163-168.

Dominic, J., McAllister, R. G., Kuo, C. S., Reddy, C. P. & Surawicz, B. (1979) Verapamil plasma levels and ven- tricular rate response in patients with atrial fibrillation and flutter. Clinical Pharmacology and Therapeutics, 26,710-714.

Dorward, P. K., Andresen, M. C., Burke, S. L., Oliver, J. R. & Korner, P. I. (1982) Rapid resetting of the aortic baroreceptors in the rabbit and its implications for short-term and longer term reflex control. Circulation Research, 50,428-439.

Duff, H. J., Roden, D. M., Maffucci, R. J. et al. (1981) Suppression of resistant ventricular arrhythmias by twice daily dosing with flecainide. American Journal of Car- diology, 48, 1133-1140.

Echt, D. S., Liebson, P. R., Mitchell, L. B. et al. (1991) Mortality and morbidity in patients receiving encainide, flecainide, or placebo. New England Journal of Medicine, 324,78 1-788.

Fitzgerald, J. D., Wale, J. & Austin, M. (1971) The haemo- dynamic effects of propranolol, dexpropranolol, ox- prenolol, practolol and sotalol in anesthetized dogs. European Journal of Pharmacology, 17, 123- 134.

Greenberg, S. & Wilson, W. R. (1974) Iproveratril: A non- specific antagonist of peripheral vascular reactivity. Canadian Journal of Physiology and Pharmacology, 52, 266-27 1.

Higgins, C. B., Vatner, S. F., Franklin, D. & Braunwald, E. (1973) Extent of regulation of the heart’s contractile state in the conscious dog by alteration in the frequency of contraction. Journal of Clinical Investigation, 52, 1187- 1194.

Hoffmann, R. P. & Grupp, G. (1969) The effects of sotalol and propranolol on contractile force and atrioventricular conduction time of the dog heart in situ. Chest, 55, 229-234.

Hoffmeister, H. M., Beyer, M., Muller, S. & Seipel, L.

(1989) Inotropic effects of class I and 111 antiarrhythmic drugs. Journal of Cellular Cardiology, 21 (Suppl. IV), S81.

Hondeghem, L. M. (1987) Antiarrhythmic agents: modu- lated receptor applications. Circulation, 75, 514-520.

Hutton, I., Lorimer, A. R., Hillis, W., McCall, D., Reid, J. & Lawrie, T. (1972) Haemodynamics and myocardial function after sotalol. British Heart Journal, 34,787-790.

Jackson, N., Verma, S. P., Frais, M. A. et al. (1985) Hemodynamic dose-response effects of flecainide in acute myocardial infarction with and without left ventricular decompensation. Clinical Pharmacology and Thera- peutics, 37, 6 19-624.

Jose, A. D. (1966) Effect of combined sympathetic and parasympathetic blockade on heart rate and cardiac function in man. American Journal of Cardiology, 18, 476-478.

Kopelman, H. A., Woosley, R. L., Lee, J. T., Roden, D. M. & Echt, D. S. (1988) Electrophysiologic effects of intra- venous and oral sotalol for sustained ventricular tachy- cardia secondary to coronary artery disease. American Journal of Cardiology, 61, 1 106-1 1 11.

Mahler, F., ROSS, J., O’Rourke, R. A. & Covell, J. W. (1975) Effects of changes in preload, afterload and inotropic state on ejection fraction and isovolumetric phase measurements of contractility in the conscious dog. American Journal of Cardiology, 35,626-634.

Mahmarian, J. J., Verani, M. S., Hohmann, T. et al. (1987) The haemodynamic effects of sotalol and quinidine: analysis by use of rest and exercise gated radionuclide angiography. Circulation, 76, 324-33 1.

Mason, D. T. (1969) Usefulness and limitations of the rate of rise of intraventricular pressure (dP/ dt) in the evalu- ation of myocardial contractility in man. American Journal of Cardiology, 23,5 16-527.

McRitchie, R. J., Blood, R. & Chalmers, J. P. (1984) Measurement of myocardial contractility in conscious rabbits. American Journal of Physiology, 246, H293-295.

Munoz, A., Karila, P., Gallay, P. et al. (1988) A randomized haemodynamic comparison of intravenous amiodarone with and without Tween 80. European Heart Journal, 9, 142- 148.

Nademanee, K., Feld, G., Hendrickson, J., Singh, P. N. & Singh, B. N. (1985) Electrophysiologic and antiarrhyth- mic effects of sotalol in patients with life-threatening ventricular tachyarrhythmias. Circulation, 72,555-564.

Nayler, W. G. & Horowitz, J. D. (1983) Calcium antagonists: A new class of drugs. Pharmacology and Therapeutics, 20,203-262.

Nayler, W. G. & Szeto, J. (1972) Effect of verapamil on contractility, oxygen utilization, and calcium exchange- ability in mammalian heart muscle. Cardiovascular Research, 6 , 120-128.

Nokin, P., Clinet, M. & Schoenfeld, P. (1983) Cardiac beta- adrenoreceptor modulation by amiodarone. Biochemical Pharmacology, 32,2473 -2477.

Packer, M., Leon, M. B., Bonow, R. O., Kieval, J., Rosing, D. R. & Subramanian, V. B. (1982) Hemodynamic and clinical effects of combined verapamil and propranolol

188 N. Twidale et al.

therapy in angina pectoris. American Journal of Cardi- ology, 50,903-912.

Platou, E. S. & Refsum, H. (1986) Acute electrophysiologic and blood pressure effects of amiodarone and its solvent in the dog. Acta Pharmacologica et Toxicologica, 58, 163-168.

Puri, P. & Bing, R. J. (1969) Effects on myocardial contractility, hemodynamics and cardiac metabolism of a new beta-adrenergic blocking drug, sotalol. Chest, 55, 235-239.

Remme, W. J., Kruyssen, H., Look, M., van Hoogenhuyze, D. & Krauss, X. (1991) Hemodynamic effects and tolerability of intravenous amiodarone in patients with impaired left ventricular function. American Heart Journal, 122,96-103.

Ruder, M. A., Ellis, T., Lebsack, C., Mead, R. H., Smith, N. A. & Winkle, R. A. (1989) Clinical experience with sotalol in patients with drug refractory ventricular arrhythmias. Journal of the American College Car- diology, 13, 145-152.

Schlepper, M. (1989) Cardiodepressive effects of anti- arrhythmic drugs. European Heart Journal, 10 (E), 73-80.

Serruys, P. W., Vanhaleweyk, G., Van Den Brand, M., Verdouw, P., Lubsen, J. & Hugenholtz, P. (1983) The haemodynamic effect of intravenous flecainide acetate in patients with coronary artery disease. British Journal of Clinical Pharmacology, 16,51-59.

Silke, B., Frais, M., Verma, S. et al. (1986) Comparative haemodynamic effects of intravenous lignocaine, diso- pyramide and flecainide in uncomplicated acute myo- cardial infarction. British Journal of Clinical Phar- macology, 22,707-714.

Singh, B. N., Jewitt, D. E., Downey, J. M., Kirk, E. S. & Sonnenblick, E. H. (1976) Effects of amiodarone and L8040, novel antianginal and antiarrhythmic drugs, on cardiac and coronary haemodynamics and on cardiac intracellular potentials. Clinical and Experimental Phar- macology and Physiology, 3,427-442.

Singh, S . N., Cohen, A., Chen, Y. et al. (1988) Sotalol for refractory sustained ventricular tachycardia and nonfatal

cardiac arrest. American Journal of Cardiology, 62, 399-402.

Tande, P. M. & Refsum, H. (1990) Class I11 antiarrhythmic action linked with positive hotropy: Acute electrophysio- logical and inotropic effects of amiodarone in vitro. Pharmacology and Toxicology, 66, 18-22.

Tande, P. M. & Refsum, H. (1988) Class I11 antiarrhythmic action linked with positive inotropy: Effect of the d-and 1-isomer of sotalol on isolated rat atria at threshold and subthreshold stimulation. Pharmacology and Toxicology, 62,272-277.

Vatner, S . F. & Braunwald, E. (1975) Cardiovascular control mechanisms in the conscious state. New England Journal of Medicine, 293,970-976.

Vaughnn Williams, E. M. (1970) Classification of anti- arrhythmic drugs. In: Symposium on Cardiac Arrhyth- mias, Elsinore, Denmark, April 23-25, 1970 (Eds E. Sand$e, E. Flensted-Jensen & K. H. Oelsen), pp. 449- 469. A. B. Astra, Sodertiilje.

Verdouw, P. D., Deckers, J. W. & Conard, G. J. (1979) Antiarrhythmic and haemodynamic actions of flecainide acetate (R-818) in the ischemic porcine heart. Journal of Cardiovascular Pharmacology, 1,473-486.

Wallenstein, S., Zucker, C. L. & Fleiss, J. L. (1980) Some statistical methods useful in circulation research. Circu- lation Research, 47, 1-9.

West, M. J., Angus, J. A. & Korner, P. I. (1975) Estimation of non-autonomic and autonomic components of iliac bed vascular resistance in renal hypertensive rabbits. Cardiovascular Research, 9,697-706.

White, S . W., Angus, J. A., McRitchie, R. J. & Porges, W. L. (1974) Evaluation of the Doppler flowmeter for measurement of blood flow in small vessels of unanes- thetized animals. Clinical and Experimental Pharma- cology and Physiology, 1,79-92.

Yabec, S. M., Kato, R. & Singh, B. N. (1986) Effects of amiodarone and its metabolite, desethylamiodarone, on the electrophysiologic properties of isolated cardiac muscle. Journal of Cardiovascular Pharmacology, 8, 197-207.