Sotalol vs metoprolol for ventricular rate control in patients with chronic atrial fibrillation who have undergone digitalization: a single-blinded crossover study

Europace (2001) 3, 73–79
doi:10.1053/eupc.2000.0140, available online at http://www.idealibrary.com on Sotalol vs metoprolol for ventricular rate control in
patients with chronic atrial fibrillation who have
undergone digitalization: a single-blinded
crossover study
G. E. Kochiadakis, E. M. Kanoupakis, M. D. Kalebubas, N. E. Igoumenidis,
K. E. Vardakis, H. E. Mavrakis and P. E. Vardas
Cardiology Department, University Hospital of Heraklion, Crete, Greece Aims To compare the effects of sotalol and metoprolol on
(116 9 bpm for sotalol vs 125 11 bpm for metoprolol, heart rate, during isotonic (ITE) and isometric (IME) P<0·001). During isometric exercise, sotalol produced a exercise and daily activities, in digitalized patients with lower maximum heart rate than did metoprolol (113 22 vs 129 18 bpm, respectively). Both agents produced a lowermean heart rate than placebo over 24 h (P<0·001 for all), Methods and Results The study had a randomized,
while sotalol produced a lower mean heart rate than single-blinded, crossover design. Twenty-three patients metoprolol during the daytime (P<0·01).
with chronic atrial fibrillation received placebo for 4weeks, followed by a 4-week period of treatment with Conclusion Sotalol is a safe and effective agent for control
sotalol and metoprolol in random order. At the end of of heart rate in digitalized patients with atrial fibrillation.
each period, the patients were assessed with 24-h ECG Sotalol is superior to metoprolol at submaximal exercise, monitoring, a cardiopulmonary exercise test and a handgrip resulting in better rate control during daily activities.
(Europace 2001; 3: 73–79)
Both agents produced a lower heart rate than placebo at  2001 The European Society of Cardiology rest and at all levels of isotonic exercise (P<0·001) withoutaffecting oxygen uptake. Sotalol produced a lower heart Key Words: Chronic atrial fibrillation, sotalol, metoprolol,
rate than metoprolol only at submaximal exercise Introduction
limited by a range of adverse reactions, of which the most serious is their negative inotropic action.
Beta-blockers have been shown to be effective for the Sotalol is a non-selective beta-blocker that also has a control of heart rate at rest and during exercise in class III action (prolongation of action potentials).
patients with chronic atrial fibrillation[. Indeed, a This property could contribute to sotalol’s ability to recent study assumes that the combination of beta- prolong AV node refractoriness and thus to control the blocker and digoxin is the most effective regimen, re- ventricular response in chronic atrial fibrillation more flecting a synergistic effect on the atrioventricular (AV) effectively than a conventional beta-blocker. Further- node of the vagomimetic actions of the cardiac glycoside more, sotalol is considered to have a less negative and the prolongation of AV node refractoriness by inotropic action than conventional beta-blockers, and beta-blockers. However, the use of beta-blockers is this could be an additional reason for its administration.
However, in spite of its theoretical advantages, the superiority of sotalol over conventional beta-blockers in Manuscript submitted 21 June 2000, revised 11 October 2000, and controlling heart rate in patients with chronic atrial fibrillation has not been proven to date. The aim of this Correspondence: Prof. P. E. Vardas, MD, PhD (London), FESC, study was to assess the effects of sotalol at rest and FACC, Cardiology Department, Heraklion University Hospital, during programmed exercise and daily activities in a P.O. Box 1352, Stavrakia Heraklion, Crete, Greece. E-mail: group of patients with chronic atrial fibrillation who had  2001 The European Society of Cardiology undergone digitalization, and to compare the results metoprolol for each patient enrolled, as determined by a with those of a conventional beta-blocker, metoprolol.
decrease in resting heart rate to c70 bpm.
Sotalol was initiated at a dose of 40 mg twice daily.
The dose was titrated up in 40–80 mg increments every 48–72 h until the target reduction in heart rate was achieved. The final maintenance dose of sotalol was adjusted downwards if adverse effects were noted by the patients, or if the rate-corrected QT interval exceeded Twenty-three patients (13 men, 10 women, mean age 63 8 years) with atrial fibrillation for >1 month (de- Metoprolol therapy was initiated at a dose of 25 mg fined in this study as chronic atrial fibrillation) made up twice daily. The dose was titrated up in 25–50 mg the study group. These patients were selected from a increments every 48–72 h with the same aim. Treatment total population of 146 inpatients with chronic atrial with any antiarrhythmic agent except digoxin was fibrillation who were examined in the authors’ depart- stopped for at least five half-lives before the study ment in a 2-year period. Fifteen of them had a history of unsuccessful electrical attempts to restore sinus rhythm, and eight patients relapsed after electrical cardioversion, Before the start of the study, all patients underwent a even though they received propafenone to prevent atrial trial cardiopulmonary exercise test in order to familiar- fibrillation recurrence. All patients were carefully evalu- ize themselves with the equipment and the procedure.
ated by history, physical examination, ECG, chest X-ray Patients exercised on a calibrated, motor-driven tread- and echocardiography, and satisfied the following selec- mill (Max-1, Marquette, Milwaukee, USA) with an tion criteria: no history or signs of heart failure; absence incremental exercise workload using a modified Naugh- of severe valvular heart disease; no evidence of ischaemic ton protocol. A rhythm strip lasting 6 s was recorded at heart disease; no evidence of renal, hepatic, endocrine, rest (after standing for 2 min) and at the end of each pulmonary or neurological disease; no history of un- minute during exercise and until the fourth minute of toward reaction to any of the medications used in the recovery, in order to evaluate the heart rate, with a present study; and the ability to undergo a treadmill paper speed of 25 mm . s 1. A six-lead ECG was also recorded at a paper speed of 50 mm . s 1 at rest, at All patients were treated with digoxin and were given submaximal and at maximal exercise. The longest QT anticoagulation therapy with acenocoumarol (INR 2·5– interval in any of the six leads was recorded and 3·5). While receiving digoxin therapy, these patients averaged from the measurement of six consecutive were required to have a resting heart rate >90 bpm.
QRS-T complexes. The QT interval was corrected (QTc) All patients gave their consent to the protocol, which for rate using Bazett’s formula, as follows: QTc= was approved by the local Ethics Committee for Human QT/√RR, where RR is the preceding RR interval.
Blood pressure was measured with a cuff sphyg- momanometer at rest and at 2-min intervals during the Gas exchange analysis and determination of anaerobic The study had a randomized, single-blinded, crossover design. It consisted of an initial 4-week placebo treat- During testing, the patients breathed atmospheric air ment period (2 tablets twice daily), followed by random- through a low-resistance mask. The partial pressures of ization to either metoprolol or sotalol treatment for a respiratory O2 and CO2 were measured using a special 4-week period, and then (after a placebo washout for gas analyser (Oxycon A, Mijnhard). The signals were 1 week) the alternative medication for the same length processed through analogue-to-digital conversion for breath-by-breath gas exchange analysis. The gas analy- The digoxin treatment was maintained throughout the ser was recalibrated before each test. Gas exchange study. The dosage was adjusted until the serum concen- variables were measured continuously and averaged at tration came within the range 0·8–2·0 ng . ml 1 before 30-s intervals during the 2-min rest period and through- out the test. The variables measured included oxygen At the end of each treatment period, patients were uptake (VO2, ml . kg 1 min 1), respiratory exchange assessed with 24-h ECG monitoring, a maximal, ratio (RER, VCO2/VO2) and oxygen pulse (oxygen symptom-limited cardiopulmonary exercise test, an iso- uptake/heart rate). These parameters were determined at metric exercise test with a handgrip dynamometer, and submaximal exercise (speed 3 mph, 0% grade), at the evaluation of subjective wellbeing and adverse events.
anaerobic threshold and at peak exercise. The gas ex- change anaerobic threshold, determined as outlined by Beaver et al., was taken as the mean of estimations Before the study, an open-label titration phase estab- performed by two independent observers who were lished the optimal therapeutic dose of sotalol or unaware of the patient’s treatment or other data.
Ventricular rate control in chronic AF patients 75 To evaluate the ventricular response to isometric exer- cise, a sustained handgrip test was performed. First, the subjects tested their maximal voluntary contraction, then, after a break of at least 2 min, the patients were asked to maintain their contraction at 30% of maximum for 150 s under continuous ECG monitoring. Heart rate was determined at rest and every 30 s on 6 s rhythm Continuous data are summarized as mean standard deviation. The effect of the three different treatment Figure 1 Time course of heart rate during isotonic
schemes on heart rate changes over time was assessed exercise in patients with chronic atrial fibrillation receiv-
with repeated measures analysis of variance with two ing sotalol (), metoprolol () and placebo (). The
factors: one for temporal effect (at 24 levels for the changes during the first 8 min were chosen because all the
circadian variation, nine levels for the isotonic exercise exercise tests contained complete data for this duration.
During the early stages of exercise (until the fourth
test and six levels for the isometric) and one for treat- minute), the difference between sotalol and metoprolol is
ment effect (three levels). The isotonic exercise test apparent.
had nine levels (one resting and eight consecutive minutes’ heart rate recording) because all patients completed at least 8 min of exercise. For the isometric test, the six levels corresponded to one at rest and five at with mean values significantly lower than those of 30-s intervals during the test. In case of significant metoprolol during the first 4 min of exercise. By the fifth findings, post hoc tests were also performed. Separate minute, the heart rate on sotalol had almost reached the subanalyses were also performed to examine the day (7:00–23:00) and night (23:01–6:59) effect. A P value The mean exercise parameters and gas exchange vari- <5% was the criterion for significance in all statistical ables at different exercise levels are shown in The mean exercise duration was similar in all treatment groups. Both sotalol and metoprolol produced a lower mean heart rate than placebo at all exercise levels (P<0·001 for all). Sotalol gave a lower mean heart rate than metoprolol at submaximal exercise (P<0·01), All patients completed the study uneventfully. No whereas at other levels, there was no difference.
patient converted to sinus rhythm, but all maintained a There were no significant differences in blood pressure resting heart rate c70 bpm. Thirteen of them had no between the three treatment groups at all exercise levels.
evidence of underlying heart disease, eight had hyperten- VO2 was similar in all three groups, at all exercise sive cardiovascular disease (signs of left ventricular hypertrophy on the echocardiogram or ECG), and two patients had surgically corrected mitral stenosis. All patients were in NYHA class I or II, and their mean Ventricular rate changes during isometric The mean drug dosages were 206 37 mg . day 1 for sotalol and 182 44 mg. day 1 for metoprolol.
Repeated measures ANOVA showed that there were also significant time, group and interaction effects (P<0·001 for all) on changes in heart rate during iso- Ventricular rate changes during isotonic For all three treatments, heart rate followed an in- creasing course during the first 90 s and then remained According to ANOVA with repeated measures, there at a plateau for the following 60 s. Heart rate was much were time, group and interaction effects (P<0.001) on higher under placebo than under the other two treat- heart rate changes during exercise. For placebo, the ments throughout the period of observation (quadratic exercise curves showed a curvilinear course, with an trend), with a steeper rise during the first 30 s. Sotalol initial sharp rise followed by a gradual increase. Meto- and metoprolol had a similar pattern of changes, more prolol gave a similar pattern, but the values were lower or less linear. However, although heart rate started out throughout the period of observation. For sotalol, the at the same level, it quickly diverged, increasing at a exercise curve showed a smoother, more linear path, slower rate under sotalol. The maximum heart rate Table 1 Exercise parameters and gas exchange variables at rest, submaximal
exercise, anaerobic threshold and maximal exercise in patients with chronic atrial
fibrillation under treatment with placebo, metoprolol and sotalol
*P<0·001 QT duration at exercise vs resting.
†P<0·01 sotalol vs metoprolol.
‡P<0·001 sotalol or metoprolol vs placebo.
BP, blood pressure; bpm, beats per minute; HR, heart rate; RER, respiratory exchange ratio; Ventricular rate control over 24 h Regardless of treatment, heart rate showed a significant circadian pattern, with higher values during the day than at night (time effect P<0·001 for all). However, the particular course of heart rate changes over the 24-h period was strongly treatment dependent (treatment and ffect P<0·001 for all). Under placebo, mean heart rate was higher throughout than in the other two groups, with a steeper rise in the morning until 13:00, then a steeper fall from the early afternoon until 23:00.
Sotalol and metoprolol had similar effects on mean HR, except during the early afternoon (13·00–16·00) when heart rate was higher in the metoprolol group Figure 2 Time course of heart rate during isometric
exercise in patients with chronic atrial fibrillation receiv-
ing sotalol (), metoprolol () and placebo (). There
are clear differences between the three treatments, in
terms of both the patterns of change and the values of
heart rate.
During the open-label phase, no patient suffered side- effects that would necessitate exclusion from the study.
Nor was it necessary to withdraw medication from any under placebo was 155 20 bpm, significantly higher patient during the study because of side-effects. There than that for metoprolol (129 18) bpm, P<0·001) and was no evidence of sotalol-associated torsade de pointes sotalol (113 22 bpm, P<0·001). The difference between or proarrhythmia during either the study or the open- metoprolol and sotalol was also significant (P<0·01).
Ventricular rate control in chronic AF patients 77 maintains a lower heart rate than does metoprolol. It is important to note that this lower heart rate under sotalol was accompanied by slightly higher levels of O2 con- sumption. A possible explanation of this is the different effects of the two agents on myocardial function, given that previous investigators have reported that sotalol has a lesser negative inotropic action than conventional beta-blockers, or even a positive inotropic effect on the myocardium. As a result, the more negative chrono- tropic effect of sotalol at that stage may be counter- balanced by a better use of the stroke volume reserve compared with metoprolol. The fact that oxygen pulse was significantly higher on sotalol than on metoprolol during only the initial stages of exercise, first of all Figure 3 Mean hourly ventricular rate over 24 h under
the three treatments. The circadian pattern is maintained
To the authors’ knowledge, there is only one study in in all three cases and the difference between the medi- which it was observed that the heart rate in patients with
cations is clear. , sotalol; , metoprolol; , placebo.
chronic atrial fibrillation increased at the start of an isometric exercise test and remained high throughout the The QT interval on sotalol was longer than on the test. This pattern corresponds to that reported from other two treatments at rest (P<0·001 for both). It previous observations in normal subjects, suggesting shortened significantly as the exercise level increased in that it is due to an initial vagal withdrawal and to the all three treatment groups, until at maximal levels there increased sympathetic tone which exists during isometric were no differences between the groups. At submaximal exercise.
exercise, however, the QT interval on sotalol was longer The present study showed that the increase in heart than on the other two treatments (P<0·001 for both, rate during isometric exercise in patients with chronic No significant changes were observed in the atrial fibrillation is large and digoxin is insufficient for corrected QT interval during exercise and there were no rate control. In contrast, agents with antiadrenergic differences between the groups either at rest or during properties seem to be able to control the heart rate response during isometric exercise in these patients.In this case, sotalol appears to be superior to conventional beta-blockers, causing a greater decrease in heart rate.
The most likely explanation for this is that the cardiac Discussion
stimulation during isometric exercise does not reach sufficient levels for the class III action of sotalol to Ventricular rate control at rest and during This study first of all confirms the findings of previous Ventricular rate control during daily studies, that sotalol, combined with digoxin, is a safe and very effective means of controlling heart rate, both at rest and during exercise, in patients with chronic atrial fibrillation. According to the present results during This study used 24-h Holter recordings to investigate the exercise, this therapeutic combination reduces heart rate effectiveness of the studied medications on ventricular at all levels, achieving a linear increase in heart rate and rate control during patients’ daily activities. The results eliminating the abrupt increase which is seen in atrial showed that during the day, when sympathetic tone is fibrillation during the early stages of exercise.It should increased, digoxin is insufficient for rate control in also be noted that, despite the significant reduction in patients with chronic atrial fibrillation. Sotalol, how- heart rate that sotalol causes, exercise performance was ever, was shown to be effective in achieving this aim, and not significantly affected. As oxygen pulse, which reflects was superior to a conventional beta-blocker, especially stroke volume, increases under sotalol, this suggests that during the daytime. The most likely explanation of this the study population were able to use their stroke is that during daily activities the patients do not reach volume reserve to compensate for the negative chrono- maximum exercise levels and so sotalol retains its class tropic effects of sotalol. The study’s major finding, III action.
however, was that the effects of sotalol on exercise show certain differences from those of conventional beta- blockers, such as metoprolol, that are targeted at low levels of exercise. Specifically, while the two agents have the same effect on exercise capacity and heart rate at Both drugs were administered after the patients had maximal levels, in the early stages of exercise, sotalol been digitalized. Digoxin has traditionally been used for control of heart rate in these patients, but its relative The present study suggests that this position could be importance has declined because, as the present findings filled by sotalol, a non-selective beta-blocker that also show, although it reduces the resting heart rate, it fails to has a class III effect. According to the results of this control exercise-induced tachycardia. How- study, this agent, combined with digoxin, is safe and is ever, the administration of digoxin in combination with more effective in controlling heart rate than conven- other medications has been proved to produce a greater tional beta-blockers in patients with chronic atrial fibril- decrease in heart rate than either digoxin or the lation during daily activities, when the patient does not additional agent alone. Furthermore, it allows for the reach maximal levels of exercise. Further studies are use of smaller dosages of the drugs and thus fewer needed to compare the effectiveness of sotalol in control- ing ventricular rate in patients with chronic atrial fibril- lation, with that of other treatments, such as calcium The present study showed that both sotalol and meto- References
prolol are relatively safe when used for rate control in [1] David D, Segni ED, Klein HO, Kaplinsky E. Inefficacy of patients with chronic atrial fibrillation. In the case of digitalis in the control of heart rate in patients with chronic sotalol, however, previous studies have reported a sig- atrial fibrillation: beneficial effect of an added beta adrenergic nificant proarrhythmic effect, especially in patients with blocking agent. Am J Cardiol 1979; 44: 1378–82.
underlying heart disease. The fact that patients [2] Atwood JE, Sullivan M, Forbes S, et al. Effect of beta- with compromised left ventricular function were ex- adrenergic blockade on exercise performance in patients with chronic atrial fibrillation. J Am Coll Cardiol 1987; 10: 314–20.
cluded from the present study is likely to have had a [3] Matsuda M, Matsuda Y, Yamagishi T, et al. Effects of significant effect on the findings regarding the safety of digoxin, propranolol and verapamil on exercise in patients sotalol. Another important factor is that the drug was with chronic isolated atrial fibrillation. Cardiovasc Res 1991; administered in combination with digoxin, so that only [4] Khalsa A, Edvardsson N, Olsson SB. Effects of metoprolol on relatively small dosages needed to be given.
heart rate in patients with digitalis treated chronic atrial fibrillation. Clin Cardiol 1978; 1: 91–5.
[5] Atwood JE, Myers J, Quaglietti S, Grumet J, Gianrossi R, Umman T. Effect of betaxolol on the hemodynamic, gas exchange, and cardiac output response to exercise in chronic atrial fibrillation. Chest 1999; 115: 1175–80.
The fact that these patients had good left ventricular [6] Farshi R, Kistner D, Sarma JS, Longmate JA, Singh BN.
Ventricular rate control in chronic atrial fibrillation during function must clearly have influenced the results con- daily activity and programmed exercise: a crossover open- cerning the effects of the two drugs on exercise perfor- label study of five drug regimens. J Am Coll Cardiol 1999; 33: mance. It is well known that conventional beta-blockers do not affect exercise capacity in atrial fibrillation [7] Hohnloser SH, Woosley RL. Sotalol. N Engl J Med 1994; patients with good left ventricular function, but reduce it [8] Anderson JL, Prystowsky EN. Sotalol: An important new in patients with atrial fibrillation and impaired left antiarrhythmic. Am Heart J 1999; 137: 388–409.
ventricular function. Given the weaker negative inotro- [9] Kowey PR, Marinchak RA, Rials SJ, Bharucha D. Pharma- pic action of sotalol, it seems likely that its advantages cologic and pharmacokinetic profile of class III antiarrhyth- over beta-blockers might become even more apparent in mic drugs. Am J Cardiol 1997; 80(8A): 16G–23G.
patients with compromised left ventricular function, [10] Beaver WL, Wasserman K, Whipp BJ. A new method for detecting anaerobic threshold by gas exchange. J Appl Physiol especially at low levels of exercise. Further studies will be necessary in order to investigate this question.
[11] Brodsky M, Saini R, Bellinger R, Zoble R, Weiss R, Powers L. Comparative effects of the combination of digoxin and dl-sotalol therapy versus digoxin monotherapy for control of ventricular response in chronic atrial fibrillation. dl-Sotalol Atrial Fibrillation Study Group. Am Heart J 1994; 127: [12] Singh S, Saini RK, DiMarco J, Kluger J, Gold R, Chen YW.
Currently, there are three major classes of pharmaco- Efficacy and safety of sotalol in digitalized patients with logical agents that are used for controlling the ventricu- chronic atrial fibrillation. The Sotalol Study Group. Am J lar rate in patients with chronic atrial fibrillation: digitalis glycosides, beta-blockersand cal- [13] Hsieh MH, Chen SA, Wen ZC, et al. Effects of antiarrhythmic drugs on variability of ventricular rate and exercise perfor- mance in chronic atrial fibrillation complicated with ventricu- patients, adequate rate control either cannot be achieved lar arrhythmias. Int J Cardiol 1998; 64: 37–45.
with the range of drugs available, or requires such large [14] Lok NS, Lau CP. Oxygen uptake kinetics and cardiopul- doses of the drug in question that the adverse reactions monary performance in lone atrial fibrillation and the effects detract from the beneficial effects. Therefore, there is still of sotalol. Chest 1997; 111: 934–40.
[15] Nagayoshi H, Janota T, Hnatkova K, Camm AJ, Malik M.
a need for a safer, better tolerated and more effective Autonomic modulation of ventricular rate in atrial fibril- drug for ventricular rate control in these patients.
lation. Am J Physiol 1997; 272: H1643–9.
Ventricular rate control in chronic AF patients 79 [16] Hohnloser SH. Proarrhythmia with class III antiarrhythmic [21] Lang R, Klein HO, Weiss E, et al. Superiority of oral drugs: types, risks, and management. Am J Cardiol 1997; 80: verapamil therapy to digoxin in treatment of chronic atrial fibrillation. Chest 1983; 83: 491–9.
[17] Reiffel JA. Impact of structural heart disease on the selection [22] Roth A, Harrison E, Mitani G, Cohen J, Rahimtoola SH, of class III antiarrhythmics for the prevention of atrial fibril- Elkayam U. Efficacy and safety of medium- and high-dose lation and flutter. Am Heart J 1998; 135: 551–6.
diltiazem alone and in combination with digoxin for control of [18] Van den Berg MP, De Langen CD, Crijns HJ, et al. Effect of heart rate at rest and during exercise in patients with chronic metoprolol on atrial fibrillatory rate, atrioventricular nodal atrial fibrillation. Circulation 1986; 73: 316–24.
concealed conduction, and ventricular response during atrial [23] Lang R, Klein HO, Di Segni E, et al. Verapamil improves fibrillation in pigs. J Cardiovasc Pharmacol 1994; 23: 846–51.
exercise capacity in chronic atrial fibrillation: double-blind [19] Lewis RV, Irvine N, McDevitt DG. Relationships between crossover study. Am Heart J 1983; 105: 820–5.
heart rate, exercise tolerance and cardiac output in atrial [24] Steinberg JS, Katz RJ, Bren GB, Buff LA, Varghese PJ.
fibrillation: the effects of treatment with digoxin, verapamil Efficacy of oral diltiazem to control ventricular response in and diltiazem. Eur Heart J 1988; 9: 777–81.
chronic atrial fibrillation at rest and during exercise. J Am [20] Goldman S, Probst P, Selzer A, Cohn K. Inefficacy of ‘therapeutic’ serum levels of digoxin in controlling the ven- [25] Lundstrom T, Ryden L. Ventricular rate control and exercise tricular rate in atrial fibrillation. Am J Cardiol 1975; 35: performance in chronic atrial fibrillation: effects of diltiazem and verapamil. J Am Coll Cardiol 1990; 16: 86–90.

Source: http://europace.oxfordjournals.org/content/3/1/73.full.pdf