Electrophysiology. Atrial fibrillation (AF) is the most common sustained cardiac arrhythmia, and rhythm control using cardioversion

Transcription

1 Electrophysiology Rate control versus rhythm control for patients with persistent atrial fibrillation with mild to moderate heart failure: Results from the RAte Control versus Electrical cardioversion (RACE) study Vincent E. Hagens, MD, a Harry J.G.M. Crijns, MD, b Dirk J. Van Veldhuisen, MD, a Maarten P. Van Den Berg, MD, a Michiel Rienstra, MD, a Adelita V. Ranchor, PhD, c Hans A. Bosker, MD, d Otto Kamp, MD, e Jan G.P. Tijssen, PhD, f Nic J.G.M. Veeger, MSc, g and Isabelle C. Van Gelder, MD, a for the RAte Control versus Electrical cardioversion for persistent atrial fibrillation study group Groningen, Maastricht, Arnhem, and Amsterdam, The Netherlands Background This study was conducted to compare rate- and rhythm-control therapy in patients with persistent atrial fibrillation (AF) and mild to moderate chronic heart failure (CHF). Rate control is not inferior to rhythm control in preventing mortality and morbidity in patients with AF. In CHF, this issue is still unsettled. Methods In this predefined analysis of the RACE study, a total of 261 patients were in New York Heart Association (NYHA) classes II and III at baseline. These patients were analyzed. The primary end point was a composite of cardiovascular mortality, hospitalization for CHF, thromboembolic complications, bleeding, pacemaker implantation, and life-threatening drug side effects. Furthermore, quality of life was compared. Results After 2.3 F 0.6 years, the primary end point occurred in 29 (22.3%) of the 130 rate-control patients and in 32 (24.4%) of the 131 rhythm-control patients. More cardiovascular deaths, hospitalization for CHF, and bleeding occurred under rate control. Thromboembolic complications, drug side effects, and pacemaker implantation were more frequent under rhythm control. Quality of life did not differ between strategies. In patients successfully treated with rhythm control, the prevalence of end points was not different from those who were in AF at study end. However, the type of end point was different: mortality, bleeding, hospitalization for heart failure, and pacemaker implantation occurred less frequently. Conclusions In patients with mild to moderate CHF, rate control is not inferior to rhythm control. However, if sinus rhythm can be maintained, outcome may be improved. A prospective randomized trial is necessary to confirm these results. (Am Heart J 2005;149: ) See related Editorial on page 957. From the a Department of Cardiology, University Medical Center Groningen, Groningen, The Netherlands, b Department of Cardiology, University Hospital, Maastricht, The Netherlands, c Department of Cardiology, Northern Center for Healthcare research, Groningen, The Netherlands, d Department of Cardiology, Rijnstate Hospital, Arnhem, The Netherlands, e Department of Cardiology, VU University Medical Center, Amsterdam, The Netherlands, f Department of Cardiology, Academical Medical Center, Amsterdam, The Netherlands, and g Department of Cardiology, Trial Coordination Center, Groningen, The Netherlands. Submitted April 29, 2004; accepted November 13, Reprint requests: Isabelle C. Van Gelder, MD, Department of Cardiology, Thoraxcenter, University Medical Center Groningen, PO Box , 9700 RB Groningen, The Netherlands. i.c.van.gelder@thorax.umcg.nl /$ - see front matter n 2005, Elsevier Inc. All rights reserved. doi: /j.ahj Atrial fibrillation (AF) is the most common sustained cardiac arrhythmia, and rhythm control using cardioversion and prophylactic antiarrhythmic drugs is often applied. 1,2 Drawbacks of cardioversion are the low success rate and the risks of life-threatening proarrhythmia. The alternative, that is, rate control, is aimed at controlling the ventricular rate at rest and during exercise. Recent clinical trials showed that rate control was not inferior to rhythm control in the prevention of morbidity and mortality in patients with persistent AF. 3-6 Similar to AF, chronic heart failure (CHF) is an ever increasing condition with a significant impact on morbidity and mortality. 7 In CHF, there is an increased risk of AF development. 8 With the restoration and maintenance of sinus rhythm, left ventricular function, exercise capacity, and maximal oxygen consumption may improve because of the restoration of atrial contribution to ventricular filling, the restoration of a regular heart beat, and the reversal of tachycardiainduced cardiomyopathy However, antiarrhythmic drug treatment to maintain sinus rhythm is less effective

2 American Heart Journal Volume 149, Number 6 Hagens et al 1107 in CHF and may even worsen the prognosis. 12,13 A recent retrospective analysis showed that there was no difference in mortality after 2 years of rate-control or rhythm-control treatments in patients with congestive heart failure in predominantly New York Heart Association (NYHA) functional classes III and IV. 14 Despite this analysis, it is still unsettled whether adequate rate control suffices with AF in CHF or rhythm control should rather be encouraged in these patients. 15 In this predefined substudy, we compared rate and rhythm control in patients with persistent AF and mild to moderate CHF included in the RACE study for persistent AF. 3 Furthermore, the quality of life (QoL) was investigated. Methods Study design In the RACE study, 522 patients with persistent AF were included from June 1, 1998, until July 1, In the present study, only patients with CHF in NYHA functional classes II and III at inclusion were selected from the original population. A total of 261 patients were analyzed with 130 and 131 patients randomized to rate- and rhythm-control groups, respectively. A detailed description of the RACE design was recently published. 3,16 In short, rate control was achieved using negative chronotropic drugs including digitalis, h-blocker, and nondihydropyridine calcium channel blocker. Rhythm control consisted of serial electrical cardioversion with institution of antiarrhythmic drugs (sotalol, class IC drugs, or amiodarone). Class IC drugs were only prescribed in case of mild CHF in the absence of coronary artery disease or (previous) overt CHF. For CHF, NYHA class was ensured at baseline in all patients. Around electrical cardioversion, oral acenocoumarol or fenprocoumon (target International Normalized Ratio [INR] ) was given. If sinus rhythm was present at 1 month, anticoagulation drugs could be stopped or changed to 80 to 100 mg of aspirin daily. In patients younger than 65 years, aspirin was allowed. All other patients received oral anticoagulation drugs. The study was approved by the institutional review boards at each participating hospital, and all patients gave written informed consent. Primary end point The primary end point of the RACE study was the composite of cardiovascular death, hospitalization for CHF, thromboembolic complication, bleeding, pacemaker implantation, or severe adverse effects of antiarrhythmic drugs. The definition for the primary end point was published previously. 3 All components of the primary end point that occurred between randomization and July 1, 2001, were recorded, with a maximum of 3 years of follow-up. A committee of experts who were unaware of the treatment assignments adjudicated all possible end points. Afterwards, all electrocardiograms were reviewed by an experienced electrophysiologist. Additional analysis of the occurrence of cardiovascular mortality was performed in patients who did not use class IC drugs in follow-up. Quality of life assessment The Medical Outcomes Study Short-Form Health Survey (SF-36) questionnaire was used to assess QoL. 16,17 It contains items to assess physical health (general health perception, physical functioning, role limitations due to adverbial physical problems, and bodily pain) and mental health (social functioning, role limitations due to emotional problems and mental health, and vitality). For each of the 8 subscales, scores are transformed to a scale ranging from 0 to 100, with lower scores representing lower QoL. For the QoL analysis, only patients who completed the self-administered QoL questionnaire at baseline and at the end of follow-up were considered. Patients who died during the follow-up of the RACE study were not analyzed. At baseline and at the end of the study, QoL was compared between the rate- and rhythm-control groups. Statistical analysis The primary objective of the RACE study was to show the noninferiority of rate control as compared with rhythm control with respect to the occurrence of the primary end point. 3 For this substudy in patients with CHF, no power calculation was performed, and all comparisons between rate control and rhythm control were considered as descriptive rather than conclusive. In these analyses, all patients were taken into account in accordance with the assigned strategy (intention-totreat principle). Furthermore, the occurrence of primary end points (per patient first occurrence) was estimated by using the Kaplan-Meier method. Baseline descriptive quantitative variables were compared between groups using a 2-tailed t test for normally distributed variables and a Wilcoxon 2-sampled test for skewed distributed variables. For the qualitative variables (categorical or ordered), group differences were evaluated using a Fisher exact test or a m 2 test. Accordingly, baseline characteristics are given in mean F SD, median and range (min-max), or percentages. For these baseline comparisons, P b.05 was considered statistically significant. Results Patient characteristics A total of 261 patients with AF in NYHA functional classes II and III for CHF were included in this study (Table I). Except for slightly more patients with hypertension in the rhythm-control strategy, no other significant differences in clinical characteristics between the groups were present. Treatment After a mean follow-up of 2.3 F 0.6 years, 47 (36%) patients in the rhythm-control group were in sinus rhythm. Use of antiarrhythmic drugs and cardioversions were comparable to the main study. In the rate-control group, 18 (14%) patients were in sinus rhythm at the end of the study. The mean resting heart rate was significantly lower in the rhythm-control group (77 F 20 beat/min) than in the rate-control group (84 F 17 beat/min). The number of patients who received oral anticoagulant therapy during follow-up ranged from 91%

3 1108 Hagens et al American Heart Journal June 2005 Table I. Patient characteristics Rate control (n = 130) Rhythm control (n = 131) Age (y) 69 F 9 69 F 8 Male 85 (65) 85 (65) Total AF duration (d) 499 ( ) 562 ( ) Duration present 34 (1-399) 31 (2-376) episode of AF (d) Complaints of AF 111 (85) 113 (86) Fatigue 61 (55) 64 (57) Dyspnea 73 (66) 73 (65) Palpitations 42 (38) 35 (31) Heart rate at inclusion 91 F F 20 (beats/min) Blood pressure (mm Hg) Systolic 142 F F 21 Diastolic 84 F F 12 Underlying diseases Coronary artery disease Previous myocardial infarction Valvular disease Cardiomyopathy 18 9 NYHA class for heart 95/5 93/7 failure (II/III) History of hypertension History of chronic obstructive lung disease Diabetes mellitus History of 13 5 cerebrovascular accident No cardiovascular disease demonstrable Treatment at inclusiont Digitalis h-blockery Verapamil or diltiazem Class IC 5 12 antiarrhythmic drugs Sotalol Amiodarone 0 1 Angiotensin-converting enzyme inhibitor Diuretics Echocardiographic parameters Left ventricular 57 F 7 55 F 8 end-diastolic diameter (mm) Left ventricular end- 41 F 9 40 F 8 systolic diameter (mm) Fractional shortening (%) 29 F F 9 Left atrial diameter, 46 F 7 46 F 7 parasternal long axis (mm) Left atrial diameter, apical view (mm) 65 F 9 64 F 9 Data are presented as mean F SD, number (%) of subjects, median value (range), or percentage of subjects. TAll patients used combinations of these drugs. yexcluding sotalol. Table II. Incidence of the primary end point and its components according to the treatment group Rate control (n = 130) Rhythm control (n = 131) Primary end point 22.3 (29) 24.4 (32) Cardiovascular mortality 11.5 (15) 6.1 (8) Sudden death 4.6 (6) 3.1 (4) Progression of heart 2.3 (3) failure Thromboembolic 1.5 (2) complications Bleeding 4.6 (6) 1.5 (2) Hospitalization for heart 5.4 (7) 3.8 (5) failure Thromboembolic 6.2 (8) 8.4 (11) complications Bleeding 8.5 (11) 3.1 (4) Adverse effects of 0.8 (1) 5.3 (7) antiarrhythmic drugs Pacemaker implants 4.6 (6) No double counts within lines; between lines double counts are possible. Data are presented as % (n). to 95% in the rate-control group and from 86% to 99% in the rhythm-control group. Under rate control, 85% of the patients were on continuous oral anticoagulation during the whole study versus 66% of the rhythmcontrol patients. Outcome primary end point The primary end point occurred in 29 (22.3%) of the 130 patients in the rate-control group and in 32 (24.4%) of the 131 patients in the rhythm-control group (Table II) Figure 1. More patients died under rate-control treatment. Six patients in the rate-control group and 4 patients in the rhythm-control group died suddenly; 1 was on flecainide, 1 was on sotalol, 1 was on propafenone, and 1 did not use antiarrhythmic drugs. Under rate control, more hospitalizations for CHF and bleeding were observed. Thromboembolic complications, pacemaker implantations, and adverse effects of antiarrhythmic drugs were more frequent under rhythm control. A total of 14 of the 19 thromboembolic complications occurred under inadequate oral anticoagulant therapy (INR below 2.5). All patients with thromboembolic complication had one or more stroke risk factors. Major bleeding occurred nearly solely under an INR well above 3.5: 8 (73%) of the 11 patients under rate control and in all 4 (100%) patients under rhythm control. A total of 8 patients suffered from severe adverse effects of antiarrhythmic drugs. Under rate control, 1 patient had digitalis intoxication with bidirectional ventricular tachycardia. Under rhythm control, 3 patients had a third-degree atrioventricular (AV)-nodal block, 2 on amiodarone (both suffered from a total AV-nodal block

4 American Heart Journal Volume 149, Number 6 Hagens et al 1109 Figure 1 Event-free survival (%) Rate control Rhythm control No. at risk Time (y) Rate control Rhythm control Kaplan-Meier survival curves for event-free survival in the ratecontrol and rhythm-control groups. Table III. SF-36 QoL scores SF-36 subscale StrategyT Baseline Study end General health Rate 49 (17) 52 (18) Rhythm 51 (18) 52 (19) Physical functioning Rate 57 (24) 58 (22) Rhythm 58 (23) 60 (26) Physical role limitations Rate 37 (43) 50 (44)y Rhythm 41 (44) 46 (44) Bodily pain Rate 76 (23) 77 (23) Rhythm 80 (22) 79 (22) Mental health Rate 72 (18) 74 (16)y Rhythm 72 (18) 76 (18)y Social functioning Rate 74 (24) 80 (22)y Rhythm 76 (23) 78 (23) Emotional role limitations Rate 67 (43) 74 (39) Rhythm 68 (43) 67 (41) Vitality Rate 56 (21) 54 (20) Rhythm 57 (20) 60 (21) Data are presented as mean F SD. TNinety rate-control and 93 rhythm-control patients were analyzed. yp b.05 compared with baseline score. in sinus rhythm necessitating pacemaker implantation; 1 patient was known to be with a right bundle branch block), and 1 on flecainide (transient AV-nodal block directly after electrical cardioversion from AF; flecainide was discontinued). The patient on flecainide also used digoxin; the 2 patients on amiodarone did not use other AV-nodal blocking agents. One patient had a rapid hemodynamically significant AV conduction during atrial flutter on flecainide; 1 had suffered from exercise-induced ventricular tachycardia during inhospital loading, thereafter, flecainide was discontinued; 1 had drug-induced heart failure on flecainide; and 1 had sick sinus syndrome on amiodarone. The occurrence of cardiovascular mortality was evaluated for the subgroup of patients who did not use class IC drugs in follow-up. For the rate control group, 14 (11.9%) of the 118 patients suffered from cardiovascular death. In the rhythm control group, 5 (6.3%) of the 80 patients died of cardiovascular cause. These percentages do not differ significantly from those of the total group (Table II). Impact of chronic sinus rhythm in the rhythm-control group Post hoc analysis in the rhythm-control group showed that the eventual rhythm did not affect the incidence of end points. However, the distribution of the components of the primary end points was importantly different according to whether the patient had sinus rhythm or AF at the end of follow-up. Of the 47 patients with sinus rhythm at study end, 11 (23.4%) reached an end point versus 21 (25%) who fell back to AF. The components of the primary end point for patients in sinus rhythm compared with those in AF at the study end were the following: cardiovascular mortality, 0% versus 9.5%; progression of CHF, 2.1% versus 4.8%; thromboembolic complication, 10.6% versus 7.1%; bleeding, 0% versus 4.8%; adverse drug effects, 8.5% versus 3.6%; pacemaker implantation, 2.1% versus 6.0%, respectively. Quality of life QoL was evaluated in 90 patients randomized to rate control and in 93 patients randomized to rhythm control (Table III). At baseline and also at the end of the study, there were no differences between the groups. During follow-up, QoL improved slightly but significantly in 3 subscales under rate control and in 1 subscale under rhythm control. The change in QoL from baseline to study end was comparable for the rate- and rhythm-control strategies. Patients with AFrelated complaints at study inclusion showed improvement of QoL ( P b.05) irrespective of the randomized strategy. Furthermore, the presence of sinus rhythm at study end was a determinant of QoL improvement ( P b.05 for the scales of physical functioning, role physical limitations, and vitality). Discussion The present study shows that in patients with AF and mild CHF the occurrence of cardiovascular morbidity and mortality is comparable between those treated with rate control and those treated with rhythm control. However, we observed an important difference in the type of end points between both groups. There was a

5 1110 Hagens et al American Heart Journal June 2005 trend for a higher mortality and major bleedings under rate control. In addition, rhythm control was associated with excellent survival if sinus rhythm could be maintained. Of note, rhythm control did not ameliorate progression of CHF. For both treatment options, QoL was comparable. Mortality The impact of AF on survival remains controversial. Several studies in patients with CHF showed that AF did not increase mortality In contrast, others showed that AF was an independent mortality risk factor This discrepancy may relate to the fact that AF seems associated with an increased mortality only in mild to moderate CHF but not in severe CHF. 24 In the present study, mortality in patients with mild CHF was higher under rate control. By contrast, the present results suggest that chronic sinus rhythm under rhythm control is associated with excellent survival. Therefore, in patients with AF and CHF, restoring and maintaining sinus rhythm may reduce cardiovascular mortality and morbidity. This is currently investigated in the AF-CHF study. 15 Thromboembolic complications and bleeding Fatal and nonfatal bleedings occurred frequently and more often under rate control. Thromboembolic complications also importantly contributed to cardiovascular morbidity and mortality. Both end points were predominantly encountered at the moment that oral anticoagulation drug was inadequate (INR too high or too low, respectively). These findings elucidate that (inadequate) treatment with oral anticoagulation drugs importantly affects outcome in AF patients. In this respect, new anticoagulation drugs, such as oral thrombin inhibitors that possibly are as effective but safer and easier to dose, may improve prognosis of AF in CHF. 25 Antiarrhythmic drugs in CHF The use of prophylactic antiarrhythmic drugs significantly contributed to end points in the rhythm-control arm and obviously not in the rate-control arm. Severe adverse effects did not occur during sotalol treatment; although, 1 patient died suddenly while on sotalol. Class IC drugs were instituted only in the absence of coronary artery disease and (previous) overt CHF. Life-threatening adverse events were encountered in 4 patients on class IC drugs, 1 during inhospital institution of the drug. In addition, 2 patients died suddenly while being treated with class IC drugs. Subanalysis of the occurrence of cardiovascular mortality for patients who did not use class IC drugs showed percentages comparable for the whole study population with more events under rate control. Amiodarone was safe and only caused nonfatal bradyarrhythmias. This was expected because amiodarone does not have a negative impact on survival in patients with CHF. 26 In patients with AF and CHF, Deedwania et al 27 described a lower mortality rate in patients who were in AF at baseline and converted to sinus rhythm on amiodarone compared with those who did not. The present and previous observations indicate that amiodarone is an effective and safe antiarrhythmic drug in patients with AF and CHF As has been recommended before, class IC drugs should not be used in any patient with CHF. 29 Quality of life No differences were observed in QoL between patients randomized to rate or rhythm control at baseline. During long-term treatment in both groups, QoL did not change importantly. Thus, no differences in QoL between the rate control- and rhythm-control groups were present at the end of follow-up. This observation is in accordance with data from the AFFIRM and PIAF investigators who could also not demonstrate any difference in QoL and AF-related symptoms between the 2 treatment approaches in patients with moderate heart failure. 4,30 Like in the CTAF, we also found that sinus rhythm at the end of the study was the most important determinant of QoL improvement. 31 In the CTAF and in the present study, improvements were present for both physical and mental health scales indicating that sinus rhythm may improve exercise tolerance and sense of vitality. Thus, rhythm control may be beneficial in improving QoL if effective. Therefore, better means of maintaining sinus rhythm may have a major general impact on QoL in patients with persistent AF and mild CHF. Limitations The present study is a prespecified substudy of the RACE study. Because the study was not powered to find noninferiority, no firm conclusions can be drawn from this analysis. Furthermore, the degree of CHF was determined by assessment of the functional class for CHF. However, this is the first randomized comparison between rate and rhythm controls for patients with AF in the setting of mild and moderate heart failure. Larger trials are warranted to show whether rate control is indeed comparable to rhythm control in these patients. Conclusions In patients with persistent AF and mild CHF, rate control was not inferior to rhythm control. However, more patients died or were hospitalized for CHF under rate control. Pacemaker implantations and adverse effects of antiarrhythmic therapy were more frequent under rhythm control. Effective rhythm control may be beneficial for the prevention of severe cardiovascular

6 American Heart Journal Volume 149, Number 6 Hagens et al 1111 events. Whether this relates to the presence of sinus rhythm itself or the selection of patients (patients who maintain sinus rhythm are less sick) remains to be determined. References 1. Feinberg WM, Blackshear JL, Laupacis A, et al. Prevalence, age distribution, and gender of patients with atrial fibrillation. Analysis and implications. Arch Intern Med 1995;155: Van Gelder IC, Crijns HJGM, Tieleman RG, et al. Chronic atrial fibrillation. Success of serial cardioversion therapy and safety of oral anticoagulation. Arch Intern Med 1996;156: Van Gelder IC, Hagens VE, Bosker HA, et al. A comparison of rate control and rhythm control in patients with recurrent persistent atrial fibrillation. N Engl J Med 2002;347: Wyse DG, Waldo AL, DiMarco JP, et al. A comparison of rate control and rhythm control in patients with atrial fibrillation. N Engl J Med 2002;347: Hohnloser SH, Kuck KH, Lilienthal J. Rhythm or rate control in atrial fibrillation Pharmacological Intervention in Atrial Fibrillation (PIAF): a randomised trial. Lancet 2000;356: Carlsson J, Miketic S, Windeler JJ, et al. Randomized trial of ratecontrol versus rhythm-control in persistent atrial fibrillation. The Strategies of Treatment of Atrial Fibrillation (STAF) study. J Am Coll Cardiol 2003;41: Ho KK, Pinsky JL, Kannel WB, et al. The epidemiology of heart failure: the Framingham Study. J Am Coll Cardiol 1993;22:6A - 13A. 8. Maisel WH, Stevenson LW. Atrial fibrillation in heart failure: epidemiology, pathophysiology, and rationale for therapy. Am J Cardiol 2003;91: Van Gelder IC, Crijns HJGM, Blanksma PK, et al. Time course of hemodynamic changes and improvement of exercise tolerance after cardioversion of chronic atrial fibrillation unassociated with cardiac valve disease. Am J Cardiol 1993;72: Daoud EG, Weiss R, Bahu M, et al. Effect of an irregular ventricular rhythm on cardiac output. Am J Cardiol 1996;78: Grogan M, Smith HC, Gersh BJ, et al. Left ventricular dysfunction due to atrial fibrillation in patients initially believed to have idiopathic dilated cardiomyopathy. Am J Cardiol 1992;69: Tuinenburg AE, Van Gelder IC, Van Den Berg MP, et al. Lack of prevention of heart failure by serial electrical cardioversion in patients with persistent atrial fibrillation. Heart 1999;82: Flaker GC, Blackshear JL, McBride R, et al. Antiarrhythmic drug therapy and cardiac mortality in atrial fibrillation. The Stroke Prevention in Atrial Fibrillation Investigators. J Am Coll Cardiol 1992;20: Al Khatib SM, Shaw LK, Lee O Connor C, et al. Is rhythm control superior to rate control in patients with atrial fibrillation and congestive heart failure? Am J Cardiol 2004;94: The AF-CHF investigators. Rationale and design of a study assessing treatment strategies of atrial fibrillation in patients with heart failure: the Atrial Fibrillation and Congestive Heart Failure (AF-CHF) trial. Am Heart J 2002;144: Hagens VE, Ranchor AV, Van Sonderen E, et al. Effect of rate or rhythm control on quality of life in persistent atrial fibrillation. Results from the Rate Control Versus Electrical Cardioversion (RACE) study. J Am Coll Cardiol 2004;43: Ware Jr JE, Sherbourne CD. The MOS 36-item Short-Form Health Survey (SF-36). I. Conceptual framework and item selection. Med Care 1992;30: Carson PE, Johnson GR, Dunkman WB, et al. The influence of atrial fibrillation on prognosis in mild to moderate heart failure. The V-HeFT studies. Circulation 1993;87:VI Mahoney P, Kimmel S, DeNofrio D, et al. Prognostic significance of atrial fibrillation in patients at a tertiary medical center referred for heart transplantation because of severe heart failure. Am J Cardiol 1999;83: Crijns HJGM, Tjeerdsma G, De Kam PJ, et al. Prognostic value of the presence and development of atrial fibrillation in patients with advanced chronic heart failure. Eur Heart J 2000;21: Middlekauff HR, Stevenson WG, Stevenson LW. Prognostic significance of atrial fibrillation in advanced heart failure. A study of 390 patients. Circulation 1991;84: Dries DL, Exner DV, Gersh BJ, et al. Atrial fibrillation is associated with an increased risk for mortality and heart failure progression in patients with asymptomatic and symptomatic left ventricular systolic dysfunction: a retrospective analysis of the SOLVD trials. J Am Coll Cardiol 1998;32: Wang TJ, Larson MG, Levy D, et al. Temporal relations of atrial fibrillation and congestive heart failure and their joint influence on mortality: the Framingham Heart Study. Circulation 2003;107: Van Den Berg MP, Van Gelder IC, Van Veldhuisen DJ. Impact of atrial fibrillation on mortality in patients with chronic heart failure. Eur J Heart Fail 2002;4: The SPORTIF III investigators. Stroke prevention with the oral direct thrombin inhibitor ximelagatran compared with warfarin in patients with non-valvular atrial fibrillation (SPORTIF III): randomised controlled trial. Lancet 2003;362: The Amiodarone Trials Meta-Analysis Investigators. Effect of prophylactic amiodarone on mortality after acute myocardial infarction and in congestive heart failure: meta-analysis of individual data from 6500 patients in randomised trials. Lancet 1997;350: Deedwania PC, Singh BN, Ellenbogen K, et al. Spontaneous conversion and maintenance of sinus rhythm by amiodarone in patients with heart failure and atrial fibrillation: observations from the veterans affairs congestive heart failure survival trial of antiarrhythmic therapy (CHF-STAT). Circulation 1998;98: Roy D, Talajic M, Dorian P, et al. Amiodarone to prevent recurrence of atrial fibrillation. N Engl J Med 2000;342: The Cardiac Arrhythmia Suppression Trial (CAST) Investigators. Preliminary report: effect of encainide and flecainide on mortality in a randomized trial of arrhythmia suppression after myocardial infarction. N Engl J Med 1989;321: Gronefeld GC, Lilienthal J, Hohnloser SH. Symptoms and quality of life in patients suffering from atrial fibrillation with and without moderate heart failure: data from a prospective trial comparing rate versus rhythm control [Abstract]. Heart Rhythm 2004;1:S35 [Abstract supplement]. 31. Dorian P, Paquette M, Newman D, et al. Quality of life improves with treatment in the Canadian Trial of Atrial Fibrillation. Am Heart J 2002;143: