Comparison of Rate Versus Rhythm Control for Atrial Fibrillation in Patients With Left Ventricular Dysfunction (from the AFFIRM Study)

Comparison of Rate Versus Rhythm Control for Atrial Fibrillation in Patients With Left Ventricular Dysfunction (from the AFFIRM Study) Ronald S. Freudenberger, MD*, Alan C. Wilson, PhD, and John B. Kostis, MD, for the AFFIRM Investigators and Committees Optimal treatment for patients with atrial fibrillation (AF) and left ventricular (LV) dysfunction is not well defined. It is unclear if sinus rhythm is of greater benefit in patients with significantly reduced ejection fraction (EF) than in patients with normal or mildly depressed LV function. The Atrial Fibrillation Follow-up Investigation of Rhythm Management (AFFIRM) study compared 2 treatment strategies: “rhythm control,” attempting to maintain sinus rhythm, principally with antiarrhythmic drugs, and “rate control,” allowing AF to persist or recur while controlling the ventricular rate. We sought to determine if rhythm control was superior to rate control for patients in the AFFIRM study with various degrees of LV dysfunction. The present study analyzed outcome data of 3,032 subjects from the AFFIRM study with LV dysfunction by 3 EF strata: 40% to 49%, 30% to 39%, and <30%. The end points were mortality, hospitalization, and a change in New York Heart Association (NYHA) class. Analyses were done by intent to treat and by final rhythm status. In conclusion, there was no significant improvement in mortality, hospitalization, and NYHA class with the strategy of rhythm control in any of the 3 EF strata. When the data were analyzed by final rhythm status, we again found no significant benefit to patients in the rhythm control arm. © 2007 Elsevier Inc. All rights reserved. (Am J Cardiol 2007; 100:247–252)

The optimal treatment for patients who have heart failure and atrial fibrillation (AF) is poorly defined. It is unclear if those with various degrees of left ventricular (LV) dysfunction benefit from 1 approach versus another. Patients with heart failure and a significantly depressed ejection fraction (EF) may be more dependent on the atrial contribution to LV filling than those with relatively preserved LV function. This is a significant issue in that the prevalence of AF increases with the severity of heart failure. Several studies have examined the question of rate control versus rhythm control.1,2 The Atrial Fibrillation Follow-up Investigation of Rhythm Management (AFFIRM) study compared 2 treatment strategies: attempting to maintain sinus rhythm, principally with antiarrhythmic drugs (i.e., “rhythm control”), and allowing AF to persist or recur while controlling the ventricular rate, principally with atrioventricular node blocking drugs (i.e., “rate control”).3 The AFFIRM investigators found that more patients in the rhythm control group than in the rate control group were hospitalized, there were more adverse drug effects in the rhythm control group, and there was no effect on mortality. We analyzed data from the AFFIRM trial to determine whether those patients with more severe LV dysfunction might show an increased benefit of rhythm control. Department of Medicine, University of Medicine and Dentistry of New Jersey, Robert Wood Johnson Medical School, New Brunswick, New Jersey. Manuscript received December 15, 2006; revised manuscript received and accepted February 26, 2007. *Corresponding author: Tel: 732-235-7005; fax: 732-235-8149. E-mail address: [email protected] (R. Freudenberger). 0002-9149/07/$ – see front matter © 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.amjcard.2007.02.101

Methods Outcome measures were overall mortality, hospitalization, and change in New York Heart Association (NYHA) class defined as the difference between the baseline visit and the last follow-up visit. Subjects were analyzed by intent to treat and by final rhythm status. In analyses based on reported final rhythm status (rather than on randomized treatment assignment), patients were classified as being in sinus rhythm if the investigator recorded sinus rhythm at the final 2 study clinic visits; otherwise, patients were classified to AF final rhythm status. Univariate analyses of categorical variables were performed with the chi-square test, and analysis for variance was used for continuous variables. KaplanMeier survival curves for time to first event were constructed for the 3 strata and the log-rank test was used to compare the 2 treatment arms. Differences in change of NYHA class from baseline between treatment arms and between final rhythm status classes were compared using a nonparametric 2-sided Wilcoxon rank-sum test (stratified by EF). Patients whose NYHA class changed by ⱖ1 class were classified as “better” or “worse” at the end of the study compared with baseline. A “last visit carried forward” approach was used in which the last available assessment was used if the 1 for the final visit was missing because of withdrawal from the study or death.4 – 6 At enrollment in the AFFIRM trial, patients were ⱖ65 years of age or had other risk factors for stroke or death. A total of 4,060 patients were enrolled and baseline transthoracic echocardiograms were obtained in 3,311 patients: 1,650 in the rate control arm and 1,661 in the rhythm control www.AJConline.org

248

The American Journal of Cardiology (www.AJConline.org)

Table 1 Baseline characteristics of AFFIRM patients with EF ⬍50% by treatment randomization and by final rhythm status Stratum Variable Overall Age ⱖ 65 yrs Men Minority Sustained AF History of CAD History of HTN History HF Mild LV dysfunction (EF 40%–49%) No. of patients Age ⱖ 65 yrs Men Minority AF History of CAD History of HTN History of HF Moderate LV dysfunction (EF 30%–39%) No. of patients Age ⱖ 65 yrs Men Minority AF History of CAD History of HTN History of HF Severe LV dysfunction (EF ⬍30%) No. of patients Age ⱖ 65 yrs Men Minority AF History of CAD History of HTN History of HF

Rate Control Group (n ⫽ 380)

Rhythm Control Group (n ⫽ 408)

p Value*

Sinus Rhythm (n ⫽ 347)

AF (n ⫽ 415)

p Value†

69% 74% 17% 56% 56% 71% 52%

72% 75% 12% 52% 54% 70% 50%

0.5 0.8 0.02 0.4 0.6 0.9 0.8

65% 75% 16% 53% 56% 73% 52%

75% 46% 12% 55% 54% 69% 49%

0.002 0.8 0.2 0.6 0.6 0.3 0.4

192 78% 72% 15% 56% 51% 76% 34%

200 76% 73% 9% 52% 53% 74% 39%

— 0.6 1.0 0.1 0.4 0.6 0.6 0.4

163 72% 72% 15% 56% 55% 79% 37%

220 79% 75% 10% 51% 49% 72% 36%

— 0.1 0.5 0.2 0.4 0.2 0.2 0.9

107 67% 75% 16% 59% 60% 69% 58%

134 69% 76% 16% 55% 51% 69% 57%

— 0.7 0.8 1.0 0.6 0.2 1.0 0.9

114 61% 78% 15% 65% 62% 72% 60%

119 75% 76% 14% 60% 59% 68% 55%

— 0.02 0.7 0.9 0.4 0.3 0.5 0.4

81 53% 75% 25% 49% 65% 60% 84%

74 66% 77% 11% 49% 64% 62% 72%

— 0.5 0.8 0.02 0.4 0.6 0.9 0.8

70 54% 76% 19% 42% 67% 60% 76%

76 64% 79% 16% 55% 63% 63% 78%

— 0.2 0.6 0.7 0.1 0.6 0.7 0.8

* Rate control versus rhythm control. † Sinus rhythm versus AF. Comparisons using the chi-square test for rate vs rhythm and sinus rhythm versus AF (26 subjects with missing data). CAD ⫽ coronary artery disease any time before randomization (completed after abstraction of the medical record, as well as examination of the patient and direct interview of the patient and/or family); HF ⫽ history of congestive heart failure; HTN ⫽ hypertension.

arm. LV function was unknown in 279 of these patients and the remaining 3,032 patients were stratified according to EF: 2,244 normal (EF of ⬎50%), 392 with mild LV dysfunction (EF of 40% to 49%), 241 with moderate LV dysfunction (EF of 30% to 39%), and 155 with severe LV dysfunction (EF of ⬍30%). The prevalence of AF at the time of echocardiography ranged from 63% to 85% in these strata. The subjects of the present analysis are the 788 patients with various degrees of LV dysfunction, 380 in the rate control arm and 408 in the rhythm control arm. Results Baseline characteristics of the cohort in the present analysis are listed in Table 1. Apart from a higher proportion of minority patients with severe LV dysfunction being randomized to the rate control group and a greater number of

patients aged ⬎65 years with moderate LV dysfunction having AF as the final rhythm status, there were no significant differences between the 2 treatment strategies or final rhythm status classes. Not unexpectedly, a history of heart failure was more prevalent in the strata with lower EF levels. There was no significant difference in mortality between the rate control and rhythm control arms in any of the 3 EF strata (Figure 1). In the mild LV dysfunction group, there were 38 deaths among 192 subjects in the rate control arm compared with 42 deaths among 200 subjects in the rhythm control arm (p ⫽ 0.72). In the moderate LV dysfunction group, there were 25 deaths among 107 subjects in the rate control arm and 29 deaths among 134 subjects in the rhythm control arm (p ⫽ 0.73). In the severe LV dysfunction group, there was no evidence of benefit from rhythm control, with 30 deaths among 81 subjects in the rate control arm com-

Arrhythmias and Conduction Disturbances/AF Rhythm Control in Low EF

249

Figure 1. Kaplan-Meier survival rates in the rhythm control group (dashed line) and the rate control group (solid line) of AFFIRM trial subjects with measured LVEF: overall population (A), EF of 40% to 49% (B), EF of 30% to 39% (C), and EF of ⬍30% (D). LVD ⫽ LV dysfunction.

pared with 33 deaths among 74 subjects in the rhythm control arm (p ⫽ 0.39). There were significantly fewer nonfatal hospitalizations associated with rate control compared with rhythm control among subjects in the mild LV dysfunction group (Figure 2), but this relation was not seen in the analyses of patients with more severe LV dysfunction. In the mild LV dysfunction group, there were 116 nonfatal hospitalizations among the 192 subjects in the rate control arm compared with 140 among 200 subjects in the rhythm control arm (p ⫽ 0.04). Among subjects with moderate LV dysfunction, there were 72 hospitalizations among 107 subjects in the rate control arm and 92 hospitalizations among 134 subjects in the rhythm control arm (p ⫽ 0.18). In the severe LV dysfunction stratum (i.e., EF ⬍30%), there were 51 hospitalizations among 81 subjects in the rate control arm and 51 among 74 subjects in the rhythm control arm (p ⫽ 0.58). With regard to change in NYHA class from baseline (Figure 3), there was a trend toward fewer subjects’ NYHA class worsening and more subjects’ NYHA class improving among patients treated with rhythm control and those in whom sinus rhythm was maintained. Nonparametric analysis (results not shown) ranking the number of levels of NYHA class change relative to baseline revealed no significant improvement in NYHA class overall with rhythm control compared with rate control (Wilcoxon 2-sided rank sum test, p ⫽ 0.08). When analysis was repeated considering the final rhythm status, again no significant benefit appeared to be derived from a final status of sinus rhythm compared with AF (p ⫽ 0.09).

In the mild and moderate LV dysfunction strata, there was significantly more use of ␤ blockers and calcium channel blockers in the rate control arm compared with the rhythm control arm, with odds ratios of 2.5 (95% confidence interval [CI] 1.4 to 4.5) and 2.7 (95% CI 1.3 to 5.8), respectively, for ␤ blockers and 2.9 (95% CI 1.6 to 5.3) and 4.8 (95% CI 1.7 to 13.4), respectively, for calcium channel blockers (Table 2). However, among patients with severe LV dysfunction, the rates of therapy with ␤ blockers and calcium channel blockers were not statistically different between the rhythm control arm and the rate control arm (Fisher’s exact test, p ⫽ 0.06 for ␤ blockers and p ⫽ 0.50 for calcium channel blockers). Discussion Our findings demonstrate that, with regard to mortality, hospitalization, and NYHA class, there is no benefit of an antiarrhythmic-based rhythm control strategy compared with a strategy of rate control, even in patients with AF and moderate or severe LV dysfunction. This is a significant finding, in that heart failure affects ⬎50% of patients with AF,7 and 42% of patients with heart failure develop AF during their lifetime.8 Furthermore, previous data demonstrated increased total and heart failure mortality in those with concomitant AF.9 There are physiologic reasons to believe that those who would benefit most from rhythm control would be those with moderate or severe LV dysfunction. When AF occurs, the cardiac output decreases in the acute and chronic phases. The loss of atrial contribution

250

The American Journal of Cardiology (www.AJConline.org)

Figure 2. Cumulative hospitalization in the rhythm control group (dashed line) and the rate control group (solid line) of AFFIRM trial participants with measured LVEF: overall population (A), EF of 40% to 49% (B), EF of 30% to 39% (C), and EF of ⬍30% (D). LVD ⫽ LV dysfunction.

Figure 3. Change in NYHA class level for AFFIRM trial participants with complete NYHA data (n ⫽ 760; 22 participants were missing NYHA data). (A) Percent improved and worsened, overall and stratified by EF: dark bars represent those randomized to the rhythm control arm and open bars represent those treated with the rate control strategy. (B) Percent improved and worsened, overall and stratified by EF: dark bars represent participants in sinus rhythm at the last 2 follow-up visits and open bars represent those remaining in AF. Subjects with no change are not represented.

Arrhythmias and Conduction Disturbances/AF Rhythm Control in Low EF Table 2 Medication use at last AFFIRM clinic visit among AFFIRM patients with EF ⬍50% by treatment randomization and by LVD stratum Medication

Mild dysfunction (EF 40%–49%) No. of patients Digoxin alone ␤ blocker* Calcium channel blocker* ␤ blocker and calcium channel blocker combined* Moderate dysfunction (EF 30%–39%) No. of patients Digoxin alone ␤ blocker* Calcium channel blocker* ␤ blocker and calcium channel blocker combined* Severe dysfunction (EF ⬍30%) No. of patients Digoxin alone ␤ blocker* Calcium channel blocker* ␤ blocker and calcium channel blocker combined*

Rate Control Group

Rhythm Control Group

p Value

188 12 (6%) 22 (12%) 21 (11%) 22 (12%)

195 8 (4%) 13 (7%) 10 (5%) 8 (4%)

— 0.001 0.0003 —

105 1 (1%) 15 (14%) 9 (9%) 8 (8%)

128 7 (5%) 10 (8%) 3 (2%) 2 (2%)

— 0.008 0.0015 —

76 33 (4%) 8 (11%) 2 (3%) 4 (5%)

70 2 (3%) 3 (4%) 2 (3%) 1 (1%)

— 0.065 0.50 —

* With or without digoxin. p Value for difference in therapy with ␤ blocker, calcium channel blocker alone or in combination. Fisher’s exact test was used where cell numbers ⬍5.

to ventricular filling has been shown to decrease cardiac output and increase LV end-diastolic pressure.10 In addition, the presence of an irregular ventricular response has been shown to decrease cardiac output.11 One would therefore surmise that the patient at greatest risk for developing symptoms or hospitalizations from the combination of AF and heart failure would be the patient with the lowest EF. There are several limitations to these findings. This is a retrospective subgroup analysis of the main trial; therefore, the results should be interpreted with caution and considered hypothesis generating. An observed lack of statistical significance in a subgroup may be a result of lack of power because of small numbers rather than absence of a relation. The absence of a finding in this analysis does not necessarily support a strategy of rate control only. It is possible that there are improvements in functional parameters that were not adequately addressed by this analysis. It is also possible that the greater use of ␤-blocking agents in the rate control arm versus the rhythm control arm may have offset any potential beneficial effects of rhythm control. The beneficial effects of ␤ blockade in the heart failure population in terms of mortality, hospitalization, and functional capacity has been well described.12–14 Perhaps greater use of ␤-blocking drugs in this low EF population would effect the outcome of this analysis. Conversely, the negative inotropic affects of antiarrhythmic drugs may have adversely affected the out-

251

come in the rhythm control arm. The ongoing Atrial Fibrillation and Congestive Heart Failure trial is designed to answer many of these important questions.15 Our analysis does suggest that a strategy of rhythm control predominantly with antiarrhythmic therapy is not superior to rate control. Perhaps the use of nonpharmacologic approaches to maintain sinus rhythm rather than antiarrhythmic therapy will result in demonstrable improvements. The use of catheter ablation appears promising, but further study is needed.16 1. Hagens VE, Crijns HJ, Van Veldhuisen DJ, Van Den Berg MP, Rienstra M, Ranchor AV, Bosker HA, Kamp O, Tijssen JG, Veeger NJ, Van Gelder IC. 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. Am Heart J 2005;149:1106 –1111. 2. Torp-Pedersen C, Moller M, Bloch-Thomsen PE, Kober L, Sandoe E, Egstrup K, Agner E, Carlsen J, Videbaek J, Marchant B, Camm AJ. Dofetilide in patients with congestive heart failure and left ventricular dysfunction. Danish Investigations of Arrhythmia and Mortality on Dofetilide Study Group. N Engl J Med 1999;341:857– 865. 3. Wyse DG, Waldo AL, DiMarco JP, Domanski MJ, Rosenberg Y, Schron EB, Kellen JC, Greene HL, Mickel MC, Dalquist JE, Corley SD, AFFIRM Investigators. A comparison of rate control and rhythm control in patients with atrial fibrillation. N Engl J Med 2002;347: 1825–1833. 4. Hjalmarson A, Goldstein S, Fagerberg B, Wedel H, Waagstein F, Kjekshus J, Wikstrand J, El Allaf D, Vitovec J, Aldershvile J, et al. Effects of controlled-release metoprolol on total mortality, hospitalizations, and well-being in patients with heart failure: the Metoprolol CR/XL Randomized Intervention Trial in congestive heart failure (MERIT-HF). MERIT-HF Study Group. JAMA 2000;283: 1295–1302. 5. Colucci WS, Packer M, Bristow MR, Gilbert EM, Cohn JN, Fowler MB, Krueger SK, Hershberger R, Uretsky BF, Bowers JA, et al. Carvedilol inhibits clinical progression in patients with mild symptoms of heart failure. US Carvedilol Heart Failure Study Group. Circulation 1996;94:2800 –2806. 6. Packer M, Colucci WS, Sackner-Bernstein JD, Liang CS, Goldscher DA, Freeman I, Kukin ML, Kinhal V, Udelson JE, Klapholz M, et al. Double-blind, placebo-controlled study of the effects of carvedilol in patients with moderate to severe heart failure. The PRECISE Trial. Prospective Randomized Evaluation of Carvedilol on Symptoms and Exercise. Circulation 1996;94:2793–2799. 7. Jordaens L, Trouerbach J, Calle P, Tavernier R, Derycke E, Vertongen P, Bergez B, Vandekerckhove Y. Conversion of atrial fibrillation to sinus rhythm and rate control by digoxin in comparison to placebo. Eur Heart J 1997;18:643– 648. 8. Wang TJ, Larson MG, Levy D, Vasan RS, Leip EP, Wolf PA, D’Agostino RB, Murabito JM, Kannel WB, Benjamin EJ. Temporal relations of atrial fibrillation and congestive heart failure and their joint influence on mortality: the Framingham Heart Study. Circulation 2003;107:2920 –2925. 9. Dries DL, Exner DV, Gersh BJ, Domanski MJ, Waclawiw MA, Stevenson LW. 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. Studies of Left Ventricular Dysfunction. J Am Coll Cardiol 1998;32:695–703. 10. Mukharji J, Rehr RB, Hastillo A, Thompson JA, Hess ML, Paulsen WJ, Vetrovec GW. Comparison of atrial contribution to cardiac hemodynamics in patients with normal and severely compromised cardiac function. Clin Cardiol 1990;13:639 – 643. 11. Daoud EG, Weiss R, Bahu M, Knight BP, Bogun F, Goyal R, Harvey M, Strickberger SA, Man KC, Morady F. Effect of an irregular ventricular rhythm on cardiac output. Am J Cardiol 1996; 78:1433–1436.

252

The American Journal of Cardiology (www.AJConline.org)

12. Poole-Wilson PA, Swedberg K, Cleland JG, Di Lenarda A, Hanrath P, Komajda M, Lubsen J, Lutiger B, Metra M, Remme WJ, et al, COMET Investigators. Comparison of carvedilol and metoprolol on clinical outcomes in patients with chronic heart failure in the Carvedilol Or Metoprolol European Trial (COMET): randomised controlled trial. Lancet 2003;362:7–13. 13. The Cardiac Insufficiency Bisoprolol Study II (CIBIS-II): a randomised trial. Lancet 1999;353:9 –13. 14. Krum H, Roecker EB, Mohacsi P, Rouleau JL, Tendera M, Coats AJ, Katus HA, Fowler MB, Packer M, COPERNICUS Study Group.

Effects of initiating carvedilol in patients with severe chronic heart failure: results from the COPERNICUS Study. JAMA 2003;289:712– 718. 15. Roy D. Rationale for the Atrial Fibrillation and Congestive Heart Failure (AF-CHF) trial. Card Electrophysiol Rev 2003;7:208 – 210. 16. Hsu LF, Jais P, Sanders P, Garrigue S, Hocini M, Sacher F, Takahashi Y, Rotter M, Pasquie JL, Scavee C, et al. Catheter ablation for atrial fibrillation in congestive heart failure. N Engl J Med 2004;351:2373– 2383.