Author + information
- Received October 14, 2012
- Revision received December 27, 2012
- Accepted January 23, 2013
- Published online May 7, 2013.
- David G. Jones, MD⁎,†,
- Shouvik K. Haldar, MBBS⁎,†,
- Wajid Hussain, MB, ChB⁎,†,
- Rakesh Sharma, PhD⁎,†,
- Darrel P. Francis, MD†,
- Shelley L. Rahman-Haley, MD⁎,
- Theresa A. McDonagh, MD⁎,†,
- S. Richard Underwood, MD⁎,†,
- Vias Markides, MD⁎,† and
- Tom Wong, MD⁎,†,⁎ ()
- ↵⁎Reprint requests and correspondence:
Dr. Tom Wong, Heart Rhythm Centre and National Institute for Health Research Cardiovascular Biomedical Research Unit, Royal Brompton and Harefield NHS Foundation Trust, Sydney Street, London SW3 6NP, United Kingdom
Objectives This study sought to compare catheter ablation with rate control for persistent atrial fibrillation (AF) in heart failure (HF).
Background The optimal therapy for AF in HF is unclear. Drug-based rhythm control has not proved clinically beneficial. Catheter ablation improves cardiac function in patients with HF, but impact on physiological performance has not been formally evaluated in a randomized trial.
Methods In a randomized, open-label, blinded-endpoint clinical trial, adults with symptomatic HF, radionuclide left ventricular ejection fraction (EF) ≤35%, and persistent AF were assigned to undergo catheter ablation or rate control. Primary outcome was 12-month change in peak oxygen consumption. Secondary endpoints were quality of life, B-type natriuretic peptide, 6-min walk distance, and EF. Results were analyzed by intention-to-treat.
Results Fifty-two patients (age 63 ± 9 years, EF 24 ± 8%) were randomized, 26 each to ablation and rate control. At 12 months, 88% of ablation patients maintained sinus rhythm (single-procedure success 68%). Under rate control, rate criteria were achieved in 96%. The primary endpoint, peak oxygen consumption, significantly increased in the ablation arm compared with rate control (difference +3.07 ml/kg/min, 95% confidence interval: 0.56 to 5.59, p = 0.018). The change was not evident at 3 months (+0.79 ml/kg/min, 95% confidence interval: −1.01 to 2.60, p = 0.38). Ablation improved Minnesota score (p = 0.019) and B-type natriuretic peptide (p = 0.045) and showed nonsignificant trends toward improved 6-min walk distance (p = 0.095) and EF (p = 0.055).
Conclusions This first randomized trial of ablation versus rate control to focus on objective exercise performance in AF and HF shows significant benefit from ablation, a strategy that also improves symptoms and neurohormonal status. The effects develop over 12 months, consistent with progressive amelioration of the HF syndrome. (A Randomised Trial to Assess Catheter Ablation Versus Rate Control in the Management of Persistent Atrial Fibrillation in Chronic Heart Failure; NCT00878384)
Atrial fibrillation (AF), the commonest arrhythmia in humans, causes a substantial burden of morbidity on patients and cost on healthcare systems. In patients with heart failure (HF), AF becomes more common and imposes greater burdens (1–3). Prevalence of AF rises from 10% in mild toward 50% in severe HF (3). Coexistence of AF with HF is associated with increased hospital stay and mortality (4,5) and further increases the stroke risk from AF 3-fold (6).
Rhythm control with antiarrhythmic drugs has not been shown to confer benefit in randomized trials, whether in patients with (7) or without HF (8,9). The lack of benefit from antiarrhythmic drugs might reflect their poor (<50%) efficacy in maintaining sinus rhythm (7–9). The risk of adverse effects might outweigh the benefits of restoring sinus rhythm, and indeed most antiarrhythmics are contraindicated in HF. Nonrandomized studies of catheter ablation, which might avoid such problems, have shown improvements in cardiac function (10–13), exercise capacity (12), and quality of life (14) in patients with HF and AF.
We designed a clinical trial to test the ability of ablation-based rhythm control to improve objective cardiovascular function in patients with HF and persistent AF. Impairment of exercise intolerance is both a hallmark symptom of HF and an important indicator of long-term survival. Given that peak oxygen consumption (VO2) is a strong prognostic indicator (15–17), we chose a change in peak VO2 as the primary endpoint. For the control arm we chose the contemporary standard-of-care for persistent AF in HF, a rigorous rate-control strategy that is at least noninferior to pharmacological rhythm control (7).
The enrollment criteria were 18 to 80 years of age, persistent AF (>7 days), symptomatic HF (New York Heart Association functional class II to IV) on optimal HF therapy, and left ventricular ejection fraction (EF) ≤35%. Exclusion criteria included cardiovascular implantable electronic device insertion or cerebrovascular event within 6 months; coronary revascularization or atrioventricular nodal ablation within 3 months; reversible causes of AF or HF including thyroid dysfunction, alcohol, primary valvular disease, or recent major surgery; prior heart transplant or on urgent transplant waiting list; pregnancy; active malignancy; severe renal impairment; single chamber pacemaker and atrioventricular block; and contraindications to general anesthesia or oral anticoagulation. Optimal HF therapy was defined as taking or having tried angiotensin-converting enzyme inhibitor (or angiotensin blocker), beta-blocker, and other therapy as recommended by their HF specialist; having had therapy for >1 month; and symptoms for >3 months. Prior failure of rhythm control by electrical or pharmacological cardioversion was not a prerequisite for enrollment, and no pre-enrollment rate-control criteria were specified.
The study, approved by the local research ethics committee in December 2008, was conducted at Royal Brompton and Harefield hospitals between April 2009 and June 2012. Patients were consecutively screened and enrolled after referral from cardiology services based locally and at linked referring hospitals. All patients provided written informed consent.
Baseline assessment comprised history, physical examination, blood tests including B-type natriuretic peptide (BNP), radionuclide ventriculography, cardiopulmonary exercise testing, 2-dimensional echocardiography, Minnesota Living with Heart Failure Questionnaire (MLHFQ), 6-min walk test, and 24-h Holter electrocardiogram (ECG).
Cardiopulmonary treadmill exercise testing was performed with the modified Bruce protocol. Peak VO2 (ml/kg/min) was defined as the mean of the highest 2 consecutive values of 15-s averages of VO2. The ventilation/carbon dioxide (ventilatory efficiency) slope was obtained by linear regression analysis of the data acquired throughout the entire period of exercise (16,18).
Because ejection fraction varies between beats in AF, we avoided methods that involve selecting beats for EF calculation (19). We chose radionuclide ventriculography, which systematically averages across hundreds of heartbeats.
Randomization and masking
After baseline investigations, eligible patients underwent 1:1 randomization by computer-generated sequence, stratified for age (above and below 50 years) and known atrioventricular block. The study was open-label; however, those conducting cardiopulmonary exercise testing, blood assays, and imaging analysis were blinded to randomization. The VO2 results were investigator-blinded until completion of follow-up.
Patients received pharmacological therapy (beta-blockers and/or digoxin) targeted to achieve a mean heart rate (assessed by apical auscultation over 30 s) ≤80 beats/min at rest before and ≤110 beats/min after a 6-min walk (7,8). If rate-control criteria were not met at baseline or during follow-up, patients re-attended at 4-week intervals for repeat assessment and adjustment of drug therapy until targets were achieved. In patients with pacemakers, if the base rate (≤80 beats/min) was not exceeded, no additional medication was prescribed for rate control. Atrioventricular node ablation and pacing was not adopted as a protocol, because it had just been reported to be inferior to pulmonary vein isolation (20).
The procedure was performed under general anesthesia. Transesophageal echocardiography was performed to exclude left atrial thrombus and to guide transseptal puncture. Patients were heparinized to maintain the activated clotting time over 300 s. Atrial anatomy was reconstructed with the NavX mapping system with an AFocusII catheter (St. Jude Medical, St. Paul, Minnesota). Radiofrequency ablation was performed with a 3.5-mm irrigated-tip catheter (ThermoCool, Biosense Webster, Diamond Bar, California) and comprised the following stepwise strategy: 1) pulmonary-vein isolation; 2) linear ablation at the left atrial roof and mitral isthmus; and 3) ablation of left atrial complex fractionated electrograms guided by high-density multipolar mapping as previously described (21). If atrial tachycardia occurred, the protocol was terminated, and the tachycardia was mapped and ablated. If AF persisted, sinus rhythm was restored by external cardioversion, followed by cavotricuspid isthmus ablation. All linear lesions were checked, and further ablation was performed as required to achieve block.
During a 2-month blanking period post-ablation, recurrent sustained atrial arrhythmia was treated by electrical cardioversion. Antiarrhythmic drugs other than non-sotalol beta-blockers were stopped post-ablation unless there was a separate indication, such as ventricular arrhythmias, but could be used if required within the blanking period. Those who developed further atrial arrhythmias after blanking were offered a second ablation procedure. A maximum of three ablations could be performed during the study period.
Follow-up and endpoints
Patients were followed up at 3, 6, and 12 months; this commenced on the date of the first procedure in the ablation group and from the date of randomization (including medication adjustment) in the rate-control group. The primary endpoint, peak VO2, was defined at 12 months and also measured at 3 months. The secondary endpoints LHFQ score, BNP, and 6-min walk distance were defined at 3, 6, and 12 months; radionuclide EF was only re-measured at 12 months to minimize radiation (NCT00878384). Freedom from atrial arrhythmia after ablation was assessed by ECG rhythm documentation at 6 to 8 weeks and all follow-ups, plus 48-h ambulatory monitoring at 6 and 12 months. Patients with implanted devices had additional arrhythmia log interrogation at follow-up. Any post-blanking recurrence of atrial arrhythmia >30 s constituted procedural failure (22). Bi-atrial areas were measured by transthoracic echocardiography (apical 4-chamber view) at baseline and at 6 and at 12 months.
Sample size calculation
We designed the study so that it would be able to detect a change of 2 ml/kg/min in a sample whose SD of difference between successive VO2 measurements was 2.5 ml/kg/min, with 2-sided alpha 0.05, power of 80%, comparing ablation with rate control. The number of patients required was 25 in each group.
Continuous baseline variables are presented as mean ± SD. Categorical variables are presented as frequency/percentage and compared with the Fisher's exact or the chi-square test. Outcomes were assessed, on an intention-to-treat basis, by independent group comparison of absolute changes from baseline. Parametric data were analyzed by the Student t test and represented as mean and 95% confidence interval (CI); nonparametric/ordinal data were analyzed by the Mann-Whitney U test and represented as median and interquartile range. A 2-sided p < 0.05 was considered statistically significant. No correction was made for assessments made at multiple time points. Arrhythmia-free survival was analyzed by Kaplan-Meier method. All calculations were performed with SPSS (version 20, SPSS, Chicago, Illinois) and endpoint graphs were constructed in R (23).
A total of 101 patients were referred for participation, and 75 attended for baseline assessment. Fifty-two patients were randomized, 26 patients to each arm (Fig. 1). Baseline characteristics are summarized in Table 1. One patient in the ablation arm withdrew consent for ablation and continued existing therapy; 1 patient in the rate-control arm requested and underwent ablation after 4 months; both were analyzed by intention to treat. One patient in the ablation arm—with chronic lung disease, dilated cardiomyopathy, and a biventricular pacemaker-defibrillator in-situ—died 11 months post-ablation due to progressive cardio-renal failure. Device interrogation showed no AF recurrence.
At 12 months peak VO2 had increased by 2.13 (−0.10 to +4.36) ml/kg/min in the ablation arm, compared with a decrease (−0.94 [−2.21 to +0.32] ml/kg/min) in the rate-control arm (mean difference +3.07 ml/kg/min [0.56 to 5.59], p = 0.018) (Fig. 2; individual responses shown in Fig. 3). At 3 months there was a nonsignificant increase of VO2 in the ablation arm (mean difference +0.79 ml/kg/min [−1.01 to +2.60], p = 0.38).
The secondary endpoints are displayed in Figure 4. The LHFQ score improved (reduced) in the ablation arm, compared with rate-control, nonsignificantly at 3 months (p = 0.196) but significantly at 6 (p = 0.015) and 12 months: median −15.5 (interquartile range: −26.75 to −7.25) compared with −5 (−16 to +9) under rate control (p = 0.019). This reflected a change from 42 ± 23 and 49 ± 21 to 21 ± 19 and 41 ± 21 at 12 months in the ablation and rate-control arms, respectively. The BNP similarly showed nonsignificant decrease at 3 months (p = 0.132) but significant reduction at 6 (p = 0.038) and 12 months (p = 0.045) in the ablation arm: median −124 (−284 to 0) pg/ml, compared with −18 (−86 to +31) pg/ml for rate control. Six-min walk distance tended to increase in both groups toward 6 months. At 12 months, ablation produced a nonsignificant increase (p = 0.095) (median: +21 m [−51 to +89 m]) compared with a decrease under rate control (median: −10 m [−73 to +15 m]). The EF significantly increased in the ablation arm from 21.5 ± 8.3% at baseline to 32.8 ± 14.3% at 12 months (p < 0.001), a change of 10.9 ± 11.5%. In the rate-control arm, EF increased from 24.9 ± 7.2% at baseline to 30.2 ± 9.4% at 12 months (p = 0.003), a change of 5.4 ± 8.5%. Accordingly, EF showed a nonsignificant trend toward improvement in the ablation arm (mean difference +5.6%, 95% CI: −0.1 to +11.3, p = 0.055) compared with rate control.
Exercise time mirrored peak VO2 with nonsignificant change at 3 months in the ablation arm (mean difference: +54 s [−31 to 139 s], p = 0.205) but significant improvement by 12 months (+133 s [19 to 246 s], p = 0.023). Ventilation/carbon dioxide slope showed a nonsignificant reduction at 3 months (−3.45 [−7.33 to 0.44], p = 0.081) and 12 months (−1.68 [−6.50 to 3.14], p = 0.49). The proportion of patients achieving satisfactory ventilatory anaerobic threshold was analyzed to compare effort between groups at follow-up: a respiratory exchange ratio ≥1.05 was achieved in 17 of 26 ablation versus 16 of 26 rate control at baseline (p = 0.77) and 16 of 25 versus 15 of 26 at 12 months (p = 0.77). In the rate control arm, change in peak respiratory exchange ratio was +0.03 ± 0.07 at 3 months and +0.30 ± 0.12 at 12 months. By comparison, in the catheter-ablation arm the change was +0.01 ± 0.12 at 3 months (p = 0.39) and 0.01 ± 0.14 at 12 months (p = 0.50). Thus, there was no evidence of differential motivation to explain the observed differences in peak VO2.
The left atrial area decreased in the ablation arm at both 6 months (mean difference compared with rate control −4.96 cm2 [−7.23 to −2.68 cm2], p = 0.001) and 12 months (−6.22 cm2 [−9.17 to −3.27 cm2], p = 0.001). The right atrial area showed a similar decrease at 6 months (−5.44 cm2 [−8.33 to −2.61 cm2], p < 0.001) but not at 12 months (−2.62 cm2 [−5.99 to 0.74 cm2], p = 0.12) (Fig. 5).
At baseline, rate-control criteria were met in 14 of 26 (54%) patients. Rate-control drugs were changed in 12: 2 were started on a regimen of beta-blockers; 9 had beta-blockers increased; 4 were commenced on digoxin; and 1 had digoxin increased. By 3 months, 23 of 26 (88%) were rate controlled. At 12 months, 2 were in sinus rhythm (1 after defibrillation at 9 months; 1 after undergoing ablation); of the remainder, 23 of 24 (96%) were rate controlled.
Of 26 patients randomized, 25 underwent catheter ablation after 1 withdrew consent. Procedure duration was 333 ± 61 min; fluoroscopy was 80 ± 19 min; and ablation was 82 ± 20 min. All patients had conduction block of the pulmonary veins and roof; 24 of 25 (96%) had mitral isthmus block (12 requiring ablation within the coronary sinus); 22 of 25 (88%) had cavotricuspid isthmus block (not attempted in 2 due to concern with regard to prolonged anesthetic).
One patient had asymptomatic flutter at final follow-up after a single procedure. Another reverted to AF by 6 months but declined further ablation, given prior complications. Five patients had additional ablation procedures for recurrent arrhythmia during follow-up: 4 for atrial tachycardia, and 1 for AF (followed by a third procedure for atrial tachycardia). Mean time to first atrial arrhythmia recurrence was 155 (95% CI: 62 to 248) days. Direct current cardioversion was required during the blanking period in 8 patients, 3 of whom remained arrhythmia-free.
By intention-to-treat, Kaplan-Meier 1-year arrhythmia-free survival was 69% after a single ablation, off a regimen of antiarrhythmic drugs (Fig. 6). Mean arrhythmia-free survival time was 304 (95% CI: 264 to 345) days. After all procedures, 22 of 25 (88%) were in sinus rhythm without further atrial arrhythmias, 1 of whom was on a regimen of sotalol and amiodarone for ventricular arrhythmias. Single-procedure success was 72%, and multi-procedural success was 92%, excluding the patient who declined ablation.
There was 1 serious procedural complication: a steam-pop caused tamponade during cavotricuspid-isthmus ablation, requiring emergency pericardiocentesis and a sternotomy to repair a perforation at the atrioventricular groove. Other complications were 1 groin hematoma, 1 chest infection 2 weeks post-ablation, and 1 patient with post-procedural pulmonary edema, which resolved within 24 h. All these patients made a full recovery and completed follow-up.
This is the first randomized clinical trial of ablation-based rhythm control, versus rate control, for persistent AF in HF to focus on objective cardiopulmonary exercise capacity. We found that ablation was associated with improvement in the primary endpoint of peak VO2 as well as quality of life and neurohormonal status. Left atrial size was reduced by ablation, although EF—which has greater natural variability and showed overall improvement in both arms—showed only a nonsignificant increase. Overall, these results suggest that rhythm control by ablation is a more effective strategy than medical rate control.
Previous studies have found no advantage from drug-based rhythm control over rate control (8,9). The AFFIRM (Atrial Fibrillation Follow-up Investigation of Rhythm Management) data showed a trend toward favorability of rhythm control in those with HF (8), but results from the AF-CHF (Atrial Fibrillation and Congestive Heart Failure) trial showed no benefit over rate control (7). There is, however, evidence that sinus rhythm is beneficial, if it can be maintained (8,24). Although this might reflect self-selection, a role has been argued for a therapy that can maintain sinus rhythm with minimal long-term side effects. Because catheter ablation might be able to achieve this and had shown promise in nonrandomized studies in the HF population (12,14), the next step was to prospectively compare an ablation-based rhythm-control strategy directly against medical rate control (7) in a well-defined cohort with nonparoxysmal AF and systolic HF, which formed the basis for the ARC-HF (A Randomised Trial to Assess Catheter Ablation Versus Rate Control in the Management of Persistent Atrial Fibrillation in Chronic Heart Failure).
The primary endpoint, peak VO2, is well-established as a prognostic indicator in HF (15–17) and as an endpoint in clinical trials (25). Recently Swank et al. (26) have shown that modest increases in peak VO2 were associated with reduced mortality and hospital stay. For every 6% increase in VO2 at repeat cardiopulmonary exercise testing after 3 months, there was a 7% reduction in all-cause mortality. In our study the overall benefit was approximately 20% (mean: 19.9% higher than rate control, 95% CI: 3.9% to 35.9%, p = 0.016)—which might have favorable prognostic implications.
Previous studies have shown that reversion to sinus rhythm increases peak VO2 and that maintenance of sinus rhythm leads to later improvement after 1 to 2 years (27). In our study, it is unlikely that the exercise advantage of ablation arose solely from being in sinus rhythm during the exercise test alone, because the same high proportion of patients were in sinus rhythm at 3 months and 12 months, yet approximately three-quarters of the 12-month advantage developed only in those intervening 9 months: an additional increment of 2.55 ml/kg/min (95% CI: 0.71 to 4.39 ml/kg/min, p = 0.008). The most likely explanation for this is progressive systemic improvement in cardiovascular physiology, largely occurring long after sinus rhythm was restored.
Minnesota LHFQ score and BNP improved in the ablation arm, which became more visible as follow-up progressed. The LHFQ score, a validated measure of therapeutic efficacy, was improved by a magnitude similar to that previously associated with favorable prognostic outcomes (28). The reduction in BNP with ablation, previously demonstrated in a non-HF population (29), might also have prognostic implications, particularly because BNP levels seem not to be independently affected by rhythm itself in HF (30). Our study did not show significant differences in the 6-min walk distance, but this—being self-paced—is open to an additional degree of variation not present on cardiopulmonary exercise testing, the latter having more extensively documented relationship to long-term survival (15–17,31). Although the improvement in EF post-ablation was consistent with previous studies (10–14,20), some improvement—perhaps reflecting increased diastolic filling consequent to lower mean heart rate—was seen in the rate-control group, and thus overall EF showed only a numerical trend toward improvement; nevertheless in head-to-head comparison cardiopulmonary exercise capacity has stronger prognostic power in HF than in EF (15). Regression of left atrial dilation occurred post-ablation. Although this has been observed before and might contribute to reduced arrhythmia recurrence (32), this is a notable finding in a HF population where atrial dilation could be less reversible. Given time, the atrium has some potential to recover, despite this advanced substrate.
Maintenance of sinus rhythm was achieved in 92% of those who underwent ablation, with 72% single-procedure success, off a regimen of antiarrhythmic medication. The high success rate might reflect the increasing efficacy of modern protocols, building upon 50% single-procedure success as reported in 2004 (12). More recently the PABA-CHF (Pulmonary Vein Antrum Isolation vs AV Node Ablation With Biventricular Pacing for Treatment of Atrial Fibrillation in Patients With Congestive Heart Failure) study, with one-half the patients having paroxysmal AF, achieved 71% freedom from AF off drugs regimen at 6 months (20). Our protocol also involved multiple linear lesions including the cavotricuspid isthmus and ablation of abnormal electrograms as per contemporary practice (33), and the inherent long procedure times provided a long period during which pulmonary-vein reconnections could be identified and corrected, which might be an additional avenue that improved outcome (34). Unlike earlier studies of AF ablation, proven failure of rhythm-control therapy was not a prerequisite for entry to the ARC-HF trial. Over one-third of patients had never undergone any attempt to restore sinus rhythm, and thus the outcomes might reflect a more amenable AF cohort than historical studies.
Since commencement of the ARC-HF trial, a randomized trial with a different ablation protocol that delivered 50% sinus rhythm maintenance and with shorter follow-up (6 months) (35) reported no significant difference in the endpoints of EF, N-terminal pro-BNP, and quality of life. As the investigators commented, this rate of procedural success reduced power greatly. Additionally, EF might not be a sufficient comprehensive marker of physiological state, considering the greater prognostic power of integrated markers such as objective exercise capacity (16). A follow-up duration of only 6 months might also be insufficient for the effect of restoration of sinus rhythm to fully manifest. The ARC-HF trial helps put the Macdonald et al. (35) study in context: the difference in headline results between the studies might be not a contradiction but merely a manifestation of the markedly different number of patients required to detect an effect.
The ARC-HF trial was not designed to evaluate event endpoints: studies specifically designed to evaluate the prognostic impact of AF ablation are now underway and are scheduled to complete in 3 to 4 years, including the CASTLE-AF (Catheter Ablation Versus Standard Conventional Treatment in Patients With Left Ventricular Dysfunction and Atrial Fibrillation) and RAFT AF (A Randomized Ablation-based Atrial Fibrillation Rhythm Control Versus Rate Control Trial in Patients With Heart Failure and High Burden Atrial Fibrillation) trials. Those larger studies will help provide more information about procedural complications, which affected 4 of 25 patients in our study, and also about the longer-term success of ablation in maintaining sinus rhythm, which is likely to be lower.
Our cohort was a little younger than those in other clinical trials in AF and HF (7) and correspondingly had a lower rate of ischemic etiology, which might limit its generalizability. However, there were no constraints other than those listed in the methods; the study was registered prospectively, and all patients formally enrolled (36). This study intentionally only addressed patients who had HF symptoms and therefore is not informative about the potential for benefit in asymptomatic patients. We do not know the greater denominator beyond the 101 who were referred to the trial; many patients might have not been referred because of geography, lack of symptoms, aversion to invasive procedures, or disinclination to volunteer. Although randomized trials have weaknesses, they remain the gold standard for evaluating therapeutic choices (37).
No imputation was made for missing data in the primary analysis to account for the patient who died before final follow-up: at sensitivity analysis, a worst-case scenario imputed a peak VO2 value equal to the lowest resting VO2 in any patient, showed resulting mean benefit of catheter ablation of +2.63 ml/kg/min (95% CI: 0.03 to 5.23 ml/kg/min, p = 0.048).
We did not attempt to identify post-hoc predictors of individualized response, because this activity is generally futile and often misleading (36).
There were some minor imbalances in the baseline characteristics of patients, to be expected in any randomized study. However, in regression analysis, neither increased beta-blocker dose under rate control (p = 0.66) nor presence of aldosterone antagonists (p = 0.14) or cardiac resynchronization therapy (p = 0.56) at baseline significantly influenced the primary endpoint.
Although the ARC-HF trial complied with the minimal ECG monitoring for persistent AF ablation recommended by the international consensus statements of 2007 (22) and 2012 (38), silent episodes of paroxysmal AF might have been missed during follow-up. However, in this population with prior persistent AF, the presence of sinus rhythm on multiple visits is highly suggestive of a major reduction in arrhythmia burden.
The ideal ablation strategy in persistent AF, particularly if longstanding, remains unknown. We systematically applied an extensive ablation strategy in this challenging cohort to minimize arrhythmia recurrence (39).
The ARC-HF trial, a randomized trial of patients with persistent AF and HF, indicates that an ablation strategy—achieving maintenance of sinus rhythm in the majority—produces improvements within 12 months in symptoms, neurohormonal status, and objective physiological exercise capacity. Progressive improvement from 3 to 12 months implies that the effects reflect more than just sinus rhythm restoration, suggesting this method of rhythm control initiates a period of progressive systemic regression of the HF syndrome. Large-scale trials have commenced to assess the prognostic impact of catheter ablation-based rhythm-control of AF in HF: the encouraging pointers for now are that, under randomized controlled conditions, 2 powerful objective prognostic markers (VO2 and BNP) respond favorably.
The authors thank the late Professor Philip Poole-Wilson for his guidance in the design of this study. They also thank the data and safety monitoring board (Professor Kim Fox, Dr. Paul Oldershaw, Prof. Peter Collins); Winston Banya and the Clinical Trials and Evaluation Unit at Royal Brompton Hospital for their assistance with randomization and statistical analysis; Mr. Matthew Ockendon for plotting of graphical data; and our patients for their participation.
The clinical trial was supported by the National Institute for Health Research cardiovascular Biomedical Research Unit at the Royal Brompton and Harefield National Health Service Foundation Trust and Imperial College London.
Dr. Jones and Dr. Haldar have received research fellow support grants from St. Jude Medical UK. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- atrial fibrillation
- B-type natriuretic peptide
- confidence interval
- ejection fraction
- heart failure
- left atrium/atrial
- pulmonary vein isolation
- peak oxygen consumption
- Received October 14, 2012.
- Revision received December 27, 2012.
- Accepted January 23, 2013.
- American College of Cardiology Foundation
- Stewart S.,
- Murphy N.F.,
- Walker A.,
- McGuire A.,
- McMurray J.J.
- Dries D.L.,
- Exner D.V.,
- Gersh B.J.,
- Domanski M.J.,
- Waclawiw M.A.,
- Stevenson L.W.
- Wang T.J.,
- Larson M.G.,
- Levy D.,
- et al.
- Chen M.S.,
- Marrouche N.F.,
- Khaykin Y.,
- et al.
- Mancini D.M.,
- Eisen H.,
- Kussmaul W.,
- Mull R.,
- Edmunds L.H. Jr..,
- Wilson J.R.
- Francis D.P.,
- Shamim W.,
- Davies L.C.,
- et al.
- Clark A.L.,
- Poole-Wilson P.A.,
- Coats A.J.
- Calkins H.,
- Brugada J.,
- Packer D.L.,
- et al.
- ↵(2012) R: A Language and Environment for Statistical Computing (R Foundation for Statistical Computing, Vienna, Austria).
- Pedersen O.D.,
- Bagger H.,
- Keller N.,
- Marchant B.,
- Kober L.,
- Torp-Pedersen C.
- Swank A.M.,
- Horton J.,
- Fleg J.L.,
- et al.
- Gosselink A.T.,
- Bulsma E.B.,
- Landsman M.L.,
- Crijns H.J.,
- Lie K.I.
- Sacher F.,
- Corcuff J.-B.,
- Schraub P.,
- et al.
- Rienstra M.,
- Van Gelder I.C.,
- Van den Berg M.P.,
- Boomsma F.,
- Van Veldhuisen D.J.
- Mudge G.H.,
- Goldstein S.,
- Addonizio L.J.,
- et al.
- Beukema W.P.,
- Elvan A.,
- Sie H.T.,
- Misier A.R.,
- Wellens H.J.
- Knecht S.,
- Hocini M.,
- Wright M.,
- et al.
- MacDonald M.R.,
- Connelly D.T.,
- Hawkins N.M.,
- et al.
- Sen S.,
- Davies J.E.,
- Malik I.S.,
- et al.
- Calkins H.,
- Kuck K.H.,
- Cappato R.,
- et al.