Treatment of Atrial Fibrillation With Antiarrhythmic Drugs or Radiofrequency AblationCLINICAL PERSPECTIVE
Two Systematic Literature Reviews and Meta-Analyses
Background— Although radiofrequency catheter ablation (RFA) has evolved from an experimental procedure to an important treatment option for atrial fibrillation, the relative safety and efficacy of catheter ablation relative to that of antiarrhythmic drug (AAD) therapy has not been established.
Methods and Results— Two separate systematic reviews were conducted: one on RFA and the other on AAD to provide accurate and broadly representative estimates of the clinical efficacy and safety of both therapies in the treatment of atrial fibrillation. Electronic searches were conducted in EMBASE and MEDLINE from 1990 to 2007. For the RFA review, all study designs were accepted. For the AAD review, articles were limited to prospective studies on the following drugs of interest: amiodarone, dofetilide, sotalol, flecainide, and propafenone. Data were extracted by 1 reviewer, with a second reviewer performing independent confirmation of extracted data. Sixty-three RFA and 34 AAD studies were included in the reviews. Patients enrolled in RFA studies tended to be younger (mean age, 55 versus 62 years), had longer duration of atrial fibrillation (6.0 versus 3.1 years), and had failed a greater number of prior drug trials (2.6 versus 1.7). The single-procedure success rate of ablation off AAD therapy was 57% (95% CI, 50% to 64%), the multiple procedure success rate off AAD was 71% (95% CI, 65% to 77%), and the multiple procedure success rate on AAD or with unknown AAD usage was 77% (95% CI, 73% to 81%). In comparison, the success rate for AAD therapy was 52% (95% CI, 47% to 57%). A major complication of catheter ablation occurred in 4.9% of patients. Adverse events for AAD studies, although more common (30% versus 5%), were less severe.
Conclusions— Studies of RFA for treatment of atrial fibrillation report higher efficacy rates than do studies of AAD therapy and a lower rate of complications.
Received October 6, 2008; accepted April 28, 2009.
The past decade has witnessed radiofrequency catheter ablation (RFA) of atrial fibrillation (AF) evolve from an experimental procedure to an important treatment option for many patients with AF. Although ablation of AF is performed commonly worldwide, its proper place in treatment algorithms remains subject to debate. Although updated American Heart Association/American College of Cardiology/European Society of Cardiology (AHA/ACC/ESC) guidelines1 state that “catheter ablation is a reasonable alternative to pharmacological therapy to prevent recurrent AF in symptomatic patients with little or no left atrial enlargement (Class 2A, Level of Evidence; C),” this document also includes a flow chart that lists catheter ablation as a second-line therapy for AF rhythm control regardless of the AF subtype or patient substrate. The majority view appears to be that ablation should be reserved for patients who have failed 1 or more trials of antiarrhythmic drug therapy; however, some investigators and practitioners have championed ablation as first-line therapy for highly symptomatic patients with AF.2,3 These differing perspectives reflect uncertainties in the relative risks and benefits of ablation and antiarrhythmic drugs (AAD) for rhythm control in AF.
Clinical Perspective on p 349
Although there has been a large number of studies performed that report the safety and efficacy of catheter ablation of AF, most are either retrospective case series or single-center clinical trials. Reported success rates for a single procedure have varied widely from 29% to 85%. To date, there have been no large, prospective, randomized multicenter clinical trials performed that compare the relative safety and efficacy of catheter ablation versus antiarrhythmic drug therapy.
To address existing uncertainties, we systematically reviewed the literature for the best available evidence on safety and efficacy of both RFA and antiarrhythmic drug therapy for AF. We hoped (1) to provide accurate and broadly representative estimates of the relative safety and efficacy of these 2 treatment options; (2) to gain insights into how the development of new ablation technologies and strategies may have affected the outcomes of catheter ablation over time; and (3) to highlight the size, strengths, and limitations of currently available data concerning these 2 treatment options for AF.
Two separate literature reviews were conducted: the first on RFA for treatment of AF and the second on treatment with AAD. The AAD review focused on the 5 antiarrhythmic drugs that are currently recommended for use for treatment of AF: amiodarone, propafenone, dofetilide, flecainide, and sotalol.1 Abstracts and unpublished literature were not included.
We classified AF according to a recent consensus statement1 as follows: paroxysmal AF is recurrent AF that terminates spontaneously within 7 days. Persistent AF is defined as AF, which is sustained beyond 7 days, or lasting less than 7 days but necessitating pharmacological or electric cardioversion. Included within the category of persistent AF is “longstanding persistent AF,” which is defined as continuous AF of greater than 1-year duration.
To identify and retrieve all potentially relevant literature describing the outcomes of RFA for treatment of AF, we conducted a literature search with assistance of reference librarians and investigators trained in systematic review procedures in MEDLINE (via PubMed), EMBASE, and Current Contents. MEDLINE was searched from January 1, 1990, to January 1, 2007, using the following search strategy:
“Heart arrhythmia” OR tachycardia OR “atrial fibrillation” OR “atrial flutter” OR “atrial tachycardia” OR “sinus tachycardia” OR WPW OR “Wolff-Parkinson-White syndrome” OR “heart arrhythmia” [MeSH] OR ventricular OR “atrial fibrillation” [MeSH] OR “tachycardia, ventricular” [MeSH] OR “atrial flutter” [MeSH]
“Catheter ablation” [MeSH] OR “catheter ablation” OR “radiofrequency ablation”
1 AND 2
Limits: Human, publication dates January 1, 1990, to January 1, 2007, excluding reviews, letters, comments, or editorials. The languages were limited to English and the 5 major Western European languages (French, German, Italian, Spanish, and Portuguese).
The above search identified use of RFA for a variety of arrhythmias, and the subgroup of studies limited to patients with AF were selected from this search strategy.
The search of EMBASE was performed using a similar strategy to the MEDLINE search. To identify studies of antiarrhythmic drug therapy, the search terms were:
Atrial fibrillation/*drug therapy OR atrial flutter/*drug therapy OR ventricular tachycardia/*drug therapy OR antiarrhythmia agents/adverse events/*therapeutic
“Atrial fibrillation”/exp OR “tachycardia, ventricular”/exp OR “atrial flutter”/exp
“Amiodarone” OR “Pacerone” OR “Cordarone” OR “aratac” OR “dofetilide” OR “Tikosyn” OR “flecainide” OR “pocard” OR “eccrine” OR “propafenone” OR “Rythmol” OR “sotalol” OR “detapac” OR “betapace af”
1 AND 2 AND 3
The search spanned from January 1, 1990, to May 1, 2007. The languages were limited to English. Because the literature on AADs is more mature and thus more extensive, the relatively small additional information that might be provided by inclusion of non-English language data were not thought to justify the substantial incremental search and retrieval resources required to assess several thousand additional abstracts. The above search identified studies treating patients with multiple types of arrhythmias, and the subgroup of studies limited to patients with AF were selected from this search strategy.
Additionally, the Cochrane Library was searched for any recent systematic review of the subjects. A manual check of the reference lists of all accepted studies and of recent reviews and meta-analyses was performed to supplement the above searches and ensure optimal and complete literature retrieval.
The study attrition diagrams for the RFA and AAD reviews are presented in Figures 1 and 2⇓. From the RFA search, 1870 abstracts were reviewed. The complete studies related to the accepted abstracts were then screened on the basis of defined inclusion and exclusion criteria (Table 1). Ablation procedures in the included studies targeted the pulmonary veins with or without additional strategies such as roof line, flutter line, or superior vena cava isolation. Additionally, studies used different catheter tip sizes and techniques (such as navigation and irrigation). Because we planned to stratify the analyses by technique, we only included studies in which results were separable by technique. For studies in which the technique was unclear, we attempted to contact authors for clarification. After an initial screening of the abstracts, 537 citations were retrieved, from which 142 met all of the inclusion and exclusion criteria. Sixty-three of these accepted studies (Figure 1) included patients with AF and provided the data for this manuscript. These 63 primary AF studies were associated with an additional 25 related studies (separate reports of the identical patient population), which occasionally contributed additional data of interest to the database.
From the AAD search, 3382 abstracts were reviewed (Figure 2). A total of 628 citations were initially retrieved, from which 44 met all of the inclusion and exclusion criteria, as outlined in Table 2. These primary studies were associated with an additional 15 related studies (separate reports of the identical patient population). Thirty-four of these studies assessed patients with AF and were included in the AAD dataset.
A complete bibliography of the primary and related studies included in the RFA and AAD reviews is available online.
Assessment of Methodologic Quality
All the studies were assigned a level of evidence using the schema of evidence assignment developed by the Centre for Evidence-Based Medicine in Oxford, United Kingdom. In addition, randomized controlled trials (RCTs) were also given a quality score based on the Jadad4 scoring method.
Data Collection and Analysis
Data were collected on paper extraction forms by 1 investigator and independently verified by a second investigator. Discrepancies were reviewed by the 2 investigators and, when necessary, by the entire group. Most discrepancies in data extraction involved the assigning of ablation technique. When the technique was unclear or when investigators could not come to agreement on a particular data variable, authors were contacted for clarification.
The study, patient, and treatment data were summarized using simple counts and sample size-weighted means.
For the RFA analysis, efficacy outcomes were reported as success (disappearance of arrhythmia) after a single procedure or success after 1 or more procedures in the time frame stated within the study. Often, studies were unclear in reporting whether success was attained after 1 or multiple procedures. In these instances, we assumed that the rate was attained after more than 1 procedure. We also planned to extract data separately for patients on and off AADs; however, a number of studies did not provide sufficient detail to judge whether success was obtained with or without AADs. We therefore divided studies into 2 categories: those in which outcomes were reported off AADs, and all others, that is, those in which patients were on AADs or AAD use was uncertain. Thus, we report 4 types of success rates: single and multiple or uncertain number procedure success off of AADs, and single and multiple or uncertain number procedure success with or uncertain AAD therapy. We also calculated the proportion of patients having any new or recurrent arrhythmia after ablation as well as the proportion requiring repeat ablation procedures.
For the AAD analysis, we reported the proportion of patients having recurrence of AF, the proportion with development of a new arrhythmia, and the total proportion having any arrhythmia while on study medication.
For the safety analysis, commonly reported events were collected to calculate a mean rate weighted by sample size for each event.
Restricted maximum likelihood random-effects meta-analyses5,6 were conducted and used to construct 95% CIs for each of the RFA success rates described above as well as for the recurrence of arrhythmia, the success with AAD, and the rate of new arrhythmia for the AAD studies. This method takes into account study sample size and incorporates heterogeneity of effects estimates across the studies. Heterogeneity is defined as the variation between study effects greater than that attributed to sampling error. The outcomes in this meta-analysis were proportions (rates) that were not comparative or normally distributed, and any significance testing that might be done would not be done on the overall rates. Calculations were done using SAS software version 8.1 and SPSS software version 14.0.
Table 3 displays the study characteristics for both datasets. Of the 63 eligible studies in the RFA dataset, 9 were RCTs, 11 were prospective comparative studies, 31 were prospective single-arm studies, and 12 were retrospective series. In all, 19 of the 63 studies (30%) provided level I or II evidence and the remainder provided level III or IV evidence. Of the 9 RCTs, 2 had a Jadad quality score of 3 or greater and 7 had a score of 1 to 2. Two thirds of the studies have been published since 2004.
Of the 34 eligible AAD studies, 24 (71%) were RCTs. There was 1 additional comparative study that was nonrandomized. The other 9 studies were single-arm trials. There were no retrospective studies included in the dataset. Twenty studies (59%) were level I evidence. Two thirds were multicenter studies and the majority (17 of 34 studies) were performed in Europe. Only 4 eligible studies were published since 2004.
Table 4 displays baseline characteristics for patients in RFA and AAD studies. Statistical comparisons across the 2 groups of studies are not appropriate; however, certain qualitative observations of differences in study patients can be made. For both RFA and AAD studies, the majority of patients were male and the mean left atrial dimensions were normal to slightly enlarged. Compared with typical patients in AAD studies, patients enrolled in RFA studies tended to be younger (mean age, 55 versus 62 years) but also had longer duration of AF (6.0 versus 3.1 years) and had failed a greater number of prior drug trials (2.6 versus 1.7). A greater overall proportion of RFA patients (70%) than AAD patients (56%) had paroxysmal AF rather than persistent or long-standing persistent AF.
Reporting of concurrent comorbid disease was inconsistent in both datasets. Thus, it is difficult to assess the true proportion of patients with these conditions in the published literature. Raw weighted averages are displayed in Table 4.
We also assessed changes in baseline characteristics over time. The only characteristic that changed markedly over time was the proportion of patients with persistent AF in the RFA population. This increased from 12% for studies published in 2002 to 42.4% (220 of 898 patients in 14 treatment groups) in 2006.
Among the 34 AAD studies, there were 11 amiodarone treatment groups, 16 propafenone treatment groups, 9 sotalol treatment groups, 2 dofetilide treatment groups, and 7 flecainide treatment groups. There were 12 groups treated with placebo/other. The target dose was largely comparable among studies of the same drug. In the RFA studies, roughly half used an electroanatomic catheter navigation system. The use of electroanatomic navigation increased markedly from 2002 (14% of studies) to 2006 (72%). Irrigated ablation catheters were used in one third of studies overall with increased use in more recent eligible studies.
An important variable in the assessment of efficacy for arrhythmia treatment is the choice of monitoring. The majority of studies for both the AAD and the RFA treatments evaluated patients with ECGs performed at specified intervals during follow-up visits. A minority of studies (fewer than half of RFA studies and less than a quarter of AAD studies) used Holter or event monitors to assess arrhythmia recurrence between visits. Efficacy outcomes for RFA are displayed in Figure 3. Forest plots for single and multiple procedure success and for treatment success and recurrence rates on AAD are presented in Figures 4, 5, and 6⇓⇓.
Efficacy Outcomes for RFA
Procedure success was often defined by authors as lack of recurrence of arrhythmia during the follow-up period. The single procedure success rate of catheter ablation of AF off AAD therapy was 57% (50% to 64%) in 31 arms with 2800 patients (Figure 4). After multiple or uncertain number of procedures, the off-AAD success rate increased to 71% (65% to 77%) in 34 arms with 3481 patients (Figure 5). The ablation success rate was 77% (73% to 81%) in 3562 patients in 42 arms after multiple or uncertain number of procedures in patients on AAD therapy and 72% in 4786 patients in 52 arms after a single procedure on AAD therapy (Figure 3). The mean follow-up period was 14 months, with a range from 2 to 30 months. The blanking period (the immediate postablation time frame during which transient episodes of arrhythmia are not counted as recurrence) ranged from 2 to 12 weeks. However, three fourths of the studies did not report a blanking period.
Stratified analyses by procedure approach or use of additional technology did not reveal major differences. The only outcomes that varied greatly by technique were procedure and fluoroscopy times. In treatment arms in which ablations were performed with an electroanatomic mapping system, the mean procedure time in 35 arms with 3768 patients was 159 minutes (range, 135 to 183 minutes) versus 202 minutes (range, 171 to 233 minutes) for 2514 patients in 24 treatment arms in which electroanatomic mapping was not used. Similar results were found for fluoroscopy times, which averaged 33 minutes (range, 26 to 40 minutes) in 34 treatment arms using mapping systems and 59 minutes (range, 44 to 74 minutes) in 21 nonnavigated arms reporting this outcome. The variability of reporting formats combined with the relatively few prospective studies precluded meaningful stratification of outcomes by study design.
Efficacy Outcomes for AAD
The overall success rate (generally defined by authors as disappearance of arrhythmia during the follow-up period) for all drug treatment groups was 52% (95% CI, 47% to 57%) in 32 treatment arms with 3180 patients. Treatment success and AF recurrence are presented by drug in Figures 6 and 7⇓; there were insufficient treatment arms to meta-analyze results separately for dofetilide. By comparison, in 8 placebo treatment arms with 655 patients, the success of treatment was 24.9% (95% CI, 15% to 34%).
Amiodarone was the drug with the greatest efficacy over placebo. The odds ratio for success of amiodarone over placebo was 6.11 (95% CI, 4.37 to 8.5), higher than other classes of drugs by 20% to 100%. The paucity of head-to-head data precluded most direct comparisons of success between specific drugs. Only direct comparisons between amiodarone versus sotalol, propafenone versus sotalol, and flecainide versus propafenone were possible. Results were significant for the amiodarone versus sotalol comparison, with an odds ratio of 2.46 (95% CI, 1.88 to 3.23) in 3 studies with 934 patients. Indirect comparison of amiodarone versus propafenone was also possible (by comparing pooled success over placebo for both drugs), yielding an odds ratio of 2.04 (95% CI, 1.26 to 3.31).
The mean time to recurrence was variable across studies, ranging from 2 weeks to more than 8 months. In some studies, drug treatment significantly increased the time to recurrence as compared with placebo.7–9 The mean follow-up period for drug treatment studies was 12 months, with a range from 1 to 48 months.
Safety Outcomes for RFA
The most common complication associated with catheter ablation of AF was symptomatic or asymptomatic pulmonary vein stenosis (defined as a >70% reduction in lumen diameter), with an overall incidence of 1.6% (Table 5). Somewhat less common were cardiac tamponade (0.7%), pericardial effusion (0.6%), periprocedural stroke (0.3%), and periprocedural transient ischemic attack (0.2%). There were no procedure-related deaths reported among the studies meeting our inclusion criteria, although case reports of fatal procedural complications have been published.10 The overall mortality rate in all treatment groups, including those with several years of follow-up, was 0.7%.
Safety Outcomes for AAD
The types of adverse events assessed in AAD studies were different from those reported in radiofrequency studies, and there was greater variability among rates in different studies than was seen in the safety data on RFA. Overall mortality rate was 2.8%; however, treatment-related deaths occurred in 0.5% of patients. The most commonly reported events were gastrointestinal problems (6.5%). In 32 active treatment arms, 10.4% of patients discontinued AADs because of adverse events, whereas 13.5% discontinued AADs because of treatment failure in 12 treatment arms that reported this outcome. Discontinuation as the result noncompliance was 4.2% in 4 active treatment arms with 457 patients, too few studies to allow meaningful comparison of compliance by drug.
Heterogeneity and Outlier Studies
There was significant heterogeneity of efficacy outcomes among studies in both datasets. Such heterogeneity would be expected with the diversity of patient populations (including variability in AF subtype) and treatment settings (including variability of techniques and catheter types), particularly in the RFA studies. Despite the heterogeneity, the 95% CIs for efficacy were fairly narrow, as can be seen in Figure 3.
Outlier studies were examined to understand which characteristics might be contributing to the heterogeneity. For example, studies that had higher or lower than expected rates of recurrent AF were evaluated. Studies with longer-term follow-up, such as 36 months11 or 48 months12 in the AAD data set, had a nearly 68% recurrence rate. Also, AAD studies with stricter definitions of success13 (eg, not having a recurrence and not experiencing adverse events) had lower success rates.
Similar trends were found in the RFA dataset. Studies with longer follow-up duration were again associated with lower rates of success. As in the AAD results, studies reporting lower success had stricter definitions of success and screened for asymptomatic AF using prolonged ECG monitoring to confirm the lack of arrhythmia.14,15
To the best of our knowledge, this is the first systematic literature review and meta-analysis of clinical studies of RFA and AAD for the treatment of AF. The results of our study reveal that studies of RFA for treatment of AF report higher overall efficacy rates than do studies of AAD therapy as well as a lower rate of complications. It is noteworthy that catheter ablation of AF was associated with higher efficacy than antiarrhythmic drug therapy, despite the fact that most of the studies of AF ablation enrolled patients who had previously failed 1 or more trials of AAD therapy. At first, these findings might suggest that catheter ablation of AF should always be the preferred treatment strategy for AF; however, we urge caution, based on important differences in trial methodologies, patient characteristics, and the relative severity of complications resulting from catheter ablation of AF versus AAD therapy. A recent meta-analysis of RCTs yielded consistent results16 despite the fact that the datasets were nonoverlapping and of vastly different sizes (owing to differences in inclusion/exclusion criteria).
Changes Over Time
No clear differences in outcome were observed over time with modifications in ablation technologies and procedure strategies. This may be explained, at least in part, by the changing demographics of patients undergoing catheter ablation across the study period—in particular, a shift away from paroxysmal AF to persistent AF and a gradual rise in the mean age of the patients. We were unable to quantify the impact of specific RFA techniques or technologies because of a paucity of comparative trials. However, the finding that ablation success has remained steady over time despite the increasing proportion of older patients and of patients with persistent AF—in whom lower success rates might be expected—could be attributed to a positive impact of advances in ablation technology and/or to a flattening of the learning curve.
The results of this study should be of value both to clinicians considering RFA versus AAD therapy for their patients’ AF but also for researchers in the field of catheter ablation. From a clinical standpoint, the results of this meta-analysis are reassuring because they reveal that catheter ablation of AF appears to have considerable efficacy for treatment of AF with an acceptable risk of major complications and they support the current AHA/ACC/ESC recommendation that ablation should be considered as a second-line therapy in patients who are drug refractory.1 The data also point to a clinical reality, namely that these previously drug-refractory patients will sometimes require >1 procedure and/or continued AAD therapy to achieve a success rate >70%.
Although this meta-analysis reveals an enormous body of data to support the value of AF ablation, it also reveals important limitations of the clinical studies that have been performed to date. Only 30% of the 63 AF ablation studies that were included in this analysis provided level I or II evidence. Although 9 of these trials were randomized and controlled, only 1 of these involved >200 patients,17 and none had >2 participating centers. Further, it is unclear whether these published results from self-selected centers of excellence are obtainable in routine practice. These limitations in the evidence base undoubtedly relate to the difficulties in conducting large-scale trials of an invasive and evolving procedure but raise questions about the generalizability of the published literature on RFA relative to AADs.
A comparison of the number and types of complications resulting from RFA versus AAD therapy reveals the challenges in weighing the risks and benefits of the 2. Although at first consideration, the 30% complication rate for AAD therapy overwhelms the 5% complication rate associated with RFA, there are significant differences in the seriousness of the complications reported. It is difficult to compare cardiac tamponade, reported during 0.8% of ablation procedures, with gastrointestinal complications, neuropathy, and thyroid dysfunction, which were observed in 6%, 5%, and 3% of patients enrolled in AAD trials. Furthermore, because most of the ablation studies were single-center case series, it is likely that adverse events were reported with less rigor than in typical randomized controlled drug trials. For example, our meta-analysis found a <1% rate of vascular access site complications—including a 0% risk of hemorrhage—which we believe to be a significant underestimate.18
Our review also emphasized the need for greater uniformity of reporting both definitions of success and monitoring for recurrent arrhythmia (including data on compliance with monitoring) in AF studies going forward. The recently published HRS Consensus Statement on Catheter Ablation of Atrial Fibrillation should be helpful in this regard.3 In addition to the variability in reporting, the review also found a paucity of data on potentially important patient subgroups, such as the elderly or those with comorbidities.
Finally, for both AAD and RFA studies, it should be recognized that “success” as defined and measured in a clinical trial setting does not necessarily equate with clinical improvement. Although the carefully documented absence of recurrent AF is an objective and meaningful definition of treatment success, patients with occasional AF recurrences may not be considered as treatment failures clinically if the frequency, duration, and/or severity of the recurrences are markedly changed by the intervention. For this reason, consensus guidelines consider infrequent, well-tolerated recurrences of AF as a reasonable definition for success on drug therapy.1 Clinical trials should and increasingly are incorporating quality of life and other patient-reported outcomes to gauge clinical improvement.
Generalizability of Findings
The quality of all research synthesis is dependent on the studies included in the data synthesis, and such analyses are subject to the limitations imposed by the criteria set for inclusion in the review. The review included a broad range of study designs, treatment settings and ablation techniques, and study inclusion criteria. Conducting meta-analyses across disparate study settings could be of concern. However, including different settings and study designs in the meta-analysis reflected the disease in the population as a whole. Thus, the review provides an estimate of “average expectations” across treatment settings and techniques.
Although a planned prospective comparative study could more reliably address the scientific goals and the unanswered questions on relative efficacy and safety of different treatment approaches, the costs and other hurdles involved with randomization between therapeutic approaches have limited this approach. The presently compiled dataset therefore provides the best available evidence analyzing important clinical end points such as the safety and efficacy of catheter ablation and AAD therapy that is relevant to patients and physicians.
Sources of Funding
This study was supported by Biosense Webster.
Dr Calkins is a consultant for Biosense Webster, CryoCor, Medtronic, and ProRhythm; he owns no stock in Biosense Webster and receives less than $10 000 per year. Dr Reynolds is a consultant for Biosense Webster and Cardiome Pharma Corp. Dr Spector is a consultant for Medtronic, Biosense Webster, and Boston Scientific; he receives research grant support from Medtronic and Biosense Webster.
Fuster V, Ryden LE, Cannon DS, et al. ACC/AHA/ESC 2006 guidelines for the management of patients with atrial fibrillation-executive summary: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines and the European Society of Cardiology Committee for Practice Guidelines (Writing Committee to Revise the 2001 Guidelines for the Management of Patients With Atrial Fibrillation). J Am Coll Cardiol. 2006; 48: 149–246.
Wazni OM, Marrouche NF, Martin DO, Verma A, Bhargava M, Saliba W, Bash D, Schweikert R, Brachmann J, Gunther J, Gutleben K, Pisano E, Potenza D, Fanelli R, Raviele A, Themistoclakis S, Rossillo A, Bonso A, Natale A. Radiofrequency ablation vs antiarrhythmic drugs as first-line treatment of symptomatic atrial fibrillation: a randomized trial. JAMA. 2005; 293: 2634–2640.
Calkins H, Brugada J, Packer DL, Cappato R, Chen SA, Crijns HJ, Damiano RJ Jr, Davies DW, Haines DE, Haissaguerre M, Iesaka Y, Jackman W, Jais P, Kottkamp H, Kuck KH, Lindsay BD, Marchlinski FE, McCarthy PM, Mont JL, Morady F, Nademanee K, Natale A, Pappone C, Prystowsky E, Raviele A, Ruskin JN, Shemin RJ. HRS/EHRA/ECAS expert consensus statement on catheter and surgical ablation of atrial fibrillation: recommendations for personnel, policy, procedures and follow-up: a report of the Heart Rhythm Society (HRS) Task Force on catheter and surgical ablation of atrial fibrillation. Heart Rhythm. 2007; 4: 816–861.
Chalmers TC, Lau J. Meta-analytic stimulus for changes in clinical trials. Stat Methods Med Res. 1993; 2: 161–172.
Pappone C, Oral H, Santinelli V, Vicedomini G, Lang CC, Manguso F, Torracca L, Benussi S, Alfieri O, Hong R, Lau W, Hirata K, Shikuma N, Hall B, Morady F. Atrio-esophageal fistula as a complication of percutaneous transcatheter ablation of atrial fibrillation. Circulation. 2004; 109: 2724–2726.
Deedwania PC, Singh BN, Ellenbogen K, Fisher S, Fletcher R, Singh SN. 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): the Department of Veterans Affairs CHF-STAT Investigators. Circulation. 1998; 98: 2574–2579.
Wnuk-Wojnar AM, Trusz-Gluza M, Czerwinski C, Wozniak-Skowerska I, Szydlo K, Hoffman A, Nowak S, Wita K, Konarska-Kuszewska E, Krauze J, Rybicka-Musialik A, Drzewiecka-Gerber A. Circumferential pulmonary vein RF ablation in the treatment of atrial fibrillation: 3-year experience of one centre. Kardiol Pol. 2005; 63: 362–370.
Pappone C, Manguso F, Vicedomini G, Gugliotta F, Santinelli O, Ferro A, Gulletta S, Sala S, Sora N, Paglino G, Augello G, Agricola E, Zangrillo A, Alfieri O, Santinelli V. Prevention of iatrogenic atrial tachycardia after ablation of atrial fibrillation: a prospective randomized study comparing circumferential pulmonary vein ablation with a modified approach. Circulation. 2004; 110: 3036–3042.
Cappato R, Calkins H, Chen SA, Davies W, Iesaka Y, Kalman J, Kim YH, Klein G, Packer D, Skanes A. Worldwide survey on the methods, efficacy, and safety of catheter ablation for human atrial fibrillation. Circulation. 2005; 111: 1100–1105.
Atrial fibrillation (AF) is a common condition that affects more than 3 million people in the United States. Many patients with symptomatic AF are treated either with antiarrhythmic drug (AAD) therapy or with catheter ablation. The relative safety and efficacy of these 2 therapeutic approaches can be determined by reviewing data from a variety of sources including anonymous surveys, nonrandomized clinical trials, randomized clinical trials, and meta-analyses. In this article, we report the results of 2 separate systematic reviews and meta-analyses of these 2 therapeutic approaches. Our results reveal that the single-procedure success rate of catheter ablation of AF is approximately 60%. The performance of additional ablation procedures and/or the addition of AAD therapy increases the success rate to approximately 75%. In comparison, the success rate for AAD therapy was approximately 50%. The complication rate of ablation was approximately 5%, and approximately one third of patients treated with AAD therapy had some type of side effect. The results of these 2 meta-analyses are reassuring in that the safety and efficacy of catheter ablation are remarkably similar to that reported in several randomized controlled clinical trials and are consistent with clinical practice. The 50% success rate and 30% incidence of side effects to AAD therapy are both somewhat greater than has been reported in randomized clinical trials but are consistent with clinical practice. It is clear that catheter ablation is more effective than AAD therapy in treating AF. However, the complications associated with catheter ablation, including stroke, pulmonary vein stenosis, cardiac tamponade, development of an atrial esophageal fistula, and, very rarely, death, are less commonly encountered with AAD therapy. For this reason, catheter ablation of AF should be considered after a patient has failed attempts at treatment with 1 or more AAD.
The online-only Data Supplement is available at http://circep.ahajournals.org/cgi/content/full/CIRCEP.108.824789/DC1.