Antiarrhythmic use from 1991 to 2007: Insights from the Canadian Registry of Atrial Fibrillation (CARAF I and II)

Transcription

1 Antiarrhythmic use from 1991 to 2007: Insights from the Canadian Registry of Atrial Fibrillation (CARAF I and II) Jason G. Andrade, MD,* Stuart J. Connolly, MD, Paul Dorian, MD, FHRS, Martin Green, MD, Karin H. Humphries, MBA, DSc,* George J. Klein, MD, Robert Sheldon, MD, PhD, # Mario Talajic, MD,** Charles R. Kerr, MD, FHRS* From the *St. Paul s Hospital and the University of British Columbia, Vancouver, British Columbia, Canada, Hamilton General Hospital, Hamilton, Ontario, Canada, St. Michael s Hospital and the University of Toronto, Toronto, Ontario, Canada, Ottawa Heart Institute, Ottawa, Ontario, Canada, Centre for Health Evaluation and Outcomes Sciences, Vancouver, British Columbia, Canada, University of Western Ontario, London, Ontario, Canada, # University of Calgary, Calgary, Alberta, Canada, and **Montréal Heart Institute, Université de Montréal, Montréal, Québec, Canada. BACKGROUND The pharmacologic management of atrial fibrillation (AF), the most common sustained cardiac arrhythmia, has been traditionally dichotomized into control of ventricular rate or re-establishment and maintenance of sinus rhythm. OBJECTIVE The purpose of this study was to evaluate the use of rate-controlling drugs and antiarrhythmic drugs (AAD) in the Canadian Registry of Atrial Fibrillation (CARAF) over a 16-year period from 1991 through METHODS 1,400 patients with new-onset paroxysmal AF who were enrolled in CARAF were included in this analysis. We assessed trends in ventricular rate-controlling medication use (digoxin, beta-blockers, and calcium channel blockers) and AAD (class IA, IC, and III antiarrhythmic agents) at baseline and follow-up visits as well as by calendar year. RESULTS AAD use increased initially from 1991 to 1994 (peak use 42.5%) before steadily declining. Sotalol use decreased (27% to 6%), whereas amiodarone use increased (1.6% to 17.9%). Ratecontrolling medication use decreased from 1991 to 1995 (54.1% to 34.1%) due to declining digoxin use (62.9% to 16.3%). After 1999, there was a continued increase in rate-controlling medication use (peak use 52.5% in 2007) due to increased beta-blocker use (17% to 45.7%). Calcium channel blockers use changed little over the duration of the study. CONCLUSION The management of AF has undergone significant shifts since 1990, reflecting the influence of drug development, prevailing belief systems, the impact of large clinical trials, and evidence-based recommendations. Monitoring of pharmacotherapy trends will provide insight into the real-world application of evidence-based guidelines as well as allow the opportunity to identify deficiencies and improve patient care. KEYWORDS Atrial fibrillation; Antiarrhythmia agents; Flecainide; Propafenone; Sotalol; Quinidine; Amiodarone; Adrenergic beta-antagonists; Calcium channel blockers; Digoxin; Digitalis glycosides; Trends ABBREVIATIONS AAD antiarrhythmic drugs; AF atrial fibrillation; BB beta-blockers; CARAF Canadian Registry of Atrial Fibrillation; CCB calcium channel blockers; LV left ventricular (Heart Rhythm 2010;7: ) 2010 Heart Rhythm Society. All rights reserved. Introduction Atrial fibrillation (AF) is the most common sustained arrhythmia seen in clinical practice. AF affects nearly 2.5 million Americans and 4.5 million Europeans, and as a result, accounts for the majority of arrhythmia-related physician visits and hospital admissions. 1 3 Although the majority of these cases are not immediately life threatening, AF is associated with significant complications, The Canadian Registry of Atrial Fibrillation has been supported by an unrestricted grant from Sanofi-Aventis Canada ( ), St. Jude Medical (2007), Proctor and Gamble Pharmaceuticals Inc. ( ), Knoll Pharmaceuticals ( ), and Dupont Pharma (1996). Address reprint requests and correspondence: Dr. Charles R. Kerr, Burrard Street, Vancouver BC V6Z 1Y6, Canada. address: ckerr@providencehealth.bc.ca. (Received December 16, 2009; accepted April 14, 2010.) such as stroke, as well as reductions in quality of life, functional status, and cardiac performance. In addition, patients with AF are at higher risk of overall mortality when compared with non-af patients. 4 The pharmacologic management of AF has been traditionally dichotomized into 2 competing strategies: control of ventricular rate or re-establishment and maintenance of sinus rhythm. Traditionally, the maintenance of sinus rhythm by means of cardioversion followed by the chronic use of antiarrhythmic drugs (AAD) was chosen as the initial therapy for AF. In contrast to the prevailing clinical practice, landmark trials published over the past 10 years have not demonstrated a significant benefit with a strategy of routine rhythm control in respect to survival, stroke prevention, or quality of life. These studies have suggested that a strategy /$ -see front matter 2010 Heart Rhythm Society. All rights reserved. doi: /j.hrthm

2 1172 Heart Rhythm, Vol 7, No 9, September 2010 focused on minimizing symptoms and complications with ventricular rate control is equally efficacious, particularly in an elderly, minimally symptomatic patient population The purpose of this study was to evaluate the use of rate-controlling medications and AADs in the Canadian Registry of Atrial Fibrillation (CARAF) over a 16-year period (1991 to 2007). Because CARAF I and CARAF II are nondirected cohorts, they provide the opportunity to evaluate the long-term trends in drug therapy for patients with AF beginning in the 10-year period prior to the publication of the American College of Cardiology/ American Heart Association/European Society of Cardiology clinical practice guidelines for the management of AF in 2001 through the publication of the landmark AF trials in the late 1990s to early 2000s As such, it provides unique insight into the use of antiarrhythmic and rate-controlling medications by primary and secondary care physicians over time. Methods Detailed methods of the CARAF cohorts have been reported elsewhere. 11,12 Briefly, CARAF I and II are prospective cohort studies that document the natural history and management of patients with new-onset AF or flutter. CARAF I enrolled 1,095 patients in 6 Canadian cities between 1991 and CARAF II added Toronto to the original 6 cities and enrolled 503 patients between 2001 and In both cohorts, patients were enrolled from emergency rooms, cardiology laboratories, hospitals, and family physician and specialist offices at the time of their first electrocardiographically documented diagnosis of AF or flutter. In both CARAF I and II, study nurses performed comprehensive baseline, 3-month, and annual follow-up evaluations. A set of standardized data forms was used to obtain clinical information. Extensive baseline data were collected, including a comprehensive history, basic laboratory tests, electrocardiograms, and echocardiograms. Patients were followed up at 3 months, 1 year, and annually by the study nurses. Electrocardiograms were repeated annually, and echocardiograms were performed every 2 years. The maximum duration of follow-up was 10 years in CARAF I and 5 years in CARAF II. All interventions for AF, including medications and doses, were thoroughly documented. Specific events including stroke and death as well as major interventions such as pacemaker implantation, ablation, and cardioversion were detailed. Recurrence of AF was evaluated at annual clinic visits and documented by both history and electrocardiogram. The study investigators were not involved in directing treatment of patients unless specifically requested by a referring physician. We assessed use of medications for ventricular rate control and maintenance of sinus rhythm. For ventricular rate control, these drugs included digoxin, 7 different beta-blockers (BB; acebutolol, atenolol, metoprolol, nadolol, pindolol, propranolol, and timolol) and 2 calcium channel blockers (CCB; verapamil and diltiazem). For restoration and maintenance of sinus rhythm, we included 3 Vaughan-Williams class I AADs (quinidine, flecainide, and propafenone), and 2 class III AADs (amiodarone and sotalol). For the purposes of this analysis, we considered patients to be receiving a rate-control strategy if they were taking a drug for ventricular rate control but not an AAD. Conversely, we considered patients to be receiving a rhythm control strategy if they were taking an AAD regardless of whether they were concomitantly taking a rate-slowing drug as well. We determined the percentage of patients taking each class of medication at baseline and each of the follow-up visits and substratified these analyses for CARAF I, CARAF II, and the combination cohort consisting of CARAF I and II. As well, we determined the percentage of patients taking each class of medication over the calendar years from 1990 to 2005 (CARAF I) and 2001 to 2008 (CARAF II). For each class of medication, we determined the baseline characteristics of patients prescribed that class of medication and compared them with the baseline characteristics of patients never prescribed that class of medication. For the purposes of this study, we excluded from analysis the 198 patients from CARAF who had AF after cardiac or thoracic surgery. Of the 1,400 patients with complete datasets at each follow-up visit, 897 patients originated from CARAF I and 503 patients from CARAF II. Results Rate versus rhythm control from 1991 to 2007 The proportion of patients in the CARAF studies receiving AADs to maintain sinus rhythm (rhythm control strategy) increased from 26.9% in 1991 to 42.4% in 1994 (Figure 1). There was a steady decline in the frequency of AAD use from a peak in 1994 through the end of follow-up in 2007 (with the exception of a slight increase in use in 2003). In both CARAF I and CARAF II, as well as in the combined Figure 1 The use of a rhythm control strategy, a rate control strategy, and no medications in AF patients by calendar year in CARAF I and II. AF atrial fibrillation; CARAF Canadian Registry of Atrial Fibrillation.

3 Andrade et al Insights From CARAF I and II 1173 Figure 2 The use of a rhythm control strategy, a rate control strategy, and no medications in AF patients from enrolment through to the end of study follow-up (10 years for CARAF I and 5 years for CARAF II). In both CARAF I and CARAF II, the prevalence of chronic AF increased over the duration of follow-up. Mirroring this increase in AF was a steady decline in AAD use. AAD antiarrhythmic drug; AF atrial fibrillation; CARAF Canadian Registry of Atrial Fibrillation. cohort, the use of AAD was highest in the first year after AF diagnosis (45.6% for the combined cohort) before steadily declining on subsequent follow-up (Figure 2). In contrast, the proportion of patients receiving exclusively a rate control strategy decreased from a peak of 54.2% in 1991 to a nadir of 34.1% in This level of use remained stable until , when there was an abrupt increase in the use of a rate control strategy. In 2000, rate control became more prevalent than rhythm control. It reached a plateau at around 45% for 4 years beginning in 2002 before again abruptly increasing in frequency through the end of the follow-up period. When CARAF I and CARAF II were analyzed separately, interesting trends emerged. In CARAF I, despite an early preference for a rhythm control strategy, the use of a rate control strategy predominated from 2000 to 2005 when a preference toward rhythm control re-emerged in the last year of follow-up. Similarly, at all time points in CARAF II, which began in 2001, the use of a rate control strategy was preferred. It is interesting to note that for all directly comparable time points, the use of a rate control strategy was approximately 25% higher in CARAF II when compared with CARAF I. As a result, the absolute difference between patients receiving a rhythm control strategy and those receiving a rate control strategy in CARAF I was dramatically less than the difference observed in CARAF II. Use of antiarrhythmic medications from 1991 to 2007 There were significant shifts in the use of individual medications throughout the study period (Figure 3). Throughout the 1990s, sotalol was the most frequently prescribed AAD, with rates double that of the second most frequently prescribed AAD (propafenone). The use of amiodarone increased throughout the study period, eventually overtaking sotalol as the most frequently prescribed AAD in In contrast, the use of sotalol and propafenone peaked in 1993 and 1995 at 27.0% and 13.3%, respectively, before steadily declining thereafter. Quinidine was used fairly frequently in 1991 (9.5%) but progressively declined throughout the study period. Flecainide was seldom used throughout the study period, with usage peaking at 2.0% in When CARAF I and CARAF II were compared, important observations emerged. In both studies, the use of amiodarone peaked in before declining thereafter. By the time enrolment began in CARAF II, the use of sotalol and propafenone was already on the decline. As a result, the use of sotalol and propafenone was slightly less frequent in CARAF II when compared with CARAF I (approximately 1% to 2% lower). Conversely, amiodarone use was slightly greater in CARAF II when compared with CARAF I (approximately 1% to 2% greater), with the greatest difference occurring in 2003 (19.8% in CARAF II and 14.4% in CARAF I). Use of ventricular rate control agents from 1991 to 2007 Significant variability was also observed in the choice of ventricular rate control agents throughout the study period (Figure 4). In the early 1990s, digoxin was the preferred agent for controlling ventricular rate, with usage 3 to 4 times that of BB and CCB (62.9% vs. 20.2% vs. 14.2%, respectively). Although CCB use remained fairly constant throughout the study period, the use of digoxin decreased dramatically from 1991 through 2007, falling from 62.9% to 16.3%. Conversely, the use of BB increased from 20.2% in 1991 to 45.7% in When CARAF I and CARAF II were compared, important observations emerged. In CARAF I, the use of BB was steady at around 20%, until 1998, when it climbed and peaked at over 30% in Similarly, CARAF II began enrolment in 2001, and like CARAF I, had peak BB usage in (53.2%). In both trials, the use of BB declined thereafter. It is interesting to note that for all time points the use of BB was approximately 50% higher in CARAF II when compared with CARAF I. In contrast, the use of digoxin was approximately 50% lower in CARAF II Figure 3 AAD use in AF patients by calendar year in CARAF I and II. AAD antiarrhythmic drug; AF atrial fibrillation; CARAF Canadian Registry of Atrial Fibrillation.

4 1174 Heart Rhythm, Vol 7, No 9, September 2010 follow-up). Conversely, the use of a combination of a BB and CCB increased from a nadir of 3.7% in 1992 to a peak of 34.6% at the end of follow-up. The use of BB, CCB, and digoxin in combination was relatively rare and did not change over time (7% of cases of combination therapy; 2% of the overall study population). The patterns of rate controlling agent use did not vary between the rate control group and the rhythm control group; however, the frequency of use was higher in the rate control group. Figure 4 Rate control medication use in AF patients by calendar year in CARAF I and II. AF atrial fibrillation; BB beta-blocker; CARAF Canadian Registry of Atrial Fibrillation; CCB nondihydropyridine calcium channel blocker. when compared with CARAF I for all time points. The use of CCB did not differ between CARAF I and CARAF II. The use of rate control agents in the rate control and rhythm control groups In patients receiving AAD to maintain sinus rhythm (rhythm control strategy), concomitant rate controlling medications were used in over 50% of the subjects. In both the rhythm control group and the rate control group, digoxin was the preferred rate control agent during the early study period (Figure 5). In both groups, the use of digoxin decreased from peak use in 1991 and was supplanted by BB as the preferred rate controlling agent in In both groups, the use of CCB was significantly less frequent. Between groups, the biggest difference was in the frequency of use of each of the agents. In 1991, digoxin was used in 86% of subjects in the rate control group, but only 60% of the rhythm control group. Likewise, BBs were used in approximately 80% of subjects in the rate control group from 2004 onward, but 60% of subjects in the rhythm control group over the same period. In a given year, an average of 17% of patients in the study population required therapy with combinations of rate controlling agents. At study onset, the utilization of combination therapy was infrequent, with the nadir use of 12.5% occurring in 1994 and Thereafter, the usage of combination therapy steadily increased and peaked in 2003, with 21.2% of the study population requiring therapy with more than 1 rate controlling agent. Combination therapy with BB and digoxin was most commonly used and observed in an average of 53% of those on combination therapy (range of yearly use 45% to 61%) (Figure 5). The second most common combination varied depending on the time point of observation. Early in the study period, the use of a combination of digoxin and CCB was second most frequent (44% of cases in 1993); however, its use declined over the study period (10% at the end of Baseline characteristics and the use of AAD or rate controlling medications In both CARAF I and CARAF II as well as the combined cohort, patients prescribed AAD were younger compared with those prescribed rate control alone (average age 62 years vs years; P.0001). Patients with a history of congestive heart failure were more likely to be prescribed rate control medications compared with those prescribed AAD; however, this was only significant in the earlier CARAF I cohort (P.02 for CARAF I; P.2 for CARAF II and the combined cohort). Patients with a clinical history of stroke or hypertension were less likely to be Figure 5 Usage of medications to control ventricular rate in patients receiving a rate control strategy (A) and a rhythm control strategy (B). AAD antiarrhythmic drug; AF atrial fibrillation; BB beta-blocker; CARAF Canadian Registry of Atrial Fibrillation; CCB nondihydropyridine calcium channel blocker; Dig digoxin.

5 Andrade et al Insights From CARAF I and II 1175 prescribed AAD when compared with those prescribed rate control alone (7.5% for AAD vs. 10.9% for rate control; P.04 and 51.4% for AAD vs. 41.6% for rate control; P.0009, respectively). Patients presenting with symptomatic AF were more likely to be prescribed AAD (77.1% vs. 66.6% prescribed rate control alone; P.0001). There was no difference in AAD use based on gender, objective left ventricular (LV) impairment (LV ejection fraction 40%) or a clinical history diabetes (P.6, P.2, and P.6, respectively). AAD use was not influenced by renal function (P.6). Patients receiving neither AAD nor rate-limiting pharmacotherapy were significantly younger (average age 55.6 years) and were less likely to have a history of hypertension (23.8%), clinical congestive heart failure (5.2%), or an objective impairment of LV systolic function (8.2%); 73.3% of this group had symptomatic AF. The use of BB, CCB, and digoxin was significantly higher in older patients (mean 63.9 vs years for BB, P.0002; 65.3 vs years for CCB, P.0001; and 64.9 vs years for digoxin, P.0001). BB and digoxin were less likely to be prescribed to men (P.0001 and P.009, respectively). Digoxin was more often used in those with a history of congestive heart failure or impaired LV function (P.0001 for both). In contrast, BB and CCB were more often prescribed to those without congestive heart failure (P.04 for BB) and preserved LV function (P.04 for CCB). Patients were more likely to receive BB or CCB if they had a history of hypertension (P.0001 and P.0001, respectively). Those with a history of diabetes mellitus were more likely to receive CCB (P.04). Discussion AF and flutter are increasingly common medical problems for which the pharmacologic management has changed significantly over the past 20 years. From these 2 large nondirected prospective cohort studies, we were able to gain insight into the evolving long-term trends in the pharmacologic management of AF beginning in the 10-year period prior to the publication of landmark AF trials in the late 1990s and early 2000s. Over the course of our study period, the management strategy for AF and flutter underwent 2 significant philosophical shifts. At the onset of our study in 1991, a strategy focused on rate control predominated, only to be superseded by a rhythm control strategy in This persisted until the year 2000 when, nearing the publication of the landmark Rate Control versus Electrical Cardioversion (RACE), Pharmacological Intervention in Atrial Fibrillation (PIAF), and Atrial Fibrillation Follow-up Investigation of Rhythm Management (AFFIRM) studies, rate control again became the predominant management strategy. 6 9 The use of medications for the maintenance of sinus rhythm has also evolved over the past 2 decades. Most notably, there was a rapid increase in the use of class III agents in favor of class I agents. In contrast to previous observational series, class Ia agents such as quinidine, although reasonably popular at study onset, were not used by ,14 In fact, the use of class Ia agents continued to decline from a peak use of 9.5% at study onset in 1991 to less than 2% from 1999 onward. Although somewhat effective in the maintenance of sinus rhythm, quinidine is associated with frequent noncardiac adverse events as well as increased allcause mortality Likewise, the Cardiac Arrhythmia Suppression Trial (CAST) trial from 1989 tainted the use of the class Ic agent flecainide. 16,21 23 This information coupled with lack of promotion likely influenced the overall low use of these agents throughout our study period. Interestingly, the use of propafenone, another class Ic agent, became increasingly prominent as the use of other class I agents declined. This likely reflected the then-novel information that new AADs, such as propafenone and sotalol, a class III AAD, were more effective in the management of AF with favorable side effect profiles and increased ease of use. 24 As such, the use of these 2 agents, which peaked in the early to mid 1990s, was largely responsible for the first peak in AAD use observed in the early part of our study period. Thereafter, their use steadily declined, only to be supplanted by the even more efficacious amiodarone in 2002, whose usage was buoyed by the publication of randomized trials demonstrating its superiority in the maintenance of sinus rhythm In both CARAF I and CARAF II, the use of AAD was highest in the first year after AF diagnosis (48.96% in CARAF I and 41.55% in CARAF II) (Figure 2). Thereafter, there was a steady decline in the frequency of AAD use, with the rates of AAD use falling to 38.51% and 22.28% at 5 years of follow-up (CARAF I and CARAF II, respectively) and 31.61% at 10 years (CARAF I). This decline in AAD use was contrasted by a progressively greater propensity toward a rate control strategy in both the CARAF I and CARAF II cohorts. This may be partly related to the publication of major trials demonstrating similar outcomes with strategies for controlling ventricular rate compared with strategies for restoring sinus rhythm, as well as alterations in the clinical presentation with patients presenting with less severe or even no symptoms. 6 9 However, it is possible that the observed declining trend in AAD use merely reflects a progression of the underlying disease. It is well known that the natural history of AF is characterized by a steady progression from paroxysmal to permanent AF. 12 This was observed in our study in both the CARAF I and CARAF II cohorts, in which the rates of permanent AF had increased from 10% to 15% at the first year of follow-up to 28.59% and 30.43% at 5 years in CARAF I and II, respectively, and 37.22% at 10 years (CARAF I). As such, it may be that the observed longitudinal decline in AAD use was merely a reflection of the acceptance of permanent AF by patients and their treating physicians, with a resultant shift toward a goal of ventricular rate control. However, it is interesting to note that at any given time point of follow-up, the use of a rhythm control strategy was approximately 10% less common in CARAF II when compared with CARAF I. This most likely reflects the evolution of clinical practice away

6 1176 Heart Rhythm, Vol 7, No 9, September 2010 from routine rhythm control after the publication of the aforementioned landmark rate control trials in the early 2000s. 6 9 The use of rate controlling medications also evolved significantly over the last 2 decades. Digoxin, initially most prevalent, declined in concert with the early decline in use of a rate control strategy. This was related in part to evolving recognition that digoxin is particularly ineffective at controlling ventricular rate with effort. 5,28 32 In contrast, the use of BB slowly increased until the late 1990s, when their use abruptly increased significantly. Temporally, this is most likely related to the dissemination of knowledge surrounding the publication of RACE, PIAF, and AFFIRM, as well as the publication of studies demonstrating their effectiveness in heart failure management. 6 9,33 Interestingly, despite their use in seminal rate control trials, the use of CCB did not change dramatically over the study period (14.3% in 1991 to 16.3% in 2007). 6 9 Some of the variability in the use of AAD and rate controlling agents can be explained by baseline patient characteristics. In the combined CARAF cohorts, younger patients and those with symptomatic AF were more likely to be prescribed AAD. It is interesting to note that patients with congestive heart failure or LV dysfunction were more likely to receive digoxin yet less likely to receive BBs, especially in the later CARAF II cohort. This is in contrast to the publication of multiple clinical trials in the late 1990s demonstrating the efficacy of beta-blockade in these populations. 33 There are several limitations to this study that should be acknowledged. The CARAF cohorts are observational in design and largely reflect patterns of practice at tertiary care institutions. As a result, the use of specific medications, although predominantly dictated by the subject s primary physicians, may have been directed by a cardiac electrophysiologist. This observation, combined with the non population-based nature of the study, may limit its generalizability. That being said, there was a consistency to the nature of patient screening and referral, and as such the population included should reasonably represent the population presenting with newly diagnosed paroxysmal AF. As well, although providing a longitudinal assessment of the use of pharmacotherapy over time, these cohorts were managed in an era prior to the publication of landmark AF ablation trials. As a result, these cohorts do not necessarily reflect the use of AAD or rate controlling agents in the era of increasing adoption of radiofrequency catheter ablation for AF. Lastly, for the purposes of this analysis, patients were considered to be receiving a rate control strategy if they were taking a drug for ventricular rate control but not an AAD. Conversely, patients were considered to be receiving a rhythm control strategy if they were taking an AAD regardless of whether they were concomitantly taking a rate-slowing drug as well. Although reasonably accurate, we could not be certain that these definitions reflect the motivations of the treating physician. For example, some patients taking amiodarone may have been in permanent AF with the amiodarone used for control of the ventricular rate. Conversely, the initial use of BB monotherapy may have been intended as a strategy of rhythm control; however, the actual rhythm control effects of BB therapy in AF remain somewhat controversial. 34 Furthermore, the definitions used herein to define a rate control strategy and a rhythm control strategy are consistent with the distinctions used in the landmark rate control versus rhythm control trials and reflect contemporary clinical practice. 6 9 Conclusion The combined CARAF cohorts provide an 18-year duration of follow-up for newly diagnosed AF patients and offer a unique insight into the use of medications for the maintenance of sinus rhythm as well as the use of medications for the control of ventricular rate. Over this time, the management of AF and flutter has undergone significant shifts reflecting the influence of drug development, prevailing belief systems, and the impact of large clinical trials. As such, continued monitoring of the trends in pharmacotherapy will provide insight into the real-world application of evidence and clinical practice guidelines, as well as allow the opportunity to identify deficiencies and improve patient care. References 1. Go AS, Hylek EM, Phillips KA, et al. Prevalence of diagnosed atrial fibrillation in adults: national implications for rhythm management and stroke prevention: the AnTicoagulation and Risk Factors In Atrial Fibrillation (ATRIA) study. JAMA 2001;285: Fuster V, Rydén LE, Cannom DS. ACC/AHA/ESC 2006 guidelines for the management of patients with atrial fibrillation. Circulation 2006;114:e257 e Waktare JE, Camm AJ. Acute treatment of atrial fibrillation: why and when to maintain sinus rhythm. Am J Cardiol 1998;81:3C 15C. 4. Wolf PA, Mitchel JB, Baker CS, et al. Impact of atrial fibrillation on mortality stroke, and medical cost. Arch Intern Med 1998;158: Fuster V, Ryden LE, Asinger RW, et al. ACC/AHA/ESC 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 and Policy Conferences (Committee to Develop Guidelines for the Management of Patients with Atrial Fibrillation): developed in collaboration with the North American Society of Pacing and Electrophysiology. J Am Coll Cardiol 2001;38: de Denus S, Sanoski CA, Carlsson J, et al. Rate vs rhythm control in patients with atrial fibrillation: a meta-analysis. Arch Intern Med 2005;165: 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: The Atrial Fibrillation Follow-up Investigation of Rhythm Management (AF- FIRM) Investigators. 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: Pappone C, Rosanio S, Augello G, et al. Mortality, morbidity, and quality of life after circumferential pulmonary vein ablation for atrial fibrillation. J Am Coll Cardiol 2003;42: Humphries KH, Kerr CR, Connolly SJ, et al. New-onset atrial fibrillation: sex differences in presentation, treatment, and outcome. Circulation 2001;103: Kerr CR, Humphries KH, Talajic M, et al. Progression to chronic atrial fibrillation after the initial diagnosis of paroxysmal atrial fibrillation: results

7 Andrade et al Insights From CARAF I and II 1177 from the Canadian Registry of Atrial Fibrillation. Am Heart J 2005; 149: Al-Khatib SM, Allen LaPointe NM, Curtis LH, et al. Outpatient prescribing of antiarrhythmic drugs from 1995 to Am J Cardiol 2003;91: Fang MC, Stafford RS, Ruskin JN, et al. National trends in antiarrhythmic and antithrombotic medication use in atrial fibrillation. Arch Intern Med 2004;164: Hine LK, Laird NM, Hewitt P, Chalmers TC. Meta-analysis of empirical long term antiarrhythmic therapy after myocardial infarction. JAMA 1989;262: Teo KK, Yusuf S, Furberg CD. Effects of prophylactic antiarrhythmic drug therapy in acute myocardial infarction: an overview of results from randomized controlled trials. JAMA 1993;270: Coplen SE, Antman EM, Berlin JA, et al. Efficacy and safety of quinidine therapy for maintenance of sinus rhythm after cardioversion: a meta-analysis of randomized control trials. Circulation 1990;82: Morganroth J, Goin J. Quinidine-related mortality in the short- to mediumterm treatment of ventricular arrhythmias: a meta-analysis. Circulation 1991; 84: Reimold S, Chalmers TC, Berlin JA, et al. Assessment of the efficacy and safety of antiarrhythmic therapy for chronic atrial fibrillation: observations on the role of trial design and implications of drug-related mortality. Am Heart J 1992;124: 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: 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: Echt DS, Liebson PR, Mitchell LB, et al. Mortality and morbidity in patients receiving encainide, flecainide, or placebo. The Cardiac Arrhythmia Suppression Trial. N Engl J Med 1991;324: Henthorn RW, Waldo AL, Anderson JL, et al. Flecainide acetate prevents recurrence of symptomatic paroxysmal supraventricular tachycardia. The Flecainide Supraventricular Tachycardia Study Group. Circulation 1991;83: Reimold SC, Cantillon CO, Friedman PL, Antman EM. Propafenone versus sotalol for suppression of recurrent symptomatic atrial fibrillation. Am J Cardiol 1993;71: Roy D, Talajic M, Dorian P, et al. Amiodarone to prevent recurrence of atrial fibrillation. Canadian Trial of Atrial Fibrillation Investigators. N Engl J Med 2000;342: Maintenance of sinus rhythm in patients with atrial fibrillation: an AFFIRM substudy of the first antiarrhythmic drug. J Am Coll Cardiol 2003;42: Kochiadakis GE, Igoumenidis NE, Marketou ME, et al. Low dose amiodarone and sotalol in the treatment of recurrent, symptomatic atrial fibrillation: a comparative, placebo controlled study. Heart 2000;84: Boriani G, Diemberger I, Martignani C, et al. The epidemiological burden of atrial fibrillation: a challenge for clinicians and health care systems. Eur Heart J 2006;27: Segal JB, McNamara RL, Miller MR, et al. The evidence regarding the drugs used for ventricular rate control. J Fam Pract 2000;49: Page RL. Clinical practice. Newly diagnosed atrial fibrillation. N Engl J Med 2004;351: Corley SD, Epstein AE, DiMarco JP, et al. Relationships between sinus rhythm, treatment, and survival in the Atrial Fibrillation Follow-Up Investigation of Rhythm Management (AFFIRM) study. Circulation 2004;109: Snow V, Weiss KB, LeFevre M, et al. Management of newly detected atrial fibrillation: a clinical practice guideline from the American Academy of Family Physicians and the American College of Physicians. Ann Intern Med 2003;139: Yan AT, Yan RT, Liu PP. Narrative review: pharmacotherapy for chronic heart failure: evidence from recent clinical trials. Ann Intern Med 2005;142: Kuhlkamp V, Schirdewan A, Stangl K, Homberg M, Ploch M, Beck OA. Use of metoprolol CR/XL to maintain sinus rhythm after conversion from persistent atrial fibrillation: a randomized, double- blind, placebo-controlled study. J Am Coll Cardiol 2000;36: