This Article Free upon publication Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Fuster, V. Articles by Torbicki, A. Search for Related Content PubMed PubMed Citation Articles by Fuster, V. Articles by Torbicki, A.

(Circulation. 2001;104:2118.)
© 2001 American Heart Association, Inc.

ACC/AHA Practice Guidelines

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

, Committee MembersValentin Fuster, MD PhD, FACC, Chair; Lars E. Rydén, MD PhD, FACC, FESC, Co-Chair; Richard W. Asinger, MD FACC; David S. Cannom, MD FACC; Harry J. Crijns, MD FESC; Robert L. Frye, MD MACC; Jonathan L. Halperin, MD FACC; G. Neal Kay, MD FACC; Werner W. Klein, MD FACC, FESC; Samuel Lévy, MD FACC, FESC; Robert L. McNamara, MD MHS, FACC; Eric N. Prystowsky, MD FACC; L. Samuel Wann, MD FACC; D. George Wyse, MD PhD, FACC; , Task Force Members; Raymond J. Gibbons, MD FACC, Chair; Elliott M. Antman, MD FACC, Vice Chair; Joseph S. Alpert, MD FACC; David P. Faxon, MD FACC; Valentin Fuster, MD PhD, FACC; Gabriel Gregoratos, MD FACC; Loren F. Hiratzka, MD FACC; Alice K. Jacobs, MD FACC; Richard O. Russell, MD FACC^*; Sidney C. Smith, Jr., MD FACC; , ESC Committee for Practice Guidelines & Policy Conferences Members; Werner W. Klein, MD FACC, FESC, Chair; Angeles Alonso-Garcia, MD FACC, FESC; Carina Blomström-Lundqvist, MD PhD, FESC; Guy de Backer, MD PhD, FACC, FESC; Marcus Flather, MD FESC; Jaromir Hradec, MD FESC; Ali Oto, MD FACC, FESC; Alexander Parkhomenko, MD FESC; Sigmund Silber, MD PhD, FESC; Adam Torbicki, MD FESC

	I. Introduction

Top I. Introduction II. Definitions III. Classification IV. Epidemiology and Prognosis V. Pathophysiological Mechanisms VI. Associated Conditions,... VII. Clinical Evaluation VIII. Management IX. Proposed Management... References

 
Atrial fibrillation (AF), the most common sustained cardiac rhythm disturbance, is increasing in prevalence as the population ages. Although it is often associated with heart disease, AF occurs in many patients with no detectable disease. Hemodynamic impairment and thromboembolic events result in significant morbidity, mortality, and cost. Accordingly, the American College of Cardiology (ACC), the American Heart Association (AHA), and the European Society of Cardiology (ESC) created a committee of experts to establish guidelines for management of this arrhythmia.

The committee was composed of 8 members representing the ACC and AHA, 4 representing the ESC, 1 from the North American Society of Pacing and Electrophysiology (NASPE), and a representative of the Johns Hopkins University Evidence-Based Practice Center representing the Agency for Healthcare Research and Quality’s report on Atrial Fibrillation in the Elderly. This document was reviewed by 3 official reviewers nominated by the ACC, 3 nominated by the AHA, and 3 nominated by the ESC, as well as by the ACC Clinical Electrophysiology Committee, the AHA ECG and Arrhythmia Committee, NASPE, and 25 reviewers nominated by the writing committee. The document was approved for publication by the governing bodies of the ACC, AHA, and ESC and officially endorsed by NASPE. These guidelines will be reviewed annually by the task force and will be considered current unless the task force revises or withdraws them from distribution.

The committee conducted a comprehensive review of the literature from 1980 to June 2000 relevant to AF using the following databases: PubMed/Medline, EMBASE, the Cochrane Library (including the Cochrane Database of Systematic Reviews and the Cochrane Controlled Trials Registry), and Best Evidence. Searches were limited to English language sources and to human subjects.

	II. Definitions

Top I. Introduction II. Definitions III. Classification IV. Epidemiology and Prognosis V. Pathophysiological Mechanisms VI. Associated Conditions,... VII. Clinical Evaluation VIII. Management IX. Proposed Management... References

 
A. Atrial Fibrillation
AF is a supraventricular tachyarrhythmia characterized by uncoordinated atrial activation with consequent deterioration of atrial mechanical function. On the electrocardiogram (ECG), AF is described by the replacement of consistent P waves by rapid oscillations or fibrillatory waves that vary in size, shape, and timing, associated with an irregular, frequently rapid ventricular response when atrioventricular (AV) conduction is intact (¹). The ventricular response to AF depends on electrophysiological properties of the AV node, the level of vagal and sympathetic tone, and the action of drugs (²). Regular RR intervals are possible in the presence of AV block or interference by ventricular or junctional tachycardia. A rapid, irregular, sustained, wide-QRS-complex tachycardia strongly suggests AF with conduction over an accessory pathway or AF with underlying bundle-branch block. Extremely rapid rates (over 200 bpm) suggest the presence of an accessory pathway.

B. Related Arrhythmias
AF can be isolated or associated with other arrhythmias, often atrial flutter or atrial tachycardia. Atrial flutter can arise during treatment with antiarrhythmic agents prescribed to prevent recurrent AF. Atrial flutter is more organized than AF, with a saw-tooth pattern of regular atrial activation called flutter (f) waves on the ECG, particularly visible in leads II, III, and aVF. Untreated, the atrial rate typically ranges from 240 to 320 beats per minute (bpm), with f waves inverted in ECG leads II, III, and aVF and upright in lead V₁. The wave of activation in the right atrium (RA) may be reversed, resulting in f waves that are upright in leads II, III, and aVF and inverted in lead V₁. Two-to-one AV block is common, producing a ventricular rate of 120 to 160 bpm. Atrial flutter can degenerate into AF, AF can initiate atrial flutter, or the ECG pattern can alternate between atrial flutter and AF, reflecting changing atrial activation.

Other atrial tachycardias, as well as AV reentrant tachycardias and AV nodal reentrant tachycardias, can also trigger AF. In other atrial tachycardias, P waves are readily identified and are separated by an isoelectric baseline in 1 or more ECG leads. The morphology of the P waves can help localize the origin of atrial tachycardias. A unique type of atrial tachycardia originates in the pulmonary veins (³), is typically more rapid than 250 bpm, and often degenerates into AF. Intracardiac mapping can help differentiate the various atrial arrhythmias.

	III. Classification

Top I. Introduction II. Definitions III. Classification IV. Epidemiology and Prognosis V. Pathophysiological Mechanisms VI. Associated Conditions,... VII. Clinical Evaluation VIII. Management IX. Proposed Management... References

 
AF has a heterogeneous clinical presentation, occurring in the presence or absence of detectable heart disease or related symptoms. For example, the term "lone AF" has been variously defined. The prognosis in terms of thromboembolism and mortality is most benign when applied to young individuals (aged less than 60 years) without clinical or echocardiographic evidence of cardiopulmonary disease (⁴). These patients have a favorable prognosis with respect to thromboembolism and mortality. By virtue of aging or the development of cardiac abnormalities, however, patients move out of the lone AF category over time, and the risks of thromboembolism and mortality rise. Lone AF is distinguished from idiopathic AF, which implies uncertainty about its origin without reference to the age of the patient or associated cardiovascular pathology. By convention, the term nonvalvular AF is restricted to cases in which the rhythm disturbance occurs in the absence of rheumatic mitral stenosis or a prosthetic heart valve.

The classification scheme recommended in this document represents a consensus driven by a desire for simplicity and clinical relevance. The clinician should distinguish a first-detected episode of AF, whether or not it is symptomatic or self-limited, recognizing that there can be uncertainty about the duration of the episode and about previous undetected episodes (Fig. 1) . When a patient has had 2 or more episodes, AF is considered recurrent. Once terminated, recurrent AF is designated paroxysmal, and when sustained, persistent. In the latter case, termination by pharmacological therapy or electrical cardioversion does not change the designation. Persistent AF can be either the first presentation or a culmination of recurrent episodes of paroxysmal AF. Persistent AF includes cases of long-standing AF (eg, greater than 1 year), in which cardioversion has not been indicated or attempted, usually leading to permanent AF (Fig. 1). The terminology defined in the preceding paragraph applies to episodes of AF that last more than 30 seconds and that are unrelated to a reversible cause. AF secondary to a precipitating condition such as acute myocardial infarction, cardiac surgery, myocarditis, hyperthyroidism, or acute pulmonary disease is considered separately. In these settings, treatment of the underlying disorder concurrently with management of the episode of AF usually eliminates the arrhythmia.

View larger version (10K): [in this window] [in a new window]   Figure 1. Patterns of atrial fibrillation. 1, episodes that generally last less than or equal to 7 days (most less than 24 h); 2, usually more than 7 days; 3, cardioversion failed or not attempted; and 4, either paroxysmal or persistent AF may be recurrent.

	IV. Epidemiology and Prognosis

Top I. Introduction II. Definitions III. Classification IV. Epidemiology and Prognosis V. Pathophysiological Mechanisms VI. Associated Conditions,... VII. Clinical Evaluation VIII. Management IX. Proposed Management... References

 
AF is the most common clinically significant cardiac arrhythmia. In one series, AF accounted for 34.5% of patients hospitalized with a cardiac rhythm disturbance (⁵). It has been estimated that 2.2 million Americans have paroxysmal or persistent AF (⁶).

A. Prevalence
The prevalence of AF is estimated at 0.4% of the general population, increasing with age (⁷). AF is uncommon in childhood except after cardiac surgery. It occurs in fewer than 1% of those under 60 years of age but in more than 6% of those over 80 years of age (^8–10) (Fig. 2) . The age-adjusted prevalence is higher in men (^10,11). Blacks have less than half the age-adjusted risk of developing AF that is seen in whites (¹²). The frequency of lone AF was less than 12% of all cases of AF in some series (^4,10,13,14) but over 30% in others (^15,16). The prevalence of AF increases with the severity of congestive heart failure (HF) or valvular heart disease.

View larger version (21K): [in this window] [in a new window]   Figure 2. Prevalence of AF in 2 American epidemiological studies. Framingham indicates the Framingham Heart Study (9); CHS, Cardiovascular Health Study (10).

B. Prognosis
The rate of ischemic stroke among patients with nonrheumatic AF averages 5% per year, which is 2 to 7 times the rate for people without AF (^{8,9,15,17–19}) (Fig. 3) . One of every 6 strokes occurs in patients with AF (²⁰). Including transient ischemic attacks and clinically silent strokes detected radiographically, the rate of brain ischemia accompanying nonvalvular AF exceeds 7% per year (^21–25). In the Framingham Heart Study, patients with rheumatic heart disease and AF had a 17-fold increased risk of stroke compared with age-matched controls (²⁶), and the attributable risk was 5 times greater than in those with nonrheumatic AF (⁹). Among AF patients from general practices in France, the ALFA Study (Etude en Activité Liberale sur le Fibrillation Auriculaire) found a 2.4% incidence of thromboembolism over a mean of 8.6 months of follow-up (¹⁵). The annual risk of stroke attributable to AF increased from 1.5% in Framingham Study participants aged 50 to 59 years to 23.5% for those aged 80 to 89 years (⁹). The total mortality rate is approximately doubled in patients with AF compared with patients in normal sinus rhythm and is linked with the severity of underlying heart disease (^8,11,18) (Fig. 3).

View larger version (21K): [in this window] [in a new window]   Figure 3. Relative risk of stroke and mortality in patients with AF compared with patients without AF. Source data are from the Framingham Heart Study (11), Regional Heart Study (8), Whitehall study (8), and Manitoba study (18).

	V. Pathophysiological Mechanisms

Top I. Introduction II. Definitions III. Classification IV. Epidemiology and Prognosis V. Pathophysiological Mechanisms VI. Associated Conditions,... VII. Clinical Evaluation VIII. Management IX. Proposed Management... References

 
A. Atrial Factors
1. Pathology of the Atrium in Patients With AF
The atria of patients with persistent AF display structural abnormalities beyond those caused by underlying heart disease (²⁷). Patchy fibrosis with juxtaposition of normal and diseased atrial fibers may account for nonhomogeneity of atrial refractoriness (^28,29). Fibrosis or fatty infiltration can also affect the sinus node and might be a reaction to inflammatory or degenerative processes that are difficult to detect. The role of inflammation in the pathogenesis of AF has not yet been evaluated, but histological changes consistent with myocarditis were reported in 66% of biopsy specimens from patients with lone AF (²⁹). Infiltration of the atrial myocardium can occur in amyloidosis, sarcoidosis, and hemochromatosis. Atrial fiber hypertrophy has been described as a major and sometimes the sole histological feature (²⁸). Progressive atrial dilatation has been demonstrated echocardiographically in patients with AF (³⁰) and, like hypertrophy, can be either a cause or a consequence of persistent AF.

2. Mechanisms of AF
Theories of the mechanism of AF involve 2 main processes: enhanced automaticity in 1 or several rapidly depolarizing foci and reentry involving 1 or more circuits (^31,32). Rapidly firing atrial foci, located in 1 or several of the superior pulmonary veins, can initiate AF in susceptible patients (^3,33). Foci also occur in the RA and infrequently in the superior vena cava or coronary sinus (^3,33,34). The focal origin appears to be more important in paroxysmal AF than in persistent AF. Ablation of such foci can be curative (³).

The multiple-wavelet hypothesis as the mechanism of reentrant AF was advanced by Moe and colleagues (^31,35), who proposed that fractionation of the wave fronts as they propagate through the atria results in self-perpetuating "daughter wavelets." The number of wavelets present at any time depends on the refractory period, mass, and conduction velocity in different parts of the atria.

Although the patterns of activation underlying the irregular atrial electrical activity of AF have traditionally been described as disorganized or random, recent evidence has emerged that AF is spatially organized. Based on mapping studies of patients undergoing surgery for the Wolff-Parkinson-White (WPW) syndrome, 3 patterns of induced AF have been identified (³⁶). Type I AF involves single wave fronts propagating across the RA. Type II AF involves 1 or 2 wave fronts, and type III AF is characterized by multiple activation wavelets propagating in different directions. Ultimately, a better understanding of electrophysiological mechanisms will lead to the development of effective preventive measures (³⁷).

B. AV Conduction
1. General Aspects
The AV node is ordinarily the factor that limits conduction during AF. The compact AV node is located anteriorly in the triangle of Koch (³⁸), surrounded by transitional cells. There appear to be 2 distinct atrial inputs to the AV node, posteriorly via the crista terminalis and anteriorly via the interatrial septum. Studies on rabbit AV nodal preparations show that during AF, propagation of impulses through the AV node to the His bundle depends in part on the relative timing of the anterior and posterior septal activation inputs to the AV node (³⁹). Other factors that affect conduction through the AV node are its intrinsic conduction and refractoriness, concealed conduction, and autonomic tone.

2. AV Conduction in the WPW Syndrome
Accessory pathways are muscle connections between the atrium and ventricle that have the capacity to conduct rapidly. Conduction over an accessory pathway during AF can result in a very rapid ventricular response that can be fatal (^2,40).

Drugs such as digitalis, calcium channel antagonists, and beta-blockers, which are usually given to slow conduction across the AV node during AF, do not block conduction over the accessory pathway and can even enhance conduction, resulting in hypotension or cardiac arrest (⁴¹). Patients who develop AF with a rapid ventricular response associated with hemodynamic instability that results from conduction over an accessory pathway should undergo immediate electrical cardioversion. In the absence of hemodynamic instability or a preexcited ventricular response, intravenous procainamide and ibutilide are drugs of choice to achieve pharmacological cardioversion or to block conduction over the accessory pathway.

C. Myocardial and Hemodynamic Consequences of AF
During AF, 3 factors can affect hemodynamic function: loss of synchronous atrial mechanical activity, irregularity of ventricular response, and inappropriately rapid heart rate. A marked decrease in cardiac output can occur with the loss of atrial contraction, especially in patients with impaired diastolic ventricular filling, hypertension, mitral stenosis, hypertrophic cardiomyopathy (HCM), or restrictive cardiomyopathies. The variation in RR intervals during AF can also result in hemodynamic impairment. A persistently rapid atrial rate can adversely affect atrial mechanical function (tachycardia-induced atrial cardiomyopathy) (^2,42). Such changes in atrial tissue might explain the delayed recovery of atrial contractility in patients after cardioversion to sinus rhythm.

A persistently elevated ventricular rate during AF (130 bpm or faster in one study) (⁴³) can produce dilated ventricular cardiomyopathy (^2,43–46). It is critically important to recognize tachycardia-induced cardiomyopathy, because control of the ventricular rate can lead to partial or even complete reversal of the myopathic process. In fact, HF can be the initial manifestation of AF. A variety of hypotheses have been proposed to explain tachycardia-mediated cardiomyopathy that involve myocardial energy depletion, ischemia, abnormalities of calcium regulation, and remodeling, but the actual mechanisms responsible for this disorder are still unclear (⁴⁷).

D. Thromboembolism
Although ischemic stroke and systemic arterial occlusion in AF are generally attributed to embolism from the left atrium (LA), the pathogenesis of thromboembolism is complex (⁴⁸). Up to 25% of AF-associated strokes can be due to intrinsic cerebrovascular diseases, other cardiac sources of embolism, or atheromatous pathology in the proximal aorta (^49,50). About half of elderly AF patients have chronic hypertension (a major risk factor for cerebrovascular disease) (¹⁹), and approximately 12% harbor cervical carotid artery stenosis. Carotid atherosclerosis is not substantially more prevalent in AF patients with stroke than in patients without AF, however, and is probably a relatively minor contributing factor (⁵¹).

1. Pathophysiology of Thrombus Formation
Thrombus associated with AF arises most frequently in the LA appendage (LAA). This cannot be examined reliably by precordial (transthoracic) echocardiography (⁵²), whereas transesophageal Doppler echocardiography provides a sensitive and specific method to assess LAA function (⁵³) and to detect thrombotic material. LAA flow velocities are reduced because of loss of organized mechanical contraction during AF (^54,55). This substrate of decreased flow within the LA/LAA has been associated with spontaneous echo contrast, thrombus formation, and embolic events (^56–62). LAA flow velocities are lower in patients with atrial flutter than what is usually seen with normal sinus rhythm but are higher than with AF. Whether this accounts for the slightly lower prevalence of LAA thrombus and perhaps a lower rate of thromboembolism associated with atrial flutter is uncertain.

In patients with AF, independent predictors of spontaneous echo contrast include LA size, LAA flow velocity (^56,63), left ventricular (LV) dysfunction, fibrinogen level (⁶²), hematocrit (^61,62), and aortic atherosclerosis (^61,62,64,65). This phenomenon might be an echocardiographic surrogate for regional coagulopathy and, when dense, is of clinical value to identify AF patients at high risk for thromboembolism (⁶⁴), but its utility for prospective risk stratification for thromboembolism beyond that achieved by clinical assessment has not been established. Although conventional clinical management is based on the presumption that thrombus formation requires continuation of AF for approximately 48 h, thrombi have been identified by transesophageal echocardiography (TEE) within shorter intervals (^66,67).

Contrary to the prevalent view that systemic anticoagulation for 4 weeks results in endocardial adherence and organization of LAA thrombus, TEE studies have verified resolution of thrombus in the majority of patients (⁶⁸). Similar observations have defined the transient nature of LA/LAA dysfunction on conversion of AF, which provides a mechanistic rationale for anticoagulation for several weeks before and after successful cardioversion.

2. Clinical Implications
Because the pathophysiology of thromboembolism in patients with AF is uncertain, the mechanisms that link risk factors to ischemic stroke in AF are also incompletely defined. The strong association between hypertension and stroke in AF is probably mediated primarily by embolism that originates in the LAA (⁴⁹), but hypertension also increases the risk of noncardioembolic strokes in AF (^49,69). Hypertension in AF patients is associated with reduced LAA flow velocity and spontaneous echo contrast, which predisposes the patient to thrombus formation (^63,64,70). Ventricular diastolic dysfunction might underlie the effect of hypertension on LA dynamics (^71,72). The effect of advancing age to increase stroke risk in AF is multifactorial. In patients with AF, aging is associated with LA enlargement, reduced LAA flow velocity, and spontaneous echo contrast, each of which predisposes to LA thrombus formation (^30,63,64). Additionally, age is a risk factor for atherosclerosis, including complex aortic arch plaque, and is associated with stroke independently of AF (⁶⁵). LV systolic dysfunction predicts ischemic stroke in AF patients receiving no antithrombotic therapy (^73–76).

	VI. Associated Conditions, Clinical Manifestations, and Quality of Life

Top I. Introduction II. Definitions III. Classification IV. Epidemiology and Prognosis V. Pathophysiological Mechanisms VI. Associated Conditions,... VII. Clinical Evaluation VIII. Management IX. Proposed Management... References

 
A. Ca uses and Associated Conditions
1. Acute Causes of AF
AF can be related to acute, temporary causes, including alcohol intake, surgery, electrocution, myocarditis, pulmonary embolism, other pulmonary diseases, and hyperthyroidism. Successful treatment of the underlying condition can eliminate AF. AF is a common early postoperative complication of myocardial infarction and of cardiac or thoracic surgery.

2. AF Without Associated Cardiovascular Disease
In younger patients, approximately 30% to 45% of paroxysmal cases and 20% to 25% of persistent cases of AF occur as lone AF (^13,15,16).

3. AF With Associated Cardiovascular Disease
Specific cardiovascular conditions associated with AF include valvular heart disease (most often mitral), coronary artery disease (CAD), and hypertension, particularly when LV hypertrophy (LVH) is present.

4. Neurogenic AF
The autonomic nervous system can trigger AF in susceptible patients through heightened vagal or adrenergic tone. Many patients experience onset of AF during periods of enhanced parasympathetic or sympathetic tone. Coumel (⁷⁷) described a group of patients that he characterized in terms of a vagal or adrenergic form of AF. Patients with pure vagal or adrenergic AF are uncommon, but if the history reveals a pattern of onset of AF with features of one of these syndromes, the clinician can select agents more likely to prevent recurrent episodes.

B. Clinical Manifestations
AF can be symptomatic or asymptomatic, even in the same patient. Symptoms vary with the ventricular rate, underlying functional status, duration of AF, and individual patient perceptions. The dysrhythmia can present for the first time with an embolic complication or exacerbation of HF. Most patients with AF complain of palpitations, chest pain, dyspnea, fatigue, or lightheadedness. Release of atrial natriuretic peptide can be associated with polyuria. AF can lead to tachycardia-mediated cardiomyopathy, especially in patients who are unaware of the arrhythmia. Syncope is an uncommon but serious complication that usually indicates sinus node dysfunction, valvular aortic stenosis, HCM, cerebrovascular disease, or an accessory AV pathway.

C. Quality of Life
Although strokes certainly account for much of the functional impairment associated with AF, the rhythm disturbance also decreases quality of life. In the Stroke Prevention in Atrial Fibrillation (SPAF) study cohort, the New York Heart Association functional classification, developed for HF, was an insensitive index of quality of life in patients with AF (⁷⁸). In another study (⁷⁹), 47 (68%) of 69 patients with paroxysmal AF considered the dysrhythmia disruptive of their lives. This perception was not associated with either the frequency or duration of symptoms. The AFFIRM trial (Atrial Fibrillation Follow-up Investigation of Rhythm Management), which is still in progress, is comparing maintenance of sinus rhythm with rate control in patients with AF, addressing many facets of quality of life, as did the smaller PIAF (Pharmacological Intervention in Atrial Fibrillation) study (⁸⁰). In selected patients, radiofrequency catheter ablation of the AV node and pacemaker insertion improved quality-of-life scores compared with medical therapy (^81–86).

Long-term oral anticoagulant therapy, which involves frequent blood testing and multiple drug interactions, is another factor with important implications for the quality of life of AF patients. Decision analysis found that only 61% of 97 patients preferred anticoagulation therapy to no treatment (⁸⁷), a considerably smaller proportion than that for whom treatment has been recommended according to published guidelines.

	VII. Clinical Evaluation

Top I. Introduction II. Definitions III. Classification IV. Epidemiology and Prognosis V. Pathophysiological Mechanisms VI. Associated Conditions,... VII. Clinical Evaluation VIII. Management IX. Proposed Management... References

 
A. Minimum Evaluation of the Patient With AF
1. Clinical History and Physical Examination
The initial evaluation of a patient with suspected or proven AF includes characterizing the pattern of the arrhythmia as paroxysmal or persistent, determining its cause, and defining associated cardiac and extracardiac factors (Table 1). A careful history will result in a well-planned, focused workup that serves as an effective guide to therapy (²). The workup of an AF patient can usually take place and therapy can be initiated in 1 outpatient encounter. Delay occurs when the rhythm has not been specifically documented and additional monitoring is necessary.

View this table: [in this window] [in a new window]   Table 1. Minimum and Additional Clinical Evaluation of Patients With Atrial Fibrillation

View this table: [in this window] [in a new window]   Table 2. Vaughan Williams Classification of Antiarrhythmic Drug Actions

The physical examination may suggest AF on the basis of irregular pulse, irregular jugular venous pulsations, and variation in the loudness of the first heart sound. Examination may also disclose associated valvular heart disease, myocardial abnormalities, or HF. The findings on examination are similar in patients with atrial flutter, except that the rhythm may be regular and rapid venous oscillations may occasionally be visible in the jugular pulse.

2. Investigations
The diagnosis of AF requires ECG documentation by at least single-lead ECG recording during the dysrhythmia, which may be facilitated by review of emergency department records, Holter monitoring, or transtelephonic or telemetric recordings. A portable ECG recording tool may help establish the diagnosis in cases of paroxysmal AF and provide a permanent ECG record of the dysrhythmia. If episodes are frequent, then a 24-h Holter monitor can be used. If episodes are infrequent, then an event recorder, which allows the patient to transmit the ECG to a recording facility when the arrhythmia occurs, may be more useful.

B. Additional Investigation of Selected Patients With AF
Additional investigation may include Holter monitoring and exercise testing, transesophageal echocardiography, and/or electrophysiological study. See Table 1 for details.

	VIII. Management

Top I. Introduction II. Definitions III. Classification IV. Epidemiology and Prognosis V. Pathophysiological Mechanisms VI. Associated Conditions,... VII. Clinical Evaluation VIII. Management IX. Proposed Management... References

 
The major issues in management of patients with AF are related to the arrhythmia itself and to prevention of thromboembolism. In patients with persistent AF, there are fundamentally 2 ways to manage the dysrhythmia: to restore and maintain sinus rhythm or to allow AF to continue and ensure that the ventricular rate is controlled.

A. Rhythm Control vs Heart Rate Control
Reasons for restoration and maintenance of sinus rhythm in patients with AF include relief of symptoms, prevention of embolism, and avoidance of cardiomyopathy.

B. Cardioversion
1. Basis for Cardioversion of AF
Cardioversion is often performed electively to restore sinus rhythm in patients with persistent AF. The need for cardioversion can be immediate, however, when the arrhythmia is the main factor responsible for acute HF, hypotension, or worsening of angina pectoris in a patient with CAD. Nevertheless, cardioversion carries a risk of thromboembolism unless anticoagulation prophylaxis is initiated before the procedure, and this risk appears to be greatest when the arrhythmia has been present more than 48 h.

2. Methods of Cardioversion
Cardioversion can be achieved by means of drugs or electrical shocks. Drugs were commonly used before electrical cardioversion became a standard procedure. The development of new drugs has increased the popularity of pharmacological cardioversion, although some disadvantages persist, including the risk of drug-induced torsade de pointes ventricular tachycardia or other serious arrhythmias. Pharmacological cardioversion is still less effective than electrical cardioversion, but the latter requires conscious sedation or anesthesia, whereas the former does not.

The risk of thromboembolism or stroke does not differ between pharmacological and electrical cardioversion. Thus, recommendations for anticoagulation are the same for both methods.

C. Pharmacological Cardioversion
Pharmacological cardioversion appears to be most effective when initiated within 7 days after the onset of AF (^88–91). Most such patients have paroxysmal AF, a first-documented episode of AF, or an unknown pattern of AF at the time of treatment. (See Section III, Classification.) A large proportion of patients with recent-onset AF experience spontaneous cardioversion within 24 to 48 h (^92–94). This is less likely to occur when AF has persisted for more than 7 days.

The relative efficacy of various drugs differs for pharmacological cardioversion of AF and atrial flutter, yet many studies of drug therapy for AF have included patients with atrial flutter. The dose, route, and rapidity of administration influence efficacy. Reference is made to the Vaughan Williams classification of antiarrhythmic drugs (⁹⁵), which has been modified to include recently available drugs (Table 2) . A summary of recommendations is presented in Tables 10through12 (see Section IX-B, Recommendations for Pharmacological and Electrical Cardioversion of AF, Tables 10–12). Although clinical use of the antiarrhythmic drugs listed has been approved by regulatory agencies, therapeutic use for AF has not been mentioned or approved in all cases in each country. The recommendations given in this document do not necessarily adhere to governmental regulations and labeling requirements.

View this table: [in this window] [in a new window]   Table 10. Recommendations for Pharmacological Cardioversion of AF Less Than or Equal to 7 Days Duration**

View this table: [in this window] [in a new window]   Table 11. Recommendations for Pharmacological Cardioversion of AF More Than 7 Days Duration**

View this table: [in this window] [in a new window]   Table 12. Recommended Doses of Drugs Proven Effective for Pharmacological Cardioversion of Atrial Fibrillation

A frequent issue related to pharmacological cardioversion is whether the antiarrhythmic drug should be started in the hospital or on an outpatient basis. The major concern is the potential for serious adverse effects, including torsade de pointes ventricular tachycardia. With the exception of those involving low-dose oral amiodarone (⁹⁶), virtually all studies of pharmacological cardioversion have been limited to hospitalized patients.

D. Electrical Cardioversion
Direct-current cardioversion involves an electrical shock synchronized with the intrinsic activity of the heart. This ensures that electrical stimulation does not occur during the vulnerable phase of the cardiac cycle (⁹⁷).

Successful cardioversion of AF depends on the nature of the underlying heart disease and the current density delivered to the atrial myocardium. The latter, in turn, depends on the voltage of the defibrillator capacitor, the output waveform, the size and position of the electrode paddles, and transthoracic impedance.

In a randomized controlled study of 301 subjects undergoing elective external cardioversion, patients were allocated to anterior-lateral (ventricular apex and right infraclavicular) or anterior-posterior (sternum and left scapular) paddle positions (⁹⁸). The energy requirement was lower and overall success was greater with the anterior-posterior configuration (87%) than with the anterior-lateral alignment (76%).

Cardioversion is performed with the patient having fasted and under adequate anesthesia. Short-acting anesthetic agents are preferred, because cardioversion patients are well suited to day care and should recover rapidly after the procedure (⁹⁹).

An initial shock of 100 J is often too low, and an initial energy of 200 J or greater is recommended for electrical cardioversion of AF. Devices that deliver current with a biphasic waveform appear to achieve cardioversion at lower energy levels than those that use a monophasic waveform. The primary success rate as measured 3 days after cardioversion in 100 consecutive subjects was 86% (¹⁰⁰); this increased to 94% when the procedure was repeated during treatment with quinidine or disopyramide. Only 23% of the patients remained in sinus rhythm after 1 year and 16% after 2 years; in those who relapsed, repeated cardioversion with antiarrhythmic medication resulted in sinus rhythm in 40% and 33% after 1 and 2 years, respectively. For patients who relapsed again, a third cardioversion resulted in sinus rhythm in 54% at 1 year and 41% at 2 years. Thus, sinus rhythm can be restored in a substantial proportion of patients by direct-current cardioversion, but the rate of relapse is high unless antiarrhythmic drug therapy is given concomitantly.

Cardioversion of patients with implanted pacemaker and defibrillator devices is safe when appropriate precautions are taken. The device should be interrogated immediately before and after cardioversion to verify appropriate function. The paddles used for external cardioversion should be positioned as distant as possible from the device, preferably in the anterior-posterior configuration.

The risks of electrical cardioversion are mainly related to embolic events and cardiac arrhythmias. Thromboembolic events have been reported in 1% to 7% of patients who did not receive anticoagulation before cardioversion (^101,102). Various brief arrhythmias might arise, especially ventricular and supraventricular premature beats, bradycardia, and short periods of sinus arrest (¹⁰³). Ventricular tachycardia and fibrillation can be precipitated in patients with hypokalemia or digitalis intoxication (^104,105). A slow ventricular response to AF in the absence of drugs that slow AV nodal conduction can indicate conduction defect. The patient should be evaluated before cardioversion with these issues in mind to avoid symptomatic bradycardia (¹⁰⁶). Transient ST-segment elevation can appear on the ECG after cardioversion (^107–108) and blood levels of creatine kinase-MB can rise even without apparent myocardial damage.

Prophylactic drug therapy to prevent early recurrence of AF should be considered individually for each patient. Should relapse (particularly early relapse) occur, antiarrhythmic therapy is recommended in conjunction with the second attempt. Further cardioversion is of limited value. In highly symptomatic patients, infrequently repeated cardioversion can be an acceptable approach.

E. Maintenance of Sinus Rhythm
1. Pharmacological Therapy to Prevent Recurrence of AF
Maintenance of sinus rhythm is relevant in patients with paroxysmal AF (in whom episodes terminate spontaneously) and persistent AF (in whom electrical or pharmacological cardioversion is necessary to restore sinus rhythm). Whether paroxysmal or persistent, AF is a chronic disorder, and recurrence is likely at some point in most patients (Fig. 4) (^109,110). Prophylactic treatment with antiarrhythmic drugs is therefore often necessary. The goal of maintenance therapy is suppression of symptoms and sometimes prevention of tachycardia-induced cardiomyopathy due to AF. It is not yet known whether maintenance of sinus rhythm prevents thromboembolism, HF, or death (^80,111).

View larger version (19K): [in this window] [in a new window]   Figure 4. Arrhythmia-free survival after electrical cardioversion in patients with persistent atrial fibrillation. The lower curve represents outcome after a single shock when no prophylactic drug therapy was given. The upper curve depicts the outcome with repeated electrical cardioversions in conjunction with antiarrhythmic drug prophylaxis. ECV indicates electrical cardioversion; SR, sinus rhythm. Reproduced with permission from van Gelder et al., Arch Intern Med 1996;156:2585–92, © 1996, American Medical Association (110).

Recurrence of AF is not equivalent to treatment failure. In several studies (^112,113), patients with recurrent AF often chose to continue drug treatment. Sometimes, reduction of the arrhythmia burden can constitute partial success, whereas to other patients, any recurrence of AF may seem intolerable. Nevertheless, most patients with AF eventually experience recurrence. Risk factors for frequent recurrence of paroxysmal AF (more than 1 episode per month) include female sex and underlying heart disease (¹¹⁴). In persistent AF, the 4-year arrhythmia-free survival was less than 10% after single-shock electrical cardioversion without prophylactic drug therapy (¹¹⁰). Predictors of recurrences within that interval included hypertension, age greater than 55 years, and AF duration greater than 3 months. In that study, serial cardioversions and prophylactic drug therapy resulted in freedom from recurrent AF in only approximately 30% of patients (¹¹⁰). Predictors of recurrence included age greater than 70 years, AF duration greater than 3 months, and HF (¹¹⁰).

2. General Approach to Antiarrhythmic Drug Therapy
Before any antiarrhythmic agent is administered, reversible cardiovascular and noncardiovascular precipitants of AF should be addressed. Prophylactic drug treatment is seldom indicated in case of a first-detected episode of AF and can also be avoided in patients with infrequent and well-tolerated paroxysmal AF. Similarly, when recurrences are infrequent and tolerated, patients experiencing breakthrough arrhythmias might not require a change in antiarrhythmic drug therapy. Beta-blockers can be effective in patients who develop AF only during exercise, but few patients have a single specific inciting cause for all episodes of AF, and a majority will not sustain sinus rhythm without antiarrhythmic drug treatment.

In patients with lone AF, a beta-blocker may be tried first, but flecainide, propafenone, and sotalol are particularly effective. Amiodarone and dofetilide are recommended as alternative therapy. Quinidine, procainamide, and disopyramide are not favored unless amiodarone fails or is contraindicated. The anticholinergic activity of long-acting disopyramide makes this a relatively attractive choice for patients with a predilection to vagally induced AF. Propafenone is not recommended in vagally mediated AF because its (weak) intrinsic beta-blocking activity might aggravate this type of paroxysmal AF. In patients with adrenergically mediated AF, beta-blockers represent first-line treatment, followed by sotalol and amiodarone. In patients with adrenergically mediated lone AF, amiodarone should be chosen later in the sequence of drug therapy because of its potential toxicity (see Fig. 11, Section IX). Overall, when treatment with a single drug fails, combinations of antiarrhythmic drugs may be tried. Useful combinations include a beta-blocker, sotalol or amiodarone, plus a type IC agent.

View larger version (23K): [in this window] [in a new window]   Figure 11. Antiarrhythmic drug therapy to maintain sinus rhythm in patients with recurrent paroxysmal or persistent atrial fibrillation. Drugs are listed alphabetically and not in order of suggested use. *For adrenergic atrial fibrillation, beta-blockers or sotalol are the initial drugs of choice. Consider nonpharmacological options to maintain sinus rhythm if drug failure occurs. HF indicates heart failure; CAD, coronary artery disease; and LVH, left ventricular hypertrophy.

A drug that is initially safe might become proarrhythmic when the patient develops CAD or HF or begins taking other medication that in combination can be arrhythmogenic. Thus, the patient should be alerted to the potential significance of such symptoms as syncope, angina pectoris, or dyspnea and warned about the use of noncardiac drugs that can prolong the QT interval. A useful source of information is the Internet site http://www.torsades.org.

Monitoring of antiarrhythmic drug treatment varies with the agent involved and with patient factors. Prospective trial data on upper limits of drug-induced increases in QRS duration or QT prolongation are not available. The following recommendations are the consensus of the writing committee. With type IC drugs, QRS widening should not exceed 150% of the pretreatment QRS duration. Exercise testing can be helpful to detect QRS widening that occurs only at rapid heart rates (use-dependent conduction slowing). For type IA or type III drugs, with the possible exception of amiodarone, the corrected QT interval in sinus rhythm should remain below 520 ms. During follow-up, plasma potassium and magnesium levels and renal function should be checked periodically, because renal insufficiency leads to drug accumulation and predisposes to proarrhythmia. In individual patients, serial noninvasive tests can be appropriate to reevaluate LV function, especially if clinical HF develops during treatment of AF.

3. Pharmacological Agents to Maintain Sinus Rhythm
Fourteen controlled trials of drug prophylaxis involving patients with paroxysmal AF have been published, and there have been 22 published trials of drug prophylaxis for maintenance of sinus rhythm in patients with persistent AF. Comparative data are not sufficient to permit subclassification by drug or etiology. Individual drugs and dosages for maintenance of sinus rhythm are given in Table 3.

View this table: [in this window] [in a new window]   Table 3. Typical Doses of Drugs Used to Maintain Sinus Rhythm in Patients With Atrial Fibrillation**

4. Out-of-Hospital Initiation of Antiarrhythmic Drugs in Patients With AF
Recommendations for out-of-hospital initiation or intermittent use of antiarrhythmic drugs differ for patients with paroxysmal and persistent AF. In patients with paroxysmal AF, the aims are to stop an attack, to prevent recurrences, or a combination of both. In patients with persistent AF, the aims of out-of-hospital drug initiation are to achieve pharmacological cardioversion of AF, thereby obviating the need for electrical cardioversion, or to enhance the success of electrical cardioversion and prevent early recurrence of AF.

Few prospective data are available on the safety of outpatient initiation of antiarrhythmic drug therapy. Proarrhythmia (Table 4) is rare in patients with normal ventricular function and baseline QT intervals (^41,115) without profound bradycardia. As long as sinus or AV node dysfunction is not suspected, propafenone or flecainide may be initiated out of hospital. Sudden death related to idiopathic ventricular fibrillation in a structurally normal heart can occur in patients with the Brugada syndrome, an inherited cardiac disease characterized by ST-segment elevation in the right precordial ECG leads and frequently accompanied by right bundle-branch block. Type I antiarrhythmic drugs can unmask this condition (^116,117). Before therapy with these agents is begun, a beta-blocker or calcium channel antagonist should be given to prevent rapid AV conduction or 1:1 AV conduction if atrial flutter develops (^118–122). Because termination of paroxysmal AF with flecainide or propafenone can be associated with bradycardia due to sinus node or AV node dysfunction, an initial conversion should be undertaken in the hospital before a patient is declared fit for outpatient "pill-in-the pocket" use of these agents for conversion of recurrences. Out-of-hospital drug termination should be avoided in patients with symptomatic sick sinus syndrome, AV conduction disturbances, or bundle-branch block.

View this table: [in this window] [in a new window]   Table 4. Types of Proarrhythmia During Treatment With Various Antiarrhythmic Drugs for Atrial Fibrillation or Atrial Flutter According to the Vaughan Williams Classification

Sotalol may be initiated in outpatients with little or no heart disease as long as the baseline uncorrected QT interval is less than 450 ms, serum electrolytes are normal, and none of the type III drug-related proarrhythmia risk factors are present. Safety is greatest when sotalol is started when the patient is in sinus rhythm. Amiodarone can usually be given safely on an outpatient basis, even in patients with persistent AF (¹²³), but in-hospital loading might be more appropriate when earlier restoration of sinus rhythm is needed, as in patients with HF. Some loading regimens involve giving either 600 mg per day for 4 weeks (¹²³) or greater than or equal to 1 g per day for 1 week (¹²⁴) followed by lower maintenance doses. Quinidine, procainamide, and disopyramide should generally not be started out of hospital. An exception can be made for disopyramide in patients without heart disease who have a normal QT interval. Currently, the standards for use of dofetilide do not permit out-of-hospital initiation.

Transtelephonic monitoring or other ECG surveillance methods can be used to monitor conduction disturbances as pharmacological antiarrhythmic therapy is initiated in patients with AF. Specifically, the PR interval (flecainide, propafenone, sotalol, and amiodarone), QRS duration (flecainide and propafenone), and QT interval (sotalol, amiodarone, and disopyramide) should be measured. As a general rule, antiarrhythmic drugs should be started at a relatively low dose with upward titration as needed, reassessing the ECG as each dose change is made. The dose of other medication used for rate control should be reduced approximately 6 weeks after initiation of amiodarone and stopped if the rate slows excessively. Concomitant drug therapies should be monitored closely, and both the patient and the physician should be alert to possible deleterious drug interactions.

5. Recurrence of AF After Cardioversion: Implications for Drug Treatment
Although it has long been known that most recurrences of AF occur within the first month after electrical cardioversion, recent research with internal atrial cardioversion (¹²⁵) and day-to-day postconversion studies (¹²⁶) have established several patterns of recurrence (Fig. 5) . In some cases, there is complete failure of direct-current countershock to elicit even a single isolated sinus or ectopic atrial beat, tantamount to a high atrial defibrillation threshold. In others, AF reappears within 1 to 2 minutes after a period of sinus rhythm (^127,128). Sometimes recurrence is delayed from 1 day to 2 weeks (¹²⁶) or more after the shock. Complete shock failure and immediate recurrences occur in approximately 25% of patients undergoing electrical cardioversion, and subacute recurrences occur in about an equal proportion within 2 weeks (¹²⁷).

View larger version (16K): [in this window] [in a new window]   Figure 5. Hypothetical illustration of cardioversion failure.Three types of recurrences after electrical cardioversion of persistent AF are shown The efficacy of drugs varies in enhancement of shock conversion and suppression of recurrences ECV indicates external cardioversion; IRAF, first recurrence of AF after cardioversion. Modified with permission from Van Gelder et al (129).

Available data suggest that starting pharmacological therapy before electrical cardioversion (Table 5) can enhance immediate success and suppress early recurrences. As a corollary, it seems appropriate to establish therapeutic plasma drug concentrations at the time of cardioversion and for a few weeks thereafter to optimize the chance of success. After cardioversion to sinus rhythm, patients receiving drugs that prolong QT interval should be observed in the hospital for 24 to 48 h to evaluate the effects of heart rate slowing and to allow prompt intervention in the event torsade de pointes develops. Pharmacological agents may be started out of hospital or in hospital immediately before electrical cardioversion. The risks of pretreatment include the possibility of a paradoxical increase in the defibrillation threshold, as has occurred with flecainide (¹³⁰); accelerated ventricular rate during loading with type IA or IC drugs in the absence of an AV nodal blocking agent (^{118–122,131}) and ventricular proarrhythmia (Table 4).

View this table: [in this window] [in a new window]   Table 5. Pharmacological Treatment Before Cardioversion in Patients With Persistent Atrial Fibrillation: Effects of Various Antiarrhythmic Drugs on Acute and Subacute Outcome of Transthoracic Direct Current Shock

Pretreatment with pharmacological agents is most appropriate in patients who have previously failed to respond to electrical cardioversion and in those who developed immediate or subacute recurrence of AF. In patients with late recurrence and those undergoing initial cardioversion of persistent AF, pretreatment is optional (Fig. 10 and 11; see Section IX).

6. Selection of Antiarrhythmic Agents in Patients With Certain Cardiac Diseases
a. Heart Failure

Patients with congestive HF are particularly prone to the ventricular proarrhythmic effects of antiarrhythmic drugs. Randomized trials have demonstrated the safety of amiodarone and dofetilide (given separately) in patients with HF (^132,133), and these are the recommended drugs for maintenance of sinus rhythm (Fig. 11; see Section IX).

b. Coronary Artery Disease

In stable patients with CAD, beta-blockers may be considered first, but their use is supported by only 2 studies (^134–135); data on their efficacy for maintenance of sinus rhythm in patients with persistent AF after cardioversion are not convincing (¹³⁴). Sotalol has substantial beta-blocking activity and can therefore be chosen as the initial antiarrhythmic agent in AF patients with ischemic heart disease, because it is associated with less long-term toxicity than amiodarone (Fig. 11; see Section IX). Both sotalol and amiodarone have reasonable short-term safety profiles, and amiodarone might be preferred in patients with HF (^136–138). Flecainide (¹³⁹) and propafenone are not recommended in these situations. Quinidine, procainamide, and disopyramide are considered third-line treatment choices in patients with CAD. Given the results of the DIAMOND-MI trial (Danish Investigations of Arrhythmias and Mortality on Dofetilide in Myocardial Infarction), dofetilide may be considered as a second-line rather than a third-line antiarrhythmic agent, but that study involved selected post-myocardial infarction patients in whom the antiarrhythmic benefit balanced the risk of proarrhythmic toxicity.

c. Hypertensive Heart Disease

Patients with LVH are at increased risk of developing torsade de pointes related to early ventricular afterdepolarizations (^140–142). Thus, a drug that does not prolong the QT interval is preferable as first-line therapy, and in the absence of CAD or marked ventricular hypertrophy, propafenone and flecainide are reasonable choices (Fig. 11; Section IX). Amiodarone prolongs the QT interval but carries a very low risk of ventricular proarrhythmia; its extracardiac toxicity profile relegates it to second-line therapy in these individuals, but amiodarone becomes first-line therapy when marked LVH is present. When amiodarone and sotalol either fail or are inappropriate, disopyramide, quinidine, or procainamide can be used as alternatives.

d. WPW Syndrome

Radiofrequency ablation of the accessory pathway is the preferred therapy for patients with preexcitation syndromes and AF, but antiarrhythmic drugs might be useful. Digoxin should be avoided because of the risk of paradoxical acceleration of the ventricular rate during AF. Beta-blockers do not decrease conduction over accessory pathways during preexcited periods of AF.

7. Nonpharmacological Correction of AF
a. Surgical Ablation
Based on mapping studies of animal and human AF, Cox et al (^143–146) developed a surgical procedure (maze operation) that controls AF in more than 90% of selected patients. The exact mechanism has not been established conclusively, but the creation of barriers to conduction within the RA and LA limits the amount of myocardium available to propagate reentrant wave fronts, thereby inhibiting sustained AF. Incisions encircling the pulmonary veins can prevent initiation of AF by isolating potentially arrhythmogenic foci near the pulmonary veins from the remainder of the atria or by isolating atrial regions with the shortest refractory periods. Modifications of the Cox maze operation involve encircling the pulmonary veins by surgical incisions within the LA and radial incisions in both atria that join the mitral and tricuspid valve annuli (^147–149).

Surgical operations for AF have been successfully combined with operative correction of a variety of structural cardiac conditions. In patients with highly symptomatic AF who require open heart operations for valvular, ischemic, or congenital heart disease, consideration should be given to concomitant maze operation for AF or flutter, although this entails additional risk. The mortality rate of an isolated maze operation is less than 1%, but mortality is higher when the procedure is combined with other types of operative repair. The morbidity associated with the maze operation includes consequences common to median sternotomy and cardiopulmonary bypass, as well as a risk of short-term fluid retention (owing to reduced release of atrial natriuretic peptide), transient reduction in LA and RA transport function, and early postoperative atrial tachyarrhythmias. In addition, when the blood supply to the sinus node is disrupted, sinus node dysfunction might require permanent pacemaker implantation. Progressive iterations of these operations have reduced the risk of this complication to less than 10%.

b. Catheter Ablation

Based on the success of surgical approaches to AF, several catheter ablation strategies have been designed to produce similar effects (^150–152). Ablation strategies limited to the RA produce marginal improvement (¹⁵⁰), whereas linear ablation in the LA has been more successful. The recognition that foci triggering AF often originate within the pulmonary veins has led to ablation strategies that target this zone or electrically isolate the pulmonary vein from the LA. Other sites of arrhythmogenic foci have been found in the superior vena cava, the RA and LA, and the coronary sinus. Ablation of these foci eliminates or reduces the frequency of recurrent AF in more than 60% of patients, but the risk of recurrent AF after a focal ablation procedure is still 30% to 50% over the first year and even higher when more than 1 pulmonary vein is involved. Thus, many patients continue to require antiarrhythmic drug therapy after ablative therapy of AF (¹⁵³). Potential complications of catheter ablation for AF include systemic embolism, pulmonary vein stenosis, pericardial effusion, cardiac tamponade, and phrenic nerve paralysis. Thus, although these procedures have produced promising results, they have not yet been widely applied.

Atrial flutter can develop not only as a distinct arrhythmia but also during antiarrhythmic drug therapy of AF, especially with type IC agents. Catheter ablation is more effective than antiarrhythmic drugs for treatment of atrial flutter, reducing the recurrence rate from 93% to 5% when used as first-line therapy (³⁷). In addition, the risk of developing AF might also be lower after catheter ablation of atrial flutter than with pharmacological therapy (29% vs 60% over the first year).

c. Suppression of AF by Pacing

Although atrium-based pacing is associated with a lower risk of AF and stroke than ventricle-based pacing for patients requiring pacemakers for bradyarrhythmias, the use of pacing as a primary therapy for prevention of recurrent AF has not been validated.

d. Internal Atrial Cardioverter/Defibrillators

There has been an interest in internal cardioversion of AF for the past 10 years. Attempts have been made to find shock waveforms that reduce the energy requirements for cardioversion, making the shock tolerable to awake patients. One cardioverter (¹⁵⁴) capable of atrial sensing and cardioversion as well as ventricular sensing and pacing was evaluated in 290 AF patients. The conversion rate to sinus rhythm was 93%. Another device, with dual-chamber sensing, pacing, and cardioversion/defibrillation capabilities and a maximum output of 27 J, is a ventricular defibrillator with atrial cardioversion capabilities. It was designed to treat both atrial and ventricular arrhythmias with pacing modalities before delivering low-energy shocks. Potential candidates for atrial cardioverter/defibrillators, mainly those with infrequent episodes of poorly tolerated AF, are often suitable for catheter ablation.

F. Rate Control During AF
1. Pharmacological Approach
An alternative to maintenance of sinus rhythm in patients with paroxysmal or persistent AF is control of the ventricular rate. In a recent randomized trial, the therapeutic strategies of rate versus rhythm control yielded similar clinical results with respect to symptoms in patients with AF, but exercise tolerance was better with rhythm control (¹¹¹). The results of other studies comparing these 2 strategies are not yet available (⁸⁰). The rate is generally considered controlled when the ventricular response is between 60 and 80 bpm at rest and between 90 to 115 bpm during moderate exercise (^156,157). Heart rate variability during AF provides additional information about the status of the autonomic nervous system, which might have independent prognostic implications (^158–161).

Negative chronotropic therapy of AF is based mainly on depression of conduction across the AV node. Sinus bradycardia and heart block occur in some patients with paroxysmal AF, particularly the elderly, as an