Antiarrhythmic drugs to maintain sinus rhythm in patients with atrial fibrillation: Recommendations

INTRODUCTION — For patients with atrial fibrillation (AF), there are two main strategies to manage the irregular rhythm and its impact on symptoms: rhythm control (restoration followed by maintenance of sinus rhythm with either antiarrhythmic drugs or catheter ablation); and rate control with atrioventricular (AV) nodal blockers. (See 'Initial management decisions' below.)

For those patients in whom a rhythm control strategy is chosen, the main goal of therapy is to reduce symptoms by decreasing the frequency and duration of episodes as well as the symptoms during recurrences [1,2]. As antiarrhythmic drugs are associated with a potential for serious adverse side effects, particularly the induction of proarrhythmia, they should be prescribed only by practitioners familiar with their use. Patients should be fully informed of both the benefits and risks associated with the use of these drugs. (See 'Drug-related arrhythmias and mortality' below.)

Rhythm control can be achieved with either antiarrhythmic drug therapy or nonpharmacologic methods. This topic provides recommendations for the former. The clinical trials describing the efficacy and toxicity (including proarrhythmia) of the different antiarrhythmic drugs are presented separately. (See "Antiarrhythmic drugs to maintain sinus rhythm in patients with atrial fibrillation: Clinical trials".)

Nonpharmacologic methods to maintain sinus rhythm (including surgery and radiofrequency ablation or cryoballoon ablation) in selected patients who are refractory to conventional therapy are discussed elsewhere. (See "Atrial fibrillation: Catheter ablation" and "Atrial fibrillation: Surgical ablation".)

INDICATIONS — There are three settings in which a rhythm control strategy for the maintenance of sinus rhythm should be considered [3]:

●Persistent symptoms (palpitations, dyspnea, lightheadedness, angina, syncope, and heart failure) despite adequate rate control.

●An inability to attain adequate rate control (to prevent tachycardia-mediated cardiomyopathy). (See "Arrhythmia-induced cardiomyopathy".)

●Patient preference. Some patients will strongly prefer to avoid either paroxysmal or persistent AF.

Maintenance antiarrhythmic drug therapy is not routinely used after cardioversion in patients with newly detected AF [3]. These issues are discussed in detail separately. (See "Management of atrial fibrillation: Rhythm control versus rate control".)

INITIAL MANAGEMENT DECISIONS — Prior to selecting and initiating antiarrhythmic drug therapy, the following issues should be considered.

●Rhythm versus rate control: The choice between a rhythm- or a rate-control strategy is determined by many factors, including patient age, the degree to which symptoms interfere with the quality of life, and concerns about antiarrhythmic drug therapy or catheter ablation.

Our recommendations for the use of these two strategies are found elsewhere. (See "Management of atrial fibrillation: Rhythm control versus rate control", section on 'Summary and recommendations'.)

●Precipitating factors: Before initiating a rhythm control strategy, any risk factors for atrial fibrillation (AF) should be addressed. Examples include hyperthyroidism, hypertension, heart failure, sleep apnea, and excess alcohol intake. (See "Epidemiology, risk factors, and prevention of atrial fibrillation" and "Epidemiology, risk factors, and prevention of atrial fibrillation", section on 'Chronic disease associations'.)

Maintenance antiarrhythmic drug therapy may not be necessary after cardioversion in a patient with a transient or reversible cause (such as cardiac surgery, pericarditis, or pulmonary embolism). An option in such patients is beta blocker therapy after restoration of sinus rhythm, which may provide modest protection against recurrent AF [4]. However, short-term antiarrhythmic therapy can be considered in this situation as the underlying cause is treated in patients who are highly symptomatic.

●Anticoagulation: The proper use of anticoagulation in the period surrounding conversion to sinus rhythm is discussed separately. (See "Prevention of embolization prior to and after restoration of sinus rhythm in atrial fibrillation".)

●Rate control: An atrioventricular (AV) nodal blocker, such as a beta blocker or a rate-slowing calcium channel blocker, is usually started before, or simultaneously with, antiarrhythmic drug therapy in patients who have demonstrated a moderate to rapid ventricular rate (≥110 beats per minute) during AF. Slowing of the rate generally improves symptoms prior to the restoration of sinus rhythm. This therapy is continued while the patient is in sinus rhythm to protect against a rapid ventricular rate should AF recur. This issue is discussed in detail separately. (See "Control of ventricular rate in patients with atrial fibrillation who do not have heart failure: Pharmacologic therapy", section on 'Evaluation and goal ventricular rate'.)

●Restoration of sinus rhythm: Many patients with AF in whom a rhythm control strategy is chosen will need sinus rhythm restored prior to the initiation of long-term antiarrhythmic drug therapy. The restoration of sinus rhythm is discussed in detail elsewhere. (See "Atrial fibrillation: Cardioversion".)

Some patients with relatively infrequent episodes of paroxysmal atrial fibrillation can be managed with antiarrhythmic therapy given only at the time of the episode. This form of outpatient "pill-in-the-pocket" therapy for recurrent AF is discussed separately. (See "Atrial fibrillation: Cardioversion", section on 'Pharmacologic cardioversion'.)

SELECTING AN ANTIARRHYTHMIC DRUG — Once the issues discussed above have been addressed, an antiarrhythmic agent can be chosen. The choice of drug is significantly influenced both by drug and patient characteristics. As with all therapeutic interventions, the choice of agent should take into account the benefit to risk ratio of the therapy chosen. (See 'Proarrhythmia' below.)

Amiodarone, dofetilide, flecainide, propafenone, sotalol, and less commonly dronedarone are the drugs we recommend to maintain sinus rhythm. (See 'Concerns about dronedarone' below.)

For additional information regarding the therapeutic use of these drugs, including information regarding dosing and side effects, the reader is referred to individual UpToDate topics on these drugs or to the individual drug monographs in our drug database.

The following points regarding antiarrhythmic drugs should be kept in mind in choosing therapy:

●Compared to other agents, amiodarone is associated with the greatest likelihood of maintaining sinus rhythm, but also with the highest risk of long-term complications [5,6]. In addition, a 2014 report raises the possibility that amiodarone use in patients taking warfarin is associated with an increased risk of stroke compared to those not taking the drug [7]. In this study, there was a lower time in the therapeutic range (of the international normalized ratio) in patients receiving amiodarone. (See "Antiarrhythmic drugs to maintain sinus rhythm in patients with atrial fibrillation: Clinical trials" and "Amiodarone: Adverse effects, potential toxicities, and approach to monitoring".)

●Quinidine, procainamide, and disopyramide are no longer recommended for patients with AF, except perhaps in patients with vagally mediated atrial fibrillation (AF), as there are more effective drugs and due to extracardiac side effects as well as the concern about proarrhythmia [8]. (See 'Proarrhythmia' below.)

●Beta blockers are modestly effective in maintaining sinus rhythm and can be tried first in selected patients, such as those without structural heart disease who are concerned about proarrhythmia [4,9,10]. Of course beta blockers may have already been initiated to slow the ventricular rate in AF. (See 'Proarrhythmia' below and "Antiarrhythmic drugs to maintain sinus rhythm in patients with atrial fibrillation: Clinical trials", section on 'Beta blockers'.)

The following patient characteristics may influence decision making:

●The clinical features of the patient, such as presence or absence of clinical heart disease. We believe it is prudent to obtain a two-dimensional echocardiogram to screen for structural heart disease (eg, left ventricular systolic dysfunction, left ventricular hypertrophy, or valvular heart disease). An exercise or nuclear stress imaging test may be used to screen for coronary heart disease and is typically done before starting a class IC agent. (See 'Atrial fibrillation without structural heart disease' below and 'Atrial fibrillation associated with structural heart disease' below.)

●The presence of paroxysmal compared to persistent AF [1,11]. As examples, our experts rarely use dofetilide for paroxysmal AF and infrequently choose dronedarone for persistent AF due to reduced efficacy compared with amiodarone.

●The presence of vagally-mediated AF [12,13]. The 2016 European Society of Cardiology AF guideline suggest that, because of its long-lasting anticholinergic activity, disopyramide may be considered in patients with vagally-induced AF (eg, occurring most often in athletic young men with slow heart rates during rest or sleep), as long as the patient does not have prostatism or glaucoma [11,14]. The combination of disopyramide and either a beta blocker or a calcium channel blocker must be used cautiously because of the additive negative inotropic effects. If disopyramide cannot be given or is not tolerated, flecainide and amiodarone represent the sequential alternatives. Our experts use disopyramide cautiously due to concern for proarrhythmia.

●For patients with adrenergically-mediated AF (eg, occurring during exercise or other activity), we suggest beta blockers as first-line therapy, followed by sotalol and amiodarone.

Antiarrhythmic drugs are associated with a potential for serious adverse side effects, particularly the induction of proarrhythmia. Thus, they should be prescribed only by practitioners familiar with their use. Patients should be fully informed of both the benefits and risk associated with the use of these drugs. (See 'Drug-related arrhythmias and mortality' below.)

As the expectation of antiarrhythmic therapy is to reduce the frequency and duration of episodes, improve quality of life, and prevent hospitalization, a recurrence of AF does not necessarily denote a failure of the medication or mandate a change to a different antiarrhythmic drug.

Atrial fibrillation without structural heart disease — Patients without structural heart disease include those with hypertension who do not have left ventricular hypertrophy. The author and reviewers of this topic generally select flecainide or propafenone as the first antiarrhythmic drug for these patients due to its relatively good side effect profile, efficacy, and ease of use. The use of these drugs in patients >70 years of age should be considered more cautiously, given the higher likelihood of underlying coronary artery disease.

In these patients, flecainide, propafenone, amiodarone, dronedarone, sotalol, and dofetilide are superior to placebo for maintaining sinus rhythm. (See "Antiarrhythmic drugs to maintain sinus rhythm in patients with atrial fibrillation: Clinical trials".)

For those patients in whom flecainide or propafenone will not be used as the preferred agent, the following points can influence the choice of antiarrhythmic drug in patients without structural heart disease:

●In the Canadian Trial of Atrial fibrillation, AFFIRM, and the SAFE-T randomized trials, amiodarone was more effective than flecainide, propafenone, or sotalol (which have nearly equivalent efficacy to each other), but has a significantly higher rate of adverse side effects. (See "Antiarrhythmic drugs to maintain sinus rhythm in patients with atrial fibrillation: Clinical trials", section on 'Amiodarone'.)

●In a meta-analysis of trials where the effect of amiodarone versus dronedarone was estimated with the use of indirect comparison and normal logistic meta-analysis models, amiodarone was found to be more effective in maintaining sinus rhythm, but at the expense of greater drug discontinuation secondary to adverse events [15]. (See "Antiarrhythmic drugs to maintain sinus rhythm in patients with atrial fibrillation: Clinical trials", section on 'Dronedarone'.)

Both amiodarone and dronedarone are associated with significant side effects. We suggest carefully discussing these with the patient prior to initiating therapy. (See "Amiodarone: Adverse effects, potential toxicities, and approach to monitoring" and "Clinical uses of dronedarone".)In the EMERALD trial, Dofetilide had a somewhat better efficacy than sotalol. (See "Antiarrhythmic drugs to maintain sinus rhythm in patients with atrial fibrillation: Clinical trials", section on 'Dofetilide'.)

Some cardiologists prefer to use low-dose amiodarone (100 to 200 mg per day), particularly in older patients, in preference to flecainide, sotalol, or dronedarone for two principal reasons: greater efficacy than sotalol and dronedarone (figure 1) [16-18], and a very low incidence of torsades de pointes [19,20]. In addition, since amiodarone has beta blocking and calcium channel blocking activity, the ventricular rate is usually slower and better tolerated if AF does recur. If amiodarone is used for rhythm control, the need for additional medications to control rate (eg, beta blockers or calcium channel blockers) may be decreased. Despite these advantages, low-dose amiodarone still has appreciable toxicity, including thyroid disease, hepatic dysfunction, lung disease, neurologic abnormalities, and bradycardia [19,20]. (See "Amiodarone: Adverse effects, potential toxicities, and approach to monitoring".)

Atrial fibrillation associated with structural heart disease — Amiodarone, sotalol, and dofetilide are the most commonly recommended first-line drugs in patients with structural heart disease (algorithm 1) [11]. Our authors and reviewers prefer either dronedarone or sotalol to amiodarone and dofetilide. Dronedarone is easier to use than sotalol (continuous monitoring of initiation required), but is less efficacious. (See "Clinical uses of sotalol".)

These drugs (with the exception of dronedarone) were used for initial therapy in almost 70 percent of patients in AFFIRM, 88 percent of whom had organic heart disease and/or hypertension [21]. Amiodarone was significantly more effective than sotalol in the CTAF, AFFIRM, and SAFE-T trials [16-18]. However, in SAFE-T, sotalol was as effective as amiodarone in the subgroup of patients with coronary heart disease [17]. (See "Antiarrhythmic drugs to maintain sinus rhythm in patients with atrial fibrillation: Clinical trials".)

Coronary heart disease — In patients with coronary heart disease who do not have heart failure, sotalol, dronedarone, dofetilide, and amiodarone are acceptable choices (table 1 and algorithm 1) [11,22,23]. We prefer sotalol due to its better extracardiac side effect profiles. Flecainide and propafenone are contraindicated in this population. (See "Antiarrhythmic drugs to maintain sinus rhythm in patients with atrial fibrillation: Clinical trials" and "Major side effects of class I antiarrhythmic drugs".)

In the Cardiac Arrhythmia Suppression Trial (CAST) of patients with drug-suppressible ventricular premature beats in the year following a myocardial infarction, flecainide increased mortality compared to placebo (figure 2) [24]. Although propafenone was not used in CAST and may not have the same potential for proarrhythmia as flecainide and encainide [25], it cannot be recommended in patients with underlying heart disease [26].

The extension of this concern to structural heart disease other than coronary artery disease stems in part from the flecainide clinical and safety database, which was used in a retrospective study demonstrating that the presence of structural heart disease including valvular heart disease, congenital heart disease, and cardiomyopathies lead to an alarming increase in proarrhythmia and death [27].

Heart failure — Amiodarone and dofetilide, are used in patients with AF and heart failure (HF) or those with a left ventricular ejection fraction less than 35 percent. Our authors and reviewers are more comfortable using dofetilide in this setting with an implantable defibrillator in place or in younger patients with less severe impairment of left ventricular systolic function. This issue is discussed in detail elsewhere. (See "The management of atrial fibrillation in patients with heart failure".)

Sotalol, propafenone, dronedarone, and flecainide should not be used in patients with heart failure, due to studies showing an increase in mortality with these agents. (See "Clinical uses of sotalol" and "Amiodarone: Clinical uses" and "Major side effects of class I antiarrhythmic drugs".)

Left ventricular hypertrophy — Patients with significant left ventricular hypertrophy (defined as left ventricular wall thickness greater 1.4 cm for the purposes of this discussion) due to hypertension, hypertrophic cardiomyopathy, or aortic stenosis have underlying subendocardial ischemia and electrophysiologic abnormalities. These increase the risk for proarrhythmia with antiarrhythmic agents. (See "Left ventricular hypertrophy and arrhythmia" and 'Proarrhythmia' below and "Left ventricular hypertrophy: Clinical findings and ECG diagnosis".)

Sotalol, flecainide and propafenone are thought to have a significant arrhythmic risk in patients with left ventricular hypertrophy (LVH). Dronedarone has been evaluated in patients with LVH and is thought to be relatively safe [23], although our experts rarely use it. Amiodarone is another therapeutic option. (See "Clinical uses of dronedarone".)

Drug-resistant atrial fibrillation — Some patients are refractory to individual antiarrhythmic agents plus an AV nodal blocker or develop side effects on doses necessary for arrhythmia prevention. Although some have suggested that combination antiarrhythmic drug therapy (eg, a class IC agent with sotalol or amiodarone, often in lower doses, or the combination of dronedarone plus ranolazine) may be an alternative, there are limited data to support such an approach and the patient may be exposed to a greater risk of proarrhythmia and other side effects [28].

As a result, combination antiarrhythmic drug therapy is not recommended. Such patients can be treated with a rate control strategy or referred for nonpharmacologic therapy to prevent recurrent AF including surgery (such as the maze operation) or catheter ablation (such as pulmonary vein isolation). (See "Atrial fibrillation: Catheter ablation" and "The role of pacemakers in the prevention of atrial fibrillation" and "Atrial fibrillation: Surgical ablation".)

INPATIENT VERSUS OUTPATIENT INITIATION — Many patients begun on antiarrhythmic drug therapy should be hospitalized for continuous electrocardiographic monitoring due to a 10 to 15 percent incidence of adverse cardiac events during the initiation of therapy [29]. (See 'Proarrhythmia' below and "Arrhythmia management for the primary care clinician", section on 'Antiarrhythmic drugs'.)

The two complications of greatest concern are bradycardia and proarrhythmia. Other adverse cardiac events can include significant QT prolongation, heart failure, rapid ventricular rate, conduction abnormalities, hypotension, and stroke. The risk is greatest in the first 24 hours and in patients with a prior myocardial infarction.

Outpatient initiation of antiarrhythmic drug therapy with the following agents may be considered:

●Flecainide or propafenone in patients in sinus rhythm who have no underlying structural heart disease, normal baseline QT intervals, and no profound bradycardia or suspected sinus or atrioventricular (AV) node dysfunction [11].

●Amiodarone or dronedarone in selected patients who have no other risk factors for torsades de pointes (eg, hypokalemia, hypomagnesemia), sinus node dysfunction, or AV conduction disease. Dronedarone and amiodarone are the only two drugs that can be initiated in outpatients while in atrial fibrillation.

Patients with an implantable cardioverter-defibrillator (ICD) represent another group in which outpatient initiation of therapy can be tried, since the ICD provides protection against the risks associated with bradyarrhythmias and tachyarrhythmias. However, one should be cognizant of the potential effects of antiarrhythmic drugs on ventricular defibrillation threshold and ventricular tachycardia cycle length, which could influence the efficacy of ICD therapy.

The initiation of antiarrhythmic drugs in patients with paroxysmal AF while they are in sinus rhythm is also associated with some risk. In a review of 409 outpatient initiation trials for a history of recurrent AF or atrial flutter, adverse cardiac events occurred in 17 (4.5 percent); these included three deaths, three permanent pacemakers for bradycardia, and 11 dose reductions for bradycardia [30].

Inpatient initiation with continuous telemetry of higher-risk drugs such as dofetilide and sotalol is typically done over a course of three days, which encompasses five half-lives allowing for achievement of steady-state plasma concentrations. In highly selected patients (eg, normal renal function, no bradycardia, and normal QT interval), sotalol can be loaded as outpatient with event monitor and closely following electrocardiogram for QT interval while in sinus rhythm.

LONG-TERM ISSUES

AF recurrence — Recurrent atrial fibrillation (AF) should not necessarily be labeled as treatment failure. Some patients will elect to continue drug therapy (and, in some cases, occasional cardioversion) because the arrhythmia burden has been substantially reduced as evidenced by episodes that are less frequent, shorter, or associated with milder symptoms. Nonpharmacologic therapies are another option in such patients. (See "Atrial fibrillation: Atrioventricular node ablation" and "Atrial fibrillation: Catheter ablation".)

If a patient has unacceptable recurrent AF on one antiarrhythmic drug, the drug is discontinued and another (and on rare occasion a third) agent is tried.

Dosing — The starting and maintenance doses for amiodarone, dronedarone, propafenone, flecainide, sotalol, disopyramide, and dofetilide are found in the drug interactions program available in UpToDate. Many factors including age, sex, weight, renal or hepatic function, and characteristics on the electrocardiogram influence the starting dose for many of these. (See "Antiarrhythmic drugs to maintain sinus rhythm in patients with atrial fibrillation: Clinical trials".)

Drug-related arrhythmias and mortality — The use of antiarrhythmic drugs is associated with possible life-threatening side effects. The greatest concerns with these agents are proarrhythmia (and consequent tachyarrhythmia) and bradycardia. Patients should be instructed to report symptoms suggestive of the development of drug related arrhythmias, such as syncope, lightheadedness or dizzy spells, or worsening exercise intolerance. (See "Arrhythmia management for the primary care clinician", section on 'Antiarrhythmic drugs' and "Arrhythmia management for the primary care clinician", section on 'Symptoms'.)

A 2012 meta-analysis of 56 studies (20,771 patients) compared one or more antiarrhythmic drugs to control or to each other [31]. Compared to controls, the use of the class IA antiarrhythmics quinidine and disopyramide (odds ratio 2.39, 95% CI 1.03-5.59) or sotalol (2.47, 95% CI 1.2-5.05) was associated with increased all-cause mortality, whereas the use of amiodarone, dronedarone, and dofetilide was not (odds ratios were not calculated for flecainide or propafenone). All antiarrhythmics studied showed increased pro-arrhythmic effects (counting both bradyarrhythmias and tachyarrhythmias attributable to treatment), with the exceptions of amiodarone, dronedarone, and propafenone.

Proarrhythmia — All of the antiarrhythmic drugs used to maintain sinus rhythm have the potential to increase ectopy or induce or aggravate monomorphic ventricular tachycardia (VT), torsades de pointes, or ventricular fibrillation (VF); this is referred to as proarrhythmia (table 1).

In addition to its baseline potential to predispose the patient to proarrhythmia, a drug that is initially safe may become proarrhythmic when the patient develops coronary heart disease or heart failure or is treated with other medications that in combination may be arrhythmogenic. Thus, the patient should be alerted to the potential significance of symptoms such as syncope and dyspnea and warned about the use of noncardiac drugs that can prolong the QT interval (table 2). (See "Acquired long QT syndrome: Definitions, pathophysiology, and causes".)

We agree with the following recommendations made according to drug class in the American College of Cardiology/American Heart Association/European Society of Cardiology guideline [11]:

●With type IC drugs, QRS widening should not be permitted to exceed 150 percent of the baseline QRS duration. Exercise testing may detect QRS widening that occurs only at rapid heart rates (use-dependent conduction slowing). Exercise testing is also a useful way to screen for exercise-induced proarrhythmia and is typically performed one to two weeks after drug initiation.

●For type IA or type III drugs (table 3), with the possible exception of amiodarone, the corrected QT interval in sinus rhythm should remain below 520 milliseconds. More specific and conservative recommendations are available for dofetilide in the package insert.

●During follow-up, serum creatinine, potassium, and magnesium concentrations should be monitored periodically because proarrhythmia is increased by renal insufficiency, which can lead to drug accumulation, hyperkalemia, and hypermagnesemia. The presence of renal insufficiency warrants dose reduction or cessation of sotalol and dofetilide. In comparison, amiodarone is metabolized in the liver and dose adjustment is probably necessary in patients with hepatic dysfunction. (See "Amiodarone: Adverse effects, potential toxicities, and approach to monitoring", section on 'Adverse hepatic effects'.)

Bradyarrhythmia — Amiodarone and dronedarone can cause both sinus bradycardia and AV nodal block, with an overall incidence of bradycardic events of about 5 percent. Sotalol, like other beta blockers, can also cause bradycardia. In some cases, permanent pacemaker placement is necessary to permit continued use of these agents. (See "Amiodarone: Adverse effects, potential toxicities, and approach to monitoring" and "Clinical uses of sotalol".)

Ambulatory monitoring — Our authors and reviewers have differing thresholds for the use of ambulatory monitoring to screen for proarrhythmia and bradycardia, ranging from screening in the highest risk cases only to screening in everyone. Some experts suggest screening all patients with an ambulatory event monitor for at least two weeks after initiation of therapy, looking for QT interval prolongation or bradyarrhythmias. The basis for this recommendation is that many events occur after three days [30]. For those experts who are more selective based on patient risk, high-risk is defined as baseline bradycardia or borderline QT prolongation, heart failure, or systolic left ventricular dysfunction. Others perform routine monitoring when sotalol, flecainide, or propafenone are chosen. Dofetilide must be initiated in a setting with continuous monitoring. (See "Clinical use of dofetilide".)

For those patients who are not referred for ambulatory monitoring, we suggest that a 12-lead electrocardiogram be obtained after the initiation of antiarrhythmic drug therapy.

Short- versus long-term therapy — Based on concerns about drug related arrhythmias and the observation that the atrial action potential normalizes after two to four weeks of sinus rhythm (after AF), the concept that short-term therapy might be as effective and safer than long-term therapy has been proposed. This concept was tested in the Flec-SL non-inferiority trial, which randomly assigned 554 patients with persistent AF and who were intended to undergo cardioversion to either four weeks or six months of flecainide (200 to 300 mg per day) [32]. All patients had successful restoration of sinus rhythm and were then followed with daily telemetric electrocardiography (and Holter monitoring whenever AF was noted on two ECGs) for six months.

The primary outcome of time to persistent AF or death occurred in 46 and 39 percent of patients, respectively, which did not meet the criteria of non-inferiority. In addition, a post-hoc analysis of patients who had not reached the primary endpoint in the first month found long-term therapy to be superior (Kaplan-Meier estimate of difference 14.3 percent; hazard ratio 0.3; p = 0.0001). We do not consider short-term therapy appropriate for most patients with persistent AF.

Concerns about dronedarone — Patients with severe heart failure (HF) (generally those with NYHA class III or IV HF, or those who have been hospitalized with HF in the past four weeks) or those with an ejection fraction of <35 percent should not receive dronedarone. (See "The management of atrial fibrillation in patients with heart failure", section on 'Antiarrhythmic drugs'.)

The Permanent Atrial fibriLLAtion outcome Study (PALLAS) was designed to test the hypothesis that dronedarone would improve major outcomes in 10,000 patients with permanent AF, over 70 percent of whom had New York Heart Association heart failure class I to III or left ventricular systolic dysfunction at baseline. The rationale was that patients with permanent AF, which affects up to 50 percent of patients with AF, have an increased risk of adverse cardiovascular outcomes including death and myocardial infarction as well as systemic embolization. The ATHENA trial showed a significant reduction in cardiovascular events with dronedarone in patients with paroxysmal or persistent AF. (See "Antiarrhythmic drugs to maintain sinus rhythm in patients with atrial fibrillation: Clinical trials", section on 'Dronedarone'.)

Patients in PALLAS were treated with standard therapies for AF and then randomly assigned to dronedarone or placebo. The study was stopped early (3236 patients enrolled), after a significantly increased risk (Hazard Ratio 2.29, 95% CI 1.34-3.94) of cardiovascular events (cardiovascular death, myocardial infarction, stroke and systemic embolism) was observed in the dronedarone arm [33]. The individual secondary end points of stroke, death from cardiovascular causes, and hospitalization for heart failure were also significantly increased in the dronedarone group. (See 'Summary and recommendations' below.)

The European Medicines Agency and the United States Food and Drug Agency (USFDA) have advised against the use of dronedarone in patients with permanent AF [34,35]. In addition, the USFDA now recommends that people taking the drug should have an electrocardiogram every three months to make sure that AF has not become permanent. For patients taking dronedarone, routine monitoring of lung and liver function is not mandated by the USFDA; however, periodic monitoring may be reasonable [36]. (See "Clinical uses of dronedarone", section on 'Maintenance of sinus rhythm'.)

Follow-up — We consider the following approach to follow-up reasonable:

●We perform an ECG one week after initiation of any antiarrhythmic drug.

●We typically see patients within three months of initiating a new antiarrhythmic drug to assess efficacy and side effects. This is in addition to commonly performing ambulatory monitoring after drug initiation.

●Patients are typically seen every 6 to 12 months unless there are particular concerns regarding QT interval prolongation, bradycardia, or other issues identified on the ECG.

Specific follow-up recommendations for individual drugs are presented separately. (See "Amiodarone: Adverse effects, potential toxicities, and approach to monitoring" and "Clinical uses of sotalol", section on 'Major side effects' and "Clinical use of dofetilide", section on 'Safety' and "Major side effects of class I antiarrhythmic drugs", section on 'Flecainide' and "Major side effects of class I antiarrhythmic drugs", section on 'Propafenone'.)

RECOMMENDATIONS OF OTHERS — Our recommendations for the use of antiarrhythmic drugs to maintain sinus rhythm in patients with AF are generally in agreement with recommendations from the American Heart Association/American College of Cardiology/Heart Rhythm Society (2014) and its 2019 focused update, as well as the European Society of Cardiology (2016) [13,37-39]. (See 'Summary and recommendations' below.)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Atrial fibrillation" and "Society guideline links: Arrhythmias in adults".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, “The Basics” and “Beyond the Basics.” The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on “patient info” and the keyword(s) of interest.)

●Basics topic (see "Patient education: Medicines for atrial fibrillation (The Basics)")

SUMMARY AND RECOMMENDATIONS

●Initial management decisions – For those patients with atrial fibrillation (AF) in whom a rhythm-control strategy is chosen, the principal goal is to reduce symptoms by decreasing the frequency and duration of episodes [1,2]. (See 'Initial management decisions' above.)

●Selecting an antiarrhythmic drug – (See 'Selecting an antiarrhythmic drug' above.)

•Beta blockers are modestly effective in maintaining sinus rhythm and can be tried first in selected patients, such as those without structural heart disease who are concerned about proarrhythmia.

•Compared to placebo, amiodarone, sotalol, dofetilide, dronedarone, flecainide, and propafenone are effective for the maintenance of sinus rhythm, but maintenance rates at one year are significantly less than 75 percent. Amiodarone is consistently more effective than the other antiarrhythmic drugs.

●Drug-related arrhythmias and mortality – In addition to less-than-optimal efficacy, serious drug-related adverse side effects limit the use of these drugs. Antiarrhythmic drug therapy should be prescribed only by practitioners familiar with their use. Patients should be fully informed of both the benefits and risks associated with the use of these drugs. (See 'Drug-related arrhythmias and mortality' above.)

●Inpatient versus outpatient initiation – Based upon the potential for drug toxicity in the form of induced bradycardia or tachycardia, many patients will need to be hospitalized for continuous electrocardiographic monitoring. Dofetilide must be initiated in a setting with continuous monitoring. (See 'Inpatient versus outpatient initiation' above.)

●Selecting an antiarrhythmic drug

•Atrial fibrillation without structural heart disease – For patients with no structural heart disease and no apparent risk for drug-induced bradycardia or tachycardia, we suggest flecainide or propafenone as the preferred antiarrhythmic drug (Grade 2B). Amiodarone, dofetilide, dronedarone, or sotalol may be used, with sotalol chosen more often by our authors and reviewers. Practitioners should choose only those agents with which they have significant familiarity. (See 'Atrial fibrillation without structural heart disease' above.)

•Coronary heart disease – For patients with coronary artery disease who do not have advanced heart failure, we suggest dronedarone or sotalol in preference to amiodarone (Grade 2B). Amiodarone is a reasonable choice in patients who prefer its greater efficacy despite its worse extracardiac side-effect profile. (See 'Coronary heart disease' above.)

•Heart failure – For patients with heart failure, we suggest amiodarone in preference to dofetilide (Grade 2B). Flecainide, propafenone, dronedarone, and sotalol are contraindicated in these patients. (See 'Heart failure' above.)

•Left ventricular hypertrophy – For patients with left ventricular hypertrophy, either amiodarone or dronedarone is generally preferred to other antiarrhythmic agents. Our authors and reviewers have differing approaches, with some choosing amiodarone more often and others choosing dronedarone more often. (See 'Left ventricular hypertrophy' above.)

●Ambulatory monitoring – After the initiation of antiarrhythmic drug therapy, screening for drug-associated arrhythmia with ambulatory monitoring should be considered, particularly for patients at high risk of drug-induced arrhythmia. This includes those with baseline bradycardia or borderline QT prolongation, heart failure, or systolic left ventricular dysfunction. (See 'Ambulatory monitoring' above.)