2. What Diagnostic Tests Should Be Done in Patients with a New Diagnosis of Atrial Fibrillation?

1. What are the differences between lone, paroxysmal, persistent and permanent atrial fibrillation (AF)?

A joint task force met in 2001 to develop guidelines for the management of AF and proposed a clinically relevant classification system [1]. Episodes of AF that occur on more than one occasion are called recurrent. Paroxysmal AF is used to identify AF that generally last less than or equal to 7 days and may be recurrent. Persistent AF refers to episodes that last longer than 7 days. If AF continues for over a year, with or without failed cardioversion, it is called Persistent. The presence or absence of symptoms do not affect the classification of AF. This system of classification excludes AF that lasts less than 30 seconds or that is secondary to a precipitating condition, such as surgery or myocardial infarction. The term “Chronic AF” is not used in this classification, but is generally what is thought of as Permanent AF. Lone AF refers to AF that occurs in the absence of echocardiographic or clinical evidence of cardiopulmonary disease and generally applies to patients less than 60 years old and implies a group of patients at low risk of thromboembolic events. Lone AF can be paroxysmal, persistent or permanent.

2. What diagnostic tests should be done in patients with a new diagnosis of atrial fibrillation?

There are multiple risk factors and disease processes that have been associated with atrial fibrillation. A study of the Framingham cohort, for example, revealed that diabetes, hypertension, myocardial infarction and age are strong risk factors of AF[2]. Tests should be directed at determining underlying etiology or an associated, treatable condition. A thorough history may reveal symptoms of cardiopulmonary disease or expose risk factors such as binge drinking. Physical exam can assess blood pressure and heart murmurs and signs of heart failure. The electrocardiogram can help establish the diagnosis of AF and assess for ventricular hypertrophy and acute myocardial infarction. Echocardiography should be performed to rule out ventricular dysfunction, valvular disease, rheumatic heart disease and hypertrophic cardiomyopathy. Patients should also be screened for diabetes, if not already established, with fasting blood sugars. Although TSH is routinely checked, a Canadian study revealed that fewer than 1% of patients with newly diagnosed AF were found by two years to have clinical hyperthyroidism and that checking TSH routinely is of low yield[3]. Further targeted testing should be performed if clinical suspicion exists for acute myocardial infarction, obstructive sleep apnea, COPD or pulmonary embolism.

3. What are the indications for hospitalization in a patient with atrial fibrillation?

Under appropriate conditions, it is safe to manage even new onset AF as an outpatient. Admission to the hospital is often considered in order to rule out acute myocardial infarction. In one study of 109 patients who were admitted with new onset AF to rule out myocardial infarction, only 5.5% ruled in[4]. In this study, all patients who lacked ST depressions > 2mm or ST elevations ruled out for a myocardial infarct. It would be reasonable to treat those patients at low risk for an acute coronary syndrome as outpatients. If the patient, however, demonstrates hemodynamic instability, signs or symptoms of heart failure, persistent rapid ventricular response or symptoms that can not be controlled in the emergency room, he or she should be admitted for further management and potential inpatient cardioversion. Although not yet validated for these purposes, a model for calculating risk of stroke or death, such as the one based on the Framingham cohort [5], could potentially be used to help decide which patients would best be served as inpatients.

4. Is rhythm control better than rate control?

It was formerly believed that maintenance of AV synchrony was beneficial and was often attained at considerable cost. The largest randomized controlled trial that challenged this assumption was the AFFIRM trial[6]. 4060 patients with asymptomatic AF were randomized to rate control with nodal agents versus rhythm control with pharmacologic or electrical cardioversion. There were no statistically significant differences in stroke or death between the two groups. However, patients in the rhythm control group had more hospitalizations and adverse drug reactions and many deaths in the rhythm control group may have been attributed to the antiarrhythmic drugs themselves. Similar findings were noted in the RACE trial, a randomized controlled trial with 522 participants[7]. It is reasonable, however, to aim for rhythm control in patients who either can not achieve adequate rate control or are symptomatic from AF, especially in the younger population that was not well represented in these trials. As newer, potentially safer approaches to rhythm control become available, such as catheter ablation techniques, the benefits of one approach over the other will need to be readdressed.

5. Are beta blockers better than calcium channel blockers for rate control?

Although rate control has been established as an appropriate approach to the management of AF in older, asymptomatic patients, the choice of which nodal agent to use is in question. The goals for patients randomized to rate control in the AFFIRM trial was a heart rate below 80 at rest and below 110 during a six minute walk test[6]. In a sub-study of the AFFIRM trial, the medical approach to rate control was evaluated[8]. Patients whose first drug was a beta-blocker, with or without digoxin, were able to achieve rate control in 70% of cases whereas those whose first drug was a calcium channel blocker, with or without digoxin, were successful in 54% of cases. Calcium channel blockers may, however, be more beneficial in selected groups, such as those prone to bronchospasm or who develop other adverse reactions to beta blockers or younger patients who may have diminished exercise capacity with beta blockers. Six minute walk tests or Holter monitoring during activities of daily living may be necessary to assess adequate rate control.

6. Which patients benefit from treatment with digoxin?

Among its many mechanisms of action, digoxin increases vagal tone and can have a direct sympathoinhibitory effect. Because it has been available for a long time, it is relatively inexpensive and requires only once a day administration, it remains a popular therapeutic agent. Alone, it is effective at rate control as a first-line agent in only 58% of patients, which is comparable to calcium channel blockers but worse than beta blockers alone[8]. In combination, it can help reduce the dose of calcium channel blocker or beta blocker required to achieve rate control and together with carvedilol, can lower the heart rate better than either agent alone in patients with systolic heart failure[9]. In the absence of heart failure, digoxin is not effective at converting patients in AF to sinus rhythm[10], but may reduce the frequency of symptomatic episodes in paroxysmal AF. In summary, digoxin may benefit patients already on calcium channel blockers or beta blockers who have not achieved adequate rate control, especially those with heart failure, and patients with frequent episodes of symptomatic paroxysmal AF.

7. Is pharmacologic cardioversion better than electrical cardioversion in patients who need rhythm control?

There are no trials that have randomized patients undergoing acute rhythm control to pharmacologic versus electrical cardioversion. The decision remains primarily the preference of the patient and health care provider. Electrical cardioversion with a biphasic waveform is effective in 80-90% of cases[11], but requires conscious sedation. A small group of patients may be at risk of developing VT, particularly in the setting of hypokalemia or digitalis intoxication. The efficacy of pharmacologic therapy varies depending on the agent chosen, but is generally not as high as electrical cardioversion. Pharmacologic cardioversion tends to be tolerated, but the risk of torsade de pointes can range from 1% – 8% , depending on the agent and dose used as well as any underlying susceptibilities, such as bradycardia and hypokalemia[12]. The primary risk of cardioversion with either approach remains thromboembolism or stroke and can be as high as 7% in those not adequately anticoagulated. The risk does not differ significantly between these two approaches.

8. How should electrical cardioversion be performed?

Unless the patient is hemodynamically unstable or there is another indication for emergent cardioversion, the patient should undergo conscious sedation. A recent multicenter controlled trial randomized 203 patients to either biphasic or monophasic shock waveforms and found that biphasic shocks were more efficacious, requiring fewer shocks and resulted in less dermal injury[11]. The amount of energy necessary for cardioversion varies depending on the waveform used, duration of AF and size of patient. For monophasic waveforms, the minimal therapy should be 100J for < 48h, 150J for < 7 days and 200J for >1 year. For biphasic waveforms, the minimal initial therapy should be 50J for < 48h, 100J for < 7 days and 150J for > 1year. This can be titrated up to the maximum energy of 360J for monophasic and 200J for biphasic waveforms. Anterior-posterior paddle positioning, over the sternum and left scapula, is more effective than the anterior-lateral approach[13]. Cardioversion in patients with pacemakers and defibrillators is safe, but devices should be interrogated immediately before and after external discharge. Long-term success improves if the cardioversion is performed after the inciting cause, if any, of AF has resolved and after initiation of therapy for maintenance of sinus rhythm.

9. Which antiarrhythmic agents should be used for acute conversion to and maintenance of sinus rhythm?

The choice of antiarrhythmic agent depends primarily on the presence or absence of any underlying cardiovascular disease, the side effect profile and the experience of the clinician with certain agents. There are few studies that have randomized patients to the different available drugs. In a meta-analysis of clinical trials measuring efficacy of pharmacologic cardioversion, it was found that ibutilide, dofetilide and flecainide were 24-29 times more likely to lead to cardioversion than standard therapy with calcium channel or beta blockers[14]. Although amiodarone and propafenone were moderately effective, sotalol was generally ineffective at inducing sinus rhythm. In this meta-analysis, all agents were approximately equally efficacious in the maintenance of normal sinus rhythm. Given the propensity of certain agents to induce ventricular arrhythmias in the presence of cardiovascular disease, guidelines from the joint task force recommend that for maintenance of sinus rhythm, flecainide, propafenone and sotalol be used as first-line agents in patients with no or minimal cardiovascular disease[1]. According to these guidelines, patients with heart failure can be started on amiodarone or dofetilide, patients with coronary disease can begin with sotalol and patients with hypertension can be started on amiodarone.

10. How is amiodarone used in the management of AF?

Amiodarone is a class III agent whose use in AF, although not FDA approved for this indication, has become exceedingly common primarily because of its low risk of ventricular arrhythmias compared with other antiarrhythmics and lack of negative ionotropic effects. The major deterrent is that its long-term use is associated with multiple toxicities, namely pulmonary, hepatic and thyroid toxicities and its use in younger populations should be limited. Although there is no standardized dosing regimen, consensus guidelines recommend an initial loading of approximately 10 grams of amiodarone and subsequent titration down to 200mg daily[1]. One approach is to begin an oral load with 400mg twice daily for one week, followed by 200mg twice daily for two weeks, then titrated down to 200mg po daily for maintenance. Another approach proven to be safe[15] is to load with 600mg daily for four weeks, followed by maintenance of 200mg daily. When used intravenously, amiodarone can be bolused at 150mg, followed by a drip of 1mg per hour for 6 hours and then maintained at 0.5mg per hour until the load can be completed orally. Toxicities from long-term use should be monitored with baseline and regular pulmonary function, liver function thyroid stimulating hormone tests.

11. Can antiarrhythmic agents be started as an outpatient?

Because of the risk of torsade de pointes associated with these medications, almost all studies of antiarrhythmics were performed during inpatient observation with careful monitoring for QT prolongation. In one study of 597 drug trials, 13.7% of patients suffered adverse cardiac events, including bradyarrhythmias in 8%, significant QT prolongation in 1.5% requiring drug discontinuation and ventricular arrhythmias in 1.3% of trials[16]. Some antiarrhythmics, such as dofetilide, are additionally restricted to prescribers who have completed appropriate educational forums[17]. In one study of 129 patients with AF or flutter, subjects were given 600mg daily of amiodarone for 4 weeks and none of the patients suffered ventricular arrhythmias[15]. Given this and similar observations, it is considered that amiodarone can be started safely on an outpatient basis.

12. What is the pill-in-pocket approach?

There are a group of patients with paroxysmal AF who have relatively mild or well tolerated symptoms during their episodes. If the episodes are infrequent enough, there may be a reluctance to initiate daily use antiarrhythmic agents. These patients may be managed with the “pill-in-the-pocket” approach, which consists of asking this group of patients to take a single dose of an antiarrhythmic when they become symptomatic instead of coming to the emergency room. In a study evaluating the safety of this approach[18], 268 patients presenting with mild or no heart disease were converted to sinus rhythm with either flecainide or propafenone in the hospital. The patients who did not have an adverse event were then followed as outpatients and advised to self-administer a single dose of flecainide or propafenone after the onset of heart palpitations. These patients had a significant fall in the number of emergency room visits and hospital admissions compared to the year prior. Only one patient experienced atrial flutter with rapid ventricular rate and 11 patients reported non-cardiac side effects. This approach is a safe option for the management of a select group of patients who would prefer not to endure long-term standing use of antiarrhythmic therapy.

13. Do angiotensin converting enzyme inhibitors (ACEI) and angiotensin receptor blockers (ARB) prevent recurrence of atrial fibrillation?

On retrospective examination of several trials, it was observed that the use of ACEI and ARBs were associated with the prevention of AF and the maintenance of sinus rhythm in patients with AF. Several potential mechanisms have been proposed and this effect is likely related to atrial remodeling as well as the modification of diseases that are associated with AF such as hypertension and congestive heart failure. In one study of 186 patients with persistent AF, participants were randomized to amiodarone versus amiodarone plus irbesartan followed by electrical cardioversion[19]. After 2 months, the group treated with irbesartan had fewer recurrences of AF. In a recent meta-analysis that reviewed 11 randomized trials of ACEI and ARBs that also reported new onset of AF, it was found that ACEI and ARBs reduced the relative risk of developing AF by 28%[20]. This effect was greatest in patients with heart failure. At this point there is not sufficient prospective evidence to support the empiric use of ACEI or ARBs in the treatment of AF, however, this may influence the choice of therapy in someone who also has hypertension or congestive heart failure. Larger, more definitive trials to address this question are underway.

14. Who should receive anticoagulation?

Whether or not a patient is anticoagulated depends on their risk of suffering a thromboembolic event, which is related to age and coexisting cardiovascular disease. In a large Danish cohort of patients with AF, the incidence of stroke rose from 1.3% per year in those aged 50-59 to 5.1% per year in those aged 80-89[21]. The stroke rate in those with lone AF, defined as younger than 60 without evidence of cardiopulmonary disease, remains at 1.3% per year [22]. Stroke rates can be reduced significantly with the use of oral anticoagulation and, to a lesser degree, with aspirin. A meta-analysis of sixteen trials, including Stroke Prevention in Atrial Fibrillation, revealed that adjusted-dose warfarin reduced stroke by 62% while ASA reduced stroke by only 22%[23]. Similar multi-trial analysis suggests that patients who are below 65 and without a history of transient ischemic attacks, diabetes mellitus and hypertension have a stroke risk of less than 1% per year without warfarin[24]. Based on these studies, it is strongly recommended that patients with AF be treated with warfarin at a goal INR 2-3. Aspirin alone may be acceptable in patients younger than 60 years who have no hypertension, diabetes or evidence of cardiovascular disease.

15. Is immediate cardioversion with TEE better than delayed cardioversion after anticoagulation?

Immediately after either electrical or pharmacological cardioversion, there is transient left atrial dysfunction known as stunning. Depending on how long the patient has been in AF, stunning can last for several weeks and predisposes patients to thromboembolism. Patients with evidence of a thrombus in the left atrial appendage seen on transesophageal echocardiogram are at a particularly high risk of thromboembolic events. The risk of an embolic event without anticoagulation has been reported to vary from 1-5% in patients with persistent AF, but is closer to 0.8% in those who have been in AF for less than 48 hours[25]. In a multicenter, randomized prospective trial, 1222 patients with AF of greater than 48 hours duration were either anticoagulated for 3 to 4 weeks before cardioversion or had earlier cardioversion guided by transesophageal echocardiography[26]. There were no significant differences in rates of embolic events, however, the group guided by transesophageal echo had a greater rate of successful restoration of sinus rhythm and fewer hemorrhagic events. The decision of one approach over the other may be guided primarily by patient preference, but this study suggested that both approaches are equally effective in reducing stroke rates.

16. Do patients withparoxysmal atrial fibrillation need to be anticoagulated for life?

Anticoagulation can be burdensome because of the close monitoring that is necessary and bruising associated with the therapy. Many patients understandably request that anticoagulation be stopped, especially patients with paroxysmal AF who have not felt any symptoms over a long period of time. The stroke risk, however, is equivalent between patients with persistent and patients with chronic AF[27]. When patients were recorded with continuous electrocardiograms over a 28 day period, however, most patients with paroxysmal AF experienced asymptomatic episodes of AF[28]. Patients will experience asymptomatic episodes even when they claim to know exactly when they are in the arrhythmia. There is no easy way to identify patients who will stay in sinus rhythm and no longer be at risk for thromboembolic events. Although it is not known definitively if asymptomatic episodes of AF are associated with increased risk of stroke, at this moment there is no evidence to suggest that stopping anticoagulation after maintenance of sinus rhythm is safe.