Atrial Fibrillation: Rate Control Strategies

Transcription

1 Atrial Fibrillation: Rate Control Strategies Olshansky, Brian Sbaity, Salam Summary The safest long-term management strategy for elderly patients with persistent or permanent and some patients with paroxysmal atrial fibrillation (AF) is careful rate control with medications in conjunction with appropriate oral warfarin anticoagulation. AV junctional ablation, with pacemaker implantation, may be necessary in patients who cannot have their rate controlled adequately, or have marked fluctuations in rate with standard medical therapies. Maintenance of sinus rhythm with repeat cardioversions, antiarrhythmic medications, device therapy and/or atrial ablation approaches are appropriate to consider for younger patients and those elderly patients who suffer from substantial symptoms shown related directly to AF. INTRODUCTION In 2000, it was estimated that atrial fibrillation (AF) affected over 2.3 million, mostly elderly, Americans (1,2). This number is estimated to more than double over the coming fifty years as the population continues to age (2). The problem also exists worldwide. AF is the most common arrhythmia requiring hospitalization (2) and it contributes to endless hours of outpatient management. AF is associated with stroke, heart failure, cardiomyopathy, impaired quality of life, various non-specific symptoms and death (1),(3),(4). The prevalence of AF increases with age reaching 10% in individuals above 80 years (5,6). AF may be an isolated, even an explainable episode, recur in paroxysms, be persistent requiring cardioversion or become permanent and not be correctable (7). This might be due to the different underlying mechanisms which include rapid ectopic activity, single-circuit re-entry, and multiple-circuit re-entry (8). Strategies to manage AF include anticoagulation to reduce the risk of stroke, antiarrhythmic drugs and ablation techniques to reduce the risk of AF recurrence and techniques and drugs to help control the ventricular response rate. Approaches to, and need for, rate control in patients with AF are discussed.

2 WHY TREAT ATRIAL FIBRILLATION? AF is associated with an increased risk of death especially for patients over age 65 (1). Data from the Framingham cohort show that AF confers an independent relative risk of death of 1.5 and 1.9 for men and women respectively, after adjusting for co-morbidities (1). However, this alone is not an adequate reason to treat AF since little, if any, compelling evidence indicates that any treatment of AF reduces the risk of death. AF is associated with an increased risk of stroke(1) (but stroke risk escalates dependant on additional medical conditions), impairment in ventricular function (9), heart failure(10), and hemodynamic impairment (11) especially when the rates are rapid and/or irregular (12) and if rapid rates persist. This can precipitate tachycardia-induced cardiomyopathy (3), and, most notably, symptoms that impair quality of life (13). For older patients, the main reason to treat AF is to reduce or eliminate symptoms, facilitate functionality and improve quality-of-life. Symptoms define the problem for younger patients as well. Younger individuals, who represent about 35% of all patients with AF, usually do not have concomitant medical conditions ( lone AF) and can ascribe all their symptoms to AF directly. Unfortunately, for most patients, little hard data document that AF treatment improves the most important outcomes such as symptom reduction and improvement in quality-of-life (11). SYMPTOMS AND ATRIAL FIBRILLATION The diverse spectrum of symptoms described by patients with AF varies by AF type, frequency and rate, patient age and associated medical conditions. Symptoms range from a chance finding with no (apparent) symptoms to palpitations, fatigue, dyspnea, angina, congestive heart failure, stroke or a new neurological abnormality. In the elderly, excess fatigue is a common symptom. In younger patients, palpitations, dyspnea and other symptoms prevail. The Etude en Activité Libérale de la Fibrillation Auriculaire (ALFA) trial addressed issues regarding symptoms in a large population of patients with AF (1) but no study addresses adequately the importance of symptom reduction or outcomes based on specific symptoms. Implications of individual symptoms vary widely. The degree of impairment due to symptoms is often not clear.

3 No guideline categorizes a need for treatment based on specific symptoms severity. No AF symptom classification exists. Other than anticoagulant prophylaxis to prevent stroke, symptoms are the primary reason to treat AF so a classification scheme would make sense It is clear that individual patients with AF have a markedly impaired quality-of-life determined by SF-36 scores and other standardized scales placing AF in the category of impairment severity in functioning similar to heart failure and coronary artery disease. AF can, and does, lead to disability. Considering treatment approaches, the primary goal to improve symptoms may be at odds with expense, time, and commitment to treatment and substantial risk to the patient. Symptoms may not be due solely to AF. The understanding a patient has or the appreciation of expected symptoms and the associated disease state can contribute to outcomes. The physician or other health care professional may over-emphasize the importance of the problem and accentuate symptoms. Symptoms may be related in part to a patient s understanding, interpretation or appreciation of the AF. Awareness of symptoms and/or the presence of AF may be over-emphasized by the physician, friends, the internet or other sources. If a physician, not the patient recognizes the problem, there is reason for concern; the patient may be over treated. If a patient has mild symptoms or has not apparent symptoms, but the physician recognizes AF, the patient and the family may have reason for concern. Over treatment may ensue. A patient may notice an irregular rate or become aware and concerned when told that AF is present. This concern may initiate a search for effective and even highly aggressive but unnecessary treatment precipitating an iatrogenic consequence. Symptoms can be underestimated; a patient may even have no complaints and be apparently fully functional but feel better than he did in years after a cardioversion to sinus rhythm. Alternatively, symptoms should not be overestimated since fatigue, a common symptom of AF, can be due to a multiplicity of issues and not be solved by correction of AF. Aggressive therapies directed at AF treatment can have deleterious effects. Symptoms may be overblown by the patient causing an unbridled enthusiasm for aggressive unnecessary treatment of no use. There may even be a relationship between AF, quality of life and neuroticism (14).

4 Alternatively, AF can be more cosmetic than real. Much AF, even in symptomatic patients, is not detected. Much of what is perceived to be AF and symptoms due to AF actually are not. AF can have dire consequences and cause syncope or even cardiac arrest. Conditions such as the Wolff- Parkinson-White Syndrome or atrial flutter, in association with AF, can lead to potentially lifethreatening rapid rates and even ventricular fibrillation. WHY AF CAUSES SYMPTOMS Symptoms can be due to an inappropriate, and abrupt, change in heart rate response especially at initiation and termination of episodes, inappropriate tachycardia, inadequate or inappropriate chronotropic response to exercise, or ensuing irregular ventricular response rate. Loss of atrial kick can be an issue particularly in those with substantial diastolic dysfunction associated with conditions such as aortic stenosis or hypertrophic cardiomyopathy. AF often coexists with other medical conditions particularity in the elderly that may also contribute to symptoms. Expectation and understanding of the presence of a perceived potentially debilitating medical condition such as AF may contribute. Perhaps this is the reason that some patients with AF complain of symptoms even in sinus rhythm. Therapies directed at AF, or even cure, will not necessarily lead to a more functional patient. ASSOCIATED CONDITIONS A causal relationship (or at least an association) between AF and other processes or medical conditions do occur but in some cases, they can be difficult to ascertain. Some specific conditions trigger or initiate AF, such as ischemia (15-17), alcohol binges(18) (19), thyrotoxicosis (20,21), cardiac surgery (22), Wolff-Parkinson-White syndrome (23), GI bleed, pulmonary embolism (24), hypothermia (25,26), electrolyte disturbances (27,28) or even anger (29). Atrial stretch can initiate AF (30). Processes that do this include diastolic dysfunction, in particular, related to hypertension and valvular heart diseases. Perhaps, this is the reason that angiotensin receptor blockers and angiotensin converting enzyme (ACE) inhibitors and even beta-blockers appear reduce AF

5 episodes in patients with hypertension (31,32) although there may be a specific genetic trait unrelated to atrial stretch (33). Treatment of conditions that affect the presence of AF often achieves priority over a direct attempt at controlling the AF. Treatment of these conditions may ultimately reduce problems related to AF. ATRIAL FIBRILLATION TREATMENT OPTIONS Treatment of AF is multidimensional. It is usually directed towards the following aims: 1.Securing patient stability 2.treating possible underlying causes of arrhythmia 3.alleviating symptoms 4.controlling ventricular rate and 5.minimizing the risk for stroke. Management strategies include heart rate control during AF ( rate control ) and attempts to maintain sinus rhythm ( rhythm control ). Anticoagulation must be considered for both treatment strategies for patients at high risk for thromboemboli and stroke. Barely any data support the concept that rhythm control with antiarrhythmic drug therapy reduces the risk for thromboemboli. Patients with isolated, identifiable AF episodes may not require long-term therapy. Patients whose AF is not due to an acute precipitant or who have frequent recurrences will require long-term, if not life-long, treatment. Rhythm control in a patient with permanent AF makes no sense. Rate control in a young patient with occasional AF makes no sense. Rate control is part of virtually any management plan for patients with AF even if the ultimate goal is to control the rhythm. Drug therapy to control the rhythm is less than optimal. AF recurrence, even with optimal antiarrhythmic medications approaches 80% so that even if the episode may be reduced with drug therapy, rate control strategies will be needed for episodic, sometimes asymptomatic, recurrences. Antiarrhythmic drugs can convert AF into a slow form of atrial flutter with one to one atrial to ventricular conduction that can cause fast ventricular rates. WHAT IS RATE CONTROL? Rate control is achieving an appropriate and necessary heart rate for a given level of exercise or at rest. No strict, widely accepted, definition exists. It is likely that a faster rate is required in AF to maintain the same level of cardiac output that would occur in sinus rhythm. The ventricular rate in

6 AF may need to be different from the rate in sinus rhythm to maintain effective cardiac output and hemodynamics. Rate control approaches differs between sedentary and active patients. Some evidence indicates that rate self adjusts in AF over time even without additional treatment in some patients. Rate control may not improve symptoms and may create new problems such as bradycardia and symptoms from the medications. Requirements for rate control vary depending on hemodynamics and ventricular function. Rate control in permanent AF may be easier to achieve because therapy only needs to account for AV nodal conduction. For other forms of AF, sinus node function may be altered adversely by the same therapy. Part of the challenge of determining rate control strategies depends on what is considered the proper rate for the patient. Certain rates are clearly unacceptable for any patients but those with poorer ventricular function may require more rapid baseline rates at rest. The rate may vary from patient to patient with exercise. Randomized trials comparing rate control to rhythm control provide no insight into an appropriate rate in AF for an individual patient. A stringent definition of rate control was used in the AF Follow-Up Investigation of Rhythm Management Trial(AFFIRM) (11) (Table 1). This definition was rigorous. No other groups or studies have defined the control of the ventricular response in AF so precisely. TABLE 1 AFFIRM Definition of Adequate Ventricular Rate Control Average heart rate at rest 80 beats/minute, and either: Heart rate maximum during a 6-minute walk 110 beats/minute, or Average heart rate during a 24-hour ambulatory Holter electrocardiographic monitor 100 beats/minute (at least 18-hours of interpretable monitoring) and no heart rate > 110% maximum predicted age-adjusted exercise heart rate

7 In practice, rate control is rarely based on such tightly controlled guidelines. This may or may not be appropriate. Likely, ventricular rates are rarely monitored carefully in practice except sporadically at clinic visits (and, even then without testing the response to exercise) The best endpoints for effective rate control in AF are not defined. The efficacy criteria developed in the AFFIRM Trial is based on a consensus of what would be considered a reasonable approach to rate control. The criteria may be too lenient or too strict or the criteria may be addressing the wrong issues (rate instead of symptoms) or defining the wrong rates. The rates may be patient, age or sex dependant. It is possible that AFFIRM turned out as it did due to the pre-selected rate control guidelines. In practice, the efficacy of rate control is rarely determined with a 6-minute walk test, a Holter monitor and a resting heart rate at every visit. Adequate rate control at rest does not ensure adequate rate control with exercise. In the AFFIRM trial results of a rate controlling drug was generally as effective at rest as it was during exertion. It remains uncertain what rate control endpoint is optimal to balance hemodynamic and symptomatic benefits with drug-induced adverse effects. Attempts to optimize drug therapy for rate control may be ineffective or may increase adverse drug effects. It may be better to settle for apparent inadequate rate response in lieu of iatrogenic complications. Rate controlling drugs are relatively safe when used in experienced hands. Rate control has now become the accepted and recommended initial strategy for AF management by several organizations (34) but is still considered one of several options by others (35). It is likely that patients who have adequate and effective rate control in AF will have fewer symptoms and be more functional. WHY CONTROL THE VENTRICULAR RATE? Acutely, rapid rates can cause rapid hemodynamic decompensation and may precipitate pulmonary edema in especially in patients with diastolic dysfunction, ventricular hypertrophy, chronic congestive heart failure, ischemia, and/or valvular lesions. Rapid rates can initiate hypotension, angina or even syncope, the latter being most common with abrupt and marked changes in rates.

8 Rarely, AF can cause hemodynamic collapse that leads to a cardiac arrest especially when there are exceedingly rapid rates. This may occur even in young individuals with normal cardiac function especially if there is enhanced AV nodal conduction (36) or an accessory AV pathway as occurs in the Wolff-Parkinson White Syndrome (37). Most commonly, and perhaps of most importance for the great majority of patients with AF, rapid rates cause intolerable and alarming symptoms for the patient and can raise the concern of the doctor. Chronically, rate control can improve symptoms, exercise capacity, and cardiac function. Even for those patients with no symptoms and no measurable hemodynamic abnormalities, rapid rates are a potential problem. Inadequate rate control can cause hemodynamic impairment, tachycardia induced cardiomyopathy and subsequent heart failure symptoms (3). It is likely that many patients with AF have inadequate rate control and develop a cardiomyopathy due to longstanding rapid rates. It is not known what heart rate places patients at risk but likely chronic average rates exceeding 100 beats/minute even at rest are the issue. While there is some evidence that tachycardia induced cardiomyopathy reverses based on improvements in ejection fraction, return of rapid rate causes rapid deterioration in cardiac performance. Nerheim et al followed 24 patients with NYHA functional class III heart failure or greater on presentation. The cause was atrial fibrillation in 13 of them. Within 6 months of rate control or correction of the rhythm, left ventricular ejection fraction improved and symptoms resolved in all. Tachycardia recurred in 5 of them. In these patients, left ventricular ejection fraction dropped precipitously and heart failure ensued within 6 months, even though the initial impairment took years(3). Thus, Aggressive rate control from the start can prevent this deleterious consequence. APPROACHES TO RATE CONTROL Acute Rate Control For the patient who presents in an emergent situation with rapid ventricular rates in AF, the approach requires understanding the clinical situation first. If an acute trigger is identified and treated, this may correct AF. If the patient is acutely ischemic, in the throes of congestive heart failure or hypotensive, the situation is more critical and the decision to control rate or to cardiovert is more urgent.

9 In most cases, rate control can be achieved with an intravenous medication. In rare circumstances, immediate cardioversion is the preferred approach. When AF causes severe hemodynamic compromise, electrical cardioversion is the treatment of choice (37). This is an exceedingly rare presentation. In most cases, rate control can be achieved with IV medications. If Wolff-Parkinson-White-Syndrome is the problem, the approach differs markedly. Aggressive attempts to maintain sinus rhythm are the rule acutely. Drugs used for acute rate control include intravenous (IV) calcium channel blockers, beta-blockers, digoxin and clonidine. No rigorous studies compare these drug classes and the approach differs based on the clinical condition. Beta-blockers and calcium channel blockers can control the rate effectively. Digoxin may be less effective as a single drug in the acute setting. Beta-blockers may make the most sense in situations in which high catecholamine states drive the rapid rate (38-45). Such conditions include acute myocardial infarction, myocardial ischemia, pregnancy, postoperative states especially after cardiac surgery, and hypertensive crisis. IV metoprolol and esmolol are the most commonly used beta-blockers to control the ventricular rate. The advantage of metoprolol is that it is relatively inexpensive. The difficulty is its long half-life. If there is an adverse effect from its use (bradycardia, heart failure, bronchospasm, hypotension), the effect will be long lasting. An advantage of esmolol is its short half-life. The drug can be titrated easily and stopped if there are adverse effects without long-term consequence as the half-life is only about nine minutes. Side effects are less likely to be a problem with esmolol due to its short (nine-minute) half-life but the drug is more expensive and harder to titrate from those who are inexperienced with its use. The disadvantage is complex drug dosing and the expense. Some hospitals restrict the use of esmolol to intensive care settings whereas IV diltiazem and metoprolol are not similarly restricted Calcium channel blockers are relatively versatile to control the ventricular rate. IV diltiazem and verapamil have been well studied for acute control of heart rate in patients with AF (46) (47-49). Diltiazem or esmolol are similar in terms of their ability to control rate and both have a similar rapid onset of action. IV diltiazem is surprisingly well tolerated in a variety of circumstances including acute congestive heart failure, myocardial ischemia, malignant hypertension, AF associated with respiratory distress and pulmonary conditions, relative hypotension and even circumstances in

10 which it appears acute cardioversion is required. This has been proven in multiple studies and is presumed to be due to the improvement of stroke volume with increased diastolic filling time (50-52). Similar to diltiazem, IV verapamil also can control the ventricular response rate to AF but the drug is more of a negative inotrope, can cause a greater degree of hypotension and is more challenging to titrate. Verapamil is not the calcium channel blocker of choice to control the ventricular rate in AF. Diltiazem remains one of the most popular IV infusions to control the ventricular response rate to AF in a variety of acute situations. IV digoxin, perhaps less effective, may be appropriate in conditions associated with acute congestive heart failure and poor ventricular systolic function. The largest study to evaluate digoxin in acute AF was the DAAF study, a randomized double-blinded study of IV digoxin versus placebo in acute AF (53-55). This study, with 289 patients enrolled, was designed to determine if digoxin increased the rate of conversion to sinus rhythm. Digoxin did not increase rate of conversion to sinus rhythm but digoxin did show a rapid decrease in ventricular rate, significant at 2 hours and this was dose dependant. Aggressive titration of IV digoxin is rarely used as the first option. The initial loading dose of digoxin is 1 mg and is independent of renal function. The time to effect can be several hours. Shorter acting glycosides, include strophanthidin-arabinoside and ouabain, used in the past, are no longer available (50,56,57). For the Wolff-Parkinson-White syndrome the above rate controlling drugs, especially digoxin are not to be given (58) as ventricular fibrillation can ensue (59). Rather, cardioversion for poorly tolerated episodes of a class I drug, such as IV procainamide, is recommended. It will block in the accessory pathway and may terminate atrial fibrillation. Low-dose clonidine is also a potential treatment for patients with rapid atrial fibrillation who are hemodynamically stable. A small randomized control trial in an ER setting showed that mg oral clonidine, at baseline and after 2 hours, achieved a mean decreases in heart rate of 38 beats/min (60). In another randomized trial, clonidine was shown to be comparable to controls

11 ventricular rate in new-onset atrial fibrillation with an efficacy comparable to that of standard agents (61). IV amiodarone can control the ventricular rate to AF in critically ill patients but it is rarely given for this indication. Compared to diltiazem, amiodarone is not as effective in controlling rate but it is even less likely to cause intolerable hypotension. Amiodarone is an alternative in patients with severe hemodynamic compromise (62). IV magnesium sulfate may help slow the ventricular rate safely and have some adjunctive role in helping to control rate. As a single intervention, magnesium is generally ineffective for rate control (63). IV amiodarone and IV magnesium sulfate may contribute to rate control if other therapies are not fully effective(63-65). While studies have compared these drug classes, the studies are small, poorly controlled and directed at specific populations of patients. The two most common approaches to control rate are single IV infusions with or without a bolus of one of the two beta-blockers mentioned or diltiazem. Chronic rate control Extensive guidelines suggest various rate control approaches with little data.(35) Rate control in AF can improve symptoms, exercise capacity and cardiac function. It remains uncertain what therapeutic endpoint provides the best compromise between hemodynamic and symptomatic benefit and drug-induced adverse effects. Adequate rate control at rest often parallels but does not ensure adequacy of rate control with exertion. Rate control does not guarantee the best exercise tolerance or best quality-of-life. Rate control may be inadequate but the patient may feel well and, vice versa. Rate control may be adequate but the patient developed side effect of the drug. Drug classes used to control the ventricular rate response in AF include digoxin, calcium channel blockers, and beta-adrenergic blockers alone and in combination. Each available drug class has specific advantages and each has the potential to cause adverse side effects. Despite policy statements and a wealth of data evaluating drugs to manage AF, no carefully controlled, comparative, trials have yet determined the best drug class to control the ventricular rate in AF (and in which patients).

12 Segal analyzed 45 repots evaluating the effects of 17 drugs on rate control (66). In seven of 12 comparisons of a beta-blocker with placebo, the beta-blocker was efficacious to control the resting heart rate. Evidence is scant but suggests that a drug specific effect is possible - nadolol and atenolol proved to be most efficacious. In trials of verapamil and diltiazem, rate was reduced at rest and with exercise compared with placebo. In seven of eight trials, digoxin slowed the resting heart rate more than placebo, but it did not significantly slow the rate during exercise in four studies. All three-drug classes showed efficacy but the relative potency of one versus another is not completely clear. Studies examining rate control in chronic AF have been small, with most studies evaluating less than 50 patients who have had and follow-up of 4 weeks or less. Trials evaluating rate control drugs have been small, and limited to assessing endpoints of heart rate and exercise response. Long-acting, dihydropyridine, calcium-channel blockers have specific safety issues in patients with coronary artery disease. For most patients with AF, chronic rate control can be achieved with one drug. In those with normal ventricular function, diltiazem, atenolol or metoprolol are appropriate. For those with impaired ventricular function and heart failure carvedilol may be the best. Amiodarone (67) and clonidine (68,69) may be effective as adjunctive therapy in the patient with difficult to control rates. In addition, Heart rate may self-adjust in some patients. Digoxin For chronic AF, digoxin had been the only drug for rate control for decades. The action appears to be due to enhancement of vagal tone (70). As such, when it slows the rate it does so with an irregular ventricular response rate. Digoxin may also exert a favorable effect on the beta-receptors (71).

13 Digoxin, at nontoxic doses, is safe but its effect on rate may be limited at rest and absent with exercise. It is likely that ventricular rate control could be achieved at rest in most patients with digoxin, but only at levels exceeding 2.0 ng/ml. At nontoxic levels, digoxin is relatively safe for patients with impaired left ventricular function and does not increase overall mortality (54). Digoxin is contraindicated in the Wolff-Parkinson-White syndrome where it can paradoxically increase the ventricular response rate to AF precipitating ventricular fibrillation. With digoxin, wide swings in heart rate, despite rate control, also may limit improvement in symptoms (and cardiac output), even at rest. Digoxin may also shorten atrial refractory period, increase anisotropy, and facilitate AF (as opposed to beta-adrenergic blockers that may have efficacy in preventing AF recurrence). In the AFFIRM Study, rate control with digoxin during exercise was similar to that with betablockers (72,73). This finding is perplexing, but it is possible that digoxin alone was given primarily to patients who were chronotropically incompetent, as evidenced by a slower initial ventricular response rate to AF. It is possible that these results did not portray a fair representation of the effects of digoxin in an unselected population. On the other hand, digoxin can be associated with adequate rate control when the drug is used. Similarly, it appeared that combinations of digoxin with other drugs were associated with improved rate control. Such results may not be achievable in an unselected population, but as prescribed clinically, rate control can be achieved with these drug combinations without undue risk. Recent data indicate that rate control using calcium antagonists is superior to digoxin in improving quality of life and exercise tolerance for patient with chronic AF (74). Calcium Channel Blockers Calcium channel blockers may provide a controlled and, perhaps, more regular ventricular rate response at rest and with exertion. While an attractive first-line approach to rate control AF, calcium channel blockers, however, may have adverse effects, cause hypotension and worsen heart failure. They may increase mortality for some patient populations especially those patients who have underlying coronary artery disease and who have had a myocardial infarction.

14 Several studies compare the effects of calcium channel blockers on rate control to digoxin for chronic AF. Lang showed that verapamil alone or in combination with digoxin, was superior to digoxin alone in decreasing heart rate at rest and with exercise (75). In a subsequent prospective double-blinded study (75), verapamil combined with digoxin improved exercise capacity. Comparing medium (240mg/day) to high-dose diltiazem (360mg/day) alone or in combination with digoxin (76) showed that medium dose diltiazem was equivalent to digoxin for control of resting rate. Higher dose diltiazem was associated with higher side effects (75%) in this crossover study of 12 patients. Combination digoxin and medium-dose diltiazem was superior to therapy with either drug alone, for both resting and exercise heart rate control. Lundstrom (77) found verapamil and diltiazem equivalent when combined with digoxin in patients with chronic AF. Crossover data comparing digoxin to diltiazem reported an increase in ventricular ectopy in patients receiving digoxin treatment compared to diltiazem treatment for rate control (78) (79), though the clinical significance of this is uncertain. Verapamil, alone or in combination with digoxin, can be superior to digoxin alone in decreasing resting and exercise rate (80). Verapamil combined with digoxin can improve exercise capacity (75,81). Diltiazem, alone, or combined with digoxin, was similar to digoxin for control of resting heart rate (76) but high dose diltiazem can have frequent adverse effects(82). Verapamil and diltiazem when combined with digoxin may have similar effects in patients with chronic AF (77). Their use may be problematic following myocardial infarction and with heart failure. Farshi (82), comparing five pharmacological regimens in 12 patients with AF, found diltiazem and digoxin combinations superior to either drug alone for 24-hour heart rate control. Beta-Blockers Beta-blockers alone, or in combination with digoxin, are very effective to control the ventricular rate at rest and during exercise. Beta-blockers can also help maintain sinus rhythm. Beta-blockers are the preferred drug class for most patients with AF requiring rate control.

15 Beta-blockers may be more useful as adjuvant therapy for other reasons in this patient population including their role in reducing mortality. In patients with heart failure, beta-blockers have additional specific advantages. Beta-blockers are an effective drug class to control rapid ventricular rate response when there is excess sympathetic stimulation. They may also help to maintain sinus rhythm. Beta-blockers may have little advantage over other rate controlling therapies for sedentary patients with low resting sympathetic tone. Beta-adrenergic blockers can also cause a variety of adverse effects that may limit their use substantially but these purported side effects may be over emphasized (16). Beta-adrenergic blockers may not only provide the best rate control but also may reduce mortality. Beta-blockers may help maintain normal sinus rhythm better than other options to control rate and may be as effective as some antiarrhythmic drugs in this regard. Few randomized controlled data document the best approach to rate control. Farshi (82) performed one of the best trials to assess rate control (at rest, with exercise and during daily activities in AF). A beta-blocker/digoxin combination provided the best but the study did not compare specific drugs in any drug class and did not test various doses. The study included only 12 patients and did not evaluate the long-term effects of the drugs or symptoms related to rate control. The best controlled randomized data is in a study by Farshi (82) comparing digoxin, diltiazem, atenolol, digoxin and diltiazem and digoxin with atenolol. The best combination for exercise occurred with digoxin and atenolol, but this was in 12 patients and the rate control was not necessarily reflective of improvement and outcomes, quality of life, mortality or any other objective measure. In this open-label, crossover, study atenolol alone was superior to treatment with digoxin or diltiazem alone for control of exercise rate, though there were no differences in 24-hour mean ventricular rate. In this study, there were no differences in exercise time with atenolol or combination atenolol/digoxin contrary to prior published studies. Most reports assess beta-blockers in combination with digoxin (83). The effect of nadolol on resting and exercise heart rates in AF patients receiving digoxin (84) was reduced compared to digoxin alone. However, nadolol decreased exercise time. Similarly, Atwood (85) found that the

16 beta-1-selective agent, celiprolol, improved rate control to exercise, but decreased exercise capacity and VO 2. An excessive dose of beta-blocker may explain that decrease in exercise tolerance.(86) Despite the many trials that evaluate beta-blockers in rate control for AF, it is unclear how clinical outcomes differ based on the beta-blocker use. Some beta-blockers have partial agonist activity and this may be important (87). Others are lipid or water soluble and half-lives differ. Some are cardioselective and some are not. Associated properties may have further benefits. The most used beta-blockers for rate control of AF are metoprolol and atenolol despite little comparative data confirming that these are the best beta- blockers to use. Carvedilol may have a specific benefit in AF patients with heart failure and who are post myocardial infarction (88-90). There have been recent studies evaluating sotalol, a beta-blocker with class III antiarrhythmic therapy, for rate control. In a recent study, metoprolol and sotalol were compared, in combination with digoxin, versus digoxin alone (91). Like previous studies, either beta-blocker when combined with digoxin, was more effective than digoxin alone, though sotalol gave greater heart rate control during daily activities and with submaximal exercise compared to metoprolol. Verapamil has been compared to Xamoterol, a beta-blocker, (92). Xamoterol was preferable to verapamil for treatment of patients with chronic AF who exhibit resting bradycardia and excessive exercise induced tachycardia. Beta-blockers may additionally help to maintain sinus rhythm better than other rate control options. Rate control in the AFFIRM Trial In the AFFIRM trial (11) (22) (72), we had the opportunity to evaluate a large number of patients and determine rate control approaches that were potentially successful. Standard drug classes were used for rate control, including digoxin, calcium channel blockers and beta-blockers. The presence of AF was defined as an episode of AF lasting at least 6 hours over the previous 6 months. There were stringent endpoints regarding successful rate control, which included a rate <80 beats/ minute at rest and a rate < 110 beats/minute with a 6-minute walk test or a 24-hour

17 Holter monitor which showed rates 100 beats/minute (at least 18-hours of interpretable monitoring) and no heart rate > 110% maximum predicted age-adjusted exercise heart rate. Over the duration of the study, rate control defined by these criteria, improved at the first few visits from 60% to almost 80% at 5 years. It appeared that rate control improved over time. Drug switches were possible within a drug class and outside the drug class. Some patients were on drug combinations. AFFIRM included a high-risk population for both stroke and death. Patients has at least one or more of the following risk factors for stroke: Age older than 65, hypertension, diabetes mellitus, congestive heart failure, prior transient ischemic attack, and left ventricular dysfunction. From the AFFIRM Study (11) data from 2027 rate control patients was evaluated (22) (72). Rate control was achieved at rest with a beta-blocker (with or without digoxin) in 75% of patients, with digoxin alone in 68% and with a calcium channel blocker (with or without digoxin) in 66% (22) (72). The best choice for effective rate control was a beta-blocker. Digoxin was associated with adequate rate control in many patients. At 5 years, about 80% of patients had successful rate control with few side effects. More patients taking digoxin or calcium channel blockers were switched to betablockers than vice versa (p<0.0001). In the AFFIRM trial, the initial treatment with the beta-blocker occurred in 24% of patients, with a calcium channel blocker in 17%, with digoxin alone in 16%, with beta-blocker and digoxin in 14% or with calcium channel blocker and digoxin in 14%. Adequate rate control was achieved in 58% with the first drug or drug combination. Overall, rate control was achieved in 70% of patients given beta-blockers as the first drug with or without digoxin, 54% of patients with calcium channel blockers with or without digoxin, and 58% of patients with digoxin alone. The average follow up was 3.5 ± 1.3 years. Multivariate analysis revealed an association with the first drug class and several clinical variables. Rate control can be difficult, and drugs had to be changed in approximately one-third of patients. Drug therapy was ultimately successful by the AFFIRM definitions in two-thirds of the patients, and rate control improved over time. More patients were switched to beta-adrenergic blockers than to other drug classes. Often combination therapy was needed. AV junctional ablation and a

18 permanent pacemaker were used in 108 patients. A pacemaker was placed for symptomatic bradycardia in 147 patients. In the AFFIRM Study, changing from one rate-controlling drug class to another was common. More patients were switched to beta-adrenergic blockers than to other drug classes. Drug combinations were associated with improved rate control. Allowing changes in rate-controlling drugs and combination therapy may explain the improved rate control seen in the AFFIRM Study over time. The success of achieving rate control in the AFFIRM Study may hinge on the flexibility of the investigators to use more than one drug class to control rate. It is possible that any drug class was not fully tested and any drug titration was not complete. In AFFIRM, no placebo was compared to other rate control attempts and no mandatory baseline was mandated during which drug therapy was prohibited. Drug selection, dose selection and titration, and drug changes were discretionary. We could not exclude the contribution of natural changes in AV nodal conduction with time. Without a placebo control or a washout period, it was not possible to assess the mechanism of improved rate control over time. Stringent rate control might not be the best approach. A substudy of the AFFIRM trial showed that rigid rate control did not improve event-free survival ( p = 0.81), overall survival ( p = 0.13), or the quality of life for patients with AF. Therefore, lenient rate control may not be inferior to stringent approach (93). The RACE II trial (Rate Control Efficacy in permanent atrial fibrillation) is an ongoing trial that addresses the question of level of rate control strictness that should be applied. It will be looking at endpoints like cardiovascular mortality, morbidity, neurohormonal activation, New York Heart Association class for heart failure, left atrial size, quality of life and costs (14). The importance of these questions stem from the fact that 108/2027 patients, or 5%, required an AV junctional ablation and 147 required a permanent pacemaker in the AFFIRM study. Thus, it is possible that fewer patients would have required a permanent pacemaker or an AV junctional ablation if a more lenient approach were taken. This occurred in the RACE trial where patients were not as rigorously evaluated for rate control with a Holter monitor or a 6-minute walk test and average rates could be up to 100 beats/ minute.

19 The event free survival was similar in both studies, but fewer pacemakers were required in the RACE trial. A study comparing morbidity and mortality endpoints showed no significant difference between RACE and AFFIRM (94). Van Gelder et al. compared the outcome of patients enrolled in the rate-control arms of AFFIRM and RACE in an attempt to assess the effect of strict rate control (95). They utilized data from patients who met a composite of overlapping inclusion and exclusion criteria. Thus, 1091 patients were evaluated (874 from AFFIRM and 217 from RACE). In AFFIRM, the rate-control strategy aimed for a resting heart rate < or =80 bpm and heart rate during daily activity of < or =110 bpm. In RACE, a more lenient approach was taken: resting heart rate <100 bpm. Primary endpoint was a composite of mortality, cardiovascular hospitalization, and myocardial infarction. Mean heart rate across all follow-up visits for patients in AF was lower in AFFIRM (76.1 vs bpm). There was no significant difference in the event-free survival for the occurrence of the primary endpoint between the two studies (64% in AFFIRM vs. 66% in RACE). The authors concluded that stringency of the approach to rate control was not associated with an important difference in clinical events. In conclusion, optimal rate control criteria are not at the present. Resting tachycardia should not be allowed. Heart rate variation with exercise should be considered individually. More prospective data is needed to best define rate control criteria. WHO SHOULD BE CONSIDERED FOR AV JUNCTIONAL ABLATION? AV junctional ablation should be considered for patients resistant to drug therapy who have rapid and/or wide fluctuations in rates. Patients will require a long-term anticoagulation therapy. (96-99). Patients improve markedly after AV junctional ablation and pacing especially if there are substantial symptoms from rapid rates (96) (100,101). Part of the improvement may be related to regularization of rate. Following ablation, patients may still need drugs to maintain sinus rhythm. Anticoagulation is generally required. AV junctional ablation is not considered a first-line approach to control the ventricular rates in AF. AV junctional ablation causes unopposed right ventricular pacing with possible adverse outcomes, including death, especially if initial pacing rates are kept low (102). In a recent trial long-term RV

20 pacing after AV node ablation for chronic AF was shown to induce LV dyssynchrony in almost 50% of patients. The development of LV dyssynchrony was associated with deterioration in heart failure symptoms, systolic LV function, and LV dilatation (103). In another study, sudden death likely or possibly related to catheter ablation occurred in seven of 334 patients (2.1%). Risk of sudden death is highest within two days after the procedure in one report.(104) Programming the ventricular rate to faster rates for the first several weeks eliminates the risk of Torsades de Pointes in these patients. The AIRCRAFT trial, a randomized trial of patients compared AV junctional ablation with a permanent pacemaker to medical therapy for patients with mild to moderately symptoms from AF. Permanent pacing did not worsen cardiac function. The ejection fraction and exercise time was the same in both groups. The peak rate was lower in the AV junctional ablation group with the same amount of exercise and daily activity. The AV junctional ablation group had less symptoms and the global subjective quality-of-life, using the ladder-of-life scale, was 6% better with AV junctional ablation. The conclusion was that ablation and pacing in symptomatic patients with permanent AF did not worsen cardiac function long-term and side effects and quality-of-life improved (105). The Post AV Nodal Ablation Evalution (PAVE) trial, evaluating bi-ventricular pacing after AV junctional ablation, showed mild improvement in functional performance and exercise capacity for patients undergoing bi-ventricular as opposed to RV pacing alone (101). This approach is not yet recommended. In addition, small studies indicate that para-hisian pacing is feasible and safe and allows an improvement in functional and hemodynamic parameters when compared with conventional right ventricular apical pacing (106,107). The PAF 2 trial evaluated effects of antiarrhythmic drugs on outcomes after ablation and pacing therapy in a small, multicenter, randomized controlled trial. Patients in the antiarrhythmic drug arm had less chronic AF but enjoyed no further benefit beyond ablation and pacing alone. Antiarrhythmic therapy was associated with serious adverse events, including heart failure and hospitalization. (99)

21 AV junctional modification using ablation to slow conduction through the AV node enjoyed a shortterm use but long-term efficacy was low and there was a risk that a pacemaker would be required. This approach is all essentially abandoned WHICH IS BETTER RATE CONTROL OR RHYTHM CONTROL? Randomized trials indicate that rate control is possibly the superior approach for elderly patients at risk for stroke (11,14). PIAF (Pharmacological Intervention in AF), STAF (STrategies in AF, AFFIRM (AF Follow-up Investigation of Rhythm Management, and RACE (RAte Control versus Electrical Cardioversion for Persistent AF), SAFE-T (The Sotalol Amiodarone (Atrial) Fibrillation Efficacy Trial) and HOT CAFÉ are among the randomized clinical trials performed and the data are consistent. (11,14,15, ) Trials are ongoing in select patient populations (112). Randomized trial data indicate that rate control is an acceptable option for patients with AF enrolled in these trials. In a meta-analysis of 5,239 patients with persistent AF or AF considered likely to recur from 5 major randomized trials, no difference was observed between rate and the rhythm control groups in all-cause mortality (13.0% vs 14.6%; OR, 0.87; 95% CI, ; P=0.09). The data were consistent from trial to trial (113). The AFFIRM study (11) included a diverse group of patients with AF at risk for stroke. Patients were randomized to standard rate control or rhythm control strategies. In both groups, the recommendation was to anticoagulate since patients were at high risk for stroke. The primary endpoint was total mortality, not symptom reduction. Four thousand sixty patients were enrolled. The mean age was 69.7 and follow-up lasted a mean of 3.5 (range 2-6) years. The diverse population included 39% females, 71% hypertensives, 38% with coronary artery disease and 23% with congestive heart failure. Many patients in the AFFIRM trial were symptomatic. Approximately 12% had more than one symptom/day, 15% had symptoms 1/day 1/week; 20% had symptoms 1/week 1/month. Only 6% were asymptomatic.

22 At the end of the study, sinus rhythm was present in 65% of the patients assigned to rhythm control and in 35% assigned to rate control strategy. These rates of sinus rhythm are higher than what was achieved in other trials. Worth mentioning is that 25% of the patients were already converted back to sinus rhythm before randomization after their qualifying episode of AF (114). There was no difference in quality-of-life between groups but there was a slight decrease in qualityof-life compared to the general population corrected for age (115). The mortality in the rate control and the rhythm control arms was similar with a trend favoring rate control (P= 0.08). Causes of death in this trial deserve careful attention. In the rate and rhythm control groups, 130 and 129 patients died of cardiac cause, respectively. However, more patients in the rhythm control group died on noncardiovascular cause. Pulmonary deaths were significantly higher in the rhythm control group (23 vs 39, for rate and rhythm controls, p = 0.04) and so was the cancer deaths ( 52 vs 81, for rate and rhythm controls, p = 0.01) (22) (72). It is also important to try to identify subgroups of patients who will specifically benefit from a certain treatment strategy. For example, in an AFFIRM subanalysis, patients > 65 years of age and those without heart failure seemed to have benefited from rate control above rhythm control ( P < 0.01) (116). Digoxin and antiarrhythmic drug use were associated with poorer survival. Warfarin and the presence of sinus rhythm were associated with a better survival. Importantly, it was not the approach (rate vs. rhythm control) but absence of AF in either rate or rhythm control strategies that was associated with a better survival. Patients in sinus rhythm may have been less ill.(11) Apparent rhythm control did not protect from stroke. The RACE study, involving patients with persistent AF (14), enrolled, 522 patients (63% male, mean age 68 years) randomized to a rate or rhythm control strategy with the hypothesis that rate control in chronic AF was not inferior to rhythm control. The rhythm control arm had more non-fatal endpoints but there was no difference in mortality between groups (6.7% in each). There was a non-significant trend to higher incidence of primary endpoint with rhythm control (22.6 vs. 17.2%). Sinus rhythm was present in 10% in the rate control arm and 40% in the rhythm control arm. A trend towards a better outcome existed in the rate control arm.