Management of Arrhythmias and Device Therapy in Heart Failure

Epidemiology

In the Framingham Heart Study, 2326 men and 2866 women were followed for 2 years, and the risk of developing permanent AF was 8.5% for men and 13.7% for women. Paroxysmal AF was seen in 8.2% of men and 20.4% of women. In those without prior or concurrent congestive heart failure or myocardial infarction, the lifetime risks for AF were approximately 16%. In diastolic heart failure, approximately 25% to 30% of patients have evidence of AF. The prevalence increases with the severity of diastolic heart failure, reaching up to 40% in advanced stages. Sustained high ventricular rates during atrial arrhythmias may result in a tachycardia-induced cardiomyopathy that in the absence of any other myopathic process may be completely reversible. However, in patients who develop AF in the setting of established heart failure, atrial fibrosis with deleterious electrical remodeling is the primary cause for the arrhythmia.

Pathophysiology

The clinical manifestations of AF relate to the loss of atrial systolic function and an irregular, ventricular response whose rate is typically determined by the conduction and refractory properties of the atrioventricular (AV) node. Left atrial systole contributes up to 25% of the cardiac output, and this fraction may increase to 50% in left ventricular failure. The loss of atrial systolic function results in impaired hemodynamic function of the heart. AF causes a fall in cardiac stroke volume of about 10% in normal subjects, with a greater decrease seen at fast ventricular rates. This loss becomes more important clinically with increasing age and progressive impairment of left ventricular systolic or diastolic function, because atrial systole makes a greater contribution toward the overall stroke volume in these conditions. In addition to the loss of AV synchrony, the irregular and often inappropriate ventricular rates seen in AF result in suboptimal ventricular filling. These may further compromise cardiac output, an effect particularly seen in patients with mitral stenosis or diastolic dysfunction. Cardioversion of AF with poorly controlled ventricular rates usually improves left ventricular ejection fraction (LVEF) and exercise capacity, but the improvement occurs gradually after the procedure.

Loss of atrial systolic function results in stasis within the left atrium, leading to intra-atrial thrombus formation and an increased risk of stroke and thromboembolism. During episodes of AF, echocardiography can detect spontaneous echo contrast (SEC) as a result of the formation of erythrocyte aggregations and thrombus in the atria. Stasis within the left atrium has been related to hemostatic abnormalities that are suggestive of a hypercoagulable state and that involve coagulation factors and abnormal endothelial and platelet function. These abnormalities of hemostasis have been related to changes in inflammatory indices and growth factors. The hypercoagulable state is often exaggerated in low flow states such as left ventricular dysfunction. Hypercoagulability is altered by antithrombotic therapy and gradually by cardioversion of AF to sinus rhythm. Prothrombotic indices in AF have been shown to be prognostically relevant, being predictive of stroke and vascular events, and can be used to refine clinical stroke risk stratification. Atrial natriuretic peptide levels are also increased in patients who have AF, which contributes to hemoconcentration and an increased risk for thrombus formation.

Prolonged AF with rapid ventricular rates produces functional, ultrastructural, and microscopic changes within the myocardium that may result in progressive left ventricular dilation and reduction of left ventricular systolic function; this is referred to as tachycardia-induced cardiomyopathy. In a patient who has chronic heart failure and is in sinus rhythm, increased intracardiac pressures may lead to atrial stretch and dilation, predisposing to both the development and the recurrence of AF. Several potential electrophysiologic mechanisms may be responsible for AF. One postulated mechanism is multiple wavelet reentry, in which wavefronts continuously sweep through the atria in a random fashion. The multiple wavelet hypothesis requires that a minimum number of wavefronts and enough atrial tissue to permit their simultaneous propagation exist. An alternate hypothesis is that there are only one or two primary reentrant circuits or rotors that are constantly forming and disappearing, but the cycle lengths in these circuits are too short to allow the rest of the atria to follow in an organized fashion, resulting in fibrillatory conduction. It has been shown clinically that AF may be produced by rapid tachycardias from either focal sources, commonly found in musculature of the pulmonary veins, or stable reentrant circuits that drive the remaining atrial tissue until degeneration to AF occurs. The pulmonary veins, which include muscular sleeves that may be electrically active, and the posterior left atrial wall are considered to be the critical structures involved in the pathogenesis of AF.

Clinical Presentation

The classification of AF focuses on temporal pattern after onset. Paroxysmal AF is a recurrence of AF that terminates spontaneously in 7 days or less. Persistent AF is recurrent AF that lasts longer than 7 days. Patients who undergo cardioversion within 48 hours of onset are described as paroxysmal and persistent if done after 48 hours. Longstanding persistent AF is continuous AF that has lasted longer than 1 year. Permanent AF is when the patient remains in longstanding persistent AF and there is no plan for rhythm control.

Among patients who have recurrent forms of AF, this temporally based clinical classification can assist management strategies, particular in relation to considering rhythm control or rate control. In paroxysmal AF, the episodes are generally self-terminating, and thus the goals of therapy are the prevention of paroxysms and the long-term maintenance of sinus rhythm. In sustained AF, the therapeutic goal is either cardioversion to sinus rhythm or heart rate control. Antithrombotic therapy for moderate or high-risk patients is an important component of both strategies.

The differentiation between these clinical categories is dependent on the history given by the patient, electrocardiogram (ECG) documentation of the episode, and the duration of the most recent previous episode of AF. Although this classification is helpful, there is considerable variability, both between patients and in the same patient, in the temporal pattern of AF episodes, and approaches to therapy must be individualized, especially in relation to symptoms. Furthermore, paroxysmal AF may become permanent (8% at 1 year, 18% at 4 years), especially with increasing age. The EURO Heart Survey showed that hypertension, age older than 75 years, previous transient ischemic attack or stroke, chronic obstructive pulmonary disease, and heart failure were independent predictors of AF progression. These investigators used these factors to develop the HATCH score, which predicts the probability of progression of AF. With an increasing HATCH score, the percentage of patients in whom AF progressed to persistent forms was significantly higher. Fifty percent of the patients with a HATCH score more than five progressed to persistent AF, compared with only 6% of the patients with a HATCH score of 0.

Diagnosis

The investigation of a patient with AF requires a careful clinical history (including a past medical history) with emphasis on certain clinical features. The history should cover whether the symptoms are sustained or intermittent and whether any complications (such as heart failure, stroke, or thromboembolism) are present. Other useful data include the date of the first episode, information about acute precipitating factors or chronic conditions linked to AF, how symptoms are relieved, the typical duration of episodes and the typical interval between them, the duration of the current or most recent episode, and current and past drug treatment for both rate and rhythm control.

At the initial consultation, basic blood tests, including full blood count, biochemistry (renal function, electrolytes), and thyroid function tests, are taken. A full blood count is useful to exclude anemia because anticoagulation may be considered. Serum urea and electrolytes are relevant for consideration of drug therapy (e.g., the dose of digoxin would be reduced in renal impairment). The risk of AF is increased by clinical and subclinical hyperthyroidism, and thus the serum thyroid stimulating hormone level should be measured in all patients who have AF, even if there are no symptoms suggestive of thyrotoxicosis.

The arrhythmia should be documented, ideally with a standard 12-lead ECG. The characteristic ECG findings in AF include rapid baseline oscillations or fibrillatory waves that vary in size, shape, and timing; the absence of discrete P waves; and an irregularly irregular ventricular rate.

The ECG may also provide a clue to the electrophysiologic features that may have caused AF (e.g., a previous myocardial infarction, left ventricular hypertrophy, or preexcitation). Often, patients present with previous symptoms suggestive of AF but are in sinus rhythm at the time of their evaluation. If symptoms occur on a daily basis, a 24-hour ambulatory ECG should provide the diagnosis. If symptoms occur less frequently, a patient-activated event recorder or an implanted loop recorder would be more likely to provide a diagnosis. Smartphone apps, smartwatch heart rate monitoring, and home blood pressure monitors are increasingly helping with arrhythmia diagnosis.

An echocardiogram provides important information for the initial evaluation of most patients with AF. Either transthoracic echocardiography or transesophageal echocardiography (TEE), or a combination of the two, may be appropriate. The initial goal of the echocardiographic evaluation should be to establish the presence or absence of structural heart disease, including valvular abnormalities, congenital anomalies, chamber dimensions, pericardial thickening or effusions, and ventricular function.

Left atrial size is an important predictor of outcome in patients with AF because the presence of significant left atrial enlargement has been shown to reduce the chances of successful cardioversion and long-term maintenance of sinus rhythm in most series. Left atrial enlargement may also increase the risk of stroke, owing to a greater potential for stasis in the dilated chamber.

Although transthoracic echocardiography is acceptable for assessing chamber size, detection of thrombi or the assessment of left atrial appendage anatomy and function requires the TEE approach. Using TEE, up to 27% of patients with AF of more than 3 days’ duration may have detectable thrombi. A prethrombotic finding, SEC (also called “smoke”), is due to erythrocyte aggregation in the low-flow state and is even more commonly seen. Other TEE indices of high stroke risk include the presence of left atrial thrombus, low left atrial appendage velocities, dense SEC, and complex aortic plaque in the descending aorta.

Invasive electrophysiologic studies have only a limited role in the routine evaluation of patients with AF unless catheter ablation is planned. Electrophysiologic studies of AF should be reserved for the following situations: when another arrhythmia (e.g., atrial flutter or atrial or supraventricular tachycardia) is thought to be the cause of the AF, when other electrophysiologic abnormalities or symptoms (e.g., preexcitation, sinus node dysfunction, syncope) require clarification or therapy, or when catheter ablation is planned.

Management of Acute Episodes of Atrial Fibrillation

Successful management of acute episodes of AF requires attention to several issues, including rate control, pharmacologic or electrical conversion, and protection against thromboembolic events. When patients present with new onset or recurrent episodes of AF, the initial step should be to assess their symptoms and hemodynamic status. Rarely, the patient will be so severely compromised that urgent electrical cardioversion will be required despite a high risk for early recurrence. In most patients, however, the first step for stabilization should be to lower the ventricular rate. In an intensive care unit or other monitored setting, intravenous β-blockers are usually the first choice in patients with heart failure and systolic dysfunction. Intravenous digoxin may be a useful adjunct, but its onset of action is delayed and its inability to lower rate is lessened during periods of high sympathetic tone. The non-dihydropyridine calcium channel blockers, diltiazem and verapamil, are effective alternatives in patients with preserved systolic function but must be used with caution, if at all, in patients with known systolic dysfunction. Intravenous amiodarone can play a dual role, since it will both slow ventricular rates and may eventually cardiovert the rhythm. The heart rate target during attempts at rate control will depend on the patient’s condition. In a resting, minimally symptomatic patient, one tries to approximate what the average rate would be in sinus rhythm (i.e., 60–100 beats/min). During periods of stress, however, this degree of rate control may not be achievable, and rates up to 110 to 120 beats/min are usually adequate to control symptoms.

Once the patient’s rate had been controlled, a decision about possible cardioversion should be made. Anticoagulation issues that will affect this decision will be discussed later. Most patients with new onset AF will be candidates for at least one attempt at cardioversion. Elective direct-current cardioversion without addition of an antiarrhythmic drug is an appropriate strategy for patients with recent onset AF in whom the risk of recurrence is thought to be only moderate or low. After cardioversion, the patient can be followed off antiarrhythmic drugs until a recurrence has been documented. AV nodal blocking agents are often continued at least during the early period after cardioversion. Although intravenous ibutilide and vernakalant (not currently available in the United States) are effective for conversion of recent onset AF, neither of these agents is advisable in patients with systolic heart failure or left ventricular hypertrophy. Intravenous or oral loading with amiodarone or oral loading with dofetilide is the best approach for pharmacologic cardioversion if long-term therapy is planned in the heart failure patient.

Chemical or electrical cardioversion of AF of more than 2 days’ duration is associated with a significant risk (2%–8%) of stroke or systemic thromboembolism in all patients with nonvalvular AF not on anticoagulants. A prudent approach is to start anticoagulation in patients without contraindications scheduled for cardioversion if the cardioversion is to be delayed or if the duration of the episode is uncertain. For episodes of greater than 48 hours’ duration, two anticoagulation strategies are acceptable. If early cardioversion is planned, the patient should undergo a transesophageal echocardiogram to exclude the presence of left atrial appendage thrombus. If no thrombus is visualized, anticoagulation is continued and cardioversion may be performed. The alternate strategy is to delay cardioversion until the patient has been adequately anticoagulated for at least 3 weeks. Although warfarin with an international normalized ratio (INR) between 2.0 and 3.0 has long been the anticoagulation standard, preliminary data suggest that a similar duration of anticoagulation with dabigatran, rivaroxaban, and apixaban would be similarly effective. The TEE-guided and delayed treatment approaches were compared in the ACUTE trial. Both approaches were associated with a low rate of thromboembolic events after cardioversion and similar probabilities for sinus rhythm maintenance and bleeding.

Chronic Management of Atrial Fibrillation

Patients with recurrent forms of AF will require a treatment program that provides rate control, an assessment of the benefits and feasibility of restoring and maintaining sinus rhythm, and mitigation of the risks for stroke and systemic embolism.

Rate Control Versus Rhythm Control

The first question that must be addressed in patients with recurrent AF is whether a rate control or a rhythm control strategy is most appropriate. Factors that should be considered include the temporal pattern (paroxysmal vs. persistent) of the arrhythmia, the frequency of episodes, the severity of symptoms, patient factors, and the probabilities for maintaining sinus rhythm or effectively controlling ventricular rates.

Although a rhythm control strategy would seem intuitively to be superior to a rate control strategy, a series of randomized trials has been unable to demonstrate this with pharmacologically based therapies. The two most relevant trials for heart failure patients were the AFFIRM trial and the AF-CHF trial. AFFIRM randomized 4060 patients, 23% of whom had heart failure, between rate control and rhythm control strategies. No difference in total mortality or stroke was seen between the two strategies, with a slight trend favoring rate control. Patients in whom sinus rhythm was maintained during the study had improved outcomes, but this likely represents a “healthy responder” phenomenon. A second trial, AF-CHF, compared rate control and rhythm control strategies in patients required to have heart failure and depressed left ventricular systolic function. There was no significant difference between the two strategies in the three primary endpoints, mortality, stroke, and heart failure hospitalizations. In addition, even when patients were grouped into those with high and low prevalence of sinus rhythm during the course of the study, no benefit on these outcomes could be demonstrated. However, it must be remembered that entry into all of the rate control versus rhythm control strategy trials required that the patient be a candidate for both approaches. Highly symptomatic patients therefore were unlikely to be randomized. Therefore, most clinicians recommend that heart failure patients with persistent symptoms related to their AF should have at least an initial attempt to restore and maintain sinus rhythm with rate control a fallback approach if rhythm control is unsuccessful or poorly tolerated.

Pharmacologic Rhythm Control

As shown in Fig. 38.1 , pharmacologic treatment options for maintaining sinus rhythm in patients with heart failure are limited. With the possible exception of disopyramide in patients with hypertrophic cardiomyopathy, class IA agents are not useful due to the high prevalence of side effects. Flecainide and propafenone are contraindicated in patients with heart failure. Dronedarone is similarly contraindicated in heart failure patients based on the data from the ANDROMEDA and PALLAS trials, both of which showed increased mortality with dronedarone therapy in patients with heart failure. Drug selection in patients with left ventricular hypertrophy is also limited by the risk for QT prolongation and torsades de pointes with class III agents like sotalol. These observations frequently leave only dofetilide (not currently marketed in the European Union) and amiodarone as treatment options for heart failure patients. Dofetilide is a relatively pure I _Kr blocker that is effective for both converting AF episodes and for maintaining sinus rhythm. In the Diamond-HF trial, dofetilide did not increase overall mortality and improved outcomes in the subgroup with AF, but careful attention to dose and effects on the QTc during therapy is required. Heart failure patients with unstable renal function would not be good candidates for dofetilide. Amiodarone, therefore, frequently remains the only reasonable choice for sinus rhythm maintenance in patients with heart failure. Amiodarone was the drug most frequently used in both AFFIRM and AF-CHF. The estimated probability of maintaining sinus rhythm with selective use of electrical cardioversion in patients on amiodarone is 60% to 80% at 2 to 3 years but decreases over time. Amiodarone has many side effects, however, and should be used at the lowest effective chronic dose, usually 200 mg daily.

A number of agents and interventions that do not have classic antiarrhythmic effects are also likely to be important to the overall management of AF in heart failure. Angiotensin-converting enzyme (ACE) inhibitors, angiotensin receptor blockers (ARBs), β-adrenergic blockers, and aldosterone antagonists may be useful in this regard because they may both delay the onset of AF initially and act synergistically with more traditional antiarrhythmic drugs long-term. Sleep apnea is another risk factor for AF that should also be treated when present.

Atrial Fibrillation Ablation in Heart Failure

AF ablation may be technically challenging in patients with congestive heart failure. The basic technique of pulmonary vein antral isolation alone is rarely successful due to left atrial enlargement, chronic left atrial hypertension, and diffuse atrial scarring ( Fig. 38.2 ). Additional linear lesions, both left and right atrial, and lesions targeting atrial electrograms that are fractionated are often placed with a modest increase in efficacy. Nevertheless, even though only intermediate success rate should be anticipated, catheter or surgical ablation may be a useful option in selected patients. Catheter ablation should probably be attempted before AV junctional ablation in younger patients without AV block because the latter procedure is irreversible and creates a situation of life-long pacemaker dependency.

The Catheter Ablation versus Antiarrhythmic Drug Therapy in Atrial Fibrillation (CABANA) Trial randomized 2204 patients with new onset AF either greater than 65 years or less than 65 years with ≥1 risk factor for stroke to initial therapy with catheter ablation or antiarrhythmic drugs. Congestive heart failure was reported in 15% of this population. The intention to treat analysis showed no difference in the composite endpoint of all-cause mortality, disabling stroke, bleeding, or cardiac arrest. Crossover between groups was substantial with 10% of the ablation group not receiving ablation and 28% of the drug group crossing over to ablation. A secondary analysis of treatment received showed significant reduction in the primary composite endpoint for patients undergoing ablation (7.0%) versus drug treatment (10.9%, HR 0.67). Patients with a history of heart failure did better than those with no prior heart failure.

The Catheter Ablation for Atrial Fibrillation with Heart Failure (CASTLE-AF) Trial randomized patients with systolic heart failure, an LVEF ≤35%, and New York Heart Association (NYHA) class II, III, IV symptoms to catheter ablation or medical therapy (rate or rhythm control). After a mean follow-up of 37.8 months, the composite endpoint of all-cause mortality or hospitalization for worsening heart failure was significantly lower in the ablation group (28.5% vs. 44.6%). There was also a 45% reduction in heart failure hospitalizations in the ablation group.