The Pharmaceutical Journal Vol 264 No 7093 p622-626
April 22, 2000 Continuing education

Atrial Fibrillation

(1) The condition

by Gregory Y. H. Lip, MD, FRCPE and Sridhar Kamath, MRCP

Atrial fibrillation is the commonest sustained cardiac rhythm disorder and an important cause of mortality and morbidity. It causes, for example, a four- to five-fold increase in risk of stroke. This is the first in a series of three articles discussing the condition and its treatment. The second article discusses antiarrhythmic agents, and the third article covers antithrombotic therapy

Until fairly recently, atrial fibrillation (AF) was often regarded as an innocuous cardiac rhythm, where patients could simply be treated with some digoxin and left well alone. Nothing could be further from the truth - AF is now recognised as an important cause of mortality and morbidity from heart failure, stroke and thromboembolism. Many patients with AF suffer from reduced exercise capacity as well as poor quality of life. Recent randomised controlled trials have clearly established the value of antithrombotic therapy in preventing stroke and thromboembolism in AF, and clinicians have also recognised the limitations of simply treating such patients with digoxin.

Epidemiology

Pathophysiology

The pathophysiology of AF is very complex. The most widely accepted mechanism is the multiple wavelet re-entry hypothesis of Moe,⁴ in which a critical number of coexisting wavefronts continuously sweep through the atria in a random, irregular, shifting fashion, repeatedly encountering excitable myocardium. This hypothesis requires a minimum number (4-6) of independent wavefronts and enough atrial tissue to permit their simultaneous propagation. However, Allessie's model⁵ suggests that AF is due to multiple wavelets continuously sweeping around the atria in patterns, which may collide with each other, extinguishing themselves or creating new wavelets and wavefronts, thereby perpetuating the arrhythmia.⁶ These rapid irregular impulses are transmitted down the atrioventricular (AV) node to the ventricle, resulting in irregular ventricular contractions.
The pathophysiological consequences of AF relate to the loss of effective atrial systolic function ("atrial kick") and atrioventricular synchrony, which effectively reduces stroke volume and hence cardiac output, especially in the presence of structural heart disease (eg, hypertensive left ventricular hypertrophy). The loss of atrial systolic function has two main sequelae: first, impaired haemodynamic function of the heart, and, secondly, intra-atrial stasis and thrombogenesis, predisposing to stroke and systemic thromboembolism. The impaired haemodynamic function accounts for the symptoms of reduced exercise tolerance and is partially compensated for by a fast heart rate, which may lead to palpitations, chest pain, giddiness or syncope, shortness of breath and anxiety.

Clinical presentation

AF can be symptomatic or asymptomatic. Both symptomatic and asymptomatic episodes of AF can occur in the same patient. In a comparison of patients with paroxysmal supraventricular tachycardia (PSVT) and patients with paroxysmal AF, most episodes of PSVT tended to be symptomatic while only one in 12 episodes of paroxysms of AF were symptomatic.⁷ Asymptomatic AF is usually discovered, incidentally, during cardiac auscultation or 12-lead ECG recording or 24-hour Holter recording undertaken for unrelated reasons.
Symptoms associated with AF can vary depending on several factors, including the ventricular rate, cardiac function, concomitant medical problems and individual patient perception. The duration of AF may be unknown in many patients, and whether AF was the cause or effect of (say) heart failure may be uncertain. Palpitations in patients with AF may be due to awareness of a fast heart rate or slow irregular rate. Very often the patient presents because of medical problems associated with AF, such as heart failure or stroke. Chest pain can be due to the precipitation of angina, while shortness of breath may be due to pulmonary oedema, and giddiness or syncope due to associated hypotension.
AF occuring in the context of acute myocardial infarction is associated with greater myocardial damage and poorer left ventricular function when compared with infarction without AF, resulting in a significantly higher in-hospital stroke and 30-day mortality.^8,9 Some patients with AF may present for the first time with symptoms relating to a stroke or thromboembolism. The latter may include systemic thromboembolisation to the leg, leading to acute limb ischaemia, or to organs such as the kidneys, causing infarction. The risk for stroke associated with AF rises from 1.5 per cent at the age of 50-59 years to 23.5 per cent at the age of 80-89 years.2 Very rarely, right sided thrombi may lead to pulmonary thromboembolism.
On physical examination, AF typically is associated with an irregularly irregular pulse, which is confirmed with an ECG showing atrial activity as irregular baseline undulations of varying amplitude and morphology, which are referred to as "f" (fibrillation) waves and could be as fast as 600/min. The ventricular complexes are irregularly irregular although the rate is rarely as fast as the atrial rate since the atrio- ventricular node is unable to conduct impulses at this fast rate unless impulses are conducted through an accessory pathway. In long standing AF, the baseline may appear smooth without any P wave. Rapid AF with a rapid ventricular response may be easily mistaken for other supraventricular arrhythmias (for example, atrial flutter or supraventricular tachycardias) or, if a bundle branch block is present, ventricular tachycardia. Subtle variation in the R-R interval is the important clue.
The chief hazard of AF is stroke, the risk of which is increased four- to five-fold. AF is also associated with a doubling of mortality in both sexes, which is decreased to 1.5-1.9-fold after adjusting for associated cardiovascular conditions.¹⁰ The presence of AF also significantly increases all-cause mortality and progressive pump failure death in patients with left ventricular systolic dysfunction, whether symptomatic or asymptomatic.¹¹

Management

Clinical classification One proposed clinical definition of AF divides patients into acute-onset and chronic AF; the latter are further divided into paroxysmal and sustained. Sustained AF is further divided into persistent and permanent AF. The differentiation between these clinical categories is dependent upon the history given by the patient, ECG documentation of the current episode and the duration of the last episode of AF.
The term "acute AF" is used to describe either an episode of AF related to a transient and reversible cause, the first onset of AF or a recent paroxysm of AF. The term "chronic AF" implies either recurrent or prolonged episodes. Chronic AF can then be subclassified as either paroxysmal (self terminating) or sustained. Persistent AF is sustained AF that can be successfully cardioverted to sinus rhythm, while permanent AF is sustained AF that is either resistant to, or not appropriate for, cardioversion.
This "3P" classification (that is, paroxysmal, persistent and permanent) allows adequate division along the lines of clinical objectives of management. It can assist management strategies by defining treatment objectives, although many patients change from one category to another and therapy must be individualised.¹²
For example, in paroxysmal AF, the episodes are generally self terminating and thus the main objective of management is the prevention of paroxysms and long-term maintenance of sinus rhythm, in addition to appropriate antithrombotic therapy. In persistent AF, the episode lasts beyond 48 hours and the possibility of cardioversion from AF to sinus rhythm remains; the objective of management is appropriate anticoagulation to reduce the risk of thromboembolism, cardioversion and antiarrhythmic therapy to maintain long-term sinus rhythm. In permanent AF, the objective of management is heart rate control and appropriate antithrombotic therapy.
Initial basic investigations and assessment in a patient presenting de novo with AF should include blood tests, chest X-ray and ECG. As with any arrhythmia, AF should first be documented before treatment is initiated. Fast AF in a young patient raises the suspicion of the Wolff-Parkinson-White syndrome, and a delta wave may be seen on the QRS complex. A chest X-ray may reveal intrathoracic pathology that may have precipitated AF.
An echocardiogram provides important information for the initial evaluation of most patients with AF. Either transthoracic (TTE) or transoesophageal (TEE) echocardiography, or a combination of the two, may be appropriate. The initial goal of the echocardiographic evaluation is to establish the presence or absence of structural heart disease and to assess ventricular function. However, the assessment of cardiac function may be difficult in the presence of fast ventricular rates. Invasive electrophysiological studies usually have a limited role in the routine evaluation of AF but may be needed in an occasional patient with electrophysiological abnormalities. If a patient with recent onset AF presents with signs or symptoms of a cerebrovascular event (stroke or transient ischaemic attack), a CT scan of the brain is useful in identifying patients with haemorrhagic stroke before starting anticoagulation.

Acute AF Patients who present with acute onset of AF should be treated with immediate cardioversion if AF is associated with significant haemodynamic compromise. Patients who present with acute-onset AF who are haemodynamically stable and asymptomatic, or who present later than 48 hours after the onset, can initially be treated with a strategy of heart rate control and anticoagulation, as an alternative to cardioversion, allowing time for assessment and evaluation of the patient.¹² In either case, effective anticoagulation in the form of heparin and/or warfarin should be initiated to decrease the risk of thromboembolic complications. Associated complications should be managed accordingly, for example, acute heart failure should be treated with diuretics.
Evaluation of the cause or precipitating factor for acute AF may be as important as treating the arrhythmia. For example, rendering the patient euthyroid in the context of thyrotoxic AF often results in spontaneous conversion to sinus rhythm in a high proportion of patients.¹³

Paroxysmal AF In paroxysmal AF, the objective of management is the suppression of paroxysms, heart rate control during paroxysms and the long-term maintenance of sinus rhythm. In view of this an appropriate drug such as a class I, class II or class III agent should be chosen, as well as appropriate use of antithrombotic therapy.¹⁴ Antithrombotic therapy should be considered in patients with paroxysmal AF in two circumstances: (i) prior to cardioversion, if such a patient has persisting AF; and (ii) as long-term therapy for prophylaxis against stroke and thromboembolism.
If a patient is experiencing only mild and infrequent symptoms it is advisable to try to avoid antiarrhythmic drug therapy if at all possible. General measures should always be considered, for example, the withdrawal of caffeine or alcohol, or stress counselling, especially if paroxysms of AF can be related to these. Occasionally, paroxysmal AF may be related to conditions which may benefit from atrial pacing, for example, the sick sinus syndrome.
Digoxin should be avoided in paroxysmal AF as the evidence suggests that this drug makes paroxysmal AF worse. In fact, clinical evidence has shown that paroxysms of AF occur more frequently, at faster heart rates and for significantly longer in patients receiving digoxin.^15,16 In the CRAFT study, digoxin reduced the frequency of symptomatic AF episodes, but the effect was small and may have been due to a reduction in the ventricular rate or irregularity rather than an antiarrhythmic action.¹⁷ Digoxin is also ineffective for the long-term maintenance of sinus rhythm or cardioversion of AF.
Verapamil is relatively ineffective in controlling paroxysmal AF, and its administration to patients with underlying Wolff-Parkinson-White syndrome who present with paroxysmal AF may occasionally lead to serious adverse effects, including ventricular fibrillation and severe haemodynamic impairment. Class I antiarrhythmics remain popular drugs for use in paroxysmal AF. For example, class Ic agents, such as flecainide, have been shown to be effective in preventing recurrences of paroxysmal AF in up to 60 per cent of patients.¹⁸ Propafenone, another class Ic compound, may also be effective in paroxysmal AF.¹⁹ Nevertheless, doubts on the safety of the class I agents have been raised by the Cardiac Arrhythmia Suppression Trial, in which patients (post myocardial infarction) given flecainide for ventricular arrhythmias had a worsened prognosis.²⁰
Sotalol, another commonly prescribed drug, combines both class II (beta-blockade) and class III (prolongation of repolarisation) antiarrhythmic effects, and is useful in paroxysmal AF. Another class III drug, amiodarone, is effective for the treatment of paroxysmal AF but its use has to be moderated by its potentially serious, albeit relatively rare, side effects.

Persistent AF and cardioversion Cardioversion of atrial fibrillation can be performed by pharmacological or electrical methods. Patients who undergo cardioversion for AF are at substantial risk of thromboembolism either due to dislodgement of already formed thrombus in the atrium or formation of new thrombus due to lack of effective mechanical atrial function, and require a course of anticoagulation. Current anticoagulation guidelines for cardioversion are discussed further in the third article of this series.
Electrical cardioversion was first introduced by Lown²¹ and is commonly used. Nevertheless, recent developments in the use of antiarrhythmic drugs have identified suitable drugs, especially class I and III antiarrhythmic agents. These are usually administered intravenously, especially for the cardioversion of acute AF, although oral loading is possible in occasional stable patients. The success rate for direct current (DC) cardioversion, using a synchronised electrical shock, is 80-90 per cent whereas the success rate for pharmacological cardioversion is in the range 40-90 per cent, depending on the type of agent used.^22,23 Internal cardioversion, where an intracardiac electrode is placed within the heart for cardioversion, is a useful technique for patients resistant to transthoracic cardioversion, although the procedure is limited to specialised centres.²⁴ The choice of the type of cardioversion depends on the speed with which cardioversion is warranted, underlying ischaemic heart disease, left ventricular function and the clinician's previous experience with either of the strategies.
Predictors of refractoriness to cardioversion or unsuccessful maintenance of sinus rhythm include the following: age, duration of arrhythmia, the presence of uncontrolled hypertension, structural heart disease (dilated impaired left ventricles, valve disease, etc) and other organic heart disease.²² Importantly, potential benefits of cardioversion of AF to sinus rhythm include improved cardiac haemodynamics at rest and exercise, improvement of symptoms and a possible reduction in the risk of stroke and the need for long-term anticoagulation.

Permanent AF and rate control In permanent AF, the effective management strategy should be heart rate control and appropriate use of antithrombotic therapy. Heart rate control can be achieved using pharmacological and non-pharmacological methods. The latter has seen many advances in the fields of electrophysiology and pacemakers.
The definition of adequate heart rate control in AF is still subject to much debate. Optimal heart rate control during AF is probably that at which cardiac output is most optimal. One definition for heart rate control in AF is when the ventricular response rate ranges from 60 to 80 bpm at rest and between 90 and 115 bpm during moderate exercise. However, the recent Royal College of Physicians of Edinburgh consensus conference on atrial fibrillation suggested that, for adequate heart rate control, the ventricular rate needs to be under 90 bpm at rest and under 180 bpm during exercise.²⁵
Various pharmacological agents are used for satisfactory ventricular rate control. Digoxin is the commonest drug for heart rate control in AF, acting primarily by vagotonic inhibition of atrioventricular nodal conduction. While effective for rate control at rest, digoxin is less effective for rate control than the beta-blockers or rate-limiting calcium channel blockers, and is less likely to control the ventricular rate during pyrexia, stress and exercise (when vagal tone is low and sympathetic tone is high).²⁶ Digoxin has little or no ability to terminate the arrhythmia, and often does not slow the heart rate with recurrent AF.
Beta-blockers, such as propranolol, atenolol, metoprolol or esmolol, can be valuable for heart rate control for AF in specific settings. Beta-blockers prevent the effects of sympathetic tone, decrease the resting heart rate and blunt the heart rate response to exercise, but may also reduce exercise tolerance. Beta-blockers are superior to the use of digoxin alone, but may be used in combination with digoxin and/or calcium channel blockers (not verapamil) in chronic AF.²⁷
Non-dihydropyridine calcium channel blockers, such as verapamil or diltiazem, both of which increase refractoriness and slow conduction in the AV node, are effective agents for heart rate control. These can be administered intravenously in the emergency setting but the negative inotropic effect, particularly with verapamil, requires their cautious use in heart failure. Both agents rarely cardiovert AF to sinus rhythm.
The class III agent amiodarone is effective in controlling the ventricular rate in AF and flutter, and should be considered, especially in patients intolerant of other agents. Amiodarone is also quite effective in the prevention of paroxysms of AF and the long-term maintenance of sinus rhythm after successful cardioversion. Rate control is achieved with amiodarone via its sympatholytic and calcium channel blocking properties, which depress atrio-ventricular conduction.
Class I agents are not usually advocated for heart rate control of chronic AF. Indeed, class Ia agents such as quinidine may even increase atrio-ventricular conduction due to their anticholinergic effects.

Antithrombotic therapy

Patients with permanent AF are at increased risk of stroke and thromboembolism and should thus be considered for antithrombotic therapy, in the absence of contraindications. In a recent meta-analysis of antithrombotic therapy in AF, the use of anticoagulation was shown to reduce the risk of stroke by 62 per cent (95 per cent confidence intervals [CI] 48-72 per cent), while aspirin reduces the risk by 22 per cent (95 per cent CI 2-38 per cent).²⁸
The risk of stroke and thromboembolism is not uniform and certain clinical and echocardiographic factors can identify patients with AF who are at high risk, who would benefit most from anticoagulation. By contrast, low risk patients with AF, which include patients with lone AF, can be treated with aspirin. This topic is considered in greater detail in the third article of this series.

Non-pharmacological measures

If pharmacological measures fail to prevent recurrence of AF and/or to control the ventricular rate, non-pharmacological strategies should be considered. These include pacemakers, electrophysiological techniques (atrioventricular node modification or ablation) and surgery. It is now well-established that atrial pacing in patients with sinus node disease confers a significantly lower risk for the development of AF, thromboembolism, heart failure and mortality when compared with ventricular pacing at eight years of follow-up, with a low incidence of progression to atrioventricular block.^29,30 In some patients with paroxysmal AF, especially those resistant to medical therapy, the atrial defibrillator (or "atrioverter") is a new device to treat paroxysms and to maintain sinus rhythm. Surgery for AF is often considered in resistant cases, alhough this is restricted to specialist units and is often performed in combination with associated valve surgery.

Future directions in AF management

Electrophysiological techniques Increasing developments in understanding the electrophysiology of AF opens the possibility of new techniques for managing AF. For example, the "maze" surgical procedure can be replicated using radiofrequency ablation, although the technique is tedious and still under evaluation. Atrioventricular node modification or ablation (with concomitant pacemaker implantation) has been shown to improve symptoms and quality of life, although some studies applying this technique report an increase in progression to persistent AF and, rarely, sudden death.^31,32 Newer pacing modalities, including new algorithms and multi-site atrial pacing hold promise for the future management of AF.

TEE guided cardioversion Transoesophageal echocardiography (TEE) has a high sensitivity and specificity in detecting left atrial thrombus.³³ TEE-guided cardioversion is a new strategy being evaluated in an attempt to simplify and minimise the anticoagulation regime required for cardioversion. The patient is heparinised and warfarin initiated (target INR 2.0 to 3.0), following which biplane or multiplane TEE is performed to assess left atrial size and to check for the presence of atrial thrombi or possible mitral stenosis. If no thrombus or other adverse features (spontaneous echocontrast, valve disease, etc) are seen, cardioversion is then performed, and warfarin continued for a minimum of four weeks. If thrombus is seen, severe spontaneous echocontrast is present, the left atrium cannot be adequately evaluated for technical reasons, or TEE is contraindicated, patients receive four weeks of warfarin prior to elective cardioversion, as per conventional management (see above). Anticoagulation with warfarin should be continued for at least four weeks after cardioversion in all patients, regardless of the cardioversion method.

Cardioversion vs rate control Cardioversion of AF has been shown to improve cardiac haemodynamics and quality of life in the short term.²² Nevertheless, there is a paucity of data comparing whether a strategy of rate control is superior to a strategy of aggressive cardioversion ("rhythm control") with respect to long-term mortality and morbidity. A number of prospective trials are being carried out to answer this question.

Provision of thromboprophylaxis While many randomised controlled trials have firmly established the role of antithrombotic therapy in AF, current provision of anticoagulation for such patients is suboptimal. Many clinicians are still reluctant to prescribe warfarin due to the risks of bleeding and the inconvenience of attendances at anticoagulation clinics to monitor treatment.
New antithrombotic drugs may be the answer, providing safe and effective thromboprophylaxis without the need for constant monitoring.
However, risk stratification is also needed to identify the patients at highest risk for stroke and thromboembolism who can be targeted for anticoagulation. Based on current evidence, most risk stratification can be performed using clinical and echocardiographic risk factors. Further refinement of risk stratification may be obtained by TEE³⁴ and, possibly, measurement of indices of thrombogenesis.³⁵

Conclusion

Patients with AF are at considerable risk of morbidity and mortality. This arrhythmia needs to be managed effectively to avoid adverse events, with individualised treatment and consideration of management objectives.

Dr Lip is consultant cardiologist and reader in medicine and Dr Kamath is research fellow in the Haemostasis, Thrombosis and Vascular Biology Unit, University Department of Medicine,City Hospital, Birmingham