CHAPTER
10
CHRONIC ATRIAL
FIBRILLATION
Mathew Maurer,
M.D.
Chronic atrial fibrillation (AF) is the most common sustained cardiac
arrhythmia. Incidence rises markedly with age,[i]
and AF is seen in five percent of patients over the age of 69.[ii],[iii]
There are an estimated 2.2 million people in the United States with AF, with a
median age of about 75 years.[iv]
Patients with AF are at increased risk of stroke and of death. In the Framingham
cohort study, the risk of stroke in patients with nonvalvular (nonrheumatic) AF
was 5.6 times greater than in comparably aged patients in sinus rhythm; in
patients with AF and rheumatic valvular disease it was 17 times higher.[v] AF is associated with
reduced life expectancy with an approximate doubling of all-cause mortality.2,[vi]
Recent analysis of the SOLVD trials found that the presence of atrial
fibrillation in patients with asymptomatic and symptomatic left ventricular
systolic dysfunction is associated with an increased risk for all-cause
mortality [relative risk 1.34].[vii]
Treatment of patients with AF has two principle objectives. One is to relieve
symptoms through restoration of sinus rhythm or control of ventricular rate; the
other is to use prophylactic therapy to reduce the risk of stroke.[viii],[ix]
Initial Evaluation
The initial
evaluation of a patient with atrial fibrillation should include (1) historical
information with attention to the presence of concurrent symptoms of congestive
heart failure, angina and palpitations, previous thromboembolic events, and a
history of hypertension, diabetes, congestive heart failure or valvular heart
disease (2) physical examination (3) electrocardiogram (4) initial laboratory
examination including an electrolyte panel, coagulation profile and TSH and (5)
transthoracic echocardiogram to assess valve function and left atrial size.[x]
Concurrent
symptoms:
AF may present as an
asymptomatic arrhythmia or, at the other extreme, in a patient who is severely
symptomatic with hemodynamic compromise. The previous focus of published
literature on antithrombotic therapy for AF has been on demonstrating its value
in decreasing stroke and death. The effects of individual symptomatic
treatments, including rate control medicines, atrioventricular node ablation,
arrhythmia termination and prevention therapies (cardioversion, antiarrhythmic
medications or operation) on morbidity and mortality are unknown.[xi],[xii]
In the absence of clinical trial data on optimal therapy for arrhythmia related symptoms,
assessment of the setting and severity of such symptoms is critical for
determining the appropriateness of therapy for an individual patient.
Symptoms usually
consist of palpitations and lightheadedness. The loss of synchronous atrial
contraction and decreased diastolic filling time during tachycardia can lead to
a hemodynamically significant decrease in cardiac output in patients with left
ventricular dysfunction, valvular or coronary heart disease. Angina pectoris may
be precipitated as a result of increased myocardial oxygen demand and shortened
diastolic coronary blood flow. In SPAF-III, among 1,411 patients, atrial
fibrillation was initially asymptomatic in 44 percent, while 32 percent had
palpitations, 10 percent had congestive heart failure, 2 percent presented with
a stroke, and 2 percent with syncope. In 11 percent, the presenting symptom was
unknown.[xiii]
While these study subjects were not necessarily representative of the general
population with AF, the data support the conclusion that symptoms from AF are
usually not disabling.
Physical
exam:
Physical findings in patients with AF include an irregularly irregular
ventricular rhythm, a variation in intensity of the first heart sound and
absence of a waves in the jugular
venous pulse. At faster ventricular rates, the auscultated apical rate may
exceed the palpable radial rate, owing to failure of many of the ventricular
contractions to generate a palpable peripheral pulse. Since AF is associated
with underlying heart disease and heart failure, the physical examination should
focus on the presence of murmurs and gallops. Occult or manifest thyrotoxicosis
should be considered in patients with recent onset AF and physical examination
should include careful assessment of the thyroid.
Labs:
Electrolyte abnormalities may precipitate or exacerbate an underlying
tendency towards AF and since correction of these abnormalities is simple and
effective, routine evaluation is indicated. A screening TSH may identify
hyperthyroidism as the etiology of AF. This can be particularly important in the
elderly, in whom clinical symptoms and signs of hyperthyroidism may be absent.[xiv]
In one large survey of patients who presented with AF, occult thyroid disease
(especially hyperthyroidism) has been identified as a principle underlying cause
in 3.5 percent.[xv]
Thus, routine thyroid function tests, while having a low yield,[xvi]
may identify a potentially treatable cause of the dysrhythmia and should be
performed. Finally, before anticoagulation is begun, knowledge of the
patient’s coagulation status is important in judging response to therapy and
in preventing complications.
EKG:
The twelve lead EKG confirms the diagnosis of AF and provides information about underlying heart disease and coexisting conduction disease; it should be performed in all patients with AF.
Transthoracic
Echocardiogram:
A transthoracic echocardiogram provides anatomic and functional information about underlying valvular heart disease, the presence and degree of left ventricular hypertrophy and an assessment of left ventricular function. Left atrial size may predict the outcome of cardioversion and subsequent maintenance of sinus rhythm.[xvii],[xviii] Echocardiographic data are also useful in making an assessment of embolic risk. Moderate to severe left ventricular systolic dysfunction, as demonstrated by two-dimensional echocardiography, is a strong predictor of stroke risk [RR 2.5], independent of clinical risk factors.[xix] Additionally, normal cardiac function implies that the risk of systemic embolism is low and long-term anticoagulation may not be required (see Anticoagulation, below).
Rate Control
Control of
resting and exercise heart rate during AF is an important therapeutic goal,
which may markedly reduce symptomatology. Restoration and maintenance of sinus
rhythm may not be indicated or possible in many patients, in whom rate control
and anticoagulation are essential. Patients with AF should have their heart
rates reduced to physiologically normal values. In the middle aged population,
this might be a resting heart rate < 90 beats/min and an exercise heart rate
< 140 beats/min. Chronic rate control can be achieved with several different
classes of agents, none of which is clearly preferable, and patient symptoms and
co-morbid conditions should be taken into account when selecting an agent for
rate control.
Digoxin:
Digoxin has been used for over a century to acutely reduce the heart rate
in AF. Numerous studies have demonstrated, however, that the predominant effect
of digoxin is on the resting heart rate and that its mechanism of action is by
activating the parasympathetic nervous system, causing sinus slowing and AV node
inhibition.[xx] Despite controlled resting
heart rates, patients on digoxin can have very rapid responses during exertion
or with paroxysms of fibrillation when sympathetic tone is high. In fact, there
is very little difference in exercise heart rates with or without digoxin
therapy,[xxi]
and increasing digoxin dosage (and blood levels) does not seem to add any
benefit with respect to nonresting rate control. Sedentary patients may be well
controlled on digoxin alone, but in many patients the drug is ineffective in
controlling heart rate, particularly during exercise.
Verapamil
and Diltiazem:
Calcium channel blockers have been shown to reduce resting and exercise
heart rates in AF. In a placebo-controlled crossover trial of verapamil, Lang
demonstrated reduction in exercise and resting heart rates and improved exercise
capacity.[xxii]
Several uncontrolled studies have shown improved heart rate response at rest and
with exercise using diltiazem. High dose calcium channel blockers, with or
without digoxin, may cause prolonged pauses, which may become symptomatic. Drug
combinations at lower doses may effect adequate rate control without symptomatic
bradycardia. These agents should be used with caution in the presence of heart
failure. Diltiazem should be preferred to verapamil in patients with mild to
moderate left ventricular dysfunction.[xxiii]
Because of concern about the cardiovascular toxicity of short-acting calcium
channel blockers, long-acting preparations should be used.
These are the preferred agents for rate control in patients with
symptomatic obstructive lung disease.
Beta
blockers:
Several controlled and uncontrolled trials have shown that beta blockers
effectively reduce exercise heart rates in patients with AF.[xxiv],[xxv]
In a randomized, double-blind crossover trial, Dibianco demonstrated that
adding naldolol to digoxin effectively controlled exercise heart rate in
patients with AF.[xxvi] No study has
conclusively demonstrated improved exercise capacity with beta blockers. As with
calcium channel blockers, symptomatic bradycardia is a concern, and these agents
should be used with caution in the presence of heart failure.
Amiodarone:
The use of amiodarone solely for the chronic management of rapid ventricular response in AF has not been evaluated in a controlled trial. However, in several trials evaluating patients with refractory or resistant AF, a consistent and significant slowing of ventricular response has been uniformly reported. While amiodarone is clearly efficacious in maintaining sinus rhythm in patients with AF, especially if they are resistant to other therapies, its toxicity precludes recommending it simply for rate control when other less toxic agents are available. Nonetheless, in patients with left ventricular dysfunction who cannot tolerate standard therapy for rate control, amiodarone may be beneficial.
Catheter
ablation therapy:
Ablation of the AV node or His bundle using radiofrequency energy delivered by catheter and insertion of a permanent rate responsive pacemaker is an alternative to medication.[xxvii] Since the procedure has not been studied in a controlled trial, and since it creates permanent heart block, it should be considered only for patients who have failed all other forms of medical therapy.
Anticoagulation
Prospective
trials have confirmed that the risk of stroke in patients with nonvalvular AF is
large - about five percent per year. More importantly, these trials have
convincingly demonstrated that low dose warfarin therapy can substantially
reduce stroke with a minimal risk of significant hemorrhage in properly chosen
patients.[xxviii],[xxix],[xxx],[xxxi],[xxxii]
In these five randomized trials, warfarin consistently decreased the risk of
stroke in patients with AF - a 68 percent reduction in risk - with virtually no
increased incidence of major bleeding. Several
recent trials have demonstrated that adjusted dose warfarin [i.e. to maintain an
INR 2.0 – 3.0] is superior to fixed mini-dose warfarin [1.25 mg/day] and the
combination of fixed mini-dose warfarin plus aspirin.[xxxiii],[xxxiv],[xxxv]
In spite of this documented efficacy, warfarin is underused in the
prevention of thromboembolism among patients with chronic atrial fibrillation,[xxxvi],[xxxvii],[xxxviii]
especially the elderly.[xxxix],[xl]
Pooled analysis
of the five studies demonstrates four clinical features which independently
predict higher stroke risk: previous transient ischemic attack, thromboembolism
or stroke (RR 2.5), increasing age (RR = 1.4 for each decade), history of
hypertension (RR=1.6) and diabetes (RR=1.7).[xli]
Those at highest risk were patients with a prior history of stroke or TIA, who
had an annual stroke rate of 12 percent - which fell to 5.1 percent per year
with warfarin. Fifteen percent of the patients in these trials had none of these
risk factors and their risk of stroke receiving no anticoagulant therapy was 1.0
percent per year. Although not significant in the multivariate analysis, the
presence of either congestive heart failure or clinical coronary artery disease
did increase the risk of stroke. Patients with either of these cardiac disorders
had stroke rates approximately three times higher than patients without any of
the risk factors.
While echocardiographic features that predict stroke were not included in the pooled analysis, the SPAF trial found that left atrial size and left ventricular dysfunction measured on two dimensional echocardiography were associated with increased stroke risk.[xlii] In addition, mitral annular calcification was associated with increased stroke risk in the BAATAF study.[xliii] Transesophageal echocardiography may be more sensitive for identifying predictors of stroke, but its clinical importance in risk stratification of patients with atrial fibrillation awaits confirmation.
(1)
Previous thromboembolic stroke, TIA, or other arterial thromboembolism?
YES ---> high risk
(2) Age > 75 and history of diabetes or
hypertension? YES ---> high risk
(3) Clinical evidence of heart valve disease
or prosthesis? YES ---> high risk
(4) Planned cardioversion? YES ---> high
risk
(5) Thyrotoxicosis? YES ---> high
risk
(6) Clinical evidence of heart failure? YES
---> high risk
(7)
Echocardiographic findings? (intracardiac thrombus, LA > 4.5 cm, impaired
LV function, valvular
disease including mitral annular calcification) YES ---> high
risk
(8)
None of the above, but age > 65 or age < 65 with history of diabetes or
hypertension, ischemic heart disease or peripheral arterial disease --->
moderate risk
(9) None of the above ---> low
risk
Figure
1 outlines an algorithm for anticoagulation in patients with AF.
Risk-stratification using the data outlined above allows recommendations for
warfarin, aspirin or no antithrombotic prophylaxis:
High/Moderate RiskIf warfarin not contraindicated, treat with warfarin. If contraindicated, treat with aspirin 75 - 325 mg/day. Target INR = 2.0 - 3.0 |
Low riskIf aspirin not contraindicated, treat with aspirin 75-325 mg/day. |
Low risk:
Patients with “lone AF” - those under 60 with no history of previous thromboembolic event, diabetes, hypertension, valvular disease or prosthesis, heart failure, thyrotoxicosis, ischemic heart disease, or echocardiographic risk factors (impaired LV function, valvular disease or intracardiac thrombus) - have a very low risk of stroke and routine anticoagulation may not be warranted.[xliv] However, many clinicians use aspirin as prophylaxis in these patients.11
Warfarin (INR
2-3):
Advancing age and more
intense anticoagulation increases the risk of major hemorrhage in patients given
warfarin for stroke prevention. The present recommendation in all patients with
chronic AF is to give warfarin therapy in doses that prolong the INR to 2.0-3.0.[xlv]
Patients with recent thromboembolic events may benefit from a slightly higher
INR and some investigators recommend a target INR of 3.0.[xlvi]
Aspirin (325
mg):
Two of the primary prevention trials[xlvii],[xlviii]and
a secondary prevention study[xlix]
investigated whether antiplatelet therapy with aspirin reduced the risk of
stroke compared to placebo. Taken overall, the studies suggest that aspirin, at
325 mg, prevents stroke less effectively than warfarin but that it is also less
likely to cause major hemorrhage. A more recent primary prevention study
directly compared aspirin 325 mg daily with warfarin. Compared with aspirin,
warfarin reduced the incidence of ischemic stroke by about 30 percent. However,
aspirin-treated patients under 75 with no risk factors other than AF had a low
rate of thromboembolism (0.5 percent annually) and treatment with warfarin was
no more effective.[l]
The recent SPAF III trial has shown that patients with AF categorized as
"low risk" based on the absence of four prespecified thromboembolic
risk factors:
(1) recent congestive heart failure or left ventricular fractional shortening of
25% or less, (2) previous thromboembolism, (3) systolic blood pressure greater
than 160 mm Hg, or (4) female gender at age older than 75 years have relatively
low rates of ischemic stroke during treatment with aspirin.[li]
Thus, in patients at low risk for stroke, warfarin may confer no benefit in
terms of absolute risk and aspirin therapy may be beneficial.11
Cardioversion
The goals of
therapy in patients with atrial fibrillation are directed at controlling the
patient’s symptoms and reducing the risk of thromboembolic complications.
Conversion of AF to sinus rhythm will directly accomplish the first goal over
the short term and theoretically will accomplish the second goal if sinus rhythm
can be maintained. Conversion of AF to sinus rhythm is associated with increased
cardiac output and improved exercise capacity. An attempt at cardioversion
should be considered in all patients with chronic AF and electrical
cardioversion is the preferred method with which to reestablish sinus rhythm.[lii]
For patients in chronic atrial fibrillation, anticoagulation with
warfarin for 3 to 4 weeks prior to cardioversion is required to reduce
thromboembolic risk. Among patients presenting with atrial fibrillation that was
clinically estimated to have lasted less than 48 hours [i.e. acute atrial
fibrillation], the likelihood of cardioversion-related clinical thromboembolism
is low [0.8%] and early cardioversion in these patients is safe.[liii]
Cardioversion
can be achieved with antiarrhythmic drug therapy, electrical cardioversion or
both. To date, there have been no studies comparing the relative benefits of
maintenance of sinus rhythm with antiarrhythmic drugs to the strategy of rate
control and anticoagulation in patients with AF. There are no data to prove that
maintenance of sinus rhythm with antiarrhythmic agents reduces thromboembolic
risk. In addition, there is increasing concern that some antiarrhythmic drugs
may confer a substantial risk of proarrhythmia in patients with AF.
Initiation of antiarrhythmic therapy is associated with a significant
risk of adverse events. Therefore, observation with electrocardiographic monitoring
is advisable for the first 24 – 48 hours, especially in the elderly or those
with previous myocardial infarction.[liv]
Thus, the benefits of chronic drug therapy need to be weighed against
each patient’s individual risks and the decision to institute therapy should
be strongly guided by the patient’s symptoms.
[i] Falk RH. Etiology and complications of atrial fibrillation: insights from pathology studies. Am J Cardiol 1998;82:10N-17M.
[ii] The National Heart Lung and Blood Institute Working Group on Atrial Fibrillation: Atrial fibrillation: current understandings and research imperatives. J Am Coll Cardiol 1993;22:1830.
[iii] Ryder KM, Benjamin EJ. Epidemiology and significance of atrial fibrillation. Am J Cardiol 1999;84:131R-138R.
[iv] Feinberg WM, Blackshear JL, Laupacis A, et al. Prevalence, age distribution, and gender of patients with atrial fibrillation. Analysis and implications. Arch Intern Med 1995;155(5):469-73.
[v] Kannel WB, Abbot RD, Savage DD et al. Epidemiologic features of chronic atrial fibrillation: the Framingham study. N Engl J Med 1982;306:1018-22.
[vi] Gajewski J and Singer RB. Mortality in an insured population with atrial fibrillation. JAMA 1981;245:1540-44.
[vii] Dries DL, Exner DV, Gersh BJ, et al. Atrial fibrillation is associated with an increased risk for
mortality and heart failure progression in patients with asymptomatic and symptomatic left ventricular systolic dysfunction: a retrospective analysis of the SOLVD trials. Studies of Left Ventricular Dysfunction.
J Am Coll Cardiol 1998 32(3):695-703
[viii] Levy S, Breithardt G, Campbell RW et al. Atrial fibrillation: current knowledge and recommendations for management. Working Group on Arrhythmias of the European Society of Cardiology. Eur Heart J 1998;19:1294-320.
[ix] Hart RG, Sherman DG, Easton JD et al. Prevention of stroke in patients with nonvalvular atrial fibrillation. Neurology 1998;51:674-81.
[x] Gillis AM, Klein GJ and MacDonald RG. Investigation of the patient with atrial fibrillation. Can J Cardiol 1996;12[suppl A]:12A-13A
[xi] National Heart, Lung and Blood Institute Working Group on Atrial Fibrillation. Atrial Fibrillation: current understandings and research imperatives. JACC 1993;22:1830-34.
[xii] Pritchett ELC. Management of atrial fibrillation. NEJM 1992;326:1264-71.
[xiii] Blackshear, JL, Kopecky SL, Litin SC et al. Management of atrial fibrillation: prevention of thromboembolism and symptomatic treatment. Mayo Clin Proc 1996;71:150-60.
[xiv] Koutras DA. Subclinical hyperthyroidism. Thyroid 1999;9:311-15.
[xv] Lip GY, Team KN and Dunn FG. Treatment of atrial fibrillation in a district general hospital. Br Heart J 1994;71:92-5.
[xvi] Krahn AD, Klein GJ, Kerr CR, et al. How useful is thyroid function testing in patients with recent-onset
atrial fibrillation? The Canadian Registry of Atrial Fibrillation Investigators. Arch Intern Med 1996;156(19):2221-4.
[xvii] Dethy M, Chassat C, Roy D, et al. Doppler echocardiographic predictors of recurrence of atrial fibrillation after cardioversion. Am J Cardiol 1988;62:723-6.
[xviii] Dittrich HC, Erikson JS, Scheiderman T, et al Echocardiographic and clinical predictors of outcome of elective cardioversion of atrial fibrillation. Am J Cardiol 1989;63:193-7.
[xix] Echocardiographic predictors of stroke in patients with atrial fibrillation: a prospective study of 1066 patients from 3 clinical trials. Arch Intern Med 1998;158(12):1316-20.
[xx] Gold H, Kwit NT, Otto H et al. On vagal and extravagal factors in cardiac slowing by digitalis in patients with atrial fibrillation. J Clin Invest 1939;18:429.
[xxi] Beasley R, Smith DA and McHaffie DJ. Exercise heart rates at different serum digoxin concentrations in patients with atrial fibrillation. Br Med J Clin Res 1985;290:9-11.
[xxii] Lang R, Klein HO, DiSegni E et al. Verapamil improves exercise capacity in chronic atrial fibrillation: double blind crossover study. Am Heart J 1983;105:820-5.
[xxiii] Heywood JT, Graham B, Marais GE, et al. Effect of intravenous diltiazem on rapid atrial fibrillation accompanied by congestive heart failure. Am J Cardiol 1991;67:1150-52.
[xxiv] David D, DiSegni E, Klein HO, et al. Inefficacy of digitalis in the control of heart rate in patients with chronic atrial fibrillation: beneficial effect of an added beta adrenergic blocking agent. Am J Cardiol 1979;44:1378-82.
[xxv] Matsuda M, Matsuda Y, Tamagishi T, et al. Effects of digoxin, propranolol and verapamil on exercise in patients with chronic isolated atrial fibrillation. Cardiovasc Res 1991;25:453-57.
[xxvi] DiBianco R, Morganroth J, Freitag JA, et al. Effects of naldolol on the spontaneous and exercise-provoked heart rate of patients with chronic atrial fibrillation receiving stable doses of digoxin. Am Heart J 1984;108;1121-27.
[xxvii] Fleck RP, Chen PS, Boyce K, et al. Radiofrequency modification of atrioventricular conduction by selective ablation of the low posterior septal right atrium in a patient with atrial fibrillation and rapid ventricular response. PACE 1993;16:377-381.
[xxviii] Ezekowitz MD, Bridgers SL, James KE, et al. Warfarin in the prevention of stroke with nonrheumatic atrial fibrillation. N Engl J Med 1992;327:1407-12.
[xxix]Peterson P, Boysen G, Godtfredsen J, et al. Placebo-controlled, randomized trial of warfarin and aspirin for prevention of thromboembolic complications in chronic atrial fibrillation: the Copenhagen AFASAK study. Lancet 1989;1:175-78.
[xxx] The Boston Area Anticoagulation Trial for Atrial Fibrillation Investigators. The effect of low-dose warfarin on the risk of stroke in patients with nonrheumatic atrial fibrillation. N Engl J Med 1990;323:1505.
[xxxi] Stroke Prevention in Atrial Fibrillation Investigators. Stroke prevention in atrial fibrillation study: final results. Circulation 1991;84:527.
[xxxii] Connolly SJ, Laupacis A, Gent M, et al. Canadian Atrial Fibrillation Anticoagulation (CAFA) Study. JACC 1991;18:349.
[xxxiii] Gullov AL, Koefoed BG, Petersen P, et al. Fixed minidose warfarin and aspirin alone and in combination vs. adjusted-dose warfarin for stroke prevention in atrial fibrillation: Second Copenhagen Atrial Fibrillation, Aspirin, and Anticoagulation Study. Arch Intern Med 1998 158(14):1513-21
[xxxiv] Gallus AS. Towards the safer use of warfarin I; an overview. J Qual Clin Pract 1999;19:55-59.
[xxxv] Stroke Prevention in Atrial Fibrillation III randomized clinical trial Adjusted-dose warfarin versus low-intensity, fixed-dose warfarin plus aspirin for high-risk patients with atrial fibrillation. Lancet 1996;348(9028):633-8/
[xxxvi] Stafford RS and Singer DE. Recent national patterns of warfarin use in atrial fibrillation. Circulation 1998;97(13):1231-3.
[xxxvii] Whittle J, Wickenheiser L and Venditti LN. Is warfarin underused in the treatment of elderly persons with atrial fibrillation? Arch Intern Med 1997;157(4):441-5.
[xxxviii] Albers GW, Yim JM, Belew KM, et al. Status of antithrombotic therapy for patients with atrial
fibrillation in university hospitals Arch Intern Med 1996;156(20):2311-6.
[xxxix] McCrory DC, Matchar DB, Samsa G, et al. Physician attitudes about anticoagulation for nonvalvular atrial fibrillation in the elderly. Arch Intern Med 1995;155(3):277-81.
[xl] Tong DC, Albers GW. Antithrombotic management of atrial fibrillation for stroke prevention in older people. Clin Geriatr Med 1999;15:645-62.
[xli] Atrial Fibrillation Investigators. Risk factors for stroke and efficacy of anti-thrombotic therapy in atrial fibrillation: analysis of pooled data from five randomized controlled trials. Arch Int Med 1994;154:1449-57.
[xlii] The Stroke Prevention in Atrial Fibrillation Investigators. Predictors of thromboembolism in atrial fibrillation II: Echocardiographic features of patients at risk. Ann Int Med 1992;116:6-12.
[xliii] The Boston Area Anticoagulation Trial for Atrial Fibrillation Investigators. The effect of low-dose warfarin on the risk of stroke in patients with nonrheumatic atrial fibrillation. N Engl J Med 1990;323:1505.
[xliv] Kopecky SI, Gersh BJ, McGoon MD, et al. The natural history of lone atrial fibrillation. A population based study over three decades. N Engl J Med 1987;317:669-674.
[xlv] Hylek EM, Skates SJ, Sheehan MA, et al. An analysis of the lowest effective intensity of prophylactic
anticoagulation for patients with nonrheumatic atrial fibrillation N Engl J Med 1996;335(8):540-6.
[xlvi] The European Atrial Fibrillation Trial Study Group. Optimal oral anticoagulant therapy in patients with nonrheumatic atrial fibrillation and recent cerebral ischemia. N Engl J Med 1995;333(1):5-10.
[xlvii] The Boston Area Anticoagulation Trial for Atrial Fibrillation Investigators. The effect of low-dose warfarin on the risk of stroke in patients with nonrheumatic atrial fibrillation. N Engl J Med 1990;323:1505.
[xlviii] Peterson P, Boysen G, Godtfredsen J, et al. Placebo-controlled, randomized trial of warfarin and aspirin for prevention of thromboembolic complications in chronic atrial fibrillation: the Copenhagen AFASAK study. Lancet 1989;1:175-78.
[xlix] EAFT (European Atrial Fibrillation Trial) Study Group. Secondary prevention in nonrheumatic atrial fibrillation after transient ischemic attack or minor stroke. Lancet 1993;342:1255-62.
[l] Warfarin versus aspirin for prevention of thromboembolism in atrial fibrillation. Stroke Prevention in Atrial Fibrillation II Study. Lancet 1994;343:687-691.
[li] Patients with nonvalvular atrial fibrillation at low risk of stroke during treatment with aspirin: Stroke Prevention in Atrial Fibrillation III Study. The SPAF III Writing Committee for the Stroke Prevention in Atrial Fibrillation Investigators JAMA 1998;279(16):1273-7.
[lii] Golzari H, Cebul RD and Bahler RC. Atrial fibrillation: restoration and maintenance of sinus rhythm and
indications for anticoagulation therapy. Ann Intern Med 1996;125(4):311-23.
[liii] Weigner MJ, Caulfield TA, Danias PG, et al. Risk for clinical thromboembolism associated with conversion to sinus rhythm in patients with atrial fibrillation lasting less than 48 hours. Ann Intern Med 1997;126(8):615-20.
[liv] Maisel WH, Kuntz KM, Reimold SC, et al. Risk of initiating antiarrhythmic drug therapy for atrial fibrillation in patients admitted to a university hospital. Ann Intern Med 1997;127(4):281-4.