Review: atrial fibrillation and stroke prevention.
Cerebrovascular disease is a common and potentially preventable cause of morbidity and mortality in elderly people. The annual incidence of stroke in the United Kingdom is approximately 195/100000, with the majority occurring in patients aged over 65 years. Atrial fibrillation (AF) is well recognized as an important and independent risk factor for stroke in elderly people[2, 3]. However, despite the results of three recent large randomized trials which demonstrated the efficacy of oral anticoagulant therapy in the primary prevention of stroke in individuals with AF[4-6] many doctors remain reluctant to prescribe warfarin for their older patients. In this review, we describe the epidemiology and pathogenesis of cerebrovascular disease in elderly subjects with AF and discuss the risks and benefits for stroke prophylaxis using anticoagulant therapy.
Epidemiology of cerebral infarction
in atrial fibrillation
AF is the most frequent disturbance of cardiac rhythm after multiple ectopic beats in the elderly. The prevalence of AF in the community rises with increasing age, from 3.3% in men and 1.7% in women aged 65-69 years to 13.3% in men and 8.1% of women aged over 85 in a survey of a Northern English community . Most cases of AF in older individuals are now of non-rheumatic (nonvalvular; NVAF) aetiology. It is well established that chronic AF is associated with an increased risk of peripheral arterial embolism, particularly in the cerebral hemispheres[10, 11]. In a retrospective review of 150 patients with AF, Sherman found that 31% had experienced a stroke or peripheral arterial embolism. Approximately 15-20% of all ischaemic strokes are associated with AF, this proportion rising to 30% in the elderly. The risk of a cardiogenic embolus is greatest in individuals with AF secondary to rheumatic heart disease; however, it is now widely accepted that NVAF is a significant risk factor for ischaemic stroke. In a 30-year follow-up of the Framingham study cohort, chronic NVAF resulted in a five-fold increased risk of first stroke when compared with sinus rhythm. This predisposition to stroke in individuals with NVAF was independent of concomitant cardiac failure and ischaemic heart disease often found in these patients. Indeed in subjects aged over 80 years, NVAF was the only cardiovascular risk factor which continued to have any effect on the incidence of stroke. The attributable risk of stroke increases significantly with age from 1.5% for subjects aged 50-59 years to 23.5% for those aged 80-89 years. Chronic NVAF is also associated with an appreciable risk of recurrent stroke disease. In a study of 59 patients aged around 70 years with NVAF who survived their initial stroke, the annual rate of recurrence remained at 20% throughout a 9-year follow-up period, the annual recurrence rate for patients in sinus rhythm being approximately 10-15%.
NVAF should be distinguished from |lone' AF; defined as AF occurring in the absence of detectable underlying cardiac disease. Lone AF in patients aged under 60 years at diagnosis is associated with a low but significant risk of embolic stroke. Analysis of the Framingham data confirmed a four-fold increased rate of stroke in patients with lone AF when compared with controls in sinus rhythm.
NVAF is a common finding in elderly subjects with clinically apparent or manifest cerebrovascular disease. However, patients with NVAF are also at increased risk of silent cerebral infarction. Petersen in a study of 29 subjects with predominantly NVAF found that although the proportion of patients with AF who had low-density areas ion CT scan was not significantly greater than controls in sinus rhythm, the total number of these silent infarcts was significantly higher in the AF group. In another retrospective study of stroke patients with AF, 13% had CT evidence of silent infarcts compared with 4% of controls in sinus rhythm; almost half of the affected AF patients had multiple infarcts. In both of these studies, the silent infarcts seen in subjects with AF tended to be peripheral and consistent with embolism. Analysis of the CT brain scans of patients with NVAF enrolled in the Stroke Prevention in Atrial Fibrillation (SPAF) study confirmed that for patients with AF age was the single most important predictor of silent infarction . In subjects aged > 65 years, the prevalence of silent infarcts was 24%, compared with 11% in those < 65 years. The clinical relevance of asymptomatic cerebral infarction in elderly subjects with NVAF is not clear. However, there is some evidence from autopsy data that AF is the most important underlying cardiac condition in patients with multi-infarct dementia. Recent clinical studies have demonstrated evidence of subclinical cognitive impairment in patients with cerebral infarction. It is conceivable therefore that recurrent silent cerebral infarction in older patients with NVAF will also result in cognitive impairment either clinical or subclinical.
The risk of thromboembolism associated with AF is greatest when the arrhythmia first develops or becomes sustained in nature. Although AF is a common electrocardiographic diagnosis in patients admitted to hospital following a stroke, being present in almost one-quarter [15, 25], it is not always certain whether this is a cause or effect relationship. Detailed analysis of the Framingham study data has however confirmed a clustering of stroke events in the months following the onset of VAF. Similarly there is an apparent increase in the incidence of stroke in older individuals following the transition from intermittent (paroxysmal) to sustained AF. Paroxysmal AF accounts for approximately one-third of cases of AF in elderly hospital inpatients. In a prolonged follow-up of 426 predominantly elderly patients with paroxysmal AF, the arrhythmia became chronic in one-third, with a median time to transition of 34 months. The annual incidence of embolic complications in patients with paroxysmal AF was 2%. This rose to 13.3% in the first year following transition to chronic AF, the majority of emboli being cerebral in distribution. Transition from paroxysmal to chronic AF and subsequent embolism was more common in patients with underlying rheumatic heart disease or previous embolic episodes. However, a substantial number of individuals with NVAF (16%) also suffered embolic complications.
Pathogenesis of cerebral infarction in
Any therapeutic intervention designed to reduce the risk of clinical or silent cerebral infarction in patients with NVAF must reflect the underlying pathogenesis. Not all strokes in patients with NVAF are cardioembolic in origin; estimates of the proportion of embolic strokes range from 19 to 75%. Mechanisms of cerebral ischaemia in NVAF include embolus from stasis-related thrombi in the left atrium or left ventricle, embolus from atheroma in the ascending aorta or co-existent cerebrovascular disease. Echocardiographic evidence of left atrial enlargement in patients with NVAF is certainly associated with an enhanced risk of stroke or silent cerebral infarction[29, 30]. Intracardiac thrombus is poorly visualized by conventional echocardiographic methods, therefore non-visualization of intracardiac thrombi does not preclude embolic stroke. Detection of intracardiac thrombi may be improved by the use of invasive trans-oesophageal echocardiography. Embolic strokes tend to be large cortical lesions which may be clinically devastating and not always preceded by transient ischaemic attacks (TIAs). The predominant CT scan finding in patients with NVAF is the presence of cortical infarcts, lacunar infarcts being relatively uncommon [20, 21]. In the Oxfordshire Community Stroke Project, AF was significantly less common in patients with lacunar infarction than in those with cortical lesions. Furthermore, Hankey and Warlow recently demonstrated that lacunar TIAs were not associated with a potential cardiac source of embolus. Ultrasonography studies of the carotid arteries in patients with NVAF and stroke have demonstrated a low prevalence of ipsilateral carotid stenosis . The relevance of this observation to the pathogenesis of lacunar infarction in NVAF remains to be determined. However the presence of such deep infarcts on the CT scans of some subjects with NVAF suggests that factors other than embolism may be important in the pathogenesis of stroke in a proportion of cases. Other risk factors which may contribute to the occurrence of stroke in NVAF include cardiac failure and hypertension although the relative importance of such co-existent conditions is uncertain due to the conflicting results from different studies.
As NVAF is associated with recurrent stroke and subclinical cerebral infarction, medical therapy should be directed towards the prevention of such cerebrovascular events. In theory, stasis-related thrombi, e.g. in the left atrium, could be prevented by anticoagulation with warfarin. In cerebrovascular disease and aortic atherosclerosis, where emboli may arise from platelet-fibrin thrombi, platelet inhibition with aspirin is more appropriate. Knowledge of the underlying pathogenesis is therefore important in deciding upon the optimum anti-thrombotic treatment.
Anticoagulation, aspirin and atrial
It is well established that anticoagulation is of value in the prevention of cardioembolic stroke in patients with atrial fibrillation associated with rheumatic heart disease or prosthetic valves[36, 37]. For the majority of patients presenting with non-rheumatic, NVAF, the role of anticoagulation or aspirin in stroke prevention has until recently been unclear.
Since 1989 three major clinical trials have been published[4-6], interim results from which confirm the value of anticoagulation for the primary prevention of stroke and peripheral embolus in patients with NVAF. These studies have also raised important questions on the use of aspirin for such patients. In 1989 the Copenhagen Atrial Fibrillation, Aspirin, Anticoagulation study (AFASAK) published results from 1007 patients (median age 74.2 years) randomized to warfarin (INR maintained 2.4-4.2), aspirin (75 mg) or placebo (mean follow-up 11 months). There was a significant reduction in all thromboembolic events (stroke, TIA, peripheral embolism) in patients receiving warfarin of 2% per annum (95% confidence limits 0.4-4.8 %) when compared with aspirin (5.5%) or placebo (5.5% per annum: 95% CL 2.9-9.4%). Vascular death was also significantly reduced in the warfarin group.
Subsequent to this study, the Stroke Prevention in Atrial Fibrillation trial (SPAF), randomized 1244 patients (mean age: men 65 years, women 68 years) to warfarin (INR 2-3.5), aspirin (325 mg) or placebo. This study was terminated prematurely following an interim analysis which demonstrated a clear beneficial treatment effect for both aspirin and warfarin. In this study the rate of cardioembolic events (ischaemic stroke and systemic embolism) was significantly reduced in patients given active treatment; warfarin or aspirin (1.6% per annum) compared with placebo (8.3%) representing an 81% risk reduction (95% CL 56-91%). For patients given aspirin there was a 49% risk reduction of cardioembolic events (95% CL 15-69%) although this benefit was confined to patients aged less than 75 years. The relatively small number of end-points in this trial precluded further analysis of factors that may explain this age related difference.
The most recently published study, the Boston Area Anticoagulation Trial for Atrial Fibrillation (BAATAF) compared the use of warfarin with placebo in patients with NVAF. This was an unblinded randomized trial in which 420 patients were randomized to open low-dose warfarin (INR 1.2-1.5) and the control group could choose to take open aspirin if they wished. Again a clear beneficial effect was demonstrated in the warfarin treatment group after a mean follow-up period of 2.2 years. In this study there was an 86% risk reduction (95% CL 51-96%) in ischaemic stroke events (fixed deficit > 24 hours) in patients given low-dose warfarin compared with the control group.
From these three clinical trials it would appear that both low-and full-dose warfarin are effective in reducing the risk of stroke, TIA and peripheral embolus in patients with NVAF in addition to reducing overall mortality. The role of aspirin is as yet unclear. From the evidence available it is possible that in patients aged less than 75 years there may be a beneficial effect with 325 mg aspirin daily although there is no evidence that lower doses (75 mg) are beneficial in patients with NVAF. There are obvious differences between these study cohorts[4, 5]; the AFASAK study cohort had a higher prevalence of cardiac failure and myocardial infarction and over half were aged over 74 years. Such patients are potentially more at risk of stasis-related thrombi for which antiplatelet therapy is less likely to be effective.
The studies described are primary prevention studies in patients with NVAF. For individuals in sinus rhythm, aspirin is of limited value in the primary prevention of stroke but is of course of proven value for the secondary prevention of vascular events[39, 40]. There is, however, no general agreement as to what the optimum dose of aspirin should be. For secondary prevention in sinus rhythm, low-dose aspirin (30 mg) appears to be as efficacious as high-dose (300 mg)[41, 17]. For primary prevention in patients with NVAF, there is no present explanation why the efficacy of aspirin may be dose- or age-related unless this is purely a |trial effect'. The role of aspirin and warfarin in secondary prevention in patients with NVAF remains unclear. Previous studies have been non-randomized and uncontrolled[42-45]. The European Atrial Fibrillation Trial (EAFT) which is a large, multicentre, randomized trial of aspirin and warfarin is presently under way to address this question.
Risks of anticoagulation
Present evidence suggests that widespread anticoagulation with warfarin in low or high dose is associated with a 69% risk reduction of stroke (95% CL 48-82%)[4-6]. Only approximately 35% of patients with atrial fibrillation ultimately experience a stroke in their lifetime[10, 12] although, as discussed, up to 24% may have CT evidence of silent cerebral infarction. Anticoagulation with warfarin in patients following a transient ischaemic attack or myocardial infarction is associated with an eight-fold risk (1% per annum) of intracerebral haemorrhage in addition to a 2% per annum risk of a potentially life-threatening non-central nervous system haemorrhage[47-49]. The annual incidence of serious bleeding complications in the three primary prevention trials discussed ranged from 0.5% to 3.2%[4-6]. In the BAA-TAF study where low-dose warfarin was used to maintain the INR between 1.2 and 1.5 there was only one fatal haemorrhage in the treatment group and 38 and 21 minor haemorrhages in the warfarin and control group, respectively.
In comparison with anticoagulation, platelet anti-aggregating agents are a simpler and safer option for widespread ischaemic stroke prevention. The risk of gastro-intestinal (GI) side-effects appears to be directly related to the dose of aspirin and ranges from 2.6% to 3.6% GI bleeds per annum 30-300 mg aspirin, respectively [17, 41]. Aspirin is also associated with an increased risk of intracerebral haemorrhage[17, 40]. In the UK-TIA study where 2435 patients were randomized to aspirin or placebo, 14 patients experienced a primary intracerebral haemorrhage whilst taking aspirin compared with two on placebo. Similarly, in the SALT study there was an increase in the risk of fatal haemorthagic stroke (4 v 0 aspirin: placebo). Overview analyses from all the secondary prevention trials show a small excess of haemorrhagic strokes associated with long-term antiplatelet therapy (approximately 0.5/ 1000 per annum). This excess is of course outweighed by the 15% (150/1000) reduction in ischaemic strokes per annum associated with aspirin therapy.
Despite the potential for preventing cardioembolic strokes in patients with NVAF, many negative attitudes still prevail regarding the widespread use of warfarin in those at greatest risk, namely elderly patients. In the BAATAF study of low-dose warfarin, there was no difference in patient wellbeing or health perception between warfarin and placebo treatment groups although not surprisingly patients who had complications of warfarin therapy had a significant reduction in health perception.
An American survey of physicians' attitudes to anticoagulants revealed that whilst the majority would anticoagulate patients with atrial fibrillation and mitral valve disease, this was not the case in patients with NVAF. Fear of haemorrhagic complications (52%) and the erroneous belief of a lack of efficacy of warfarin (24%) in primary prevention were the principal reasons for such attitudes.
Age over 65 has been implicated as a risk factor for anticoagulant-induced haemorrhage [51, 52]. Several studies have however demonstrated no age-related increase in haemorrhagic complications[53, 54]. Provided that patients without known contra-indications to anticoagulation are chosen and, regular monitoring is undertaken then anticoagulant therapy is not contra-indicated by age alone.
[1.] Oxfordshire Community Stroke Project. Incidence of stroke in Oxfordshire: first year's experience of a Community Stroke Register. Br Med J 1983;2:713-16. [2.] Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation: a major contributor to stroke in the elderly: the Framingham study. Arch Int Med 1987;147:1561-4. [3.] Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation as an independent risk factor for stroke: the Framingham study. Stroke 1991;22:983-8. [4.] Petersen P, Boyson G, Godtfredsen J, Anderson ED, Anderson B. Placebo controlled, randomized trial of warfarin and aspirin for prevention of thromboembolic complications in chronic atrial fibrillation: the Copenhagen AFASAK study. Lancet 1989;i:175-9. [5.] Stroke Prevention in Atrial Fibrillation Study Group Investigators. Preliminary report of the stroke prevention in atrial fibrillation study. N Engl J Med 1990;322:863-8. [6.] The Boston Area Anticoagulation Trial for Atrial Fibrillation Investigators. The effect of low dose warfarin on the risk of stroke in patients with non-rheumatic atrial fibrillation. N Engl J Med 1990;323:1505-11. [7.] Kutner M, Nixon G, Silverstone F. Physicians' attitudes towards oral anticoagulants and antiplatelet agents for stroke prevention in elderly patients with atrial fibrillation. Arch Intern Med 1991;151:1950-3. [8.] Evans JG. Risk factors for stroke in the elderly. MD Dissertation. University of Cambridge, 1985. [9.] Heger JJ. Cardiac arrhythmias in the elderly. In: Noble RJ, Rothbaum DA, eds. Geriatric cardiology. (Cardiovascular Clinics Series) Philadelphia, Davies, 1981;145-59. [10.] Hinton RC, Kistle JP, Fallon JT, et al. Influence of etiology of atrial fibrillation on incidence of systemic embolism. Am J Cardiol 1977;40:509-15. [11.] Kannel WB, Abbott RD, Savage DD, McNamara PM. Epidemiological features of chronic atrial fibrillation: the Framingham study. N Engl J Med 1982;306:1018-22. [12.] Sherman DG, Goldman L, Whiting RB, Jurgensen K, Katse M, Easton Dj. Thromboembolism in patients with atrial fibrillation. Arch Neurol 1984;41:708-10. [13.] Halperin JL, Hart RG. Atrial fibrillation and stroke: new ideas, persisting dilemmas. Stroke 1988;19:937-41. [14.] Wolf PA, Dawber TR, Thomas HE, Kannel WB. Epidemiologic assessment of chronic atrial fibrillation and risk of stroke: the Framingham study. Neurology 1978;28:973-7. [15.] Wolf PA, Kannel WB, McGee DL, Meeks SL, Bharucha NE, McNamara PM. Duration of atrial fibrillation and imminence of stroke: the Framingham study. Stroke 1983;14:664-7. [16.] Sage JI, Van Uitert RL. Risk of recurrent stroke in patients with atrial fibrillation and nonvalvular heart disease. Stroke 1983;14:537-40. [17.] UK-TIA Study Group. United Kingdom transient ischaemic attack (UK-TIA) aspirin trial: interim results. Br Med J 1988;296:316-20. [18.] Kopecky SL, 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;317:669-74. [19.] Brand FN, Abbott RD, Kannel WB, Wolf PA. Characteristics and prognosis of lone atrial fibrillation: 30 year follow up of the Framingham study. JAMA 1985:254:3449-53. [20.] Petersen P, Madsen EB, Brun B, et al. Silent cerebral infarction in chronic atrial fibrillation. Stroke 1987;18:1098-1100. [21.] Kempster PA, Gerraty RP, Gates PC. Asymptomatic cerebral infarction in patients with chronic atrial fibrillation. Stroke 1988;19:955-7. [22.] Feinberg WM, Seeger JF, Carmody RF, Anderson DC, Hart RG, Pearce LA. Epidemiologic features of asymptomatic cerebral infarction in patients with nonvalvular atrial fibrillation. Arch Int Med 1990;150:2340-4. [23.] Ratcliffe PJ, Wilcock GK. Cerebrovascular disease in dementia: the importance of atrial fibrillation. Postgrad Med J 1985;61:201-4. [24.] Babikian VL, Wolfe V, Linn R, Knoefel JE, Albert ML. Cognitive changes in patients with multiple cerebral infarcts. Stroke 1990;21:1013-18. [25.] Friedman Pj. Atrial fibrillation after stroke in the elderly. Stroke 1991;22:209-14. [26.] Petersen P. Thromboembolic complications in atrial fibrillation. Stroke 1990;21:4-13. [27.] Treseder AS, Sastry BSD, Thomas TPL, Yates MA, Pathy MSJ. Atrial fibrillation and stroke in elderly hospitalized patients. Age Ageing 1986; 15:89-92. [28.] Petersen P, Godtfredsen J. Embolic complications in paroxysmal atrial fibrillation. Stroke 1986;17:622-6. [29.] Chesebro JH, Fuster V, Halperin JL. Atrial fibrillation-risk marker for stroke. N Engl J Med 1990;323:1556-8. [30.] Caplan LR, D'Cruz I, Hier DB, Reddy H, Shah S. Atrial size, atrial fibrillation and stroke. Ann Neurol 1986; 19:158-61. [31.] Tegeler CH, Hart RG, Sherman DG, Quiroga GS, Easton JD. Atrial fibrillation and stroke study[Abstract]. Stroke 1986;17:131. [32.] Aschenberg W, Schluter M, Kremer P, Schroder E, Siglow V, Bleifeld W. Transoesophageal 2-dimensional echocardiography for the detection of left atrial appendage thrombus. Y Am Coll Cardiol 1986;7:163-6. [33.] Lodder J, Bamford JM, Sandercock PAG, Jones LN, Warlow CP. Risk factors amongst subgroups of cerebral infarction in the Oxfordshire Community Stroke Project: are hypertension or cardiac embolism likely causes of lacunar infarction? Stroke 1990;21:375-81. [34.] Hankey GJ, Warlow CP. Lacunar transient ischaemic attacks: a clinically useful concept? Lancet 1991;337:335-8. [35.] Weinberger J, Rothlauf E, Materese L, Halperin J. Non-invasive evaluation of the extracranial carotid arteries in patients with cerebrovascular events and atrial fibrillation. Arch Int Med 1988;148:1785-8. [36.] Dunn M, Alexander J, deSilva R, Hildner F. Antithrombotic therapy in atrial fibrillation. Chest 1989;95:118-25S. [37.] Levine HJ, Pauker SG, Salzman EW. Antithrombotic therapy in valvular heart disease. Chest 1989;95:98-106S. [38.] Peto R, Gray R, Collins R, et al. Randomised trial of prophylactic daily aspirin in British male doctors. Br Med J 1988;296:313-16. [39.] Antiplatelet Trialists' Collaboration. Secondary prevention of vascular disease by prolonged antiplatelet treatment. Br Med J 1988;296:320-31. [40.] The SALT Collaborative Group. Swedish Aspirin Low-dose Trial (SALT) of 75 mg aspirin as secondary prophylaxis after cerebrovascular ischaemic events. Lancet 1991; 338:1345-9. [41.] The Dutch TIA Trial Study Group. A comparison of two doses of aspirin (30 mg vs 283 mg a day) in patients after a transient ischaemic attack or minor ischaemic stroke. N Engl J Med 1991;325:1261-6. [42.] Rothruck JF, Dittrich HC, McAllen S, et al. Acute anticoagulation following cardioembolic stroke. Stroke 1989;20:730-4. [43.] Koller RL. Recurrent embolic cerebral infarction and anticoagulation. Neurology 1982; 32:283-5. [44.] Szekely P. Systemic embolism and anticoagulation prophylaxis in rheumatic heart disease. Br Med J 1964;1:209-12. [45.] Carter AB. The immediate treatment of cerebral embolism. Q J Med 1957;26:335-47. [46.] The EAFT Study Group' European Atrial Fibrillation Trial. The secondary preventive effect of anticoagulants and acetylsalicylic acid in patients with non-rheumatic atrial fibrillation and a TIA or non-disabling ischaemic stroke. January 1990 [Unpublished]. [47.] Whisnant JP, Cartlidge NEF, Elveback LR. Carotid and vertebrobasilar TIA: effect of anticoagulants, hypertension and cardiac disorders on survival and stroke occurrence. Ann Neurol 1978;3:107-15. [48.] Sixty-plus Reinfarction Study Research Group. Risks of longterm anticoagulant therapy in elderly patients after myocardial infarction. Lancet 1982;i:64-8. [49.] Forfar JC. Prediction of haemorrhage during longterm oral coumarin anticoagulation by excessive prothrombin ratio. Am Heart J 1982; 103:445-7. [50.] Lancaster TR, Singer DE, Sheeham MA, et al. The impact of long-term warfarin therapy on quality of life. Arch Intern Med 1991;151: 1944-9. [51.] Landefeld CS, Goldman L. Major bleeding in outpatients treated with warfarin: Incidence and prediction by factors known at the start of outpatient therapy. Am J Med 1989;87:144-52. [52.] Forfar JC. A 7-year analysis of haemorrhage in patients on long-term anticoagulant treatment. Br Heart J 1979;42:128-32. [53.] Gurtwitz JH, Goldberg RJ, Holden A, et al. Age-related risks of long-term oral anticoagulation therapy. Arch Intern Med 1988;148:1733-6. [54.] Wickramasinghe LSP, Basu SK, Bansal SK. Long-term oral anticoagulant therapy in elderly patients. Age Ageing 1988;17:388-96.
|Printer friendly Cite/link Email Feedback|
|Author:||O'Connell, Janice E.; Cassidy, Timothy P.; Gray, Christopher S.|
|Publication:||Age and Ageing|
|Date:||Sep 1, 1992|
|Previous Article:||Pulmonary function in gastro-oesophageal reflux disease of elderly people.|
|Next Article:||Alzheimer's Disease Treatment and Family Stress: Directions for Research.|