Optimal intensity of warfarin therapy for secondary prevention of stroke in patients with nonvalvular atrial fibrillation : a multicenter, prospective, randomized trial. Japanese Nonvalvular Atrial Fibrillation-Embolism Secondary Prevention Cooperative Study Group

Population-based distribution of plasminogen activity and estimated prevalence and relevance to thrombotic diseases of plasminogen deficiency in the Japanese: the Suita Study

Reduced plasminogen activity with a normal level of antigen is commonly observed in Japanese individuals. The first reported patient with plasminogen deficiency was accompanied with deep vein thrombosis. The present study examines whether heterozygous or homozygous deficiency of plasminogen is a risk factor for thrombotic disease. This study measures the plasminogen activity of 4517 individuals in the general population, determines the cut-off to define plasminogen deficiency, and identifies plasminogen deficiencies in the control groups and thrombotic disease groups. In another study, we examined the phenotypes of consecutive patients with homozygous plasminogen deficiency detected in our hospital. We found 173 and two of 4517 individuals to have heterozygous and homozygous deficiency with normal plasminogen antigen level, respectively, and 19 to have heterozygous deficiency with reduced antigen levels. The incidence of plasminogen deficiency in an age- and sex-matched control group (13/324, 4.01% for deep vein thrombosis or 13/330, 3.94% for stroke) selected from the 4517 individuals was not significantly different from those in patients with deep vein thrombosis (3/108, 2.78%) or cardioembolic stroke (6/110, 5.55%). Among 19 patients with homozygous plasminogen deficiency showing about 10% plasminogen activity, none had deep vein thrombosis. These findings indicate that neither heterozygous nor homozygous plasminogen deficiency constitutes a significant risk factor for thrombotic disease.

Investigation of the optimal treatment strategy for atrial fibrillation in Japan

The Japanese Rhythm Management Trial for Atrial Fibrillation (J-RHYTHM study) is a randomized comparative evaluation of rate control and rhythm control, both combined with antithrombotic therapy, as therapeutic strategies for the treatment of atrial fibrillation (AF). This study differs from the earlier AFFIRM and RACE studies in that it has a composite primary end-point representing mortality and also physical/psychological disablement (total mortality, symptomatic cerebral infarction, systemic embolism, major bleeding, hospitalization for heart failure requiring intravenous administration of diuretics, and patient disablement). Patients' will to change the therapeutic strategy to the other is also considered as an end-point representing disablement under the assigned strategy. The secondary end-point includes quality of life scores and the efficacy and safety of drugs used in treating AF. The J-RHYTHM study emphasizes patient-reported experience and perception of AF-specific disablement, and the safety of antiarrhythmics available in Japan; it will follow 2600 patients treated at more than 150 sites in Japan for a 3-year period.

Traitement anticoagulant dans la fibrillation auriculaire

Atrial fibrillation is the most frequent arrhythmia in the general population, and it increases with age. The prevention of thromboembolic events, the most important complication of the disease, is a major problem. Antivitamin K is to date the most efficient therapeutic class for the prevention of these events. Although they allow a decrease in stroke by at least 50%, they are associated with an increased haemorrhagic risk (annual incidence ranging from 7% to 22%). This risk makes oral anticoagulant treatment underused in high risk patients, particularly in the elderly populations. Optimisation of the management of patients with atrial fibrillation should be based on an individual evaluation of the thromboembolic and haemorrhagic risks. Several stratifications have been performed to identify the risk predictors of thromboembolic and haemorrhagic events in patients with atrial fibrillation, allowing an evaluation of the benefit/risk ratio of antithrombotic treatments and using indices such as NNT (number of patients needed to treat to avoid an event) and NNH (number of patients needed to harm with haemorrhagic event). The available data do not allow, however, to evaluate precisely the individual level of haemorrhagic risk.

Cardioembolic stroke: an update

Cardioembolic stroke accounts for approximately 15% of all strokes and is thought to be one of the more preventable types of strokes. Features that have been reported to support cardioembolism as a mechanism for ischemic stroke have included documented cardiac source of embolism, maximal neurologic deficit at onset, multiple cerebrovascular territories involved, enhanced tendency toward hemorrhagic transformation, enhanced risk of syncope or seizure associated with presentation, and lower likelihood of premonitory transient ischemic attacks. Features that tend to make cardioembolic stroke less likely include significant cerebral atherosclerosis, step-wise progression of the neurologic deficit within a finite period of time, vascular distribution such as entire internal carotid artery territory with combined middle cerebral artery and anterior cerebral artery involvement or watershed distribution, and premonitory transient ischemic attacks. A number of cardiac conditions can promote thromboembolism, and there is risk stratification reflective of the specific condition or coexistent conditions. Anticoagulant therapy generally has been found to be the most effective means of preventing cardiogenic brain embolism, but the intensity of anticoagulation needs to be optimized to reflect the risk-to-benefit ratio for the particular patient.

[Oral anticoagulants in the elderly]

PURPOSE

Atrial fibrillation and venous thromboembolism are particularly frequent in the elderly. Whether or not prescribe oral anticoagulant treatment in the elderly is therefore a common question for the physician. Despite the benefits of anticoagulation demonstrated in clinical trials, oral anticoagulant therapy is underused in the elderly.

CURRENT KNOWLEDGE AND KEY POINTS

Indications for oral anticoagulation are discussed specifically in the elderly with a literature review. Only the length of anticoagulation treatment after a venous thromboembolism remained a purpose of discussion regarding the severity of the pathology. The frequency of systemic thromboembolism in nonvalvular atrial fibrillation is increasing with age. Oral anticoagulation reduces this risk. This benefice is to compare with the increasing rate of major bleeding complications in the elderly and in patient who had stroke, hyper-tension, diabetes mellitus or gastrointestinal bleeding. The objective of this article is to focus on the mode to measure oral anticoagulant benefice/risk ratio in the elderly and to propose several ways to minimize the risk for bleeding.

FUTURE PROSPECTS AND PROJECTS

The potential drug side effect severity of oral anticoagulation must lead to find the "reasonable" clinical indications in term of benefice/risk ratio and what measures should be take to increase the safety of oral anticoagulation in the elderly. The comprehensive geriatric evaluation should be considered as a decision-aid tool in long-term oral anticoagulation in the frail elderly. Anticoagulation clinics, informatics'-prescription coupled with dose-adaptation decision-aid adapted to the elderly should be helpful in this research of quality. Finally, prescribers education supports must insist on the early course of therapy that is at higher risk of bleeding.

Stroke prevention in patients with atrial fibrillation

We evaluated the antithrombotic therapy and eligibility for anticoagulation before stroke in 30 patients with atrial fibrillation (AF) admitted to a district hospital in Kochi, Japan from 1992 to 1998. The mean age was 77+/-10 years old. Subtypes of ischemic stroke were classified as possibly cardioembolic in 26 (87%) patients and lacunar in four (13%). Eight (26.7%) patients died in the acute phase and 15 (50%) were disabled at discharge. Most patients were eligible for anticoagulation before stroke because of previously known AF (80%), high risk for stroke (80%), absence of contraindications (83.3%), and good clinical compliance (90%). The prescription rate of warfarin was, however, less than 20% even in high risk patients who needed anticoagulation. In conclusion, underuse of warfarin and high eligibility for anticoagulation in stroke patients with AF suggest that the chance of stroke prevention may be lost in many patients with AF in clinical practice.

Medical therapy for secondary prevention of stroke

The identification and modification of risk factors for stroke and their appropriate management can lead to reduction of stroke incidence. The real impact on recurrences of risk factors associated with lifestyles has not been thoroughly investigated, and the possible role of their modification in secondary prevention is principally extrapolated from primary prevention studies. On the other hand, several pathological conditions such as hypertension, atrial fibrillation, carotid stenosis, and diabetes are known to favour the risk of recurrence. Available antiplatelet regimens offer only partial protection against stroke and more efficacious antithrombotic agents would be useful. There is no doubt that warfarin is effective in preventing recurrence in stroke patients with atrial fibrillation. However, a careful etiological subtyping of stroke is recommended before starting treatment.

Correction of INR by prothrombin complex concentrate and vitamin K in patients with warfarin related hemorrhagic complication

We investigated the effect of prothrombin complex concentrate (PCC, median 500 IU) and vitamin K (10-20 mg) or either on blood coagulation and clinical findings in 17 patients with major hemorrhagic complication during warfarin treatment. Their international normalized ratio (INR) at admission was median 2.7 (2.0-above 10.0). In 11 patients treated with PCC and vitamin K, INR decreased to median 1.13 (0.91-1.36) 10 min after the administration with elevation of plasma levels of coagulant factors II, VII, IX, X and protein C.INR decreased abruptly after the administration of PCC without vitamin K in two patients but it increased again 12-24 h after, with decrease of coagulant factors levels. In one of them, a hematoma of the brain enlarged with INR re-increase 12-24 h after the administration. In four patients treated with vitamin K alone, INR decreased slowly from 2.69 (1.03-3.35) to 1.28 (1.25-1.44) 12-24 h after the administration in parallel with gradual increase of the coagulant factors.PCC administration with or without vitamin K seems to be more effective in rapidly correcting increased INR levels than vitamin K treatment without PCC. PCC without vitamin K may result in re-increase of INR and clinical deterioration.

D-dimer: a characteristic of the coagulation state of each patient with chronic atrial fibrillation

BACKGROUND AND OBJECTIVE

It is accepted that patients with atrial fibrillation (AF) are characterised by increased levels of plasmatic D-dimers, with a wide inter-individual variability depending on the patients and therapeutic characteristics, but it has not been established if this level was predictive of the risk of arterial thromboembolic event. In order to answer such a question, it has to be established if the D-dimer level in a given patient is characteristic of such a patient (stable over time) if also fluctuating with time (and useless to characterise the patient).

METHODS AND RESULTS

One hundred thirty clinically stable patients with chronic AF were recruited (anticoagulant: group 1, antiaggregant aspirin: group 2, no antithrombotic: group 3). During the follow-up of patients without clinical events (n=63), it is notable that in patients with D-dimer levels <500 ng/ml, these remained <1000 ng/ml, in patients with levels between 500 and 1000 ng/ml, these did not reach 1590 ng/ml, and in those with D-dimers >1000 ng/ml, the levels remained relatively stable. Mean age and D-dimer levels were lower in group 1 (74.4 years and 509.1 ng/ml, respectively) than in group 2 (82.4 years, p=0.0003 and 1015.7 ng/ml, p<0.0001, respectively) and in group 3 (79.3 years and 1289.3 ng/ml, p<0.0001, respectively). The effect of the antithrombotic therapy was independent of the age of patients (p=0.017).

CONCLUSION

D-dimer levels in patients with chronic AF remain in the same range over time. They are lower on anticoagulant therapy than on antiaggregant or no antithrombotic therapy, irrespective of age. Thus, D-dimers appear to be a useful parameter for assessing the degree of hypercoagulability of patients whatever their age.

Neurologic manifestations of cardiogenic embolism: an update

Cardiogenic embolism is increasingly appreciated as an important and preventable cause of stroke. Several potential sources of embolism have been identified with the advent of transoesophageal echocardiography. Their role as independent risk factors for stroke and management implications based on recent evidence, along with characterization of schemes for antithrombotic management of patients with atrial fibrillation are reviewed.

Aortic spontaneous echocardiographic contrast and hemostatic markers in patients with nonrheumatic atrial fibrillation

OBJECTIVES

To determine the relationship between spontaneous echocardiographic contrast (SEC) in the descending thoracic aorta and plasma levels of hemostatic markers in patients with nonrheumatic atrial fibrillation (AF).

DESIGN AND SETTINGS

A cross-sectional study at a university hospital.

PATIENTS AND MEASUREMENTS

In 91 consecutive patients (mean +/- SE age, 70 +/- 1 years; 68 men) with nonrheumatic AF who underwent transesophageal echocardiography, plasma levels of markers for platelet activity (platelet factor 4 [PF4] and beta-thromboglobulin [beta-TG]), thrombotic status (thrombin-antithrombin III complex [TAT]), and fibrinolytic status (D-dimer and plasmin-alpha(2)-plasmin inhibitor complex [PIC]) were determined.

RESULTS

Forty-three patients who had aortic SEC (AoSEC) were older (72 years vs 68 years; p < 0.05) and had a higher prevalence of chronic AF (88% vs 52%; p < 0.05) than 48 patients without AoSEC. TAT, PIC, and D-dimer levels were significantly higher in patients with AoSEC than in those without AoSEC, whereas PF4 and beta-TG levels were not different between the two groups. Although the prevalence of cerebral embolism did not differ between the two groups (23% vs 29%), the prevalence of peripheral embolism was higher in patients with AoSEC than in those without AoSEC (10% vs 0%; p < 0.05). Multivariate analysis revealed mitral regurgitation (odds ratio, 7.53; p < 0.02), SEC in the left atrium (odds ratio, 2.14; p < 0.02), and aortic atherosclerosis (odds ratio, 1.87; p < 0.04) emerged as independent predictors of AoSEC.

CONCLUSIONS

Patients with nonrheumatic AF who have AoSEC appear to have enhanced coagulation activity but not platelet activity. Intensive anticoagulation treatment might be required for these patients.

A comparison of warfarin and aspirin for the prevention of recurrent ischemic stroke

BACKGROUND

Despite the use of antiplatelet agents, usually aspirin, in patients who have had an ischemic stroke, there is still a substantial rate of recurrence. Therefore, we investigated whether warfarin, which is effective and superior to aspirin in the prevention of cardiogenic embolism, would also prove superior in the prevention of recurrent ischemic stroke in patients with a prior noncardioembolic ischemic stroke.

METHODS

In a multicenter, double-blind, randomized trial, we compared the effect of warfarin (at a dose adjusted to produce an international normalized ratio of 1.4 to 2.8) and that of aspirin (325 mg per day) on the combined primary end point of recurrent ischemic stroke or death from any cause within two years.

RESULTS

The two randomized study groups were similar with respect to base-line risk factors. In the intention-to-treat analysis, no significant differences were found between the treatment groups in any of the outcomes measured. The primary end point of death or recurrent ischemic stroke was reached by 196 of 1103 patients assigned to warfarin (17.8 percent) and 176 of 1103 assigned to aspirin (16.0 percent; P=0.25; hazard ratio comparing warfarin with aspirin, 1.13; 95 percent confidence interval, 0.92 to 1.38). The rates of major hemorrhage were low (2.22 per 100 patient-years in the warfarin group and 1.49 per 100 patient-years in the aspirin group). Also, there were no significant treatment-related differences in the frequency of or time to the primary end point or major hemorrhage according to the cause of the initial stroke (1237 patients had had previous small-vessel or lacunar infarcts, 576 had had cryptogenic infarcts, and 259 had had infarcts designated as due to severe stenosis or occlusion of a large artery).

CONCLUSIONS

Over two years, we found no difference between aspirin and warfarin in the prevention of recurrent ischemic stroke or death or in the rate of major hemorrhage. Consequently, we regard both warfarin and aspirin as reasonable therapeutic alternatives.

Secondary prevention of stroke in patients with nonvalvular atrial fibrillation: optimal intensity of anticoagulation

Nonvalvular atrial fibrillation (NVAF) is frequently seen in elderly people and has become a main cause of cardioembolic stroke. The efficacy of anticoagulation for primary prevention of stroke or transient ischaemic attacks (TIAs) in patients with NVAF has been established by prospective, randomised and controlled trials. Warfarin decreased the frequency of all strokes by 68% and the rate of the combined outcome of stroke, systemic embolism or death by 48%. Anticoagulation with warfarin using international normalised ratios (INRs) ranging from 2.0 to 3.0 is recommended for patients with NVAF, who have any of the risk factors identified by the Atrial Fibrillation Investigators (AFI) [previous stroke or TIA, history of hypertension, diabetes mellitus, advanced age (> or = 65 years old), congestive heart failure and coronary artery disease], the American College of Chest Physicians (ACCP) [increased age (> 75 years old), prior stroke, hypertension and heart failure], or the Stroke Prevention in Atrial Fibrillation (SPAF) investigators [women > 75 years old, prior stroke, systolic blood pressure > 160mm Hg, recent heart failure, and fractional shortening < 25% on echocardiography]. For the secondary prevention of stroke, the efficacy of adjusted-dose warfarin therapy has been demonstrated by 2 major randomised trials. SPAF III (INR 2.0 to 3.0) demonstrated a lower incidence of ischaemic stroke or systemic embolism (3.4 %/year) compared with low fixed-dose warfarin plus aspirin (acetylsalicylic acid) [11.9%]. The European Atrial Fibrillation Trial [EAFT] (INR 2.5 to 4.0) showed a lower incidence of all stroke (4.0 %/year) with adjusted-dose warfarin compared with placebo (12.0 %/year). The incidence of major bleeding in the adjusted-dose warfarin group in SPAF III and EAFT was 2.4 and 2.8 %/year, respectively. EAFT incidence rates for the occurrence of a first ischaemic or haemorrhagic complication analysed by INR range indicated that the rate was lowest at INRs of 2.0 to 2.9, and higher with INRs of 3.0 to 3.9. Therefore, the optimal intensity of anticoagulation for prevention of recurrent stroke seems to be an INR of between 2.0 and 3.0, as for primary prevention. Retrospective and prospective studies from Japan reported that in the elderly, haemorrhagic complications occur frequently with INRs above 2.6 and major ischaemic events cannot be prevented at INRs below 1.6. Therefore, an INR target between 1.6 and 2.6 may be an alternative for secondary prevention of stroke in elderly patients with NVAF who have a potential risk of bleeding, to avoid both major ischaemic and haemorrhagic events. Antiplatelets may be administered in patients who are unable to manage taking warfarin properly or who have a high risk of falling and subsequently sustaining a head injury, although the efficacy of antiplatelets for secondary prevention of stroke in NVAF has not yet been established.

Anticoagulation in the elderly

Oral anticoagulation therapy has demonstrated benefit in the treatment and prevention of a variety of thromboembolic disorders. Most individuals who receive oral anticoagulant therapy are elderly patients with nonvalvular atrial fibrillation and acute or recurrent venous thromboembolism. Anticoagulation in elderly patients poses unique challenges for the practicing clinician because they are simultaneously at higher risk for recurrent thromboembolism and major bleeding, including catastrophic intracranial hemorrhage. The pharmacology of warfarin in the elderly is reviewed, including important drug interactions and current dosing recommendations for elderly patients. Evidence of the benefits and risks of oral anticoagulation therapy are reviewed for patients with atrial fibrillation and venous thromboembolism. This information should enable practitioners to better assess the relative risks and benefit of oral anticoagulation therapy to guide treatment decisions in the elderly.

A transesophageal echocardiographic study on risk factors for stroke in elderly patients with atrial fibrillation: a comparison with younger patients

STUDY OBJECTIVES

Atrial fibrillation (AF) becomes an increasingly important cause of stroke as patients get older. The aim of the study was to determine whether risk factors of cerebral embolism among elderly patients with AF differed from those of younger patients by using transesophageal echocardiography (TEE).

DESIGN AND SETTING

Cross-sectional study at a university hospital.

METHODS

Cardiovascular lesions with the potential for thromboembolism in patients with AF were investigated using TEE. Left atrial spontaneous echocardiographic contrast (SEC), peak flow velocity in the left atrial appendage (LAA-flow), and aortic atherosclerosis of the thoracic aorta were assessed in 67 elderly (> or = 70 years old) and 135 younger (< 70 years old) patients. All patients underwent either brain CT (n = 54) or MRI (n = 148) to assess presence of cerebral infarction.

RESULTS

Cerebral infarction due to embolism was noted in 113 patients with AF. There was a higher prevalence of cerebral embolism in elderly patients when compared with younger patients (78% vs 45%; p < 0.001). Cerebral embolism found in younger patients was associated with high grade of SEC and lower LAA-flow (p < 0.05). In addition to these TEE findings, aortic atherosclerosis was more severe in elderly patients with cerebral embolism than in those without cerebral embolism (p < 0.0001). By multivariate logistic analysis, LAA-flow was an independent predictor of cortical infarction in younger patients, but not in elderly patients, whereas aortic atherosclerosis was a useful marker in predicting embolic risk in elderly patients.

CONCLUSIONS

TEE findings indicative of left atrial blood stasis were useful to identify the embolic risk of younger patients with AF, while atherosclerosis of the thoracic aorta appears to be an important marker for cerebral embolism in elderly patients.

Effect of low-intensity warfarin therapy on left atrial thrombus resolution in patients with nonvalvular atrial fibrillation: a transesophageal echocardiographic study

The presence of left atrial thrombus (LAT) is associated with an increased risk of embolic stroke. However, it has yet to be established definitively whether low-intensity warfarin therapy (INR: 1.5-2.0) can prevent LAT formation in patients with nonvalvular atrial fibrillation (NVAF). The present study analyzed the clinical and transesophageal echocardiography (TEE) features of 123 such patients to identify risk factors for LAT formation and the efficacy of prophylactic low-intensity warfarin therapy. Left atrial thrombi were found in 35 patients (28%) in whom systemic hypertension (49% vs 23%; p<0.01) and ischemic heart disease (17% vs 3%; p<0.01) were more frequent. Left ventricular ejection fraction (54+/-14% vs 60+/-11%; p<0.05), left ventricular end-diastolic dimension (51+/-7 mm vs 48+/-5 mm; p<0.05), spontaneous echo contrast (2.2+/-0.7 vs 1.4+/-0.9; p<0.01), left atrial diameter (50+/-6 mm vs 43+/-7 mm; p<0.01), left atrial appendage blood velocity (22.3+/-8.7 cm/s vs 37.2+/-21.5 cm/s; p<0.01) and the incidence of left ventricular hypertrophy (37% vs 15%; p<0.01) were also significantly different between the groups. Fourteen patients received continuous warfarin therapy (target INR: 1.5-2.0) and on the follow-up TEE study the left atrial thrombus resolved in 10 (71%). There were no thromboembolic events or major hemorrhagic complications in these patients, so it was concluded that low-intensity warfarin therapy is efficacious in treating LAT formation in patients with NVAF.

Antithrombotic therapy in atrial fibrillation

Atrial fibrillation is a common condition affecting elderly individuals; as many as 10% of people older than age 80 years have AF. AF is also a potent risk factor for ischemic stroke, raising the risk of stroke fivefold. A set of consistent randomized controlled trials has demonstrated that long-term anticoagulation can largely reverse the risk of stroke attributable to AF. In these trials, anticoagulation generally proved quite safe, raising the risk of intracranial hemorrhage by less than 0.5% per year. The anticoagulation target for AF is INR 2 to 3 with INR 2.5 as the specific goal. The trials were much less consistent about the efficacy of aspirin, although it seems that aspirin has a small stroke-preventive effect. The recommended dose of aspirin is 325 mg per day. Because it raises the risk of hemorrhage and adds the burden of frequent monitoring of INR values, anticoagulation is recommended for those patients with AF at higher risk of stroke. Such higher risk is conferred by the following risk factors: (1) a history of a prior stroke, TIA, or other systemic embolic event; (2) a history of hypertension; (3) diabetes mellitus; (4) left ventricular dysfunction; (5) mitral stenosis; and (6) older age. The exact age threshold conferring sufficiently increased risk is uncertain, with some research indicating the threshold should be age 65 years, and other research indicating the threshold should be age 75 years. For lower-risk patients, aspirin is recommended. Future research should focus on the oldest patients with AF. These individuals face the highest risk of ischemic stroke without anticoagulation and the highest risk of major hemorrhage with anticoagulation. Only small numbers of such elderly patients were included in the randomized trials. Future research should also focus on improved risk stratification, allowing better targeting of anticoagulation. Discoveries of new antithrombotic agents and new drugs and devices for preservation of sinus rhythm could radically improve stroke-preventive strategies for AF.

Anticoagulants or antiplatelet therapy for non-rheumatic atrial fibrillation and flutter

BACKGROUND

Atrial fibrillation (AF) carries a high risk of stroke and other thromboembolic events. Appropriate use of drugs to prevent thromboembolism in patients with AF involves comparing the patient's risk of stroke to the risk of hemorrhage from medication use.

OBJECTIVES

To quantify risk of stroke, major hemorrhage and death from using medications that have been rigorously evaluated for prevention of thromboembolism in AF.

SEARCH STRATEGY

Articles were identified through the Cochrane Collaboration's CENTRAL database and MEDLINE until December 1999.

SELECTION CRITERIA

Included Randomized controlled trials of drugs to prevent thromboembolism in adults with non-postoperative AF. Excluded RCTS of patients with rheumatic valvular disease.

DATA COLLECTION AND ANALYSIS

Data were abstracted by two reviewers. Odds ratios from all qualitatively similar studies were combined, with weighting by study size, to yield aggregate odds ratios for stroke, major hemorrhage, and death for each drug.

MAIN RESULTS

Fourteen articles were included in this review. Warfarin was more efficacious than placebo for primary stroke prevention [aggregate odds ratio (OR) of stroke=0.30 [95% Confidence Interval (C.I.) 0.19,0.48]], with moderate evidence of more major bleeding [ OR= 1.90 [95% C.I. 0.89,4.04].]. Aspirin was inconclusively more efficacious than placebo for stroke prevention [OR=0.68 [95% C.I. 0.29,1.57]], with inconclusive evidence regarding more major bleeds [OR=0.81[95% C.I. 0.37,1.78]]. For primary prevention, assuming a baseline risk of 45 strokes per 1000 patient-years, warfarin could prevent 30 strokes at the expense of only 6 additional major bleeds. Aspirin could prevent 17 strokes, without increasing major hemorrhage. In direct comparison, there was moderate evidence for fewer strokes among patients on warfarin than on aspirin [aggregate OR=0.64[95% C.I. 0.43,0.96]], with only suggestive evidence for more major hemorrhage [OR =1.58 [95% C.I. 0.76,3.27]]. However, in younger patients, with a mean age of 65 years, the absolute reduction in stroke rate with warfarin compared to aspirin was low (5.5 per 1000 person-years) compared to an older group (15 per 1000 person-years). Low-dose warfarin or low-dose warfarin with aspirin was less efficacious for stroke prevention than adjusted-dose warfarin.

REVIEWER'S CONCLUSIONS

The evidence strongly supports warfarin in AF for patients at average or greater risk of stroke, although clearly there is a risk of hemorrhage. Although not definitively supported by the evidence, aspirin may prove to be useful for stroke prevention in sub-groups with a low risk of stroke, with less risk of hemorrhage than with warfarin. Further studies are needed of low- molecular weight heparin and aspirin in lower risk patients.