0
We're unable to sign you in at this time. Please try again in a few minutes.
Retry
We were able to sign you in, but your subscription(s) could not be found. Please try again in a few minutes.
Retry
There may be a problem with your account. Please contact the AMA Service Center to resolve this issue.
Contact the AMA Service Center:
Telephone: 1 (800) 262-2350 or 1 (312) 670-7827  *   Email: subscriptions@jamanetwork.com
Error Message ......
Clinical Cardiology |

Oral Anticoagulants vs Aspirin in Nonvalvular Atrial Fibrillation:  An Individual Patient Meta-analysis FREE

Carl van Walraven, MD, MSc, FRCPC; Robert G. Hart, MD; Daniel E. Singer, MD; Andreas Laupacis, MD, MSc, FRCPC; Stuart Connolly, MD; Palle Petersen, MD, DMSc; Peter J. Koudstaal, MD, PhD; Yuchiao Chang, PhD; Beppie Hellemons, MD, PhD
[+] Author Affiliations

Author Affiliations: Clinical Epidemiology Unit, Ottawa Health Research Institute (Dr van Walraven); Institute for Clinical Evaluative Sciences (Drs van Walraven and Laupacis); Canadian Institutes for Health Research (Dr Laupacis), Ottawa, Ontario; Department of Medicine, University of Texas, San Antonio (Dr Hart); Clinical Epidemiology Unit, General Medicine Division, Massachusetts General Hospital, Boston (Drs Singer and Chang); McMaster University, Hamilton, Ontario (Dr Connolly); Hvidovre University Hospital, Copenhagen, Denmark (Dr Petersen); Department of Neurology, Erasmus MC, Rotterdam (Dr Koudstaal) and University of Maastricht, Maastricht (Dr Hellemons), the Netherlands. Dr van Walraven is an Ontario Ministry of Health Career Scientist.


Clinical Cardiology Section Editor: Michael S. Lauer, MD, Contributing Editor.


JAMA. 2002;288(19):2441-2448. doi:10.1001/jama.288.19.2441.
Text Size: A A A
Published online

Context Patients with nonvalvular atrial fibrillation (AF) have an increased risk of stroke and other vascular events.

Objective To compare the risk of vascular and bleeding events in patients with nonvalvular AF treated with vitamin K –inhibiting oral anticoagulants or acetylsalicylic acid (aspirin).

Design Pooled analysis of patient-level data from 6 published, randomized clinical trials.

Patients A total of 4052 patients with AF randomly assigned to receive therapeutic doses of oral anticoagulant or aspirin with or without low-dose oral anticoagulants.

Main Outcome Measures Ischemic and hemorrhagic stroke, other cardiovascular events, all-cause death, and major bleeding events. Person-year incidence rates were calculated to provide crude comparisons. Relative efficacy was assessed using proportional hazards modeling stratified by study. The variation of the oral anticoagulant's relative effect by pertinent patient factors was explored with interaction terms. All analyses were conducted using the intention-to-treat principle.

Results Patients receiving oral anticoagulant and aspirin were balanced for important prognostic factors. There was no significant heterogeneity between trials in the relative efficacy of oral anticoagulant vs aspirin for any outcome. Patients receiving oral anticoagulant were significantly less likely to experience any stroke (2.4 vs 4.5 events per 100 patient-years; hazard ratio [HR], 0.55; 95% confidence interval [CI], 0.43-0.71), ischemic stroke (HR, 0.48; 95% CI, 0.37-0.63), or cardiovascular events (HR, 0.71; 95% CI, 0.59-0.85) but were more likely to experience major bleeding (2.2 vs 1.3 events per 100 patient-years; HR, 1.71; 95% CI, 1.21-2.41). The reduction in ischemic stroke risk was similar in patients with paroxysmal AF (1.5 vs 4.7 events per 100 patient-years; HR, 0.32; 95% CI, 0.16-0.61; P<.001). Treating 1000 patients with AF for 1 year with oral anticoagulant rather than aspirin would prevent 23 ischemic strokes while causing 9 additional major bleeds. Overall all-cause survival did not differ but appeared to improve for oral anticoagulant patients 3 years after therapy was started.

Conclusions Compared with aspirin, oral anticoagulant significantly decreases the risk of all strokes, ischemic strokes, and cardiovascular events for patients with nonvalvular chronic or paroxysmal AF but modestly increases the absolute risk of major bleeding. The balance of benefits and risks varies by patient subgroup.

Figures in this Article

Nonvalvular atrial fibrillation (AF) more than quadruples the risk of stroke.1 Stroke prophylaxis in patients with atrial fibrillation has been examined extensively in randomized trials,216 analyses of individual patient-level data,1719 traditional meta-analyses,2031 and review articles.3242 The clinical trials, and their quantitative reviews, show that treatment with vitamin K–dependent oral anticoagulants significantly decrease the risk of stroke in AF patients by more than 50%. Acetylsalicylic acid (aspirin) also appears to decrease stroke risk but to a lesser extent.20 Because oral anticoagulant use is more troublesome43 and has a greater likelihood of complications,44 it is important to determine the relative efficacy and safety of oral anticoagulant vs aspirin overall and in relevant patient subgroups.

Six published clinical trials have randomized AF patients to oral anticoagulant or a regimen containing aspirin.27,9 Three of these studies found that oral anticoagulant was significantly more efficacious than aspirin at reducing thromboembolic outcomes.2,5,9 However, the conclusions from quantitative reviews of randomized trials comparing oral anticoagulant and aspirin-containing regimens have varied extensively from oral anticoagulants being "substantially more efficacious than aspirin"20 to finding "moderate evidence for fewer strokes among patients on oral anticoagulants than on aspirin"31 to "considerable uncertainty about the value of long-term anticoagulation compared to antiplatelet treatment."45 This uncertainty of the relative benefit of oral anticoagulant vs aspirin in AF might explain the relatively low use of oral anticoagulant in usual practice.46

All prior meta-analyses have been limited to using summary data available in published trial reports. We analyzed the pooled individual patient data from all published randomized trials comparing oral anticoagulant and aspirin for AF. These data are distinctive by permitting true time-to-event analyses, exploration of subgroup effects, and comparison of the efficacy of oral anticoagulant vs aspirin on a number of important cardiovascular outcomes in addition to ischemic stroke.

Included Studies: Design and Treatment

This study included patient-level data from every published clinical trial in which patients with AF were randomly assigned to full-dose oral anticoagulant or aspirin. These trials included the Atrial Fibrillation, Aspirin, Anticoagulation (AFASAK) Studies 12 and 2,3 Primary Prevention of Arterial Thromboembolism in patients with Nonrheumatic Atrial Fibrillation in Primary Care (PATAF),4 the European Atrial Fibrillation Trial (EAFT),5 and the Stroke Prevention in Atrial Fibrillation (SPAF) Studies 1,6 2,7 and 3.9 Since SPAF 1 patients randomized to receive oral anticoagulant or aspirin all continued into SPAF 2, SPAF 1 data are combined with SPAF 2 as in the original publication.7 To our knowledge, there are no other published studies in which AF patients were randomized to receive a full-dose of oral anticoagulant or aspirin.

All patients were adults with nonvalvular AF. The exclusion criteria varied only slightly between the studies. In general, patients were excluded if they had clinical indications for, or contraindications to, oral anticoagulation or aspirin. These contraindications included pregnancy, alcoholism, renal or hepatic failure, thrombocytopenia, or bleeding disorders. With the exception of the AFASAK studies,2,3 patients were also excluded if they had a recent acute coronary event or cardiac revascularization. All studies also excluded patients with thyrotoxicosis except for AFASAK 1, which included only 28 such patients.

Patients were randomly assigned to receive a full-dose of oral anticoagulant or aspirin with or without low-dose oral anticoagulant. Patients in the oral anticoagulant group were treated primarily with coumarin derivatives including warfarin sodium and 4-hydroxycoumarin. The target international normalized ratio (INR) varied slightly in each trial (Table 1). The daily aspirin dose ranged between 75 and 325 mg (Table 1).

Table Graphic Jump LocationTable 1. Description of Patients and Studies*

In our analyses, we combined patients treated with aspirin alone with patients treated with aspirin plus low-dose warfarin because there were no significant differences in outcomes between such patients.3 We also reasoned that, if low-dose warfarin has a previously unmeasured effect on cardiovascular outcomes, including such patients would likely decrease the differences between the oral anticoagulant and aspirin groups for cardiovascular outcomes. The AFASAK 2 and SPAF 3 trials were the only ones in which patients receiving aspirin also received low-dose warfarin in dosages of 1.25 mg/d and a median of 2.0 mg/d, respectively, and with negligible prolongation of the INR. To ensure that adding low-dose warfarin to aspirin did not bias our study, we repeated the analyses after excluding patients taking aspirin plus low-dose warfarin.

Baseline Factors

Patient clinical features were collected by research coordinators and physicians before the initiation of therapy (see eTable 1). These included previous stroke or transient ischemic attack, hypertension, congestive heart failure, diabetes, and coronary artery disease. All studies classified patients as hypertensive if they were taking medications given to lower blood pressure. In the AFASAK studies, patients were considered to have congestive heart failure if it was graded as moderate or severe. We used the Atrial Fibrillation Investigators risk stratification to classify patient stroke risk.18 Using these data, we labeled the patients with hypertension, diabetes, or prior cerebral ischemia as high risk. Patients without these risk factors were considered low risk if they were younger than 65 years. All others were considered moderate risk.

Table Graphic Jump LocationeTable 1. Definition of Baseline Factors*
Outcomes

We compared the relative efficacy of oral anticoagulant and aspirin for 6 outcomes: ischemic or hemorrhagic stroke; ischemic stroke alone; hemorrhagic stroke alone (including subarachnoid and subdural hemorrhage); aggregate cardiovascular events (including ischemic stroke, myocardial infarction, systemic embolism, or cardiovascular death); major bleeding (including intracranial and systemic bleeding); and all-cause death. Cardiovascular deaths included those due to stroke, myocardial infarction, congestive heart failure, pulmonary embolism, or systemic embolism. Ischemic stroke and major bleeds were classified as lethal if death occurred within 30 days of the event. (For outcome criteria for each trial, see eTable 2.)

Patients were followed up for a mean of 1.9 years (Table 1). Patients were routinely seen every 3 to 6 months by study investigators or when an outcome event was suspected. With the exception of patients in AFASAK 1, a central events committee that was blinded to therapeutic group reviewed and confirmed all events. Of all strokes, 97% were assessed by neuroimaging, almost exclusively by computed tomography.

Analysis

Crude event rates were calculated as events per 100 person-years of observation and compared using the normal approximation of the Poisson distribution as described by Rothman and Greenland.47 Kaplan-Meier curves were generated to compare time to each outcome, and the log-rank statistic determined the significance of differences between the curves. We used proportional hazards regression modeling to estimate the relative effect of oral anticoagulant on all 6 outcomes. Each model was stratified by study, thereby allowing different hazard functions to be estimated for each study and then a pooled hazard ratio (HR) across all studies was estimated. Such modeling helps adjust for factors that might be unique to a particular study.48,49 We tested the proportional hazards assumption for all Cox models using an interaction term between the predictor variable and time.49 If the P value for this term was less than .05, we concluded that the assumption of proportional hazards was not met over time.

We assessed heterogeneity between studies for each outcome by visually comparing HRs for each study with the overall estimate and by calculating the DerSimonian and Laird Q statistic for heterogeneity.50 To measure interactions between treatment group and baseline factors, appropriate interaction terms were included in the model. All analyses were conducted using SAS 8.1 software (Cary, NC).

Table 1 describes the patients according to the study in which the patients were enrolled. Atrial fibrillation was nonparoxysmal in 83% and was present for more than a year in 67% of patients. Sixty-five percent of patients were classified as having high, 27% as having moderate, and 8% as having low risk of stroke. Reflecting different patient selection criteria between studies, the prevalence of patient characteristics varied between trials (Table 1). After pooling patients from all 6 trials, patients receiving oral anticoagulant and aspirin were well matched (Table 2).

Table Graphic Jump LocationTable 2. Patient Features by Treatment Group*

Compared with aspirin, oral anticoagulant significantly decreased the rate of all stroke, ischemic stroke, and cardiovascular events (Table 3). The decrease in the rate of all stroke was due to a large decrease in ischemic stroke with only a small absolute increase in hemorrhagic stroke. Of all ischemic strokes, 12.5% were lethal. Oral anticoagulants also decreased the lethal ischemic stroke rate (0.5 vs 0.2 events per 100 patient-years; P = .01). The significant decrease in cardiovascular events seen in the oral anticoagulant group was due primarily to decreased rates of ischemic stroke and myocardial infarction.

Table Graphic Jump LocationTable 3. Comparison of Oral Anticoagulants and Aspirin for Several Outcomes*

The use of oral anticoagulant significantly increased the rate of major bleeding (Table 3), with 15.3% of all major bleeding episodes being lethal. Hemorrhagic stroke accounted for 21.9% of all major bleeding and 52.4% of fatal bleeding events. Patients receiving oral anticoagulant had a nonsignificantly increased rate of lethal hemorrhages. Overall mortality did not differ between patient groups.

Life-table analyses (Figure 1) confirmed the crude-rate comparisons and demonstrated, with the exception of all-cause death, a consistent relative beneficial effect of oral anticoagulant over time. All-cause death was the only outcome for which the assumption of proportionality was not satisfied. This is reflected in the survival plots for death showing a difference between treatment groups starting 2½ years into observation (Figure 1).

Figure 1. Outcomes Survival Curves
Graphic Jump Location
In each plot, the horizontal axis represents time in years. The P value is a log-rank statistic. All strokes included ischemic and hemorrhagic events. Cardiovascular events included ischemic strokes, myocardial infarctions, systemic emboli, and vascular death. Major bleeding events included intracranial and major systemic bleeds.

The outcomes for each trial are presented in Figure 2. The relative effect of oral anticoagulant for each outcome was similar across studies with the exception of major bleeds. For major bleeding, EAFT and SPAF 3 provided particularly high and low HRs, respectively. However, the DerSimonian and Laird Q statistic for heterogeneity50 in treatment effect was not significant for any outcome, including major bleeding.

Figure 2. Relationship of Therapy With Outcomes in Individual Trials
Graphic Jump Location
For each study, the relative effect of oral anticoagulants vs aspirin (with or without low-dose warfarin) is presented for all 6 outcomes as a hazard ratio. Hazard ratios below 1 indicate that oral anticoagulant decreases the risk of the event. Hazard ratios whose 95% confidence interval (error bars) excludes 1 are statistically significant at the 5% level. Because hemorrhagic strokes were uncommon events, a hazard ratio could not be estimated for each study individually. The P value for the DerSimonian and Laird Q statistic, as a measure of heterogeneity, is presented for each outcome in the top right-hand corner. AFASAK indicates Atrial Fibrillation, Aspirin, Anticoagulation study12,3; EAFT, European Atrial Fibrillation Trial5; PATAF, Primary Prevention of Atrial Thromboembolism in patients with Nonrheumatic Atrial Fibrillation in Primary Care4; SPAF, Stroke Prevention in Atrial Fibrillation studies.7,9

To determine whether our conclusions changed without patients assigned to receive aspirin plus low-dose warfarin, we repeated the analysis after excluding such patients. This analysis excluded SPAF 3 and the combined aspirin-low-dose warfarin group of AFASAK 2, resulting in 2837 patients. We noted similar results for all stroke, 2.4 events among those taking oral anticoagulant vs 3.8 taking aspirin per 100 patient-years (HR, 0.65; 95% CI, 0.49-0.86; P = .003); ischemic stroke, 2.0 vs 3.7 events per 100 patient years (HR, 0.56; 95% CI, 0.41-0.76; P<.001), cardiovascular events, 5.3 vs 6.9 events per 100 patient-years (HR, 0.76; 95% CI, 0.62-0.93; P = .02); and death, 4.7 vs 5.1 (HR, 0.92; 95% CI, 0.74-1.15; P = .41). There was a slight increase of the difference between the 2 groups in hemorrhagic stroke, 0.5 vs 0.2 events per 100 patient-years (HR, 2.26; 95% CI, 0.93-5.50; P = .06) and major bleeding, 2.2 vs 1.2 events per 100 patient-years (HR, 1.93; 95% CI, 1.30-2.88; P = .001). The HRs excluding those who received combined aspirin-low-dose warfarin are similar to HRs attained with the entire cohort (Table 3). When the patients taking aspirin and low-dose warfarin were excluded, cardiovascular outcome rates did not change extensively in the oral anticoagulant group but were noticeably lower in the aspirin group. This is likely due to the exclusion of high-risk patients enrolled in the SPAF 3 study.

Effect of Treatment in Patient Subgroups

We explored the interaction of treatment effect with relevant patient features on ischemic stroke and major bleeding (Figure 3). In almost all patient subgroups, oral anticoagulant was significantly more efficacious than aspirin for reducing the risk of ischemic stroke. Although we noted some variability in the relative benefit of oral anticoagulant between patient subgroups, most of these interactions were not significant with P>.10. There were 2 possible exceptions. The relative benefit of oral anticoagulant vs aspirin in ischemic stroke prevention appeared greater for patients younger than 75 years vs those who were 75 years or older (P for interaction = .08) and for women vs men (P for interaction = .04). The increased risk of major bleeding for patients taking oral anticoagulants appeared consistent in all patient subgroups and none of the interaction terms had a P<.10.

Figure 3. The Effect of Patient Baseline Factors on the Efficacy of Oral Anticoagulants or Aspirin on Ischemic Stroke and Major Bleeding
Graphic Jump Location
For subgroups in each patient factor (left), outcomes are presented as relative risks with 95% confidence intervals (error bars). Relative hazards that are less than 1 indicate that oral anticoagulants are associated with a decreased risk of the outcome. The event rates for each subgroup, expressed as the number of events per 100 person-years of observation, are presented to the left of each plot. Patients were classified as being at high risk of stroke if they had hypertension, diabetes, or prior cerebral ischemia. Patients without these risk factors were classified as being at low risk if they were younger than 65 years. All others were classified as being at moderate risk.18 TIA indicates transient ischemic attack.

The absolute rate reduction in ischemic stroke by oral anticoagulant vs aspirin varied with stroke risk (Figure 3). The absolute risk reduction in ischemic stroke achieved with oral anticoagulant was greater in patients at highest baseline risk of stroke. Patients with a previous history of stroke or transient ischemic attack had an absolute risk reduction of 6.0% per year (number needed to treat [NNT], 17) while those patients without previous cerebrovascular disease had an absolute risk reduction of 1.2% per year (NNT, 83). Patients who were classified as being at high risk of ischemic stroke by the AF investigators' criteria had an absolute risk reduction of 3.3% per year (NNT, 30), whereas those at low risk of ischemic stroke only had an absolute risk reduction of 0.4% per year (NNT not significant). In contrast to ischemic stroke, the absolute risk increase of major bleeding with oral anticoagulant vs aspirin was more consistent between patient subgroups.

Compared with aspirin, oral anticoagulant significantly decreased the risk of all strokes, ischemic strokes, and cardiovascular events in patients with nonvalvular paroxysmal or chronic AF. The absolute risk reductions of these events were not substantially offset by small but statistically significant increases in major hemorrhage. In these patients, the overall risk of stroke or cardiovascular events was substantially higher than that of major bleeding.

These results support the use of oral anticoagulant as first-line preventive therapy for high-risk patients with nonvalvular AF. Among trial participants, the absolute rate reduction for all stroke (2.1 events per 100 patient-years), ischemic stroke (2.3 events per 100-patient-years), and cerebrovascular events (2.2 events per 100 patient-years) was considerable. Since the absolute rate increase of major bleeding with oral anticoagulant is 0.9 events per 100 patient-years, treating 1000 AF patients for 1 year with oral anticoagulant rather than aspirin would prevent 23 ischemic strokes while causing 9 additional major bleeding episodes.

We believe that our results both highlight and clarify the trade-off between potential harms and benefits of oral anticoagulant prophylaxis for AF patients. Compared with aspirin, oral anticoagulant decreased the risk of ischemic stroke but increased the risk of major bleeding. We also found that the overall absolute risk of ischemic stroke was approximately twice that for major bleeding. Several studies show that patients with AF perceive stroke as a much more serious outcome than bleeding. Hing et al51 also found that patients had a strong aversion to strokes and would take warfarin if it reduced the absolute risk of stroke by 1% annually. The AF patients that Gage et al52 interviewed gave a median utility value of 0.03 for moderate strokes and 0.0 for major stroke, with death having a utility value of 0.52 In a study by Solomon et al,53 patients equated the utility of severe stroke with that of death. Devereaux et al54 found that the average patient would take warfarin if it reduced the absolute risk of stroke by at least 0.9 events per year but would only stop the drug if it increased the risk of bleeding by 8.7 events per year. These findings show that patients fear the consequences of having a stroke more than they fear the consequences of being prone to severe bleeding, which supports the use of oral anticoagulant treatment for such patients.

However, our results also highlight the importance of determining baseline risk to identify AF patients most likely to benefit from oral anticoagulant. There were some notable variations in the absolute difference of stroke rates (Figure 3). For patients at higher risk of ischemic stroke, such as the elderly or those with previous cerebrovascular disease, the absolute risk reduction in ischemic stroke with oral anticoagulant supercedes the associated bleeding risk. In contrast, patients at low risk of ischemic stroke have only a small absolute risk reduction in ischemic stroke. Therefore, proper risk stratification is essential to identify patients who have the best chance of benefiting from oral anticoagulant. This is especially true for patients without prior cerebrovascular disease for whom the absolute benefit of oral anticoagulant is less (Figure 3). Several risk stratification schemes have been published18,43,55,56 and further research is required to better refine our estimates of stroke risk in AF patients.

Three meta-analyses have compared oral anticoagulant with antiplatelet therapy for patients with AF.20,31,45 These analyses have summarized reported aggregate rates of trial outcome events. Our analysis used individual patient data, which allows a more complete and accurate comparison of outcome risk between oral anticoagulant and aspirin for patients with AF. Individual patient data permitted full survival analyses of a variety of relevant outcome events, not just ischemic stroke. We were able to provide absolute as well as relative comparisons of person-year rates of outcomes. Finally, analysis of pooled individual patient data allowed the exploration of treatment effects in relevant patient subgroups, such as those with paroxysmal AF. Because we pooled individual patient data, we had the opportunity to study 672 patients with paroxysmal AF. It is noteworthy that these patients had a similar reduction in stroke risk with oral anticoagulant (Figure 3).

Our results differ most from the meta-analysis by Taylor et al,30 which concluded that "the evidence for current clinical practice in long-term anticoagulation for patients with nonrheumatic atrial fibrillation is not strong." In addition to the methodological differences between the studies, other factors explain why our findings were considerably different. Taylor et al excluded the EAFT study5 and the high-risk component of the SPAF 3 study,9 both of which showed considerable advantage of oral anticoagulant over aspirin for all strokes, ischemic strokes, and cardiovascular events (Figure 2). The EAFT study was excluded from the analysis by Taylor et al because data were not presented to allow direct comparison of patients randomized to oral anticoagulant and aspirin.57 Our access to individual patient data avoided this problem. The SPAF 3 study was excluded because patients taking low-dose warfarin were included with those taking aspirin. We believe that it was preferable to include these patients in our analysis because including patients taking low-dose warfarin in an analysis with those taking aspirin patients would, if anything, decrease differences in cardiovascular events. Excluding these patients from the analysis did not alter our results. The only study that randomized patients to receive aspirin or low-intensity oral anticoagulant with aspirin found no significant differences in outcomes when low-dose warfarin was added to aspirin for patients with AF.3 Finally, their meta-analysis dichotomized outcomes into exclusive fatal and nonfatal events. This greatly decreases their statistical power to identify differences between treatment groups.

A potential limitation of our analyses stems from their uncertain applicability to AF patients outside of clinical trials. When compared with other patients, those enrolled in clinical trials tend to be healthier and more compliant. In addition, study investigators might more effectively monitor oral anticoagulant therapy than usual clinical caregivers. As a consequence, both bleeding and stroke rates might be higher among patients anticoagulated in general clinical practice. However, a recently published systematic review of anticoagulated AF patients in actual clinical practice found stroke and bleeding rates that were very similar to those found in randomized trials,58 supporting the external validity of our conclusions.

In summary, oral anticoagulant is more effective than aspirin in decreasing the risk of stroke and other cardiovascular events in patients with nonvalvular AF. Although oral anticoagulant also increases the risk of major bleeding, the frequency of strokes and the gravity of their consequences exceed that of therapy-associated bleeding events in most patients with AF. As a result, oral anticoagulant should be the preferred prophylactic treatment for patients with AF at a significant risk of thromboembolism.

Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation as an independent risk factor for stroke: the Framingham Study.  Stroke.1991;22:983-988.
Petersen P, Boysen G, Godtfredsen J, Andersen ED, Andersen B. Placebo-controlled, randomised trial of warfarin and aspirin for prevention of thromboembolic complications in chronic atrial fibrillation: the Copenhagen AFASAK study.  Lancet.1989;1:175-179.
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;158:1513-1521.
Hellemons BS, Langenberg M, Lodder J.  et al.  Primary prevention of arterial thromboembolism in non-rheumatic atrial fibrillation in primary care: randomised controlled trial comparing two intensities of coumarin with aspirin.  BMJ.1999;319:958-964.
EAFT (European Atrial Fibrillation Trial) Study Group.  Secondary prevention in non-rheumatic atrial fibrillation after transient ischaemic attack or minor stroke.  Lancet.1993;342:1255-1262.
 Stroke Prevention in Atrial Fibrillation Study: final results.  Circulation.1991;84:527-539.
 Warfarin vs aspirin for prevention of thromboembolism in atrial fibrillation: Stroke Prevention in Atrial Fibrillation II Study.  Lancet.1994;343:687-691.
The SPAF III Writing Committee for the Stroke Prevention in Atrial Fibrillation Investigators.  Patients with nonvalvular atrial fibrillation at low risk of stroke during treatment with aspirin: Stroke Prevention in Atrial Fibrillation III Study.  JAMA.1998;279:1273-1277.
 Adjusted-dose warfarin vs low-intensity, fixed-dose warfarin plus aspirin for high-risk patients with atrial fibrillation: Stroke Prevention in Atrial Fibrillation III randomised clinical trial.  Lancet.1996;348:633-638.
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;323:1505-1511.
Connolly SJ, Laupacis A, Gent M, Roberts RS, Cairns JA, Joyner C. Canadian Atrial Fibrillation Anticoagulation (CAFA) Study.  J Am Coll Cardiol.1991;18:349-355.
Ezekowitz MD, Bridgers SL, James KE.  et al. for the Veterans Affairs Stroke Prevention in Nonrheumatic Atrial Fibrillation Investigators.  Warfarin in the prevention of stroke associated with nonrheumatic atrial fibrillation.  N Engl J Med.1992;327:1406-1412.
Diener H, Cunha L, Forbes C.  et al.  European stroke prevention study 2: dipyridamole and acetylsalicylic acid in the prevention of stroke.  J Neurol Sci.1996;143:1-13.
Morocutti C, Amabile G, Fattapposta F.  et al. for the SIFA (Studio Italiano Fibrillazione Atriale) Investigators.  Indobufen vs warfarin in the secondary prevention of major vascular events in nonrheumatic atrial fibrillation.  Stroke.1997;28:1015-1021.
Posada IS, Barriales V.for the LASAF Pilot Study Group.  Alternate-day dosing of aspirin in atrial fibrillation.  Am Heart J.1999;138:137-143.
Pengo V, Zasso A, Barbero F.  et al.  Effectiveness of fixed minidose warfarin in the prevention of thromboembolism and vascular death in nonrheumatic atrial fibrillation.  Am J Cardiol.1998;82:433-437.
The Atrial Fibrillation Investigators.  The efficacy of aspirin in patients with atrial fibrillation: analysis of pooled data from 3 randomized trials.  Arch Intern Med.1997;157:1237-1240.
Atrial Fibrillation Investigators.  Risk factors for stroke and efficacy of antithrombotic therapy in atrial fibrillation: analysis of pooled data from five randomized controlled trials.  Arch Intern Med.1994;154:1449-1457.
 Echocardiographic predictors of stroke in patients with atrial fibrillation: a prospective study of 1066 patients from 3 clinical trials.  Arch Intern Med.1998;158:1316-1320.
Hart RG, Benavente O, McBride R, Pearce LA. Antithrombotic therapy to prevent stroke in patients with atrial fibrillation: a meta-analysis.  Ann Intern Med.1999;131:492-501.
Benavente O, Hart R, Koudstaal P, Laupacis A, McBride R. Oral anticoagulants for preventing stroke in patients with nonvalvular atrial fibrillation and no previous history of stroke or transient ischemic attacks.  Cochrane Database Syst Rev.2000;(2):CD001927.
Antithrombotic Trialists' Collaboration.  Collaborative meta-analysis of randomised trials of antiplatelet therapy for prevention of death, myocardial infarction, and stroke in high risk patients.  BMJ.2002;324:71-86.
Ezekowitz MD, Levine JA. Preventing stroke in patients with atrial fibrillation.  JAMA.1999;281:1830-1835.
Green CJ, Hadorn DC, Bassett K, Kazanjian A. Anticoagulation in chronic nonvalvular atrial fibrillation: a critical appraisal and meta-analysis.  Can J Cardiol.1997;13:811-815.
Hart RG, Halperin JL. Atrial fibrillation and stroke: concepts and controversies.  Stroke.2001;32:803-808.
Howard PA, Duncan PW. Primary stroke prevention in nonvalvular atrial fibrillation: implementing the clinical trial findings.  Ann Pharmacother.1997;31:1187-1196.
Koudstaal PJ. Anticoagulants vs antiplatelet therapy for preventing stroke in patients with nonrheumatic atrial fibrillation and a history of stroke or transient ischemic attacks.  Cochrane Database Syst Rev.2000;(2):CD000187.
Koudstaal PJ. Anticoagulants for preventing stroke in patients with nonrheumatic atrial fibrillation and a history of stroke or transient ischemic attacks.  Cochrane Database Syst Rev.2000;(2):CD000185.
Singer DE. Overview of the randomized trials to prevent stroke in atrial fibrillation.  Ann Epidemiol.1993;3:563-567.
Taylor FC, Cohen H, Ebrahim S. Systematic review of long term anticoagulation or antiplatelet treatment in patients with non-rheumatic atrial fibrillation.  BMJ.2001;322:321-326.
Segal JB, McNamara RL, Miller MR.  et al.  Anticoagulants or antiplatelet therapy for non-rheumatic atrial fibrillation and flutter.  Cochrane Database Syst Rev.2001;(1):CD001938.
 Practice parameter: Stroke prevention in patients with nonvalvular atrial fibrillation. Report of the Quality Standards Subcommittee of the American Academy of Neurology.  Neurology.1998;51:671-673.
Albers GW, Atwood JE, Hirsh J, Sherman DG, Hughes RA, Connolly SJ. Stroke prevention in nonvalvular atrial fibrillation.  Ann Intern Med.1991;115:727-736.
Albers GW, Sherman DG, Gress DR, Paulseth JE, Petersen P. Stroke prevention in nonvalvular atrial fibrillation: a review of prospective randomized trials.  Ann Neurol.1991;30:511-518.
Albers GW. Atrial fibrillation and stroke: three new studies, three remaining questions.  Arch Intern Med.1994;154:1443-1448.
Albers GW. Choice of antithrombotic therapy for stroke prevention in atrial fibrillation: warfarin, aspirin, or both?  Arch Intern Med.1998;158:1487-1491.
Dunn M, Alexander J, de Silva R, Hildner F. Antithrombotic therapy in atrial fibrillation.  Chest.1989;95(2 suppl):118S-127S.
Fuster V, Ryden LE, Asinger RW.  et al.  ACC/AHA/ESC guidelines for the management of patients with atrial fibrillation: executive summary. A Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines and the European Society of Cardiology Committee for Practice Guidelines and Policy Conferences (Committee to Develop Guidelines for the Management of Patients With Atrial Fibrillation): developed in Collaboration With the North American Society of Pacing and Electrophysiology.  J Am Coll Cardiol.2001;38:1231-1266.
Gage BF, Fihn SD, White RH. Warfarin therapy for an octogenarian who has atrial fibrillation.  Ann Intern Med.2001;134:465-474.
Gottlieb LK, Salem-Schatz S. Anticoagulation in atrial fibrillation: does efficacy in clinical trials translate into effectiveness in practice?  Arch Intern Med.1994;154:1945-1953.
Hart RG, Sherman DG, Easton JD, Cairns JA. Prevention of stroke in patients with nonvalvular atrial fibrillation.  Neurology.1998;51:674-681.
Kitchens JM, Flegel KM. Atrial fibrillation, stroke and anticoagulation: what is to be done?  J Gen Intern Med.1986;1:126-129.
Albers GW, Dalen JE, Laupacis A, Manning WJ, Petersen P, Singer DE. Antithrombotic therapy in atrial fibrillation.  Chest.2001;119(suppl 1):194S-206S.
The Stroke Prevention in Atrial Fibrillation Investigators.  Bleeding during antithrombotic therapy in patients with atrial fibrillation.  Arch Intern Med.1996;156:409-416.
Taylor FC, Ramsay ME, Renton A, Cohen H. Methods for managing the increased workload in anticoagulant clinics.  BMJ.1996;312:286.
Sudlow M, Thomson R, Thwaites B, Rodgers H, Kenny RA. Prevalence of atrial fibrillation and eligibility for anticoagulation in the community.  Lancet.1998;352:1167-1171.
Greenland S, Rothman KJ. Introduction to categorical statistics. In: Rothman KJ, Greenland S, eds. Modern Epidemiology. Philadelphia, Pa: Lippincott-Raven Publishers; 1998:253-280.
Harrell FE. Cox proportional hazards regression model. In: Regression Modelling Strategies. New York, NY: Springer; 2001:465-507.
Allison PD. Estimating Cox-regression models with PROC PHREG. In: Survival Analysis Using the SAS System. Cary, NC: SAS Institute Inc; 2000:111-184.
Takkouche B, Cadarso-Suarez C, Spiegelman D. Evaluation of old and new tests of heterogeneity in epidemiologic meta-analysis.  Am J Epidemiol.1999;150:206-215.
Man-Son-Hing M, Laupacis A, O'Connor A.  et al.  Warfarin for atrial fibrillation: the patient's perspective.  Arch Intern Med.1996;156:1841-1848.
Gage BF, Cardinalli AB, Owens DK. The effect of stroke and stroke prophylaxis with aspirin or warfarin on quality of life.  Arch Intern Med.1996;156:1829-1836.
Solomon NA, Glick HA, Russo CJ, Lee J, Schulman KA. Patient preferences for stroke outcomes.  Stroke.1994;25:1721-1725.
Devereaux PJ, Anderson DR, Gardner MJ.  et al.  Differences between perspectives of physicians and patients on anticoagulation in patients with atrial fibrillation: observational study.  BMJ.2001;323:1218-1222.
Gage BF, Waterman AD, Shannon W, Boechler M, Rich MW, Radford MJ. Validation of clinical classification schemes for predicting stroke: results from the national registry of atrial fibrillation.  JAMA.2001;285:2864-2870.
Stroke Prevention in Atrial Fibrillation Investigators.  Risk factors for thromboembolism during aspirin therapy in patients with atrial fibrillation: the stroke prevention in atrial fibrillation study.  J Stroke Cerebrovasc Dis.1995;5:147-157.
Ebrahim S, Taylor F. Review: anticoagulants may be better than antiplatelet agents for nonfatal stroke but not other vascular or fatal events in nonrheumatic AF.  ACP J Club.2001;135:A17.
Evans A, Kalra L. Are the results of randomized controlled trials on anticoagulation in patients with atrial fibrillation generalizable to clinical practice?  Arch Intern Med.2001;161:1443-1447.

Figures

Figure 1. Outcomes Survival Curves
Graphic Jump Location
In each plot, the horizontal axis represents time in years. The P value is a log-rank statistic. All strokes included ischemic and hemorrhagic events. Cardiovascular events included ischemic strokes, myocardial infarctions, systemic emboli, and vascular death. Major bleeding events included intracranial and major systemic bleeds.
Figure 2. Relationship of Therapy With Outcomes in Individual Trials
Graphic Jump Location
For each study, the relative effect of oral anticoagulants vs aspirin (with or without low-dose warfarin) is presented for all 6 outcomes as a hazard ratio. Hazard ratios below 1 indicate that oral anticoagulant decreases the risk of the event. Hazard ratios whose 95% confidence interval (error bars) excludes 1 are statistically significant at the 5% level. Because hemorrhagic strokes were uncommon events, a hazard ratio could not be estimated for each study individually. The P value for the DerSimonian and Laird Q statistic, as a measure of heterogeneity, is presented for each outcome in the top right-hand corner. AFASAK indicates Atrial Fibrillation, Aspirin, Anticoagulation study12,3; EAFT, European Atrial Fibrillation Trial5; PATAF, Primary Prevention of Atrial Thromboembolism in patients with Nonrheumatic Atrial Fibrillation in Primary Care4; SPAF, Stroke Prevention in Atrial Fibrillation studies.7,9
Figure 3. The Effect of Patient Baseline Factors on the Efficacy of Oral Anticoagulants or Aspirin on Ischemic Stroke and Major Bleeding
Graphic Jump Location
For subgroups in each patient factor (left), outcomes are presented as relative risks with 95% confidence intervals (error bars). Relative hazards that are less than 1 indicate that oral anticoagulants are associated with a decreased risk of the outcome. The event rates for each subgroup, expressed as the number of events per 100 person-years of observation, are presented to the left of each plot. Patients were classified as being at high risk of stroke if they had hypertension, diabetes, or prior cerebral ischemia. Patients without these risk factors were classified as being at low risk if they were younger than 65 years. All others were classified as being at moderate risk.18 TIA indicates transient ischemic attack.

Tables

Table Graphic Jump LocationTable 1. Description of Patients and Studies*
Table Graphic Jump LocationeTable 1. Definition of Baseline Factors*
Table Graphic Jump LocationTable 2. Patient Features by Treatment Group*
Table Graphic Jump LocationTable 3. Comparison of Oral Anticoagulants and Aspirin for Several Outcomes*

References

Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation as an independent risk factor for stroke: the Framingham Study.  Stroke.1991;22:983-988.
Petersen P, Boysen G, Godtfredsen J, Andersen ED, Andersen B. Placebo-controlled, randomised trial of warfarin and aspirin for prevention of thromboembolic complications in chronic atrial fibrillation: the Copenhagen AFASAK study.  Lancet.1989;1:175-179.
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;158:1513-1521.
Hellemons BS, Langenberg M, Lodder J.  et al.  Primary prevention of arterial thromboembolism in non-rheumatic atrial fibrillation in primary care: randomised controlled trial comparing two intensities of coumarin with aspirin.  BMJ.1999;319:958-964.
EAFT (European Atrial Fibrillation Trial) Study Group.  Secondary prevention in non-rheumatic atrial fibrillation after transient ischaemic attack or minor stroke.  Lancet.1993;342:1255-1262.
 Stroke Prevention in Atrial Fibrillation Study: final results.  Circulation.1991;84:527-539.
 Warfarin vs aspirin for prevention of thromboembolism in atrial fibrillation: Stroke Prevention in Atrial Fibrillation II Study.  Lancet.1994;343:687-691.
The SPAF III Writing Committee for the Stroke Prevention in Atrial Fibrillation Investigators.  Patients with nonvalvular atrial fibrillation at low risk of stroke during treatment with aspirin: Stroke Prevention in Atrial Fibrillation III Study.  JAMA.1998;279:1273-1277.
 Adjusted-dose warfarin vs low-intensity, fixed-dose warfarin plus aspirin for high-risk patients with atrial fibrillation: Stroke Prevention in Atrial Fibrillation III randomised clinical trial.  Lancet.1996;348:633-638.
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;323:1505-1511.
Connolly SJ, Laupacis A, Gent M, Roberts RS, Cairns JA, Joyner C. Canadian Atrial Fibrillation Anticoagulation (CAFA) Study.  J Am Coll Cardiol.1991;18:349-355.
Ezekowitz MD, Bridgers SL, James KE.  et al. for the Veterans Affairs Stroke Prevention in Nonrheumatic Atrial Fibrillation Investigators.  Warfarin in the prevention of stroke associated with nonrheumatic atrial fibrillation.  N Engl J Med.1992;327:1406-1412.
Diener H, Cunha L, Forbes C.  et al.  European stroke prevention study 2: dipyridamole and acetylsalicylic acid in the prevention of stroke.  J Neurol Sci.1996;143:1-13.
Morocutti C, Amabile G, Fattapposta F.  et al. for the SIFA (Studio Italiano Fibrillazione Atriale) Investigators.  Indobufen vs warfarin in the secondary prevention of major vascular events in nonrheumatic atrial fibrillation.  Stroke.1997;28:1015-1021.
Posada IS, Barriales V.for the LASAF Pilot Study Group.  Alternate-day dosing of aspirin in atrial fibrillation.  Am Heart J.1999;138:137-143.
Pengo V, Zasso A, Barbero F.  et al.  Effectiveness of fixed minidose warfarin in the prevention of thromboembolism and vascular death in nonrheumatic atrial fibrillation.  Am J Cardiol.1998;82:433-437.
The Atrial Fibrillation Investigators.  The efficacy of aspirin in patients with atrial fibrillation: analysis of pooled data from 3 randomized trials.  Arch Intern Med.1997;157:1237-1240.
Atrial Fibrillation Investigators.  Risk factors for stroke and efficacy of antithrombotic therapy in atrial fibrillation: analysis of pooled data from five randomized controlled trials.  Arch Intern Med.1994;154:1449-1457.
 Echocardiographic predictors of stroke in patients with atrial fibrillation: a prospective study of 1066 patients from 3 clinical trials.  Arch Intern Med.1998;158:1316-1320.
Hart RG, Benavente O, McBride R, Pearce LA. Antithrombotic therapy to prevent stroke in patients with atrial fibrillation: a meta-analysis.  Ann Intern Med.1999;131:492-501.
Benavente O, Hart R, Koudstaal P, Laupacis A, McBride R. Oral anticoagulants for preventing stroke in patients with nonvalvular atrial fibrillation and no previous history of stroke or transient ischemic attacks.  Cochrane Database Syst Rev.2000;(2):CD001927.
Antithrombotic Trialists' Collaboration.  Collaborative meta-analysis of randomised trials of antiplatelet therapy for prevention of death, myocardial infarction, and stroke in high risk patients.  BMJ.2002;324:71-86.
Ezekowitz MD, Levine JA. Preventing stroke in patients with atrial fibrillation.  JAMA.1999;281:1830-1835.
Green CJ, Hadorn DC, Bassett K, Kazanjian A. Anticoagulation in chronic nonvalvular atrial fibrillation: a critical appraisal and meta-analysis.  Can J Cardiol.1997;13:811-815.
Hart RG, Halperin JL. Atrial fibrillation and stroke: concepts and controversies.  Stroke.2001;32:803-808.
Howard PA, Duncan PW. Primary stroke prevention in nonvalvular atrial fibrillation: implementing the clinical trial findings.  Ann Pharmacother.1997;31:1187-1196.
Koudstaal PJ. Anticoagulants vs antiplatelet therapy for preventing stroke in patients with nonrheumatic atrial fibrillation and a history of stroke or transient ischemic attacks.  Cochrane Database Syst Rev.2000;(2):CD000187.
Koudstaal PJ. Anticoagulants for preventing stroke in patients with nonrheumatic atrial fibrillation and a history of stroke or transient ischemic attacks.  Cochrane Database Syst Rev.2000;(2):CD000185.
Singer DE. Overview of the randomized trials to prevent stroke in atrial fibrillation.  Ann Epidemiol.1993;3:563-567.
Taylor FC, Cohen H, Ebrahim S. Systematic review of long term anticoagulation or antiplatelet treatment in patients with non-rheumatic atrial fibrillation.  BMJ.2001;322:321-326.
Segal JB, McNamara RL, Miller MR.  et al.  Anticoagulants or antiplatelet therapy for non-rheumatic atrial fibrillation and flutter.  Cochrane Database Syst Rev.2001;(1):CD001938.
 Practice parameter: Stroke prevention in patients with nonvalvular atrial fibrillation. Report of the Quality Standards Subcommittee of the American Academy of Neurology.  Neurology.1998;51:671-673.
Albers GW, Atwood JE, Hirsh J, Sherman DG, Hughes RA, Connolly SJ. Stroke prevention in nonvalvular atrial fibrillation.  Ann Intern Med.1991;115:727-736.
Albers GW, Sherman DG, Gress DR, Paulseth JE, Petersen P. Stroke prevention in nonvalvular atrial fibrillation: a review of prospective randomized trials.  Ann Neurol.1991;30:511-518.
Albers GW. Atrial fibrillation and stroke: three new studies, three remaining questions.  Arch Intern Med.1994;154:1443-1448.
Albers GW. Choice of antithrombotic therapy for stroke prevention in atrial fibrillation: warfarin, aspirin, or both?  Arch Intern Med.1998;158:1487-1491.
Dunn M, Alexander J, de Silva R, Hildner F. Antithrombotic therapy in atrial fibrillation.  Chest.1989;95(2 suppl):118S-127S.
Fuster V, Ryden LE, Asinger RW.  et al.  ACC/AHA/ESC guidelines for the management of patients with atrial fibrillation: executive summary. A Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines and the European Society of Cardiology Committee for Practice Guidelines and Policy Conferences (Committee to Develop Guidelines for the Management of Patients With Atrial Fibrillation): developed in Collaboration With the North American Society of Pacing and Electrophysiology.  J Am Coll Cardiol.2001;38:1231-1266.
Gage BF, Fihn SD, White RH. Warfarin therapy for an octogenarian who has atrial fibrillation.  Ann Intern Med.2001;134:465-474.
Gottlieb LK, Salem-Schatz S. Anticoagulation in atrial fibrillation: does efficacy in clinical trials translate into effectiveness in practice?  Arch Intern Med.1994;154:1945-1953.
Hart RG, Sherman DG, Easton JD, Cairns JA. Prevention of stroke in patients with nonvalvular atrial fibrillation.  Neurology.1998;51:674-681.
Kitchens JM, Flegel KM. Atrial fibrillation, stroke and anticoagulation: what is to be done?  J Gen Intern Med.1986;1:126-129.
Albers GW, Dalen JE, Laupacis A, Manning WJ, Petersen P, Singer DE. Antithrombotic therapy in atrial fibrillation.  Chest.2001;119(suppl 1):194S-206S.
The Stroke Prevention in Atrial Fibrillation Investigators.  Bleeding during antithrombotic therapy in patients with atrial fibrillation.  Arch Intern Med.1996;156:409-416.
Taylor FC, Ramsay ME, Renton A, Cohen H. Methods for managing the increased workload in anticoagulant clinics.  BMJ.1996;312:286.
Sudlow M, Thomson R, Thwaites B, Rodgers H, Kenny RA. Prevalence of atrial fibrillation and eligibility for anticoagulation in the community.  Lancet.1998;352:1167-1171.
Greenland S, Rothman KJ. Introduction to categorical statistics. In: Rothman KJ, Greenland S, eds. Modern Epidemiology. Philadelphia, Pa: Lippincott-Raven Publishers; 1998:253-280.
Harrell FE. Cox proportional hazards regression model. In: Regression Modelling Strategies. New York, NY: Springer; 2001:465-507.
Allison PD. Estimating Cox-regression models with PROC PHREG. In: Survival Analysis Using the SAS System. Cary, NC: SAS Institute Inc; 2000:111-184.
Takkouche B, Cadarso-Suarez C, Spiegelman D. Evaluation of old and new tests of heterogeneity in epidemiologic meta-analysis.  Am J Epidemiol.1999;150:206-215.
Man-Son-Hing M, Laupacis A, O'Connor A.  et al.  Warfarin for atrial fibrillation: the patient's perspective.  Arch Intern Med.1996;156:1841-1848.
Gage BF, Cardinalli AB, Owens DK. The effect of stroke and stroke prophylaxis with aspirin or warfarin on quality of life.  Arch Intern Med.1996;156:1829-1836.
Solomon NA, Glick HA, Russo CJ, Lee J, Schulman KA. Patient preferences for stroke outcomes.  Stroke.1994;25:1721-1725.
Devereaux PJ, Anderson DR, Gardner MJ.  et al.  Differences between perspectives of physicians and patients on anticoagulation in patients with atrial fibrillation: observational study.  BMJ.2001;323:1218-1222.
Gage BF, Waterman AD, Shannon W, Boechler M, Rich MW, Radford MJ. Validation of clinical classification schemes for predicting stroke: results from the national registry of atrial fibrillation.  JAMA.2001;285:2864-2870.
Stroke Prevention in Atrial Fibrillation Investigators.  Risk factors for thromboembolism during aspirin therapy in patients with atrial fibrillation: the stroke prevention in atrial fibrillation study.  J Stroke Cerebrovasc Dis.1995;5:147-157.
Ebrahim S, Taylor F. Review: anticoagulants may be better than antiplatelet agents for nonfatal stroke but not other vascular or fatal events in nonrheumatic AF.  ACP J Club.2001;135:A17.
Evans A, Kalra L. Are the results of randomized controlled trials on anticoagulation in patients with atrial fibrillation generalizable to clinical practice?  Arch Intern Med.2001;161:1443-1447.

Letters

CME
Meets CME requirements for:
Browse CME for all U.S. States
Accreditation Information
The American Medical Association is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. The AMA designates this journal-based CME activity for a maximum of 1 AMA PRA Category 1 CreditTM per course. Physicians should claim only the credit commensurate with the extent of their participation in the activity. Physicians who complete the CME course and score at least 80% correct on the quiz are eligible for AMA PRA Category 1 CreditTM.
Note: You must get at least of the answers correct to pass this quiz.
You have not filled in all the answers to complete this quiz
The following questions were not answered:
Sorry, you have unsuccessfully completed this CME quiz with a score of
The following questions were not answered correctly:
Commitment to Change (optional):
Indicate what change(s) you will implement in your practice, if any, based on this CME course.
Your quiz results:
The filled radio buttons indicate your responses. The preferred responses are highlighted
For CME Course: A Proposed Model for Initial Assessment and Management of Acute Heart Failure Syndromes
Indicate what changes(s) you will implement in your practice, if any, based on this CME course.

Multimedia

Some tools below are only available to our subscribers or users with an online account.

Web of Science® Times Cited: 287

Related Content

Customize your page view by dragging & repositioning the boxes below.

Articles Related By Topic
Related Collections
PubMed Articles
JAMAevidence.com

Users' Guides to the Medical Literature
Clinical Resolution

Users' Guides to the Medical Literature
Clinical Resolution