This Article Abstract Full Text (PDF) All Versions of this Article: 110/16/2287    most recent 01.CIR.0000145172.55640.93v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Gage, B. F. Articles by Petersen, P. Search for Related Content PubMed PubMed Citation Articles by Gage, B. F. Articles by Petersen, P. Related Collections Anticoagulants Antiplatelets Coumarins Thrombosis risk factors Embolic stroke Risk Factors for Stroke Related Article

(Circulation. 2004;110:2287-2292.)
© 2004 American Heart Association, Inc.

Arrhythmia/Electrophysiology

Selecting Patients With Atrial Fibrillation for Anticoagulation

Stroke Risk Stratification in Patients Taking Aspirin

Brian F. Gage, MD, MSc; Carl van Walraven, MD, FRCPC, MSc; Lesly Pearce, MS; Robert G. Hart, MD; Peter J. Koudstaal, MD; B.S.P. Boode, MD; Palle Petersen, MD, PhD

From the Division of General Medical Sciences, Washington University School of Medicine, St. Louis, Mo (B.F.G.); Departments of Medicine and of Epidemiology and Community Medicine, University of Ottawa, and Ottawa Health Research Institute, Ottawa, Canada (C.v.W.); Biostatistical Consultant, Minot, ND (L.P.); Department of Medicine (Neurology), University of Texas Health Science Center, San Antonio, Tex (R.G.H.); Department of Neurology, Erasmus Medical Center, Rotterdam, the Netherlands (P.J.K.); Department of General Practice, University of Maastricht, Maastricht, the Netherlands (B.S.P.B.); and Department of Neurology, Copenhagen University Hospital, Copenhagen, Denmark (P.P.).

Correspondence to Brian F. Gage, MD, MSc, Division of General Medical Sciences, Washington University School of Medicine Campus Box 8005, 660 S Euclid Ave, St. Louis, MO 63110. E-mail bgage{at}im.wustl.edu

Received December 30, 2003; de novo received March 27, 2004; revision received June 30, 2004; accepted August 3, 2004.

	Abstract

Top Abstract Introduction Methods Results Discussion Appendix References

 
Background— The rate of stroke in atrial fibrillation (AF) depends on the presence of comorbid conditions and the use of antithrombotic therapy. Although adjusted-dose warfarin is superior to aspirin for reducing stroke in AF, the absolute risk reduction of warfarin depends on the stroke rate with aspirin. This prospective cohort study tested the predictive accuracy of 5 stroke risk stratification schemes.

Methods and Results— The study pooled individual data from 2580 participants with nonvalvular AF who were prescribed aspirin in a multicenter trial (Atrial Fibrillation, Aspirin, Anticoagulation I study [AFASAK-1], AFASAK-2, European Atrial Fibrillation Trial, Primary Prevention of Arterial Thromboembolism in patients with nonrheumatic Atrial Fibrillation in primary care study, and Stroke Prevention and Atrial Fibrillation [SPAF]-III high risk or SPAF-III low risk). There were 207 ischemic strokes during 4887 patient-years of aspirin therapy. All schemes predicted stroke better than chance, but the number of patients categorized as low and high risk varied substantially. AF patients with prior cerebral ischemia were classified as high risk by all 5 schemes and had 10.8 strokes per 100 patient-years. The CHADS₂ scheme (an acronym for Congestive heart failure, Hypertension, Age >75, Diabetes mellitus, and prior Stroke or transient ischemic attack) successfully identified primary prevention patients who were at high risk of stroke (5.3 strokes per 100 patient-years). In contrast, patients identified as high risk by other schemes had 3.0 to 4.2 strokes per 100 patient-years. Low-risk patients identified by all schemes had 0.5 to 1.4 strokes per 100 patient-years of therapy.

Conclusions— Patients with AF who have high and low rates of stroke when given aspirin can be reliably identified, allowing selection of antithrombotic prophylaxis to be individualized.

Key Words: anticoagulants • aspirin • atrial fibrillation • risk factors • stroke

	Introduction

Top Abstract Introduction Methods Results Discussion Appendix References

 
The rate of stroke in nonvalvular atrial fibrillation (AF) ranges widely and depends on the presence of prior cerebral ischemia, comorbid conditions, and use of antithrombotic therapy. Without antithrombotic therapy, the rate varies from fewer than 2 to more than 10 strokes per 100 patient-years.^1–7 Although adjusted-dose warfarin is superior to aspirin for reducing stroke in AF patients, the absolute risk reduction is determined by the stroke risk with aspirin therapy.^8,9

Thus, quantifying the risk of stroke is crucial for determining which AF patients would benefit most from warfarin therapy. Patients whose risk is less than 2 strokes per 100 patient-years with aspirin have little to gain from warfarin therapy; for these patients, the risks of warfarin tend to outweigh any benefits.^9–11 A risk stratification scheme that reliably identified these low-risk patients could spare them the risks, inconvenience, and costs associated with anticoagulation. In contrast, for AF patients whose risk exceeds 4 strokes per 100 patient-years of aspirin therapy, warfarin therapy consistently improves quality-adjusted survival.^4,10,12 Between these 2 extremes are patients for whom the key issue is whether stroke risk can be quantified reliably so that antithrombotic therapy can be judiciously selected on the basis of stroke risk, hemorrhage risk, and individual preferences.^10,13,14

Risk stratification schemes that accurately and reliably stratify stroke risk could influence the antithrombotic management of millions of people who have AF. Because patients with a prior stroke or transient ischemic attack (TIA) are at high risk of stroke, the greatest need is to quantify stroke risk in the AF primary prevention population. Here, we use prospective data to test the predictive accuracy of 5 widely available stroke risk stratification schemes and provide reliable guidance to clinicians selecting antithrombotic therapy.

	Methods

Top Abstract Introduction Methods Results Discussion Appendix References

 
Risk Stratification Schemes
Multivariate analyses of prospective cohorts of AF patients who were prescribed aspirin or no antithrombotic therapy yielded independent predictors of stroke that formed the basis of 5 previously published risk stratification schemes (Appendix).^2,4,15,16 In 1994, the Atrial Fibrillation Investigators (AFI) conducted a multivariate analysis of pooled data from 1593 untreated AF participants in 5 randomized clinical trials.⁴ Participants with prior cerebral ischemia (either stroke or TIA), hypertension, or diabetes mellitus were at high risk of stroke; patients without these risk factors were at moderate risk of stroke if older than 65 years and at low risk otherwise (Table 1).

View this table: [in this window] [in a new window]   TABLE 1. Stroke Risk Stratification Schemes

The Stroke Prevention and Atrial Fibrillation (SPAF) investigators developed a classification scheme from 854 SPAF I and II participants treated with aspirin.¹⁵ Four factors independently predicted a high risk of stroke: prior cerebral ischemia, the combination of age greater than 75 years plus female gender, left ventricular dysfunction (defined as recent clinical heart failure or left ventricular fractional shortening 25% by echocardiography), and systolic blood pressure >160 mm Hg (Table 1). SPAF participants with a history of hypertension but blood pressure 160 mm Hg were found to have a moderate risk (3 strokes per 100 patient-years), ^5,15 and participants with none of these factors were at low risk of stroke (Table 1).

In 1998 and 2001, the American College of Chest Physicians (ACCP) Consensus Conference on Antithrombotic Therapy classified patients as high risk if they had prior cerebral ischemia (or systemic embolism), hypertension, congestive heart failure (either clinical heart failure or poor systolic function on echocardiography), age >75 years, or at least 2 moderate-risk factors.^16,17 The moderate-risk factors were age 65 to 75 years, diabetes mellitus, and coronary artery disease. Patients with 1 moderate-risk factor were classified as moderate risk, and patients with none of the risk factors were classified as low risk.

In 2001, an amalgamation of the AFI and SPAF schemes led to the CHADS₂ scheme.² The CHADS₂ acronym was derived from the individual stroke risk factors: Congestive heart failure, Hypertension, Age >75 years, Diabetes mellitus, and prior Stroke or TIA. Two points were given for prior stroke or TIA (hence, the subscripted "2"), and 1 point was assigned for each of the other factors. The point system was designed to simplify the determination of stroke risk in general practice. CHADS₂ stroke rates (Table 1) were calculated in 1733 Medicare beneficiaries in the National Registry of Atrial Fibrillation using International Classification of Diseases codes 434 (occlusion of cerebral arteries), 435 (transient cerebral ischemia), and 436 (acute, but ill-defined, cerebrovascular disease). The expected stroke rate (95% CI) per 100 patient-years without antithrombotic therapy increases by a factor of 1.5 for each 1-point increase in the CHADS₂ score: 1.9 (1.2 to 3.0) for a score of 0; 2.8 (2.0 to 3.8) for 1; 4.0 (3.1 to 5.1) for 2; 5.9 (4.6 to 7.3) for 3; 8.5 (6.3 to 11.1) for 4; 12.5 (8.2 to 17.5) for 5; and 18.2 (10.5 to 27.4) for 6.

In 2003, Wang et al¹⁸ developed a risk classification scheme based on 868 Framingham participants, some of whom were taking warfarin or aspirin therapy. Using the coefficients from a Cox proportional survival model, they developed a point system based on age (0 to 10 points), gender (6 points for female; 0 for male), blood pressure (0 to 4 points), diabetes mellitus (4 points), and prior stroke or TIA (6 points) to develop a scheme to predict the combination of ischemic plus hemorrhagic stroke (Table 1). Whether the Framingham scheme will predict ischemic stroke in other AF populations is not clear.

For a variety of reasons, we evaluated only these 5 risk stratification schemes and excluded others. We excluded schemes that were based entirely on retrospective data.^19,20 We excluded 2 schemes^21,22 that were based on data used in the present analysis and another because the other study focused exclusively on secondary prevention.⁷ We excluded the initial SPAF I scheme⁶ because it was superseded by the subsequent SPAF and AFI schemes, both of which used SPAF I data. We excluded schemes that required the use of echocardiography to risk-stratify patients^23–27 because we did not have echocardiographic results on all participants and because we wished to validate a scheme that could predict stroke on the basis of clinical risk factors.

Description of the Validation Population
Participants with nonvalvular AF who took aspirin at dosages ranging between 75 and 325 mg daily in 6 prospective trials made up the validation cohort (Table 2). To validate the risk stratification schemes in an independent cohort of AF patients, patient data that were used to derive any of the 4 classification schemes were excluded from this analysis.

View this table: [in this window] [in a new window]   TABLE 2. Study Participants

We used patient data from 6 prospective randomized trials. In 4 trials, participants were prescribed aspirin alone: the Atrial Fibrillation, Aspirin, Anticoagulation I (AFASAK-I; n=336) study,²⁸ the Primary Prevention of Arterial Thromboembolism in patients with nonrheumatic Atrial Fibrillation in primary care study (PATAF; n=319), ²⁹ the European Atrial Fibrillation Trial (EAFT; n=404), ³⁰ and the low-risk SPAF III study (n=891; 1 additional SPAF patient was excluded for missing information).⁵ From the fifth trial, AFASAK-2, we included participants who were prescribed aspirin, either alone (n=169) or in combination with an ineffective, 1.25-mg dose of warfarin (n=171).³¹ From the sixth trial, high-risk SPAF-III,³² we included participants (n=290) who were prescribed aspirin in combination with low-dose warfarin (median dose 2 mg/d) if their international normalized ratio never exceeded 1.4 during follow-up. Adherence to aspirin therapy exceeded 85% in the studies in which it was reported.^5,29,33

Research coordinators and physicians recorded baseline patient characteristics at the time of enrollment in the original trials. We classified participants into the appropriate strata of each scheme using baseline clinical characteristics (Appendix). These characteristics included age, previous stroke or TIA, hypertension, cardiac disease, and diabetes. Patients were classified as hypertensive if they were taking medications given to lower blood pressure or if blood pressure exceeded 140/90 mm Hg. Because echocardiographic information was not available, we were unable to apply one of the stroke risk factors (decreased systolic function) of the ACCP and SPAF schemes. Likewise, because we could not ascertain the duration of congestive heart failure, we substituted any congestive heart failure for recent heart failure in the SPAF and CHADS₂ schemes.

During follow-up in these trials, physicians assessed patients with suspected stroke. To identify etiology, a brain CT scan was done in 98% of incident neurological events. Strokes were defined as neurological deficits that persisted for more than 24 hours and that were not associated with an intracranial hemorrhage.

Statistical Analysis
Incidence rates for ischemic stroke were calculated as the number of strokes per 100 patient-years of observation.³⁴ For this calculation, observation started when participants were randomized to aspirin and ended when they experienced an ischemic stroke or were censored. Participants were censored for nonstroke death or study termination. To calculate 95% CIs, we used the Poisson distribution.

We used the weighted -statistic to compare the agreement between schemes corrected for chance agreement.³⁵ Because the -statistic requires the same number of risk strata for all schemes, we collapsed CHADS₂ scores into 3 strata (0, 1 to 2, and 3 to 6) and the Framingham scores into 3 strata (0 to 7, 8 to 13, and >13) for this comparison.

We quantified the discrimination of each stratification scheme using the log-rank test, the Wald ² statistic from a Cox proportional hazards model, and the c-statistic.³⁶ Discrimination is the ability of the stratification schemes to separate the AF population into strata that have distinct stroke rates.³⁷ A c-statistic of 1.0 indicates perfect discrimination, which means that patients with different stroke risks are always correctly distinguished, whereas a value of 0.5 is noninformative. We bootstrapped 200 samples (with replacement) of the ² and c-statistic values using a publicly available macro (http://ftp.sas.com/techsup/download/stat/) and compared the differences values between schemes by paired t tests. Statistical analyses were done with SAS software (SAS Institute Inc). All statistical tests were 2 sided, and probability values <0.05 were considered significant.

The Institutional Review Board of participating institutions approved the study.

	Results

Top Abstract Introduction Methods Results Discussion Appendix References

 
Characteristics of the Validation Cohort
The 2580 participants had a mean age of 72 years, 38% were female, and 22% had suffered a prior stroke or TIA (Table 2). The most frequent risk factor was hypertension, which was present in 46%, and the least common was diabetes, with a 13% prevalence. Participants were followed up for a mean of 1.9 years (maximum 6.6 years). During 4887 patient-years of follow-up, there were 207 ischemic strokes, for an overall incidence rate of 4.2 strokes per 100 patient-years during aspirin therapy. The rate among the 2014 primary prevention participants without prior cerebral ischemia was 2.5 strokes per 100 patient-years of aspirin; for the 566 participants with a prior stroke or TIA, the rate was 10.8 per 100 patient-years of aspirin.

Stroke Rates According to Predicted Risk for Each Scheme
All schemes stratified the risk of ischemic stroke significantly better than chance (log-rank P<0.001 for all schemes), but the number of AF patients categorized as at high, moderate, and low risk varied substantially (Table 3). The agreement between schemes was variable, with weighted -values ranging from a low of 0.13 (ACCP versus Framingham) to a high of 0.58 (ACCP versus AFI).

View this table: [in this window] [in a new window]   TABLE 3. Validation of Stratification Schemes for Primary Prevention of Stroke in 2014 Participants Prescribed Aspirin

Comparison of the Classification Schemes
The stroke rates (95% CI) per 100 patient-years of aspirin rose with increasing CHADS₂ scores: 0.8 (0.4 to 1.7; n=469) with 0 points; 2.2 (1.6 to 3.1; n=752) with 1 point; 4.5 (3.5 to 5.9; n=670) with 2 points; 8.6 (6.8 to 11.0; n=428) with 3 points; 10.9 (7.8 to 15.2; n=200) with 4 points; 12.3 (6.6 to 22.9; n=56) with 5 points; and 13.7 (2 to 97; n=5) with 6 points. Rates (95% CI) per 100 patient-years of aspirin in other high-risk patients were lower than in the highest CHADS₂ cohorts: 6.1 (5.3 to 7.1) by AFI criterion, 6.5 (5.6 to 7.6) by SPAF, 5.1 (4.4 to 5.8) by ACCP, and 7.9 (6.5 to 9.7) for Framingham score >13.

Among primary prevention participants, CHADS₂ identified participants at high risk for stroke: primary prevention participants with 3 or 4 points averaged 5.3 (95% CI 3.3 to 8.4) strokes per 100 patient-years. In contrast, patients identified by other schemes as high risk had rates of 3.0 to 4.2 strokes per 100 patient-years (Table 3). The use of a higher Framingham threshold (>15 rather than >13 points; Table 3) identified 144 participants whose stroke rate was only 3.9.

A Cox proportional hazards model quantified the ability to discriminate between low- and high-risk patients by the likelihood ratio ² test. The ² (SD) was 67 (16) for AFI, 73 (16) for SPAF, 44 (11) for ACCP, 98 (19) for CHADS₂, and 89 (20) for Framingham (P<0.001 for CHADS₂ versus the other schemes). Collapsing CHADS₂ into 3 strata (0, 1 to 2, and 3 to 6) yielded a ² value of 86, and collapsing Framingham into 3 strata (0 to 7, 8 to 13, and >13) yielded a value of 69. When only primary prevention patients were analyzed, ² values were lower, but the pattern was similar: AFI 18, SPAF 17, ACCP 17, CHADS₂ 22 (20 with 3 strata), and Framingham 16 (P<0.001 for CHADS₂ versus other schemes).

The c-statistics (SD) were 0.63 (0.01) for AFI, 0.64 (0.01) for SPAF, 0.58 (0.01) for ACCP, 0.70 (0.02) for CHADS₂, and 0.69 (0.02) for Framingham (P<0.001 for CHADS₂ versus other schemes). When participants with a prior stroke or TIA were excluded, c-statistics (SD) were 0.61 (0.02) for AFI, 0.61 (0.02) for SPAF, 0.58 (0.02) for ACCP, 0.63 (0.03) for CHADS₂, and 0.62 (0.03) for Framingham.

Identification of Patients Whose Stroke Rate Was Low With Aspirin Therapy
ACCP criteria classified the fewest participants as low risk (n=182). In contrast, SPAF classified 668 participants as low risk (Table 3). Primary prevention participants with a Framingham score of 4 or less (n=502) or 7 or less (n= 983) both averaged 1.4 strokes per 100 patient-years. Primary prevention participants with 7 or fewer Framingham points who were not considered low risk by SPAF averaged 2.2 strokes per 100 patient-years; primary prevention participants with 7 or fewer Framingham points who had 1 or more CHADS₂ points averaged 1.9 strokes per 100 patient-years.

	Discussion

Top Abstract Introduction Methods Results Discussion Appendix References

 
This study of 2580 participants to whom aspirin had been prescribed confirms that AF populations with high and low stroke risks can be identified prospectively. Patients with a prior stroke or TIA averaged 10.8 strokes per 100 patient-years despite aspirin therapy. For these patients, it is clear that the benefits of anticoagulant therapy outweigh the risks.^30,32,38

Primary prevention patients whose stroke risk exceeds 4 per 100 patient-years of aspirin also benefit from warfarin therapy.^4,10,12 These patients were reliably identified by a CHADS₂ score 3. Such patients averaged 5.3 strokes per 100 patient-years of aspirin. The number needed to treat with warfarin instead of aspirin for 1 year to prevent 1 stroke would be 30 for these patients.^9,12 High-risk primary prevention patients identified by the other schemes had stroke rates of only 3.0 to 4.2.

In contrast, all schemes successfully identified low-risk patients whose stroke rate was 1.4 or lower per 100 patient-years of aspirin, but the agreement between schemes was poor. Experts and patients typically prefer aspirin to warfarin when the risk is less than 2 strokes per 100 patient-years of aspirin.^9–11 For these AF patients, the number needed to treat with warfarin for 1 year to prevent 1 stroke exceeds 100. The ability to characterize low-risk AF patients with confidence allows clinicians to identify patients who can safely be treated with aspirin, sparing them the risk of bleeding, cost, and inconvenience from anticoagulant therapy.^39,40 Although the Framingham scheme identified the largest fraction of low-risk patients (almost half of the primary prevention cohort had a Framingham score of 7 or less), the additional low-risk patients identified had 2 strokes per 100 patient-years, a rate substantially greater than other low-risk cohorts.

For patients whose stroke risk is 2 to 4 per 100 patient-years of aspirin therapy, many experts offer warfarin, ^41,42 whereas others offer aspirin,^43,44 depending on risk of hemorrhage and patient preferences. In clinical trials, warfarin increased the risk of major hemorrhage 1.7-fold compared with aspirin.⁹ Outside of trials, the risk of hemorrhage was greater,^45,46 depending on how warfarin was monitored⁴⁷ and risk factors for hemorrhage.⁴⁸ How patients trade off the risk of stroke, risk of hemorrhage, and the aggravation of taking and monitoring anticoagulant therapy depends on individual preferences.^10,13,14

The use of data from clinical trial cohorts confers both strengths and limitations to the present study. One strength is that similar sets of comorbid conditions were collected at baseline across different trials. A second strength is that ischemic strokes were identified prospectively by clinical examination and confirmed by computerized tomography. A third is that all patients received aspirin, which allowed us to quantify the stroke rate with this ubiquitous, inexpensive therapy. A fourth strength is that none of the patients included in these analyses were included as part of the derivation cohorts for any of the schemes. Finally, because the schemes were derived primarily from patients assigned to no antithrombotic therapy, the present study demonstrates that the schemes (especially CHADS₂) are valid predictors of stroke in patients prescribed aspirin.

One limitation is that participants with contraindications to warfarin therapy were included in only 1 of the 6 trials. The inclusion of more of these patients would have provided greater generalizability, but such patients were excluded from clinical trials. Second, echocardiographic results were not available, which would have allowed us to assess whether they would have improved the predictive accuracy of the schemes that consider left ventricular systolic dysfunction as a stroke risk factor. Thus, the SPAF criteria for impaired left ventricular function could not be tested, and a history of heart failure was used instead. A finding of significant systolic dysfunction by echocardiography primarily would be relevant to patients at low risk of stroke on the basis of clinical factors.^24,49

Recent retrospective studies of other AF populations further validate CHADS₂. For example, enrollees of Kaiser Permanente (Northern California) who were not prescribed warfarin (4% of whom had a prior stroke) had a very low stroke rate (0.5 per 100 patient-years) if their CHADS₂ score was 0. Their stroke rates were greater with greater CHADS₂ scores: 1.5 for 1 point, 2.5 for 2 points, 5.3 for 3 points, 6.0 for 4 points, and 6.9 for 5 or 6 points.⁴⁶ Although other schemes were not evaluated by these studies, it confirms the ability of CHADS₂ to identify low- and high-risk patients reliably.

In the future, a more accurate prediction rule for stroke may be possible by incorporating additional factors. For example, left ventricular systolic dysfunction detected by 2D transthoracic echocardiography is an independent risk factor for stroke in AF.²⁴ Also, it seems likely that hormone replacement therapy^21,50 and cigarette smoking⁵¹ increase the risk of stroke in AF, whereas modest alcohol consumption may decrease it.²¹ Finally, future studies will determine whether biochemical markers of inflammation (eg, C-reactive protein) or endothelial dysfunction (eg, von Willebrand factor) will help clinicians predict stroke in the AF population.

	Appendix

Top Abstract Introduction Methods Results Discussion Appendix References

 

View this table: [in this window] [in a new window]   TABLE 4. Definitions Used by Stroke Risk Stratification Schemes

	Acknowledgments

 
This work was supported by an award from the American Heart Association (AHA #0270099N). The data used for this study were funded by a variety of sources. The National Institutes of Health (R01 NS 24224) funded the SPAF III studies (principal investigator: Dr Hart). The Danish Heart Foundation funded the Atrial Fibrillation, Aspirin, Anticoagulation I and II trials (principal investigators: Dr Petersen). The Zorg Onderzoek Nederland Prevention fund (grant 002817010) funded the Primary Prevention of Arterial Thromboembolism in patients with nonrheumatic Atrial Fibrillation (principal investigator: Dr Boode). The Netherlands Heart Foundation, Bayer Germany, the UK Stroke Association, University Hospital Utrecht, and University Hospital Rotterdam funded the European Atrial Fibrillation Trial (principal investigator: Dr Koudstaal). We thank Dr Andreas Laupacis for his leadership in combining the patient-level data from these trials. We thank Dr Gregory Albers for his comments on an earlier draft of this manuscript and Dr Elena Deych for performing the bootstrapping statistical analysis. The writing committee for this article consisted of the first 4 authors (Dr Gage, Dr van Walraven, L. Pearce, and Dr Hart); all authors provided critical review of the manuscript. We are grateful for the assistance of Dr Annette Lemche.

Disclosure

Several of the authors were involved in the development of the risk stratification schemes tested in these analyses: AFI (Drs Hart and Petersen), SPAF (Drs Hart and L. Pearce), and CHADS₂ (Dr Gage).

	References

Top Abstract Introduction Methods Results Discussion Appendix References

 

1. Pearce LA, Hart RG, Halperin JL. Assessment of three schemes for stratifying stroke risk in patients with nonvalvular atrial fibrillation. Am J Med. 2000; 109: 45–51.[CrossRef][Medline] [Order article via Infotrieve]
   
2. Gage BF, Waterman AD, Shannon W, et al. Validation of clinical classification schemes for predicting stroke: results from the National Registry of Atrial Fibrillation. JAMA. 2001; 285: 2864–2870.[Abstract/Free Full Text]
   
3. Kopecky SL, Gersh BJ, McGoon MD, et al. Lone atrial fibrillation in elderly persons: a marker for cardiovascular risk. Arch Intern Med. 1999; 159: 1118–1122.[Abstract/Free Full Text]
   
4. Atrial Fibrillation Investigators. Risk factors for stroke and efficacy of antithrombotic therapy in atrial fibrillation. Arch Intern Med. 1994; 154: 1449–1457.[Abstract]
   
5. Stroke Prevention Atrial Fibrillation III Writing Committee. 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.[Abstract/Free Full Text]
   
6. Stroke Prevention in Atrial Fibrillation Investigators. Predictors of thromboembolism in atrial fibrillation, I: clinical features of patients at risk. Ann Intern Med. 1992; 116: 1–5.[Medline] [Order article via Infotrieve]
   
7. van Latum JC, Koudstaal PJ, Venables GS, et al. Predictors of major vascular events in patients with a transient ischemic attack or minor ischemic stroke and with nonrheumatic atrial fibrillation. Stroke. 1995; 26: 801–806.[Abstract/Free Full Text]
   
8. Gage BF, Cardinalli AB, Albers GW, et al. Cost-effectiveness of warfarin and aspirin for prophylaxis of stroke in patients with nonvalvular atrial fibrillation. JAMA. 1995; 274: 1839–1845.[Abstract]
   
9. van Walraven C, Hart RG, Singer DE, et al. Oral anticoagulants vs aspirin in nonvalvular atrial fibrillation: an individual patient meta-analysis. JAMA. 2002; 288: 2441–2448.[Abstract/Free Full Text]
   
10. Gage BF, Cardinalli AB, Owens D. Cost-effectiveness of preference-based antithrombotic therapy for patients with nonvalvular atrial fibrillation. Stroke. 1998; 29: 1083–1091.[Abstract/Free Full Text]
    
11. Glasziou PP, Irwig L. An evidence based approach to individualising treatment. BMJ. 1995; 311: 1356–1359.[Free Full Text]
    
12. Hart RG, Benavente O, McBride R, et al. Antithrombotic therapy to prevent stroke in patients with atrial fibrillation: a meta-analysis. Ann Intern Med. 1999; 131: 492–501.[Abstract/Free Full Text]
    
13. Thomson R, Parkin D, Eccles M, et al. Decision analysis and guidelines for anticoagulant therapy to prevent stroke in patients with atrial fibrillation. Lancet. 2000; 355: 956–962.[CrossRef][Medline] [Order article via Infotrieve]
    
14. Protheroe J, Fahey T, Montgomery AA, et al. The impact of patients’ preferences on the treatment of atrial fibrillation: observational study of patient based decision analysis. BMJ. 2000; 320: 1380–1384.[Abstract/Free Full Text]
    
15. 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.
    
16. Albers GW, Dalen JE, Laupacis A, et al. Antithrombotic therapy in atrial fibrillation. Chest. 2001; 119: 194S–206S.[Free Full Text]
    
17. Laupacis A, Albers G, Dalen J, et al. Antithrombotic therapy in atrial fibrillation. Chest. 1998; 114: 579S–589S.[Medline] [Order article via Infotrieve]
    
18. Wang TJ, Massaro JM, Levy D, et al. A risk score for predicting stroke or death in individuals with new-onset atrial fibrillation in the community: the Framingham Heart Study. JAMA. 2003; 290: 1049–1056.[Abstract/Free Full Text]
    
19. Moulton AW, Singer DE, Hass J. Risk factors for stroke in patients with nonrheumatic atrial fibrillation: a case-control study. Am J Med. 1991; 91: 156–161.[CrossRef][Medline] [Order article via Infotrieve]
    
20. Flegel KM, Hanley J. Risk factors for stroke and other embolic events in patients with non-rheumatic atrial fibrillation. Stroke. 1989; 20: 1000–1004.[Abstract/Free Full Text]
    
21. Hart RG, Pearce LA, McBride R, et al. Factors associated with ischemic stroke during aspirin therapy in atrial fibrillation: analysis of 2012 participants in the SPAF I-III clinical trials. Stroke. 1999; 30: 1223–1229.[Abstract/Free Full Text]
    
22. van Walraven C, Hart RG, Wells GA, et al. A clinical prediction rule to identify patients with atrial fibrillation and a low risk for stroke while taking aspirin. Arch Intern Med. 2003; 163: 936–943.[Abstract/Free Full Text]
    
23. Cabin HS, Clubb KS, Hall C, et al. Risk for systemic embolization of atrial fibrillation without mitral stenosis. Am J Cardiol. 1990; 65: 1112–1116.[CrossRef][Medline] [Order article via Infotrieve]
    
24. Atrial Fibrillation Investigators. 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.[Abstract/Free Full Text]
    
25. Stollberger C, Chnupa P, Kronik G, et al. Transesophageal echocardiography to assess embolic risk in patients with atrial fibrillation. Ann Intern Med. 1998; 128: 630–638.[Abstract/Free Full Text]
    
26. Stroke Prevention in Atrial Fibrillation Investigators. Predictors of thromboembolism in atrial fibrillation, II: echocardiographic features of patients at risk. Ann Intern Med. 1992; 116: 6–12.[Medline] [Order article via Infotrieve]
    
27. Stroke Prevention in Atrial Fibrillation Investigators Committee on Echocardiography. Transesophageal echocardiographic correlates of thromboembolism in high-risk patients with nonvalvular atrial fibrillation. Ann Intern Med. 1998; 128: 639–647.[Abstract/Free Full Text]
    
28. Petersen P, Boysen G, Godtfredsen J, et al. 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.[Medline] [Order article via Infotrieve]
    
29. 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.[Abstract/Free Full Text]
    
30. The 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.[Medline] [Order article via Infotrieve]
    
31. Gulløv 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.[Abstract/Free Full Text]
    
32. Stroke Prevention in Atrial Fibrillation Investigators. Adjusted-dose warfarin versus 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.[CrossRef][Medline] [Order article via Infotrieve]
    
33. Stroke Prevention in Atrial Fibrillation Investigators. Stroke Prevention in Atrial Fibrillation (SPAF) Study: final results. Circulation. 1991; 84: 527–539.[Abstract/Free Full Text]
    
34. Rothman KJ, Greenland S. Modern Epidemiology. 2nd ed. Philadelphia, Pa: Lippincott-Raven; 1998.
    
35. Kraemer HC, Periyakoil VS, Noda A. Kappa coefficients in medical research. Stat Med. 2002; 21: 2109–2129.[CrossRef][Medline] [Order article via Infotrieve]
    
36. Therneau TM, Grambsch PM. Modeling Survival Data: Extending the Cox Model. New York, NY: Springer; 2000.
    
37. Justice AC, Covinsky KE, Berlin JA. Assessing the generalizability of prognostic information. Ann Intern Med. 1999; 130: 515–524.[Medline] [Order article via Infotrieve]
    
38. Morocutti C, Amabile G, Fattapposta F, et al. Indobufen versus warfarin in the secondary prevention of major vascular events in nonrheumatic atrial fibrillation. Stroke. 1997; 28: 1015–1021.[Abstract/Free Full Text]
    
39. Gage BF, Cardinalli AB, Owens D. The effect of stroke and stroke prophylaxis with aspirin or warfarin on quality of life. Arch Intern Med. 1996; 156: 1829–1836.[Abstract]
    
40. Hart RG, Halperin JL. Atrial fibrillation and thromboembolism: a decade of progress in stroke prevention. Ann Intern Med. 1999; 131: 688–695.[Abstract/Free Full Text]
    
41. Singer DE. Patients with atrial fibrillation at low risk of stroke. JAMA. 1998; 280: 882–883.[Free Full Text]
    
42. Albers G. Choice of antithrombotic therapy for stroke prevention in atrial fibrillation: warfarin, aspirin, or both? Arch Intern Med. 1998; 158: 1487–1491.[Free Full Text]
    
43. Lip GY. Thromboprophylaxis for atrial fibrillation. Lancet. 1999; 353: 4–6.[CrossRef][Medline] [Order article via Infotrieve]
    
44. Sweeney KG, Gray DP, Steele R, et al. Use of warfarin in non-rheumatic atrial fibrillation: a commentary from general practice. Br J Gen Pract. 1995; 45: 153–158.[Medline] [Order article via Infotrieve]
    
45. Landefeld SC, Beyth R. Anticoagulant-related bleeding: clinical epidemiology, prediction and prevention. Am J Med. 1993; 95: 315–328.[CrossRef][Medline] [Order article via Infotrieve]
    
46. Go AS, Hylek EM, Chang Y, et al. Anticoagulation therapy for stroke prevention in atrial fibrillation: how well do randomized trials translate into clinical practice? JAMA. 2003; 290: 2685–2692.[Abstract/Free Full Text]
    
47. Beyth RJ, Quinn L, Landefeld CS. A multicomponent intervention to prevent major bleeding complications in older patients receiving warfarin: a randomized, controlled trial. Ann Intern Med. 2000; 133: 687–695.[Abstract/Free Full Text]
    
48. Beyth RJ, Milligan PE, Gage BF. Risk factors for bleeding in patients taking coumarins. Curr Hematol Rep. 2002; 1: 41–49.[Medline] [Order article via Infotrieve]
    
49. Kalra L, Perez I, Melbourn A. Risk assessment and anticoagulation for primary stroke prevention in atrial fibrillation. Stroke. 1999; 30: 1218–1222.[Abstract/Free Full Text]
    
50. Rossouw JE, Anderson GL, Prentice RL, et al. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results from the Women’s Health Initiative randomized controlled trial. JAMA. 2002; 288: 321–333.[Abstract/Free Full Text]
    
51. Hylek EM, Skates SJ, Sheehan MA, et al. An analysis of the lowest effective intensity of prophylactic anticoagulation for patients with nonrheumatic atrial fibrillation. N Engl J Med. 1996; 335: 540–546.[Abstract/Free Full Text]
    

Related Article:

Circulation 2004 110: 2277. [Full Text]

This article has been cited by other articles:

				P M Rothwell Prognostic models Practical Neurology, August 1, 2008; 8(4): 242 - 253. [Abstract] [Full Text] [PDF]

				J. D. Douketis, P. B. Berger, A. S. Dunn, A. K. Jaffer, A. C. Spyropoulos, R. C. Becker, and J. Ansell The Perioperative Management of Antithrombotic Therapy: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edition) Chest, June 1, 2008; 133(6_suppl): 299S - 339S. [Abstract] [Full Text] [PDF]

				D. E. Singer, G. W. Albers, J. E. Dalen, M. C. Fang, A. S. Go, J. L. Halperin, G. Y. H. Lip, and W. J. Manning Antithrombotic Therapy in Atrial Fibrillation: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edition) Chest, June 1, 2008; 133(6_suppl): 546S - 592S. [Abstract] [Full Text] [PDF]

				Stroke Risk in Atrial Fibrillation Working Group Comparison of 12 Risk Stratification Schemes to Predict Stroke in Patients With Nonvalvular Atrial Fibrillation Stroke, June 1, 2008; 39(6): 1901 - 1910. [Abstract] [Full Text] [PDF]

				J. S. Healey, R. G. Hart, J. Pogue, M. A. Pfeffer, S. H. Hohnloser, R. De Caterina, G. Flaker, S. Yusuf, and S. J. Connolly Risks and Benefits of Oral Anticoagulation Compared With Clopidogrel Plus Aspirin in Patients With Atrial Fibrillation According to Stroke Risk: The Atrial Fibrillation Clopidogrel Trial With Irbesartan for Prevention of Vascular Events (ACTIVE-W) Stroke, May 1, 2008; 39(5): 1482 - 1486. [Abstract] [Full Text] [PDF]

				G. Y.H. Lip The Balance Between Stroke Prevention and Bleeding Risk in Atrial Fibrillation: A Delicate Balance Revisited Stroke, May 1, 2008; 39(5): 1406 - 1408. [Full Text] [PDF]

				C. L Garwood and T. L Corbett Use of Anticoagulation in Elderly Patients with Atrial Fibrillation Who Are at Risk for Falls Ann. Pharmacother., April 1, 2008; 42(4): 523 - 532. [Abstract] [Full Text] [PDF]

				M. I. Aguilar and R. Hart Oral Anticoagulants Versus Antiplatelet Therapy for Preventing Stroke in Patients With Nonvalvular Atrial Fibrillation and No History of Stroke or Transient Ischemic Attacks Stroke, April 1, 2008; 39(4): 1399 - 1400. [Full Text] [PDF]

				A D S Ryding and A P Banning Which patients receiving warfarin can be treated safely with a drug-eluting stent? Heart, March 1, 2008; 94(3): 275 - 277. [Full Text] [PDF]

				N U Weir An update on cardioembolic stroke Postgrad. Med. J., March 1, 2008; 84(989): 133 - 142. [Abstract] [Full Text] [PDF]

				M. C. Fang, A. S. Go, Y. Chang, L. Borowsky, N. K. Pomernacki, D. E. Singer, and for the ATRIA Study Group Comparison of Risk Stratification Schemes to Predict Thromboembolism in People With Nonvalvular Atrial Fibrillation J. Am. Coll. Cardiol., February 26, 2008; 51(8): 810 - 815. [Abstract] [Full Text] [PDF]

				N.A. M. Estes III, J. L. Halperin, H. Calkins, M. D. Ezekowitz, P. Gitman, A. S. Go, R. L. McNamara, J. V. Messer, J. L. Ritchie, S. J.W. Romeo, et al. ACC/AHA/Physician Consortium 2008 Clinical Performance Measures for Adults With Nonvalvular Atrial Fibrillation or Atrial Flutter: A Report of the American College of Cardiology/American Heart Association Task Force on Performance Measures and the Physician Consortium for Performance Improvement (Writing Committee to Develop Clinical Performance Measures for Atrial Fibrillation) Developed in Collaboration With the Heart Rhythm Society J. Am. Coll. Cardiol., February 26, 2008; 51(8): 865 - 884. [Full Text] [PDF]

				N. A. M. Estes III, J. L. Halperin, H. Calkins, M. D. Ezekowitz, P. Gitman, A. S. Go, R. L. McNamara, J. V. Messer, J. L. Ritchie, S. J.W. Romeo, et al. ACC/AHA/Physician Consortium 2008 Clinical Performance Measures for Adults With Nonvalvular Atrial Fibrillation or Atrial Flutter: A Report of the American College of Cardiology/American Heart Association Task Force on Performance Measures and the Physician Consortium for Performance Improvement (Writing Committee to Develop Clinical Performance Measures for Atrial Fibrillation) Circulation, February 26, 2008; 117(8): 1101 - 1120. [Full Text] [PDF]

				J. S. Taggar, T. Watson, and G. Y.H. Lip Can Patients With Atrial Fibrillation Be Optimally Risk Stratified for Stroke and Thromboembolism? Stroke, February 1, 2008; 39(2): e24 - e25. [Full Text] [PDF]

				E. A. Nutescu ASHP Therapeutic Position Statement on Antithrombotic Therapy in Chronic Atrial Fibrillation Am. J. Health Syst. Pharm., November 1, 2007; 64(21): 2281 - 2291. [Full Text] [PDF]

				L. Friberg, N. Hammar, H. Pettersson, and M. Rosenqvist Increased mortality in paroxysmal atrial fibrillation: report from the Stockholm Cohort-Study of Atrial Fibrillation (SCAF) Eur. Heart J., October 1, 2007; 28(19): 2346 - 2353. [Abstract] [Full Text] [PDF]