U.S. flag

An official website of the United States government

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Sanders GD, Lowenstern A, Borre E, et al. Stroke Prevention in Patients With Atrial Fibrillation: A Systematic Review Update [Internet]. Rockville (MD): Agency for Healthcare Research and Quality (US); 2018 Oct. (Comparative Effectiveness Reviews, No. 214.)

Cover of Stroke Prevention in Patients With Atrial Fibrillation: A Systematic Review Update

Stroke Prevention in Patients With Atrial Fibrillation: A Systematic Review Update [Internet].

Show details

Evidence Summary

Introduction

This systematic review is an update of an earlier report published in 2013 which evaluated questions related to stroke prevention in patients with atrial fibrillation (AF) and atrial flutter. Given evidence that has emerged since the publication of the 2013 report, this review focuses on updating and expanding the earlier work in three key areas: (1) evaluating the accuracy and utility of clinical tools and imaging tools to predict thromboembolic event risk, (2) evaluating the accuracy and utility of clinical tools used to predict bleeding risk, and (3) exploring the comparative safety and effectiveness of various interventions to prevent thromboembolic events in patients with nonvalvular atrial fibrillation (Figure A). In addition, this review explores the strengths and weaknesses of shared decisionmaking tools available to aid patients and clinicians in selecting an intervention to prevent stroke.

Figure A. Analytic framework.

Figure A

Analytic framework. Abbreviations: AF=atrial fibrillation; DVT=deep vein thrombosis; KQ=Key Question; ICH=intracranial hemorrhage; PE=pulmonary embolism

Results/Key Findings

Accuracy and Utility of Clinical and Imaging Tools To Predict Stroke Risk

  • CHADS2 score (continuous): Based on a meta-analysis of 14 studies (10 low risk of bias, 4 medium risk of bias, 761,128 patients), there is moderate strength of evidence (SOE) that the continuous CHADS2 score provides limited prediction of stroke events (c-statistic of 0.69; 95% confidence interval [CI] 0.66 to 0.73).
  • CHADS2 score (categorical): Based on a meta-analysis of 16 studies (11 low risk of bias, 5 medium risk of bias, 548,464 patients), there is moderate SOE that the categorical CHADS2 score provides limited prediction of stroke events (c-statistic of 0.66; 95% CI 0.63 to 0.69).
  • CHA2DS2-VASc (continuous): Based on a meta-analysis of 17 studies (13 low risk of bias, 4 medium risk of bias; 511,481 patients), there is moderate SOE that the continuous CHA2DS2-VASc score provides limited prediction of stroke events (c-statistic of 0.67; 95% CI 0.64 to 0.70).
  • CHA2DS2-VASc (categorical): Based on a meta-analysis of 13 studies (8 low risk of bias, 5 medium risk of bias; 496,683 patients), there is low SOE that the categorical CHA2DS2-VASc score provides limited prediction of stroke events (c-statistic of 0.64; 95% CI 0.58 to 0.70).
  • Framingham score (categorical): Based on a meta-analysis of 6 studies (5 low risk of bias, 1 medium risk of bias; 282,572 patients), there is moderate SOE that the categorical Framingham score provides limited prediction of stroke events (c-statistic of 0.63; 95% CI 0.62 to 0.65).
  • ABC score (categorical): Based on a meta-analysis of 4 studies (4 low risk of bias, 25,614 patients), there is moderate SOE that the categorical ABC score provides limited prediction of stroke events (c-statistic of 0.67; 95% CI 0.63 to 0.71).
  • Echocardiography: There is insufficient evidence for the relationship between findings on echocardiography (transthoracic) and subsequent stroke based on 5 studies (3 low risk of bias, 2 medium risk of bias; 1,228 patients) that reported discrepant results.
  • Comparative accuracy: CHADS2 and CHA2DS2-VASc have the most evidence predicting stroke events accurately when directly compared with other scores. This finding was, however, statistically significant only for the comparison with the Framingham categorical score. Other comparisons were not possible given limited data.
  • Limitations: Included studies used heterogeneous populations; some participants were on and some were off antiplatelets and anticoagulants at baseline. Also, few studies used clinical validation in their report of stroke rates, instead relying on administrative data, chart review, or other measures that did not use consistent definitions and were not similar across studies, complicating synthesis of their findings. Furthermore, although event rates were consistently reported, c-statistics and measures of calibration, strength of association, and diagnostic accuracy were inconsistently reported.
  • The outcome of impact on clinical decisionmaking (diagnostic thinking, therapeutic efficacy, and patient outcome efficacy) was not assessed by any studies.

Accuracy and Utility of Clinical Tools To Predict Bleeding Risk

  • AF patients on warfarin: 13 studies (10 low risk of bias, 2 medium risk of bias, 1 high risk of bias; 197,312 patients) compared different risk scores (Bleeding Risk Index [BRI], HEMORR2HAGES, HAS-BLED, ATRIA, ABC) in predicting major bleeding events. These studies differed markedly in population, major bleeding rates, and statistics reported for evaluating risk prediction scores for major bleeding events. Evidence favors HAS-BLED based on two studies demonstrating that it has statistically significantly higher prediction (by c-statistic) for major bleeding events than other scores among patients on warfarin, but the majority of comparative studies which evaluated HAS-BLED showed no statistically significant differences in prediction abilities, reducing the strength of evidence (moderate SOE).
  • Chronic kidney disease (CKD) and major bleeding: Eight studies (7 low risk of bias, 1 medium risk of bias; 322,010 patients) evaluated the risk of major bleeding in patients with CKD. All studies demonstrated increased risk of bleeding in patients with CKD (moderate SOE) although do not formally evaluate the use of a tool incorporating CKD.
  • AF patients on warfarin: 1 study (low risk of bias; 48,599 patients) compared HEMORR2HAGES and HAS-BLED in predicting intracranial hemorrhage (ICH). This study showed no statistically significant difference in prediction abilities between the two scores (low SOE).
  • AF patients on aspirin alone: Three studies (2 low risk of bias, 1 medium risk of bias; 177,538 patients) comparing different combinations of bleeding risk scores (BRI, HEMORR2HAGES, and HAS-BLED) in predicting major bleeding events showed no statistically significant differences (low SOE).
  • AF patients not on therapy: Six studies (4 low risk of bias, 2 medium risk of bias; 310,607 patients) comparing different combinations of bleeding risk scores (BRI, HEMORR2HAGES, HAS-BLED, and ATRIA) in predicting major bleeding events showed no statistically significant differences (low SOE).
  • Limitations: Although studies consistently reported event rates and c-statistics, measures of tool calibration, strength of association, and diagnostic accuracy were inconsistently reported.
  • The outcome of impact on clinical decisionmaking (diagnostic thinking and therapeutic efficacy) was not assessed by any studies.

Comparative Safety and Effectiveness of Interventions To Prevent Thromboembolic Events

  • Acetylsalicylic acid (ASA) versus vitamin K antagonist (VKA; warfarin): Based on 5 observational studies involving 251,578 patients, warfarin reduces the risk of nonfatal and fatal ischemic stroke compared with aspirin (moderate SOE); however, based on 3 studies involving 212,770 patients, warfarin is also associated with increased rates of major bleeding complications compared with aspirin (moderate SOE)
  • ASA+clopidogrel versus ASA: In patients not eligible for warfarin, two good quality RCTs involving 8,147 patients showed lower rates of any stroke (HR 0.72, 95% CI 0.62 to 0.83) for combination therapy of aspirin and clopidogrel compared to ASA alone (moderate SOE). In the largest RCT (7,554 patients), the combination of aspirin and clopidogrel was associated with higher rates of major bleeding than aspirin alone (HR 1.57, 95% CI 1.29 to 1.92) (moderate SOE).
  • Warfarin versus clopidogrel: Based on 1 large observational, good quality study involving 54,636 patients, warfarin reduces the risk of nonfatal and fatal ischemic stroke compared with clopidogrel monotherapy, with no evidence of differences in major bleeding (moderate SOE).
  • ASA+clopidogrel versus warfarin: Based on two large, good-quality RCTs involving 60,484 patients, warfarin is superior to aspirin plus clopidogrel for the prevention of stroke or systemic embolism (high SOE). In one good quality RCT of 6,706 patients, warfarin is superior to aspirin plus clopidogrel for the reduction in any minor bleeding (moderate SOE) however warfarin increased hemorrhagic stroke risk compared to ASA+ clopidogrel (moderate SOE). There was no evidence of a difference between therapies for MI, death from vascular causes or all-cause mortality (moderate SOE for both outcomes).
  • Clopidogrel+warfarin versus warfarin: Clopidogrel+warfarin shows a trend toward a benefit on stroke prevention (low SOE) and is associated with increased risk of nonfatal and fatal bleeding compared with warfarin alone (moderate SOE). These findings are based on 1 good-quality observational study involving 52,349 patients.
  • Warfarin+aspirin+clopidogrel versus warfarin: Triple therapy increases the risk of nonfatal and fatal bleeding (moderate SOE) and also shows a trend toward increased ischemic stroke (low SOE) compared with warfarin alone. These findings are based on 1 good-quality observational study involving 52,180 patients
  • Thrombin inhibitors (dabigatran) versus warfarin: Based on 1 large good-quality RCT involving 18,113 patients and 35 observational studies involving 1,737,961 patients we found:
    • Dabigatran at a 150mg dose is superior to warfarin in reducing the incidence of the composite outcome of stroke (including hemorrhagic) or systemic embolism (RR 0.66, 95% CI 0.53 to 0.82), with no statistically significant difference in the occurrence of major bleeding (RR 0.93, 95% CI 0.81 to 1.07) (high SOE for both outcomes), all-cause mortality(RR 0.88, 95% CI 0.77 to 1.00) (low SOE), or myocardial infarction (MI) risk (low SOE).
    • Dabigatran at a 110mg dose is similar to warfarin for the composite outcome of stroke or systemic embolism (RR 0.91, 95% CI 0.74 to 1.11) (moderate SOE). It is associated with a reduction in the risk of major bleeding (RR 0.80, 95% CI 0.69 to 0.93) when compared with warfarin (high SOE), but there is no evidence of a difference in all-cause mortality or MI risk (low SOE for both outcomes). Note the 110mg dose is currently not approved for stroke prevention in patients with AF in the US.
    • Observational studies were inconsistent with RCT evidence for the outcomes of all-cause mortality (observational studies demonstrated a benefit for patients on dabigatran, while RCT studies suggested no evidence of a difference on either dose) and MI risk (observational studies did not show a difference, RCT studies suggested an increase with the 150mg dose of dabigatran).
  • Xa inhibitor (apixaban) versus ASA: Apixaban is superior to aspirin in reducing the incidence of stroke or systemic embolism (HR 0.45, 95% CI 0.32 to 0.62) with similar major bleeding risk (HR 1.13, 95% CI 0.74 to 1.75), in patients who are not suitable for warfarin (moderate SOE for both outcomes). These findings are based on 1 good quality RCT involving 5,599 patients.
  • Xa inhibitor (apixaban) versus warfarin: Apixaban is superior in reducing the incidence of (1) stroke or systemic embolism (HR 0.79, 95% CI 0.66 to 0.95) (high SOE), (2) the risk of major bleeding (0.69, 95% CI 0.60 to 0.80) (high SOE), and (3) all-cause mortality (low SOE) when compared with warfarin. These findings are based on 1 large good-quality RCT involving 18,201 patients, and 29 observational studies with 1,251,855 patients.
  • Xa inhibitor (rivaroxaban) versus warfarin: Rivaroxaban is similar to warfarin in preventing stroke or systemic embolism (HR 0.88, 95% CI 0.74 to 1.03) (moderate SOE), with similar rates of major bleeding (low SOE) and all-cause mortality (moderate SOE). These findings are based on 1 large, good-quality RCT involving 14,264 patients and 26 observational studies with 1,483,949 patients. Inconsistent with the RCT findings, observational studies supported a reduction in stroke or systemic embolism and a trend towards a reduction in ischemic or uncertain stroke, while also providing evidence of a small increase in the risk of major bleeding.
  • Xa inhibitor (edoxaban) versus warfarin: Edoxaban (either 60mg or 30mg dose) is superior in reducing hemorrhagic stroke (low dose HR 0.33, 95% CI 0.22 to 0.50; high dose HR 0.54, 95% CI 0.38 to 0.77) (moderate SOE) and the risk of major bleeding (moderate SOE) though did not differ in overall stroke risk (moderate SOE), myocardial infarction (moderate SOE) or all-cause mortality (moderate SOE for high dose). There was low SOE that low dose edoxaban (30 mg) reduced all-cause mortality. These findings are based on 1 large, good-quality RCT involving 21,105 patients. Note that the 60 mg once-daily dose of edoxaban is approved by the FDA to treat only NVAF patients with creatinine clearance (CrCL) >50 to ≤ 95 mL/min, while 30 mg once-daily dose of edoxaban is approved to treat NVAF in patients with renal dysfunction (CrCL 15 to 50 mL/min).
  • Percutaneous left atrial appendage (LAA) closure versus warfarin: LAA shows a trend toward a benefit over warfarin for all strokes (including ischemic or hemorrhagic) and all-cause mortality (low SOE for both outcomes). Although LAA with percutaneous closure results in less frequent major bleeding than warfarin (low SOE), it is also associated with a higher rate of adverse safety events such as pericardial effusion and device embolization (moderate SOE). These findings are based on 1 good-quality RCT involving 707 patients and 4 observational studies involved 1,430 patients.

Discussion

Additional details about this systematic review are described in Table A.

Observational Studies Versus RCT Evidence

  • Within the included set of observational studies, use of direct oral anticoagulants and comparative effectiveness analyses of the different oral anticoagulants often have inconsistent findings. These inconsistencies likely resulted from confounding, selection bias, different endpoint definitions, rigor and completeness of followup, and variations in decisionmaking practice between trial populations and real world scenarios.
  • When considered together, the findings from observational and RCT studies were inconsistent related to all-cause mortality and myocardial infarction for dabigatran versus warfarin.
    • The observational studies demonstrated a benefit in all-cause mortality for patients on dabigatran compared with warfarin. RCT evidence, however did not demonstrate evidence of a difference. In addition, observational studies did not show a difference in myocardial infarction while RCT studies suggested an increase with dabigatran.
  • Xa inhibitors (all-cause mortality): The observational studies did not show a reduction in all-cause mortality across Xa inhibitors, whereas RCTs showed reduction in all-cause mortality across Xa inhibitors.
  • Other RCT findings were supported by existing observational studies.

Shared Decisionmaking Tools

  • While many publications have described decision support tools for anticoagulation for patients with nonvalvular AF, these tools are all early in development, haven’t been validated, and the tools are not in clinical use.
  • Future studies are required to evaluate how decision aids influence actual choices and clinical outcomes.

Key Limitations and Research

Gaps

  • For risk prediction tools, further studies are needed that: (1) report complete data across the full continuous range of scores; (2) use validated clinical outcomes for stroke and bleeding; and (3) compare all available risk scores using consistent and appropriate statistical evaluations such as c-statistics.
  • There is a need for a tool that could be used for decisionmaking about antithrombotic therapy in AF patients taking into account both thromboembolic and bleeding risks.
  • Additional studies utilizing prospectively constructed databases (registries) with longer-term outcomes data that compare all available risk prediction tools would be of great use in better clarifying which risk score system is superior in predicting major bleeding or thromboembolic risk.
  • It is important to have new studies with head-to-head comparisons of direct oral anticoagulants (DOACs). Given variability in patient populations, concomitant therapies, and underlying patient care, indirect comparisons across RCTs in this field is of limited use.
  • There are also many novel invasive treatments for treating AF such as left atrial appendage (LAA) closure devices but the evidence remains sparse about these interventions in terms of stroke prevention. Studies need to be conducted in patients who receive these procedures to determine if and how anticoagulation strategies should be modified in patients receiving these procedures.
  • An area worthy of further study is the use of the direct oral anticoagulants in specific populations of patients such as those with severe kidney disease (end-stage renal disease), older adults, patients with comorbid diseases, or frail patients.
Table A. Summary of review characteristics.

Table A

Summary of review characteristics.

Views

  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (5.8M)

In this Page

Other titles in this collection

Recent Activity

ClearTurn OffTurn On

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...