Text Size:    +   -

Anticoagulation for Individuals with Stroke and Atrial Fibrillation

October 2017

 Note: These recommendations focus on atrial fibrillation in the context of secondary prevention of stroke. For information on the primary prevention of stroke in individuals with non-valvular atrial fibrillation (AF), please refer to the Canadian Cardiovascular Society Atrial Fibrillation Guidelines 2016: Prevention of Stroke and Systemic Thromboembolism in Atrial Fibrillation and Flutter (October 2016).

Note: These recommendations are applicable to ischemic stroke and transient ischemic attack. 

Nonvalvular atrial fibrillation refers to atrial fibrillation in the absence of rheumatic mitral stenosis, a mechanical or bioprosthetic heart valve, or mitral valve repair (CCS 2016).

DOAC – Direct Oral AntiCoagulant

7.1 Detection of Atrial Fibrillation

  1. Patients with suspected transient ischemic attack or ischemic stroke should have a 12-lead ECG to assess cardiac rhythm and identify atrial fibrillation or flutter or evidence of structural heart disease (e.g. myocardial infarction, left ventricular hypertrophy) [Evidence Level B].
  2. For patients being investigated for an acute embolic ischemic stroke or TIA of undetermined source, ECG monitoring at least 24 hours is recommended as part of the initial stroke work-up to detect paroxysmal atrial fibrillation in patients who would be potential candidates for anticoagulant therapy [Evidence Level A].
  3. For patients being investigated for an acute embolic ischemic stroke or TIA of undetermined source whose initial short-term ECG monitoring does not reveal atrial fibrillation but a cardioembolic mechanism is suspected, prolonged ECG monitoring for at least 2 weeks is recommended to improve detection of paroxysmal atrial fibrillation in selected patients who are not already receiving anticoagulant therapy but would be potential anticoagulant candidates [Evidence Level A].

7.2 Prevention of recurrent stroke in patients with non-valvular atrial fibrillation

  1. Patients with transient ischemic attack or ischemic stroke and non-valvular atrial fibrillation should receive oral anticoagulation [Evidence Level A]. Refer to Appendix Four for additional information on selection of anticoagulant medications.
    1. In most patients requiring anticoagulants for atrial fibrillation, direct non-vitamin K oral anticoagulants (DOAC) such as apixaban, dabigatran, edoxaban, or rivaroxaban should be prescribed in preference over warfarin [Evidence Level A].
    2. For patients already receiving warfarin with good International Normalized Ratio (INR) control (Range 2.0 – 3.0, with TTR >70%), continuing warfarin is a reasonable anticoagulation option [Evidence Level B].
    3. When selecting choice of oral anticoagulants, patient specific criteria should be considered [Evidence level C].  Refer to Summary Table for Selection of Anticoagulant Agents for Management of Atrial Fibrillation after stroke or transient ischemic attack , available at www.strokebestpractices.ca.
  2. For patients with acute ischemic stroke and atrial fibrillation, routine use of bridging with heparin is not recommended [Evidence Level B].
    1. Bridging with antiplatelet therapy is suggested until the patient is anticoagulated [Evidence Level C]. Refer to Prevention of Stroke Section 6 on Antiplatelet Therapy for Ischemic Stroke and Transient Ischemic Attack for additional recommendations on antithrombotic therapy.
  3. For patients with ischemic stroke or TIA and atrial fibrillation who are unable to take oral anticoagulant therapy (DOAC or warfarin), aspirin alone is recommended [Evidence Level A]. (New for 2017)
    1. The addition of clopidogrel to aspirin therapy, compared with aspirin therapy alone, may be reasonable and decisions should be individualized based on patient bleeding risk [Evidence Level B].
  4. For patients with a mechanical heart valve, warfarin is recommended for stroke prevention with careful INR monitoring; non-vitamin K oral anticoagulants are contraindicated [Evidence Level B].
  5. For patients in whom long-term anticoagulant therapy is contraindicated, a left atrial appendage closure procedure may be considered [Evidence Level B].

Clinical Considerations (new 2017):

  1. The optimal timing to start anticoagulant therapy after stroke has not been defined by clinical trial evidence, and should be based on individual benefit/risk assessment taking into account the clinical circumstances, infarct size, imaging appearances, age, comorbidities, and estimated stroke recurrence risk.
  2. According to expert consensus, a general approach to the target timing of initiation of oral anticoagulant therapy poststroke is as follows: 1 day post-event after a TIA, 3 days poststroke after a mild stroke, 6 days poststroke after a moderate stroke, and 12 days poststroke after a severe stroke. (Heidbuchel et al. EHRA practical guide on the use of new oral anticoagulants in patients with non-valvular atrial fibrillation: executive summary. European heart journal. 2013;34(27):2094-2106).

7.3 Enhancing anticoagulant therapy effectiveness in practice and minimizing bleeding complications

  1. Medication adherenceshould be continually assessed and reinforced for patients on all oral anticoagulants at each follow-up visit [Evidence Level B].
    1. For patients with atrial fibrillation taking warfarin, careful dosing and consistent international normalized ratio monitoring is recommended to minimize adverse events; warfarin efficacy is dependent on maintaining therapeutic INR control (INR range 2.0 to 3.0; if presence of mechanical valve range is 2.5 to 3.5), and declines significantly when the international normalized ratio falls below 2.0 [Evidence Level A].
    2. Patients who are prescribed a DOAC should be reassessed at intervals and educated regarding the short half-life of this class of drugs, the importance of daily medication adherence and the dangers of missed doses or prolonged interruptions of therapy [Evidence Level C].
  2. For patients prescribed apixaban, dabigatran, edoxaban, or rivaroxaban, creatinine clearance should be routinely monitored at least once annually, and when there is a change in health status [Evidence Level C]. Refer to Appendix Four for Selection of Anticoagulant Agents for Management of Atrial Fibrillation after stroke or transient ischemic attack.
    1. Dose adjustments or a change in selected agent may be required based on changes in renal function if detected.
    2. More frequent monitoring of renal function (every 6 months or more frequently) may be considered for patients with renal impairment or a dehydrating illness (e.g., creatinine creatinine) for medication adjustment if required [Evidence Level C].
  3. Concomitant antiplatelet therapy with oral anticoagulant therapy is not routinely recommended in patients with atrial fibrillation due to increased bleeding risk unless there is a specific additional medical indication [Evidence Level B].

Notes: (New 2017)

The final, definitive version of this paper has been published in International Journal of Stroke by SAGE Publications Ltd. Copyright © 2017 World Stroke Organization.


Atrial fibrillation is a significant risk factor for stroke, with one in six patients with ischemic stroke found to have atrial fibrillation. Stroke caused by atrial fibrillation is highly preventable if patients are treated with anticoagulants.  Detecting AF following a stroke or transient ischemic attack is important since, once identified, it can be effectively treated. Since most patients do not undergo prolonged screening, AF is often undetected and hence, the condition is generally under-diagnosed.New classes of drugs are available that have demonstrated benefits over previous standard therapy with warfarin.

System Implications
  • Increased public awareness of atrial fibrillation as a risk factor for stroke.
  • Establishment of stroke prevention clinics to improve secondary stroke prevention including management of atrial fibrillation in patients with stroke and transient ischemic attack (effective, consistent prevention with early recognition of risk factors and timely, targeted interventions).
  • A process for appropriate outpatient monitoring of patients’ international normalized ratio and follow-up communication with patients taking anticoagulants.
  • Optimization of comprehensive strategies at the local, regional and provincial levels to prevent the recurrence of stroke.
  • Stroke prevention awareness and education about secondary prevention for primary care practitioners and specialists who manage stroke patients during the acute phase and after discharge from acute care.
  • For patients taking warfarin, access to a dedicated anticoagulant management clinic is associated with better patient outcomes compared to routine medical care. 
  • Universal access to cost-effective pharmaceuticals, regardless of ability to pay or geography, through private and/or public drug coverage plans which can help manage atrial fibrillation.
Performance Measures
  1. Proportion of acute ischemic stroke patients with atrial fibrillation who are treated with anti-coagulant therapy.
  2. Proportion of eligible stroke and transient ischemic attack patients with atrial fibrillation prescribed anticoagulant therapy on discharge from acute care.
  3. Proportion of eligible stroke and transient ischemic attack patients with atrial fibrillationprescribed anticoagulant therapy after a visit to a secondary prevention clinic.
  4. Proportion of atrial fibrillation patients taking anticoagulant therapy at the time of hospital admission for acute ischemic stroke or transient ischemic attack.
  5. Proportion of atrial fibrillation patients with stroke or transient ischemic attack on antiplatelet therapy and not prescribed anticoagulant therapy.
  6. Proportion of atrial fibrillation patients with stroke or transient ischemic attack continuing on anticoagulant therapy at 3 months, 6 months, and 1 year following initiation of therapy.
  7. For atrial fibrillation patients on warfarin, the proportion with an international normalized ratio in the therapeutic range at three months.

Measurement Notes

  • Performance measure 3: reasons why patients with atrial fibrillation and stroke are not on anticoagulants should be collected and reported. These may include contraindications, compliance issues and physician prescribing patterns, among others. This additional information will help to inform the direction for quality improvement initiatives.
  • If there is documentation of atrial fibrillation, the chart should be reviewed for medications prescribed to the patient at the time of discharge, specifically including warfarin, dabigatran, rivaroxaban, apixaban or heparin.  Performance measures should be stratified to include proportions prescribed each of these medications.
  • Data sources may include discharge summary, history and physical examination, ,  primary care provider  ( or physician/nurse practitioner orders, nurses’ notes from inpatient chart, stroke prevention clinic documents, and primary care charts.
  • To measure whether the patient’s International Normalized Ratio was in the therapeutic range, laboratory reports or other reliable documentation are required to verify the International Normalized Ratio levels, and these should be reviewed over a period of time rather than as one single measure.
  • Providing a prescription does not ensure patient adherence with medication administration. Adherence can be determined through patient self-report and through International Normalized Ratio measurements over time.
  • At this time, adherence to the new oral anticoagulants cannot be objectified in the same way as having INR in therapeutic range
Implementation Resources and Knowledge Transfer Tools

Health Care Provider Information

Patient Information

Summary of the Evidence, Evidence Tables and References

Atrial Fibrillation and Stroke Evidence Tables and Reference List

Detecting Atrial Fibrillation
Atrial fibrillation (AF) is a common arrhythmia, which is associated with an increased risk of ischemic stroke. Detecting AF following a stroke or TIA is important since, once identified, it can be effectively treated. However, AF is under-diagnosed because it is frequently paroxysmal and asymptomatic and patients do not routinely undergo prolonged screening. The results from four RCTs and numerous observational studies have demonstrated that prolonged post-stroke ECG monitoring using wearable or insertable devices is effective for improving the detection of paroxysmal AF (numbers needed to screen range from 8-14), with longer monitoring durations associated with an increased probability of AF detection. In the Event Monitor Belt for Recording Atrial Fibrillation after a Cerebral Ischemic Event (EMBRACE) trail (Gladstone et al. 2014), a 30-day ambulatory cardiac event monitor was found to be superior to repeat 24-hour Holter monitoring in identifying AF in 572 patients aged 52 to 96 years (mean=72.5 years) without known AF, who had sustained a cryptogenic ischemic stroke or TIA within the previous 6 months. Atrial fibrillation lasting ≥30 seconds was detected in 16.1% of patients, using the cardiac event monitor compared with 3.2% of patients in the Holter group (absolute difference, 12.9%; 95% CI 8.0 to 17.6; p<0.001; number needed to screen= 8). The cardiac event monitor was also more likely to identify cases of AF lasting longer than ≥2.5 minutes (9.9% vs. 2.5%, absolute difference, 7.4%, 95% CI, 3.4 to 11.3; p<0.001). By 90 days, oral anticoagulant therapy had been prescribed for more patients in the intervention group (18.6% vs. 11.1%, p=0.01). Three-quarters of AF cases identified in the intervention group were detected within the first 2 weeks of monitoring. An economic evaluation, based on AF rates and anticoagulation treatment from the EMRACE trial (Yong et al. 2016) suggests that prolonged cardiac monitoring was highly cost-effective ($2,166/QALY). De Angelis et al. (2016) also reported that under certain conditions, cardiac monitoring can be cost-effective. Similar findings were reported in the Cryptogenic Stroke and Underlying AF (CRYSTAL-AF) trial (Sanna et al. 2014) when patients (mean age of 61.5 years) received long-term monitoring with an insertable cardiac monitor (ICM).  At 6 months, the rate of detection of AF was significantly higher among patients assigned to the ICM group (8.9% vs. 1.4%, HR=6.4, 95% CI 1.9- 21.7, p<0.001), compared with those who received standard monitoring using ECG monitoring on a schedule at the discretion of their treating physician. Similar results were reported at 12 months (12.4% vs. 2.0%, HR=7.3, 95% CI 2.6- 20.8, p<0.001). The yield of detecting atrial fibrillation with prolonged cardiac monitoring is greatest with increased age.  No RCT data is available on individuals under the age of 40 years; the decision to order prolonged cardiac monitoring in young patients could be considered based on clinical circumstances. 

A UK trial (Higgins et al. 2013) that randomized 100 patients with no history of AF and in sinus rhythm, reported that a strategy of 7-day ECG monitoring in the acute phase post-stroke was superior to standard care for the detection of paroxysmal AF (18% vs. 2%; p<0.05). Significantly more patients that received additional monitoring were started on anticoagulants. The Finding Atrial Fibrillation in Stroke - Evaluation of Enhanced and Prolonged Holter Monitoring (FIND-AF) trial randomized 398 patients over age 60 years (average age 73 years) and found that a strategy of 10-day Holter monitoring started within the first week post stroke and repeated at 3 months and 6 months was superior to standard care, which consisted of an average of 73 hours of inpatient telemetry plus an average of 24 hours of Holter monitoring (Wachter et al. 2016). At 6 months, detection of AF was significantly higher in the prolonged monitoring group (13.5% vs. 4.5%; absolute difference 9%, 95% CI 3.5-14.6, p=0.002; NNS=11). Similar findings were reported in the Cryptogenic Stroke and Underlying AF (CRYSTAL-AF) trial (Sanna et al. 2014) when patients (mean age of 61.5 years) received long-term monitoring with an insertable cardiac monitor (ICM).  At 6 months, the rate of detection of AF was significantly higher among patients assigned to the ICM group (8.9% vs. 1.4%, HR=6.4, 95% CI 1.9- 21.7, p<0.001), compared with those who received standard monitoring using ECG monitoring on a schedule at the discretion of their treating physician. Similar results were reported at 12 months (12.4% vs. 2.0%, HR=7.3, 95% CI 2.6- 20.8, p<0.001).

The terms valvular and nonvalvular heart disease reflect overly simplistic definitions, and do not sufficiently discriminate disorders with similar pathogenesis of thromboembolisms or thromboembolic risk, and importantly do not clearly define treatment needs (De Caterina et al. 2014, Macle et al. 2015). While phase III studies comparing individual novel oral anticoagulants (NOACs) with warfarin used variable definitions of valvular atrial fibrillation, more accurate definitions are being sought to better guide treatment for patients who should not be treated with a NOAC.  Severe mitral stenosis and mechanical heart valves are the only conditions where there is consensus regarding the avoidance of anticoagulation with NOACs.  Data are limited from any of the pivotal NOAC trials to provide clear reassurance with regards to the treatment of other cardiac disorders, especially severe native valvular lesions that might merit anticoagulant prophylaxis (De Caterina et al. 2014, Di Biase 2016). The 2016 update of the Canadian Cardiovascular Society guidelines for the management of atrial fibrillation state that “the use of NOACs is contraindicated in the presence of mechanical heart valves, rheumatic mitral stenosis, or moderate and severe nonrheumatic mitral stenosis”.

Warfarin is well established as an effective medication for reducing the risk of stroke in patients with AF and atrial flutter and has been evaluated in a variety of adjusted-dose regimens, alone and in combination with ASA, as well as in low intensity and fixed, mini-dose treatment plans. A systematic review & meta-analysis (Hart et al. 2007) included the results of 29 trials involving 28,044 patients who had non-valvular atrial fibrillation. Six of the included trials compared placebo with adjusted-dose warfarin (2,900 participants, 20% with previous stroke or TIA). Treatment with adjusted dose warfarin was associated with a 64% reduction in all strokes (ARR= 2.7%/year, NNT=37 for primary prevention; ARR=8.4%/year, NNT=12 for secondary prevention of stroke) and a 67% reduction for ischemic stroke. Mean INRs ranged from 2.0 – 2.6 in primary prevention studies and was 2.9 in the only secondary prevention study included. In trials that compared the effectiveness of warfarin with other antiplatelets, including clopidogrel and dipyridamole, the use of warfarin was associated with a 37% reduction in all strokes (95% CI 23%- 48%). An increased risk of intracranial hemorrhage was found to be associated with the use of adjusted-dose warfarin, although it was very small (absolute risk=0.2%/year).

The observational study, Patient-Centered Research into Outcomes Stroke Patients Prefer and Effectiveness Research (PROSPER, Xian et al. 2015) included 12,552 patients with acute ischemic stroke and documented persistent or paroxysmal AF/flutter who were admitted to hospitals in the United States from 2009-2011. The risk of major cardiovascular evets among those discharged with warfarin was compared with those not treated with any oral anticoagulant. Over the following 2 years, fewer patients discharged on warfarin experienced a major event (54.7% vs. 66.8%; HR=0.87, 99% CI 0.78-0.98, p=0.003) and spent more days at home (47.6 days, 99% CI 26.9-68.2, p<0.001).

The Birmingham Atrial Fibrillation Treatment of the Aged (BAFTA) study recruited 973 patients (12.5% with previous stroke or TIA aged 75 years or greater from primary care and randomly assigned them to receive adjusted-dose warfarin (INR 2.0 - 3.0) or ASA (75 mg once daily) and followed them for a mean of 2.7 years (Mant et al. 2007). The primary endpoint was fatal or disabling stroke (ischemic or hemorrhagic), other intracranial hemorrhage, or clinically significant systemic embolism. There were fewer primary events among participants assigned to warfarin (21 strokes, 2 other intracranial hemorrhages, and 1 systemic embolus), compared to those assigned to ASA (48 primary events: 44 strokes, 1 other intracranial hemorrhage, and 3 systemic emboli). The corresponding annual risks were 1.8% vs. 3.8%, RRR=52%, 95% CI 20-72%, p=0.003. To prevent one event each year, the number needed to treat was 50. The annual risk of extracranial hemorrhage was 1.4% for patients assigned warfarin and 1.6% for those assigned ASA. A Cochrane review authored by Saxena & Koudstaal (2004) also examined the effectiveness of oral anti-coagulants with antiplatelet therapy in individuals with non-rheumatic (non-valvular) AF and history of previous stroke or TIA. Two RCTs were included. The European Atrial Fibrillation Trial (EAFT) included 455 patients within three months of TIA or minor stroke who were randomly assigned to warfarin (INR 2.5 to 4.0) or ASA (300 mg/day) and followed for a mean of 2.3 years (EAFT 1993). The Studio Italiano Fibrillazione Atriale (SIFA) trial included 916 patients within 15 days of TIA or minor stroke who were randomized to open-label warfarin (INR 2.0 to 3.5) or indobufen (a reversible platelet cyclooxygenase inhibitor, 100 or 200 mg twice a day), and followed for one year (Morocutti 1997). Pooled analysis of the 2 trials revealed a significant protective effect in favour of anti-coagulant therapy over antiplatelet therapy for all vascular events (OR=0.67, 95%CI 0.50, 0.91) and for recurrent stroke (OR=0.49, 95% CI 0.33, 0.72). In terms of absolute risk, anticoagulant therapy was associated with a risk of approximately 4% per year in both studies, whereas the risk was 10%/year and 5%/year for individuals assigned to treatment with antiplatelet therapy in the EAFT and SIFA study, respectively. Warfarin use was not associated with significant increases in the risk of intracranial bleeding. Although major extracranial bleeding complications occurred more often in patients on warfarin (OR=5.16, 95% CI 2.08–12.83), the absolute difference was small (2.8% vs. 0.9%/year in EAFT and 0.9% vs. 0%/year in SIFA).

Novel Anticoagulants
In response to some of the management challenges associated with warfarin use such as the need for frequent monitoring and food and drug interactions, several new (novel) oral anticoagulants have been developed. Dabigatran, one such agent, is a direct thrombin inhibitor with a serum half-life of 12 to 17 hours. The landmark Randomized Evaluation of Long-term anticoagulant therapy (RE-LY) trial (Connolly et al. 2009), included 18,113 patients with AF and at least one other stroke risk factor. Patients were randomly allocated to receive dabigatran (110 mg or 150 mg twice daily) or warfarin (adjusted to an INR of 2.0-3.0) and followed for a median of two years. The primary outcome was a composite of stroke or systemic embolism. Both doses of dabigatran were found to be non-inferior to warfarin therapy in terms of risk for stroke or systemic embolism. In addition, the fixed dose of 150 mg was superior to warfarin therapy for the primary study outcome (RR=0.66, 95% CI 0.53, 0.82, p<0.001). However, when the subgroup of patients with previous TIA/stroke were analysed separately, neither the 110 mg dose of dabigatran nor the 150 mg dose was associated with significant reductions in risk for recurrent events when compared with warfarin (p=0.65 and 0.34, respectively). Compared to warfarin, the risks for major bleeding events, including life-threatening bleeding, intracranial bleeding, and gastrointestinal bleeding, were reduced in the 110 mg group only (RR=0.80, 95% CI 0.69, 0.93, p = 0.003), while the 150 mg dose was associated with increased risk for gastrointestinal bleeding (RR=1.50, 95% CI 1.19, 1.89, p<0.001). It should be noted that patients in the warfarin group had a therapeutic INR only about 64% of the time, which is consistent with other clinical trials. To achieve a stroke rate similar to the dabigatran 150 mg twice daily group, it is estimated that patients assigned to warfarin in RE-LY needed to have a therapeutic INR 80% of the time, a degree of control unlikely to be achieved in clinical trials or clinical practice.  

In a long-term extension of the RE-LY trial (Connolly et al. 2013), 5,851 pparticipants who had been assigned to either of the dabigatran dosing schedules in the original trial could continue in the RELY-ABLE study if they did not discontinue study medication at trial termination. Participants continued to receive the same dose of dabigatran (still blinded to the dose condition) as they had throughout the original trial. Patients enrolled in the warfarin condition did not continue in the trial. Median duration of follow-up for the patients enrolled in RELY-ABLE was 5.5 years. During the study period, annual rates of stroke or systemic embolism were 1.46% and 1.6% in the 150 mg and 110 mg dose groups, respectively. The risk of this combined outcome was not significantly different between groups (HR=0.91, 95% CI 0.69-1.20). Similarly, annual rates of ischemic stroke were 1.15% in the 150 mg group and 1.24% in the 110 mg group (HR=0.92, 95% CI 0.67, 1.27), with low incidences of hemorrhagic stroke and myocardial infarction in both groups. There was a significantly increased risk of bleeding events associated with the higher dose of dabigatran (3.74% vs. 2.99%; HR=1.26, 95% CI 1.04-1.53), although gastrointestinal bleeding events were similar in both groups (1.54% and 1.56%/year). Mortality was similar in both dose conditions (3.1% and 3.02% per year). Dyspeptic symptoms were reported in approximately 5% of patients in each group. In subgroup analysis, Diener et al. (2010) examined treatment effects between patients with and without previous history of stroke or TIA. No interactions were reported for any of the outcomes of interest, including stroke, ICH, ischemic or unknown stroke, disabling or fatal stroke, MI, vascular death, or death from any cause.

Three Factor Xa inhibitors, rivaroxiban, apixaban and edoxaban (recently approved in Canada), have been investigated in large clinical trials. In the Rivaroxaban Once Daily Oral Direct Factor Xa Inhibition Compared with Vitamin K Antagonism for Prevention of Stroke and Embolism Trial in Atrial Fibrillation (ROCKET-AF, Patel et al. 2011), 14,264 patients with elevated risk for stroke were randomized to receive fixed-dose rivaroxiban (20 mg daily or 15 mg daily in patients with reduced creatinine clearance) or adjusted-dose warfarin (target INR pf 2.0 to 3.0). The median length of treatment was 590 days. Stroke or systemic embolism occurred less frequently in patients who received rivaroxiban (1.7% vs. 2.2% per year; HR= 0.79; 95% CI 0.66- 0.96, p<0.001 for non-inferiority). There were fewer incidences of intracranial hemorrhage in the rivaroxaban group (HR=0.67, 95% CI 0.47, 0.93; p=0.02), while the risk of major bleeding from a gastrointestinal site was increased (3.2% vs. 2.2%, p<0.001). A post hoc analysis from the ROCKET-AF trial demonstrated no significant difference in treatment effectiveness or risk for adverse events between groups of individuals with or without previous stroke or TIA, suggesting that rivaroxaban may be considered as a potential alternative to warfarin in secondary prevention of stroke (Hankey et al. 2012).
The Apixaban for Reduction in Stroke and Other Thromboembolic Events in Atrial Fibrillation (ARISTOTLE) trial (Granger et al. 2011) randomized 18,201 patients with AF and at least one other risk factor for stroke to treatment with apixaban (5 mg twice daily) or dose-adjusted warfarin (target INR 2.0-3.0). The primary outcome of stroke or systemic embolism occurred in significantly fewer patients in the apixaban group (212 vs 265; HR= 0.79; 95% CI 0.66- 0.95; p<0.001 for non-inferiority and p=0.01 for superiority). There was no between group difference for ischemic stroke alone (p=0.42); however, treatment with apixaban was associated with a significant reduction in risk for hemorrhagic stroke when compared to warfarin (HR=0.51, 95% CI 0.35-0.75; p<0.001). There was a significant reduction in risks of death from any cause and fatal or disabling stroke associated with apixaban (HR=0.89, 95% CI 0.80- 0.99; p=0.047 and HR=0.71; 95% CI, 0.54-0.94, respectively). Intracranial bleeding occurred more often in individuals assigned to treatment with warfarin (HR=0.42, 95% CI 0.3-0.58; p<0.001). The risk of major bleeding was significantly lower in the apixaban group (HR= 0.69; 95% CI, 0.60- 0.80; p<0.001). Overall, apixaban was found to be superior to warfarin in preventing stroke or systemic embolism, caused less bleeding, and resulted in lower mortality. In the subgroup analysis of patients with previous stroke or TIA (Easton et al. 2012), the rate of stroke or systemic embolism was similar between groups (2.46 per 100 patient-years of follow-up in the apixaban vs. 3.24 in the warfarin group; HR= 0.76, 95% CI 0.56-1·03, p for interaction=0·71).

Apixaban has also been compared with ASA in patients with AF. In the Apixaban Versus Acetylsalicylic Acid to Prevent Strokes in Atrial Fibrillation Patients Who Have Failed or Are Unsuitable for Vitamin K Antagonist Treatment (AVERROES, Connolly et al. 2011) trial, 5,599 patients were randomized to receive apixaban 5 mg twice daily or ASA at a dose of 81 to 324 mg daily. The median length of follow-up was 1.1 years. The primary efficacy outcome was the occurrence of stroke (ischemic or hemorrhagic) or systemic embolism. The trial was terminated early given the clear benefit demonstrated in favour of apixaban. There were significantly fewer primary outcome events recorded in the apixaban condition than in the ASA condition (113 vs. 51, HR=0.45, 95% CI 0.32-0.62; p<0.001). For stroke events in particular, there were significantly fewer ischemic events in individuals treated with apixaban (HR=0.37, 95% CI 0.25-0.55; p<0.001), although there were no significant between group differences in hemorrhagic stroke (p=0.45). There was no difference in the incidence of major bleeding events between groups.

The Effective Anticoagulation with Factor Xa Next Generation in Atrial Fibrillation–Thrombolysis in Myocardial Infarction 48 (ENGAGE AF-TIMI 48) trial (Giugliano et al. 2013) assessed the use of edoxaban versus warfarin in patients with atrial fibrillation. The trial randomized 21,105 patients to receive dose-adjusted warfarin, high-dose edoxaban (60mg), or low-dose edoxaban (30mg). The target INR for the warfarin group was 2.0-3.0 and the median duration of the treatment was 2.5 years. Blinding was maintained throughout the study by using sham INR values for the edoxaban group and by giving all patients a placebo tablet in addition to their active medication. The primary efficacy outcome was the occurrence of stroke or systemic embolic event and the primary safety outcome was the occurrence of major bleeding during treatment. Patients in the high-dose and low-dose edoxaban groups experienced non-inferior rates of stroke and systemic embolic events compared to the patients receiving warfarin (HR 0.79, 97.5% CI 0.63 to 0.99, p<0.001 and HR 1.07, 97.5% CI 0.87 to 1.31, p=0.005). A superiority analysis for the annualized rate of stroke or systemic embolic event found no evidence for the superiority of either high-dose edoxaban (HR 0.87, 97.5% CI 0.73 to 1.04, p=0.08) or low-dose edoxaban (HR 1.13, 97.5% CI 0.96 to 1.34, p=0.10) compared to warfarin. The safety profile of edoxaban was supported by significantly lower annualized rates of bleeding events for both high-dose and low-dose treatment regimens compared to warfarin (HR 0.8, 95% CI 0.71 to 0.91, p<0.001 and HR 0.47, 95% CI, 0.41 to 0.55, p<0.001).

Mechanical Heart Valves
Lifelong anticoagulation is usually required for patients with prosthetic heart valve replacement due to the risk of thromboembolic complications; however, questions remain regarding the most appropriate regimens. Current Canadian guidelines recommend target INRs of 2.5-3.0, depending on the location of the replacement valve with a vitamin K antagonist (VKA). Puskas et al. (2014) evaluated whether a less aggressive target for anticoagulation could be as effective. In this study, 425 patients with elevated risk of thromboembolism, including chronic atrial fibrillation or left ventricular ejection fraction<30% were recruited in the Prospective Randomized On-X Valve Anticoagulation Clinical Trial (PROACT). In addition to receiving 81 mg aspirin daily, patients were randomized to a lower-dose warfarin group with a target INR of 1.5-2.0, or to a standard therapy group with a target INR=2.0-3.0 through self-management three months following aortic valve replacement. After a mean duration of just under 4 years, there were significantly fewer major, minor and total bleeding events in the lower-dose warfarin group (10 vs. 25, RR=0.45, 95% CI 0.21-0.94, p=0.032; p 8 vs. 25, RR=0.36, 95% CI 0.16-0.79, p=0.011 and 18 vs. 50, RR=0.40, 95% CI 0.24-0.69, p<0.001, respectively). The risks of hemorrhagic, ischemic stroke and TIA were similar between groups (1 vs. 2, RR=0.56, 95% CI 0.001-10.7, p=0.63; 5 vs. 5, RR=1.12, 95% CI 0.32-3.87, p=0.859 and 9 vs. 6, RR=1.68, 95% CI 0.60-4.72. p=0.326, respectively). The potential benefit of dabigatran was examined in the Randomized, Phase II Study to Evaluate the Safety and Pharmokinetics of Oral Dabigatran Etexilate in Patients after Heart Valve Replacement (RE-ALIGN). This trial randomized patients to warfarin with a target INR of 2-3, or 2.5-3.5 depending on thromboembolic risk, following aortic and/or mitral valve replacement, or 2 escalating doses of dabigatran for 12 weeks (Eikelboom et al. 2013). The trial was stopped early due to an excess of thromboembolic and bleeding events in the dabigatran group. Among patients in whom treatment was initiated within 7 days of valve replacement, there were 9 strokes and 2 TIAs in the dabigatran group and 0 strokes and 2 TIAs in the warfarin group, respectively. The addition of antiplatelets to VKA therapy following heart valve replacement was the topic of a Cochrane review (Massel & Little 2013), which included the results from 13 trials. The addition of either aspirin or dipyridamole significantly reduced the risk of thromboembolic events (OR= 0.43, 95% CI 0.32- 0.59, p < 0.00001) and total mortality (OR= 0.57, 95% CI 0.42- 0.78, p = 0.0004); however, the risk of major bleeding was increased significantly (OR=1.58, 95% CI 1.14- 2.18, p= 0.006).

Timing or Anticoagulation Following Ischemic Stroke
Results from the Early Recurrence and cerebral bleeding in patients with acute ischemic stroke and Atrial Fibrillation (RAF) study (Paciaroni et al. 2015) suggest that the optimal window for initiation or resumption of treatment with anticoagulants is between 4-14 days following stroke.  Of 1,029 patients admitted with acute ischemic stroke and known or newly diagnosed AF, significantly fewer patients treated with oral anticoagulants had a primary outcome event (composite of stroke, TIA, symptomatic systemic embolism, symptomatic cerebral bleeding, and major extracerebral bleeding at 90 days) compared with patients treated with either LMWHs alone or LMWH followed by oral anticoagulants (7% vs. 16.8% and 12.3%, respectively, p=0.003). Adjusted for age, sex, CHA2DS2-VASc score, lesion size, reperfusion therapy, and NIHSS on admission, patients who had been initiated on treatment with anticoagulants between 4 and 14 days had a significantly reduced risk of the primary outcome and in ischemic events compared with patients who had their treatments initiated before 4 or after 14 days from stroke onset (HR=0.53, 95% CI 0.30–0.93, p=0.025 and HR=0.43, 95% CI 0.19–0.97, p=0.043, respectively).

Left Atrial Appendage (LAA) Devices
In patients with non-valvular AF, embolic stroke can occur through the formation of a thrombus in the left atrium. Several devices are available to exclude blood flow from the LAA, reducing stroke risk. The WATCHMAN device has been evaluated (for non-inferiority) in several large RCTs. In the Watchman Left Atrial Appendage System for Embolic Protection in Patients With AF (PROTECT-AF, Holmes et a. 2009), 707 patients with a CHADS2 score of ≥1 were randomized to undergo LAA occlusion with the WATCHMAN device (n=463) or to continuing warfarin therapy (n=244). After a mean duration of follow-up of 18 months, the event rate/100 patient- years for the primary outcome (a composite of the occurrence of stroke, cardiovascular or unexplained death, or systemic embolism), was 3.0 for the intervention group vs. 4.9 for the control group (RR=0.62, 95%, Cr I 0.35 to1.25), which met the threshold for non-inferiority. However, the risk of events related to excessive bleeding was significantly higher in the intervention group (7.4 vs. 4.4/100 patient-years). The Watchman LAA Closure Device in Patients With Atrial Fibrillation Versus Long Term Warfarin Therapy (PREVAIL) study (Holmes et al. 2014) was similar to PROTECT-AF, in terms of treatment contrasts and eligibility criteria. In this trial, which included 407 participants, the mean age was slightly older and the proportion of patients with a CHADS2 score of ≥2 was higher. While the results of this trial failed to demonstrate non-inferiority of the WATCHMAN device compared with warfarin for the reduction of the early primary efficacy endpoint (a composite of ischemic or hemorrhagic stroke, systemic embolism and cardiovascular death), evidence of non-inferiority was reached for the late primary efficacy endpoint (events excluding the first 7 days post procedure).