Note: These recommendations are applicable to ischemic stroke and/or transient ischemic attack.
10.1 Patent Foramen Ovale (PFO)
- Patients with a recent ischemic stroke suspected to be related to a PFO should have an evaluation by healthcare professionals with stroke and cardiovascular expertise [Evidence Level C].
- For carefully selected patients with a recent ischemic stroke attributed to a PFO, PFO device closure plus long-term antiplatelet therapy is recommended over long-term antithrombotic therapy alone provided all the following criteria are met [Evidence Level A]:
- Age 18-60 years;
- The diagnosis of the index stroke event is confirmed by imaging as a non-lacunar embolic ischemic stroke.
- The patient has been evaluated by a neurologist or healthcare professional with stroke expertise, and the PFO is felt to be the most likely cause for the index stroke event following a thorough etiological evaluation that has excluded alternate likely etiologies.
- (New for 2020): It is reasonable to recommend against PFO closure for patients who have none of the following higher-risk anatomical features on echocardiography: (a) atrial septal aneurysm; (b) large right-to-left shunt (e.g., >20 microbubbles); and (c) large diameter PFO (e.g., ≥ 2 mm) [Evidence Level B].
- For patients requiring long-term anticoagulation for other reasons, the benefit of PFO closure is uncertain, and treatment decisions should be based on individual patient characteristics and risk versus benefit profile [Evidence Level C].
- For patients with a recent ischemic stroke attributed to a PFO who do not undergo PFO closure and are aged 60 years or younger, either antiplatelet or anticoagulant therapy is recommended for secondary stroke prevention, unless there is a separate evidence-based indication for chronic anticoagulant therapy [Evidence Level B].
Section 10.1 Clinical Considerations
Warfarin can reduce recurrent stroke; however, this benefit may be outweighed by the increased risk of major hemorrhage.
The role of DOACs is unknown in this population.
10.2 Aortic Arch Atheroma
- Aortic arch atheroma should be managed according to the stroke prevention recommendations included in all relevant sections of the Secondary Prevention of Stroke Module [Evidence Level C].
- In the ARCH trial, no significant difference was found in individuals treated with dual antiplatelet therapy (acetylsalicylic acid plus clopidogrel) as compared with warfarin; the effectiveness of anticoagulant therapy compared with antiplatelet therapy in this context is uncertain and the choice should be individualized [Evidence Level B].
10.3 Heart Failure, Decreased Left Ventricular Ejection Fraction, Cardiac Thrombus
- For patients with ischemic stroke or transient ischemic attack who are in sinus rhythm and have a left atrial or left ventricular thrombus demonstrated by echocardiography or other imaging modality, anticoagulant therapy is recommended for greater than 3 months [Evidence Level C].
- For patients with ischemic stroke or transient ischemic attack who are in sinus rhythm and have severe left ventricular dysfunction (ejection fraction ≤35%) without evidence of left atrial or left ventricular thrombus, the net benefit of anticoagulant therapy (with either vitamin K antagonists or DOACs) compared with antiplatelet therapy is uncertain, and the choice of management strategies should be individualized [Evidence Level B].
Patent Foramen Ovale
For years, the role of percutaneous closure of a patent foramen ovale (PFO) for secondary stroke prevention had been controversial, because (1) PFO is common in the general population (25%), (2) they are most often incidental rather than pathogenic, and (3) the initial RCTs of PFO closure versus medical therapy were inconclusive. In 2017 and 2018, the publication of three new RCTs and long-term follow-up of an earlier one, demonstrated that among carefully selected patients PFO closure was superior to medical therapy for prevention of stroke recurrence (Saver JL, et al. Stroke 2018;49:1541-8). Meta-analyses of the PFO trials (Mir H et al. BMJ Open 2018;8:e023761; Kent DM et al. J Am Coll Cardiol 2016;67:907-17; Stortecky et al. Eur Heart J 2015;36:120-8) support PFO closure for carefully selected patients aged 60 years or younger with an unexplained embolic stroke event who are found to have a PFO and who do not require chronic anticoagulant therapy for another reason. PFO is not recommended for patients whose PFO is most likely incidental rather than causal for the index stroke event. In such circumstances, antiplatelet therapy alone is recommended for secondary stroke prevention, unless there is a separate evidence-based indication for anticoagulant therapy. None of the PFO trials (except one) enrolled patients over age 60 years. Therefore, based on the available trial evidence, PFO closure cannot be recommended for elderly patients with stroke. PFO closure is not recommended for primary stroke prevention.
The recommendation for PFO closure places high value on avoidance of disabling ischemic stroke, especially for young patients with a long-life expectancy, and avoidance of bleeding complications associated with long-term anticoagulant therapy. The recommendation places less value on the fact that a PFO is associated with a relatively low annual absolute risk of stroke recurrence (1.2% per year), there is often uncertainty regarding whether or not the PFO was truly the causal factor, the fact that PFO closure can only prevent PFO-related strokes, and in a minority of cases PFO closure is incomplete leaving a residual shunt. While the annual absolute risk reduction associated with PFO closure is small, the long-term benefit over decades for young patients may become quite large. Patient counselling and shared decision-making, taking into account patient values and preferences, are encouraged and decision aids have been developed (MAGICapp: Patent foramen ovale closure, antiplatelet therapy or anticoagulation therapy for management of cryptogenic stroke; available at: https://app.magicapp.org/#/guideline/2649). Patients considering PFO closure need to understand the benefits and risks of the procedure and the alternatives and accept an up-front risk of procedural complications that is relatively small but potentially serious, including atrial fibrillation. In the meta-analysis by Mir et al., the overall incidence of procedural complications was 3.6%, and there was an 1.8% absolute increase in the risk of persistent atrial fibrillation and a 1.2% increase in the risk of transient atrial fibrillation (Mir H et al. BMJ Open 2018;8:e023761).
When PFO closure is being contemplated, careful patient selection is essential. The committee advocates consultation with a stroke specialist and an experienced interventional cardiologist with a low rate of procedural complications. A consensus statement on technical standards for operators has been published (Horlick et al. Catheter Cardiovasc Interv 2019;93:859-874). The appropriateness of PFO closure depends on (1) an accurate diagnosis of the index stroke event and exclusion of mimics, and (2) a thorough etiological workup to exclude alternate more likely causes for the stroke event. Given that transient ischemic attack can be difficult to diagnose (over-diagnosis of transient ischemic attack is a common problem in clinical practice because many conditions may mimic a transient ischemic attack), and the fact that patients none of the PFO trials (except for one) enrolled patients with transient ischemic attack as the index event, the committee had concerns about the potential for inappropriate PFO closure procedures for patients with transient ischemic attack mimics. Therefore, transient ischemic attack was removed from the general recommendations above regarding PFO closure. If PFO closure is offered to a patient with transient ischemic attack, the transient ischemic attack diagnosis should be convincing, i.e., a high degree of certainty that the index event was a genuine cerebral ischemic event, ideally with cortical symptoms and verified by a neurologist; every effort should be made to obtain brain imaging evidence for confirmation of acute embolic ischemia.
In patients with ischemic stroke or transient ischemic attack who have either left atrial or left ventricular thrombus, anticoagulant therapy with VKA is recommended for ≥ 3 months (Level of Evidence C). 13 In the WARCEF randomized trial of 2305 patients with heart failure (left ventricular function ≤35%) and in sinus rhythm, warfarin reduced the risk of ischemic stroke but the benefit was offset by an increase in major bleeding. In a meta-analysis of patients with congestive heart failure with reduced ejection fraction, warfarin was compared with antiplatelet therapy and there was a small absolute reduction in the risk of ischemic stroke in those treated with warfarin, but this was accompanied by an increased risk of major hemorrhage (mostly intracranial hemorrhage), and was not associated with any accompanying reduction in death, MI, or hospitalization due to heart failure. In the COMMANDER HF trial of patients with reduced left ventricular ejection fraction (40%) in sinus rhythm, rivaroxaban at a dose of 2.5mg twice daily was not associated with a significantly lower rate of the composite outcome of death, MI or stroke compared with placebo, although stroke events were reduced.
People with stroke, their family and caregivers expressed the desire to learn more about the connection between the heart and the brain. Many people with stroke were not aware of the conditions discussed in this section, such as a PFO, and the potential connections to stroke. Some people with stroke did not know if they had been investigated for these conditions and if they were, what the results where.
People who have experienced a stroke and have concurrent heart conditions have described the challenges of siloed healthcare systems on their care. For example, some health professionals only wanted to focus on their stroke related conditions and would not discuss other factors, such as heart conditions. In addition, there have been experiences of inconsistent information sharing between specialties and an opportunity for more integrated and coordinated care for people with a broader range of vascular risk factors affecting both heart and brain.
- Integration of care for people with heart and stroke conditions to efficiently manage appointments and ensure coordination of care.
- Support for ongoing research into etiology for patients with cryptogenic stroke.
- Support for research to further investigate the impact of PFO closure versus medical therapy.
- Proportion of people who experience an acute stroke who have a patent foramen ovale.
- Proportion of people with acute ischemic stroke and PFO who are treated with antiplatelet agents or anticoagulants (and specific type of anticoagulant – LMWH, warfarin or DOAC).
- Proportion of people with acute ischemic stroke and PFO who undergo a PFO closure procedure.
- Median time from stroke to PFO closure.
- Proportion of people who experience an acute stroke who also have a diagnosis of heart failure.
- Median time from diagnosis of heart failure to stroke event.
- Mortality rates for people with pre-existing heart failure who experience an acute stroke (stratified by ischemic or hemorrhage).
- All performance measures should be stratified by age, sex, type of stroke and time from heart diagnosis to stroke occurrence.
- PFO may only be detected after a stroke has occurred.
Health Care Provider Information
- Heart & Stroke: Post-Stroke Checklist
- Heart & Stroke: Your Stroke Journey
- Heart & Stroke: Online and Peer Support
- Thrombosis Canada
- Canadian Cardiovascular Society
- Canadian Heart Failure Society
- Canadian Cardiovascular Harmonized National Guidelines Endeavour (C-CHANGE): guideline for the prevention and management of cardiovascular disease in primary care: 2018 update
- Canadian Cardiovascular Harmonized National Guidelines Endeavour (C-CHANGE) guideline for the prevention and management of cardiovascular disease in primary care: 2018 update | CMAJ
- Heart & Stroke: Post-Stroke Checklist
- Heart & Stroke: Online and Peer Support
- Heart & Stroke: Heart Conditions Resource
- Heart & Stroke: Working with your doctor
- Heart & Stroke Patent Foramen Ovale
- Heart & Stroke: Living Well With Heart Disease
- Heart & Stroke: Living With Heart Failure
- Canadian Heart Failure Society: Patient Resources
Patent foramen ovale (PFO)
Three earlier RCTs, CLOSURE 1 (Furlan et al. 2012), the PC Trial (Meier et al. 2013), and RESPECT (Carroll et al. 2013) investigated the effectiveness of PFO closure for reducing the risk of stroke recurrence and mortality following cryptogenic stroke, compared to medical management. Across the three trials, no significant reductions in the risk of the primary outcomes, which included recurrent stroke or transient ischemic attack and death, were associated with closure in their respective intention-to-treat analyses. The associated hazard ratios (HR were: 0.78 (95% CI 0.45 to 1.35, p=0.37) in CLOSURE 1, during 2-years follow-up; 0.63, (95% CI 0.24 to 1.62, p=0.34) in the PC trial after a mean of 4.1 years of follow-up, and 0.49, (95% CI 0.22 to 1.11, p=0.08) in RESPECT after a mean follow-up of 2.6 years. Whereas the authors of CLOSURE 1 and the PC trials both observed similar findings in per protocol-based analyses, the authors of RESPECT reported that in a per protocol analysis, PFO closure was associated with a significant reduction in the composite outcome of recurrent ischemic stroke or death, compared to medical therapy (HR= 0.37, 95% CI 0.14 to 0.96, p=0.03). There was no significant increase in the risk of serious adverse events in the intervention arm of any of the trials.
Results from recent trials including CLOSE (Mas et al. 2017), REDUCE (Sondergaard et al. 2017), DEFENSE-PFO (Lee et al. 2018), and long-term results of the RESPECT trial (Saver et al. 2017), have demonstrated that among carefully selected patients PFO closure is superior to medical therapy for prevention of stroke recurrence.
In the CLOSE trial, Mas et al. (2017) enrolled 633 patients aged 16–60 years (mean age approximately 43 years) who had experienced a recent stroke with no identifiable cause other than a PFO, following detailed etiological work- up by a neurologist. The PFO had to be associated with either an atrial septal aneurysm (excursion >10 mm) or large interatrial shunt (>30 microbubbles in the left atrium within three cardiac cycles after opacification of the right atrium). Mean duration of follow-up was 5.3±2.0 years. The rate of recurrent stroke was 0 in the PFO-closure group compared with 6.0% in the antiplatelet-only group (mostly aspirin) (HR= 0.03; 95% CI 0-0.26; p<0.001; NNT=20 to prevent 1 stroke in 5 years; 95% CI 17-25). The rate of procedural complications in the PFO closure group was 5.9%, mostly consisting of atrial fibrillation (4.6% in the closure group vs. 0.9% in the antiplatelet group, p=0.02); most cases of atrial fibrillation were transient and did not recur during follow-up.
The REDUCE trial (Sondergaard et al. 2017) enrolled 664 patients aged 18–59 years (mean age 45.2 years) with a PFO with a right-to-left shunt (spontaneous or during Valsalva maneuver), of whom 81% had moderate (6-25 microbubbles) or large (>25 microbubbles) interatrial shunts. Median follow-up was 3.2 years. The risk of ischemic stroke was significantly lower in the PFO closure group compared with the antiplatelet-only group (1.4% vs. 5.4%, HR=0.23, 95% CI 0.09-0.62; p=0.002; NNT=28 to prevent 1 stroke in 2 years). Serious device-related adverse events occurred in 1.4% of patients. The frequency of new-onset atrial fibrillation or flutter was significantly higher in the PFO closure group (6.6% vs 0.4%, p<0.01); most cases of atrial fibrillation in the closure group were transient.
The RESPECT trial enrolled patients aged 18–60 years (mean age 45.9 years) with a cryptogenic stroke and PFO. In the extended follow-up analysis, during a median duration of follow-up of 5.9 years, the risk of recurrent ischemic stroke was significantly lower in the PFO closure group compared with the medical therapy group (antiplatelet therapy or warfarin) (3.6% vs. 5.8%; HR=0.55, 95% CI 0.31-0.999, p=0.046, NNT=42 to prevent 1 stroke in 5 years). In subgroup analysis, the benefit of closure appeared to be driven by those with an atrial septal aneurysm or a “substantial” shunt size (grade 3).
Lee et al. (2018) randomized 120 patients aged 18-80 years, with a cryptogenic ischemic stroke occurring within the previous 6 months with no identifiable cause other than a high risk PFO with right-to-left shunting to receive medical therapy with antiplatelet or anticoagulation alone, or PFO closure. Antiplatelet therapy included aspirin (100 mg/day), aspirin in combination with clopidogrel (75 mg/day), or aspirin in combination with cilostazol (200 mg/day). Dual antiplatelet therapy was used by most patients in both groups at 30 days and 6 months. At 12 months, more patients in the PFO group were taking a single antiplatelet, compared with the medical therapy group. Anticoagulants were used by ≤25% of patients at any assessment point. There were significantly more primary outcome events (recurrent stroke and major bleeding within 2 years) in the medication-only group (6 vs. 0, p=0.013), of which 5 were ischemic strokes. The authors estimated that the NNT to avoid one stroke at 2 years, was 10.
Clinical clues to the diagnosis of a PFO-related stroke event include stroke symptom onset preceded by a Valsalva maneuver, waking up with a stroke, sleep apnea, deep vein thrombosis or pulmonary embolism, respiratory symptoms at the time of stroke onset, and recent prolonged travel or immobilization (Ozdemir AO et al. J Neurol Sci 2008;275:121-7). History-taking should elicit whether there is a personal or family history of venous thromboembolism or high-risk thrombophilia. The Risk of Paradoxical Embolism (RoPE) score can help higher scores (younger age, cortical infarct, and absence of traditional vascular risk factors) predict that a PFO is more likely to be stroke-related than incidental (Kent DM et al. Neurology 2013;81:619-625).
There have been conflicting data on the significance of PFO size and uncertainty whether or not an isolated small PFO/small shunt without an atrial septal aneurysm should be closed. In meta-analysis of the PFO trials, the benefit of PFO closure over medical therapy was maximal in subgroups with a large PFO or substantial shunt or atrial septal aneurysm; in subgroups without any of those anatomical features, a significant benefit of PFO closure was not observed (Mas J-L et al. Archives of Cardiovascular Disease 2019;112:532-542; Turc G et al. J Am Heart Assoc 2018;7; Pristipino C et al. European Heart Journal 2018; Saver JL, et al. Stroke 2018;49:1541-8). However, it has been pointed out that the observed effect of shunt size may be confounded by a higher proportion of antiplatelet therapy rather than anticoagulation in the medical arms of the trials that enrolled more patients with larger shunts (Mir H et al. BMJ Open 2018;8: e023761). According to Saver et al., the NNT to prevent 1 stroke over 5 years with PFO closure plus antiplatelet therapy (versus antiplatelet therapy alone) is 24 overall. For PFO closure versus medical therapy, the NNT to prevent 1 stroke in 5 years is lower among patients with an atrial septal aneurysm (NNT=13) and among patients with moderate-large shunts (NNT=18) (Saver JL, et al. Stroke 2018;49:1541-8).
There remains uncertainty about the relative effectiveness of anticoagulant therapy versus antiplatelet therapy for patients with a stroke and PFO (without atrial fibrillation) who do not undergo PFO closure. In the PFO subgroup analyses from the ESUS trials, there was a (non-significant) trend in favour of DOAC therapy over aspirin in the NAVIGATE ESUS trial (Kasner SE et al. Lancet Neurol 2018;17:1053-1060) but not in the RESPECT ESUS trial (Diener HC et al. NEJM 2019;380:1906-1917). In the CLOSE trial, there were numerically fewer recurrent stroke events among patients assigned to the anticoagulation arm (in which 93% received a vitamin K antagonist and 7% received a DOAC) compared with the antiplatelet arm, but this comparison was underpowered. An updated meta-analysis of 5 trials found a non-significant trend in favour of anticoagulation over antiplatelet therapy (Sagris D, et al. Stroke 2019;50). Anticoagulation has the advantage of providing protection against both arterial and venous thromboembolic events beyond the PFO, and hence would probably be preferred for patients with unprovoked deep vein thrombosis or pulmonary embolism or a high-risk thrombophilia.
When the results of 6 RCTs (CLOSURE, PC, RESPECT, GORE-Reduce, CLOSE and DEFENSE PFO) were pooled (Turc et al. 2018), the risk of recurrent stroke is significantly lower in the PFO group, compared with antithrombotic therapy (37/1,889 vs. 79/1,671; RR= 0.36, 95% CI 0.17-0.79, p=0.01).
Sex and Gender Considerations: In a single centre study including patients admitted for recent ischemic stroke and referred for transesophageal echocardiography (TEE), the prevalence of PFO was similar in men and women (32.4% vs. 28.2%, p= 0.15, respectively). (Gupta et al. 2008). In a pooled analysis of 5 RCTs, the risk of recurrent stroke was significantly lower in men who had undergone PFO closure compared with medical management (OR=0.32, 95% CI=0.14–0.73, p=0.01), but was not in women (OR=0.84, 95% CI=0.47–1.51, p=0.56) (Agasthi et al. 2019).
Aortic Arch Atheroma
The definitive management of patients with aortic arch plaques is unclear. Typically, monotherapy with an antiplatelet agent or oral anticoagulation is used to prevent further events in patients with a prior ischemic stroke. Amarenco et al. (2014) tested the hypothesis that dual antiplatelet therapy would be superior to oral anticoagulation. The Aortic Arch Related Cerebral Hazard Trial (ARCH) included 351 patients with a previous ischemic stroke, transient ischemic attack, or peripheral embolism with plaque in the thoracic aorta >4 mm and no other identified embolic source. Patients were randomized to receive 75 to 150 mg/d aspirin + 75 mg/d clopidogrel or dose-adjusted warfarin with a target INR of 2.5 (2-3) for the duration of the trial. After a median of 3.4 years of follow-up, the risk of the primary outcome, a composite of cerebral infarction, myocardial infarction, peripheral embolism, vascular death, or intracranial hemorrhage was not significantly lower in the dual therapy group (7.6% vs. 11.3%, HR=0.76, 95% CI 0.36-1.61, p=0.50). There was no significant difference in the occurrence of major hemorrhages between groups (2.3% for dual therapy vs. 3.4% for warfarin, p=0.2).
Risk of Recurrent Stroke Associated with Heart Failure
Heart failure is known to be associated with increased risk of recurrent stroke. Katsanos et al. (2016) included the results from 7 studies (n=9,173) that reported the recurrence of ischemic stroke in patients with heart failure. The definitions used for heart failure were based on medical history (n=3), ejection fraction (n=1), Framingham criteria (n=1) or were not reported (n=3). Within the included studies, the percentage of patients with heart failure ranged from 4.8% to 33.9%. The mean follow-up durations across the included studies ranged from 7 days to 5 years. The risk of recurrent stroke was significantly increased among patients with heart failure (RR=1.96, 95% CI 1.49 -2.60, p<0.0001). Using data from the Canadian Stroke Registry, Pongmoragot et al. (2016) compared the outcomes of 12,396 patients admitted to hospital following an ischemic stroke with heart failure versus those without. Heart failure was defined either as pre-existing, or pulmonary edema present at the time of arrival to hospital. While the number of patients with stroke recurrence at 30 days did not differ between groups (3.9% vs. 3.2%, p=0.194), stroke fatality at discharge, 30 days and 1 year was significantly higher for patients with heart failure. Heart failure was also an independent predictor of death or disability at discharge (OR=1.18, 95% CI 1.01-1.37), 30-day survival (HR=1.22, 95% CI 1.05-1.41) and 30-day readmission (OR=1.32, 95% CI 1.05-1.65), after adjusting for age, sex, stroke severity and medical comorbidities.
Stroke Prevention for Patients in Heart Failure
The effectiveness of anticoagulation compared with antiplatelet therapy for stroke prevention in patients with heart failure in sinus rhythm remains unclear. Although several trials have compared their relative effectiveness, the superiority of any one approach has not been demonstrated. Most recently, the COMMANDER-HF trial (Zannad et al. 2018) randomized 5,081 participants in sinus rhythm with chronic heart failure, a left ventricular ejection fraction (LVEF) of ≤40%, with coronary artery disease and who had been treated for an episode of worsening heart failure within the previous 21 days, to receive 2.5 mg rivaroxaban twice daily or matching placebo in addition to standard care (including mono or dual antiplatelet therapy). After a median of 21.1 months, there was no difference between groups in the risk of the primary composite outcome of death from any cause, MI, or stroke (rivaroxaban 13.44 events/100 persons years vs. placebo 14.27 events/100-person years; HR=0.94, 95% CI 0.84–1.05, p=0.27), although the risk of stroke was reduced significantly in the rivaroxaban group (1.08 vs. 1.63 events/100-person years; HR=0.66, 95% CI 0.47–0.95).
The Warfarin versus Aspirin in Reduced Cardiac Ejection Fraction (WARCEF) trial included 2,305 patients with LVEF ≤35% (Homma et al. 2012). Patients were randomized to receive 325 mg aspirin daily or warfarin with a target INR of 2.75 for the study duration. After an average of 3.5 years, the rates for the primary outcome, a composite outcome of time to first event of ischemic stroke, intracerebral hemorrhage or death from any cause, were similar between groups (7.47 and 7.93 events/100 patient years for warfarin and aspirin, respectively: HR for warfarin=0.93, 95% CI 0.79-1.10, p=0.40). Although warfarin was associated with a significantly reduced risk of ischemic stroke (HR=0.52, 95% CI 0.33-0.82, p=0.005), the risks of major and minor hemorrhages were significantly increased. A subgroup analysis of the WARCEF trial (Homma et al. 2013) found that patients <60 years treated with warfarin had a significantly lower risk of the primary outcome (HR=0.63, 95% CI 0.48-0.84, p=0.003) compared with aspirin therapy, while there was no significant treatment effect for patients 60 years or older. Patients <60 years treated with warfarin had a significantly lower risk of the primary outcome plus any major hemorrhage (HR=0.68, 95% CI 0.52-0.89, p=0.005). Patients ≥60 years treated with warfarin had a higher risk (HR=1.25, 95% CI 1.02-1.53, p=0.03) compared with aspirin. Investigators of the Warfarin and Antiplatelet Therapy in Chronic Heart Failure (WATCH) Trial compared 162 mg aspirin daily versus 75 mg clopidogrel daily versus warfarin, with target INR of 2.5 to 3.0 in patients in heart failure with a LVEF ≤35% (Massie et al. 2009). The risk of the primary outcome was similar between groups (20.7% aspirin vs. 21.6% clopidogrel vs. 19.6% warfarin). While warfarin was associated with a decreased risk of nonfatal and total stroke compared with either antiplatelet agent, the risk of bleeding events was significantly higher among patients in the warfarin group compared with clopidogrel.