INTRODUCTION — An ischemic stroke may occur in patients with atrial fibrillation (AF) either as the initial presenting manifestation of AF or despite appropriate antithrombotic prophylaxis. In such patients, a cardiac embolus most commonly originating from the left atrial appendage is a common cause of ischemic stroke. (See "Clinical diagnosis of stroke subtypes", section on 'Brain ischemia'.)

Issues specific to stroke in patients with AF will be reviewed here. The risk of atheroembolism (including stroke), the role of anticoagulant prophylaxis (primary prevention) in patients with AF, and the general evaluation and management of the patient with stroke are presented elsewhere. (See "Atrial fibrillation in adults: Use of oral anticoagulants" and "Overview of the evaluation of stroke" and "Approach to reperfusion therapy for acute ischemic stroke" and "Early antithrombotic treatment of acute ischemic stroke and transient ischemic attack".)

STROKE CHARACTERISTICS — Strokes due to the embolization of thrombus from the left atrial appendage in patients with atrial fibrillation (AF) present with the characteristics of ischemic stroke. (See "Clinical diagnosis of stroke subtypes", section on 'Distinguishing stroke subtypes'.)

AF is associated with more severe ischemic strokes and "longer" transient ischemic attacks (TIAs) than emboli from carotid disease, presumably due to embolization of larger thrombi with AF [1,2]. This relationship was illustrated in a report comparing ischemic brain events in patients with AF and those with carotid disease in two major trials: The ratio of hemispheric events to retinal events was 25:1 with AF compared with 2:1 with carotid disease [1]. As a result, patients with AF who suffer an ischemic stroke appear to have a worse outcome (more disability, greater mortality) than those who have an ischemic stroke in the absence of AF, even after adjustment for the advanced age of patients with AF-related stroke [3-5]. The "longer" TIAs typical in AF patients are more often associated with abnormal magnetic resonance diffusion imaging and would be classified as strokes by the revised American Heart Association definition [6]. (See "Definition, etiology, and clinical manifestations of transient ischemic attack".)

In addition to causing symptomatic stroke with major deficits, AF is also associated with silent cerebral infarctions and TIAs [7-9]. The frequency of silent cerebral infarction was evaluated in a report of 516 patients (predominantly men) with nonrheumatic AF in the Veterans Administration SPINAF trial; computed tomography scanning was performed initially and, in the absence of neurologic symptoms, at the end of follow-up [7]. One or more silent cerebral infarctions were seen at presentation in 15 percent; the estimated rate of new silent cerebral infarcts was about 1.3 percent per year at up to three years of follow-up with a similar "silent" event rate for placebo and warfarin.

Anticoagulated AF patients who experience ischemic stroke typically have smaller infarcts with a lower mortality rate compared with AF patients with strokes who are not anticoagulated [10]. This is likely explained by a higher fraction of nonembolic strokes among anticoagulated AF patients and small size of embolic strokes: anticoagulation greatly reduced the likelihood of large stroke due to left atrial emboli, so that the remaining strokes are from cerebral small artery disease or other mechanisms [10].

RISK OF RECURRENT STROKE — Patients who have had a prior embolic event already have the most potent high-risk factor for subsequent stroke. The risk of recurrent stroke in the first few weeks after the initial event is 3 to 5 percent based upon large numbers of patients observed in the control arms of randomized trials [11,12]. In addition, a risk of up to 12 percent per year has been reported in nonanticoagulated patients in the first two to three years after a stroke [13,14].

Due to the high risk of recurrent embolism, lifelong anticoagulation is strongly recommended for secondary prevention (these patients have a minimum CHA₂DS₂-VASc score of 2 for which chronic anticoagulation is strongly recommended).

EVALUATION — Patients with atrial fibrillation (AF) who suffer a stroke are likely to have had a cardioembolic event. On the other hand, AF is common in older adults, who often are at risk for other types of stroke. Thus, the presence of AF in a stroke patient does not always mean that there is a causal relationship [15]. As a result, all patients with a stroke, even in the setting of AF, need consideration of other causes of stroke, especially if they would result in different treatment.

The evaluation is generally the same as the evaluation in other patients with acute stroke, including brain and neurovascular imaging, cardiac monitoring during the acute phase, and echocardiography. As for other patients with a suspected embolic stroke, transesophageal echocardiography (TEE) may be used to identify embolic sources (intracardiac or aortic), which may be particularly helpful for patients at increased risk for complications of anticoagulation. However, a TEE is not used to exclude AF as the cause of embolic stroke, since residual atrial thrombi may or may not be present. (See "Overview of the evaluation of stroke" and "Initial assessment and management of acute stroke".)

Differential diagnosis — For those AF patients with stroke in whom an embolus seems likely, other sources than the left atrial appendage (LAA) need to be considered. Embolism refers to particles of debris originating elsewhere that block arterial access to a particular brain region. Embolic strokes may arise from a source in the heart, aorta, or large vessels (table 1). (See "Etiology, classification, and epidemiology of stroke", section on 'Embolism' and "Clinical diagnosis of stroke subtypes", section on 'Brain ischemia'.)

Thromboembolism of aortic atheroma is discussed separately. (See "Thromboembolism from aortic plaque".)

Echocardiography — Transthoracic echocardiographic evaluation is recommended for most stroke patients, primarily to investigate the conditions associated with AF. Because chronic anticoagulation with warfarin or one of the direct oral anticoagulants (DOAC; also referred to as non-vitamin K oral anticoagulants [NOAC]) is strongly recommended in all eligible patients with AF and stroke, echocardiography often will not have a significant impact on anticoagulant management decisions. (See "Role of echocardiography in atrial fibrillation" and "Epidemiology of and risk factors for atrial fibrillation" and "Atrial fibrillation in adults: Use of oral anticoagulants".)

The LAA and thus LAA thrombus are rarely seen on transthoracic echocardiography but are easily visualized/detected by TEE. Approximately 45 percent of patients presenting with an acute embolic event in the setting of new onset AF will have residual LAA thrombus [16,17]. Even when not seen on TEE, an intracardiac thrombus is presumed to have been present in all patients with AF who have had a recent thromboembolic event independent of anticoagulation status. This hypothesis is based in part upon the observations that microscopic thrombus can be identified in most patients with chronic sustained AF at autopsy [18] and that patients with a recent thromboembolism and newly recognized AF are significantly more likely to have spontaneous echo contrast, a marker of stasis, than similar patients without a thromboembolic event (87 versus 48 percent) [17].

Thus, diagnostic evaluation by TEE to search for a residual intraatrial thrombus is not essential since the absence of a thrombus will not alter the long-term clinical (anticoagulation) management. However, TEE to confirm absence of residual thrombus prior to cardioversion may be reasonable for those in whom a rhythm strategy is going to be pursued. (See "Role of echocardiography in atrial fibrillation" and "Rhythm control versus rate control in atrial fibrillation".)

Cardiac monitoring — For those in sinus rhythm without a history of AF, cardiac monitoring is recommended for at least the first 24 hours after the onset of ischemic stroke to identify AF or atrial flutter [19]. However, paroxysmal AF may not be detected on short-term cardiac monitoring such as continuous telemetry and 24- or 48-hour Holter monitors. Ambulatory cardiac monitoring for several weeks is suggested for all adult patients with a cryptogenic ischemic stroke or cryptogenic TIA. (See "Overview of the evaluation of stroke", section on 'Monitoring for subclinical atrial fibrillation'.)

REPERFUSION THERAPY — All patients with acute ischemic stroke should be evaluated for possible reperfusion therapy. The immediate goal of reperfusion therapy is to restore blood flow to the regions of brain that are ischemic but not yet infarcted. (See "Approach to reperfusion therapy for acute ischemic stroke".)

●Intravenous thrombolysis with alteplase (tPA) improves functional outcome at three to six months when given within 4.5 hours of ischemic stroke onset. However, current anticoagulant use with an international normalized ratio >1.7 or PT >15 seconds or evidence of intracranial hemorrhage on neuroimaging are absolute contraindications to treatment with intravenous alteplase (table 2). (See "Intravenous thrombolytic therapy for acute ischemic stroke: Therapeutic use".)

●Mechanical thrombectomy is indicated for patients with acute ischemic stroke caused by an intracranial large artery occlusion in the proximal anterior circulation who can be treated within 24 hours of the time last known to be well. (See "Mechanical thrombectomy for acute ischemic stroke".)

Specific data on the effectiveness of thrombolytic therapy in ischemic stroke are limited for patients with atrial fibrillation (AF), although such patients account for 20 to 30 percent of those participating in clinical trials [20,21]. As an example, the National Institute of Neurological Disorders and Stroke (NINDS) trial included 115 patients with AF (18 percent) [20]. No subgroup analysis of these patients has been reported, although there was no evidence of a treatment interaction between a history of AF and benefit from alteplase. The large size and worse prognosis of AF-associated acute ischemic stroke accentuates both the risks and the benefits of fibrinolysis [21]. (See "Approach to reperfusion therapy for acute ischemic stroke".)

ACUTE ANTITHROMBOTIC THERAPY — In patients with atrial fibrillation who suffer an ischemic stroke, acute antithrombotic therapy (algorithm 1 and algorithm 2) may be warranted both to reduce disability and the risk of early recurrent stroke, which is 3 to 5 percent in the first two weeks [11,12]. These benefits must be balanced against the risk of intracranial bleeding with antithrombotic therapy. The management of acute antithrombotic therapy in patients with stroke is discussed in detail elsewhere. (See "Early antithrombotic treatment of acute ischemic stroke and transient ischemic attack".)

LONG-TERM ANTICOAGULATION — Warfarin is the most studied antithrombotic therapy for prevention of recurrent stroke in patients with atrial fibrillation (AF) and an ischemic stroke and its efficacy is well demonstrated (see 'Warfarin' below).

Direct oral anticoagulants (DOAC), also referred to as non-vitamin K oral anticoagulants (NOAC), have been less extensively studied but appear to be equally efficacious, though preliminary secondary analysis of data from the AUGUSTUS trial that enrolled patients with AF and recent acute coronary syndrome suggested that apixaban was superior to vitamin K antagonists with regard to intracranial bleeding and stroke/TIA/thromboembolism in those with a prior stroke/TIA/thromboembolism [22]. (See 'Dabigatran' below and 'Rivaroxaban' below and 'Apixaban' below and 'Edoxaban' below.)

Patients with a previous intracranial hemorrhage may be candidates for anticoagulation, depending upon their risk of recurrent ischemic stroke and intracranial bleeding. (See "Spontaneous intracerebral hemorrhage: Secondary prevention and long-term prognosis", section on 'Anticoagulation'.)

Timing after acute ischemic stroke — Anticoagulation with warfarin or a DOAC is recommended in eligible patients. For medically stable patients with a small- or moderate-sized infarct with no intracranial bleeding, warfarin can be initiated soon (after 24 hours) after admission with minimal risk of transformation to hemorrhagic stroke. We prefer to wait for 48 hours to start a DOAC in these patients, as DOACs have a more rapid anticoagulant effect.

Although once widely practiced, early treatment with heparin for patients with atrial fibrillation who have an acute cardioembolic stroke should generally be avoided as studies have shown that such treatment causes more harm than good. (See "Early antithrombotic treatment of acute ischemic stroke and transient ischemic attack", section on 'Atrial fibrillation'.)

Withholding anticoagulation for two weeks is generally recommended for those with large ischemic strokes [23], symptomatic hemorrhagic transformation, or poorly controlled hypertension. Patients who are not treated with warfarin earlier may benefit from aspirin until therapeutic anticoagulation is achieved [24]. (See "Early antithrombotic treatment of acute ischemic stroke and transient ischemic attack".)

Patients with lacunar infarction — The optimal therapy is not known for AF patients with hypertension who experience a small subcortical "lacunar" infarct deemed as likely to be due to cerebral small artery disease as opposed to a cardiac embolus [25]. Anticoagulation is recommended, given the large benefit of warfarin or DOACs consistently documented in randomized trials and the inability to define stroke mechanism with certainty.

Warfarin — Aspirin alone offers inadequate protection, with a stroke risk that averaged 10 percent per year in a pooled analysis of individual participants in six randomized trials [26]. Compared with aspirin, treatment with adjusted-dose warfarin (international normalized ratio 2 to 3) reduced this risk to 4 percent per year.

In an analysis from the European Atrial Fibrillation Trial (EAFT) and Stroke Prevention in Atrial Fibrillation (SPAF) III of 834 patients with prior nondisabling ischemic stroke or prior transient ischemic attack (TIA) at study entry, the long-term risk of recurrent stroke was lower in patients with a prior TIA than in those with a completed ischemic stroke [27]. However, the reduction in recurrent stroke risk with warfarin therapy was comparable in both groups: 3 versus 7 percent per year with aspirin in patients with a TIA and 4 versus 11 percent per year in those with a completed ischemic stroke.

Direct oral anticoagulants

Apixaban — Apixaban was compared with adjusted-dose warfarin in the ARISTOTLE trial, involving over 18,000 patients, and was found to be superior to warfarin in preventing stroke or systemic embolism (1.3 versus 1.6 percent) [28]. Apixaban also caused less major bleeding compared with warfarin (2.1 versus 3.1 percent) and resulted in lower overall mortality (3.5 versus 3.9 percent) [28].

Among the 3436 subjects with a prior stroke or TIA, there was no significant difference in stroke and systemic embolism between apixaban and warfarin (apixaban 2.5 versus warfarin 3.2 percent).

Dabigatran — Dabigatran was compared with adjusted-dose warfarin in a trial involving over 18,000 AF patients [29]. Dabigatran compared favorably to warfarin, including in the subset of participants with prior stroke.

Edoxaban — Edoxaban was compared with warfarin in the ENGAGE TIMI 48 trial of over 21,000 patients [30]. Edoxaban was found to be noninferior with regard to the primary efficacy end point and caused less bleeding. Outcomes were similar in the subgroup of patients with prior stroke or transient ischemic attack as in the entire cohort.

Rivaroxaban — Rivaroxaban was compared with adjusted-dose warfarin in the ROCKET AF trial of over 14,000 patients and was found to be non-inferior (1.7 versus 2.2 percent per year, respectively) with regard to the primary composite end point of stroke or non-central nervous systemic embolism [31].

Similar to the entire cohort, there was no significant difference in the primary composite outcome between rivaroxaban and warfarin in the subgroup of patients (52 percent) with prior stroke (2.79 versus 2.96 percent per year) [32].

Anticoagulant-intolerant patients — Dual antiplatelet therapy may be a reasonable alternative to therapy with aspirin alone in high-risk patients with AF who cannot be treated with warfarin or DOAC (see 'Dabigatran' above and 'Rivaroxaban' above and 'Apixaban' above and 'Edoxaban' above), or because of strong patient preference following careful consideration of the advantages of oral anticoagulation. This issue is discussed separately.

STROKE WHILE ON ANTICOAGULATION — All patients with AF who have an ischemic stroke despite oral anticoagulation with warfarin or a direct oral anticoagulant (DOAC) should have a thorough evaluation to determine if the most likely stroke mechanism is cardioembolic due to AF or noncardioembolic due to large artery atherosclerosis, small vessel disease, or another cause of ischemic stroke. (See 'Evaluation' above.)

Subtherapeutic intensity of anticoagulation is the most common cause of treatment failure for patients taking warfarin, and missed doses should be suspected for patients taking a DOAC. For patients with stroke while on warfarin anticoagulation with a therapeutic international normalized ratio (INR) or on DOAC anticoagulation with good compliance, a noncardioembolic stroke mechanism (eg, lacunar, large artery stenosis, malignancy) is often the cause.

●Warfarin – In patients with AF who suffer ischemic stroke during warfarin anticoagulation, the intensity of anticoagulation is most often subtherapeutic (INR less than 2), and continuing warfarin (after two weeks of temporary interruption for patients with large infarcts) with renewed efforts to keep the INR in the 2 to 3 therapeutic range or consideration of a change to a DOAC is recommended. When ischemic stroke occurs with a therapeutic INR (2 to 3), it is reasonable to perform a transesophageal echocardiogram (TEE) to assess for left atrial appendage (LAA) thrombus to confirm the presence of residual thrombus, though the ischemic stroke is often "lacunar" related to cerebral small artery disease, rather than cardioembolic etiology. Nevertheless, we favor increasing the target INR to 2.5 to 3.5, or switching from warfarin to a DOAC rather than routine addition of antiplatelet therapy. The addition of antiplatelet therapy is known to increase major hemorrhage (and particularly brain hemorrhage) and the benefit is less well defined.

The management of patients with AF who develop a warfarin-associated intracranial hemorrhage is discussed separately. (See "Reversal of anticoagulation in intracranial hemorrhage".)

●DOAC – While data are limited, ischemic stroke during therapy with a DOAC (eg, apixaban, dabigatran, edoxaban, or rivaroxaban) for AF has been associated with several factors, including treatment at doses lower than recommended. One study compared 713 cases of ischemic stroke or transient ischemic attack (TIA) during DOAC treatment with unmatched controls (consecutive outpatients with AF) who did not have cerebrovascular events during DOAC treatment [33]. In multivariable analysis, ischemic cerebrovascular events were associated with off-label under-dosing of DOAC, atrial enlargement, hyperlipidemia, and high CHA2DS2-VASc score. When ischemic stroke or TIA occurs during DOAC therapy, it is reasonable to perform a TEE to assess for LAA thrombus and if present, reassess compliance. Regardless, continuing anticoagulation therapy (after two weeks of temporary interruption with large infarcts) is generally indicated. It is important to verify that the correct DOAC dose is prescribed and that the patient is compliant. If a thrombus is present despite appropriate dosing and compliance, it is reasonable to change to another DOAC, but optimal treatment is uncertain and no consensus exists. A retrospective study suggested that for patients with left ventricular thrombus, warfarin may be superior to DOAC for reducing the risk of stroke or systemic embolism [34]. Analogous data for AF patients with left atrial appendage thrombi are not available.

Note that there are legitimate reasons for DOAC dose reductions, which differ according to the specific agent. In general, clinical settings in which dose modification may be indicated include older age, low body weight, renal insufficiency, and/or concomitant use of interacting drugs. These are reviewed separately. (See "Direct oral anticoagulants (DOACs) and parenteral direct-acting anticoagulants: Dosing and adverse effects".)

ADDITIONAL RISK REDUCTION STRATEGIES

Control of hypertension — Blood pressure control is an important component of the management of patients with AF who have had a stroke. Antihypertensive therapy, preferably including an angiotensin converting enzyme inhibitor, reduces the risk of warfarin-associated intracranial hemorrhage and may reduce the rate of recurrent stroke. (See "Reversal of anticoagulation in intracranial hemorrhage".)

The latter benefit was suggested in a secondary analysis from the PROGRESS trial, which demonstrated the benefit of blood pressure lowering (using perindopril-indapamide) among both hypertensive and nonhypertensive patients who had a previous stroke or TIA [35]. (See "Antihypertensive therapy for secondary stroke prevention".)

Among the subset of 476 patients with AF, perindopril-based therapy produced a mean 7.3/3.4 mmHg reduction in blood pressure compared to placebo and a 34 percent reduction in the incidence of recurrent stroke (13.6 versus 18.9 percent), a difference that was not statistically significant because of the small number of recurrent events [36]. However, there was a significant 38 percent reduction in all major vascular events (one major vascular event prevented in every 11 patients treated for five years), providing a strong rationale for blood pressure lowering.

Revascularization for carotid artery stenosis — About 10 percent of AF patients with ischemic stroke or TIA have a cervical carotid stenosis of 50 percent or greater diameter stenosis, slightly more than half of which are ipsilateral to the neurological symptoms. Based on estimates of attributable risk, ipsilateral stenosis of at least 70 percent stenosis is equally likely to be the cause of cerebral ischemia as is cardiogenic embolism. Consequently, carotid endarterectomy or stenting seems reasonable for AF patients with high-grade ipsilateral stenosis, followed by chronic anticoagulation and antiplatelet therapy, although this approach is empiric, without good supporting evidence. The management of symptomatic carotid artery disease is discussed elsewhere. (See "Management of symptomatic carotid atherosclerotic disease".)

Statin therapy — For most patients with ischemic stroke, we start statin therapy. Statin therapy reduces the risk of recurrent ischemic stroke and cardiovascular events among patients with stroke of atherosclerotic origin, although the efficacy of statin therapy specifically for patients with ischemic stroke attributed to AF has not been well studied. However, a report of 6116 patients with ischemic stroke who were discharged on a statin found that outpatient adherence to statin therapy was associated with a reduced risk of recurrent ischemic stroke for patients with AF as well as those without AF, even after adjustment for time in the therapeutic range of the international normalized ratio among patients with AF taking warfarin [37]. Many patients with AF have concomitant atherosclerotic disease, and statin therapy is recommended for patients with atherosclerotic cardiovascular disease (such as prior acute coronary syndrome, myocardial infarction, stable or unstable angina, coronary or other arterial revascularization, ischemic stroke, TIA, or peripheral arterial disease) (see "Overview of secondary prevention of ischemic stroke", section on 'LDL-C lowering therapy'). In addition, and in the absence of defined atherosclerotic cardiovascular disease (CVD), many patients are at high risk for a CVD event due to age and the presence of hypertension. (See "Management of elevated low density lipoprotein-cholesterol (LDL-C) in primary prevention of cardiovascular disease", section on 'Older patients'.)

Lifestyle modification — A number of behavioral and lifestyle modifications may be beneficial for reducing the risk of ischemic stroke and cardiovascular disease. These include smoking cessation, limited alcohol consumption, weight control, regular aerobic physical activity, salt restriction, and a Mediterranean diet. (See "Overview of secondary prevention of ischemic stroke", section on 'Lifestyle modification'.)

SUMMARY AND RECOMMENDATIONS

●Ischemic stroke may be the presenting manifestation of atrial fibrillation (AF) in some patients, while in others it may occur despite appropriate anticoagulant prophylaxis.

●AF is associated with more severe ischemic strokes and "longer" transient ischemic attacks than emboli from carotid disease, presumably due to embolization of larger thrombi in AF. AF is associated with particularly severe ischemic strokes, mostly caused by relatively large emboli from the left atrial appendage. (See 'Stroke characteristics' above.)

●Evaluation of patients with AF and acute stroke is similar to other patients with stroke. All patients with acute ischemic stroke should be assessed to see if they are eligible for reperfusion therapy. (See 'Evaluation' above and 'Reperfusion therapy' above.)

●Neuroimaging with cranial computed tomography or magnetic resonance imaging should be obtained before commencing reperfusion or antithrombotic therapy to confirm the absence of intracranial hemorrhage and to assess the size of any cerebral infarction. The size of the infarction impacts the timing of warfarin or direct oral anticoagulant (DOAC; also referred to as non-vitamin K oral anticoagulants [NOAC]) initiation.

●Chronic anticoagulation with warfarin (target international normalized ratio [INR] 2 to 3) or a DOAC is the most effective long-term therapy for prevention of recurrent stroke in patients with AF. Patients with a previous intracranial hemorrhage may be candidates for anticoagulation depending upon their risk of recurrent ischemic stroke and intracranial hemorrhage. (See 'Long-term anticoagulation' above.)

●For patients who develop an ischemic stroke while on anticoagulation, subtherapeutic intensity of anticoagulation (eg, low INR; inappropriate low-dose DOAC, or missed DOAC doses) is the most common cause of treatment failure. In such cases, it is reasonable to perform a transesophageal echocardiogram to assess for left atrial appendage (LAA) thrombus, and if present despite warfarin therapy, to increase the target INR to 2.5 to 3.5 or switch to a DOAC. If an LAA thrombus is detected despite DOAC therapy at the correct dose with good patient compliance, it may be reasonable to switch to warfarin. (See 'Stroke while on anticoagulation' above.)