Original Paper Received: April 15, 2013 Accepted: September 10, 2013 Published online: November 9, 2013
Cerebrovasc Dis 2013;36:373–382 DOI: 10.1159/000355571
Is Atrial Fibrillation Always a Culprit of Stroke in Patients with Atrial Fibrillation plus Stroke? Suk Jae Kim a Sookyung Ryoo a Soonwook Kwon a Yun Kyung Park a Jun Pyo Kim a Ga Yeon Lee b Oh Young Bang a
Departments of a Neurology and b Cardiology, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, South Korea
Key Words Ischemic stroke · Stroke subtype · Atrial fibrillation · Cardioembolism · Transthoracic echocardiography · Tissue Doppler imaging
Abstract Background: Some ischemic strokes in patients with atrial fibrillation (AF) are caused by noncardioembolic etiologies (AF-unrelated stroke), but not AF itself (AF-related stroke). However, most clinical trials on the risk of stroke in AF have not distinguished between these. We investigated the frequency and features of AF-unrelated versus AF-related strokes in patients with AF plus ischemic stroke. We hypothesized that certain clinical factors, including chronicity of AF, treatment at the time of stroke onset and echocardiographic findings, may help to discriminate between AF-related and AF-unrelated strokes. The mechanisms and antithrombotic medications at the time of stroke recurrence in the two groups were also examined. Methods: Consecutive patients with ischemic stroke within 7 days of symptom onset and with AF were included. Patients were classified according to the previously published criteria. Clinical factors including CHADS2 and CHA2DS2-VASc scores and transthoracic echocardiographic (TTE) findings were evaluated. Results: Of 522 patients, 424 (81.2%) were grouped as AF-related stroke and the remaining 90 (17.2%) were classified as AF-unrelated
© 2013 S. Karger AG, Basel 1015–9770/13/0366–0373$38.00/0 E-Mail [email protected] www.karger.com/ced
stroke. Among the patients with AF-unrelated stroke, 51 (9.8%) were categorized as possible large artery atherosclerosis and 38 (7.3%) as possible small artery occlusion; 1 patient (0.2%) was assigned to miscellaneous cause. The AF-related and AF-unrelated strokes had similar CHADS2 and CHA2DS2VASc scores. However, compared to AF-unrelated stroke, AFrelated stroke was independently associated with female sex (odds ratio, OR, 2.19; 95% confidence interval, CI, 1.18–4.05), sustained AF (OR, 2.09; 95% CI, 1.21–3.59), inadequate anticoagulation at stroke onset (OR, 3.21; 95% CI, 1.33–7.75) and left ventricular dysfunction on TTE (OR, 2.84; 95% CI, 1.40–5.74). We identified 26 patients who experienced 2 strokes during the study period. The initial stroke subtype was a strong predictor of the recurrent stroke mechanism (p < 0.001). Among 17 events of AF-related recurrent stroke in these subpopulation, only 2 strokes (11.8%) occurred in a setting of adequate anticoagulation, whereas 4 out of 9 patients (44.4%) who had AF-unrelated strokes at recurrence were sufficiently anticoagulated at the time of admission (p = 0.138). Conclusion: AF is not always a culprit of stroke in patients with AF plus ischemic stroke; approximately one sixth of these cases are unrelated to AF and have distinct characteristics compared to AFrelated stroke. There are significant differences in terms of some clinical and TTE parameters between AF-related and AF-unrelated stroke. Future studies are warranted to optimize strategies for risk stratification, treatment and prevention of stroke in these patients. © 2013 S. Karger AG, Basel
Oh Young Bang, MD, PhD Department of Neurology and the Stroke and Cerebrovascular Center Samsung Medical Center, Sungkyunkwan University 50 Irwon-dong, Gangnam-gu, Seoul 135-710 (South Korea) E-Mail nmboy @ unitel.co.kr
Introduction
The risk of stroke occurrence in patients with atrial fibrillation (AF) increases with the presence of other risk factors including advanced age, hypertension, diabetes, congestive heart failure, prior stroke or transient ischemic attack (TIA), myocardial infarction, peripheral artery disease, aortic plaque and female sex [1, 2]. Since most of these risk factors also enhance the risk of other types of stroke such as large artery atherosclerosis (LAA) or small artery occlusion (SAO), one can assume that some of the ischemic strokes in patients with AF are caused by noncardioembolic mechanisms (AFunrelated stroke), but not AF per se (AF-related stroke) [3–5]. However, most clinical trials on the risk of stroke in patients with AF did not distinguish between these [6–9]. Although vitamin K antagonist therapy (warfarin) is highly effective in reducing the risk of stroke and death in patients with AF [10], some strokes occur while the patient takes warfarin. There may be two major possibilities: inadequate anticoagulation at the time of stroke onset or insufficient efficacy of warfarin for the prevention of AFunrelated stroke. In recent years, there has been increasing interest in novel oral anticoagulants (NOACs) to overcome the unpredictable anticoagulant response of warfarin [7–9]. Indeed, new stroke guidelines issued in 2012 recommended warfarin or NOACs for the prevention of stroke in patients with AF [11]. However, anticoagulants may have limited effects for the prevention of AF-unrelated stroke in patients with ischemic stroke plus AF [3]. In this study, we investigated the frequency and characteristics of AF-related versus AF-unrelated strokes in patients with AF plus ischemic stroke. We hypothesized that certain clinical factors, including chronicity of AF, treatment at the time of stroke onset and echocardiographic findings, may help to discriminate between AFrelated and AF-unrelated strokes. The mechanisms and antithrombotic medications at the time of stoke recurrence in the two groups were also examined.
Methods Patient Selection, Evaluation and Grouping This retrospective analysis was performed on data collected from a prospective registry of patients with ischemic stroke who presented to a university hospital stroke center within 7 days after symptom onset between June 2006 and March 2012 (n = 3,472). From this initial population, we excluded 2,900 patients without
374
Cerebrovasc Dis 2013;36:373–382 DOI: 10.1159/000355571
AF as well as subjects with prosthetic heart valve (n = 26) or valvular AF (n = 21) and those whose diagnosis of stroke was not proven owing to the lack of brain imaging (n = 3; fig. 1). Valvular AF was defined as AF with rheumatic valvular heart disease, significant (greater than moderate) aortic valve disease (stenosis or regurgitation), mitral regurgitation, or any degree of mitral stenosis other than rheumatic origin. All patients received standard medical therapy in accordance with conventional guidelines. The local institutional review board approved this study and written informed consent was not obtained due to the retrospective study design. We evaluated patients according to a protocol that included demographic data, medical history, vascular risk factors and diagnostic testing including neuroimaging (brain magnetic resonance imaging, MRI, and MR angiography, MRA), electrocardiograms and transthoracic echocardiograms (TTE). The vascular imaging was conducted using 3D time-of-flight MRA for the intracranial arterial system and contrast-enhanced MRA including the extracranial internal carotid artery and the vertebral artery for the extracranial arterial system by 3.0-tesla MRI scanners. In addition, some of the patients were also evaluated by computed tomography (CT) angiography or ultrasound to improve diagnostic accuracy for the detection of significant stenosis. Medical history and vascular risk factors included hypertension, diabetes, hyperlipidemia, congestive heart failure, coronary heart disease, peripheral artery disease, smoking, history of stroke or TIA, and previous use of antiplatelet or warfarin medications (patients with new anticoagulants such as dabigatran etexilate, rivaroxaban and apixaban were not included). Fasting blood samples were drawn within 24 h of admission and were analyzed for glucose, lipid panel and a standard battery of biochemical and hematological tests including prothrombin time. Since achieving international normalized ratio (INR) levels equally distributed in the range of 1.6–2.5 is predicted to provide 90% of the protection of INRs between 2.0 and 3.0 for primary prevention of stroke in patients with nonvalvular AF, patients were considered to be adequately anticoagulated at the time of stroke onset if the INR levels were ≥1.6 at admission [12]. Five risk factors are considered when determining the CHADS2 score and the basis of the acronym: C denotes congestive heart failure or left ventricular (LV) ejection fraction (EF) ≤40%, H denotes hypertension, A denotes age ≥75 years, D denotes diabetes, and S2 denotes history of stroke or TIA – 2 points are given for a prior stroke or TIA, and 1 point is assigned for each of the other factors [1]. We investigated an additional four stroke risk factors for calculating the new CHA2DS2-VASc score: A2 denotes an age ≥75 years; V denotes vascular disease including prior myocardial infarction, peripheral artery disease or aortic plaque; A denotes age between 65 and 74 years, and Sc denotes the sex category (female) – 2 points were added for age ≥75 years and a prior stroke or TIA, and 1 point was given for each of the other factors [2]. In this study, the presenting stroke did not add points to both CHADS2 and CHA2DS2-VASc scores. Based on these scoring systems, patients were stratified as low (CHADS2 or CHA2DS2VASc = 0), intermediate (CHADS2 or CHA2DS2-VASc = 1), or high (CHADS2 or CHA2DS2-VASc ≥2) risks for stroke, as described previously [13]. The diagnosis of AF was based on electrocardiographic findings and/or 24-hour Holter monitoring during the hospitalization, assessed by a cardiologist, or on data from medical history with an
Kim/Ryoo/Kwon/Park/Kim/Lee/Bang
Ischemic stroke
Cerebrovasc Dis 2013;36:373–382 DOI: 10.1159/000355571
Is Atrial Fibrillation Always a Culprit of Stroke in Patients with Atrial Fibrillation plus Stroke? Suk Jae Kim a Sookyung Ryoo a Soonwook Kwon a Yun Kyung Park a Jun Pyo Kim a Ga Yeon Lee b Oh Young Bang a
Departments of a Neurology and b Cardiology, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, South Korea
Key Words Ischemic stroke · Stroke subtype · Atrial fibrillation · Cardioembolism · Transthoracic echocardiography · Tissue Doppler imaging
Abstract Background: Some ischemic strokes in patients with atrial fibrillation (AF) are caused by noncardioembolic etiologies (AF-unrelated stroke), but not AF itself (AF-related stroke). However, most clinical trials on the risk of stroke in AF have not distinguished between these. We investigated the frequency and features of AF-unrelated versus AF-related strokes in patients with AF plus ischemic stroke. We hypothesized that certain clinical factors, including chronicity of AF, treatment at the time of stroke onset and echocardiographic findings, may help to discriminate between AF-related and AF-unrelated strokes. The mechanisms and antithrombotic medications at the time of stroke recurrence in the two groups were also examined. Methods: Consecutive patients with ischemic stroke within 7 days of symptom onset and with AF were included. Patients were classified according to the previously published criteria. Clinical factors including CHADS2 and CHA2DS2-VASc scores and transthoracic echocardiographic (TTE) findings were evaluated. Results: Of 522 patients, 424 (81.2%) were grouped as AF-related stroke and the remaining 90 (17.2%) were classified as AF-unrelated
© 2013 S. Karger AG, Basel 1015–9770/13/0366–0373$38.00/0 E-Mail [email protected] www.karger.com/ced
stroke. Among the patients with AF-unrelated stroke, 51 (9.8%) were categorized as possible large artery atherosclerosis and 38 (7.3%) as possible small artery occlusion; 1 patient (0.2%) was assigned to miscellaneous cause. The AF-related and AF-unrelated strokes had similar CHADS2 and CHA2DS2VASc scores. However, compared to AF-unrelated stroke, AFrelated stroke was independently associated with female sex (odds ratio, OR, 2.19; 95% confidence interval, CI, 1.18–4.05), sustained AF (OR, 2.09; 95% CI, 1.21–3.59), inadequate anticoagulation at stroke onset (OR, 3.21; 95% CI, 1.33–7.75) and left ventricular dysfunction on TTE (OR, 2.84; 95% CI, 1.40–5.74). We identified 26 patients who experienced 2 strokes during the study period. The initial stroke subtype was a strong predictor of the recurrent stroke mechanism (p < 0.001). Among 17 events of AF-related recurrent stroke in these subpopulation, only 2 strokes (11.8%) occurred in a setting of adequate anticoagulation, whereas 4 out of 9 patients (44.4%) who had AF-unrelated strokes at recurrence were sufficiently anticoagulated at the time of admission (p = 0.138). Conclusion: AF is not always a culprit of stroke in patients with AF plus ischemic stroke; approximately one sixth of these cases are unrelated to AF and have distinct characteristics compared to AFrelated stroke. There are significant differences in terms of some clinical and TTE parameters between AF-related and AF-unrelated stroke. Future studies are warranted to optimize strategies for risk stratification, treatment and prevention of stroke in these patients. © 2013 S. Karger AG, Basel
Oh Young Bang, MD, PhD Department of Neurology and the Stroke and Cerebrovascular Center Samsung Medical Center, Sungkyunkwan University 50 Irwon-dong, Gangnam-gu, Seoul 135-710 (South Korea) E-Mail nmboy @ unitel.co.kr
Introduction
The risk of stroke occurrence in patients with atrial fibrillation (AF) increases with the presence of other risk factors including advanced age, hypertension, diabetes, congestive heart failure, prior stroke or transient ischemic attack (TIA), myocardial infarction, peripheral artery disease, aortic plaque and female sex [1, 2]. Since most of these risk factors also enhance the risk of other types of stroke such as large artery atherosclerosis (LAA) or small artery occlusion (SAO), one can assume that some of the ischemic strokes in patients with AF are caused by noncardioembolic mechanisms (AFunrelated stroke), but not AF per se (AF-related stroke) [3–5]. However, most clinical trials on the risk of stroke in patients with AF did not distinguish between these [6–9]. Although vitamin K antagonist therapy (warfarin) is highly effective in reducing the risk of stroke and death in patients with AF [10], some strokes occur while the patient takes warfarin. There may be two major possibilities: inadequate anticoagulation at the time of stroke onset or insufficient efficacy of warfarin for the prevention of AFunrelated stroke. In recent years, there has been increasing interest in novel oral anticoagulants (NOACs) to overcome the unpredictable anticoagulant response of warfarin [7–9]. Indeed, new stroke guidelines issued in 2012 recommended warfarin or NOACs for the prevention of stroke in patients with AF [11]. However, anticoagulants may have limited effects for the prevention of AF-unrelated stroke in patients with ischemic stroke plus AF [3]. In this study, we investigated the frequency and characteristics of AF-related versus AF-unrelated strokes in patients with AF plus ischemic stroke. We hypothesized that certain clinical factors, including chronicity of AF, treatment at the time of stroke onset and echocardiographic findings, may help to discriminate between AFrelated and AF-unrelated strokes. The mechanisms and antithrombotic medications at the time of stoke recurrence in the two groups were also examined.
Methods Patient Selection, Evaluation and Grouping This retrospective analysis was performed on data collected from a prospective registry of patients with ischemic stroke who presented to a university hospital stroke center within 7 days after symptom onset between June 2006 and March 2012 (n = 3,472). From this initial population, we excluded 2,900 patients without
374
Cerebrovasc Dis 2013;36:373–382 DOI: 10.1159/000355571
AF as well as subjects with prosthetic heart valve (n = 26) or valvular AF (n = 21) and those whose diagnosis of stroke was not proven owing to the lack of brain imaging (n = 3; fig. 1). Valvular AF was defined as AF with rheumatic valvular heart disease, significant (greater than moderate) aortic valve disease (stenosis or regurgitation), mitral regurgitation, or any degree of mitral stenosis other than rheumatic origin. All patients received standard medical therapy in accordance with conventional guidelines. The local institutional review board approved this study and written informed consent was not obtained due to the retrospective study design. We evaluated patients according to a protocol that included demographic data, medical history, vascular risk factors and diagnostic testing including neuroimaging (brain magnetic resonance imaging, MRI, and MR angiography, MRA), electrocardiograms and transthoracic echocardiograms (TTE). The vascular imaging was conducted using 3D time-of-flight MRA for the intracranial arterial system and contrast-enhanced MRA including the extracranial internal carotid artery and the vertebral artery for the extracranial arterial system by 3.0-tesla MRI scanners. In addition, some of the patients were also evaluated by computed tomography (CT) angiography or ultrasound to improve diagnostic accuracy for the detection of significant stenosis. Medical history and vascular risk factors included hypertension, diabetes, hyperlipidemia, congestive heart failure, coronary heart disease, peripheral artery disease, smoking, history of stroke or TIA, and previous use of antiplatelet or warfarin medications (patients with new anticoagulants such as dabigatran etexilate, rivaroxaban and apixaban were not included). Fasting blood samples were drawn within 24 h of admission and were analyzed for glucose, lipid panel and a standard battery of biochemical and hematological tests including prothrombin time. Since achieving international normalized ratio (INR) levels equally distributed in the range of 1.6–2.5 is predicted to provide 90% of the protection of INRs between 2.0 and 3.0 for primary prevention of stroke in patients with nonvalvular AF, patients were considered to be adequately anticoagulated at the time of stroke onset if the INR levels were ≥1.6 at admission [12]. Five risk factors are considered when determining the CHADS2 score and the basis of the acronym: C denotes congestive heart failure or left ventricular (LV) ejection fraction (EF) ≤40%, H denotes hypertension, A denotes age ≥75 years, D denotes diabetes, and S2 denotes history of stroke or TIA – 2 points are given for a prior stroke or TIA, and 1 point is assigned for each of the other factors [1]. We investigated an additional four stroke risk factors for calculating the new CHA2DS2-VASc score: A2 denotes an age ≥75 years; V denotes vascular disease including prior myocardial infarction, peripheral artery disease or aortic plaque; A denotes age between 65 and 74 years, and Sc denotes the sex category (female) – 2 points were added for age ≥75 years and a prior stroke or TIA, and 1 point was given for each of the other factors [2]. In this study, the presenting stroke did not add points to both CHADS2 and CHA2DS2-VASc scores. Based on these scoring systems, patients were stratified as low (CHADS2 or CHA2DS2VASc = 0), intermediate (CHADS2 or CHA2DS2-VASc = 1), or high (CHADS2 or CHA2DS2-VASc ≥2) risks for stroke, as described previously [13]. The diagnosis of AF was based on electrocardiographic findings and/or 24-hour Holter monitoring during the hospitalization, assessed by a cardiologist, or on data from medical history with an
Kim/Ryoo/Kwon/Park/Kim/Lee/Bang
Ischemic stroke
