Clinical Characteristics of Cardioembolic Transient Ischemic Attack: Comparison with Noncardioembolic Transient Ischemic Attack Takeshi Hayashi, MD, Yoshihide Seahara, MD, Yuji Kato, MD, Takuya Fukuoka, MD, Ichiro Deguchi, MD, Yasuko Ohe, MD, Hajime Maruyama, MD, Yohsuke Horiuchi, MD, Hiroyasu Sano, MD, Yuito Nagamine, MD, and Norio Tanahashi, MD
Background: Previous studies show that 6%-31% of transient ischemic attacks (TIA) were caused by cardiogenic cerebral embolism (cardioembolic TIA). As prompt initiation of therapy is essential in TIA to prevent subsequent strokes, determining their cause is important. In this study, we aim to determine the features of cardioembolic TIA and to compare them with those of noncardioembolic etiology. Methods: We retrospectively reviewed patients with a tissue-defined TIA who were admitted to our hospital from April 2007 to August 2013. The etiology was categorized according to Trial of Org 10172 in Acute Stroke Treatment, and TIA of cardioembolic origin and cervicocerebrovascular etiology (noncardioembolic TIA) were included in this study. Those with 2 or more possible causes or undetermined etiologies were excluded. Age, sex, comorbidities, ABCD2 score, and CHADS2 score were assessed and compared between the 2 groups. Results: There were no significant differences in the neurologic symptoms and their duration, morbidities of hypertension, diabetes, and dyslipidemia between the 2 groups. Coronary and peripheral artery diseases were more common in the cardioembolic TIA group (18.4% vs. 6.9%). Incidences of prior stroke and cerebral infarction determined by MRI were similar between the 2 groups. The ABCD2 score showed a similar distribution, but the CHADS2 score was significantly different; the cardioembolic TIA group showed a higher score (P 5 .005). Conclusions: Clinical features are similar in tissue-defined TIA of cardioembolic and noncardioembolic etiologies. The CHADS2 score can be useful in assessing the probability of cardioembolic TIA. Key Words: Cardioembolism—clinical score—etiology—transient ischemic attack. Ó 2014 by National Stroke Association
From the Department of Neurology and Cerebrovascular Medicine, Saitama Medical University, International Medical Center, Hidaka, Japan. Received February 16, 2014; accepted April 2, 2014. Address correspondence to Takeshi Hayashi, MD, Department of Neurology and Cerebrovascular Medicine, Saitama Medical University, International Medical Center, 1397-1 Hidaka, Saitama 350-1298, Japan. E-mail: [email protected]. 1052-3057/$ - see front matter Ó 2014 by National Stroke Association http://dx.doi.org/10.1016/j.jstrokecerebrovasdis.2014.04.005
Ten to 20% of patients who experience a transient ischemic attack (TIA) will have a stroke within 3 months, and approximately 50% within 48 hours.1-3 Moreover, a recent study revealed that more than 2% of patients with TIA experienced a stroke within 12 hours.4 It is important to administer antithrombotic therapy immediately because prompt and appropriate therapy can help prevent more than 80% of subsequent strokes.5,6 However, clinicians often hesitate over which treatment to choose, as the etiology of TIA is sometimes
Journal of Stroke and Cerebrovascular Diseases, Vol. 23, No. 8 (September), 2014: pp 2169-2173
2169
T. HAYASHI ET AL.
2170
undiagnosed when stroke-preventive therapy is initiated. The National Institute for Clinical Excellence7 and Japanese8 guidelines recommend that all TIA patients be promptly administered 160-300 mg of aspirin per day, even when the etiology has not yet been determined, but the optimal method to prevent subsequent strokes may vary among patients. Approximately 5% of patients with TIA received anticoagulation therapy and another 5% underwent urgent carotid revascularization in a TIA clinic.5 As with cerebral infarction, the optimal medical treatment can differ based on TIA etiology. Approximately 6%-31% of TIAs are caused by a cardiogenic cerebral embolism (cardioembolic TIA).9,10 Determining TIA etiology is important before administering therapy, but it is difficult in some cases. In this study, we aimed to compare the features of a cardioembolic TIA with those of a noncardioembolic TIA. These kind of studies have been carried out previously, although a time-based definition of TIA was used.11,12 Indeed, the present study with tissue-defined TIA presented some unique findings.
Methods This study was approved by the Institutional Review Board at Saitama Medical University International Medical Center. All patients who were admitted to our hospital with transient focal neurologic deficits underwent magnetic resonance imaging (MRI), unless contraindicated. If the MRI revealed no fresh lesions on diffusion-weighed imaging, a second MRI was performed 3-7 days later. Those without high-intensity lesions on both diffusion-weighed images were diagnosed with tissue-defined TIA. Patients with lasting focal neurologic deficits on admission that resolved within 24 hours were also included, if repeat MRI after resolution of the neurologic deficits showed no new lesions.13 In total, 171 patents had tissue-defined TIA from April 2007 to August 2013. Etiology was categorized according to the Trial of Org 10172 in Acute Stroke Treatment.14 A cardiogenic source was investigated using 12-lead electrocardiography (ECG) and transthoracic echocardiography in all cases. In certain cases, we performed 24-hour Holter ECG, bedside ECG monitoring for several days, and transesophageal echocardiography when needed. Among the 171 tissue-defined TIA patients, 38 had cardioembolic TIA and 116 had either large artery atherosclerosis or small vessel disease (noncardioembolic TIA). Patients with a TIA of 2 or more possible causes (eg, atrial fibrillation [AF] and carotid artery stenosis) or undetermined etiology were not included in this study. We retrospectively reviewed the medical records of patients, including demographic information (age, sex), clinical history of symptomatic stroke, stroke risk factors (hypertension, diabetes mellitus, dyslipidemia), symp-
toms (presence of hemiparesis and/or dysarthria during the TIA), and duration of symptoms. Patients with a blood pressure of $140/90 mm Hg or those who were receiving antihypertensive medication were defined as having hypertension.15 Diabetes mellitus was defined as a blood glucose level of $200 mg/dL and a glycated hemoglobin level of $6.5% on admission or treatment with antidiabetic medication.15 Dyslipidemia was diagnosed if the patient had any of the following: lowdensity lipoprotein cholesterol level $140 mg/dL, high-density lipoprotein cholesterol level #40 mg/dL, triglyceride level $150 mg/dL, or treatment with lipidlowering medication.15 We investigated old cerebral infarction (including asymptomatic episodes) using MRI; high-intensity lesions on T2-weighed images and fluid-attenuated inversion recovery images were considered old infarctions.13 We also reviewed vascular diseases other than cerebrovascular disease, such as coronary heart or peripheral artery diseases. ABCD2 and CHADS2 scores, which are stroke risk stratification tools for TIA and nonvalvular AF patients, respectively, were assessed in all cases. These variables were compared between the cardioembolic and noncardioembolic TIA groups. Statistical analysis was performed using the PASW Statistics software (version 18; SPSS Inc, Chicago, IL). Age differences between the groups were analyzed using the Wilcoxon test; differences in other variables were assessed using the chi-square test. As the CHADS2 score was found to be higher in the cardioembolic TIA group (shown in Results), the predictive value of this scale was quantified using the area under the receiver operating characteristics curve with a 95% confidence interval.
Results Among the 171 tissue-defined TIA patients, 38 were eventually diagnosed with cardioembolic TIA, and 116 were diagnosed with either large artery atherosclerosis or small vessel disease; these were defined as the noncardioembolic TIA group. Two or more possible causes (eg, AF and carotid artery stenosis) were identified in 7 cases, and the etiology could not be determined in 14 cases; both groups were not included in our study. The pathogenesis of cardioembolic TIA is presented in Table 1. Among 38 patients, 25 had nonvalvular AF (17 permanent and 8 paroxysmal). Five of the 25 AF patients had concurrent heart failure. Seven patients had a myocardial infarction with impaired wall contraction, 2 had overt mitral valve disease, 2 had a patent foramen ovale, 1 had dilated cardiomyopathy, and 1 had sick sinus syndrome. The clinical characteristics of patients with cardioembolic and noncardioembolic TIA are presented in Table 2. In total, the mean age of the patients was 67.4 6 11.7 years, and 55.2% were men. Thirty-three
CARDIOEMBOLIC AND NONCARDIOEMBOLIC TIA
2171
Table 1. Causes of cardioembolic TIA Nonvalvular AF Permanent AF Paroxysmal AF Without heart failure With heart failure Myocardial infarction Mitral valve disease Patent foramen ovale Cardiomyopathy Sick sinus syndrome
25 17 8 20 5 7 2 2 1 1
Abbreviations: AF, atrial fibrillation; TIA, transient ischemic attacks.
patients (21.4%) had a clinical history of symptomatic stroke, 105 (68.2%) had hypertension, 34 (22.1%) had diabetes mellitus, and 55 (35.7%) had dyslipidemia. A total of 102 patients (66.2%) had a history of hemiparesis and 69 patients (44.8%) experienced dysarthria during their attack. The duration of symptoms was less than 10 minutes in 36 patients (22.4%), 10-60 minutes in 67 patients (43.5%), and more than 60 minutes in 51 patients (33.1%). Radiologic investigation revealed an old cerebral infarction in 80 patients (52.0%). Fifteen patients (9.7%) had either coronary heart disease or peripheral artery disease. All variables were compared between the 2 groups. Except for vascular diseases such as coronary or peripheral artery diseases, there were no significant differences between the 2 groups (Table 2). Vascular diseases (other than stroke) were more frequent in the cardioembolic TIA group (P 5 .038), but this reflected cases with acute myocardial infarction, as this group included 7 cases of myocardial infarction-related TIA (Table 1). Thus, patient
demographics, clinical history, symptoms and their duration, comorbidities, and MRI findings do not contribute to differentiating cardioembolic TIA from those of noncardioembolic etiologies. The ABCD2 score, a well-known stroke risk stratification scale for TIA, was assessed in all patients. As shown in Table 3, there was a similar distribution for both groups, indicating that this score gives no information for differential diagnosis. The CHADS2 score, which is widely used for stroke risk stratification in nonvalvular AF patients, was also assessed in all patients and is shown in Table 4. Contrary to the ABCD2 score, there was a statistically significant difference in the CHADS2 score between the 2 groups (P 5 .005); a higher score indicated a higher probability of cardioembolic TIA. To quantify the predictive accuracy of the CHADS2 score in the diagnosis of cardioembolic TIAs, the receiver operating characteristics curve was analyzed (Fig 1). The area under the curve was .640 (95% confidence interval, .539-.742). The CHADS2 score was shown to provide useful information for the diagnosis of cardioembolic TIA, but its predictive power is limited.
Discussion Of all patients who experience an ischemic stroke, 7%40% had a prior TIA.16,17 Prompt initiation of appropriate therapy for TIA would decrease a permanently disabling stroke to a substantial degree. The optimal treatment for the acute phase of cerebral infarction may vary among patients. Similarly, in patients with TIA, the best means of treatment would differ depending on its etiology; for example, anticoagulation would be preferred over antiplatelet therapy for the treatment of cardioembolic
Table 2. Clinical characteristics of patients with cardioembolic and noncardioembolic TIA
Age, mean 6 SD Male sex, n (%) Risk factors, n (%) Hypertension Diabetes mellitus Dyslipidemia Clinical features, n (%) Hemiparesis Dysarthria Duration, n (%) ,10 min 10-60 min .60 min Vascular disease, n (%) Prior stroke, n (%) Old infarction by MRI, n (%)
Total
Cardioembolic TIA
Noncardioembolic TIA
P value
67.4 6 11.7 85 (55.2)
69.3 6 13.2 23 (60.5)
66.8 6 11.1 62 (53.4)
.258 .446
105 (68.2) 34 (22.1) 55 (35.7)
27 (71.1) 9 (23.7) 11 (28.9)
78 (67.2) 25 (21.6) 44 (37.9)
.818 .783 .316
102 (66.2) 69 (44.8)
26 (68.4) 13 (34.2)
76 (65.5) 56 (48.3)
.743 .130
36 (22.4) 67 (43.5) 51 (33.1) 15 (9.7) 33 (21.4) 80 (52.0)
7 (18.4) 21 (55.3) 10 (26.3) 7 (18.4) 8 (21.1) 18 (47.4)
29 (25.0) 46 (39.7) 41 (35.3) 8 (6.9) 25 (21.6) 62 (53.4)
.242 .038 .948 .515
Abbreviations: MRI, magnetic resonance imaging; SD, standard deviation; TIA, transient ischemic attacks.
T. HAYASHI ET AL.
2172
Table 3. ABCD2 score of cardioembolic and noncardioembolic TIA patients
ABCD2 0 1 2 3 4 5 6 7 Total
Total
Cardioembolic TIA
Noncardioembolic TIA
P value
0 2 9 25 56 42 17 3 154
0 0 2 7 14 10 3 2 38
0 2 7 18 42 32 14 1 116
.656
Abbreviation: TIA, transient ischemic attacks.
TIAs.18 In fact, the American College of Chest Physicians Evidence-Based Clinical Practice Guidelines recommended the early use of novel oral anticoagulants for cardioembolic TIA patients to prevent subsequent strokes.19 Our study showed that only 17 of 38 patients (44.7%) with cardioembolic TIA had permanent AF, indicating that more than 50% of cardioembolic TIA patients are at risk of being misdiagnosed with a noncardioembolic TIA if only a 12-lead ECG is used. It is therefore necessary to improve the diagnostic accuracy for patients who present with this condition. In cases of permanent stroke, disturbed consciousness was more frequent in patients with cardiogenic embolism than in those with large artery or small vessel diseases.20 The temporal profile of symptoms also differed; abrupt onset and immediate completion of symptoms were more common in cardioembolism than in large artery or small vessel diseases.20,21 In TIA, on the other hand, differences in symptoms and their temporal profiles were less extensively investigated. Kimura et al12 reported that aphasia was observed more frequently in Table 4. CHADS2 score of cardioembolic and noncardioembolic TIA patients
Total CHADS2 0 21 1 54 2 42 3 25 4 10 5 2 6 0 Total 154
Cardioembolic TIA
Noncardioembolic TIA
P value
3 8 15 5 5 2 0 38
18 46 27 20 5 0 0 116
.005
Abbreviation: TIA, transient ischemic attacks.
Figure 1. The receiver operating characteristics curve for the CHADS2 score in predicting the cardioembolic etiology in TIA. The areas under the curve with 95% confidence interval were .640 (.539-.742).
TIA patients with AF, perhaps because cardioembolic TIA tends to involve the cerebrocortical vasculature. In our study, however, we did not assess aphasia, because it was quite difficult in a retrospective study (reviewed by Kim).22 Instead, we assessed dysarthria and hemiparesis; neither showed differences between the 2 groups (Table 2). As our criteria for tissue-defined TIA might have excluded cardioembolic TIA resulting from a relatively large embolus, the differences in symptoms, which had possibly been confirmed for a time-defined TIA could have been negated. The duration of symptoms was similar between the 2 groups in our study (Table 2). Again, this may have been because of our definition of a TIA. A previous study with time-defined TIA demonstrated that emboligenic TIA showed longer duration of symptoms.11 As TIA of longer duration had a greater likelihood of a discernible lesion by MRI,23,24 the exclusion of patients with MRI-positive TIA could have negated the difference of symptom duration. Although no single parameter other than comorbidity of coronary or peripheral artery diseases showed a significantly different frequency, we found that the CHADS2 score differed between the 2 groups; patients with a higher score had a higher probability of cardioembolic etiology (Table 4). This score is used as a stroke risk stratification tool for nonvalvular AF, but will be useful in predicting the cardioembolic source in patients with TIA. Unfortunately, the diagnostic value is modest (Fig 1), but adding this information to laboratory data and 12-lead ECG findings may improve the diagnostic accuracy. This study has a limitation inevitable to a study of tissue-defined TIA. Patients with a contraindication for MRI were not included in this study; those with a pacemaker, implanted cardiac defibrillator, or prosthetic valve
CARDIOEMBOLIC AND NONCARDIOEMBOLIC TIA
were excluded. A previous study showed that prosthetic valves and pacemakers were the second and third most common causes of cardioembolic TIA.11 Therefore, the exclusion of these patients might have influenced our results. Nevertheless, as the definition of TIA changed from time-defined to tissue-defined one, the finding obtained in this study will provide information to help differentiate cardioembolic from noncardioembolic TIAs.
References 1. Wu CM, McLaughlin K, Lorenzetti DL, et al. Early risk of stroke after transient ischemic attack: a systematic review and meta-analysis. Arch Intern Med 2007;167:2417-2422. 2. Johnston SC, Gress DR, Browner WS, et al. Short-term prognosis after emergency department diagnosis of TIA. JAMA 2000;284:2901-2906. 3. Hill MD, Yiannakoulias N, Jeerakathil T, et al. The high risk of stroke immediately after transient ischemic attack: a population-based study. Neurology 2004;62:2015-2020. 4. Chandratheva A, Mehta Z, Geraghty OC, et al. Population-based study of risk and predictors of stroke in the first few hours after a TIA. Neurology 2009;72:1941-1947. 5. Lavallee PC, Meseguer E, Abboud H, et al. A transient ischaemic attack clinic with round-the-clock access (SOSTIA): feasibility and effects. Lancet Neurol 2007;6:953-960. 6. Luengo-Fernandez R, Gray AM, Rothwell PM. Effect of urgent treatment for transient ischaemic attack and minor stroke on disability and hospital costs (EXPRESS study): a prospective population-based sequential comparison. Lancet Neurol 2009;8:235-243. 7. National collaborating centre for chronic conditions (UK). Stroke: National clinical guideline for diagnosis and initial management of acute stroke and transient ischemic attack (TIA). National Institute for Health and Clinical Excellence: Guidance. London: Royal College of Physicians (UK) 2008. 8. Shinohara Y, Yanagihara T, Abe K, et al. Cerebral infarction/transient ischemic attack (TIA). J Stroke Cerebrovasc Dis 2011;20:S31-S73. 9. Bogousslavsky J, Hachinski VC, Boughner DR. Cardiac and arterial lesions in carotid transient ischemic attacks. Arch Neurol 1986;43:223-228. 10. Sempere AP, Duarte J, Cabezas C, et al. Etiopathogenesis of transient ischemic attacks and minor ischemic strokes: a community-based study in Segovia, Spain. Stroke 1998; 29:40-45. 11. Kimura K, Minematsu K, Yasaka M, et al. The duration of symptoms in transient ischemic attack. Neurology 1999; 52:976-980. 12. Kimura K, Minematsu K, Wada K, et al. Clinical characteristics in transient ischemic attack patients with atrial fibrillation. Intern Med 2003;42:255-258.
2173 13. Hayashi T, Kato Y, Nagoya H, et al. Prediction of ischemic stroke in patients with tissue-defined transient ischemic attack. J Stroke Cerebrovasc Dis 2014;23:1368-1373. 14. Madden KP, Karanjia PN, Adams HP Jr, et al. Accuracy of initial stroke subtype diagnosis in the TOAST study. Trial of ORG 10172 in Acute Stroke Treatment. Neurology 1995;45:1975-1979. 15. Deguchi I, Ohe Y, Fukuoka T, et al. Relationship of obesity to recanalization after hyperacute recombinant tissueplasminogen activator infusion therapy in patients with middle cerebral artery occlusion. J Stroke Cerebrovasc Dis 2012;21:161-164. 16. Easton JD, Saver JL, Albers GW, et al. Definition and evaluation of transient ischemic attack: a scientific statement for healthcare professionals from the American Heart Association/American Stroke Association Stroke Council, Council on Cardiovascular Surgery and Anesthesia, Council on Cardiovascular Radiology and Intervention, Council on Cardiovascular Nursing, and the Interdisciplinary Council on Peripheral Vascular Disease. The American Academy of Neurology affirms the value of this statement as an educational tool for neurologists. Stroke 2009;40:2276-2293. 17. Feinberg WM, Albers GW, Barnett HJ, et al. Guidelines for the management of transient ischemic attacks. From the Ad Hoc Committee on Guidelines for the Management of Transient Ischemic Attacks of the Stroke Council of the American Heart Association. Circulation 1994; 89:2950-2965. 18. Uehara T, Minematsu K. Guidelines for management of patients with transient ischemic attack. In: Uchiyama S, Amarenco P, Minematsu K, et al, eds. TIA as acute cerebrovascular syndrome. Tokyo: Karger 2014:103-114. 19. Lansberg MG, O’Donnell MJ, Khatri P, et al. Antithrombotic and thrombolytic therapy for ischemic stroke: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians EvidenceBased Clinical Practice Guidelines. Chest 2012;141(2 Suppl):e601S-e636S. 20. Bogousslavsky J, Van Melle G, Regli F. The Lausanne Stroke Registry: analysis of 1,000 consecutive patients with first stroke. Stroke 1988;19:1083-1092. 21. Foulkes MA, Wolf PA, Price TR, et al. The Stroke Data Bank: design, methods, and baseline characteristics. Stroke 1988;19:547-554. 22. Kim JS. Symptoms of transient ischemic attack. In: Uchiyama S, Amarenco P, Minematsu K, et al, eds. TIA as acute cerebrovascular syndrome. Tokyo: Karger 2014:82-102. 23. National Institute of Neurological Disorders and Stroke Ad Hoc Committee. Classification of cerebrovascular diseases III. Stroke 1990;21:637-676. 24. Shah SH, Saver JL, Kidwell CL, et al. A multicenter pooled, patient-level data analysis of diffusionweighted MRI in TIA patients. Stroke 2007;38:463.
Background: Previous studies show that 6%-31% of transient ischemic attacks (TIA) were caused by cardiogenic cerebral embolism (cardioembolic TIA). As prompt initiation of therapy is essential in TIA to prevent subsequent strokes, determining their cause is important. In this study, we aim to determine the features of cardioembolic TIA and to compare them with those of noncardioembolic etiology. Methods: We retrospectively reviewed patients with a tissue-defined TIA who were admitted to our hospital from April 2007 to August 2013. The etiology was categorized according to Trial of Org 10172 in Acute Stroke Treatment, and TIA of cardioembolic origin and cervicocerebrovascular etiology (noncardioembolic TIA) were included in this study. Those with 2 or more possible causes or undetermined etiologies were excluded. Age, sex, comorbidities, ABCD2 score, and CHADS2 score were assessed and compared between the 2 groups. Results: There were no significant differences in the neurologic symptoms and their duration, morbidities of hypertension, diabetes, and dyslipidemia between the 2 groups. Coronary and peripheral artery diseases were more common in the cardioembolic TIA group (18.4% vs. 6.9%). Incidences of prior stroke and cerebral infarction determined by MRI were similar between the 2 groups. The ABCD2 score showed a similar distribution, but the CHADS2 score was significantly different; the cardioembolic TIA group showed a higher score (P 5 .005). Conclusions: Clinical features are similar in tissue-defined TIA of cardioembolic and noncardioembolic etiologies. The CHADS2 score can be useful in assessing the probability of cardioembolic TIA. Key Words: Cardioembolism—clinical score—etiology—transient ischemic attack. Ó 2014 by National Stroke Association
From the Department of Neurology and Cerebrovascular Medicine, Saitama Medical University, International Medical Center, Hidaka, Japan. Received February 16, 2014; accepted April 2, 2014. Address correspondence to Takeshi Hayashi, MD, Department of Neurology and Cerebrovascular Medicine, Saitama Medical University, International Medical Center, 1397-1 Hidaka, Saitama 350-1298, Japan. E-mail: [email protected]. 1052-3057/$ - see front matter Ó 2014 by National Stroke Association http://dx.doi.org/10.1016/j.jstrokecerebrovasdis.2014.04.005
Ten to 20% of patients who experience a transient ischemic attack (TIA) will have a stroke within 3 months, and approximately 50% within 48 hours.1-3 Moreover, a recent study revealed that more than 2% of patients with TIA experienced a stroke within 12 hours.4 It is important to administer antithrombotic therapy immediately because prompt and appropriate therapy can help prevent more than 80% of subsequent strokes.5,6 However, clinicians often hesitate over which treatment to choose, as the etiology of TIA is sometimes
Journal of Stroke and Cerebrovascular Diseases, Vol. 23, No. 8 (September), 2014: pp 2169-2173
2169
T. HAYASHI ET AL.
2170
undiagnosed when stroke-preventive therapy is initiated. The National Institute for Clinical Excellence7 and Japanese8 guidelines recommend that all TIA patients be promptly administered 160-300 mg of aspirin per day, even when the etiology has not yet been determined, but the optimal method to prevent subsequent strokes may vary among patients. Approximately 5% of patients with TIA received anticoagulation therapy and another 5% underwent urgent carotid revascularization in a TIA clinic.5 As with cerebral infarction, the optimal medical treatment can differ based on TIA etiology. Approximately 6%-31% of TIAs are caused by a cardiogenic cerebral embolism (cardioembolic TIA).9,10 Determining TIA etiology is important before administering therapy, but it is difficult in some cases. In this study, we aimed to compare the features of a cardioembolic TIA with those of a noncardioembolic TIA. These kind of studies have been carried out previously, although a time-based definition of TIA was used.11,12 Indeed, the present study with tissue-defined TIA presented some unique findings.
Methods This study was approved by the Institutional Review Board at Saitama Medical University International Medical Center. All patients who were admitted to our hospital with transient focal neurologic deficits underwent magnetic resonance imaging (MRI), unless contraindicated. If the MRI revealed no fresh lesions on diffusion-weighed imaging, a second MRI was performed 3-7 days later. Those without high-intensity lesions on both diffusion-weighed images were diagnosed with tissue-defined TIA. Patients with lasting focal neurologic deficits on admission that resolved within 24 hours were also included, if repeat MRI after resolution of the neurologic deficits showed no new lesions.13 In total, 171 patents had tissue-defined TIA from April 2007 to August 2013. Etiology was categorized according to the Trial of Org 10172 in Acute Stroke Treatment.14 A cardiogenic source was investigated using 12-lead electrocardiography (ECG) and transthoracic echocardiography in all cases. In certain cases, we performed 24-hour Holter ECG, bedside ECG monitoring for several days, and transesophageal echocardiography when needed. Among the 171 tissue-defined TIA patients, 38 had cardioembolic TIA and 116 had either large artery atherosclerosis or small vessel disease (noncardioembolic TIA). Patients with a TIA of 2 or more possible causes (eg, atrial fibrillation [AF] and carotid artery stenosis) or undetermined etiology were not included in this study. We retrospectively reviewed the medical records of patients, including demographic information (age, sex), clinical history of symptomatic stroke, stroke risk factors (hypertension, diabetes mellitus, dyslipidemia), symp-
toms (presence of hemiparesis and/or dysarthria during the TIA), and duration of symptoms. Patients with a blood pressure of $140/90 mm Hg or those who were receiving antihypertensive medication were defined as having hypertension.15 Diabetes mellitus was defined as a blood glucose level of $200 mg/dL and a glycated hemoglobin level of $6.5% on admission or treatment with antidiabetic medication.15 Dyslipidemia was diagnosed if the patient had any of the following: lowdensity lipoprotein cholesterol level $140 mg/dL, high-density lipoprotein cholesterol level #40 mg/dL, triglyceride level $150 mg/dL, or treatment with lipidlowering medication.15 We investigated old cerebral infarction (including asymptomatic episodes) using MRI; high-intensity lesions on T2-weighed images and fluid-attenuated inversion recovery images were considered old infarctions.13 We also reviewed vascular diseases other than cerebrovascular disease, such as coronary heart or peripheral artery diseases. ABCD2 and CHADS2 scores, which are stroke risk stratification tools for TIA and nonvalvular AF patients, respectively, were assessed in all cases. These variables were compared between the cardioembolic and noncardioembolic TIA groups. Statistical analysis was performed using the PASW Statistics software (version 18; SPSS Inc, Chicago, IL). Age differences between the groups were analyzed using the Wilcoxon test; differences in other variables were assessed using the chi-square test. As the CHADS2 score was found to be higher in the cardioembolic TIA group (shown in Results), the predictive value of this scale was quantified using the area under the receiver operating characteristics curve with a 95% confidence interval.
Results Among the 171 tissue-defined TIA patients, 38 were eventually diagnosed with cardioembolic TIA, and 116 were diagnosed with either large artery atherosclerosis or small vessel disease; these were defined as the noncardioembolic TIA group. Two or more possible causes (eg, AF and carotid artery stenosis) were identified in 7 cases, and the etiology could not be determined in 14 cases; both groups were not included in our study. The pathogenesis of cardioembolic TIA is presented in Table 1. Among 38 patients, 25 had nonvalvular AF (17 permanent and 8 paroxysmal). Five of the 25 AF patients had concurrent heart failure. Seven patients had a myocardial infarction with impaired wall contraction, 2 had overt mitral valve disease, 2 had a patent foramen ovale, 1 had dilated cardiomyopathy, and 1 had sick sinus syndrome. The clinical characteristics of patients with cardioembolic and noncardioembolic TIA are presented in Table 2. In total, the mean age of the patients was 67.4 6 11.7 years, and 55.2% were men. Thirty-three
CARDIOEMBOLIC AND NONCARDIOEMBOLIC TIA
2171
Table 1. Causes of cardioembolic TIA Nonvalvular AF Permanent AF Paroxysmal AF Without heart failure With heart failure Myocardial infarction Mitral valve disease Patent foramen ovale Cardiomyopathy Sick sinus syndrome
25 17 8 20 5 7 2 2 1 1
Abbreviations: AF, atrial fibrillation; TIA, transient ischemic attacks.
patients (21.4%) had a clinical history of symptomatic stroke, 105 (68.2%) had hypertension, 34 (22.1%) had diabetes mellitus, and 55 (35.7%) had dyslipidemia. A total of 102 patients (66.2%) had a history of hemiparesis and 69 patients (44.8%) experienced dysarthria during their attack. The duration of symptoms was less than 10 minutes in 36 patients (22.4%), 10-60 minutes in 67 patients (43.5%), and more than 60 minutes in 51 patients (33.1%). Radiologic investigation revealed an old cerebral infarction in 80 patients (52.0%). Fifteen patients (9.7%) had either coronary heart disease or peripheral artery disease. All variables were compared between the 2 groups. Except for vascular diseases such as coronary or peripheral artery diseases, there were no significant differences between the 2 groups (Table 2). Vascular diseases (other than stroke) were more frequent in the cardioembolic TIA group (P 5 .038), but this reflected cases with acute myocardial infarction, as this group included 7 cases of myocardial infarction-related TIA (Table 1). Thus, patient
demographics, clinical history, symptoms and their duration, comorbidities, and MRI findings do not contribute to differentiating cardioembolic TIA from those of noncardioembolic etiologies. The ABCD2 score, a well-known stroke risk stratification scale for TIA, was assessed in all patients. As shown in Table 3, there was a similar distribution for both groups, indicating that this score gives no information for differential diagnosis. The CHADS2 score, which is widely used for stroke risk stratification in nonvalvular AF patients, was also assessed in all patients and is shown in Table 4. Contrary to the ABCD2 score, there was a statistically significant difference in the CHADS2 score between the 2 groups (P 5 .005); a higher score indicated a higher probability of cardioembolic TIA. To quantify the predictive accuracy of the CHADS2 score in the diagnosis of cardioembolic TIAs, the receiver operating characteristics curve was analyzed (Fig 1). The area under the curve was .640 (95% confidence interval, .539-.742). The CHADS2 score was shown to provide useful information for the diagnosis of cardioembolic TIA, but its predictive power is limited.
Discussion Of all patients who experience an ischemic stroke, 7%40% had a prior TIA.16,17 Prompt initiation of appropriate therapy for TIA would decrease a permanently disabling stroke to a substantial degree. The optimal treatment for the acute phase of cerebral infarction may vary among patients. Similarly, in patients with TIA, the best means of treatment would differ depending on its etiology; for example, anticoagulation would be preferred over antiplatelet therapy for the treatment of cardioembolic
Table 2. Clinical characteristics of patients with cardioembolic and noncardioembolic TIA
Age, mean 6 SD Male sex, n (%) Risk factors, n (%) Hypertension Diabetes mellitus Dyslipidemia Clinical features, n (%) Hemiparesis Dysarthria Duration, n (%) ,10 min 10-60 min .60 min Vascular disease, n (%) Prior stroke, n (%) Old infarction by MRI, n (%)
Total
Cardioembolic TIA
Noncardioembolic TIA
P value
67.4 6 11.7 85 (55.2)
69.3 6 13.2 23 (60.5)
66.8 6 11.1 62 (53.4)
.258 .446
105 (68.2) 34 (22.1) 55 (35.7)
27 (71.1) 9 (23.7) 11 (28.9)
78 (67.2) 25 (21.6) 44 (37.9)
.818 .783 .316
102 (66.2) 69 (44.8)
26 (68.4) 13 (34.2)
76 (65.5) 56 (48.3)
.743 .130
36 (22.4) 67 (43.5) 51 (33.1) 15 (9.7) 33 (21.4) 80 (52.0)
7 (18.4) 21 (55.3) 10 (26.3) 7 (18.4) 8 (21.1) 18 (47.4)
29 (25.0) 46 (39.7) 41 (35.3) 8 (6.9) 25 (21.6) 62 (53.4)
.242 .038 .948 .515
Abbreviations: MRI, magnetic resonance imaging; SD, standard deviation; TIA, transient ischemic attacks.
T. HAYASHI ET AL.
2172
Table 3. ABCD2 score of cardioembolic and noncardioembolic TIA patients
ABCD2 0 1 2 3 4 5 6 7 Total
Total
Cardioembolic TIA
Noncardioembolic TIA
P value
0 2 9 25 56 42 17 3 154
0 0 2 7 14 10 3 2 38
0 2 7 18 42 32 14 1 116
.656
Abbreviation: TIA, transient ischemic attacks.
TIAs.18 In fact, the American College of Chest Physicians Evidence-Based Clinical Practice Guidelines recommended the early use of novel oral anticoagulants for cardioembolic TIA patients to prevent subsequent strokes.19 Our study showed that only 17 of 38 patients (44.7%) with cardioembolic TIA had permanent AF, indicating that more than 50% of cardioembolic TIA patients are at risk of being misdiagnosed with a noncardioembolic TIA if only a 12-lead ECG is used. It is therefore necessary to improve the diagnostic accuracy for patients who present with this condition. In cases of permanent stroke, disturbed consciousness was more frequent in patients with cardiogenic embolism than in those with large artery or small vessel diseases.20 The temporal profile of symptoms also differed; abrupt onset and immediate completion of symptoms were more common in cardioembolism than in large artery or small vessel diseases.20,21 In TIA, on the other hand, differences in symptoms and their temporal profiles were less extensively investigated. Kimura et al12 reported that aphasia was observed more frequently in Table 4. CHADS2 score of cardioembolic and noncardioembolic TIA patients
Total CHADS2 0 21 1 54 2 42 3 25 4 10 5 2 6 0 Total 154
Cardioembolic TIA
Noncardioembolic TIA
P value
3 8 15 5 5 2 0 38
18 46 27 20 5 0 0 116
.005
Abbreviation: TIA, transient ischemic attacks.
Figure 1. The receiver operating characteristics curve for the CHADS2 score in predicting the cardioembolic etiology in TIA. The areas under the curve with 95% confidence interval were .640 (.539-.742).
TIA patients with AF, perhaps because cardioembolic TIA tends to involve the cerebrocortical vasculature. In our study, however, we did not assess aphasia, because it was quite difficult in a retrospective study (reviewed by Kim).22 Instead, we assessed dysarthria and hemiparesis; neither showed differences between the 2 groups (Table 2). As our criteria for tissue-defined TIA might have excluded cardioembolic TIA resulting from a relatively large embolus, the differences in symptoms, which had possibly been confirmed for a time-defined TIA could have been negated. The duration of symptoms was similar between the 2 groups in our study (Table 2). Again, this may have been because of our definition of a TIA. A previous study with time-defined TIA demonstrated that emboligenic TIA showed longer duration of symptoms.11 As TIA of longer duration had a greater likelihood of a discernible lesion by MRI,23,24 the exclusion of patients with MRI-positive TIA could have negated the difference of symptom duration. Although no single parameter other than comorbidity of coronary or peripheral artery diseases showed a significantly different frequency, we found that the CHADS2 score differed between the 2 groups; patients with a higher score had a higher probability of cardioembolic etiology (Table 4). This score is used as a stroke risk stratification tool for nonvalvular AF, but will be useful in predicting the cardioembolic source in patients with TIA. Unfortunately, the diagnostic value is modest (Fig 1), but adding this information to laboratory data and 12-lead ECG findings may improve the diagnostic accuracy. This study has a limitation inevitable to a study of tissue-defined TIA. Patients with a contraindication for MRI were not included in this study; those with a pacemaker, implanted cardiac defibrillator, or prosthetic valve
CARDIOEMBOLIC AND NONCARDIOEMBOLIC TIA
were excluded. A previous study showed that prosthetic valves and pacemakers were the second and third most common causes of cardioembolic TIA.11 Therefore, the exclusion of these patients might have influenced our results. Nevertheless, as the definition of TIA changed from time-defined to tissue-defined one, the finding obtained in this study will provide information to help differentiate cardioembolic from noncardioembolic TIAs.
References 1. Wu CM, McLaughlin K, Lorenzetti DL, et al. Early risk of stroke after transient ischemic attack: a systematic review and meta-analysis. Arch Intern Med 2007;167:2417-2422. 2. Johnston SC, Gress DR, Browner WS, et al. Short-term prognosis after emergency department diagnosis of TIA. JAMA 2000;284:2901-2906. 3. Hill MD, Yiannakoulias N, Jeerakathil T, et al. The high risk of stroke immediately after transient ischemic attack: a population-based study. Neurology 2004;62:2015-2020. 4. Chandratheva A, Mehta Z, Geraghty OC, et al. Population-based study of risk and predictors of stroke in the first few hours after a TIA. Neurology 2009;72:1941-1947. 5. Lavallee PC, Meseguer E, Abboud H, et al. A transient ischaemic attack clinic with round-the-clock access (SOSTIA): feasibility and effects. Lancet Neurol 2007;6:953-960. 6. Luengo-Fernandez R, Gray AM, Rothwell PM. Effect of urgent treatment for transient ischaemic attack and minor stroke on disability and hospital costs (EXPRESS study): a prospective population-based sequential comparison. Lancet Neurol 2009;8:235-243. 7. National collaborating centre for chronic conditions (UK). Stroke: National clinical guideline for diagnosis and initial management of acute stroke and transient ischemic attack (TIA). National Institute for Health and Clinical Excellence: Guidance. London: Royal College of Physicians (UK) 2008. 8. Shinohara Y, Yanagihara T, Abe K, et al. Cerebral infarction/transient ischemic attack (TIA). J Stroke Cerebrovasc Dis 2011;20:S31-S73. 9. Bogousslavsky J, Hachinski VC, Boughner DR. Cardiac and arterial lesions in carotid transient ischemic attacks. Arch Neurol 1986;43:223-228. 10. Sempere AP, Duarte J, Cabezas C, et al. Etiopathogenesis of transient ischemic attacks and minor ischemic strokes: a community-based study in Segovia, Spain. Stroke 1998; 29:40-45. 11. Kimura K, Minematsu K, Yasaka M, et al. The duration of symptoms in transient ischemic attack. Neurology 1999; 52:976-980. 12. Kimura K, Minematsu K, Wada K, et al. Clinical characteristics in transient ischemic attack patients with atrial fibrillation. Intern Med 2003;42:255-258.
2173 13. Hayashi T, Kato Y, Nagoya H, et al. Prediction of ischemic stroke in patients with tissue-defined transient ischemic attack. J Stroke Cerebrovasc Dis 2014;23:1368-1373. 14. Madden KP, Karanjia PN, Adams HP Jr, et al. Accuracy of initial stroke subtype diagnosis in the TOAST study. Trial of ORG 10172 in Acute Stroke Treatment. Neurology 1995;45:1975-1979. 15. Deguchi I, Ohe Y, Fukuoka T, et al. Relationship of obesity to recanalization after hyperacute recombinant tissueplasminogen activator infusion therapy in patients with middle cerebral artery occlusion. J Stroke Cerebrovasc Dis 2012;21:161-164. 16. Easton JD, Saver JL, Albers GW, et al. Definition and evaluation of transient ischemic attack: a scientific statement for healthcare professionals from the American Heart Association/American Stroke Association Stroke Council, Council on Cardiovascular Surgery and Anesthesia, Council on Cardiovascular Radiology and Intervention, Council on Cardiovascular Nursing, and the Interdisciplinary Council on Peripheral Vascular Disease. The American Academy of Neurology affirms the value of this statement as an educational tool for neurologists. Stroke 2009;40:2276-2293. 17. Feinberg WM, Albers GW, Barnett HJ, et al. Guidelines for the management of transient ischemic attacks. From the Ad Hoc Committee on Guidelines for the Management of Transient Ischemic Attacks of the Stroke Council of the American Heart Association. Circulation 1994; 89:2950-2965. 18. Uehara T, Minematsu K. Guidelines for management of patients with transient ischemic attack. In: Uchiyama S, Amarenco P, Minematsu K, et al, eds. TIA as acute cerebrovascular syndrome. Tokyo: Karger 2014:103-114. 19. Lansberg MG, O’Donnell MJ, Khatri P, et al. Antithrombotic and thrombolytic therapy for ischemic stroke: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians EvidenceBased Clinical Practice Guidelines. Chest 2012;141(2 Suppl):e601S-e636S. 20. Bogousslavsky J, Van Melle G, Regli F. The Lausanne Stroke Registry: analysis of 1,000 consecutive patients with first stroke. Stroke 1988;19:1083-1092. 21. Foulkes MA, Wolf PA, Price TR, et al. The Stroke Data Bank: design, methods, and baseline characteristics. Stroke 1988;19:547-554. 22. Kim JS. Symptoms of transient ischemic attack. In: Uchiyama S, Amarenco P, Minematsu K, et al, eds. TIA as acute cerebrovascular syndrome. Tokyo: Karger 2014:82-102. 23. National Institute of Neurological Disorders and Stroke Ad Hoc Committee. Classification of cerebrovascular diseases III. Stroke 1990;21:637-676. 24. Shah SH, Saver JL, Kidwell CL, et al. A multicenter pooled, patient-level data analysis of diffusionweighted MRI in TIA patients. Stroke 2007;38:463.
