Research CHADS2 score/CHA2DS2-VASc score and major artery occlusion in cardioembolic stroke patients with nonvalvular atrial fibrillation Ichiro Deguchi*, Takeshi Hayashi, Yasuko Ohe, Yuji Kato, Takuya Fukuoka, Hajime Maruyama, Yohsuke Horiuchi, Hiroyasu Sano, Yuito Nagamine, and Norio Tanahashi Objective The associations between the CHADS2 score/ CHA2DS2-VASc score, and the presence of cerebral vessel occlusion on admission were examined in cardioembolic stroke patients with nonvalvular atrial fibrillation. Methods The subjects were 546 consecutive patients hospitalized between April 2007 and December 2012 with onset of cardioembolic stroke associated with nonvalvular atrial fibrillation within 24 h. The associations between the CHADS2 score/CHA2DS2-VASc score and the presence of occluded cerebral vessels on magnetic resonance angiography were evaluated retrospectively. Occluded cerebral vessels were classified into the internal carotid artery, middle cerebral artery (M1, M2), basilar artery, and other (anterior cerebral artery [A1], posterior cerebral artery [P1], vertebral artery). Results Major artery occlusion was seen in 52% of patients with CHADS2 score 0, 52% of patients with score 1, 57% with score 2, 75% with score 3, and 75% with score ≥4. As for the CHA2DS2-VASc score, major artery occlusion was seen in 62% of patients with score 0, 49% with score 1, 53% with score 2, 53% with score 3, 65% with score 4, 71% with score 5, and 82% with score ≥6. The incidence of concurrent major arterial occlusion increased as both scores rose. When classified by occluded blood vessel, the incidence of concurrent internal carotid artery occlusion increased as both the CHADS2 and CHA2DS2-VASc scores increased. Conclusion As the CHADS2 and CHA2DS2-VASs scores increased, the incidence of concurrent major arterial occlusion, particularly internal carotid artery occlusion, increased in patients with cardioembolic stroke associated with nonvalvular atrial fibrillation. Key words: CHADS2 score, CHA2DS2-VASc score, major artery occlusion
Introduction The CHADS2 score and the CHA2DS2-VASc score are widely used to evaluate the risk of cerebral embolism onset in patients with nonvalvular atrial fibrillation (NVAF) (1,2). Moreover, it has been reported that both the CHADS2 and CHA2DS2-VASc scores predict early and 90-day outcomes (3,4). We have also reported that the neurological deficit on admission was more severe and the 90-day clinical outcomes were worse as the CHADS2 and CHA2DS2-VASc scores increased in patients with initial cardioembolic stroke associated with NVAF (5,6). We thus hypothesized that large arteries tend to be occluded in cardioembolic stroke Correspondence: Ichiro Deguchi*, Department of Neurology, Saitama Medical University International Medical Center, 1397-1 Yamane, Hidaka, Saitama 350-1298, Japan. E-mail: [email protected] Received: 26 March 2013; Accepted: 20 October 2013; Published online 21 January 2014 Conflicts of interest: None declared. DOI: 10.1111/ijs.12227
576
Vol 9, July 2014, 576–579
patients with high CHADS2 or CHA2DS2-VASc scores, and this relationship was then examined retrospectively in the present study.
Subjects and methods In the period between April 2007 and December 2012, 813 cardioembolic stroke patients were admitted to the hospital within 24 h of onset. Of these patients, 546 consecutive patients with NVAFassociated cardioembolic stroke who underwent magnetic resonance imaging (MRI) and magnetic resonance angiography (MRA) at the time of their hospital visit were included in the study. The associations between the CHADS2 score/CHA2DS2-VASc score and the presence of cerebral vessel occlusion on MRI and MRA on admission were evaluated retrospectively. In all patients, cranial MRI scans were performed at the time of admission. Patients who had significant stenosis (50% or more) or intracranial vascular stenosis in the proximal portion of the lesion, which suggests the presence of atherothrombotic infarction, were excluded. The diagnosis of chronic atrial fibrillation (AF) or paroxysmal AF was made based on a 12-lead electrocardiogram, 24-h Holter monitoring, and patient monitoring or a history of paroxysmal AF. All patients underwent echocardiography to evaluate valvular disease. NVAF was defined as AF with no history of rheumatic mitral valve disease, a prosthetic heart valve, or mitral valve reconstruction (7). Occluded blood vessels were classified into the internal carotid artery (ICA), middle cerebral arteries M1 and M2, basilar artery (BA), and other (anterior cerebral artery A1, posterior cerebral artery P1, vertebral artery). CHADS2 scores were classified as 0, 1, 2, 3, and ≥4, and CHA2DS2-VASc scores were classified as 0, 1, 2, 3, 4, 5, and ≥6. Determination of the occluded vessel The MRI devices used were a 1·5-T Achieva Nova Dual (Philips, Best, The Netherlands) and a 1·5-T Magnetom Avanto (Siemens, Erlangen, Germany). MRA was performed from the branching point of the ICA to the inside of the cranium according to the following conditions: Achieva using the three-dimensional time of flight (3D-TOF) method (TR 20 ms; TE 6·0 ms; flip angle 16°; FOV 200 mm × 200 mm; slice thickness 0·65 mm); and Avanto using the 3D-TOF method (TR 22 ms; TE 7·0 ms; flip angle 18°; FOV 220 mm × 220 mm; slice thickness 0·7 mm). Vessel occlusion was defined as signal loss and was determined by two neurologists and one blinded neuroradiologist who did not know the patients’ backgrounds. This study was approved by the Ethics Committee at Saitama Medical University International Medical Center. © 2014 The Authors. International Journal of Stroke © 2014 World Stroke Organization
Research
I. Deguchi et al. Statistical analyses For statistical analysis, IBM SPSS Statistics 20 software was used (IBM Corporation, Armonk, NY, USA). The Wilcoxon rank-sum test and Fisher’s exact test (two sided) were used to compare the background characteristics between groups. The CHADS2/ CHA2DS2-VASc scores and occluded blood vessels were compared using the chi-square test. For each analysis, a significant difference was defined as P < 0·05.
Results The patients’ clinical characteristics are shown in Table 1. The CHADS2 score was 0 in 52 (9·5%), 1 in 146 (26·7%), 2 in 153 (28·0%), 3 in 114 (20·9%), and ≥4 in 81 (14·8%) patients. The Table 1 Clinical characteristics Age (years), mean ± SD Age 65–74 years, n (%) Age≥75 years, n (%) Sex (female), n (%) Hypertension, n (%) Diabetes mellitus, n (%) Congestive heart failure/LV dysfunction, n (%) Vascular disease, n (%) CHADS2 score, n (%) 0 1 2 3 ≥4 CHA2DS2-VASc score, n (%) 0 1 2 3 4 5 ≥6 Onset to MRI, mean ± SD, min Antithrombotic treatment at admission, n (%) Anticoagulant alone Anticoagulant plus an antiplatelet drug Antiplatelet drug None Occluded vessel None ICA, n (%) (MCA) M1, n (%) M2, n (%) BA, n (%) Other, n (%)
75·8 ± 10·1 168 (30·8) 313 (57·3) 222 (40·7) 375 (68·7) 119 (21·8) 126 (23·1) 55 (10·1) 52 (9·5) 146 (26·7) 153 (28·0) 114,(20·9) 81 (14·8) 13 (2·4) 49 (9·0) 107 (19·6) 122 (22·3) 102 (18·7) 86 (15·6) 67 (12·3) 260·8 ± 260·6 269 (49·3) 98 (17·9) 16 (2·9) 155 (28·4) 277 (50·7) 210, (38·5) 98 (17·9) 130 (23·8) 61 (11·2) 21 (3·8) 26 (4·8)
CHADS2, congestive heart failure, hypertension, age ≥ 75 years, diabetes mellitus, and stroke/transient ischemic attack; CHA(2)DS(2)VASc, congestive heart failure/left ventricular dysfunction, hypertension, age of 75 years or older or 65–74 years, diabetes mellitus, stroke/transient ischemic attack, vascular disease and female sex; LV, left ventricular; MRA, magnetic resonance angiography; ICA, internal carotid artery; MCA, middle cerebral artery; BA, basilar artery; other, anterior cerebral artery, posterior cerebral artery, vertebral artery. SD, standard deviation.
© 2014 The Authors. International Journal of Stroke © 2014 World Stroke Organization
CHA2DS2-VASc score was 0 in 13 (2·4%), 1 in 49 (9·0%), 2 in 107 (19·6%), 3 in 122 (22·3%), 4 in 102 (18·7%), 5 in 86 (15·6), and ≥6 in 67 (12·3). Major artery occlusion was observed in 336 patients (61·5%). In addition, chronic occlusion at a major artery other than the stroke-causing vessel was observed in 24 patients. The comparison of the background characteristics of the patients with initial and those with recurrent cerebral infarctions is shown in Table 2. The proportion of 65- to 74-year-old patients, usage of antithrombotic agents, and presence of major artery occlusion were significantly greater in the recurrent cerebral infarction group. Concerning the association between the CHADS2/CHA2DS2VASc scores and the presence of occluded blood vessels (Figs 1 and 2), there were 27 patients (52%) with occlusion and CHAD2 score 0, 76 (52%) with score 1, 87 (57%) with score 2, 85 (75%) with score 3, and 61 (75%) with score ≥4. As for the CHA2DS2VASc score, 8 patients (62%) had occlusion with score 0, 24 (49%) with score 1, 57 (53%) with score 2, 65 (53%) with score 3, 66 (65%) with score 4, 61 (71%) with score 5, and 55 (82%) with score ≥6. These results showed that the incidences of concurrent major arterial occlusion differed according to both scores (P < 0·001), with the incidences increasing as the scores became higher. In the comparison between groups with and without an occluded vessel (ICA vs. no occlusion, M1 vs. no occlusion, M2 vs. no occlusion, BA vs. no occlusion, other vs. no occlusion), the CHADS2 scores for ICA and for M1 (ICA; P < 0·001, M1; P = 0·013) and the CHA2DS2-VASc score for ICA (P < 0·001) each had different rates of complications depending on the score. There was no significant difference in onset-MRI time between the absence and presence of an occluded vessel (P = 0·289).
Discussion The results of this study demonstrated that the incidence of concurrent major artery occlusion, especially occlusion of the ICA, increased as the CHADS2 score and CHA2DS2-VASc score increased. To the best of our knowledge, this is the first report on the relationships between the CHADS2 score/CHA2DS2-VASc score and the presence of occluded blood vessels. It is known that the factors comprising the CHADS2 score and the CHA2DS2-VASc score not only promote arteriosclerosis, but also cause endothelial dysfunction (8–11). The mechanism of thrombus formation in patients with AF involves enhanced coagulation and atrial endothelial dysfunction, as well as dynamic circulatory blood flow stasis in the left atrium (blood flow change) (12). Thus, as the CHADS2 score and the CHA2DS2-VASc score increase, endothelial function in the atrium would be more strongly impaired, promoting not only thrombus formation but also thrombus enlargement. Furthermore, when thrombus formation occurs in a major artery, thrombi can more easily be dissolved via the function of endogenous tissue plasminogen activator when the severity of endothelial dysfunction is mild. Cardioembolic stroke leads to higher in-hospital mortality than other subtypes of cerebral infarction, and AF in particular is associated with higher in-hospital mortality, not only in the overall ischemic stroke population, but also in patients with the Vol 9, July 2014, 576–579
577
Research
I. Deguchi et al.
Table 2 Patients’ background characteristics by initial and recurrent cerebral infarction.
Age (years), mean ± SD Age 65–74 years, n (%) Age ≥ 75 years, n (%) Sex (Female), n (%) Hypertension, n (%) Diabetes mellitus, n (%) Congestive heart failure/ LV dysfunction, n (%) Vascular disease, n (%) Onset to MRI, mean ± SD, min Antithrombotic treatment at admission, n (%) Anticoagulant alone Anticoagulant plus an antiplatelet drug Antiplatelet drug None Occluded vessel Absent Present
Recurrent cardioembolic stroke
Initial cardioembolic stroke
n = 114
n = 432
P value
75·9 ± 8·7 45 (39·5) 61 (53·5) 50 (43·9) 84 (73·7) 31 (27·2)
75·7 ± 10·3 122 (28·2) 252 (58·3) 172 (39·8) 291 (67·4) 88 (20·4)
0·822* 0·023† 0·395† 0·454† 0·231† 0·126†
34 (29·8) 15 (13·2) 295·6 ± 310 96 (84·2) 47 (41·2) 7 (6·1) 42 (36·8) 18 (15·8)
94 (21·8) 40 (9·3) 251·1 ± 245·4 173 (40·0) 51 (11·8) 9 (2·1) 113 (26·2) 259 (60·0)
0·082† 0·223† 0·416*
Introduction The CHADS2 score and the CHA2DS2-VASc score are widely used to evaluate the risk of cerebral embolism onset in patients with nonvalvular atrial fibrillation (NVAF) (1,2). Moreover, it has been reported that both the CHADS2 and CHA2DS2-VASc scores predict early and 90-day outcomes (3,4). We have also reported that the neurological deficit on admission was more severe and the 90-day clinical outcomes were worse as the CHADS2 and CHA2DS2-VASc scores increased in patients with initial cardioembolic stroke associated with NVAF (5,6). We thus hypothesized that large arteries tend to be occluded in cardioembolic stroke Correspondence: Ichiro Deguchi*, Department of Neurology, Saitama Medical University International Medical Center, 1397-1 Yamane, Hidaka, Saitama 350-1298, Japan. E-mail: [email protected] Received: 26 March 2013; Accepted: 20 October 2013; Published online 21 January 2014 Conflicts of interest: None declared. DOI: 10.1111/ijs.12227
576
Vol 9, July 2014, 576–579
patients with high CHADS2 or CHA2DS2-VASc scores, and this relationship was then examined retrospectively in the present study.
Subjects and methods In the period between April 2007 and December 2012, 813 cardioembolic stroke patients were admitted to the hospital within 24 h of onset. Of these patients, 546 consecutive patients with NVAFassociated cardioembolic stroke who underwent magnetic resonance imaging (MRI) and magnetic resonance angiography (MRA) at the time of their hospital visit were included in the study. The associations between the CHADS2 score/CHA2DS2-VASc score and the presence of cerebral vessel occlusion on MRI and MRA on admission were evaluated retrospectively. In all patients, cranial MRI scans were performed at the time of admission. Patients who had significant stenosis (50% or more) or intracranial vascular stenosis in the proximal portion of the lesion, which suggests the presence of atherothrombotic infarction, were excluded. The diagnosis of chronic atrial fibrillation (AF) or paroxysmal AF was made based on a 12-lead electrocardiogram, 24-h Holter monitoring, and patient monitoring or a history of paroxysmal AF. All patients underwent echocardiography to evaluate valvular disease. NVAF was defined as AF with no history of rheumatic mitral valve disease, a prosthetic heart valve, or mitral valve reconstruction (7). Occluded blood vessels were classified into the internal carotid artery (ICA), middle cerebral arteries M1 and M2, basilar artery (BA), and other (anterior cerebral artery A1, posterior cerebral artery P1, vertebral artery). CHADS2 scores were classified as 0, 1, 2, 3, and ≥4, and CHA2DS2-VASc scores were classified as 0, 1, 2, 3, 4, 5, and ≥6. Determination of the occluded vessel The MRI devices used were a 1·5-T Achieva Nova Dual (Philips, Best, The Netherlands) and a 1·5-T Magnetom Avanto (Siemens, Erlangen, Germany). MRA was performed from the branching point of the ICA to the inside of the cranium according to the following conditions: Achieva using the three-dimensional time of flight (3D-TOF) method (TR 20 ms; TE 6·0 ms; flip angle 16°; FOV 200 mm × 200 mm; slice thickness 0·65 mm); and Avanto using the 3D-TOF method (TR 22 ms; TE 7·0 ms; flip angle 18°; FOV 220 mm × 220 mm; slice thickness 0·7 mm). Vessel occlusion was defined as signal loss and was determined by two neurologists and one blinded neuroradiologist who did not know the patients’ backgrounds. This study was approved by the Ethics Committee at Saitama Medical University International Medical Center. © 2014 The Authors. International Journal of Stroke © 2014 World Stroke Organization
Research
I. Deguchi et al. Statistical analyses For statistical analysis, IBM SPSS Statistics 20 software was used (IBM Corporation, Armonk, NY, USA). The Wilcoxon rank-sum test and Fisher’s exact test (two sided) were used to compare the background characteristics between groups. The CHADS2/ CHA2DS2-VASc scores and occluded blood vessels were compared using the chi-square test. For each analysis, a significant difference was defined as P < 0·05.
Results The patients’ clinical characteristics are shown in Table 1. The CHADS2 score was 0 in 52 (9·5%), 1 in 146 (26·7%), 2 in 153 (28·0%), 3 in 114 (20·9%), and ≥4 in 81 (14·8%) patients. The Table 1 Clinical characteristics Age (years), mean ± SD Age 65–74 years, n (%) Age≥75 years, n (%) Sex (female), n (%) Hypertension, n (%) Diabetes mellitus, n (%) Congestive heart failure/LV dysfunction, n (%) Vascular disease, n (%) CHADS2 score, n (%) 0 1 2 3 ≥4 CHA2DS2-VASc score, n (%) 0 1 2 3 4 5 ≥6 Onset to MRI, mean ± SD, min Antithrombotic treatment at admission, n (%) Anticoagulant alone Anticoagulant plus an antiplatelet drug Antiplatelet drug None Occluded vessel None ICA, n (%) (MCA) M1, n (%) M2, n (%) BA, n (%) Other, n (%)
75·8 ± 10·1 168 (30·8) 313 (57·3) 222 (40·7) 375 (68·7) 119 (21·8) 126 (23·1) 55 (10·1) 52 (9·5) 146 (26·7) 153 (28·0) 114,(20·9) 81 (14·8) 13 (2·4) 49 (9·0) 107 (19·6) 122 (22·3) 102 (18·7) 86 (15·6) 67 (12·3) 260·8 ± 260·6 269 (49·3) 98 (17·9) 16 (2·9) 155 (28·4) 277 (50·7) 210, (38·5) 98 (17·9) 130 (23·8) 61 (11·2) 21 (3·8) 26 (4·8)
CHADS2, congestive heart failure, hypertension, age ≥ 75 years, diabetes mellitus, and stroke/transient ischemic attack; CHA(2)DS(2)VASc, congestive heart failure/left ventricular dysfunction, hypertension, age of 75 years or older or 65–74 years, diabetes mellitus, stroke/transient ischemic attack, vascular disease and female sex; LV, left ventricular; MRA, magnetic resonance angiography; ICA, internal carotid artery; MCA, middle cerebral artery; BA, basilar artery; other, anterior cerebral artery, posterior cerebral artery, vertebral artery. SD, standard deviation.
© 2014 The Authors. International Journal of Stroke © 2014 World Stroke Organization
CHA2DS2-VASc score was 0 in 13 (2·4%), 1 in 49 (9·0%), 2 in 107 (19·6%), 3 in 122 (22·3%), 4 in 102 (18·7%), 5 in 86 (15·6), and ≥6 in 67 (12·3). Major artery occlusion was observed in 336 patients (61·5%). In addition, chronic occlusion at a major artery other than the stroke-causing vessel was observed in 24 patients. The comparison of the background characteristics of the patients with initial and those with recurrent cerebral infarctions is shown in Table 2. The proportion of 65- to 74-year-old patients, usage of antithrombotic agents, and presence of major artery occlusion were significantly greater in the recurrent cerebral infarction group. Concerning the association between the CHADS2/CHA2DS2VASc scores and the presence of occluded blood vessels (Figs 1 and 2), there were 27 patients (52%) with occlusion and CHAD2 score 0, 76 (52%) with score 1, 87 (57%) with score 2, 85 (75%) with score 3, and 61 (75%) with score ≥4. As for the CHA2DS2VASc score, 8 patients (62%) had occlusion with score 0, 24 (49%) with score 1, 57 (53%) with score 2, 65 (53%) with score 3, 66 (65%) with score 4, 61 (71%) with score 5, and 55 (82%) with score ≥6. These results showed that the incidences of concurrent major arterial occlusion differed according to both scores (P < 0·001), with the incidences increasing as the scores became higher. In the comparison between groups with and without an occluded vessel (ICA vs. no occlusion, M1 vs. no occlusion, M2 vs. no occlusion, BA vs. no occlusion, other vs. no occlusion), the CHADS2 scores for ICA and for M1 (ICA; P < 0·001, M1; P = 0·013) and the CHA2DS2-VASc score for ICA (P < 0·001) each had different rates of complications depending on the score. There was no significant difference in onset-MRI time between the absence and presence of an occluded vessel (P = 0·289).
Discussion The results of this study demonstrated that the incidence of concurrent major artery occlusion, especially occlusion of the ICA, increased as the CHADS2 score and CHA2DS2-VASc score increased. To the best of our knowledge, this is the first report on the relationships between the CHADS2 score/CHA2DS2-VASc score and the presence of occluded blood vessels. It is known that the factors comprising the CHADS2 score and the CHA2DS2-VASc score not only promote arteriosclerosis, but also cause endothelial dysfunction (8–11). The mechanism of thrombus formation in patients with AF involves enhanced coagulation and atrial endothelial dysfunction, as well as dynamic circulatory blood flow stasis in the left atrium (blood flow change) (12). Thus, as the CHADS2 score and the CHA2DS2-VASc score increase, endothelial function in the atrium would be more strongly impaired, promoting not only thrombus formation but also thrombus enlargement. Furthermore, when thrombus formation occurs in a major artery, thrombi can more easily be dissolved via the function of endogenous tissue plasminogen activator when the severity of endothelial dysfunction is mild. Cardioembolic stroke leads to higher in-hospital mortality than other subtypes of cerebral infarction, and AF in particular is associated with higher in-hospital mortality, not only in the overall ischemic stroke population, but also in patients with the Vol 9, July 2014, 576–579
577
Research
I. Deguchi et al.
Table 2 Patients’ background characteristics by initial and recurrent cerebral infarction.
Age (years), mean ± SD Age 65–74 years, n (%) Age ≥ 75 years, n (%) Sex (Female), n (%) Hypertension, n (%) Diabetes mellitus, n (%) Congestive heart failure/ LV dysfunction, n (%) Vascular disease, n (%) Onset to MRI, mean ± SD, min Antithrombotic treatment at admission, n (%) Anticoagulant alone Anticoagulant plus an antiplatelet drug Antiplatelet drug None Occluded vessel Absent Present
Recurrent cardioembolic stroke
Initial cardioembolic stroke
n = 114
n = 432
P value
75·9 ± 8·7 45 (39·5) 61 (53·5) 50 (43·9) 84 (73·7) 31 (27·2)
75·7 ± 10·3 122 (28·2) 252 (58·3) 172 (39·8) 291 (67·4) 88 (20·4)
0·822* 0·023† 0·395† 0·454† 0·231† 0·126†
34 (29·8) 15 (13·2) 295·6 ± 310 96 (84·2) 47 (41·2) 7 (6·1) 42 (36·8) 18 (15·8)
94 (21·8) 40 (9·3) 251·1 ± 245·4 173 (40·0) 51 (11·8) 9 (2·1) 113 (26·2) 259 (60·0)
0·082† 0·223† 0·416*
