Dement Geriatr Cogn Disord 2015;40:130–136 DOI: 10.1159/000379744 Accepted: February 2, 2015 Published online: June 12, 2015
© 2015 S. Karger AG, Basel 1420–8008/15/0404–0130$39.50/0 www.karger.com/dem
Original Research Article
Cerebral Microbleeds and Cognitive Function in Ischemic Stroke or Transient Ischemic Attack Patients Zhaolu Wang a Adrian Wong a Wenyan Liu a Jie Yang c Winnie C.W. Chu b Lisa Au a Alexander Lau a Anne Chan a Yunyun Xiong d Yannie Soo a Thomas Leung a Lawrence K.S. Wong a Vincent C.T. Mok a Departments of a Medicine and Therapeutics and b Radiology and Organ Imaging, The Chinese University of Hong Kong, Hong Kong, c Key Laboratory of Neurogenetics and Channelopathies of Guangdong Province and the Ministry of Education of China, Institute of Neuroscience and the Second Affiliated Hospital of Guangzhou Medical University, Guangzhou, and d Department of Neurology, Jinling Hospital, Nanjing University School of Medicine, Nanjing, China
Key Words Microbleeds · Cognitive impairment · Stroke Abstract Background: We explored the association between cerebral microbleeds (CMBs) and cognitive impairment in patients with ischemic stroke/transient ischemic attack (TIA). Methods: A total of 488 ischemic stroke/TIA patients received magnetic resonance imaging. Montreal Cognitive Assessment (MoCA) was used to evaluate global cognitive function and cognitive domains. The association of CMB quantity with cognitive function and the impact of CMB locations (strictly lobar, strictly deep, and mixed regions) on cognitive impairment were examined in regression models with adjustments for confounders. Results: A total of 113 subjects (23.2%) had ≥1 CMB. Strictly lobar, strictly deep, and mixed CMBs were identified in 36, 40, and 37 patients, respectively. The presence of ≥5 CMBs or strictly deep CMBs was associated with the MoCA total score (p = 0.007 and 0.020, respectively). Of all MoCA domains tested, a lower score in the attention domain was related to the presence of ≥5 CMBs (p = 0.014) and strictly deep CMBs (p = 0.028). Conclusion: CMBs were associated with cognitive dysfunction in stroke/TIA patients, especially in the attention domain. This association was mainly driven by CMBs in the deep region, underlining the role of hypertensive microangiopathy in stroke© 2015 S. Karger AG, Basel related cognitive impairment.
Vincent C.T. Mok Department of Medicine and Therapeutics, 9/F Clinical Sciences Building Prince of Wales Hospital, The Chinese University of Hong Kong Shatin, New Territories, Hong Kong, SAR (China) E-Mail vctmok @ cuhk.edu.hk
Downloaded by: University of Pittsburgh 198.143.32.65 - 1/27/2016 6:36:22 PM
Z.W. and A.W. are co-first authors, with equal contribution to the manuscript.
131
Dement Geriatr Cogn Disord 2015;40:130–136 DOI: 10.1159/000379744
© 2015 S. Karger AG, Basel www.karger.com/dem
Wang et al.: Cerebral Microbleeds and Cognitive Function in Ischemic Stroke or Transient Ischemic Attack Patients
Introduction
Cerebral microbleeds (CMBs) are considered a marker of cerebral small vessel disease, which can be visualized as small, rounded, homogenous and hypointense lesions on T2*weighted gradient-recalled echo (T2*-GRE) or susceptibility-weighted imaging (SWI)magnetic resonance imaging (MRI), corresponding to small perivascular hemosiderin deposits from leakage through cerebral small vessels [1, 2]. The locations of CMBs probably represent different underlying pathologies. Strictly lobar CMBs are presumed to represent pathologies associated with cerebral amyloid angiopathy (CAA), whereas non-strictly lobar CMBs represent hypertensive vasculopathy [1]. CMBs are relatively common in subjects with cognitive impairment or vascular dementia, and a few studies observed an association of CMBs with a higher incidence of vascular dementia [3, 4]. However, controversy still exists, and some prospective studies did not find the relationship between CMBs and cognitive decline [5, 6]. Besides, how CMBs locations affect cognitive function is also under debate. A British study found an association of strictly lobar CMBs with cognition [7], whereas 2 Japanese studies observed the association of strictly deep or mixed CMBs with global cognitive dysfunction or vascular dementia [4, 8]. Ischemic stroke patients have a high prevalence of CMBs (around 23% on T2*-GRE [7]) and also a high incidence of cognitive impairment [7]. However, few studies investigated the association of CMBs with post-stroke cognitive dysfunction [6, 7, 9]. By recruiting patients with ischemic stroke or transient ischemic attack (TIA), this study investigated the association of CMBs with general cognitive functions/cognitive domains after adjusting for confounders. Furthermore, we explored whether this association, if any, was driven by strictly lobar or strictly deep/infratentorial CMBs. Methods Subjects The subjects were participants of the ongoing STRIDE (STroke Registry Investigating cognitive DEcline) study [10], which has recruited 1,062 consecutive acute ischemic stroke/TIA patients admitted to the Prince of Wales Hospital in Hong Kong between 2009 and 2010, aiming at investigating mechanisms of cognitive decline over a 5-year period. Patients with a known history of dementia before stroke/TIA were excluded (patients with mild cognitive impairment were not excluded). The detailed inclusion and exclusion criteria have been described before [10]. In this article, we included 488 ischemic stroke/TIA patients who received SWI-MRI (n = 218) or GRE-MRI (n = 270) at baseline from the STRIDE study. This study was approved by the Joint Chinese University of Hong Kong – New Territories East Cluster Clinical Research Ethics Committee, and written informed consent was obtained.
Cognitive Assessment Cognitive functions were assessed using Montreal Cognitive Assessment (MoCA) 3–6 months after stroke/TIA. The MoCA subscores were divided into five cognitive domains: memory (ranging from 0 to 12), visuospatial (ranging from 0 to 4), language (ranging from 0 to 5), attention (ranging from 0 to 4) and executive function (ranging from 0 to 5) [12]. According to predefined criteria, impaired performance was classified as scores of ≤7 in memory, ≤3 in visuospatial, ≤4 in language, ≤3 in attention, or ≤4 in executive domain [12].
Downloaded by: University of Pittsburgh 198.143.32.65 - 1/27/2016 6:36:22 PM
Demographic and Vascular Risk Factors We collected data on basic demography (age, sex, and education in years) and vascular risk factors (diabetes mellitus, hyperlipidemia, hypertension, and stroke/TIA history) at 3–6 months after stroke/TIA, as we described before [10]. Ischemic stroke was classified into large-artery atherosclerosis, small-artery occlusion, cardioembolism (CE), stroke of other determined cause and stroke of undetermined cause according to the TOAST criteria [11].
132
Dement Geriatr Cogn Disord 2015;40:130–136 DOI: 10.1159/000379744
© 2015 S. Karger AG, Basel www.karger.com/dem
Wang et al.: Cerebral Microbleeds and Cognitive Function in Ischemic Stroke or Transient Ischemic Attack Patients
Neuroimaging Examinations Among the 488 patients, 270 received GRE-MRI scanning (1.5 T, Sonata; Siemens Medical, Erlangen, Germany), with a standard protocol including repetition time 350 ms, echo time 30 ms, field of view 230 mm, matrix 256 × 256, and slice thickness 5 mm with a 0.5-mm gap. SWI-MRI scanning (Achieva 3.0 T TX Series; Philips Medical System, Best, The Netherlands) was performed on 218 patients, with a standard protocol including repetition time 16.5 ms, echo time 23.4 ms, field of view 230 mm, matrix 256 × 256, and slice thickness 2.5 mm without a gap. Besides, all patients received T1-weighted imaging, T2-weighted imaging, fluid-attenuated inversion recovery, and diffusion-weighted imaging using standard protocols. CMBs were defined as focal areas of a very low signal intensity of maximum 10 mm in diameter. Signal voids caused by sulcal vessels, symmetric calcification in the basal ganglia, choroid plexus, pineal calcification, and signal averaging from bone were excluded [1]. The locations of CMBs were recorded as below when counting CMBs on images: lobar (cortical gray and subcortical or periventricular white matter), deep (deep gray matter: basal ganglia and thalamus; white matter of the corpus callosum, internal, external, and extreme capsule), and infratentorial (brain stem and cerebellum) [1]. Ventricular-brain ratio (VBR, a marker of global atrophy), medial temporal lobe atrophy (MTLA, yes or no), lacunes (≥1 vs. none), and the severity of white matter hyperintensities (WMH) were rated on MRI, as we described elsewhere [10]. The total volume of acute infarcts on diffusion-weighted imaging was calculated as the product of pixel number that acute infarcts involved and the pixel size of the template image in the standard space. All ratings were performed blinded to patients’ clinical data. Both intrarater agreements and interrater agreements between two raters were excellent for rating CMB in the total brain, in the lobe, and in the deep or infratentorial regions as well as VBR, MTLA, lacunes, and WMH on 30 randomly selected patients (the intraclass correlation coefficient ranged from 0.81 to 0.99). Statistics MoCA total scores were transformed by reflect and natural logarithm to reduce negative skewness. Linear regression was used to investigate the relationship between CMBs and MoCA total scores. Logistic regression was used to investigate the relationship between CMBs and each MoCA domain. Age, sex, education, diabetes, hyperlipidemia, hypertension, VBR, MTLA, lacunes, WMH, and volume of acute infarction were first screened in univariable regression models, and variables with a p value
© 2015 S. Karger AG, Basel 1420–8008/15/0404–0130$39.50/0 www.karger.com/dem
Original Research Article
Cerebral Microbleeds and Cognitive Function in Ischemic Stroke or Transient Ischemic Attack Patients Zhaolu Wang a Adrian Wong a Wenyan Liu a Jie Yang c Winnie C.W. Chu b Lisa Au a Alexander Lau a Anne Chan a Yunyun Xiong d Yannie Soo a Thomas Leung a Lawrence K.S. Wong a Vincent C.T. Mok a Departments of a Medicine and Therapeutics and b Radiology and Organ Imaging, The Chinese University of Hong Kong, Hong Kong, c Key Laboratory of Neurogenetics and Channelopathies of Guangdong Province and the Ministry of Education of China, Institute of Neuroscience and the Second Affiliated Hospital of Guangzhou Medical University, Guangzhou, and d Department of Neurology, Jinling Hospital, Nanjing University School of Medicine, Nanjing, China
Key Words Microbleeds · Cognitive impairment · Stroke Abstract Background: We explored the association between cerebral microbleeds (CMBs) and cognitive impairment in patients with ischemic stroke/transient ischemic attack (TIA). Methods: A total of 488 ischemic stroke/TIA patients received magnetic resonance imaging. Montreal Cognitive Assessment (MoCA) was used to evaluate global cognitive function and cognitive domains. The association of CMB quantity with cognitive function and the impact of CMB locations (strictly lobar, strictly deep, and mixed regions) on cognitive impairment were examined in regression models with adjustments for confounders. Results: A total of 113 subjects (23.2%) had ≥1 CMB. Strictly lobar, strictly deep, and mixed CMBs were identified in 36, 40, and 37 patients, respectively. The presence of ≥5 CMBs or strictly deep CMBs was associated with the MoCA total score (p = 0.007 and 0.020, respectively). Of all MoCA domains tested, a lower score in the attention domain was related to the presence of ≥5 CMBs (p = 0.014) and strictly deep CMBs (p = 0.028). Conclusion: CMBs were associated with cognitive dysfunction in stroke/TIA patients, especially in the attention domain. This association was mainly driven by CMBs in the deep region, underlining the role of hypertensive microangiopathy in stroke© 2015 S. Karger AG, Basel related cognitive impairment.
Vincent C.T. Mok Department of Medicine and Therapeutics, 9/F Clinical Sciences Building Prince of Wales Hospital, The Chinese University of Hong Kong Shatin, New Territories, Hong Kong, SAR (China) E-Mail vctmok @ cuhk.edu.hk
Downloaded by: University of Pittsburgh 198.143.32.65 - 1/27/2016 6:36:22 PM
Z.W. and A.W. are co-first authors, with equal contribution to the manuscript.
131
Dement Geriatr Cogn Disord 2015;40:130–136 DOI: 10.1159/000379744
© 2015 S. Karger AG, Basel www.karger.com/dem
Wang et al.: Cerebral Microbleeds and Cognitive Function in Ischemic Stroke or Transient Ischemic Attack Patients
Introduction
Cerebral microbleeds (CMBs) are considered a marker of cerebral small vessel disease, which can be visualized as small, rounded, homogenous and hypointense lesions on T2*weighted gradient-recalled echo (T2*-GRE) or susceptibility-weighted imaging (SWI)magnetic resonance imaging (MRI), corresponding to small perivascular hemosiderin deposits from leakage through cerebral small vessels [1, 2]. The locations of CMBs probably represent different underlying pathologies. Strictly lobar CMBs are presumed to represent pathologies associated with cerebral amyloid angiopathy (CAA), whereas non-strictly lobar CMBs represent hypertensive vasculopathy [1]. CMBs are relatively common in subjects with cognitive impairment or vascular dementia, and a few studies observed an association of CMBs with a higher incidence of vascular dementia [3, 4]. However, controversy still exists, and some prospective studies did not find the relationship between CMBs and cognitive decline [5, 6]. Besides, how CMBs locations affect cognitive function is also under debate. A British study found an association of strictly lobar CMBs with cognition [7], whereas 2 Japanese studies observed the association of strictly deep or mixed CMBs with global cognitive dysfunction or vascular dementia [4, 8]. Ischemic stroke patients have a high prevalence of CMBs (around 23% on T2*-GRE [7]) and also a high incidence of cognitive impairment [7]. However, few studies investigated the association of CMBs with post-stroke cognitive dysfunction [6, 7, 9]. By recruiting patients with ischemic stroke or transient ischemic attack (TIA), this study investigated the association of CMBs with general cognitive functions/cognitive domains after adjusting for confounders. Furthermore, we explored whether this association, if any, was driven by strictly lobar or strictly deep/infratentorial CMBs. Methods Subjects The subjects were participants of the ongoing STRIDE (STroke Registry Investigating cognitive DEcline) study [10], which has recruited 1,062 consecutive acute ischemic stroke/TIA patients admitted to the Prince of Wales Hospital in Hong Kong between 2009 and 2010, aiming at investigating mechanisms of cognitive decline over a 5-year period. Patients with a known history of dementia before stroke/TIA were excluded (patients with mild cognitive impairment were not excluded). The detailed inclusion and exclusion criteria have been described before [10]. In this article, we included 488 ischemic stroke/TIA patients who received SWI-MRI (n = 218) or GRE-MRI (n = 270) at baseline from the STRIDE study. This study was approved by the Joint Chinese University of Hong Kong – New Territories East Cluster Clinical Research Ethics Committee, and written informed consent was obtained.
Cognitive Assessment Cognitive functions were assessed using Montreal Cognitive Assessment (MoCA) 3–6 months after stroke/TIA. The MoCA subscores were divided into five cognitive domains: memory (ranging from 0 to 12), visuospatial (ranging from 0 to 4), language (ranging from 0 to 5), attention (ranging from 0 to 4) and executive function (ranging from 0 to 5) [12]. According to predefined criteria, impaired performance was classified as scores of ≤7 in memory, ≤3 in visuospatial, ≤4 in language, ≤3 in attention, or ≤4 in executive domain [12].
Downloaded by: University of Pittsburgh 198.143.32.65 - 1/27/2016 6:36:22 PM
Demographic and Vascular Risk Factors We collected data on basic demography (age, sex, and education in years) and vascular risk factors (diabetes mellitus, hyperlipidemia, hypertension, and stroke/TIA history) at 3–6 months after stroke/TIA, as we described before [10]. Ischemic stroke was classified into large-artery atherosclerosis, small-artery occlusion, cardioembolism (CE), stroke of other determined cause and stroke of undetermined cause according to the TOAST criteria [11].
132
Dement Geriatr Cogn Disord 2015;40:130–136 DOI: 10.1159/000379744
© 2015 S. Karger AG, Basel www.karger.com/dem
Wang et al.: Cerebral Microbleeds and Cognitive Function in Ischemic Stroke or Transient Ischemic Attack Patients
Neuroimaging Examinations Among the 488 patients, 270 received GRE-MRI scanning (1.5 T, Sonata; Siemens Medical, Erlangen, Germany), with a standard protocol including repetition time 350 ms, echo time 30 ms, field of view 230 mm, matrix 256 × 256, and slice thickness 5 mm with a 0.5-mm gap. SWI-MRI scanning (Achieva 3.0 T TX Series; Philips Medical System, Best, The Netherlands) was performed on 218 patients, with a standard protocol including repetition time 16.5 ms, echo time 23.4 ms, field of view 230 mm, matrix 256 × 256, and slice thickness 2.5 mm without a gap. Besides, all patients received T1-weighted imaging, T2-weighted imaging, fluid-attenuated inversion recovery, and diffusion-weighted imaging using standard protocols. CMBs were defined as focal areas of a very low signal intensity of maximum 10 mm in diameter. Signal voids caused by sulcal vessels, symmetric calcification in the basal ganglia, choroid plexus, pineal calcification, and signal averaging from bone were excluded [1]. The locations of CMBs were recorded as below when counting CMBs on images: lobar (cortical gray and subcortical or periventricular white matter), deep (deep gray matter: basal ganglia and thalamus; white matter of the corpus callosum, internal, external, and extreme capsule), and infratentorial (brain stem and cerebellum) [1]. Ventricular-brain ratio (VBR, a marker of global atrophy), medial temporal lobe atrophy (MTLA, yes or no), lacunes (≥1 vs. none), and the severity of white matter hyperintensities (WMH) were rated on MRI, as we described elsewhere [10]. The total volume of acute infarcts on diffusion-weighted imaging was calculated as the product of pixel number that acute infarcts involved and the pixel size of the template image in the standard space. All ratings were performed blinded to patients’ clinical data. Both intrarater agreements and interrater agreements between two raters were excellent for rating CMB in the total brain, in the lobe, and in the deep or infratentorial regions as well as VBR, MTLA, lacunes, and WMH on 30 randomly selected patients (the intraclass correlation coefficient ranged from 0.81 to 0.99). Statistics MoCA total scores were transformed by reflect and natural logarithm to reduce negative skewness. Linear regression was used to investigate the relationship between CMBs and MoCA total scores. Logistic regression was used to investigate the relationship between CMBs and each MoCA domain. Age, sex, education, diabetes, hyperlipidemia, hypertension, VBR, MTLA, lacunes, WMH, and volume of acute infarction were first screened in univariable regression models, and variables with a p value
