ORIGINAL ARTICLE
White matter microstructural changes in pure Alzheimer’s disease and subcortical vascular dementia
a Department of Neurology, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul; bNeuroscience Center, Samsung Medical Center, Seoul; cDepartment of Biomedical Engineering, Hanyang University, Seoul; dDepartment of Neurology, Ilsong Institute of Life Science, Hallym University, Anyang; eRadiology, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul; and fNuclear Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea
Keywords:
Alzheimer’s disease, diffuse tensor imaging, subcortical vascular dementia, white matter microstructural changes Received 11 September 2014 Accepted 12 November 2014 European Journal of Neurology 2015, 22: 709–716 doi:10.1111/ene.12645
Background and purpose: Recent studies have demonstrated that Alzheimer’s disease (AD) and subcortical vascular dementia (SVaD) have white matter (WM) microstructural changes. However, previous studies on AD and SVaD rarely eliminated the confounding effects of patients with mixed Alzheimer’s and cerebrovascular disease pathologies. Therefore, our aim was to evaluate the divergent topography of WM microstructural changes in patients with pure AD and SVaD. Methods: Patients who were clinically diagnosed with AD and SVaD were prospectively recruited. Forty AD patients who were Pittsburgh compound B (PiB) positive [PiB(+) AD] without WM hyperintensities and 32 SVaD patients who were PiB negative [PiB( ) SVaD] were chosen. Fifty-six cognitively normal individuals were also recruited (NC). Tract-based spatial statistics of diffuse tensor imaging were used to compare patterns of fractional anisotropy (FA) and mean diffusivity (MD). Results: Compared with the NC group, the PiB(+) AD group showed decreased FA in the bilateral frontal, temporal and parietal WM regions and the genu and splenium of the corpus callosum as well as increased MD in the left frontal and temporal WM region. PiB( ) SVaD patients showed decreased FA and increased MD in all WM regions. Direct comparison between PiB(+) AD and PiB( ) SVaD groups showed that the PiB( ) SVaD group had decreased FA across all WM regions and increased MD in all WM regions except occipital regions. Conclusion: Our findings suggest that pure AD and pure SVaD have divergent topography of WM microstructural changes including normal appearing WM.
Introduction Alzheimer’s disease (AD) and subcortical vascular dementia (SVaD) are the two most common causes of dementia. AD is characterized by accumulation of amyloid plaques and neurofibrillary tangles in the Correspondence: Sang Won Seo, MD, PhD, Department of Neurology, Sungkyunkwan University School of Medicine, Samsung Medical Center, 50 Ilwon-dong, Gangnam-gu, Seoul 135-710, Korea (tel.: +82 2 3410 1233; fax: +82 2 3410 0052; e-mail: [email protected]).
© 2015 EAN
cortex, which in turn lead to cortical atrophy [1]. SVaD is characterized by ischaemic changes in the white matter (WM) or deep nuclei caused by small vessel disease [2]. Clinically, AD patients have more impaired episodic memory function whilst SVaD patients have more impaired frontal/executive function [3]. However, pathological studies have suggested that patients clinically diagnosed with SVaD have co-associated AD pathology and vice versa [4,5]. With the advent of diffusion tensor imaging (DTI), which is a sensitive tool for detecting microstructural
709
EUROPEAN JOURNAL OF NEUROLOGY
Y. J. Kima,b, H. K. Kwonc, J.-M. Leec, Y. J. Kimd, H. J. Kima,b, N.-Y. Junga,b, S. T. Kime, K. H. Leef, D. L. Naa,b and S. W. Seoa,b
710
Y. J. KIM ET AL.
changes in WM [6], recent studies have demonstrated that AD might have WM microstructural changes [7]. Previous studies also suggested that SVaD had extensive WM microstructural abnormalities, especially in normal-appearing WM [8]. However, because previous studies on AD and SVaD rarely eliminated the confounding effects of patients with mixed Alzheimer’s and cerebrovascular disease pathologies, they did not evaluate the divergent topography of WM microstructural changes in patients with pure AD and SVaD. Pittsburgh compound B (PiB) positron emission topography (PET) can detect amyloid pathologies [9]; therefore, it can help discriminate between patients with mixed amyloid and vascular pathologies and those with pure AD and SVaD. Previous amyloid imaging studies suggested that 90% and 30% of clinically diagnosed AD and SVaD patients, respectively, had significant amyloid burdens [10,11]. Nearly 40% of AD patients also have a moderate degree of white matter hyperintensities (WMH) [12]. Therefore, PiB(+) AD without WMH and PiB( ) SVaD refer to pure AD and pure SVaD, respectively. In this study, the divergent topography of WM microstructural changes in patients with PiB(+) AD without WMH and PiB( ) SVaD was investigated. Considering the previous neuropsychological and neuroimaging findings [13,14], it was hypothesized that PiB (+) AD without WMH would have WM microstructural changes predominantly in the posterior regions whilst PiB( ) SVaD would have WM microstructural changes predominantly in the anterior regions.
Methods Participants
In all, 140 patients who were clinically diagnosed with SVaD (n = 70) or AD (n = 70) and who underwent 11C-PiB-PET scan and magnetic resonance imaging (MRI) at Samsung Medical Center between September 2008 and August 2011 were prospectively recruited. All patients with SVaD fulfilled the Diagnostic and Statistical Manual of Mental Disorders Fourth Edition criteria for VaD and had severe WMH on MRI. Severe WMH on MRI was defined as a 10-mm cap or band as well as a 25-mm deep WM lesion. Patients with WMH on MRI scan because of radiation injury, multiple sclerosis, vasculitis or leukodystrophy, using their clinical history and other information such as blood test results if necessary, were excluded. The diagnosis of AD was made on the basis of criteria for probable AD proposed by the National Institute of Neurological and Communicative Disorders and Stroke/Alzheimer’s Disease and Related Disorders Association [15]. The
final patient sample consisted of 32/70 (45.7%) patients with SVaD who tested negative for PiB retention [PiB ( ) SVaD] and 40/70 (57.1%) patients with AD who tested positive for PiB retention [PiB(+) AD]. Fifty-six normal cognition (NC) subjects from the neurology clinic at Samsung Medical Center who had no history of neurological or psychiatric illnesses and no abnormalities detected during neurological examination were also recruited. They were determined to be cognitively normal after undergoing the same neuropsychological testing and MRI scanning. Written informed consent was obtained from each participant, and the institutional review board of the Samsung Medical Center approved the study protocol. Magnetic resonance imaging techniques
Standardized T2, three-dimensional T1 turbo field echo, three-dimensional fluid-attenuated inversion recovery (FLAIR) and DTI images were acquired from all subjects at Samsung Medical Center using the same 3.0 T MRI scanner (Achieva; Philips Medical Systems, Eindhoven, the Netherlands). Detailed imaging parameters are described in Data S1. Diffusion tensor imaging processing
Diffusion tensor imaging data were processed using FMRIB’s Software Library (FSL v5.0.2.1) (http:// www.fmrib.ox.ac.uk/fsl). Motion artifacts and eddy current distortions were corrected by normalizing each diffusion-weighted volume to the non-diffusionweighted volume (b0) using the affine registration method in FMRIB’s Linear Image Registration Tool (FLIRT v6.0) [16]. The diffusion-weighted images were skull stripped with the Brain Extraction Tool from the FSL tool [17]. The DTIFIT tool (part of FSL) calculated the diffusion tensor from the diffusion-weighted images using non-linear estimation of the diffusion tensor model [18]. Then, the fractional anisotropy (FA) [19] and the mean diffusivity (MD) ((k1 + k2 + k3)/2) were extracted for each voxel based on diagonal elements (k1, k2 and k3) of the diffusion tensor. These maps (FA, MD) were used for tractbased spatial statistics (TBSS) analysis. Diffusion tensor matrices from the sets of diffusion-weighted images were generated using a general linear fitting algorithm. Subsequently, FA and MD were calculated for each voxel according to standard methods. Tract-based spatial statistics analysis
The FA and MD maps of DTI preprocessing results were used in TBSS (v1.2) [20]. All FA images were
© 2015 EAN
WM CHANGES IN PURE AD AND SVAD
aligned onto a standard FMRIB58 FA template provided by FSL software, using a non-linear registration algorithm implemented in the TBSS package. The FA images, which were aligned on the FMRIB58 FA template, were averaged to create a skeletonized mean FA image. Each subject’s aligned FA image projects onto the skeleton by filling the skeleton with highest FA values from the nearest relevant center of fiber tracts. A threshold FA value of 0.2 was chosen to exclude voxels of adjacent gray matter or cerebrospinal fluid. For MD analysis, MD images were also processed by applying the FA non-linear registration and projecting them onto the skeleton using identical projection methods to those inferred from the original FA data. Then, voxel-wise statistics were calculated across subjects on the skeleton-space FA and MD images. Positron emission tomography acquisition
[11C] Pittsburgh compound B positron emission topography scanning was performed at Samsung or Asan Medical Center using a Discovery STe PET/CT scanner (GE Medical Systems, Milwaukee, WI, USA) in a three-dimensional scanning mode that examined 35 slices of 4.25-mm thickness spanning the entire brain. Detailed methods are described in Data S2. Pittsburgh compound B positron emission topography data analysis
Pittsburgh compound B positron emission topography images were co-registered to individual MRIs, which were normalized to a T1-weighted MRI template. The quantitative regional values of PiB retention on the spatially normalized PiB images were obtained using automated voxel of interest analysis using the automated anatomical labeling atlas. Data processing was performed using SPM version 5 (SPM5; Wellcome Trust Centre for Neuroimaging, University College London, London, UK) under Matlab 6.5 (Mathworks, Natick, MA, USA). Detailed methods for the calculation of global PiB retention ratio are described in Data S3. Statistical analysis
Demographic and patient characteristics data were analyzed using one-way ANOVA for continuous variables and chi-squared tests for dichotomous variables. Two-sided P < 0.05 was considered statistically significant. To test for localized differences across groups, voxel-wise statistical analysis of individual skeleton
© 2015 EAN
711
images was performed based on a non-parametric permutation test [21] using ‘randomise v2.9’ (part of FSL). Age, gender and level of education were included as covariates in the analysis of covariance (ANCOVA), and the null distribution was built up over 5000 permutations. For control over multiple comparison correction, threshold-free cluster enhancement was used with the ‘2D’ parameter settings [22]. The results for FA and MD were considered significant for family-wise error (FWE) corrected P < 0.01.
Results Patient characteristics
The patient characteristics are described in Table 1. There were no differences in clinical dementia rating sum-of-boxes between PiB(+) AD and PiB( ) SVaD patients. Comparisons of metrics between NC, PiB(+) AD and PiB( ) SVaD groups
Tract-based spatial statistics revealed decreased FA in the PiB(+) AD group compared to the NC group in the bilateral frontal, temporal and parietal WM regions and the genu and splenium of the corpus callosum (Fig. 1a). Compared to NC, the PiB( ) SVaD group exhibited decreased FA in all WM regions (Fig. 1b). Direct comparisons between PiB(+) AD and PiB( ) SVaD groups showed that PiB( ) SVaD groups had decreased FA in all WM regions whilst there were no regions where PiB(+) AD had decreased FA (Fig. 1c). Tract-based spatial statistics revealed increased MD in the PiB(+) AD group compared to the NC group in the left frontal and temporal WM regions (Fig. 2a). Compared to NC, the PiB( ) SVaD group revealed
Table 1 Demographics of subjects Characteristics a
Age, years Female, nb Education, yearsa MMSEa CDR-SOBa
NC (56)
PiB(+) AD (40)
PiB( ) SVaD (32)
62.5 7.5 40 (71.4) 12.5 5.0
67.0 9.1* 26 (65.0)* 10.4 5.3
71.3 7.3* 15 (46.9)* 8.6 4.6*
28.8 1.4 0.58 0.45
17.4 6.1* 4.85 2.29*
21.6 4.6*† 5.97 3.84*
NC, normal controls; PiB, Pittsburgh compound B; AD, Alzheimer’s disease; SVaD, subcortical vascular dementia; MMSE, Mini-Mental State Examination; CDR-SOB, clinical dementia rating sum-ofboxes. aValues are means SD; bnumber of cases with percentages in parentheses; *P < 0.05 between NC and PiB(+) AD or PiB( ) SVaD; †P < 0.05 between PiB(+) AD and PiB( ) SVaD.
712
Y. J. KIM ET AL.
(a)
(b)
(c)
Figure 1 Tract-based spatial statistics results of FA images. Green represents MNI152 standard brain and the skeleton image. Red yellow represents decreased FA in PiB(+) AD patients compared with NC subjects (a), PiB( ) SVaD patients compared with NC subjects (b) and PiB( ) SVaD patients compared with PiB(+) AD patients (c). P < 0.01, FWE corrected for multiple comparisons.
increased MD in all WM regions except occipital WM regions (Fig. 2b). Direct comparisons between the PiB (+) AD and PiB( ) SVaD groups showed that PiB( ) SVaD groups had increased MD in all WM regions except occipital WM regions, whilst there were no regions where PiB(+) AD had increased MD (Fig. 2c). To further illustrate the magnitude of these effects, Fig. 3 presents DTI metrics in the representative regions from superior frontal, medial temporal and lateral parietal WM regions for the NC, PiB(+)
AD and PiB( ) SVaD groups. These results were consistent with the statistical maps of differences in DTI metrics between the groups.
Discussion A novel finding of WM microstructural changes in PiB(+) AD without WMH and PiB( ) SVaD is reported. Although previous DTI studies have shown that AD and SVaD patients had WM microstructural
© 2015 EAN
WM CHANGES IN PURE AD AND SVAD
713
(a)
(b)
(c)
Figure 2 Tract-based spatial statistics results of MD images. Green represents MNI152 standard brain and the skeleton image. Blue light blue represents increased MD in PiB(+) AD patients compared with NC subjects (a), PiB( ) SVaD patients compared with NC subjects (b) and PiB( ) SVaD patients compared with PiB(+) AD patients (c). P < 0.01, FWE corrected for multiple comparisons.
changes, they did not exclude patients with combined AD and SVaD pathologies. Therefore, our study is the first showing that pure AD and pure SVaD have divergent topography of WM microstructural changes including normal appearing WM. In the present study, PiB(+) AD without WMH showed altered microstructural changes not only in the temporo-parietal WM regions but also in the frontal WM regions. Specifically, our finding that PiB(+) AD without WMH had increased MD in the medial
© 2015 EAN
temporal region and decreased FA in the splenium of the corpus callosum is consistent with previous studies [23]. AD pathology characteristically involved the cortical regions adjacent to these WM regions, which in turn led to the involvement of WM through Wallerian degeneration [23]. Interestingly, frontal WM involvement in these patients was also observed. Although previous studies showed microstructural changes in the frontal WM region [24], these might be due to combined cerebrovascular disease markers. Frontal
714
Y. J. KIM ET AL.
(a)
(b)
(c)
(d)
(e)
(f)
Figure 3 Diffusion tensor imaging metrics in the representative regions for NC, PiB(+) AD and PiB( ) SVaD groups. Mean FA values from superior frontal (a), medial temporal (b) and lateral parietal white matter regions (c). Mean MD values from superior frontal (d), medial temporal (e) and lateral parietal white matter regions (f).
WM involvement may be related to myelin breakdown as well as Wallerian degeneration [25]. That is, toxic amyloid burdens distributed predominantly in the frontal regions may affect myelin sheath or oligodendrocytes, leading to WM microstructural changes in the frontal region. Indeed, our suggestion may be supported by our finding that PiB(+) AD without WMH had decreased FA predominantly in the frontal-parietal WM regions. According to the retrogenesis hypothesis, AD pathology might preferentially affect late-myelinating WM tracts that connect frontal and parietal lobes [25]. Pittsburgh compound B negative SVaD patients showed microstructural changes in all brain WM regions compared to NC. Although clinically diagnosed SVaD patients were reported to show decreased FA in the anterior and posterior brain WM regions [8], it was hypothesized that posterior WM involvement might be driven by combined AD pathology. Because PiB(+) SVaD patients were excluded in this study, it is the first to show that pure SVaD patients have decreased FA in the anterior and posterior brain WM regions. The pathobiology of WM involvement in pure SVaD patients remains unclear. Previous pathological studies suggested that WMH correlated with the severity of rarefaction of myelin, axonal loss or gliosis
[26]. The pathological correlates are also reported to affect microstructural metrics of DTI [6]. In fact, WMH volumes were negatively associated with FA values [27]. Moreover, decreased FA values were observed in the normal appearing WM of cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy [28]. Finally, these microstructural alterations might propagate to distant WM regions through intra-hemispheric cortico-cortical disconnection caused by Wallerian degeneration, dying back or myelin breakdown [23]. Indeed, our findings of posterior involvements in pure SVaD patients might be supported by previous studies from our group [9,12,29]. These findings suggest that PiB ( ) SVaD patients had cortical atrophy in temporoparietal regions as well as the frontal region and corresponding cognitive impairments including memory and visuospatial dysfunctions. There are some limitations to this study. First, because postmortem studies were not performed, other pathologies could not be measured including other AD (soluble amyloid and neurofibrillary tangles), cardiovascular disease (microinfarct) or possible combined degenerative dementia (DLB and FTD) pathologies, which are also associated with microstructural changes in the WM. Secondly, PiB PET scanning was not performed in the NC group.
© 2015 EAN
WM CHANGES IN PURE AD AND SVAD
Previous studies showed that approximately 10% 40% of NC subjects could be PiB(+) [30]. Although there has been some controversy on the relationship between PiB retention and WM microstructural changes in cognitively normal subjects, there could have been some underestimation of the degree of WM microstructural changes in subjects with PiB(+) AD and PiB( ) SVaD. Finally, PiB PET may not be sufficiently sensitive to detect soluble amyloid oligomers or a very low level of compact plaques. However, our findings suggested that pure AD and SVaD patients showed divergent topography of WM microstructural change and provided a better understanding of their cognitive symptoms.
4. 5.
6.
7.
8.
Acknowledgements This study was supported by the Basic Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (NRF-2013R1A1A2065365), by the Korean Healthcare Technology R&D Project Ministry for Health & Welfare Affairs (HI10C2020 and HIC120713), by the Korea Ministry of Environment (MOE) as the Environmental Health Action Program (2014001360002), by the KOSEF NRL program grant (MEST; 20110028333), by Samsung Medical Center (CRL108011&CRS110-14-1) and by the Converging Research Center Program through the Ministry of Science, ICT and Future Planning, Korea (2013K000338).
9.
10.
11.
12.
13.
14.
Disclosure of conflicts of interest The authors declare no financial or other conflicts of interest.
15.
Supporting Information Additional Supporting Information may be found in the online version of this article:
16.
Data S1. Imaging parameters for MRI acquisition. Data S2. PET acquisition. Data S3. Calculation of the global PiB retention ratio.
17.
18.
References
19.
1. Walsh DM, Selkoe DJ. A beta oligomers a decade of discovery. J Neurochem 2007; 101: 1172–1184. 2. Roman GC, Erkinjuntti T, Wallin A, Pantoni L, Chui HC. Subcortical ischaemic vascular dementia. Lancet Neurol 2002; 1: 426–436. 3. Graham NL, Emery T, Hodges JR. Distinctive cognitive profiles in Alzheimer’s disease and subcortical vascular
© 2015 EAN
20.
21.
715
dementia. J Neurol Neurosurg Psychiatry 2004; 75: 61–71. Chui HC. Subcortical ischemic vascular dementia. Neurol Clin 2007; 25: 717–740, vi. Park JH, Seo SW, Kim C, et al. Effects of cerebrovascular disease and amyloid beta burden on cognition in subjects with subcortical vascular cognitive impairment. Neurobiol Aging 2014; 35: 254–260. Nitkunan A, McIntyre DJ, Barrick TR, et al. Correlations between MRS and DTI in cerebral small vessel disease. NMR Biomed 2006; 19: 610–616. Barber R, Scheltens P, Gholkar A, et al. White matter lesions on magnetic resonance imaging in dementia with Lewy bodies, Alzheimer’s disease, vascular dementia, and normal aging. J Neurol Neurosurg Psychiatry 1999; 67: 66–72. Kim SH, Park JS, Ahn HJ, et al. Voxel-based analysis of diffusion tensor imaging in patients with subcortical vascular cognitive impairment: correlates with cognitive and motor deficits. J Neuroimaging 2011; 21: 317–324. Klunk WE, Engler H, Nordberg A, et al. Imaging brain amyloid in Alzheimer’s disease with Pittsburgh compound-B. Ann Neurol 2004; 55: 306–319. Cho H, Seo SW, Kim JH, et al. Amyloid deposition in early onset versus late onset Alzheimer’s disease. J Alzheimers Dis 2013; 35: 813–821. Lee JH, Kim SH, Kim GH, et al. Identification of pure subcortical vascular dementia using 11C-Pittsburgh compound B. Neurology 2011; 77: 18–25. Park HK, Na DL, Han SH, et al. Clinical characteristics of a nationwide hospital-based registry of mild-to-moderate Alzheimer’s disease patients in Korea: a CREDOS (Clinical Research Center for Dementia of South Korea) study. J Korean Med Sci 2011; 26: 1219–1226. Kim CH, Seo SW, Kim GH, et al. Cortical thinning in subcortical vascular dementia with negative 11C-PiB PET. J Alzheimers Dis 2012; 31: 315–323. Yoon CW, Shin JS, Kim HJ, et al. Cognitive deficits of pure subcortical vascular dementia vs. Alzheimer disease: PiB-PET-based study. Neurology 2013; 80: 569– 573. McKhann G, Drachman D, Folstein M, Katzman R, Price D, Stadlan EM. Clinical diagnosis of Alzheimer’s disease: report of the NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer’s Disease. Neurology 1984; 34: 939–944. Jenkinson M, Smith S. A global optimisation method for robust affine registration of brain images. Med Image Anal 2001; 5: 143–156. Smith SM, Zhang Y, Jenkinson M, et al. Accurate, robust, and automated longitudinal and cross-sectional brain change analysis. NeuroImage 2002; 17: 479–489. Pierpaoli C, Basser PJ. Toward a quantitative assessment of diffusion anisotropy. Magn Reson Med 1996; 36: 893–906. Basser PJ, Mattiello J, LeBihan D. MR diffusion tensor spectroscopy and imaging. Biophys J 1994; 66: 259–267. Smith SM, Jenkinson M, Johansen-Berg H, et al. Tractbased spatial statistics: voxelwise analysis of multi-subject diffusion data. NeuroImage 2006; 31: 1487–1505. Nichols TE, Holmes AP. Nonparametric permutation tests for functional neuroimaging: a primer with examples. Hum Brain Mapp 2002; 15: 1–25.
716
Y. J. KIM ET AL.
22. Smith S, Nichols T. Threshold-free cluster-enhancement addressing the problem of threshold dependence in cluster inference. In: Proc. Human Brain Mapping Conference, Chicago 2007: S92. 23. Coleman M. Axon degeneration mechanisms: commonality amid diversity. Nat Rev Neurosci 2005; 6: 889–898. 24. Sexton CE, Kalu UG, Filippini N, Mackay CE, Ebmeier KP. A meta-analysis of diffusion tensor imaging in mild cognitive impairment and Alzheimer’s disease. Neurobiol Aging 2011; 32: 2322. e5–e18. 25. Benitez A, Fieremans E, Jensen JH, et al. White matter tract integrity metrics reflect the vulnerability of latemyelinating tracts in Alzheimer’s disease. Neuroimage Clin 2013; 4: 64–71. 26. Fazekas F, Kleinert R, Offenbacher H, et al. Pathologic correlates of incidental MRI white matter signal hyperintensities. Neurology 1993; 43: 1683–1689.
27. Zhan W, Zhang Y, Mueller SG, et al. Characterization of white matter degeneration in elderly subjects by magnetic resonance diffusion and FLAIR imaging correlation. NeuroImage 2009; 47(Suppl. 2): T58–T65. 28. Chabriat H, Pappata S, Poupon C, et al. Clinical severity in CADASIL related to ultrastructural damage in white matter: in vivo study with diffusion tensor MRI. Stroke 1999; 30: 2637–2643. 29. Kim HJ, Ye BS, Yoon CW, et al. Cortical thickness and hippocampal shape in pure vascular mild cognitive impairment and dementia of subcortical type. Eur J Neurol 2014; 21: 744–751. 30. Pike KE, Savage G, Villemagne VL, et al. Beta-amyloid imaging and memory in non-demented individuals: evidence for preclinical Alzheimer’s disease. Brain 2007; 130: 2837–2844.
© 2015 EAN
Copyright of European Journal of Neurology is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
White matter microstructural changes in pure Alzheimer’s disease and subcortical vascular dementia
a Department of Neurology, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul; bNeuroscience Center, Samsung Medical Center, Seoul; cDepartment of Biomedical Engineering, Hanyang University, Seoul; dDepartment of Neurology, Ilsong Institute of Life Science, Hallym University, Anyang; eRadiology, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul; and fNuclear Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea
Keywords:
Alzheimer’s disease, diffuse tensor imaging, subcortical vascular dementia, white matter microstructural changes Received 11 September 2014 Accepted 12 November 2014 European Journal of Neurology 2015, 22: 709–716 doi:10.1111/ene.12645
Background and purpose: Recent studies have demonstrated that Alzheimer’s disease (AD) and subcortical vascular dementia (SVaD) have white matter (WM) microstructural changes. However, previous studies on AD and SVaD rarely eliminated the confounding effects of patients with mixed Alzheimer’s and cerebrovascular disease pathologies. Therefore, our aim was to evaluate the divergent topography of WM microstructural changes in patients with pure AD and SVaD. Methods: Patients who were clinically diagnosed with AD and SVaD were prospectively recruited. Forty AD patients who were Pittsburgh compound B (PiB) positive [PiB(+) AD] without WM hyperintensities and 32 SVaD patients who were PiB negative [PiB( ) SVaD] were chosen. Fifty-six cognitively normal individuals were also recruited (NC). Tract-based spatial statistics of diffuse tensor imaging were used to compare patterns of fractional anisotropy (FA) and mean diffusivity (MD). Results: Compared with the NC group, the PiB(+) AD group showed decreased FA in the bilateral frontal, temporal and parietal WM regions and the genu and splenium of the corpus callosum as well as increased MD in the left frontal and temporal WM region. PiB( ) SVaD patients showed decreased FA and increased MD in all WM regions. Direct comparison between PiB(+) AD and PiB( ) SVaD groups showed that the PiB( ) SVaD group had decreased FA across all WM regions and increased MD in all WM regions except occipital regions. Conclusion: Our findings suggest that pure AD and pure SVaD have divergent topography of WM microstructural changes including normal appearing WM.
Introduction Alzheimer’s disease (AD) and subcortical vascular dementia (SVaD) are the two most common causes of dementia. AD is characterized by accumulation of amyloid plaques and neurofibrillary tangles in the Correspondence: Sang Won Seo, MD, PhD, Department of Neurology, Sungkyunkwan University School of Medicine, Samsung Medical Center, 50 Ilwon-dong, Gangnam-gu, Seoul 135-710, Korea (tel.: +82 2 3410 1233; fax: +82 2 3410 0052; e-mail: [email protected]).
© 2015 EAN
cortex, which in turn lead to cortical atrophy [1]. SVaD is characterized by ischaemic changes in the white matter (WM) or deep nuclei caused by small vessel disease [2]. Clinically, AD patients have more impaired episodic memory function whilst SVaD patients have more impaired frontal/executive function [3]. However, pathological studies have suggested that patients clinically diagnosed with SVaD have co-associated AD pathology and vice versa [4,5]. With the advent of diffusion tensor imaging (DTI), which is a sensitive tool for detecting microstructural
709
EUROPEAN JOURNAL OF NEUROLOGY
Y. J. Kima,b, H. K. Kwonc, J.-M. Leec, Y. J. Kimd, H. J. Kima,b, N.-Y. Junga,b, S. T. Kime, K. H. Leef, D. L. Naa,b and S. W. Seoa,b
710
Y. J. KIM ET AL.
changes in WM [6], recent studies have demonstrated that AD might have WM microstructural changes [7]. Previous studies also suggested that SVaD had extensive WM microstructural abnormalities, especially in normal-appearing WM [8]. However, because previous studies on AD and SVaD rarely eliminated the confounding effects of patients with mixed Alzheimer’s and cerebrovascular disease pathologies, they did not evaluate the divergent topography of WM microstructural changes in patients with pure AD and SVaD. Pittsburgh compound B (PiB) positron emission topography (PET) can detect amyloid pathologies [9]; therefore, it can help discriminate between patients with mixed amyloid and vascular pathologies and those with pure AD and SVaD. Previous amyloid imaging studies suggested that 90% and 30% of clinically diagnosed AD and SVaD patients, respectively, had significant amyloid burdens [10,11]. Nearly 40% of AD patients also have a moderate degree of white matter hyperintensities (WMH) [12]. Therefore, PiB(+) AD without WMH and PiB( ) SVaD refer to pure AD and pure SVaD, respectively. In this study, the divergent topography of WM microstructural changes in patients with PiB(+) AD without WMH and PiB( ) SVaD was investigated. Considering the previous neuropsychological and neuroimaging findings [13,14], it was hypothesized that PiB (+) AD without WMH would have WM microstructural changes predominantly in the posterior regions whilst PiB( ) SVaD would have WM microstructural changes predominantly in the anterior regions.
Methods Participants
In all, 140 patients who were clinically diagnosed with SVaD (n = 70) or AD (n = 70) and who underwent 11C-PiB-PET scan and magnetic resonance imaging (MRI) at Samsung Medical Center between September 2008 and August 2011 were prospectively recruited. All patients with SVaD fulfilled the Diagnostic and Statistical Manual of Mental Disorders Fourth Edition criteria for VaD and had severe WMH on MRI. Severe WMH on MRI was defined as a 10-mm cap or band as well as a 25-mm deep WM lesion. Patients with WMH on MRI scan because of radiation injury, multiple sclerosis, vasculitis or leukodystrophy, using their clinical history and other information such as blood test results if necessary, were excluded. The diagnosis of AD was made on the basis of criteria for probable AD proposed by the National Institute of Neurological and Communicative Disorders and Stroke/Alzheimer’s Disease and Related Disorders Association [15]. The
final patient sample consisted of 32/70 (45.7%) patients with SVaD who tested negative for PiB retention [PiB ( ) SVaD] and 40/70 (57.1%) patients with AD who tested positive for PiB retention [PiB(+) AD]. Fifty-six normal cognition (NC) subjects from the neurology clinic at Samsung Medical Center who had no history of neurological or psychiatric illnesses and no abnormalities detected during neurological examination were also recruited. They were determined to be cognitively normal after undergoing the same neuropsychological testing and MRI scanning. Written informed consent was obtained from each participant, and the institutional review board of the Samsung Medical Center approved the study protocol. Magnetic resonance imaging techniques
Standardized T2, three-dimensional T1 turbo field echo, three-dimensional fluid-attenuated inversion recovery (FLAIR) and DTI images were acquired from all subjects at Samsung Medical Center using the same 3.0 T MRI scanner (Achieva; Philips Medical Systems, Eindhoven, the Netherlands). Detailed imaging parameters are described in Data S1. Diffusion tensor imaging processing
Diffusion tensor imaging data were processed using FMRIB’s Software Library (FSL v5.0.2.1) (http:// www.fmrib.ox.ac.uk/fsl). Motion artifacts and eddy current distortions were corrected by normalizing each diffusion-weighted volume to the non-diffusionweighted volume (b0) using the affine registration method in FMRIB’s Linear Image Registration Tool (FLIRT v6.0) [16]. The diffusion-weighted images were skull stripped with the Brain Extraction Tool from the FSL tool [17]. The DTIFIT tool (part of FSL) calculated the diffusion tensor from the diffusion-weighted images using non-linear estimation of the diffusion tensor model [18]. Then, the fractional anisotropy (FA) [19] and the mean diffusivity (MD) ((k1 + k2 + k3)/2) were extracted for each voxel based on diagonal elements (k1, k2 and k3) of the diffusion tensor. These maps (FA, MD) were used for tractbased spatial statistics (TBSS) analysis. Diffusion tensor matrices from the sets of diffusion-weighted images were generated using a general linear fitting algorithm. Subsequently, FA and MD were calculated for each voxel according to standard methods. Tract-based spatial statistics analysis
The FA and MD maps of DTI preprocessing results were used in TBSS (v1.2) [20]. All FA images were
© 2015 EAN
WM CHANGES IN PURE AD AND SVAD
aligned onto a standard FMRIB58 FA template provided by FSL software, using a non-linear registration algorithm implemented in the TBSS package. The FA images, which were aligned on the FMRIB58 FA template, were averaged to create a skeletonized mean FA image. Each subject’s aligned FA image projects onto the skeleton by filling the skeleton with highest FA values from the nearest relevant center of fiber tracts. A threshold FA value of 0.2 was chosen to exclude voxels of adjacent gray matter or cerebrospinal fluid. For MD analysis, MD images were also processed by applying the FA non-linear registration and projecting them onto the skeleton using identical projection methods to those inferred from the original FA data. Then, voxel-wise statistics were calculated across subjects on the skeleton-space FA and MD images. Positron emission tomography acquisition
[11C] Pittsburgh compound B positron emission topography scanning was performed at Samsung or Asan Medical Center using a Discovery STe PET/CT scanner (GE Medical Systems, Milwaukee, WI, USA) in a three-dimensional scanning mode that examined 35 slices of 4.25-mm thickness spanning the entire brain. Detailed methods are described in Data S2. Pittsburgh compound B positron emission topography data analysis
Pittsburgh compound B positron emission topography images were co-registered to individual MRIs, which were normalized to a T1-weighted MRI template. The quantitative regional values of PiB retention on the spatially normalized PiB images were obtained using automated voxel of interest analysis using the automated anatomical labeling atlas. Data processing was performed using SPM version 5 (SPM5; Wellcome Trust Centre for Neuroimaging, University College London, London, UK) under Matlab 6.5 (Mathworks, Natick, MA, USA). Detailed methods for the calculation of global PiB retention ratio are described in Data S3. Statistical analysis
Demographic and patient characteristics data were analyzed using one-way ANOVA for continuous variables and chi-squared tests for dichotomous variables. Two-sided P < 0.05 was considered statistically significant. To test for localized differences across groups, voxel-wise statistical analysis of individual skeleton
© 2015 EAN
711
images was performed based on a non-parametric permutation test [21] using ‘randomise v2.9’ (part of FSL). Age, gender and level of education were included as covariates in the analysis of covariance (ANCOVA), and the null distribution was built up over 5000 permutations. For control over multiple comparison correction, threshold-free cluster enhancement was used with the ‘2D’ parameter settings [22]. The results for FA and MD were considered significant for family-wise error (FWE) corrected P < 0.01.
Results Patient characteristics
The patient characteristics are described in Table 1. There were no differences in clinical dementia rating sum-of-boxes between PiB(+) AD and PiB( ) SVaD patients. Comparisons of metrics between NC, PiB(+) AD and PiB( ) SVaD groups
Tract-based spatial statistics revealed decreased FA in the PiB(+) AD group compared to the NC group in the bilateral frontal, temporal and parietal WM regions and the genu and splenium of the corpus callosum (Fig. 1a). Compared to NC, the PiB( ) SVaD group exhibited decreased FA in all WM regions (Fig. 1b). Direct comparisons between PiB(+) AD and PiB( ) SVaD groups showed that PiB( ) SVaD groups had decreased FA in all WM regions whilst there were no regions where PiB(+) AD had decreased FA (Fig. 1c). Tract-based spatial statistics revealed increased MD in the PiB(+) AD group compared to the NC group in the left frontal and temporal WM regions (Fig. 2a). Compared to NC, the PiB( ) SVaD group revealed
Table 1 Demographics of subjects Characteristics a
Age, years Female, nb Education, yearsa MMSEa CDR-SOBa
NC (56)
PiB(+) AD (40)
PiB( ) SVaD (32)
62.5 7.5 40 (71.4) 12.5 5.0
67.0 9.1* 26 (65.0)* 10.4 5.3
71.3 7.3* 15 (46.9)* 8.6 4.6*
28.8 1.4 0.58 0.45
17.4 6.1* 4.85 2.29*
21.6 4.6*† 5.97 3.84*
NC, normal controls; PiB, Pittsburgh compound B; AD, Alzheimer’s disease; SVaD, subcortical vascular dementia; MMSE, Mini-Mental State Examination; CDR-SOB, clinical dementia rating sum-ofboxes. aValues are means SD; bnumber of cases with percentages in parentheses; *P < 0.05 between NC and PiB(+) AD or PiB( ) SVaD; †P < 0.05 between PiB(+) AD and PiB( ) SVaD.
712
Y. J. KIM ET AL.
(a)
(b)
(c)
Figure 1 Tract-based spatial statistics results of FA images. Green represents MNI152 standard brain and the skeleton image. Red yellow represents decreased FA in PiB(+) AD patients compared with NC subjects (a), PiB( ) SVaD patients compared with NC subjects (b) and PiB( ) SVaD patients compared with PiB(+) AD patients (c). P < 0.01, FWE corrected for multiple comparisons.
increased MD in all WM regions except occipital WM regions (Fig. 2b). Direct comparisons between the PiB (+) AD and PiB( ) SVaD groups showed that PiB( ) SVaD groups had increased MD in all WM regions except occipital WM regions, whilst there were no regions where PiB(+) AD had increased MD (Fig. 2c). To further illustrate the magnitude of these effects, Fig. 3 presents DTI metrics in the representative regions from superior frontal, medial temporal and lateral parietal WM regions for the NC, PiB(+)
AD and PiB( ) SVaD groups. These results were consistent with the statistical maps of differences in DTI metrics between the groups.
Discussion A novel finding of WM microstructural changes in PiB(+) AD without WMH and PiB( ) SVaD is reported. Although previous DTI studies have shown that AD and SVaD patients had WM microstructural
© 2015 EAN
WM CHANGES IN PURE AD AND SVAD
713
(a)
(b)
(c)
Figure 2 Tract-based spatial statistics results of MD images. Green represents MNI152 standard brain and the skeleton image. Blue light blue represents increased MD in PiB(+) AD patients compared with NC subjects (a), PiB( ) SVaD patients compared with NC subjects (b) and PiB( ) SVaD patients compared with PiB(+) AD patients (c). P < 0.01, FWE corrected for multiple comparisons.
changes, they did not exclude patients with combined AD and SVaD pathologies. Therefore, our study is the first showing that pure AD and pure SVaD have divergent topography of WM microstructural changes including normal appearing WM. In the present study, PiB(+) AD without WMH showed altered microstructural changes not only in the temporo-parietal WM regions but also in the frontal WM regions. Specifically, our finding that PiB(+) AD without WMH had increased MD in the medial
© 2015 EAN
temporal region and decreased FA in the splenium of the corpus callosum is consistent with previous studies [23]. AD pathology characteristically involved the cortical regions adjacent to these WM regions, which in turn led to the involvement of WM through Wallerian degeneration [23]. Interestingly, frontal WM involvement in these patients was also observed. Although previous studies showed microstructural changes in the frontal WM region [24], these might be due to combined cerebrovascular disease markers. Frontal
714
Y. J. KIM ET AL.
(a)
(b)
(c)
(d)
(e)
(f)
Figure 3 Diffusion tensor imaging metrics in the representative regions for NC, PiB(+) AD and PiB( ) SVaD groups. Mean FA values from superior frontal (a), medial temporal (b) and lateral parietal white matter regions (c). Mean MD values from superior frontal (d), medial temporal (e) and lateral parietal white matter regions (f).
WM involvement may be related to myelin breakdown as well as Wallerian degeneration [25]. That is, toxic amyloid burdens distributed predominantly in the frontal regions may affect myelin sheath or oligodendrocytes, leading to WM microstructural changes in the frontal region. Indeed, our suggestion may be supported by our finding that PiB(+) AD without WMH had decreased FA predominantly in the frontal-parietal WM regions. According to the retrogenesis hypothesis, AD pathology might preferentially affect late-myelinating WM tracts that connect frontal and parietal lobes [25]. Pittsburgh compound B negative SVaD patients showed microstructural changes in all brain WM regions compared to NC. Although clinically diagnosed SVaD patients were reported to show decreased FA in the anterior and posterior brain WM regions [8], it was hypothesized that posterior WM involvement might be driven by combined AD pathology. Because PiB(+) SVaD patients were excluded in this study, it is the first to show that pure SVaD patients have decreased FA in the anterior and posterior brain WM regions. The pathobiology of WM involvement in pure SVaD patients remains unclear. Previous pathological studies suggested that WMH correlated with the severity of rarefaction of myelin, axonal loss or gliosis
[26]. The pathological correlates are also reported to affect microstructural metrics of DTI [6]. In fact, WMH volumes were negatively associated with FA values [27]. Moreover, decreased FA values were observed in the normal appearing WM of cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy [28]. Finally, these microstructural alterations might propagate to distant WM regions through intra-hemispheric cortico-cortical disconnection caused by Wallerian degeneration, dying back or myelin breakdown [23]. Indeed, our findings of posterior involvements in pure SVaD patients might be supported by previous studies from our group [9,12,29]. These findings suggest that PiB ( ) SVaD patients had cortical atrophy in temporoparietal regions as well as the frontal region and corresponding cognitive impairments including memory and visuospatial dysfunctions. There are some limitations to this study. First, because postmortem studies were not performed, other pathologies could not be measured including other AD (soluble amyloid and neurofibrillary tangles), cardiovascular disease (microinfarct) or possible combined degenerative dementia (DLB and FTD) pathologies, which are also associated with microstructural changes in the WM. Secondly, PiB PET scanning was not performed in the NC group.
© 2015 EAN
WM CHANGES IN PURE AD AND SVAD
Previous studies showed that approximately 10% 40% of NC subjects could be PiB(+) [30]. Although there has been some controversy on the relationship between PiB retention and WM microstructural changes in cognitively normal subjects, there could have been some underestimation of the degree of WM microstructural changes in subjects with PiB(+) AD and PiB( ) SVaD. Finally, PiB PET may not be sufficiently sensitive to detect soluble amyloid oligomers or a very low level of compact plaques. However, our findings suggested that pure AD and SVaD patients showed divergent topography of WM microstructural change and provided a better understanding of their cognitive symptoms.
4. 5.
6.
7.
8.
Acknowledgements This study was supported by the Basic Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (NRF-2013R1A1A2065365), by the Korean Healthcare Technology R&D Project Ministry for Health & Welfare Affairs (HI10C2020 and HIC120713), by the Korea Ministry of Environment (MOE) as the Environmental Health Action Program (2014001360002), by the KOSEF NRL program grant (MEST; 20110028333), by Samsung Medical Center (CRL108011&CRS110-14-1) and by the Converging Research Center Program through the Ministry of Science, ICT and Future Planning, Korea (2013K000338).
9.
10.
11.
12.
13.
14.
Disclosure of conflicts of interest The authors declare no financial or other conflicts of interest.
15.
Supporting Information Additional Supporting Information may be found in the online version of this article:
16.
Data S1. Imaging parameters for MRI acquisition. Data S2. PET acquisition. Data S3. Calculation of the global PiB retention ratio.
17.
18.
References
19.
1. Walsh DM, Selkoe DJ. A beta oligomers a decade of discovery. J Neurochem 2007; 101: 1172–1184. 2. Roman GC, Erkinjuntti T, Wallin A, Pantoni L, Chui HC. Subcortical ischaemic vascular dementia. Lancet Neurol 2002; 1: 426–436. 3. Graham NL, Emery T, Hodges JR. Distinctive cognitive profiles in Alzheimer’s disease and subcortical vascular
© 2015 EAN
20.
21.
715
dementia. J Neurol Neurosurg Psychiatry 2004; 75: 61–71. Chui HC. Subcortical ischemic vascular dementia. Neurol Clin 2007; 25: 717–740, vi. Park JH, Seo SW, Kim C, et al. Effects of cerebrovascular disease and amyloid beta burden on cognition in subjects with subcortical vascular cognitive impairment. Neurobiol Aging 2014; 35: 254–260. Nitkunan A, McIntyre DJ, Barrick TR, et al. Correlations between MRS and DTI in cerebral small vessel disease. NMR Biomed 2006; 19: 610–616. Barber R, Scheltens P, Gholkar A, et al. White matter lesions on magnetic resonance imaging in dementia with Lewy bodies, Alzheimer’s disease, vascular dementia, and normal aging. J Neurol Neurosurg Psychiatry 1999; 67: 66–72. Kim SH, Park JS, Ahn HJ, et al. Voxel-based analysis of diffusion tensor imaging in patients with subcortical vascular cognitive impairment: correlates with cognitive and motor deficits. J Neuroimaging 2011; 21: 317–324. Klunk WE, Engler H, Nordberg A, et al. Imaging brain amyloid in Alzheimer’s disease with Pittsburgh compound-B. Ann Neurol 2004; 55: 306–319. Cho H, Seo SW, Kim JH, et al. Amyloid deposition in early onset versus late onset Alzheimer’s disease. J Alzheimers Dis 2013; 35: 813–821. Lee JH, Kim SH, Kim GH, et al. Identification of pure subcortical vascular dementia using 11C-Pittsburgh compound B. Neurology 2011; 77: 18–25. Park HK, Na DL, Han SH, et al. Clinical characteristics of a nationwide hospital-based registry of mild-to-moderate Alzheimer’s disease patients in Korea: a CREDOS (Clinical Research Center for Dementia of South Korea) study. J Korean Med Sci 2011; 26: 1219–1226. Kim CH, Seo SW, Kim GH, et al. Cortical thinning in subcortical vascular dementia with negative 11C-PiB PET. J Alzheimers Dis 2012; 31: 315–323. Yoon CW, Shin JS, Kim HJ, et al. Cognitive deficits of pure subcortical vascular dementia vs. Alzheimer disease: PiB-PET-based study. Neurology 2013; 80: 569– 573. McKhann G, Drachman D, Folstein M, Katzman R, Price D, Stadlan EM. Clinical diagnosis of Alzheimer’s disease: report of the NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer’s Disease. Neurology 1984; 34: 939–944. Jenkinson M, Smith S. A global optimisation method for robust affine registration of brain images. Med Image Anal 2001; 5: 143–156. Smith SM, Zhang Y, Jenkinson M, et al. Accurate, robust, and automated longitudinal and cross-sectional brain change analysis. NeuroImage 2002; 17: 479–489. Pierpaoli C, Basser PJ. Toward a quantitative assessment of diffusion anisotropy. Magn Reson Med 1996; 36: 893–906. Basser PJ, Mattiello J, LeBihan D. MR diffusion tensor spectroscopy and imaging. Biophys J 1994; 66: 259–267. Smith SM, Jenkinson M, Johansen-Berg H, et al. Tractbased spatial statistics: voxelwise analysis of multi-subject diffusion data. NeuroImage 2006; 31: 1487–1505. Nichols TE, Holmes AP. Nonparametric permutation tests for functional neuroimaging: a primer with examples. Hum Brain Mapp 2002; 15: 1–25.
716
Y. J. KIM ET AL.
22. Smith S, Nichols T. Threshold-free cluster-enhancement addressing the problem of threshold dependence in cluster inference. In: Proc. Human Brain Mapping Conference, Chicago 2007: S92. 23. Coleman M. Axon degeneration mechanisms: commonality amid diversity. Nat Rev Neurosci 2005; 6: 889–898. 24. Sexton CE, Kalu UG, Filippini N, Mackay CE, Ebmeier KP. A meta-analysis of diffusion tensor imaging in mild cognitive impairment and Alzheimer’s disease. Neurobiol Aging 2011; 32: 2322. e5–e18. 25. Benitez A, Fieremans E, Jensen JH, et al. White matter tract integrity metrics reflect the vulnerability of latemyelinating tracts in Alzheimer’s disease. Neuroimage Clin 2013; 4: 64–71. 26. Fazekas F, Kleinert R, Offenbacher H, et al. Pathologic correlates of incidental MRI white matter signal hyperintensities. Neurology 1993; 43: 1683–1689.
27. Zhan W, Zhang Y, Mueller SG, et al. Characterization of white matter degeneration in elderly subjects by magnetic resonance diffusion and FLAIR imaging correlation. NeuroImage 2009; 47(Suppl. 2): T58–T65. 28. Chabriat H, Pappata S, Poupon C, et al. Clinical severity in CADASIL related to ultrastructural damage in white matter: in vivo study with diffusion tensor MRI. Stroke 1999; 30: 2637–2643. 29. Kim HJ, Ye BS, Yoon CW, et al. Cortical thickness and hippocampal shape in pure vascular mild cognitive impairment and dementia of subcortical type. Eur J Neurol 2014; 21: 744–751. 30. Pike KE, Savage G, Villemagne VL, et al. Beta-amyloid imaging and memory in non-demented individuals: evidence for preclinical Alzheimer’s disease. Brain 2007; 130: 2837–2844.
© 2015 EAN
Copyright of European Journal of Neurology is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
