Neurobiology of Aging 36 (2015) 485e491
Contents lists available at ScienceDirect
Neurobiology of Aging journal homepage: www.elsevier.com/locate/neuaging
Hippocampal volume and shape in pure subcortical vascular dementia Geon Ha Kim a, b, Jae Hong Lee c, Sang Won Seo b, Jeong Hun Kim d, Joon-Kyung Seong e, f, Byoung Seok Ye g, Hanna Cho h, Young Noh i, Hee Jin Kim b, Cindy W. Yoon j, Seung Jun Oh k, Jae Seung Kim k, Yearn Seong Choe l, Kyung Han Lee l, Sung Tae Kim m, Jung Won Hwang n, Jee Hyang Jeong a, Duk L. Na b, * a
Department of Neurology, Ewha Womans University Mokdong Hospital, Ewha Womans University School of Medicine, Seoul, Korea Department of Neurology, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea c Department of Neurology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Korea d Department of Computer and Radio Communications Engineering, Korea University, Seoul, Korea e Department of Biomedical Engineering, Korea University, Seoul, Korea f Department of Brain and Cognitive Engineering, Korea University, Seoul, Korea g Department of Neurology, Yonsei University College of Medicine, Seoul, Korea h Department of Neurology, Gangnam Severance Hospital, Yonsei University College of Medicine, Seoul, Korea i Department of Neurology, Gachon University Gil Medical Center, Incheon, Korea j Department of Neurology, Inha University Hospital, Inha University School of Medicine, Incheon, Korea k Department of Nuclear Medicine, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Korea l Department of Nuclear Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea m Department of Radiology, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea n Samsung Advanced Institute for Health Sciences and Technology, Sungkyunkwan University, Seoul, Korea b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 2 August 2013 Received in revised form 1 August 2014 Accepted 7 August 2014 Available online 28 September 2014
The purposes of the present study were to explore whether hippocampal atrophy exists in pure subcortical vascular dementia (SVaD) as defined by negative 11C-Pittsburg compound-B (PiB) positron emission tomography and to compare hippocampal volume and shape between PiB SVaD and PiB positive (PiBþ) Alzheimer’s disease (AD) dementia. Hippocampal volume and shape were compared among 40 patients with PiB SVaD, 34 with PiBþ AD, and 21 elderly with normal cognitive function (NC). The normalized hippocampal volume of PiB SVaD was significantly smaller than NC but larger than that of PiBþ AD (NC > PiB SVaD > PiBþ AD). Both PiB SVaD and PiBþ AD patients had deflated shape changes in the cornus ammonis (CA) 1 and subiculum compared with NC. However, direct comparison between PiB SVaD and PiBþ AD demonstrated more inward deformity in the subiculum of the left hippocampus in PiBþ AD. PiB SVaD patients did have smaller hippocampal volumes and inward shape change on CA 1 and subiculum compared with NC, suggesting that cumulative ischemia without amyloid pathology could lead to hippocampal atrophy and shape changes. Ó 2015 Elsevier Inc. All rights reserved.
Keywords: Alzheimer’s disease Vascular dementia Hippocampus PiB-PET
1. Introduction Vascular dementia (VaD) is one of the most common causes of dementia after Alzheimer’s disease (AD), accounting for perhaps 1/5 as many cases as AD (Du et al., 2002). Subcortical vascular dementia (SVaD) is regarded as one of the most common types of VaD, which is caused by ischemia of cerebral white-matter and * Corresponding author at: Department of Neurology, Samsung Medical Center, Sungkyunkwan University School of Medicine, 81 Irwon-ro, Gangnam-gu, Seoul 135-710, Korea. Tel.: þ82 2 3410 3591/3599; fax: þ82 2 3410 0052. E-mail address: [email protected] (D.L. Na). 0197-4580/$ e see front matter Ó 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.neurobiolaging.2014.08.009
multiple lacunar infarcts (Roman et al., 2002). Although ischemia is the primary underpinning pathology of SVaD, patients with clinically diagnosed SVaD often exhibit both vascular and concomitant AD pathology (Jellinger and Attems, 2007; Korczyn, 2002). Recently developed 11C-Pittsburg compound-B positron emission tomography (PiB-PET) is a sensitive method to detect amyloid burden in living subjects (Klunk et al., 2004). Therefore, PiB imaging may discriminate patients with relatively pure SVaD (PiB negative) from those with mixed pathology (PiB positive). Hippocampal atrophy is a key characteristic in patients with AD (Burton et al., 2009; Gosche et al., 2002). Previous studies also reported that hippocampal atrophy is present in patients with VaD
486
G.H. Kim et al. / Neurobiology of Aging 36 (2015) 485e491
(Du et al., 2002; Laakso et al., 1996). However, it has not yet been clarified whether hippocampal atrophy in VaD might result from the accumulating burden of ischemia or combined AD pathology (Du et al., 2002; Laakso et al., 1996; Zarow et al., 2005) because previous studies did not discriminate the patients with pure VaD from mixed AD combined with small-vessel disease (mixed dementia) at the premortem stage (Du et al., 2002; Laakso et al., 1996; van de Pol et al., 2011; Zarow et al., 2005). The hippocampus is an intricate structure that includes the subfields cornus ammonis (CA) 1-4, dentate gyrus, and subiculum. Therefore, volumetric measures alone may not be enough to clarify the specific regional changes of the hippocampus in patients with dementia. In this regard, hippocampal shape analysis has been used to detect subtle changes in hippocampal subfields even in the early stages of dementia (Costafreda et al., 2011; Csernansky et al., 2005; Frisoni et al., 2008). The purposes of the present study, therefore, were to explore whether hippocampal atrophy exists in patients with pure SVaD as defined by the absence of amyloid deposition in PiB-PET (PiB SVaD) and to compare the hippocampal shape between patients with PiB SVaD and PiBþ AD. 2. Materials and methods 2.1. Patients with SVaD From September 2008 to September 2010, we have been prospectively recruiting new or follow-up patients with SVaD to participate in the Amyloid PET Imaging for Subcortical Vascular Dementia study (Lee et al., 2011) conducted by the Memory Disorder Clinic at the Samsung Medical Center or at the Asan Medical Center in Seoul, Korea. Over this period of time, a total of 516 patients were evaluated: 248 patients were diagnosed with probable AD according to published criteria (National Institute of Neurological and Communicative Disorders and Stroke and the Alzheimer’s Disease and Related Disorders Association) (McKhann et al., 1984) with minimal evidence of cerebrovascular disease, 147 patients were diagnosed with possible AD but had significant ischemia, and 121 patients were diagnosed with clinically probable SVaD. All SVaD patients fulfilled the following criteria: (1) 50 age 85 years; (2) Korean version of Mini-Mental State Examination (KMMSE) score 10; (3) Diagnostic and Statistical Manual of Mental Disordersdfourth edition criteria for VaD (American Psychiatric Association, 1994); and (4) severe white-matter hyperintensity (WMH) on magnetic resonance imaging (MRI) (a cap or band 10 mm and a deep white-matter lesion 25 mm as modified from the Fazekas ischemia criteria, Fazekas et al., 1993). We excluded patients with other structural lesions on brain MRI such as territorial infarction, intracranial hemorrhage, hydrocephalus, or WMHs associated with radiation, multiple sclerosis, or vasculitis. Patients completed laboratory tests including apolipoprotein E (APOE) genotyping and underwent detailed neuropsychological tests (Ahn et al., 2010). All diagnostic tests were performed 3 months before or after the PiB scan. Among 121 SVaD patients, 61 patients agreed to participate in the Amyloid PET Imaging for Subcortical Vascular Dementia study. Out of 61, 45 with SVaD were the same as those participants from our study that has been previously published (Lee et al., 2011). The 61 patients were not significantly different from the remaining 60 patients with SVaD who did not enroll in this study in terms of age (74.3 vs. 75.9 years), sex (% female, 59 vs. 60), education (9.3 vs. 7.8 years), clinical dementia rating (CDR) (1.2 vs. 1.2), and sum of boxes (6.7 vs. 5.6), whereas the MMSE scores of study participants were significantly higher than those in SVaD patients who did not participate in
this study (20.8 vs 18.5, p ¼ 0.02). Out of 61 patients with SVaD, 41 (67.2%) patients were negative for amyloid uptake (PiB), whereas 20 (32.8%) were positive PiBþ. Among the 41 patients, 1 participant with PiB SVaD was excluded from hippocampal volumetric analysis because of error in hippocampal mesh extraction. Therefore, the final number of PiB SVaD patients was 40. 2.2. Patients with AD dementia Out of 248 patients with probable AD dementia with minimal evidence of cerebrovascular disease, only 63 patients with AD underwent PiB-PET scanning. Minimal evidence of cerebrovascular disease was defined as a cap or band 10 mm and a deep whitematter lesion 10 mm, as modified from the Fazekas ischemia criteria (Fazekas et al., 1993). Of the 63 AD patients, 14 patients with early-onset AD, including 2 autosomal dominant familial AD with positive presenilin 1 mutation, were excluded from this study. None of the remaining 49 patients had a family history suggestive of an autosomal dominant AD. Fifteen patients were also excluded for the following reasons: 9 patients were PiB negative and 6 patients had inadequate MRI imaging such as severe motion artifact to analyze hippocampal shape analysis. Therefore, the final sample size for the group of AD patients consisted of 34 subjects. These 34 patients did not differ from the 214 patients who were not included in the study in terms of age (71.6 vs. 73.6 years), sex (% female, 61.8 vs. 64.7), MMSE scores (17.8 vs. 18.1), and the years of education (10.4 vs. 9.1 years), whereas CDR (0.8 vs. 1.3) and sum of boxes (5.0 vs. 7.0) of the study participants were significantly better than those of the nonstudy samples (p < 0.001). 2.3. Individuals with normal cognitive impairment There were 21 individuals with normal cognitive impairment (NC) who served as controls for the analysis of the hippocampus (NC). These controls were spouses of outpatients in the Memory Disorder Clinic of Samsung Medical Center and had neither a history of neurological and psychiatric illnesses nor abnormalities on neurological examinations. These NC exhibited normal performances on the MMSE and neuropsychological tests. The individuals who had significant WMH on their MRI were excluded in this group. 2.4. Standard protocol approvals, registrations, and patient consents We obtained written consent from each participant, and the Institutional Review Board of Asan Medical Center and Samsung Medical Center approved the study protocol. 2.5. Image acquisition and processing 2.5.1. MR image acquisition All patients were sent to Samsung Medical Center for MR images that were acquired employing 5 different techniques (3-dimensional T1 turbo field echo [TFE], fluid-attenuated inversion recovery, T1 reference, T2 fast field echo) using the same imaging protocols and the same 3.0-T MRI scanner (Philips 3.0-T Achieva; The Netherlands). Three-dimensional T1 TFE MR images were acquired with the following imaging parameters: sagittal slice thickness 1.0 mm, no gap, repetition time (TR) 9.9 ms, echo time (TE) 4.6 ms, flip angle 8 , and a matrix size of 480 480 pixels. The second technique was fluid-attenuated inversion recovery MR images, which were acquired with an axial slice thickness of 2 mm, no gap, TR of 11,000.0 ms, TE of 125.0 ms, flip angle of 90 , and a matrix size of 512 512 pixels. The third technique was T1 reference MR imaging; images
G.H. Kim et al. / Neurobiology of Aging 36 (2015) 485e491
were acquired with an axial slice thickness of 4 mm, no gap, TR of 545 ms, TE of 10 ms, flip angle of 70 , and a matrix size of 512 512 pixels. The third technique was T2 MR images, which were acquired using an axial slice thickness of 5.0 mm, interslice thickness of 1.5 mm, TR of 3000.0 ms, TE of 80.0 ms, flip angle of 90 , and a matrix size of 512 512 pixels. Finally, T2 fast field echo (FFE) images were obtained using the following parameters: axial slice thickness 5.0 mm, interslice thickness 2 mm, TR 669 ms, TE 16 ms, flip angle 18 , and a matrix size of 560 560 pixels. All axial sections were obtained parallel to the anterior-posterior commissure lines. 2.5.2. PiB-PET All patients completed the PiB-PET scan at Samsung Medical Center or Asan Medical Center. All subjects completed a same type of PET scan with a Discovery STe PET/CT scanner (GE Medical Systems, Milwaukee, WI, USA). The detailed radiochemistry profiles and scanning protocol are described in a previous study (Lee et al., 2011). 2.6. Hippocampal volume and shape analysis Our hippocampal shape analysis method was based on boundary surfaces of the hippocampus; thus, the volume of each hippocampus was measured from its boundary surface instead of counting the number of voxels contained within the structure. The shape analysis method consisted of 3 steps: hippocampal surface construction, surface registration, and surface deformity computation. The first step constructed surface meshes for each hippocampus. Thus, this step constructed 2 surface meshes for both hemispheres in each subject. The T1 TFE images of each subject were first processed to obtain the anatomic parcellations of hippocampal structures by the FreeSurfer software package (version 5.1; Athinoula A. Martinos Center at the Massachusetts General Hospital, Harvard Medical School; http://www.surfer.nmr.mgh.harvard.edu/). After parcellations, the hippocampal mesh surfaces were then extracted from the labeled images for each subject by employing the Laplacianbased surface modeling system (Kim and Park, 2012) (Supplementary Fig. 1). The results of all automated constructed hippocampal surfaces were visually inspected by a single rater (J-KS) who was blinded for subjects’ cognitive status. The visual investigation found an error in 1 patient with PiB SVaD (0.9%) among all 106 participants in this study.
487
We could not perform a manual correction for the errors from this participant; therefore, we excluded the data in this analysis. In the second step after we generated a surface mesh for each subject, we performed a surface-based registration using the template surface mesh. For the registration, we employed the method that was proposed in the previous study (Cho et al., 2011). Each surface mesh from each participant was registered to the template mesh of normal control group. The surface registration provided vertex correspondences for all subcortical surface meshes. The final step measured relative deformation of the subcortical surface meshes against the template. The surface deformity was calculated for each vertex by measuring the spatial displacement along its outward normal direction. Given that a surface mesh represented hippocampus, we estimated its volume based on the discrete form of the divergence theorem (Alyassin et al., 1994). Because the hippocampal surface mesh is a closed surface, we used the Visualization Toolkit implementation for the volume measurement (http://www.vtk.org/doc/ release/5.2/html/a00827.html). Hippocampal volume (mm3) was calculated on each individual’s hippocampal mesh on the standard space. The ICV of each participant was calculated through an automated process using the mri_segstats command in FreeSurfer (for more details, http://surfer.nmr.mgh.harvard.edu/fswiki/eTIV). 2.7. Neuropsychological tests All participants received standardized neuropsychological tests using the Seoul Neuropsychological Screening Battery (Ahn et al., 2010). This battery comprised test for attention, language, calculation, praxis, visuospatial/constructive function, verbal/visual memory, and frontal/executive function. 2.8. Statistical analyses Statistical analyses were performed using the Statistical Package for the Social Sciences 18.0 (SPSS Inc, Chicago, IL, USA). Descriptive statistics of the initial workup were performed using demographic and clinical scores. Student t test was used to assess continuous variables and chi-square test to assess dichotomous variables. The analysis of covariance was used to compare hippocampal volumes
Table 1 Demographic and clinical characteristics of PiB SVaD and PiBþ AD patients Characteristics
Mean SD NC (n ¼ 21)
PiB SVaD (n ¼ 40)
PiBþ AD (n ¼ 34)
Age (y) Gender (M:F) Education K-MMSE CDR CDR SOB Hypertension, n (%) Diabetes mellitus, n (%) Hyperlipidemia, n (%) Cardiac disease, n (%) Lacunes Microbleeds APOE ε4 carrier, n (%)c ICV (cm3) PiB retention (SUV) WMH volume (cm3)
70.9 4.4 8:13 10.7 4.9 28.7 1.3 0.2 0.2 0.3 0.5 4 (19.0) 4 (19.0) 4 (19.0) 0 (0) 0.3 0.8 0.3 0.6 1 (20.0) 1323.6 102.8 d 1.6 1.1
71.8 7.1 19:21 9.2 4.8 22.0 4.6a 0.9 0.6a 5.8 3.6a 32 (80.0)a 10 (25.0) 19 (47.5) 6 (15.0)a 20.7 18.8a 8.9 10.4a 7 (18.9)a 1382.2 136.7 1.2 0.1 39.7 13.2a
71.6 5.8 13:21 10.7 5.7 17.8 4.5a,b 0.9 0.4a 5.12.3a 15 (44.1)a,b 2 (5.9) b 13 (38.2) 1(2.9) 0.2 0.6b 0.1 0.3b 20 (58.8)a,b 1352.2 153.2 2.3 0.3b 3.1 3.4b
Key: AD, Alzheimer’s disease; APOE, apolipoprotein E; CDR, clinical dementia rating; F, female; ICV, intracerebral volume; K-MMSE, Korean version of the Mini-Mental State Examination; M, male; NC, normal controls; PiB, 11C-Pittsburg compound-B; SD, standard deviation; SOB, sum of box; SUV, standardized uptake value; SVaD, subcortical vascular dementia; WMH, white-matter hyperintensity. a Bonferroni post hoc test differs from NC. b Bonferroni post hoc test differs from PiB SVaD. c APOE genotyping was performed in 37 patients with PiB SVaD, 34 with AD, and 5 with NC.
488
G.H. Kim et al. / Neurobiology of Aging 36 (2015) 485e491
Table 2 Neuropsychological tests in PiB SVaD and PiB+ AD patients compared with NC Neuropsychological tests
NC (n ¼ 21) Attention Digit span: forward Digit span: backward Language and calculation K-BNT Calculation Visuospatial function RCFT Memory SVLT: immediate recall SVLT: delayed recall SVLT: recognition RCFT: immediate recall RCFT: delayed recall RCFT: recognition Frontal/executive function COWAT: semantic COWAT: phonemic Stroop test: word reading Stroop test: color reading
Pairwise p valuea
Mean SD
PiB SVaD (n ¼ 40)
+
PiB AD (n ¼ 34)
NC vs. PiB SVaD
NC vs. PiB+ AD
PiB SVaD vs. PiB+ AD
6.7 1.2 4.2 0.6
5.1 1.2 2.7 1.0
5.2 1.2 2.8 1.0
Contents lists available at ScienceDirect
Neurobiology of Aging journal homepage: www.elsevier.com/locate/neuaging
Hippocampal volume and shape in pure subcortical vascular dementia Geon Ha Kim a, b, Jae Hong Lee c, Sang Won Seo b, Jeong Hun Kim d, Joon-Kyung Seong e, f, Byoung Seok Ye g, Hanna Cho h, Young Noh i, Hee Jin Kim b, Cindy W. Yoon j, Seung Jun Oh k, Jae Seung Kim k, Yearn Seong Choe l, Kyung Han Lee l, Sung Tae Kim m, Jung Won Hwang n, Jee Hyang Jeong a, Duk L. Na b, * a
Department of Neurology, Ewha Womans University Mokdong Hospital, Ewha Womans University School of Medicine, Seoul, Korea Department of Neurology, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea c Department of Neurology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Korea d Department of Computer and Radio Communications Engineering, Korea University, Seoul, Korea e Department of Biomedical Engineering, Korea University, Seoul, Korea f Department of Brain and Cognitive Engineering, Korea University, Seoul, Korea g Department of Neurology, Yonsei University College of Medicine, Seoul, Korea h Department of Neurology, Gangnam Severance Hospital, Yonsei University College of Medicine, Seoul, Korea i Department of Neurology, Gachon University Gil Medical Center, Incheon, Korea j Department of Neurology, Inha University Hospital, Inha University School of Medicine, Incheon, Korea k Department of Nuclear Medicine, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Korea l Department of Nuclear Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea m Department of Radiology, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea n Samsung Advanced Institute for Health Sciences and Technology, Sungkyunkwan University, Seoul, Korea b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 2 August 2013 Received in revised form 1 August 2014 Accepted 7 August 2014 Available online 28 September 2014
The purposes of the present study were to explore whether hippocampal atrophy exists in pure subcortical vascular dementia (SVaD) as defined by negative 11C-Pittsburg compound-B (PiB) positron emission tomography and to compare hippocampal volume and shape between PiB SVaD and PiB positive (PiBþ) Alzheimer’s disease (AD) dementia. Hippocampal volume and shape were compared among 40 patients with PiB SVaD, 34 with PiBþ AD, and 21 elderly with normal cognitive function (NC). The normalized hippocampal volume of PiB SVaD was significantly smaller than NC but larger than that of PiBþ AD (NC > PiB SVaD > PiBþ AD). Both PiB SVaD and PiBþ AD patients had deflated shape changes in the cornus ammonis (CA) 1 and subiculum compared with NC. However, direct comparison between PiB SVaD and PiBþ AD demonstrated more inward deformity in the subiculum of the left hippocampus in PiBþ AD. PiB SVaD patients did have smaller hippocampal volumes and inward shape change on CA 1 and subiculum compared with NC, suggesting that cumulative ischemia without amyloid pathology could lead to hippocampal atrophy and shape changes. Ó 2015 Elsevier Inc. All rights reserved.
Keywords: Alzheimer’s disease Vascular dementia Hippocampus PiB-PET
1. Introduction Vascular dementia (VaD) is one of the most common causes of dementia after Alzheimer’s disease (AD), accounting for perhaps 1/5 as many cases as AD (Du et al., 2002). Subcortical vascular dementia (SVaD) is regarded as one of the most common types of VaD, which is caused by ischemia of cerebral white-matter and * Corresponding author at: Department of Neurology, Samsung Medical Center, Sungkyunkwan University School of Medicine, 81 Irwon-ro, Gangnam-gu, Seoul 135-710, Korea. Tel.: þ82 2 3410 3591/3599; fax: þ82 2 3410 0052. E-mail address: [email protected] (D.L. Na). 0197-4580/$ e see front matter Ó 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.neurobiolaging.2014.08.009
multiple lacunar infarcts (Roman et al., 2002). Although ischemia is the primary underpinning pathology of SVaD, patients with clinically diagnosed SVaD often exhibit both vascular and concomitant AD pathology (Jellinger and Attems, 2007; Korczyn, 2002). Recently developed 11C-Pittsburg compound-B positron emission tomography (PiB-PET) is a sensitive method to detect amyloid burden in living subjects (Klunk et al., 2004). Therefore, PiB imaging may discriminate patients with relatively pure SVaD (PiB negative) from those with mixed pathology (PiB positive). Hippocampal atrophy is a key characteristic in patients with AD (Burton et al., 2009; Gosche et al., 2002). Previous studies also reported that hippocampal atrophy is present in patients with VaD
486
G.H. Kim et al. / Neurobiology of Aging 36 (2015) 485e491
(Du et al., 2002; Laakso et al., 1996). However, it has not yet been clarified whether hippocampal atrophy in VaD might result from the accumulating burden of ischemia or combined AD pathology (Du et al., 2002; Laakso et al., 1996; Zarow et al., 2005) because previous studies did not discriminate the patients with pure VaD from mixed AD combined with small-vessel disease (mixed dementia) at the premortem stage (Du et al., 2002; Laakso et al., 1996; van de Pol et al., 2011; Zarow et al., 2005). The hippocampus is an intricate structure that includes the subfields cornus ammonis (CA) 1-4, dentate gyrus, and subiculum. Therefore, volumetric measures alone may not be enough to clarify the specific regional changes of the hippocampus in patients with dementia. In this regard, hippocampal shape analysis has been used to detect subtle changes in hippocampal subfields even in the early stages of dementia (Costafreda et al., 2011; Csernansky et al., 2005; Frisoni et al., 2008). The purposes of the present study, therefore, were to explore whether hippocampal atrophy exists in patients with pure SVaD as defined by the absence of amyloid deposition in PiB-PET (PiB SVaD) and to compare the hippocampal shape between patients with PiB SVaD and PiBþ AD. 2. Materials and methods 2.1. Patients with SVaD From September 2008 to September 2010, we have been prospectively recruiting new or follow-up patients with SVaD to participate in the Amyloid PET Imaging for Subcortical Vascular Dementia study (Lee et al., 2011) conducted by the Memory Disorder Clinic at the Samsung Medical Center or at the Asan Medical Center in Seoul, Korea. Over this period of time, a total of 516 patients were evaluated: 248 patients were diagnosed with probable AD according to published criteria (National Institute of Neurological and Communicative Disorders and Stroke and the Alzheimer’s Disease and Related Disorders Association) (McKhann et al., 1984) with minimal evidence of cerebrovascular disease, 147 patients were diagnosed with possible AD but had significant ischemia, and 121 patients were diagnosed with clinically probable SVaD. All SVaD patients fulfilled the following criteria: (1) 50 age 85 years; (2) Korean version of Mini-Mental State Examination (KMMSE) score 10; (3) Diagnostic and Statistical Manual of Mental Disordersdfourth edition criteria for VaD (American Psychiatric Association, 1994); and (4) severe white-matter hyperintensity (WMH) on magnetic resonance imaging (MRI) (a cap or band 10 mm and a deep white-matter lesion 25 mm as modified from the Fazekas ischemia criteria, Fazekas et al., 1993). We excluded patients with other structural lesions on brain MRI such as territorial infarction, intracranial hemorrhage, hydrocephalus, or WMHs associated with radiation, multiple sclerosis, or vasculitis. Patients completed laboratory tests including apolipoprotein E (APOE) genotyping and underwent detailed neuropsychological tests (Ahn et al., 2010). All diagnostic tests were performed 3 months before or after the PiB scan. Among 121 SVaD patients, 61 patients agreed to participate in the Amyloid PET Imaging for Subcortical Vascular Dementia study. Out of 61, 45 with SVaD were the same as those participants from our study that has been previously published (Lee et al., 2011). The 61 patients were not significantly different from the remaining 60 patients with SVaD who did not enroll in this study in terms of age (74.3 vs. 75.9 years), sex (% female, 59 vs. 60), education (9.3 vs. 7.8 years), clinical dementia rating (CDR) (1.2 vs. 1.2), and sum of boxes (6.7 vs. 5.6), whereas the MMSE scores of study participants were significantly higher than those in SVaD patients who did not participate in
this study (20.8 vs 18.5, p ¼ 0.02). Out of 61 patients with SVaD, 41 (67.2%) patients were negative for amyloid uptake (PiB), whereas 20 (32.8%) were positive PiBþ. Among the 41 patients, 1 participant with PiB SVaD was excluded from hippocampal volumetric analysis because of error in hippocampal mesh extraction. Therefore, the final number of PiB SVaD patients was 40. 2.2. Patients with AD dementia Out of 248 patients with probable AD dementia with minimal evidence of cerebrovascular disease, only 63 patients with AD underwent PiB-PET scanning. Minimal evidence of cerebrovascular disease was defined as a cap or band 10 mm and a deep whitematter lesion 10 mm, as modified from the Fazekas ischemia criteria (Fazekas et al., 1993). Of the 63 AD patients, 14 patients with early-onset AD, including 2 autosomal dominant familial AD with positive presenilin 1 mutation, were excluded from this study. None of the remaining 49 patients had a family history suggestive of an autosomal dominant AD. Fifteen patients were also excluded for the following reasons: 9 patients were PiB negative and 6 patients had inadequate MRI imaging such as severe motion artifact to analyze hippocampal shape analysis. Therefore, the final sample size for the group of AD patients consisted of 34 subjects. These 34 patients did not differ from the 214 patients who were not included in the study in terms of age (71.6 vs. 73.6 years), sex (% female, 61.8 vs. 64.7), MMSE scores (17.8 vs. 18.1), and the years of education (10.4 vs. 9.1 years), whereas CDR (0.8 vs. 1.3) and sum of boxes (5.0 vs. 7.0) of the study participants were significantly better than those of the nonstudy samples (p < 0.001). 2.3. Individuals with normal cognitive impairment There were 21 individuals with normal cognitive impairment (NC) who served as controls for the analysis of the hippocampus (NC). These controls were spouses of outpatients in the Memory Disorder Clinic of Samsung Medical Center and had neither a history of neurological and psychiatric illnesses nor abnormalities on neurological examinations. These NC exhibited normal performances on the MMSE and neuropsychological tests. The individuals who had significant WMH on their MRI were excluded in this group. 2.4. Standard protocol approvals, registrations, and patient consents We obtained written consent from each participant, and the Institutional Review Board of Asan Medical Center and Samsung Medical Center approved the study protocol. 2.5. Image acquisition and processing 2.5.1. MR image acquisition All patients were sent to Samsung Medical Center for MR images that were acquired employing 5 different techniques (3-dimensional T1 turbo field echo [TFE], fluid-attenuated inversion recovery, T1 reference, T2 fast field echo) using the same imaging protocols and the same 3.0-T MRI scanner (Philips 3.0-T Achieva; The Netherlands). Three-dimensional T1 TFE MR images were acquired with the following imaging parameters: sagittal slice thickness 1.0 mm, no gap, repetition time (TR) 9.9 ms, echo time (TE) 4.6 ms, flip angle 8 , and a matrix size of 480 480 pixels. The second technique was fluid-attenuated inversion recovery MR images, which were acquired with an axial slice thickness of 2 mm, no gap, TR of 11,000.0 ms, TE of 125.0 ms, flip angle of 90 , and a matrix size of 512 512 pixels. The third technique was T1 reference MR imaging; images
G.H. Kim et al. / Neurobiology of Aging 36 (2015) 485e491
were acquired with an axial slice thickness of 4 mm, no gap, TR of 545 ms, TE of 10 ms, flip angle of 70 , and a matrix size of 512 512 pixels. The third technique was T2 MR images, which were acquired using an axial slice thickness of 5.0 mm, interslice thickness of 1.5 mm, TR of 3000.0 ms, TE of 80.0 ms, flip angle of 90 , and a matrix size of 512 512 pixels. Finally, T2 fast field echo (FFE) images were obtained using the following parameters: axial slice thickness 5.0 mm, interslice thickness 2 mm, TR 669 ms, TE 16 ms, flip angle 18 , and a matrix size of 560 560 pixels. All axial sections were obtained parallel to the anterior-posterior commissure lines. 2.5.2. PiB-PET All patients completed the PiB-PET scan at Samsung Medical Center or Asan Medical Center. All subjects completed a same type of PET scan with a Discovery STe PET/CT scanner (GE Medical Systems, Milwaukee, WI, USA). The detailed radiochemistry profiles and scanning protocol are described in a previous study (Lee et al., 2011). 2.6. Hippocampal volume and shape analysis Our hippocampal shape analysis method was based on boundary surfaces of the hippocampus; thus, the volume of each hippocampus was measured from its boundary surface instead of counting the number of voxels contained within the structure. The shape analysis method consisted of 3 steps: hippocampal surface construction, surface registration, and surface deformity computation. The first step constructed surface meshes for each hippocampus. Thus, this step constructed 2 surface meshes for both hemispheres in each subject. The T1 TFE images of each subject were first processed to obtain the anatomic parcellations of hippocampal structures by the FreeSurfer software package (version 5.1; Athinoula A. Martinos Center at the Massachusetts General Hospital, Harvard Medical School; http://www.surfer.nmr.mgh.harvard.edu/). After parcellations, the hippocampal mesh surfaces were then extracted from the labeled images for each subject by employing the Laplacianbased surface modeling system (Kim and Park, 2012) (Supplementary Fig. 1). The results of all automated constructed hippocampal surfaces were visually inspected by a single rater (J-KS) who was blinded for subjects’ cognitive status. The visual investigation found an error in 1 patient with PiB SVaD (0.9%) among all 106 participants in this study.
487
We could not perform a manual correction for the errors from this participant; therefore, we excluded the data in this analysis. In the second step after we generated a surface mesh for each subject, we performed a surface-based registration using the template surface mesh. For the registration, we employed the method that was proposed in the previous study (Cho et al., 2011). Each surface mesh from each participant was registered to the template mesh of normal control group. The surface registration provided vertex correspondences for all subcortical surface meshes. The final step measured relative deformation of the subcortical surface meshes against the template. The surface deformity was calculated for each vertex by measuring the spatial displacement along its outward normal direction. Given that a surface mesh represented hippocampus, we estimated its volume based on the discrete form of the divergence theorem (Alyassin et al., 1994). Because the hippocampal surface mesh is a closed surface, we used the Visualization Toolkit implementation for the volume measurement (http://www.vtk.org/doc/ release/5.2/html/a00827.html). Hippocampal volume (mm3) was calculated on each individual’s hippocampal mesh on the standard space. The ICV of each participant was calculated through an automated process using the mri_segstats command in FreeSurfer (for more details, http://surfer.nmr.mgh.harvard.edu/fswiki/eTIV). 2.7. Neuropsychological tests All participants received standardized neuropsychological tests using the Seoul Neuropsychological Screening Battery (Ahn et al., 2010). This battery comprised test for attention, language, calculation, praxis, visuospatial/constructive function, verbal/visual memory, and frontal/executive function. 2.8. Statistical analyses Statistical analyses were performed using the Statistical Package for the Social Sciences 18.0 (SPSS Inc, Chicago, IL, USA). Descriptive statistics of the initial workup were performed using demographic and clinical scores. Student t test was used to assess continuous variables and chi-square test to assess dichotomous variables. The analysis of covariance was used to compare hippocampal volumes
Table 1 Demographic and clinical characteristics of PiB SVaD and PiBþ AD patients Characteristics
Mean SD NC (n ¼ 21)
PiB SVaD (n ¼ 40)
PiBþ AD (n ¼ 34)
Age (y) Gender (M:F) Education K-MMSE CDR CDR SOB Hypertension, n (%) Diabetes mellitus, n (%) Hyperlipidemia, n (%) Cardiac disease, n (%) Lacunes Microbleeds APOE ε4 carrier, n (%)c ICV (cm3) PiB retention (SUV) WMH volume (cm3)
70.9 4.4 8:13 10.7 4.9 28.7 1.3 0.2 0.2 0.3 0.5 4 (19.0) 4 (19.0) 4 (19.0) 0 (0) 0.3 0.8 0.3 0.6 1 (20.0) 1323.6 102.8 d 1.6 1.1
71.8 7.1 19:21 9.2 4.8 22.0 4.6a 0.9 0.6a 5.8 3.6a 32 (80.0)a 10 (25.0) 19 (47.5) 6 (15.0)a 20.7 18.8a 8.9 10.4a 7 (18.9)a 1382.2 136.7 1.2 0.1 39.7 13.2a
71.6 5.8 13:21 10.7 5.7 17.8 4.5a,b 0.9 0.4a 5.12.3a 15 (44.1)a,b 2 (5.9) b 13 (38.2) 1(2.9) 0.2 0.6b 0.1 0.3b 20 (58.8)a,b 1352.2 153.2 2.3 0.3b 3.1 3.4b
Key: AD, Alzheimer’s disease; APOE, apolipoprotein E; CDR, clinical dementia rating; F, female; ICV, intracerebral volume; K-MMSE, Korean version of the Mini-Mental State Examination; M, male; NC, normal controls; PiB, 11C-Pittsburg compound-B; SD, standard deviation; SOB, sum of box; SUV, standardized uptake value; SVaD, subcortical vascular dementia; WMH, white-matter hyperintensity. a Bonferroni post hoc test differs from NC. b Bonferroni post hoc test differs from PiB SVaD. c APOE genotyping was performed in 37 patients with PiB SVaD, 34 with AD, and 5 with NC.
488
G.H. Kim et al. / Neurobiology of Aging 36 (2015) 485e491
Table 2 Neuropsychological tests in PiB SVaD and PiB+ AD patients compared with NC Neuropsychological tests
NC (n ¼ 21) Attention Digit span: forward Digit span: backward Language and calculation K-BNT Calculation Visuospatial function RCFT Memory SVLT: immediate recall SVLT: delayed recall SVLT: recognition RCFT: immediate recall RCFT: delayed recall RCFT: recognition Frontal/executive function COWAT: semantic COWAT: phonemic Stroop test: word reading Stroop test: color reading
Pairwise p valuea
Mean SD
PiB SVaD (n ¼ 40)
+
PiB AD (n ¼ 34)
NC vs. PiB SVaD
NC vs. PiB+ AD
PiB SVaD vs. PiB+ AD
6.7 1.2 4.2 0.6
5.1 1.2 2.7 1.0
5.2 1.2 2.8 1.0
