![](/uploads/1/2/5/7/125745166/516490207.jpg)
Alzheimer’s disease dementia (ADD) and subcortical vascular dementia (SVaD) both show cortical thinning and white matter (WM) microstructural changes. We evaluated different patterns of correlation between gray matter (GM) and WM microstructural changes in pure ADD, pure SVaD, and mixed dementia.
Yes, your brain is a muscle, but about 70% of its gray and white matter is composed of fat. And it's not just any fat; it's the long-chain polyunsaturated fats that are part of the cell membranes.
We enrolled 40 Pittsburgh compound B (PiB) positive ADD patients without WM hyperintensities (pure ADD), 32 PiB negative SVaD patients (pure SVaD), 23 PiB positive SVaD patients (mixed dementia), and 56 normal controls. WM microstructural integrity was quantified using fractional anisotropy (FA), axial diffusivity (DA), and radial diffusivity (DR) values.
We used sparse canonical correlation analysis to show correlated regions of cortical thinning and WM microstructural changes. In pure ADD patients, lower FA in the frontoparietal area correlated with cortical thinning in the left inferior parietal lobule and bilateral paracentral lobules.
In pure SVaD patients, lower FA and higher DR across extensive WM regions correlated with cortical thinning in bilateral fronto-temporo-parietal regions. In mixed dementia patients, DR and DA changes across extensive WM regions correlated with cortical thinning in the bilateral fronto-temporo-parietal regions.
Our findings showed that the relationships between GM and WM degeneration are distinct in pure ADD, pure SVaD, and mixed dementia, suggesting that different pathomechanisms underlie their correlations. Alzheimer’s disease dementia (ADD) and subcortical vascular dementia (SVaD) are the two representative causes of dementia. ADD is characterized by amyloid plaques and neurofibrillary tangles accumulating in the gray matter (GM), which leads to neuronal death and cortical thinning. SVaD is characterized by ischemic changes in the white matter (WM) 2. However, it is well known that ADD patients also have WM microstructural changes, and SVaD patients have cortical atrophy as well.The mechanism of WM-integrity disruption, along with GM degeneration (cortical thinning), in ADD and SVaD has not been well defined.
In ADD patients, secondary Wallerian degeneration or WM degeneration independent of GM lesions has been suggested as a possible mechanism for WM microstructural changes,. In SVaD, secondary axonal and trans-synaptic degeneration following subcortical injury or global ischemia in the cortex, as well as incidentally combined amyloid pathologies, have been proposed as the underlying mechanisms of cortical thinning. Despite these many hypotheses, there is a paucity of information on the correlation between WM and GM degeneration in ADD and SVaD patients.To evaluate correlation between GM and WM degeneration, we first investigated the patterns of WM microstructural changes (using DTI) and GM degeneration (using cortical thickness analysis) in patients with pure ADD, pure SVaD, and mixed dementia, which were determined by Pittsburgh compound B (PiB) PET result and WM hyperintensities degree as described in the Methods. We have previously reported that both pure ADD and pure SVaD patients showed decreased FA and increased MD. In this study, we further analyzed axial diffusivity (DA) and radial diffusivity (DR), which are diffusivities parallel or perpendicular to the WM fibers that provide more information regarding axonal or myelin damage, respectively,.
Then, we explored the correlation between WM microstructural changes and GM degeneration in each group using sparse canonical correlation analysis (sCCA), which makes it possible to combine different imaging modalities. We hypothesized that in pure SVaD patients, changes in WM would be related to myelin breakdown and subsequent axonal damage. Thus, in pure SVaD, we expected an increase in DR and a decrease in FA to correlate with cortical thinning. On the other hand, we hypothesized that in pure ADD patients, cortical thinning would be the primary change leading to secondary Wallerian degeneration in adjacent WM. Thus, in pure ADD, we expected decreases in both DA and FA to correlate to cortical thinning.
Finally, we hypothesized that mixed dementia would show both ADD and SVaD characteristic and that an increase in DR and a decrease in DA would correlate to cortical thinning. White matter diffusivity properties in patients with PiB(+) ADD, PiB(−) SVaD, and PiB(+) SVaDWe compared WM microstructural changes of each dementia group to the NC group. FA changes in PiB(+) ADD and PiB(−) SVaD group compared to NC were reported in our previous study and thus, these results are not described here. The PiB(+) ADD group showed a focal higher DR in the left frontal WM region. DA was lower bilaterally in the parietal and occipital subcortical WM regions adjacent to the cortex, while it was higher in small areas including the corona radiata WM regions (Fig. ). TBSS in patients with PiB(+) ADD, PiB(−) SVaD, and PiB(+) SVaD compared to normal controls (NC).
Warm color (red-yellow) indicates lower DTI indices in the dementia group compared to the NC group while cold color (blue-light blue) indicates higher DTI indices in the dementia group compared to the NC group. These results are overlaid on the MNI152 standard brain and the mean skeleton image (green color, FA 0.2). Abbreviations: FWE, family-wise error; FA, fractional anisotropy; DR, radial diffusivity; DA, axial diffusivity; NC, normal controls; PiB, Pittsburgh compound B; ADD, Alzheimer’s disease dementia; SVaD, subcortical vascular dementia; TBSS, tract-based spatial statistics; DTI, diffusion tensor imaging; MNI, Montreal Neurological Institute. The PiB(−) SVaD group showed more extensive WM microstructural changes. DR was higher in all WM regions.
DA was bilaterally higher in the centrum semiovale and corona radiata WM regions, while it was bilaterally lower in the parietal and occipital subcortical WM regions adjacent to the cortex (Fig. ).WM microstructural changes of the PiB(+) SVaD group were similar to the PiB(−) SVaD group. FA was lower and DR was higher in all WM regions. DA was bilaterally higher in the centrum semiovale and corona radiata WM regions, while it was bilaterally lower in the parietal and occipital subcortical WM regions adjacent to the cortex (Fig. ). Details of the TBSS parameters are described in Table. In the PiB(−) SVaD group, there were marked correlations between all DTI indices (especially FA) and cortical thickness.
Variation of FA in extensive WM regions correlated with variation of cortical thickness in the bilateral temporo-parietal, medial frontal, and right frontal areas. Variation of DR in the frontal, temporal and parietal WM regions correlated with variation of cortical thickness in the bilateral lateral frontal, left lateral temporo-parietal, left paracentral lobule, and right medial frontal areas. Variation of DA in the bilateral parietal and occipital WM regions correlated with variation of cortical thickness in the inferior temporal, and bilateral anterior and posterior cingulate areas (Fig. ).In the PiB(+) SVaD group, there were marked correlations between all DTI indices (especially DR and DA) and cortical thickness.
Variation of FA in the anterior corpus callosum and scattered WM regions including centrum semiovale correlated with variation of cortical thickness in the bilateral frontal areas. Variation of DR in extensive WM regions correlated with variation of cortical thickness in the bilateral temporo-parietal, medial frontal, and right frontal areas. Variation of DA in extensive WM regions correlated with variation of cortical thickness in the bilateral frontal and right parietal areas (Fig. ). We report a novel finding regarding correlation between WM microstructural changes and cortical thinning in pure ADD, pure SVaD, and mixed dementia. The first major finding is that in the pure ADD group, disruption of WM integrity was minimal with lower DA in WM adjacent to the cortex. In pure SVaD and mixed dementia, there was extensive disruption of WM integrity with higher DR and overall higher DA, but lower DA in WM adjacent to the cortex. The second major finding is that cortical thinning in pure SVaD strongly correlated with changes in FA and DR, while cortical thinning in mixed dementia strongly correlated with changes in DR and DA.
Taken together, these findings suggest that the relationship between GM and WM degeneration differs according to the underlying pathobiology.In pure ADD, TBSS revealed minimal disruption of WM integrity with lower DA in WM adjacent to the cortex, higher DA in small areas of deep WM, and higher DR in the left frontal WM area. Our finding that DA was especially lower in WM adjacent to the cortex in the pure ADD group is consistent with our hypothesis that Wallerian degeneration following neuronal death in the GM would damage axonal fibers leading to low DA. Meanwhile, we also found higher DA in small areas of deep WM regions as observed in previous studies,. This might be explained by complex pathobiologies: both axonal and myelin loss, subsequent increase in membrane permeability and glial alteration might cause water diffusion in unanticipated directions, leading to an increase in DA.In pure SVaD and mixed dementia, TBSS showed extensive disruption of WM integrity with higher DR and DA.
These results are consistent with previous studies showing that increases in DR and DA were associated with chronic WM degeneration,. Interestingly, we found a unique pattern of DA change according to the distance from the cortex. DA was lower in the WM adjacent to the cortex in the posterior region, while it was higher in the deep WM regions, which might be related to different underlying pathobiologies. Although a similar pattern was found in the pure ADD group, DA was predominantly increased in the pure SVaD and mixed dementia groups. One possible explanation might be related to vulnerability to chronic ischemia. That is, in both pure SVaD and mixed dementia, deep WM regions, including periventricular WM areas, are more vulnerable to chronic ischemia which is represented by WMH on MRI.
Previous pathology studies showed that WMH regions had low myelin content and axonal loss or damage,. This leads to a decrease in tract volume and an increase in the surrounding extracellular fluid, contributing to increased isotropic diffusivity with correspondingly higher DR and DA,. On the other hand, WM regions adjacent to the cortex in pure SVaD and mixed dementia are relatively spared from chronic ischemia,. Lower DA in this region may be due to secondary Wallerian degeneration from combined cortical atrophy. Alternatively, the direction of DA change might reflect the stage of WM degeneration. A previous study showed that DA initially decreased after axonal damage and subsequently increased over time. Therefore, higher DA in deep WM regions might be related to chronic WM degeneration, while lower DA adjacent to cortex might be related to recently developed WM degeneration.sCCA revealed a disease-specific correlation pattern between GM and WM degeneration.
Although the pure ADD group showed cortical thinning and significantly lower FA, there was little correlation between the two. In addition, no correlation was observed between lower DA (as determined by TBSS analysis) and cortical thinning. Although we may not have found significant correlations due to a small sample size, our findings suggest that the overall WM microstructural changes in pure ADD may occur independent of cortical pathology.In the pure SVaD group, lower FA and higher DR (as determined by TBSS analysis) correlated with cortical thinning, while in the mixed dementia group, higher DR and higher DA (as determined by TBSS analysis) correlated with cortical thinning. In both the pure SVaD and mixed dementia groups, the cortical regions that correlated with WM integrity were mainly located in the lateral temporal, inferior parietal, and the lateral and medial frontal areas, which indicated that neurodegeneration in those areas paralleled WM degeneration.
In both groups, chronic ischemia may primarily result in myelin breakdown, which is represented by higher DR. The degeneration process may propagate in a retrograde manner and result in neuronal cell death. Alternatively, disconnection of the cortical-subcortical loop may result in secondary cortical thinning. It is also possible that chronic ischemia may cause GM-WM correlation by simultaneously attacking GM and WM.
However, the fact that different DTI indices correlated with cortical thinning in the pure SVaD and mixed dementia groups suggests that different pathomechanisms may underlie the GM-WM correlation in each disease. A higher vascular burden, such as lacunes, in pure SVaD might result in a more pronounced DR increase and FA decrease when compared to mixed dementia.
In mixed dementia, the presence of an amyloid burden in addition to the chronic ischemic condition may alter the pattern of GM-WM correlation such that axial (DA) or radial (DR) diffusivity, rather than directionality of diffusion (FA), matters more for the association with widespread cortical thinning. Additionally, there was little correlation between lower FA and cortical thinning in mixed dementia patients, which suggested that these might occur independently in mixed dementia patients.There are several limitations in this study. Firstly, we did not conduct PiB-PET scanning on the NC group.
Because previous studies showed that approximately 10 to 30% of cognitively normal subjects are PiB positive, we might have underestimated the degree of WM microstructural changes and cortical thinning in the patient groups. Secondly, the mean ages of the NC and dementia groups were different. However, since the patients were consecutively recruited, we believe this reflected participant characteristics. To minimize the effect of age difference across the groups, we performed the analysis after adjusting for age. In addition, in the GM-WM correlation analysis, we calculated each patient’s W score, which is an age-adjusted Z score relative to the control group,.
Finally, sample size was relatively small and the number of subjects was not even across the four groups. Thus, we might have missed some important information because of low statistical power.In conclusion, we revealed distinct GM-WM relationships in the dementia spectrum of ADD and SVaD (pure ADD, pure SVaD, and mixed dementia). We suggest that the relationship between GM and WM degeneration differs according to the underlying pathobiology of each condition. ParticipantsWe prospectively recruited 139 patients who were clinically diagnosed with SVaD (n = 70) or ADD (n = 69). Patients underwent 11C-PiB-PET scanning and MRI between September 2008 and August 2011.
All SVaD patients fulfilled the Diagnostic and Statistical Manual of Mental Disorders - Fourth Edition (DSM-IV) criteria for vascular dementia and had severe WMH on MRI, which was defined as periventricular WMH (caps or band) ≥10 mm and deep WMH ≥25 mm in maximal diameter according to the modified Fazekas criteria. Patients with WMH due to radiation injury, multiple sclerosis, vasculitis, or leukodystrophy (based on clinical history and other information such as blood test results) were excluded. Among the 70 SVaD patients (47 PiB(−) SVaD and 23 PiB(+) SVaD), 15 patients with PiB(−) SVaD were excluded due to preprocessing errors such as failure in brain mask extraction or registration to a common space.ADD was diagnosed based on the criteria for probable AD proposed by the National Institute of Neurological and Communicative Disorders and Stroke-Alzheimer’s Disease and Related Disorders Association (NINDS-ADRDA).
Among the 69 patients with ADD, we excluded 7 patients who were negative for PiB PET, 14 patients who exceeded mild WMH (periventricular WMH.
This study aimed to discriminate between neuroinflammation and neuronal degeneration in the white matter (WM) and gray matter (GM) of patients with Parkinson’s disease (PD) using free-water (FW) imaging. Analysis using tract-based spatial statistics (TBSS) of 20 patients with PD and 20 healthy individuals revealed changes in FW imaging indices (i.e., reduced FW-corrected fractional anisotropy (FA T), increased FW-corrected mean, axial, and radial diffusivities (MD T, AD T, and RD T, respectively) and fractional volume of FW (FW) in somewhat more specific WM areas compared with the changes of DTI indices. The region-of-interest (ROI) analysis further supported these findings, whereby those with PD showed significantly lower FA T and higher MD T, AD T, and RD T (indices of neuronal degeneration) in anterior WM areas as well as higher FW (index of neuroinflammation) in posterior WM areas compared with the controls.
![Gray Matter Disease Gray Matter Disease](/uploads/1/2/5/7/125745166/190462912.jpg)
Results of GM-based spatial statistics (GBSS) analysis revealed that patients with PD had significantly higher MD T, AD T, and FW than the controls, whereas ROI analysis showed significantly increased MD T and FW and a trend toward increased AD T in GM areas, corresponding to Braak stage IV. These findings support the hypothesis that neuroinflammation precedes neuronal degeneration in PD, whereas WM microstructural alterations precede changes in GM. IntroductionParkinson’s disease (PD) is characterized by widespread aggregation of α-synuclein-immunoreactive inclusions in the form of Lewy pathology.
Lewy pathology was recently demonstrated to trigger reactive microgliosis before nigral degeneration in animal models of PD. Distinguishing between neuroinflammation and neuronal degeneration in vivo may provide a better understanding of the progression of PD pathology.Diffusion tensor imaging (DTI) has been widely used in the evaluation of brains of patients with PD. DTI indices such as fractional anisotropy (FA), mean diffusivity (MD), axial diffusivity (AD), and radial diffusivity (RD) characterize the orientation and distribution of the random movements of water molecules, diffusion magnitude, diffusional directionality perpendicular to the axon, and diffusional directionality along the axon, respectively. Despite their sensitivity, DTI indices are not tissue specific. Furthermore, the assumption of a single-tissue compartment per voxel such that partial volume effect averaging in a voxel from free water (FW) can introduce a bias in the interpretation of DTI indices. In the human brain, FW is present as cerebrospinal fluid (CSF); however, it may also accumulate within the extracellular spaces of the brain parenchyma owing to brain pathologies, such as neuroinflammation, tumor, and trauma ,. Studies involving the use of DTI in patients with PD reported reduced FA , and increased MD ,.
A decrease in FA accompanied by increased MD may be attributed to neuronal degeneration and/or neuroinflammation. Thus, the use of these indices might not be useful in differentiating between these pathologies.Conversely, FW imaging was developed to quantify the contribution of FW and eliminate bias when estimating tissue microstructures and enabled differentiation between alterations in the tissues themselves, such as neuronal degeneration, as measured by FW-corrected DTI indices (FA T, MD T, AD T, and RD T, respectively), and extracellular FW changes, such as neuroinflammation, as measured by the fractional volume of FW. FW imaging is achieved by adopting a two-compartment model and fitting 2 tensors into the diffusion data. In PD, FW imaging has thus far only been used to evaluate the substantia nigra.
FW within the substantia nigra is considered to be a promising biomarker for distinguishing patients with PD from healthy individuals and as a biomarker for disease progression ,.We hypothesize that applying FW correction to DTI data will improve the detection of abnormalities associated with PD in DTI indices. Specifically, the use of FW imaging might allow the discrimination between neuroinflammation and neuronal degeneration in white matter (WM) and gray matter (GM) in PD. To test this hypothesis, we compared 20 patients with PD to 20 control individuals using DTI and FW imaging. SubjectsTwenty patients with PD in Hoehn and Yahr stage 1–2 and 20 age- and sex-matched controls with no history of neurologic or psychiatric disorders and no abnormal signals on structural magnetic resonance imaging (MRI) were included in this retrospective case–control study. Patients with PD were diagnosed by specialists based on the clinical diagnostic criteria for PD by the Movement Disorder Society.
Disease severity was assessed using non-motor and motor scores of the Movement Disorder Society’s Unified Idiopathic PD Rating Scale (UPDRS) parts I and III, respectively. All patients with PD remained free from atypical parkinsonism and exhibited good response (30% in UPDRS part III score with change in treatment or a clearly documented history of marked changes from a reliable patient or caregiver ) to anti-parkinsonian therapy for 18 months or more after the initial diagnosis. At the time of MRI and clinical examinations, all patients were taking levodopa in combination with a dopamine decarboxylase inhibitor (benserazide or carbidopa).
All patients with PD underwent single-photon computed tomography imaging of dopamine transporters and demonstrated deficits in specific binding ratio (less than 95% of the lower limit of prediction intervals for healthy Japanese population). The clinical phenotypes of PD, including tremor-dominant ( n = 6), postural instability/gait difficulty ( n = 8), and intermediate ( n = 6), were assessed using UPDRS part III in all patients with PD. Furthermore, in all patients with PD, rapid eye movement sleep behavior disorder (RBD) ( n = 8) was assessed using the RBD single-question screen (RBD1Q). Summarizes the demographic and clinical characteristics of healthy controls and patients with PD. The ethics committee approved this study, and all participants signed a written informed consent.
HCAll PDRight-Sided Onset PDLeft-Sided Onset PDP Value (HC vs. All PDP Value (R vs.
L Onset PD)P Value (HC vs. R Onset PD)P Value (HC vs. HC, healthy controls; LED, levodopa equivalent dose; L, left; MDS-UPDRS, Movement Disorder Society’s Unified Parkinson’s Disease Rating Scale; PD, Parkinson’s disease; R, right; SBR, specific binding ratio; SD, standard deviation. Statistical analyses were performed using the χ 2 test (°) or unpaired Student’s t-test (.).
MDS-UPDRS part I subscores: (1) cognitive impairment, (2) hallucinations, (3) depression, (4) anxiety, (5) apathy, (6) dopamine dysregulation syndrome, (7) sleep problems, (8) daytime sleepiness, (9) pain, (10) urinary problem, (11) constipation, (12) light headedness on standing, (13) fatigue. Acquisition of MRI DataAll MRI data were acquired using a 3-T scanner (MAGNETOM Prisma; Siemens Healthcare, Erlangen, Germany) with a 64-channel head coil. Whole-brain diffusion-weighted imaging and 3D magnetization-prepared 180° radio-frequency pulses and rapid gradient-echo (MP-RAGE) T1-weighted imaging were obtained for all subjects. Whole-brain diffusion-weighted images were acquired using spin-echo planar imaging with the following parameters: repetition time, 3300 ms; echo time, 70 ms; flip angle, 90°; diffusion gradient directions, 64; b-values, 0 and 1000 s/mm 2; field of view, 229 × 229 mm; matrix size, 130 × 130; resolution, 1.8 × 1.8 mm; slice thickness, 1.6 mm; and acquisition time, 3 min 55 s. 3D MP-RAGE T1-weighted images were acquired using the following parameters: repetition time, 2300 ms; echo time, 2.32 ms; inversion time, 900 ms; field of view, 240 × 240 mm; matrix size, 256 × 256; resolution, 0.9 × 0.9 mm; slice thickness, 0.9 mm; and acquisition time, 6 min 25 s. Diffusion MRI PreprocessingAll diffusion MRI data from 64 different axial, sagittal, and coronal directions were visually checked.
Moreover, all datasets were free from severe artifacts such as gross geometric distortion, signal dropout, or bulk motion. Diffusion MRI data were then corrected for susceptibility-induced geometric distortions, eddy current distortions, and inter-volume subject motion using EDDY and TOPUP toolboxes.Single-tensor FA, MD, AD, and RD maps were generated using the DTIFIT tool implemented in FMRIB Software Library version 5.0.9 (FSL; Oxford Centre for Functional MRI of the Brain, Oxford, UK; ). Meanwhile, an in-house MATLAB (MathWorks, Natick, MA, USA) script was used to fit a regularized bi-tensor model and generate maps for FA T, MD T, AD T, RD T, and FW. A more detailed description of the methods is discussed elsewhere. TBSSWM was analyzed using the skeleton projection step of TBSS using the following steps. First, FA maps of all subjects were aligned to the standard Montreal Neurological Institute (MNI) space (MNI152) with the FMRIB non-linear registration tool. Next, a mean FA image was generated and thinned to create the mean FA skeleton, which represented the centers of all tracts common to the groups.
Next, the threshold of the mean FA skeleton was set to FA 0.20 to include major WM pathways and exclude peripheral tracts and GM. Finally, the aligned FA map of each subject was projected onto the FA skeleton. The same process was applied to other single-tensor DTI (MD, AD, and RD) and bi-tensor FW imaging (FW, FA T, MD T, AD T, and RD T) maps such that the maps were projected onto the mean FA without the initial registration. GBSSGM was analyzed using GBSS , a GM analog of TBSS, using the following steps. First, the Brain Extraction Tool was used to remove non-brain voxels from each subject’s 3D T1-weighted images. Next, each skull-stripped 3D T1-weighted image was affine- and non-linearly aligned to an MNI152 standard space at a 1-mm resolution using the FMRIB linear image registration tool and the FMRIB non-linear registration tool , respectively. Next, field bias was corrected, and GM, WM, and CSF segmentations were obtained using the FMRIB automated segmentation tool.
The resulting GM image was then used to create a median GM skeleton with a threshold of 0.2 to minimize the contribution of voxels from WM and CSF. Next, b0 maps of each subject were affine-aligned to their 3D T1-weighted images (epi-reg). After all maps were affine- and non-linearly aligned into an MNI152 brain common space at a 1-mm resolution , the aligned maps of each subject were projected onto the median GM skeleton map. Region-of-Interest AnalysisWM and GM were further evaluated using automatic region-of-interest (ROI) analyses. Maps showing significant clusters on TBSS and GBSS analyses were localized using Johns Hopkins University’s ICBM-DTI-81 WM labels and tractography atlases and the Desikan–Killiany atlas, respectively. Voxel-Based MorphometryVoxel-based morphometry was used to obtain WM and GM volumetry. First, 3D T1-weighted images were segmented into GM, WM, and CSF through a unified tissue segmentation model using the Statistical Parametric Mapping 12 software (Wellcome Department of Imaging Neuroscience, London, UK; ) running on a MATLAB 2014a platform (MathWorks; ).
Segmented GM and WM images were then spatially normalized to the customized template in the standardized anatomic space using the Diffeomorphic Anatomical Registration Through Exponentiated Lie Algebra (DARTEL) algorithm. To preserve GM and WM volumes within each voxel, Jacobean determinants derived from spatial normalization using DARTEL and an 8 mm full-width at half maximum Gaussian kernel were used to modulate and smooth the images, respectively. Statistical AnalysisAll statistical analyses were performed using IBM SPSS Statistics for Windows, version 22.0 (IBM Corporation, Armonk, NY, USA), except for general linear model analysis, where the FSL was used. The Shapiro–Wilk test was used to assess data normality, whereas demographic data were analyzed using unpaired Student’s t-test and the χ 2 test for continuous and categorical variables, respectively. Statistical significance for all two-tailed tests was set at 0.05.For TBSS and GBSS analyses, a general linear model framework including unpaired Student’s t-test (healthy controls vs.
The graphics of LiEat are so high and seems that it is just like realistic game and attract the players also, the LiEat`s ranking and its marketing is good and had received positive feed backs from regular players.
Download LiEat by clicking the download button below.
![Lieat download Lieat download](http://sikv.web.fc2.com/lieat/img/hal.png)
All patients with PD) and one-way analysis of variance (healthy controls vs. Patients with right-sided PD vs. Those with left-sided PD), with age and sex as covariates, and the randomize tool was used with 5000 permutations to compare all diffusion indices between groups. Results were then corrected for multiple comparisons by controlling for family wise error (FWE) and applying threshold-free cluster enhancement. An FWE-corrected P value of 0.05 was considered to indicate statistical significance. The randomize tool was also applied during voxel-wise correlation analysis of each index with disease duration, UPDRS part I score, UPRDRS part I subscores (cognitive, neuropsychiatric, sleep disorder, sensory and others, and autonomic), or UPDRS part III score.Unpaired Student’s t-test was used to assess differences between the healthy control and PD groups according to the ROI of each metric that showed significant group differences in the TBSS analysis of WM and GM. Bonferroni’s correction was used for multiple comparisons in WM ( n = 2, anterior and posterior) and GM ( n = 3, stages IV, V, and VI) with the level of significance for two-tailed P values set at 0.025 (0.05/2) and 0.017 (0.05/3), respectively.
The effect size was then calculated using Cohen’s d to evaluate the statistical power of the relationship determined during group comparisons. Average diffusion metrics of the WM and GM ROIs were then correlated with disease duration or UPDRS part I score, UPDRS part I subscores (cognitive, neuropsychiatric, sleep disorder, sensory and others, and autonomic), or UPDRS part III score. Considering the exploratory nature of this analysis, Bonferroni correction was not applied.The volumes were compared between the patients with PD and healthy controls using a generalized linear model for analysis of covariance, with age, sex, and total intracranial volume as covariates using the FWE rate set at p = 0.05. TBSSand show the results of the TBSS analysis for DTI (FA, MD, AD, and RD) and FW (FA T, MD T, AD T, RD T, and FW) imaging indices.
The patients with PD exhibited significantly lower FA and FA T and higher MD, MD T, AD, AD T, RD, RD T, and FW ( p. Comparison of DTI (FA, MD, AD, and RD) and FW imaging (FA T, MD T, AD T, RD T, and FW) indices between healthy controls and patients with PD.
TBSS analyses show that patients with PD have significantly ( p. ROIand show the ROI analysis results for DTI (FA, MD, AD, and RD) and FW imaging (FA T, MD T, AD T, RD T, and FW) indices in the anterior and posterior WM tracts. Within the anterior WM tracts, the patients with PD exhibited significantly lower FA and FA T and higher MD, RD, MD T, AD T, and RD T compared with the healthy controls. Moreover, the patients with PD tended to have higher AD compared with the healthy controls.
Within the posterior WM tracts, the patients with PD exhibited significantly lower FA and higher RD and FW compared with the healthy controls. Moreover, the patients with PD tended to have higher MD compared with the healthy controls. Comparison of FW imaging (MD T, AD T, and FW) indices between the healthy controls and the patients with PD. GBSS analyses showed that the patients with PD had significantly higher MD T, AD T, and FW (red-yellow voxels) compared with the healthy controls ( p. GBSSand show the GBSS analysis results of MD T, AD T, and FW. The patients with PD exhibited significantly higher MD T, AD T, and FW ( p. ( A) Region-of-interest analyses of the anterior (ACR, ATR, and forceps minor) and posterior (PCR, PTR, and forceps major) white matter areas.
( B) Mean values for DTI (FA, MD, AD, and RD) and FW imaging (FA T, MD T, AD T, RD T, and FW) indices of the anterior and posterior white matter areas in healthy controls (white bars) and patients with PD (gray bars). ACR, anterior corona radiata; AD, axial diffusivity, AD T, FW-corrected axial diffusivity; ATR, anterior thalamic radiation; DTI, diffusion tensor imaging; FA, fractional anisotropy; FA T, free water-corrected fractional anisotropy; FW, free water; HC, healthy controls; MD, mean diffusivity; MD T, free water-corrected mean diffusivity; PCR, posterior corona radiata; PD, Parkinson’s disease; PTR, posterior thalamic radiation; RD, radial diffusivity; RD T, free water-corrected radial diffusivity. ModalityContrastCluster SizeAnatomical RegionPeak t-ValuePeak MNI Coordinates ( X, Y, Z)MD THC. ROIand show the ROI analysis results for MD T, AD T, and FW in GM areas corresponding to Braak stages IV–VI.
The patients with PD demonstrated significantly higher MD T and FW and tended to have higher AD T compared with the healthy controls in the areas corresponding to Braak stage IV. Among the Braak stage V areas, the patients with PD tended to have higher MD T, AD T, and FW compared with the healthy controls. No significant differences in GM were found between the patients with PD and healthy controls within the areas corresponding to Braak stage VI.
DiscussionIn the present study, we investigated WM and GM alterations by comparing the patients with PD to the healthy controls using DTI and FW images. In the TBSS analysis, the changes in the FW imaging indices (increased FW and reduced FA T and increased MD T, AD T, and RD T) were in somewhat more specific WM areas compared with the changes in DTI indices (reduced FA and increased MD, AD, and RD). These findings are further supported by our ROI analysis to specifically analyze the WM tracts running through the anterior and posterior portions of the brain. The significantly reduced FA T and the significantly increased MD T, AD T, and RD T were observed only in the anterior WM tracts with a higher effect size than the DTI indices; the increased FW was observed only in the posterior WM tracts of the patients with PD. Furthermore, in the GBSS analysis, increases in MD T, AD T, and FW were observed in the GM of the patients with PD, whereas no significant differences were found in any of the DTI indices.
The ROI analysis demonstrated that the changes in these measures corresponded with Braak stage IV.In the WM, the observed increase in increased FW, which is expected during neuroinflammation , is in line with the reported inflammatory processes related to the activation of astrocytes and microglia, which are linked to α-synuclein aggregation, in PD. Furthermore, the reduced FA T and increased MD T, AD T, and RD T, which usually result from axonal damage and demyelination , might be the result of the aggregation of Lewy neurites that are associated with impaired axonal transport with subsequent microstructural changes in the axon and surrounding myelin. The current study also demonstrated that the FW indices were able to derive more precise estimations of localized WM degeneration in PD compared with the DTI indices. An explanation might be that partialling out FW eliminated the influence of CSF on WM tracts running closely adjacent to the ventricles and brain pathologies such as neuroinflammation, thereby increasing the specificity of FW imaging indices ,. Thus, the current results suggest that the changes in the DTI indices within the posterior WM tracts of the patients with PD were largely influenced by neuroinflammation.
In line with the current results, a study on patients with schizophrenia reported changes in DTI indices in specific areas after excluding the influence of extracellular FW. Furthermore, a study recently demonstrated the correlation between FW obtained using FW imaging and the 18-kDa translocator protein (TSPO), using positron emission tomography with 11CDPA-713 that binds TSPO, in patients with traumatic brain injury.Anterior brain has been suggested to be more prone to Lewy pathology than posterior brain ,. The prefrontal cortex is an area where Lewy pathology occurs at a relatively early stage of PD.
Additionally, PD is considered to be a prion-like disorder where misfolded α-synuclein spreads from one neuron to another from the anterior to the posterior along WM. Given that most of the current study patients (85%) were in Hoehn and Yahr stage 2, it is possible that, at the time of the examination, the Lewy neurites had progressed from the anterior portions to reach the posterior portions of the brain, with simultaneous neurodegeneration induced by neuroinflammation in the anterior portions. Taken together, the current results demonstrated that the microstructural changes in the WM of patients with PD were preceded by neuroinflammation and followed by neurodegeneration.
These findings are supported by the previously demonstrated exacerbation of dopaminergic neurodegeneration in the substantia nigra pars compacta by the chronic release of pro-inflammatory cytokines. Moreover, animal models of PD demonstrated that Lewy pathology triggered reactive microgliosis prior to nigral degeneration.Increased MD T, AD T, and FW, which are indicative of axonal degeneration and neuroinflammation , in the GM of patients with PD, were demonstrated using GBSS analysis.
The lack of significant differences in the DTI indices in GM suggests that the FW imaging indices provided greater sensitivity in the detection of GM abnormalities in the patients with PD. DTI is not the preferred method for the evaluation of GM, especially the cortex, other than the substantia nigra and striatum, mainly because of its inability to thoroughly describe microstructural abnormalities in GM due to water diffusion isotropy. Furthermore, the use of diffusion MRI for the evaluation of GM, especially cortical areas, has been limited by the partial volume effect of the WM and CSF adjacent to the cortical GM. FW imaging removes the isotropic FW compartment in the GM, thus leading to a more anisotropic estimation of the single tensor in one voxel. In the present study, the partial volume effect of the WM and CSF was further minimized using GBSS analysis.
In line with a previous study , the current results demonstrated that GBSS analysis enables the regional characterization of GM pathology in PD because GBSS analysis involves the aggregation of diffusion MRI indices in regions surrounding the skeleton created at the center of the cortical GM, thus minimizing the partial volume effect.Unlike the widespread microstructural changes observed in WM, the GBSS results identified significant changes within more limited GM areas in the patients with PD. Furthermore, no significant difference in FA T of the GM was observed between the patients with PD and the healthy controls. Decreases in GM FA, an index of neuron integrity, was previously reported to appear later than increases in MD during the progression of PD.
Taken together, the current results indicate that WM microstructural changes in PD occur early and may precede changes in GM. Our findings agree with those of previous studies that reported widespread changes in WM microstructure in patients with early PD with no or limited GM alteration ,.
These results are also consistent with a recent hypothesis regarding pathological progression in PD, wherein pre-synaptic terminal damage, impaired axonal transport, and/or altered axonal structure precede cell body damage, otherwise known as the “dying back” pattern of degeneration. In addition, the changes in FW imaging measures were also demonstrated in the thalamus areas. Previous histopathological and imaging studies demonstrated the involvement of thalamus in PD patients with RBD ,. Thus, the microstructural changes of the thalamus might be related to RBD in patients with PD.Furthermore, the ROI analyses in the present study indicated that the most striking difference in the extent and the distribution of GM damage, specifically, neuroinflammation (indexed by FW) and axonal degeneration (indexed by MD T and AD T) , corresponded with Braak stage IV. At Braak stage III, patients with PD exhibit typical features described in Hoehn and Yahr stage 1 (i.e., tremors, rigidity, and bradykinesia) typically on one side of the body.
However, as the disease enters Braak stage IV, the clinical features become bilateral (Hoehn and Yahr stage 2). Considering that 85% of the participants in the present study were in Hoehn and Yahr stage 2, the study findings agree with the neuropathological and clinical patterns of PD.
Additionally, among the changes in MD T, AD T, and FW, FW showed the strongest effect size. This finding, along with that for WM, also suggest that neuroinflammation may be preceding axonal degeneration in the GM of patients with PD.No differences in the WM and GM volumes were observed between the patients with PD and healthy controls. The current study results support the conclusions from several studies wherein no volume changes were detected in early PD , suggesting that volumetric alterations in brain structures may occur later in the disease course, as described in studies on advanced PD. This finding also lends further support for the suggestion that diffusion MRI indices are potentially more sensitive as biomarkers for detecting microstructural changes in PD prior to tissue loss detected during volumetric MRI.None of the indices included herein correlated with disease duration. This finding is consistent with those of histopathological studies showing that the aggregation of Lewy pathology occurs before the appearance of symptoms. However, medication effects may also explain the observed lack of correlation with the scores of UPDRS parts I and III. and Wen et al.
previously reported a correlation between DTI indices and the scores of UPDRS part III in the brains of untreated patients with PD. In contrast, we and another group showed a lack of such correlation in the brains of treated patients with PD. Furthermore, one study demonstrated that the UPDRS lacks the sensitivity to discriminate the severity of symptoms of early PD.Some limitations of the present study should be noted. First, the small sample size limits the statistical power, which may have led to the nonsignificant results in some measures including correlation analysis.
Second, this was a case–control study. Third, although the patients with PD in the current study were defined to be Hoehn and Yahr stage 1–2, most of the patients were in Hoehn and Yahr stage 2. Further studies including patients with preclinical or earlier-stage PD and longitudinal studies involving larger cohorts will be particularly informative in clarifying the utility of FW indices as biomarkers for diagnosis and disease progression and in resolving the temporal sequence of events of neuroinflammation and neurodegeneration. Fourth, the current study lacked histopathological verification. Future studies should also include TSPO-positron emission tomography imaging to verify neuroinflammation in PD. Fifth, in the current study, we did not categorize the patients with PD based on the clinical PD phenotypes such as non-tremor and tremor-dominant subtypes, whereas motor phenotypes in PD are recognized to have a distinct neural basis , which should be considered for further investigation as well. Sixth, we did not specifically evaluate patients with PD based on clinical symptoms such as RBD or hyposmia.
Additionally, polysomnography and the odor stick identification test , objective assessments that confirm the clinical diagnosis of RBD and hyposmia, respectively, were not performed in the patients with PD. However, considering that both symptoms appear early and are common in PD, further studies should be conducted with objective evaluation focusing on these symptoms. Studies investigating microstructural changes in patients with RBD, de novo PD, and PD with RBD, using FW imaging will also be crucial for a better understanding of the pathology underlying PD. Finally, TBSS and GBSS lack the sensitivity to detect peripheral effects located outside the skeleton, and the skeleton projection step may also introduce bias to the projected parameters. ConclusionsThe current study results provide novel evidence that FW imaging in PD may be useful in determining the etiology of microstructural changes, more specifically related to neurodegeneration and neuroinflammation. Furthermore, FW imaging provided more precise estimations of localized microstructural changes in PD compared with DTI.
These findings also demonstrated that neuroinflammation preceded neurodegeneration in PD, whereas changes in the WM microstructure preceded those in the GM. However, considering the limitations of the current study, the findings should be clinically interpreted with caution. Particularly, longitudinal studies and histopathological verification are necessary to validate the current study findings. Was supported by Brain/MINDS program from the Japan Agency for Medical Research and Development (AMED); JSPS KAKENHI Grant Number 18H02772 and JP16H06280; and a Grant-in-Aid for Special Research in Subsidies for ordinary expenses of private schools from The Promotion and Mutual Aid Corporation for Private Schools of Japan.
Was supported by Brain/MINDS Beyond program from AMED (JP19dm0307024); JSPS KAKENHI (19K17244). Was supported by a National Health & Medical Research Council (NHMRC) Senior Principal Research Fellowship (1105825). Was supported by the Australian National Health and Medical Research Council (NHMRC) Senior Research Fellowship B (1136649).
![](/uploads/1/2/5/7/125745166/516490207.jpg)