G Model EURR-6857; No. of Pages 6
ARTICLE IN PRESS European Journal of Radiology xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
European Journal of Radiology journal homepage: www.elsevier.com/locate/ejrad
Diffusion tensor imaging differentiates vascular parkinsonism from parkinsonian syndromes of degenerative origin in elderly subjects Jérémy Deverdun a,b,c,1 , Sophie Menjot de Champfleur a,d,1 , Simon Cabello-Aguilar a,c , Florence Maury e , Franc¸ois Molino b,f , Mahmoud Charif e , Nicolas Leboucq a , Xavier Ayrignac e , Pierre Labauge e , Alain Bonafe a,c,g , Giovanni Castelnovo h , Emmanuelle Le Bars a,c , Christian Geny e,i,j , Nicolas Menjot de Champfleur a,c,f,∗ a
Department of Neuroradiology, Montpellier University Hospital Center, Gui de Chauliac Hospital, Montpellier, France Laboratoire Charles Coulomb, CNRS UMR 5221 - Université Montpellier II, Montpellier, France c I2FH, Institut d’Imagerie Fonctionnelle Humaine, Hôpital Gui de Chauliac, CHRU de, Montpellier, France d Clinique du Parc, Castelnau-le-Lez, France e Department of Neurology, Montpellier University Hospital Center, Gui de Chauliac Hospital, Montpellier, France f Institut de Génomique Fonctionnelle, UMR 5203 - INSERM U661 - Université Montpellier II - Université, Montpellier I, France g Team “Plasticity of Central Nervous System, Stem Cells and Glial Tumors”, U1051, Institut of Neurosciences of Montpellier, Saint Eloi Hospital, Montpellier, France h Department of Neurology, Caremeau University Hospital Center, Nîmes, France i EuroMov, 700 Avenue du Pic Saint Loup - 34090, Montpellier, France j Movement to Health (M2H), Montpellier-1 University, France b
a r t i c l e
i n f o
Article history: Received 16 April 2014 Received in revised form 30 June 2014 Accepted 11 July 2014
a b s t r a c t Background and Purpose: The etiologic diagnosis of parkinsonian syndromes is of particular importance when considering syndromes of vascular or degenerative origin. The purpose of this study is to find differences in the white-matter architecture between those two groups in elderly patients. Materials and Methods: Thirty-five patients were prospectively included (multiple-system atrophy, n = 5; Parkinson’s disease, n = 15; progressive supranuclear palsy, n = 9; vascular parkinsonism, n = 6), with a mean age of 76 years. Patients with multiple-system atrophy, progressive supranuclear palsy and Parkinson’s disease were grouped as having parkinsonian syndromes of degenerative origin. Brain MRIs included diffusion tensor imaging. Fractional anisotropy and mean-diffusivity maps were spatially normalized, and group analyses between parkinsonian syndromes of degenerative origin and vascular parkinsonism were performed using a voxel-based approach. Results: Statistical parametric-mapping analysis of diffusion tensor imaging data showed decreased fractional anisotropy value in internal capsules bilaterally in patients with vascular parkinsonism compared to parkinsonian syndromes of degenerative origin (p = 0.001) and showed a lower mean diffusivity in the white matter of the left superior parietal lobule (p = 0.01). Fractional anisotropy values were found decreased in the middle cerebellar peduncles in multiplesystem atrophy compared to Parkinson’s disease and progressive supranuclear palsy. The mean diffusivity was increased in those regions for these subgroups. Conclusion: Clinically defined vascular parkinsonism was associated with decreased fractional anisotropy in the deep white matter (internal capsules) compared to parkinsonian syndromes of degenerative origin. These findings are consistent with previously published neuropathological data. © 2014 Elsevier Ireland Ltd. All rights reserved.
Abbreviations: DeP, Parkinsonian syndromes of degenerative origin; DTI, diffusion tensor imaging; FA, fractional anisotropy; FLAIR, fluid-attenuated inversion recovery; MD, mean diffusivity; MSA, multiple-system atrophy; PD, Parkinson’s disease; PSP, progressive supranuclear palsy; VP, vascular parkinsonism; WM, white matter. ∗ Corresponding author at: Department of Neuroradiology, University Hospital Center, Gui de Chauliac Hospital, 80 Avenue Augustin Fliche, 34295 Montpellier Cedex 5, France. Tel.: +33 4 67 33 78 87; fax: +33 4 67 33 68 38; mobile: +33 6 64 93 10 25. E-mail address: nicolasdechampfl[email protected] (N. Menjot de Champfleur). 1 These authors contributed equally to this work. http://dx.doi.org/10.1016/j.ejrad.2014.07.012 0720-048X/© 2014 Elsevier Ireland Ltd. All rights reserved.
Please cite this article in press as: Deverdun J, et al. Diffusion tensor imaging differentiates vascular parkinsonism from parkinsonian syndromes of degenerative origin in elderly subjects. Eur J Radiol (2014), http://dx.doi.org/10.1016/j.ejrad.2014.07.012
G Model EURR-6857; No. of Pages 6
ARTICLE IN PRESS J. Deverdun et al. / European Journal of Radiology xxx (2014) xxx–xxx
2
1. Introduction Parkinson’s disease (PD) is the most common, but not the only, cause of parkinsonian syndromes. This diagnosis is specially challenging in elderly subjects due to the frequency in this population of other causes of parkinsonian syndromes [1], comorbidities and the heterogeneity of clinical expression and evolution. Other parkinsonian syndromes could be separated into degenerative, toxic or vascular syndromes and include progressive supranuclear palsy (PSP), parkinsonian multiple-system atrophy (MSA), corticobasal degeneration syndrome, dementia with Lewy’s bodies, and secondary causes (i.e., toxic origins). Parkinsonian syndromes of degenerative origin (DeP) comprise PD, PSP and MSA. The differentiation of PD from other parkinsonian syndromes is of particular importance because their progression, prognosis and treatment might be very different. In elderly subjects, vascular parkinsonism (VP) may be diagnosed based on Zijlman’s criteria. VP accounts for 4.4% to 12% of all cases of parkinsonism [2]. VP is typically associated with macroscopic infarcts in the basal ganglia or microscopic small-vessel disease, leading to changes in deep white matter (WM) [3]. Computerized tomography and MRI support the concept of VP [4], and neuroimaging changes appear to be correlated with changes seen on post-mortem examination [5]. There are no clear biomarkers for differential diagnosis between VP and DeP, and SPECT imaging [6] and conventional MRI are often insufficient, although MRI morphometric and volumetric studies of the brain seem promising [7]. Diffusion-weighted imaging and especially diffusion tensor imaging (DTI) permit the quantification of brain-water movements and allow a specific and indirect examination of structural changes in the WM. Diffusion tensor imaging is differently restricted in various tissues; in WM, for example, the directions of diffusion are more limited. Anisotropy represents the limited directionality of diffusion and is higher in WM than in the less organized gray matter. This difference allows the calculation of FA and MD values for the entire cerebral parenchyma and the generation of parametric maps. Fractional anisotropy (FA) ranges from 0 to 1, where 0 is isotropic diffusion (no directional organization) and 1 is anisotropic diffusion (well-organized tissues like WM). Mean diffusivity (MD) is a measure of the average molecular motion, independent of tissue directionality. It is modified by cellular size and integrity [8]. Applications of DTI have been increasing in recent years because it may help qualify and quantify structural changes in cerebral WM with greater sensitivity than the usual visual evaluation of WM hyperintensities used with conventional MRI data [9]. DTI parameters are altered in PD [10], and DTI may distinguish DePs [11,12]. Recent works also show that VP is associated with architectural abnormalities in WM that are visible via DTI [13]. To the best of our knowledge, DTI parameters have not been compared between VP and other DePs. The main goal of this study is to compare DTI-derived indices between elderly subjects with DeP and VP. In parallel, we will also investigate the use of DTI to distinguish between the different subgroups of DeP, comparing our findings to the published literature.
were not usable for one subject. We thus have a total of 35 subjects in the MRI study. Fig. 1 summarizes the study participation. 2.2. Clinical evaluation Diagnoses were verified by a neurologist experienced with parkinsonian syndromes (C.G.) according to established guidelines: the UK Parkinson’s Disease Society Brain Bank criteria for idiopathic PD, the National Institute of Neurological Disorders and Stroke (NINDS) and the Society for Progressive Supranuclear Palsy (SPSP) criteria for PSP, Gilman’s criteria for MSA, and Zijlmans’s criteria for VP. All patients were examined within 2 months before the MRI exam. Clinical data included age, disease duration, and age at onset. Neurological impairment was assessed using the Hoehn and Yahr scale and MDS-UPDRS. No patient had a history of head injury, stroke, intra-cerebral bleeding, exposure to neuroleptic drugs before onset of symptoms, or psychiatric comorbidity. According to the previously listed criteria, patients were classified in 4 subgroups: sixteen patients met the criteria for idiopathic PD (10 males/6 females; mean age 74.3 ± 7.8 y-o); 5 patients met the criteria for MSA (5 males; mean age 73 ± 7.6 y-o); 11 patients met the criteria for PSP (6 males/5 females; mean age 75.3 ± 4.2 yo); and 7 patients met the criteria for VP (2 males/5 females; mean age 82.6 ± 5 years). Because PD, MSA and PSP are of neurodegenerative etiology, i.e., are characterized by neuronal loss and brain atrophy, we grouped them as parkinsonian syndromes of degenerative origin to clearly differentiate them from parkinsonian syndromes of vascular origin. 2.3. Neuropsychological evaluation Neuropsychological evaluation was performed to rule out dementia with Lewy bodies within 2 months before the MRI exam. The evaluation was composed of the Mattis Dementia-Rating Scale, the Grober and Buschke verbal-learning test, the Span Task, the Stroop Word-Color Test, the Trail-Making Test, the Rey-Osterrieth Complex-Figure Test and a semantic-processing task (LEXIS test). It also included the Frontal Assessment Battery, which estimates frontal subcortical-dysfunction syndrome, the Isaac test for verbal fluency and the clock-drawing test to assess visuospatial activities. Finally, educational attainment (EA) was recorded. Global cognitive evaluation was assessed using the Mini Mental-Status Examination. 2.4. MRI acquisitions Brain MRIs were performed on a 1.5T magnet (Siemens Avanto, Erlangen, Germany). The acquisition protocol contained an anatomic 3D T1 weighted sequence, a fluid-attenuated inversion recovery (FLAIR) acquisition, susceptibility-weighted imaging (repetition time = 62 ms, echo time = 12.26 ms and 52.65 ms, flip angle = 15◦ , slice thickness = 2 mm) and a DTI acquisition (64 directions, 55 slices, thickness = 2.5 mm, repetition time = 6700 ms, echo time = 82 ms, number of excitations = 1, voxel: 2.5 × 2.5 × 2.5 mm, b = 1000 s/mm2 ).
2. Materials and methods 2.5. Data analysis 2.1. Population Forty-two patients with late-onset parkinsonian syndromes without dementia were prospectively recruited for this study, and each individual gave informed consent. Three patients were excluded because of an uncertain diagnosis. Thirty-nine patients (23 males/16 females; mean age 76 years, range 56–90 years) were included, but 3 could not continue with the MRI and the MRI data
2.5.1. Morphologic analysis 2.5.1.1. Susceptibility-weighted imaging analysis. One boardcertified neuroradiologist (S.M.C.) independently counted the number of cerebral microbleeds on susceptibility-weighted imaging and was blinded to individual diagnoses. The total number of cerebral microbleeds was compared between groups (DeP and VP) and subgroups (PSP, MSA, DP and VP).
Please cite this article in press as: Deverdun J, et al. Diffusion tensor imaging differentiates vascular parkinsonism from parkinsonian syndromes of degenerative origin in elderly subjects. Eur J Radiol (2014), http://dx.doi.org/10.1016/j.ejrad.2014.07.012
G Model EURR-6857; No. of Pages 6
ARTICLE IN PRESS J. Deverdun et al. / European Journal of Radiology xxx (2014) xxx–xxx
3
Fig. 1. Study flow diagram.
2.5.1.2. FLAIR analysis. The age-related white-matter-changes rating scale of the European Task Force was used as a general measure of the severity of cerebral WM hyperintensities [14]. In short, this scale uses a four-point severity rating (0, normal; 1, punctate; 2, beginning confluence; 3, diffuse involvement).
2.5.2. Diffusion tensor imaging data analysis Diffusion tensor imaging data were first corrected for distortions due to eddy currents using FSL software (http://www.fmrib.ox. ac.uk/fsl/index.html). Then, FA and MD parametric maps were calculated for each subject. SPM software was then used (SPM8, Statistical Parametric Mapping, Wellcome Department of Cognitive Neurology, Institute of Neurology, London, UK; http://www.fil.ion.ucl.ac.uk/spm/, Matlab 2010b,The MathWorks, Natick, MA, USA). All volumes were reoriented according to the anterior commissure position. Using a customized DARTEL template, FA and MD maps were normalized for each subject. Data were spatially smoothed with a 6-mm full width at half maximum isotropic Gaussian kernel.
3. Results 3.1. Population The four subgroups showed a subtle difference in age (p = 0.05, Kruskal–Wallis test; PSP = 75.3 y-o ± 4.2 y-o; PD = 74.3 y-o ± 7.8 y-o; MSA = 73.0 y-o ± 7.6 y-o; VP = 82.6 y-o ± 5.0 y-o); this difference was less pronounced between the pooled DeP and VP groups (p = 0.07, Kruskal–Wallis test). 3.2. Clinical and neuropsychological evaluation There were no statistically significant differences in clinical score between subgroups (Table 1). 3.3. Morphological analysis We found no statistically significant difference in the global number of cerebral microbleeds between VP and DeP or between the subgroups of DeP. We also found no statistically significant difference in the presence of age-related white-matter changes between VP and DeP or between the subgroups of DeP.
2.6. Statistical analysis 3.4. DTI analysis Diffusion tensor imaging data analysis used a voxel-based approach (voxel-based diffusion tensor imaging) to compare FA and MD values between the VP and DeP groups (and the subgroups) using a two-sided Student’s t-test. We used a voxel-wise threshold of p < 0.01 and a cluster-size threshold of 80 for the cluster extent. Multiple-comparisons correction was applied if possible using the false-discovery-rate procedure (p < 0.05 with cluster level >80). Group comparisons for quantitative parameters were performed with the Kruskal–Wallis non-parametric test and the Wilcoxon rank-sum test with the Holm correction for multiple comparisons.
Both the right and left internal capsules (IC) and the corona radiata showed significantly lower FA values in VP compared to the DeP groups (p < 0.001 with false-discovery-rate-procedure correction and cluster-extent threshold = 350 voxels; Fig. 2; Table 2). No significant differences in MD values were observed in the IC, but in the left superior parietal lobule, decreased MD values were observed in the VP group compared to the DeP group (p < 0.001; Fig. 3; Table 2). Differences were observed between MSA and the PD and PSP subgroups. These differences were significant in the middle cerebral peduncles and pons. FA values were lower (p < 0.01) and MD values
Please cite this article in press as: Deverdun J, et al. Diffusion tensor imaging differentiates vascular parkinsonism from parkinsonian syndromes of degenerative origin in elderly subjects. Eur J Radiol (2014), http://dx.doi.org/10.1016/j.ejrad.2014.07.012
G Model
ARTICLE IN PRESS
EURR-6857; No. of Pages 6
J. Deverdun et al. / European Journal of Radiology xxx (2014) xxx–xxx
4
Table 1 Demographic data and clinical scores for each group of subjects. MSA Mean age ± SD (y-o) Gender (F/M) Disease duration ± SD (years) HY MMSE FIVE WORDS CLOCK ISAAC TEST FAB TEST
73.0 ± 7.6 5 (0/5) 4±2 3±1 25 ± 5 9±1 6±2 33 11 ± 3
PD
PSP
74.3 ± 7.8 16(6/10) 7±7 3±1 25 ± 4 9±1 6±2 32 ± 5 14 ± 3
75.3 ± 4.2 11(5/6) 4±3 3±1 23 ± 5 9±1 4±3 22 ± 5 11 ± 5
VaP
p values
82.6 ± 5.0 7(5/2) 3±2 3±1 25 ± 2 10 ± 1 5±3 29 ± 7 14 ± 3
0.0485 0.523332 0.6461 0.6635 0.8109 0.3658 0.02881 0.4296
MMSE, Mini Mental-State Examination; HY, Hoehn and Yahr scale, SD, standard deviation; F, female; M, male.
Fig. 2. Normalized average axial FA images showing, as an overlay, significant clusters of reduced FA in patients with VP compared to DeP. Clusters were considered significant at p < 0.001 (false-discovery-rate-procedure correction, cluster-extent threshold = 350 voxels). FA is decreased bilaterally in internal capsules and the corona radiata. The images are in MNI space.
were higher (p < 0.01) in the former, but no differences were found between PSP and PD. 4. Discussion The principal aim of this study was to identify DTI biomarkers that can be used to differentiate between DeP and VP in elderly patients. We used an explorative voxel-based approach on DTI images obtained from 35 patients. We identified in VP patients, when compared to DeP patients, 1) a significant decrease in FA values in the deep WM in both internal capsules and the corona radiate; 2) a decreased MD value in the left superior parietal lobule; and 3) no significant difference between the two groups in white-matter hyperintensities. We are aware that a limit of this study is the unbalanced size of the DeP (29 patients) and VP (6 patients) groups. This is due to the difficulty in diagnosing unambiguously purely vascular parkinsonian syndromes. Nevertheless to our knowledge, this work is the first voxel-based diffusion tensor-imaging study comparing VP and DeP in elderly
patients, and our findings suggest that patients with VP and DeP should be treated as distinct groups. The literature involving the application of DTI to parkinsonian syndromes is mainly concerned with comparisons between DeP subgroups. In summary, FA appears to be decreased in PSP patients in several WM tracts (i.e., superior longitudinal fasciculus, arcuate fasciculus, corpus callosum, internal capsule, and posterior thalamic radiations) [15] and in both superior cerebellar peduncles [16]. In MSA patients, FA is decreased and MD is increased in the middle cerebellar peduncles [16]. Mean diffusivity was increased in the putamens, globus pallidus, and caudate nucleus [17] in PSP compared to PD. The cerebellar WM showed decreased FA or increased MD in PD compared to healthy volunteers [18]. WM abnormalities were recently reported with the use of DTI in VP compared to healthy subjects, with a reduced FA in the left thalamus, right frontal subcortical WM, and left anterior limb of the internal capsule and an increased MD in the bilateral frontal subcortical WM [13]. Our results could be attributed to the subtle but statistically significant age difference between the VP and the DeP groups.
Table 2 Peaks for significant clusters of decreased FA and increased MD in patients with VP compared to DeP. Hemisphere
Decreased FA in VP vs DeP
Decreased MD in VP vs DeP
Location
Cluster size
a
p-value
L
• Internal capsule (posterior limb) •Corona radiata
877
0.001 (FDR correction)
R
• Internal capsule •Corona radiata
717a
0.001 (FDR correction)
L
• Superior parietal lobule
90a
0.038 (uncorrected)
Z score
MNI coordinates X
Y
Z
4.06 3.73 3.58 3.83 3.60 3.57
−20 −28 −25 32 28 25
−12 5 25 −25 22 28
10 22 12 22 12 −5
3.87 3.75
−28 −20
−50 −50
65 60
FA, fractional anisotropy; MD, mean diffusivity; DeP, parkinsonian syndromes of degenerative origin; VP, vascular parkinsonism; FDR, multiple-comparisons correction using the false-discovery-rate method. a All significant maxima are part of a single large cluster.
Please cite this article in press as: Deverdun J, et al. Diffusion tensor imaging differentiates vascular parkinsonism from parkinsonian syndromes of degenerative origin in elderly subjects. Eur J Radiol (2014), http://dx.doi.org/10.1016/j.ejrad.2014.07.012
G Model EURR-6857; No. of Pages 6
ARTICLE IN PRESS J. Deverdun et al. / European Journal of Radiology xxx (2014) xxx–xxx
5
statistically significant with respect to other subgroups, despite previous modifications demonstrated in the substantia nigra [21]. Finally, in contrast to previous observations [18], we did not observe any significant decrease in MD in PD patients, possibly because we studied an older population with longer disease durations or because of small-sample-size effects. We used susceptibility-weighted imaging to assess cerebral microbleeds. This technique is now known to be more sensitive to lesions containing deoxyhemoglobin and hemosiderin [22] than the usual T2 Gradient Recalled Echo sequence (T2*). We found no statistically significant difference in the global number of cerebral microbleeds between groups or subgroups. Cerebral microbleeds are common in the elderly, and their prevalence increases with age, reaching 40% prevalence in patients over 80 years old [23]. Once again, the absence of any significant difference in the number of cerebral microbleeds between VP and DeP suggests that our population is homogeneous in terms of age-related signal abnormalities. 5. Conclusion Fig. 3. Axial (a and b) and frontal (c and d) normalized average MD images showing as an overlay significant clusters of reduced MD in patients with VP compared to DeP. Clusters were considered significant at p < 0.001 (uncorrected, cluster-extent threshold = 90 voxels). MD is decreased in the left superior parietal lobule in VP compared to DeP. The images are in MNI space.
Nevertheless, the absence of any significant difference between VP and DeP by FLAIR analysis on the age-related white-matter-changes rating scale shows that our population is statistically homogeneous in terms of age-related signal abnormalities. Indeed, previous studies of VP subjects using imaging techniques and histopathological examination showed specific signal abnormalities in the subcortical white or gray matter. Vascular parkinsonism is now clearly linked to subcortical white- or gray-matter lesions [4] identified in neuroimaging studies that correlate with histopathologic findings of arteriosclerosis, demyelination and lacunar infarcts [19]. Subcortical lesions in VP may involve the gray nuclei (putamen and thalamus predominantly) and the deep WM, especially in watershed areas; those deep WM lesions are related to pallor areas in the frontal and occipital WM [5]. The local reduction in FA observed in internal capsules and the corona radiata is likely to reflect the specific micro-structural disturbance of WM pathways that carry information to the cerebellum and to the basal ganglia [3]. These WM lesions may explain the disorganization of pathways observed in DTI, but we observe a DTI signature in the VP versus DeP subjects, even in the absence of any significant difference in lesion structure on FLAIR. This difference hints that there may be a specific DTI biomarker of the VP group that was not previously reported. DTI analysis revealed a lower FA and higher MD in the middle cerebellar peduncles bilaterally in MSA patients compared to PD and PSP patients’. This finding is consistent with previous studies that showed decreased FA values in this group in the right cerebellum and in the middle cerebellar peduncles [16]. These modifications in anisotropy were previously described in comparisons of MSA to healthy subjects by a region-of-interest approach [20] and can be explained by physiopathology, which leads to cerebellar dysfunction. Dysfunction is caused by degenerative changes in the olivopontocerebellar system, assuming that the loss of FA is parallel to neuronal changes in the brain. Our purely data-driven method supports these published results. FA and MD changes in the putamen between MSA and PD patients are still controversial, and we did not find any significant changes in these parameters using a data-driven approach. FA changes in PD patients were not
We conducted a prospective study to determine the clinical utility of DTI in the differentiation of VP and DeP (including PD, PSP and MSA), focusing on elderly subjects. Few studies have been conducted in this population because of the difficulties in establishing age-matched parkinsonian groups. We found significantly decreased anisotropy in the deep WM in clinically defined VP compared to DeP. The positive diagnosis of vascular parkinsonism remains an important challenge in clinical practice. This work is part of a more general effort to develop an accurate etiological diagnosis of parkinsonian syndromes, to optimize therapy through the use of VP biomarkers. These diagnostic ambiguities are particularly significant for elderly subjects due to the multiple symptoms of “Parkinson’s disease”, which are often linked to aging effects or to other neurological pathologies (e.g., normal-pressure hydrocephalus). Considering the aging population, these diagnostic difficulties are being encountered more frequently, necessitating the development of new biomarkers like that proposed here. References [1] Benito-Leon J, Bermejo-Pareja F, Rodriguez J, et al. Prevalence of PD and other types of parkinsonism in three elderly populations of central Spain. Mov Disord 2003;18(3):267–74. [2] Benito-Leon J, Bermejo-Pareja F, Morales-González JM, et al. Incidence of Parkinson disease and parkinsonism in three elderly populations of central Spain. Neurology 2004;62(5):734–41. [3] Zijlmans JC, Daniel SE, Hughes AJ, Revesz T, Lees AJ. Clinicopathological investigation of vascular parkinsonism, including clinical criteria for diagnosis. Mov Disord 2004;19(6):630–40. [4] Demirkiran M, Bozdemir H, Sarica Y. Vascular parkinsonism: a distinct, heterogeneous clinical entity. Acta Neurol Scand 2001;104(2):63–7. [5] Yamanouchi H, Nagura H. Neurological signs and frontal white matter lesions in vascular parkinsonism. A clinicopathologic study. Stroke 1997;28(5):965–9. [6] Aarsland D, Svenningsson P. Using biomarkers to disentangle different causes of Parkinsonism. J Neurol Neurosurg Psychiatr 2013;84(2):119. [7] Choi S-M, Kim BC, Nam T-S, et al. Midbrain atrophy in vascular Parkinsonism. Eur Neurol 2011;65(5):296–301. [8] Basser PJ, Pierpaoli C. Microstructural and physiological features of tissues elucidated by quantitative-diffusion-tensor MRI. J Magn Reson Ser B 1996;111(3):209–19. [9] Della Nave R, Foresti S, Pratesi A, et al. Whole-brain histogram and voxelbased analyses of diffusion tensor imaging in patients with leukoaraiosis: correlation with motor and cognitive impairment. AJNR Am J Neuroradiol 2007;28(7):1313–9. [10] Kim D-E, Bae H-J, Lee S-H, Kim H, Yoon B-W, Roh J-K. Gradient echo magnetic resonance imaging in the prediction of hemorrhagic vs ischemic stroke: a need for the consideration of the extent of leukoariosis. Archiv Neurol 2002;59(3):425–9. [11] Focke NK, Helms G, Pantel PM, et al. Differentiation of typical and atypical Parkinson syndromes by quantitative MR imaging. AJNR Am J Neuroradiol 2011;32(11):2087–92.
Please cite this article in press as: Deverdun J, et al. Diffusion tensor imaging differentiates vascular parkinsonism from parkinsonian syndromes of degenerative origin in elderly subjects. Eur J Radiol (2014), http://dx.doi.org/10.1016/j.ejrad.2014.07.012
G Model EURR-6857; No. of Pages 6 6
ARTICLE IN PRESS J. Deverdun et al. / European Journal of Radiology xxx (2014) xxx–xxx
[12] Paviour D, Thornton J, Lees A, Jager H. Diffusion-weighted magnetic resonance imaging differentiates Parkinsonian variant of multiple-system atrophy from progressive supranuclear palsy. Movement Disorders: Official J Movement Disorder Soc 2007;22(1):68–74. [13] Wang H-C, Hsu J-L, Leemans A. Diffusion tensor imaging of vascular parkinsonism: structural changes in cerebral white matter and the association with clinical severity. Archiv Neurol 2012:1–9. [14] Wahlund LO, Barkhof F, Fazekas F, et al. A new rating scale for age-related white matter changes applicable to MRI and CT. Stroke 2001;32(6):1318–22. [15] Padovani A, Borroni B, Brambati SM, et al. Diffusion tensor imaging and voxel based morphometry study in early progressive supranuclear palsy. J Neurol Neurosurg Psychiatr 2006;77(4):457–63. [16] Blain CR, Barker GJ, Jarosz JM, et al. Measuring brain stem and cerebellar damage in parkinsonian syndromes using diffusion tensor MRI. Neurology 2006;67(12):2199–205. [17] Seppi K, Schocke MF, Esterhammer R, et al. Diffusion-weighted imaging discriminates progressive supranuclear palsy from PD, but not from the parkinson variant of multiple system atrophy. Neurology 2003;60(6):922–7.
[18] Zhang K, Yu C, Zhang Y, et al. Voxel-based analysis of diffusion tensor indices in the brain in patients with Parkinson’s disease. Eur J Radiol 2011;77(2): 269–73. [19] van Swieten JC, van den Hout JH, van Ketel BA, Hijdra A, Wokke JH, van Gijn J. Periventricular lesions in the white matter on magnetic resonance imaging in the elderly. A morphometric correlation with arteriolosclerosis and dilated perivascular spaces. Brain: J Neurol 1991;114(Pt 2):761–74. [20] Shiga K, Yamada K, Yoshikawa K, Mizuno T, Nishimura T, Nakagawa M. Local tissue anisotropy decreases in cerebellopetal fibers and pyramidal tract in multiple system atrophy. J Neurol 2005;252(5):589–96. [21] Chan L, Rumpel H, Yap K, et al. Case control study of diffusion tensor imaging in Parkinson’s disease. J Neurol Neurosurg Psychiatr 2007;78(12):1383–6. [22] menjot n, Langlois C, Ankenbrandt WJ, et al. MRI evaluation of cerebral cavernous malformations with susceptibility-weighted imaging. Neurosurgery 2010. [23] Poels MMF, Vernooij MW, Ikram MA, et al. Prevalence and risk factors of cerebral microbleeds: an update of the Rotterdam scan study. Stroke 2010;41(10 Suppl):S103–6.
Please cite this article in press as: Deverdun J, et al. Diffusion tensor imaging differentiates vascular parkinsonism from parkinsonian syndromes of degenerative origin in elderly subjects. Eur J Radiol (2014), http://dx.doi.org/10.1016/j.ejrad.2014.07.012
ARTICLE IN PRESS European Journal of Radiology xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
European Journal of Radiology journal homepage: www.elsevier.com/locate/ejrad
Diffusion tensor imaging differentiates vascular parkinsonism from parkinsonian syndromes of degenerative origin in elderly subjects Jérémy Deverdun a,b,c,1 , Sophie Menjot de Champfleur a,d,1 , Simon Cabello-Aguilar a,c , Florence Maury e , Franc¸ois Molino b,f , Mahmoud Charif e , Nicolas Leboucq a , Xavier Ayrignac e , Pierre Labauge e , Alain Bonafe a,c,g , Giovanni Castelnovo h , Emmanuelle Le Bars a,c , Christian Geny e,i,j , Nicolas Menjot de Champfleur a,c,f,∗ a
Department of Neuroradiology, Montpellier University Hospital Center, Gui de Chauliac Hospital, Montpellier, France Laboratoire Charles Coulomb, CNRS UMR 5221 - Université Montpellier II, Montpellier, France c I2FH, Institut d’Imagerie Fonctionnelle Humaine, Hôpital Gui de Chauliac, CHRU de, Montpellier, France d Clinique du Parc, Castelnau-le-Lez, France e Department of Neurology, Montpellier University Hospital Center, Gui de Chauliac Hospital, Montpellier, France f Institut de Génomique Fonctionnelle, UMR 5203 - INSERM U661 - Université Montpellier II - Université, Montpellier I, France g Team “Plasticity of Central Nervous System, Stem Cells and Glial Tumors”, U1051, Institut of Neurosciences of Montpellier, Saint Eloi Hospital, Montpellier, France h Department of Neurology, Caremeau University Hospital Center, Nîmes, France i EuroMov, 700 Avenue du Pic Saint Loup - 34090, Montpellier, France j Movement to Health (M2H), Montpellier-1 University, France b
a r t i c l e
i n f o
Article history: Received 16 April 2014 Received in revised form 30 June 2014 Accepted 11 July 2014
a b s t r a c t Background and Purpose: The etiologic diagnosis of parkinsonian syndromes is of particular importance when considering syndromes of vascular or degenerative origin. The purpose of this study is to find differences in the white-matter architecture between those two groups in elderly patients. Materials and Methods: Thirty-five patients were prospectively included (multiple-system atrophy, n = 5; Parkinson’s disease, n = 15; progressive supranuclear palsy, n = 9; vascular parkinsonism, n = 6), with a mean age of 76 years. Patients with multiple-system atrophy, progressive supranuclear palsy and Parkinson’s disease were grouped as having parkinsonian syndromes of degenerative origin. Brain MRIs included diffusion tensor imaging. Fractional anisotropy and mean-diffusivity maps were spatially normalized, and group analyses between parkinsonian syndromes of degenerative origin and vascular parkinsonism were performed using a voxel-based approach. Results: Statistical parametric-mapping analysis of diffusion tensor imaging data showed decreased fractional anisotropy value in internal capsules bilaterally in patients with vascular parkinsonism compared to parkinsonian syndromes of degenerative origin (p = 0.001) and showed a lower mean diffusivity in the white matter of the left superior parietal lobule (p = 0.01). Fractional anisotropy values were found decreased in the middle cerebellar peduncles in multiplesystem atrophy compared to Parkinson’s disease and progressive supranuclear palsy. The mean diffusivity was increased in those regions for these subgroups. Conclusion: Clinically defined vascular parkinsonism was associated with decreased fractional anisotropy in the deep white matter (internal capsules) compared to parkinsonian syndromes of degenerative origin. These findings are consistent with previously published neuropathological data. © 2014 Elsevier Ireland Ltd. All rights reserved.
Abbreviations: DeP, Parkinsonian syndromes of degenerative origin; DTI, diffusion tensor imaging; FA, fractional anisotropy; FLAIR, fluid-attenuated inversion recovery; MD, mean diffusivity; MSA, multiple-system atrophy; PD, Parkinson’s disease; PSP, progressive supranuclear palsy; VP, vascular parkinsonism; WM, white matter. ∗ Corresponding author at: Department of Neuroradiology, University Hospital Center, Gui de Chauliac Hospital, 80 Avenue Augustin Fliche, 34295 Montpellier Cedex 5, France. Tel.: +33 4 67 33 78 87; fax: +33 4 67 33 68 38; mobile: +33 6 64 93 10 25. E-mail address: nicolasdechampfl[email protected] (N. Menjot de Champfleur). 1 These authors contributed equally to this work. http://dx.doi.org/10.1016/j.ejrad.2014.07.012 0720-048X/© 2014 Elsevier Ireland Ltd. All rights reserved.
Please cite this article in press as: Deverdun J, et al. Diffusion tensor imaging differentiates vascular parkinsonism from parkinsonian syndromes of degenerative origin in elderly subjects. Eur J Radiol (2014), http://dx.doi.org/10.1016/j.ejrad.2014.07.012
G Model EURR-6857; No. of Pages 6
ARTICLE IN PRESS J. Deverdun et al. / European Journal of Radiology xxx (2014) xxx–xxx
2
1. Introduction Parkinson’s disease (PD) is the most common, but not the only, cause of parkinsonian syndromes. This diagnosis is specially challenging in elderly subjects due to the frequency in this population of other causes of parkinsonian syndromes [1], comorbidities and the heterogeneity of clinical expression and evolution. Other parkinsonian syndromes could be separated into degenerative, toxic or vascular syndromes and include progressive supranuclear palsy (PSP), parkinsonian multiple-system atrophy (MSA), corticobasal degeneration syndrome, dementia with Lewy’s bodies, and secondary causes (i.e., toxic origins). Parkinsonian syndromes of degenerative origin (DeP) comprise PD, PSP and MSA. The differentiation of PD from other parkinsonian syndromes is of particular importance because their progression, prognosis and treatment might be very different. In elderly subjects, vascular parkinsonism (VP) may be diagnosed based on Zijlman’s criteria. VP accounts for 4.4% to 12% of all cases of parkinsonism [2]. VP is typically associated with macroscopic infarcts in the basal ganglia or microscopic small-vessel disease, leading to changes in deep white matter (WM) [3]. Computerized tomography and MRI support the concept of VP [4], and neuroimaging changes appear to be correlated with changes seen on post-mortem examination [5]. There are no clear biomarkers for differential diagnosis between VP and DeP, and SPECT imaging [6] and conventional MRI are often insufficient, although MRI morphometric and volumetric studies of the brain seem promising [7]. Diffusion-weighted imaging and especially diffusion tensor imaging (DTI) permit the quantification of brain-water movements and allow a specific and indirect examination of structural changes in the WM. Diffusion tensor imaging is differently restricted in various tissues; in WM, for example, the directions of diffusion are more limited. Anisotropy represents the limited directionality of diffusion and is higher in WM than in the less organized gray matter. This difference allows the calculation of FA and MD values for the entire cerebral parenchyma and the generation of parametric maps. Fractional anisotropy (FA) ranges from 0 to 1, where 0 is isotropic diffusion (no directional organization) and 1 is anisotropic diffusion (well-organized tissues like WM). Mean diffusivity (MD) is a measure of the average molecular motion, independent of tissue directionality. It is modified by cellular size and integrity [8]. Applications of DTI have been increasing in recent years because it may help qualify and quantify structural changes in cerebral WM with greater sensitivity than the usual visual evaluation of WM hyperintensities used with conventional MRI data [9]. DTI parameters are altered in PD [10], and DTI may distinguish DePs [11,12]. Recent works also show that VP is associated with architectural abnormalities in WM that are visible via DTI [13]. To the best of our knowledge, DTI parameters have not been compared between VP and other DePs. The main goal of this study is to compare DTI-derived indices between elderly subjects with DeP and VP. In parallel, we will also investigate the use of DTI to distinguish between the different subgroups of DeP, comparing our findings to the published literature.
were not usable for one subject. We thus have a total of 35 subjects in the MRI study. Fig. 1 summarizes the study participation. 2.2. Clinical evaluation Diagnoses were verified by a neurologist experienced with parkinsonian syndromes (C.G.) according to established guidelines: the UK Parkinson’s Disease Society Brain Bank criteria for idiopathic PD, the National Institute of Neurological Disorders and Stroke (NINDS) and the Society for Progressive Supranuclear Palsy (SPSP) criteria for PSP, Gilman’s criteria for MSA, and Zijlmans’s criteria for VP. All patients were examined within 2 months before the MRI exam. Clinical data included age, disease duration, and age at onset. Neurological impairment was assessed using the Hoehn and Yahr scale and MDS-UPDRS. No patient had a history of head injury, stroke, intra-cerebral bleeding, exposure to neuroleptic drugs before onset of symptoms, or psychiatric comorbidity. According to the previously listed criteria, patients were classified in 4 subgroups: sixteen patients met the criteria for idiopathic PD (10 males/6 females; mean age 74.3 ± 7.8 y-o); 5 patients met the criteria for MSA (5 males; mean age 73 ± 7.6 y-o); 11 patients met the criteria for PSP (6 males/5 females; mean age 75.3 ± 4.2 yo); and 7 patients met the criteria for VP (2 males/5 females; mean age 82.6 ± 5 years). Because PD, MSA and PSP are of neurodegenerative etiology, i.e., are characterized by neuronal loss and brain atrophy, we grouped them as parkinsonian syndromes of degenerative origin to clearly differentiate them from parkinsonian syndromes of vascular origin. 2.3. Neuropsychological evaluation Neuropsychological evaluation was performed to rule out dementia with Lewy bodies within 2 months before the MRI exam. The evaluation was composed of the Mattis Dementia-Rating Scale, the Grober and Buschke verbal-learning test, the Span Task, the Stroop Word-Color Test, the Trail-Making Test, the Rey-Osterrieth Complex-Figure Test and a semantic-processing task (LEXIS test). It also included the Frontal Assessment Battery, which estimates frontal subcortical-dysfunction syndrome, the Isaac test for verbal fluency and the clock-drawing test to assess visuospatial activities. Finally, educational attainment (EA) was recorded. Global cognitive evaluation was assessed using the Mini Mental-Status Examination. 2.4. MRI acquisitions Brain MRIs were performed on a 1.5T magnet (Siemens Avanto, Erlangen, Germany). The acquisition protocol contained an anatomic 3D T1 weighted sequence, a fluid-attenuated inversion recovery (FLAIR) acquisition, susceptibility-weighted imaging (repetition time = 62 ms, echo time = 12.26 ms and 52.65 ms, flip angle = 15◦ , slice thickness = 2 mm) and a DTI acquisition (64 directions, 55 slices, thickness = 2.5 mm, repetition time = 6700 ms, echo time = 82 ms, number of excitations = 1, voxel: 2.5 × 2.5 × 2.5 mm, b = 1000 s/mm2 ).
2. Materials and methods 2.5. Data analysis 2.1. Population Forty-two patients with late-onset parkinsonian syndromes without dementia were prospectively recruited for this study, and each individual gave informed consent. Three patients were excluded because of an uncertain diagnosis. Thirty-nine patients (23 males/16 females; mean age 76 years, range 56–90 years) were included, but 3 could not continue with the MRI and the MRI data
2.5.1. Morphologic analysis 2.5.1.1. Susceptibility-weighted imaging analysis. One boardcertified neuroradiologist (S.M.C.) independently counted the number of cerebral microbleeds on susceptibility-weighted imaging and was blinded to individual diagnoses. The total number of cerebral microbleeds was compared between groups (DeP and VP) and subgroups (PSP, MSA, DP and VP).
Please cite this article in press as: Deverdun J, et al. Diffusion tensor imaging differentiates vascular parkinsonism from parkinsonian syndromes of degenerative origin in elderly subjects. Eur J Radiol (2014), http://dx.doi.org/10.1016/j.ejrad.2014.07.012
G Model EURR-6857; No. of Pages 6
ARTICLE IN PRESS J. Deverdun et al. / European Journal of Radiology xxx (2014) xxx–xxx
3
Fig. 1. Study flow diagram.
2.5.1.2. FLAIR analysis. The age-related white-matter-changes rating scale of the European Task Force was used as a general measure of the severity of cerebral WM hyperintensities [14]. In short, this scale uses a four-point severity rating (0, normal; 1, punctate; 2, beginning confluence; 3, diffuse involvement).
2.5.2. Diffusion tensor imaging data analysis Diffusion tensor imaging data were first corrected for distortions due to eddy currents using FSL software (http://www.fmrib.ox. ac.uk/fsl/index.html). Then, FA and MD parametric maps were calculated for each subject. SPM software was then used (SPM8, Statistical Parametric Mapping, Wellcome Department of Cognitive Neurology, Institute of Neurology, London, UK; http://www.fil.ion.ucl.ac.uk/spm/, Matlab 2010b,The MathWorks, Natick, MA, USA). All volumes were reoriented according to the anterior commissure position. Using a customized DARTEL template, FA and MD maps were normalized for each subject. Data were spatially smoothed with a 6-mm full width at half maximum isotropic Gaussian kernel.
3. Results 3.1. Population The four subgroups showed a subtle difference in age (p = 0.05, Kruskal–Wallis test; PSP = 75.3 y-o ± 4.2 y-o; PD = 74.3 y-o ± 7.8 y-o; MSA = 73.0 y-o ± 7.6 y-o; VP = 82.6 y-o ± 5.0 y-o); this difference was less pronounced between the pooled DeP and VP groups (p = 0.07, Kruskal–Wallis test). 3.2. Clinical and neuropsychological evaluation There were no statistically significant differences in clinical score between subgroups (Table 1). 3.3. Morphological analysis We found no statistically significant difference in the global number of cerebral microbleeds between VP and DeP or between the subgroups of DeP. We also found no statistically significant difference in the presence of age-related white-matter changes between VP and DeP or between the subgroups of DeP.
2.6. Statistical analysis 3.4. DTI analysis Diffusion tensor imaging data analysis used a voxel-based approach (voxel-based diffusion tensor imaging) to compare FA and MD values between the VP and DeP groups (and the subgroups) using a two-sided Student’s t-test. We used a voxel-wise threshold of p < 0.01 and a cluster-size threshold of 80 for the cluster extent. Multiple-comparisons correction was applied if possible using the false-discovery-rate procedure (p < 0.05 with cluster level >80). Group comparisons for quantitative parameters were performed with the Kruskal–Wallis non-parametric test and the Wilcoxon rank-sum test with the Holm correction for multiple comparisons.
Both the right and left internal capsules (IC) and the corona radiata showed significantly lower FA values in VP compared to the DeP groups (p < 0.001 with false-discovery-rate-procedure correction and cluster-extent threshold = 350 voxels; Fig. 2; Table 2). No significant differences in MD values were observed in the IC, but in the left superior parietal lobule, decreased MD values were observed in the VP group compared to the DeP group (p < 0.001; Fig. 3; Table 2). Differences were observed between MSA and the PD and PSP subgroups. These differences were significant in the middle cerebral peduncles and pons. FA values were lower (p < 0.01) and MD values
Please cite this article in press as: Deverdun J, et al. Diffusion tensor imaging differentiates vascular parkinsonism from parkinsonian syndromes of degenerative origin in elderly subjects. Eur J Radiol (2014), http://dx.doi.org/10.1016/j.ejrad.2014.07.012
G Model
ARTICLE IN PRESS
EURR-6857; No. of Pages 6
J. Deverdun et al. / European Journal of Radiology xxx (2014) xxx–xxx
4
Table 1 Demographic data and clinical scores for each group of subjects. MSA Mean age ± SD (y-o) Gender (F/M) Disease duration ± SD (years) HY MMSE FIVE WORDS CLOCK ISAAC TEST FAB TEST
73.0 ± 7.6 5 (0/5) 4±2 3±1 25 ± 5 9±1 6±2 33 11 ± 3
PD
PSP
74.3 ± 7.8 16(6/10) 7±7 3±1 25 ± 4 9±1 6±2 32 ± 5 14 ± 3
75.3 ± 4.2 11(5/6) 4±3 3±1 23 ± 5 9±1 4±3 22 ± 5 11 ± 5
VaP
p values
82.6 ± 5.0 7(5/2) 3±2 3±1 25 ± 2 10 ± 1 5±3 29 ± 7 14 ± 3
0.0485 0.523332 0.6461 0.6635 0.8109 0.3658 0.02881 0.4296
MMSE, Mini Mental-State Examination; HY, Hoehn and Yahr scale, SD, standard deviation; F, female; M, male.
Fig. 2. Normalized average axial FA images showing, as an overlay, significant clusters of reduced FA in patients with VP compared to DeP. Clusters were considered significant at p < 0.001 (false-discovery-rate-procedure correction, cluster-extent threshold = 350 voxels). FA is decreased bilaterally in internal capsules and the corona radiata. The images are in MNI space.
were higher (p < 0.01) in the former, but no differences were found between PSP and PD. 4. Discussion The principal aim of this study was to identify DTI biomarkers that can be used to differentiate between DeP and VP in elderly patients. We used an explorative voxel-based approach on DTI images obtained from 35 patients. We identified in VP patients, when compared to DeP patients, 1) a significant decrease in FA values in the deep WM in both internal capsules and the corona radiate; 2) a decreased MD value in the left superior parietal lobule; and 3) no significant difference between the two groups in white-matter hyperintensities. We are aware that a limit of this study is the unbalanced size of the DeP (29 patients) and VP (6 patients) groups. This is due to the difficulty in diagnosing unambiguously purely vascular parkinsonian syndromes. Nevertheless to our knowledge, this work is the first voxel-based diffusion tensor-imaging study comparing VP and DeP in elderly
patients, and our findings suggest that patients with VP and DeP should be treated as distinct groups. The literature involving the application of DTI to parkinsonian syndromes is mainly concerned with comparisons between DeP subgroups. In summary, FA appears to be decreased in PSP patients in several WM tracts (i.e., superior longitudinal fasciculus, arcuate fasciculus, corpus callosum, internal capsule, and posterior thalamic radiations) [15] and in both superior cerebellar peduncles [16]. In MSA patients, FA is decreased and MD is increased in the middle cerebellar peduncles [16]. Mean diffusivity was increased in the putamens, globus pallidus, and caudate nucleus [17] in PSP compared to PD. The cerebellar WM showed decreased FA or increased MD in PD compared to healthy volunteers [18]. WM abnormalities were recently reported with the use of DTI in VP compared to healthy subjects, with a reduced FA in the left thalamus, right frontal subcortical WM, and left anterior limb of the internal capsule and an increased MD in the bilateral frontal subcortical WM [13]. Our results could be attributed to the subtle but statistically significant age difference between the VP and the DeP groups.
Table 2 Peaks for significant clusters of decreased FA and increased MD in patients with VP compared to DeP. Hemisphere
Decreased FA in VP vs DeP
Decreased MD in VP vs DeP
Location
Cluster size
a
p-value
L
• Internal capsule (posterior limb) •Corona radiata
877
0.001 (FDR correction)
R
• Internal capsule •Corona radiata
717a
0.001 (FDR correction)
L
• Superior parietal lobule
90a
0.038 (uncorrected)
Z score
MNI coordinates X
Y
Z
4.06 3.73 3.58 3.83 3.60 3.57
−20 −28 −25 32 28 25
−12 5 25 −25 22 28
10 22 12 22 12 −5
3.87 3.75
−28 −20
−50 −50
65 60
FA, fractional anisotropy; MD, mean diffusivity; DeP, parkinsonian syndromes of degenerative origin; VP, vascular parkinsonism; FDR, multiple-comparisons correction using the false-discovery-rate method. a All significant maxima are part of a single large cluster.
Please cite this article in press as: Deverdun J, et al. Diffusion tensor imaging differentiates vascular parkinsonism from parkinsonian syndromes of degenerative origin in elderly subjects. Eur J Radiol (2014), http://dx.doi.org/10.1016/j.ejrad.2014.07.012
G Model EURR-6857; No. of Pages 6
ARTICLE IN PRESS J. Deverdun et al. / European Journal of Radiology xxx (2014) xxx–xxx
5
statistically significant with respect to other subgroups, despite previous modifications demonstrated in the substantia nigra [21]. Finally, in contrast to previous observations [18], we did not observe any significant decrease in MD in PD patients, possibly because we studied an older population with longer disease durations or because of small-sample-size effects. We used susceptibility-weighted imaging to assess cerebral microbleeds. This technique is now known to be more sensitive to lesions containing deoxyhemoglobin and hemosiderin [22] than the usual T2 Gradient Recalled Echo sequence (T2*). We found no statistically significant difference in the global number of cerebral microbleeds between groups or subgroups. Cerebral microbleeds are common in the elderly, and their prevalence increases with age, reaching 40% prevalence in patients over 80 years old [23]. Once again, the absence of any significant difference in the number of cerebral microbleeds between VP and DeP suggests that our population is homogeneous in terms of age-related signal abnormalities. 5. Conclusion Fig. 3. Axial (a and b) and frontal (c and d) normalized average MD images showing as an overlay significant clusters of reduced MD in patients with VP compared to DeP. Clusters were considered significant at p < 0.001 (uncorrected, cluster-extent threshold = 90 voxels). MD is decreased in the left superior parietal lobule in VP compared to DeP. The images are in MNI space.
Nevertheless, the absence of any significant difference between VP and DeP by FLAIR analysis on the age-related white-matter-changes rating scale shows that our population is statistically homogeneous in terms of age-related signal abnormalities. Indeed, previous studies of VP subjects using imaging techniques and histopathological examination showed specific signal abnormalities in the subcortical white or gray matter. Vascular parkinsonism is now clearly linked to subcortical white- or gray-matter lesions [4] identified in neuroimaging studies that correlate with histopathologic findings of arteriosclerosis, demyelination and lacunar infarcts [19]. Subcortical lesions in VP may involve the gray nuclei (putamen and thalamus predominantly) and the deep WM, especially in watershed areas; those deep WM lesions are related to pallor areas in the frontal and occipital WM [5]. The local reduction in FA observed in internal capsules and the corona radiata is likely to reflect the specific micro-structural disturbance of WM pathways that carry information to the cerebellum and to the basal ganglia [3]. These WM lesions may explain the disorganization of pathways observed in DTI, but we observe a DTI signature in the VP versus DeP subjects, even in the absence of any significant difference in lesion structure on FLAIR. This difference hints that there may be a specific DTI biomarker of the VP group that was not previously reported. DTI analysis revealed a lower FA and higher MD in the middle cerebellar peduncles bilaterally in MSA patients compared to PD and PSP patients’. This finding is consistent with previous studies that showed decreased FA values in this group in the right cerebellum and in the middle cerebellar peduncles [16]. These modifications in anisotropy were previously described in comparisons of MSA to healthy subjects by a region-of-interest approach [20] and can be explained by physiopathology, which leads to cerebellar dysfunction. Dysfunction is caused by degenerative changes in the olivopontocerebellar system, assuming that the loss of FA is parallel to neuronal changes in the brain. Our purely data-driven method supports these published results. FA and MD changes in the putamen between MSA and PD patients are still controversial, and we did not find any significant changes in these parameters using a data-driven approach. FA changes in PD patients were not
We conducted a prospective study to determine the clinical utility of DTI in the differentiation of VP and DeP (including PD, PSP and MSA), focusing on elderly subjects. Few studies have been conducted in this population because of the difficulties in establishing age-matched parkinsonian groups. We found significantly decreased anisotropy in the deep WM in clinically defined VP compared to DeP. The positive diagnosis of vascular parkinsonism remains an important challenge in clinical practice. This work is part of a more general effort to develop an accurate etiological diagnosis of parkinsonian syndromes, to optimize therapy through the use of VP biomarkers. These diagnostic ambiguities are particularly significant for elderly subjects due to the multiple symptoms of “Parkinson’s disease”, which are often linked to aging effects or to other neurological pathologies (e.g., normal-pressure hydrocephalus). Considering the aging population, these diagnostic difficulties are being encountered more frequently, necessitating the development of new biomarkers like that proposed here. References [1] Benito-Leon J, Bermejo-Pareja F, Rodriguez J, et al. Prevalence of PD and other types of parkinsonism in three elderly populations of central Spain. Mov Disord 2003;18(3):267–74. [2] Benito-Leon J, Bermejo-Pareja F, Morales-González JM, et al. Incidence of Parkinson disease and parkinsonism in three elderly populations of central Spain. Neurology 2004;62(5):734–41. [3] Zijlmans JC, Daniel SE, Hughes AJ, Revesz T, Lees AJ. Clinicopathological investigation of vascular parkinsonism, including clinical criteria for diagnosis. Mov Disord 2004;19(6):630–40. [4] Demirkiran M, Bozdemir H, Sarica Y. Vascular parkinsonism: a distinct, heterogeneous clinical entity. Acta Neurol Scand 2001;104(2):63–7. [5] Yamanouchi H, Nagura H. Neurological signs and frontal white matter lesions in vascular parkinsonism. A clinicopathologic study. Stroke 1997;28(5):965–9. [6] Aarsland D, Svenningsson P. Using biomarkers to disentangle different causes of Parkinsonism. J Neurol Neurosurg Psychiatr 2013;84(2):119. [7] Choi S-M, Kim BC, Nam T-S, et al. Midbrain atrophy in vascular Parkinsonism. Eur Neurol 2011;65(5):296–301. [8] Basser PJ, Pierpaoli C. Microstructural and physiological features of tissues elucidated by quantitative-diffusion-tensor MRI. J Magn Reson Ser B 1996;111(3):209–19. [9] Della Nave R, Foresti S, Pratesi A, et al. Whole-brain histogram and voxelbased analyses of diffusion tensor imaging in patients with leukoaraiosis: correlation with motor and cognitive impairment. AJNR Am J Neuroradiol 2007;28(7):1313–9. [10] Kim D-E, Bae H-J, Lee S-H, Kim H, Yoon B-W, Roh J-K. Gradient echo magnetic resonance imaging in the prediction of hemorrhagic vs ischemic stroke: a need for the consideration of the extent of leukoariosis. Archiv Neurol 2002;59(3):425–9. [11] Focke NK, Helms G, Pantel PM, et al. Differentiation of typical and atypical Parkinson syndromes by quantitative MR imaging. AJNR Am J Neuroradiol 2011;32(11):2087–92.
Please cite this article in press as: Deverdun J, et al. Diffusion tensor imaging differentiates vascular parkinsonism from parkinsonian syndromes of degenerative origin in elderly subjects. Eur J Radiol (2014), http://dx.doi.org/10.1016/j.ejrad.2014.07.012
G Model EURR-6857; No. of Pages 6 6
ARTICLE IN PRESS J. Deverdun et al. / European Journal of Radiology xxx (2014) xxx–xxx
[12] Paviour D, Thornton J, Lees A, Jager H. Diffusion-weighted magnetic resonance imaging differentiates Parkinsonian variant of multiple-system atrophy from progressive supranuclear palsy. Movement Disorders: Official J Movement Disorder Soc 2007;22(1):68–74. [13] Wang H-C, Hsu J-L, Leemans A. Diffusion tensor imaging of vascular parkinsonism: structural changes in cerebral white matter and the association with clinical severity. Archiv Neurol 2012:1–9. [14] Wahlund LO, Barkhof F, Fazekas F, et al. A new rating scale for age-related white matter changes applicable to MRI and CT. Stroke 2001;32(6):1318–22. [15] Padovani A, Borroni B, Brambati SM, et al. Diffusion tensor imaging and voxel based morphometry study in early progressive supranuclear palsy. J Neurol Neurosurg Psychiatr 2006;77(4):457–63. [16] Blain CR, Barker GJ, Jarosz JM, et al. Measuring brain stem and cerebellar damage in parkinsonian syndromes using diffusion tensor MRI. Neurology 2006;67(12):2199–205. [17] Seppi K, Schocke MF, Esterhammer R, et al. Diffusion-weighted imaging discriminates progressive supranuclear palsy from PD, but not from the parkinson variant of multiple system atrophy. Neurology 2003;60(6):922–7.
[18] Zhang K, Yu C, Zhang Y, et al. Voxel-based analysis of diffusion tensor indices in the brain in patients with Parkinson’s disease. Eur J Radiol 2011;77(2): 269–73. [19] van Swieten JC, van den Hout JH, van Ketel BA, Hijdra A, Wokke JH, van Gijn J. Periventricular lesions in the white matter on magnetic resonance imaging in the elderly. A morphometric correlation with arteriolosclerosis and dilated perivascular spaces. Brain: J Neurol 1991;114(Pt 2):761–74. [20] Shiga K, Yamada K, Yoshikawa K, Mizuno T, Nishimura T, Nakagawa M. Local tissue anisotropy decreases in cerebellopetal fibers and pyramidal tract in multiple system atrophy. J Neurol 2005;252(5):589–96. [21] Chan L, Rumpel H, Yap K, et al. Case control study of diffusion tensor imaging in Parkinson’s disease. J Neurol Neurosurg Psychiatr 2007;78(12):1383–6. [22] menjot n, Langlois C, Ankenbrandt WJ, et al. MRI evaluation of cerebral cavernous malformations with susceptibility-weighted imaging. Neurosurgery 2010. [23] Poels MMF, Vernooij MW, Ikram MA, et al. Prevalence and risk factors of cerebral microbleeds: an update of the Rotterdam scan study. Stroke 2010;41(10 Suppl):S103–6.
Please cite this article in press as: Deverdun J, et al. Diffusion tensor imaging differentiates vascular parkinsonism from parkinsonian syndromes of degenerative origin in elderly subjects. Eur J Radiol (2014), http://dx.doi.org/10.1016/j.ejrad.2014.07.012
