Clinical Imaging 39 (2015) 20–25
Contents lists available at ScienceDirect
Clinical Imaging journal homepage: http://www.clinicalimaging.org
Gray matter alteration in patients with restless legs syndrome: a voxel-based morphometry study Yongmin Chang a, 1, Hyuk Won Chang b, 1, Huijin Song c, Jeonghun Ku d, Christopher J. Earley f, Richard P. Allen f, Yong Won Cho e,⁎ a
Department of Molecular Medicine, Kyungpook National University and Hospital, Daegu, Republic of Korea Department of Radiology, Dongsan Medical Center, Keimyung University School of Medicine, Daegu, Republic of Korea Department of Medical & Biological Engineering, Kyungpook National University and Hospital, Daegu, Republic of Korea d Department of Biomedical Engineering, Dongsan Medical Center, Keimyung University School of Medicine, Daegu, Republic of Korea e Department of Neurology, Dongsan Medical Center, Keimyung University School of Medicine, Daegu, Republic of Korea f Department of Neurology, Johns Hopkins University, Hopkins Bayview Medical Center, Baltimore, MD, USA b c
a r t i c l e
i n f o
Article history: Received 26 March 2014 Received in revised form 27 June 2014 Accepted 22 July 2014 Keywords: Restless legs syndrome MRI Voxel-based morphometry Gray matter
a b s t r a c t The purpose of this study was to demonstrate whether or not restless legs syndrome (RLS) is associated with any morphological change in gray matter. Forty-six RLS subjects and 46 controls were enrolled. We performed voxel-based morphometry analysis and compared the results of the two groups. The RLS subjects showed significant regional decreases of gray matter volume in the left hippocampal gyrus, both parietal lobes, medial frontal areas and cerebellum (uncorrected, Pb .001). We found that RLS patients showed structural alteration in the brain and alterations in certain parts of the brain in RLS patients are relevant to RLS. © 2015 Elsevier Inc. All rights reserved.
1. Introduction Restless legs syndrome (RLS) is a sensorimotor neurological disorder, characterized by an urge to move the legs that is intimately associated with an uncomfortable sensation in the leg [1]. Although iron insufficiency and dopamine have key roles in the pathophysiology of this disease, substantially less is known about the neuroanatomical basis of this disease [2]. Also, it is known that RLS is a central nerve system, mainly brain, disorder. As a brain disorder, structural brain imaging studies would be interesting, but prior studies of structural brain images in RLS patients, using voxel-based morphometry (VBM) and diffusion tensor imaging, have reported contrasting results [3–10]. Some studies showed no specific brain alteration, and others showed some structural brain alterations (Table 1). Obviously, these discrepancies might be explained by methodological differences between the studies. However, these methodological differences alone might not be sufficient to explain the different results between the studies. The heterogeneity of RLS might also be reflected in the differences between the studies. Recently, a study showed that postmortem RLS brains have widespread impairment in myelin [5]. These postmortem areas were also those that demonstrated reduced volume with voxel-based ⁎ Corresponding author. Department of Neurology, Keimyung University School of Medicine, 194 Dongsan-dong, Jung-gu, Daegu 700-712, Korea. Tel.: +82 53 250 7831; fax: +82 53 250 7840. E-mail address: [email protected] (Y.W. Cho). 1 Two first authors contributed equally to this work. http://dx.doi.org/10.1016/j.clinimag.2014.07.010 0899-7071/© 2015 Elsevier Inc. All rights reserved.
analyses and were consistent with that seen with experimentally induced brain iron deficiency [5]. Thus, there is supported evidence that some structural alteration exists in the brain of RLS patients, and it may be related with the pathophysiology of RLS. However, as previous studies had contrasting results and there had been no such study in the Asian RLS population, we decided to do this study. Mainly, we are learning about the understanding of the pathophysiological mechanism of RLS. In addition there is no biomarker to diagnosis RLS, so far to the authors' knowledge. Therefore, the VBM study would be valuable in understanding the pathophysiology, particularly the neuroanatomical basis of this disease. We have speculated that some structural brain alternation does exist in patients with RLS. The purpose of our study is to demonstrate whether or not RLS is associated with any brain alteration using VBM and if any clinical variables influence these alterations. 2. Materials and methods 2.1. Study subjects This study recruited 46 subjects above the age of 18 who were diagnosed as primary RLS and who visited a University Hospital Outpatient Sleep Disorders Center. We enrolled all consecutive RLS patients who consented to participate in this study and 46 age- and gender-matched healthy controls. The healthy controls were selected after screening for any sleep disorders including RLS or serious
Y. Chang et al. / Clinical Imaging 39 (2015) 20–25
21
Table 1 Summary of the brain volumetric studies in patients with RLS Year, author
Magnet strength
TR/TE voxel size
No. of patients/control
Analysis
Results
2005, Etgen et al.
1.5 T
51/51
SPM1 software
2007, Hornyak et al.
1.5 T
11.08/4 1×1×1.08 1×1×1
14/14
SPM2 software
2007, Unrath et al.
1.5 T
9.7/3.93 1×1×1
63/40
SPM2 software
2010, Celle et al.
1.5 T
1900/3.95 2×2×2
17/54
SPM2 software
2011, Connor et al.
3T
9.87/4.59 1×1×1
23/23
SPM5 software
2012, Comley et al.
1.5 T
16/16
FSL-VBM software
2012, Margariti et al.
1.5 T
11/11
SPM5 Software
2012, Rizzo et al.
1.5 T
20/5 1.02×1.02×1 25/4.6 0.86×0.86×1 5.1/12.5
GM increase in pulvinar bilaterally (statistical test: GLM; threshold at Pb.001 uncorrected) GM increase in left hippocampus and middle orbitofrontal gyrus (statistical test: ANCOVA; threshold at Pb.05 corrected for multiple comparisons) GM decreased in the bihemispheric primary somatosensory cortex, which additionally extended into left-sided primary motor areas (statistical test: ANCOVA; threshold at Pb.05 corrected for small-volume. No significant results at Pb.05 corrected for multiple comparisons) No significant results at Pb.05 corrected for multiple comparisons. Statistical test: unpaired Student's t test (GM increase in left inferior occipital and left calcarine cortex and GM decrease in right superior temporal cortex at Pb.001 uncorrected). Reductions in WM volumes in small areas of the genu of the corpus callosum, anterior cingulum and precentral gyrus (statistical test: paired Student's t test; threshold at Pb.001 uncorrected) No significant results at Pb.05 corrected for multiple comparisons. Statistical test: unpaired Student's t test No significant results at Pb.05 corrected for multiple comparisons
20/20
This study
3T
6/2.2
46/46
SPM8 and FSL-VBM software VBM using SPM8 VBM8 toolbox
No significant results in all analysis at Pb.05 corrected for multiple comparisons Reduced volume in some areas of the left hippocampal gyrus, both parietal lobes, medial frontal areas, lateral temporal areas and cerebellum (uncorrected, Pb.001)
ANCOVA, analysis of covariance.
medical disorders. The diagnosis was based on diagnostic standards set by the National Institutes of Health workshop on RLS [1] and was made during face-to-face interviews utilizing the validated Korean-language version [11] of the John Hopkins Telephone diagnostic questionnaire [12]. We excluded the RLS mimics during face-to-face interviews and physical examinations and laboratory tests in person. All subjects were diagnosed by a sleep specialist (C.Y.W.) who is an expert in RLS. We also excluded comorbidities to RLS or secondary RLS caused by pregnancy or by other diseases such as chronic kidney disease or peripheral neuropathy. However, subjects with only peripheral iron deficiency without definite cause were included. We also excluded other comorbid sleep disorders such as primary insomnia, sleep disordered breathing, circadian sleep disorders and parasomnia through analysis of the sleep questionnaires. The questionnaires used in this study underwent a validation process in Korean populations [13–15]. The severity of RLS symptoms was evaluated from the validated Koreanlanguage version [16] of the International RLS scale (K-IRLS) [17]. All RLS subjects had moderate to severe RLS symptoms (K-IRLS≥15).
Matic toolbox (http://irc.cchmc.org/software/tom.php) was used to create a template of TPM. Images were corrected for a bias field in homogeneities and registered using linear 12-parameter affine and nonlinear transforms, and tissue was classified into GM, WM and CSF within the same generative model [18]. The segmentation procedure was performed by DARTEL for highdimensional warping [19] by accounting for partial volume effect [20], by applying adaptive maximum a posterior estimations [21] and by a hidden Markov random filed model [22]. The GM images were modulated to account for volume changes resulting from the normalization process with an in house created TPM template. Finally, images were smoothed with a Gaussian kernel of 8-mm isolation window (full width at half maximum). Image analyses were made and data recorded by a single rater blinded to the clinical diagnosis of the patient. The study was approved by the institutional ethics committee of a regional hospital. Informed consent was obtained from all participating subjects. 2.3. Statistical analysis
2.2. MR image acquisition and analysis methods Magnetic resonance imaging (MRI) was performed on a GE VHi scanner operating at 3.0T (GE Medical Systems, Milwaukee, WI, USA). A three-dimensional (3D) anatomical MRI was obtained on each subject using a T1-weighted 3D spoiled gradient recalled (3D-SPGR) sequence (repetition time (TR)=6 ms, echo time (TE)=2.2 ms, flip angle=20°, field of view=240 mm, 256×256, 152 axial slices, slice thickness=2 mm thick). A T1-weighted 3D-SPGR sequence was employed in the current study because this sequence provides high resolution and good contrast between gray and white matter (WM) structures for a VBM. All 3D-SPGR data were processed in the same method using a VBM toolbox version 8.0 (http://dbm.neuro.uni-jena.de) with SPM8 (Institute of Neurology, London, UK; http://www.fil.ion.ucl.ac.uk/spm). Tissue probability maps [TPM; gray matter (GM), WM and cerebrospinal fluid (CSF)] were created for optimizing registration with obtained T1-weighted images using 126 normal Korean male subjects' T1 anatomical data (mean age±S.D.: 34.25±11.14). The Template-O-
The general linear model (GLM) testing was used to find any structural changes of GM throughout the whole brain. A whole brainbased statistical approach instead of a region of interest (ROI)-based method was chosen since this type of approach is free from any a Table 2 Demographic and clinical characteristics Characteristics
RLS (n=46)
Controls (n=46)
p
Age (years) Gender, male (%)/female (%) RLS severity (K-IRLS) Symptom duration (months) Onset type (early/late) Serum ferritin Medications Yes (dopamine agonists) No
55.9±11.4 14 (30.4)/32 (69.6) 26.9±7.0 123.9±101.4 21 (45.7%)/25 (54.3%) 53.13±50.07
53.0±11.9 18 (39.1)/28 (60.9)
.278 .381
13 (28.3%) 33 (71.7%)
22
Y. Chang et al. / Clinical Imaging 39 (2015) 20–25
priori hypotheses and the pathology of RLS cannot be hypothesized to certain brain regions. All clusters that survived at a significant level, Pb.001, uncorrected for multiple comparison, voxel level=32, were considered for interpretation. An independent t test was used to compare brain morphology for age- and gender-matched pairs from the two groups. We used partial correlational analysis controlling for age and gender to investigate the relation of any significantly reduced volume in ROIs to clinical variables and in assessing the association between the RLS patient's voxel-based volumes of the entire brain and clinical variables, such as the K-IRLS total, a subfactor of the K-IRLS (symptom severity related), and the disease duration. The GLM testing was used to find any structural changes of GM throughout the whole brain. 3. Results The mean age of the idiopathic RLS patients was 55.9±11.4 years, and there were 32 females (69.6%). The mean age of the controls was 53.0±11.9 years, and there were 28 females (60.9%). The mean K-IRLS was 26.9±7.0, and the mean RLS duration was 123.9±101.4 months. Thirteen of them had dopaminergic treatments (Table 2). All RLS patients showed no gross structural abnormalities of the brain. A comparison of RLS to controls subjects showed significant regional decreases of GM volume in the left hippocampal gyrus, both parietal lobes, medial frontal areas, lateral temporal areas and
Left
cerebellum (uncorrected, rb0.001; Fig. 1). These areas showed no correlation with clinical variables using partial correlational analyses controlling for age and gender. However, the covariate analysis, based on a VBM after adjusting for age and gender, showed significant negative correlations between symptom severity (the K-IRLS total score; r=−0.500, Pb.001, the K-IRLS symptom related factor; r=−0.535, Pb .001) and specific area (Culmen) of the cerebellum (Fig. 2), and also negative correlations between the disease duration (r=−0.746, Pb.001) and some brain areas, such as both medial frontal areas, the left cuneus, both temporal areas, the left thalamus and the right precentral gyrus (Fig. 3) in RLS patients. With further analysis using a less restrictive threshold (uncorrected, Pb.005), there were also negative correlations between the left precentral gyrus and symptom severity (the K-IRLS total score; r= −0.462, the K-IRLS symptom related factor; r=−0.468, Pb .01) in RLS patients. 4. Discussion In this study, RLS patients were found to have structural alterations compared to the controls and that some of these brain alterations correlated to symptom duration or severity. The comparison of the RLS patients versus healthy controls showed significant regional decreases of GM volume in the left hippocampal gyrus, both parietal lobes and medial frontal areas and lateral temporal areas, and the
Right
Fig. 1. The left hippocampal gyrus and bilateral parietal, medial frontal lobes, temporal lobes and cerebellum were reduced compared to healthy controls. Uncorrected P=.001, voxel level=32.
Y. Chang et al. / Clinical Imaging 39 (2015) 20–25
Left
23
Right
B
A
Fig. 2. The correlation areas with RLS symptoms. Uncorrected P= .001, voxel level=32. (A) Alternated brain area, culmen. (B) Correlation graph of the K-IRLS total scores (r= −0.500, Pb.001).
cerebellum. We think that these findings are relevant to RLS, even though the statistical power is not high. The hippocampus is a part of the limbic system and is important in forming new memories and connecting emotions and spatial orientation [23]. The feeling of discomfort and its relevant persistent memory in RLS may be related with this alteration. The frontal cortex generates a large number of motor or mental plans, and the most appropriate are selected for execution. Stimulation of the frontal cortex brings an urge to move involving signals being sent to the
parietal area [24]. There have been reports that stimulation of the parietal lobe evokes a desire to move, which is similar to RLS symptoms [25,26]. The various interactions between these two areas may be related to an alteration in the frontal cortex. The temporal gyrus is connected with the frontal cortex and insula [27]. This area is involved in perception of pain-related stimuli and its processing. All these functions may be related with the volume changes observed in this study. The cerebellum is important for sensory processing [28]. The cerebellum may act as an integral network and may inhibit the
Right
Left
A
B
Fig. 3. The correlation areas with RLS symptom duration. Uncorrected P=.001, voxel level=32. (A) Alternated brain areas, both medial frontal areas, the left cuneus, both temporal areas, the left thalamus and the right precentral gyrus. (B) Correlation graph of RLS symptom duration (r=−0.746, Pb.001).
24
Y. Chang et al. / Clinical Imaging 39 (2015) 20–25
processing of uncomfortable sensory symptoms of RLS [29]. A failure of this inhibition may provide a link to reports that certain heredoataxias are present with RLS [30]. In a functional MRI study of RLS, the cerebellum and thalamus were activated during sensory leg discomfort, which supports evidence that the cerebellum is involved in the pathogenesis of RLS [31]. However, it still remains open if the volume changes in these brain regions are specific to RLS. Therefore, the abovementioned interpretation why each implicated region could cause or result from RLS must be considered with caution. Though these changed volume areas relative to controls showed no correlation with clinical variables, when we did correlation analysis based on a VBM in the patient group with clinical variables, we found some interesting correlation areas, such as the culmen (cerebellum), both medial frontal areas, the left cuneus, both temporal areas, the left thalamus and the right precentral gyrus. We speculate that these correlated areas using voxel-based analysis were changing depending on clinical course, but not remarkably changed relative to controls. The culmen of the cerebellum (though not the exact same area in the groups' comparison) showed a negative correlation with the symptom severity. This result was similar when using both the K-IRLS total scores and symptom severity–related factor. That is, the more severe symptoms in RLS patients reflected more alteration in the cerebellar areas. This seems to indicate that the cerebellum may be one of the major circuits affecting RLS symptoms. Both precentral gyri showed a negative correlation with the disease duration or symptom severity. The longer the duration or the more severe the symptoms of RLS, the volumes in these areas were more reduced. Several VBM studies reported alterations of the primary sensory motor area [5,10]. In the study of Rizzo et al. [9], though they concluded that there were no significant results using VBM and DTI, using the less restrictive threshold, they reported that the sensory motor WM area was correlated with symptom severity. These findings are similar to our study. The main complaints of RLS patients are abnormal sensations in their legs with an urge to move. And these symptoms are relieved with movement. A previous study reported that RLS is associated with an impairment of the central somatosensory processing [32]. Also, neurophysiological studies, including the transcranial magnetic stimulation, reported that alterations in movement-related cortical plasticity in RLS and RLS symptoms are related to cortical sensorimotor dysfunction [33,34]. Our study, as well as previous studies, has shown that an alteration of the sensorimotor area is related to the pathophysiology of RLS. However, we did not find any significant volume differences for the primary dopamine areas (caudate and putamen) compared to healthy controls. It contrasts with the Parkinson data, another dopamine dysfunctional disease [35]. It may be related to a different pathophysiology in RLS, such as increased dopaminergic activity in these areas, increased tyrosine hydroxylase and a presumed increase in synaptic dopamine levels [36]. The volume difference and relation to clinical variables found in this study may be related to RLS pathophysiology; it is not, however, certain whether these changes are a primary neuronal change or reflect a consequence of a chronic increase in afferent input of behaviorally relevant information. The low iron content in the brain may possibly have had an effect in the changes. A limitation of our study was not excluding the effect of medication in the results. However, all our subjects were primary RLS patients and were diagnosed by a direct face-to-face interview. So the sample is of relatively homogenous and advanced patients. Also, we used VBM analysis using Korean healthy brain templates, and these findings are constant even using covariate analysis. Also, we have a larger number of subjects than most previous studies, and we used 3-T MRI, in comparison with using 1.5-T MRI as in most previous studies. Our study also included a large number of comparisons. The significance for differences was set at Pb.001 to take into account the large number of comparisons. These findings, nonetheless, need
replication before they can be considered conclusive. They, however, open interesting areas for future study. 5. Conclusions We found that some brain areas, including the hippocampus and cerebellum, had reduced volume relative to controls and that the length of the disease duration and severity of symptoms could be related to the alterations of the brain structure. In the future, studies on whether RLS treatment can change brain alteration, and how to prevent it would be interesting. Acknowledgments This work was supported by the Basic Science Research Program through the National Research Foundation of Korea funded by the Ministry of Education, Science and Technology (2011-0008025). References [1] Allen RP, Picchietti D, Hening WA, Trenkwalder C, Walters AS, Montplaisi J. Restless legs syndrome: diagnostic criteria, special considerations, and epidemiology. A report from the restless legs syndrome diagnosis and epidemiology workshop at the National Institutes of Health. Sleep Med 2003;4:101–19. [2] Earley CJ, Allen RP, Hening W. Restless legs syndrome and periodic leg movements in sleep. Handb Clin Neurol 2011;99:913–48. [3] Celle S, Roche F, Peyron R, Faillenot I, Laurent B, Pichot V, et al. Lack of specific gray matter alterations in restless legs syndrome in elderly subjects. J Neurol 2010;257:344–8. [4] Comley RA, Cervenka S, Palhagen SE, Panagiotidis G, Matthews JC, Lai RY, et al. A comparison of gray matter density in restless legs syndrome patients and matched controls using voxel-based morphometry. J Neuroimaging 2012;22:28–32. [5] Connor JR, Ponnuru P, Lee BY, Podskalny GD, Alam S, Allen RP, et al. Postmortem and imaging based analyses reveal CNS decreased myelination in restless legs syndrome. Sleep Med 2011;12:614–9. [6] Etgen T, Draganski B, Ilg C, Schroder M, Geisler P, Hajak G, et al. Bilateral thalamic gray matter changes in patients with restless legs syndrome. Neuroimage 2005;24:1242–7. [7] Hornyak M, Ahrendts JC, Spiegelhalder K, Riemann D, Voderholzer U, Feige B, et al. Voxel-based morphometry in unmedicated patients with restless legs syndrome. Sleep Med 2007;9:22–6. [8] Margariti PN, Astrakas LG, Tsouli SG, Hadjigeorgiou GM, Konitsiotis S, Argyropoulou MI. Investigation of unmedicated early onset restless legs syndrome by voxel-based morphometry, T2 relaxometry, and functional MR imaging during the night-time hours. AJNR Am J Neuroradiol 2012;33:667–72. [9] Rizzo G, Manners D, Vetrugno R, Tonon C, Malucelli E, Plazzi G, et al. Combined brain voxel-based morphometry and diffusion tensor imaging study in idiopathic restless legs syndrome patients. Eur J Neurol 2012;19:1045–9. [10] Unrath A, Juengling FD, Schork M, Kassubek J. Cortical grey matter alterations in idiopathic restless legs syndrome: an optimized voxel-based morphometry study. Mov Disord 2007;22:1751–6. [11] Cho Y, Shin W, Yun C, Hong S, Kim J, Allen R, et al. Epidemiology of restless legs syndrome in Korean adults. Sleep 2008;31:219–23. [12] Hening WA, Allen RP, Washburn M, Lesage S, Earley CJ. Validation of the Hopkins telephone diagnostic interview for restless legs syndrome. Sleep Med 2008;9: 283–9. [13] Cho YW, Lee MY, Yun CH, Shin WC, Hong SB, Kim JH. The reliability and validity of the Korean version of paradigm of questions for epidemiology studies of restless legs syndrome and the Johns Hopkins telephone diagnostic interview form for the restless legs syndrome. J Korean Neurol Assoc 2007;25:494–9. [14] Cho YW, Lee JH, Son HK, Lee SH, Shin C, Johns MW. The reliability and validity of the Korean version of the Epworth sleepiness scale. Sleep Breath 2011;15:377–84. [15] Sohn SI, Lee MY, Cho YW. The reliability and validity of the Korean version of the Pittsburgh Sleep Quality Index. Sleep Breath 2012;16:803–12. [16] Yang J, Kim D, Lee J, Park K, Jung K, Shin W, et al. The reliability and validity of the Korean versions of the International Restless Legs Scale and the Restless Legs Syndrome Quality of Life Questionnaire. J Korean Neurol Assoc 2010;28:263–9. [17] Walters AS, LeBrocq C, Dhar A, Hening W, Rosen R, Allen RP, et al. Validation of the International Restless Legs Syndrome Study Group rating scale for restless legs syndrome. Sleep Med 2003;4:121–32. [18] Ashburner J, Friston KJ. Unified segmentation. Neuroimage 2005;26:839–51. [19] Ashburner J. A fast diffeomorphic image registration algorithm. Neuroimage 2007; 38:95–113. [20] Tohka J, Zijdenbos A, Evans A. Fast and robust parameter estimation for statistical partial volume models in brain MRI. Neuroimage 2004;23:84–97. [21] Rajapakse JC, Giedd JN, Rapoport JL. Statistical approach to segmentation of singlechannel cerebral MR images. IEEE Trans Med Imaging 1997;16:176–86. [22] Cuadra MB, Cammoun L, Butz T, Cuisenaire O, Thiran JP. Comparison and validation of tissue modelization and statistical classification methods in T1weighted MR brain images. IEEE Trans Med Imaging 2005;24:1548–65. [23] Fanselow MS, Dong HW. Are the dorsal and ventral hippocampus functionally distinct structures? Neuron 2010;65:7–19.
Y. Chang et al. / Clinical Imaging 39 (2015) 20–25 [24] Fried I, Katz A, McCarthy G, Sass KJ, Williamson P, Spencer SS, et al. Functional organization of human supplementary motor cortex studied by electrical stimulation. J Neurosci 1991;11:3656–66. [25] Desmurget M, Reilly KT, Richard N, Szathmari A, Mottolese C, Sirigu A. Movement intention after parietal cortex stimulation in humans. Science 2009;324:811–3. [26] Neuroscience Haggard P. The sources of human volition. Science 2009;324:731–3. [27] Moulton EA, Becerra L, Maleki N, Pendse G, Tully S, Hargreaves R, et al. Painful heat reveals hyperexcitability of the temporal pole in interictal and ictal migraine states. Cereb Cortex 2011;21:435–48. [28] Jueptner M, Rijntjes M, Weiller C, Faiss JH, Timmann D, Mueller SP, et al. Localization of a cerebellar timing process using PET. Neurology 1995;45:1540–5. [29] Trenkwalder C, Paulus W. Why do restless legs occur at rest? Pathophysiology of neuronal structures in RLS. Neurophysiology of RLS (part 2). Clin Neurophysiol 2004;115:1975–88. [30] Schols L, Haan J, Riess O, Amoiridis G, Przuntek H. Sleep disturbance in spinocerebellar ataxias: is the SCA3 mutation a cause of restless legs syndrome? Neurology 1998;51:1603–7.
25
[31] Bucher SF, Seelos KC, Oertel WH, Reiser M, Trenkwalder C. Cerebral generators involved in the pathogenesis of the restless legs syndrome. Ann Neurol 1997;41:639–45. [32] Schattschneider J, Bode A, Wasner G, Binder A, Deuschl G, Baron R. Idiopathic restless legs syndrome: abnormalities in central somatosensory processing. J Neurol 2004;251:977–82. [33] Scalise A, Pittaro-Cadore I, Golob EJ, Gigli GL. Absence of postexercise and delayed facilitation of motor cortex excitability in restless legs syndrome: evidence of altered cortical plasticity? Sleep 2006;29:770–5. [34] Tyvaert L, Houdayer E, Devanne H, Bourriez JL, Derambure P, Monaca C. Cortical involvement in the sensory and motor symptoms of primary restless legs syndrome. Sleep Med 2009;10:1090–6. [35] Pitcher TL, Melzer TR, Macaskill MR, Graham CF, Livingston L, Keenan RJ, et al. Reduced striatal volumes in Parkinson's disease: a magnetic resonance imaging study. Translational Neurodegeneration 2012;1:17. [36] Connor JR, Wang XS, Allen RP, Beard JL, Wiesinger JA, Felt BT, et al. Altered dopaminergic profile in the putamen and substantia nigra in restless leg syndrome. Brain 2009;132:2403–12.
Contents lists available at ScienceDirect
Clinical Imaging journal homepage: http://www.clinicalimaging.org
Gray matter alteration in patients with restless legs syndrome: a voxel-based morphometry study Yongmin Chang a, 1, Hyuk Won Chang b, 1, Huijin Song c, Jeonghun Ku d, Christopher J. Earley f, Richard P. Allen f, Yong Won Cho e,⁎ a
Department of Molecular Medicine, Kyungpook National University and Hospital, Daegu, Republic of Korea Department of Radiology, Dongsan Medical Center, Keimyung University School of Medicine, Daegu, Republic of Korea Department of Medical & Biological Engineering, Kyungpook National University and Hospital, Daegu, Republic of Korea d Department of Biomedical Engineering, Dongsan Medical Center, Keimyung University School of Medicine, Daegu, Republic of Korea e Department of Neurology, Dongsan Medical Center, Keimyung University School of Medicine, Daegu, Republic of Korea f Department of Neurology, Johns Hopkins University, Hopkins Bayview Medical Center, Baltimore, MD, USA b c
a r t i c l e
i n f o
Article history: Received 26 March 2014 Received in revised form 27 June 2014 Accepted 22 July 2014 Keywords: Restless legs syndrome MRI Voxel-based morphometry Gray matter
a b s t r a c t The purpose of this study was to demonstrate whether or not restless legs syndrome (RLS) is associated with any morphological change in gray matter. Forty-six RLS subjects and 46 controls were enrolled. We performed voxel-based morphometry analysis and compared the results of the two groups. The RLS subjects showed significant regional decreases of gray matter volume in the left hippocampal gyrus, both parietal lobes, medial frontal areas and cerebellum (uncorrected, Pb .001). We found that RLS patients showed structural alteration in the brain and alterations in certain parts of the brain in RLS patients are relevant to RLS. © 2015 Elsevier Inc. All rights reserved.
1. Introduction Restless legs syndrome (RLS) is a sensorimotor neurological disorder, characterized by an urge to move the legs that is intimately associated with an uncomfortable sensation in the leg [1]. Although iron insufficiency and dopamine have key roles in the pathophysiology of this disease, substantially less is known about the neuroanatomical basis of this disease [2]. Also, it is known that RLS is a central nerve system, mainly brain, disorder. As a brain disorder, structural brain imaging studies would be interesting, but prior studies of structural brain images in RLS patients, using voxel-based morphometry (VBM) and diffusion tensor imaging, have reported contrasting results [3–10]. Some studies showed no specific brain alteration, and others showed some structural brain alterations (Table 1). Obviously, these discrepancies might be explained by methodological differences between the studies. However, these methodological differences alone might not be sufficient to explain the different results between the studies. The heterogeneity of RLS might also be reflected in the differences between the studies. Recently, a study showed that postmortem RLS brains have widespread impairment in myelin [5]. These postmortem areas were also those that demonstrated reduced volume with voxel-based ⁎ Corresponding author. Department of Neurology, Keimyung University School of Medicine, 194 Dongsan-dong, Jung-gu, Daegu 700-712, Korea. Tel.: +82 53 250 7831; fax: +82 53 250 7840. E-mail address: [email protected] (Y.W. Cho). 1 Two first authors contributed equally to this work. http://dx.doi.org/10.1016/j.clinimag.2014.07.010 0899-7071/© 2015 Elsevier Inc. All rights reserved.
analyses and were consistent with that seen with experimentally induced brain iron deficiency [5]. Thus, there is supported evidence that some structural alteration exists in the brain of RLS patients, and it may be related with the pathophysiology of RLS. However, as previous studies had contrasting results and there had been no such study in the Asian RLS population, we decided to do this study. Mainly, we are learning about the understanding of the pathophysiological mechanism of RLS. In addition there is no biomarker to diagnosis RLS, so far to the authors' knowledge. Therefore, the VBM study would be valuable in understanding the pathophysiology, particularly the neuroanatomical basis of this disease. We have speculated that some structural brain alternation does exist in patients with RLS. The purpose of our study is to demonstrate whether or not RLS is associated with any brain alteration using VBM and if any clinical variables influence these alterations. 2. Materials and methods 2.1. Study subjects This study recruited 46 subjects above the age of 18 who were diagnosed as primary RLS and who visited a University Hospital Outpatient Sleep Disorders Center. We enrolled all consecutive RLS patients who consented to participate in this study and 46 age- and gender-matched healthy controls. The healthy controls were selected after screening for any sleep disorders including RLS or serious
Y. Chang et al. / Clinical Imaging 39 (2015) 20–25
21
Table 1 Summary of the brain volumetric studies in patients with RLS Year, author
Magnet strength
TR/TE voxel size
No. of patients/control
Analysis
Results
2005, Etgen et al.
1.5 T
51/51
SPM1 software
2007, Hornyak et al.
1.5 T
11.08/4 1×1×1.08 1×1×1
14/14
SPM2 software
2007, Unrath et al.
1.5 T
9.7/3.93 1×1×1
63/40
SPM2 software
2010, Celle et al.
1.5 T
1900/3.95 2×2×2
17/54
SPM2 software
2011, Connor et al.
3T
9.87/4.59 1×1×1
23/23
SPM5 software
2012, Comley et al.
1.5 T
16/16
FSL-VBM software
2012, Margariti et al.
1.5 T
11/11
SPM5 Software
2012, Rizzo et al.
1.5 T
20/5 1.02×1.02×1 25/4.6 0.86×0.86×1 5.1/12.5
GM increase in pulvinar bilaterally (statistical test: GLM; threshold at Pb.001 uncorrected) GM increase in left hippocampus and middle orbitofrontal gyrus (statistical test: ANCOVA; threshold at Pb.05 corrected for multiple comparisons) GM decreased in the bihemispheric primary somatosensory cortex, which additionally extended into left-sided primary motor areas (statistical test: ANCOVA; threshold at Pb.05 corrected for small-volume. No significant results at Pb.05 corrected for multiple comparisons) No significant results at Pb.05 corrected for multiple comparisons. Statistical test: unpaired Student's t test (GM increase in left inferior occipital and left calcarine cortex and GM decrease in right superior temporal cortex at Pb.001 uncorrected). Reductions in WM volumes in small areas of the genu of the corpus callosum, anterior cingulum and precentral gyrus (statistical test: paired Student's t test; threshold at Pb.001 uncorrected) No significant results at Pb.05 corrected for multiple comparisons. Statistical test: unpaired Student's t test No significant results at Pb.05 corrected for multiple comparisons
20/20
This study
3T
6/2.2
46/46
SPM8 and FSL-VBM software VBM using SPM8 VBM8 toolbox
No significant results in all analysis at Pb.05 corrected for multiple comparisons Reduced volume in some areas of the left hippocampal gyrus, both parietal lobes, medial frontal areas, lateral temporal areas and cerebellum (uncorrected, Pb.001)
ANCOVA, analysis of covariance.
medical disorders. The diagnosis was based on diagnostic standards set by the National Institutes of Health workshop on RLS [1] and was made during face-to-face interviews utilizing the validated Korean-language version [11] of the John Hopkins Telephone diagnostic questionnaire [12]. We excluded the RLS mimics during face-to-face interviews and physical examinations and laboratory tests in person. All subjects were diagnosed by a sleep specialist (C.Y.W.) who is an expert in RLS. We also excluded comorbidities to RLS or secondary RLS caused by pregnancy or by other diseases such as chronic kidney disease or peripheral neuropathy. However, subjects with only peripheral iron deficiency without definite cause were included. We also excluded other comorbid sleep disorders such as primary insomnia, sleep disordered breathing, circadian sleep disorders and parasomnia through analysis of the sleep questionnaires. The questionnaires used in this study underwent a validation process in Korean populations [13–15]. The severity of RLS symptoms was evaluated from the validated Koreanlanguage version [16] of the International RLS scale (K-IRLS) [17]. All RLS subjects had moderate to severe RLS symptoms (K-IRLS≥15).
Matic toolbox (http://irc.cchmc.org/software/tom.php) was used to create a template of TPM. Images were corrected for a bias field in homogeneities and registered using linear 12-parameter affine and nonlinear transforms, and tissue was classified into GM, WM and CSF within the same generative model [18]. The segmentation procedure was performed by DARTEL for highdimensional warping [19] by accounting for partial volume effect [20], by applying adaptive maximum a posterior estimations [21] and by a hidden Markov random filed model [22]. The GM images were modulated to account for volume changes resulting from the normalization process with an in house created TPM template. Finally, images were smoothed with a Gaussian kernel of 8-mm isolation window (full width at half maximum). Image analyses were made and data recorded by a single rater blinded to the clinical diagnosis of the patient. The study was approved by the institutional ethics committee of a regional hospital. Informed consent was obtained from all participating subjects. 2.3. Statistical analysis
2.2. MR image acquisition and analysis methods Magnetic resonance imaging (MRI) was performed on a GE VHi scanner operating at 3.0T (GE Medical Systems, Milwaukee, WI, USA). A three-dimensional (3D) anatomical MRI was obtained on each subject using a T1-weighted 3D spoiled gradient recalled (3D-SPGR) sequence (repetition time (TR)=6 ms, echo time (TE)=2.2 ms, flip angle=20°, field of view=240 mm, 256×256, 152 axial slices, slice thickness=2 mm thick). A T1-weighted 3D-SPGR sequence was employed in the current study because this sequence provides high resolution and good contrast between gray and white matter (WM) structures for a VBM. All 3D-SPGR data were processed in the same method using a VBM toolbox version 8.0 (http://dbm.neuro.uni-jena.de) with SPM8 (Institute of Neurology, London, UK; http://www.fil.ion.ucl.ac.uk/spm). Tissue probability maps [TPM; gray matter (GM), WM and cerebrospinal fluid (CSF)] were created for optimizing registration with obtained T1-weighted images using 126 normal Korean male subjects' T1 anatomical data (mean age±S.D.: 34.25±11.14). The Template-O-
The general linear model (GLM) testing was used to find any structural changes of GM throughout the whole brain. A whole brainbased statistical approach instead of a region of interest (ROI)-based method was chosen since this type of approach is free from any a Table 2 Demographic and clinical characteristics Characteristics
RLS (n=46)
Controls (n=46)
p
Age (years) Gender, male (%)/female (%) RLS severity (K-IRLS) Symptom duration (months) Onset type (early/late) Serum ferritin Medications Yes (dopamine agonists) No
55.9±11.4 14 (30.4)/32 (69.6) 26.9±7.0 123.9±101.4 21 (45.7%)/25 (54.3%) 53.13±50.07
53.0±11.9 18 (39.1)/28 (60.9)
.278 .381
13 (28.3%) 33 (71.7%)
22
Y. Chang et al. / Clinical Imaging 39 (2015) 20–25
priori hypotheses and the pathology of RLS cannot be hypothesized to certain brain regions. All clusters that survived at a significant level, Pb.001, uncorrected for multiple comparison, voxel level=32, were considered for interpretation. An independent t test was used to compare brain morphology for age- and gender-matched pairs from the two groups. We used partial correlational analysis controlling for age and gender to investigate the relation of any significantly reduced volume in ROIs to clinical variables and in assessing the association between the RLS patient's voxel-based volumes of the entire brain and clinical variables, such as the K-IRLS total, a subfactor of the K-IRLS (symptom severity related), and the disease duration. The GLM testing was used to find any structural changes of GM throughout the whole brain. 3. Results The mean age of the idiopathic RLS patients was 55.9±11.4 years, and there were 32 females (69.6%). The mean age of the controls was 53.0±11.9 years, and there were 28 females (60.9%). The mean K-IRLS was 26.9±7.0, and the mean RLS duration was 123.9±101.4 months. Thirteen of them had dopaminergic treatments (Table 2). All RLS patients showed no gross structural abnormalities of the brain. A comparison of RLS to controls subjects showed significant regional decreases of GM volume in the left hippocampal gyrus, both parietal lobes, medial frontal areas, lateral temporal areas and
Left
cerebellum (uncorrected, rb0.001; Fig. 1). These areas showed no correlation with clinical variables using partial correlational analyses controlling for age and gender. However, the covariate analysis, based on a VBM after adjusting for age and gender, showed significant negative correlations between symptom severity (the K-IRLS total score; r=−0.500, Pb.001, the K-IRLS symptom related factor; r=−0.535, Pb .001) and specific area (Culmen) of the cerebellum (Fig. 2), and also negative correlations between the disease duration (r=−0.746, Pb.001) and some brain areas, such as both medial frontal areas, the left cuneus, both temporal areas, the left thalamus and the right precentral gyrus (Fig. 3) in RLS patients. With further analysis using a less restrictive threshold (uncorrected, Pb.005), there were also negative correlations between the left precentral gyrus and symptom severity (the K-IRLS total score; r= −0.462, the K-IRLS symptom related factor; r=−0.468, Pb .01) in RLS patients. 4. Discussion In this study, RLS patients were found to have structural alterations compared to the controls and that some of these brain alterations correlated to symptom duration or severity. The comparison of the RLS patients versus healthy controls showed significant regional decreases of GM volume in the left hippocampal gyrus, both parietal lobes and medial frontal areas and lateral temporal areas, and the
Right
Fig. 1. The left hippocampal gyrus and bilateral parietal, medial frontal lobes, temporal lobes and cerebellum were reduced compared to healthy controls. Uncorrected P=.001, voxel level=32.
Y. Chang et al. / Clinical Imaging 39 (2015) 20–25
Left
23
Right
B
A
Fig. 2. The correlation areas with RLS symptoms. Uncorrected P= .001, voxel level=32. (A) Alternated brain area, culmen. (B) Correlation graph of the K-IRLS total scores (r= −0.500, Pb.001).
cerebellum. We think that these findings are relevant to RLS, even though the statistical power is not high. The hippocampus is a part of the limbic system and is important in forming new memories and connecting emotions and spatial orientation [23]. The feeling of discomfort and its relevant persistent memory in RLS may be related with this alteration. The frontal cortex generates a large number of motor or mental plans, and the most appropriate are selected for execution. Stimulation of the frontal cortex brings an urge to move involving signals being sent to the
parietal area [24]. There have been reports that stimulation of the parietal lobe evokes a desire to move, which is similar to RLS symptoms [25,26]. The various interactions between these two areas may be related to an alteration in the frontal cortex. The temporal gyrus is connected with the frontal cortex and insula [27]. This area is involved in perception of pain-related stimuli and its processing. All these functions may be related with the volume changes observed in this study. The cerebellum is important for sensory processing [28]. The cerebellum may act as an integral network and may inhibit the
Right
Left
A
B
Fig. 3. The correlation areas with RLS symptom duration. Uncorrected P=.001, voxel level=32. (A) Alternated brain areas, both medial frontal areas, the left cuneus, both temporal areas, the left thalamus and the right precentral gyrus. (B) Correlation graph of RLS symptom duration (r=−0.746, Pb.001).
24
Y. Chang et al. / Clinical Imaging 39 (2015) 20–25
processing of uncomfortable sensory symptoms of RLS [29]. A failure of this inhibition may provide a link to reports that certain heredoataxias are present with RLS [30]. In a functional MRI study of RLS, the cerebellum and thalamus were activated during sensory leg discomfort, which supports evidence that the cerebellum is involved in the pathogenesis of RLS [31]. However, it still remains open if the volume changes in these brain regions are specific to RLS. Therefore, the abovementioned interpretation why each implicated region could cause or result from RLS must be considered with caution. Though these changed volume areas relative to controls showed no correlation with clinical variables, when we did correlation analysis based on a VBM in the patient group with clinical variables, we found some interesting correlation areas, such as the culmen (cerebellum), both medial frontal areas, the left cuneus, both temporal areas, the left thalamus and the right precentral gyrus. We speculate that these correlated areas using voxel-based analysis were changing depending on clinical course, but not remarkably changed relative to controls. The culmen of the cerebellum (though not the exact same area in the groups' comparison) showed a negative correlation with the symptom severity. This result was similar when using both the K-IRLS total scores and symptom severity–related factor. That is, the more severe symptoms in RLS patients reflected more alteration in the cerebellar areas. This seems to indicate that the cerebellum may be one of the major circuits affecting RLS symptoms. Both precentral gyri showed a negative correlation with the disease duration or symptom severity. The longer the duration or the more severe the symptoms of RLS, the volumes in these areas were more reduced. Several VBM studies reported alterations of the primary sensory motor area [5,10]. In the study of Rizzo et al. [9], though they concluded that there were no significant results using VBM and DTI, using the less restrictive threshold, they reported that the sensory motor WM area was correlated with symptom severity. These findings are similar to our study. The main complaints of RLS patients are abnormal sensations in their legs with an urge to move. And these symptoms are relieved with movement. A previous study reported that RLS is associated with an impairment of the central somatosensory processing [32]. Also, neurophysiological studies, including the transcranial magnetic stimulation, reported that alterations in movement-related cortical plasticity in RLS and RLS symptoms are related to cortical sensorimotor dysfunction [33,34]. Our study, as well as previous studies, has shown that an alteration of the sensorimotor area is related to the pathophysiology of RLS. However, we did not find any significant volume differences for the primary dopamine areas (caudate and putamen) compared to healthy controls. It contrasts with the Parkinson data, another dopamine dysfunctional disease [35]. It may be related to a different pathophysiology in RLS, such as increased dopaminergic activity in these areas, increased tyrosine hydroxylase and a presumed increase in synaptic dopamine levels [36]. The volume difference and relation to clinical variables found in this study may be related to RLS pathophysiology; it is not, however, certain whether these changes are a primary neuronal change or reflect a consequence of a chronic increase in afferent input of behaviorally relevant information. The low iron content in the brain may possibly have had an effect in the changes. A limitation of our study was not excluding the effect of medication in the results. However, all our subjects were primary RLS patients and were diagnosed by a direct face-to-face interview. So the sample is of relatively homogenous and advanced patients. Also, we used VBM analysis using Korean healthy brain templates, and these findings are constant even using covariate analysis. Also, we have a larger number of subjects than most previous studies, and we used 3-T MRI, in comparison with using 1.5-T MRI as in most previous studies. Our study also included a large number of comparisons. The significance for differences was set at Pb.001 to take into account the large number of comparisons. These findings, nonetheless, need
replication before they can be considered conclusive. They, however, open interesting areas for future study. 5. Conclusions We found that some brain areas, including the hippocampus and cerebellum, had reduced volume relative to controls and that the length of the disease duration and severity of symptoms could be related to the alterations of the brain structure. In the future, studies on whether RLS treatment can change brain alteration, and how to prevent it would be interesting. Acknowledgments This work was supported by the Basic Science Research Program through the National Research Foundation of Korea funded by the Ministry of Education, Science and Technology (2011-0008025). References [1] Allen RP, Picchietti D, Hening WA, Trenkwalder C, Walters AS, Montplaisi J. Restless legs syndrome: diagnostic criteria, special considerations, and epidemiology. A report from the restless legs syndrome diagnosis and epidemiology workshop at the National Institutes of Health. Sleep Med 2003;4:101–19. [2] Earley CJ, Allen RP, Hening W. Restless legs syndrome and periodic leg movements in sleep. Handb Clin Neurol 2011;99:913–48. [3] Celle S, Roche F, Peyron R, Faillenot I, Laurent B, Pichot V, et al. Lack of specific gray matter alterations in restless legs syndrome in elderly subjects. J Neurol 2010;257:344–8. [4] Comley RA, Cervenka S, Palhagen SE, Panagiotidis G, Matthews JC, Lai RY, et al. A comparison of gray matter density in restless legs syndrome patients and matched controls using voxel-based morphometry. J Neuroimaging 2012;22:28–32. [5] Connor JR, Ponnuru P, Lee BY, Podskalny GD, Alam S, Allen RP, et al. Postmortem and imaging based analyses reveal CNS decreased myelination in restless legs syndrome. Sleep Med 2011;12:614–9. [6] Etgen T, Draganski B, Ilg C, Schroder M, Geisler P, Hajak G, et al. Bilateral thalamic gray matter changes in patients with restless legs syndrome. Neuroimage 2005;24:1242–7. [7] Hornyak M, Ahrendts JC, Spiegelhalder K, Riemann D, Voderholzer U, Feige B, et al. Voxel-based morphometry in unmedicated patients with restless legs syndrome. Sleep Med 2007;9:22–6. [8] Margariti PN, Astrakas LG, Tsouli SG, Hadjigeorgiou GM, Konitsiotis S, Argyropoulou MI. Investigation of unmedicated early onset restless legs syndrome by voxel-based morphometry, T2 relaxometry, and functional MR imaging during the night-time hours. AJNR Am J Neuroradiol 2012;33:667–72. [9] Rizzo G, Manners D, Vetrugno R, Tonon C, Malucelli E, Plazzi G, et al. Combined brain voxel-based morphometry and diffusion tensor imaging study in idiopathic restless legs syndrome patients. Eur J Neurol 2012;19:1045–9. [10] Unrath A, Juengling FD, Schork M, Kassubek J. Cortical grey matter alterations in idiopathic restless legs syndrome: an optimized voxel-based morphometry study. Mov Disord 2007;22:1751–6. [11] Cho Y, Shin W, Yun C, Hong S, Kim J, Allen R, et al. Epidemiology of restless legs syndrome in Korean adults. Sleep 2008;31:219–23. [12] Hening WA, Allen RP, Washburn M, Lesage S, Earley CJ. Validation of the Hopkins telephone diagnostic interview for restless legs syndrome. Sleep Med 2008;9: 283–9. [13] Cho YW, Lee MY, Yun CH, Shin WC, Hong SB, Kim JH. The reliability and validity of the Korean version of paradigm of questions for epidemiology studies of restless legs syndrome and the Johns Hopkins telephone diagnostic interview form for the restless legs syndrome. J Korean Neurol Assoc 2007;25:494–9. [14] Cho YW, Lee JH, Son HK, Lee SH, Shin C, Johns MW. The reliability and validity of the Korean version of the Epworth sleepiness scale. Sleep Breath 2011;15:377–84. [15] Sohn SI, Lee MY, Cho YW. The reliability and validity of the Korean version of the Pittsburgh Sleep Quality Index. Sleep Breath 2012;16:803–12. [16] Yang J, Kim D, Lee J, Park K, Jung K, Shin W, et al. The reliability and validity of the Korean versions of the International Restless Legs Scale and the Restless Legs Syndrome Quality of Life Questionnaire. J Korean Neurol Assoc 2010;28:263–9. [17] Walters AS, LeBrocq C, Dhar A, Hening W, Rosen R, Allen RP, et al. Validation of the International Restless Legs Syndrome Study Group rating scale for restless legs syndrome. Sleep Med 2003;4:121–32. [18] Ashburner J, Friston KJ. Unified segmentation. Neuroimage 2005;26:839–51. [19] Ashburner J. A fast diffeomorphic image registration algorithm. Neuroimage 2007; 38:95–113. [20] Tohka J, Zijdenbos A, Evans A. Fast and robust parameter estimation for statistical partial volume models in brain MRI. Neuroimage 2004;23:84–97. [21] Rajapakse JC, Giedd JN, Rapoport JL. Statistical approach to segmentation of singlechannel cerebral MR images. IEEE Trans Med Imaging 1997;16:176–86. [22] Cuadra MB, Cammoun L, Butz T, Cuisenaire O, Thiran JP. Comparison and validation of tissue modelization and statistical classification methods in T1weighted MR brain images. IEEE Trans Med Imaging 2005;24:1548–65. [23] Fanselow MS, Dong HW. Are the dorsal and ventral hippocampus functionally distinct structures? Neuron 2010;65:7–19.
Y. Chang et al. / Clinical Imaging 39 (2015) 20–25 [24] Fried I, Katz A, McCarthy G, Sass KJ, Williamson P, Spencer SS, et al. Functional organization of human supplementary motor cortex studied by electrical stimulation. J Neurosci 1991;11:3656–66. [25] Desmurget M, Reilly KT, Richard N, Szathmari A, Mottolese C, Sirigu A. Movement intention after parietal cortex stimulation in humans. Science 2009;324:811–3. [26] Neuroscience Haggard P. The sources of human volition. Science 2009;324:731–3. [27] Moulton EA, Becerra L, Maleki N, Pendse G, Tully S, Hargreaves R, et al. Painful heat reveals hyperexcitability of the temporal pole in interictal and ictal migraine states. Cereb Cortex 2011;21:435–48. [28] Jueptner M, Rijntjes M, Weiller C, Faiss JH, Timmann D, Mueller SP, et al. Localization of a cerebellar timing process using PET. Neurology 1995;45:1540–5. [29] Trenkwalder C, Paulus W. Why do restless legs occur at rest? Pathophysiology of neuronal structures in RLS. Neurophysiology of RLS (part 2). Clin Neurophysiol 2004;115:1975–88. [30] Schols L, Haan J, Riess O, Amoiridis G, Przuntek H. Sleep disturbance in spinocerebellar ataxias: is the SCA3 mutation a cause of restless legs syndrome? Neurology 1998;51:1603–7.
25
[31] Bucher SF, Seelos KC, Oertel WH, Reiser M, Trenkwalder C. Cerebral generators involved in the pathogenesis of the restless legs syndrome. Ann Neurol 1997;41:639–45. [32] Schattschneider J, Bode A, Wasner G, Binder A, Deuschl G, Baron R. Idiopathic restless legs syndrome: abnormalities in central somatosensory processing. J Neurol 2004;251:977–82. [33] Scalise A, Pittaro-Cadore I, Golob EJ, Gigli GL. Absence of postexercise and delayed facilitation of motor cortex excitability in restless legs syndrome: evidence of altered cortical plasticity? Sleep 2006;29:770–5. [34] Tyvaert L, Houdayer E, Devanne H, Bourriez JL, Derambure P, Monaca C. Cortical involvement in the sensory and motor symptoms of primary restless legs syndrome. Sleep Med 2009;10:1090–6. [35] Pitcher TL, Melzer TR, Macaskill MR, Graham CF, Livingston L, Keenan RJ, et al. Reduced striatal volumes in Parkinson's disease: a magnetic resonance imaging study. Translational Neurodegeneration 2012;1:17. [36] Connor JR, Wang XS, Allen RP, Beard JL, Wiesinger JA, Felt BT, et al. Altered dopaminergic profile in the putamen and substantia nigra in restless leg syndrome. Brain 2009;132:2403–12.
