Microstructural MRI basis of the cognitive functions in patients with Spinocer-
ebellar ataxia type 2
G. Olivito, M. Lupo, C. Iacobacci, S. Clausi, S. Romano, M. Masciullo, M.
Molinari, M. Cercignani, M. Bozzali, M. Leggio
Reference: NSC 18072
To appear in: Neuroscience
Received Date: 20 June 2017
Accepted Date: 5 October 2017
Please cite this article as: G. Olivito, M. Lupo, C. Iacobacci, S. Clausi, S. Romano, M. Masciullo, M. Molinari, M.
Cercignani, M. Bozzali, M. Leggio, Microstructural MRI basis of the cognitive functions in patients with
Spinocerebellar ataxia type 2, Neuroscience (2017), doi: https://doi.org/10.1016/j.neuroscience.2017.10.007
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Microstructural MRI basis of the cognitive functions in patients with
Spinocerebellar ataxia type 2
G. Olivito1,2, M. Lupo1, C. Iacobacci1,3, S. Clausi1,4, S. Romano5, M. Masciullo6, M. Molinari7, M.
Cercignani2,8, M. Bozzali2,8, M. Leggio1,4.
1. Ataxia Laboratory, IRCCS Santa Lucia Foundation, Rome, Italy;
2. Neuroimaging Laboratory, IRCCS Santa Lucia Foundation, Rome, Italy;
3. PhD Program in Behavioral Neuroscience, Sapienza University of Rome, Rome, Italy
4.Department of Psychology, Sapienza University of Rome, Italy;
5. Department of Neurosciences, Mental Health and Sensory Organs (NESMOS), “Sapienza” University of Rome -
Sant'Andrea Hospital, Rome, Italy ;
6. SPInalREhabilitation Lab, IRCCS Fondazione Santa Lucia, Rome, Italy;
7. Neurorehabilitation 1 and Spinal Center, Robotic Neurorehabilitation Lab, IRCCS Santa Lucia Foundation, Rome,
8. Clinical Imaging Science Center, Brighton and Sussex Medical School, Brighton, UK.
Keywords: DTI; Tractography; Cerebellum; Cerebellar Peduncles; White Matter; Cognition.
Dr. Giusy Olivito, PhD
Ataxia Laboratory, IRCCS Santa Lucia Foundation, Via Ardeatina 306, 00179, Rome, Italy;
Neuroimaging Laboratory, IRCCS Santa Lucia Foundation, Via Ardeatina 306, 00179, Rome, Italy.
Telephone number: #39-06-51501547
Spinocerebellar ataxia type 2 (SCA2) is an autosomal dominant neurodegenerative disease
involving the cerebellum. The particular atrophy pattern results in some typical clinical features
mainly including motor deficits. In addition, the presence of cognitive impairments, involving
language, visuospatial and executive functions, has been also shown in SCA2 patients and it is now
widely accepted as a feature of the disease. The aim of the study is to investigate the
microstructural patterns and the anatomo-functional substrate that could account for the cognitive
symptomatology observed in SCA2 patients. In the present study, Diffusion tensor imaging (DTI)
based-tractography was performed to map the main cerebellar white matter bundles, such as Middle
and Superior Cerebellar Peduncles, connecting cerebellum with higher-order cerebral regions.
Damage-related diffusivity measures were used to determine the pattern of pathological changes of
cerebellar white matter microstructure in patients affected b y SCA2 and correlated with the
patients’ cognitive scores. Our results provide the first evidence that white matter (WM) diffusivity
is altered in the presence of the cerebellar cortical degeneration associated with SCA2 thus resulting
in a cerebello-cerebral dysregulation that may account for the specificity of cognitive
symptomatology observed in patients.
Spinocerebellar ataxia type 2 (SCA2) is an autosomal dominant neurodegenerative disease
characterized by a progressive cerebellar syndrome, typically affecting motor functions (Takahashi
et al., 2010). The cognitive performances of SCA2 patients have been exhaustively investigated and
it has been shown that patients affected by SCA2 present not only motor impairment but also a
cognitive symptomatology (Klinke et al., 2010; Fancellu et al., 2013; Moriarty et al., 2016), mainly
involving visuospatial and executive functions (LePira et al., 2002; Kawai et al., 2008).
From a neuropathological point of view, SCA2 present with a macroscopic pattern
of olivopontocerebellar atrophy as well with a pattern of neuronal loss in several and in cerebellar
cortex, and a diffuse damage of the brainstem and cerebellar white matter (WM) (Durr et al.,
1995; Gilman et al., 1996; Iwabuchi et al., 1999; Estrada et al., 1999; Pang et al., 2002). These
features have been depicted in vivo by Magnetic Resonance Imaging (MRI) studies using voxel
based morphometry (VBM) and diffusion tensor imaging (DTI) (Mandelli et al., 2007; Della Nave
et al., 2008a,b). Specifically, the cerebellar vermis and hemispheres show a pattern of extensive GM
loss with sparing of the vermian lobules I,II (lingula) and X (nodulus) and of the hemispheric
lobules I,II (lingula) and Crus II. Cerebellar WM damage has been shown to affect mainly the
peridentate regions and middle cerebellar peduncle (MCP) (Della Nave et al., 2008b). Consistent
with the hypothesis that cerebellar atrophy may affect also the regions connected with the
cerebellum (Dayan et al., 2016), several supratentorial areas have been found to be altered in SCA2,
such as the right orbito-frontal cortex, right temporo-mesial cortex, the primary sensorimotor cortex
bilaterally, the right thalamus, the left precentral gyrus and inferior frontal operculum as well as
inferior parietal and post-central gyri (Brenneis et al., 2003; Della Nave et al., 2008a). The
supratentorial atrophy can be related to both a primary degenerative process associated to SCA2
disease and to secondary effect resulting from the cerebellar deafferentation (Brenneis et al., 2003).
Furthermore, the interruption of cerebello-thalamo-cortical pathways has been reported as the
mechanism responsible for crossed cerebello-cerebral diaschisis (Broich et al., 1987; Boni et al.,
1992; Komaba et al., 2000).
Thus, it is possible to hypothesize that a disruption of cerebello-cerebral pathway is responsible for
structural and functional alteration of cortical areas. Middle (MCP) and Superior cerebellar
peduncles (SCP) are respectively the feedback and feedforward limbs of the cerebello-cortical
system through which the cerebellum receives information from cerebral regions and then sends
back the cerebellar-processed information to accomplish functions successfully. Therefore it is
reasonable that white matter alterations of the peduncles may reflect alteration in the cerebello-
cortical interactions and may be responsible for patients’ cognitive symptomatology.
DTI has proven to be a valuable tool for investigating brain WM since it can probe tissue
microstructure by assessing the displacement of water molecules within specific WM tracts (Basser
et al., 1994). In the brain the motion of water molecules is hindered by the local microstructure, as
they tend to diffuse in preferred directions corresponding to white matter fiber bundles orientation.
The diffusion tensor (DT) model is a simplistic diffusion MRI (dMRI) model which assumes only
one fiber direction per voxel. It is commonly used to quantify the diffusion process with DT-derived
metrics such as fractional anisotropy (FA), mean diffusivity (MD), radial diffusivity (RD) and axial
diffusivity (AD), which relate to the tendency of water molecules to move in a particular direction
(Alexander et al., 2007; Feldman et al., 2010). The 3D connectivity patterns of WM could be
investigating by using WM tractography, a well-established approach which follows coherent
spatial patterns in the major eigenvectors of the diffusion tensor field (Alexander et al., 2007) thus
providing a model of brain connectivity, through which brain disconnection can be studied.
Using different tracing methods, tractography algorithms are capable of generating anatomically
plausible estimates of WM trajectories in the human brain (Alexander et al., 2000, 2007). In
particular, more advanced diffusion imaging methods that allow multiple fiber directions to be
estimated, such as QBall Imaging (QBI) (Tuch et al., 2004) and the persistent angular structure
(PAS) (Jansons and Alexander, 2003) may be particularly suitable in the case of the cerebellum
since they are able to better resolve intersecting crossing WM regions, such as MCP and SCP
(Alexander et al.,2007; see also Dayan et al., 2016). Alterations of MCP and SCP dMRI metrics
(Della Nave et al, 2008b, 2011; Vieira Karuta et al., 2015) have been demonstrated in degenerative
ataxia patients, but no studies specifically used tractography to reconstruct these tracts and assessed
voxel-wise diffusivity alterations and their impact on cognitive outcomes of SCA2 patients.
The aim of this work is to investigate the microstructural organization of MCP and SCP,
reconstructed via tractography and to investigate the relationship between dMRI metrics and
cognitive subscores of SCA2 patients thus providing a novel contribution to understanding the
relationship between cerebello-cerebral disconnection likely to be associated with SCA2
neurodegenerative processes and the specific cognitive symptomatology.
Nine patients affected by SCA2 [F/M=6/3; mean age ± SD = 47,6± 10.2 years], were recruited from
the Ataxia Lab of Foundation Santa Lucia Hospital. Both in-patients (admitted for rehabilitation)
and out-patients (followed up at the clinic) were included. At the time of assessment, all patients
had more than 6 months of illness from the genetically confirmed diagnosis. T2-weighted MRI
scans, acquired as part of this research study, were visually inspected by an expert neuroradiologist
to ensure the absence of any extra-cerebellar lesion.
The neurological examination of the patients showed that they all presented with a pure cerebellar
syndrome, except CA-3 (see Table 1) that presented the Babinski sign. A quantification of
cerebellar motor signs was also performed using the International Cooperative Ataxia Rating Scale
(ICARS, Trouillas et al., 1997), whose global score ranges from 0 (absence of any motor deﬁcit) to
100 (presence of motor deﬁcit at the highest degree). Demographic and clinical characteristics of
the patients are reported in Table 1.
A group of 25 healthy subjects (HS) [F/M=19/6] ranging from 40 to 60 years of age [mean age ±
SD = 53.8 ± 5.9 years] with no history of neurological or psychiatric illness were also recruited as
control group for the MRI protocol.
This research study was approved by the Ethics Committee of Santa Lucia Foundation, according to
the principles expressed in the Declaration of Helsinki. Written informed consent was obtained
from each subject.
SCA2 patients underwent a neuropsychological evaluation according to the evidence that in this
pathology specific cognitive domain are mainly involved, in particular visuospatial, executive and
attentional abilities (LePira et al., 2002; Kawai et al., 2008). For the cognitive assessment, the
following tests were used and then grouped by the functional domains that were measured: (Lezak,
- Wechsler Adult Intelligent Scale–revised Intelligent Quotient (Wechsler, 1981; Orsini and
Laicardi, 1997, 2003) and Raven’47 progressive matrices test (Raven, 1949) to analyze the
- Rey-Osterrieth Complex Figure Test (recall and copy) (Caffarra, 2002), forward and backword
Corsi (Corsi, 1972), and Wechsler Adult Intelligent Scale -revised block design subtest (Wechsler,
1981; Orsini and Laicardi, 1997, 2003) to analyze the visuospatial ability;
- Stroop Test (“time effect” and “error effect”) (Caffarra, 2002), semantic, phonological and verbal
fluency (Borkowskyet al., 1967), Wisconsinn Card Sorting Test (WCST) (Heaton et al., 2002), and
Tower of London procedure (TOL) (Krikorian et al., 1994), to analyze the executive functions;
-Trail Making Test B-A (Giovagnoli et al., 1996) to analyze the attention abilities.
The results of the neuropsychological assessment are reported in Table 2.
MRI acquisition protocol
The MRI examination was performed by using a 3T scanner (Magnetom Allegra, Siemens,
Erlangen, Germany) and the following scans were acquired: 1) dual-echo turbo spin echo [TSE]
(TR = 6190 ms, TE = 12/109 ms); 2) fast-FLAIR (TR = 8170 ms, 204TE = 96 ms, TI = 2100 ms);
3) 3D Modiﬁed Driven Equilibrium Fourier Transform (MDEFT) scan (TR = 1338 ms, TE = 2.4
ms, Matrix = 256 × 224×176, in-plane FOV=250×250mm2, slice thickness=1 mm); 4) diffusion
weighted Spin-Echo Echo Planar Imaging (SE EPI) along 61 non-collinear directions (TR=7 s,
TE=85 ms, b factor=1000 s/mm2, 45 contiguous slices volumes with a 2.3mm3 isotropic
reconstructed voxel size). Nine volumes without diffusion weighting (b=0) were also acquired.
MRI imaging and data analyses
Affine registration to the first non-diffusion weighted volume using FSL was done on DTI volumes
to correct for eddy currents and small head movements (Smith et al., 2004). After brain
segmentation with the BET utility (Smith, 2002), the diffusion tensor (DT) coefficients were
computed in Camino (Cook et al., 2006) to generate whole brain maps of the dMRI metrics,
including FA and MD. Additionally, to better characterize the tissue microstructure changes AD
and RD were also analyzed. Each FA volume was registered to the native space MDEFT volume
with a linear registration first, followed by a non-linear transformation. The target for the linear
registration was the skull-stripped MDEFT, while the original volume (including skull). was the
target for the non-linear transformation. The registration was achieved using the tools FLIRT
(Jenkinson et al., 2002) and FNIRT (Andersson et al., 2008) from FSL. This “FA to MDEFT”
transformation was combined with each individual “MDEFT to MNI” transformation, obtained by
non-linear registration of the MDEFT to the ICBM152 MNI template. This resulted in the final
transformation from each participant’s DTI space to the ICBM152 MNI template.
diffusion MRI based Tractography
It has been pointed out previously (Ye et al., 2013) that tractography based on DTI is not able to
adequately segment the SCP, and particularly the decussation of the SCP. Here MCP and SCP were
reconstructed using tractography based on two multi-fiber models implemented in Camino.
Specifically, QBall (Tuch, 2004) was used for MCP, as it provides less false positive fiber
components while PAS MRI (Janson and Alexander, 2003) was used for left (LSCP) and right
(RSCP) SCP, as it deals more effectively than QBall with fiber crossing. This procedure was
optimized in a previous study from our group (Dayan et al., 2016).
Once the multi-fiber directions were estimated, probabilistic tractography was carried out based on
these data using the PICo algorithm. N = 10000 tracking iterations were performed from each voxel
of the seed Region of Interest (ROI) with stopping criteria of FA ≤ 0:1 and curving angle ≤ 80°.
Five ROIs were manually drawn on the FA map images for MCP tracking. Cerebellar tract
reconstruction was performed using the same approach as in Dayan and colleagues (2016). The
SCP was segmented separately for each cerebellar hemisphere and two endpoint ROIs were chosen
so as to select all the fibers that continue posteriorly to the seed ROI, centered in the dentate
nucleus, and to include both the red nucleus and its medial area, contralaterally, where the SCP is
known to decussate (see Dayan et al., 2016).
In order to obtain a binary map of the “average tract”, every subject’s reconstructed MCP, LSCP
and RSCP maps were binarised using a probability threshold for probability index of connectivity
(PICo) maps computed by in-house software to minimize the amount of tract volume variation with
PICo threshold. These images were then warped into standard space using the FA to ICBM152
MNI space transformation previously calculated, and averaged. The resulting maps were
thresholded to retain only those voxels that were common to at least 50% of subjects.
For each tests, individual raw scores were converted to obtain a mean Z-score for each functional
domain. For the tests that lacked published normative data, individual z-scores were calculated with
reference to specific control groups using the following formula: (subject raw score – population
mean)/population standard deviation (SD). Demographic and performance data of control groups
are reported in Table 3.
Published normative data were used for the following tests: Rey-Osterrieth Complex Figure Test,
(recall and copy versions), Raven’47 progressive matrices and Trail Making Test. No control
subject had history of neurological or psychiatric illness, and all controls were well matched with
regard to age and education (independent-sample t-test: p= not significant).
Voxel-wise analysis on white matter tracts
A voxel-wise analysis was performed in order to compare FA and MD changes differences in the
white matter between SCA2 patients and HS, restricting the comparison to the voxels of the MCP
and SCP, based on the average tract masks obtained as described above. T-contrasts were evaluated
with voxel significance set at p < 0.001 and corrected for family-wise error (FWE) at cluster level
with significance level chosen for p < 0.05. In order to better characterize the tissue microstructure
both FA and MD were used, while AD and RD were analyzed to help the interpretation of changes
to FA and MD (Alexander et al., 2007).
To remove the effect of confounding variables, the analysis was adjusted for age, since statistically
significant difference was found between patients and controls as assessed by the t-test analysis
(T=-2.23422; p=0.03). Although there was no difference in gender distribution between groups
(chi-square= 0.2962, df=1, p=0.58), sex was also set as covariate.
In order to investigate the relationship between WM damage and cognitive impairment Spearman
rank-order correlation coefficient was used to analyze possible correlations between individual
values of WM diffusivity, extracted using FSL command line from the FMRIB software library
(FSL, www.fmrib.ox.ac.uk/fsl/) and the correspondent neuropsychological scores.
Subjects with cerebellar damage had negative Z-scores for all cognitive domain except for
Attention (0.14) (Fig 1).
The results of tractography were visually evaluated in every participant. In order to be deemed
successful, the segmentation had to fulfill the following criteria: the MCP included the transverse
pontine fibers both posterior and anterior to the corticospinal tracts and the SCP decussation was
visible at the level of the midbrain, as expected from known anatomy (see also Dayan et al., 2016).
Based on this procedure, MCP and SCPs were successfully reconstructed in all patients and HS. Fig
2 shows the 3D fiber reconstruction for the average MCP and SCP of both groups of subjects.
Voxel-wise comparisons between patients and HS were performed for each diffusion metric
separately in each tract (MCP, LSCP, RSCP). WM analysis showed a widespread pattern of WM
diffusivity alterations to affect MCP, LSCP and RSCP. Indeed, when compared to controls, SCA2
patients showed a significant decrease of FA and a significant increase of MD in all tracts examined.
When compared to controls, SCA2 patients also showed a significant increase of both AD and RD
to affect both the MCP and SCP, bilaterally.
Results are illustrated in Fig 3 and detailed statistics are reported in Table 4.
Regarding the relationship between WM damage and patients’ cognitive scores, Spearman’s
correlation analysis showed mean LSCP MD to be negatively correlated with the Visuospatial
ability (R= -0.67; p=0.05) and mean RSCP MD to be negatively correlated with Executive function
(R= -0.67; p=0.05).
No correlation was detected between FA and cognitive impairment.
In the present study we aimed to investigate the pattern of WM changes associated with cerebellar
degeneration in SCA2 patients. dMRI based tractography was used to reconstruct the main
cerebellar WM tracts, namely MCP and SCP, and then to evaluate DTI metrics within those tracts.
Specifically, FA decrease and MD increase were found bilaterally in MCP and SCP of SCA2
patients compared to controls. This pattern is consistent with the presence of microstructural white
matter damage, although at this stage we can only speculate on the underlying pathology. Since the
examination of multiple DTI measures may provide more specific information about the tissue
microstructure (Alexander et al., 2007), AD and RD were also examined, in order to investigate the
combination of AD and RD changes that may underlie the decreased FA in the examined tracts. In
the present study, an increase of both AD and RD was also found to affect bilaterally MCP and SCP
of SCA2 patients.
FA is widely recognized as a marker of so-called white matter integrity and thus it is used as the
primary measure of tissue microstructural damage. Both demyelination and axonal loss can result in
reduced FA (Beaulieu, 2002; Song et al., 2002): indeed, when axons packing is not sufficiently
dense, more intercellular water will result in a less restriction of diffusion and therefore lowering
FA (Feldman et al., 2010). Although FA is a highly sensitive measure of tissue microstructure, it is
a non-specific biomarker of neuropathology. Further insight can be gained from the evaluation of
MD. Conversely to FA, MD has been shown to negatively correlate with fiber integrity (Beaulieu,
2002; Song et al., 2002). Although the exact pathological correlates of increased MD cannot be
established, it has been proposed that it predominantly reflects wallerian degeneration. Present
findings of MD increase thus might reflect fiber degeneration within the cerebellar peduncle as
often reported in SCA2 patients (Durr et al., 1995; Gilman et al., 1996; Iwabuchi et al., 1999;
Estrada et al., 1999; Pang et al., 2002).
While RD increase is largely accepted to correlate with myelin disruption (Alexander et al., 2007;
Feldman et al., 2010), human studies found increased AD in association with axonal degeneration
(Hasan et al., 2008; Roosendaal et al., 2009; Metwalli et al., 2010). Consistently, our data suggest
an increase in AD of SCA2 patients, which may reflect fiber degeneration. Similar results were
observed in patients affected by Friedriech’s ataxia (FRDA) (Della Nave et al., 2011). Indeed, it
has been postulated that the RD values increase with loss of myelin integrity in chronic
pathological conditions while the AD values increase in areas with reduced axonal density or
caliber (Hasan et al., 2008; Kumar et al., 2008, 2010; Della Nave et al., 2011). Further support to
this interpretation comes from DTI studies on murine models of acute lesions (Song et al., 2003;
Budde et al., 2007, 2009; Kim et al., 2010) suggesting that AD changes are characterized by a
time-dependent course with initial decrease during the acute phase followed by normalization or
increase in more advanced stages as axon fragments are cleared (Concha et al., 2006).
Consistently, increases of AD have been shown in chronically degenerated white matter bundles in
humans (Pierpaoli et al., 2001; Glenn et al., 2003). Thus, a significant increase of AD may feature
either in direct WM damage, i.e stroke and Multiple Sclerosis, (Bammer et al., 2000; Pierpaoli et
al., 2001) or in the more advanced stage of wallerian degeneration (Mandelli et al., 2007). DTI
metrics modifications have been already reported in the inferior, middle and superior cerebellar
peduncles of SCA2 patients (Della Nave et al., 2008b; Hernandez-Castillo et al., 2015). Those
studies, however, were based on tract-based spatial statistics (TBSS) (Smith et al., 2006), thus
investigating voxel-wise differences of the WM along the core of tracts. By contrast, we evaluated
the DTI Metrics for the whole tract, restricting the analysis to the fibers of interest tracked in each
subject (see also Roine et at., 2015).
A further element of novelty in this study is the observed relationship between cerebellar
microstructural damage and cognitive abilities in SCA2 patients, which was not investigated
previously. Interestingly, we found that cognitive functions typically altered in SCA2, i.e executive
and visuospatial functions (LePira et al., 2002; Kawai et al., 2008), correlate with damage of right
and left SCP respectively. This datum is in line with cerebellar lateralization of functions (Stoodley
and Schmahmann, 2008). According to the crossed cerebello-cerebral projections, negative Z-
scores of visuospatial functions negatively correlated with increased MD values in the left SCP, the
main cerebellar output tract connecting the cerebellum with right cerebral cortex, known to be
predominantly involved in visuospatial processing (Baillieux et al., 2010). The correlation between
executive functions and MD values in the right SCP is consistent with the evidence that left
prefrontal cortex plays a crucial role in many critical aspects of executive processing such as
strategy (Grafman et al., 2005; Vallesi et al., 2012), hypothesis generation (Reverberi et al., 2005),
and manipulations of goal hierarchies (Kaller, et al., 2011; Reverberi et al., 2005; Crescentini et al.,
2011; Langdon and Warrington, 2000).
Overall, our results show a specific pattern of white matter microstructural damage, likely to be
associated with SCA2 neurodegenerative processes, as expressed by the decrease of FA and the
increase of MD affecting bilateral MCP and SCP. This suggests that in SCA2 the cerebellar
atrophy also affects the diffusivity of the main cerebellar white matter tracts thus resulting in a
cerebello-cerebral dysregulation that may account for the cognitive symptomatology of SCA2
patients. Consistent with this interpretation, a previous DTI study in FDRA individuals (Zalensky et
al., 2014) has specifically shown that the interruption of the cerebellar afferent and efferent
connections leads to secondary functional effects in distant cortical and subcortical regions, a
phenomenon referred to as reverse cerebellar diaschisis, thus resulting in the cerebellar cognitive
affective syndrome (Schmahmann and Sherman 1998).
Furthermore, the fact that we found a precise lateralization of cognitive functions and structural
alterations indicating significant correlations of visuospatial functions with RSCP MD and
executive functions with LSCP MD, provides additional support to our conclusions.
In light of our results, we hypothesize a combination of demyelination, as expressed by increased
RD, and axonal changes, as expressed by AD increase, to affect the tracts examined. Taken
together our data suggest some heterogeneity of microstructural WM damage in SCA2 as
suggested also by Della Nave and colleagues (2008b).
However, it has to be considered that in the case of more complex diseases, where a combination
of demyelination, axon loss, gliosis, and inflammation may affect brain regions, the use of
integrated approaches with other imaging measures (e.g., T1, T2, magnetization transfer,
perfusion, fast/slow diffusion, spectroscopy) could improve the DTI information on the
neuropathology and the interpretation of diffusivity changes (Alexander et al., 2007). According
to this evidence, our interpretation of combined diffusivity changes in SCA2 patients needs
further investigation and should be confirmed in future studies.
It is also important to discuss some methodological limitations. First of all, it is important to
reiterate that typically diffusion MRI data are acquired with a resolution of 2-3 mm3, which is too
coarse to capture fine anatomical details. In addition, all tractography algorithms are at high risk
of both false positives and false negatives. Furthermore, reconstructing the SCP is notoriously
difficult because of the crossing occurring at their decussation. Several tractographic
reconstructions have failed (Zhang et al., 2008; Ye et al., 2012). Ye and colleagues (2013) have
proposed a method based on random forest classification, which can be trained to segment the
SCPs based on the shape of the diffusion tensor in every voxel. Instead, we used two
sophisticated models of diffusion, namely Q-ball and PAS MRI to rely exclusively on
tractography for the segmentation. Overall, this approach allowed us to independently analyze
which voxels along the tracts presented with DTI metrics changes.
The present findings are largely consistent with MRI studies that have investigated dMRI metrics
in patients affected by cerebellar atrophy of different etiology. Indeed, patterns of diffusivity WM
changes to affect FA and MD of MCP and/or SCP have been evidenced in SCA1, SCA2
(Mandelli et al., 2007) and SCA6 (Ying et al., 2006; Ye et al., 2013) patients. With the exception
of the study from Ye and colleagues (2013), that examines the FA and MD in SCA6 within the
whole volumes of the MCP and SCPs, all these studies measured dMRI metrics of the MCP and
SCPs in a Regions of Interest (ROI) limited to a single slice. The use of single slice can introduce
important bias and variations in the metric averaged over that region, depending on the particular
ROI position (Dayan et al., 2016). The use of tractography allows the whole bundle to be taken
into account, overcoming this limitation.
Furthermore, the present study demonstrates the sensitivity of dMRI to detect the microstructural
alterations linked to cognitive symptomatology in SCA2 patients. The additional correlation with
cognitive scores further support the idea that DTI metrics may serve as clinical imaging
biomarker to differentiate between different kinds of cerebellar neurodegenerative diseases, as
previously suggested in another study o f our group with mixed cerebellar ataxias (Dayan et al.,
In conclusion, we advance the hypothesis that microstructural WM damage in cerebellar
peduncles may account for the specificity of the cognitive impairment observed in SCA2 patients.
This work was supported by the Ministry of Education, Universities and Research (MIUR).-
(Grant Number C26A1329AR) to ML, and Ministry of Health (Grant Number RF-2011-
02348213) to MM and (Grant Number GR-2013-02354888) to SC.
Giusy Olivito, Maria Leggio, Marco Molinari and Marco Bozzali contributed to the study
conception and design and supervised development of work;
Marcella Masciullo and Silvia Romano contributed to recruitment and enrollment of patients;
Michela Lupo, Claudia Iacobacci and Silvia Clausi contributed to the acquisition and
interpretation of neuropsyhological data;
Michela Lupo contributed to the analysis of neuropsychological data;
Mara Cercignani contributed to the implementation of MRI protocol and analysis;
Giusy Olivito contributed to acquisition of MRI protocol, preprocessing and analysis of MRI
Giusy Olivito contributed to the writing of the original manuscript;
All co-authors contributed to final editing and critical revision of the original manuscript.
Conflict of Interest
Conflicts of interest: none'.
Alexander AL, Hasan K, Kindlmann G, Parker DL, Tsuruda JS (2000), A geometric comparison of
diffusion anisotropy measures. Magn Reson Med 44:283–291.
Alexander AL, Lee JE, Lazar M, Field AS (2007), Diffusion Tensor Imaging of the Brain.
Neurotherapeutics 4(3): 316–329.
Andersson J, Smith S, Jenkinson M (2008), Fnirt-fmrib’s non-linear 24 image registration tool. In:
14th Annual Meeting of the Organization for Human Brain Mapping; 15–9.
Baillieux H, De Smet HJ, Dobbeleir A, Paquier PF, De Deyn PP, Mariën P (2010), Cognitive and
affective disturbances following focal cerebellar damage in adults: a neuropsychological and
SPECT study. Cortex 46(7):869-79.
Bammer R, Augustin M, Strasser-Fuchs S, Seifert T, Kapeller P, Stollberger, R, Ebner F, Hartung
HP, Fazekas F (2000), Magnetic resonance diffusion tensor imaging for characterizing diffuse and
focal abnormalities in multiple sclerosis. Magn Reson Med 44:583–591.
Basser PJ, Mattiello J, LeBihan D (1994), MR diffusion tensor spectroscopy and imaging. Biophys
Beaulieu C (2002), The basis of anisotropic water diffusion in the nervous system - a technical
review. NMR Biomed 15:435–55.
Boni S, Valle G, Ciuffi RP, Sonetti MG, Perrone E, Tofani A, et al. (1992), Crossed cerebello-
cerebral diaschisis: a SPECT study. Nucl Med Commun 13:824–831.
Borkowsky JG, Benton AL, Spreen O (1967), Word fluency and brain-damage. Neuropsychologia
Brenneis C, Bosch SM, Schocke M, Wenning GK, Poewe W (2003), Atrophy pattern in SCA2
determined by voxel-based morphometry. Neuroreport 14:1799-1802.
Broich K, Hartmann A, Biersack HJ, Horn R (1987), Crossed cerebello-cerebral diaschisis in a
patient with cerebellar infarction. Neurosci Lett 83(1-2):7-12.
Budde MD, Kim JH, Liang HF et al. (2007), Toward accurate diagnosis of white matter pathology
using diffusion tensor imaging. Magn Reson Med 57:688–695.
Budde MD, Xie M, Cross AH et al. (2009), Axial diffusivity is the primary correlate of axonal
injury in the EAE spinal cord: a quantitative pixelwise analysis. J Neurosci 29:2805–2813.
Caffarra P, Vezzadini G, Dieci F, Zonato F, Venneri A (2002), Rey-Osterrieth complex figure:
normative values in an Italian population sample. Neurol Sci 22:443-7.
Concha L, Gross DW, Wheatley M et al. (2006), Diffusion tensor imaging of time dependent
axonal and myelin degradation after corpus callosotomy in epilepsy patients. NeuroImage 32:1090–
Cook PA, Bai Y, Gilani NS, Seunarine KK, Hall MG, Parker GJ, et al. (2006), Camino: Open-
source diffusion-MRI reconstruction and processing. In: 14th Scientific Meeting of the International
Society for Magnetic Resonance in Medicine; Seattle, USA.
Corsi PM (1972), Human memory and the medial temporal regions of the brain. Dissert Abst Int
Crescentini C, Seyed-Allaei S, de Pisapia N, Jovicich J, Amati D, Shallice T (2011), Mechanisms of
rule acquisition and rule following in inductive reasoning. Journal of Neuroscience 31:7763-7774.
Dayan M, Olivito G, Molinari M, Cercignani M, Bozzali M, Leggio M (2016), Impact of cerebellar
atrophy on cortical gray matter and cerebellar peduncles as assessed by voxel-based morphometry
and high angular resolution diffusion imaging. Funct Neurol 31(4):239-248.
Della Nave R, Ginestroni A, Diciotti S, Salvatore E, Soricelli A, Mascalchi M (2011), Axial
diffusivity is increased in the degenerating superior cerebellar peduncles of Friedreich's ataxia.
Della Nave R, Ginestroni A, Tessa C, Cosottini M, Giannelli M, Salvatore E, Sartucci F, De
Michele G, Dotti MT, Piacentini S, Mascalchi M (2008a), Brain structural damage in
spinocerebellar ataxia type 2. A voxel-based morphometry study. Mov Disord 23(6):899-903.
Della Nave R, Ginestroni A, Tessa C, Salvatore E, De Grandis D, Plasmati R, Salvi F, De Michele
G, Dotti MT, Piacentini S, Mascalchi M (2008b), Brain white matter damage in SCA1 and SCA2.
An in vivo study using voxel-based morphometry, histogram analysis of mean diffusivity and tract-
based spatial statistics. Neuroimage. 43(1):10-9.
Durr A, Smadja D, Cancel G, Lezin A, Stevanin G, Mikol J, et al. (1995), Autosomal dominant
cerebellar ataxia type I in Martinique (French West Indies),. Clinical and neuropathological analysis
of 53 patients from three unrelated SCA2 families. Brain 118:1573–81.
Estrada R, Galarraga J, Orozco G, Nodarse A, Auburger G (1999), Spinocerebellar ataxia 2
(SCA2),: morphometric analyses in 11 autopsies. Acta Neuropathol (Berl) 97: 306–10.
Fancellu R, Paridi D, Tomasello C, Panzeri M, Castaldo A, Genitrini S, Soliveri P, Girotti F (2013),
Longitudinal study of cognitive and psychiatric functions in spinocerebellar ataxia types 1 and 2. J
Neurol 260: 3134-43.
Feldman HM, Yeatman JD, Lee ES, Barde LHF, Gaman-Bean S (2010), Diffusion Tensor Imaging:
A Review for Pediatric Researchers and Clinicians. J Dev Behav Pediatr 31(4): 346–356.
Gilman S, Sima AA, Junck L, Kluin KJ, Koeppe RA, Lohman ME, et al. (1996), Spinocerebellar
ataxia type1 with multiple system degeneration and glial cytoplasmic inclusions. Ann Neurol 39:
Giovagnoli AR, Del Pesce M, Mascheroni S, Simoncelli M, Laiacona M, Capitani E (1996), Trail
making test: normative values from 287 normal adult controls. Ital J Neurol Sci 17:305-9.
Glenn OA, Henry RG, Berman JI, Chang PC, Miller SP, Vigneron DB, et al. (2003), DTI-based
three-dimensional tractography detects differences in the pyramidal tracts of infants and children
with congenital hemiparesis. J Magn Reson Imaging 18:641– 648.
Grafman J, Spector L, Rattermann M J (2005), Planning and the brain. In R. Morris & G. Ward
(Eds.), The cognitive psychology of planning. Psychology Press 181-188.
Hasan KM, Halphen C, Boska MD, Narayana PA (2008), Diffusion tensor metrics, T2 relaxation,
and volumetry of the naturally aging human caudate nuclei in healthy young and middle-aged
adults: possible implications for the neurobiology of human brain aging and disease. Magn Reson
Heaton RK, Chelune GJ, Talley JL, Kay GG, Curtiss G (2000), WCST: Wisconsin Card Sorting
Test. Forma completa revisionata. Adattamento italiano a cura di Hardoy MC, Carta MG, Hardoy
MJ e Cabras PL. Ed. It. O.S. Organizzazioni Speciali. Firenze.
Hernandez-Castillo CR, Galvez V, Mercadillo R, Diaz R, Campos-Romo A, Fernandez-Ruiz J.
(2015), Extensive White Matter Alterations and Its Correlations with Ataxia Severity in SCA 2
Patients. PLoS One, 10, e0135449.
Iwabuchi K, Tsuchiya K, Uchihara T, Yagishita S (1999), Autosomal dominant spinocerebellar
degenerations. Clinical, pathological, and genetic correlations. Rev Neurol (Paris) 155:255–70.
Janson KM, Alexander DC (2003), Persistent angular structure: new insights from diffusion
magnetic resonance imaging data. Inf Process Med Imaging 19:1031-1046.
Jenkinson M, Bannister P, Brady M, et al. (2002), Improved optimization for the robust and
accurate linear registration and motion correction of brain images. NeuroImage 17:825-41.
Kaller CP, Rahm B, Spreer J, Weiller C, Unterrainer JM (2011), Dissociable contributions of left
and right dorsolateral prefrontal cortex in planning. Cerebral Cortex 21:307-317.
Kawai Y, Suenaga M, Watanabe H, Sobue G (2009), Cognitive impairment in spinocerebellar
degeneration. Review. Eur Neurol 61(5):257-68.
Kim JH, Loy DN, Wang Q et al. (2010), Diffusion tensor imaging at 3 hours after traumatic spinal
cord injury predicts long-term locomotory recovery. J Neurotrauma 27:587–598.
Klinke I, Minnerop M, Schmitz-Hübsch T, Hendriks M, Klockgether T, Wüllner U, Helmstaedter C
(2010), Neuropsychological features of patients with Spinocerebellar Ataxia (SCA), Types 1,2,3,
and 6. Cerebellum 9:433-442.
Komaba Y, Osono E, Kitamura S, Katayama Y (2000), Crossed cerebellocerebral diaschisis in
patients with cerebellar stroke. Acta Neurol Scand 101(1):8-12.
Krikorian R, Bartok J, Gay N (1994), Tower of London procedure: a standard method and
developmental data. J Clin Exp Neuropsychol 16(6):840-50.
Kumar R, Macey PM, Woo MA, Alger JR, Harper RM (2008), Diffusion tensor imaging
demonstrates brainstem and cerebellar abnormalities in congenital central hypoventilation
syndrome. Pediatr Res 64(3):275–280.
Kumar R, Macey PM, Woo MA, Harper RM (2010), Rostral brain axonal injury in congenital
central hypoventilation syndrome. J Neurosci Res 88(10):2146–2154.
Langdon D, Warrington EK (2000), The role of the left hemisphere in verbal and spatial reasoning
tasks. Cortex 36:691-702.
Le Pira F, Zappalà G, Saponara R, Domina E, Restivo D, Reggio E, Nicoletti A, Giuffrida S (2002),
Cognitive findings in spinocerebellar ataxia type 2: relationship to genetic and clinical variables. J
Neurol Sci 201(1-2):53-7.
Lezak MD (1995), Neurpsychological assessment. 3rd edition. New York: Oxford University Press.
Mandelli ML, De Simone T, Minati L, Bruzzone MG, Mariotti C, Fancellu R, Savoiardo M, Grisoli
M (2007), Diffusion tensor imaging of spinocerebellar ataxias types 1 and 2. AJNR Am J
Metwalli NS, Benatar M, Nair G, Usher S, Hu X, Carew JD (2010), Utility of axial and radial
diffusivity from diffusion tensor MRI as markers of neurodegeneration in amyotrophic lateral
sclerosis. Brain Research 1348:156-164.
Moriarty A, Cook A, Hunt H, Adams ME, Cipolotti L, Giunti P (2016), A longitudinal investigation
into cognition and disease progression in spinocerebellar ataxia types 1, 2, 3, 6, and 7. Orphanet J
Rare Dis 11: 82.
Orsini A, Laicardi C (1997), WAIS-R. Contributo alla taratura Italiana, Firenze Organizzazioni
Orsini A, Laicardi C (2003), Wais-r e terza età, Firenze Organizzazioni Speciali.
Pang JT, Giunti P, Chamberlain S, An SF, Vitaliani R, Scaravilli T, Martinian L, Wood NW,
Scaravilli F, Ansorge O (2002), Neuronal intranuclear inclusions in SCA2: a genetic, morphological
and immunohistochemical study of two cases. Brain 125:656–663.
Pierpaoli C, Barnett A, Pajevic S, Chen R, Penix Virta A, Basser P (2001), Water diffusion changes
in Wallerian degeneration and their dependance on white matter architecture. NeuroImage 13:1174–
Raven JC (1949), Progressive matrices, Sets A, Ab, B: Board and Book forms, LondonLewis.
Reverberi C, Lavaroni A, Gigli GL, Skrap M, Shallice T (2005), Specific impairments of rule
induction in different frontal lobe subgroups. Neuropsychologia 43:460-472.
Roine U, Salmi J, Roine T, Wendt TN, Leppämäki S, Rintahaka P, Tani P, Leemans A, Sams M
(2015), Constrained spherical deconvolution-based tractography and tract-based spatial statistics
show abnormal microstructural organization in Asperger syndrome. Mol Autism 6:4
Roosendaal SD, Geurts JJG, Vrenken H, Hulst HE, Cover KS, Castelijns JA, Pouwels PJW,
Barkhof F (2009), Regional DTI differences in multiple sclerosis patients. Neuroimage 44(4):1397-
Schmahmann JD, Sherman JC (1998), The cerebellar cognitive affective syndrome. Brain 121:561–
Smith S, Jenkinson M, Woolrich M, Beckmann C, Behrens T, Johansen-Berg H, et al. (2004),
Advances in functional and structural MR image analysis and implementation as fsl. Neuroimage
23, S208–Conference on Mathematics in Brain Imaging, Jul 12-23, Los Angeles, CA.
Smith SM (2002), Fast robust automated brain extraction. Hum Brain Mapp 17:143-55.
Smith SM, Jenkinson M, Johansen-Berg H, Rueckert D, Nichols TE, Mackay CE, et al. (2006),
Tract-based spatial statistics: voxelwise analysis of multi-subject diffusion data. Neuroimage 31:
Song SK, Sun SW, Ramsbottom MJ, Chang C, Russell J, Cross AH (2002), Dysmyelination
revealed through MRI as increased radial (but unchanged axial), diffusion of water. Neuroimage
Song SK, Sun SW, Ju WK et al. (2003), Diffusion tensor imaging detects and differentiates axon
and myelin degeneration in mouse optic nerve after retinal ischemia. NeuroImage 20:1714–1722.
Takahashi T, Katada S, Onodera O (2010), Polyglutamine diseases: wheredoes toxicity come from?
what is toxicity? where are we going? J Mol Cell Biol 2:180-191.
Trouillas P, Takayanagi T, Hallett M, Currier RD, Subramony SH, Wessel K, Bryer A, Diener HC,
Massaquoi S, Gomez CM, Coutinho P, Ben Hamida M, Campanella G, Filla A, Schut L, Timann D,
Honnorat J, Nighoghossian N, Manyam B (1997), International Cooperative Ataxia Rating Scale
for pharmacological assessment of the cerebellar syndrome. The Ataxia Neuropharmacology
Committee of the World Federation of Neurology. J Neurol Sci 145:205-11.
Tuch DS (2004), Q-ball imaging. Magn Reson Med 52:1358-1372.
Vallesi A (2012), Organisation of executive functions: Hemispheric asymmetries, Journal of
Cognitive Psychology 24(4):367-386.
Vieira Karuta SC, Raskin S, de Carvalho Neto A, Gasparetto EL, Doring T, Teive HA (2015),
Diffusion tensor imaging and tract-based spatial statistics analysis in Friedreich's ataxia patients.
Parkinsonism Relat Disord 21(5):504-8.
Zalesky A, Akhlaghi H, Corben LA, Bradshaw JL, Delatycki MB, Storey E, Georgiou-Karistianis
N, Egan GF (2014), Cerebello-cerebral connectivity deficits in Friedreich ataxia. Brain Structure
and Function 219(3):969-81.
Wechsler D (1981), Scala di intelligenza Wechsler per adulti rivisitata (WAIS-R), Manuale, Firenze
Ye C, Bazin PL, Bogovic JA, Ying SH, Prince JL (2012), Labeling of the cerebellar peduncles
using a supervised Gaussian classifier with volumetric tract segmentation. Proceedings of SPIE
Medical Imaging 8314:143.
Ye C, Bogovic JA, Ying SH, Prince JL (2013), Segmentation of the complete superior cerebellar
peduncles using a multi-object geometric deformable model. 2013 IEEE 10th International
Symposium on Biomedical Imaging (ISBI) 49–52.
Ying SH, Choi SI, Perlman SL, Baloh RW, Zee DS, Toga AW (2006), Pontine and cerebellar
atrophy correlate with clinical disability in SCA2. Neurology 66:424-426.
Zhang H, Avants BB, Yushkevich PA, Woo JH, Wang S, McCluskey LF, Elman LB, Melhem ER,
Gee JC (2007), High-dimensional spatial normalization of diffusion tensor images improves the
detection of white matter differences: an example study using amyotrophic lateral sclerosis. IEEE
Transactions on Medical Imaging 26(11):1585–1597.
Fig.1 Neuropsychological assessment. Mean and Standard Error of cognitive functions in the SCA
2 group expressed in Z-scores (mean Z-scores are reported). Neuropsychological functions
are grouped according to the cognitive domains assessed.
Fig.2 DTI-based tractography of middle and superior cerebellar peduncles. 3D reconstruction
of the average tract of MCP (red) LSCP (green) and RSCP (blue) with voxels belonging to at least
50% of the subjects. Reconstructed tracts are superimposed on the Spatially Unbiased Template of
the Cerebellum and Braistem (SUIT) (Diedrichsen et al., 2009). The decussation of the SCPs at
level of the ventral brainstem is clearly visible. A: anterior view; P: posterior view; S: superior
Fig.3 Voxel-wise analysis of white matter tracts. Regions showing altered Fractional Anisotropy
(FA), Mean Diffusivity (MD) , Axial Diffusivity (AD) and Radial diffusivity (RD) in patients
compared to controls. Only clusters of significant diffusivity changes that survived after correction
for multiple comparisons are reported. Results are shown in different colors for the middle
cerebellar peduncle (red), left superior cerebellar peduncle (green), and right superior cerebellar
peduncle (blue), Clusters are superimposed on MNI (the Montreal Neurological Institute) template
in coronal, sagittal and axial slices.
*FWE uncorrected at p<0.001 at cluster level
1) Cerebellar microstructural damage affects Middle and Superior Cerebellar Peduncle in
2) Diffusivity changes predominantly reflect the white matter degeneration typically observed
in cerebellar atrophies
3) Diffusivity changes may reflect a dysregulation of cerebello-cerebral interaction
4) The pattern of cerebellar white matter damage is associated with impaired cognitive
performances of patients
Table 1. Demographic characteristics and motor deficit scores of the patients.
Education Gender Years of illness CGA Repeats ICARS TOTAL
Table 1. The table reports for each patients age, education, gender, years of illness, CGA repeats
and total motor scores as assessed by the International Cooperative Ataxia Rating Scale (ICARS)
(Trouillas et al., 1997) Means scores and standard deviations (SD) are also reported.
Table 2. Patients’ Neuropsychological raw scores.
CA1 74 30 34 4.5 4 - 12 37.5 0
9 39 15 9 19 25 2
CA2 81 29 33 13 5 4 12 15.5 0 2
5 27 19 28 48 25 -
CA3 85 33 32 6.5 6 4 16 14.0
8 -0.5 1
9 44 24 7 18 24 45
CA4 91 35 31 13.5 4 4 22
1 29 13 4 7 30 48
CA5 82 30 27 8.5 5 4 8 31 0 2
3 27 12 8 16 32 49
CA6 98 29 28 8.5 5 5 28 9.5 0 3
0 42 19 5 8 27 68
CA7 75 30 30 9.5 6 6 10 25.5 1
2 23 6 18 44 27 156
CA8 91 34 33 22 6 5 23 12.5 0 1
8 19 10 6 18 34 192
CA9 84 34 35 16 7 6 21 39.5 0 2
1 24 10 15 29 34 66
TIQ : Total Intellectual Quotient; PM: Progressive matrices; fRey: Rey-Osterrieth Complex Figure;
FC: forward Corsi; BC: backword Corsi; BD: block design subtest; SF: semantic fluency; FF:
phonological fluency; VF: verbal fluency ; WCST: Wisconsinn Card Sorting Test (PErr:
perseverative errors; TErr: total errors); TOL: Tower of London; TMT : Trail Making Test.
Table 3. Demographic and performances data of the different control groups for each test.
Forward digit span
ward digit span
Tower of London
Stroop Time effect
WCST n° errors
“ n° perseverative errors
Table 4. Statistics of whole brain voxel-wise analysis within the reconstructed tracts
d at p<