Demyelination and degeneration in the injured human spinal cord detected with
diffusion and magnetization transfer MRI
J. Cohen-Adada,b,⁎, M-M. El Mendilia, S. Lehéricyc, P-F. Pradatd, S. Blanchoe, S. Rossignolf, H. Benalia
aUMR-678, INSERM-UPMC, Pitié-Salpêtrière Hospital, Paris, France
bA.A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Harvard Medical School, Charlestown, MA, USA
cCentre for Neuroimaging Research (CENIR), Centre de Recherche de l'Institut du Cerveau et de la Moelle epiniere, UPMC, UMR-S975, INSERM U975, CNRS UMR 7225,
Groupe Hospitalier Pitie-Salpetriere, Paris, France
dFédération des Maladies du Système Nerveux, AP-HP, Pitié-Salpêtrière Hospital, Paris, France
eInstitut pour la Recherche sur la Moelle Epinière et l'Encéphale, France
fGRSNC, Faculty of Medicine, Université de Montréal, Montreal, QC, Canada
a b s t r a c ta r t i c l ei n f o
Received 9 September 2010
Revised 15 November 2010
Accepted 17 November 2010
Available online 11 January 2011
Spinal cord injury
Characterizing demyelination/degeneration of spinal pathways in traumatic spinal cord injured (SCI) patients
is crucial for assessing the prognosis of functional rehabilitation. Novel techniques based on diffusion-
weighted (DW) magnetic resonance imaging (MRI) and magnetization transfer (MT) imaging provide
sensitive and specific markers of white matter pathology. In this paper we combined for the first time high
angular resolution diffusion-weighted imaging (HARDI), MT imaging and atrophy measurements to evaluate
the cervical spinal cord of fourteen SCI patients and age-matched controls. We used high in-plane resolution
to delineate dorsal and ventrolateral pathways. Significant differences were detected between patients and
controls in the normal-appearing white matter for fractional anisotropy (FA, pb0.0001), axial diffusivity
(pb0.05), radial diffusivity (pb0.05), generalized fractional anisotropy (GFA, pb0.0001), magnetization
transfer ratio (MTR, pb0.0001) and cord area (pb0.05). No significant difference was detected in mean
diffusivity (p=0.41), T1-weighted (p=0.76) and T2-weighted (p=0.09) signals. MRI metrics were
remarkably well correlated with clinical disability (Pearson's correlations, FA: pb0.01, GFA: pb0.01, radial
diffusivity: p=0.01, MTR: p=0.04 and atrophy: pb0.01). Stepwise linear regressions showed that measures
of MTR in the dorsal spinal cord predicted the sensory disability whereas measures of MTR in the ventro-
lateral spinal cord predicted the motor disability (ASIA score). However, diffusion metrics were not specific to
the sensorimotor scores. Due to the specificity of axial and radial diffusivity and MT measurements, results
suggest the detection of demyelination and degeneration in SCI patients. Combining HARDI with MT imaging
is a promising approach to gain specificity in characterizing spinal cord pathways in traumatic injury.
© 2011 Elsevier Inc. All rights reserved.
Sensorimotor impairments after spinal cord injuries (SCI) largely
the white matter that are distributed throughout the various quadrants
of the spinal cord. The dorsal columns are clearly delineated and contain
tracts ascend more laterally and carry sensory information to the
cerebellum (ventral and dorsal spinocerebellar tracts) or the thalamus
(lateral and spinothalamic tracts). Major descending tracts are located
laterally (corticospinal and rubrospinal) and ventrally and carry
information mainly from the vestibular system, the reticular system
and some direct ipsilateral corticospinal projections.
Traumatic lesions (including primary and secondary lesions) not
only can induce a physical discontinuity of the tracts but also
anterograde wallerian demyelination as well as some retrograde
degeneration. After SCI, some pathways may be preserved and
contribute to recovery of function. This could be achieved by
regeneration of pathways or sprouting of undamaged pathways
(Bareyre et al., 2004; Maier and Schwab, 2006; Rossignol et al., 2007).
Whereas in the first case, pathways are replaced by regenerated
fibers, in the second case, new connections are either made or
strengthened through existing structures. Thus, damage to the
corticospinal tract can be in part offset by sprouting new connections
through propriospinal or reticulospinal pathways, which then act
more or less as a new (or enhanced relay) between the cortex and the
NeuroImage 55 (2011) 1024–1033
Abbreviations: SCI, spinal cord injury; DW, Diffusion-Weighted; MT, Magnetization
Transfer; HARDI, High Angular Resolution Diffusion Imaging; FA, Fractional Anisotropy;
GFA, Generalized Fractional Anisotropy; MD, Mean Diffusivity; DTI, Diffusion Tensor
Imaging; QBI, Q-Ball Imaging; FOV, Field Of View.
⁎ Corresponding author. A.A. Martinos Center for Biomedical Imaging, Massachusetts
General Hospital, 149 Thirteen St, Charlestown, MA 02129, USA. Fax: +1 617 726 1383.
E-mail address: email@example.com (J. Cohen-Adad).
1053-8119/$ – see front matter © 2011 Elsevier Inc. All rights reserved.
Contents lists available at ScienceDirect
journal homepage: www.elsevier.com/locate/ynimg
spinal cord. It is thus important to develop prognostic imaging tools
that will allow the characterization of the damaged tracts and the
state of residual tracts.
Diffusion-weighted (DW) magnetic resonance imaging (MRI)
exploits signal attenuation of water molecules that diffuse preferen-
tially along white matter axons. It is possible to model this diffusion
profile using diffusion tensor imaging (DTI) (Basser et al., 1994) and
derive metrics of fractional anisotropy (FA), mean diffusivity (MD),
axial and radial diffusivities that provide sensitive biomarkers for
characterizing abnormality in the white matter. Animal studies
notably showed that axial and radial diffusivities are good predictors
of axonal loss and demyelination, respectively (Budde et al., 2007).
DTI has been applied to assess the severity of the spinal cord injury
(Agosta et al., 2007; Budde et al., 2007; Cohen-Adad et al., 2008a;
DeBoy et al., 2007; Deo et al., 2006; Ducreux et al., 2007; Ellingson
et al., 2008; Fujiyoshi et al., 2007; Kim et al., 2007; Lammertse et al.,
2007; Nevo et al., 2001; Ohgiya et al., 2007; Plank et al., 2007; Ries
et al., 2000; Schwartz et al., 2005; Shen et al., 2007; Thurnher and
Bammer, 2006; Valsasina et al., 2005; Vargas et al., 2007). As an
extension to DTI, high angular resolution diffusion imaging (HARDI)
and Q-Ball imaging (QBI) can represent more than one diffusion
direction, thereby alleviating limitations of the diffusion tensor in
presenceof crossing fibers(Tuch, 2004). HARDI hasproven efficient in
the detection of subtle axonal connections in the spinal cord (Cohen-
Adad et al., 2008b; Lundell et al., 2009) and HARDI-based metrics such
as the generalized fractional anisotropy (GFA) might be a good
surrogate of white matter pathology, as suggested in previous work
(Barmpoutis et al., 2009; Cohen-Adad et al., 2009b). One limitation of
DTI/HARDI in the characterization of white matter integrity however,
is the lack of specificity for determining demyelination and axonal
loss. Several physical parameters can influence diffusion metrics
including myelination, axonal density, axonal diameter, or orientation
of fiber bundles (Beaulieu, 2002; Sen and Basser, 2005). Therefore
combining DW-MRI with an independent measure that is sensitive to
demyelination would increase the reliability of diagnosis.
Magnetization transfer (MT) contrast is based on the interaction
between hydrogen protons bounded to macromolecules (e.g. lipid
constituted of axons myelin sheet), thereby providing an indirect
surrogate for myelin content (Kucharczyk et al., 1994; Pike et al.,
2000). One advantage of MT is its specificity to demyelination and
degeneration, as assessed by histopathology (Mottershead et al.,
2003; Schmierer et al., 2004). Using high resolution MT measure-
ments in the spinal cord, it is possible to assess demyelination of
specific spinal pathways, as shown in MS patients (Zackowski et al.,
2009). It should however be stressed that MTR is a semi-quantitative
measure that not only depends on the size of the macromolecular
pool but also on the exchange rate between the bound and mobile
proton pools, decreasing its specificity for myelin imaging
(McCreary et al., 2009). Hence, combining measures of MT and
DW-MRI is a means to become more specific to white matter
pathology (Reich et al., 2007). Moreover, the high reproducibility of
MT and DW-MRI in the human cervical cord at 3T suggests that these
measures would provide robust assessment of white matter
pathology (Smith et al., 2009).
The goal of this study was to assess the state of spinal tracts in
patients with chronic SCI by combining HARDI, magnetization
transfer imaging and measures of cord atrophy. We performed
correlations and stepwise regressions between MRI metrics and
clinical parameters. Our hypotheses were:
1. HARDI metrics and MTR (measured in the normal appearing tissue
as assessed using conventional T2 contrast) and cord area differ
between SCI patients and age-matched controls.
2. HARDI metrics and MTR correlate with clinical disability score and
are specific to the tracts involved, i.e., dorsal region for sensory
scores and ventrolateral regions for the motor scores.
Materials and methods
Patients with chronic cervical SCI were recruited (N=14,
age=45±14 years, three women, delay after injury =25±35 years)
(see Table 1). Exclusion criteria were: significant acute and chronic
medical conditions, significant psychiatric or neurological history
(other than SCI for patients), use of psychoactive drugs, osteosynth-
esis material in the spine and standard contraindications to MRI.
Most patients presented spasticity and were treated with baclofen
(Lioresal, 30 mg/day). Neuropathic pains were treated in twelve
patients using pregabalin (Lyrica). All patients were clinically
assessed and scored on the motor and sensory ASIA score (ASIA,
2002) within the week of MRI acquisition. The ASIA motor (ASIAm)
ranges from 0 to 50 for each limb (maximum score of 100). The ASIA
sensory score (ASIAs) only involved the “light touch” test to assess
superficial sensitivity, and ranges from 0 to 56 for each limb
(maximum score of 112). Patients were compared to age-matched
controls (N=14, age=45±17 years, five women). The local ethics
committee of our institution approved all experimental procedures
of the study, and written informed consent was obtained from each
participant. Spinal lesions were identified by an experienced
neuroradiologist (S.L.) using high-resolution T2-weighted images
(see MRI acquisition). The presence of signal hyperintensity and/or
cord compression in the spinal cord region was evaluated (see
Table 1). Regions of normal appearing spinal cord white matter were
identified for further analyses with HARDI and MT data.
Subjects were positioned head-first supine, with a 2 cm thick
pillow to lift the head and no pillow below the neck. This strategy was
used to limit the natural cervical cord lordosis at around C3–C4, i.e.,
excessive cord curvature in the antero-posterior (A-P) direction.
Straightening the spinal cord during positioning ensured limited
partial volume effects when imaging in axial orientation. A typical
example of subject positioning is illustrated in Fig. 1. Following
positioning, pulse oxymeter probe was put on a finger. Before each
scan the subject was asked not to swallow (or minimally).
Acquisitions were conducted on a 3T MRI system (TIM Trio,
Siemens Healthcare, Erlangen, Germany). RF excitation was per-
formed using the body coil and detection was achieved using a
combination of 12-channel head-coil, 4-channel neck coil and 24-
Demography of SCI patients with T2 findings. R: Right, L: Left. In case the injury is
present both dorsally and ventrally (or left and right), the preferential localization of
hypersignal is indicated by the “N” sign.
Gender AgeHypersignal T2ASIAASIAm total/
C3–C4. Bilateral; RNL
C5–C6. Dorsal, L in C6
C3–C4. Lateral bilateral
C3–C4. Dorsolateral, RNL
C5–C6. No signal change
J. Cohen-Adad et al. / NeuroImage 55 (2011) 1024–1033
channel spine matrix (only the 3 most rostral elements of the spine
matrix were used). A rapid localizer image was first acquired in the
three orthogonal plans to ensure proper slice orientation for DW and
MT imaging. Total imaging time was approximately 40 min.
Anatomical scans were conducted to evaluate the anatomical
integrity of the spinal cord. We used a T2-weighted SPACE (Sampling
Perfection with Application optimized Contrasts using different flip
angle Evolution) sequence. This sequence is a 3D turbo spin echo with
slab selective excitation pulses. It provides high SNR due to the 3D
acquisition, high resolution due to the isotropic acquisition, short
acquisition times by combining parallel acquisition with high turbo
factors and low specific absorption rate due to low flip angles.
Parameters were: sagittal orientation, one slab of 52 slices, field of
view (FOV)=280×280 mm2, TR=1500 ms, TE=120 ms, voxel
size=0.9×0.9×0.9 mm3, flip angle=140°, parallel acquisition with
R=3 acceleration factor and generalized autocalibrating partially
parallel acquisitions reconstruction (GRAPPA) (Griswold et al., 2002),
phase encoding direction head-foot, phase oversampling 80%, slice
oversampling 7.7%, bandwidth=744 Hz/Pixel, turbo factor=69 and
acquisition time ~6 min.
HARDI data were acquired using a single shot EPI sequence with
monopolar DW scheme to achieve low TE (Callot et al., 2009; Stejskal
and Tanner, 1965). Eight axial slices were prescribed to cover C2 to T2
vertebral levels (see Fig. 1A). Slices were centred in the middle of each
vertebral body to minimize B0inhomogeneities (Cohen-Adad et al.,
2010a; Cooke et al., 2004). The acquisition was cardiac-gated with
slices acquired during the quiescent phase of cardiac-related motion
of the spinal cord (Summers et al., 2006). Parameters were:
FOV=128 mm, TR=~700 ms (depending on heart beat),
TE=96 ms, voxel size=1×1×5 mm3, parallel acquisition: R=2
with 24 reference lines and GRAPPA reconstruction, phase encoding
direction A-P, number of diffusion-weighting directions=64, b-
value=1000 s/mm2, bandwidth=1086 Hz/Pixel, echo spa-
cing=1.04 ms and number of repetitions=4. Two vertical saturation
bands were used. One was positioned ventrally over the trachea to
limit flow effects and motion due to swallowing. The other saturation
band was set dorsally and aimed at suppressing signal from the non-
spinal cord tissue close to the surface coils and producing high
intensity signal. Manual shimming was estimated within a parallel-
epiped closely fitting the cervical spinal cord (see Fig. 1B).
Magnetization transfer imaging
T1-weighted 3D gradient echo images with slab-selective excita-
tion were acquired with and without magnetization transfer
saturation pulse (Gaussian envelop, duration=9984 μs, frequency
offset=1200 Hz). Parameters were: axial orientation, 52 slices
(spaced with 20% gap), FOV=230×230 mm2, TR=28 ms,
TE=3.2 ms, voxel size=0.9×0.9×2 mm3, flip angle=23°, phase
encoding direction right-left, phase partial Fourier 6/8, band-
width=400 Hz/Pix, acquisition time ~5 min for each volume
(~10 min for both volumes with and without MT pulse).
The cordarea wasmeasuredfromtheT2-SPACE images at thelevel
C1–C2. The plane perpendicular to the spinal cord was resampled to
maximize the accuracy of area measurements, as done in Lundell et al.
(2011). The cord area was then measured using the semi-automatic
method described in Losseff et al. (1996).
Motion correction was doneusingFSL FLIRT(Jenkinson et al.,2002).
HARDI data were first split along the Z direction and motion correction
was applied slice-by-slice to account for the non-rigid motion of
structures across slices, notably induced by the B0fluctuations close to
the lungs (Van de Moortele et al., 2002). The motion correction
algorithm minimized the correlation coefficient ratio between each
image (includingtheb=0 image)and themeanDWimage usingthree
degrees of freedom (Tx, Ty and Rz), as suggested elsewhere to be
optimal for axial spinal cord EPI (Cohen-Adad et al., 2010a). The data
were then averaged across repetitions.
Diffusion tensor and its related metrics were estimated voxel-wise
using FSL (Smith et al., 2004). Of all metrics computed, the fractional
anisotropy (FA), the first eigenvalue (axial diffusivity, λ//) and the
average of the 2nd and 3rd eigenvalues (radial diffusivity, λ⊥) were
further considered for analysis. Q-Ball diffusion orientation distribu-
tion functions (ODF) were estimated using the method described in
Fig. 1. Localizer image demonstrating typical patient positioning with minimum lordosis. A: Slice prescription for diffusion-weighted acquisition, covering C2 to T2 vertebral levels.
Saturation bands were set ventrally and dorsally to the spinal cord to limit aliasing and ghosting artifacts. B: Box prescription for manual shimming, ensuring good B0homogeneity
within the imaged portion of the spinal cord.
J. Cohen-Adad et al. / NeuroImage 55 (2011) 1024–1033
(Descoteaux et al., 2007). The HARDI signal was expressed by means
of the spherical harmonic basis, which allows an analytical solution
for the Funk–Radon transform to obtain the diffusion ODF. We used
spherical harmonic decomposition of order 4 and a regularization
parameter of 0.006 (Descoteaux et al., 2006). From the reconstructed
ODF, we computed the generalized fractional anisotropy (GFA) (Tuch,
2004). As an extension of the FA, the GFA is defined as the standard
deviation divided by the root mean square of the ODF. Hence, it is a
measurement of anisotropy generalized throughout more than three
T1-weighted volumes with and without MT pulse were co-
registered using the non-linear method available in FSL FNIRT
(Smith et al., 2004). Magnetization transfer ratio (MTR) was
computed voxel-wise following the equation [(S0−SMT)/S0]×100,
whereS0andSMTare theT1-weightedimage withoutandwiththeMT
pre-saturation pulse, respectively. Two patients were discarded from
this analysis due to important motion within either the saturated or
the non-saturated T1-weighted acquisition. To compare T1-weighted
signal between the two populations, images without MT pulse were
normalized by the signal in the cerebrospinal fluid to account for B1
Due to the large gap between slices, we did not quantify diffusion
metrics using tractography. Moreover, with coexistent pathology,
tract-specific ROI definition based on tractography (Ciccarelli et al.,
2007; Van Hecke et al., 2008) or fuzzy-logic (Ellingson et al., 2007) is
potentially biased. As suggested in (Xu et al., 2010), we used
geometry-based ROI definition to isolate the dorsal and ventrolateral
quadrants of the cord. Fig. 2 illustrates the definition of ROI for both
DWI and MT analyses. For each modality (HARDI and MT), ROIs were
created by selecting voxels in the dorsal, ventral, left and right aspects
of the spinal cord, as done in (Ciccarelli et al., 2007; Cohen-Adad et al.,
2008a; Onu et al., 2010). ROIs were selected at vertebral levels
presenting no abnormality on the T2-weighted SPACE image. All
vertebral levels were used to define ROIs in healthy controls,
according to a previous study showing that DTI and MT metrics are
similar across vertebral levels in the cervical spinal cord (Smith et al.,
2009). To avoid any bias in the definition of the ROI, i.e., circularity
induced by defining ROIs based on HARDI or MT metrics, ROIs were
defined on the mean diffusion-weighted images (for HARDI analysis)
and on the T1-weighted image (for MT analysis). Although T1-
weighted images offered better white/grey matter contrast with
higher in-plane resolution than diffusion-weighted images, we chose
not to apply T1-based ROIs on diffusion-weighted images, because of
patient motion and susceptibility-related distortions in EPI.
Statistical analysis was performed using Matlab (The Mathworks,
MA, USA) and the Statistical Package for the Social Sciences (SPSS Inc,
Chicago, IL, USA).
To detect differences in HARDI metrics and MTR between patients
and controls in regions of normal appearing white matter, we ran a 2-
tailed Student T-test for several MRI metrics: FA, GFA, axial and radial
diffusivities, MD, MTR, T1- and T2-weighted signals (normalized by
CSF) and cord area. Due to the heterogeneity of lesion location across
patients, metrics were averaged between all sub-ROIs (dorsal and
Correlations between MRI measurements and clinical disability scores
To establish correlations between MRI metrics and clinical
disability scores in patients, we first conducted Pearson's correlation
measures between the total MRI metrics in the normal appearing
white matter (dorsal and ventrolateral regions) and the global clinical
scores (ASIA motor+sensory). Then, tract-specific MRI metrics were
tested with regards to the motor or sensory ASIA score. Namely, we
hypothesized the severity of the sensory or motor disability to be
predicted by more severe white matter pathology in the dorsal and
ventrolateral aspect of the cord, respectively. Regression analyses
were performed for each MRI metric where column-specific infor-
mation was available (FA, axial and radial diffusivity, MD, GFA, MTR).
A forward stepwise model was used, with the test score as the
dependent variable (ASIAm or ASIAs) and MRI parameters and age as
independent variables. The probability for a predictor to enter the
stepwise model was based on a Fisher test, with a p-value set to 0.05.
Relationship between lesion location, MRI metrics and disability
We further tested the relationship between the actual location of
the lesion in each individual (based on the anatomical T2-weighted
image) and the ASIA score and MRI metrics. To test whether patients
presenting a more dorsal lesion have lower sensory score coupled
with low FA or MTR, we categorized the location of the lesion in each
patient as being more dorsal (1) or ventral (0). This binary vector was
then fitted to a logistic regression model using both sensory and
motor ASIA scores, as well as ventrolateral and dorsal MRI metrics as
Significant differences were detected between SCI patients and
controls for metrics measured in the normal-appearing spinal cord
Fig. 2. ROIs were anatomically defined on the mean diffusion-weighted data (for HARDI metrics) and on the T1-weighted image (for MTR). ROIs were selected in the ventro-lateral
(red) and dorsal (green) aspect of the spinal cord, to include most descending and ascending tracts, respectively.
J. Cohen-Adad et al. / NeuroImage 55 (2011) 1024–1033
(see Fig. 3). Cord area was 77.5±3.2 mm2in controls and 68.8±
12.1mm2in patients. Student's T-test showed differences in FA
(pb0.0001), axial diffusivity (p=0.0138), radial diffusivity
(p=0.0135), GFA (pb0.0001), MTR (pb0.0001) and atrophy index
(p=0.0201). No significant difference was detected in MD (p=0.41),
T1-weighted (p=0.76) and T2-weighted (p=0.09) signals.
Correlation with clinical scores
Fig. 4 showsHARDI,MTandatrophymeasuresplottedagainsttotal
ASIA score in SCI patients. Significant correlations were detected for
FA (pb0.01), GFA (pb0.01), radial diffusivity (p=0.01), MTR
(p=0.04) and atrophy (pb0.01). The sign of correlations was
consistent with previous hypotheses, i.e., decrease in FA, GFA, MTR
and cord area and increase in radial diffusivity. Table 2 shows the full
correlation table between motor and sensory scores, atrophy index,
age, disease duration and MRI metrics measured in the ventrolateral
and dorsal aspects (FA, GFA, Radial Diffusivity and MTR).
Stepwise regressions were conducted to evaluate the tract-based
specificity to motor or sensory deficits (see Table 3). MRI metrics were
measured in the dorsal and ventrolateral aspects of the spinal cord.
Results show that dorsal measures of FA (pb0.05), GFA (pb0.01),
radial diffusivity (pb0.05) and MTR (pb0.05) predicted sensory
disability whereas ventrolateral measures of MTR predicted motor
disability (pb0.05). However, ventrolateral measures of FA, GFA and
radial diffusivity explained sensory disability, which suggest a
somewhat lower specificity of diffusion measures.
Relationship between lesion location, MRI metrics and disability
We performed a logistic regression to test the relationship
between the location of the lesion in each individual, the ASIA score
and the MRI metrics. As a result, no predictor was significantly
correlated with the localization of the lesion (using p-threshold of
0.05). We think this reflects the somewhat difficult task in
categorizing patients based on the presentation of their lesions on
the T2 image, given the presence of white matter degeneration and
possibly local cytotoxic events in the normal appearing white
Fig. 3. Boxplots of MRI metrics averaged in the normal appearing spinal cord white matter in the control and the patient groups. Group differences were assessed using Student T-
test. In each box, the central mark is the median, the edges of the box are the 25th and 75th percentiles, the whiskers extend to the most extreme data points not considered outliers,
and outliers are plotted individually.
J. Cohen-Adad et al. / NeuroImage 55 (2011) 1024–1033
This study shows that HARDI and MT measurements at 3T in the
cervical spinal cord of chronic SCI patients detected spinal cord
damage in regions where conventional imaging was negative.
Moreover,some metrics (FA, GFA, radial diffusivity,MTR and atrophy)
were remarkably well correlated with clinical disability. Stepwise
linear regressions showed that measures of MTR in the dorsal spinal
cord explained sensory disability (ASIA score) whereas measures of
MTR in the ventro-lateral spinal cord explained motor disability.
However, diffusion metrics were not specific to the sensorimotor
scores. To our knowledge, this is the first study that combines DTI/QBI
metrics, MTR and atrophy measurements in the injured spinal cord.
DTI and MTR findings in the normal-appearing white matter
for most diffusion metrics (FA, MD, GFA and radial diffusivity), MTR
and atrophy. No significant change was detected for the T1- and
validates the first hypothesis, which suggested that diffusion metrics
and MTR could detect signal changes in the normal appearing tissue. It
also confirms previous studies where DTI (Cohen-Adad et al., 2008a;
DeBoy et al., 2007; Kim et al., 2007; Zhang et al., 2009) and MTR
injury distal to the site of the lesion. These changes may be associated
with global demyelination and degeneration in the spinal cord of SCI
patients, rostral and caudal to the injury, notably due anterograde and
retrograde Wallerian degeneration (Beirowski et al., 2005). Secondary
pathological processes including ischemia, inflammation and
excitotoxic events may also have occurred (Park et al., 2004; Tator
and Fehlings, 1991).
We also investigated the influence of including normal appearing
white matter both rostrally and caudally to the lesion site, when
determining the average MT and diffusion metrics. We did not
observe difference between the three conditions (rostrally, caudally,
or both). Therefore, we decided to include both the rostral and caudal
portionof thecordfor tworeasons:1) toincreasethestatisticalpower
Fig. 4. Pearson's correlations between total clinical score and MRI metrics.
Pearson's correlations table. DD: Disease Duration; λ⊥: Radial diffusivity; vl: ventrolateral, d: dorsal. Levels of significance are indicated as *: Pb0.05; **: Pb0.005.
ASIAm ASIAsAgeDD AtrophyFAvlFAd GFAvlGFAd
J. Cohen-Adad et al. / NeuroImage 55 (2011) 1024–1033
of the analysis by including more voxels and 2) degeneration of tract
were expected both rostrally and caudally to the lesion, as been
observed in animal models (Zhang et al., 2009).
Specificity to white matter demyelination and degeneration
Here we detected significant changes in DTI axial and radial
diffusivity between controls and patients. Previous studies conducted
in animal models of de/dysmyelination and axonal damage showed
that axial and radial diffusivity could respectively predict axonal
degeneration and demyelination (Budde et al., 2007, 2008, 2009;
Hofling et al., 2009; Mac Donald et al., 2007; Song et al., 2002; Xie
et al., 2010; Zhang et al., 2009). Other studies however reported
possible dependences between axial and radial diffusivities, thereby
limiting the specificity of these measures (Herrera et al., 2008; Sun
et al., 2006; Wheeler-Kingshott and Cercignani, 2009). One argument
refers to the pathophysiology of axon degeneration, as this process is
known to be associated with demyelination in several pathologies
such as in MS (Schmierer et al., 2007) or in SCI (Zhang et al., 2009).
Another argument is related to the biophysical properties of DW-MRI:
diffusion metrics depend on several physical parameters including
myelination, axonal density, axonal diameter and orientation of fiber
bundles (Beaulieu, 2002; Sen and Basser, 2005; Wheeler-Kingshott and
Cercignani, 2009). A study combining DTI with multiecho T2 measure-
ments in MS patients did not find significant correlation between
myelin water content and any of the DTI metrics (Kolind et al., 2008).
Here we combined DW and MT imaging to gain confidence in the
identification of pathological processes affecting the white matter.
Significant change in MTR was detected between controls and
patients in the spinal cord white matter. Previous studies showed
that MTR has greater pathological specificity for changes in myelin
content than conventional MRI does. It was notably shown that MTR
decreases with acute demyelination (Chen et al., 2007; Deloire-
Grassin et al., 2000; Inglese et al., 2002; McCreary et al., 2009; Merkler
et al., 2005; Schmierer et al., 2004). However, MTR is a semi-
quantitative measure that not only depends on the size of the
macromolecular pool but also on the exchange rate between the
bound and mobile proton pools (McCreary et al., 2009). To overcome
the multi-parametric dependence of MTR, quantitative MT (qMT)
may provide a more direct surrogate of myelin content (Davies et al.,
2003; Schmierer et al., 2007). However, estimation of qMT requires
assumptions in the number of proton pools in the sample and
necessitates multiple MT measurements as well as an independent
measurement of T1 (Levesque and Pike, 2009). In the present study
we quantified MTR rather than qMT due to constraints in imaging
time for SCI patients.
One benefit of combining DTI and MTR was that it provided two
sets of markers sensitive to white matter pathology based on different
biophysical properties, thereby increasing the reliability of diagnosis.
However, as above-mentioned, these markers may still be influenced
by multiple biophysical parameters. Complementary techniques may
be added to the imaging protocol to further improve the specificity to
various sub-types of white matter pathology. Notably, myelin water
fraction estimated from the short T2 relaxation of water trapped
within myelin sheets is another measure that correlates with myelin
content (Kozlowski et al., 2008; Laule et al., 2006). Despite several
difficulties related to spinal cord motion and low SNR, this technique
has been successfully applied in the human spinal cord at 1.5T (Minty
et al., 2009).
Correlation with clinical disability
For this study we recruited patients with chronic SCI to avoid
additional source of signal variation caused by hemorrhage or edema
(mean time between lesion and imaging was 20±24 months). Also,
clinical parameters are more stable in chronic patients, which is
required for establishing accurate correlations with MRI parameters.
Most MRI parameters measured in the normal appearing spinal cord
correlated with the total clinical disability score (ASIA motor
+sensory), which partly validated the second hypothesis. Highest
correlations were observed for the FA (pb0.01), GFA (pb0.01) and
atrophy (pb0.01). The sign of correlations was consistent with
previous studies, i.e., decrease in FA, GFA, MTR and cord area and
increase in radial diffusivity.
We also tested whether the degeneration of descending pathways
detected in the ventrolateral spinal cord would mostly correlate with
motor disability score (ASIAm) whereas degeneration of ascending
pathways detected in the dorsal spinal cord would mostly correlate
with sensory disability score (ASIAs). Results of stepwise regression
analysis indeed showed that dorsal measures of FA, GFA, radial
diffusivity and MTR explained sensory disability whereas ventrolat-
eral measures of MTR explained motor disability. The hypothesis is
therefore only partially confirmed, due to the somewhat lower
specificity of ventrolateral HARDI measures in regards to the motor
disability, as these measures were rather correlated to sensory
deficits. Several arguments could explain these discrepancies. First,
although relatively high spatial resolution was used for HARDI and
MTR measurements, partial volume effect was still present, adding a
systematic error to the correlations. Second, ASIA motor and sensory
scores, although providing quantitative measures of clinical disability,
only describe global trends in each individual — partly influenced by
the rater and patient condition, and are therefore subject to some
inaccuracies. The use of sensorimotor evoked potential may provide
higher correlations (Lindberg et al., 2007). Moreover, ASIA scores are
not specific to the spinal cord and could also be affected by lesions in
the brain, which sometimes occur in traumatic injuries. Third, the
simplistic sub-division of the spinal cord to distinguish ascending and
descending pathways was somewhat inaccurate. For instance, we
chose to consider lateral voxels for correlation with motor disability.
However, tracts located laterally also transmit ascending information
(e.g., spinocerebellar and spinothalamic tracts). Although we man-
aged to minimize the overlap by excluding most lateral voxels (where
some ascending fibers are located), it is possible that changes in
diffusion and MT metrics observed in patients experiencing pain
might have been partly caused by modifications in the spinothalamic
tracts. Future developments seeking to improve the spatial resolution
will help in that matter.
High angular resolution diffusion imaging
One original aspect of this study was the use of another
reconstruction method than DTI to estimate anisotropy measures in
the spinal cord white matter and to assess its usefulness in
pathological condition. We reconstructed the Q-Ball diffusion ODF
from single-shell measurements using spherical harmonic
Results of stepwise linear regressions that assessed the specificity of MRI metrics with
motor and sensory disability. Dependent variables were clinical scores (ASIAm or
ASIAs). Independent variables were the age and MRI metrics (FA, radial diffusivity, axial
diffusivity, MD, GFA, and MTR) measured in the ventrolateral (VL) or dorsal (D) aspect
of the spinal cord. Significant predictors are reported along with their p-value. “–”
indicates that no variable entered the stepwise regression. Motor disability (ASIAm)
was predicted by the ventrolateral MTR and the dorsal FA, GFA and radial diffusivity,
whereas sensory disability (ASIAs) was predicted by the dorsal FA, GFA, radial
diffusivity and MTR.
J. Cohen-Adad et al. / NeuroImage 55 (2011) 1024–1033
decomposition (Anderson, 2005; Descoteaux, 2007; Hess et al., 2006)
and estimated the generalized fractional anisotropy (GFA), which is a
measure similar to FA (Tuch, 2004). Results showed that like the FA,
the GFA measured in the normal appearing white matter showed
significant differences between controls and patients (pb0.0001).
More interestingly, it showed stronger correlations with clinical score
(total ASIA score). Similar results were observed in animal models of
SCI, where the GFA along with DTI metrics showed significant changes
distal to the site of the lesion (Cohen-Adad et al., 2009a,b,c). However,
drawing definite conclusions about the usefulness of the GFA for
detecting white matter abnormalities is still limited by the interpre-
tation of the biophysical processes underlying the diffusion ODF, since
contrarily to the tensor model there is no direct relationship between
the diffusivity and the value on the diffusion ODF. The use of non-
Gaussian metrics such as the diffusion Kurtosis (Hui et al., 2008;
Jensen et al., 2005) or metrics derived from multi-tensor modeling
(Hosey et al., 2005; Kreher et al., 2005) or higher order tensor
(Barmpoutis et al., 2009; Ghosh et al., 2008) may be useful in the
course of finding biomarkers for characterizing white matter path-
ways in the central nervous system. Notably, metrics based on the
directionality of diffusion may provide new phenotypes of patholog-
ical processes in the whitematterandcould potentiallyhelpfollowing
the course of anatomical reorganization in the central nervoussystem,
as suggested in recent work (Barmpoutis et al., 2009; Cohen-Adad
et al., 2009c).
Limitations and future developments
Although relatively high spatial resolution was employed here
(1×1 mm2in-plane for HARDI and 0.9×0.9 mm2for MT imaging),
partial volume effect was still present between the white and the grey
matter and the cerebrospinal fluid. We do not think this was critical
for addressing the first hypothesis of this study, which compared
HARDI metrics and MTR between two populations having the same
spatialresolution.However,partialvolume effect wasmorecriticalfor
the delineation of white matter sub-quadrants to isolate ascending
and descending tracts, which may partly explain the inconsistencies
observed in the regression analysis between HARDI metrics and the
ASIAm score. The slightly higher spatial resolution of MT measure-
ments and the absence of susceptibility-related distortions may partly
explain that both sensory and motor ASIA scores were respectively
predicted by the dorsal and ventrolateral MTR. The significant atrophy
measured in the spinal cord of SCI patients might have also biased the
measurement of MRI metrics in the spinal cord, due to an increased
partial volume effect. This limitation highlights the need for higher
spatial resolution. The ongoing research that could help increasing the
spatial resolution includes the development of new receive coils for
higher sensitivity (Cohen-Adad et al., 2010b), pulse sequences for
minimizing distortions in DW-MRI (Saritas et al., 2008; Wilm et al.,
2008) and ultra-high field MRI.
For this study we recruited patients with cervical SCI. The goal was
to demonstrate the proof-of-principle for in vivo detection of
demyelination/degeneration in the normal appearing spinal cord
white matter using HARDI and MT measurements. Although theoret-
ically feasible in the thoraco-lumbar region, the proposed method
necessitates further optimization for best results. Challenges in
imaging the thoraco-lumbar spinal cord include: (i) the limited
access for surrounding the imaged region with small coil elements as
could be done in the neck. Typically, an arrangement of posterior
surface coils would be used for the thoraco-lumbar spinal cord,
limiting parallel imaging capabilities when phase encoding is set to
antero-posterior direction (as in most studies); (ii) respiratory-
induced fluctuations create ghosting patterns lowering the quality
of images in this region. Moreover, the movements of the chest induce
local fluctuations in the B0field creating phase errors in EPI-based
diffusion images (Van de Moortele et al., 2002; van Gelderen et al.,
2007); (iii) the natural curvature of the cord at the thoraco-lumbar
level renders more difficult imaging planes perpendicular to the cord,
therefore the use of a sequence allowing to tilt slices within one slab
would be valuable to keep each slice perpendicular to the cord (Xu
et al., 2010).
High angular resolution diffusion-weighted imaging (HARDI),
magnetization transfer ratio and cord atrophy are sensitive markers
of spinal cord pathology and clinical disability in patients with spinal
cord injury. Moreover, tract-specific information could be derived to
predict sensory and motor disability in the normal appearing white
matter. Multi-parametric MRI provides sensitive markers of demye-
lination and degeneration, opening the door to longitudinal studies
for testing therapeutic strategies in spinal cord injury.
We thank Dr. Maxime Descoteaux for providing the code to
compute the Q-Ball ODF and Dr. Henrik Lundell for providing the code
to measure the cord area. We also thank Drs. Stéphane Ouary, Olivier
Freund, Kevin Nigaud, Alexandre Vignaud and Eric Bardinet for helping
with the project. We thank Drs. Thierry Albert, Bertrand Baussart,
Caroline Hugeron, Hugues Pascal Moussellard, Frédéric Petit and Marc-
Antoine Rousseau for helping with patient recruitment and we thank
all subjects. We also thank the reviewers for their helpful comments
that greatly improved the quality of the manuscript. This study was
supported by the Association Française contre les Myopathies (AFM)
and by the Institut pour la Recherche sur la Moelle épinière et
l'Encéphale (IRME). S.R. received a special fellowship from the IRME to
participate in these studies during a sabbatical leave in Paris.
Agosta, F., Absinta, M., Sormani, M.P., Ghezzi, A., Bertolotto, A., Montanari, E., Comi, G.,
Filippi, M., 2007. In vivo assessment of cervical cord damage in MS patients: a
longitudinal diffusion tensor MRI study. Brain 130, 2211–2219.
Anderson, A.W., 2005. Measurement of fiber orientation distributions using high
angular resolution diffusion imaging. Magn. Reson. Med. 54, 1194–1206.
ASIA, 2002. American Spinal Injury Association. International standards for neurolog-
ical classification of spinal cord injury. revised 2000, reprinted 2002 ASIA, Chicago.
Bareyre, F.M., Kerschensteiner, M., Raineteau, O., Mettenleiter, T.C., Weinmann, O.,
Schwab, M.E., 2004. The injured spinal cord spontaneously forms a new intraspinal
circuit in adult rats. Nat. Neurosci. 7, 269–277.
Barmpoutis, A., Hwang, M.S., Howland, D., Forder, J.R., Vemuri, B.C., 2009. Regularized
positive-definite fourth order tensor field estimation from DW-MRI. Neuroimage
Basser, P.J., Mattiello, J., LeBihan, D., 1994. Estimation of the effective self-diffusion
tensor from the NMR spin echo. J. Magn. Reson. 103, 247–254.
Beaulieu, C., 2002. The basis of anisotropic water diffusion in the nervous system — a
technical review. NMR Biomed. 15, 435–455.
Beirowski, B., Adalbert, R., Wagner, D., Grumme, D.S., Addicks, K., Ribchester, R.R.,
Coleman, M.P., 2005. The progressive nature of Wallerian degeneration in wild-
type and slow Wallerian degeneration (WldS) nerves. BMC Neurosci. 6, 6.
Budde, M.D., Kim, J.H., Liang, H.F., Schmidt, R.E., Russell, J.H., Cross, A.H., Song, S.K., 2007.
Toward accurate diagnosis of white matter pathology using diffusion tensor
imaging. Magn. Reson. Med. 57, 688–695.
Budde, M.D., Kim, J.H., Liang, H.-F., Russell, J.H., Cross, A.H., Song, S.-K., 2008. Axonal
injury detected by in vivo diffusion tensor imaging correlates with neurological
disability in a mouse model of multiple sclerosis. NMR Biomed. 21, 589–597.
Budde, M.D., Xie, M., Cross, A.H., Song, S.-K., 2009. Axial diffusivity is the primary
correlate of axonal injury in the experimental autoimmune encephalomyelitis
spinal cord: a quantitative pixelwise analysis. J. Neurosci. 29, 2805–2813.
Callot, V., Duhamel, G., Vignaud, A., Cozzone, P., 2009. Toward a better description of the
gray matter spinal cord by using highly resolved diffusion-weighted and
morphologic T2*-weighted MRI. Proceedings 17th Scientific Meeting, International
Society for Magnetic Resonance in Medicine, Honolulu, p. 1302.
Chen, J.T., Kuhlmann, T., Jansen, G.H., Collins, D.L., Atkins, H.L., Freedman, M.S.,
O'Connor, P.W., Arnold, D.L., 2007. Voxel-based analysis of the evolution of
magnetization transfer ratio to quantify remyelination and demyelination with
J. Cohen-Adad et al. / NeuroImage 55 (2011) 1024–1033
histopathological validation in a multiple sclerosis lesion. Neuroimage 36,
Ciccarelli, O., Wheeler-Kingshott, C.A., McLean, M.A., Cercignani, M., Wimpey, K.,
Miller, D.H., Thompson, A.J., 2007. Spinal cord spectroscopy and diffusion-based
tractography to assess acute disability in multiple sclerosis. Brain 130,
Cohen-Adad, J., Benali, H., Hoge, R.D., Rossignol, S., 2008a. In vivo DTI of the healthy and
injured cat spinal cord at high spatial and angular resolution. Neuroimage 40,
Cohen-Adad, J., Descoteaux, M., Rossignol, S., Hoge, R.D., Deriche, R., Benali, H., 2008b.
Detection of multiple pathways in the spinal cord using q-ball imaging. Neuro-
image 42, 739–749.
Cohen-Adad, J., Ghosh, A., Leblond, H., Descoteaux, M., Deriche, R., Benali, H., Rossignol,
S., 2009a. Comparison of DTI and Q-Ball imaging metrics in a cat model of spinal
cord injury. 14th Annual Meeting of the Organization for Human Brain Mapping
(HBM), San Francisco, USA. p. 51.
Cohen-Adad, J., Leblond, H., Delivet-Mongrain, H.,Ghosh, A., Descoteaux,M., Deriche, R.,
Benali, H., Rossignol, S., 2009b. High angular resolution diffusion MRI of spinal cord
injured cats. 39th Annual meeting of the Society for Neuroscience, Chicago, USA.
Cohen-Adad, J., Leblond, H., Ghosh, A., Descoteaux, M., Deriche, R., Benali, H., Rossignol,
S., 2009c. Evaluation of q-ball metrics for assessing the integrity of the injured
spinal cord. Proceedings of the 17th Annual Meeting of ISMRM, Honolulu, USA, vol.
Cohen-Adad, J., Gauthier, C.J., Brooks, J.C.W., Slessarev, M., Han, J., Fisher, J.A., Rossignol,
S., Hoge, R.D., 2010a. BOLD signal responses to controlled hypercapnia in human
spinal cord. Neuroimage 50, 1074–1084.
Cohen-Adad, J., Mareyam, A., Polimeni, J.R., Wald, L.L., 2010b. Improving diffusion and
functional MRI of the brain and spinal cord using a new 32ch coil. Proceedings of
the 16th Annual Meeting of OHBM, Barcelona, Spain. p. 3694.
Cooke, F.J., Blamire, A.M., Manners, D.N., Styles, P., Rajagopalan, B., 2004. Quantitative
proton magnetic resonance spectroscopy of the cervical spinal cord. Magn. Reson.
Med. 51, 1122–1128.
Davies, G.R., Ramani, A., Dalton, C.M., Tozer, D.J., Wheeler-Kingshott, C.A., Barker, G.J.,
Thompson, A.J., Miller, D.H., Tofts, P.S., 2003. Preliminary magnetic resonance study
of the macromolecular proton fraction in white matter: a potential marker of
myelin? Mult. Scler. 9, 246–249.
DeBoy, C.A., Zhang, J., Dike, S., Shats, I., Jones, M., Reich, D.S., Mori, S., Nguyen, T.,
Rothstein, B., Miller, R.H., Griffin, J.T., Kerr, D.A., Calabresi, P.A., 2007. High
resolution diffusion tensor imaging of axonal damage in focal inflammatory and
demyelinating lesions in rat spinal cord. Brain 130, 2199–2210.
Deloire-Grassin, M.S., Brochet, B., Quesson, B., Delalande, C., Dousset, V., Canioni, P.,
Petry, K.G., 2000. In vivo evaluation of remyelination in rat brain by magnetization
transfer imaging. J. Neurol. Sci. 178, 10–16.
Deo, A.A., Grill, R.J., Hasan, K.M., Narayana, P.A., 2006. In vivo serial diffusion tensor
imaging of experimental spinal cord injury. J. Neurosci. Res. 83, 801–810.
Descoteaux, M., 2007. A Linear and Regularized ODF estimation algorithm to recover
multiple fibers in Q-Ball Imaging. INRIA Research Report, pp. 1–48.
Descoteaux, M., Angelino, E., Fitzgibbons, S., Deriche, R., 2006. Apparent diffusion
coefficients from high angular resolution diffusion imaging: estimation and
applications. Magn. Reson. Med. 56, 395–410.
Descoteaux, M., Angelino, E., Fitzgibbons, S., Deriche, R., 2007. Regularized, fast, and
robust analytical Q-ball imaging. Magn. Reson. Med. 58, 497–510.
Ducreux, D., Fillard, P., Facon, D., Ozanne, A., Lepeintre, J.F., Renoux, J., Tadie, M.,
Lasjaunias, P., 2007. Diffusion tensor magnetic resonance imaging and fiber
tracking in spinal cord lesions: current and future indications. Neuroimaging Clin.
N. Am. 17, 137–147.
Ellingson, B.M., Ulmer, J.L., Schmit, B.D., 2007. Gray and white matter delineation in the
human spinal cord using diffusion tensor imaging and fuzzy logic. Acad. Radiol. 14,
Ellingson, B.M., Ulmer, J.L., Kurpad, S.N., Schmit, B.D., 2008. Diffusion tensor MR
imaging in chronic spinal cord injury. AJNR Am. J. Neuroradiol. 29, 1976–1982.
Fujiyoshi, K., Yamada, M., Nakamura, M., Yamane, J., Katoh, H., Kitamura, K., Kawai, K.,
Okada, S., Momoshima, S., Toyama, Y., Okano, H., 2007. In vivo tracing of neural
tracts in the intact and injured spinal cord of marmosets by diffusion tensor
tractography. J. Neurosci. 27, 11991–11998.
Ghosh, A., Descoteaux, M., Deriche, R., 2008. Riemannian framework for estimating
symmetric positive definite 4th order diffusion tensors. Med. Image Comput.
Comput. Assist. Interv. 11, 858–865.
Griswold, M.A., Jakob, P.M., Heidemann, R.M., Nittka, M., Jellus, V., Wang, J., Kiefer, B.,
Haase, A., 2002. Generalized autocalibrating partially parallel acquisitions (GRAP-
PA). Magn. Reson. Med. 47, 1202–1210.
Herrera, J.J., Chacko, T., Narayana, P.A., 2008. Histological correlation of diffusion tensor
imaging metrics in experimental spinal cord injury. J. Neurosci. Res. 86, 443–447.
Hess, C.P., Mukherjee, P., Han, E.T., Xu, D., Vigneron, D.B., 2006. Q-ball reconstruction of
multimodal fiber orientations using the spherical harmonic basis. Magn. Reson.
Med. 56, 104–117.
Hofling, A.A., Kim, J.H., Fantz, C.R., Sands, M.S., Song, S.-K., 2009. Diffusion tensor
imaging detects axonal injury and demyelination in the spinal cord and cranial
nerves of a murine model of globoid cell leukodystrophy. NMR Biomed. 22,
Hosey, T., Williams, G., Ansorge, R., 2005. Inference of multiple fiber orientations in high
angular resolution diffusion imaging. Magn. Reson. Med. 54, 1480–1489.
Hui, E.S., Cheung, M.M., Qi, L., Wu, E.X., 2008. Towards better MR characterization of
neural tissues using directional diffusion kurtosis analysis. Neuroimage 42,
Inglese, M., Salvi, F., Iannucci, G., Mancardi, G.L., Mascalchi, M., Filippi, M., 2002.
Magnetization transfer and diffusion tensor MR imaging of acute disseminated
encephalomyelitis. AJNR Am. J. Neuroradiol. 23, 267–272.
Jenkinson, M., Bannister, P., Brady, M., Smith, S., 2002. Improved optimization for the
robust and accurate linear registration and motion correction of brain images.
Neuroimage 17, 825–841.
Jensen, J.H., Helpern, J.A., Ramani, A., Lu, H., Kaczynski, K., 2005. Diffusional kurtosis
imaging: the quantification of non-gaussian water diffusion by means of magnetic
resonance imaging. Magn. Reson. Med. 53, 1432–1440.
Kim, J.H., Loy, D.N., Liang, H.F., Trinkaus, K., Schmidt, R.E., Song, S.K., 2007. Noninvasive
diffusion tensor imaging of evolving white matter pathology in a mouse model of
acute spinal cord injury. Magn. Reson. Med. 58, 253–260.
Kolind, S.H., Laule, C., Vavasour, I.M., Li, D.K.B., Traboulsee, A.L., Mädler, B., Moore, G.R.
W., MacKay, A.L., 2008. Complementary information from multi-exponential T2
relaxation and diffusion tensor imaging reveals differences between multiple
sclerosis lesions. Neuroimage 40, 77–85.
Kozlowski, P., Liu, J., Yung, A., Tetzlaff, W., 2008. High-resolution myelin water
measurements in rat spinal cord. Magn. Reson. Med. 59, 796–802.
Kreher, B.W., Schneider, J.F., Mader, I., Martin, E., Hennig, J., Il'yasov, K.A., 2005.
Multitensor approach for analysis and tracking of complex fiber configurations.
Magn. Reson. Med. 54, 1216–1225.
Kucharczyk, W., Macdonald, P.M., Stanisz, G.J., Henkelman, R.M., 1994. Relaxivity and
magnetization transfer of white matter lipids at MR imaging: importance of
cerebrosides and pH. Radiology 192, 521–529.
Lammertse, D., Dungan, D., Dreisbach, J., Falci, S., Flanders, A., Marino, R., Schwartz, E.,
2007. Neuroimaging in traumatic spinal cord injury: an evidence-based review for
clinical practice and research. J. Spinal Cord Med. 30, 205–214.
Laule, C., Leung, E., Lis, D.K., Traboulsee, A.L., Paty, D.W., MacKay, A.L., Moore, G.R., 2006.
Myelin water imaging in multiple sclerosis: quantitative correlations with
histopathology. Mult. Scler. 12, 747–753.
Levesque, I.R., Pike, G.B., 2009. Characterizing healthy and diseased white matter using
quantitative magnetization transfer and multicomponent T(2) relaxometry: A
unified view via a four-pool model. Magn. Reson. Med. 62, 1487–1496.
Lindberg, P.G., Skejo, P.H., Rounis, E., Nagy, Z., Schmitz, C., Wernegren, H., Bring, A.,
Engardt, M., Forssberg, H., Borg, J., 2007. Wallerian degeneration of the
corticofugal tracts in chronic stroke: a pilot study relating diffusion tensor
imaging, transcranial magnetic stimulation, and hand function. Neurorehabil.
Neural Repair 21, 551–560.
Losseff, N.A., Wang, L., Lai, H.M., Yoo, D.S., Gawne-Cain, M.L., McDonald, W.I., Miller,
D.H., Thompson, A.J., 1996. Progressive cerebral atrophy in multiple sclerosis. A
serial MRI study. Brain 119 (Pt 6), 2009–2019.
Lundell, H., Dyrby, T., Ptito, M., Nielsen, J.B., 2009. Crossing fibers in lateral white matter
of the cervical spinal cord detected with diffusion MRI in monkey postmortem.
Proceedings of the 17th Annual Meeting of ISMRM, Honolulu, USA, vol. 1497.
Lundell, H., Barthelemy, D., Skimminge, A., Dyrby, T.B., Biering-Sørensen, F., Nielsen, J.B.,
2011. Independent spinal cord atrophy measures correlate to motor and sensory
deficits in individuals with spinal cord injury. Spinal Cord 49, 70–75.
Mac Donald, C.L., Dikranian, K., Bayly, P., Holtzman, D., Brody, D., 2007. Diffusion tensor
imaging reliably detects experimental traumatic axonal injury and indicates
approximate time of injury. J. Neurosci. 27, 11869–11876.
Maier, I.C., Schwab, M.E., 2006. Sprouting, regeneration and circuit formation in the
injured spinal cord: factors and activity. Philos. Trans. R. Soc. Lond. B Biol. Sci. 361,
McCreary, C.R., Bjarnason, T.A., Skihar, V., Mitchell, J.R., Yong, V.W., Dunn, J.F., 2009.
Multiexponential T2 and magnetization transfer MRI of demyelination and
remyelination in murine spinal cord. Neuroimage 45, 1173–1182.
Merkler, D., Boretius, S., Stadelmann, C., Ernsting, T., Michaelis, T., Frahm, J., Brück, W.,
2005. Multicontrast MRI of remyelination in the central nervous system. NMR
Biomed. 18, 395–403.
Minty, E.P., Bjarnason, T.A., Laule, C., Mackay, A.L., 2009. Myelin water measurement in
the spinal cord. Magn. Reson. Med. 61, 883–892.
Mottershead, J.P., Schmierer, K., Clemence, M., Thornton, J.S., Scaravilli, F., Barker, G.J.,
Tofts, P.S., Newcombe, J., Cuzner, M.L., Ordidge, R.J., McDonald, W.I., Miller, D.H.,
2003. High field MRI correlates of myelin content and axonal density in multiple
sclerosis — a post-mortem study of the spinal cord. J. Neurol. 250, 1293–1301.
Nevo, U., Hauben, E., Yoles, E., Agranov, E., Akselrod, S., Schwartz, M., Neeman, M., 2001.
Diffusion anisotropy MRI for quantitative assessment of recovery in injured rat
spinal cord. Magn. Reson. Med. 45, 1–9.
Ohgiya, Y., Oka, M., Hiwatashi, A., Liu, X., Kakimoto, N., Westesson, P.L., Ekholm, S.E.,
2007. Diffusion tensor MR imaging of the cervical spinal cord in patients with
multiple sclerosis. Eur. Radiol. 17, 2499–2504.
Onu, M., Gervai, P., Cohen-Adad, J., Lawrence, J., Kornelsen, J., Tomanek, B., Sboto-
Frankenstein, U.N., 2010. Human cervical spinal cord funiculi: investigation with
magnetic resonance diffusion tensor imaging. J. Magn. Reson. Imaging 31, 829–837.
Park, E., Velumian, A.A., Fehlings, M.G., 2004. The role of excitotoxicity in secondary
mechanisms of spinal cord injury: a review with an emphasis on the implications
for white matter degeneration. J. Neurotrauma 21, 754–774.
Pike, G.B., De Stefano, N., Narayanan, S., Worsley, K.J., Pelletier, D., Francis, G.S., Antel,
J.P., Arnold, D.L., 2000. Multiple sclerosis: magnetization transfer MR imaging of
white matter before lesion appearance on T2-weighted images. Radiology 215,
Plank, C., Koller, A., Mueller-Mang, C., Bammer, R., Thurnher, M.M., 2007. Diffusion-
weighted MR imaging (DWI) in the evaluation of epidural spinal lesions.
Neuroradiology 49, 977–985.
Reich, D.S., Smith, S.A., Zackowski, K.M., Gordon-Lipkin, E.M., Jones, C.K., Farrell, J.A.,
Mori, S., van Zijl, P.C., Calabresi, P.A., 2007. Multiparametric magnetic resonance
J. Cohen-Adad et al. / NeuroImage 55 (2011) 1024–1033
imaging analysis of the corticospinal tract in multiple sclerosis. Neuroimage 38, Download full-text
Ries, M., Jones, R.A., Dousset, V., Moonen, C.T., 2000. Diffusion tensor MRI of the spinal
cord. Magn. Reson. Med. 44, 884–892.
Rossignol, S., Schwab, M., Schwartz, M., Fehlings, M.G., 2007. Spinal cord injury: time to
move? J. Neurosci. 27, 11782–11792.
Saritas, E.U., Cunningham, C.H., Lee, J.H., Han, E.T., Nishimura, D.G., 2008. DWI of the
spinal cord with reduced FOV single-shot EPI. Magn. Reson. Med. 60, 468–473.
Schmierer, K., Scaravilli, F., Altmann, D.R., Barker, G.J., Miller, D.H., 2004. Magnetization
transfer ratio and myelin in postmortem multiple sclerosis brain. Ann. Neurol. 56,
Schmierer, K., Tozer, D.J., Scaravilli, F., Altmann, D.R., Barker, G.J., Tofts, P.S., Miller, D.H.,
2007. Quantitative magnetization transfer imaging in postmortem multiple
sclerosis brain. J. Magn. Reson. Imaging 26, 41–51.
Schwartz, E.D., Duda, J., Shumsky, J.S., Cooper, E.T., Gee, J., 2005. Spinal cord diffusion
tensor imaging and fiber tracking can identify white matter tract disruption and glial
scar orientation following lateral funiculotomy. J. Neurotrauma 22, 1388–1398.
Sen, P.N., Basser, P.J., 2005. A model for diffusion in white matter in the brain. Biophys. J.
Shen, H., Tang, Y., Huang, L., Yang, R., Wu, Y., Wang, P., Shi, Y., He, X., Liu, H., Ye, J., 2007.
Applications of diffusion-weighted MRI in thoracic spinal cord injury without
radiographic abnormality. Int. Orthop. 31, 375–383.
Smith, S.M., Jenkinson, M., Woolrich, M.W., Beckmann, C.F., Behrens, T.E., Johansen-
Berg, H., Bannister, P.R., De Luca, M., Drobnjak, I., Flitney, D.E., Niazy, R.K., Saunders,
J., Vickers, J., Zhang, Y., De Stefano, N., Brady, J.M., Matthews, P.M., 2004. Advances
in functional and structural MR image analysis and implementation as FSL.
Neuroimage 23 (Suppl 1), S208–S219.
Smith, S.A., Jones, C.K., Gifford, A., Belegu, V., Chodkowski, B., Farrell, J.A.D., Landman, B.A.,
Reich, D.S., Calabresi, P.A., McDonald, J.W., Van Zijl, P.C.M., 2010. Reproducibility of
tract-specific magnetization transfer and diffusion tensor imaging in the cervical
spinal cord at 3 tesla. NMR Biomed. 23, 207–217.
Song, S.K., Sun, S.W., Ramsbottom, M.J., Chang, C., Russell, J., Cross, A.H., 2002.
Dysmyelination revealed through MRI as increased radial (but unchanged axial)
diffusion of water. Neuroimage 17, 1429–1436.
Stejskal, E.O., Tanner, J.E., 1965. Spin diffusion measurements: spin echoes in the
presence of a time-dependent field gradient. J. Chem. Phys. 42, 288–292.
Summers, P., Staempfli, P., Jaermann, T., Kwiecinski, S., Kollias, S., 2006. A preliminary
study of the effects of trigger timing on diffusion tensor imaging of the human
spinal cord. AJNR Am. J. Neuroradiol. 27, 1952–1961.
Sun, S.-W., Liang, H.-F., Trinkaus, K., Cross, A.H., Armstrong, R.C., Song, S.-K., 2006.
Noninvasive detection of cuprizone induced axonal damage and demyelination in
the mouse corpus callosum. Magn. Reson. Med. 55, 302–308.
Tator, C.H., Fehlings, M.G., 1991. Review of the secondary injury theory of acute spinal
cord trauma with emphasis on vascular mechanisms. J. Neurosurg. 75, 15–26.
Thurnher, M.M., Bammer, R., 2006. Diffusion-weighted magnetic resonance imaging of
the spine and spinal cord. Semin. Roentgenol. 41, 294–311.
Tuch, D.S., 2004. Q-ball imaging. Magn. Reson. Med. 52, 1358–1372.
Valsasina, P., Rocca, M.A., Agosta, F., Benedetti, B., Horsfield, M.A., Gallo, A., Rovaris, M.,
Comi, G., Filippi, M., 2005. Mean diffusivity and fractional anisotropy histogram
analysis of the cervical cord in MS patients. Neuroimage 26, 822–828.
Van de Moortele, P.F., Pfeuffer, J., Glover, G.H., Ugurbil, K., Hu, X., 2002. Respiration-
induced B0fluctuations and their spatial distribution in the human brain at 7 Tesla.
Magn. Reson. Med. 47, 888–895.
van Gelderen, P., de Zwart, J.A., Starewicz, P., Hinks, R.S., Duyn, J.H., 2007. Real-time
shimming to compensate for respiration-induced B0fluctuations. Magn. Reson.
Med. 57, 362–368.
Van Hecke, W., Leemans, A., Sijbers, J., Vandervliet, E., Van Goethem, J., Parizel, P.M.,
2008. A tracking-based diffusion tensor imaging segmentation method for the
detection of diffusion-related changes of the cervical spinal cord with aging. J.
Magn. Reson. Imaging 27, 978–991.
Vargas, M.I., Delavelle, J., Jlassi, H., Rilliet, B., Viallon, M., Becker, C.D., Lövblad, K.-O.,
2008. Clinical applications of diffusion tensor tractography of the spinal cord.
Neuroradiology 50, 25–29.
Wheeler-Kingshott, C., Cercignani, M., 2009. About “axial” and “radial” diffusivities.
Magn. Reson. Med. 61, 1255–1260.
Wilm, B.J., Gamper, U., Henning, A., Pruessmann, K.P., Kollias, S.S., Boesiger, P., 2009.
Diffusion-weighted imaging of the entire spinal cord. NMR Biomed. 22, 174–181.
Xie, M., Tobin, J.E., Budde, M.D., Chen, C.-I., Trinkaus, K., Cross, A.H., Mcdaniel, D.P.,
Song, S.-K., Armstrong, R.C., 2010. Rostrocaudal analysis of corpus callosum
demyelination and axon damage across disease stages refines diffusion tensor
imaging correlations with pathological features. J. Neuropathol. Exp. Neurol. 69,
Xu, J., Klawiter, E., Shimony, J.S., Snyder, A.Z., Naismith, R., Priatna, A., Benzinger, T.,
Cross, A., Song, S.-K., 2010. Toward reproducible tract-specific in vivo diffusion
quantification in human cervical spinal cord. Proceedings of the 18th Annual
Meeting of ISMRM, Stockholm, Sweden, vol. 2458.
Zackowski, K.M., Smith, S.A., Reich, D.S., Gordon-Lipkin, E., Chodkowski, B.A.,
Sambandan, D.R., Shteyman, M., Bastian, A.J., van Zijl, P.C., Calabresi, P.A., 2009.
Sensorimotor dysfunction in multiple sclerosis and column-specific magnetization
transfer-imaging abnormalities in the spinal cord. Brain 132, 1200–1209.
Zhang, J., Jones, M., DeBoy, C.A., Reich, D.S., Farrell, J.A., Hoffman, P.N., Griffin, J.W.,
Sheikh, K.A., Miller, M.I., Mori, S., Calabresi, P.A., 2009. Diffusion tensor magnetic
resonance imaging of Wallerian degeneration in rat spinal cord after dorsal root
axotomy. J. Neurosci. 29, 3160–3171.
J. Cohen-Adad et al. / NeuroImage 55 (2011) 1024–1033