- Email: [email protected]
Tract-Specific Diffusion Tensor Imaging Reveals Laterality of Neurological Symptoms in Patients with Cervical Compression Myelopathy
Accepted Manuscript Tract-specific Diffusion Tensor Imaging Reveals Laterality of Neurological Symptoms in Patients with Cervical Compression Myelopathy Satoshi Maki, Masao Koda, Junya Saito, Sho Takahashi, Taigo Inada, Koshiro Kamiya, Mitsutoshi Ota, Yasushi Iijima, Yoshitada Masuda, Koji Matsumoto, Masatoshi Kojima, Kazuhisa Takahashi, Takayuki Obata, Masashi Yamazaki, Takeo Furuya PII:
S1878-8750(16)30813-0
DOI:
10.1016/j.wneu.2016.08.129
Reference:
WNEU 4540
To appear in:
World Neurosurgery
Received Date: 22 July 2016 Revised Date:
29 August 2016
Accepted Date: 30 August 2016
Please cite this article as: Maki S, Koda M, Saito J, Takahashi S, Inada T, Kamiya K, Ota M, Iijima Y, Masuda Y, Matsumoto K, Kojima M, Takahashi K, Obata T, Yamazaki M, Furuya T, Tract-specific Diffusion Tensor Imaging Reveals Laterality of Neurological Symptoms in Patients with Cervical Compression Myelopathy, World Neurosurgery (2016), doi: 10.1016/j.wneu.2016.08.129. 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.
Maki
ACCEPTED MANUSCRIPT
Tract-specific Diffusion Tensor Imaging Reveals Laterality of Neurological Symptoms in
2
Patients with Cervical Compression Myelopathy
3
Satoshi Maki1,*, Masao Koda1, Junya Saito1 , Sho Takahashi2, Taigo Inada1, Koshiro
4
Kamiya1, Mitsutoshi Ota1, Yasushi Iijima1, Yoshitada Masuda3, Koji Matsumoto3, Masatoshi
5
Kojima3, Kazuhisa Takahashi1, Takayuki Obata4, Masashi Yamazaki5, Takeo Furuya1
6
1
7
1-8-1 Inohana Chuou-ku Chiba City, Chiba 260-8670, Japan
8
2
9
1-8-1 Inohana Chuou-ku Chiba City, Chiba 260-8677, Japan
RI PT
1
Department of Orthopaedic Surgery, Chiba University Graduate School of Medicine
SC
Clinical Research Center, Chiba University Hospital
10
3
11
1-8-1 Inohana Chuou-ku Chiba City, Chiba 260-8677, Japan
12
4
13
4-9-1 Anagawa, Inage-ku, Chiba City, Chiba 263-8555, Japan
14
5
15
1-1-1 Tennodai, Tsukuba City, Ibaragi, 305-8575, Japan
M AN U
Department of Radiology, Chiba University Hospital
Center for Charged Particle Therapy, National Institute of Radiological Sciences
TE D
Department of Orthopaedic Surgery, Faculty of Medicine, University of Tsukuba
16
*Address correspondence and reprint requests to:
18
Satoshi Maki, MD, PhD
19
Department of Orthopaedic Surgery, Chiba University Graduate School of Medicine
20
1-8-1 Inohana, Chuou-ku, Chiba 260-8670, Japan
21
Tel: 043-226-2117, Fax: 043-226-2116
22
E-mail: [email protected]
AC C
EP
17
23 24
Keywords Diffusion tensor imaging, Spinal cord, Cervical compression myelopathy,
25
Laterality, Asymmetry index 1
Maki
ACCEPTED MANUSCRIPT Abstract
2
Backbround: Patients with cervical compression myelopathy (CCM) generally present
3
bilateral neurological symptoms in their extremities. However, a substantial portion of
4
patients with CCM exhibit laterality of neurological symptoms. The aim of this study was to
5
assess the correlation between intrinsic structural damage and laterality of symptoms using
6
spinal cord diffusion tensor imaging (DTI) of the corticospinal tract.
7
Methods: We enrolled 10 healthy volunteers and 40 patients with CCM in this study. We
8
evaluated motor function using the American Spinal Injury Association (ASIA) motor score
9
for left and right extremities. For DTI acquisitions, a 3.0 T MR system with
SC
RI PT
1
diffusion-weighted spin-echo sequence was used. Regions-of-interest (ROIs) in the lateral
11
column (LC) tracts were determined. We determined the correlations between fractional
12
anisotropy (FA) and ASIA motor scores. An FA asymmetry index was calculated using left
13
and right ROIs.
14
Results: Four patients exhibited laterality of symptoms in their extremities, for which left and
15
right ASIA scores correlated moderately with FA in the left and right LCs respectively (left: ρ
16
= 0.64, P < 0.001; right: ρ = 0.67, P < 0.001). The area under the receiver-operator
17
characteristic curve showed the FA asymmetry index indicated laterality of symptoms.
18
Conclusions: Using tract-specific DTI, we demonstrated that microstructural damages in the
19
left and right corticospinal tracts correlated with corresponding neurological symptoms in the
20
ipsilateral side and the FA asymmetry index could indicate laterality in neurological
21
symptoms of patients with CCM.
AC C
EP
TE D
M AN U
10
22 23
Keywords Diffusion tensor imaging; Spinal cord; Cervical compression myelopathy;
24
Laterality; Asymmetry index
25 2
Maki
ACCEPTED MANUSCRIPT Introduction
2
Patients with cervical compression myelopathy (CCM), including cervical spondylotic
3
myelopathy, ossification of the longitudinal ligament, and cervical disk herniation, generally
4
present bilateral extremity symptoms. Five types of spinal cord lesions in patients with CCM
5
were proposed by Crandall et al.1 The classification is widely accepted and used because it
6
effectively describes neurological symptoms based on neuroanatomy. The most frequently
7
encountered Crandall classification is transverse syndrome, which generally presents bilateral
8
symptoms. Brown-Sequard syndrome is a unilateral cord lesion with ipsilateral motor deficit
9
and contralateral analgesia below the lesion.2 However, a substantial portion of patients with
SC
RI PT
1
CCM exhibit laterality of neurological symptoms without showing contralateral analgesia
11
below the lesion. Few studies regarding laterality of neurological symptoms in patients with
12
CCM have been reported.3
13
It is difficult to quantify spinal cord dysfunction by conventional magnetic resonance imaging
14
(MRI). However, diffusion tensor imaging (DTI) can detect microstructural changes in the
15
white matter of the spinal cord and provide quantitative diffusion parameters. Fractional
16
anisotropy (FA) and mean diffusivity (MD) are derived from the diffusion matrix and
17
represent the degree of anisotropy and magnitude of the diffusion process. FA and MD reflect
18
tissue integrity in the spinal cord of patients with CCM. This hypothesis is supported by
19
several studies.4, 5 Maki et al. have reported that FA in lateral and posterior columns in
20
tract-specific DTI correlated with gait disturbance in patients with CCM.6
21
The aim of this study was to assess the correlation between intrinsic structural damage and
22
laterality of symptoms using spinal cord diffusion tensor imaging (DTI) of the corticospinal
23
tract.
AC C
EP
TE D
M AN U
10
24 25
Materials and Methods 3
Maki
ACCEPTED MANUSCRIPT Participants
2
All participants provided written, informed consent before participating in the research,
3
which was previously approved by our institutional review board. We enrolled 10 healthy
4
volunteers and 40 patients with CCM in this study. We excluded potential participants with a
5
history of neurological injury, spinal surgery, or neurodegenerative disease causing any spinal
6
cord disorder. Healthy volunteers included 6 men and 4 women with a mean age of 43.7 ± 8.6
7
years. In the group of patients with CCM, there were 26 men and 14 women with a mean age
8
of 65.3 ± 11.0 years. The clinical diagnoses included 18 patients with cervical spondylotic
9
myelopathy, 19 with ossification of the posterior longitudinal ligament, 2 with cervical disk
SC
herniation, and 1 with atlantoaxial dislocation because of rheumatoid arthritis.
M AN U
10
RI PT
1
11 Clinical Evaluation
13
We evaluated motor function using the American Spinal Injury Association (ASIA) motor
14
score for left and right upper and lower extremities. Left and right motor scores ranged from
15
0 to 50.7
16
TE D
12
Conventional MRI
18
The patients were studied using a 3.0 T MRI system (Discovery MR 750, GE Healthcare,
19
Waukesha, WI). An 8-ch head, neck, and spine receiver coil was used. Conventional MRI
20
consisted of T1-weighted (T1W) and T2-weighted (T2W) sagittal images, and T2W axial
21
images were also obtained. For sagittal imaging, the acquisition settings were as follows:
22
field-of-view (FOV) = 24 × 24 cm2, slice thickness = 3 mm echo time (TE)/time of repetition
23
(TR) = 8.4/500 ms (T1W) and TE/TR = 105/2710 ms (T2W). For axial imaging, the
24
acquisition settings were as follows: FOV = 16 × 16 cm2, slice thickness = 4 mm, TE/TR =
25
105/4000 ms.
AC C
EP
17
4
Maki
ACCEPTED MANUSCRIPT 1 Diffusion Tensor Imaging
3
For DTI acquisitions, diffusion-weighted spin-echo single-shot echo-planar imaging was
4
used, with diffusion gradients applied in 6 noncollinear spatial directions and 1 baseline
5
image with b = 700 s/mm2. Reduced FOV technique was applied to reduce susceptibility
6
artifacts and achieve higher resolution.8, 9 The thickness of each slice was 5 mm and the
7
sequence parameters for DTI were: TE = 74.9 ms, TR = 3000 ms, FOV = 14 × 4.2 cm2,
8
number of excitations = 16, and the acquisition data matrix = 176 × 54. Transverse images
9
covering the cervical spinal cord from C1 to T1 were acquired, each of which was placed at
SC
the center of either a vertebra or intervertebral disk.
M AN U
10
RI PT
2
11 Image Analysis
13
We identified the level of maximum compression on T2W axial images, and divided the
14
cross-sectional area (CSA) of the spinal cord into left and right halves using the line bisecting
15
the center of the spinous process and vertebra. We measured left and right CSAs using Image
16
J software, version 1.45 (NIH, Bethesda, MD) (Fig. 1).
17
We quantified diffusion tensor parameters using DTIStudio software (Johns Hopkins Medical
18
Institute, Johns Hopkins University). Five levels from C2/3 to C6/7 were measured in control
19
group. The level of maximum compression was measured in patients with CCM.
20
Regions-of-interest (ROI) for the lateral column (LC) tracts were determined based on the
21
geometry of the cord on the axial image of the FA map and FA and MD were measured (Fig.
22
2). The size of ROIs was standardized at 6-8 pixels. We analyzed the correlation between FA
23
and MD and left and right ASIA motor scores.
AC C
EP
TE D
12
24 25
Reliability of ROI Measurement for DTI Analysis 5
Maki
ACCEPTED MANUSCRIPT 1
ROI measurements were made twice by 2 experienced spinal surgeons at 2-week intervals.
2
Intraclass correlation coefficients (ICC) were used to determine the interobserver and
3
intraobserver reliabilities. We provide 95% prediction limits for the errors in measurements.
4 Statistical Analyses
6
A Mann–Whitney U test was used to compare means of FA and MD values and asymmetry
7
indices between patients with CCM and healthy controls. Correlations between DTI
8
parameters and left and right ASIA motor scores were determined using Spearman’s
9
correlation coefficients. The FA symmetry index for healthy volunteers and patients with
SC
RI PT
5
CCM was calculated. The FA asymmetry index is the percentage of the absolute difference
11
between the FA in the left and right ROIs relative to the sum of the respective FA (asymmetry
12
index = |FAright – FAleft|/(FAright + FAleft) × 100). The asymmetry indices for MD and CSA
13
values were also calculated. The prediction of asymmetrical symptoms using the FA, MD and
14
CSA asymmetry index was determined using the area under a receiver–operator characteristic
15
(ROC) curve. P < 0.05 was considered significant. All the analyses were conducted using
16
JMP software, version 10.0.2 (SAS Institute, Cary, NC).
17
EP
TE D
M AN U
10
Results
19
Clinical Symptoms
20
The ASIA motor score for the left side was 47.6 (range 36–50) and for the right side was 47.7
21
(range 38–50). The difference between the left and right ASIA scores was 0.1 ± 2.3. Because
22
the asymmetry of symptoms between left and right was more than 2 standard deviations
23
(SDs) (i.e. more than 4 points difference in left and right ASIA motor score), 4 of 40 (10%)
24
patients had laterality of neurological symptoms. None of these patients presented
25
Brown-Sequard syndrome, and were considered to have transverse syndrome.
AC C
18
6
Maki
ACCEPTED MANUSCRIPT 1 FA, MD and asymmetry indices
3
DTI analyses revealed that the FA in patients with CCM was significantly lower than that in
4
healthy volunteers in both left and right LC (Table 1). The MD in patients with CCM was
5
significantly higher than that in healthy volunteers in both left and right LC (Table 1). The FA
6
and MD asymmetry indices in patients with CCM were significantly higher than those in
7
healthy volunteers (Table 2).
8 9
Correlation between ASIA Score and DTI Indices
SC
RI PT
2
In patients with CCM, left and right ASIA scores correlated moderately with FA in the left
11
and right LCs (left: ρ = 0.64, P < 0.001; right: ρ = 0.67, P < 0.001; Fig. 3). However, left and
12
right ASIA scores were not correlated with MD in the right or left LC respectively (left: ρ =
13
-0.13, P = 0.43; right: ρ = -0.26, P = 0.10; Fig. 3).
M AN U
10
TE D
14
Reliability of ROI Measurement for DTI Analysis
16
The ICC for interobserver reliability of ROI measurements was 0.87 for the FA value and
17
0.76 for the MD value. The ICC for intraobserver reliability of ROI measurements was 0.90
18
for the FA value and 0.93 for the MD value. The inter- and intraobserver reliability of tract
19
specific ROI measurements was considered reliable for analyzing DTI of the patients with
20
CCM.
AC C
21
EP
15
22
Cross-sectional Area (CSA) of the Spinal Cord
23
The mean left and right CSAs of the spinal cord in patients with CCM were 25.4 ± 8.2 mm2
24
and 26.6 ± 7.9 mm2 respectively at the most apparent level of cord compression. The left
25
CSAs showed only a weak correlation with left ASIA motor scores and right CSAs showed 7
Maki
ACCEPTED MANUSCRIPT 1
no correlation with right ASIA motor scores (left: ρ = 0.39, P = 0.01; right: ρ = 0.28, P =
2
0.08; Fig.4). The asymmetry index for CSA was 11.6 ± 9.1%.
3 Detection of the Laterality of Neurological Symptoms
5
The area under the receiver–operator characteristic (ROC) curve for the FA asymmetry index
6
was 0.86, which indicated that the FA asymmetry index could moderately indicate the
7
laterality of neurological symptoms in the left and right extremities. On the other hand, the
8
area under the ROC curve for the MD and CSA asymmetry index were 0.57 and 0.54,
9
respectively, which indicated that both MD and CSA asymmetry index had low accuracy for
SC
detecting the laterality of neurological symptoms of left and right extremities.
M AN U
10
RI PT
4
11 DISCUSSION
13
The present study shows that it is feasible to demonstrate correlation between left and right
14
microstructural damages in spinal cord and corresponding spinal cord long tract symptoms
15
independently and to reveal asymmetry of neurological symptoms in patients with CCM
16
using tract-specific DTI. A certain portion of patients with CCM exhibited asymmetrical
17
symptoms in their upper and lower extremities.
18
Tract-specific analysis of the spinal cord is gaining popularity because high resolution images
19
can be acquired by reduced FOV techniques.8, 9 However, most studies of DTI for evaluating
20
the spinal cord at the tract level target demyelinating disease in patients whose spinal cord
21
morphology is preserved. There are few DTI studies evaluating patients with CCM at the
22
tract level.6,10–13 Even fewer DTI studies focus on laterality of the neurological symptoms in
23
patients with demyelinating disease and CCM. Meyenburg et al. reported that DTI revealed
24
asymmetrical changes in the spinal cord of patients with multiple sclerosis, which
25
corresponded to asymmetrical electrophysiological deficits for both arms and legs, and
AC C
EP
TE D
12
8
Maki
ACCEPTED MANUSCRIPT
reflected a specific structure–function relationship in the spinal cord.14 Yoo et al. analyzed 9
2
patients with CCM using DTI and sorted DTI parameters into affected and unaffected sides
3
according to the presence or the absence of symptoms. They found a significant decrease in
4
FA of anterior and lateral ROIs in the affected side compared with a control group. However,
5
their patient population was relatively small and the definition of affected and unaffected side
6
was not well described.15 Cui et al. conducted column specific analysis of the spinal cord in
7
patients with cervical spondylotic myelopathy. In their report, they presented a patient with
8
left hand clumsiness and spasticity of the left lower limb showing low FA in DTI of the
9
lateral and posterior column on the left side of the spinal cord as a representative case.10 The
10
present study focused on laterality of symptoms in patients with CCM and revealed left and
11
right decreases in FA in the LCs corresponding to left and right limb motor dysfunction (Fig.
12
5). By contrast, spinal cord CSA was not correlated with ASIA motor score. Correlation
13
between CSA and clinical symptoms of CCM remains uncertain and it seems difficult to
14
predict the severity of the symptoms using CSA.16 Kadanka et al. reported that there was a
15
significant difference in the modified JOA score of patients with cervical spondylotic
16
myelopathy between those with CSA <50 mm2 and those with CSA >60 mm2.17 Golash et al.
17
demonstrated that narrowness of the spinal cord CSA was well correlated with severity of
18
established myelopathy, but was an unsatisfactory prognosticator for mild degrees of
19
myelopathy.18
20
In the present study, 10% of the patients with CCM exhibited asymmetrical symptoms in
21
their extremities. There are few studies referring to laterality of neurological symptoms in
22
patients with CCM. Mihara et al. found that 12% of patients with CCM had laterality of
23
symptoms in their extremities, which was defined as 4 times or more difference in a left and
24
right limb performance test (a finger grip and release test and a triangle step test).19 They
25
compared surgical outcomes between patients with cervical myelopathy who had either
AC C
EP
TE D
M AN U
SC
RI PT
1
9
Maki
ACCEPTED MANUSCRIPT symmetrical or asymmetrical symptoms.3 JOA score recovery rates in patients with
2
asymmetrical symptoms were less than those of patients with mild symmetrical symptoms,
3
but were greater than those of patients with severe symmetrical symptoms. The investigators
4
hypothesized that cervical myelopathy might progress through a period of asymmetrical
5
symptoms before producing severe symmetrical symptoms. However, in the present study, we
6
identified patients with relatively severe impairment as having asymmetrical symptoms. The
7
difference can be attributed to the definition of asymmetrical symptoms, which were defined
8
by more than two SDs difference in ASIA motor score in the current study, and not by a
9
performance test. Because 4 points loss in ASIA motor score represents substantial motor
10
weakness, patients with asymmetrical symptoms present poor functional status. Previous
11
neurohistopathological studies of patients with cervical spondylotic myelopathy showed that
12
the distribution of the lesion in the spinal cord was not homogeneous, which supports the
13
hypothesis that degeneration of the LC can occur asymmetrically.20 A proportion of our
14
patients presented asymmetrical symptoms in their extremities even though they were
15
classified as having a transverse syndrome, but not Brown-Sequard syndrome.
16
Decrease in FA and increase in MD at the compression site were seen in patients with CCM
17
compared with healthy volunteers, which was consistent with previous studies.21–24 Decrease
18
in FA is considered to reflect demyelination and axon damage in the spinal cord. 23-26 The
19
current study showed a moderate correlation between decrease in FA and left and right ASIA
20
motor score. The LC contains the lateral corticospinal tract, which is the principal descending
21
motor fiber tract conducting voluntary movement in the human spinal cord. Therefore,
22
damage in the LC, which is reflected by a decrease in FA, is considered to result in decreased
23
ASIA motor score. By contrast, the present study showed no correlation between MD and
24
ASIA motor score, which is consistent with previous findings.11, 21 Increase in MD is
25
considered to occur at the initial stage of CCM, and is possibly related to edema.15 MD may
AC C
EP
TE D
M AN U
SC
RI PT
1
10
Maki
ACCEPTED MANUSCRIPT
be useful to indicate the presence of myelopathy, but has difficultly predicting the severity of
2
myelopathy.4, 27 Budzik et al. reported that assessment using FA is more useful than MD
3
because of the higher sensitivity of FA for identifying microstructural changes compared with
4
MD.11 Thus, analysis of FA using tract-specific DTI enables a better understanding of the
5
pathology of CCM in the transverse plane of the spinal cord.
6
There are severallimitations to the present study. First, the age of the healthy volunteers and
7
patients with CCM does not match perfectly. FA in the cervical spinal cord decreases with
8
age.28, 29 Although this may contribute to the higher FA seen in the spinal cord of healthy
9
volunteers, it is the asymmetry of the FA that is important in this study. Second, the results of
SC
RI PT
1
the study do not provide guidance on how patients with decreased FA or with asymmetrical
11
symptoms respond to following surgical intervention. This is beyond the scope of our
12
investigation, yet a study is currently underway to determine the association between DTI
13
parameters and functional outcomes after surgery. Third, because automatic selection of tracts
14
in compressed spinal cord is not available, we had to prescribe the ROI manually. In patients
15
with severely compressed spinal cords, there was a possibility of poor reproducibility of ROI
16
placement even though we showed relatively high reliability and reproducibility using ICC
17
analyses. Nevertheless, to our knowledge, this is the first report showing correlations between
18
left and right FA of specific spinal cord tracts and severity of motor dysfunction in left and
19
right extremities of patients with CCM.
TE D
EP
AC C
20
M AN U
10
21
Conclusions
22
Using tract-specific DTI, we demonstrated that microstructural changes in the left and right
23
corticospinal tracts correlated with corresponding neurological symptoms in the ipsilateral
24
side and the FA asymmetry index could indicate laterality in neurological symptoms of
25
patients with CCM. Thus, tract-specific DTI can be used to assess pathological changes in the 11
Maki
ACCEPTED MANUSCRIPT 1
axial plane of the spinal cord in a quantitative manner and enable a better understanding of
2
the relationship between deterioration of the tracts of the spinal cord and the symptoms of
3
CCM.
4 Acknowledgment
6
Funding: This work was supported by Initiative for Accelerating Regulatory Science in
7
Innovative Drug, Medical Device, and Regenerative Medicine the Ministry of Health, Labour
8
and Welfare; National Mutual Insurance Federation of Agricultural Cooperatives; and The
9
General Insurance Association of Japan grant funds.
AC C
EP
TE D
M AN U
SC
RI PT
5
12
Maki
ACCEPTED MANUSCRIPT References
1. Crandall PH, Batzdorf U: Cervical spondylotic myelopathy. J Neurosurg 1966;25:57-66. 2. Brown-Sequard CE. De la transmission des impressions sensitives par la moelle epiniere. CR Soc Biol 192-194, 1849.
RI PT
3. Mihara H, Kondo S, Katoh S, Ishida K, Ono M, Hachiya M: Neurological laterality and surgical outcome in cervical myelopathy patients. J Spine Res 2:1454-1460, 2011.
4. Demir A, Ries M, Moonen CTW, Vital JM, Dehais J, Arne P, Caillé J-M, Dousset V:
SC
Diffusion-weighted MR imaging with apparent diffusion coefficient and apparent diffusion tensor maps in cervical spondylotic myelopathy1. Radiology 229:37-43, 2003.
M AN U
5. Hori M, Okubo T, Aoki S, Kumagai H, Araki T: Line scan diffusion tensor MRI at low magnetic field strength: feasibility study of cervical spondylotic myelopathy in an early clinical stage. J Magn Reson Imaging 23:183-188, 2006.
6. Maki S, Koda M, Ota M, Oikawa Y, Kamiya K, Inada T, Furuya T, Takahashi K, Masuda Y,
TE D
Matsumoto K, Kojima M, Obata T, Yamazaki M: Reduced field-of-view diffusion tensor imaging of the spinal cord shows motor dysfunction of the lower extremities in patients with cervical compression myelopathy. Spine (Phila Pa 1976) August 2015 [Epub ahead of pront].
EP
7. Maynard FM, Bracken MB, Creasey G, Ditunno JF, Donovan WH, Ducker TB, Garber SL,
AC C
Marino RJ, Stover SL, Tator CH, Waters RL, Wilberger JE, Young W: International standards for neurological and functional classification of spinal cord injury. American Spinal Injury Association. Spinal cord Off J Int Med Soc Paraplegia 35:266-274, 1997. 8. Saritas EU, Cunningham CH, Lee JH, Han ET, Nishimura DG: DWI of the spinal cord with reduced FOV single-shot EPI. Magn Reson Med 60:468-473, 2008. 9. Saritas EU, Lee D, Çukur T, Shankaranarayanan A, Nishimura DG: Hadamard slice encoding for reduced-FOV diffusion-weighted imaging. Magn Reson Med 72:1277-90, 2014. 10. Cui JL, Li X, Chan TY, Mak KC, Luk KDK, Hu Y: Quantitative assessment of 13
Maki
ACCEPTED MANUSCRIPT
column-specific degeneration in cervical spondylotic myelopathy based on diffusion tensor tractography. Eur Spine J 24:41-47, 2015. 11. Budzik JF, Balbi V, Le Thuc V, Duhamel A, Assaker R, Cotten A: Diffusion tensor imaging and fibre tracking in cervical spondylotic myelopathy. Eur Radiol 21:426-433, 2011.
RI PT
12. Wen CY, Cui JL, Mak KC, Luk KDK, Hu Y: Diffusion tensor imaging of somatosensory tract in cervical spondylotic myelopathy and its link with electrophysiological evaluation. Spine J 14:1493-1500, 2014.
SC
13. Wang KY, Idowu O, Thompson CB, Orman G, Myers C, Riley LH, Carrino JA, Flammang A, Gilson W, Sadowsky CL, Izbudak I: Tract-specific diffusion tensor imaging in cervical
M AN U
spondylotic myelopathy before and after decompressive spinal surgery: preliminary results. Clin Neuroradiol June 2015 [Epub ahead of pront].
14. Von Meyenburg J, Wilm BJ, Weck A, Petersen J, Gallus E, Mathys J, Schaetzle E, Schubert M, Boesiger P, von Meyenburg K, Goebels N, Kollias S: Spinal cord diffusion-tensor
TE D
Imaging and motor-evoked potentials in multiple sclerosis patients: microstructural and functional asymmetry. Radiology 267:869-879, 2013. 15. Yoo WK, Kim TH, Hai DM, Sundaram S, Yang YM, Park MS, Kim YC, Kwak YH, Ohn
EP
SH, Kim SW: Correlation of magnetic resonance diffusion tensor imaging and clinical findings
AC C
of cervical myelopathy. Spine J 13:867-876, 2013. 16. Nouri A, Martin AR, Mikulis D, Fehlings MG: Magnetic resonance imaging assessment of degenerative cervical myelopathy: a review of structural changes and measurement techniques. Neurosurg Focus 40:E5, 2016. 17. Kadanka Z, Kerkovsky M, Bednarik J, Jarkovsky J: Cross-sectional transverse area and hyperintensities on magnetic resonance imaging in relation to the clinical picture in cervical spondylotic myelopathy. Spine (Phila Pa 1976) 32:2573-2577, 2007. 18. Golash D, Birchall RD, Laitt A: Significance of CSF area measurements in cervical 14
Maki
ACCEPTED MANUSCRIPT spondylitic myelopathy. Br J Neurosurg 15:17-21, 2001.
19. Mihara H, Kondo S, Murata A, Ishida K, Niimura T, Hachiya M: A New Performance Test for Cervical Myelopathy. Spine (Phila Pa 1976) 35:32-35, 2010. 20. Ito T, Oyanagi K, Takahashi H, Takahashi HE, Ikuta F: Cervical spondylotic myelopathy.
RI PT
Clinicopathologic study on the progression pattern and thin myelinated fibers of the lesions of seven patients examined during complete autopsy. Spine (Phila Pa 1976) 21:827-833, 1996. 21. Ellingson BM, Salamon N, Grinstead JW, Holly LT: Diffusion tensor imaging (DTI)
SC
predicts functional impairment in mild to moderate cervical spondylotic myelopathy. Spine J 14:2589-97, 2014.
M AN U
22. Jones JG, Chen SY, Lebel RM, Hsieh PC, Law M: Diffusion tensor imaging correlates with the clinical assessment of disease severity in cervical spondylotic myelopathy and predicts outcome following surgery. AJNR Am J Neuroradiol 34:471-478, 2013. 23. Vedantam A, Jirjis MB, Schmit BD, Wang MC, Ulmer JL, Kurpad SN: Diffusion tensor
TE D
imaging of the spinal cord: insights from animal and human studies. Neurosurgery 74:1-8, 2014.
24. Martin AR, Aleksanderek I, Cohen-Adad J, Tarmohamed Z, Tetreault L, Smith N, Cadotte
EP
DW, Crawley A, Ginsberg H, Mikulis DJ, Fehlings MG: Translating state-of-the-art spinal
AC C
cord MRI techniques to clinical use: A systematic review of clinical studies utilizing DTI, MT, MWF, MRS, and fMRI. NeuroImage Clin 10:192-238, 2016. 25. Budde MD, Kim JH, Liang H-F, Schmidt RE, Russell JH, Cross AH, Song S-K: Toward accurate diagnosis of white matter pathology using diffusion tensor imaging. Magn Reson Med 57:688-695, 2007. 26. DeBoy CA, Zhang J, Dike S, Shats I, Jones M, Reich DS, Mori S, Nguyen T, Rothstein B, Miller RH, Griffin JT, Kerr DA, Calabresi PA: High resolution diffusion tensor imaging of axonal damage in focal inflammatory and demyelinating lesions in rat spinal cord. Brain 15
Maki
ACCEPTED MANUSCRIPT 130:2199-2210, 2007.
27. Uda T, Takami T, Tsuyuguchi N, Sakamoto S, Yamagata T, Ikeda H, Nagata T, Ohata K: Assessment of cervical spondylotic myelopathy using diffusion tensor magnetic resonance imaging parameter at 3.0 tesla. Spine (Phila Pa 1976) 38:407-414, 2013.
RI PT
28. Vedantam A, Jirjis MB, Schmit BD, Wang MC, Ulmer JL, Kurpad SN: Characterization and limitations of diffusion tensor imaging metrics in the cervical spinal cord in neurologically intact subjects. J Magn Reson Imaging 38:861-867, 2013.
SC
29. Agosta F, Laganà M, Valsasina P, Sala S, Dall’Occhio L, Sormani MP, Judica E, Filippi M:
EP
AC C
Figure captions
TE D
Neuroimage 36:728-735, 2007.
M AN U
Evidence for cervical cord tissue disorganisation with aging by diffusion tensor MRI.
Figure 1. Measurement of the left and right cross-sectional area (CSA) of the spinal cord using Image J software, version 1.45 (NIH, Bethesda, MD) The CSA was manually demarcated and divided into left and right halves by the line bisecting the center of the spinous process and vertebra.
Figure 2. Fractional anisotropy (FA) maps of (a) a healthy volunteer and (b) a patient with cervical compression myelopathy (CCM). In the patients with CCM, regions-of-interest 16
Maki
ACCEPTED MANUSCRIPT
(ROIs) were defined on axial images of the FA map at the level of maximum compression. ROIs were placed in the lateral columns (white circles).
Figure 3. Spearman’s correlation between American Spinal Injury Association (ASIA) motor
RI PT
score and fractional anisotropy (FA). Left and right ASIA score correlated moderately with FA of the left and right lateral column (LC) respectively. ASIA motor score was not
SC
correlated with mean diffusivity (MD) of the LC of the spinal cord.
Figure 4. Spearman’s correlation between American Spinal Injury Association (ASIA) motor
M AN U
score and cross-sectional area (CSA) of spinal cord. The left CSA showed only a weak correlation with left ASIA motor score and the right CSA was not correlated with right ASIA motor score.
TE D
Figure 5. A 73-year-old woman with severe myelopathy (Japanese Orthopaedic Association (JOA) score; 10 points) because of cervical spondylotic myelopathy. Her American Spinal Injury Association (ASIA) motor scores were 48 on the left and 41 on the right.
EP
A: T2-weighted sagittal image.
AC C
B: T2-weighted axial image at the level of maximum compression (C4-5). C: Fractional anisotropy (FA) map at the level of maximum compression. FA in the lateral column was 0.45 on the left and 0.29 on the right.
17
ACCEPTED MANUSCRIPT
Table 1. The mean FA and MD values in CCM patients and healthy volunteer
FA in right LC MD in left LC MD in right LC
Healthy volunteers
0.56
0.71
(range 0.32–0.76)
(range 0.62–0.77)
0.55
0.71
(range 0.29–0.75)
(range 0.61–0.78)
1.6×10-3
1.1×10-3
(range 1.0–2.4×10-3)
(range 0.82–1.5×10-3)
1.6×10-3
1.1×10-3
(range 0.75–1.4×10-3)
M AN U
(range 1.0–2.7×10-3)
P <0.001
RI PT
FA in left LC
CCM patients
SC
DTI parameters
<0.001 <0.001 <0.001
DTI: diffusion tensor imaging, CCM: cervical compression myelopathy, FA: fractional
AC C
EP
TE D
anisotropy, LC: lateral column, MD: mean diffusivity.
ACCEPTED MANUSCRIPT
Table 2. The mean FA and MD asymmetry indices in CCM patients and healthy volunteer
CCM patients
Healthy volunteers
P
FA asymmetry index
5.0 ± 4.9%
0.8 ± 0.5%
<0.001
MD asymmetry index
6.3 ± 5.7%
1.7 ± 1.4%
RI PT
Asymmetry index
0.004
AC C
EP
TE D
M AN U
SC
CCM: cervical compression myelopathy, FA: fractional anisotropy, MD: mean diffusivity.
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Fractional anisotropy correlated with left and right symptoms independently.
Fractional anisotropy asymmetry index could indicate laterality in symptoms. Ten percent of the patients exhibited laterality of symptoms in their extremities.
SC M AN U TE D EP AC C
RI PT
ACCEPTED MANUSCRIPT
Abbreviations: CCM: cervical compression myelopathy DTI: diffusion tensor imaging
RI PT
FA: fractional anisotropy ASIA: American Spinal Injury Association ROI: region of interest LC: lateral column
SC
MD: mean diffusivity CSA: cross sectional area
T2W: T2-weighted, FOV: field of view, TE: echo time TR: time of repetition
TE D
ICC: intraclass correlation coefficients
M AN U
T1W: T1-weighted,
AC C
EP
ROC: receiver operator characteristic
ACCEPTED MANUSCRIPT
Conflict of interest
AC C
EP
TE D
M AN U
SC
RI PT
The authors declare that there is no conflict of interest regarding the publication of this paper
S1878-8750(16)30813-0
DOI:
10.1016/j.wneu.2016.08.129
Reference:
WNEU 4540
To appear in:
World Neurosurgery
Received Date: 22 July 2016 Revised Date:
29 August 2016
Accepted Date: 30 August 2016
Please cite this article as: Maki S, Koda M, Saito J, Takahashi S, Inada T, Kamiya K, Ota M, Iijima Y, Masuda Y, Matsumoto K, Kojima M, Takahashi K, Obata T, Yamazaki M, Furuya T, Tract-specific Diffusion Tensor Imaging Reveals Laterality of Neurological Symptoms in Patients with Cervical Compression Myelopathy, World Neurosurgery (2016), doi: 10.1016/j.wneu.2016.08.129. 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.
Maki
ACCEPTED MANUSCRIPT
Tract-specific Diffusion Tensor Imaging Reveals Laterality of Neurological Symptoms in
2
Patients with Cervical Compression Myelopathy
3
Satoshi Maki1,*, Masao Koda1, Junya Saito1 , Sho Takahashi2, Taigo Inada1, Koshiro
4
Kamiya1, Mitsutoshi Ota1, Yasushi Iijima1, Yoshitada Masuda3, Koji Matsumoto3, Masatoshi
5
Kojima3, Kazuhisa Takahashi1, Takayuki Obata4, Masashi Yamazaki5, Takeo Furuya1
6
1
7
1-8-1 Inohana Chuou-ku Chiba City, Chiba 260-8670, Japan
8
2
9
1-8-1 Inohana Chuou-ku Chiba City, Chiba 260-8677, Japan
RI PT
1
Department of Orthopaedic Surgery, Chiba University Graduate School of Medicine
SC
Clinical Research Center, Chiba University Hospital
10
3
11
1-8-1 Inohana Chuou-ku Chiba City, Chiba 260-8677, Japan
12
4
13
4-9-1 Anagawa, Inage-ku, Chiba City, Chiba 263-8555, Japan
14
5
15
1-1-1 Tennodai, Tsukuba City, Ibaragi, 305-8575, Japan
M AN U
Department of Radiology, Chiba University Hospital
Center for Charged Particle Therapy, National Institute of Radiological Sciences
TE D
Department of Orthopaedic Surgery, Faculty of Medicine, University of Tsukuba
16
*Address correspondence and reprint requests to:
18
Satoshi Maki, MD, PhD
19
Department of Orthopaedic Surgery, Chiba University Graduate School of Medicine
20
1-8-1 Inohana, Chuou-ku, Chiba 260-8670, Japan
21
Tel: 043-226-2117, Fax: 043-226-2116
22
E-mail: [email protected]
AC C
EP
17
23 24
Keywords Diffusion tensor imaging, Spinal cord, Cervical compression myelopathy,
25
Laterality, Asymmetry index 1
Maki
ACCEPTED MANUSCRIPT Abstract
2
Backbround: Patients with cervical compression myelopathy (CCM) generally present
3
bilateral neurological symptoms in their extremities. However, a substantial portion of
4
patients with CCM exhibit laterality of neurological symptoms. The aim of this study was to
5
assess the correlation between intrinsic structural damage and laterality of symptoms using
6
spinal cord diffusion tensor imaging (DTI) of the corticospinal tract.
7
Methods: We enrolled 10 healthy volunteers and 40 patients with CCM in this study. We
8
evaluated motor function using the American Spinal Injury Association (ASIA) motor score
9
for left and right extremities. For DTI acquisitions, a 3.0 T MR system with
SC
RI PT
1
diffusion-weighted spin-echo sequence was used. Regions-of-interest (ROIs) in the lateral
11
column (LC) tracts were determined. We determined the correlations between fractional
12
anisotropy (FA) and ASIA motor scores. An FA asymmetry index was calculated using left
13
and right ROIs.
14
Results: Four patients exhibited laterality of symptoms in their extremities, for which left and
15
right ASIA scores correlated moderately with FA in the left and right LCs respectively (left: ρ
16
= 0.64, P < 0.001; right: ρ = 0.67, P < 0.001). The area under the receiver-operator
17
characteristic curve showed the FA asymmetry index indicated laterality of symptoms.
18
Conclusions: Using tract-specific DTI, we demonstrated that microstructural damages in the
19
left and right corticospinal tracts correlated with corresponding neurological symptoms in the
20
ipsilateral side and the FA asymmetry index could indicate laterality in neurological
21
symptoms of patients with CCM.
AC C
EP
TE D
M AN U
10
22 23
Keywords Diffusion tensor imaging; Spinal cord; Cervical compression myelopathy;
24
Laterality; Asymmetry index
25 2
Maki
ACCEPTED MANUSCRIPT Introduction
2
Patients with cervical compression myelopathy (CCM), including cervical spondylotic
3
myelopathy, ossification of the longitudinal ligament, and cervical disk herniation, generally
4
present bilateral extremity symptoms. Five types of spinal cord lesions in patients with CCM
5
were proposed by Crandall et al.1 The classification is widely accepted and used because it
6
effectively describes neurological symptoms based on neuroanatomy. The most frequently
7
encountered Crandall classification is transverse syndrome, which generally presents bilateral
8
symptoms. Brown-Sequard syndrome is a unilateral cord lesion with ipsilateral motor deficit
9
and contralateral analgesia below the lesion.2 However, a substantial portion of patients with
SC
RI PT
1
CCM exhibit laterality of neurological symptoms without showing contralateral analgesia
11
below the lesion. Few studies regarding laterality of neurological symptoms in patients with
12
CCM have been reported.3
13
It is difficult to quantify spinal cord dysfunction by conventional magnetic resonance imaging
14
(MRI). However, diffusion tensor imaging (DTI) can detect microstructural changes in the
15
white matter of the spinal cord and provide quantitative diffusion parameters. Fractional
16
anisotropy (FA) and mean diffusivity (MD) are derived from the diffusion matrix and
17
represent the degree of anisotropy and magnitude of the diffusion process. FA and MD reflect
18
tissue integrity in the spinal cord of patients with CCM. This hypothesis is supported by
19
several studies.4, 5 Maki et al. have reported that FA in lateral and posterior columns in
20
tract-specific DTI correlated with gait disturbance in patients with CCM.6
21
The aim of this study was to assess the correlation between intrinsic structural damage and
22
laterality of symptoms using spinal cord diffusion tensor imaging (DTI) of the corticospinal
23
tract.
AC C
EP
TE D
M AN U
10
24 25
Materials and Methods 3
Maki
ACCEPTED MANUSCRIPT Participants
2
All participants provided written, informed consent before participating in the research,
3
which was previously approved by our institutional review board. We enrolled 10 healthy
4
volunteers and 40 patients with CCM in this study. We excluded potential participants with a
5
history of neurological injury, spinal surgery, or neurodegenerative disease causing any spinal
6
cord disorder. Healthy volunteers included 6 men and 4 women with a mean age of 43.7 ± 8.6
7
years. In the group of patients with CCM, there were 26 men and 14 women with a mean age
8
of 65.3 ± 11.0 years. The clinical diagnoses included 18 patients with cervical spondylotic
9
myelopathy, 19 with ossification of the posterior longitudinal ligament, 2 with cervical disk
SC
herniation, and 1 with atlantoaxial dislocation because of rheumatoid arthritis.
M AN U
10
RI PT
1
11 Clinical Evaluation
13
We evaluated motor function using the American Spinal Injury Association (ASIA) motor
14
score for left and right upper and lower extremities. Left and right motor scores ranged from
15
0 to 50.7
16
TE D
12
Conventional MRI
18
The patients were studied using a 3.0 T MRI system (Discovery MR 750, GE Healthcare,
19
Waukesha, WI). An 8-ch head, neck, and spine receiver coil was used. Conventional MRI
20
consisted of T1-weighted (T1W) and T2-weighted (T2W) sagittal images, and T2W axial
21
images were also obtained. For sagittal imaging, the acquisition settings were as follows:
22
field-of-view (FOV) = 24 × 24 cm2, slice thickness = 3 mm echo time (TE)/time of repetition
23
(TR) = 8.4/500 ms (T1W) and TE/TR = 105/2710 ms (T2W). For axial imaging, the
24
acquisition settings were as follows: FOV = 16 × 16 cm2, slice thickness = 4 mm, TE/TR =
25
105/4000 ms.
AC C
EP
17
4
Maki
ACCEPTED MANUSCRIPT 1 Diffusion Tensor Imaging
3
For DTI acquisitions, diffusion-weighted spin-echo single-shot echo-planar imaging was
4
used, with diffusion gradients applied in 6 noncollinear spatial directions and 1 baseline
5
image with b = 700 s/mm2. Reduced FOV technique was applied to reduce susceptibility
6
artifacts and achieve higher resolution.8, 9 The thickness of each slice was 5 mm and the
7
sequence parameters for DTI were: TE = 74.9 ms, TR = 3000 ms, FOV = 14 × 4.2 cm2,
8
number of excitations = 16, and the acquisition data matrix = 176 × 54. Transverse images
9
covering the cervical spinal cord from C1 to T1 were acquired, each of which was placed at
SC
the center of either a vertebra or intervertebral disk.
M AN U
10
RI PT
2
11 Image Analysis
13
We identified the level of maximum compression on T2W axial images, and divided the
14
cross-sectional area (CSA) of the spinal cord into left and right halves using the line bisecting
15
the center of the spinous process and vertebra. We measured left and right CSAs using Image
16
J software, version 1.45 (NIH, Bethesda, MD) (Fig. 1).
17
We quantified diffusion tensor parameters using DTIStudio software (Johns Hopkins Medical
18
Institute, Johns Hopkins University). Five levels from C2/3 to C6/7 were measured in control
19
group. The level of maximum compression was measured in patients with CCM.
20
Regions-of-interest (ROI) for the lateral column (LC) tracts were determined based on the
21
geometry of the cord on the axial image of the FA map and FA and MD were measured (Fig.
22
2). The size of ROIs was standardized at 6-8 pixels. We analyzed the correlation between FA
23
and MD and left and right ASIA motor scores.
AC C
EP
TE D
12
24 25
Reliability of ROI Measurement for DTI Analysis 5
Maki
ACCEPTED MANUSCRIPT 1
ROI measurements were made twice by 2 experienced spinal surgeons at 2-week intervals.
2
Intraclass correlation coefficients (ICC) were used to determine the interobserver and
3
intraobserver reliabilities. We provide 95% prediction limits for the errors in measurements.
4 Statistical Analyses
6
A Mann–Whitney U test was used to compare means of FA and MD values and asymmetry
7
indices between patients with CCM and healthy controls. Correlations between DTI
8
parameters and left and right ASIA motor scores were determined using Spearman’s
9
correlation coefficients. The FA symmetry index for healthy volunteers and patients with
SC
RI PT
5
CCM was calculated. The FA asymmetry index is the percentage of the absolute difference
11
between the FA in the left and right ROIs relative to the sum of the respective FA (asymmetry
12
index = |FAright – FAleft|/(FAright + FAleft) × 100). The asymmetry indices for MD and CSA
13
values were also calculated. The prediction of asymmetrical symptoms using the FA, MD and
14
CSA asymmetry index was determined using the area under a receiver–operator characteristic
15
(ROC) curve. P < 0.05 was considered significant. All the analyses were conducted using
16
JMP software, version 10.0.2 (SAS Institute, Cary, NC).
17
EP
TE D
M AN U
10
Results
19
Clinical Symptoms
20
The ASIA motor score for the left side was 47.6 (range 36–50) and for the right side was 47.7
21
(range 38–50). The difference between the left and right ASIA scores was 0.1 ± 2.3. Because
22
the asymmetry of symptoms between left and right was more than 2 standard deviations
23
(SDs) (i.e. more than 4 points difference in left and right ASIA motor score), 4 of 40 (10%)
24
patients had laterality of neurological symptoms. None of these patients presented
25
Brown-Sequard syndrome, and were considered to have transverse syndrome.
AC C
18
6
Maki
ACCEPTED MANUSCRIPT 1 FA, MD and asymmetry indices
3
DTI analyses revealed that the FA in patients with CCM was significantly lower than that in
4
healthy volunteers in both left and right LC (Table 1). The MD in patients with CCM was
5
significantly higher than that in healthy volunteers in both left and right LC (Table 1). The FA
6
and MD asymmetry indices in patients with CCM were significantly higher than those in
7
healthy volunteers (Table 2).
8 9
Correlation between ASIA Score and DTI Indices
SC
RI PT
2
In patients with CCM, left and right ASIA scores correlated moderately with FA in the left
11
and right LCs (left: ρ = 0.64, P < 0.001; right: ρ = 0.67, P < 0.001; Fig. 3). However, left and
12
right ASIA scores were not correlated with MD in the right or left LC respectively (left: ρ =
13
-0.13, P = 0.43; right: ρ = -0.26, P = 0.10; Fig. 3).
M AN U
10
TE D
14
Reliability of ROI Measurement for DTI Analysis
16
The ICC for interobserver reliability of ROI measurements was 0.87 for the FA value and
17
0.76 for the MD value. The ICC for intraobserver reliability of ROI measurements was 0.90
18
for the FA value and 0.93 for the MD value. The inter- and intraobserver reliability of tract
19
specific ROI measurements was considered reliable for analyzing DTI of the patients with
20
CCM.
AC C
21
EP
15
22
Cross-sectional Area (CSA) of the Spinal Cord
23
The mean left and right CSAs of the spinal cord in patients with CCM were 25.4 ± 8.2 mm2
24
and 26.6 ± 7.9 mm2 respectively at the most apparent level of cord compression. The left
25
CSAs showed only a weak correlation with left ASIA motor scores and right CSAs showed 7
Maki
ACCEPTED MANUSCRIPT 1
no correlation with right ASIA motor scores (left: ρ = 0.39, P = 0.01; right: ρ = 0.28, P =
2
0.08; Fig.4). The asymmetry index for CSA was 11.6 ± 9.1%.
3 Detection of the Laterality of Neurological Symptoms
5
The area under the receiver–operator characteristic (ROC) curve for the FA asymmetry index
6
was 0.86, which indicated that the FA asymmetry index could moderately indicate the
7
laterality of neurological symptoms in the left and right extremities. On the other hand, the
8
area under the ROC curve for the MD and CSA asymmetry index were 0.57 and 0.54,
9
respectively, which indicated that both MD and CSA asymmetry index had low accuracy for
SC
detecting the laterality of neurological symptoms of left and right extremities.
M AN U
10
RI PT
4
11 DISCUSSION
13
The present study shows that it is feasible to demonstrate correlation between left and right
14
microstructural damages in spinal cord and corresponding spinal cord long tract symptoms
15
independently and to reveal asymmetry of neurological symptoms in patients with CCM
16
using tract-specific DTI. A certain portion of patients with CCM exhibited asymmetrical
17
symptoms in their upper and lower extremities.
18
Tract-specific analysis of the spinal cord is gaining popularity because high resolution images
19
can be acquired by reduced FOV techniques.8, 9 However, most studies of DTI for evaluating
20
the spinal cord at the tract level target demyelinating disease in patients whose spinal cord
21
morphology is preserved. There are few DTI studies evaluating patients with CCM at the
22
tract level.6,10–13 Even fewer DTI studies focus on laterality of the neurological symptoms in
23
patients with demyelinating disease and CCM. Meyenburg et al. reported that DTI revealed
24
asymmetrical changes in the spinal cord of patients with multiple sclerosis, which
25
corresponded to asymmetrical electrophysiological deficits for both arms and legs, and
AC C
EP
TE D
12
8
Maki
ACCEPTED MANUSCRIPT
reflected a specific structure–function relationship in the spinal cord.14 Yoo et al. analyzed 9
2
patients with CCM using DTI and sorted DTI parameters into affected and unaffected sides
3
according to the presence or the absence of symptoms. They found a significant decrease in
4
FA of anterior and lateral ROIs in the affected side compared with a control group. However,
5
their patient population was relatively small and the definition of affected and unaffected side
6
was not well described.15 Cui et al. conducted column specific analysis of the spinal cord in
7
patients with cervical spondylotic myelopathy. In their report, they presented a patient with
8
left hand clumsiness and spasticity of the left lower limb showing low FA in DTI of the
9
lateral and posterior column on the left side of the spinal cord as a representative case.10 The
10
present study focused on laterality of symptoms in patients with CCM and revealed left and
11
right decreases in FA in the LCs corresponding to left and right limb motor dysfunction (Fig.
12
5). By contrast, spinal cord CSA was not correlated with ASIA motor score. Correlation
13
between CSA and clinical symptoms of CCM remains uncertain and it seems difficult to
14
predict the severity of the symptoms using CSA.16 Kadanka et al. reported that there was a
15
significant difference in the modified JOA score of patients with cervical spondylotic
16
myelopathy between those with CSA <50 mm2 and those with CSA >60 mm2.17 Golash et al.
17
demonstrated that narrowness of the spinal cord CSA was well correlated with severity of
18
established myelopathy, but was an unsatisfactory prognosticator for mild degrees of
19
myelopathy.18
20
In the present study, 10% of the patients with CCM exhibited asymmetrical symptoms in
21
their extremities. There are few studies referring to laterality of neurological symptoms in
22
patients with CCM. Mihara et al. found that 12% of patients with CCM had laterality of
23
symptoms in their extremities, which was defined as 4 times or more difference in a left and
24
right limb performance test (a finger grip and release test and a triangle step test).19 They
25
compared surgical outcomes between patients with cervical myelopathy who had either
AC C
EP
TE D
M AN U
SC
RI PT
1
9
Maki
ACCEPTED MANUSCRIPT symmetrical or asymmetrical symptoms.3 JOA score recovery rates in patients with
2
asymmetrical symptoms were less than those of patients with mild symmetrical symptoms,
3
but were greater than those of patients with severe symmetrical symptoms. The investigators
4
hypothesized that cervical myelopathy might progress through a period of asymmetrical
5
symptoms before producing severe symmetrical symptoms. However, in the present study, we
6
identified patients with relatively severe impairment as having asymmetrical symptoms. The
7
difference can be attributed to the definition of asymmetrical symptoms, which were defined
8
by more than two SDs difference in ASIA motor score in the current study, and not by a
9
performance test. Because 4 points loss in ASIA motor score represents substantial motor
10
weakness, patients with asymmetrical symptoms present poor functional status. Previous
11
neurohistopathological studies of patients with cervical spondylotic myelopathy showed that
12
the distribution of the lesion in the spinal cord was not homogeneous, which supports the
13
hypothesis that degeneration of the LC can occur asymmetrically.20 A proportion of our
14
patients presented asymmetrical symptoms in their extremities even though they were
15
classified as having a transverse syndrome, but not Brown-Sequard syndrome.
16
Decrease in FA and increase in MD at the compression site were seen in patients with CCM
17
compared with healthy volunteers, which was consistent with previous studies.21–24 Decrease
18
in FA is considered to reflect demyelination and axon damage in the spinal cord. 23-26 The
19
current study showed a moderate correlation between decrease in FA and left and right ASIA
20
motor score. The LC contains the lateral corticospinal tract, which is the principal descending
21
motor fiber tract conducting voluntary movement in the human spinal cord. Therefore,
22
damage in the LC, which is reflected by a decrease in FA, is considered to result in decreased
23
ASIA motor score. By contrast, the present study showed no correlation between MD and
24
ASIA motor score, which is consistent with previous findings.11, 21 Increase in MD is
25
considered to occur at the initial stage of CCM, and is possibly related to edema.15 MD may
AC C
EP
TE D
M AN U
SC
RI PT
1
10
Maki
ACCEPTED MANUSCRIPT
be useful to indicate the presence of myelopathy, but has difficultly predicting the severity of
2
myelopathy.4, 27 Budzik et al. reported that assessment using FA is more useful than MD
3
because of the higher sensitivity of FA for identifying microstructural changes compared with
4
MD.11 Thus, analysis of FA using tract-specific DTI enables a better understanding of the
5
pathology of CCM in the transverse plane of the spinal cord.
6
There are severallimitations to the present study. First, the age of the healthy volunteers and
7
patients with CCM does not match perfectly. FA in the cervical spinal cord decreases with
8
age.28, 29 Although this may contribute to the higher FA seen in the spinal cord of healthy
9
volunteers, it is the asymmetry of the FA that is important in this study. Second, the results of
SC
RI PT
1
the study do not provide guidance on how patients with decreased FA or with asymmetrical
11
symptoms respond to following surgical intervention. This is beyond the scope of our
12
investigation, yet a study is currently underway to determine the association between DTI
13
parameters and functional outcomes after surgery. Third, because automatic selection of tracts
14
in compressed spinal cord is not available, we had to prescribe the ROI manually. In patients
15
with severely compressed spinal cords, there was a possibility of poor reproducibility of ROI
16
placement even though we showed relatively high reliability and reproducibility using ICC
17
analyses. Nevertheless, to our knowledge, this is the first report showing correlations between
18
left and right FA of specific spinal cord tracts and severity of motor dysfunction in left and
19
right extremities of patients with CCM.
TE D
EP
AC C
20
M AN U
10
21
Conclusions
22
Using tract-specific DTI, we demonstrated that microstructural changes in the left and right
23
corticospinal tracts correlated with corresponding neurological symptoms in the ipsilateral
24
side and the FA asymmetry index could indicate laterality in neurological symptoms of
25
patients with CCM. Thus, tract-specific DTI can be used to assess pathological changes in the 11
Maki
ACCEPTED MANUSCRIPT 1
axial plane of the spinal cord in a quantitative manner and enable a better understanding of
2
the relationship between deterioration of the tracts of the spinal cord and the symptoms of
3
CCM.
4 Acknowledgment
6
Funding: This work was supported by Initiative for Accelerating Regulatory Science in
7
Innovative Drug, Medical Device, and Regenerative Medicine the Ministry of Health, Labour
8
and Welfare; National Mutual Insurance Federation of Agricultural Cooperatives; and The
9
General Insurance Association of Japan grant funds.
AC C
EP
TE D
M AN U
SC
RI PT
5
12
Maki
ACCEPTED MANUSCRIPT References
1. Crandall PH, Batzdorf U: Cervical spondylotic myelopathy. J Neurosurg 1966;25:57-66. 2. Brown-Sequard CE. De la transmission des impressions sensitives par la moelle epiniere. CR Soc Biol 192-194, 1849.
RI PT
3. Mihara H, Kondo S, Katoh S, Ishida K, Ono M, Hachiya M: Neurological laterality and surgical outcome in cervical myelopathy patients. J Spine Res 2:1454-1460, 2011.
4. Demir A, Ries M, Moonen CTW, Vital JM, Dehais J, Arne P, Caillé J-M, Dousset V:
SC
Diffusion-weighted MR imaging with apparent diffusion coefficient and apparent diffusion tensor maps in cervical spondylotic myelopathy1. Radiology 229:37-43, 2003.
M AN U
5. Hori M, Okubo T, Aoki S, Kumagai H, Araki T: Line scan diffusion tensor MRI at low magnetic field strength: feasibility study of cervical spondylotic myelopathy in an early clinical stage. J Magn Reson Imaging 23:183-188, 2006.
6. Maki S, Koda M, Ota M, Oikawa Y, Kamiya K, Inada T, Furuya T, Takahashi K, Masuda Y,
TE D
Matsumoto K, Kojima M, Obata T, Yamazaki M: Reduced field-of-view diffusion tensor imaging of the spinal cord shows motor dysfunction of the lower extremities in patients with cervical compression myelopathy. Spine (Phila Pa 1976) August 2015 [Epub ahead of pront].
EP
7. Maynard FM, Bracken MB, Creasey G, Ditunno JF, Donovan WH, Ducker TB, Garber SL,
AC C
Marino RJ, Stover SL, Tator CH, Waters RL, Wilberger JE, Young W: International standards for neurological and functional classification of spinal cord injury. American Spinal Injury Association. Spinal cord Off J Int Med Soc Paraplegia 35:266-274, 1997. 8. Saritas EU, Cunningham CH, Lee JH, Han ET, Nishimura DG: DWI of the spinal cord with reduced FOV single-shot EPI. Magn Reson Med 60:468-473, 2008. 9. Saritas EU, Lee D, Çukur T, Shankaranarayanan A, Nishimura DG: Hadamard slice encoding for reduced-FOV diffusion-weighted imaging. Magn Reson Med 72:1277-90, 2014. 10. Cui JL, Li X, Chan TY, Mak KC, Luk KDK, Hu Y: Quantitative assessment of 13
Maki
ACCEPTED MANUSCRIPT
column-specific degeneration in cervical spondylotic myelopathy based on diffusion tensor tractography. Eur Spine J 24:41-47, 2015. 11. Budzik JF, Balbi V, Le Thuc V, Duhamel A, Assaker R, Cotten A: Diffusion tensor imaging and fibre tracking in cervical spondylotic myelopathy. Eur Radiol 21:426-433, 2011.
RI PT
12. Wen CY, Cui JL, Mak KC, Luk KDK, Hu Y: Diffusion tensor imaging of somatosensory tract in cervical spondylotic myelopathy and its link with electrophysiological evaluation. Spine J 14:1493-1500, 2014.
SC
13. Wang KY, Idowu O, Thompson CB, Orman G, Myers C, Riley LH, Carrino JA, Flammang A, Gilson W, Sadowsky CL, Izbudak I: Tract-specific diffusion tensor imaging in cervical
M AN U
spondylotic myelopathy before and after decompressive spinal surgery: preliminary results. Clin Neuroradiol June 2015 [Epub ahead of pront].
14. Von Meyenburg J, Wilm BJ, Weck A, Petersen J, Gallus E, Mathys J, Schaetzle E, Schubert M, Boesiger P, von Meyenburg K, Goebels N, Kollias S: Spinal cord diffusion-tensor
TE D
Imaging and motor-evoked potentials in multiple sclerosis patients: microstructural and functional asymmetry. Radiology 267:869-879, 2013. 15. Yoo WK, Kim TH, Hai DM, Sundaram S, Yang YM, Park MS, Kim YC, Kwak YH, Ohn
EP
SH, Kim SW: Correlation of magnetic resonance diffusion tensor imaging and clinical findings
AC C
of cervical myelopathy. Spine J 13:867-876, 2013. 16. Nouri A, Martin AR, Mikulis D, Fehlings MG: Magnetic resonance imaging assessment of degenerative cervical myelopathy: a review of structural changes and measurement techniques. Neurosurg Focus 40:E5, 2016. 17. Kadanka Z, Kerkovsky M, Bednarik J, Jarkovsky J: Cross-sectional transverse area and hyperintensities on magnetic resonance imaging in relation to the clinical picture in cervical spondylotic myelopathy. Spine (Phila Pa 1976) 32:2573-2577, 2007. 18. Golash D, Birchall RD, Laitt A: Significance of CSF area measurements in cervical 14
Maki
ACCEPTED MANUSCRIPT spondylitic myelopathy. Br J Neurosurg 15:17-21, 2001.
19. Mihara H, Kondo S, Murata A, Ishida K, Niimura T, Hachiya M: A New Performance Test for Cervical Myelopathy. Spine (Phila Pa 1976) 35:32-35, 2010. 20. Ito T, Oyanagi K, Takahashi H, Takahashi HE, Ikuta F: Cervical spondylotic myelopathy.
RI PT
Clinicopathologic study on the progression pattern and thin myelinated fibers of the lesions of seven patients examined during complete autopsy. Spine (Phila Pa 1976) 21:827-833, 1996. 21. Ellingson BM, Salamon N, Grinstead JW, Holly LT: Diffusion tensor imaging (DTI)
SC
predicts functional impairment in mild to moderate cervical spondylotic myelopathy. Spine J 14:2589-97, 2014.
M AN U
22. Jones JG, Chen SY, Lebel RM, Hsieh PC, Law M: Diffusion tensor imaging correlates with the clinical assessment of disease severity in cervical spondylotic myelopathy and predicts outcome following surgery. AJNR Am J Neuroradiol 34:471-478, 2013. 23. Vedantam A, Jirjis MB, Schmit BD, Wang MC, Ulmer JL, Kurpad SN: Diffusion tensor
TE D
imaging of the spinal cord: insights from animal and human studies. Neurosurgery 74:1-8, 2014.
24. Martin AR, Aleksanderek I, Cohen-Adad J, Tarmohamed Z, Tetreault L, Smith N, Cadotte
EP
DW, Crawley A, Ginsberg H, Mikulis DJ, Fehlings MG: Translating state-of-the-art spinal
AC C
cord MRI techniques to clinical use: A systematic review of clinical studies utilizing DTI, MT, MWF, MRS, and fMRI. NeuroImage Clin 10:192-238, 2016. 25. Budde MD, Kim JH, Liang H-F, Schmidt RE, Russell JH, Cross AH, Song S-K: Toward accurate diagnosis of white matter pathology using diffusion tensor imaging. Magn Reson Med 57:688-695, 2007. 26. DeBoy CA, Zhang J, Dike S, Shats I, Jones M, Reich DS, Mori S, Nguyen T, Rothstein B, Miller RH, Griffin JT, Kerr DA, Calabresi PA: High resolution diffusion tensor imaging of axonal damage in focal inflammatory and demyelinating lesions in rat spinal cord. Brain 15
Maki
ACCEPTED MANUSCRIPT 130:2199-2210, 2007.
27. Uda T, Takami T, Tsuyuguchi N, Sakamoto S, Yamagata T, Ikeda H, Nagata T, Ohata K: Assessment of cervical spondylotic myelopathy using diffusion tensor magnetic resonance imaging parameter at 3.0 tesla. Spine (Phila Pa 1976) 38:407-414, 2013.
RI PT
28. Vedantam A, Jirjis MB, Schmit BD, Wang MC, Ulmer JL, Kurpad SN: Characterization and limitations of diffusion tensor imaging metrics in the cervical spinal cord in neurologically intact subjects. J Magn Reson Imaging 38:861-867, 2013.
SC
29. Agosta F, Laganà M, Valsasina P, Sala S, Dall’Occhio L, Sormani MP, Judica E, Filippi M:
EP
AC C
Figure captions
TE D
Neuroimage 36:728-735, 2007.
M AN U
Evidence for cervical cord tissue disorganisation with aging by diffusion tensor MRI.
Figure 1. Measurement of the left and right cross-sectional area (CSA) of the spinal cord using Image J software, version 1.45 (NIH, Bethesda, MD) The CSA was manually demarcated and divided into left and right halves by the line bisecting the center of the spinous process and vertebra.
Figure 2. Fractional anisotropy (FA) maps of (a) a healthy volunteer and (b) a patient with cervical compression myelopathy (CCM). In the patients with CCM, regions-of-interest 16
Maki
ACCEPTED MANUSCRIPT
(ROIs) were defined on axial images of the FA map at the level of maximum compression. ROIs were placed in the lateral columns (white circles).
Figure 3. Spearman’s correlation between American Spinal Injury Association (ASIA) motor
RI PT
score and fractional anisotropy (FA). Left and right ASIA score correlated moderately with FA of the left and right lateral column (LC) respectively. ASIA motor score was not
SC
correlated with mean diffusivity (MD) of the LC of the spinal cord.
Figure 4. Spearman’s correlation between American Spinal Injury Association (ASIA) motor
M AN U
score and cross-sectional area (CSA) of spinal cord. The left CSA showed only a weak correlation with left ASIA motor score and the right CSA was not correlated with right ASIA motor score.
TE D
Figure 5. A 73-year-old woman with severe myelopathy (Japanese Orthopaedic Association (JOA) score; 10 points) because of cervical spondylotic myelopathy. Her American Spinal Injury Association (ASIA) motor scores were 48 on the left and 41 on the right.
EP
A: T2-weighted sagittal image.
AC C
B: T2-weighted axial image at the level of maximum compression (C4-5). C: Fractional anisotropy (FA) map at the level of maximum compression. FA in the lateral column was 0.45 on the left and 0.29 on the right.
17
ACCEPTED MANUSCRIPT
Table 1. The mean FA and MD values in CCM patients and healthy volunteer
FA in right LC MD in left LC MD in right LC
Healthy volunteers
0.56
0.71
(range 0.32–0.76)
(range 0.62–0.77)
0.55
0.71
(range 0.29–0.75)
(range 0.61–0.78)
1.6×10-3
1.1×10-3
(range 1.0–2.4×10-3)
(range 0.82–1.5×10-3)
1.6×10-3
1.1×10-3
(range 0.75–1.4×10-3)
M AN U
(range 1.0–2.7×10-3)
P <0.001
RI PT
FA in left LC
CCM patients
SC
DTI parameters
<0.001 <0.001 <0.001
DTI: diffusion tensor imaging, CCM: cervical compression myelopathy, FA: fractional
AC C
EP
TE D
anisotropy, LC: lateral column, MD: mean diffusivity.
ACCEPTED MANUSCRIPT
Table 2. The mean FA and MD asymmetry indices in CCM patients and healthy volunteer
CCM patients
Healthy volunteers
P
FA asymmetry index
5.0 ± 4.9%
0.8 ± 0.5%
<0.001
MD asymmetry index
6.3 ± 5.7%
1.7 ± 1.4%
RI PT
Asymmetry index
0.004
AC C
EP
TE D
M AN U
SC
CCM: cervical compression myelopathy, FA: fractional anisotropy, MD: mean diffusivity.
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Fractional anisotropy correlated with left and right symptoms independently.
Fractional anisotropy asymmetry index could indicate laterality in symptoms. Ten percent of the patients exhibited laterality of symptoms in their extremities.
SC M AN U TE D EP AC C
RI PT
ACCEPTED MANUSCRIPT
Abbreviations: CCM: cervical compression myelopathy DTI: diffusion tensor imaging
RI PT
FA: fractional anisotropy ASIA: American Spinal Injury Association ROI: region of interest LC: lateral column
SC
MD: mean diffusivity CSA: cross sectional area
T2W: T2-weighted, FOV: field of view, TE: echo time TR: time of repetition
TE D
ICC: intraclass correlation coefficients
M AN U
T1W: T1-weighted,
AC C
EP
ROC: receiver operator characteristic
ACCEPTED MANUSCRIPT
Conflict of interest
AC C
EP
TE D
M AN U
SC
RI PT
The authors declare that there is no conflict of interest regarding the publication of this paper