See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/260483648

Generalized Tonic-Clonic Seizures: Aberrant Interhemispheric Functional and

Anatomical Connectivity

Article  in  Radiology · March 2014

DOI: 10.1148/radiol.13131638 · Source: PubMed

CITATIONS READS
51 400

6 authors, including:

Gong-Jun Ji Zhiqiang Zhang

Anhui Medical University Nanjing University
55 PUBLICATIONS   657 CITATIONS    238 PUBLICATIONS   3,992 CITATIONS   

SEE PROFILE SEE PROFILE

Qun Xu Yu-Feng Zang

HPLC Applications Lab, ThermoFisher Scientific, Hangzhou Normal University
852 PUBLICATIONS   16,443 CITATIONS    311 PUBLICATIONS   21,300 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

neuroimaging of FGIDs View project

The physiological meanning of white matter BOLD signal during resting state View project

All content following this page was uploaded by Yu-Feng Zang on 18 May 2015.

The user has requested enhancement of the downloaded file.

FYI Radiology Manuscript Proofs
Manuscript Proofs
Your manuscript has been edited for RSNA style and usage, which are based on widely
accepted conventions of grammar and the American Medical Association Manual of Style.
Please read the attached copy carefully to ensure that your meaning has been retained.

Author Queries
Queries are indicated in the margins and listed at the end of the page proofs. Please keep
your changes to a minimum and correct only inaccurate items. Figure sizing and many
layout decisions have already been made. Annotate the PDF proofs with your corrections
using Acrobat comment and markup tools. We recommend that you retain a copy of the
corrected manuscript.

Figures
Check the fidelity and orientation of the figures. Check the placement of keys (arrows, letters,
etc) on the illustrations; if repositioning is necessary, indicate the proper location on the figure
itself. Illustrations become property of RSNA and will be stored for 2 years after publication.
They will be returned only upon written request from the corresponding author.

Return and Publication

Return corrected page proofs within 72 hours by uploading the proofs to The Author Center.
Your paper cannot be published until we have received your corrections. Do not impose any
security on your document.

Reprints
Please e-mail inquiries regarding reprints to reprints@rsna.org. Specify the number of copies
and color choice (black & white or color).

—RSNA Publications
 ORIGINAL RESEARCH
Generalized Tonic-Clonic
[AQ1] Seizures: Aberrant Interhemispheric

n NEURORADIOLOGY
Functional and Anatomical
[AQ2] Connectivity1
[AQ3] Gong-Jun Ji, PhD
Purpose: To characterize the interhemispheric functional and ana-
Zhiqiang Zhang, MD
tomic connectivity in patients with idiopathic generalized
Qiang Xu, MS
epilepsy and generalized tonic-clonic seizures (GTCS).
Yu-Feng Zang, MD
Wei Liao, PhD Materials and This retrospective study was approved by the local institu-
Guangming Lu, MD Methods: tional review board and was HIPAA compliant. All partic-
ipants provided written informed consent. Resting-state
functional and structural magnetic resonance images were
acquired in 52 patients with GTCS and 65 healthy control
subjects. The functional connectivity between bilateral ho-
motopic voxels was calculated. Homotopic regions showing
abnormal functional connectivity in patients were adopted
as regions of interest for an analysis of diffusion-tensor
imaging tractography. The fractional anisotropy and fiber
length were compared between groups. Two-sample t test
and nonparametric correlation analysis were used.

Results: Compared with control subjects, patients showed in-

creased interhemispheric functional connectivity between
the bilateral cuneus (P = .0008, corrected) and the ante-
rior cingulate cortex (P = .0003, corrected) and decreased
[AQ5] functional connectivity between the bilateral olfactory cortex
(P = .00005, corrected), inferior frontal gyrus (P = .00005,
corrected), supramarginal gyrus (P = .0002, corrected),
and temporal pole (P = .0003, corrected). Furthermore,
the fiber length of the commissural fiber bundles con-
necting the bilateral anterior cingulate cortex (t = 22.30,
P = .03, uncorrected) and the bilateral cuneus was shorter
in patients than in control subjects (t = 23.19, P = .002,
uncorrected).

Conclusion: Our findings show that the bilateral anterior cingulate

cortex may be critical to the pathophysiology of patients
with GTCS and suggest that the corresponding commis-
1
From the Center of Cognitive and Brain Disorders, sural fiber bundle in the genu of the corpus callosum is a
Hangzhou Normal University, 126 Wenzhou Road, Gongshu potential target for future surgical treatment in patients
District, Hangzhou, Zhejiang 310015, China (G.J.J., Y.F.Z., with intractable GTCS.
W.L.); and Department of Medical Imaging, Jinling Hospital,
Clinic School of Medical College, Nanjing University, q
RSNA, 2014
Nanjing, Jiangsu, China (Z.Z., Q.X., G.L.). Received July
13, 2013; revision requested September 10; revision
received November 1; accepted November 27; final version
accepted December 11. Z.Z., Y.F.Z., and W.L. supported by
the National Natural Science Foundation of China (grants
81271553, 81020108022, and 81201155). Y.F.Z. supported
by the Qian Jiang Distinguished Professor program. W.L.
supported by the China Postdoctoral Science Foundation
(grant 2013M532229). Address correspondence to W.L.
[AQ4] (e-mail: weiliao.wl@gmail.com).

q
RSNA, 2014

Radiology: Volume 000: Number 0—   2014 n radiology.rsna.org 1

NEURORADIOLOGY: Tonic-Clonic Seizures and Interhemispheric Connectivity
I
diopathic generalized epilepsy (IGE)
encompasses a group of seizure
types that are newly conceptualized
as originating in a specific region and
rapidly engaging bilaterally distributed
networks (1). Generalized tonic-clonic
seizure (GTCS), a common phenotype
of IGE, is characterized by generalized
spike-wave discharges (2.5–5 Hz) in-
volving the bilateral hemispheres during
seizures (2). Simultaneous electroen-
cephalography and functional magnetic
resonance (MR) imaging study results
have shown that the hemodynamic
changes associated with generalized
spike-wave discharge are bilateral and
fairly symmetric (3).
Abnormal bihemispheric synchro-
nization in patients with GTCS during
generalized spike-wave discharges is
implicated as a cause of enhanced in-
terhemispheric communication or ex-
citability (4). Results of animal studies
(5,6) in rodents and felines document-
ed that cutting the corpus callosum
could largely confine the spike-wave
discharges to a single hemisphere,
which strongly reduces bilateral syn-
chronization. Corpus callosotomy is an
optional treatment for human patients
with refractory generalized epilepsy.
Follow-up studies have demonstrated
that the frequency of generalized sei-
zures can be contained effectively and
decreased (7,8) and that unbalanced
interhemispheric functional connec-
tivity can return to normal (9). These
findings demonstrate that the initiation
of generalized seizures requires fluent
interhemispheric communication that
is modulated by the corpus callosum
Advances in Knowledge
n The bilateral anterior cingulate
cortices showed a prominent
increase in interhemispheric
functional connectivity (t . 1.98,
P , .05) in patients with idio-
pathic generalized epilepsy.
n Commissure tracts connecting
the bilateral cingulate cortices,
which pass through the genu of
the corpus callosum, were
shorter in patients (t = 22.30, P
= .03) than in control subjects.
2 radiology.rsna.org
and other commissural pathways (eg,
anterior commissure). Thus, character-
izing interhemispheric synchronization
in patients with GTCS is crucial to un-
derstanding the pathophysiology of this
disorder.
To date, authors of only a few
studies investigated interhemispheric
synchronization in epilepsy by using
resting-state functional MR imaging
(10). Bai et al (10) used resting-state
functional connectivity to analyze inter-
hemispheric synchronization between
16 pairs of predefined seizure-related
regions in patients with childhood ab-
sence epilepsy and found increased con-
nectivity in the bilateral orbitofrontal
cortex. However, little is known about
the interhemispheric functional syn-
chronization and associated anatomic
connectivity in patients with GTCS.
To investigate the functional con-
nectivity between hemispheres in a
voxel-wise manner, Zuo et al (11) de-
veloped a method called voxel-mirrored
homotopic connectivity (VMHC). This
method was subsequently used to show
the aberrant interhemispheric func-
tional connectivity in patients with vari-
ous brain disorders such as autism (12)
and cocaine addiction (13).
In this retrospective study, we ad-
opted VMHC to explore alterations of
interhemispheric functional connec-
tivity in patients with GTCS. Accord-
ing to clinical and basic research study
(7,8,14) results, the frontal lobe may be
critical in the generation of generalized
seizures. Thus, we predicted abnormal
functional connectivity between the bi-
lateral frontal areas. While testing this
hypothesis, we also investigated whether
the functional changes were associ-
ated with corresponding alterations of
Implications for Patient Care
n The bilateral anterior cingulate
cortices may be critical to under-
standing the pathophysiology of
generalized tonic-clonic seizures.
n The genu of the corpus callosum
is a potential target for surgical
treatment in patients with intrac-
table generalized tonic-clonic
seizures.
Ji et al
anatomic connectivity. In particular, in
this study we aim to characterize thor-
oughly the interhemispheric functional
and anatomic connectivity changes in
patients with IGE and GTCS.
Materials and Methods
This retrospective study was approved
by the local institutional review board
and conducted in compliance with the
Health Insurance Portability and Ac-
countability Act of 1996. All partici-
pants provided written informed con-
sent before functional MR imaging or
neurologic evaluations.
Participants
The patients in the current study were
consecutively enrolled from Jinling Hos-
pital, Nanjing, China (from December
2009 to January 2012). Patients were
diagnosed according to International
League against Epilepsy criteria (15).
Three experienced neurologists (F.Y.,
G.C., and Q.T. with 15, 35, and 50 years
of experience with clinical epilepsy ex-
aminations, respectively) screened the
patients and inspected their electro-
encephalographic data. The inclusion
criteria for patient recruitment were as
follows: (a) manifestation of typical clin-
ical symptoms of GTCS; (b) presence
of generalized polyspike-wave on the [AQ6]
Published online before print
10.1148/radiol.13131638 Content codes:
Radiology 2014; 000:1–9
Abbreviations:
ACC = anterior cingulate cortex
GTCS = generalized tonic-clonic seizures
IGE = idiopathic generalized epilepsy
ROI = region of interest
VMHC = voxel-mirrored homotopic connectivity
Author contributions:
Guarantors of integrity of entire study, G.J.J., Z.Z., W.L.,
G.L.; study concepts/study design or data acquisition
or data analysis/interpretation, all authors; manuscript
drafting or manuscript revision for important intellectual
content, all authors; approval of final version of submitted
manuscript, all authors; literature research, G.J.J., Z.Z.,
Q.X., W.L.; clinical studies, Z.Z., G.L.; experimental studies,
G.J.J., Z.Z., Y.F.Z., G.L.; statistical analysis, G.J.J., Y.F.Z.,
W.L.; and manuscript editing, G.J.J., Z.Z., Y.F.Z., W.L.
Conflicts of interest are listed at the end of this article.
n Radiology: Volume 000: Number 0—   2014
 NEURORADIOLOGY: Tonic-Clonic Seizures and Interhemispheric Connectivity
patient’s scalp and sphenoidal electro-
encephalographic (including video elec-
troencephalography); (c) no evidence
of a cause of secondary GTCS such as
trauma, tumor, or intracranial infec-
tion; (d) no focal abnormality at struc-
tural MR imaging; and (e) right hand-
edness. A total of 60 patients fulfilled
these inclusion criteria. The exclusion
criteria were as follows: (a) history of
addictions or neurologic diseases other
than epilepsy, (b) history of partial sei-
zures, (c) falling asleep during resting-
state functional MR imaging, and (d)
head translation or rotation parameters
exceeding 6 1 mm or 6 1 degree. Fi-
nally, 52 patients with GTCS (33 males
[AQ7] and 19 females; mean age, 24.27 years
6 7.04) were included in this study.
Eight patients were excluded, two for
falling asleep and six for head motion
exceeding 6 1 mm or 6 1 degree. A
subset of these patients (n = 20) had
participated in one of our earlier stud-
ies (15). Among the 52 included pa-
tients, 40 patients were treated with
antiepileptic drugs (monotherapy, 13;
polytherapy, 27; valproic acid, 21; car-
bamazepine, 13; topiramate, seven;
phenytoin, four; phenobarbitone, four;
traditional Chinese herbal medicine,
three; lamotrigine, two; oxcarbazepine,
two; clonazepam, one).
Sixty-eight healthy control subjects
were recruited initially from the staff
of the Jinling Hospital, Nanjing, China.
They had no history of neurologic dis-
orders or psychiatric illnesses and no
gross abnormalities on brain MR im-
ages. Three of these 68 control subjects
were excluded because they fell asleep
or moved their heads excessively dur-
ing MR imaging. Thus, 65 healthy con-
trol subjects (36 males and 29 females;
mean age, 25.49 years 6 7.15) were in-
cluded. No significant difference in age
(t = 20.33, P = .75) or sex (x2 = 1.25, P
= .26) was found between groups. De-
mographic and clinical information of
the 52 patients and 65 control subjects
is detailed in Table 1.
Data Acquisition
All experiments were performed with
a clinical 3-T whole-body MR imager
(TIM Trio; Siemens Medical Solutions,
Radiology: Volume 000: Number 0—   2014
Table 1
Characteristics of Patients with Generalized Tonic-Clonic Seizures and Control
Subjects
Characteristic GTCS (n = 52)
Sex (M/F) 33/19*
Age (y) 24.27 6 7.04
Handedness (R/L) 52/0*
Duration (y) 7.69 6 7.83
Frequency (times/y) 5.15 6 9.06
Onset age (y) 17.58 6 6.11
Note.—Unless otherwise indicated, data are means 6 standard deviation.
* Data are proportions of patients.
†
Two-sample two-tailed t test.
‡
x2 test
Erlangen, Germany) by using a stan-
dard birdcage head transmit and re-
ceive coil. Data were acquired during
the interictal periods without combined
electroencephalographic confirmation.
Foam padding was used to minimize
head motion for all subjects. Functional
images were acquired by using a single-
shot, gradient-recalled echo-planar im-
aging sequence (repetition time msec/
echo time msec, 2000/30; flip angle,
90 degrees). Images of 30 transverse
sections (field of view, 240 3 240 mm2;
in-plane matrix, 64 3 64; section thick-
ness, 4 mm; intersection gap, .4 mm;
and voxel size, 3.75 3 3.75 3 4 mm3)
aligned along the anterior–posterior
commissure line were acquired.
For each subject, a total of 250 vol-
umes were acquired, resulting in an im-
aging time of 500 seconds. The subjects
were instructed to rest with their eyes
closed, not think of anything in particu-
lar, and not fall asleep. The high-resolu-
tion three-dimensional T1-weighted an-
atomic images were acquired in sagittal
orientation by using a magnetization-
prepared rapid gradient-echo sequence
(2300/2.98; flip angle, 9 degrees; field
of view, 256 3 256 mm2; matrix size,
256 3 256; section thickness, 1 mm
without intersection gap; voxel size,
0.5 3 0.5 3 1 mm3; sections, 176).
In addition to resting-state functional
MR imaging and three-dimensional T1
imaging, we tried to acquire diffusion
data for each subject included in the
n radiology.rsna.org
Ji et al
Control Subjects (n = 65) t Value P Value
36/29* 0.78† 0.38‡
25.49 6 7.15 20.93† 0.36†
65/0* ... ...
... ... ...
... ... ...
... ... ...
study. However, we had limited access
to the MR imager because this imager
is used mainly for clinical examinations.
As a result, we collected diffusion data
in only 20 patients (14 females, all right
handed; age, 24.3 years 6 8.2) and 40
controls (22 females, all right-handed;
age, 23.9 years 6 5.1). No significant
difference in age (t = 20.33, P = .75)
or sex (x2 = 1.25, P = .26) were found
between groups.
The anatomic connectivity analysis
was based on the diffusion data of these
subjects. Diffusion-tensor imaging data
from the whole brain were obtained by
using spin-echo echo-planar imaging.
This included 30 volumes with diffusion
gradients applied along 30 noncollin-
ear directions (b = 1000 sec/mm2) and
one volume without diffusion weighting
(b = 0 sec/mm2). Each volume con-
sisted of 45 contiguous axial sections
(6100/93; number of signals acquired,
4; flip angle, 90 degrees; field of view,
240 3 240 mm2; matrix size, 256 3 256; [AQ8]
voxel size, 0.94 3 0.94 3 3 mm3).
Data Processing
Functional images.—Functional images
were preprocessed by two authors (G.J.J.
and W.L., with 4 and 7 years of experi-
ence in functional brain network analysis,
respectively) by using the Data Process-
ing Assistant for Resting-State Functional
MR Imaging toolkit (16), which synthe-
sizes procedures in the Resting State
Functional MR imaging toolkit (REST;
3
NEURORADIOLOGY: Tonic-Clonic Seizures and Interhemispheric Connectivity
http://www.restfmri.net) (17) and SPM8
(www.fil.ion.ucl.ac.uk/spm). The first 10
images were excluded to ensure steady-
state longitudinal magnetization, and the
remaining images were then corrected for
temporal differences and head motion.
After subject selection, neither transla-
tion nor rotation parameters in any given
data set exceeded 6 1 mm or 6 1 de-
gree. We then coregistered the individ-
ual T1 images to functional images. The
T1 images were segmented (gray matter,
white matter, and cerebrospinal fluid)
and normalized to the Montreal Neu-
rologic Institute space by using a 12-pa-
rameter nonlinear transformation. These
transformation parameters were applied
to the functional images. The normalized
images were resectioned at a resolution
of 3 3 3 3 3 mm3. To account for dif-
ferences in the geometric configuration
of the cerebral hemispheres, we further
transformed the preprocessed functional
images to a symmetric space. To achieve
this, we used the following procedure:
(a) The normalized gray matter images
were averaged for all subjects to create
a mean normalized gray matter image,
(b) this image was then averaged with its
left-right mirrored version to generate a
group-specific symmetrical template, and
(c) normalized gray-matter images were
registered to the symmetric template and
applied to the nonlinear transformation
to the normalized functional images. Fi-
nally, we spatially smoothed images with
a 6-mm full-width at half-maximum iso-
tropic Gaussian kernel.
VMHC analysis was carried out by
using the Resting State Functional MR
Imaging toolkit (REST; http://www.rest-
fmri.net) (17). Several preprocessing
steps were taken to remove the sources
of possible spurious variance from each
voxel’s time series: (a) removing linear
trends, (b) filtering the temporal band-
pass (0.01–0.08 Hz), and (c) regressing
out nine nuisance signals (global mean,
white matter, cerebrospinal fluid signals,
and six head-motion parameters). We
computed Pearson correlations between
the time series of every pair of sym-
metrical interhemispheric voxels. The
resulting correlations for each paired
voxel constituted a VMHC brain map
(Fisher z transformed) and were used
4 radiology.rsna.org
for subsequent group-level analyses.
We also calculated the global VMHC for
each subject and compared it between
groups. Global VMHC was calculated by
averaging VMHC values in all brain vox-
els within a mask (there was only one
correlation for each pair of homotopic
voxels). The mask was unilateral hemi-
spheric gray matter from the symmetric
template produced before.
Diffusion-tensor images.—Data were
preprocessed and analyzed by using the
Pipeline for Analyzing Brain Diffusion
Images toolkit (PANDA; http://www.
nitrc.org/projects/panda) (18), which
synthesizes procedures in FSL (http://fsl.
fmrib.ox.ac.uk/fsl) and the diffusion tool-
kit (19). For each subject, diffusion-ten-
sor images were geometrically corrected
by using an unweighted B0 image (b =
0 sec/mm2) and a filed map. Diffusion-
tensor images were coregistered to the
B0 image by using affine transformations
to minimize head movements. Diffusion-
tensor models were estimated by using
the linear least-squares fitting method
at each voxel by using the diffusion tool-
kit. Whole-brain fiber tracking was per-
formed in the native diffusion space for
each subject by using the fiber assignment
with the continuous tracking algorithm
(20) embedded in the diffusion toolkit.
Path tracking proceeded until either the
fractional anisotropy was less than 0.15
or the angle between the current and the
previous path segment was greater than
35 degrees, as in our previous studies
(21). Fibers less than 10 mm or with ob-
vious false paths were discarded.
Several regions showed abnormal in-
terhemispheric functional connectivity in
patients. These regions were selected as
regions of interest (ROIs) for an analysis
of diffusion-tensor imaging data. Before
the fibers connecting bilateral ROIs were
tracked, several procedures were per-
formed: (a) We transformed these ROIs
from the normalized symmetric space to
each individual’s native functional space,
(b) we coregistered the mean functional
image (native functional space) to the
B0 image (native diffusion space) and
applied the transformation to all ROIs,
and (c) we dilated the ROIs by one voxel
into the white matter to ensure that they
were in contact with the fibers. Fiber
Ji et al
bundles connecting symmetrical ROIs in
the two hemispheres were then extracted
from the total set of fibers. This was ac-
complished by using software (TrackVis;
www.trackvis.org) (22). To determine
through which part of the corpus callo-
sum the abnormal commissure fibers
pass, we normalized the tracts of all sub-
jects from individual spaces to the Mon-
treal Neurologic Institute space. Specifi-
cally, (a) we coregistered the T1 image
to the B0 image, (b) we normalized the
coregistered T1 image to the Montreal
Neurologic Institute space and obtained
the transformation matrices, and (c) we
extracted the commissure fibers connect-
ing bilateral ROIs in individual space and
warped them to the Montreal Neurologic
Institute space by using the transforma-
tion matrices defined above. Finally, a
population-based probabilistic map of
commissure tracts was produced.
Statistical Analysis
The statistical significance of VMHC
within groups was analyzed by using a
one-sample t test (P , .05, corrected
with a single voxel height of P , .05 and
cluster volume . 3213 mm3 by using the
a software program; AFNI AlphaSim;
http://afni.nimh.nih.gov /pub/dist/doc/
manual/AlphaSim.pdf).The significant
differences in VMHC between groups
were analyzed by using the two-sample
t test (P , .05, corrected with a single
voxel height of P , .01 and a cluster vol-
ume . 918 mm3 by using the AlphaSim
program). Fiber length and fractional an-
isotropy of the commissural tracts con-
necting the bilateral ROIs were compared
between groups by using a two-sample t
test, and the multiple comparisons prob-
lem was corrected by means of Bonfer-
roni correction. Pearson correlation
analysis was performed between the im-
aging data and clinical measurements in
the patients. To account for the potential
influence of outliers, we also used the
Shepherd p correlation analysis (23).
Results
VMHC
Within-group results indicated that
both patients (Fig 1A) and control
n Radiology: Volume 000: Number 0—   2014
 NEURORADIOLOGY: Tonic-Clonic Seizures and Interhemispheric Connectivity Ji et al

Figure 1

[AQ11] Figure 1: MR images show interhemispheric functional connectivity within and between groups. Regions show significant interhemi- [AQ12]
spheric functional connectivity in patients, A, and control subjects, B, respectively (P , .05, AlphaSim corrected). C, Homotopic regions
show increased (hot color) or decreased (cool color) functional connectivity in patient group (P , .05, AlphaSim corrected).

subjects (Fig 1B) had robust homotopic
functional connectivity with regional dif-
ferences in strength. Although patients
and control subjects did not differ on
global VMHC (patients, 0.55 6 0.08;
control subjects, 0.54 6 0.07; t = 0.53,
P = .60), some brain regions showed
abnormal VMHC in patients with IGE
(P , .05, AlphaSim corrected) (Fig 1C;
Table 2). Specifically, interhemispheric
functional connectivity for the bilateral
cuneus and bilateral anterior cingulate
cortex (ACC) was higher in patients
than in control subjects. Conversely,
interhemispheric connectivities of the
bilateral olfactory cortex, bilateral infe-
rior frontal gyrus, bilateral supramar-
ginal gyrus, and bilateral temporal pole
were weaker in the patients. We did not
find a significant correlation between
the interhemispheric functional connec-
tivity in these regions and any clinical
measure (including seizure frequency,
disease duration, and onset age).
Diffusion-Tensor Imaging
Three commissural tracts that connect-
ed the bilateral ACC, bilateral cuneus,
and bilateral olfactory cortex were de-
tected in all subjects. The tracts that
connected the other three bilateral ROIs
only could be found in a few subjects
(n = 8) by using our diffusion data. For

Radiology: Volume 000: Number 0—   2014 n radiology.rsna.org 5

 NEURORADIOLOGY: Tonic-Clonic Seizures and Interhemispheric Connectivity
Table 2
Regions Showing Aberrant Interhemispheric Functional Connectivity between Two
Groups
Brodman
Brain Regions Area*
GTCS . control subjects
Anterior cingulate cortex 32
Visual cortex 18
GTCS , control subjects
Inferior frontal cortex 44
Olfactory cortex 34
Supramarginal cortex 22
Temporal pole 38
[AQ14] * Data are number of patients.
this reason, we performed a between-
group analysis for only the three com-
missural tracts connecting the bilateral
ACC, cuneus, and olfactory cortex.
Between-group analysis indicated
that the commissural fibers connect-
ing the bilateral ACC (t = 22.30, P =
.03) and bilateral cuneus (t = 23.19,
P = .002) were shorter and the frac-
tional anisotropy values of the con-
necting bilateral olfactory cortex (t =
22.03, P = .047) were smaller in the
patients (Fig 2). However, only the
length of the fibers connecting the bi-
lateral cuneus still showed group dif-
ferences after Bonferroni correction.
To clearly show through which part
of the corpus callosum these fibers
went, we produced a population-based
(n = 60) probabilistic map. We found
that the fibers connecting the bilateral
olfactory cortex, ACC, and cuneus
were located in the rostra, genu, and
splenium of the corpus callosum, re-
spectively. We correlated the parame-
ters of these commissural fibers with
clinical measurements. By using the
Pearson correlation, we found that the
fractional anisotropic value of fibers
connecting the bilateral ACC were
negatively correlated with disease du-
ration (r = 0.53, P = .02) and seizure
frequency (r = 0.53, P = .02), whereas
no significant results were found by
using the Shepherd correlation (all,
P . .05) (Fig 3). The correlation with
6 radiology.rsna.org
Ji et al
interhemispheric communication. In our
study, we observed increased functional
connectivity between the bilateral ACC
in patients during the interictal state.
Montreal Neurologic This may be driven by a subthreshold
Institute Coordinates excitatory state before or after seizures
(10). Furthermore, this finding is con-
X Y Z Peak t Value Cluster Size (mm3)
sistent with the suggestion that mesial
frontal regions (26,27) are critical for
6 36 27 3.78 1107 seizure generation.
15 2102 12 3.46 1377 Results of a recent study (28) indi-
cated that the abnormal global synchro-
42 9 27 24.24 1053 nization originates in discrete areas of
18 6 215 24.24 999 the frontal lobe. Here we also found
60 251 21 23.88 2430 a change in anatomic connectivity be-
48 18 236 23.76 1107
tween the bilateral ACC, but this did
survive multiple comparison correction.
It is widely assumed that structural con-
nectivity is relatively stable over time,
whereas the functional connectivity is
seizure onset age was positive and ap- more variable (21). Therefore, we rea-
proached significance (r = 0.41, P = soned that anatomic connectivity may
.07). be less affected in patients with GTCS.
The presence of both anatomic and
functional changes suggested that bilat-
Discussion eral ACC is important in the pathophys-
We investigated interhemispheric func- iology of IGE. The enhanced connectiv-
tional and anatomic connectivity in pa- ity between bilateral ACC may provide
tients with GTCS by using resting-state new targets for the surgical treatment
functional MR imaging and diffusion of intractable IGE. Results of follow-up
tractographic techniques, respectively. studies (7,8) demonstrated that ante-
Patients showed decreased functional rior callosotomy effectively reduces sei-
connectivity between the bilateral olfac- zure frequency. To minimize cognitive
tory cortex, inferior frontal gyrus, su- impairment (such as cerebral discon-
pramarginal gyrus, and temporal pole as nection syndrome) after surgery (29),
well as increased functional connectivity reducing the extent of the resection
between the bilateral ACC and cuneus. may be necessary in the future. Our
Anatomic connectivity changes were in results suggest that commissural fibers
accordance with these alterations in connecting the bilateral ACC may be a
functional connectivity. The presence target for selective corpus callosotomy.
of specific hypo- and hyperconnectivity Interhemispheric connectivity
changes indicated that GTCS does not changes were found in other brain re-
homogeneously affect the entire brain. gions. For instance, we observed in-
The increased connectivity between the creased interhemispheric functional
bilateral ACC suggests that this brain synchronization and anatomic connec- [AQ9]
region is important in the pathophysi- tivity efficiency in the visual cortex. This
ology of GTCS. may be correlated with the visual aura
Resting-state functional MR imaging experienced by some patients with IGE
enabled us to gain further insight into (30). Previous functional MR imaging
the functional organization of the brain (31) and neuropsychologic (32) studies
in patients with GTCS. Functional ab- showed attention deficits in patients
normalities in patients with GTCS have with GTCS. We observed decreased
been reported with regard to local ac- homotopic connectivity between the
tivity (24), network integrity (25), and bilateral attention-related regions (ie,
whole-brain architecture (21). How- the inferior frontal gyrus and the su-
ever, little is known about changes in pramarginal gyrus). We also found
n Radiology: Volume 000: Number 0—   2014
NEURORADIOLOGY: Tonic-Clonic Seizures and Interhemispheric Connectivity Ji et al

Figure 2

Figure 2: Diffusion tractographic images (left) from a single control subject show between-group comparison for commissure fiber [AQ13]
parameters. Fibers connecting, A, bilateral olfactory cortex, B, anterior cingulate cortex, and, C, cuneus were located in the rostra,
genu, and splenium of the corpus callosum, respectively. Inset shows probabilistic maps of each commissural tract constructed from
60 subjects. Scatterplots show the between-group difference of these three tracts. NC = control subjects; PA = patients; ∗ 5 P , .05,
uncorrected; and ∗∗ = P , .05, Bonferroni corrected.

Figure 3

Figure 3: Scatterplots show correlation between fractional anisotropy of the fibers connecting the bilateral anterior cingulate cortex and
clinical characteristics. Solid line and dashed lines represent the best-fit line of Pearson and Shepherd correlation, respectively.  = outliers.

decreased connectivity between the bi- are part of a social processing pathway that characterize patients with GTCS
lateral temporal pole and the bilateral (33), this reduced connectivity may be (34). These findings provided valuable
olfactory cortex. Because these regions related to the impaired social abilities information that can support our better

Radiology: Volume 000: Number 0—   2014 n radiology.rsna.org 7

NEURORADIOLOGY: Tonic-Clonic Seizures and Interhemispheric Connectivity
understanding of cognitive impairment
in patients with GTCS.
Finally, the limitations of our investi-
gations are worth mentioning. First, we
conducted an interictal-state study that
allowed us to investigate the effect of
GTCS on brain function. It is challenging
to acquire functional MR imaging data of
acceptable quality during the ictal state
due to body movements. In the future,
high-density electroencephalographic
recordings might be used to investigate
the link between interictal discharge and
homotopic connectivity. Second, our
findings were confounded by the use of
antiepileptic drugs, which can produce
cognitive impairment (35). Further stud-
ies must be conducted to clarify the ef-
fect of antiepileptic drugs on anatomic
and functional organization of the human
brain. Third, we did not investigate intra-
hemispheric communication. Because in-
ter- and intrahemispheric communication
are correlated, future investigations are
warranted to examine both in patients
with GTCS, which might further advance
our understanding of the pathophysiology
of this disease (36).
In our study, we found robust en-
hanced functional synchronization and
anatomic connectivity efficiency be-
tween the bilateral ACC, suggesting that
the bilateral ACC may be important in
the pathophysiology of GTCS, and the
commissural fiber could be a potential
target for future surgical treatment in
patients with intractable GTCS.
Acknowledgments: We thank Dr. Guanghui
Chen (G.C.), Dr. Fang Yang (F.Y.), and Dr.
Qifu Tan (Q.T.) for their help with revising the
manuscript.
Disclosures of Conflicts of Interest: G.J.J. No
relevant conflicts of interest to disclose. Z.Z. No
relevant conflicts of interest to disclose. Q.X.
No relevant conflicts of interest to disclose.
Y.F.Z. No relevant conflicts of interest to dis-
close. W.L. No relevant conflicts of interest to
disclose. G.L. No relevant conflicts of interest
to disclose.
References
1. Berg AT, Scheffer IE. New concepts in classi-
fication of the epilepsies: entering the 21st
century. Epilepsia 2011;52(6):1058–1062.
2. Engel J Jr; International League Against Epi-
lepsy (ILAE). A proposed diagnostic scheme
8 radiology.rsna.org
for people with epileptic seizures and with
epilepsy: report of the ILAE Task Force on
Classification and Terminology. Epilepsia
2001;42(6):796–803.
3. Bai X, Vestal M, Berman R, et al. Dynamic
time course of typical childhood absence
seizures: EEG, behavior, and functional
magnetic resonance imaging. J Neurosci
2010;30(17):5884–5893.
4. Bloom JS, Hynd GW. The role of the corpus
callosum in interhemispheric transfer of in-
formation: excitation or inhibition? Neurop-
sychol Rev 2005;15(2):59–71.
5. Musgrave J, Gloor P. The role of the corpus
callosum in bilateral interhemispheric syn-
chrony of spike and wave discharge in fe-
line generalized penicillin epilepsy. Epilepsia
1980;21(4):369–378.
6. Vergnes M, Marescaux C, Lannes B, Depau-
lis A, Micheletti G, Warter JM. Interhemi-
spheric desynchronization of spontaneous
spike-wave discharges by corpus callosum
transection in rats with petit mal-like epi-
lepsy. Epilepsy Res 1989;4(1):8–13.
7. Jenssen S, Sperling MR, Tracy JI, et al.
Corpus callosotomy in refractory idi-
opathic generalized epilepsy. Seizure
2006;15(8):621–629.
8. Tanriverdi T, Olivier A, Poulin N, Ander-
mann F, Dubeau F. Long-term seizure out-
come after corpus callosotomy: a retrospec-
tive analysis of 95 patients. J Neurosurg
2009;110(2):332–342.
9. Pizoli CE, Shah MN, Snyder AZ, et al. Rest-
ing-state activity in development and main-
tenance of normal brain function. Proc
Natl Acad Sci U S A 2011;108(28):11638–
11643.
10. Bai X, Guo J, Killory B, et al. Resting func-
tional connectivity between the hemispheres
in childhood absence epilepsy. Neurology
2011;76(23):1960–1967.
11. Zuo XN, Kelly C, Di Martino A, et al. Grow-
ing together and growing apart: regional
and sex differences in the lifespan develop-
mental trajectories of functional homotopy.
J Neurosci 2010;30(45):15034–15043.
12. Anderson JS, Druzgal TJ, Froehlich A,
et al. Decreased interhemispheric func-
tional connectivity in autism. Cereb Cortex
2011;21(5):1134–1146.
13. Kelly C, Zuo XN, Gotimer K, et al. Reduced
interhemispheric resting state functional
connectivity in cocaine addiction. Biol Psy-
chiatry 2011;69(7):684–692.
14. Holmes MD, Brown M, Tucker DM. Are
“generalized” seizures truly generalized?
Evidence of localized mesial frontal and
Ji et al
frontopolar discharges in absence. Epilepsia
2004;45(12):1568–1579.
15. Proposal for revised classification of epi-
lepsies and epileptic syndromes. Commis-
sion on Classification and Terminology of
the International League Against Epilepsy.
Epilepsia 1989;30(4):389–399.
16. Chao-Gan Y, Yu-Feng Z. DPARSF: A MAT-
LAB toolbox for “pipeline” data analysis
of resting-state fMRI. Front Syst Neurosci
2010;4:13.
17. Song XW, Dong ZY, Long XY, et al. REST: a
toolkit for resting-state functional magnetic
resonance imaging data processing. PLoS
ONE 2011;6(9):e25031.
18. Cui Z, Zhong S, Xu P, He Y, Gong G.
PANDA: a pipeline toolbox for analyzing
brain diffusion images. Front Hum Neurosci
2013;7:42.
19. Wang R, Beener T, Sorensen AG, Weeden
VJ. Diffusion toolkit: a soft-ware package
for diffusion imaging data processing and
tractography [abstr]. In: Proceedings of
the Fifteenth Meeting of the International
Society for Magnetic Resonance in Medi-
cine. Berkeley, Calif: International Society
for Magnetic Resonance in Medicine; May
19–25, 2007; Berlin, Germany; 3720.
20. Chao YP, Cho KH, Yeh CH, Chou KH, Chen
JH, Lin CP. Probabilistic topography of hu-
man corpus callosum using cytoarchitectur-
al parcellation and high angular resolution
diffusion imaging tractography. Hum Brain
Mapp 2009;30(10):3172–3187.
21. Zhang Z, Liao W, Chen H, et al. Altered
functional-structural coupling of large-scale
brain networks in idiopathic generalized ep-
ilepsy. Brain 2011;134(Pt 10):2912–2928.
22. Liao W, Zhang Z, Pan Z, et al. Default mode
network abnormalities in mesial temporal
lobe epilepsy: a study combining fMRI and
DTI. Hum Brain Mapp 2011;32(6):883–895.
23. Schwarzkopf DS, De Haas B, Rees G. Better
ways to improve standards in brain-behavior
correlation analysis. Front Hum Neurosci
2012;6:200.
24. Zhong Y, Lu G, Zhang Z, Jiao Q, Li K, Liu
Y. Altered regional synchronization in epi-
leptic patients with generalized tonic-clonic
seizures. Epilepsy Res 2011;97(1-2):83–91.
25. Wang Z, Zhang Z, Jiao Q, et al. Impairments
of thalamic nuclei in idiopathic generalized
epilepsy revealed by a study combining mor-
phological and functional connectivity MRI.
PLoS ONE 2012;7(7):e39701.
26. Beghi M, Beghi E, Cornaggia CM, Gobbi G.
Idiopathic generalized epilepsies of adoles-
cence. Epilepsia 2006;47(Suppl 2):107–110.
n Radiology: Volume 000: Number 0—   2014
NEURORADIOLOGY: Tonic-Clonic Seizures and Interhemispheric Connectivity
27. Anderson J, Hamandi K. Understanding
juvenile myoclonic epilepsy: Contribu-
tions from neuroimaging. Epilepsy Res
2011;94(3):127–137.
28. Yan B, Li P. The emergence of abnormal hy-
persynchronization in the anatomical struc-
tural network of human brain. Neuroimage
2013;65:34–51.
29. Andersen B, Rogvi-Hansen B, Kruse-Larsen C,
Dam M. Corpus callosotomy: seizure and psy-
chosocial outcome. A 39-month follow-up of
20 patients. Epilepsy Res 1996;23(1):77–85.
30. Gungor-Tuncer O, Baykan B, Altindag E,
Bebek N, Gurses C, Gokyigit A. Prevalence
and characteristics of visual aura in idio-
Radiology: Volume 000: Number 0—   2014 n radiology.rsna.org
pathic generalized epilepsy. Epilepsy Behav
2012;25(4):573–576.
31. Wang Z, Lu G, Zhang Z, et al. Altered rest-
ing state networks in epileptic patients with
generalized tonic-clonic seizures. Brain Res
2011;1374:134–141.
32. Henkin Y, Sadeh M, Kivity S, Shabtai
E, Kishon-Rabin L, Gadoth N. Cognitive
function in idiopathic generalized epi-
lepsy of childhood. Dev Med Child Neurol
2005;47(2):126–132.
33. Olson IR, Plotzker A, Ezzyat Y. The Enig-
matic temporal pole: a review of findings
on social and emotional processing. Brain
2007;130(Pt 7):1718–1731.
Ji et al
34. Camfield P, Camfield C. Idiopathic general-
ized epilepsy with generalized tonic-clonic
seizures (IGE-GTC): a population-based
cohort with .20 year follow up for medi-
cal and social outcome. Epilepsy Behav
2010;18(1-2):61–63.
35. Ijff DM, Aldenkamp AP. Cognitive side-
effects of antiepileptic drugs in children.
In: Vinken PJ, Bruyn GW, eds. Handbook
of clinical neurology. Vol 111. LOCATION:
PUBLISHER, 2013; 707–718. [AQ10]
36. Gee DG, Biswal BB, Kelly C, et al. Low
frequency fluctuations reveal integrated
and segregated processing among the cere-
bral hemispheres. Neuroimage 2011;54(1):
517–527.
9
NEURORADIOLOGY: Tonic-Clonic Seizures and Interhemispheric Connectivity Ji et al

Author: Read proofs carefully. This is your ONLY opportunity to make changes. NO fur-
ther alterations will be allowed after this point.

Author Queries

[AQ1]: Your manuscript was edited by Jean Winkler, manuscript editor at the
RSNA. E-mail: jwinkler@rsna.org.
[AQ2]: Title has been edited per RSNA style to mention the condition or body part
examined first.
[AQ3]: Please review all author names and academic degrees for accuracy.
[AQ4]: May we print your e-mail address for correspondence?
[AQ5]: In the comparison that showed decreased functional connectivity, it is not
clear which of the four brain areas are being compared. Were the compar-
isons made among all the areas or was one area compared with the other
three?
[AQ6]: “polyspike-wave” is used as an adjective here, but there is no noun.
“Polyspikes and waves” or “polyspike-wave discharges” or “polyspike-wave
patterns?
[AQ7]: Were any of the patients or control subjects younger than 18 years old?
[AQ8]: High spatial resolution or high temporal resolution. Please provide a more
specific term.
[AQ9]: It is not clear what the connectivity is between. Do you mean “between the
parts of the bilateral ACC”? or “the sides of the bilateral ACC?” or just “in
the ACC”
[AQ10]: Please provide the publisher and the location of the publisher for reference
35.
[AQ11]: RSNA style requires a description of the type of image at the beginning of a
figure legend. Please add the plane of imaging.
[AQ12]: RSNA style reserves the use of “significant” for the statistical sense. If this is
not what you meant, please change to “substantial” or something similar.
[AQ13]: Figure legend edited per RSNA style. Please review for accuracy.
[AQ14]: Is the footnote about the Brodman area correct?

Radiology: Volume 000: Number 0—   2014 n radiology.rsna.org 1

View publication stats