Accepted Manuscript
Title: The presence of short and sharp MEG spikes implies
focal cortical dysplasia
Author: Yuki Ueda Kiyoshi Egawa Tomoshiro Ito Fumiya
Takeuchi Midori Nakajima Kosuke Otsuka Naoko Asahina
Kayoko Takahashi Shingo Nakane Shinobu Kohsaka Hideaki
Shiraishi
PII:
DOI:
Reference:

S0920-1211(15)00100-X
http://dx.doi.org/doi:10.1016/j.eplepsyres.2015.04.020
EPIRES 5373

To appear in:

Epilepsy Research

Received date:
Revised date:
Accepted date:

14-9-2014
10-3-2015
24-4-2015

Please cite this article as: <doi>http://dx.doi.org/10.1016/j.eplepsyres.2015.04.020</doi>
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.

The presence of short and sharp MEG spikes implies focal cortical

ip
t

dysplasia

cr

Yuki Uedaa, Kiyoshi Egawaa, Tomoshiro Itoa, Fumiya Takeuchib, Midori Nakajimaa,

us

Kosuke Otsukaa, Naoko Asahinaa, Kayoko Takahashic, Shingo Nakanec, Shinobu
Kohsakaa, Hideaki Shiraishia

Department of Pediatrics, Hokkaido University Graduate School of Medicine,

Department of Health Science, Hokkaido University Graduate School of Medicine,

d

b

M

North 15 West 7, Kita-ku, Sapporo, Japan

an

a

c

Ac
ce
pt
e

North 12 West 5, Kita-ku, Sapporo, Japan

Division of Magnetoencephalography, Hokkaido University Hospital, North 14

West 5, Kita-ku, Sapporo, Japan

Yuki Ueda, Department of Pediatrics, Hokkaido University Graduate School of
Medicine, yuki_ueda@med.hokudai.ac.jp
Kiyoshi Egawa, Department of Pediatrics, Hokkaido University Graduate School of
Medicine, egakiyo@huhp.hokudai.ac.jp
Tomoshiro Ito, Department of Pediatrics, Hokkaido University Graduate School of
Medicine, itou_tomoshiro@kitami.jrc.or.jp;

1
Page 1 of 29

Fumiya Takeuchi, Department of Health Science, Hokkaido University Graduate
School of Medicine, take@hs.hokudai.ac.jp;

ip
t

Midori Nakajima, Department of Pediatrics, Hokkaido University Graduate School of
Medicine, midori.nakajima@yahoo.com;

cr

Kosuke Otsuka, Department of Pediatrics, Hokkaido University School of Medicine,

us

kounet@med.hokudai.ac.jp;

an

Naoko Asahina, Department of Pediatrics, Hokkaido University School of Medicine,
asahi-na@med.hokudai.ac.jp;

d

Hospital, tk5623@med.hokudai.ac.jp;

M

Kayoko Takahashi, Division of Magnetoencephalography, Hokkaido University

Ac
ce
pt
e

Shingo Nakane, Division of Magnetoencephalography, Hokkaido University Hospital,
sinakane@med.hokudai.ac.jp;

Shinobu Kohsaka, Department of Pediatrics, Hokkaido University School of Medicine,
shinobu-masako@ac.auone-net.jp,

Hideaki Shiraishi, Department of Pediatrics, Hokkaido University School of Medicine,
siraisi@med.hokudai.ac.jp.
Fumiya Takeuchi is now at Center for Advanced Research and Education,
Asahikawa Medical University, Midorigaoka-higashi 2-1-1-1, Asahikawa, Japan.

2
Page 2 of 29

Corresponding Author:

ip
t

Yuki Ueda
Address: Department of Pediatrics, Hokkaido University Graduate School of

Ac
ce
pt
e

d

M

an

E-mail address: yuki_ueda@med.hokudai.ac.jp

us

Tel: +81-11-706-5954, Fax: +81-11-706-7898

cr

Medicine, North 15 West 7, Kita-ku, Sapporo, Japan.

3
Page 3 of 29

Abstract
Purpose: This study focused on the characteristic needle-like epileptic spikes of short
duration and steep shape seen on magnetoencephalography (MEG) in patients

ip
t

diagnosed with focal cortical dysplasia (FCD) morphologically. We aimed to validate
the analysis of MEG spike morphology as a noninvasive method of identifying the

cr

presence and location of FCD.

us

Methods: MEG was collected by 204-channel helmet-shaped gradiometers. We
analyzed MEG spike sources for 282 patients with symptomatic localization-related

an

epilepsy. MEG showed clustered equivalent current dipoles when superimposed on
their three-dimensional-magnetic resonance images (MRI) in 85 patients.

M

Fifty-seven patients were excluded from our study, because they had destructive brain
lesions or an insufficient number of spikes for statistical analysis. Twenty-eight

d

patients (18 males, 10 females; aged 1-34 years) were finally matched to our inclusion

Ac
ce
pt
e

criteria, and were categorized into three groups: FCD (7 patients), non-FCD (10
patients), and non-lesion (11 patients), based on the MRI findings. We measured the
duration, amplitude, and tilt manually for at least 15 spikes per patient, and compared
the three groups using a one-way analysis of variance, followed by the Tukey test
when statistically significant (p < 0.05). In 17 patients with visible MRI lesions, we
investigated the correlation between the depth of the lesion and the tilt using the
Pearson product moment correlation.

Results: The average spike duration was significantly shorter in the FCD and
non-lesion groups than in the non-FCD group (p < 0.05). The average amplitude was
not significantly different between the three groups. The average spike tilt was
significantly steeper in the FCD group than in the non-FCD group (p = 0.0058). There

4
Page 4 of 29

was no significant difference between FCD and non-lesion patients in both duration
and tilt. Our additional study revealed a significant negative correlation between the
depth of the lesion and the average tilt (p = 0.0009).
Significance: MEG epileptiform discharges of short duration and steep tilt

ip
t

characterize FCD, especially when located at the superficial neocortical gyrus. We

cr

speculate that this particular spike morphology results from the intrinsic

epileptogenicity of FCD. Morphological analysis of MEG spikes can evaluate the

us

etiology of epileptogenic lesions and detect a strong, localized epileptogenic focus

M

an

such as that typically observed in FCD.

Keywords: focal cortical dysplasia, epilepsy, magnetoencephalography, spike

Ac
ce
pt
e

d

morphology, waveform analysis

Abbreviations: ECD, equivalent current dipole; ECoG, electrocorticography; FCD,
focal cortical dysplasia; MEG, magnetoencephalography; MRI, magnetic resonance
imaging; MTLE, medial temporal lobe epilepsy; EEG, electroencephalography; SEM,
standard error of the mean.

5
Page 5 of 29

1. Introduction
Focal cortical dysplasia (FCD) occurs during fetal neuronal migration, and is
characterized by abnormal cortical organization with or without cellular abnormalities.
FCD may also progress pathologically to become intractable focal epilepsy (Taylor et

ip
t

al., 1971). According to the classification by Palmini et al. (2004), FCD type IA is

cr

pathologically characterized by the presence of neocortical dyslamination, and FCD
type IB is typified by the presence of abnormal hypertrophic pyramidal-like cells.

us

FCD type II is distinguished from type I by the presence of dysmorphic neurons in the
gray and/or white matter (FCD type IIA), while FCD type IIB has the additional

an

presence of balloons cells.

M

Recently, surgical resection strategies have provided favorable outcomes in
patients with intractable epilepsy and FCD (Alexandre et al., 2006; Cohen-Gadol et

d

al., 2004; Hong et al., 2000; Kloss et al., 2002; Krai et al., 2003). Although progressive

Ac
ce
pt
e

resolution of magnetic resonance imaging (MRI) contributes to the pre-surgical
diagnosis of FCD, MRI sometimes misses the presence of FCD, particularly in those
patients with type I or IIA histology. Magnetoencephalography (MEG) is a relatively
modern diagnostic tool used to identify the location of epileptogenic foci in patients
with epilepsy, while equivalent current dipole (ECD) source analysis is a conventional
procedure used to define the localized epileptogenic zone in patients with intractable
epilepsy (Iida et al., 2005; Otsubo et al., 2001; Stefan et al., 2003). MEG ECD
localizations are well correlated to the MRI findings of FCD (Bast et al., 2004;
Ishibashi et al., 2002; Morioka et al., 1999), as FCD has an intrinsic epileptogenic
potential (Palmini et al., 1995). However MEG spike sources sometimes show an
aberrant epileptogenic area with localized epileptic currents in patients without

6
Page 6 of 29

definite MRI findings. These findings might imply the presence of MRI-invisible
epileptogenic lesions, such as FCD type I or IIA. The huge advantage of MEG
analysis is that it is a noninvasive neuroimaging tool with high spatial and temporal
resolution. So MEG is especially powerful for advanced clinical investigation in

ip
t

epileptic children.

cr

This study analyzed the shape and characteristic morphology of MEG spikes in
epileptic patients with FCD to define the etiology of the epilepsy and diagnose FCD

Ac
ce
pt
e

d

M

an

us

directly from the MEG spike morphology.

7
Page 7 of 29

2. Methods
MEG was performed in 427 patients at Hokkaido University Hospital from May
1996 to March 2012. MEG analyses were conducted for the diagnosis of epileptic

patients with symptomatic localization-related epilepsy.

ip
t

syndrome as well as for pre-surgical evaluation. We reviewed MEG data of 282

cr

MEG was detected by 204-channel helmet-shaped gradiometers (Vector View

System, Elekta AB., Stockholm, Sweden) and MEG spike sources were analyzed by

us

dipole-fit software (xfit, Neuromag Oy, Helsinki, Finland). We recorded the MEG data
at a sampling rate of 600 Hz, and then the data were band-pass filtered between 1 and

an

30 Hz to visualize and select the epileptic MEG spikes. Spike sources were

M

demonstrated as ECDs and superimposed onto the subjective three-dimensional-MRI.
We used the goodness of fit value to estimate how well the measured MEG currents

d

correlated to the estimation from a pre-defined ideal brain model. Significant ECDs

Ac
ce
pt
e

were represented when they showed > 70% goodness of fit and between 100 and 700
nAm dipole moments.

The distribution of ECDs was classified as clustered if more than 70% of the

ECDs were located in neighboring Brodmann’s areas (Ito et al., 2014). Clustered
ECDs were found in 85 patients. From this data set, we excluded 13 patients having
destructive brain lesions. Forty-four patients with insufficient epileptic spikes for
analysis in MEG recordings (< 15) were also excluded.
Twenty-eight patients (18 males, 10 females; aged 1-34 years, mean age 17.3
years) were finally matched to our inclusion criteria (Fig.1), and were categorized into
three groups: FCD group (n = 7, aged 1-34 years, mean age 14.4 years); non-FCD
group (n = 10, aged 13-34 years, mean age 18.9 years); and non-lesion group (n = 11,

8
Page 8 of 29

aged 8-31 years, mean age 17.5 years), according to the MRI findings (Table 1). In the
FCD group, two patients were diagnosed pathologically with FCD type IIA or type IA.
In the non-FCD group, seven of ten patients were diagnosed with medial temporal
lobe epilepsy (MTLE) with hippocampal sclerosis, and the other three patients had

ip
t

cavernous hemangioma, glioma, and amygdala enlargement, respectively. Ten patients

cr

diagnosed as FCD by MRI findings did not match our inclusion criteria, because MEG
did not show clustered dipoles, and most (7 out of 10) were less than 5 years (aged

us

1-13 years, mean age 5.1 years).

Representative clustered MEG spikes were selected to analyze the spike

an

morphology. We used the spike duration, amplitude, and the spike tilt as

M

morphological parameters. The duration was defined as the period between the onset
and the end of each trough of the spike source around the peaks. As the parameter

d

reflecting the sharpness of the spike, tilt was defined as the amplitude between the

Ac
ce
pt
e

spike onset and the peak, divided by the period between the onset and the peak. We
measured the duration, amplitude, and tilt manually for at least 15 spikes for each
patient, and calculated the average and standard error of the mean for each parameter.
These averages were compared statistically among the three groups using a one-way
analysis of variance, followed by the Tukey-Kramer honestly significant difference
test when statistically significant (p < 0.05). The cut-off value for each parameter for
differentiating FCD from non-FCD was also evaluated by a receiver operating
characteristic (ROC) curve, when statistically significant.
Furthermore we analyzed the depth of the lesions in order to measure its
influence on spike morphology. In the 17 patients from the FCD and non-FCD groups
with visible MRI lesions, we measured the depth of the lesion from the surface of the

9
Page 9 of 29

scalp with OsiriX (OsiriX Foundation, CA, USA), a type of Digital Imaging
Communications in Medicine viewer software. Depth was measured from three
directions (axial, coronal, and sagittal), and the shortest one was taken. In addition, we

Ac
ce
pt
e

d

M

an

us

cr

MEG spikes using the Pearson product moment correlation.

ip
t

investigated the correlation between the depth of the lesion and the average tilt of the

10
Page 10 of 29

3. Results
3.1. Morphological analysis
The average spike duration was 66.6 ± 3.7 ms for the FCD group, 95.2 ± 3.0 ms
for the non-FCD group, and 79.2 ± 4.9 ms for the non-lesion group. The spike duration

ip
t

in the FCD and non-lesion groups was significantly shorter than in the non-FCD group

cr

(p = 0.0003, p = 0.0202, respectively), while there was no difference between the FCD
and non-lesion groups (Fig. 2A). The cut-off value determined from the ROC curve

us

for differentiating FCD from non-FCD was 80.9 ms (sensitivity, 1.0; specificity, 1.0).
The average spike amplitude was not significantly different among the three

an

groups (FCD, 824.6 ± 96.8 fT/cm; non-FCD, 598.9 ± 34.5 fT/cm; non-lesion, 684.8 ±

M

71.8 fT/cm).

The average spike tilt was 27.5 ± 4.0 fT/cm/ms for the FCD group, 15.1 ± 1.2

d

fT/cm/ms for the non-FCD group, and 19.5 ± 2.2 fT/cm/ms for the non-lesion group.

Ac
ce
pt
e

The tilt in the FCD group was steeper than that in the non-FCD group (p = 0.0058).
ROC curve analysis revealed the tilt cut-off value for differentiating FCD from
non-FCD was 18.9 fT/cm/ms (sensitivity, 1.0; specificity, 0.9). There was no
difference between the FCD and non-lesion groups, or between the non-FCD and
non-lesion groups (Fig. 2B).

Data from all patients were plotted on a scatter graph showing duration and tilt

(Fig. 2C), which allowed visualization of the morphological differences between the
three groups.
In 17 patients with visible MRI lesions, there was a negative correlation between
depth of the lesion and tilt (F1,16 = 17.19, p = 0.0009; Fig. 3).

11
Page 11 of 29

3.2. Representative cases
Patient 2: FCD group (Fig. 4)
This patient was a 14-year-old boy with parietal lobe epilepsy. His initial seizure
occurred at the age of eight years, and was described as a right-sided hemiconvulsion

ip
t

with somatosensory auras in the right upper and lower limbs. Electroencephalography

cr

(EEG) demonstrated intermittent spikes for the C3, P3, and Cz electrodes, while MEG
showed repetitive spikes, and ECDs were clustered at the left post-central gyrus

us

(Brodmann’s area 1, 2, 3). Although his previous MRI had not indicated any apparent
abnormalities, his subsequent MRI was reviewed carefully taking into account the

an

MEG findings. The MR T2-weighted image showed a slightly high signal where the

M

ECDs were clustered and in the subcortical white matter. The patient underwent partial
resection to avoid total sensory palsy. Thereafter, his convulsive seizures ceased and

d

he has only experienced simple partial seizures with abnormal sensation in the right

Ac
ce
pt
e

arm. Pathologically, the FCD was classified as type IIA. We analyzed the morphology
of the clustered MEG spikes according to our definition and found an average duration
of 59.5 ms and an average tilt of 22.3 fT/cm/ms.

Patient 10: non-FCD group (Fig. 4)

This representative case was a 13-year-old girl who was diagnosed with medial

temporal lobe epilepsy with left hippocampal sclerosis. Her initial seizure occurred at
11 years of age and was described as a complex partial seizure with oral and limb
automatism. EEG demonstrated sporadic spikes for the F7 and T3 electrodes, while
MEG showed epileptic spikes in the left temporal area, and ECDs were clustered in a
horizontal direction in the anterior part of the inferior temporal gyrus (Brodmann’s

12
Page 12 of 29

area 20). She underwent selective amygdalohippocampectomy and has since been
seizure-free. Analysis of the MEG spikes revealed an average duration of 89.3 ms, and

Ac
ce
pt
e

d

M

an

us

cr

ip
t

an average tilt of 12.7 fT/cm/ms.

13
Page 13 of 29

4. Discussion
Previous MEG studies in epileptic patients have focused mainly on the source
location and orientation of the epileptic current (Ebersole and Pacia, 1996; Fukao et
al., 2010; Ishitobi et al., 2005; Kakisaka et al., 2009; Oishi et al., 2006; Park et al.,

ip
t

2004; Stefan et al., 2000), with the results correlating well with the epileptogenic focus

cr

that was validated by the operative outcome or electrocorticography (ECoG) findings.
Accordingly, MEG is currently recognized as an indispensable neuroimaging tool for

us

epileptic surgery.

In our current study, we concentrated on the shape of epileptic spikes in MEG

an

recordings, based on the findings of unique needle-like spikes in patients with active

M

epileptogenic foci, such as those with FCD. Ishii et al. (2008) also focused on the
morphology of MEG spikes and applied the Synthetic Aperture

d

Magnetometry-kurtosis (SAM(g2)) method to pinpoint the epileptogenic zone in

Ac
ce
pt
e

patients with FCD. SAM(g2) is a spatially-filtered source localization MEG technique
designed to provide an automated analysis of ‘spikiness’ by using the excess kurtosis
statistic (Kirsch et al., 2006). They used this procedure to compare the interictal
MEG-SAM(g2) analysis with corresponding ECoG results, and found that the area of
highest kurtosis value on MEG-SAM(g2) was localized within the ictal onset zone,
thus histopathologically confirming FCD. Their findings overlap with the
morphological characteristics of FCD spikes in our current study.
We have confirmed that MEG spikes of FCD patients had significantly shorter
spike duration and steeper tilt than those of non-FCD patients. We suspect that these
morphological characteristics would be related to the involvement and synchronization
of restricted cortices. FCD has intrinsic epileptogenicity and its spike sources must be

14
Page 14 of 29

derived from within these foci, so the epileptic current should be generated in a narrow
area in the cerebral cortex.
A magnetic field is influenced by distance according to Biot and Savart law, and
decreases in amplitude as the distance between the spike source and the detector

ip
t

increases (Barkley & Baumgartner, 2003; de Jongh et al, 2005). MEG showed higher

cr

sensitivity to epileptic discharges in neocortical area than to discharges in deep tissues
in previous studies of simultaneous MEG and ECoG recordings (Mikuni et al, 1997;

us

Oishi et al, 2002). On simultaneous MEG and intracranial depth electrode EEG
recording, MEG spikes with medial temporal focus were significantly lower in

an

amplitude than that with neocortical focus, however, the amplitude of substrate

M

intracranial EEG spikes was similar (Santiuste et al, 2008). In our study the average
amplitude was not significantly different among the three groups, while there was a

d

significant difference in the average tilt (amplitude/duration ratio) between the FCD

Ac
ce
pt
e

and non-FCD groups. These findings suggest spike tilt is a more sensitive parameter
reflecting steepness of spikes than amplitude alone. Our additional study demonstrated
that the tilt was well correlated to the depth of the lesion. The short and steep spike
shape tends to be exhibited in patients with FCD at the superficial neocortical gyrus.
This study showed no apparent difference in the shape of MEG spikes between

FCD and non-lesion patients, with short duration MEG spikes recorded in both groups.
MRI does not always show the presence of FCD, especially FCD type I and IIA,
which are also major causes of intractable localization-related epilepsy for which
surgery leads to seizure control (Fauser et al., 2004; Krsek et al., 2009; Widjaja et al.,
2008). Therefore, it is possible that our findings show the presence of FCD that was
undetected by MRI among the non-lesion group. However, our study has a critical

15
Page 15 of 29

limitation – because none of the non-lesion group patients came to surgery, we could
not confirm MRI-invisible FCD by pathologic findings from the surgical specimens.
Our clinical goal is to noninvasively evaluate epileptogenic lesions in candidates for
epilepsy surgery, and to define the presence of FCD even in cases where the MRI

ip
t

lesions are subtle, since the outcome of surgery in these patients is not necessarily

cr

unfavorable (Alacrón et al., 2006; Chapman et al., 2005; Cukiert et al., 2001). Indeed,
among our patients, some FCD lesions were identified on MRI only after carefully

us

reviewing where the ECDs of MEG were clustered. However, an abnormal MRI is one
of the predictors of seizure freedom in the surgical management of FCD, so epilepsy

an

surgery for patients with no MRI-evident lesion is still challenging, and careful

M

interpretation of MEG results is needed even when ECDs are well clustered.
Occasionally, clustered ECDs might have other origins, and reflect propagated

d

epileptic current, such like in patients with MTLE. We therefore propose that a

Ac
ce
pt
e

detailed analysis of MEG spike morphology could enable a more accurate detection of
true epileptogenic lesions, such as those typically seen in FCD. MEG is one of the
unique tools for the functional analysis to the organic lesion on cerebral cortex
non-invasively. Conventional researches of MEG study for the symptomatic
epileptogenic lesion have mentioned the clustering of epileptic spike sources. Our
current view for the epileptic focus has demonstrated another aspect of MEG analysis
for symptomatic epilepsy.

5. Conclusions
In conclusion, morphological analysis of MEG spikes has the advantage of
noninvasively evaluating the etiology of epileptogenic lesions. This approach is
16
Page 16 of 29

helpful for detecting a strong, localized epileptogenic focus such as that typically

Ac
ce
pt
e

d

M

an

us

cr

ip
t

observed in FCD, even if the MRI findings are subtle.

17
Page 17 of 29

6. Acknowledgements
We thank Prof. Tadashi Ariga of the Department of Pediatrics, Hokkaido University

7. Conflicts of Interest

us

None of the authors have any conflicts of interest to disclose.

cr

ip
t

Graduate School of Medicine, for his valuable editorial opinion.

We confirm that we have read the Journal’s position on issues involved in ethical

Ac
ce
pt
e

d

M

an

publication and affirm that this report is consistent with those guidelines.

18
Page 18 of 29

8. References
Alarcón, G., Valentin, A., Watt, C., Selway, R.P., Lacruz, M.E., Elwes, R.D.C., Jarosz,
J.M., Honavar, M., Brunhuber, F., Mullatti, N., Bodi, I., Salinas, M., Binnie,
C.D., Polkey, C.E., 2006. Is it worth pursuing surgery for epilepsy in patients

ip
t

with normal neuroimaging? J. Neurol. Neurosurg. Psychiatry 77, 474-480.

cr

Alexandre, V., Jr., Walz, R., Bianchin, M.M., Valasco, T.R., Terra-Bustamante, V.C.,

Wichert-Ana, L., Araújo, D., Jr., Machado, H.R., Assirati, J.A., Jr., Carlotti, C.G.,

us

Jr., Santos, A.C., Serafini, L.N., Sakamoto, A.C., 2006. Seizure outcome after
surgery for epilepsy due to focal cortical dysplastic lesions. Seizure 15, 420-427.

an

Barkley, G.L., Baumgartner, C., 2003. MEG and EEG in epilepsy. J. Clin.

M

Neurophysiol. 20, 163-178.

Bast, T., Oezkan, O., Rona, S., Stippich, C., Seitz, A., Rupp, A., Fauser, S., Zentner, J.,

d

Rating, D., Scherg, M., 2004. EEG and MEG source analysis of single and

Ac
ce
pt
e

averaged interictal spikes and averaged interictal spikes reveals intrinsic
epileptogenicity in focal cortical dysplasia. Epilepsia 45, 621-631.

Chapman, K., Wyllie, E., Najm, I., Ruggieri, P., Bingama, W., Lüders, J., Katagal, P.,
Lachhwani, D., Dinner, D., Lüders, H.O., 2005. Seizure outcome after epilepsy
surgery in patients with normal preoperative MRI. J. Neurol. Neurosurg.
Psychiatry 76, 710-713.

Cohen-Gadol, A.A., Ozduman, K., Bronen, R.A., Kim, J.H., Spencer, D.D., 2004.
Long-term outcome after epilepsy surgery for focal cortical dysplasia. J.
Neurosurg. 101, 55-65.
Cukiert, A., Buratini, J.A., Machado, E., Sousa, A., Vieira, J.O., Argentoni, M.,
Forster, C., Baldauf, C., 2001. Result of surgery in patients with refractory

19
Page 19 of 29

extratemporal epilepsy with normal or nonlocalizing magnetic resonance
findings investigated with subdural grids. Epilepsia 42, 889-894.
de Jongh, A., de Munck, J.C., Goncalves, S.I., Ossenblok, P., 2005. Differences in
MEG/EEG epileptic spike yields explained by regional differences in

ip
t

signal-to-noise ratios. J. Clin. Neurophysiol. 22, 153-158.

cr

Ebersole, J.S., Pacia, S.V., 1996. Localization of temporal lobe foci by ictal EEG
patterns. Epilepsia 37, 386-399.

us

Fauser, S., Schulze-Bonhage, A., Hanegger, J., Carmona, H., Huppertz, H.J., Pantazis,
G., Rona, S., Bast, T., Strobi, K., Steinhoff, B.J., Korinthenberg, R., Rating, D.,

an

Volk, B., Zentner, J., 2004. Focal cortical dysplasia: surgical outcome in 67

2406-2418.

M

patients in relation to histological subtypes and dual pathology. Brain 127,

d

Fukao, K., Inoue, Y., Yagi, K., 2010. Magnetoencephalographic correlates of different

Ac
ce
pt
e

types of aura in temporal lobe epilepsy. Epilepsia 51, 1846-1851.
Hong, S.C., Kang, K.S., Seo, D.W., Hong, S.B., Lee, M., Nam, D.H., Lee, J.I., Kim,
J.S., Shin, H.J., Park, K., Eoh, W., Suh, Y.L., Kim, J.H., 2000. Surgical treatment
of intractable epilepsy accompanying cortical dysplasia. J. Neurosurg. 93,
766-773.

Iida, K., Otsubo, H., Matsumoto, Y., Ochi, A., Oishi, M., Holowka, S., Pang, E., Elliot,
I., Weiss, S.K., Chuang, S.H., Snead, O.C., III, Rutka, J.T., 2005. Characterizing
magnetic spike sources by using magnetoencephalography-guided
neuronavigation in epilepsy surgery in pediatric patients. J. Neurosurg. 102,
187-196.
Ishibashi, H., Simos, P.G., Wheless, J.W., Baumgartner, J.E., Kim, H.L., Castillo,

20
Page 20 of 29

E.M., Davis, R.N., Papanicolaou, A.C., 2002. Localization of ictal and interictal
bursting epileptogenic activity in focal cortical dysplasia: agreement of
magnetoencephalography and electrocorticography. Neurol. Res. 24, 525-530.
Ishii, R., Canuet, L., Ochi, A., Xiang, J., Imai, K., Chan, D., Iwase, M., Takeda, M.,

ip
t

Snead, O.C., III, Otsubo, H., 2008. Spatially filtered magnetoencephalography

cr

compared with electrocorticography to identify intrinsically epileptogenic focal
cortical dysplasia. Epilepsy Res. 81, 228-232.

us

Ishitobi, M., Nakasato, N., Yamamoto, K., Iinuma, K., 2005. Opercular to

interhemispheric source distribution of benign rolandic spikes of childhood.

an

Neuroimage 25, 417-423.

M

Ito, T., Otsubo, H., Shiraishi, H., Yagyu, K., Takahashi, Y., Ueda, Y., Takeuchi, F.,
Takahashi, K., Nakane, S., Kohsaka, S., Saitoh, S., 2014. Advantageous

d

information provided by magnetoencephalography for patients with neocortical

Ac
ce
pt
e

epilepsy. Brain Dev. doi: 10.1016/j.braindev.2014.04.006.
Kakisaka, Y., Nakasato, N., Haginoya, K., Kanno, A., Tsuchiya, S., 2009.
Sensorimotor seizures of pediatric onset with unusual posteriorly oriented
rolandic spikes. Epilepsy Res. 84, 153-158.

Kirsch, H.E., Robinson, S.E., Mantle, M., Nagarajan, S., 2006. Automated localization
of magnetoencephalographic interictal spikes by adaptive spatial filtering. Clin.
Neurophysiol. 117, 2264-2271.

Kloss, S., Pieper, T., Pannek, H., Holthausen, H., Tuxhom, I., 2002. Epilepsy surgery
in children with focal cortical dysplasia (FCD). Neuropediatrics 33, 21-26.
Krai, T., Clusmann, H., Blümcke, I., Fimmers, R., Ostertun, B., Kurthen, M.,
Schramm, J., 2003. Outcome of epilepsy surgery in focal cortical dysplasia. J.

21
Page 21 of 29

Neurol. Neurosurg. Psychiatry 74, 183-188.
Krsek, P., Pieper, T., Karimeier, A., Hildebrandt, M., Kolodziejczyk, D., Winkler, P.,
Pauli, E., Blümcke, I., Holthausen, H., 2009. Different presurgical characteristics
and seizure outcomes in children with focal cortical dysplasia type I or II.

ip
t

Epilepsia 50, 125-137.

cr

Mikuni, N., Nagamine, T., Ikeda, A., Terada, K., Taki, W., Kimura, J., Kikuchi, H.,

Shibasaki, H., 1997. Simultaneous recording of epileptiform discharges by MEG

us

and Subdural electrodes in temporal lobe epilepsy. Neuroimage 5, 298-306.
Morioka, T., Nishio, S., Ishibashi, H., Muraishi, M., Hisada, K., Shigeto, H.,

an

Yamamoto, T., Fukui, M., 1999. Intrinsic epileptogenicity of focal cortical

M

dysplasia as revealed by magnetoencephalography and electrocorticography.
Epilepsy Res. 33, 177-187.

d

Oishi, M., Otsubo, H., Kameyama, S., Morita, N., Masuda, H., Kitayama, M., Tanaka,

Ac
ce
pt
e

R., 2002. Epileptic spikes: magnetoencephalography versus simultaneous
electrocorticography. Epilepsia 43, 1390-1395.

Oishi, M., Kameyama, S., Masuda, H., Tohyama, J., Kanazawa, O., Sasagawa, M.,
Otsubo, H., 2006. Single and multiple clusters of magnetoencephalographic
dipoles in neocortical epilepsy: significance in characterizing the epileptogenic
zone. Epilepsia 47, 355-364.

Otsubo, H., Ochi, A., Elliott, I., Chuang, S.H., Rutka, J.T., Jay, V., Aung, M., Sobel,
D.F., Snead, O.C., III, 2001. MEG predicts epileptic zone in lesional
extrahippocampal epilepsy: 12 pediatric surgery cases. Epilepsia 42, 1523-1530.
Palmini, A., Gambardella, A., Andermann, F., Dubeau, F., da Costa, J.C., Olivier, A.,
Tampieri, D., Gloor, P., Quensney, F., Andermann, E., Paglioli, E., Paglioli-Neto,

22
Page 22 of 29

E., Andermann, L.C., Leblanc, R., Kim, H.I., 1995. Intrinsic epileptogenicity of
human dysplastic cortex as suggested by corticography and surgical results. Ann.
Neurol. 37, 476-487.
Palmini, A., Najm, I., Avanzini, G., Babb, T., Guerrini, R., Foldvary-Schaefer, N.,

ip
t

Jackson, G., Lüders, H.O., Prayson, R., Spreafico, R., Vinters, H.V., 2004.

cr

Terminology and classification of cortical dysplasias. Neurology 62, S2-S8.

Park, H.M., Nakasato, N., Iwasaki, M., Shamoto, H., Tominaga, T., Yoshimoto, T.,

us

2004. Comparison of magnetoencephalographic spikes with and without

concurrent electroencephalographic spikes in extratemporal epilepsy. Tohoku J.

an

Exp. Med. 203, 165-174.

M

Santiuste, M., Nowak, R., Russi, A., Tarancon, T., Oliver, B., Ayats, E., Scheler, G.,
Graetz, G., 2008. Simultaneous magnetoencephalography and intracranial EEG

d

registration: technical and clinical aspects. J. Clin. Neurophysiol. 25, 331-339.

Ac
ce
pt
e

Stefan, H., Hammel, C., Hopgengärtner, R., Pauli, E., Tilz, C., Ganslandt, O., Kober,
H., Möler, A., Buchfelder, M., 2000. Magnetoencephalography in extratemporal
epilepsy. J. Clin. Neurophysiol. 17, 190-200.

Stefan, H., Hummel, C., Scheler, G., Genow, A., Druschky, K., Tilz, C., Kaltenhäuser,
M., Hopfengärtner, R., Buchfelder, M., Romstöck, J., 2003. Magnetic brain
source imaging of focal epileptic activity: a synopsis of 455 cases. Brain 126,
2396-2405.

Taylor, D.C., Falconer, M.A., Bruton, C.J., Corsellis, J.A., 1971. Focal dysplasia in the
cerebral cortex in epilepsy. J. Neurol. Neurosurg. Psychiatry 34, 369-387.
Turkdogan, D., Duchowny, M., Resnick, T., Jayakar, P., 2005. Subdural EEG patterns
in children with taylor-type cortical dysplasia: comparison with nondysplastic

23
Page 23 of 29

lesions. J. Clin. Neurophysiol. 22, 37-42.
Widjaja, E., Otsubo, H., Raybaud, C., Ochi, A., Chan, D., Rutka, J.T., Snead, O.C., III,
Halliday, W., Sakuta, R., Galicia, E., Shelef, I., Chuang, S.H., 2008.
Characteristics of MEG and MRI between Taylor’s Focal cortical dysplasia (type

ip
t

II) and other cortical dysplasia: surgical outcome after complete resection of

cr

MEG spike source and MR lesion in pediatric cortical dysplasia. Epilepsy Res.

Ac
ce
pt
e

d

M

an

us

82, 147-155.

Figure captions

24
Page 24 of 29

Figure 1
Flowchart of inclusion criteria for analysis. Diagnoses other than symptomatic
localization-related epilepsy (SLRE) are as follows and were initially excluded from
our database: idiopathic generalized epilepsy (n = 20), idiopathic localization-related

ip
t

epilepsy (n = 25), symptomatic generalized epilepsy (n=25), unclassified (n = 41) and
non-epilepsy (n = 34). Cases with clustered equivalent current dipoles (ECDs) but

cr

having destructive brain lesions were also excluded as follows: infarction (n = 6),

us

post-lobectomy (n = 3), broad white matter lesion (n = 2) and post-encephalopathy (n
= 1). Fully matched cases (n = 28) were divided into three groups according to the

an

magnetic resonance image (MRI) findings as shown in the bottom panels.

Figure 2

M

MEG, magnetoencephalography; FCD, focal cortical dysplasia.

Results of waveform analysis. Data are mean ± standard error of the mean (SEM). (A)

Ac
ce
pt
e

d

Comparison of magnetoencephalography (MEG) spike durations showed that the
average spike durations in the focal cortical dysplasia (FCD) and non-lesion groups
were significantly shorter than in the non-FCD group (*p = 0.0003, *p = 0.0202,
respectively), but there was no significant difference between the FCD and non-lesion
groups. (B) Comparison of tilt revealed that the average tilt in the FCD group was
steeper than that in the non-FCD group (*p = 0.0058). There was no difference
between the FCD and non-lesion groups, or between the non-FCD and non-lesion
groups. (C) A scatter graph between duration (X-axis) and tilt (Y-axis) for all patients
allowed visualization of the morphological differences among the three groups. The
FCD plots (red dots) were distributed in the middle upper area, representing short
duration and steep tilt. In contrast, the non-FCD plots (white dots) tended to be in the
right lower area, representing long duration and gentle tilt. The non-lesion plots (blue
25
Page 25 of 29

dots) were spread over both areas. A horizontal dotted line on the Y-axis indicates the
cut-off value (18.9 fT/cm/ms) determined from the ROC curve for differentiating FCD
from non-FCD.
Figure 3

ip
t

Correlation between lesion depth and spike tilt. In 17 patients with a lesion visible via

cr

magnetic resonance imaging, there was a negative correlation between the depth of the
lesion and the average tilt of magnetoencephalography spikes. (F1, 16 = 17.19, p =

us

0.0009).
Figure 4

an

Waveform analysis of magnetoencephalography (MEG) spikes in representative cases.

M

Panels A, B and C show data from a representative patient from the focal cortical
dysplasia group (Patient 2), while panels D, E, F show data from, a representative

d

patient from the non-focal cortical dysplasia group (Patient 10). (A) Magnetic

Ac
ce
pt
e

resonance image (MRI) (T2-weighted) showing high signal in the left postcentral
gyrus and subsequent white matter (yellow arrow). (B) Clustered equivalent current
dipoles (ECDs) were located in the left sensory area, within the MRI-detected lesion
(red dots). (C) Distribution of the 204-channel MEG spikes, and an enlarged
representative spike in the left central area showing a short, steep shape. (D) MRI
(fluid-attenuated inversion recovery) showing left hippocampal atrophy (yellow
arrow). (E) Clustered ECDs were located in a horizontal direction in the anterior part
of the inferior temporal gyrus (red dots), a typical pattern of source localization of
MEG in medial temporal lobe epilepsy. (F) Distribution of the 204-channel MEG
spikes, and an enlarged representative spike in the left temporal area showing a
slightly longer, blunt shape. The waveform analysis parameters (dotted lines) were t

26
Page 26 of 29

(duration), defined as the period between the onset and the end of each trough of the
spike source around the peaks; t1, the period between the onset of the spike and the
peak; and h, the amplitude between the spike onset and the peak. Tilt was calculated as

Ac
ce
pt
e

d

M

an

us

cr

Table 1. Patient profiles, MRI, MEG, and pathological findings.

ip
t

t1/h, and reflected spike sharpness.

27
Page 27 of 29

Lesion-side

Brodmann’s area of

MRI

Pathological

/Age (y)

/Location

clustered ECDs

findings

findings

1

Male/6

Left/F

4

FCD

not done

2

Male/14

Left/P

1, 2, 3

FCD

FCD IIA

3

Female/34

Right/F

47

FCD

not done

4

Female/6

Left/P

40

FCD

not done

5

Male/1

Left/P

19

FCD

not done

6

Female/23

Left/F

6

7

Male/17

Left/F

9,10

8

Male/34

Left/T

20

9

Male/29

Right/F

10

Female/13

Left/T

11

Female/16

Right/T

Non-

12

Male/14

Right/T

FCD

13

Male/20

Left/T

14

Male/14

Right/T

15

Male/16

Left/T

16

Female/16

17

cr

not done

FCD

FCD IA

us

FCD

HS

13, 14, 15, 16

Tumor

CH

20

HS

HS

20

HS

HS

21

HS

HS

21

HS

HS

20

HS

HS

27

HS

HS

Right/T

38

AE

not arrived

Male/17

Right/P

1, 2, 3

Tumor

Glioma

18

Male/8

Left/F

4

normal

not done

19

Male/8

Left/F

46

normal

not done

20

Male/9

Right/F

6

normal

not done

21

Female/16

Left/P

40

normal

not done

22

Female/19

Right/P

39

normal

not done

23

Male/17

Right/T

37

normal

not done

24

Female/23

Left/F

6

normal

not done

25

Female/12

Left/T

22

normal

not done

26

Male/31

Left/T

21

normal

not done

27

Male/21

Right/O

18

normal

not done

28

Male/29

Left/T

29

normal

not done

d

M

an

HS

Ac
ce
pt
e

FCD

Patient

ip
t

Sex

Group

Non-

lesion

MRI, magnetic resonance imaging; MEG, magnetoencephalography; FCD, focal cortical
dysplasia; ECDs, equivalent current dipoles; F, frontal; P, parietal; T, temporal; O, occipital;
HS, hippocampal sclerosis; CH, cavernous hemangioma; AE, amygdala enlargement.

28
Page 28 of 29

29

Page 29 of 29

d

Ac
ce
pt
e
us

an

M

cr

ip
t