ARTICLE IN PRESS+ModelEPIRES-5197; No. of Pages 9

Epilepsy Research (2014) xxx, xxx—xxx

jo ur nal ho me p ag e: www.elsev ier .com/ locate /ep i lepsyres

Cortical involvement in focal epilepsieswith epileptic spasms

Sabine de la Vaissièrea,b, Mathieu Milhe, Didier Scavardac,Romain Carrond, Anne Lépinea, Agnes Trébuchonf,Martine Gavaret f, Russell Hewett f, Virginie Laguittona,Géraldine Daquina, Nathalie Villeneuvea,e,Fabrice Bartolomeia,f,∗

a Hôpital Henri Gastaut, Marseille, Franceb Service de Pédiatrie, Clocheville Hospital, CHRU Tours F-37000, Francec Service de Neurochirurgie Pédiatrique, Assistance Publique des Hôpitaux de Marseille,Marseille F-13005, Franced Service de Neurochirurgie fonctionnelle et Stéréotaxie, Assistance Publique des Hôpitaux de Marseille,Marseille F-13005, Francee Service de Neuropédiatrie, Assistance Publique des Hôpitaux de Marseille, Marseille F-13005, Francef Service de Neurophysiologie Clinique, Assistance Publique des Hôpitaux de Marseille,Marseille F-13005, France

Received 29 April 2014; received in revised form 8 July 2014; accepted 21 August 2014

KEYWORDSEpileptic spasms;Intracerebral EEG;Epileptogenicityindex;Epileptic surgery;

Summary The pathophysiological mechanisms of epileptic spasms are still poorly understood.The role of subcortical structures has been suggested on the basis of non-localized EEG featuresand from experimental data. The description of asymmetric spasms associated with lateralizedEEG patterns has challenged this view and raises the possibility of a cortical origin. This studyinvestigated the cortical organization of partial seizures associated with epileptic spasms inchildren undergoing intracerebral EEG recordings for presurgical evaluation. Eleven children

Supplementary motor with drug resistant epileptic spasms and for whom depth electrode recordings were performedwere retrospectively studied. In all children several features suggested a focal origin. Cortical

area

Please cite this article in press as: de la Vaissière, S., et al., Cortical involvement in focal epilepsies with epileptic spasms.Epilepsy Res. (2014), http://dx.doi.org/10.1016/j.eplepsyres.2014.08.008

involvement was studied using the ‘‘Epileptogenicity Index’’ (EI). A focal origin was finallydemonstrated in 10/11 patients. Seven patients demonstrated pre-ictal changes in the seizureonset zone area. EI analysis showed maximal values in the temporal (n = 5), parietal (n = 1) orfrontal (n = 5) cortices. EEG changes were also observed in the premotor cortex during spasms

∗ Corresponding author at: Service de Neurophysiologie Clinique, CHU Timone-264 Rue st Pierre, 13005 Marseille, France.Tel.: +33 491385833; fax: +33 491385826.

E-mail address: fabrice.bartolomei@ap-hm.fr (F. Bartolomei).

http://dx.doi.org/10.1016/j.eplepsyres.2014.08.0080920-1211/© 2014 Published by Elsevier B.V.

ARTICLE IN PRESS+ModelEPIRES-5197; No. of Pages 9

2 S. de la Vaissière et al.

in patients with frontal or parietal seizures and in 3/5 patients with temporal lobe seizures. Goodsurgical outcome (class I or II) was obtained in 7/10 patients.Seizures associated with epileptic spasms may originate from various cortical regions. Premo-tor/motor cortices are probably involved in determining ictal clinical changes.© 2014 Published by Elsevier B.V.

I

Strsrc

oeslToss1mm

solFh(

i2fetwt

pihpd

M

P

TpMbws

mouso

ddcnd

plcMwtaehmEb

srthusdo

It

Fioa

taduring the transition between ictal and interictal activity)and takes into account the delay of appearance of this dis-charge with respect to seizure onset (Bartolomei et al.,2008, 2010, 2011). The purpose of this index is to provide

ntroduction

ince the most recent classification proposed by the ILEA,he clinical entity of epileptic spasms (ES) has been sepa-ated from West syndrome and constitutes a distinct type ofeizure (Berg et al., 2010). However their pathophysiologyemain unknown such that at present it has been difficult tolass them as either generalized or partial seizures.

The following clinical description is commonly rec-gnized: an axial contraction, more often flexion thanxtension, brief and sudden, lasting from 0.2 to 2 s, occa-ionally recurring in short clusters separated by periodsasting between 5 and 20 s (Holmes and Vigevano, 1997).hey occur typically in infancy (6—12 months) but they mayccur at a later stage of life (Gobbi et al., 1987). In addition,ome patients with ES may have other types of epilepticeizures, most commonly partial seizures (Kubota et al.,999; Pachatz et al., 2003). In these cases surgical treatmentay lead to seizure remission and neurocognitive improve-ent (Jonas et al., 2005; Ricard-Mousnier et al., 2012)Epileptic spasms were initially considered as originating

ub-cortically (Gastaut et al., 1964), but a cortical originf the spams has alternatively been suggested, in particu-ar from the results of cortical surgery (Asano et al., 2001).urthermore, more recent animal models also support theypothesis of cortical involvement in the genesis of ESScantlebury et al., 2010).

Few studies have analyzed intracerebral EEG recordingsn this context (Asano et al., 2005; Ricard-Mousnier et al.,012). A corticographic study of 62 spasms from patients suf-ering from tuberous sclerosis suggested that spasms werelicited by a cortical onset (Asano et al., 2005). Two pat-erns of discharges were described in this study, a first typeith a localized spike followed by fast activity or a second

ype without a focal spike.Our objective was to study the cortical organization of

artial seizures associated with ES in children undergoingntracerebral EEG recordings for presurgical evaluation. Weave analyzed the ictal IcEEG characteristics in children withharmacoresistant ES, for whom intracerebral EEG recor-ings were proposed due to a suspected focal onset.

aterial and methods

atients and EEG recordings

his retrospective study analyzed the IcEEG of children withharmacoresistant ES recorded at the hospital La Timone inarseilles, France. They were selected from 280 intracere-

Please cite this article in press as: de la Vaissière, S., et al., CorEpilepsy Res. (2014), http://dx.doi.org/10.1016/j.eplepsyres.

ral investigations between 2002 and 2012. All the patientsith ongoing ES that had IcEEG were included in this retro-

pective study.

qtp

Non-invasive pre-surgical recordings (pre-surgical assess-ent Phase 1) were performed at the hospitals Henri Gastaut

r La Timone in Marseilles. The EEG trace was performedsing 20 scalp electrodes, following the international 10—20ystem. Additional polygraphic recordings were composedf EMG of both deltoids and ECG.

Pre-surgical assessment phase 2 consisted of IcEEG recor-ings over a number of days. It is performed followingiscussion by a multidisciplinary team, utilizing the clini-al and imaging data, and the results from phase 1. Theumber and position of the implanted electrodes were alsoiscussed.

The implantation of the intracerebral electrodes waserformed in the neurosurgery departments. The adequateocalization of the electrodes in the cerebral space washecked using 1.5 MRI or using a fusion of preimplatationRI and CT scan with electrodes in place. ICEEG recordingsere performed using intracerebral multiple contact elec-

rodes (10—15 contacts: 2 mm diameter; 0.8 mm and 1.5 mmpart). The anatomical targeting and number of necessarylectrodes was established in each patient according to theypotheses for localization of the epileptogenic zone deter-ined by the clinical data and EEG recordings from phase 1.

ach electrode comprised of multiple contact points num-ered 1 to 15.

The signals were recorded on a 196 channels DeltamedTM

ystem. They were sampled at 512 Hz or 1024 Hz andecorded on hard disk (16bits/sample) using no digital fil-er. The only filter present in the acquisition process was aigh-pass analog filter (cut-off frequency equal to 0.16 Hz)sed to remove very slow non-physiological variations thatometimes contaminate the baseline. The video-EEG recor-ings were prolonged as long as necessary to capture severalf the patient’s habitual seizures.

ntracerebral EEG signal analysis: determination ofhe Epileptogenicity Index (EI)

or each seizure, the corresponding IcEEG trace was stud-ed and the Epileptogenic Index (EI) was calculated. Thebjective was to better characterize the regions involvedt seizure onset.

This quantification has been proposed in order to charac-erize the propensity of a given brain structure to generate

‘rapid discharge’ (the high frequency oscillations observed

tical involvement in focal epilepsies with epileptic spasms.2014.08.008

uantified information about the behavior of brain struc-ures recorded from signals they generate during the seizurerocess. This index summarizes two pieces of information

Please cite

this article

in press

as: de

la Vaissière,

S., et

al., Cortical

involvement

in focal

epilepsies w

ith epileptic

spasms.

Epilepsy Res.

(2014), http://dx.doi.org/10.1016/j.eplepsyres.2014.08.008

AR

TIC

LE

IN P

RE

SS

+Model

EPIRES-5197;

No.

of Pages

9

Cortical involvem

ent during

epileptic spasm

s

3

Table 1 Main clinical characteristics of studied patients. Cognitive evaluation was assessed for each patient using first the Weschler Intelligence Scale for Children (WISC IV)(Parkin and Beaujean, 2012). Intelligence quotients using the Wisc IV were Total IQ (TIQ), Verbal IQ (VIQ), Non verbal IQ (NVIQ).

Age atonset

NeuropsychologicalTesting (preop)TIQ, VIQ, NVIQ

NeuropsychologicalTesting (postop)TIQ, VIQ, NVIQ

NormalMRI

Histopathology HypometabolismPET-scan

Age atiEEG

Age atsurgery

Surgery Outcome/FU(year)

P1 5 months 60,75,58 ND No Right frontal MCD(heterotopia)

Left frontal 12 years 3months

12 years 9months

Right frontallobectomy

IV(3)

P2 12 months NA NA No Left temporalDNET

Left temporal 7 years 8 years Left temporallobectomy

III(4)

P3 Neonatalperiod

45, 45, 45 ND No Non specific Left frontal 8 years 4months

9 years 2months

Right frontal(premotor cortex)cortectomy

IV(3)

P4 8 months 96, 116, 90 86,116,77 No Right temporalDNET

Right temporalposterior

7 years 3months

8 years 5months

Lesionectomy IA(2)

P5 2 years 5months

NA 47,47,45 Yes Right frontal FCD(type 1histopathology)

No 6 years 4months

7 years 11months

Right prefrontalcortectomy

IA(2)

P6 5 months 93, 104, 81 78,90,67 Yes Normal (FCD type1 histopathology)

NP 7 years 1months

17 years Right prefrontalcortectomy

IIA(10)

P7 9 months 49, 58, 58 ND Yes — Left temporal 7 years NF NF —P8 4 years NA 50,63,50 No Left temporal

DNETLeft temporal 6 years 4

months6 years 10months

Lefttemporo-frontalcortectomy

IB(4)

P9 7 years 85, 86, 81 88,92,82 No Right parietalDNET

Right internalparietal

11 years 4months

11 years 9months

Parietal cortectomyaNF Lesionectomy

IA(1)

P10 2 years 72, 76, 69 58, 79,79 No Right temporomesial DNET

Right temporal 5 years 9months

6 years Lesionectomy IB(1)

P11 2 years 3months

78, 82, 82 70, 82,72 No Right temporalmesial DNET

Right temporal 7 years 2months

7 years 6months

ATL+ Lesionectomy IA(1)

Abbreviations: NA: not applicable, non testable because of major cognitive or/and behavioral problems. ND: not done. IcEEG: intracerebral EEG figures.

ARTICLE IN PRESS+ModelEPIRES-5197; No. of Pages 9

4 S. de la Vaissière et al.

Figure 1 Example of Stereotactic EEG (SEEG) recordings in a patients with a DNET located in the left temporo-basal cortex (Pt8). (A) SEEG scheme showing the location of the electrodes and MRI showing the lesion. (B) Infraclinical discharges affecting thel d wic

isddaftrtimm

ipss‘

R

C

Tpi

f1afmianattditdtweae

ai

esion site (electrodes L′ and OT′) (C) Ictal discharge associateortex (electrode SA′) in addition to the temporal region.

nto a single quantity: (1) whether or not the recorded braintructure is involved in the generation of a high frequencyischarge and (2) when involved, whether or not this rapidischarge is delayed with respect to rapid discharges gener-ted by other structures. A normalized value is used rangingrom 0 to 1. If there is no involvement of the brain struc-ure, the EI = 0 whereas if the brain structure generates aapid discharge and the time to seizure onset is minimal,he EI = 1. An EI between 0 and 1 corresponds to secondarynvolvement of the brain structure concerned (for detailedethodology, see Bartolomei et al. (2008) and supportingethods)This index therefore allows a semi-automatic character-

zation of the anatomical zone from which the epilepticrocess originates. In this study it was calculated for eacheizure in each child recorded by IcEEG. The data was sub-equently integrated into graphical form representing theepileptogenic’ zone.

esults

linical data

Please cite this article in press as: de la Vaissière, S., et al., CorEpilepsy Res. (2014), http://dx.doi.org/10.1016/j.eplepsyres.

he clinical data has been summarized in Table 1. Elevenatients (four girls, seven boys) presenting with spams werenvestigated in our unit between 2002 and 2012 and selected

acp

th clinical spams. This discharge is localized to the premotor

or this study. The average age at time of IcEEG was 7 years0 months (range 5 years 9 months—12 years 3 months). Theverage age of onset of epilepsy was 22.5 months (rangingrom birth to 7 years). The epilepsies of all the children com-enced with ES. Temporary epilepsy remission was reported

n one patient with appropriate neurodevelopment associ-ted with the period of seizure freedom. At the preoperativeeuropsychological evaluation, only 3 patients had scored at

normal/subnormal level (see details in Table 1). Hypsary-hmia was initially present in one patient. Of the 11 patients,hree patients had normal Magnet Resonance Imaging (MRI)espite repeated imaging. In one, post-operative histolog-cal analysis demonstrated a dysplastic lesion. Imaging inhe other patients revealed focal malformation of corticalevelopment (n = 2) or a dysembryoplastic neuroepithelialumor (DNET) (n = 6). The lesions, identified or suspected,ere situated in the frontal (n = 4), temporal (n = 6) or pari-tal (n = 1) regions. Metabolic imaging (PET scan) revealedreas of hypometabolism corresponding to a focal region inight patients.

The description of the spasms recorded at presurgicalssessment (Phases 1 and 2) for each patient is reportedn supporting Table 1 and supporting Table 2.

tical involvement in focal epilepsies with epileptic spasms.2014.08.008

An axial contraction was present in all patients. The aver-ge duration was 0.5 s, ranging between 0.2 and 3 s. Shortlusters of spasms were observed in seven patients. Only oneatient presented with asymmetrical spasms.

ARTICLE IN PRESS+ModelEPIRES-5197; No. of Pages 9

Cortical involvement during epileptic spasms 5

Figure 2 (A) SEEG scheme showing the location of the electrodes in a patient with temporal lobe seizures and spasms (Pt 2).The electrodes explore the temporal region. (NA′, B′, TB′, TP′) and the frontal lobe (SA′, OF′ et OR′). Seizures initially involve theinternal temporal region (internal leads of electrodes TP′, NA′, B′ and TB′) before spreading to the SMA regions (blue arrow, internalleads SA′ electrode). Clinical spasm appears subsequently (red arrow). (B) Quantification of the rapid discharge at seizure onsetin two internal regions (Hip and EC) and two frontal regions (SMA and Orbitofrontal cortex, OFC). The changes in energy ratio ofhigh frequencies (25—90 Hz, (ER(n)) are depicted in a color scale. Increased in ER is seen first in Hip and EC and secondarily (delay

refe

(sPl

FItaarcta(

OPa

800 ms) in the OFC and SMA regions. (For interpretation of thethe web version of this article.)

Intracerebral EEG recordings

IcEEG was obtained in the 11 consecutive patients. Theresults have been separated into location of the epilepto-genic zone as defined by IcEEG findings.

Temporal lobe casesSeizures were localized to the temporal region in 5 cases(cases P2, P4, P8, P10, P11). Complete sequences of seizuresand spasms were recorded in 3 patients (cases P2, P8,P10) while only brief facial minimal contractions (that didnot reproduced the complete ES features) was recording inanother patient (4). No spasms were recorded during IcEEGin patient 11.

In the three cases with IcEEG recorded typical spasms,the occurrence of spasms was associated with a dischargeaffecting both the temporal lobe and the premotor cor-tex. Involvement of the premotor cortex was delayed (mean264 ms ±0.09). Spasms were generally grouped into clus-ters. Figs. 1—2 show two examples of this pattern. In thesetwo cases, the ES coincided with a discharge extending the

Please cite this article in press as: de la Vaissière, S., et al., CorEpilepsy Res. (2014), http://dx.doi.org/10.1016/j.eplepsyres.2

limit of the temporal lobe, in particular affecting the elec-trode exploring the premotor region. The appearance ofictal spasms (cases P2, P8, P10) or more subtle manifesta-tions (case P4) was preceded by repetitive rapid discharges

spct

rences to color in this figure legend, the reader is referred to

with no clinical changes) into the lesion (cases P4 and P8,ee Fig. 1b) or by partial seizures (cases P2 and P10). In case11, isolated subclinical discharges were recorded in theesion and the hippocampal region with no clinical changes.

rontal casesn 4 cases, the origin of the ictal discharge was found to behe frontal cortex (cases P1, P3, P5, P6). In all these cases,

preictal phase was observed, characterized by the appear-nce and the gradual extension in the frontal lobe of isolatedapid discharges and/or spikes. Examples of such preictalhanges are depicted in Fig. 3. Spasms were found to behe main manifestation of seizures in these cases and weressociated with a discharge affecting the premotor cortexFigs. 3—4).

ther casesatient 9 presented with seizures involving both parietalnd premotor cortex. Clusters of spasms followed partial

tical involvement in focal epilepsies with epileptic spasms.014.08.008

eizures and were characterized by brief discharges overarietal and premotor cortices. Patient P7 had an isolatedluster of spasms with a discharge affecting a large part ofhe cortical regions with no clear focal onset.

ARTICLE IN PRESS+ModelEPIRES-5197; No. of Pages 9

6 S. de la Vaissière et al.

Figure 3 SEEG recordings in patient 1 (frontal case) A/MRI showing the position of electrodes on the right side. PM explores thepremotor area (internal leads PM2-3) and the lateral premotor cortex. SA explores the supplementary motor area (SA4-5) and thelateral premotor cortex. FO is an oblique electrode reaching the orbitofrontal cortex (OR1-2). LP is reaching the paracentral lobulecrossing the central sulcus. On the left side (not shown) there one electrode (PM′). A records the amygdala (NA2-3). CC explores themiddle part of the cingulate gyrus (CC1-2) and the lateral prefrontal cortex. CR records the anterior cingulate region (internal leads,CR1-2) and the lateral prefrontal cortex. L is an electrode exploring the lesion (gray matter heterotopia). B/Preictal discharges(blue arrow) occurred in the orbitofrontal region (FO 1—2) with no clinical manifestations (spikes and rapid activities) 2 min beforethe ES occurrence. C/Clinical seizure limited to an ES and associated with a rapid discharge in the orbitofrontal cortex (FO1-2), thea 4-5)

t this a

E

Tsvl

lvlt

tPwec

S

Fmr

scicldAbT(cwovlt

D

nterior cingulate cortex (CCR1-2) and the premotor region (SAhis figure legend, the reader is referred to the web version of

pileptogenicity index analysis

he Epileptogenicity profiles of the analyzed patients (62eizures) are indicated in Fig. 5a. They reflect the averagealues of recorded seizures in the 10 patients in whom aocalized onset was found.

EI values were often maximal at the sites of the suspectedesion or lesion reported on MRI. In the temporal cases, thealues in the premotor cortex were found to be generallyow and lower than in the frontal cases (mean 0.05 ± 007 inemporal cases versus 0.48 ± 0.2 in extra-temporal cases).

Most of the patients disclosed a focal pattern with epilep-ogenicity restricted to lesional sites (cases P10, P4, P8, P10,11, P5, P9). Only 3 cases disclosed more complex patternsith high values outside the lesion site (P1, P3, P6). Inter-stingly, these three cases corresponded to extra-temporalases and were not rendered seizure free after surgery.

urgery and outcome

Please cite this article in press as: de la Vaissière, S., et al., CorEpilepsy Res. (2014), http://dx.doi.org/10.1016/j.eplepsyres.

ollowing IcEEG, 10 patients benefited from surgical treat-ent, two patients required repeat surgery due to seizure

ecurrence and consisted in an extension of the previous

Tsii

(blue arrows). (For interpretation of the references to color inrticle.)

urgery according to the IcEEG results (to the premotorortex more posteriorly in P1 and to the frontal regionn P8). The surgery, guided by the results of the IcEEG,onsisted of a corticectomy encompassing the MRI evidentesion and/or the region considered epileptogenic in six chil-ren and purely a lesionectomy in three children (Table 1).verage age at time of surgery was 9 years 1 month (rangingetween 6 years 10 months to 12 years 9 months 17y 9).he average duration of post-surgical follow-up was 3 years1—10). Among the 10 operated patients, 7 have a good out-ome (5 Engel IA, 1 Engel Ib, 1 Engel IIA). Among the patientsith frontal lobe epilepsy and poor outcome, one had post-peratory contralateral seizures that were recorded duringideo EEG session (P1) and one had a resection that wasimited to the anterior part of the epileptogenic zone givenhe risk of motor deficit (P2).

iscussion

tical involvement in focal epilepsies with epileptic spasms.2014.08.008

his study investigated the cortical organization of partialeizures associated with epileptic spasms. Patients werencluded in this study when IcEEG was decided after non-nvasive data in which a possible focal origin of the ES was

ARTICLE IN PRESS+ModelEPIRES-5197; No. of Pages 9

Cortical involvement during epileptic spasms 7

Figure 4 SEEG recordings in patient 3 (frontal case). (A) MRI showing the position of electrodes on the left side. CR′ records theanterior cingulate region (internal leads, CR′1—2) and the lateral prefrontal cortex. CC′ explores the middle part of the cingulategyrus (CC′1—2) and the lateral prefrontal cortex. PM′ explores the premotor area (internal leads) and the lateral premotor cortex.SA′ explores the supplementary motor area (internal leads, SA′2—3) and the lateral premotor cortex. OR′ is an oblique electrodereaching the orbitofrontal cortex. LP′ is reaching the paracentral lobule crossing the central sulcus. On the right side (not shown)there are two symetrical electrodes SA and PM. (B) SEEG traces showing the discharge associated with a spasm (EMG recordings

e eleuenc

sgondvme

wlahdwc

gwmIigtpI

of the right deltoide). A rapid discharge is seen in the left sidcortices. Below is shown the changes in energy ratio of high freq

suspected. This is a well recognized situation (Asano et al.,2005; Fogarasi et al., 2003; Gobbi et al., 1987; Holmes andVigevano, 1997; Kamei et al., 1996; Kubota et al., 1999;Pachatz et al., 2003; Ricard-Mousnier et al., 2012) but toour knowledge, there is no report of depth electrodes inves-tigations in such cases with the exception of a recent casereport (Ricard-Mousnier et al., 2012). Visual analysis wassupplemented by signal EEG analysis based on the epilep-togenic index method that is particularly suited to detectthe increase in high frequencies at seizure onset (Aubertet al., 2009; Bartolomei et al., 2008, 2010, 2011; Boniniet al., 2013).

In the present series, ES occurred with a variable tem-poral relationship with partial seizures (PS): being part ofthe seizures; preceding the partial seizures; or followingthe seizures. This has been already reported. For examplein a previous paper (Kubota et al., 1999), authors reportedeight cases associated with various etiology, that had ESassociated with PS. Three types of seizure sequence wereidentified: PS followed several seconds later by ES (twopatients), alternating PS and ES starting with PS (threepatients), and PS gradually replaced by ES with overlappingof the two (three patients).

In another report, (Pachatz et al., 2003) investigatedwith video EEG 13 children with partial seizures and ES.Authors identified three groups with different seizure pat-terns regarding the temporal association of ES and partial

Please cite this article in press as: de la Vaissière, S., et al., CorEpilepsy Res. (2014), http://dx.doi.org/10.1016/j.eplepsyres.2

seizures: (a) PS followed by ES; (b) PS appearing during acluster of ES without interrupting the cluster; and (c) com-plex seizure interaction with a succession of ES and partialseizures in a close but variable temporal association.

dceb

ctrodes particularly affecting premotor, motor and prefrontalies (ER(n)) from which is calculated the epileptogenicity index.

In this context, our series also confirms that there is nopecific etiological factor associated with ES from focal ori-in even if we found a majority of patients, having latenset ES, absence of hypsarythmia and presenting with aeurodevelopmental tumors (6/11 cases). Early injury orevelopmental disorders were generally found in the pre-ious series including cortical dysplasia or tubers, cerebralalformations, perinatal anoxic-ischemic injuries (Pachatz

t al., 2003).In this series, we also observed that ES may be associated

ith seizures from various cortical origins since temporalobe seizures as well as extra temporal lobe seizures may bessociated with this clinical manifestation. EI values wereigh and restricted to the lesional site in 7 patients. Thisemonstrates that despite the clinical symptoms and oftenide diffusion of ictal EEG patterns, the epileptic diseasean be quite focal in these children.

A recurring issue has been to question the cortical ori-in of the clinical phenomenon. In our frontal lobe casesith ES, premotor and/or primary motor regions involve-ent was found to be associated with the ES occurrence.

n these cases, the ES are finally very close to patternsnvolved in motor cortex seizures as recorded during presur-ical evaluation. In particular, tonic postural seizures arehe most frequent pattern of seizures involving premotor orremotor and primary motor cortices (Bonini et al., 2013).n the case with parietal epilepsy, the ES were also coinci-

tical involvement in focal epilepsies with epileptic spasms.014.08.008

ent with a discharge affecting both parietal and premotorortex. This pattern has been already described in pari-tal seizures and is closely related to the large connectivityetween parietal and premotor cortices (Bartolomei et al.,

ARTICLE IN PRESS+ModelEPIRES-5197; No. of Pages 9

8 S. de la Vaissière et al.

Figure 5 (A) Profiles of epileptogenicity in patients with ES and Temporal lobe seizures. Mean values of EI is indicated for 15brain regions explored by depth electrodes. EI profiles show maximal epileptogenicity in temporal regions and lower values in thefrontal regions. (B) Profiles of epileptogenicity in patients with ES and extra-temporal seizures (13 brain regions are depicted).Abbreviations: EC: entorhinal cortex, Amy: amygdala, Hip:hippocampus, TP: temporo-polar cortex, TBC: temporobasal cortex,Ins: insula, STG: superior temporal gyrus, PHG: parahippocampal gyrus, Fus: fusiform gyrus, SMA: supplementary motor area, BA6:lateral premotor cortex, aCG: anterior cingulate gyrus BA32, pCG: cingulate gyrus BA 24, OFC: orbitofrontal cortex, preF: prefrontalcortex, Rol: rolandic cortex, Par(SPL): parietal cortex, superior parietal lobe; contra: contralateral side. (C) Mean (±SD) of EI valuesi n mo( nd, t

2ooaTpt

fd(ssecpcbws

tofi

i(awgneenf(

n lesional sites (blue bars), non lesional sites (orange bars) and iFor interpretation of the references to color in this figure lege

011; Salanova et al., 1995). In patients with temporal loberigin the mechanisms of spasms are less evident. In threef our cases, propagation of the discharges to the premotorrea suggested an involvement of the cortical motor system.herefore, a common pathway including an involvement ofremotor cortex seems to be involved in the ES whateverhe cortical origin.

This pathophysiological hypothesis agrees with worksrom animal models of ES. A model (Velisek et al., 2010)emonstrating the role of corticotropin-releasing hormoneCRH) supports the hypothesis of a subcortical origin forpasms and hypsarrythmia. The other models display resultsupporting the hypothesis of a cortical origin. Lee at al (Leet al., 2008) described a rat model with ES secondary toortical developmental abnormalities. Another group alsoublished an animal model in which cortical lesions were

Please cite this article in press as: de la Vaissière, S., et al., CorEpilepsy Res. (2014), http://dx.doi.org/10.1016/j.eplepsyres.

reated with injections of lipopolysaccharide and doxoru-icine (Scantlebury et al., 2010). The rats initially presentedith ES, though then developed other types of epileptic

eizures. Finally another model was a knockout mouse for

aam

tor cortices. Values are obtained from the 62 analyzed seizures.he reader is referred to the web version of this article.)

he ARX gene involved in neuronal migration. The mutationf this gene is associated with cerebral malformations andunctional abnormalities of interneurons and thus supportsnvolvement of the cortex in ES (Marsh et al., 2009).

Surgical outcome was good in our series. 70% were largelymproved. These results compare well with previous dataAsano et al., 2001; Jonas et al., 2005; Pachatz et al., 2003)nd seem particularly favorable in temporal lobe cases,here the relationship between lesion and the epilepto-enic zone was more simple and direct. Larger epileptogenicetworks as found in the cases with poor outcome mayxplain the limited effect of surgery. The extension of thepileptogenic networks has been already shown to have aegative impact on the surgical prognosis in other forms ofocal epilepsies (Aubert et al., 2009; Bartolomei et al., 2010)Bonini et al., 2013)

tical involvement in focal epilepsies with epileptic spasms.2014.08.008

In conclusion, this study confirms that seizures associ-ted with spasms may originate from various cortical regionsnd reflects the primary or delayed involvement of pre-otor cortices. Good surgical outcome may be obtained

IN+Model

G

G

H

J

K

K

L

M

P

P

R

S

S

ARTICLEEPIRES-5197; No. of Pages 9

Cortical involvement during epileptic spasms

when the epileptogenic zone is well restricted to lesionalsites.

Appendix A. Supplementary data

Supplementary data associated with this article can befound, in the online version, at http://dx.doi.org/10.1016/j.eplepsyres.2014.08.008.

References

Asano, E., Chugani, D.C., Juhasz, C., Muzik, O., Chugani, H.T., 2001.Surgical treatment of West syndrome. Brain Dev. 23, 668—676.

Asano, E., Juhasz, C., Shah, A., Muzik, O., Chugani, D.C., Shah,J., Sood, S., Chugani, H.T., 2005. Origin and propagation ofepileptic spasms delineated on electrocorticography. Epilepsia46, 1086—1097.

Aubert, S., Wendling, F., Regis, J., McGonigal, A., Figarella-Branger,D., Peragut, J.C., Girard, N., Chauvel, P., Bartolomei, F., 2009.Local and remote epileptogenicity in focal cortical dysplasiasand neurodevelopmental tumours. Brain 132, 3072—3086.

Bartolomei, F., Chauvel, P., Wendling, F., 2008. Epileptogenicity ofbrain structures in human temporal lobe epilepsy: a quantifiedstudy from intracerebral EEG. Brain 131, 1818—1830.

Bartolomei, F., Cosandier-Rimele, D., McGonigal, A., Aubert, S.,Regis, J., Gavaret, M., Wendling, F., Chauvel, P., 2010. Frommesial temporal lobe to temporoperisylvian seizures: a quan-tified study of temporal lobe seizure networks. Epilepsia 51,2147—2158.

Bartolomei, F., Gavaret, M., Hewett, R., Valton, L., Aubert, S.,Regis, J., Wendling, F., Chauvel, P., 2011. Neural networksunderlying parietal lobe seizures: a quantified study from intra-cerebral recordings. Epilepsy Res. 93, 164—176.

Berg, A.T., Berkovic, S.F., Brodie, M.J., Buchhalter, J., Cross, J.H.,van Emde Boas, W., Engel, J., French, J., Glauser, T.A., Mathern,G.W., Moshe, S.L., Nordli, D., Plouin, P., Scheffer, I.E., 2010.Revised terminology and concepts for organization of seizuresand epilepsies: report of the ILAE Commission on Classificationand Terminology, 2005—2009. Epilepsia 51, 676—685.

Bonini, F., McGonigal, A., Wendling, F., Regis, J., Scavarda, D.,Carron, R., Chauvel, P., Bartolomei, F., 2013. Epileptogenicnetworks in seizures arising from motor systems. Epilepsy Res.

Please cite this article in press as: de la Vaissière, S., et al., CorEpilepsy Res. (2014), http://dx.doi.org/10.1016/j.eplepsyres.2

106, 92—102.Fogarasi, A., Hegyi, M., Tegzes, A., 2003. The predictive value of the

type of seizure onset in infants with epileptic spasms and partialseizures within a single ictal event. Epilepsia 44, 1605—1606.

V

PRESS9

astaut, H., Roger, J., Soulayrol, R., Pinsard, N., 1964.L’encéphalopathie myoclonique infantile avec hypsarythmie(syndrome de West). Masson, Paris.

obbi, G., Bruno, L., Pini, A., Giovanardi Rossi, P., Tassinari, C.A.,1987. Periodic spasms: an unclassified type of epileptic seizurein childhood. Dev. Med. Child Neurol. 29, 766—775.

olmes, G., Vigevano, F., 1997. Infantile spasms. In: Engel,J., Pedley, T. (Eds.), Epilepsy: A Comprehensive Textbook.Lippincott-Raven, New York.

onas, R., Asarnow, R.F., LoPresti, C., Yudovin, S., Koh, S., Wu, J.Y.,Sankar, R., Shields, W.D., Vinters, H.V., Mathern, G.W., 2005.Surgery for symptomatic infant-onset epileptic encephalopathywith and without infantile spasms. Neurology 64, 746—750.

amei, A., Ichinohe, S., Ito, M., Fujiwara, T., 1996. A case of infan-tile spasms: epileptic apnea as partial seizures at onset. BrainDev. 18, 239—241.

ubota, T., Aso, K., Negoro, T., Okumura, A., Natsume, J., Takada,H., Itomi, K., Watanabe, K., Yamamoto, N., 1999. Epilepticspasms preceded by partial seizures with a close temporal asso-ciation. Epilepsia 40, 1572—1579.

ee, C.L., Frost Jr., J.D., Swann, J.W., Hrachovy, R.A., 2008. A newanimal model of infantile spasms with unprovoked persistentseizures. Epilepsia 49, 298—307.

arsh, E., Fulp, C., Gomez, E., Nasrallah, I., Minarcik, J., Sudi,J., Christian, S.L., Mancini, G., Labosky, P., Dobyns, W., Brooks-Kayal, A., Golden, J.A., 2009. Targeted loss of Arx resultsin a developmental epilepsy mouse model and recapitulatesthe human phenotype in heterozygous females. Brain 132,1563—1576.

achatz, C., Fusco, L., Vigevano, F., 2003. Epileptic spasms andpartial seizures as a single ictal event. Epilepsia 44, 693—700.

arkin, J., Beaujean, A., 2012. The effects of Wechsler Intelli-gence scale for children-four edition cognitive abilities on mathachievement. J. School Psychol. 50, 113—128.

icard-Mousnier, B., Dorfmuller, G., Fohlen, M., Jeanguillaume, C.,Nguyen, S., Delalande, O., Bulteau, C., 2012. Late-onset epilep-tic spasms may be cured by focal cortical resective surgery.Epileptic Disord. 14, 313—320.

alanova, V., Andermann, F., Rasmussen, T., Olivier, A., Quesney,L.F., 1995. Parietal lobe epilepsy. Clinical manifestations andoutcome in 82 patients treated surgically between 1929 and1988. Brain 118 (Pt 3), 607—627.

cantlebury, M.H., Galanopoulou, A.S., Chudomelova, L., Raffo, E.,Betancourth, D., Moshe, S.L., 2010. A model of symptomatic

tical involvement in focal epilepsies with epileptic spasms.014.08.008

infantile spasms syndrome. Neurobiol. Dis. 37, 604—612.elisek, L., Chachua, T., Yum, M.S., Poon, K.L., Veliskova, J., 2010.

Model of cryptogenic infantile spasms after prenatal corticoste-roid priming. Epilepsia 51 (Suppl. 3), 145—149.