Insular and insulo-opercular epilepsy in childhood: An SEEG study

Seizure 23 (2014) 300–308

Contents lists available at ScienceDirect

Seizure journal homepage: www.elsevier.com/locate/yseiz

Insular and insulo-opercular epilepsy in childhood: An SEEG study S. Dylgjeri a, D. Taussig a,*, M. Chipaux a, A. Lebas a,b, M. Fohlen a, C. Bulteau a,c,d, J. Ternier a, S. Ferrand-Sorbets a, O. Delalande a, J. Isnard e, G. Dorfmu¨ller a,c,d a

Service de Neurochirurgie Pe´diatrique, Fondation Rothschild, 25-29, rue Manin, 75940 Paris Cedex 19, France Department of Neurophysiology, Rouen University Hospital, 1 rue de Germont, 76031 Rouen, France c Inserm, U663, Paris F-75015, France d University Paris Descartes, 75005 Paris, France e Hospices civils de Lyon, Hoˆpital Neurologique; Service de Neurologie Fonctionnelle et d’Epileptologie, 69677 Bron, France b

A R T I C L E I N F O

A B S T R A C T

Article history: Received 2 July 2013 Received in revised form 7 January 2014 Accepted 8 January 2014

Purpose: In recent years, there have been series analysing the electro-clinical correlations of insular epilepsy in adult populations. In contrast, the ictal semiology in children with insular epilepsy is poorly described. Considering that early and successful surgery may greatly improve the cognitive outcome and quality of life, it is worthwhile to deepen our knowledge of insular epilepsy in children. Methods: We retrospectively evaluated ten children with drug-resistant focal insular epilepsy who had been consecutively explored with stereoelectroencephalography (SEEG), followed by individually tailored resective surgery that included part of the insula in all cases. A detailed anatomo-electro-clinical analysis of non-invasive EEG and SEEG data was performed. At least one of the electrodes explored the insular cortex. SEEG analysis confirmed that the insular cortex was included in the ictal onset zone. Results: Epilepsy onset was mostly during the first year of life, characterized by subtle seizures as well as spasms and myoclonic seizures. Later on, neurovegetative signs and asymmetric tonic and hypermotor seizures (HMS) dominated the ictal semiology. The epileptogenic zone was frequently wider than insular with frontal and central predominance. In eight patients, the tailored resection included a lesion. In seven patients, an Engel class 1 outcome as well as neuropsychological and behavioural improvement was obtained. Conclusions: SEEG is feasible and useful in children with drug-resistant insular epilepsy which is often characterized by autonomic symptoms as the initial symptoms and should be suspected in cases with HMS, asymmetric tonic seizures and even asymmetric spasms. Early propagation is mostly frontal and central. Analysis of a larger population is required to refine these findings. ß 2014 British Epilepsy Association. Published by Elsevier Ltd. All rights reserved.

Keywords: Insula SEEG Focal epilepsy Epilepsy surgery Invasive exploration Child

1. Introduction Although the first description of the insula by Johann Christian Reil dates back to the eighteenth century, ‘‘surgical’’ interest in the insula only arose in the 1940s and 1950s with Guillaume and Mazars1 as well as Penfield and Faulk.2 In the following years, several surgical series were published.3,4 Isnard et al. first pioneered the use of depth electrode recordings within the insular cortex which was considered until then dangerous and, therefore, unfeasible.5 At present, the role of stereo-electroencephalography (SEEG) has been widely demonstrated to be fundamental in defining insular semiology through anatomo-electroclinical correlations.6–9 More-

over, results of electrical stimulations have improved our knowledge about functions of this deeply located lobe.2,10–12 On the other hand, in recent decades, epilepsy surgery in children has demonstrated favourable outcomes in terms of seizure control and improved neuropsychological development.13 For those children, the safety and utility of the SEEG technique has been well-established.14,15 The aim of this retrospective study consists in a detailed analysis of anatomo-electro-clinical correlations as well as surgical results in a small group of children operated on for drug-refractory focal insular epilepsy after SEEG exploration. 2. Patients and methods

* Corresponding author. Tel.: +33 1 48 03 69 43; fax: +33 1 48 03 65 52. E-mail addresses: [email protected], [email protected] (D. Taussig).

Among 93 patients operated on for drug resistant focal epilepsy at the Paediatric Epilepsy Surgery Center of the Rothschild

1059-1311/$ – see front matter ß 2014 British Epilepsy Association. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.seizure.2014.01.008

S. Dylgjeri et al. / Seizure 23 (2014) 300–308

Foundation in Paris between January 2009 and April 2012, 10 patients fulfilled the following criteria: (i) SEEG recording included the insular cortex. (ii) Parts of the insular cortex were within the epileptogenic onset zone. (iii) Patients underwent surgical resection that included parts of the insular cortex. None of the patients had a selective insular resection. All patients underwent a rigorous evaluation performed in three steps as described previously15: (i) Pre-surgical evaluation: detailed history, neurological examination, scalp electroencephalography (EEG) and long-term video-EEG monitoring (VEEG) and stereo-electroencephalography (SEEG), high-resolution MRI, and neuropsychological testing. PET-scan was performed in seven patients. (ii) Tailored resection; pathological classification according to Blu¨mcke et al.16 In these series, we used adjunctive intraoperative electrocorticography in case 7. (iii) Follow-up: seizure outcome according to Engel’s classification,17 neurological examination, EEG, high-resolution MRI and neuropsychological testing. We do not routinely perform early post-operative MRI. During this period of time, 73 patients with non-insular epilepsy were implanted with at least one insular electrode. In cases of temporal, frontal and parietal epilepsies, an exploration with at least one insular electrode is the rule in our centre. No patients had an insular resection without invasive monitoring. In all patients, a long-term 10–20 VEEG associated with polygraphy captors was performed. For each patient, all types of seizures were recorded and recognized by the parents. In three patients, this recording required antiepileptic drug withdrawal. SEEG exploration planning was performed according to noninvasive data and consisted in an extensive covering, including the hypothetical epileptogenic(s) zone(s) and the lesion when it was present. Intracerebral multiple contact electrodes (electrodes DIXI1 or ALCIS1) were placed with the aid of a frameless stereotactic robot-guided system (MedTech1), based on a high-resolution, contrast-enhanced T1-MRI. In order to precisely localize the position of the electrodes, a post-operative three-dimensional CT-scan was merged with the preoperative MRI. For electrical stimulations, the following parameters were used: 50 Hz, pulse width 1 ms, duration 3–5 s, intensity increased gradually from 3 to 7 mA.

301

We (SD) reviewed the VEEG and SEEG data independently of the neurophysiologist involved in the presurgical evaluation (AL, MC or DT). We classified the ictal semiology at seizure onset based on clinical data reported by the parents to the neuropaediatricians according to the ILAE classification.18 Spasms and myoclonic seizures were separately considered from the clear-cut electroclinical pattern of secondarily generalized seizures. We noted the semiological changes over time when they occurred. For the VEEG, we analyzed background activity and epileptic interictal abnormalities, the temporal relationship between the clinical and electrical onset, as well as the localization and propagation of ictal discharge. We analyzed the SEEG recordings according to the SEEG criteria proposed by Kahane et al.19 We paid particular attention to electro-clinical relationships between insular gyri and neighbouring structures. Concerning insular electrical stimulations, we did not take into account functional stimulations, which were not feasible in the majority of patients, and we considered only the evoked seizures with the same electro-clinical pattern as spontaneous seizures. 3. Results 3.1. Demographic and anatomo–clinical data The demographic and anatomo–clinical data of the patients are given in Table 1. There were five males and five females. None of them had a family history of epilepsy, foetal distress, febrile convulsions or head trauma. One child had tuberous sclerosis complex (TSC). One patient had previously undergone lesional surgery when he was 11 months old. At the time of SEEG, two of the patients had not yet been lateralized. Neurological disabilities were present (either isolated or in combination) in five patients, consisting in dyspraxia, dysarthria, left hemiparesis, neglect of the left hand. Only one patient had normal neuropsychological test results. Neurodevelopmental delay was global in seven patients, whereas only two presented a delayed language acquisition. Three patients were considered able to describe subjective manifestations and the sensations evoked by electrical stimulations. Four patients did not speak at all, and three had poor language development rendering them unable to describe feeling sensations. Two patients had a behavioural disorder, both characterized by emotional troubles and aggressiveness. The cerebral MRI showed a lesion in eight patients, suggesting focal cortical dysplasia (FCD) in seven of them (five including the insula, one non-insular; one insular in a patient with TSC), and a tumoral lesion in one patient including the insula. The borders of

Table 1 Demographic and imaging data. Patients

Sex

Previous surgery

Side

Lat

MRI

PETSCAN hypometabolism

Neurol deficit

Neuropsych development

Behavioural disorder

Pt. 1 Pt. 2

F M

No No

R R

RH ND

Lesional (right INS-op) Lesional (INS); TSC

Dyspraxia; dysarthria Hemineglect

LD GD

No No

Pt. 3 Pt. 4 Pt. 5

F F M

No No Yes

L R R

RH ND RH

Cryptogenic Lesional (right INS-op) Lesional (right INS-op)

No No Neglect of LH

LD GD GD

Yes No No

Pt. 6

M

No

L

LH

Lesional (left INS-BG)

Focal (right F) Bilateral; multifocal (right INS-PT, left T) Multifocal (left FO, left INS) No data Multifocal (right INS-F; right INS-T) Multifocal (left INS-FT; left F)

GD

No

Pt. 7 Pt. 8

F F

No No

R R

RH RH

Focal (right prefrontal sulcus) No data

Normal GD

No No

Pt. 9 Pt. 10

M M

No No

R R

RH RH

Cryptogenic Lesional (right INS-op-F and head NC) Lesional (right GFs-GFm sulcus) Lesional (right INS-op-centro-F)

Dyspraxia; right hemiparesis No Left hemiparesis

Focal (right prefrontal areas) No data

No No

Impaired GD

Yes No

Pt, patient; FC, febrile convulsions; Lat, lateralization; R, right; L, left; RH, right-handed; LH, left-handed; ND, not defined; F, frontal lobe; T, temporal lobe; C, central region; P, region parietal lobe; O, occipital lobe; INS, insula; V, vertex; op, operculum; FO, fronto-orbital; GFm; middle frontal gyrus; GFs, superior frontal gyrus; BG, basal ganglia; NC, nucleus caudatum; BCPS, brachio-crural pyramidal syndrome; LD, language delay; GD, global delay; TSC, tuberous sclerosis complex.

S. Dylgjeri et al. / Seizure 23 (2014) 300–308

302 Table 2 Epilepsy history. Patients

Epilepsy onset (in months)

Seizure frequency

Semiology of focal seizures

Spasms and/or myoclonic seizures

Secondary generalized seizures

Falls

Intensive care unit

Pt. 1 Pt. 2

30 1

S/D S/D

Arrest, NVS Pseudo-absence

No No

Yes No

No No

13 18 2 13 36 22 Days 44 2 Days

S/D (morpheic) S/D S/D S/D S/D (morpheic) S/D S/D (morpheic) S/D

NVS; PT-C PT PT S-S Behavioural changes; HMS NVS; PT-C S-S PT-C

No Yes (asymmetric spasms and myoclonic seizures) No Yes (asymmetric spasms) No No No No No No

Yes Yes Yes Yes No Yes Yes No

Yes No No No No No No No

No Yes No Yes No Yes No No

Pt. Pt. Pt. Pt. Pt. Pt. Pt. Pt.

3 4 5 6 7 8 9 10

S/D, several/day seizures; NVS, neurovegetative symptoms; PT-C, focal tonic–clonic seizures; PT, focal tonic seizures; S-S, subtle seizures; HMS, hypermotor seizures; TSC, tuberous sclerosis complex.

the suspected dysplasia were ill-defined. In only one case, the FDGPET scan showed well-localized hypometabolism which was concordant with the cerebral MRI localization, whereas in the six others it was more extended or multifocal.

became richer, particularly with better reporting of subjective components thanks to language development.

3.2. Epilepsy history data

The VEEG recording data are given in Table 3. The patient with TSC was the only one with features of an encephalopathy. The epileptic interictal abnormalities were focal in one patient only, regional multifocal in four, and multifocal in five. They were located in different lobes, except for the occipital lobe which was never implicated. This epileptic interictal activity involved the frontal lobe in eight patients, the central region or the temporal lobe in seven patients, and the parietal lobe in two. Focal seizures were recorded in all patients. Asymmetric spasms occurred before, after, or independently of the focal seizure in two of them. One of those patients also had myoclonic seizures. Two of the patients had asymmetric bilateral ictal discharges; the others had extended ictal discharges including at least two other lobes. An ictal onset with low voltage fast activity was not observed in any of the patients. It was always a rhythmic slow-wave discharge which occurred before the clinical onset in eight patients and before or after the clinical onset in the two remaining ones (see Supplementary Figures 2 and 3). Supplementary material related to this article can be found, in the online version, at http://dx.doi.org/10.1016/j.seizure.2014.01.008.

3.3. VEEG analysis

They are given in Table 2. The mean age at epilepsy onset was 1.3 years (from 2 days to 44 months). All patients had a high seizure frequency (several per day) with diurnal and nocturnal seizures; in three of them, nocturnal prevalence was clear. At epilepsy onset, all patients presented with focal seizures. Among them, two patients had subtle seizures with isolated autonomic features; in three, the ictal features were dominated by neurovegetative symptoms including, either separately or in combination, cardiac and respiratory frequency changes, facial rubefaction or pallor, and mydriasis associated with partial tonic–clonic seizures or arrest (3 patients); three patients had partial tonic seizures or tonic–clonic seizures. Hypermotor seizures (HMS) occurred in one patient and pseudo-absences in another. Two patients also presented with asymmetric spasms; one of them also had myoclonic seizures. Six of the patients had secondarily generalized seizures. Two of the patients fell during the seizures. Three of the patients were admitted to an intensive care unit for seizure recrudescence or epileptic status. Over time, the ictal semiology

Table 3 VEEG recording analysis. Patients

AEDs tapered

Background activity

Interictal spikes

Seizure types

Ictal discharge

Temporal relationships between electrical and clinical onset of seizures (electrical onset before, after or concomitant with clinical onset)

Pt. 1 Pt. 2

Yes No

Norm Abnorm (TSC)

Multifocal; bilateral (right FT; left F) Multifocal (right CPT and Vertex)

No No No No Yes

Norm Norm Norm Norm Norm

Multifocal (left FCPT, left FCT) Multifocal (right FT, right FC) Focal (right FC) Focal (left FT) Focal (right FC and vertex)

No No Yes

Norm Norm Norm

Multifocal (right FCT) Focal (right FT) Focal (right CT)

ST ST ST

Extended (right FCP and V) Asymmetric bilateral (predominant right CPT) Extended (left FCPT) Extended (right FCT) Extended (right FCT) Extended (left FT) Extended; asymmetric bilateral (FC prevalently right) Extended (right FCT) Extended (right FT) Extended (right CT)

Before Before

3 4 5 6 7

ST OT, asymmetric spasms and myoclonic seizures ST OT, asymmetric spasms OT ST ST

Pt. Pt. Pt. Pt. Pt.

Pt. 8 Pt. 9 Pt. 10

Variable Before Before Before Variable Before Before Before

AEDs, antiepileptic drugs; Norm, normal; Abnorm, abnormal; TSC, tuberous sclerosis complex; F, frontal lobe; T, temporal lobe; C, central lobe; P, parietal lobe; O, occipital lobe; INS, insula; V, vertex; OT, one type of focal seizures; ST, several types of focal seizures; NVS, neurovegetative symptoms.

INS-F,C,T INS-F,C,P,T 10.3 2.1 Pt. 9 Pt. 10

INS-F,C,T,P 2.8 Pt. 8

INS-bilateral F,C 13.5 Pt. 7

INS-F,C,T 5.6 Pt. 6

INS-C,F,T,P 10.3 Pt. 5

INS-F,C,T,P 1.6 Pt. 4

Tot elect (cts) N, total electrodes (contacts); INS elect (cts) N, electrodes (contacts) within insula; F, frontal lobe; T, temporal lobe; C, central region; P, parietal lobe; O, occipital lobe; INS, insula; IS, insular sulcus; ISG, insular short gyrus; ILG, insular long gyrus; NVS, neurovegetative symptoms; FT, focal tonic seizure; FC, focal clonic seizure; HMS, hypermotor seizure.

No No No No HMS; FT FC; FT Motor Motor 2 (5) 4 (11) 10 (123) 11 (124)

Both Yes FT NVS 3 (6) 11 (107)

No No FT NVS 3 (5) 11 (134)

No Yes HMS NVS 5 (13) 14 (191)

No No HMS Probable pain, NVS 4 (8) 8 (97)

Spasms Yes FT Motor 4 (12)

Possible fear, NVS 6 (12) 13 (156) 7.6 Pt. 3

INS-C,T,F,P,O

Sup middle and posterior ISG + inf first ILG + all second ILG Sup posterior ISG + all first ILG + inf second ILG Sup posterior ISG + sup central IS + sup second ILG + inf first ILG sup anterior, middle and posterior ISG, all first ILG + inf second ILG Sup anterior and middle ISG + central IS/ sup first ILG Sup anterior and middle ISG + middlesup first ILG Sup anterior and middle ISG Sup anterior and middle ISG, both ILG

13 (162)

No Yes

Yes Yes

HSM Eye deviation, immobility of the right part of the body HMS Abdominal pain, NVS Alteration of consciousness 6 (12) 6 (16) 11 (132) 13 (165) Sup all ISG + all both ILG Sup middle ISG + central IS + both ILG INS-F,C,P,T INS-P,C,F,T 6.9 3.4

Patients

The surgical data, histological results, seizure outcome, and post-operative neurological and developmental data are given in Table 5. Post-operative MRI is provided in Fig. 1. The age of the patients at the time of surgery ranged from 1.7 to 13.9 years (mean 6.6 years). According to the SEEG results, a tailored surgery was performed, concerning the right hemisphere in eight patients and the left in two patients. None of the patients had a selective insular

Table 4.1 SEEG recording analysis.

3.5. Surgical data

Age at SEEG (years)

SEEG planning

INS gyri explored

Tot elect (cts) N

INS elect (cts) N

First clinical feature

Semiology of focal seizure

Eyelid clonic movements during focal seizure

The SEEG recording data are given in Tables 4.1 and 4.2. The mean age at SEEG was 6.4 years (range 1.6–13.5). The number of electrodes for each patient ranged between 8 and 14 (for a total number of contacts between 97 and 191). In each patient, at least two electrodes explored the insula (range between two and six electrodes with five to sixteen contacts). The electrodes explored insulo-fronto-centro-temporo-parietal areas in seven patients (one had a supplementary electrode in the temporo-occipital junction); insulo-fronto-centro-temporal areas (two patients); asymmetric bilateral insulo-fronto-central areas (one patient). In accordance with the SEEG planning and anatomical constraints, the insular gyri explored were: anterior insular short gyrus (six patients); middle insular short gyrus (eight patients); posterior insular short gyrus (five patients); central insular sulcus (two patients); first insular long gyrus (nine patients); second insular long gyrus (seven patients). The location of the insular electrodes is shown in Fig. 1. No complications occurred following the electrode implantation or removal. The SEEG was considered informative in all patients. As mentioned above, at the time of SEEG, only three patients would have been fully able to describe the subjective ictal manifestation, but one of them only had morpheic seizures without any subjective component. The two others as well as another patient had seizures limited to subjective manifestations or which began with an aura. Eight patients had more than one electroclinical seizure type. The first clinical feature was neurovegetative in six patients. The frequent presence of homolateral or bilateral eyelid clonic movements, isolated or in association with other symptoms and signs, attracted our attention (6/10 patients). They could not be correlated to the involvement of a particular insular area. Partial tonic or clonic seizures were recorded in six patients, isolated or in combination. HMS were recorded in five patients. Asymmetric spasms were recorded in four patients; myoclonic seizures in two patients. The lesional zone (LZ) included only insular gyri (three patients); the insula and other cortical areas (four patients); areas outside the insula (one patient; fronto-polar region). No LZ was identified in two patients. The interictal zone (IZ) included only insular gyri (one patient); the insula and other cortical areas (nine patients). It appears in Fig. 1. The ictal onset zone (IOZ) included only insular gyri (four patients); the insula and other cortical areas (five patients). In the remaining patient, there were insular IOZ, lesional (outside the insula) IOZ and sometimes concomitant activation of the insula and of the lesion. The epileptogenic zone (EZ) was not restricted to the insula in any patient. Insular stimulation: electrical stimulations (ES) within the insular cortex were performed in eight of the patients; six of them were considered able to cooperate. The electro–clinical pattern of the spontaneous seizures was reproduced in two patients.

Pt. 1 Pt. 2

Epileptic spasm/ myoclonic seizures

3.4. SEEG analysis

303

Spasms Both

S. Dylgjeri et al. / Seizure 23 (2014) 300–308

304

Table 4.2 SEEG recording analysis. Lesional zone

IZ

IOZ

EZ

Language

INS-ES

Seizures INS-ES

Pt. 1

Sup all ISG + first ILG + opF + opM + opP

Insulo-opercular (posterior ISG + first ILG + opM + opP + opF)

INS-CF (sup all INS + opM + opP + opF + lat GFO + C)

Yes

Yes

No

Pt. 2

Lesion + sup first ILG

Sup all ISG and first ILG + opF + opM + opP + preCrG + postCrG Lesion + sup both ILG

Insular (posterior ISG + both ILG)

No

No

No

Pt. 3

inf both ILG

First ILG, posterior and middle ISG

Insular (inf first ILG + sup posterior and middle ISG)

Yes

Yes

No

Pt. 4

Sup posterior ISG + inf both ILG + opM + opF + lat FO Sup central IS + opP

Insulo-opercular (sup posterior ISG + opM + opF) Insular (sup central IS + inf first ILG)

No

No

No

Pt. 5

Sup posterior ISG + inf both ILG + opM + opF + lat FO Sup central IS + opP

Poor

Yes

No

Pt. 6

None

inf first ILG + sup middle and anterior ISG + RG + NA + Hi

Insular (inf first ILG + sup middle ISG)

Poor

Yes

Yes

Pt. 7

None

sup middle ISG + post-central IS/first ILG + opP

Insulo-opercular (sup middle ISG + post-central IS/first ILG + opP)

Yes

Yes

No

Pt. 8

Sup anterior ISG + middle-sup first ILG

Sup anterior ISG + middle-sup first ILG + GCa + SSMA + paracentral lobule

Insulo-opercular (middle-sup first ILG + sup anterior and middle ISG + opM)

No

Yes

Yes

Pt. 9

Fronto-polar region

Fronto-polar region + anterior and middle ISG

Insular (sup anterior and middle ISG); lesional (GFs-GFm sulcus); both (insular-lesional)

Poor

Yes

No

Pt. 10

Sup posterior ISG + sup first ILG + GFm + preCrG

All ISG + opM + opP + SSMA + preCrG

Insulo-operculo-central (sup middle ISG + opM + GFm)

INS-C (central IS + both ILG + opM + opP + preCrG) INS-CT (both ILG + posterior and middle ISG + opM + opP + NA, Hi, TGs); INS-CF (both ILG + posterior and middle ISG + opM + opP + GCa + GFm) INS-CF (posterior ISG + first ILG + opM + opF + GFm + GFO lateral + GFi) INS-CF (sup central IS + all first ILG + GCa + preCrG) INS-FT (inf both ILG + sup middle ISG + GFO + NA); INS-F (inf both ILG + sup middle ISG + GCa) INS-CF (sup middle ISG + post-central IS/first ILG  SSMA proper + postCrG + preCrG) INS-CF (middle-sup first ILG + sup anterior and middle ISG + opM + GCa); INS-CFT (middle-sup first ILG + sup anterior and middle ISG + opM + GCa + NA + Hi) INS-FC (sup anterior and middle ISG + GFs-GFm sulcus + SSMA + preCrG); INS-F (sup anterior and middle ISG + GFs-GFm sulcus + SSMA) INS-CF (sup middle ISG + opM + GFi + GFm)

No

Yes

No

F, frontal lobe; T, temporal lobe; C, central region; P, parietal lobe; O, occipital lobe; INS, insula; op, operculum; M, motor; P, parietal, F, frontal; FO, fronto-orbital; ISG, insular short gyrus; ILG, insular long gyrus; preCrG, precentral gyrus; postCrG, postcentral gyrus; RG, rectus gyrus; GCa, gyrus cinguli anterior; SSMA, sensorimotor supplementary area; GFm; middle frontal gyrus; GFs, superior frontal gyrus; GFi, inferior frontal gyrus; NA, nucleus amygdale; Hi, hippocampus; GTs, superior temporal gyrus; sup, superior part of. . .; inf, inferior part of. . .

S. Dylgjeri et al. / Seizure 23 (2014) 300–308

Patients

S. Dylgjeri et al. / Seizure 23 (2014) 300–308

305

Fig. 1. Anatomo-electro-clinical correlations. For each patient, the entry point and target of the electrodes exploring the insula are shown using sagittal MRI (picture on the left and in the centre). We outlined the lesion, when ill-defined on the MRI (particularly in young patients), with a dark dotted line. In patient 6, notice the oblique electrode. In blue is the irritative zone. In patient 6, the irritative zone was widespread with multifocal spike-waves recorded on all electrodes. We outlined the area of continuous spikewaves. The picture on the right is the post-operative MRI showing the insular resection. MRI was performed at least three months post-operatively. Due to post-operative modifications, the resection sometimes looks smaller than it was. Notice in patients 3 and 5 the insufficient insular resection.

resection, notably because of surgical constraints. The resective surgery was: insulo-opercular in three patients; insulo-operculofrontal in three patients; insulo-operculo-central in two patients; insulo-frontal in one patient; insulo-temporal in one patient. The resection was wider than the lesion seen on the MRI in all patients but one (patient 1). Histological analyses revealed focal cortical dysplasia (FCD) in the majority of the patients (FCD 2B – six patients; FCD 1B – one patient; FCD 1A – one patient), a dysembryoplastic neuroepithelial tumour in one patient and TSC in the remaining one. Two patients suffered from a post-operative neurological deficit due to a surgery-related vascular lesion. Both of them had hemiparesis with further improvement. Patient 7 presently has a mild weakness of the hand whilst Patient 8 has spastic hemiparesis. Patient 2 had an exacerbation of hemineglect of the body which resolved itself a few months after surgery. The follow-up ranged from 8 to 47 months (mean 18 months). According to Engel’s classification, seizure outcome resulted in: class I – seven patients; IV – two patients; the patient with TSC was classed as Engel’s III with post-operative non-disabling seizures which were different from the preoperative ones, probably originating from another cortical area. In both patients who showed improvement but had persistent seizures, the postoperative MRI demonstrated that the insular resection was more limited than planned. Upon conducting a neuropsychological follow-up examination, all patients demonstrated significant improvement of attention

and concentration which was immediately noticeable after surgery. Due to the small size and the heterogeneity of the population in terms of age and development it is difficult to summarize the different tests. Behavioural disorders in two patients improved as well. Currently six children are attending a regular schooling system and four a specialized education system. Supplementary Figure 1 presents an illustrative case. Supplementary material related to this article can be found, in the online version, at http://dx.doi.org/10.1016/j.seizure.2014.01.008. 4. Discussion In the literature, all surgical series dealing with insular epilepsy surgery are mainly comprised of adult populations. In children, only a few case reports have been published. Levitt et al. reported the first invasive exploration performed in a two-year-old girl with insular epilepsy.20 We are presenting the first series of children with insular epilepsy confirmed by SEEG, as well as by the excellent outcome after tailored cortical resection including the insular cortex. Indeed, in our population, seven patients belong to Engel’s class I. In both patients showing improvement but with persistent seizures, the post-operative MRI demonstrated that the insular resection was more limited than planned leading us to believe that the unsatisfactory outcome was probably not due to misinterpretation or limitations of the SEEG data. We chose to analyze the

No modification Improv I 12 No

3.6

7.9

1.7

9.9

6.1

13.9

3.2

10.3

2.9

Pt. 2

Pt. 3

Pt. 4

Pt. 5

Pt. 6

Pt. 7

Pt. 8

Pt. 9

Pt. 10

F, frontal lobe; T, temporal lobe; C, central region; P, parietal lobe; O, occipital lobe; INS, insula; op, operculum; M, motor; P, parietal, F, frontal; FO, fronto-orbital; ISG, insular short gyrus; ILG, insular long gyrus; preCrG, precentral gyrus; postCrG, postcentral gyrus; RG, rectus gyrus; GCa, gyrus cinguli anterior; SSMA, sensorimotor supplementary area; GFm; middle frontal gyrus; GFs, superior frontal gyrus; GFi, inferior frontal gyrus; NA, nucleus amygdale; Hi, hippocampus; GTs, superior temporal gyrus; sup, superior; inf, inferior; Lx, lesioectomy; Cx, cortectomy; FU-follow-up; Neurol def, neurological deficit, Improv, improvement.

FCD2B

FCD2B

FCD1B

Improv Improv I 47

No modification Improv I 28

No modification Improv I 10

Yes (left hemiparesis) Yes (left hemiparesis) No FCD2B

No modification Improv I 8 No FCD2B

No modification Improv IV 11 No FCD2B

No modification Improv I 18 No

18 No FCD2A

FCD2B

Improv Improv

No modification Improv

Different seizures than pre-surgery IV 9 Yes TSC

Improv I No 7.2 Pt. 1

Insulo-operculo-frontal Lx–Cx (sup INS + opF + opM) Insulo-opercular Lx–Cx (central IS + both ILG + opM + opP) Insulo-temporal Cx (both ILG + posterior and middle ISG + NA, Hi, TGs) Insulo-operculo-frontal Lx–Cx (all ISG + opM + opF + GFm) Insulo-operculo-central Lx–Cx (central IS + first ILG + opM + lesional part of preCrG) Insulo-frontal Lx–Cx (all short I SG + first ILG + FO) Insulo-opercular Cx (sup middle ISG + central IS/first ILG + opP) Insulo-opercular Lx–Cx (all ISG + first ILG + opF + opM) Insulo-operculo-frontal Lx–Cx (anterior and middle ISG + opF + GFi + GFm + pathological GFm-GFs sulcus) Insulo-operculo-central Lx–Cx (all ISG + opM + lesional part of preCrG)

DNT

15

Neuropsych outcome Outcome (Engel class) FU (in months) Post-surg neurol def Pathology Surgical resection (anatomical limits) Age at surgery (years) Patients

Table 5 Surgical data.

No modification

S. Dylgjeri et al. / Seizure 23 (2014) 300–308

Behavioural outcome

306

cases explored during a limited time period (January 2009 to April 2012) in order to ensure the homogeneity of the implantation. We are conscious of the fact that this choice has the disadvantage of a short-lasting outcome. Moreover, we are aware of the heterogeneity of the population in relation to age at SEEG (ranging from 20 months to 13 years). Unfortunately the huge developmental differences do not allow to give clear-cut data about neuropsychological presurgical pattern or post-operative evolution from that point of view. In our series, all patients suffered from severe epilepsy. Indeed, the mean age at seizure onset was 1.3 years and within the first 2 months of life for four patients. Despite multi-drug treatment, all children had a high seizure frequency (several per day) and three of them were admitted to intensive care units for clusters of seizures or status epilepticus. Nine of the ten patients had a developmental delay, as documented by neuropsychological evaluation. All these features led us to consider a surgical approach as soon as possible despite the complexity of electroclinical correlations and the wellknown risk of post-operative neurological deficit. Ictal features are in most cases of little help in determining the onset area in focal epilepsy in children. MRI can provide an orientation, completed with VEEG supplementary arguments. During the VEEG, only one of our patients had a focal spike focus over the frontal lobe, whereas the other ones had multifocal abnormalities, regional or in remote areas, as described in the literature. Several authors showed that the ictal discharge in insular seizures could mimic frontal, temporal, central or parietal epilepsy, misleading the diagnosis.5,8,20 The ictal discharge in our patients always included at least two lobes. The clinical semiology oriented to the frontal lobe; interictal spikes and the ictal discharge included the frontal lobe in 80% of cases. In none of our patients were the abnormalities exclusively temporal. The rhythmic slowwave ictal discharge oriented to a depthly located origin of the focal seizures. In all patients, an invasive exploration was mandatory to establish the insular onset of the seizures. The rationale for the invasive exploration could have been: absence of lesions (2 patients), an ill-defined lesion (4 patients), a clear-cut lesion with poor electroclinical correlations (4 patients). Due to the location of the insula, depth electrodes had to be used. SEEG is an appropriate method because it allows multiple trajectories, covering the insular cortex. A combination of subdural and depth electrodes has also been used by several groups, but we are aware of an elevated risk of complications due to the technical limitations.21 For depth electrode implantation, we preferred the transopercular approach initially described by Isnard et al. since in adults, the operculum is always involved in insular seizures.9 A single patient had an oblique approach as described by Afif et al.22 In all of our patients, no SEEG-related complications occurred. SEEG allowed the ictal semiology of insular seizures to be correlated with the anatomical data. However, due to vascular constraints, the superior part of the middle insular short gyrus and the two long gyri are areas which are more often explored, whereas the inferior part of the insula is often impossible to reach. We compared the features found in our own paediatric population to those described in adults in recent years. From the literature, we took into account the ictal semiology already reported and the results of electrical stimulations of the insular cortex in nonepileptic subjects. Because of our patients’ young age, developmental delay, and poor development of language, one cannot expect to reveal the subjective manifestations which are most specific for insular epilepsy. None of the patients had the typical sequence described by Isnard et al. which consists in the sensation of laryngeal constriction and paresthesias, often unpleasant, affecting large cutaneous territories and occurring during full consciousness,

S. Dylgjeri et al. / Seizure 23 (2014) 300–308

followed by a complex partial seizure.6 In total, three patients described subjective manifestations. Two patients had probable pain, one abdominal (P1) and one in the throat (P5), associated with neurovegetative symptoms, and another patient presented signs of fear associated with perception of tachycardia (P3). Those sensations could be isolated as simple partial seizures or occur before a longer seizure. Patient 1’s abdominal sensation could be related to visceromotor phenomena elicited by stimulation within the inferior and anterior portions of the insula.2 Patient 5’s painful sensations resemble the one described by Isnard et al. in one of their patients.23 The epileptogenic zone in our patients included the posterior insula. Isnard et al. argued that the posterior insula may play a leading role in the triggering of the pain matrix network and in the resulting emergence of subjective pain experience. Six patients had neurovegetative features at onset. The correlating IOZ consistently included the first insular long gyrus, associated with the posterior short gyrus in three patients and the middle short gyrus in the other three. Ostrowsky et al.,10 and Afif et al.,11 evoked viscerosensory features by stimulating the anterior insula, whereas Stephani et al. elicited viscerosensory symptoms by stimulating the first insular long gyrus.12 The objective part of the seizure could be a tonic or clonic unilateral seizure (six patients). The semiology suggested supplementary motor area (SMA) seizures. In three of them, the ictal discharge almost immediately involved the SMA, which was not the case in a fourth patient. In the two others, the SMA was not properly explored. Five of our patients had HMS characterized by both horizontal movements of the trunk whilst lying down or seated on the bed and dystonic controlateral arm posturing. Several authors described this semiology in patients with an insular epilepsy that had started during childhood.7,9,24 It is noteworthy that all our patients presented an asymmetric pedalling involving the leg ipsilateral to the epileptic discharge whilst the contralateral side was dystonic. In the three cases analyzed by Ryvlin et al., the seizures originated in the antero-superior part of the insula with spreading to the supplementary sensorimotor area (SSMA) and the anterior cingulate gyrus.7 In our patients, as well as in those described by Proserpio et al.,9 the seizures did not always originate from the anterior insula; the fast spreading to the anterior cingulum was, however, consistently observed. The occasionally occurring daily auras described by Ryvlin et al. were observed in three of our five patients. We noted in six patients ipsilateral or bilateral clonic movements of the eyelids, isolated or in association with other symptoms/signs. We do not have sufficient electrophysiological arguments proving that this phenomenon has a ‘‘motor’’ origin. Five of the six patients had an electrode which was anatomically located in the vicinity of the frontal eyefield. However, we did not perform electrical stimulation to confirm this localization. No particular involvement of this area was seen in those five patients. Moreover, in our experience, stimulation of the frontal eyefield does not evoke this kind of blinking. One could hypothesize that an ‘‘autonomic’’ origin would explain this phenomenon, related to the autonomic component of the third cranial nerve. We could not correlate this feature with a specific onset zone within the insula. As far as we know, no data exist in the literature about this semiology. A peculiar type of seizure during childhood is asymmetric spasms and myoclonic seizures with a focal cortical origin and which can be effectively cured by focal surgery.25–27 During SEEG, epileptic spasms were recorded in four patients and myoclonic seizures in two of them. There was no constant implication of any cortical structure in any of the patients. Those spasms associated with focal epilepsies were always related to bursts of focal activity located in the area of the interictal

307

sharp waves progressively spreading to other areas as already described (personal data28) and not specifically to SSMA. We are aware that not all the patients have pure insular epilepsy, but the insula was always a part of the IOZ, and in all the patients, ictal semiology could be related to a network involving the insula. The IOZ was equally localized within the insula and in the insulo-opercular cortex. The semiology in the four patients in whom the ictal onset zone was purely insular does not appear to be different from the semiology of the six others, but the small size of the group does not provide conclusive results. One patient had two IOZs, one localized within a fronto-polar lesion and the other within the insula with a fast spreading of ictal discharge between the two areas. In our population, we have a clear prevalence of fronto-central propagation (8/10 patients) that could be linked to brain maturation in young patients. Analysis of a larger population is necessary in order to confirm those results and to provide reliable neuropsychological data. Signal analysis could contribute greatly in correlating the clinical semiology to the network organization in these patients. 5. Conclusions To our knowledge, we are presenting the first surgical series which analyses the anatomo-electro-clinical correlations in a child population operated on for insular epilepsy following SEEG exploration. This study allows us to evidence several features: First: insular epilepsy is severe focal epilepsy characterized by an early onset and with a significant impact on neuropsychological development in children. Second: the absent or insufficient description of the subjective components of the seizures makes the diagnosis of insular epilepsy difficult, especially at time of epilepsy onset, due to absence of language. When faced with young patients who have focal epilepsy, clinical features such as a strange reaction, arrest or suffering, and neurovegetative symptoms should suggest an insular onset. Third: the most frequent differential diagnosis of an insular seizure onset in children is frontal lobe epilepsy. Fourth: non-invasive assessments are mandatory but insufficient to lead straight to surgery. Multifocal or bilateral inter-ictal spikes in VEEG involving the frontal lobe are suggestive of insular onset; asymmetric spasms or myoclonic movements are associated with focal seizures and bi- or multi-lobar rhythmic slow-wave ictal discharge. Fifth: SEEG is a safe and reliable presurgical method in this population. Sixth: a tailored resection based on a SEEG recording can lead to an excellent seizure outcome and neuropsychological improvement in children with insular epilepsy. The 20% post-operative permanent neurological deficit has to be balanced with the spontaneous evolution of this epilepsy. Obviously there is a ‘‘learning curve’’ for the surgeon operating in this area, as patients operated on more recently by the same surgical team have not experienced those deficits. Conflicts of interest None of the authors has any conflicts of interest to disclose. We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines. Acknowledgements We are indebted to the EEG technologists Mrs. M. Benghezal and Mrs. I. Cheramy. We thank the physicians who referred the

308

S. Dylgjeri et al. / Seizure 23 (2014) 300–308

patients to epilepsy surgery. We thank Mrs. Aisha YU, native English speaker, for careful revisions of the language. References 1. Guillaume M, Mazars G. Technique de re´section de l’insula dans les e´pilepsies insulaires. Rev Neurol 1949;81:900–3. 2. Penfield W, Faulk M. The insula. Brain 1955;78:445–70. 3. Duffau H, Capelle L, Lopes M, Bitar A, Sichez J-P, van Effenterre R. Medically intractable epilepsy from insular low-grade gliomas: improvement after an extended lesionectomy. Acta Neurochir 2002;144:563–72. 4. Von Lehe M, Wellmer J, Urbach H, Schramm J, Elger C, Clusmann H. Insular lesionectomy for refractory epilepsy: management and outcome. Brain 2009;132:1048–56. 5. Isnard J, Gue´not M, Ostrowsky K, Sindou M, Mauguie`re F. The role of insular cortex in temporal lobe epilepsy. Ann Neurol 2000;48:614–23. 6. Isnard J, Guenot M, Sindou M, Mauguiere F. Clinical manifestations of insular lobe seizures: a stereo-electroencephalographic study. Epilepsia 2004;45:1079– 90. 7. Ryvlin P, Minotti L, Demarquay G, Hirsch E, Arzimanoglou A, Hoffman D, et al. Nocturnal hypermotor seizures, suggesting frontal lobe epilepsy, can originate in the insula. Epilepsia 2006;47:755–65. 8. Nguyen DK, Nguyen DB, Malak R, Leroux JM, Carmant L, Saint-Hilaire JM, et al. Revisiting the role of the insula in refractory partial epilepsy. Epilepsia 2009;50: 510–20. 9. Proserpio P, Cossu M, Francione S, Tassi L, Mai R, Didato G, et al. Insularopercular seizures manifesting with sleep-related paroxysmal motor behaviors: a stereo-EEG study. Epilepsia 2011;52:1781–91. 10. Ostrowsky K, Isnard J, Ryvlin P, Guenot M, Fischer C, Mauguiere F. Functional mapping of the insular cortex: clinical implication in temporal lobe epilepsy. Epilepsia 2000;41:681–6. 11. Afif A, Minotti L, Kahane P, Hoffmann D. Anatomofunctional organization of the insular cortex: a study using intracerebral stimulation in epileptic patients. Epilepsia 2010;51:2305–15. 12. Stephani C, Vaca F, Maciunas R, Koubeissi M, Lu¨ders HO. Functional neuroanatomy of the insular lobe. Brain Struct Funct 2011;216:137–49. 13. Cross J, Jayakar P, Nordli D, Delalande O, Duchowny M, Wieser H, et al. Proposed criteria for referral and evaluation of children for epilepsy surgery: recommendations of the Subcommission for Pediatric Epilepsy Surgery. Epilepsia 2006;47:952–9.

14. Cossu M, Cardinale F, Colombo N, Mai R, Nobili L, Sartori I, et al. Stereoelectroencephalography in the presurgical evaluation of children with drug-resistant focal epilepsy. J Neurosurg 2005;103(4 Suppl):333–43. 15. Taussig D, Dorfmu¨ller G, Fohlen M, Jalin C, Bulteau C, Ferrand-Sorbets S, et al. Invasive explorations in children younger than 3years. Seizure 2012;21:631–8. 16. Blu¨mcke I, Thom M, Aronica E, Armstrong D, Vinters H, Palmini A, et al. The clinico-pathological spectrum of focal cortical dysplasias: a consensus classification proposed by an ad hoc Task Force of the ILAE Diagnostic Methods Commission. Epilepsia 2011;52:158–74. 17. Engel JJ, Van Ness P, Rasmussen T, Ojemann L. Outcome with respect to epileptic seizures. In: Engel JJ, editor. Surgical treatment of the epilepsies. New York: Raven Press; 1993. p. 609–21. 18. Berg A, Berkovic S, Brodie M, Buchhalter J, Cross J, van Emde Boas. et al. Revised terminology and concepts for organization of seizures and epilepsies: report of the ILAE commission on classification and terminology, 2005–2009. Epilepsia 2010;51:676–85. 19. Kahane P, Landre E, Minotti L, Francione S, Ryvlin P. The Bancaud and Talairach view on the epileptogenic zone: a working hypothesis. Epileptic Disord 2006;8(Suppl. 2):S16–26. 20. Levitt M, Ojemann J, Kuratani J. Insular epilepsy masquerading as multifocal cortical epilepsy as proven by depth electrode. J Neurosurg Pediatr 2010;5:365– 7. 21. Desai A, Bekelis K, Darcey TM, Roberts DW. Surgical techniques for investigating the role of the insula in epilepsy: a review. Neurosurg Focus 2012;32:E6. 22. Afif A, Chabardes S, Minotti L, Kahane P, Hoffmann D. Safety and usefulness of insular depth electrodes implanted via an oblique approach in patients with epilepsy. Neurosurgery 2008;62:ONS471–8. 23. Isnard J, Magnin M, Jung J, Mauguiere F, Garcia-Larrea L. Does the insula tell our brain that we are in pain? Pain 2011;152:946–51. 24. Dobesberger J, Ortler M, Unterberger I, Walser G, Falkenstetter T, Bodner T, et al. Successful surgical treatment of insular epilepsy with nocturnal hypermotor seizures. Epilepsia 2008;49:159–62. 25. Asano E, Juhasz C, Shah A, Muzik O, Chugani D, Shah J, et al. Origin and propagation of epileptic spasms delineated on electrocorticography. Epilepsia 2005;46:1086–97. 26. Akiyama T, Donner E, Go C, Ochi A, Snead 3rd C, Rutka J, et al. Focal-onset myoclonic seizures and secondary bilateral synchrony. Epilepsy Res 2011;95: 168–72. 27. Moseley B, Nickels K, Wirrell E. Surgical outcomes for intractable epilepsy in children with epileptic spasms. J Child Neurol 2012;27:713–20. 28. Cossu M, Cardinale F, Castana L, Nobili L, Sartori I, Lo Russo G. Stereo-EEG in children. Child Nerv Syst 2006;22:766–78.