www.elsevier.com/locate/clinph

Clinical Neurophysiology 118 (2007) 255–261

Complete change of seizure and spike lateralization in temporallobe epilepsy at two separate monitorings

Areti Koukou a,b, Sophie Dupont a,c,d, William Szurhaj e, Michel Baulac a,c,d,Philippe Derambure e, Claude Adam a,d,f,*

a AP-HP, Epileptology Unit, Hopital de La Pitie-Salpetriere, Service de Neurologie 1, Hopital de La Pitie-Salpetriere 47-83 Bd de 1’Hopital 75651,

Paris Cedex 13, Franceb Attica’s Psychiatric Hospital, Neurology, Athens, Greece

c INSERM EMI 0224 ‘Cortex & Epilepsie’, Hopital de La Pitie-Salpetriere, Paris, Franced Universite Pierre et Marie Curie, Paris, France

e Clinical neurophysiology department, R. Salengro Hospital, Lille, Francef CNRS UPR640 ‘Neurosciences Cognitives et Imagerie Cerebrale’, Hopital de La Pitie-Salpetriere, Paris, France

Accepted 3 October 2006Available online 1 December 2006

Abstract

Objective: To report complete change of seizure and spike lateralization over time in bilateral temporal lobe epilepsies (TLE).Methods: Repetition of video-EEG monitorings in 115 patients; 2 cases are reported in detail; 113 other severe partial epilepsies wereincluded to estimate retrospectively the frequency of the reported phenomenon.Results: In 2 cases, two video-EEG monitorings, separated by several months, revealed the first time one unilateral TL (temporal lobe)seizure and spike focus and the second time a distinct seizure and spike focus located in the opposite TL. The second monitoring wasplanned for these two patients because of the presence of a discordant lesion or, in the absence of a lesion, of some bilateral or discordantfunctional (EEG, SPECT and PET) abnormalities. No patient among the other 113 cases had this video-EEG pattern.Conclusions: In TLE, two video-EEG sessions may be necessary to disclose two opposite TL epileptogenic foci.Significance: In rare bilateral TLE cases, the expression of seizure and spike foci can alternate between hemispheres.� 2006 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.

Keywords: Temporal lobe; Epilepsy; Lateralization; Epileptic focus; Video-EEG; Intracranial EEG

1. Introduction

Presurgical evaluation of severe pharmacoresistant par-tial epilepsies often comprises a single prolonged video-EEG session. Together with clinical and neuroimagingdata, information from video-EEG monitoring helps definethe site of the epileptogenic cerebral region. Video-EEGprovides ictal and interictal data which have a diagnosticand sometimes prognostic value in the preoperative explo-ration of temporal lobe epilepsies (TLE) (Blume et al.,

1388-2457/$32.00 � 2006 International Federation of Clinical Neurophysiolo

doi:10.1016/j.clinph.2006.10.005

* Corresponding author. Tel.: +33 1 42 16 18 01; fax: +33 1 44 24 52 47.E-mail address: claude.adam@psl.aphp.fr (C. Adam).

2001; Cendes et al., 2000; Chung et al., 1991; Dohertyand Cole, 2001; Gilliam et al., 1997; Hirsch et al.,1991a,b; Holmes et al., 2000, 1996; Hufnagel et al., 1994;Lieb et al., 1981; Pataraia et al., 1998; Schulz et al., 2000;Sirven et al., 1997; So et al., 1989a,b).

The first step of video-EEG recordings is non-invasive.Bitemporal, independent ictal and interictal epilepticabnormalities on scalp EEGs indicate bilateral hyperexcit-ability and are associated with a less favorable post-opera-tive outcome in TLE (Blume et al., 2001; Chung et al.,1991; Gilliam et al., 1997; Holmes et al., 2000; Hufnagelet al., 1994; Lieb et al., 1981; Schulz et al., 2000; Soet al., 1989b). Pre-operative records normally disclose

gy. Published by Elsevier Ireland Ltd. All rights reserved.

Fig. 1. Sagittal and coronal T1-weighted MRI images of Patient 1showing the right temporal polar dermoid cyst and in particular itsposterior part in relation with medial temporal limbic structures.

256 A. Koukou et al. / Clinical Neurophysiology 118 (2007) 255–261

them. It is recommended that at least 4–6 attacks berecorded to capture all potential seizure foci (Haut et al.,1997; Sirven et al., 1997). However, the temporal distribu-tion of occurrence of seizures can vary. One of two epilep-togenic foci located in opposite temporal lobes (TL) can besilent over a prolonged period. Interictal epileptic activity(IEA) may also be absent for long periods (Holmes andDodrill, 1995). These temporal fluctuations might causeone focus, involving interictal or ictal events or both, tobe missed. We present two cases which illustrate this prob-lem and try to evaluate its incidence from a retrospectivereview of patient charts.

2. Methods, patients, results

2.1. Case reports

Two recent cases are described in which at least two pro-longed, continuous video-EEGs revealed opposite ictal andinterictal lateralizations. No lateralizing clinical signs werepresent during seizure monitoring for either patient. Med-ication was progressively reduced during monitoring ses-sions and the maximal reduction reached is indicated.

Patient 1, a 23-year-old male without abnormal history,first developed seizures at 6 years of age. After a period ofonly nasal sensations and peri-umbilical pains, the attacksevolved to a loss of contact with oro-alimentary automa-tisms after the same initial sensations. A right temporalpolar dermoid cyst was discovered in MRI images(Fig. 1), soon after the beginning of his epilepsy. Five yearsago, a simple puncture of the cyst relieved the patient fromseizures for six months. Afterwards similar seizures reap-peared more and more frequently, sometimes without aura.A first video-EEG monitoring of 4 days at 23 years of age,performed in Lille with reduced medication (50%), revealeda clear EEG expression of five typical complex partial sei-zures (with an habitual nasal aura and an automatic nose-rubbing, lasting about 1 min) and four nasal auras out ofsix (Fig. 2) on the left TL region (two auras had no EEGmanifestation). An interictal spike focus was localised inthe same left TL region (Fig. 2). No abnormality was pres-ent on the right TL. Eight months later, a second continuous

video-EEG was carried out over 8 days in Paris with thesame reduction in medication. Nine complex partial sei-zures (with the same nasal sensation and a descendingwarmth sensation followed by nose-rubbing and chewing,during 1–3 min) were detected in the right TL on EEGand accompanied by a very active spike focus at the samesite (Fig. 3). No other abnormal activity was noticed. Aresection of the dermoid cyst and the surrounding cortex(involving the amygdala and a small part of the hippocam-pal head) was done on tumoral and epileptologic groundsand led to rare residual auras over 6 months (not yet a sig-nificant time period).

Patient 2, a 45-year-old female with no pathologic histo-ry, began to experience seizures at 37 years of age. At first,they consisted of an epigastric sensation of warmth and a

deja-vecu followed by a loss of contact, staring and oro-al-imentary automatisms. Three years later, the sensationswere no longer percieved before rupture of contact, isolat-ed episodes of a deja-vecu appeared and the episodesbecame nocturnal. No abnormalities were evident inMRI-scans. Three video-EEGs were all performed in Paris.During the first one recorded over 9 days with completeremoval of treatment, one nocturnal paucisymptomatic cri-sis (consisting of a single movement while in bed) wasexpressed by the left TL region. Interictal epileptic abnor-malities were bi-temporal, and right-sided dominant. Dur-ing a second video-EEG monitoring (5 days, 2/3 reduction ofmedication), one year later, five paucisymptomatic seizures(sometimes with brief oro-alimentary automatisms; threeduring sleep) but one with a secondary tonic–clonic gener-alisation were expressed initially on the left TL region aswere interictal epileptic abnormalities (Fig. 4). A PETscanshowed a bi-temporal abnormality as did the ictal SIS-COM SPECT. One year later, a video-SEEG with bitempo-ral depth and subdural electrodes (Adam et al., 1996),

Fig. 2. First surface video-EEG recording of patient 1. Interictal epileptic spikes were present only on the left TL (first part of A). Left TL expression of asimple partial seizure:the whole seizure is depicted (beginning in A and finishing in B).

A. Koukou et al. / Clinical Neurophysiology 118 (2007) 255–261 257

Fig. 3. Second surface video-EEG recording of patient 1. Spikes were this time only on the right TL (first part of A) and seizures were exclusively oralmost exclusively right-sided (A: seizure onset; B (13 s later): seizure end).

258 A. Koukou et al. / Clinical Neurophysiology 118 (2007) 255–261

Fig. 4. Second surface video-EEG of Patient 2. TL interictal spiking activity (left side of the figure) was exclusively ipsilateral to the left TL seizure onset(right side). Recording with CPz reference.

Fig. 5. Simultaneous video-SEEG and surface EEG from Patient 2. Seizure originating from right hippocampal structures and expressing on surface EEGon the right TL region. ‘‘AHip’’ electrodes are postero–anterior intracerebral electrodes exploring medio-TL and temporo–occipital structures. The otherelectrodes are sub-dural exploring TL neocortex. Recording with CPz reference. Surface EEG Amplitude calibration (SEEG divided by 5). Abbreviations:AHip, amygdalo–hippocampal; ant, anterior; L, left; mid, middle; pos, posterior; R, right; T, temporal.

A. Koukou et al. / Clinical Neurophysiology 118 (2007) 255–261 259

260 A. Koukou et al. / Clinical Neurophysiology 118 (2007) 255–261

recorded simultaneously with surface EEG, lasted 11 dayswith 3/4 reduction of treatment. Fifteen simple or complexpartial seizures (rarely with an epigastric sensation fol-lowed by some motor and chewing automatisms; five dur-ing sleep) originated from the right hippocampal contacts(Fig. 5) and scalp EEG expression was right temporal.Complex partial seizures recorded with depth electrodeswere a little longer than those (without taking account ofthe secondarily generalized seizure) recorded during sur-face monitoring (1 min 20 s versus 50 s). Right interictalspikes consisted in activities synchronized over a largeamygdalo-hippocampo-neocortical volume, present in thesurface EEG and evident both day and night. Left hippo-campus interictal EEG anomalies were detected only dur-ing sleep and were not visible in the surface EEG. Thesefindings did not support a surgical solution. In the follow-ing discussion, only the two last monitorings areconsidered.

2.2. Retrospective study

To estimate the frequency of the present pattern, wehave retrospectively reviewed charts of Paris patientsrecorded between 1991 and 2005. From 757 of 1018 avail-able charts, we found 113 other patients with two indepen-dent monitoring sessions including at least one seizure persession. They included 75 temporal cases (49 with an MRI-detectable lesion, mostly hippocampal sclerosis), 19 fron-tals, 5 occipitals, 1 parietal and 13 of undetermined focus.No cases were comparable to those described here. Fur-thermore we found no cases with a partial (between 50%and 100%) shift of seizure lateralization. This data yieldsan incidence of 2/115 (1.7%) for a complete change in sei-zure and spike lateralization in this sample of intractablesevere epilepsies.

3. Discussion

We report two patients with TLE who would have beeninterpreted as unilateral, from ictal and interictal events, ata first video-EEG monitoring. However, a second video-EEG session revealed a second epileptic focus located inthe opposite TL while the first one had disappeared. Witha time period of a year or less between the two sessions,and no change in their epilepsy, we suggest these casesshould be interpreted as bilateral TLE with a specialexpression over time and not the result of a secondary epi-leptogenesis occurring between the recordings (Morrell,1989).

Two recent publications have underlined occasional dif-ficulties in identifying a second seizure focus in contralater-al TL. They concluded that recording of multiple seizuresreduces the possibility of error: four (Sirven et al., 1997)to six seizures (Haut et al., 1997) have been suggested.Occasionally more seizures are needed: in one patient 12seizures had to be recorded before the first discordant event(Haut et al., 1997). The same phenomenon can occur for

IEA: 7 days are sometimes required to detect a secondinterictal focus (Holmes and Dodrill, 1995). Our cases goeven further, showing that just one focus can be expressedin a single video-EEG recording session to the exclusion ofa second focus. The second, similar but contralateral focuswas apparent to the exclusion of the first in a secondrecording session that took place several months later.

This temporal dissociation of two opposite independentfoci might result from an insufficiently long monitoring(Haut et al., 1997). Nonetheless, the single patient fromthe 14 cases studied by Haut et al. who needed 13 seizuresfor a second focus to emerge differs from our patients bythe coexistence of independent bilateral seizures at theend of records and the existence of a single interictal focus.In our bilateral TLE cases, both ictal and interictal epilepticactivities were generated unilaterally in all records and didnot coexist with contralateral events, which is a pivotalcharacteristic of our observations. Even during the secondsession, the focus remained strictly unilateral –contralateralto the first one.

In the absence of comparable data in the literature, wecan attempt to exclude alternative explanations for ourobservations. Multiple arguments diminish the probabilityof a false contralateral lateralization of TL seizures (Engeland Crandall, 1983; Mintzer et al., 2004; Sammaritanoet al., 1987). First this phenomenon occurs rarely (5/105),and presents differently: mainly in severe hippocampal scle-rosis, involving all or almost all seizures whereas IEA arefrequently unilateral (mostly on the lesional side) andsometimes bilateral (Mintzer et al., 2004). Second, in ourcases some seizures remained confined to one TL on sur-face EEG with no contralateral propagation (in 7 partialseizures (mostly simple): five in the first session of patient1 (Fig. 2), one in the second session of patient 1 and onein the first session of patient 2) and the other seizures prop-agating contralaterally retained a clear lateralization on theonset side. It is very doubtful that seizures with such clearlydifferent EEG patterns originate from the same region(Foldvary et al., 2001). Third, the concordance of ictaland interictal activities in our patients at a given time rein-forces their mutual localizing value: the presence of unilat-eral IEA increases the ability of ictal recordings to correctlylateralize the epileptogenic TL in 90%–96% of TLEpatients (Steinhoff et al., 1995). Fourth, scalp localizationwas verified by simultaneous intracranial EEG for oneEEG pattern in patient 2 (Fig. 2). Two other factors mightintervene. A major effect of seizure clustering can be dis-missed: no seizure was clustered within eight hours (Hautet al., 1997), except in patient 1 with seven out of nine sei-zures clustered during 26 h in the second monitoring (so 2were not clustered during this session). Medication with-drawal might also have influenced the data, but it is unclearwhy one hemisphere should have been favoured over theother (So, 1991).

Our data concern two patients observed twice in a rela-tively short period. It remains unclear if the alternatingbehavior of the two epileptic foci we observed in each case

A. Koukou et al. / Clinical Neurophysiology 118 (2007) 255–261 261

would continue over a much longer period and there is lit-tle data on mechanisms that might account for such aright/left periodic imbalance of hyperexcitability.

A practical consequence of our findings is that a bilater-al TLE may be mis-interpreted as unilateral from a singlesession and would need a second recording to be diag-nosed. This can be regarded as a sampling error over timein a bilateral TLE. However since it is a very rare event, itmay not be reasonable to routinely monitor every patienttwice. So, when should a second surface session be consid-ered? Our observations may provide some pointers. In TLpatients with no lesion visible in MRI-scans, the decision tomonitor a second time is not straightforward. It dependson an ensemble of data for each patient including history,electro-clinical observations, neuropsychological and func-tional imaging tests which may reveal bilateral or discor-dant abnormalities such as the spike focus in an initialvideo-EEG and the PET and ictal SPECT results in oursecond case. When an intracranial exploration is contem-plated without a second surface monitoring, an apparentlyunilateral surface seizure and spike focus expression shouldnot automatically rule out the use of bilateral recordings.Caution is also needed in patients with a clear discordancebetween lesional site and monitoring results. In those cases,further explorations can search for seizures originatingfrom the lesional side or may disclose underlying mecha-nisms such as a false lateralization. Finally, with only onevideo-EEG monitoring, the presence of unilateral interictaland ictal TL abnormalities, even concordant with a lesion,cannot totally rule out an unnoticed contralateral focus,which may sometimes be the most important one.

In conclusion, we have observed in bilateral TLE a pat-tern of seizure and spike foci alternating between hemi-spheres. This time-dependent pattern can hide one out ofthe two epileptogenic foci in bilateral TLE and may justifymultiple video-EEG sessions in selected patients.

Acknowledgement

We thank Dr R. Miles for suggestions on the text.

References

Adam C, Clemenceau S, Semah F, Hasboun D, Samson S, Aboujaoude N,et al. Variability of presentation in medial temporal lobe epilepsy: astudy of 30 operated cases. Acta Neurol Scand 1996;94:1–11.

Blume WT, Holloway GM, Wiebe S. Temporal epileptogenesis: localizingvalue of scalp and subdural interictal and ictal EEG data. Epilepsia2001;42:508–14.

Cendes F, Li LM, Watson C, Andermann F, Dubeau F, Arnold DL. Isictal recording mandatory in temporal lobe epilepsy? not when theinterictal electroencephalogram and hippocampal atrophy coincide.Arch Neurol 2000;57:497–500.

Chung MY, Walczak TS, Lewis DV, Dawson DV, Radtke R. Temporallobectomy and independent bitemporal interictal activity: what degreeof lateralization is sufficient? Epilepsia 1991;32:195–201.

Doherty CP, Cole AJ. The requirement for ictal EEG recordings prior totemporal lobe epilepsy surgery. Arch Neurol 2001;58:678–80.

Engel Jr J, Crandall PH. Falsely localizing ictal onsets with depth EEGtelemetry during anticonvulsant withdrawal. Epilepsia 1983;24:344–55.

Foldvary N, Klem G, Hammel J, Bingaman W, Najm I, Luders H. Thelocalizing value of ictal EEG in focal epilepsy. Neurology2001;57:2022–8.

Gilliam F, Bowling S, Bilir E, Thomas J, Faught E, Morawetz R, et al.Association of combined MRI, interictal EEG, and ictal EEG resultswith outcome and pathology after temporal lobectomy. Epilepsia1997;38:1315–20.

Haut SR, Legatt AD, O’Dell C, Moshe SL, Shinnar S. Seizurelateralization during EEG monitoring in patients with bilateral foci:the cluster effect. Epilepsia 1997;38:937–40.

Hirsch LJ, Spencer SS, Spencer DD, Williamson PD, Mattson RH.Temporal lobectomy in patients with bitemporal epilepsy defined bydepth electroencephalography. Ann Neurol 1991a;30:347–56.

Hirsch LJ, Spencer SS, Williamson PD, Spencer DD, Mattson RH.Comparison of bitemporal and unitemporal epilepsy defined by depthelectroencephalography. Ann Neurol 1991b;30:340–6.

Holmes MD, Dodrill CB. Recording time required to demonstratebilateral, independent, interictal basal-temporal epileptiform patterns.Epilepsia 1995;36:151.

Holmes MD, Dodrill CB, Wilenski AJ, Ojemann LM, Ojemenn GA.Unilateral focal preponderance of interictal epileptiform discharges asa predictor of seizure origin. Arch Neurol 1996;53:228–32.

Holmes MD, Born DE, Kutsy RL, Wilensky AJ, Ojemann GA, OjemannLM. Outcome after surgery in patients with refractory temporal lobeepilepsy and normal MRI. Seizure 2000;9:407–11.

Hufnagel A, Elger CE, Pels H, Zentner J, Wolf HK, Schramm J, et al.Prognostic significance of ictal and interictal epileptiform activity intemporal lobe epilepsy. Epilepsia 1994;35:1146–53.

Lieb JP, Engel JJ, Gevins A, Crandall PH. Surface and deep EEGcorrelates of surgical outcome in temporal lobe epilepsy. Epilepsia1981;22:515–38.

Mintzer S, Cendes F, Soss J, Andermann F, Engel JJ, Dubeau F, et al.Unilateral hippocampal sclerosis with contralateral temporal scalpictal onset. Epilepsia 2004;45:792–802.

Morrell F. Varieties of human secondary epileptogenesis. J Clin Neuro-physiol 1989;6:227–75.

Pataraia E, Lurger S, Serles W, Lindinger G, Aull S, Leutmezer F, et al.Ictal scalp EEG in unilateral mesial temporal lobe epilepsy. Epilepsia1998;39:608–14.

Sammaritano M, de Lotbiniere A, Andermann F, Olivier A, Gloor P,Quesney LF. False lateralization by surface EEG of seizure onset inpatients with temporal lobe epilepsy and gross focal cerebral lesions.Ann Neurol 1987;21:361–9.

Schulz R, Luders HO, Hoppe M, Tuxhorn I, May T, Ebner A. InterictalEEG and ictal scalp EEG propagation are highly predictive of surgicaloutcome in mesial temporal lobe epilepsy. Epilepsia 2000;41:564–70.

Sirven JI, Liporace JD, French JA, O’Connor MJ, Sperling MR. Seizuresin temporal lobe epilepsy: I. Reliability of scalp/sphenoidal ictalrecording. Neurology 1997;48:1041–6.

So NK. Effect of anticonvlusant withdrawal on spiking and EEG seizureactivity. In: Luders HO, editor. Epilepsy surgery. New York: RavenPress; 1991. p. 355–60.

So N, Gloor P, Quesney LF, Jones-Gotman M, Olivier A, Andermann F.Depth electrode investigations in patients with bitemporal epileptiformabnormalities. Ann Neurol 1989a;25:423–31.

So N, Olivier A, Andermann F, Gloor P, Quesney LF. Results of surgicaltreatment in patients with bitemporal abnormalities. Ann Neurol1989b;25:432–9.

Steinhoff BJ, So NK, Lim S, Luders HO. Ictal scalp EEG in temporal lobeepilepsy with unitemporal versus bitemporal interictal epileptiformdischarges. Neurology 1995;45:889–96.