John Libbey Eurotext

Towards a definition of the “practical” epileptogenic zone: a case of epilepsy with dual pathology Volume 8, supplément 2, Supplement 2, September 2006

Auteur(s) : Serge Chassagnon1, Maria Paola Valenti1, Cécile Sabourdy1, Philippe Esposito1, Pierre Kehrli1, Alexis Arzimanoglou2,5, Philippe Ryvlin3,5, Philippe Kahane4,5, Edouard Hirsch1,5

1Departments of Neurology and of Neurosurgery, Hôpitaux Universitaires de Strasbourg
2University Hospital Robert-Debré, AP-HP, Paris
3Department of Functional Neurology and Epileptology, Neurological University Hospital, Hospices Civils de Lyon
4Epilepsy Unit, Neurology Dept. Grenoble University Hospital, Grenoble
5CTRS-IDEE, Hospices Civils de Lyon, France

In order to define the “practical actual” epileptogenic zone at each step of the pre-surgical evaluation (Lüders et al. 2006), a hypothesis has to be made based upon the anamnestic ictal phenomenology, clinical ictal phenomenology, scalp interictal and ictal video-EEG recordings, anatomical and functional brain neuroimaging findings. The hypothesis raised at each step may have to be modified on the basis of information provided by the next step of the investigation. Thus, the clinical-electroencephalographic-anatomical-functional correlations lead the clinician to construct a hypothesis based upon the “theoretical” and “practical” epileptogenic zone. In some cases, such a hypothesis needs to be confirmed by depth recordings; the stereo-EEG method is based on the tailored implantation of depth electrodes (see Kahane et al. 2006). During the pre-surgical evaluation, the following five areas (Carreño and Lüders 2001) are measured using different diagnostic techniques: the irritative zone (“area of cortex which generates interictal activities”); the seizure-onset zone (“area of cortex that initiates clinical seizures”); the symptomatogenic zone (“area of cortex which, when activated, produces the initial ictal symptoms or signs”); the epileptogenic lesion (“macroscopic lesion which is causes the epileptic seizures by secondary hyperexcitability of adjacent cortex”); the functional deficit zone (“dysfunctional area of cortex during the interictal period, producing neuropsychological deficits”). The team in charge of the pre-surgical evaluation uses this information to suggest the location and extent of the practical, epileptogenic zone. We report on a case with dual pathology, illustrating this step-by-step procedure of the pre-surgical evaluation.

Anamnesis and seizure phenomenology

Marceline is a 34-year-old, right-handed, female patient. There is no family history of epilepsy, or any personal history of febrile seizures. At the age of three years, she presented an acute, symptomatic seizure after a severe head trauma. She then remained seizure-free for more than 20 years. At 27 years of age, she started experiencing episodes described as an epigastric aura, a sensation of fear, polypnea, right-hand automatisms, verbal automatisms, urge to urinate, without loss of contact. There was no post-ictal aphasia, no prolonged confusion, but amnesia of the seizure event was frequent. Despite several antiepileptic drug trials (carbamazepine, oxcarbazepine, levetiracetam, topiramate, lamotrigine), on monotherapy or in combination, the frequency of the seizures varied from four to 10 per month.

What can we learn at this stage from the analysis of the symptomatogenic zone?

Can we formulate an hypothesis on the “practical” epileptogenic zone?

Symptomatogenic zone: initial onset and propagation signs

Epigastric aura and fear, although rarely reported in extra-temporal epilepsy, they are frequently associated with a symptomatogenic zone in the hippocampo-amygdalo-insular network (Fried et al. 1995, Kotagal et al. 1995, Maillard et al. 2004).

Polypnea can also be related to the hippocampo-amygdalo-insular network, particularly the insular cortex (Isnard et al. 2000, 2004; Ostrowsky 2000).

Right-hand automatisms suggest a symptomatogenic zone in the left hemisphere and can be related to an ictal imbalance between temporal and frontal lobe (Maillard et al. 2004).

Urge to urinate suggests a symptomatogenic zone in the non-dominant hemisphere, with an involvement of temporal or fronto-temporal regions (Baumgartner et al. 2000, Loddenkemper et al. 2003).

Absence of loss of contact with verbal automatisms is not specific to any lobe, but it probably suggests the absence of extended propagation towards other cortical or sub-cortical structures. (Gloor 1986, Lux et al. 2002, Blumenfeld and Taylor, 2003, Maillard et al. 2004).

Absence of post-ictal aphasia. Early post-ictal neuropsychological examination and modalities of recovery can provide valuable information regarding the eventual functional deficit related to the seizure onset zone and propagation. Post-ictal aphasia is usually observed following focal seizures primarily or secondarily involving the dominant hemisphere. The absence of post-ictal aphasia, although not specific, might point to an epileptogenic zone in the non-dominant hemisphere (Saint-Hilaire and Lee 2000, Maillard et al. 2004).

Absence of prolonged confusion and amnesia of seizure events. Temporal lobe seizures frequently spread to the contra-lateral hemisphere and are usually associated either with post-ictal, prolonged confusion or transient specific amnesia or isolated amnesia of part or all the seizure, even in the absence of loss of contact or consciousness during the event. It can be explained by a bilateral activation or inhibition of the Papez circuit (Gloor, 1986, Lux et al. 2002, Blumenfeld and Taylor, 2003).

Hypothesis on the “practical” epiletogenic zone

On the basis of all the above arguments, the actual seizure-onset zone could be the right temporo-mesial region, with a spread to the ipsilateral insula and temporal neocortex, and to the contra-lateral hemisphere.

Video ictal phenomenology

Four informative seizures were recorded, which were very similar to those previously described by the patient and her family ( (figure 1) ). The additional observation made during review of videos was that the hand automatisms were bilateral, and in some seizures oral automatisms were present. No focal tonic or dystonic movement was observed. There was neither post-ictal confusion nor aphasia, but we were able to confirm a partial amnesia of the episode.

Our initial hypothesis on the “practical” epiletogenic zone is not modified following video-EEG recording of ictal phenomenology.

Inter ictal EEG

Inter-ictal EEG (figures 2 and 3) showed left temporal spikes during awakening and sleep, and bilateral sharp waves.

What additional information do we have following analysis of the irritative zone?

Do we have to modify our initial hypothesis on the “practical” epiletogenic zone?

The irritative zone clearly involved the left temporal lobe with left temporal spikes. This was not in total contradiction with our initial hypothesis (right temporal lobe seizure spreading to the left). However, it was surprising not to find only sharp waves in the right temporal region, which was presumed to be the seizure-onset zone.

Ictal EEG

See ( figure 4 ).

What additional data did we obtain from the analysis of ictal EEG ?

Although not definitive, initial depression of the EEG activity over the right temporal lobe (TP8, T4) with concomitant tachycardia was compatible with the hypothesis of a seizure-onset zone in the temporo-mesial structures. Bilateral theta activity predominating on the right hemisphere was also in accordance with the hypothesis of a right neocortical propagation followed by propagation to the left hemisphere. Consequently, at this step, our hypothesis on the “practical” epileptogenic zone is still valid.

Structural neuroimaging MRI (epileptogenic lesion)

There was a clear, dual pathology with right hippocampal sclerosis and a right posterior, possibly post-traumatic, scar. These new elements make it difficult to known if the right posterior, post-traumatic scar could participate in the actual seizure-onset zone or in the potential seizure-onset zone ( (figure 5) ).

Inter ictal and ictal SPECT (Functional deficit zone and actual seizure-onset zone)

Analysis of the interictal SPECT showed hypoperfusion predominantly in the right temporo-polar and temporo-mesial regions with less striking findings in the posterior temporal neocortical region ( (figure 6) ).

Ictal SPECT showed a normalization of the previously hypoperfused regions. These data are in accordance with our previous hypothesis on the prominent role of the right temporal lobe

FDG and flumazenil PET (Functional deficit zone)

The FDG PET showed a right temporo-mesial hypometabolism involving the temporal pole, amygdala and right hippocampus. Flumazenil PET shows a more limited diminution of Flumazenil fixation to the right hippocampus. These data, showing a functional deficit zone limited to right temporo-mesial structures, suggest that the actual seizure onset could be limited to right mesio-temporal structures but the potential epileptogenic zone could be larger ( (figure 7) ).

Neuropsychological evaluation and MRI spectroscopy of the temporal posterior lesion (functional deficit zone)

Right-handed patient; verbal IQ 80; visual IQ 96; global IQ 85; Weschler: verbal memory 91; visual memory 102.

MRI spectroscopy failed to demonstrate a specific abnormality in the region of the right posterior lesion.

There was a difference between this right-handed patient verbal IQ and visual IQ. However, the Weschler test showed no major memory deficit. Consequently, we could exclude that the discrepancy between her verbal and visual IQ was related to a functional deficit of the left temporal lobe.

Conclusions of the non-invasive phase

Hypothesis on the “practical” epiletogenic zone: from all the above arguments, the actual seizure onset could be the right temporo-mesial region, with a spread to the insula, temporal neocortex and the contra-lateral temporal lobe.

Two possibilities were discussed:

  • A right temporal lobectomy or hippocampo-amygdalectomy without further investigations;
  • An invasive stereo-EEG recording in order to distinguish or differentiate the actual practical epileptogenic zone from the potential epileptogenic zone that might also involve: the right insula, right fronto-orbital region, right posterior temporal lesion, or even the left temporal lobe.

Because of the discrepancies already discussed, it was decided to perform stereo-EEG monitoring, with a tailored implantation of depth electrodes. The aim of the stereo-EEG recording was to better define the localization of the interictal EEG abnormalities, the seizure-onset zone and the early propagation zone ( (figure 8) ).

Interictal stereo-EEG data (irritative zone)

Interictal EEG abnormalities, spikes, poly-spikes were recorded in the right temporal pole, amygdala, anterior and posterior part of the hippocampus and para-hippocampal gyrus. Rare spikes were recorded over the left hippocampus. No SEEG abnormalities were recorded over the insula, occipto-temporal junction, fronto-orbital region.

Ictal EEG data (actual seizure-onset zone and early propagation zone)

The seizure-onset zone involved, the amygdala, anterior hippocampus, right temporal pole and the anterior part of the temporal neocortex. Within 30-40 seconds, there was propagation to the fronto-orbital region and left hippocampus. Fast activity of low voltage, being probably indicative of the “actual” seizure-onset, was only seen in the amygdala-hippocampus-temporal pole complex, while the insula and the lesion (the occipito-temporal junction) were secondarily involved (60 seconds later), mostly on a slow mode of discharge.


Stereo-EEG recordings allowed to further investigate the “potential epileptogenic zone” during pre-surgical evaluation (Carreño and Lüders 2001). The “actual epileptogenic zone” was considered to be limited to the right anterior and mesial temporal lobe (pole, hippocampo-amygdalo-parahippocampal cortices), excluding the right temporal posterior scar. Epilepsy surgery of the “actual epileptogenic zone” was performed. Two years after surgery the patient remains seizure-free.