Localizing significance of temporal intermittent rhythmic delta activity (TIRDA) in drug-resistant focal epilepsy

Clinical Neurophysiology 114 (2003) 70–78 www.elsevier.com/locate/clinph

Localizing significance of temporal intermittent rhythmic delta activity (TIRDA) in drug-resistant focal epilepsy Giancarlo Di Gennaro a, Pier Paolo Quarato a, Paolo Onorati b,c,*, Giovanni B. Colazza a, Francesco Mari a,d, Liliana G. Grammaldo a, Olga Ciccarelli e, Nicolo` G. Meldolesi a, Fabio Sebastiano f, Mario Manfredi a,d, Vincenzo Esposito a,g a Epilepsy Surgery Unit, IRCCS “NEUROMED”, Pozzilli (IS), Italy Child Developmental Center, San Raffaele Pisana-Tosinvest Sanita`, Rome, Italy c Department of Human Physiology and Pharmacology, University “La Sapienza”, Rome, Italy d Department of Neurological Sciences, University “La Sapienza”, Rome, Italy e NMR Research Unit, Institute of Neurology, Queen Square, London, UK f Department of Computer and Systems Science, University “La Sapienza”, Rome, Italy g Department of Neurosurgery, University “La Sapienza”, Rome, Italy b

Accepted 15 October 2002

Abstract Objective: Temporal intermittent rhythmic delta activity (TIRDA) is an EEG pattern characterized by sinusoidal trains of activity, ranging from 1 to 3.5 Hz, and well localized over the temporal regions. It is considered to be an indicator of temporal lobe epilepsy (TLE), but full agreement between different authors has still not been reached. The aim of this study was therefore to assess the role of TIRDA in localizing the epileptogenic zone, which was estimated using anatomo-electro-clinical correlations obtained from non-invasive pre-surgical investigations, in a large group of patients affected by drug-resistant partial epilepsy. Methods: The occurrence of TIRDA was investigated using a prolonged Video-EEG recording of 129 patients affected by drug-resistant partial epilepsy that underwent a non-invasive pre-surgical protocol. Patients were divided into 3 groups: TLE only, extratemporal epilepsy, and multilobar epilepsy including temporal lobe. According to the epileptogenic zone identified using anatomo-clinical-radiological correlations, 3 different subgroups of TLE were identified: mesial, lateral, and mesio-lateral. Statistical analysis was performed in order to evaluate the relationship between TIRDA and the epileptogenic zone, and neuroradiological, neuropathological, EEG interictal and ictal findings. Results: The pattern of TIRDA was observed in 52 out of the 129 (40.3%) patients studied. Significant correlations were found between TIRDA and: (i) mesial and mesio-lateral TLE; (ii) mesial temporal sclerosis; (iii) interictal epileptiform discharge localized over the anterior temporal regions; and (iv) 5–9 Hz temporal ictal discharge. Conclusions: Our research shows that TIRDA plays a role in localizing the epileptogenic zone, suggesting that this pattern might be considered as an EEG marker of an epileptogenesis that involves the mesial structures of the temporal lobe. However, further studies investigating the relationship between intracranial EEG monitoring and simultaneous scalp EEG recording are needed in order to confirm our findings and improve our understanding of the significance of TIRDA. q 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Temporal intermittent rhythmic delta activity; Temporal lobe epilepsy; Pre-surgical evaluation; EEG; Epilepsy surgery

1. Introduction Epilepsy surgery in patients affected by drug-resistant partial epilepsy is planned using clinical and paraclinical data, including past medical history, features of seizures, and neuroradiological and electrophysiological findings * Corresponding author. Department of Human Physiology and Pharmacology, University of Rome “La Sapienza”, P.le Aldo Moro 5, 00185 Rome, Italy. Tel.: 139-06-49910896; fax: 139-06-49910851. E-mail address: [email protected] (P. Onorati).

(Munari et al., 1990; Spencer, 1997; Duchowny, 1997; Rosenow and Luders, 2001). Long-term intensive VideoEEG monitoring is a fundamental tool in non-invasive presurgical protocol, providing both EEG interictal data and ictal electroclinical correlations (Gates, 1986; Bancaud and Talairach, 1991; Sperling et al., 1992; Wieser and Williamson, 1993; Kaplan and Lesser, 1996; Clemens and Menes, 2000; Holmes et al., 2000; Mintzer et al., 2002). Despite several reports that evaluated the reliability of interictal focal activity (both slow waves and epileptiform discharges) and ictal patterns for the localization of epilep-

1388-2457/02/$ - see front matter q 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S13 88- 2457(02)0033 2-2

CLINPH 2002627

G. Di Gennaro et al. / Clinical Neurophysiology 114 (2003) 70–78

togenesis (Ebersole and Pacia, 1996; Ebersole, 1997; Pacia and Ebersole, 1997; Duncan, 1998; Adachi et al., 1998; Foldvary et al., 2001), there are few data investigating the localizing value of temporal intermittent rhythmic delta activity (TIRDA), which represents a scalp recorded EEG pattern of uncertain interictal or ictal significance. The pattern of TIRDA was first described by Reiher et al. (1989) and defined as short bursts of 10 s or more of repetitive, rhythmic, saw-toothed or sinusoidal, 1–4 Hz activity of 50–100 mV in amplitude, predominantly running over the anterior temporal regions. The authors identified this pattern

71

as suggestive of complex partial epilepsy. Gambardella et al. (1995a) recently described TIRDA as trains of rhythmic delta activity of very short duration (1–6 s), and suggested that TIRDA may indicate temporal lobe epilepsy (TLE). In another study the same authors (Gambardella et al., 1995b) evaluated the relationship of TIRDA with spike focus and mesiotemporal atrophy, which was assessed by volumetric magnetic resonance imaging (MRI), in a group of TLE patients. They reported a striking concordance between TIRDA, mesiotemporal atrophy and side of spiking. However, because the structural abnormality does not

Fig. 1. Example of TIRDA occurring over right temporal regions as trains of rhythmic 1.5–2 activity of 50–75 mV amplitude in a patient with left M-TLE.

72

G. Di Gennaro et al. / Clinical Neurophysiology 114 (2003) 70–78

always correspond to the epileptogenic zone, the association reported between TIRDA and mesiotemporal atrophy in patients with TLE does not necessarily suggest a close relationship with epileptogenesis. In order to identify the optimal strategy of surgery in patients with TLE, it is useful to distinguish between mesial and lateral (neocortical) TLE using non-invasive diagnostic procedures. Indeed, in the case of cryptogenic lateral TLE it is often recommended to perform further intracranial investigations, whereas in the case of mesial TLE the choice might be between selective amygdalo-hippocampectomy and anterior temporal lobectomy (Bartolomei et al., 1999; Foldvary et al., 1997). However, the localizing value of TIRDA in respect to mesial and lateral subtypes of TLE has not been widely investigated. Moreover, the data that have been published so far concerning the localizing value of TIRDA are inconsistent. Indeed, recent studies (Geyer et al., 1999), which confirmed the association between TIRDA and TLE, have reported the rare occurrence of TIRDA in extratemporal epilepsy (ETE) and recommended caution when using TIRDA as a presurgical localizing finding. The aim of this study was to verify the occurrence of TIRDA in a group of 129 patients affected by drug-resistant partial epilepsy studied with a non-invasive, pre-surgical diagnostic protocol. We investigated if TIRDA is a reliable indicator of epileptogenesis involving specific cortical structures. The possibility that TIRDA may reflect a nonspecific lesional interictal activity or a subclinical seizure is also discussed. 2. Patients and methods 2.1. Patients One hundred twenty-nine patients affected by medically refractory partial epilepsy were studied. There were 58 female and 71 male patients. The mean age was 32.6 years (SD 19) and the mean duration of epilepsy was 19.1 years (SD 9.8, range 1–46 years). They were referred to the Epilepsy Surgery Unit of IRCCS NEUROMED, Pozzilli (IS), Italy, between September 1998 and June 2001. 2.2. Pre-surgical diagnostic protocol All patients underwent a non-invasive pre-surgical protocol that included: (1) continuous long-term intensive, diurnal and nocturnal, Video-EEG monitoring (Telefactor Corp, Conshohoken, PA); (2) neuropsychological evaluation, including the Wechsler Adult Intelligence Scale, Aachener Aphasie Test, Words of Rey, Figure of Rey, Corsi Span, Digit Span, Raven’s Progressive Matrices, Benton Lines, Trail making test and Semantic Verbal Fluency; (3) psychiatric evaluation; (4) brain imaging, including computed tomography scan and MRI; in particular, the MRI protocol consisted of T1- and T2-weighted images acquired in the

axial, coronal and sagittal planes and fluid-attenuated inversion recovery (FLAIR) images; (5) multimodal evoked potentials, transcranial magnetic stimulation (Rossini et al., 2002) and visual field examination, which were performed only in well selected cases; (6) mono or bilateral Wada test that was performed in 6 patients with uncertain hemispheric dominance (i.e. mancinism, disagreement between ictal and/or postictal language impairment and lateralization of ictal discharge). The Video-EEG recording technique, performed by collodion fixed scalp electrodes (International 10–20 System), was in accordance with international guidelines developed by the American Electroencephalographic Society for Video-EEG monitoring in epileptic patients (American Electroencephalography Society, 1986). The electrocardiogram was simultaneously recorded in order to exclude the possible occurrence of vascular pulsation artefacts that could resemble the TIRDA sinusoidal shape. Scalp EEG analysis was done using bipolar longitudinaltransverse and referential montages. In order to facilitate seizure occurrence during the VideoEEG recording, sleep deprivation was performed in all patients (Marinig et al., 2000) and medication was tapered. During the diurnal Video-EEG, patients were monitored by a neurologist and a technician, who observed the occurrence of seizures and examined their features, including the presence of auras, language dysfunction, loss of contact, automatisms, and motor or sensory symptoms. During the nocturnal recordings, patients were instead monitored by a telemetric procedure and seizures were detected by an online computer (Crespel et al., 2000). The EEG recording of each patient was examined in order to evaluate the presence of background abnormalities, interictal slow and epileptiform activity, ictal changes and postictal slowing. The ictal and postictal EEG changes were then compared with the features of the seizures. 2.3. Analysis of TIRDA In the EEGs recorded for each patient during the admission to the Unit the occurrence of TIRDA was investigated. They were defined as a pattern of sinusoidal trains of activity ranging from 1 to 3.5 Hz, with an amplitude of 50–100 mV, and well localized over the antero-medial temporal regions (F7, T3, T5 and/or F8, T4, T6) of one or, asynchronously, both hemispheres (Figs. 1 and 2) (Reiher et al., 1989; Normand et al., 1995; Gambardella et al., 1995b). As previously suggested (Gambardella, 1995a), in order to avoid misleading observations caused by postictal EEG activation, the abnormalities recorded within 8 h after auras or complex partial seizures were not considered for analysis of TIRDA occurrence. TIRDA was classified as unilateral if at least 95% of the pattern was localized on one side. The TIRDA analysis was performed by two examiners (G.D.G., P.P.Q.), who were blind to the reciprocal findings,

G. Di Gennaro et al. / Clinical Neurophysiology 114 (2003) 70–78

73

Fig. 2. Example of TIRDA occurring over left temporal regions and, asynchronously, over right temporal regions in the same patient.

using a high definition monitor (2024 £ 1860 d/i) and two filter settings, i.e. LLF 1.3 and HFF 50 Hz. In case of interobserver disagreement, a final consensus was achieved after discussion. 2.4. Diagnostic criteria for patients’ classification Using the information provided by the non-invasive presurgical protocol, we developed a diagnostic grid (Table 1) based on the well known anatomical, EEGraphic and clinical criteria (O’Brien et al., 1996; Ebersole and Pacia, 1996; Pacia and Ebersole, 1997; Gil-Nagel and Risinger, 1997; Foldvary et al., 1997, 2001; Moriarity et al., 2001; Henkel et al., 2002) in order to identify the patients affected by TLE in accordance with the precise localization of the epileptogenic zone. These criteria were considered as suggestive of epileptogenesis in the mesial or lateral temporal lobe, and consequently grouped into mesial and temporal clusters (Table 1). In cases where the anatomical criteria could not be met, because the brain imaging was negative, only the presence of the EEGraphic and clinical criteria was taken into account for the classification. If the patients completely or partially fulfilled the criteria for mesial and lateral TLE clusters, they were sorted into the

following 3 groups: 1. Mesial TLE (M-TLE): when the anatomical criterion for mesial cluster, the EEGraphic criterion for mesial cluster, and at least two of the clinical criteria for mesial cluster were met. 2. Lateral TLE (L-TLE): when the anatomical criterion for lateral cluster, the EEGraphic criterion for lateral cluster, and at least two of the clinical criteria for lateral cluster were met. 3. Mesio-lateral TLE (ML-TLE): when at least one criterion for mesial cluster and at least one criterion for lateral cluster were met. Following this classification, 64 patients were diagnosed as affected by TLE only, and in particular 28 patients (43.7%) were diagnosed as affected by M-TLE, 9 (14.1%) by L-TLE, and 27 (42.2%) by ML-TLE. Patients that did not fulfil the TLE criteria, i.e. with initial visual, focal sensory or focal motor symptoms or signs, or with ictal EEG onset out of temporal regions, or with a lesion out of the temporal lobe and ictal EEG onset corresponding to that lesion, were diagnosed as affected by ETE (40 patients) or multilobar epilepsy including temporal lobe

74

G. Di Gennaro et al. / Clinical Neurophysiology 114 (2003) 70–78

Table 1 Grid of anatomical, EEGraphic and clinical criteria used for classification of TLE patients Mesial cluster (Mc)

Lateral cluster (Lc)

(A) Anatomical criteria (O’Brien et al., 1996)

Mesial temporal sclerosis or hippocampal atrophy and/or medial structural lesion (B) Ictal EEGraphic criteria (Ebersole and Pacia, 1996; Pacia 5–9 Hz discharge, well localized on and Ebersole, 1997; Foldvary et al., 2001) temporal regions (C) Clinical criteria (O’Brien et al., 1996; Foldvary et al., (1) Rising epigastric aura, isolated or 1997; Moriarity et al., 2001; Henkel et al., 2002; Gil-Nagel associated with other vegetative and Risinger, 1997) symptoms; (2) early oro-alimentary automatisms (‘machonnement’); (3) dystonia of contralateral superior limb

(MLITLE) (25 patients). In these patients intracranial investigations are required for identification of the epileptogenic zone.

f fD 2 fB fC ffi f ¼ pAffiffiffiffiffiffiffiffiffiffiffi pp 0 £ qq 0

2.5. Epilepsy surgery and outcome

x2 ¼

Surgery was recommended in all 64 patients affected by TLE only. Fifty patients (20 with M-TLE, 8 with L-TLE, and 22 with ML-TLE) underwent the surgery, 4 patients refused, and 10 patients are still waiting. Previous TLE epilepsy surgery carried out in our Unit consisted of the standard lobectomy as developed by Falconer (1971). Since March 2000, the surgery has consisted only of the removal of the epileptogenic zone, by reference to the new diagnostic grid (Table 2). In cases where amygdalohippocampectomy was planned, a resection of the temporal tip was performed in order to facilitate a complete removal of the hippocampal tail. Histological examination was performed in all cases after surgery (Table 3). Out of the 50 patients that had epilepsy surgery, 23 patients completed 1 year of follow-up and were all seizure-free (Engel class IA) (Engel, 1987). Out of 27 patients that did not complete 1 year of follow-up, 3 developed rare auras and one complex partial seizures (Table 2). 2.6. Statistical analysis The relationship between TIRDA and the epileptogenic zone, and the neuroradiological, neuropathological, interictal and ictal EEG findings, was evaluated in the form of a 2 £ 2 contingency table. In order to define the strength and the statistical significance of the relation the Crame´ r, or f , coefficient for dichotomous variables and the chi-square, or x 2, were calculated according to the following equations:

Focal atrophy of temporal lateral neocortex and/or lateral structural lesion 2–5 Hz discharge well localized on temporal regions (1) Psychic/experiential aura; (2) auditive or vestibular symptoms; (3) staring without oro-alimentary automatisms

ð1Þ

K X ð fri 2 fi Þ2 fri i¼1

ð2Þ

In Eq. (1) fA is the number of subjects showing TIRDA and simultaneously positive with respect to the considered variable, fB is the number of subjects showing only TIRDA, whereas fC is the number of subjects positive with respect to the variable, and fD represents the number of subjects which show neither TIRDA nor positive current variable; also p ¼ f A 1 f B, p 0 ¼ f A 1 f C, q ¼ f C 1 f D, q 0 ¼ f B 1 f D. In Eq. (2) fri indicates the expected or theoretical frequencies while fi represents the observed frequencies of the data series; K represents the number of considered categories (K ¼ 4). The significance level for the statistical test was set at P , 0:01 and P , 0:001. 3. Results Three hundred eighty-seven seizures were recorded and 665 h of Video-EEG were examined. The pattern of TIRDA was observed in 52 out of the 129 (40.3%) patients studied. 3.1. Relationship between TIRDA and the epileptogenic zone TIRDA was observed in 24 out of the 28 (85.7%) M-TLE patients (f ¼ 0:37, P , 0:001), in 18 out of the 27 (66.7%) ML-TLE patients (f ¼ 0:28, P , 0:01), and in 10 out of the

Table 2 TLE patients underwent neurosurgery: surgical resection and follow-up TLE patients underwent neurosurgery

Standard temporal lobectomy

M-TLE (20/28) 10 (50%) L-TLE (8/9) 4 (50%) ML-TLE (22/27) 22 (100%)

Amygdalo-hippocampectomy (associated with resection of temporal tip)

Lesionectomy

Follow-up .1 year Follow up ,1 year

10 (50%) – –

– 4 (50%) –

13 (13 sz-free) 4 (4 sz-free) 6 (6 sz-free)

7 (6 sz-free) 4 (3 sz-free) 16 (14 sz-free)

G. Di Gennaro et al. / Clinical Neurophysiology 114 (2003) 70–78

75

Table 3 TLE patients underwent neurosurgery: histopathology TLE subgroups

Mesial temporal sclerosis

M-TLE (20) L-TLE (8) MS-TLE (22)

16 11

Focal cortical atrophy

Dysembryoplastic neuroepithelial tumors

Dysplasias

Low grade gliomas

Cavernomas

4 1

1 1

2 2

4 2 7

1

25 (40%) MLITLE patients. The EEGs recorded in L-TLE and ETE patients did not show any TIRDA (ETE: f ¼ 20:55, P , 0:001) (Fig. 3). 3.2. Relationship between TIRDA and neuroradiological findings TIRDA was observed in 35 out of the 47 (74.5%) patients with signal abnormalities in the mesio-temporal structures (f ¼ 0:53, P , 0:001), in 4 out of the 16 (25%) patients with signal abnormalities in the lateral temporal structures, in 3 out of the 7 (42.9%) patients with signal abnormalities involving the mesio-lateral structures of the temporal lobe, and in 5 out of the 10 (50%) patients with lesions involving multilobar structures, including the temporal lobe. 3.3. Relationship between TIRDA and neuropathological findings TIRDA was observed in 30 out of the 35 (85.7%) patients with mesial temporal sclerosis (f ¼ 0:47, P , 0:001), in one out of the 7 (14.3%) patients with focal cortical atrophy (f ¼ 20:38, P , 0:01), and in 6 out of the 16 (37.5%) patients with neoplasias (f ¼ 20:34, P , 0:01). Out of the latter 6 patients, 4 had the tumour localized in the mesial temporal structures and 2 in the mesio-lateral temporal structures. TIRDA was also recorded in 2 out of the 17 (12%) patients with disorders of cortical organization, and in 2 out of the 4 (50%) patients with cavernomas. No patients with brain vascular lesions showed TIRDA.

areas. Patients with epileptiform abnormalities localized only in the extratemporal regions did not show any TIRDA. 3.4.3. Ictal discharge in patients with TLE TIRDA was observed in 23 out of the 31 (74%) patients with ictal discharges ranging from 5 to 9 Hz (f ¼ 0:45, P , 0:001), and in 16 out of the 33 (48%) patients with ictal discharges ranging from 2 to 5 Hz. 4. Discussion We have investigated the occurrence of TIRDA in 129 patients with refractory partial epilepsy who underwent a pre-surgical non-invasive protocol. In accordance with previous studies (Gambardella, 1995a), the pattern of TIRDA was observed in 66% of patients affected by TLE while no patients affected by ETE showed TIRDA, suggesting that there is a significant association between this pattern and drug-resistant TLE. In order to classify patients with TLE, which represents an heterogeneous clinical group, according to the localization of the epileptogenic zone, we developed a diagnostic grid that is based on correlations among anatomical, electrical and clinical data, in accordance with principles firstly planned by Bancaud et al. (1965) and later developed by Munari et al. (1990), obtained by non-invasive investigations. According to many studies that evaluated the localizing value of a single diagnostic criterion (Williamson et al.,

3.4. Relationship between TIRDA and EEG findings 3.4.1. Interictal slow waves TIRDA was observed in 37 out of the 58 (64%) patients with slow waves localized in the temporal regions (f ¼ 0:43, P , 0:001), and in 15 out of the 31 (48%) patients with slow waves localized both in the temporal regions and the extratemporal areas. Patients with slow waves localized only in the extratemporal regions did not show any TIRDA. 3.4.2. Interictal epileptiform discharges TIRDA was observed in 39 out of the 61 (64%) patients with interictal epileptiform discharges localized in the temporal regions (f ¼ 0:46, P , 0:001), and in 13 out of the 29 (45%) patients with epileptiform abnormalities localized both in the temporal regions and the extratemporal

Fig. 3. The significance level for the statistical correlation between TIRDA and different TLE patients groups was set at P , 0:01 and P , 0:001 (21 , f , 1).

76

G. Di Gennaro et al. / Clinical Neurophysiology 114 (2003) 70–78

1993; Ebersole and Pacia, 1996; O’Brien et al., 1996; Bleasel et al., 1997; Pacia and Ebersole, 1997; Foldvary et al., 1997, 2001; Mohamed et al., 2001; Henkel et al., 2002; GilNagel and Risinger, 1997; Moriarity et al., 2001), the data obtained from pre-surgical evaluation were grouped in clusters strongly suggesting the location of epileptogenesis in mesial or lateral aspects of the temporal lobe and, on the basis of our experience, specific modalities of a combination of different criteria were utilized for localization. With respect to other non-invasive diagnostic protocols (Sperling et al., 1992; Cendes et al., 2000) for TLE surgery, that allow selection of those patients who would benefit from standard temporal lobectomy and identification of those patients who would be invasively investigated or rejected, our diagnostic grid permits the refinement of the classification of TLE in mesial, temporal and mesio-temporal subtypes. The advantage of using this grid is that the surgery can be specifically planned, as far as possible, according to the extension of the epileptogenic zone in the 3 different TLE subtypes (i.e. amygdalo-hippocampectomy in M-TLE, standard lobectomy in ML-TLE, and lesionectomy in lesional L-TLE), avoiding standardized and unnecessary, more extensive surgery. The excellent outcome (.90% of patients seizure-free), considering both cases with a follow-up of at least 1 year and those with a shorter outcome, showed by the patients who underwent temporal resection, confirms the methodology we used in our study as a reliable diagnostic protocol of TLE. In contrast with a previous report (Geyer et al., 1999), we did not find the pattern of TIRDA in patients affected by ETE. Moreover, TIRDA was recorded in 40% of patients with MLITLE (i.e. fronto-temporal epilepsy, temporo-occipital epilepsy, etc.). A possible explanation for this difference may be the classification used in the Geyer et al. (1999) study, which did not distinguish between ETE and MLITLE. The data published so far are controversial regarding the localization of the epileptogenic zone using TIRDA. Reiher et al. (1989) first described TIRDA in a group of patients affected by epilepsy, defining them as ‘an accurate interictal indicator’ of complex partial seizures. In accordance with previous reports, Normand et al. (1995) considered this pattern as a reliable marker of epileptogenic activity. The Gambardella et al. (1995a) study found a correlation between intermittent delta activity recorded over the temporal regions and TLE ‘when the pathology is the mesiotemporal atrophy’. In contrast, Geyer et al. (1999) revealed the occurrence of lateralized TIRDA in a small percentage (4%) of patients affected by ETE, questioning the significance of this pattern as a pre-surgical localizing finding. Koutroumanidis et al. (1998) explored the relationship between interictal regional slow activity (IRSA) and cerebral glucose metabolism investigated by means of interictal 18-F fluorodeoxyglucose positron emission tomography (FDG-PET) in a TLE patient group. They showed that

IRSA strongly correlated with hypometabolism localized in the lateral aspect of the temporal lobe. These data are only apparently conflicting with our results because they are unlikely comparable for several reasons. Firstly, the methodological approach used by Koutroumanidis et al. (1998) did not find differences between TIRDA and other temporal slow activities, i.e. polymorphic temporal delta, which were considered together in the analysis of their results. Secondly, the cerebral region of hypometabolism recognized by interictal FDG-PET scans, indicating a dysfunctional cortical region, may be related to the epileptogenic zone but may not necessarily coincide with it. In fact, the epileptogenic zone is a ‘theoretical’ concept and may be identified only correlating with anatomical, electrical and clinical data (Rosenow and Luders, 2001). Finally, the patient classification adopted from the authors did not refer to TLE subtypes. Nowadays, the epileptiform nature of this pattern is still unclear. Does TIRDA accurately reflect the repetitive (ictal) spiking activity occurring in the deep structures? Although the Reiher et al. (1989) study concluded that TIRDA did not have the appearance of ‘larval seizure’ and was not associated with the clinical features of the seizures, it could not exclude the ictal significance of this pattern. According to this view, previous isolated observations (Wieser, 1980) in a TLE patient during a psychomotor status epilepticus investigated by means of co-recorded scalp-EEG and intracerebral electrodes (stereo-EEG) showed that ictal epileptiform discharges (spike-and-wave complexes) in mesial temporal lobe structures were depicted as slow rhythmic delta runs in the scalp EEG. Although this observation suggests that depth ictal discharges may appear in surface EEG as delta activity, this finding cannot immediately be applied to the physiopathology of the TIRDA, which is an EEG pattern not related to clinical seizures. In the Normand et al. (1995) study, intra-operative recordings were performed in two patients affected by drug-resistant TLE using cortical strips and depth electrodes. Spikes in the deep leads and rhythmical delta activity resembling TIRDA in the cortical leads were recorded. However, in this study a simultaneous scalp EEG recording was not performed and, therefore, a comparison between scalp recorded TIRDA and the corresponding electrographic (ictal?) events in the deep structures could not be made. Our study shows the ability of TIRDA to localize the epileptogenic zone, suggesting that this pattern might be considered as an EEG marker of an epileptogenesis that involves the mesial structures of the temporal lobe. In fact, TIRDA was observed in the majority of patients affected by mesial (85.7%) TLE (f ¼ 0:37, P , 0:001), mesio-lateral (66.7%) TLE (f ¼ 0:28, P , 0:01), and mesial temporal sclerosis (85.7%) (f ¼ 0:47, P , 0:001), while no patients affected by L-TLE and ETE (f ¼ 20:55, P , 0:001) showed TIRDA. Moreover, a strong association was found between the ictal discharges ranging from 5 to 9

G. Di Gennaro et al. / Clinical Neurophysiology 114 (2003) 70–78

Hz (f ¼ 0:45, P , 0:001), which indicate the onset of seizures in the mesial temporal structures (Ebersole and Pacia, 1996; Pacia and Ebersole, 1997; Foldvary et al., 2001), and the occurrence of TIRDA. Future studies with EEG-linked functional MRI techniques (Goldman et al., 2000; Schomer et al., 2000), permitting the analysis of the correlations between TIRDA and possible brain region activation, could improve our understanding of the generators of the pattern. However, further studies investigating the relationship between intracranial EEG monitoring (Munari et al., 1997) and simultaneous scalp EEG recording are needed to confirm these findings.

Acknowledgements We are grateful to the team of neurophysiopathology technicians (Antonella Addesso, Lucia Cacciola, Nicola Cusano, and Carolina Falco) for performing meticulous Video-EEG recordings in all patients included in the study.

References Adachi N, Alarcon G, Binnie CD, Elwes RDC, Polkey CE, Reynolds EH. Predictive value of interictal epileptiform discharges during non-REM sleep on scalp EEG recordings for the lateralization of epileptogenesis. Epilepsia 1998;39:628–632. American Electroencephalography Society. Guidelines for long-term monitoring for epilepsy. J Clin Neurophysiol 1986;63:107–111. Bancaud J, Talairach J. Se´ me´ iologie clinique des crises du lobe temporal. Crises epileptiques et epilepsies du lobe temporal, tome II, Gentilly: Documentation Medicale Labaz, 1991. Bancaud J, Talairach J, Bonis A, Schaub C, Szikla G, Morel P, BordasFerrer M. La ste´ re´ o-e´ lectroence´ phalographie dans l’e´ pilepsie, Paris: Masson, 1965. Bartolomei F, Wendling F, Vignal JP, Kochen S, Bellanger JJ, Badier JM, Le Bouquin-Jeannes R, Chauvel P. Seizures of temporal lobe epilepsy: identification of subtypes by coherence analysis using stereo-electroencephalography. Clin Neurophysiol 1999;110:1741–1754. Bleasel A, Kotagal P, Kankirawatana P, Rybicki L. Lateralizing value and semiology of ictal limb posturing and version in temporal lobe and extratemporal epilepsy. Epilepsia 1997;38:168–174. Cendes F, Li LM, Watson C, Andermann F, Dubeau F, Arnold DL. Is ictal recording mandatory in temporal lobe epilepsy? Not when the interictal electroencephalogram and hippocampal atrophy coincide. Arch Neurol 2000;57:497–500. Clemens B, Menes A. Sleep spindle asymmetry in epileptic patients. Clin Neurophysiol 2000;111(12):2155–2159. Crespel A, Coubes P, Baldy-Moulinier M. Sleep influence on seizures and epilepsy effects on sleep in partial frontal and temporal lobe epilepsies. Clin Neurophysiol 2000;111(Suppl 2):S54–S59. Duchowny M. Identification of surgical candidates and timing of operation. In: Wyllie E, editor. The treatment of epilepsy: principles and practice, Baltimore, MD: Williams and Wilkins, 1997. pp. 967–975. Duncan JS. Interictal focal activity in temporal lobe epilepsy. J Neurol Neurosurg Psychiatry 1998;65:149. Ebersole JS. Localization of epileptic foci in temporal lobe epilepsy. Electroenceph clin Neurophysiol 1997;103:47. Ebersole JS, Pacia SV. Localization of temporal lobe foci by ictal EEG patterns. Epilepsia 1996;37:386–399. Engel Jr J. Outcome with respect to epileptic seizures. In: Engel Jr J, editor.

77

Surgical treatment of the epilepsies, New York: Raven Press, 1987. pp. 553–571. Falconer MA. Anterior temporal lobectomy for epilepsy. In: Logue V, editor. Operative surgery, London: Butterworths, 1971. pp. 142–149. Foldvary N, Lee N, Thwaites G, Mascha E, Hammel J, Kim H, Friedman AH, Radtke RA. Clinical and electrographic manifestations of lesional neocortical temporal lobe epilepsy. Neurology 1997;49:757–763. Foldvary N, Klem G, Hammel J, Bingaman W, Najm I, Luders H. The localizing value of ictal EEG in focal epilepsy. Neurology 2001;57:2022–2028. Gambardella A, Gotman J, Cendes F, Andermann F. Focal intermittent delta activity in patients with mesiotemporal atrophy: a reliable marker of the epileptogenic focus. Epilepsia 1995a;36:122–129. Gambardella A, Gotman J, Cendes F, Andermann F. The relation of spike foci and of clinical seizure characteristics to different patterns of mesial temporal atrophy. Arch Neurol 1995b;52:287–293. Gates RJ. Epilepsy presurgical evaluation in era of intensive neurodiagnostic monitoring. In: Gummit RJ, editor. Intensive neurodiagnostic monitoring, Advances in neurology, 46. New York: Raven Press, 1986. pp. 227–248. Geyer JD, Bilir E, Faught RE, Kuzniecky R, Gilliam F. Significance of interictal temporal lobe delta activity for localization of the primary epileptogenic region. Neurology 1999;52:202–205. Gil-Nagel A, Risinger MW. Ictal semiology in hippocampal versus extrahippocampal temporal lobe epilepsy. Brain 1997;120:183–192. Goldman RI, Stern JM, Engel Jr. J, Cohen MS. Acquiring simultaneous EEG and functional MRI. Clin Neurophysiol 2000;111(11):1974–1980. Henkel A, Noachtar S, Pfander M, Luders HO. The localizing value of the abdominal aura and its evolution: a study in focal epilepsies. Neurology 2002;58:271–276. Holmes MD, Kutsy RL, Ojemann GA, Wilensky AJ, Ojemann LM. Interictal, unifocal spikes in refractory extratemporal epilepsy predict ictal origin and postsurgical outcome. Clin Neurophysiol 2000;111(10): 1802–1808. Kaplan PW, Lesser RP. Noninvasive EEG. In: Wyllie E, editor. The treatment of epilepsy: principles and practice, 2nd ed. Baltimore, MD: Williams and Wilkins, 1996. pp. 976–987. Koutroumanidis M, Bigne CD, Elwes RD, Polkey CE, Seed P, Alarcon G, Cocs T, Barrington S, Marsden P, Maisey MN, Panayiotopulos CP. Interictal regional slow activity in temporal lobe epilepsy correlates with lateral temporal hypometabolism as imaged with 18FDG PET: neurophysiological and metabolic implications. J Neurol Neurosurg Psychiatry 1998;65(2):170–176. Marinig R, Pauletto G, Dolso P, Valente M, Bergonzi P. Sleep and sleep deprivation as EEG activating methods. Clin Neurophysiol 2000;111(Suppl 2):S47–S53. Mintzer S, Nicholl JS, Stern JM, Engel J. Relative utility of sphenoidal and temporal surface electrodes for localization of ictal onset in temporal lobe epilepsy. Clin Neurophysiol 2002;113(6):911–916. Mohamed A, Wyllie E, Ruggieri P, Kotagal P, Babb T, Hilbig A, Wylie C, Ying Z, Staugaitis S, Najm I, Bulacio J, Foldvary N, Lu¨ ders H, Bingaman W. Temporal lobe epilepsy due to hippocampal sclerosis in pediatric candidates for epilepsy surgery. Neurology 2001;56:1643–1649. Moriarity JL, Boatman D, Krauss GL, Storm PB, Lenz FA. Human “memories” can be evoked by stimulation of the lateral temporal cortex after ipsilateral medial temporal lobe resection. J Neurol Neurosurg Psychiatry 2001;71:549–551. Munari C, Tassi L, Lo Russo G, Canevini MP, Hamasaki T. La terapia chirurgica nell’epilessie parziali: schematizzazione metodologica o approccio personalizzato. Boll Lega Ital Epil 1990;70–71:61–67. Munari C, Francione S, Tassi L, Kahane P, Hoffman D, Lo Russo G, Minottti L. Indications and strategies of stereo-electro-encephalography in severe partial symptomatic epilepsies. Electroenceph clin Neurophysiol 1997;103:16–17. Normand MM, Wszolek ZK, Klass DW. Temporal intermittent rhythmic delta activity in electroencephalograms. J Clin Neurophysiol 1995;12:280–284.

78

G. Di Gennaro et al. / Clinical Neurophysiology 114 (2003) 70–78

O’Brien TJ, Kilpatrick C, Murrie V, Vogrin S, Morris K, Cook MJ. Temporal lobe epilepsy caused by mesial temporal sclerosis and temporal neocortical lesions. A clinical and electroencephalographic study of 46 pathologically proven cases. Brain 1996;119:2133–2141. Pacia SV, Ebersole JS. Intracranial EEG substrates of scalp ictal patterns from temporal lobe foci. Epilepsia 1997;38:642–654. Reiher J, Beaudry M, Leduc CP. Temporal intermittent rhythmic delta activity (TIRDA) in the diagnosis of complex partial epilepsy: sensitivity, specificity and predictive value. Can J Neurol Sci 1989;16:398– 401. Rosenow F, Luders H. Presurgical evaluation of epilepsy. Brain 2001;124:1683–1700. Rossini PM, Berardelli A, Cantello R. Neuromagnetic methods and transcranial magnetic stimulation for testing sensorimotor cortex excitability. In: Guerrini R, Aicardi J, Andermann F, Hallet M, editors. Epilepsy and movement disorders, Cambridge: Cambridge University Press, 2002. pp. 59–76. Schomer DL, Bonmassar G, Lazeyras F, Seeck M, Blum A, Anami K,

Schwartz D, Belliveau JW, Ives J. EEG-linked functional magnetic resonance imaging in epilepsy and cognitive neurophysiology. J Clin Neurophysiol 2000;17(1):43–58. Spencer SS. Selection of candidates for temporal lobe resection. In: Wyllie E, editor. The treatment of epilepsy: principles and practice, Baltimore, MD: Williams and Wilkins, 1997. pp. 1036–1049. Sperling MR, O’Connor MJ, Saykin AJ, Phillips MD, Bridgeman PA, French JA, Gonatas N. A non-invasive protocol for anterior temporal lobectomy. Neurology 1992;42:416–422. Wieser HG. Temporal lobe or psychomotor status epilepticus. A case report. Electroenceph clin Neurophysiol 1980;48:558–572. Wieser HG, Williamson PD. Ictal semiology. In: Engel J, editor. Surgical treatment of the epilepsies, New York: Raven Press, 1993. pp. 161–172. Williamson PD, French JA, Thadani VM, Kim JH, Novelly RA, Spencer SS, Spencer DD, Matson RH. Characteristics of medial temporal lobe epilepsy: II. Interictal and ictal scalp electroencephalography, neuropsychological testing, neuroimaging, surgical results, and pathology. Ann Neurol 1993;34:781–787.