Extratempml epilepsy: Clinical psentation, p q x m t w e EEG localization and surgical outcome

L.F. Quesney Montreal Neurological Hospital and Institute

Introduction Victor Horsleys first operation for epilepsy was performed on May 25, 1886, at the National Hospital for the Paralyzed and Epileptic in Queen Square ( I ) . The young patient (James B) was run over by a cab in Edinburgh at age 7 sustaining a depressed comminuted fracture with loss of brain substance, which required surgical exploration. The patient began having seizures at the age of 15. The clinical seizure pattern was compatible with partial motor tonicclonic seizures involving initially the right lower extremity and subsequently the right arm. Based on the clinical presentation, a localizing diagnosis of a discharging lesion localized in the posterior end of the superior frontal sulcus was made. At surgery, a scar tissue was found in exactly the same anatomical location. Interestingly, Wilder Penfield’s first operation for focal epilepsy in November 1928, was also ‘performed in a young patient with post-traumatic partial seizures who was initially submitted to a small surgical resection of an area of cortex near the motor strip, as defmed by electrical stimulation (2). This patient required re-operation. During the last decade a steady increase in surgery for medically intractable partial seizures has been witnessed. Several factors have contributed to this trend, namely a better identification of ideal candidates for surgical therapy .(3), sigrdkant technological developments in EEG and neuro-imaging diagnostic procedures resulting in a more reliable pre-operative localizing of the epileptogenic zone and/or lesion (4) and lastly, a

Address: L.F. Quesney Montreal Neurological Hospital and Institute McGill University 3801 University Street Montreal, Quebec, H3A 2B4 Canada

satisfactory outcome with respect to seizures after surgical therapy (5). The current review will concentrate on the clinical and EEG localizational problems of frontal, parietal and occipital lobe epilepsy. Different surgical approaches and results will be reviewed and discussed.

Frontal lobe epilepsy I. Anatomo-clinical compartmentalization Patients with frontal lobe epilepsy represent 18 9% of large neurosurgical series (6), see Table 9. The clinical localization of the anatomical substrate responsible for the generation of the patient’s habitual seizures is a difficult task. This limitation can be explained if one realizes that the constellation of ictal behavioural manifestations commonly observed in frontal lobe epilepsy are protean and they could originate in many sites of the frontal lobe (7-12). In addition, recent studies have demonstrated that the anatomical substrate responsible for the genesis of frontal lobe seizures in frontal lobe epilepsy, may range in size from relatively small to large epileptogenic zones resembling a continuum (13-15). One extreme of this continuum comprises rather small epileptogenic zones as documented in patients who became and persisted seizurefree after selective surgical removal of restricted frontal lobe regions (16). The other extreme of this continuum includes patients with large epileptogenic zones, frequently involving several frontal gyri and at times, exhibiting a multi-lobar distribution (1 3). The epileptic disturbance in patients with large epileptogenic zones is not uniform (1 3, 17, 18) and this may explain why patients portraying multi-lobar or extensive epileptogenic lesions, such as for instance porencephalic cysts, could present with EEGs disclosing focal interictal epileptic features. Several attempts during the last decade to correlate the clinical manifestations of frontal lobe seizures with

81

Quesney specific anatomical substrate or compartments within the frontal lobe, failed to reach universal consensus (9, 10, 16, 19-29). A recent review of the clinical manifestations of frontal lobe epilepsy by Bancaud and Talairach (1 1) based on 648 seizures of frontal lobe origin recorded in 210 patients submitted to acute or chronic SEEG investigation, does provide reliable electrophysiological and clinical grounds for a compartmentalization of frontal lobe epilepsy. A schematic representation of the different frontal lobe compartments considered by these authors is illustrated in Figure 1. lnlermea d o r m la1 Frontal reg

Area 6 I

Table[.

Seizures originating from areas 4 and 6 (1 54 seizures in 43 patients) Area 4 (39 patients, 126 seizures) Isolated myoclonic jerks Partial motor seizures with Jacksonian march Partial complex motor seizures (Kojewnikoff’s syndrome, Rasmussen’s syndrome, startle epilepsy) Supplementary motor area (4 patients, 28 seizures) Speech arrest Vocalization Palilalia (dominant hemisphere) Abduction and lifting contralateral upper limb Adversive movement of head and eyes M*E

LATERAL ASPECT

“/“t

Orblo-Frontal

Intermed media Frontal req

SMA (area 6)

Chauvel et al. (30). The authors reviewed 50 patients with partial motor seizures submitted to TsEEG investigation. According to the former group, the anatomical substrate of seizure onset in SMA attacks often involves the paracentral lobule in addition to the supplementary motor cortex. The low incidence of seizures arising exclusively from the SMA, a finding which is in agreement with other reports (12, 26) is also mentioned by the authors.

Lo0 Daracenlral8s

MEDIAL ASPECT

Polar F reg

Table II.

Medial intermediate frontal seizures (1 6 1 seizures in 39 patients) Orbito Fronlal

Fg 1. Schemak anatornlcal division of the frontal lobe used for amiysis of seizure semiolqy. (reproduced Mpermissionfrom Bancadet al. 1W).

Seizures arising from areas 4 and 6 were the most numerous ( 1 54 seizures in 43 patients) and as expected, the main clinical manifestations were motor in nature (Table 1).

A recent report (12) performed in 40 patients who became and remained seizure-free after selective surgical removals in the dorso-lateral frontal lobe convexity, showed that supplementary motor area (SMA) seizures disclose a significant correlation with electrographic seizure onset in the fronto-parasagittal convexity. An extensive and thorough review of the clinical and EEG presentation of somato-motor seizures is provided by

82

Frontal absences Consciousness disorders (loss of contact) Speech arrest Movement arrest Simple gestures (automatisms) Conjugate deviation of eyes and head Immediate recovery of consciousness Complex motor seizures Deficits of consciousness Conjugate eye deviation and head adversion Abduction and raising of upper limb Head and trunk manifestations Autonomic manifestations Simple gestural activity (automatisms) Tonic-clonic “generalization”

Extratemporal Epilepsy Table 111.

Behavioral manifestations in patients with frontal lobe epilepsy

Region

N

Warning

PMC

PMT

Heador eyeturning

AF PS FO

21 10 9

9 (42%) 9 (90%) 6 (67%)

6 (29%) 6 (60%) 4 (44%)

2 (9.5%) 5 (50%) 2 (22%)

5 (24%) 4 (40%) 4 (44%)

SMA

Arrest of activity

Aphasia

Automatism

1 (5%) 2 (20%) 0

5 (24%) 2 (20%) 1 (11%)

5 (24%) 3 (30%) 1 (11%)

7 (33%) 0 2 (22%)

PMC, partial motor clonic seizures; PMT, partial motor tonic seizures; SMA, supplementary motor area seizures; AF, anterofrontal; PS, parasagittal; FO,fronto-opercular. (reproducedwib pnnksionfmm Quesney et al, 1992)

Another frontal lobe compartment responsible for the generation of specific seizure patterns according to Bancaud and Talairach (1 I), is the medial intermediate frontal lobe region. The clinical symptomatology of seizures arising in this area (Table 2) comprises brief lasting complex partial seizures which the authors report as “frontal absences”, as well as more elaborated motor seizures ranging from conjugated eye and head adversion to secondary generalized tonic-clonic seizures. Ictal behavioural manifestations resembling absence like seizures and manifested mainly by “arrest of activity” have also been described in seizures arising from the dorsolateral convexity of the frontal lobe (12, 26). The clinical manifestations of seizures originating in the dorso-lateral intermediate convexity were also reviewed by the group from Paris basing their observations on 61 ictal events

recorded in 25 patients submitted to sEEG investigation (11). The most frequent ictal pattern consisted of contralateral eye and head deviation. The latter ictal manifestation can be seen in seizures arising in the parasagittal convexity as well as in the anterior frontal region as illustrated elsewhere (12, 26; see Table 3). Seizures arising in the anterior cingular cortex produced “spectacular manifestations” to use the very same expression employed by Bancaud and Talairach (1 l), who reviewed 66 ictal events recorded in 16 patients. The clinical patterns are reportedly manifold (Table 4) including fear, screaming, aggressive verbalization, complex automatic behaviour and neuro-vegetative disturbances. Consciousness is disturbed, but some contact with the external environment is preserved, as well as partial reactivity to external cues.

Table IV.

Table V.

Orbitofrontal seizures (37 seizures in 18 patients)

Anterior cingulate gyrus seizures (Area 24):66 seizures in 16 patients ~

Fear Dread Screaming Aggressive words Archaic complex motor behavior Emotional symptoms Complex gestural movements Autonomic disorders Unformed visual hallucinations Incomplete loss of consciousness

(reproduced & pem7issiin frm Bancaud et al, 1992)

Olfactory hallucinations and illusions Visceral sensory symptoms Abdominal Epigastric Thoracic Cardiac Esophageal ,Laryngeal Pharyngeal Gestural motor disorders Mumbling Thymic alterations “Autonomic seizures” Respiratory disorders (apnea) Cardiovascular disorders (changes in heart rate, face rubefaction, or pallor) Thermoregulatory disorders Urogenital disorders (micturition) Alimentarv disorders Ihunaer. thirst) (reprcducedwib permission from E!arcaud et al, 1992)

83

3

-

S

p

l

I

F8-F4 F4-Fz P

EOG

Ilmc(v

-

u

-

w

4

S02-F8

2sec

+L-Jr-.-

4

F7-SO1 so 1-so2

F3-F7

Fz-F~

EKG

T4-C4 c4-cz \

T

C3-T3

cz-c3

B F 84-2531

Extratemporal Epilepsy One of the areas of current controversy resides in the clinical presentation of orbito-frontal seizures. Bancaud and Talairach (1 1) based on an analysis of 37 seizures recorded in 18 patients, identified two main clinical presentation patterns, namely: olfactory illusionshallucinations, and seizures with predominant neurovegetative components (Table 5). Ludwig et al. (21), upon reviewing 8 patients with orbito-frontal seizures also identified two main ictal patterns, namely psychomotor attacks and seizures consisting of eye-head turning followed by secondary generalization. More recently, Munari and Bancaud (3 1) reported that seizures arising in the orbito-frontal region as documented by sEEG recordings in 16 of the 60 patients included in his study, are not associated with overly discernible ictal behavioural manifestations, except for autonomic changes such as tachycardia, flushing, hyperpnoea, and mydriasis. Similar findings have been reported elsewhere (32). International consensus has been reached in accepting that ictal behavioural manifestations in frontal lobe epilepsy are largely due to seizure spread to neighbouring or even more distant brain regions. Coexistence of different clinical seizure patterns, is not uncommon in patients with seizures arising in the orbitofrontal and cingular cortex (see below, illustrative cast BF). The latter might be related to a widespread seizure generator (33), or it might be due to seizure propagation through different pathways recruiting different anatomical structures, thus resulting in multiple ictal patterns.

II EEG Problems The EEG mapping of interictal and ictal frontal lobe epileptogenic zones is not devoid of technical and biophysical problems which equally afflict EEG recordings obtained with extracranial and intracranial electrodes, as well as intraoperative electrocortigographic (ECoG) tracings (8, 9, 13-15, 18, 33, 38). Our main task will be to illustrate the limitations inherent to extracranial EEG recordings in frontal lobe epilepsy. The poor localization of the interictal epileptic abnormality and the rather unreliable EEG localization of ictal onsets commonly reported in patients with frontal lobe epilepsy (8,9, 13, 34-38), is dependent upon some of the following factors: A. Only a small portion of the frontal lobe’s anatomy, namely its cortical surface, is accessible to EEG recording with routine extracranial electrodes or with cortical ECoG leads, a situation which leads to an inherent risk of sampling error whenever attempting to record EEG potentials generated in the frontal lobe’s depth. The orbitofrontal cortex, the mesial interhemispheric convexity, the cingular cortex and to a similar extent the depth of the cerebral sulci within the frontal lobe are largely inaccessible to recording with extracranial electrodes (9, 13,21-23, 39, 40). Small epileptogenic foci involving these

anatomical structures may not be recorded, they may show as interictal epileptic discharges in the ipsilateral temporal lobe or they may reflect as widespread epileptic discharges involving the fronto-centro-parietal or the fronto-centrotemporal convexity, thus giving the false impression of a large epileptogenic zone. A graphic illustration of the latter (Figure 2A) was documented in BF, a 34 year old male patient with onset of recurrent complex partial seizures at age 30 without any known etiological or precipitating factors. Most seizures started with an ill-defined cephalic sensation followed by staring and semi-purposeful, stereotyped, perseverative automatisms. During some of these, the patient would for example continue to stir the coffee with a spoon, in an automatic fashon. Some other seizures consisted of paleness, arrest of activity, confusion and orofacial automatisms such as chewing or swallowing. Neurological examination was normal and a contrast enhanced CT scan showed a right orbito-frontal lesion (Figure 2B). ‘ Several pre-operative EEG recordings obtained with sphenoidal and supra-orbital electrodes showed either right temporal epileptiform discharges, or sharp-slow wave complexes simultaneously recorded from the right temporal and frontal lobes as illustrated in Figure 2A. Depth electrode investigation (Figure 2C) demonstrated a regional seizure onset involving the right orbito-frontal and cingular cortex, a finding which was confirmed in six of the patient’s habitual attacks recorded during the intracranial EEG investigation. B. The existence of a functional network of pathways permitting seizure spread within and outside the frontal lobe, may account for a regional or even a multi-lobar distribution of the interictal spiking and of ictal onsets, thus defying an accurate EEG localization. Major pathways providing functional connectivity within the frontal and temporal lobes include the uncinate and arcuate fasciculus as well as the cingulum (41).

85

Quesney

Fg 21: W recordingof one of the paknrs habihralseizures. The ~ G idmk W the fime of elWqaphc seizure onset imrolving the thd h t a l gyms (RF07-9) and be angular artex (RFSl-3).Depth e ! d & recording siteswere as fdlows: RA1= fght amygdala, RA9 surface contact in the second temporal gym,axMs 3,s and 7 for thi and all other depth elecbodes are intermedate. RBI = fgM antetbr h v p u s , RB9 = temporal neomrtexabng the second temporal QYRIS, RFSl= fgM antelior afgulatetqbn, RFS9 = m k 4surface of the second trontal gyms. RFOI = Mht oibiiefmntalcortex, RF09 = cmtd surface of the third frontal gyms. Not& late seizure spcead to the amygdalcd and hippocampalsbuctures of the right kmlsphere.

86

Extratemporal Epilepsy Temporal lobe like ictal behavioural manifestations are frequently “added” to the frontal lobe ictal features, as a result of seizure spread to the temporal lobe. We have previously reported (8) that the incidence of automatism is 4 % when the seizure discharge remains confined to the frontal lobe and reaches 55 % when the ictal discharge spreads to the temporal lobe(s). From a physiopathological perspective, secondary bilateral synchrony and secondary epileptogenesis (20,24, 39, 42-44) may play a distinctive role in the genesis of bifrontal or generalized spike-and-wave activity, which is so often observed in patients with unilateral frontal lobe epilepsy (see Table 6). Secondary bilateral synchrony is more likely to occur when the primary epileptic focus is located in the mesial parasagittal convexity, in the orbitofrontal cortex or in the cingular region (39, 40, 44).

Subjects and Methods The anatomical distribution of the interictal epileptic abnormality was studied in 113 patients with frontal lobe epilepsy submitted to long-term monitoring or serial prolonged EEG recordings with extracranial electrodes. Four sub-groups of patients were identified (38): Sub Group A: Includes 34 patients (mean age: 25.4 years) who became and persisted seizure-free after restricted surgical removal of the anterior frontal region (AF= 18 patients), the parasagittal convexity (PS = 10 patients) or the frontal-opercular region (FO = 6 patients). The postsurgical follow-up ranged from 2 to 46 years. This group represents a “pure culture” of the EEG and clinical manifestations of frontal lobe seizures originating in these regions (16). A more detailed report on these patients is available elsewhere (12, 26).

Sub Group B: Comprises 22 patients (mean age: 23 years) presenting with poorly controlled complex partial seizures of frontal lobe origin who underwent pre-operative longterm EEG monitoring. The distribution of the interictal and ictal findings in these patients is reported in further details elsewhere (36). Sub Group C: Consists of 12 patients with frontal lobe epilepsy (mean age: 27 years), in whom pre-operative EEG investigation with extracranial electrodes failed to provide reliable localization of the anatomical substrate of seizure onset (see 14 for more details). Subsequently, these patients underwent stereotactic implantation of chronic intracerebral and epidural electrodes (33). Sub Group D: Comprises 45 children (median age 1 1 years, range: 6 months to 15 years) who underwent preoperative EEG investigation and subsequent frontal lobe excisions at the Montreal Neurological Hospital between 1940 and 1980 (14, 18). Patients belonging to groups B and C were submitted to pre-operative long-term EEG-video monitoring utilizing a 16 channel cable telemetry system according to a technique described elsewhere (45-48). The main emphasis of the EEG investigation was placed in the localization of ictal onsets and in its correlation with the distribution of the interictal epileptic abnormality. The pre-operative EEG investigation in patients belonging to groups A and D consisted mostly of prolonged (2-4 hours) daytime EEG recordings with special extracranial electrodes. Seizures were recorded only in 9/34 patients of group A and in 21/45 patients of group D. The localization of the interictal epileptic abnormality (spikes, spike-and-wave, sharp-slow waves) obtained during extracranial EEG recordings was classified in the

Tabk VI.

Distribution of Interictal Spiking in Patients with Frontal Lobe Epilepsy Group

Patients

A

34 22 12 45 113

B C

D

F 3 3 6 2 14 (12%)

ML

HS

BI

BS

14 12 4 6 -

2 4 4 18 -

4 0 0 6 -

0 1 5 4 -

36 (32%)

28 (24%)

10 (9%)

10 (9%)

16 0 6 20 42 (37%)

L

The distribution of the interictal epileptic abnormality obtained during extracranial EEG recordings or at electrocorticography was classified in the following categories: Focal (F), lobar (L), multilobar (ML), hemispheric (HS), bifrontal independent (BI) and bilaterally synchronous (BS). (reprcduced & permissionfrom Quesney et aJ, 1991)

87

Quesney following anatomical categories: focal (F), lobar (L), multilobar (ML), hemispheric (HS), bifrontal independent (BI) or bilateral synchronous (BS) (see Table 6).

Results Distribution of the interictal spiking The anatomical distribution of the interictal epileptic abnormality in adult and paediatric patients with frontal lobe epilepsy is detailed in Table 6. In the adult group, only a few patients belonging to sub-group A presented with focal spiking (3/34 = 9 %). Fourteen patients belonging to this group exhibited lobar spiking (41 %). Multilobar or hemispheric spiking was recorded in 6/34 patients (1 8 %). Sixteen patients (47 %) exhibited either bifrontal or generalized bilaterally synchronous interictal epileptic abnormality, most commonly of the spike-and-slow wave type. A similar result was obtained in patients belonging to sub-group B (Table 6) revealing a definite predominance for lobar spiking (12/22 patients = 55 %). Patients belonging to sub-group C (Table 6) presented with a rather even distribution of the interictal spiking in the focal lobar and multilobar anatomical categories. The distribution of the interictal epileptic abnormality in children with frontal lobe epilepsy is summarized in Table

6, section D. Patients belonging to this group presented predominantly with multilobar spiking (18/45 patients = 40 %). Only two patients (4 %) exhibited focal interictal epileptic activity in their EEGs. The poor localization of the interictal epileptic disturbance seems to be an age dependent phenomenon since it was mainly observed in children under ten years of age (18). The incidence of bilateral synchrony was approximately the same as that observed in group A of the adult population. A good indicator of the localizing effectiveness obtained by means of the interictal epileptic abnormality in frontal lobe epilepsy is available in the summary of Table 6, comprising 113 patients belonging to the four different subgroups previously discussed. Only 14 of these patients (12 %) presented with focal interictal spiking. In contrast, 28 patients (24 %) exhibited multilobar epileptic discharges and the most prevalent interictal pattern consisted of bilateral synchronous epileptiform potentials, as documented in 42 patients (37 %).

lctal EEG findings The effectiveness of long-term EEG monitoring in the localization/lateralization of seizure onset in patients belonging to group B was far from satisfactory (36). Only a minority of seizures recorded with extracranial

Table VII.

FOIIOW-IJP Patients with non-tumoral lesions operated upon from 1928 through 1980 Group Seizure free since discharge

0

Became seizure free after some early attacks ._.___._._._._._.-.___._._

1

Seizure free 3 or more years then rare or occasional attacks 3 -.

4

_____._.-.-.-.-.-.-._.___

112 pts. (44%)

Operative deaths 4pts. Died in first 2 postop. years 6 pts. Inadequate follow-up data 16 pts.

Total (reproducedWUI permissionfrcin Rasrnussen, 1991)

88

145 pts. (56%)

(30%)

Marked reduction of seizure tendency Moderate of less reduction of seizure tenency

68 pts. (26%)

283 pts.

257 pts. with follow -up data of 2-49 yrs. (median 16 yrs.)

Extratemporal Epilepsy electrodes exhibited a focal (22 %) or a lateralized (1 1 %) EEG onset. Bilateral or generalized seizure onset was the most frequent finding (37 %). Approximately 11 % of frontal lobe seizures were not associated with d e f ~ t ictal e EEG changes. Ictal recordings obtained with extracranial electrodes were available in 8/12 patients of group C. A focal onset was documented only in 1/8 patients. A regional seizure onset (FCT) was recorded in 3/8 patients and a bifrontal (BiF) seizure onset was identified in 4/8 patients. Most of the recorded seizures disclosed generalized ictal EEG changes at their onset. It is also important to realize that several of the patients’ habitual seizures including warnings and single partial motor attacks, were not associated with definite ictal electrographic changes. Seizures were recorded in 21/45 patients of group D. A focal or regional EEG onset was documented in 8/21 children (38 %) and in the remaining 13 patients, the seizures could not be lateralized. Our fmdings are comparable to other reports in the literature (49, 50) illustrating a poor yield of seizure onset localization in patients with frontal lobe epilepsy subjected to long-term EEG monitoring with extracranial electrodes.

Surgical Approach and Results The results of frontal cortical excisions on seizure trend were extracted from a study comprising 257 patients with non-tumoral lesions operated on at the Montreal Neurological Hospital between 1928 - 1980 (Table 7) (6). Sixty eight patients (26 %) became and remained seizure free (follow up category 0 and 1). Seventy seven additional patients (30 %) experienced a marked reduction of the seizure tendency (follow up categories 2 and 3). Thus 56 % of the patients, experienced a complete or a marked reduction of seizures following frontal cortical excisions. A less successful group comprising 1 12 patients (44 %) experienced a moderate or less than moderate reduction in seizure trend. The results of different types of surgical excisions upon seizure tendency is summarized in Table 8. .Frontal Lobe Epilepsy - Results of Varbus Types of Cork4 Ex&ions. 253 Pakm hnon-bmmleplleptogenc lesions operated upon 1928 through 1980.

The latter study showed no positive correlation between a given pattern of surgical removal and satisfactory outcome (6). The low incidence of seizure free cases (26 %) in this series, may be related not only to the previously discussed limitations defying a reliable pre-operative localization of the epileptic zone, but also to an incomplete surgical removal of the epileptogenic brain tissue, particularly if it involves non resectable cortical areas. The former experience refers to patients without brain tumours. Stereotactic lesional surgery (lesionectomy) with or without resection of the “margins” has proven to be a safe and effective procedure in terms of seizure control, particularly in patients with extratemporal epilepsy due to lesions closely located to a non-resectable brain region

(51). Anterior callosotomy has also been employed as a surgical therapy in frontal lobe epilepsy (52). The indication for such an approach is dependent upon sEEG recordings demonstrating frontal lobe bilateral synchrony or localized seizure onset in one frontal lobe with rapid seizure propagation to homologous contralateral structures, in the absence of an underlying pathology. Anterior callosotomy has also been performed in patients in whom a large frontal corticectomy could not be performed due to encroachment with non-resectable cortical areas. Anterior callosotomy should be considered as a palliative surgical technique in patients with medically resistant frontal lobe epilepsy.

Localizationof the seizure generator in parietal and occipital lobe epilepsy Clinical and EEG presentation Patients with parietal and occipital lobe epilepsy represent 6 and 1 % of large neurosurgical series respectively (53; see Table 9). The clinicaland EEG presentation of occipital lobe epilepsy has been the subject of two recent reviews (54,55). Williamson’s study was performed on 25 patients (12 females, 13 males) with medically intractable occipital

Table VIII.

FU Groups FU Groups FU Group Number of 0+1 2+3 4 Patients (seizure free)

(marked reduction)

(moderate to no reduction)

~

Frontal Lobectorny 19 (32%) Anterior Frontal 22 (47%) Convexity 12 (35%) Parasaggital 9 (18%) Frontal Plus 6 (10%)

19 (32%)

Total pts.

72(29%)

68(26%)

13 (27%) 7 (21%) 18 ( 3 6 9 ) 15 (24%)

(reproduced\Nilh permissionfmm Rasmussen, 1991)

21 12 15 23 42

(369) (25%) (44%) (46%) (66%)

59 (100%) 47 (1009) 34 (100%) 50 (1009) 63 (100%)

113(45%) 253(100%)

Talk IX

Montreal NeurologicalInstitute Surgical Seizure Series Anatomical Classification. patients operated upon 1929through 1480

Temporal lobe Frontal lobe Central (sensorimotor) region Parietal lobe Occipital lobe Large rnultrlobe lesions

Total

121 0 pts (56%) 402 pts (1 8%) 151 pts (7%) 141 pts (6%) 30 pts (lo//.) 243 pts (1l o l o ) 21 77 pts

(reproduced\Nilh permission from Rasmussen, 1591)

89

Pz-P4 \ P4- P6 P6-T6 \ T6-T8 T3-T1 A T1 -T5 T 5 T O 1 5 T O l O l T T4-T2 T2-T6 -I. 70 MV T6-TO2 T02-02

Y

P5-P3 P P3-Pz

-

T7-T5 5 T5-P5

D.H. 86-0801

4

J

v1

8

a

Extratemporal Epilepsy

Fg 3B Referentiald m r a p h y ( E M ) baarg in the same pabent showirg regional $king d~ an@ihi& predominanceat elecbodes 8,4,12 and 3. Numencalsin Ihe bmin @ram represent Ihe ambrnkal W i n of he recordingdecb&.

91

Quesney lobe epilepsy subjected to pre-operative EEG monitoring and to surgical therapy at the Yale Epilepsy Center (55). The mean age of seizure onset was 11 years and the mean duration of seizures was 15 years. Awareness of the initial ictal signs and symptoms would have predicted an occipital lobe seizure origin in 22 of the 25 patients reported. The reliability of the initial ictal behavioural manifestations regarding the localization of the seizure generator was also reported in a review of 29 patients with occipital lobe epilepsy submitted to surgical therapy at the Montreal Neurological Hospital (54). The Yale group (55)proposes that the nature of the ictal pattern in occipital lobe epilepsy is dependent upon the mechanism of seizure spread. A predominant temporal lobe like ictal pattern compatible with a complex partial seizure disorder including automatisms, will often be the main clinical presentation of occipital lobe attacks undergoing seizure spread to the temporal lobe. Conversely, a predominant frontal lobe seizure pattern consisting of asymmetric tonic or clonic motor attacks, will commonly occur following seizure spread to the frontal lobe. Furthermore, a clinical presentation comprising multiple seizure patterns can be observed not only in different patients, but also in individual cases. A recent review of the clinical manifestations of parietal lobe seizures based on 800 ictal events recorded in 145 patients submitted to sEEG investigation is the subject of a scholar review available elsewhere (56).

Despite recent technological developments in longterm EEG monitoring, the localization of epileptogenic zones in the parietal and occipital lobes is not devoid of similar technical and biophysical limitations as those reported for the frontal lobe. Several reports in the literature have demonstrated a poor localization of the interictal epileptic abnormality, as well as an unreliable seizure onset localization in patients with occipital lobe epilepsy (53-55). This finding may be dependent upon some of the following limitations: A. As in frontal lobe epilepsy, only a small portion of the occipital lobe is accessible to EEG recording with scalp electrodes, a situation which involves significant risk' of sampling error whenever attempting to record EEG potentials generated in the occipital lobe's depth. Small epileptogenic foci involving the mesial occipital structures or lying in the depth of some occipital lobe sulci, may not be at all recorded, or they may reflect as widespread epileptic discharges involving the temporo-occipital or parieto-occipital regions, thus appearing as a large epileptogenic zone (54, 55, 57). B. The existence of a functional network of pathways permitting seizure spread outside the occipital lobe, may explain the regional and at times multilobar distribution of the interictal spiking and of the ictal onsets, thus defying an accurate EEG localization. The inferior longitudinal fasciculus provide functional connectivity between the infra-calcarine portion of the occipital lobe and the temporal lobe (41). The superior longitudinal fasciculus, as

Table X .

Central, Parietal and Occipital Non-Tumoral Epileptogenic Lesions Results of Cortical Excisions Seizure free since discharge Became seizure free after some early attacks Seizure free 3 or more years, then rare or occasional attacks Marked reduction of seizure tendency

28 pts. 14 pts.

-.-.-

Moderate or less reduction of seizure tendency

81 pts.

Operative deaths Died in first two postoperative years Inadequate follow-up data

Total (reproducedMI permission from Rasrnussen, 1991)

92

31 pts. 32 pts.

4 pts. 3 pts. 10 pts.

203 pts.

]

1

63 pts. (34%) 42 pts. (23%) (44%)

105 pts (56%)

186 pts. with follow-up data of 2-51 yrs. (median 18 yrs.)

Extratemporal Epilepsy well as the superior and inferior occipito-frontal fasciculus connect the supra-calcarine portion of the occipital lobe with frontal lobe structures. A widespread distribution of the interictal epileptic abnormality is also a common trend in parietal lobe epilepsy (57; Figures 3A and B). A mechanism of secondary bilateral synchrony and secondary epileptogenesis (42, 43) may play a distinctive role in the genesis of bilaterally synchronous and of bilateral independent occipital epileptiform discharges. A recent electrophysiological study provides persuasive evidence that the bilateral synchronization of occipital lobe electrical discharges is mediated by the corpus callosum (58).

Surgical Approach and Results The results of surgical excisions in 203 patients with nontumoral central, parietal and occipital lobe epilepsy is summarized in Table 10 (53). Sixty three patients (34 %) became seizure free after discharge or after some early attacks. Forty two patients (23 %) experienced a marked reduction of seizure trend. A moderate or less than moderate reduction in seizures was documented in 81 patients (44 %). Sixteen of the 25 cases of occipital lobe epilepsy recently described by Williamson et al. (55) presented with well circumscribed lesions. Excellent results were reported in 14/16 patients who underwent lesionectomy. A recent report of 29 patients with occipital lobe epilepsy submitted to surgical resection of the epileptogenic brain tissue (59) documented that nine patients remained seizure free, five other became seizure free after some early attacks and eight additional patients exhibited a marked reduction in seizure Frequency. Thus, at least 75 % of patients with occipital lobe epilepsy benefited from surgical excision of epileptic brain tissue in the occipital lobe.

Summary To summarize, the most commonly encountered problems in the pre-operative EEG localization of the epileptogenic zone in patients with extra-temporal epilepsy are dependent upon: 1) Poor EEG localization of the interictal epileptic abnormality and poor localization of seizure onset. 2) Presence of a widespread epileptogenic area during interictal and ictal tracings, often involving two or more lobes synchronously or independently, thus suggesting a multilobar or a multifocal epileptic disorder. 3) Absence or paucity of interictal epileptiform discharges is not an uncommon finding and clinical seizures without accompanying ictal EEG manifestations are also not uncommon. 4) Different paths of seizure propagation might occur even in individual patients. This may reflect clinically as

different seizure patterns resembling a multifocal seizure problem. From this review, one could envisage that the future research challenge in extra-temporal epilepsy resides in the development of a technology or system which will provide a better localization of the seizure generator, as well as a better identification of the mechanisms of seizure spread, ideally indicating anatomical pathways involved.

Acknowledgements The author is a recipient of a Killam Scholarship granted to perform research in epilepsy at the Montreal Neurological Institute, McGill University. The author thanks Mr. Charles Hodge and Miss Susan Kaupp for the photography work. Typing of the manuscript was done by Manon Gagnon.

References 1. Horsley Sir V. Brain surgery. Br Med J 1886: 2: 670-75. 2. Feindel W. Development of surgical therapy of epilepsy at the Montreal Neurological Institute. Can J Neurol Sci 1991: 18: 549-53. 3. Andermann F. Identification of candidates for surgical treatment of epilepsy. In: Engel J Jr, ed. Surgical Treatment of the Epilepsies. New York: Raven Press 1987: 51-70. 4. Engel J Jr,. Approaches to localization of the epileptogenic lesions. In: Engel J Jr, ed. Surgical Treatment of the Epilepsies. New York: Raven Press 1987: 75-95. 5. Engel J Jr,. Outcome with respect to epileptic seizures. In: Engel J Jr, ed. Surgical Treatment of the Epilepsies. New York: Raven Press 1987: 553-71. 6. Rasmussen T. Tailoring of cortical excisions for frontal lobe epilepsy. Can J Neurol Sci 1991: 18: (4):606-10. 7. Geier S, Bancaud J. Talairach J, Bonis A, Szikla G , Enjelvin M. The seizures of frontal lobe epilepsy. Neurol 1977: 27: 951-58. 8. Quesney LF, Kneger C, Leitner C, Gloor P, Olivier A. Frontal lobe epilepsy: clinical and electrographic presentation. In: Porter RJ et al. eds. Advances in Epileptology: XVth Epilepsy International Symposium. New York: Raven Press 1984 503-8. 9. Quesney LF. Seizures of frontal lobe origin. In: Pedley TA et al. eds. Recent Advances in Epilepsy - 3. Churchill Livingstone. 1986: 81110.

10. Williamson PD, Wieser HG, Delgado-Escueta AV. Clinical characteristics of partial seizures. In: Engel J Jr, ed. Surgical Treatment of the Epilepsies. New York: Raven Press 1987: 101-20. 11. Bancaud J, Talairach J. Clinical semiology of frontal lobe seizures. In: Chauvel Pet al. eds. Advances in Neurology, Vol57. New Y o r k Raven Press 1992: 3-58. 12. Quesney LF, Constain M, Rasmussen T. Seizures from the dorsolateral frontal lobe. In: Chauvel P et al. eds. Advances in Neurology, Vol. 57. New York: Raven Press 1992: 233-43. 13. Quesney LF, Olivier A. Preoperative EEG evaluation in frontal lobe epilepsy. Acta Neurol Scand 1988: 78 (Suppl 117): 61-72. 14. Quesney LF, Constain M, Fish DR, Rasmussen T. Frontal lobe epilepsy: A field of recent emphasis. Am J EEG Techno1 Aset. 1990: 30: 177-93. 15. Quesney LF, Constain M, Rasmussen T, Stefan H, Olivier A. How large are frontal lobe epileptogenic zones? EEG, ECoG and sEEG evidence. In: Chauvel P et al. eds. Advances in Neurology, Vol57. New York: Raven Press 1992: 31 1-23.

93

Quesney 16. Rasmussen T. Characteristics of a pure culture of frontal lobe epilepsy. Epilepsia 1983: 24: 482-93. 17. Gloor P. Commentary: Approaches to localization of the epileptogenic lesion. In: Engel J Jr, ed. New York: Raven Press 1987: 97-100. 18. Quesney LF, Fish DR, Rasmussen T. Extracranial EEG and electrocorticography in children with medically refractory partial seizures. J Epilepsy: 1990: 3 (Suppl): 55-67. 19. Ajmone-Marsan C, Goldhammer L. Clinical ictal patterns and electrographic data in cases of partial seizures of frontal-centralparietal origin. In: Brazier M A B ed. Epilepsy, its Phenomenon in Man. New York: Academic Press: 1973: 235-58. 20. Goldring S. The Role of prefrontal cortex in grand ma1 convulsion. Arch Neurol 1972: 26: 109-19. 21. Ludwig BI, Ajmone-Marsan C, Van Buren J. Cerebral seizures of probable orbitofrontal origin. Epilepsia 1975: 16: 141-58. 22. Ludwig BI, Ajmone-Marsan C. Van Buren J. Depth and direct cortical recording in seizure disorders of extratemporal origin. Neurology 1976: 2 6 1085-99. 23. Mazars G. Criteria for identifying cingulate epilepsies. Epilepsia 1970: 41-7. 24. Pedley TA, Tharp BR, Herman K. Clinical and electroencephalographic characteristics of midline parasagittal foci. AM Neurol 1981: 9: 142-49. 25. Commission on Classification and terminology of the International League Against Epilepsy. Proposal for Revised Classification of Epilepsies and Epileptic Syndromes. Epilepsia 1989: 30 (4): 389-99. 26. Quesney LF, Constain M, Fish DR, Rasmussen T. The clinical differentiation of seizures arising in the parasagittal and anterolateral dorsal frontal convexities. Arch Neurol 1990 47: 67779. 27. Sutherling WW. Risinger MW, Crandall PH et al.. Focal functional anatomy of dorsolateral frontocentral seizures. Neurology 1990: 40: 87-98. 28. Williamson PD, Spencer SS. Clinical and EEG features of complex partial seizures of extratemporal origin. Epilepsia 1986: 27 (Suppl 2): S46463. 29. Williamson PD, Spencer DD, Spencer S S , Novelly RA, Mattson RH. Complex partial seizures of frontal lobe origin. AM Neurol 1985: 18: 497-504. 30. Chauvel P, Trottier S , Vignal JP, Bancaud J. Somatomotor seizures of frontal lobe origin. In: Chauvel P et al. eds. Advances in Neurology. New York: Raven Press 1992: 57: 185-232. 3 1. Munari C, Bancaud J. Electroclinical symptomatology of partial seizures of orbital frontal origin. In: Chauvel P et al. eds. Advances in Neurology. New York: Raven Press 1992: 57: 257-65. 32. Rougier A, Loiseau P. Orbital frontal epilepsy: a case report. J Neurol Neurosurg Psych 1992: 51: 146-7. 33. Quesney LF, Constain M, Rasmussen T. Olivier A, Palmini A. Presurgical EEG investigation in frontal lobe epilepsy. Proceedings of the Consensus Development Co.:nference on Surgery for Epilepsy. March 19-20, 1990. NIH Bethesda. h press. 34. Fegersten L, Roger A. Frontal epileptogenic foci and their clinical correlations. Electroenceph Clin Neurophysiol I961 : 13: 905-13. 35. Quesney LF, Katsarkas A, Gloor P, Andermann F. Contribution of naso- ethmoidal electrode recording in the electrographic exploration of frontal and temporal lobe epilepsy. In: Dam M et al. eds. Advances in Epileptology: XIIth Epilepsy International Symposium. New York: Raven Press 1981: 293-304. 36. Quesney LF, Gloor P. Localization of epileptic foci. In: Gotman J et al. eds. Long-Term Monitoring in Epilepsy. Amsterdam: Elsevier 1985: (SUPPI37): 165-200. 37. Quesney LF. Extracranial EEG evaluation. In: Engel J Jr, ed. Surgical Treatment of the Epilepsies. New York: Raven Press 1987: 129-66.

94

38. Quesney LF. Preoperative electroencephalographic investigation in frontal lobe epilepsy: electroencephalographic and electrocorticographic recordings. Can J Neurol Sci 1992: 18: 559-63. 39. Ralston BL. Cingulate epilepsy and secondary bilateral synchrony. Electroenceph Clin Neurophysiol 1961: 13: 591-8. 40. Tharp BR. Orbital frontal seizures. A unique electroencephalographic and clinical syndrome. Epilepsia 1972: 13: 627-42. 41. Schneider RC, Crosby EC, Farhat SM. Extratemporal lesions triggering the temporal-lobe syndrome. J Neurosurg 1965: 22: 24663. 42. Morrell F. Secondary epileptogenesis in man. Arch Neurol 1985: 42: 3 18-35. 43. Morrell F. Varieties of human secondary epileptogenesis. J Clin Neurophysiol 1989: 6(3): 227-75. 44. Tukel K, Jasper H. The electroencephalogram in parasagittal lesions. Electroencephalograph Clin Neurophysiol 1952: 4: 481-94. 45. Gotman J, Gloor P. Automatic recognition and quantification of interictal epileptic activity in the human scalp. Electroenceph Clin Neurophysiol 1976: 41 : 5 13-29. 46. Gotman J, Ives JR, Gloor P. Automatic recognition of interictal epileptic activity in prolonged EEG recordings. Electroenceph Clin Neurophysiol 1979: 40: 510-20. 47. Gotman J. Automatic recognition of epileptic seizures in the EEG. Electroenceph Clin Neurophysiol 1982: 54: 530-40. 48. Gotman J, Ives JR, Gloor P, Quesney LF, Bergsma P. Monitoring at the Montreal Neurological Institute. Ln: Gotman J et al. eds. Long Term Monitoring in Epilepsy (EEG Suppl37). Amsterdam: Elsevier 1985: 327-40. 49. Spencer SS, Williamson PD, Bridges SL, Mattson RH, Cicochetti DV, Spencer DD. Reliability and accuracy of localization by scalp ictal EEG. Neurology 1985: 35: 1567-75. 50. Swartz BE, Walsh GO, Delgado-Escueta AV, Zolo P. Surface ictal electroencephalographic patterns in frontal vs temporal lobe epilepsy. Can J Neurol Sci 1991: 18: 549-53. 5 1. Cascino GD, Kelly PJ, Hirschorn KA, Marsh WR, Sharbrough FW. Stereotactic resection of intra-axial cerebral lesions in partial epilepsy. Mayo Clinic Proceedings 1990: 65: 1053-60. 52. Turmel A, Giard N, Bouvier G et al.. Frontal lobe seizures and epilepsy: indications for cortectomies or callosotomies. In:Chauvel Pet al. eds. Advances in Neurology. New York: Raven Press 1992: 57: 689-705. 53. Rasmussen T. Surgery for central, parietal and occipital epilepsy. Can J Neurol Sci: 1991: 18: 611-6. 54. Salanova V, Andermann F, Olivier A, Rasmussen T, Quesney LF. Occipital lobe epilepsy: electroclinical manifestations in 29 patients treated surgically. 43rd Annual Meeting AAN. Neurology 1991: 41 (Suppl 1): 262. 55. WilliamsonPD,Thadani VM,Darcey TM,SpencerDD,SpencerSS, Mattson RH. Occipital lobe epilepsy: clinical characteristics, seizure spread patterns and results of surgery. AM Neurol 1992: 31: 3-13. 56. Bancaud J, Talairach J. Munari C, Giallonardo T, Brunet P. Introduction I'itude clinique des crises ipileptiques ritrolandiques. Can J Neurol Sci 1991: 18: 566-9. 57. Williamson PD, Boon PA, Thadani VM et al. Parietal lobe epilepsy: diagnostic considerations and results of surgery. Ann Neurol 1992: 31: 193-201. 58. Engel AK, Konig P, Kreiter AK, Singer W. Interhemispheric synchronization of oscillatory neuronal responses in cat visual cortex. Science 1991: 52: 1177-9. 59. Olivier A, Salanova V, Andermann F, Rasmussen T, Quesney LF. Electrocorticography (ECoG), cortical stimulation, and surgical outcome in 29 patients with occipital lobe epilepsy. 43rd Annual Meeting AAN. Neurology 1991: 41 (Supp 11): 365.