вход по аккаунту


Depth electrode investigations in patients with bitemporal epileptiform abnormalities.

код для вставкиСкачать
Depth Electrode Investigations in Patients
with Bitemporal Epilepthorm Abnormahties
Norman So, MD, Pierre Gloor, MD, PhD, L. Felipe Quesney, MD, PhD, Marilyn Jones-Gotman, PhD,
Andre Olivier, MD, PhD, and Frederick Andermann, MD
Fifty-seven patients showing bitemporal independent epileptiform abnormalities on extracranial electroencephalograms (EEGs) in whom the epileptogenic zone could not be localized or lateralized by extracranial EEG and other
noninvasive tests were investigated with stereotactic depth electrode recordings. In a majority of 44 patients (77%),
seizures originated exclusively or with a strong predominance in one temporal lobe only. Of the remaining 13 patients
(23%), 8 had seizures originating independently in either temporal lobe without significant lateralized predominance,
and 5 had multiple seizure patterns, which were often diffuse. The patterns of seizure onset as recorded by depth
electrodes tended to vary even in the same patient. Electrical stimulation studies and the determination of afterdischarge thresholds were of limited utility for lateralization of seizure onset.
So N, Gloor P, Quesney LF, Jones-Gotman M, Olivier A, Andermann F. Depth electrode investigations in
patients with bitemporal epileptiform abnormalities.
Ann Neurol 1989;25:423-431
Bilateral independent interictal epileptiform abnormalities involving both temporal regions in extracranial
electroencephalography (EEG) studies are found in 20
to 35% of patients with temporal lobe epilepsy 11-41.
This often makes it difficult to lateralize the epileptogenic zone responsible for a patient’s seizures based on
the results of extracranial EEG and other noninvasive
studies. Frequently such inconclusive findings in a patient evaluated for surgical treatment lead to a decision
not to operate.
Intracranial depth electrode EEG investigation offers the possibility of demonstrating that seizures arise
from only one temporal lobe in some of these patients
and thus helps to identify appropriate candidates for
surgical treatment. Earlier studies from our institution
on the use of ster?otactic depth electrodes for the investigation of pat ents with bitemporal epileptiform
abnormalities sho7 red encouraging results in this regard 15, 61. We now report on our complete series of
57 consecutive patients investigated for this problem
between 1972 and 1986.
The study group consisted of 57 patients investigated with
bilateral stereotacticallyimplanted depth electrodes who suffered from medically intractable seizures of presumed temporal lobe origin, and who showed bilateral independent
epileptiform discharges over the temporal regions as the major interictal epileptiform abnormality in extracranial EEG
From the Montreal Neurological Institute, Department of Neurology and Neurosurgery, McGill University, Montreal, Quebec,
studies. Stereotactic depth electrode investigation was undertaken because of a lack of consistent agreement in the lateralization and occasionally the precise temporal localization of
the epileptogenic zone by extracranial interictal and ictal
EEG findings, clinical assessment, imaging studies, and neuropsychological evaluation. Patients with nonlateralized
bitemporal epileptiform abnormalities in the extracranial
EEG represent approximately 70% of all patients investigated by intracranial EEG recordings at our institution. Patients with extratemporal foci or multifocal abnormalities
were excluded from this report.
Stereotactic Depth Electrode Investigations
The technique of stereotactic depth electrode implantation
used at the Montreal Neurological Institute has been described previously 177. Electrodes were made of semiflexible
tresses of stainless-steelwires, with 9 contacts located 5 mm
apart. Each contact formed a wire loop of 0.52 mm diameter
and 0.1 mm thickness. For the temporal lobes, 2 or usually 3
pairs of electrodes were inserted symmetrically by an orthogonal approach. They penetrated the temporal lobe horizontally through the second temporal convolution, and their tips
reached the amygdala, the anterior hippocampus, and in
some patients the midportion of the hippocampus and
parahippocampal gyms (Fig 1). It was therefore possible to
record the intracranial EEG from the temporal limbic structures as well as from surface and deep sulcal temporal neocortex. In 39 of the 57 patients (68%), additional extratemporal electrodes were implanted because significant participation of extratemporal structures could not be entirely
excluded. Thus in 38 patients, one or more pairs of elec-
Received for publication Jul 11, 1988, and in revised form Oct 7.
Accepted for publication Oct 14, 1988.
Address correspondence to Dr Gloor, Montreal Neurological Institute, 3801 University St, Montreal, Quebec H3A 2B4, Canada.
from analysis, because it was often difficult to separate these
from frequently recurring interictal discharges. Seizures precipitated by electrical stimulation were not included in the
final analysis and chemical activation of seizures was not
SDEEGs were analyzed visually by experienced electroencephalographers (P.G. or L.F.Q., and in a few patients by Dr
Rachel Ochs). Bitemporal independent interictal spiking in
SDEEGs was seen in all patients and was usually abundant.
Its relative lateralization was, however, highly variable,
changing with the sleep-wake cycle and after seizures. This
made it unsuitable for determination of the principal epileptogenic zone($. The latter was thus defined on the basis of
the site or sites of seizure onset, which was determihed for
each clinical attack. SDEEG-recorded seizure onsets were
classified as:
Fig I . Commonly used electrode arrangement (only left side illustrated; electrodes also symmetrically inserted on the right side).
Electrodes are directed respectively at the lt$t amygdala (LA),
anterior hippocampus (LB), left midhippocampuslparahippocampal gyrus (LC), ldt superior mesial frontal cortex (LFM), and
left orbitalfrontal cortex (LFO). (From Gloor P et al {lo).)
trodes were symmetrically inserted horizontally into the
frontal lobes, usually directed at the orbitofrontal, anterior
cingulate, and supplementary motor regions. One patient
had additional electrodes implanted unilaterally into the posterior temporal and occipital regions because extracranial
EEGs had revealed epileptiform abnormalities over those
The stereotactic depth EEG (SDEEG) was recorded directly or by cable telemetry on 16 or occasionally on 32
channels together with video monitoring almost continuously throughout the 24 hours of each day, usually for several weeks. Montage selection always included channels recording from temporal limbic and neocortical strucrures
bilaterally. No simultaneous extracranial EEG was recorded
in order to minimize the risk of infection. Anticonvulsant
medications were usually partially or completely withdrawn
in the course of SDEEG recording. This has been shown not
to result in any significant false localization or lateralization of
seizure onset {8, 91. Only spontaneous seizures with clinical
symptoms or signs, the initial electrographic changes of
which were recorded and not made uninterpretable by artifact, were studied. For this purpose, auras accompanied by
ictal electrographic changes were rated as seizures. Sustained
rhythmic discharges of various kinds without clinical accompaniments labeled as “electrographic seizures” were excluded
424 Annals of Neurology
Vol 25
No 5
May 1989
1. Focal temporal when the seizure discharge started in one
discrete anatomical area within the temporal lobe, for example, the amygdala
2. Regional temporal when the ictal discharge started simultaneously in 2 or more adjacent anatomical areas within the
temporal lobe, for example, limbic involvement of both
the amygdala and the hippocampus or simultaneous limbic and neocortical involvement
3. Extratemporal for those seizures originating in an area outside of the temporal lobe
4. Widespread unilateral when ictal electrographic onset involved more than one lobe in one hemisphere, for example, frontotemporal onset
5. Bilateral simultaneous when there was simultaneous electrographic onset in the two hemispheres, either in homologous regions or diffusely
Electrical Stimulation Studies
Electrical stimulation through the depth electrode contacts
was performed in 46 patients as described previously [lo}.
Results of electrical stimulation of each patient were reviewed to determine the area with the lowest threshold for
afterdischarges (ADS)and the ability of stimulation to reproduce the patient’s habitual aura(s) and seizure(s). The lowest
current intensity required for eliciting an electrical AD on
stimulating the hippocampal regions (hippocampal AD
threshold) was compared for the two sides. Similarly the
lowest current intensity for eliciting electrical ADS in any
of the temporal limbic structures (limbic AD threshold),
namely the amygdala, anterior hippocampus, and midhippocampal and parahippocampal regions on one side, was
compared with that on the other. Responses resembling the
patient’s habitual warning(s) were rated as auras, whether or
not they were accompanied by ADS. ADS accompanied by
altered responsiveness, impaired memory, automatisms, or
convulsive phenomena were rated separately as electrically
induced clinical seizures.
Extracranial EEG Studies
EEGs using the 10-20 electrode system were recorded on 16
channels directly or by cable telemetry. Flexible, silver-wire
sphenoidal electrodes suitable for long-term use were inserted in all cases. Prolonged recordings were performed in
wakefulness and in sleep. In more recent years, intensive
monitoring by 16-channel cable telemetry and video recording, incorporating computerized automatic spike and seizure
detection programs, has been routinely employed 111). A
ratio of lateralization was established for bitemporal independent interictal epileptiform abnormalities. This was based on
the number of recordings, obtained serially over weeks to
years, showing a lateralized predominance of interictal
epileptiform discharges as determined by visual analysis.
Within each recording, the frequency ratio of discharges on
the two sides was usually less than 4 : 1. The presence of
independent extratemporal epileptiform discharges and of
diffuse or generalized epileptiform discharges was also
noted. A ratio of lateralization was similarly obtained for
nonepileptiform abnormalities, when these were localized or
lateralized as in the case of slow-wave activity. Seizures recorded by extracranial EEG were classified as showing consistent or conflicting lateralization, or as showing bilateral,
diffuse, or uninterpretable onset.
Clinical, Imaging, and Neuropsychological Findings
Etiological factors were determined from case histories,
pathology reports of surgical specimens, and sometimes by
questionnaire and telephone reports. The clinical seizure
symptoms that were known before SDEEG monitoring were
not used as lateralizing evidence because considerable controversy still surrounds the lateralizing significance of certain
clinical ictal phenomena in temporal lobe seizures. Signs of
cerebral dysfunction found on neurological examination
were used as lateralizing evidence suggesting structural cerebral pathology. Unilateral thickening of the skull vault,
middle fossa asymmetry or calcified lesions on skull x-rays,
ventricular asymmetry and dilatation on pneumoencephalography, and focal lesions or focal atrophic changes on computed tomographic (CT) scans were examples of abnormalities that provided lateralizing information. Mild
ventricular asymmetry on CT scan without other accompanying evidence of cerebral atrophy was not considered
definitely abnormal. To date too few magnetic resonance
studies have been performed in this group of patients for
meaningful analysis. Neuropsychological evaluation and intracarotid sodium amobarbital (Amytal) testing followed established protocols 112, 131. The results of neuropsychological assessment were used to determine the localization and
lateralization of preoperative cerebral dysfunction. Memory
impairment after amobarbital injection was taken as evidence
for contralateral mesial temporal lobe dysfunction.
The chi square test (with correction for continuity when
appropriate) was used for the statistical analysis of discrete
variables and the Student’s t test was employed for continuous variables.
Of the 57 patients in the study, 29 were men and 28
were women. Their ages at the time of SDEEG studies
ranged from 14 to 50 years (mean, 28.9). The duration
of SDEEG recording ranged from 8 to 30 days (mean,
17.5). The number of spontaneous clinical seizures (including auras) recorded by SDEEG in each patient
ranged from 4 to 115 (mean, 21.2). The large number
(n = 56)
50 55 60 6 5 70 75 80 85 90 95 100
Percent Predominance
Fig 2. Percent predominance is the percentage ratio of the number of seizures with ictal onset in the most active epileptogenic
zone (temporal, extratemporal, or widespread unilateral) divided
by the total number of spontaneous seizures recorded by stereotactic depth electroencephalographyin the patient. One patient was
excluded (see text).
in some patients was accounted for by the fact that all
auras accompanied by detectable ictal discharges in the
SDEEG were rated as seizures.
Complications arising from stereotactic depth electrode insertions were encountered in 3 patients (5%).
Two patients developed brain abscesses. This occurred
early during the period of implantation in 1, and the
infection responded to antibiotic therapy. In the other
patient, a brain abscess that required surgical drainage
developed 6 years after electrode removal in the temporal lobe, where an electrode fragment had remained
behind. Both patients recovered with appropriate
treatment. One patient, in a postictal confusional state,
pulled out some electrode strands but experienced no
clinical deficits. No intracranial hemorrhage or death
Spontaneous Seizures Recorded by SDEEG
For each patient the predominance of seizures arising
from the most active epileptogenic zone was expressed
as a percentage of all of his or her recorded spontaneous seizures. The frequency distribution of predominance is illustrated in Figure 2. (One patient was excluded from this figure because her most frequent
seizure type had a diffuse generalized onset, and she
was placed in Group D [see the following paragraph}.)
Based on the localization, lateralization, and predominance of the epileptogenic zones as defined by SDEEG
spontaneous seizure onsets, patients were divided into
4 groups.
Group A (unilateral temporal) included 25 patients
(44%) in whom all seizures originated exclusively in
So et al: Bitemporal Epileptiform Abnormalities I: SDEEG 425
Table 1 . Percentage Distribution of Depth Ictal Onset Patterns
Regional Temporal
Focal Temporal
Am Hipp Neo
A: Unilateral
B: Unilateral predominance
C: Bitemporal
D: Multiple
temporal Unilateral
+ Neo
amygdala; hipp = hippocampus; neo = neocortex.
one temporal lobe. Group B (unilateral temporal predominance) was composed of 19 patients (33%) in
whom 80% or more of all seizures originated in one
temporal lobe. Occasional seizures arose from the contralateral temporal lobe in 12 patients. Infrequent seizures of extratemporal onset were found in 2 patients
(arising from the frontal and occipital lobes, respectively). Three patients had occasional seizures of widespread unilateral frontotemporal onset, and 8 had infrequent seizures of bilateral simultaneous onset. This
group included 2 patients who had fewer than 80% of
their seizures arising from one temporal lobe, but all of
their remaining seizures showed bitemporal simultaneous onset with a strong asymmetrical predominance on
the side from which their seizures with a unilateral
temporal onset arose; no seizures arose independently
from the contralateral temporal lobe. Group C (bitemporal independent without significant lateralized predominance) comprised 8 patients (14%) in whom seizures originated independently in each temporal lobe
but failed to attain 80% predominance on either side.
Infrequent seizures of widespread unilateral fronroremporal onset were recorded in 1 patient and of
bilateral simultaneous onset in another. Groap D (multiple seizure patterns without an exclusively temporal
origin) was composed of 5 patients (9%). Although
these patients all showed some seizures with a temporal onset, others had extratemporal, widespread unilateral, or bilateral simultaneous onset, and no one pattern accounted for 80% or more of the seizures
recorded in the patient.
As many as 44 patients (77%) belonged to Groups
A and B and had seizures that originated exclusively or
with a strong predominance in one temporal lobe. The
largest group (Group A), numbering 25 patients,
showed a strictly unilateral temporal seizure onset. The
smaller number of patients in Groups C and D tended
to cluster on the left of the distribution curve, close to
the 50% mark (see Fig 2).
Thirty-nine of 5 7 patients (68%) had extratemporal
electrodes implanted, usually into the frontal lobes.
Extratemporal electrodes were employed in 68% of
Group A, 58% of Group B, and 75% of Group C
patients (x2 = 0.87, p = 0.65). There is therefore no
426 Annals of Neurology Vol 25 No 5 May 1989
significant bias in any of these groups to have seizures
erroneously locahzed to the temporal lobe from lack of
sampling elsewhere. In Group D, extratemporal electrodes were employed in all cases, in keeping with the
definition of this group. Even with the common practice of frontal electrode insertion, only 1 patient in
Group D had a signlficant number of his recorded
seizures (51%) arising from the frontal lobe.
In a majority of 48 patients (84%), the precise pattern of ictal electrographic onset showed some variability, even when seizures originated exclusively in one
temporal lobe. Most commonly (41 patients, 72%), a
mixture of focal and regional patterns of seizure onset
in the temporal lobe was observed. Thirteen patients
(23%) each had seizures of focal onset arising separately from the amygdala or the hippocampus, and
other seizures of regional onset from the temporal
limbic structures, sometimes simultaneously involving
the temporal neocortex. Stereotyped ictal onset patterns were only seen in 9 patients (16%),who were all
in Group A, and consisted of focal hippocampal onsets
in 6 and regional temporal limbic onsets involving the
amygdala and hippocampus simultaneously in 3. The
distribution of SDEEG ictal onset patterns within each
group is shown in Table 1. Focal temporal onsets were
more frequent in Group A (58%) than in Group B
(28%) or Group C (25%). By comparison, regional
temporal onsets were more frequent in Group B
(66%) and Group C (72%) than in Group A (42%)
(x2 = 81.90, p < 0.001). Focal temporal onsets were
seen more frequently in the hippocampus than in the
amygdala or the temporal neocortex, whereas regional
temporal onsets most commonly involved the temporal limbic structures (amygdala and hippocampus) simultaneously. Seizures with a temporal neocortical onset were uncommon. They occurred in 10 patients but
accounted for no more than 3% of recorded seizures
in any SDEEG group.
Independent extratemporal onsets and widespread
unilateral onsets involving more than one lobe were
mainly seen in Group D patients, with only rare instances occurring in Groups B and C. Bilateral simultaneous onsets occurred in 12 patients (21%),),8 in
Group B, 1 in Group C, and 3 in Group D. In patients
Table 2. Cowelation of Tests with Stereotactic Depth Electroencephalography (SDEEG) Group
(% each group undergoing test)
Concordant Lateralization"
(% each group lateralized)
Extracranial EEG interictal
epileptiform (n = 57)
Extracranial EEG nonepileptiform (n = 53)
Extracranial EEG ictal (n = 5 5 )
Asymmetrical hippocampal
AD threshold (n = 39)
Asymmetrical limbic AD
threshold (n = 43)
Abnormal neurological
signs (n = 56)
Abnormal imaging (n = 56)
Psychological dysfunction
13 (52)
11 (58)
3 (38)
2 (40)
29 (51)
15 (63)
17 (32)
7 (30)
A and B
6 (86)
8 (57)
7 (29)
16 (83)
6 (33)
7 (70)
3 (38)
4 (80)
2 (40)
4 (80)
18 (33)
31 (79)
3 (43)
12 (75)b
19 (86)
7 (64)
4 (80)
3 (60)
33 (77)
13 (68)b
2 (28)b
15 (58)b
4 (16)
2 (11)
8 (14)
4 (100)
2 (100)
6 (100)
12 (48)
15 (60)
8 (44)
16 (84)
3 (38)
5 (63)
5 (100)
24 (43)
41 (72)
9 (75)
9 (60)
8 (57)
5 (42)
0 (0)
14 (44)
5 (63)
7 (88)
10 (63)
16 (80)
19 (61)
(n = 57)
Amobarbital memory dysfunction (n = 32)
7 (54)
"Concordancewith side of exclusive or predominant SDEEG seizure onset.
bElevated A D ipsilateral to side of exclusive or predominant SDEEG seizure onset
AD = afterdischarge.
Table 3. Clinical Responses Obtained During Electrical Stimulation
Habitual Auras
Group A
Group B
Clinical Seizures
Group C
Group D
Group A
Group B
Group C
Group D
*On stimulation of one temporal lobe only, but concordance could not be determined in Group C and D patients.
with such bilateral simultaneous onsets, other seizures
of focal temporal or regional temporal onset were always recorded as the more frequent seizure pattern. In
the 12 Group B and 8 Group C patients who had
seizures recorded independently from each temporal
lobe, focal and regional temporal onsets could arise
from either side. There was no tendency for seizures
to originate in homologous structures in these patients
or for one electrographic pattern of seizure onset to
occur on one side and a different pattern on the other.
Electrical Stimulation Studies
Electrical Stimulation was performed in 46 patients.
Studies were adequate for the determination of hippocampal AD threshold in 39 patients and of limbic
AD threshold in 43. Hippocampal AD thresholds
were asymmetrical in 31 (79%) and symmetrical in 8
(2 1%). When asymmetrical, the hippocampal AD
threshold was found elevated ipsilateral to the side of
exclusive or predominant SDEEG ictal onset in 12 of
16 (75%) Group A patients, but in only 2 of 7 (29%)
Group B patients. Asymmetrical thresholds also occurred in Group C and D patients. Correlation of hippocampal and limbic AD threshold asymmetry with
SDEEG ictal onset groups is presented in Table 2.
The habitual aura(s) was reproduced in 33 patients
(72%) on electrical stimulation, usually of the temporal limbic structures. A variety of other subjective
sensations, which were, however, not recognized as
habitual auras, were elicited in 20 patients. Habitual
auras occurred exclusively on stimulation of one temporal lobe in 17 patients (37%) and on stimulation of
either temporal lobe in 16 (35%). When the aura occurred on stimulation of only one temporal lobe, it was
concordant with the side of SDEEG ictal onset in 8 of
10 Group A patients (80%) and discordant in 2
(20%). Habitual auras on stimulation of only one temporal lobe were also seen in Group C and D patients
who did not have seizures arising predominantly from
one side (Table 3). In contrast, habitual auras were also
obtained on stimulation of either temporal lobe in all
patient groups, including 6 of 16 patients in Group A
So et al: Bitemporal Epileptiform Abnormalities I: SDEEG 427
(38%), who by definition had all SDEEG recorded
seizures arising from one temporal lobe only.
Electrical stimulation elicited seizures in 22 patients
(48%). In 19 patients seizures were induced by stimulation of the limbic structures: the amygdala in 9, the
hippocampus in 6, and either of the two structures in
4. In 2 patients seizures followed stimulation of either
limbic or temporal neocortical structures separately.
One patient had a seizure elicited only on temporal
neocortical stimulation. Seizures were produced by
stimulation of one temporal lobe in 17 patients (37%)
and by stimulation of either temporal lobe in 5 ( 11 %)
(see Table 3). In Group A, electrical stimulation induced seizures from the side of exclusive SDEEG ictal
onset in 4 patients, from the contralateral side in 4, and
from either side in 1. Of the 5 instances in which
seizures were elicited by stimulation of the side contralateral to that of exclusive seizure onset in Group A,
2 patients recognized that these seizures were different
from their habitual attacks, and in the remaining 3
instances, electrical discharge spread to the other temporal lobe. In Group B, seizures were electrically induced from the side ipsilateral to that of predominant
SDEEG ictal onset in 2 patients, from the contralateral
side in 3, and from either side in 2. All electrically
induced seizures in Groups B, C, and D except for 1
were recognized as habitual seizures either by the patient or by the observing physician, regardless of the
side of stimulation. The 1 exception occurred in a
Group D patient.
Results of Noninvasive Studies and Correlation with
SDEEG Groups
CLINICAL FEATURES. The mean age of clinical seizure onset was lower in Group A and B patients (10.6
years) than in Groups C and D (15.4 years), but the
difference did not reach statistical significance. The
mean duration of the seizure disorder and the proportion of patients with or without a major etiological
factor did not differ significantly between Groups A
and B and Groups C and D. A major etiological factor
was found in 25 patients (44%) whereas no definite
etiological factor could be established in 32. Major
etiological factors included a history of one or more
early convulsions before age 3 (frequently associated
with fever), severe head trauma, central nervous system infection, severe perinatal anoxia, and foreign tissue lesion. A history of early convulsion(s) before age
3 (usually febrile) in 18 patients was the single most
common major etiological factor. Its incidence did not
differ significantly among the groups. By contrast it is
notable that the 6 patients who were eventually found
to have a foreign tissue lesion either showed an exclusive (2 in Group A) or predominant (4 in Group B)
seizure onset in the temporal lobe that harbored the
Annals of Neurology
Vol 25
No 5 May 1989
Abnormal laterahzed neurological signs such as hemiparesis, hemisensory impairment, hemiatrophy, and
unilateral impairment of coordinated fine movements
were present in 8 patients (14%). Each of these patients also had concordant lateralized abnormalities as
seen on imaging studies.
EXTRACRANLAL EEG. On average, 15 extracranial
EEG studies (range, 3-40) were performed in each
patient before electrode implantation. Extracranial interictal epileptiform EEG abnormalities showing a 4 : 1
(80%) or greater ratio of predominance on one side as
previously defined were considered lateralized. This
was found in 29 patients (51%). Forty patients (70%)
had extracranial interictal epileptiform EEG abnormalities confined to the temporal regions. Thirteen
(23%) had occasional bilaterally synchronous generalized spike and wave discharges, and 4 (7%) had independent extratemporal foci in addition to the predominant bitemporal abnormalities (temporal-plus). Using
the same criteria, nonepileptiform abnormalities were
lateralized in only 32% of patients.
Seizures were recorded by extracranial EEG in all
except 2 patients (mean, 6.22; range, 0-27), with 3 or
more seizures recorded in 35 patients. Ictal lateralization was possible in only 18 patients (33%). Seizure
onsets arising independently from one or the other
side resulted in conflicting lateralization in 20 patients
(36%). Nonlateralized bilateral or diffuse onsets occurred in 12 patients (22%) and uninterpretable recordings in 5 (9%).
IMAGING STUDIES. Imaging studies showed lateralized findings in 24 patients (43%), bilateral abnormalities in 6 ( I l s ) , and normal results in 26 (46%).
The data for one patient were lost. The abnormalities
seen were usually those of a longstanding atrophic process. Foreign tissue lesions were demonstrated radiographically in only 2 of the 6 patients who harbored
showed evidence for unilateral temporal dysfunction in
8 patients (14%) and evidence for bitemporal dysfunction with predominance on one side in 33 (58%). Thus
lateralized temporal dysfunction was found in 41 patients (72%), bitemporal dysfunction without lateralization in 13 (23%), and normal tests in 3 (5%). Bilateral amobarbital tests valid for the assessment of
memory function were completed in 32 patients, while
no tests were performed on 6. Thirteen patients had
invalid tests on one side and 6 had incomplete unilateral testing only. Of those who completed bilateral
amobarbital tests for memory, mesial temporal lobe
dysfunction was lateralized in 14 (44%) and bilateral
dysfunction was present in 6 patients (19%). Twelve
patients (38%) had normal tests on both sides.
The presence or absence of lateralized abnormalities
as seen on each of the noninvasive tests (see Table 2 )
was not significantly correlated with the subsequent
finding of exclusive or predominant SDEEG ictal onset
(Groups A and B) in one temporal lobe (x2 analysis).
Within SDEEG Groups A and B, concordance with
the side of exclusive or predominant seizure onset was
highest for abnormal neurological signs (100%) and
abnormal imaging findings (80%)when these could be
lateralized. However, lateralized abnormalities were
also found in patients who were later shown not to
have a strong predominance for seizures to originate in
one temporal lobe (Groups C and D). The rate of
concordant lateralization was poor for other tests (see
Table 2). Unexpectedly, the concordance of extracranial interictal epileptiform and nonepileptiform EEG
abnormalities was considerably higher for Group B
than for Group A patients. The topographical distribution of extracranial interictal epileptiform EEG abnormalities was not significantly correlated with the results
of SDEEG recordings. Thirty-two of 40 patients
(80%) with temporal-only discharges in extracranial
EEGs were classified in Groups A or B, as were 12 of
17 patients (70%) who had temporal-plus discharge
(x2 = 0.19, p = 0.66).
An attempt was made to see if combinations of variables based on the results of noninvasive tests could
accurately identify patients in Groups A or B and,
furthermore, correctly lateralize the side of exclusive
or predominant SDEEG ictal onset. This goal was
achieved when there was agreement on the side of
lateralization by extracranial EEG interictal epileptiform abnormality, imaging abnormality, and amobarbital test memory dysfunction or abnormal neurological signs. However, it allowed us to identify only 4 of
the 44 patients who were eventually placed in Groups
A and B.
Our study shows that stereotactic depth electrode investigation was able to reveal that seizures originated
exclusively or with a strong predominance in one temporal lobe in 77% of patients with bitemporal independent epileptiform abnormalities in extracranial
EEGs. It must be emphasized that our patients represent a highly selected group in whom, even with the
help of ancillary clinical and laboratory data, a reliable
diagnosis of the side and sometimes site of seizure
origin could not be made before depth electrodes were
inserted. The conclusions of this study therefore cannot be extrapolated to patients with temporal lobe
epilepsy as a whole.
In the largest single group of 25 patients (44%), all
of the seizures arose exclusively from one temporal
lobe. Another 33% of the patients showed a strong
predominance for their seizures to originate in one
temporal lobe. Only 14% had seizures arising independently from each temporal lobe without significant
lateralized predominance on one side. A few remaining patients (9%) had multiple seizure patterns which
were often diffuse. An 80% cutoff was selected to
separate patients with strong unilateral temporal predominance from those without. This selection was
mainly based on the bimodal frequency distribution of
the predominance of seizure onset (see Fig 2), but also
took into account the results of earlier studies on a
smaller number of patients 15, 61. The current results
confirm the original impression gained in those earlier
studies. The utility of SDEEG recording for the investigation of this bitemporal problem had not been previously established, although it was accepted implicitly
as one of the main indications for its use 114-161.
Most patients showed some variability in their patterns of SDEEG ictal onset, even when seizures
originated exclusively in one temporal lobe. Only 9
patients (16%) demonstrated a stereotyped ictal onset
pattern. The observation that seizures arising from the
same temporal lobe in the same patient could start
independently in the amygdala, hippocampus, or temporal limbic structures regionally suggests that seizures
can arise from different, and at times larger or smaller,
areas within a wider epileptogenic zone. The determinants for this variability are unknown, however.
Whether some of this variability is related to failure to
sample from a hidden generator, from which the seizure discharge may secondarily spread along different
pathways, is a hypothesis that could not be conclusively tested by any presently known method. The
weight of circumstantial evidence, however, seems to
argue against it. Yet the restricted “tunnel vision” of
depth electrode recording must be kept in mind and
precise localization of ictal onset to one of the recorded structures should not be assumed too readily
117, 181.
In keeping with an earlier report 1191, SDEEGrecorded regional temporal seizure onsets were more
common than focal temporal onsets. The latter were
most frequently seen in the hippocampus. This hippocampal predominance may be partly due to a methodological bias: Strictly amygdaloid onsets can sometimes be missed because amygdaloid discharges are
liable to form “closed fields” that are more likely to
escape detection [l8}. On the other hand, evidence
exists that the site of seizure onset is correlated with
the site of maximal pathological damage, which most
commonly seems to involve the hippocampus in temporal lobe epilepsy 1201, although the amygdala has to
date not been studied with equal care. Focal temporal
onsets were more common in Group A patients, in
whom all seizures originated in one temporal lobe, as
So et al: Bitemporal Epileptiform Abnormalities I: SDEEG 429
compared with Groups B and C patients. In the latter
two groups, who all had some independent seizures
arising from other sites, regional temporal onsets predominated. The significance of these findings is unclear. If the pattern of regional temporal onset is a
reflection of a more widespread epileptogenic zone,
then it may go hand in hand with an increased tendency for independent bitemporal seizures and other
seizure patterns, as found in Group B and C patients.
No reliable AD threshold asymmetry was seen on
electrical stimulation of homologous mesial temporal
lobe structures. Even when care was taken to select
only habitual warnings for analysis, electrical stimulation often produced auras from either temporal lobe,
as has also been reported by Halgren and associates
{21}. When an aura occurred on stimulation of only
one side, however, it was more likely to be concordant
with the side of exclusive or predominant ictal onset.
Clinical seizures occurred with almost equal likelihood
on stimulation of either temporal lobe, although seizures elicited from the lobe contralateral to that of
spontaneous seizure onset were sometimes recognizably dissimilar from the patient’s habitual attacks. It is
therefore doubtful that electrical stimulation studies of
AD threshold, and of electrically induced auras and
seizures, could be relied on for lateralization of the
epileptogenic zone in the investigation of patients with
bitemporal seizures. The different results reported by
other investigators 122-241 may be related to the fact
that all of our patients had evidence for significant bitemporal abnormalities. Patients with well-defined,
unilateral temporal abnormalities may account for the
sometimes high rates of concordant lateralization reported by others on electrical stimulation studies {22,
The limited sensitivity of noninvasive tests, which
have a relatively low rate of concordant lateralization
with the results of SDEEG recording, was not unexpected, since our patients were selected for SDEEG
study precisely because all other means had failed to
demonstrate consistently the epileptogenic zone responsible for their seizures. Iateralization based on abnormal neurological signs and imaging findings showed
higher rates of concordance with the side of exclusive
or predominant SDEEG ictal onset as compared with
other tests. Even with the benefit of retrospective analysis using multiple variables, accurate identification of
patients with seizures arising exclusively or predominantly from one temporal lobe (Groups A and B) and
correct lateralization of the side of seizure onset were
potentially possible in only 4 of 44 patients, who might
thus have been spared from invasive intracranial recording.
Our results should not be taken as an argument
against the clinical utility of noninvasive testing in the
presurgical evaluation of patients with intractable epi430 Annals of N e u r o lo g y
Vol 25 No 5 May 1989
lepsy. Reliance on the role of noninvasive tests has to
take into account the quality and the clarity of the
information obtained, as well as their ability to yield
concordant localizing evidence in each individual clinical problem [25, 261.
Our study has not addressed the possible contribution of magnetic resonance imaging and positron emission tomography to the investigation of patients with
bitemporal abnormalities because too few patients in
this series had received these tests. It has also not
addressed the question in those patients with seizures
arising independently from either temporal lobe of
whether the clinical seizure patterns would be sufficiently different to make it possible to predict bilateral
independent onsets on clinical grounds. It is our impression that in only a small number of patients were
the seizures arising from one temporal lobe manifestly
different in their clinical pattern from those arising
from the opposite side. These are issues that clearly
deserve more detailed analysis.
We are grateful to Dr Rachel Ochs, who interpreted the SDEEGs in
some patients, and to Ms Gisele Robillard for work in the preparation of the manuscript.
1. Jasper H, Permisset B, Flanigin H. EEG and cortical electrograms in patients with temporal lobe seizures. Arch Neurol
Psychiatry 1951;65:272-290
2. Gibbs FA, Gibbs EL. Atlas of electroencephalography. Cambridge: Addison-Wesley, 1952
3. Gastaut H. So-called “psychomotor” and “temporal” epilepsya critical study. Epilepsia 1953;2:59-76
4. Rovit FU,Gloor P, Rasmussen T. Sphenoidal electrodes in the
electrographic study of patients with temporal lobe epilepsy: an
evaluation. J Neurosurg 1961;18:151-158
5 . Gloor P, Olivier A, Ives J. Prolonged seizure monitoring with
stereotaxically implanted depth electrodes in patients with bilateral interictal temporal epileptic foci: how bilateral is bitemporal
epilepsy? In: Wada JA, Penry JK, eds. Advances in epileptology: Xth Epilepsy International Symposium. New York: Raven
Press, 1980:83-88
6. Olivier A, Gloor P, Quesney LF,Andermann F. The indications
for and the role of depth electrode recording in epilepsy. Appl
Neurophysiol 1983;46:33-36
7. Olivier A, Marchand E, Peters T, Tyler J. Depth implantation at
the Montreal Neurological Institute and Hospital. In. Engel J Jr,
ed. Surgical treatment of the epilepsies. New York Raven
Press, 1987:595-601
8. Spencer SS, Spencer DD, Williamson PD, Mattson RH. Ictal
effects of anticonvulsant medication withdrawal in epileptic patients. Epilepsia 198l ;22:297-307
9. Marciani MG, Gotman J. Effects of drug withdrawal on location
of seizure onset. Epilepsia 1986;27:423-43 1
10. Gloor P, Olivier A, Quesney LF, et al. The role of the limbic
system in experiential phenomena of temporal lobe epilepsy.
Ann Neurol 1982;12:129-144
11. Gotman J, Ives JR, Gloor P, et al. Monitoring at the Montreal
Neurological Institute. In: Gotman J, Ives JRZ, Gloor P, eds.
Long-term monitoring in epilepsy. Amsterdam: Elsevier, 1985:
12. Milner B. Psychological aspects of focal epilepsy and its neurosurgical management. In: Purpura DP, Penry JK, Walter RD,
eds. Neurosurgical management of the epilepsies. Advances in
neurology, vol8. New York: Raven Press, 1975:299-321
13. Jones-Gotman M. Commentary: evaluation: testing hippocampal function. In Engel J Jr, ed. Surgical treatment of the epilepsies. New York: Raven Press, 1987:203-211
14. Rand RW, Crandall PH, Walter R. Chronic stereotactic implantation of depth electrodes for psychomotor epilepsy. Acta Neurochir 1964;11:609-630
15. van Buren JM, Ajmone Marsan C, Mutsuga N. Temporal-lobe
seizures with additional foci treated by resection. J Neurosurg
1975 ;43 596-607
16. Spencer SS, Spencer DD, Williamson PD, Mattson RH. The
localizing value of depth electroencephalography in 32 patients
with refractory epilepsy. Ann Neurol 1982;12:248-253
17. Gloor P. Electroencephalography and the role of intracerebral
depth electrode recordings in the selection of patients for surgical treatment of epilepsy. In: Porter RJ, Mattson R, Ward AA,
Dam M, eds. Advances in epileptology: XVth Epilepsy International Symposium. New York Raven Press, 1984:433-437
18. Gloor P. Volume conductor principles: their application ro rhe
surface and depth electroencephalogram. In: Wieser HG, Elger
CE, eds. Presurgical evaluation of epileptics. Berlin and Heidelberg: Springer-Verlag, 1987:59-68
19. Quesney LF.Clinical and EEG features of complex partial seizures of temporal lobe origin. Epilepsia 1986;27:S27-S45
20. Babb TL.,Lieb JP, Brown WJ, et al. Distribution of pyramidal
cell density and hyperexcitability in the epileptic human hippocampal formation. Epilepsia 1984;25:72 1-728
21. Halgren E, Walter RD, Cherlow DG, Crandall PH. Mental
phenomena evoked by electrical stimulation of the human hippocampal formation and amygdala. Brain 1978;101:83-117
22. Cherlow DG, Dymond AM, Crandall PH, et al. Evoked response and after-discharge thresholds to electrical stimulation in
temporal lobe epileptics. Arch Neurol 1977;34:527-531
23. Wieser HG, Bancaud J, Talairach J, et al. Comparative value of
spontaneous and chemically and electrically induced seizures in
establishing the lateralization of temporal lobe seizures. Epilepsia 1979;20:47-59
24. Bernier GP, St-Hilaire J-M, Girard N, et al. Commentary: intracranial electrical stimulation. In Engel J Jr, ed. Surgical treatment of the epilepsies. New York: Raven Press, 1987:323-334
25. Engel J Jr, Rausch R, Lieb JP, et al. Correlation of criteria used
for localizing epileptic foci in patients considered for surgical
therapy of epilepsy. Ann Neurol 1981;9:215-224
26. Gloor P. Postscript: when are noninvasive tests enough? In Engel J Jr, ed. Surgical treatment of the epilepsies. New York:
Raven Press, 1987:259-261
So et al: Bitemporal Epileptiform Abnormalities I: SDEEG 431
Без категории
Размер файла
930 Кб
electrode, investigation, patients, depth, abnormalities, bitemporal, epileptiform
Пожаловаться на содержимое документа