close

Вход

Забыли?

вход по аккаунту

?

Correlation of criteria used for localizing epileptic foci in patients considered for surgical therapy of epilepsy.

код для вставкиСкачать
ORIGINAL ARTICLES
Correlation of Criteria Used
for Localuing Epileptic Foci in Patients
Lonsidered for 3urgcal lherapy of EpIlepSy
Jerome Engel, Jr, MD, PhD,"? Rebecca Rausch, PhD,*Jeffrey P. Lieb, PhD,'
David E. Kuhl, M D , t $ and Paul H. Crandall, MD"?
~~
~~
Criteria for anterior temporal lobectomy, performed on seven patients with partial complex seizures, were derived
from a battery of fourteen presurgical tests. Seven tests were routine studies aimed at identifying a focus of epileptic excitability, while seven were designed to reveal areas of focal functional deficit. Conflicting information was
frequently obtained from the tests of epileptic excitability, suggesting that it is probably inaccurate to view patients with partial complex seizures as having a single epileptogenic focus. Presurgical evaluation must therefore be
aimed at identifying the focus most responsible for the patient's habitual seizures. Tests of focal functional deficit
provided useful nonconflicting confirmatory information in each of the seven patients studied. The most reliable
information was obtained from depth electrode implantation, and this procedure should be considered essential
except when all evidence of surface-recorded epileptic excitability, including ictal onset, and evidence of focal
functional deficit agree.
Engel J Jr, Rausch R, Lieb JP, Kuhl DE, Crandall PH: Correlation of criteria used for localizing epileptic foci
in patients considered for surgical therapy of epilepsy. Ann Neurol 9215-224, 1981
Sixty years after Horsely initially reported the successful surgical treatment of partial epilepsy, in 1886
[20], developments in electrophysiology made it possible to perform temporal lobectomies for epilepsy
on the basis of electroencephalographic (EEG) evidence alone [4, 211. When clinical signs and symptoms are nonlocalizing and radiographic evidence of a
focal structural lesion is lacking, monitoring of electrical activity from either scalp, cortical, or intracerebral electrodes has been considered the most reliable
means to identify a focal region of abnormal function
[ 181. Almost always, the functional abnormality
sought is a focus of increased excitability as indicated
by the presence of localized interictal spikes [l, 12,
17,21, 22, 331 or the site of onset of ictal events [18,
44, 451. It is well known, however, that epileptiform
excitatory events propagate and may become unreliable for localization. This may be one reason why
some patients, surgically treated for intractable seizures, are not improved when the site of excision is
determined by the location of interictal spikes [l, 5,
14, 15, 17, 21, 22, 33, 35,401 o r even by identification of the site of seizure onset with depth electrodes
[91.
Few have routinely utilized electrophysiological o r
other techniques to help identify epileptic foci by
defining areas of abnormally decreased function. Although it has been shown that epileptic lesions of the
mesiotemporal region can be correlated with unilaterally decreased, pharmacologically induced fast activity in the appropriate sphenoidal [12, 221 or intracerebral [7] electrode, decreased paroxysmal activity
recorded from electrodes acutely inserted into the
amygdala and hippocampus intraoperatively [121, and
elevated afterdischarge thresholds obtained from
electrodes chronically implanted by stereotaxic techniques into the same areas [8], these findings are not
commonly considered in the standard evaluation of
epileptic patients for surgical therapy.
Since May, 1977, seven studies designed to define
focal areas of functional deficit (decreased function)
were combined with seven routine scalp and
stereotaxic depth EEG (SEEG) telemetry procedures
aimed at identifying focal areas of epileptiform excitability (increased function) to allow fuller evaluation of patients being considered at UCLA for the
surgical therapy of intractable partial complex epilepsy. All of the studies performed were taken into
consideration in determining the advisability of temporal lobectomy and the site of excision. These
From the 'Reed Neurological Research Center, the +Brain Research Institute, and the $Department of Radiological Sciences,
UCLA School of Medicine, Los Angeles, CA 90024.
Address reprint requests to Dr Engel, Reed Neurological Research Center, UCLA School of Medicine, LOS Angeles, CA
90024.
Received Apr 24, 1980,and in revised form July 31. Accepted for
publication Aug 4, 1980.
0364-5134/81/030215-10$01.25 @ 1980 by the American Neurological Association 215
fourteen criteria were correlated for the first seven
patients to complete the new protocol. Since all have
now been followed for more than one year postoperatively, some conclusions regarding surgical outcome can be made ([14, 401 and our own unpublished data).
1956), one had the lobectomy postponed for medical reasons, one chose to have the lobectomy performed
elsewhere, and two were not recommended for anterior
temporal lobectomy because a resectable focus could not
be identified (one of these had an amygdalotomy without
benefit).
Materials and Methods
Pa tient Population
Surface and Depth E E G
Criteria for acceptance into the program at UCLA include
the following: (1) a partial seizure pattern diagnosed on the
basis of clinical or EEG findings, preferably with behavioral
features at one time or another that suggest automatisms;
(2) frequent seizure episodes that are resistant to adequate
drug therapy and seriously interfere with daily life; (3) absence of evidence indicating progressive central nervous
system disease or a space-occupying lesion (in the latter
case depth electrode evaluation is not considered necessary
and surgical treatment is determined according to the demonstrated lesion); and ( 4 ) absence of severe health problems, psychosis, or marked mental retardation.
The new protocol for presurgical evaluation included
two weeks of scalp and sphenoidal EEG telemetry and
video monitoring (phase I) followed at a later date by
stereotaxic implantation of depth electrodes for SEEG
telemetry and video monitoring (phase 11) for an additional
four weeks. During these hospitalizations a number of
other procedures were also performed to identify focal
epileptic excitability and regional deficits in function. Informed consent was obtained from all patients upon entering the evaluation program. Seven patients completed
the full protocol and underwent the standard anterior temporal lobectomy 1101 between May, 1977, and May, 1979.
Basic data from these patients are shown in Table 1. Five
other patients who also entered phase I1 during this time
did not complete the protocol for the following reasons and
are not included in the data presented: one patient died
from intracerebral bleeding as a result of the implantation
procedure (one of two deaths among 129 implants since
During phase I, scalp recordings were made with gold disc
electrodes applied with paste or, for telemetry, with collodion according to the international 10-20 system. Each
sphenoidal electrode consisted of fifteen braided 50-gauge
stainless steel wires, coated with Teflon and bared for 1 m m
at the tip. These were threaded through a 3.8 cm, 22-gauge
hypodermic needle and inserted according to the technique
described by Pampiglione and Kerridge [34] except that
the needle was removed immediately, leaving only the wire
in place. During phase 11, depth recordings were made with
concentric bipolar stainless steel electrodes bared at the tip
and stereotaxically implanted bilaterally into amygdala, pes
hippocampi, gyms hippocampi, and, in some cases, orbital
frontal and supplementary motor cortices [ 101. Simultaneous skull and sphenoidal recordings were also made. Skull
electrodes consisted of stainless steel nails hammered into
the outer table of the skull at the time of surgery according
to the international 10-20 system 1101.
Telemetry
O n e or more prolonged baseline hard-wire recordings
were made from all patients prior to initiation of scalp or
depth telemetry in order to survey the electrodes and
choose the appropriate montages. Electrodes were then
attached to a light-weight 14-channel telemetry transmitter (Bio-Medical Monitoring Systems, Los Angeles, CA)
which was secured to the patient's head with gauze bandage. Twelve channels were used for EEG recording and 2
for time code data. Radio signals were received in a separate room and relayed to a 14-channel Ampex magnetic
tape recorder. Analog data were stored on tape, but all ictal
Table 1. General Information
Seizure Information
Patient No.,
Age (yr),
and Sex
Etiological
Factors
Neurological
Examination
Radiological
Examination
Duration
(vr)
104. 40, F
107. 24, M
109. 32, F
None
None
None
Poor memory
Poor memory
Negative
13
16
25
111. 45, M
Birth anoxia
Negative
112. 21, F
Febrile convulsions
Positive family
history
? Head trauma
Negative
Negative
Negative
Brainstem atroPhY
Right frontal
craniotomy"
Negative
113. 20, F
115. 23, M
Poor memory
Negative
Right parietal
calcificationb
Negative
44
19
12
10
Aura
TvDe
Freouencv
Epigastric rising
None
Flashing in right visual field
Sensation of movement
Nausea
PC, SG
PC
PC. SG
2-3lwk
3-411-2 wk
3-4lday
PC, SG
Manylday
PC
"Twinkle" in head,
headache
Floating sensation
PC. SG
llday
25 auraslday
3-4lday
PC
2-3lwk
"Surgical procedure performed as a result of trauma sustained after the onset of seizures (automobile accident caused by a seizure).
hThought to be unrelated to the epileptic focus; scalp electrodes and skull nails over this area consistently failed to record interictal or ictal
epileptiform events.
PC
=
partial complex; SG
=
secondarily generalized.
216 Annals of Neurology Vol 9 N o 3 March 1981
and selected segments of interictal data were also transferred to paper for visual analysis utilizing a 16-channel
Grass Model 6 EEG machine.
A Sony video camera continuously recorded the patient's
behavior, and this was stored o n a time-lapse Panasonic
videotape recorder at six frames per second. Videotapes
could thus b e reviewed in real time or at nine times real
time. A time code generator displayed actual clock time
digitally every second on the video image and, in binary
code, every ten seconds on the EEG so that behavioral and
electrical events could be precisely correlated. In order for
an electrical event to b e considered the origin of a behavioral seizure, it was necessary to demonstrate that the event
preceded the ictal behavior as observed on videotape.
Monitoring was continued 24 hours a day for approximately two weeks during phase I and for an average of four
weeks during phase 11, although the patient was occasionally off telemetry for other studies during this latter stage
of the evaluation. Montages were periodically varied to
best display the electrical phenomena observed and to
sample all derivations. Anticonvulsant medications were
generally tapered during this time but were rarely completely withdrawn.
Ictal events were identified for study in several ways: (1)
patients were under continuous observation from staff
either directly or via video monitors placed at the nursing
station so that unusual behavior could b e noted; ( 2 ) to indicate an aura or the beginning of a seizure, the patient had a
signal button to push which called the nurse and also placed
a mark on the EEG; (3) during phase 11, an automatic seizure detector which called the nurse and printed out the
clock time could be triggered by ictal activity appearing
in the SEEG record [ 3 ] ; and (4) telemetry tapes were
randomly searched to obtain samples of interictal activity
and subclinical electrical seizure discharges.
lnterictal Spike Analysis
The frequency and location of interictal spikes seen o n
routine EEGs as well as during scalp and sphenoidal telemetry were determined for this study by visual analysis of
available EEG records and by the review of old EEG reports from UCLA and elsewhere. Interictal spike data obtained from the depth electrode recordings, however, were
also analyzed in more detail utilizing a computer. Automatic spike detection was carried out off-line o n all-night
sleep recordings on a PDP-12 minicomputer, utilizing an
assembly language computer program (COUNTR) 127,281
for which spike recognition variables had previously been
optimized by comparing its performance with that of human
observers [28]. The all-night sleep records were segmented
into four stages: waking, light sleep, deep sleep, and REM
sleep. The mean rate of spike activity for each depth
channel was computed within and across sleep stages. Also
computed was the ratio of spike activity in each sleep stage
to waking in order co estimate the autonomy of spike activity across physiological states [16, 27, 411.
T h iopental Activation
Thiopental activation was performed in 5 patients while
depth electrodes were in place and in 2 with only scalp and
sphenoidal electrodes. During EEG monitoring, 25 mg of
thiopental was injected intravenously every 30 seconds until
the corneal reflex was abolished [12, 221. The patient was
then allowed to wake up. The EEG was visually analyzed
for activation of focal spike activity as well as for focal attenuation in barbiturate-induced fast activity. A 50% decrease in amplitude of fast activity at one sphenoidal electrode 50% of the time was considered to b e a significant
attenuation.
Af terdischarge Threshold Determinations
Thresholds for eliciting afterdischarges were determined
for specific sites by stimulating via individual concentric
bipolar depth electrodes in increments of 0.5 ma up to a
maximum of 5 ma. Electrical stimulation was generated by
a specially constructed constant-current stimulator coupled
to the depth electrodes and consisted of symmetrically
biphasic pulses with a pulse duration of 1 msec, train duration of 20 sec, and frequency of 60 Hz. In any patient the
maximum charge per phase was 0.5 pC, with a maximum
charge density of 55.5 pClcm2.
Intracarotid Amobarbital Studies
Intracarotid amytal studies, based on the technique of
Wada [43], were performed both to lateralize language
dominance and to determine the ability of each hemisphere
independently to support short-term memory [6, 23, 311.
Following injection of 125 mg of sodium amobarbital into
the internal carotid artery, patients were administered a
continuous performance task that surveyed both language
and memory functions. Simultaneous recording of EEG
was made to aid in the determination of spread and duration of drug effectiveness. For the present study, memory
function was evaluated by testing the patients' ability, approximately 12 minutes after the initial injection, to recognize pictures of common objects from the continuous performance task administered within 3 to 4 minutes after
injection.
Positron Emission Computed Tomography
Positron emission computed tomography (PECT) [3 7 I was
performed in six of the seven patients with the ECAT positron tomograph (ORTEC, Inc. Life Sciences, Oak Ridge,
T N ) , utilizing '*F-fluorodeoxyglucose and ':'N-ammonia as
indicators of abnormalities in local cerebral glucose utilization [39] and relative perfusion [38], respectively. PECT
serves as an in vivo analog of quantitative autoradiography.
It is a noninvasive scanning method that produces a similar
but less detailed cross-section picture of brain radioactivity
in humans following intravenous injection of a labeled indicator. In this study, radioactive "F and 13N were produced in the UCLA nuclear medical cyclotron. The use of
this technique in patients with seizure disorders has been
described in more detail elsewhere I1 3, 24-26],
Results
Surgery improved the condition of all the patients
reported here, and all showed pathological lesions in
the resected temporal lobe specimen (Table 2). None
of the patients had any evidence of a focal strzlctural
Engel et al: Criteria for Epilepsy Surgery
217
Table 2. Surgical Results
Patient
No.
Date of
Surgery
Side
Pathological Findings
Postoperative Seizures
104
2/78
Right
Hippocampal sclerosis
107
109
6/78
9/78
Left
Right
111
112
113
1/79
3/79
4179
Left
Right
Right
Hippocampal sclerosis
Tuberous sclerosis, white matter of
midtemporal gyrus
Hippocampal sclerosis
Hippocampal sclerosis
Hippocampal sclerosis
1 partial complex seizure 5/78,
occasional auras
Seizure freea
2 partial complex seizures 7/79
115
5/79
Right
Meningioma, midtemporal gyms
Seizure free
Seizure free, occasional auras
3 partial complex seizures
1/80, 2/80, 4/80, occasional
auras
Seizure free
aExcludes nocturnal grand mal seizures induced in 11/78 by sudden withdrawal of all anticonwlsant medications. H e had never had a
generalized seizure before and has had none since, indicating that this was a drug-withdrawal effect.
Table 3. Epileptic Excitability from Surface StudieJ
Interictal Spike Frequency
Patient
No.
Side of
Operation
104
107
-
Spontaneous
From
Thiopental
Seizure Onset
Right
Left
Right > left
f i g h t only
Right = left
Right only
5 seizures, right sphenoidal
7 seizures and 2 auras, right
109
Right
Left > right
Right = left
111
112
Left
Right
Left > right
Mostly normal; telemetry showed
a few on right
Right only
None
113
Right
Right only
Right = left
115
Right
Mostly normal; one EEG showed a
few o n right
None
lesion in either temporal lobe on the basis of routine
radiographic studies, including CT scan, angiography,
and pneumoencephalography, although one (Patient
113) had an unrelated parietal abnormality and
another (Patient 111) had a frontal lobe lesion acquired after the onset of seizures (see Table 1). Neurological examination also failed to indicate any localizing or lateralizing neurological deficits (see Table
1). Decision to perform surgery and selection of the
side for temporal lobectomy were determined entirely on the basis of the studies reported here. The
studies were aimed at identifying abnormalities in
neuronal function, as distinct from abnormalities of
structure, and can be divided into those delineating
abnormally increased function (focal epileptic excitability) and those indicating abnormally decreased
function (focal function deficit). These two types of
evaluation can be further divided into those that required depth electrodes and those that did not.
218 Annals of Neurology
Vol 9 No 3 March 1981
sphenoidal
4 seizures, bilateral onset, 2
spread to right, 2 to left
4 seizures, left sphenoidal
3 seizures, left sphenoidal
5 seizures, 4 auras, uncertain origin
14 seizures, 2 auras, uncertain
origin
5 seizures, right sphenoidal
Focal Epileptic Excitability
SURFACE S T U D I E S . Data obtained without the use
of depth electrodes are displayed in Table 3. Spontaneous interictal EEG spike activity was greater
contralateral to the side of eventual lobectomy in two
of the seven patients (Nos. 107 and 109). EEGs performed prior to admission were all normal in one patient (No. 112) though some interictal spikes were
seen during scalp and sphenoidal telemetry. Another
patient (No. 115) had no interictal spikes on hardwire or telemetry recording during this admission,
but a previous EEG done elsewhere was reported as
showing some interictal spikes.
Thiopental activated EEG spike activity contralatera1 to the side of eventual resection in two patients
(Nos. 107 and I l l ) , one of whom had more spontaneous spikes on the other side (Patient 111). T h e
remaining five patients showed either bilateral independent spikes of equal frequency or no spikes.
Localizing information was obtained from surface
telemetry recordings of seizure onset in five patients.
In three (Patients 104,1 1 1, and 1 1 5) the site of onset
was the temporal lobe that was later resected, but in
two (Patients 109 and 113), other data led to an
eventual contralateral temporal lobectomy . This apparent false localization will be the subject of a more
complete report. Seizures appeared to begin bilaterally in two patients (Nos. 109 and 113).
DEPTH STUDIES. Localizing and lateralizing data
obtained from our SEEG investigations of epileptiform excitability are shown in Table 4. Bilaterally
independent interictal spikes were invariably recorded from the temporal depth. Spontaneous interictal spike frequency from the depth was greater
contralateral to the side of eventual resection in four
of the seven patients (Nos. 107,109,111, and 113).
A measure of spike autonomy [16,27,411 or lack of
reactivity to alterations in the sleep-wake cycle increased the yield of apparent correct localization
from three of the seven patients to five. No
lateralizing information was obtained from this approach in one patient (No. 109),and the contralateral
side was still identified in another (Patient 107).This
technique and the results obtained are more fully reported elsewhere [27].
Thiopental activation was performed with depth
electrodes in place in only five of the seven patients,
and in all but one (No. log), interictal spike discharges were maximally activated on the side of
lobectomy.
SEEG-recorded ictal onsets indicated the temporal
lobe that was designated for resection in six patients.
In one patient (No. 1 1 5 ) , however, two seizures appeared to begin in the temporal depth on the side
eventually resected, while three other seizures appeared to originate in the contralateral temporal
depth. Consequently, other data contributed greatly
to the decision to operate and to selection of the site
of resection.
Focal Functional Dejicit
SURFACE STUDIES. Localizing and lateralizing data
obtained from studies designed to identify focal
functional deficits without the use of intracerebra1 electrodes appear in Table 5. A decrease i n ,
thiopental-induced fast activity recorded at one
sphenoidal electrode correctly identified the presumed epileptic focus in four of the seven patients,
while no information was obtained from the other
three (Patients 104,1 1 1, and 1 1 5).
Intracarotid administration of amobarbital was
performed to ascertain language dominance and to
evaluate the capacity of each temporal lobe to support memory. The left hemisphere was dominant for
language, as expected, in five of the seven patients.
The fact that the right hemisphere was language
dominant in one patient (No. 107),who was righthanded, was taken to indicate transfer of function
suggestive of a defective left hemisphere. In another
patient (No. 113),who was left-handed, the presence
of bilateral language function was not considered important with regard to lateralization. Intracarotid
amobarbital administration revealed poor memory
function on the side of eventual lobectomy in three
patients (Nos. 104, 107, and 109) and no major
deficits in the remaining four.
PECT was performed in only six of the seven patients, and in all but one (No. 107), regional decreased metabolism correlated well with the demonstrated pathological lesion. Bilateral deficits were
demonstrated, however, in one patient (No. 113).
Some of these patients are included in other reports
dealing with PECT studies of epilepsy [13,24-26].
DEPTH STUDIES. Information regarding focal decreased function that could be obtained only with
Table 4. Epileptic Excitability from Depth Studies
Interictal SDike Freouencv
Patient No.
Side of
Operation
Spontaneous
From
Thiopental
Interictal Spike
Autonomy
104
107
109
Right
Left
Right
Right > left
Right > left
Left > right
Right > left
Not done
Right = left
Right
Right
Both
111
Left
Right > left
Left > right
Left
112
113
115
Right
Right
Right
Right > left
Left > right
Right > left
Right > left
Right > left
N o t done
Right
Right
Right
Seizure Onset
6 seizures, right; 4 auras, right
8 seizures, left
5 seizures, right; 1 aura, right; 3
auras, uncertain
5 seizures, left; 5 auras, left; 20
seizures, uncertain
1 seizure, right; 15 auras, right
7 seizures, right; 6 auras, right
2 seizures, right; 3 seizures, left
Engel et al: Criteria for Epilepsy Surgery
219
Table 5 . Functional Deficit from Surface StudieJ
Intracarotid Amobarbital Testing
Patient No.
Side of
Operation
Decreased Thiopental
Fast Activity
Language
Memory
PECT
104
107
109
111
Right
Left
Right
Left
Symmetrical
Left decreased
Right decreased
Symmetrical
Left
Right
Left
Left
Not done
Normal
Right temporal decrease
Left temporal decrease
112
Right
Right decreased
Left
113
Right
Right decreased
Bilateral
115
Right
Symmetrical
Left
Right poor, left good
Right good, left poor
Right poor, left good
Right not done, left
good
Right good, left
good
Right good, left
good
Right good, left
good
Right temporoparietooccipital decrease
Right temporoparietooccipital decrease > left temporal decrease
Right lateral temporal
decrease
Table 6. Functional Deficit from Depth Studies
Patient No.
Side of
Operation
Baseline
Rhythms
Thiopental-induced
Fast Activity
Afterdischarge
Threshold
104
107
109
111
112
113
115
Right
Left
Right
Left
Right
Right
Right
Symmetrical
Left decreased
Symmetrical
Left decreased
Right decreased
Right decreased
Right decreased
Symmetrical
Not done
Symmetrical
Symmetrical
Right decreased
Right decreased
Not done
Undetermined
Left elevated
Not done
Left elevated
Right elevated
Right elevated
Not done
depth electrode recordings is shown in Table 6. An
attenuation of normal rhythmic activity was recorded
from the temporal depth in all but two patients (Nos.
104 and 109), and this was always on the side of resection. Thiopental activation was performed with
depth electrodes in only five of the seven patients,
and barbiturate-induced fast activity was decreased
on the side of surgery in two (Patients 112 and 113).
Afterdischarge thresholds were determined in five
patients. An elevation in afterdischarge threshold
correctly indicated the presence of hippocampal sclerosis in four patients (Nos. 107, 111, 112, and 113),
while thresholds could not be determined in one
(Patient 104) for technical reasons.
Discussion
The postoperative follow-up of these patients is relatively short, and seizures could still return to some
degree. However, our own experience indicates that
the interval of one to three years is sufficient to indicate that surgery has generally been beneficial, and
this conclusion is supported by the findings of others
[ 14, 401. Furthermore, pathological lesions were
found in each of the resected temporal lobe specimens. Five patients showed hippocampal sclerosis,
which has been reported to occur bilaterally to
220 Annals of Neurology
Vol 9 N o 3 March 1981
some degree in 20% of institutionalized patients with
severe epilepsy [291. The relatively intact mental
status of the patients reported here argues against the
presence of a major lesion in the contralateral hippocampus, although some bilaterally decreased function was suggested by PECT in one patient (No. 113)
and by poor memory in three (Patients 104, 107, and
113). The finding of tuberous sclerosis in Patient 109
leaves open the possibility of additional undetected
lesions in this woman, although she has no other
signs or symptoms of the disorder and family history
is negative. Other areas of epileptogenicity were indicated by the occurrence of a few seizures postoperatively in three patients (Nos. 104, 109, and 113)
and occasional auras in another (Patient 112). Indeed, evidence of bilateral irritability was present in
all patients since all had bilaterally independent interictal spikes recorded from the temporal depth.
This was also almost invariably true of depth electrode recordings from the previous 103 patients
studied in the UCLA series (unpublished data).
Furthermore, seizures do continue to some degree in
many patients despite clear pathological evidence
that a structural lesion was resected [14, 151.
All the evidence cited here indicates that patients
with chronic partial complex seizures, unlike most
animals with experimental focal neocortical epilepsy,
appear to demonstrate multifocal abnormalities. This
may be the result of functional changes such as secondary epileptogenesis [321 and kindling 1191,
structural damage induced by frequent seizures [30],
or multiple lesions incurred at the time of the initial
insult. Indeed, there is even evidence from primate
studies that bilateral mesiotemporal lesions may be
necessary for the manifestation of complex partial
seizures [42]. It is therefore probably inaccurate to
view these patients as having a single epileptogenic
lesion. The likelihood is rather that the epileptogenic
brain is abnormal in many ways and in many areas,
and an attempt must be made to locate and excise the
focus most responsible for initiating the patient’s
habitual seizures. For the purposes of this discussion,
those studies which identified this presumed most responsible focus may be referred to as “correctly” localizing, and those that identified other areas as
“falsely” localizing. It must be recognized, however,
that designation of a single correct or false localization is an oversimplification.
In order to put the results reported here into
proper perspective, many other factors should be
mentioned that commonly enter into the decision to
perform temporal lobectomy for epilepsy. Localizing
clinical signs and symptoms are of diagnostic importance, but, as in this study, such findings are often
absent in patients with partial complex epilepsy.
Careful observation of the behavioral seizure can
provide localizing information [ 111, although this can
at times be misleading (as with Patient 112). Radiographic evidence of a structural lesion may provide
clues for localization, but it is generally appreciated
that a demonstrated structural lesion does not always
correlate with an epileptogenic focus [18] (as with
Patients 111 and 113). Pharmacological [2] and electrical stimulation [36] techniques are used to activate
seizures, and the results have been compared with
spontaneous seizures [46]. Our own experience in
this regard is insufficient to allow comment. Electrocorticography at the time of lobectomy is used to
define an epileptogenic area [ 181, but since a standard
en bloc resection [ 101 is performed at UCLA on the
basis of information obtained from chronic SEEG
recordings, electrocorticography at surgery is usually not considered necessary [45]. Other potential
sources of localizing information presently under
study but not considered in this report include
psychometric evaluations, careful descriptions of the
aura, and postictal examination of the patient. The
importance of the results reported here can be best
appreciated only after one accepts that each patient
should be viewed individually and that many factors
must figure into the formula that eventually determines if and where the surgeon will operate.
The localizing information derived from each
study, for each patient, is summarized in Table 7. Individual variations may have little meaning in this
small sample. However, the demonstration of frequent conflicting results for some studies should be
considered as having clinical importance. Although
the relative homogeneity of this patient population
with respect to surgical outcome and pathological
findings precludes correlating these variables with
the results of specific studies, it does appear that evidence of decreased function was more prominent
when hippocampal sclerosis was present. However,
the fact that decreased function in mesial temporal
structures also could be demonstrated in two patients
with lateral temporal lesions (Nos. 109 and 115) suggests that the functional deficit revealed by some of
these studies does not depend entirely upon structural damage.
Conflicting information was obtained from studies
of epileptic excitability for every patient but one (Patient 104). The most consistent contralateral localization of epileptic excitability was obtained in
Patient 107, who also demonstrated the most consistent evidence of decreased function on the ipsilateral
side. The severity of neuronal disruption at the site
of seizure generation may have retarded the spread,
and therefore the manifestation, of epileptiform
electrical activity in that area.
Interictal spike activity recorded from the depth
electrodes often appeared to be less localizing than
that from scalp and sphenoidal derivations. This
agrees with a previous analysis of spike activity recorded during electrocorticography [ 123. There may,
however, be more information, not reported here,
which can be derived from the pattern of interictal
events that were recorded from the depth electrodes
(Lieb et al, unpublished data). The opposite finding,
that thiopental-activated EEG spikes recorded
from the surface never added relevant localizing
information, may in part reflect the fact that these
particular studies were not initially performed to
evaluate scalp and sphenoidal spike activity, and
the tracings may have been inadequate for this purpose.
Since the site of seizure origin in the depth has
been the single most important criterion at UCLA for
determining site of resection, it is not suprising that
this information correlated highly with the side of the
lobectomy. In view of the findings in Patient 115,
however, even this technique can produce conflicting
data. The decision in this case to operate contralateral to the side demonstrating the most seizure onsets
during SEEG telemetry was determined by several factors besides the evidence of decreased function on the other side. This will be the subject of a
more detailed report elsewhere, but in brief, the
Engel et al: Criteria for Epilepsy Surgery
221
Table 7. Summary of Results of Testing t o Localize Epileptic Foci
Factor Studied
Percent
Ipsilateral
Percent
Contralateral
I
0
I
71
0
43
29
29
29
C
I
I
I
57
14
I
ND
I
43
80
71
86
67
17
55
57
0
14
14
..
26
I
I
0
57
0
I
0
0
I
0
0
I
0
0
I
0
0
I
14
43
83
0
0
0
I
ND
0
0
I
0
I
I
I
I
I
ND
71
40
0
0
0
I
ND
I
I
I
ND
80
0
17
0
83
0
50
0
43
71
55
...
...
0
71
0
40
0
Patient
Patient
Patient
Patient
Patient
Patient
Patient
104
107
109
111
112
113
115
0
2
HippoTuberous
campal
sclesclerosis
rosis
0
Hippocampal
sclerosis
0
3
Hippocampal
sclerosis
Hippocampal
sclerosis
0
Meningioma
I
0
I
C
C
C
0
0
I
C
I
I
0
C
I
0
0
I
I
I
I
86
0
C
C
I
C
0
0
I
C
I
I
17
83
14
29
I
71
29
I
I
I
I
71
14
0
I
I
0
0
I
0
ND
I
I
0
0
0
Postoperative seizures 1
HippoLesion
campal
sclerosis
EPILEPTIC EXCITABILITY
Surface recording
Interictal spike
activity
Spontaneous
With thiopental
Seizure onset
C
Depth recording
Interictal spike
activity
Spontaneous
With thiopental
Autonomy
Seizure onset
Percent ipsilateral
Percent contralateral
ND
C
~
FUNCTIONAL DEFICIT
Surface recording
Thiopental-induced
fast activity
Intracarotid amobarbital testing
Language
Memory
PECT
Depth recording
Baseline rhythms
Thiopental-induced
fast activity
Afterdischarge
threshold
Percent ipsilateral
Percent contralateral
I
0
0
I = localization ipsilateral rn lobectomy; C = localization contralateral to lobectorny; 0 = no information obtained; ND = study nor done.
contralaterally initiated seizures were atypical for this
patient and appeared to represent an entirely different phenomenon from the spontaneous seizures that
occurred as a result of his chronic epileptic conditon.
This patient demonstrates that evidence of bilateral
independent SEEG-recorded seizure onsets need not
be a contraindication to anterior temporal lobectomy. Furthermore, sampling error must be taken
into consideration. The seizure origin recorded during SEEG telemetry in Patient 107, who had the most
222 Annals of Neurology
Vol 9 No 3 March 1981
marked evidence of focal decreased function, consisted of very low voltage fast activity reflected in
only a few electrodes [ 131, and could easily have been
missed if another montage had been used.
Although the studies designed to identify areas of
regionally decreased function frequently yielded no
information, the fact that conflicting information did
not occur is important. The results confirm previous
findings that unilateral decreases in pharmacologically induced fast activity recorded from sphenoidal
[12, 221 or depth [7]electrodes and elevations in afterdischarge threshold [8] correlate with a structural
lesion in the mesial temporal area.
PECT correctly identified the pathological lesion
in five of the six patients studied. The failure of
PECT to identify the large area of sclerosis in one
patient (No. 107) may be attributed to the fact that
he was among the first patients studied with this new
technique and the scan was of suboptimal quality.
PECT shows great promise of becoming an important
diagnostic tool in partial epilepsy [13, 24-261.
Reduction in normal baseline rhythmic activity recorded with depth electrodes appeared to be an accurate test of localized decreased function. Bipolar
chain linkage montages were often required to demonstrate attenuation of baseline rhythms. The fact
that only concentric bipolar montages were used for
the first three depth thiopental activations might
explain why these revealed no asymmetries in
barbiturate-induced fast activity even though one of
these patients (No. 111) did show a decrease in
spontaneous fast activity with a chain linkage montage.
Conclusions
Although the sample is small, the results presented
here indicate that identification of an epileptic lesion
for the purposes of surgical resection should include
not only studies aimed at defining an area of epileptic
excitability but also confirmatory evidence of focal
decreased function. The functional deficit demonstrated by these studies, more than merely reflecting
structural damage missed by radiological evaluation,
could result from physiological disruption of normal
neuronal integration inherent in the epileptic process. If this is so, epileptic lesions might always demonstrate both abnormal excitation and functional
deficit. There is reason to be encouraged about the
usefulness of these confirmatory tests, although it is
unlikely that the perfect correlation between functional deficit and side of lesion demonstrated here
will be sustained when larger series are reported.
Ictal onset recorded from SEEG telemetry still appears to yield the best localizing data, but development of confirmatory tests of functional deficit may
obviate the need for this potentially dangerous surgical procedure in some patients.
A definitive set of criteria for the surgical therapy
of partial epilepsy cannot be derived from a simple
formula incorporating the studies discussed in this
report or any others. Each patient is unique and must
be evaluated according to the peculiarities of his o r
her own case. Despite the inability to ascribe statistical significance to the values provided here, some
general conclusions can be drawn that might allow
certain studies to be given greater consideration than
others when determining the site of excision. Data
obtained from depth electrodes, revealing both
epileptic excitability and functional deficit, appear to
be the most reliable but do require a surgical procedure. PECT is a noninvasive test that shows great
promise but is still largely untested and available at
only a few centers. With regard to other techniques
not requiring depth electrodes, if all evidence of
surface-recorded epileptic excitability, including site
of ictal onset, and evidence of decreased function
agree, it may not be necessary to risk depth electrode
investigations. When surface evidence of epileptic
excitability and decreased function conflict, however,
depth electrode implantation is essential, not only to
identify the site of seizure origin with more accuracy,
but also to provide further evidence of focal functional deficits. If surface studies reveal only a focus of
epileptic excitability, the possibility of false localization must still be entertained, particularly when attempting to distinguish between a lesion in the temporal lobe and a potential extratemporal lesion. In
this case, depth electrode implantation should also be
considered necessary to resolve the issue. Although
tests of focal functional deficit produced no conflicting information in this small series of patients, localization on this basis alone should never be considered sufficient for temporal lobectomy . Epileptic
excitability must still be demonstrated in order to
identify a lesion as epileptogenic.
Supported by Grants NS-02808 and GM-4839 from the National
Institutes of Health and Contract DE-AM03-76-SF00012 from the
Department of Energy.
The authors are indebted to a great number of individuals
who assisted in one or more of the studies reported here, including
D r M. V. Lucia, D r D. 0. Walter, Dr M. Phelps, E. Carr, E. Mariani, J. Thomson, P. Glennon, K. Gentler, M. Wiliiams, c. Ary,
J. Miller, and F. Aguilar.
References
1. Ajmone Marsan C , Baldwin M: Electrocorticography. In
Baldwin M, Bailey P (eds): Temporal Lobe Epilepsy.
Springfield, IL, Thomas, 1958, pp 368-395
2. Ajmone Marsan C , Ralston B L The epileptic seizure: its
functional morphology and diagnostic significance: a clinical electrographic analysis of Metrazol-induced attacks.
Springfield, IL, Thomas, 1957
3. Babb TL, Mariani E, Crandall PH: An electronic circuit for
detection of EEG seizures recorded with implanted electrodes.
Electroencephalogr Clin Neurophysiol 37:305-308, 1974
4. Bailey P, Gibbs FA: The surgical treatment of psychomotor
epilepsy. JAMA 145:365-370, 1951
5. Bengzon ARA, Rasmussen T, Gloor P, Dussault J, Stephens
M: Prognostic factors in the surgical treatment of temporal
lobe epileptics. Neurology (Minneap) 18:717-731, 1968
6. Blume WT, Grabow JD, Darley FL.,Aronson AE: Intracarotid
amobarbital test of language and memory before temporal
lobectomy for seizure control. Neurology (Minneap)
23:812-819, 1973
7. Brazier MAB, Crandall PH, Walsh GO: Enhancement of EEG
lateralizing signs in temporal lobe epilepsy: a trial of
diazepam. Exp Neurol 51:241-258, 1976
Engel et al: Criteria for Epilepsy Surgery
223
8. Cherlow DG, Dymond AM, Crandall P H , Walter RD,
Serafetinides EA: Evoked response and after-discharge
thresholds to electrical stimulation in temporal lobe epileptics. Arch Neurol 34:527-531, 1977
9. Crandall PH: Postoperative management and criteria for
evaluation. Adv Neurol 8:265-280, 1975
10. Crandall PH, Walter RD, Rand RW: Clinical application of
studies on stereotactically implanted electrodes in temporal
lobe epilepsy. J Neurosurg 202327-840, 1963
11. Delgado-Escueta AV: Epileptogenic paroxysms: modern approaches and clinical correlations. Neurology 29:1014-1022,
1979
12. Engel J Jr, Driver MV, Falconer MA: Electrophysiological
correlates of pathology and surgical results in temporal lobe
epilepsy. Brain 98:129-156, 1975
13. Engel J Jr, Kuhl DE, Phelps ME: A comparison of electrical
and metabolic studies of brain function in epileptic patients:
simultaneous EEG and positron emission computed tomography. Trans Am Neurol Assoc (in press)
14. Falconer MA, Serafetinides EA: A follow-up study of surgery
in temporal lobe epilepsy. J Neurol Neurosurg Psychiatry
26:154-165, 1963
15. Falconer MA, Taylor DC: Surgical treatment of drug-resistant
epilepsy due to mesial temporal sclerosis: etiology and
significance. Arch Neurol 19353-361, 1968
16. Gentilomo A, Colicchio G, Pola P, Rossi GF, Scerrati M:
Brain depth recording of interictal epileptic potentials during
sleep in man. In Levin P, Koella WP (eds): Sleep 1974. Basel,
Karger, 1975, pp 444-446
17. Gibbs FA, Amador L, Rich C: Electroencephalographic
findings and therapeutic results in surgical treatment of
psychomotor epilepsy. In Baldwin M, Bailey P (eds): Temporal Lobe Epilepsy. Springfield, IL, Thomas, 1958, pp 358367
18. G l m r P: Contributions of electroencephalography and electrocorticography to the neurosurgical treatment of the epilepsies. Adv Neurol 8:59-106, 1975
19. Goddard GV, McIntyre DC, Leech CK: A permanent change
in brain function resulting from daily electrical stimulation.
Exp Neurol 25:295-330, 1969
20. Horsley V: Brain-surgery. Br Med J 2:670-675, 1886
21. Jasper H, Pertuisset B, Flanigin H : EEG and cortical electrograms in patients with temporal lobe seizures. AMA Arch
Neurol Psychiatry 65:272-290, 1951
22. Kennedy WA, Hill D: The surgical prognostic significance of
the electroencephalographic prediction of Ammon's horn
sclerosis in epileptics. J Neurol Neurosurg Psychiatry
21:24-30, 1958
23. Kl6ve H, Grabow JD, Trites RL: Evaluation of memory functions with intracarotid sodium amytal. Trans Am Neurol
Assoc 94:76-80, I969
24. Kuhl DE, Engel J Jr, Phelps ME: Emission computed tomography: applications in stroke and epilepsy. In Bes A, Geraud
G (eds): Cerebral Circulation and Neurotransmitters (Proceedings of the International Congress on Cerebral Circulation, Toulouse, France, Sept 27-29, 1979). Amsterdam, Excerpta Mcdica, 1980, pp 37-43
25. Kuhl DE, Engel J Jr, Phelps ME, Selin C: Patterns of local
cerebral metabolism and perfusion in partial epilepsy by emission computed tomography of '*F-fluorodeoxyglucose and
13NN-ammonia.Acta Neurol Scand 6O:suppl 72:538-539,
1979
26. Kuhl DE, Engel J Jr, Phelps ME, Selin C: Epileptic patterns of
local cerebral metabolism and perfusion in humans determined by emission computed tomography of '*FDG and
'3NH:,. Ann Neurol 8:348-360, 1980
224
Annals of Neurology
Vol 9 No 3
March 1981
27. Lieb JP, Joseph JP, Engel J Jr, Walker J, Crandall PH: Sleep
state and seizure foci related to depth spike activity in patients
with temporal lobe epilepsy. Electroencephalogr Clin Neurophysiol 49: 5 38-5 5 7 , 1980
28. Lieb JP, Woods SC, Siccardi A, Crandall PH, Walter D O ,
Leake B: Quantitative analysis of depth spiking in relation to
seizure foci in patients with temporal lobe epilepsy. Electroencephalogr Clin Neurophysiol44:64 1-663, 1978
29. Margerison JH, Corsellis JAN: Epilepsy and the temporal
lobes: a clinical, electroencephalographic and neuropathological study of the brain in epilepsy, with particular reference to
the temporal lobes. Brain 89:499-530; 1966
30. Meldrum BS, Horton RW, Brierley JB: Epileptic brain damage in adolescent baboons following seizures induced by allylglycine. Brain 97:407-418, 1974
31. Milner B, Branch C, Rasmussen T: Study of short-term memory after intracarotid injection of sodium amytal. Trans Am
Neurol Assoc 87:224-226, 1962
32. Morrell F: Secondary epileptogenic lesions. Epilepsia 1:538560, 1960
33. Morris AA: Temporal lobectomy with removal of uncus, hippocampus, and amygdala: results for psychomotor epilepsy
three to nine years after operation. AMA Arch Neurol Psychiatry 76:479-496, 1956
34. Pampiglione G, Kerridge J: EEG abnormahties from the temporal lobe studied with sphenoidal electrodes. J Neurol
Neurosurg Psychiatry 19:117-129, 1956
35. Penfield W, Flanigin H : Surgical therapy of temporal lobe seizures. AMA Arch Neurol Psychiatry 64:491-500, 1950
36. Penfield W, Jasper H : Epilepsy and the Functional Anatomy
of the Human Brain. Boston, Little, Brown, 1954
37. Phelps ME, Hoffman EJ, Huang SC, et al: ECAT a new computerized tomographic imaging system for positron-emitting
radiopharmaceuticals. J Nucl Med 19:635-647, 1978
38. Phelps ME, Hoffman EJ, Raynaud C: Factors which affect cerebral uptake and retention of "NH,. Stroke 8:694-702,
1977
39. Phelps ME, Huang SC, Hoffman EJ, et al: Tomographic measurement of local cerebral metabolic rate in humans with (lRF)
fluorodeoxyglucose: validation of method. Ann Neurol 6:
371-388, 1979
40. Rasmussen T, Branch C: Temporal lobe epilepsy, indications
for and results of surgical therapy. Postgrad Med 31:9-14,
1962
41. Rossi GF: Problems of analysis and interpretation of electrocerebral signals in human epilepsy. A neurosurgeon's view.
In Brazier MAB (ed): Epilepsy: Its Phenomena in Man
(UCLA Forum in Medical Sciences No. 17). New York,
Academic, 1973, pp 259-285
42. Soper HV, Strain GM, Babb TL, Lieb JP, Crandall PH:
Chronic alumina temporal lobe seizures in monkeys. Exp
Neurol 62:99-121, 1978
43. Wada J: A new method for the determination of the side of
cerebral speech dominance. A preliminary report on the intracarotid injection of sodiumamytal in man. Igaku to Seibutsugaku (Japanese) 14:221-222, 1949
44. Walter RI>: Tactical considerations leading to surgical treatment of limbic epilepsy. In Brazier MAB (ed): Epilepsy: Its
Phenomena in Man (UCLA Forum in Medical Sciences No.
17). New York, Academic, 1973, pp 99-119
45. Walter RD: Principles of clinical investigation of surgical candidates. Adv Neurol 8:49-58, 1975
46. Wieser H G , Bancaud J, Talairach J, Bonis A, Szikla G: Comparative value of spontaneous and chemically and electrically
induced seizures in establishing the lateralization of temporal
lobe seizures. Epilepsia 20:47-60, 1979
Документ
Категория
Без категории
Просмотров
2
Размер файла
983 Кб
Теги
correlation, patients, foci, epileptic, localizing, surgical, criterias, consider, used, therapy, epilepsy
1/--страниц
Пожаловаться на содержимое документа