close

Вход

Забыли?

вход по аккаунту

?

Epilepsy surgery in the posterior cortex.

код для вставкиСкачать
Epilepsy Surgery in the Posterior Cortex
Warren T. Blume, MD, FRCP(C), Sharon E. Whiting, MD, FRCP(C), and John P. Girvin, MD, FRcS(C)
Fourteen (74%) of 19 patients obtained a significant reduction in seizures after posterior corticectomy; 6 (32%) were
seizure-free over a median follow-up of 3.7 years (range, 1 to 14 years). Surgery included limited resections of the
occipital lobe in 16 patients, posterior temporal region in 11, and posterior portion of parietal lobe in 7. Surgical
failure related to probable multiple areas of epileptogenesis (4 patients), or limited resections (2 patients) to preserve
to avoid dyslexia (1 patient). Of 14 patients without a complete hemianopia preoperatively,
increased visual field deficit, 2 (14%) of which were hemianopia. Four (36%) of 11 occipital
a new or increased visual field deficit: quadrantanopia in 3 and hemianopia in 1. Visual
phenomena were the most common initial ictal symptoms, occurring in 13 (68%) of the 19 patients. Twelve patients
had complex partial seizures: in 2, always without warning; in 7 , always following an aura, usually visual; and in 3
patients, with or without warning. Scalp electroencephalography identified the origin of most recorded seizures in 12
(63%) of the 19 patients. A principal interictal spike focus appeared in 15 patients (79%), and always correlated with
the epileptogenic lobe as defined by scalp and/or subdural-recorded seizures (14 patients) or by clinical analysis and
computed tomography (1 patient).
Blume WT, Whiting SE, Girvin JP. Epilepsy surgery in the posterior cortex. Ann Neurol 1991;29:638-645
Seizures arising from the occipital lobe or adjacent portions of the temporal and parietal lobes constitute only
a small proportion of focal epilepsies, yet they may be
therapy-resistant. Thus, Gibbs and Gibbs { 11 found an
occipital epileptiforrn focus in only 268 (8%)of 3,271
patients with focal seizures. Similarly, occipital resections constituted only 23 (1.8%) of 1,267 surgical procedures for epilepsy at the Montreal Neurological
Institute (MNI) from 1728 to 1771 [Z]. However,
occipital seizure disorders may continue for many
years: Seizures and electroencephalographic (EEG) abnormahties persisted for at least 10 years in 95% of
Ludwig and Ajmone Marsads 55 patients with occipital
EEG foci (31. This suggests that effective surgical therapy would be welcomed by these patients. Two concerns about epilepsy surgery in the posterior cortex
have likely decreased its use: threat of visual field loss
[4}and doubt about the ability of scalp EEG to localize
the epileptogenic focus [ S , 61.
Our series of 19 patients demonstrates that surgical
resection of part of the posterior cortex can improve
seizure control in the majority of patients, without
causing significant visual field loss. The series also
shows that ictal and interictal scalp EEG can reliably
localize the epileptogenic focus in most instances.
Materials and Methods
All patients of the University Hospital Epilepsy Unit who
underwent surgery involving the occipital, posterior temporal, and/or posterior parietal region for therapy-resistant sei-
From the Epilepsy Unit, University Hospital, The University of
Western Ontario, London, Ontario, Canada.
Received May 4, 1989. Accepted for publication Dec 17, 1990.
zures from 1974 to 1987 and who were followed for at least
1 year were included. Patients with brain tumors were included if the clinical presentation was an intractable seizure
disorder and the neoplasm was discovered during investigation. The median follow-up period was 3.7 years (average,
5.5 years; range, 1 to 14 years). The Epilepsy Unit files and
EEG database C73 provided seizure data, neurological findings (especially visual fields), EEG data, and computed tomography (CT)and/or magnetic resonance imaging (MRI)
results (Table 1).
A median of 12.5 preoperative scalp EEGs (average, 13.2;
range, 3 to 29) localized seizure origin or other epileptiform
activity. Except in early years, most were in-hospital telemetered recordings. Epochs of sleep occurred on more than two
recordings for all patients. Subdural recordings were performed on 5 patients to clarify ambiguities on scalp recordings. Except in 2 children, surgery was done under local anesthesia to permit outlining of language areas in dominant
hemisphere resections and continuous assessment of visual
fields during the procedure. Electrocorticography (ECoG),
performed on all patients inuaoperatively,guided the surgical resection by distribution of spikes.
In patients requiring subdural ECoG, strips were employed. These consisted of seven annular platinum contacts
at 10- to 20-mm intervals along a linear array, which were
led to a seven-pin connector by stainless-steelwires through
Silastic-coated Teflon tubing. Intraoperative ECoGs were
recorded by carbon-coated ball electrodes in monopolar fashion. Cortical exposures were determined by ictal features,
preoperative EEG and subdural recordings, and evidence of
lesions on scans.
Surgical outcome was characterized as follows: (1) seizure-
Address correspondence to Dr Blume, EEG Department, University
Hospital, 339 Windermere Road, London, Ontario, Canada N 6 A
5A5.
638 Copyright 0 1991 by the American Neurological Association
Table 1. Fatierit Data
lengrh
Sire 01
Surgery
of
F0l;oWEffect of up
~ u r ~ e r y ' (yr)
Ane
ar
Age at
Onset
(yr)
Operation
iyrl
I2
I!,
2
25
29
25
4
23
Initial
Seizure
Fearures'
Subsequent
Featuresb
Principal
lnrericral
EEG Spikes
lrtd
EEGC
Seizure
Imaging
Etiology
1. visual agnoaa
2. Pdcating visual
dimming, RUQ
1. Visual illusmn
2. CPS
1. Grand rnal
None
None
L occipital
tumor (CT)
L occipital
2
9
1. Tonic
2. CPS
02-76
02-1'6
1/13
R temporal
arrophy I C T I .
R 1~nm~phc.n.
atrophy IMRI)
R occipita-
5
1.2
i. Moving red balls,
1. Grand mal
None
None
2
2
1. CPS
2. Eyes right
P4-76-02
5, all
Left occipital
AVhl i.LT1
Normal ICT)
L occipital
R lower quadrant
1. Diffuse visual
blurring
2. Pleasurable sen-
R posterior
2
3.7
postenor
temporalanrerior
tcmpnral
remporalpaneral
ambGuuus
Satlo"
3. Visual experience
11
16
I. CRS
1. White bdu. RUQ
1. CPS
2. Formed visual
phenomenon
3 Visual blurring
4. Grand mal
5. Fear
I
35
16
1 Flashing green
light R U Q
2. Black spot, cenrer
5 . Disorientation
4. Abilonunal sensation
5. Grandmal
1 R a m tun^
2. Grand mal
T5
1. Flahing lights. R
1. R hemitomc
2 . CPS
75
field
2 mo
2. Diffuse blindness
i G r r d mal
1. Apnea
2. CPS
2
14
28
6
1. Bilateral llghts
2. V~sualexperiential
3. Deji vu
22
1. Visual dlusion
2. Feadul abdominal
sensation
3. corporeal XI,.
L hernrsphrrc
1'5
Unknown
(normal
mcroscopy:
L occipiralposterior
trnipural
1
4.2
Porencephaly
L orcrpltalposterior
temporal
paned
2
10.7
I'osenccphniitls
or pcrinatd
insult
I. orclpltal
5
11.8
5
4.1
atrophv IAEGI
414
L oc'lpltd
15
isubclimcal)
calcification
(AEG)
1. Eyes rlyht or
left
2. Blinking
3. Aronic diftuse.
conic diffuse.
L chic
Multiple
loci,
maximum
L hemispllrrr
T3-01
4/6.'
L megaloenccphaly ICT)
llniiateral
mcg:alracepidy
L occipital
1. CPS
2. Grand mal
F8-T4
F8-74
I/ I
R postcriur
Cavernous
angioma
I
1
1. CPS
P4-02
UZ-P4-T6
4!bL
R portenor
anterior
temporal
R Dccipitalparier al
1
32
I . Vwon dims
2. CPS
T6-02
Tb-02
3
2.9
1 . Eyea ryht
2. Grand mal
None
1
27
1. R motor
2. R visual
blurhg
3. Grand mal
ow3
Ol-TS
2IL
None
R mripitdposterior
temporal
L occlplral
4
81
1 . CPS
F7-T3-AI
F -73-AI
L occipitalposterior
temporal
panerd
L psrenor
temporal
3
1.6
L occipital
3
1.:
L occipitalparietal
3
11.8
L occipital
3
L occipitalposrcnor
temporal
parietal
5
7.9
R occipitalposterior
temporal
5
1.?
2. Abdominal
temporal cavernous
angioma iCT)
Normd (CTJ
Astrocvtorna
posrerior
temporalpanetal
sensatiw
satinn
4. Grandmal
8
1. Tingling in arms
21
29
1. Visual blurring,
32
flickering, right
9
23
1. R arm tingling
2. Grand mal
48
55
1 Experiential phcnomenon
2. CPS
1. Cephalic sensanon
10
19
2. AWominal
sensation
I.CPS
Normal ICTI
HallIUt,XlX*
Postoperative A V H ,
lefr occipital iCT)
AVM
L parietalorripitd
porencephaly rCT)
Porenccphaly
L postenor
Asrrocyromd
2 '4
1'1
01-TS
temporal glioma
(CT)
Normal (CT)
01-TS
2>125
3
18
0 1-T5
1. Cephalic sensation
2. Diffuse
visual
L hemiatrophy
iAEG)
Ul-T5
l/ 1
Cortl'al
dysplsm
Purenrcphaly
dimrmng
3. CPS
4. Grand mal
3
33
10 mo
12
14
I. Visual dimming
2. Diffuse & h a
3. Grandmal
1. Diffusc atomic
2. Diffuse myoclonic
3. Right clonic
4. Diffuse tonic
5. Grand d
1. Dilluse visual
blurring
2 Grand mal
25
1. Head, eyes
IJnknown
righr
2. Grandmal
3. AMominal
sensation
:noIillai
m'ruscopy)
T3
T3
=
left: R
7
Porencephaly
parietal-remporal
pxcnccphaly (CTj
1. CPS
2 Eyes, head left
3 D~ffusctonic
T6-02
Ambiguous Normal iCTJ
origin
*lnrtial seizure feature, art. the features at seizure onset
bSubsequenc seizure frarures are the phenomena appearing later in a seizure
'Irtal EEG fractions are proportions of recorded seizures clearly onginsting from arca.
dothers ahovc from contralared hemisphere.
'Clinical seizure when scqucnrial spckcs spread fmm P4 to 02, Tb.
'1 = seizure-free, no medication, 2 = seizure-free, one medication; 1 = > 7096 improvement; 4
L
L occipitd-
213
whr; CPS = complex partial ceizure: AVM
=
meriovenous malformation: RUQ
=
=
SD-90fi; mprovcment, 5
= e
.
w b t upper quadrant: AEG
=
50CG rmprovemcnr.
$r encephalogram.
Table 2. Symptoms cat Seizure Onset
(No. of Patients)
Visual, unformed
Contralateral only
Bilateral only
Both
Visual, formed
Never visual
Other experiential
Abdominal, cephalic,
corporeal
Unilateral sensory
Unilateral motor
Bilateral motor
Complex partial,
no warning
Never warning
11
3
5
3
6
6
4
5
1
free with no antiepileptic drug (AED); (2) seizure-free with
one AED; (3) greater than 90% reduction of seizure quantity; ( 4 ) 50 to 90% reduction; ( 5 ) less than 50% reduction;
and (6) worse.
Results
Age
The seizure disorder began at a median age of 9 years
(range, 2 months to 48 years), with investigation and
surgery occurring at a median age of 22 years (range,
2 to 55 years).
Seizure Onset
Visual phenomena were the most common initial ictal
symptoms, occurring in 13 (6896) of the 19 patients
(Table 2). Unformed phenomena such as diffuse blurring or flashing white or colored lights initiated at least
some attacks in 11 patients. Formed visual phenomena
such as “a large man in a living room playing cards”
appeared in some seizures of 6 patients, 4 of whom
had other seizures beginning with unformed visual
phenomena. S i x patients had unilateral unformed visual phenomena, all contralateral to EEG-recorded seizures. Of these, 3 also had bilateral visual phenomena;
5 additional patients had only bilateral phenomena. No
patient had ipsilateral visual symptoms.
Visual phenomena were never the initiation of the
seizures in 6 patients. In 2, visual phenomena followed
focal (1 patient) or diffuse (1 patient) somatosensory
phenomena with posterior parietal and posterior temporal originating seizures, respectively. A third patient,
with scalp and subdural EEG-documented left occipital
seizures, had only a cephalic sensation prior to his complex partial attacks. The attacks of a 55-year-old man
with a left posterior temporal glioma began by “feeling
outside himself looking in” but he denied that this was
a visual experience. The other patients were too young
640 Annals of Neurology Vol 29 No 6 June 1991
(1 patient) and mentally subnormal (1 patient) to indicate a visual symptom.
Three of the 5 patients whose seizures occasionally
began with abdominal, cephalic, or corporeal sensations more commonly had visual-initiating seizures.
The 1 patient with early unilateral motor phenomena
was unable to give a history.
Eight of 9 patients with occasional bilateral motor
(tonic, clonic, andlor tonic-clonic) seizures without
warning more commonly had partial seizures; a visual
onset was the most common in 6.
Four of 5 patients with at least some complex partial
seizures (CPS) without warning had electrographically
recorded clinical seizures; 3 patients’ seizures began in
the posterior temporal-occipital region. A fourth patient’s seizures ostensibly originated in the left anterior
temporal region although a left posterior temporal glioma must have played a role. A fifth patient, whose
EEG-recorded seizures had an ambiguous origin, had
right posterior temporal-occipital-parietal spikes.
All but 1 patient, including 4 of the 5 patients who
also had some CPS without warning, had at least some
seizures with auras.
Subsequent Seizure Phenomena
CPS followed a simple partial onset in 10 patients; in
7 patients these usually followed a visual warning; all
seven had unilateral posteriorly originating seizures on
EEG (5 patients) or subdural (2 patients) recordings.
The 2 patients with apparently anterior temporal originating seizures on EEG had posterior temporal lesions.
One patient had only a subclinical electrographic seizure from the left posterior temporal region.
Bilateral motor phenomena (tonic and/or clonic) appeared in the course of the seizures of 10 patients,
usually preceded by visual phenomena in 9 patients.
Unilateral motor (eyes {4 patients], head 121, limbs
141)attacks were preceded by visual symptoms in 5 of
the 7 patients; a sixth patient could not give a history.
Visual ictal symptoms preceded abdominal or diffuse
sensations in 2 of 3 patients. Somatosensory phenomena preceded visual seizures in 2 patients as described
above.
Complex Partial and Generalized Motor Seizures
Twelve patients had CPS: with warnings in all (7 patients), some (3), or none (2). Fifteen patients had generalized motor seizures, either at onset andtor subsequently.
Neurological Examination and Intelligence
Unilateral upper motor neuron signs, all contralateral
to seizure origins, were present in 6 patients. Preoperative neuropsychological testing showed severe subnormality (intelligence quotient < 50) in 2 patients, mild
Table 3 . Interictal Spikes and Seizure Origin (No. of Patients)"
1. PSF same as clinical seizure origin as recorded by:
EEG only
EEG and subdural recording
I
WJJV
I SEG
Fig I . Seiznre from ldt posterior temporal-occipital region (TT01) as sequential high-jiwquency rhythmic waves becoming sequential spikes. A jlashing green light appeared in the patient's
right upper visual field at the time of this recorded seizure.
to moderate retardation in 2, and normal intelligence
in 15.
EEG
The origin of at least one clinically typical seizure was
identified by scalp EEG in 15 patients (79%) (Fig 1).
All seizure origins were clearly depicted in 7 (47%,)of
the 15, or therefore 3?% of the 19 patients. Fifty percent or more of the attacks had a distinct origin in an
additional 5 patients (26%). Thus, scalp EEG identified
the ictal source in a majority of seizures in 12 (80%)
of 15 patients with recorded seizures, or 63% of the
19 patients. Most EEG-recorded clinical seizures of 3
patients had an ambiguous origin. A sixteenth patient
had a subclinical electrographic seizure that correlated
with the site of interictal spikes and an air encephalographic-detected lesion. Three patients did not have
multiple EEGs to record a seizure, as CT lesions corresponded to the clinical ictal phenomena.
The seizures of 14 patients originated from the same
lobe or region (occipital-temporal or occipital-parietal)
as the principal or exclusive spike focus (Table 3). Typical ictal symptoms accompanied these recorded seizures in 13 patients; the seizure was subclinical in 1.
In a fifteenth patient, principal spikes originated in the
lobes (parietal-occipital) implicated by the seizure history and CT but no seizure was recorded. Thus, a principal interictal spike focus was present in 15 patients
(79%) and always indicated the epileptogenic lobe.
Eleven patients (58%) had anterior temporal spikes
on at least two recordings; in 5 these appeared ipsilateral to posterior seizure origin and in 6 they were bilateral. Nine (82%) of these 11 patients had CPS. Three
(37%) of the other 8 patients had CPS. All patients
with anterior temporal spikes had more abundant
spikes posteriorly, and posterior originating seizures
were recorded in 10 of these 11 patients. Four of these
patients, including the only one without a recorded
seizure, had posteriorly situated radiological lesions.
Subdural recording only
Subtotal
2. PSF same as subclinical seizure orgin
3. PSF in lobe implicated by seizure type and CT lesion; no recorded seizure
4. Multiple spike foci in same hemisphere as recorded seizure
5. No spikes but EEG-recorded seizure
6. No spikes or EEG-recorded seizures
Total
8
3
2
13
1
1
1
1
2
19
aPSF = principal spike focus; most accive spike focus for most scalp
EEG recordings; ratio of EEGs in which PSF predominated to those
where the second most active focus predominated ranged from 17:0
to 13:2 among the 15 patients with a PSF.
Subdural EEG verified the putative epileptogenic
lobe, based on clinical analysis and scalp EEG data, in
all 5 patients in whom it was performed. One patient
with a left occipital seizure onset as determined by
scalp EEG was studied because of apparent rapid seizure propagation contralaterally and anteriorly (Fig 2).
His subdurally recorded seizures arose from the left
occipital region (Fig 3). In 4 others electrodes were
implanted to study whether apparently right posterior
originating seizures arose instead from the anterior
temporal region. Seizures of all 4 arose from the posterior region.
Radiology
CT scans showed focal posterior lesions in 7 patients:
atrophic lesions (3 patients), tumors (2), and vascular
abnormalities (2). Hemispheric megalencephaly was
noted in 1 patient. CT scans of 7 patients were normal;
tumors were found in 2 of these 7 patients at operation.
In 1 patient, MRI disclosed a large temporal lobe,
which remains incompletely explained even by pathological examination. Air encephalograms ( 3 patients)
showed hemispheric atrophy in 2 patients and an occipital calcified scar in 1; none were normal.
Thus, radiological examinations disclosed unilateral
posterior abnormalities in 8, other unilateral abnormalities in 4 (hemispheric 13 patients}, temporal {l patient]) and normal findings in 7.
All focal and hemispheric radiological lesions correlated to lobe and hemisphere of seizure origin as determined by recorded seizures (7 patients), clinical analysis only (3), and clinical analysis with interictal spikes
(2). Seizures of 2 of the 7 patients arose from the
midtemporal region just anterior to posterior temporal
lesions.
Blume et al: Epilepsy Surgery in Posterior Cortex 641
Fig 2. (A)L4t occipital (01) originating seizure as highfrequency polyspikes resembling beta, with extensive ldt hemisphere propagation but with mininal initial right hemisphere
involvement. lctal potentials are distinct from normal anterior
beta. (B) Same seizure. 10 seconds later; bilateral posterior .sequential spikes, principally left, with bilateral rhythmic theta
anteriorly. This unusual propagation occasioned subdural recordings.
i 4-0,
1516-
17-
7475-
76-
Etiology
Etiology was assessed by combining the neurological
history with pathological findings, both of which were
available for all 19 patients.
Five patients had a presumed perinatal neurological
insult, judged from either a difficult birth or an unexpected neurological deficit at birth; 4 of these had porencephaly. Five patients had cortical developmental
abnormalities: nonvascular hamartomas (2 patients),
cortical dysplasias (2), and unilateral megalencephaly (1
patient). Astrocytomas were found in 3 patients and an
oligodendroglioma in 1. Vascular malformations were
found in 3 patients (AVMs in 2 and cavernous angioma
in 1). Two patients had nonspecific findings on microscopic examinations and no convincing etiological factors in their histories.
Locutions of Szlrgevy
Corticectomies involved limited resections of the occipital lobe in 16 patients, posterior temporal region in
11, and posterior portions of the parietal lobe in 7. The
642 Annals of Neurology Vol 27 No 6 June 1991
77-
C
Fig 3. (A)Location of LeJt convexity subdural electrodes (same
patient as in Figure 2). Left inferior occipital-temporaland
right hemisphere subdural electrodes are not shown. IB, C) Interictal polyspikes at ldt occipital concexity near occipital pole
(16, 17, 77) become more frequent and Propagate to 15, 14,
and 7 4 t o 76 at seizure onset. Two seconds later, prominent
attenuation occurs at 16 and 17: and diffusely; and high-frequency waves appear at 75 and 16: then hzgh-frequency waves
occur at 16.17, and 77. There waJ minimal anterior and contralateral propa~ationinat shown).
3 patients without occipital resections had posterior
temporal resections. Anterior temporal lobectomies
were carried out in 2 patients who also had posterior
temporal and/or occipital lobectomies.
The precise regions for resection were determined
by clinical, electrophysiological, and lesional factors.
Table 4. Eflect of Surgery (No. of Patients)
1.
2.
3.
4.
5.
Seizure-free with no medication
Seizure-free with one medication
>90% improvement
50-90s improvement
Nochange
Table 5 . Viszlal FieMs (No. of Patients)
2
4
7
1
5
19
Length of follow-up (yr)
Median
Average
Range
3.7
5.5
1-14
Twelve patients had lesions, identified preoperatively
(neuroimaging) and by gross inspection intraoperatively prior to resection. Discrete lesions (one lobe or
less) were found in 6 patients, but such exclusively
determined the sites of resection in only 2 patients. In
4 other patients with discrete lesions, active EEG
spikes (3 patients) and/or EEG-recorded seizures (3
patients) at the peripheries of such lesions extended
the resections. Six of these 12 patients had extensive
lesions (more than one lobe), and electrophysiological
findings determined the degree of resection within
each lesion: EEG spikes (6 patients), EEG-recorded
seizures (6), and subdurally recorded seizures (1 patient).
Seven patients had no demonstrable radiological lesion and therefore the resections depended on clinical
and electrophysiological data: EEG spikes (7 patients),
EEG-recorded seizures (5), and subdurally recorded
seizures (4).
Sites of resection were determined at least in part
by scalp interictal and ictal EEGs (17 patients), ECoG
(14 patients), and subdural EEG (5 patients).
Outcome
SEIZURES.
Table 4 gives results and follow-up length.
Six patients (32%) were seizure-free; 4 of the 6 take
nontoxic amounts of a single anticonvulsant. Eight additional patients (42%) achieved a worthwhile result
(>90%, 50 to 90% improvement), leaving 5 patients
(26%) without significant benefit. No patient was
worse after surgery.
Outcome did not differ between patients with no
gross lesion and those with discrete lesions. However,
3 (60%) of 5 patients with extensive lesions obtained
no benefit as compared to only 2 (14%) of 14 patients
with smaller or no lesions.
Whether or not seizures began with visual phenomena had no clear effect on outcome: 10 (77%) of 13
patients who had seizures beginning with visual symptoms were helped (350% improvement), while 4
(67%) of 6 patients without visual symptoms were
helped.
The seizure disorder generally began earlier among
Full before and after operation
Same deficit before and after operation
New deficit
Upper quadrant
Lower quadrant
Hemianopia (patchy)
the 5 patients who were not helped (median, 2 years;
average, 4.4 years; range, 9 weeks to 12 years) than
among those 14 who improved after surgery (median,
9.5 years; average, 13.1 years: range, 1 to 48 years) (t
= 2.50, two-tailed t test;p < 0.05). However, duration
of the seizure disorder among patients who were
helped (15-year median) differed minimally from the
duration in those who were not helped (13-year
median).
The following factors may account for lack of surgical effectiveness among the 5 who were not improved.
Four had active interictal spikes distant from the resection area: multiple temporal and extratemporal spikes
(4 patients) and anterior temporal spikes with CPS by
history (3 patients). Two of these had active bitemporal
spikes. Two patients had limited resections to preserve
visual fields ( 2 patients) and the dominant angular gyrus
(1 patient) and because of technical difficulty resecting
an inferior occipital epileptogenic region (1 patient).
Despite such limitations in individual patients, there
was no correlation between extent of resection and
outcome.
However, the presence of ipsilateral or bitemporal
spikes on two or more EEG recordings had no significant effect on surgical outcome in the group as a whole:
7 (64%) of 11 patients with such spikes gained at least
a 50% seizure reduction while 7 (87%) of 8 patients
without such spikes did so. In contrast, only 1 of 4
patients with multiple independent spikes (three or
more foci including nonhomologous foci in each hemisphere) on most EEG recordings were helped by surgery, while 13 of 15 patients with less frequently appearing or no multiple independent spike foci were
helped. Clearly, the significance of spikes outside of
the resection area remains unresolved.
Duration of follow-up for the 14 patients who were
helped (median, 3.4 years; average, 5.5 years) did not
differ significantly from those 5 who were not helped
(median, 4.1 years; average, 5.7 years).
Eight of the 19 patients had preoperative visual field deficits; 5 of them had hemianopia.
As indicated in Table 5, 6 (32%) of the 19 patients
obtained a new deficit from the surgical resection. The
upper quadrantopia in 2 patients possibly resulted from
associated anterior temporal resections. The lower
quadrantopia in 1 is incomplete; the other (complete)
VISUAL FIELDS.
Blume et al: Epilepsy Surgery in Posterior Cortex 643
underwent resection of a left occipital tumor. Two patients obtained right homonymous hemianopia after a
left posterior temporal resection and left occipital resection each. Three (19%) of 16 occipital resections
were associated with quadrantopia, and one (6%) with
an hemianopia.
Of 14 patients with visual fields at risk preoperatively (full fields, 11 patients; quadrantopia, 3), surgery
resulted in a deficit to 6 (43%). Fortunately, only 2 of
these 6 had hemianopia. Of 11 patients entering partial
occipital resection with fields at risk, only 1 (9%) developed a hemianopia, and 3 (27%) obtained quadrantopia.
Two of 6 patients with a new visual field deficit remain seizure-free with medication, 2 have a greater
than 90% reduction, and 2 were not helped.
Discussion
The only previous series of posterior corticectomies
for epilepsy, also of 19 cases, is that of Rasmussen {23
whose results were similar. Five of his patients and 6
of ours remain seizure-free while 13 patients in their
series and 14 in ours obtained a reduction in seizure
tendency.
Presumed causes of our incomplete success include
limited resections of an epileptogenic region to avoid
visual field loss and/or dyslexia (2 patients) and potentially epileptogenic areas distant from resection area (4
patients).
Concern about surgically induced visual field loss
will always limit posterior cortex resections and thus
their effectiveness for seizure control. However, our
series indicates that this impediment is less than what
others suggest [4]. Only 2 (14%) of 14 patients with
a preoperative field at risk (i.e., less than a hemianopia)
developed hemianopia.
Should more of our patients have had anterior temporal lobectomies instead of, or in addition to, posterior corticectomies? Clinical and experimental epilepsy
data have shown a link between occipital and anterior
temporal regions. Ludwig and Ajmone Marsan {3]
found automatisms in 16 (29%) of 55 patients with
epileptiform activity involving primarily the occipital
regions. Although the auras in those patients were
most commonly visual (26, 47%), visceral sensations
appeared first in 7 and nonvisual experiential phenomena in 4.Thirteen (68%) of our patients had initial
visual auras, while 5 (26%) had some seizures beginning with abdominal, cephalic, or corporeal sensations.
Ajmone Marsan and Ralston {81 induced seizures in
patients with occipital epileptogenic lesions with pentylenetetrazol and found EEG evidence of ictal spread
anteriorly in the temporal regions producing symptoms
appropriate to that region. Bancaud {91 and Takeda
and colleagues { 101 demonstrated such anterior propagation with intracerebral electrodes. Olivier and associ644 Annals of Neurology Vol 29 No 6 June 1391
ates [111 reported a patient with occipital seizures that
propagated to the anterior temporal region and whose
seizure disorder markedly improved after anterior temporal lobectomy. Collins and Caston [12), using '*C-2deoxyglucose to map penicillin-induced occipital epileptic activity in rats, showed propagation to entorhinal
areas and the hippocampus bilaterally, producing automatisms.
Indeed, 11 (58%) of our 19 patients had anterior
temporal spikes on two or more recordings and 9
(82%) of the 11 had CPS. Resection included the anterior temporal region in 2 of these 11 patients. However, greater evidence for posterior than for anterior
epileptogenesis led to the posterior resections: interictal spikes (1 1 patients), recorded seizures (10 patients),
and neuroimaging-detected lesions (4 patients). Moreover, CPS were more common as subsequent than initial ictal phenomena among the 12 patients with CPS.
Of the 5 unimproved patients, 3 had active anterior
temporal spikes and CPS, 2 of which were bilateral.
Moreover, 1 unimproved patient did have an anterior
temporal lobectomy. These data suggest that the most
epileptogenic anterior temporal lobe may not always
be ipsilateral to the posterior focus.
Value of EEG
Scalp EEG recorded a distinct origin of at least a majority of seizures in 12 (63%) of our 19 patients, and
origin of at least one clinical seizure was identified in
15 patients (79%). A principal spike focus was present
in 15 patients (79%) and always indicated the epileptogenic lobe. The considerable number of preoperative
EEGs (median, 12.5) may have clarified these relationships. Nonetheless, subdural EEG was deemed essential to the investigation in 5 patients; it clarified ambiguous seizure origins on scalp EEG in 4 patients and
confirmed an unusual propagation in a fifth patient.
These data indicate that while invasive EEG may
occasionally be required to clarify the origin and spread
of ictal activity, multiple ictal and interictal scalp EEGs
provide data that cannot be discounted and that may
suffice.
Our thanks to Maria Raffa who carefully prepared the manuscript.
References
1. Purpura DP, Penry JK, Walter RD, eds. Atlas of electroencephalography, vol 2: Epilepsy. Cambridge, MA: Addison-Wesley,
1952:222-224
2. Rasmussen T.Surgery for epilepsy arising in regions other than
the temporal and frontal lobes. Adv Neurol 1975;8:207-226
3. Ludwig BI, Ajmone Marsan C. Clinical ictal patterns in epileptic
patients with occipital electroencephalographic foci. Neurology
1 ~ 5 ; 2 5 : 4 6 3 - 4 ?1
4. Olivier A. Commentary: cortical resections. In: Engel J, ed. Surgical treament of the epilepsies. New York: Raven Press,
198?:405-416
5 . Williamson PD, Spencer SS. Clinical and EEG features of com-
6.
7.
8.
9.
plex partial seizures of extratemporal origin. Epilepsia 1986;27
(suppl 2):S46-S63
King DW, Laxer KD. Postscript: can the indications for specific
invasive approaches be identified? In: EngelJ, ed. Surgical treatment of the epilepsies. New York: Raven Press, 1987:371-376
Lemieux JF. Vera RS, Blume WT. Database system for EEG
laboratories. Presented at the Annual Meeting of the American
EEG Society, New Orleans, Oct 21, 1983
Ajmone Marsan C, Ralston BL. The epileptic seizure. Its functional morphology and diagnostic significance. Springfield, 1L:
Thomas, 1957:2 11-2 15
Bancaud J. Les crises epileptiques d’origine occipitale (etude
stereo- electroencephalographique). Rev Oroneuroophtalmol
1969;41:299-3 11
10. Takeda A, Bancaud J, Talairach J, et al. Concerning epileptic
attacks of occipital origin. Electroencephalogr Clin Neurophysiol 1970;28:647-648
11. Olivier A, Gloor P, Andermann F, Ives J. Occipitotemporal
epilepsy studied with stereotaxically implanted depth electrodes
and successfully treated by temporal resection. Ann Neurol
1982;11~428-432
12. Collins RC, Caston TV. Functional anatomy o i occipital lobe
seizures: an experimental study in rats. Neurology 1979;29:
705-716
Blume et al: Epilepsy Surgery in Posterior Cortex
645
Документ
Категория
Без категории
Просмотров
0
Размер файла
711 Кб
Теги
posterior, surgery, corte, epilepsy
1/--страниц
Пожаловаться на содержимое документа