close

Вход

Забыли?

вход по аккаунту

?

Comparative localization of foci in partial epilepsy by PCT and EEG.

код для вставкиСкачать
.
Lomparatwe lm&aOon
of Epileptic Foci in Partial Epilepsy
by PCT and EEG
A
T
1.
Jerome Engel, Jr, MD, PhD,*t§Ti David E. Kuhl, MD,§ Michael E. Phelps, PhD,§"
and Paul H. Crandall, MD*$r/
One or more interictal positron computed tomograms of 18F-fluorodeoxyglucosewere obtained on 50 patients with
partial seizure disorders. Ictal as well as interictal electroencephalographic (EEG) data were available for all 50
patients, with scalp, sphenoidal, and depth electrode recordings done on 27 and scalp and sphenoidal recordings
alone on 23. Thirty-five patients demonstrated one or more abnormal interictal zones of hypometabolism, while
combined EEG studies were localizing for 36. There were considerable disagreements between the location of
metabolic deficits and the epileptic focus revealed by individuul scalp and depth EEG recorded ictal and interictal
epileptiform activity; however, there was good correlation between the site of focal hypometabolism and the
epileptic focus determined by the combined results of all electrophysiological studies. When focal hypometabolism
and focal nonepileptiform EEG abnormalities (i.e., slow waves and attenuation of fast rhythms) were both present
in the same patient, their localization agreed completely. Metabolic and combined electrophysiological techniques
both occasionally produced false positive as well as false negative results. When used together, the EEG can confirm
that a hypometabolic zone is epileptogenic, while FDG scans may indicate whether an epileptic EEG focus represents a lesion or propagation from a distant site.
Engel J Jr, Kuhl DE, Phelps ME, Crandall PH: Comparative localization of epileptic foci in partial
epilepsy by PCT and EEG. Ann Neurol 12:529-537, 1982
Localization of the epileptic focus in partial epilepsy
is particularly important when surgical therapy is
anticipated [ 191. This has largely depended on electrophysiological measures such as electroencephalography (EEG) [2, 5, 6, 10, 11, 131, intraoperative
electrocorticography [ 1, 2, 131, and chronic direct
recordings from stereotaxically implanted depth
electrodes (SEEG) [l, 5, 10, 13, 16, 17, 20, 211.
These various electrophysiological techniques, however, are known to give rise to occasional conflicting
results and false localization [l, 5, 10, 13, 221,
perhaps due to propagation of epileptiform activity.
Interictal positron computed tomography (PCT) of
'*F-fluorodeoxyglucose (FDG) commonly demonstrates areas of decreased local cerebral metabolic
rate for glucose (LCMR,,,) in patients with partial
seizure disorders [4, 7-10, 151 and appears to have
the potential for providing independent confirmatory
information about the site of the epileptogenic lesion
[4, 101.
A study was carried out to determine how often
localizing information obtained from PCT studies
From the Departments of 'Neurology, +Anatomy, and $Surgery,
the $Laboratory of Nuclear Medicine, the "Division of Biophysics, and the T B r i n Research Institute, UCLA School of
Medicine, Los Angeles, CA 90024.
with FDG agrees with information obtain'ed from
electrophysiological measures routinely used to
identify an epileptic focus. In addition, when localization with PCT and EEG do not agree, it is important to attempt to determine the possible reasons for
the discrepancy. Fifty patients with partial epilepsy
were evaluated with ictal as well as interictal SEEG or
EEG recordings (or both) and with interictal FDG
scans. Both EEG and SEEG recordings were carried
out on 27 patients, and 23 had EEG alone.
Methods
Patient Population
The 50 patients reported here are the same as those included in the study described in a companion article [9].
Forty-seven were under evaluation at UCLA for possible
surgical resection of an epileptogenic focus. The patients
ranged in age from 4 to 45 years. Eight patients in this series had abnormalities o n x-ray computed tomographic
(XCT) scans, 4 of which were focal [9].Structural abnormalities have not been as commonly revealed by XCT in
our series as has been reported by others [ 121. Two focal
abnormalities were later determined not to be epilepReceived Jan 26, 1982, and in revised form Mar 25. Accepted for
publication Mar 26, 1982.
Address reprint requests to Dr Engel, Reed Neurological Re.
search Center, UCLA School of Medicine, Los Angeles, CA
90024.
0364-5 134/82/120529-09$01.25 @ 1982 by the American Neurological Association
529
hypometabolic zones in 10 [9]. Interictal surface
EEG studies were normal in 5 patients, while bilateral activity was either synchronous or else independent but equivalent on the two sides in 5 others.
In the remaining 40 patients, o n e or more interictal
spike foci were identified. EEG-recorded ictal episodes were all abnormal, but localization of ictal onsets could be determined for only 23 patients. All
interictal and ictal SEEG recordings were also abnormal, but in 5 interictal studies and 4 ictal studies,
abnormalities were either bilaterally equivalent o r
too diffuse to be localizing.
Correlation between the site of hypometabolism
and localization independently based upon individual
interictal and ictal EEG and SEEG data was relatively
poor (Tables 1, 2). However, when all available EEG
and SEEG data were combined to arrive at the best
electrophysiological localization of the epileptogenic
focus, there was good agreement with t h e location of
the zone of hypometabolism seen on F D G scan
(Table 3). I n some cases m o r e than o n e abnormality
was identified by o n e o r both techniques, but if o n e
abnormality was identified by both tests, these were
considered to be in agreement for the purposes of
the distribution shown in Table 3. By this analysis,
correlation between P C T and EEG localization was
highly significant (p < 0.01, simple chi-square). T h e
percentage of patients for whom localization could
not be obtained was essentially the same for both
tests.
togenic, while 2 appeared to involve the epileptic focus and
consisted of a n occipital cyst and a small, postirradiation
occipital glioma. To date, 29 of t h e patients included here
have undergone surgical resection of t h e suspected epileptogenic lesion 14 I.
Electroencephalogrupb.y
Twenty-three patients had only surface EEG recordings,
while both surface and SEEG recordings were carried o u t
during separate admissions o n 27 patients. In most cases,
chronic sphenoidal electrodes were also employed during
long-term monitoring. lctal recordings were obtained o n all
patients included in this study, 47 during radiotelemetry
monitoring and 3 during prolonged routine EEG studies.
The techniques of EEG recording 19, 101 and radiotelemetry monitoring [ 5 , 101 have been described elsewhere.
EEG and SEEG interictal spike foci were localized by
visual analysis of both direct and telemetry recordings.
Since EEG usually and SEEG always demonstrated bilateral
independent interictal spike discharges, lateralization of a
temporal lobe interictal spike focus in this study refers to
the side of predominant interictal spiking. The site of ictal
onset was determined from multiple iccal recordings and
was generally consistent for most seizures, unless otherwise
mentioned.
Positron Computed Tomogrupby
Tomographic studies were performed and evaluated as described previously 14,91. The presence and localization of
an abnormality were determined by two investigators (J. E.
and D. E. K.), o n e of whom (D. E. K.) did not have prior
knowledge of the electrophysiological localization. Visual
rather than quantitative measurements were chosen for this
study 14, 91.
Possible C a s e s of Complete Disugveements i n Location
T h e location of abnormalities based o n F D G scan and
combined EEGiSEEG recordings completely disagreed for 3 patients. O n e had only right occipital
hypometabolism (Fig l A ) , while the EEG showed an
ipsilateral temporal interictal spike focus and ictal
onset (Fig 2); SEEG studies were not performed. T h e
Results
Correlation of PCT with EEGISEEG
F D G scans demonstrated a single zone of abnormal
hypometabolism in 2 5 patients and two discrete
Table I . Correlation between Sites cf Hypometabolic Areas Found by PCT
and Epileptogenir Foci Loralized by Surfate E E G Recordings
Site of PCT Abnormality
Unilateral
Temporal
Temporal plus"
Occipital
Frontal
Hemisphere
Bilateral"
Normal
Total
Agree
18
11
4
3
6
3
3
3
1
2
2
0
0
0
2
1
2
0
0
1
0
1
4
15
50
2
17
Disagree
3
10
25
"Five temporal plus frontal, 1 temporal plus central.
"Three biteniporal, 1 unilateral temporal plus contralateral occipital.
530 Annals o f Neurology
EEG Ictal Onset
EEG lnterictal Spikes
No. of
Patients
Vol 12 N o 6 December 1982
Nondiag.
Agree
9
3
1
0
Disagree
2
1
2
Nondiag.
7
2
0
1
1
0
1
1
0
3
0
3
12
8
13
10
27
0
4
Table 2. Correlation beticen Sites of Hypometabolic Area.r Found by PC?'
and Epileptogenic Foci Localized b~ Stereohzxii- EEG Rerordings
Site of PCT Abnormality
Unilateral
Temporal
Temporal plusa
Occipital
Bilateralh
Agree
12
10
1
3
0
0
0
3
1
Normal
Total
Disagree
1
7
0
3
4
27
10
12
4
SEEG Ictal Onset
SEEG Interictal Spikes
N o . of
Patients
Nondiag.
Agree
Disagree
1
0
0
9
2
0
3
1
1
1
0
2
Nondiag.
1
0
0
1
3
0
5
2
5
11
12
4
"All temporal plus frontal.
hThree bitemporal, 1 unilateral temporal plus contralateral occipital
Table 3 . Correlation between Sites of Hypometabolic Areas
Found by PCT and Epileptogenic Foci Localization
Based on Best EEGISEEG Criteria"
PCT
Agree
Disagree
Nondiagnostic
Total
Abnormal
2 8 (18)
3 (1)
4
35 (19)
Normal
0
5 (2)
10
15 ( 2 )
Total
28 (18) 8 (3)
14
50 (21)
"Numbers in parentheses indicate lesions confirmed histologically
or at surgery.
Fig 1 . FDG .icans from the .3patients whose zones of-hypometaboli.\m did not coriwpond t o the localization of the
epileptogenicfocus according t o the best EEG csiteria. 14113
patients had normal X C T scans. ( A )This patient bad right occipital hypometabolim (arrow), but E E G telemetry revealed
right temporal interictal spikes and a right temporal it-tal on.ret
isee F i g 2). ( B ) FDG .studie.! revealed left temporal hypometaboli.tm (arrow) i n thi.c patient, although E E G telemetry demonstrated predominant right mesiotemporal interictcrl
.spike.r and a right temporai ictal onset. SEEG-recorded interictal .spikes UWK more prominent in the left mesiotempord
Jtructures, but it iuas thought that ictal discharges began most
consistently in the right posterior hippocampal gyru.! before
spreading to the left .side (.see Fig .3). (C)The initial FDG
scan i n this patient, pedormed with depth electrodes i n place,
X C T scan was normal in this patient, but there was
a left homonymous hemianopia and an angiogram
suggested a right parietal mass. Intraoperative corticography identified several right temporal and
suprasylvian spike foci, and the cortex appeared diffusely abnormal. Temporal biopsy revealed a heterotopia. A larger resection was not attempted, and
the true extent of the structural lesion is unknown.
T h e second patient had relative hypometabolism in
the left temporal region (Fig 1B) and bilateral independent EEG- and SEEG-recorded ictal onsets, pre-
revealed a zone of hypometabolism (arrow) high in the right
temporal area and a smaller zone in the left temporal lobe 1 irction not shown). Ho?6WKr, all SEEG-recorded seizuivs hegdn i n
a regionalfashion from left me.ciotempoi*alstructures isee F i g
4). (0)
A repeat FDG Jcan from the patient dejiribed in C.
pedormed one year follouing ldt anterior tempord Iobertomy,
demonstrated re.!olution of the right-sided hypometabolic zone.
Thi.r section was choserr t o conform t o the .section in C , and i.c
above the section best demonitidrig the lobertomy;howevev,
some left-iided h~pometabo~ism
ian be seen posterior!y (arrow).
This may be the result of surgical interruption o f affei-ent fibers
to the left posterior temporal and occipital areas. Thix is con.ri.stent with the presence,postoperatively, of a right superior
quadiwritanopia.
Engel et al: PCT and EEG Localization
531
dominately right-sided. This patient had a right anterior temporal lobectomy and has not benefited
from surgery. Since lobectomy was performed at another hospital, by suction, adequate pathological
evaluation is not available. In retrospect, the SEEGrecorded ictal onsets may not have been clearly
localized to the right temporal lobe (Fig 3).
The third patient (No. 117; also see [ 4 ] )had large
right and smaller left temporal zones of hypometabolism (Fig lC), but SEEG-recorded ictal onsets
(Fig 4 ) were consistently from the left temporal
region, contralateral to the larger FDG scan abnormality. This patient is seizure free following a left
anterior temporal lobectomy, but the resected
specimen contained no pathological abnormalities.
The right temporal zone of hypometabolism has subsequently disappeared (Fig 1D). As discussed in the
preceding paper [ 4 ] ,the first study was done when
intracerebral electrodes were in place, and the
hypometabolism might have been electrode induced.
This explanation, however, remains to be proved.
In summary, evidence suggests that 1 patient had
multiple lesions, the EEG studies may have been
wrong in the second, and in the third the F D G scan
may have been falsely localizing, perhaps as a result
of depth electrode induced transient metabolic
changes.
Norrnal PCT Scans with E E G I S E E G Localization
FDG studies may have missed the epileptogenic lesion entirely in 5 patients who had well-defined corn-bined EEG and SEEG localization. All 5 have undergone surgery and have benefited from the procedure,
532 AnnalsofNeurology
Vol I 2
F i g 2. EEG-recorded ictal onJetfiom the patient demihed in
Figure 1A. Although chewing movements produced muscle artifact ichir-hpurtJ,i ohscwes the on.iet of irtal EEG activity, 5
to 6 pet, second rL7.ythmic activity can he appreciated i n the
light hemisphere, icith phu.re reversal at the right sphenoidal
electrode, and ulso in the right temporal region. (Calibration: 1 ier, 100 pV.)
but only 2 demonstrated structural lesions. The F D G
scans were technically suboptimal in 4 patients. One,
who had mesial temporal sclerosis in the resected
specimen, was the first patient studied in this series,
and the scan could not be adequately interpreted.
Another had hippocampal sclerosis demonstrated at
surgery but not confirmed histologically, and the scan
did not completely include the temporal lobes.
Focally Abnormal PCT Scuns
without E E G I S E E G Localization
SEEG or EEG studies (or both) were ultimately
nondiagnostic i n 4 patients in whom areas of
hypometabolism were identified. Two of these
showed hypometabolism involving most or part of
one hemisphere (Fig 5A, B), but lateralization of the
ictal onset could not be determined from scalp and
sphenoidal EEG studies. The remaining 2 patients, 1
with unequal bitemporal hypometaboiism (Fig 5 C )
and the other with a unilateral temporal defect (Fig
5D), also underwent SEEG telemetry. In both, seizures were found to originate independently and
with equal frequency from both temporal lobes. Because EEG localization was not achieved, no surgery
No 6 December 1982
RSM
ROF
RMH
RMG
RMG
v
RPG
LSM 4
LOF
LMH
LMG
w
f l L W
?
A
LMG
LPG
!
\w
F i g 3 . SEEG-recorded ictal onset fiwn the patient de.wibed in
Figire I B . This recording was made from bipolar conr-rntric
and multicontart elertrodes implanted into right (R) and left
(L) supplenzentary motor (SM) and orbitofrontal (OF) cortex,
mid (M) and posterior (P)hippocanzpus (H),and hipporanipal
gyrus (G).Ictal discharges were first seen i n the right posterior
hippocampal g y m approximately one second before ictal actirs
it31 r-ou/d be apprer-iatedi n the left hippocanipus and hipDocantpal g-yrus. Arrows indicate presumed time of onsct of ictal
actirity in eash channel. (Calibration: 1 ser.)
was performed and the actual sites of the epileptogenic lesions are unknown.
Correlation of P C T Localization with the Site of
Nonepileptic Focal E E G ISEEG Ahnomalities
Although there was no significant correlation between the presence of hypometabolic zones and the
presence of focal nonepileptic EEGISEEG abnormalities [ 9 ] , 18 patients had zones of hypometabolism
on FDG scans and also had continuous focal slow
wave activity or attenuation of normal baseline
rhythms on either EEG or SEEG recordings.
Thiopental activation revealed focal attenuation of
barbiturate-induced fast activity in 12 patients with
hypometabolic zones, 5 of whom had exhibited no
focal nonepileptiform EEG or SEEG abnormalities
during nondrugged states. In all 23 patients, the lo-
cation of the focal nonepileptiform EEG or SEEG
abnormalities corresponded to the location of a zone
of hypometabolism. Five of these patients had two
zones of hypometabolism, and in 4 the site of focal
slow or attenuated activity corresponded to the
hypometabolic zone that was also demonstrated to be
the ultimate site of the epileptiform EEG or SEEG
abnormality. The fifth had SEEG-recorded ictal onsets that occurred with equal frequency in either
remporal lobe, and also had bilateral temporal zones
of hypometabolism. In this patient the nonepileptiform SEEG-recorded abnormalities were maximal
in the more hypometabolic temporal lobe. Twelve of
the 2 3 patients underwent surgical resection of the
hypometabolic zone; all have benefited, and all demonstrated pathological lesions in the resected specimen. In 2 others, XCT scans showed a glioma and
hemispheric atrophy at the site or side of the
metabolic and EEG abnormalities.
Complementary Use of PCT and
EEGISEEG Data ,for Surgical Etjaluation
Twenty-one of the 29 patients in this study who
underwent surgery were found to have structural lesions (see Table 3). One of these 21 had a biopsy, 1
had only partial resection, and another had mesial
temporal sclerosis seen during resection but not histologically confirmed because the tissue specimen
was incomplete. The remaining 18 with complete reEngel et al: PCT and EEG Localization
533
RAM
RAH
c
-
/
w
5
RMH
1
‘
7
sections and pathologically verified lesions either are
seizure free or have experienced a substantial reduction in seizure frequency after surgical therapy (Table
4). Consequently, it may be concluded that an
epileptogenic lesion was correctly localized and excised in these “best cases.”
The FDG scan was normal in 1 of these 18 patients
(No. 107), whose lesion was ultimately identified
electrophysiologically. In 4 others (Patients 111,
113, 127, and 133), PCT revealed two zones of
hypometabolism and EEG/SEEG studies determined
which zone was the origin of all or most of the ictal
events. Conflicting electrophysiological data were
obtained for 8 patients; these included apparent false
localization of the site of ictal EEG onset for 3 (Patients 107, 112, and 132), SEEG onset for l (Patient
115 ) , and interictal EEG or SEEG spikes (or both) for
6 (Patients 107, 109, 111, 113, I, and 132). Focal
FDG abnormalities helped to resolve these conflicts
in 7 of the 8 patients. The site of ictal SEEG onsec
was the most reliable indicator of the location of the
epileptogenic lesion; however, since October, 1979,
F i g 4. SEEG-recorded i c t d onset from the patient described in
Figure 1C. Thi.r recording was made from bipolar concentric
electrodes placed in the structures indicuted in Figure .3, as well
us in umygdala (AM)and anterior hippocampus (AH).Surfuce recordings from lateral temporal and sphenoidul electrodes
are also included. The tracing demonstrates thiJ pdtient’J.
stereotyped regional ictal on.ret invobzng ull lej9 temporul depth
electrodeJ-simultaneous(y. iCalidrution: 1 .rec.i
surgery has been performed on 5 patients of this series (Patients I through IV and VI) without depth
electrode evaluation, due in large part to the
confirmatory localization provided by FDG scans.
Discussion
The purpose of this study was to evaluate the ability
of FDG scans to provide localizing information in
partial epilepsy and to compare this capability with
that of more traditional electrophysiological studies.
In the few instances when conflicts between these
two types of functional evaluation were identified,
534 Annals of NeuroloAT Vol 12 No 6 December 1982
Fi g 5 . F D G scuns from the 4 patients u:hn demonstruted zones
of hypometubolism when EEG studies fuiled t n lnculize an
epileptogenicfocus. None have had surgery. iA) This patient
bud u right frnntul zone of h.ypometubolism (arrow). E E G
telemetry, however, shouled only bifrontul synchronous znterictal spikes and ictal onsets. The X C T scan was normal. ( B ) The
X C T scan in this patient ulus normal, h14t there is a lurge area
of hypometabolism in the right hemisphere extending from the
posterior frontal lobe t o the occipital lobe (arrows). E E G telmnie t r y demond truted independent lefi und right tenzporul Jpikes;
neither the site nor side of ictul onset could be determined. iC)
The X C T scan in rhis patient u'us norniul, but be had a lurge
left temporul zone of hypometabolism und u smuller znne in the
right temporal lobe (arrows). EEG and SEEG telemetry indicuted that ictul 0n.ret.c occurred independently, und with eyuul
frequency. from either temporul lobe. ( D ) Thi.r putient bud un
enlarged cisterna niagna seen on X C T s u n , hut she ulso had a
zone of hypometuboli.tm in the right posterior tenzporul lobe
(arrow). EEG und SEEG telemetry indicated thut both temporul lobes were equully capuble of generuting ictulevents.
Table 4 Correlation nf PCT and EEGISEEG Ftndzngi ILvth Pathologtially Verzjed Lwtvns
Patient
No.
107
109
111
112
113
115
116
I
I1
122
111
IV
127
130
132
133
135
VI
Date of
Operation
Side of
Operation
6178
9178
1179
RT
RT
LT
RT
RT
RT
LT
RT
RT
RO
RT
LT
LT
LT
LT
LT
RT
RT
3179
4179
5179
7/79
10179
11/79
4/80
5180
6180
8180
11/80
418 1
318 1
618 1
618 1
Agree
Disagree
Nondiagnostic
"Outcome: 1 = seizure free; 2
-
EEG
PCT
IIS
10
NEF
IIS
I0
NEF
Outcomea
N
RT
LET
RT
R > LT
RT
LT
RT
RT
RO
RT
LT
LET
LT
LT
LT > RO
RT
RT
D
D
A
A
D
A
0
0
A
D
0
A
D
D
D
A
0
O
A
A
0
A
A
0
A
A
A
A
A
D
A
A
0
A
A
A
A
0
D
A
0
A
A
0
0
0
0
. . .
. . .
. . . . . .
. . . . . .
O
A
A
A
A
A
A
A
A
A
. . .
. . .
. . . . . .
. . . . . .
0
0
D
0
0
0
O
0
A
O
A
A
A
A
A
A
A
A
A
0
. . .
. . . . . .
1
3
1
1
3
1
1
1
1
2
1
1
1
1
1
1
1
1
13
0
12
6
7
12
1
0
A
0
A
A
3
3
5
=
rare seizures; 3
=
8
.
3
7
SEEG
0
12
A
D
A
A
A
D
5
1
A
A
A
A
A
A
Pathological
Findings
MTS
Heterotopia
MTS
MTS
MTS
Tumor
MTS
MTS
Angiorna
Tumor
MTS
MTS
MTS
Tumor
MTS
MTS
MTS
MTS
11
0
2
worthwhile improvement.
11.5 = interictal spikes; I 0 = ictal onset; NEF = nonepileptic focal; A = agree; D = disagree; 0
temporal; F = frontal; 0 = occipital; N = normal; MTS = mesial temporal sclerosis.
=
nondiagnostic; R
=
right, L
=
left; T =
Engel et a[: PCT and EEG Localization
535
other data, including surgical outcome and pathological analysis, were often sufficient to allow some conclusions regarding the probability of obtaining false
negative and false positive localizing information
from PCT with FDG as opposed to EEGISEEG
studies.
Loculization Is Equally Good
with PCT or EEGISEEG
The frequent disagreements between findings on
FDG scans and individual interictal and ictal EEG
and SEEG studies reflect the degree of conflicting results obtained among these various electrophysiological techniques [ 101. However, when all interictal and
ictal EEG and SEEG studies were combined to determine the best electrophysiological criteria for
localization, the diagnostic yield for PCT and EEG
was essentially the same, and localizations were well
correlated. In most cases the disagreements between
metabolic and electrophysiological methods of localization resulted from identification of more than one
potential lesion by one or the other of these techniques.
Complete Di.ragreement between PCT
and EEGISEEG Was Rare
The fact that 3 patients showed complete disagreement between PCT and EEG localization is of some
concern. In 1 patient, probable false localization by
EEG and SEEG studies may be attributed to the
limits of resolution of these techniques for recording
rapidly propagated epileptiform discharges. Retrospective analysis could also support the conclusion
that a definitive electrophysiological localization was
unjustified in this patient. PCT was apparently falsely
localizing for another patient, in whom a transient
clinically insignificant contralateral zone of hypometabolism may have been induced by the depth
electrodes. Since no pathological abnormalities could
be demonstrated in the brain tissue removed from
this patient, even though she is now seizure free,
the exact location of the epileptogenic lesion cannot be considered proved. The possibility was considered that this contralateral zone of hypometabolism might have been due to a ”mirror focus” [18]
that resolved following successful surgery and
cessation of seizures. This explanation appears unlikely, however, because a contralateral abnormality
observed in another patient with depth electrodes
in place disappeared after electrode removal, before
the patient underwent surgery, without any change in
seizure frequency [4].The actual site of the epileptogenic lesion in the third patient was never determined, but multiple lesions were present, allowing
the possibility that both EEG and PCT were partly
correct.
536 Annals of Neurology
Vol 12
False Negutives Occurred with Both Tests
This study suggests that epileptogenic lesions which
are well localized by combined EEG and SEEG recordings and confirmed by surgical removal of a
structural abnormality and relief from seizures can be
completely missed by PCT with FDG. However, for
the 2 patients in whom this was the case, this result
could be attributed to technically inadequate scans.
For 3 others in whom PCT findings were negative, no
pathological confirmation was obtained, although all
have benefited from surgery. It is impossible to know
at this stage whether epileptogenic lesions revealed
by EEG o r SEEG were below the level of resolution
of both the F D G scans and the microscope. Undoubtedly, PCT will not identify all focal epileptogenic lesions, but the rate of false negatives in the
face of technically adequate scans remains to be determined. On the other hand, EEGISEEG studies
were nondia‘gnostic in 4 patients who revealed
hypometabolic zones. Unfortunately, none of these
patients met the criteria for surgery [ 5 ] , and the
metabolic deficits have not been proved to be the site
of epileptogenic lesions.
PCT Localization Carrelutes Better with the
Locution of Nonepileptir EEGISEEG Abnormalities
In contrast to the inconsistent localizing information
obtained from individual observations of EEG- and
SEEG-recorded focal epileptiform interictal and ictal
activity (which may have been propagated from distant lesions), there was excellent correlation between
the location of a zone of hypometabolism and the site
of focal nonepileptic abnormalities recorded by EEG
and SEEG. Both abnormalities were not always present in the same patient but, when they were, their
locations always coincided. This outcome suggests
that the pathological mechanisms underlying the
focal hypometabolism seen on FDG scan may have
more in common with the mechanisms of focal EEG
slowing and attenuation than with those of epilepriform EEG activity. Such a view agrees with evidence that the former can indicate the site of structural damage [3, 6, 10, 141 while epileptiform activity
can be recorded at a distance from the primary
epileptogenic lesion [ 101. The finding of pathological
changes in all these patients who underwent surgery
is consistent with this interpretation.
PCT and EEGISEEG Provide Complementary
Lorulizing Information for Surgical Therapy
The brains of patients with partial epilepsy often
contain diffuse or multiple areas of potential epileptogenicity [lo]. Consequently, the actual site of the
epileptogenic lesion or lesions may be uncertain even
after surgery. Eighteen “best cases” were chosen in
which pathological confirmation and good outcome
No 6 December 1982
strongly suggested that the lesion most responsible
for each patient's habitual seizures was in the resected specimen. These cases demonstrate how FDG
scan abnormalities aid in localization of the epileptogenic lesion when EEGiSEEG findings are inconsistent, and how EEGISEEG verifies the epileptogenic nature of a focal metabolic defect. Although
ictal SEEG onset remains the most reliable localizing
finding in our series, we have been able to avoid the
risk of depth electrode implantation in some patients
(since October, 1979) with good results due to the
added confirmatory information provided by PCT [ 5,
lo]. PCT of FDG has now become an important part
of the presurgical evaluation protocol at UCLA [ 5 ] .
It can be concluded that metabolic and combined
electrophysiological techniques are equally well
localizing, but that occasionally both are also capable
of producing false positive as well as false negative
results. FDG scans and combined EEG techniques
should be considered complementary; the EEG can
confirm that a hypometabolic zone is epileptogenic,
while the FDG scan may indicate that an epileptic
EEG focus is likely to represent a lesion, and not
propagation from a distant site. The ability to use
these two independent measures of cerebral function
together adds another dimension to localization of
epileptogenic lesions for surgical therapy.
Supported by Contract DE-AM03-76-SF00012 from the Department of Energy and by Grants 7R01-GM-24839, 1PO1-NS15654, and 2P5O-NS-02808 from the US Public Health Service.
Presented in part at the 105th Annual Meeting of the American
Neurological Association, Boston, MA, Sept 7-10, 1980.
We thank R. F. Ackermann, PhD, E. J. Hoffman, PhD, S. C.
Huang, DSc, N. S. MacDonald, PhD, G. D. Robinson, Jr, PhD,
Jorge Barrio, PhD, J. Miller, NMRT, F. Aguilar, NMRT, R.
Sumida, NMRT, E. Carr, AB, A. Ricci, A B , M. Williams, REEGT,
and D. Quinonez, REEGT, for their assistance.
4. Engel J Jr, Brown WJ, Kuhl DE, er al: Pathological findings
underlying focal temporal lobe hypometabolism in partial
epilepsy. Ann Neurol 12:518-528, 1982 (this issue)
5 . Engel J Jr, Crandall PH, Rausch R: Surgical treatment of the
partial epilepsies. In Rosenberg RN, Grossman RG, Schochet
S, et a1 (eds): The Clinical Neurosciences. New York, Churchill Livingstone (in press)
6. Engel J Jr, Driver M V , Falconer MA: Electrophysiological
correlates of pathology and surgical results in temporal lobe
epilepsy. Brain 98:129-156, 1975
7. 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 105:74-76, 1981
8 . Engel J Jr, Kuhl DE, Phelps ME: Patterns of ictal and interictal
local cerebral metabolic rate studied in man with positron
computed tomography. I n Akimoto H , Kazamatsuri H , Seino
M (eds): Advances in Epileptology: XJIIth Epilepsy International Symposium. New York, Raven, 1982, pp 145-1110
9. Engel J Jr, Kuhl DE, Phelps ME, Mazziotta JC: Interictal
cerebral glucose metabolism in partial epilepsy and its relation
to EEG changes. Ann Neurol 12:510-517, 1082 (this issue)
10. 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
11. Falconer MA, Serafetinides EA: A follow-up study of surgery
in temporal lobe epilepsy. J Neurol Neurosurg Psychiatry
26:154-165, 1963
12. Gastaut H, Gastaut JL: Computerized transverse axial romography in epilepsy. Epilepsia 17:325-336, 1976
13. Gloor P: Contributions of electroencephalography and electrocorticography to the neurosurgical treatment of the epilepsies. Adv Neurol 8:59-105, 1975
14 Gloor P, Ball G, Schaul N : Brain lesions that produce delta
waves in the EEG. Neurology (Minneap) 27:326-333, 1977
15 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
'WH,.
Ann Neurol 8:348-360, 1980
16 Lieb JP, Engel J Jr, Brown WJ, et al: Neuropathological
findings following temporal lobectomy related to surface and
deep EEG patterns. Epilepsia 22:539-549, 1981
17 Lieb JP, Engel J Jr, Gevins AS, Crandall PH: Surface and deep
EEG correlates of surgical outcome in temporal lobe epilepsy.
Epilepsia 22:515-538, 1981
18 Morrell F: Secondary epileptogenic lesions. Epilepsia 1:538-
560, 1960
19 Purpura DP, Penry JK, Walter RD (eds):Neurosurgical Man-
References
Ajmone Marsan C: Depth electrography and electrocorticography. In Aminoff MJ (ed): Electrodiagnosis in Clinical
Neurology. New York, Churchill Livingstone, 1980, pp
167-196
Bailey P, Gibbs FA: The surgical treatment of psychomotor
epilepsy. JAMA 145:365-370, 1951
Ball GJ, Gloor P, Schaul N : The cortical electromicrophysiology of pathological delra waves in the electroencephalogram
of cats. Electroencephalogr Clin Neurophysiol 43:346-361,
1977
azement of the Epilepsies.
Adv Neurol, vol 8, 1975
.
20. Spencer SS: Depth electroencephalography in selection of
refractory epilepsy for surgery. Ann Neurol9:207-2 14, 1981
2 1. Talairach J, Bancaud J, Szikla G, et al: Approche nouvelle d e
la neurochirurgie de I'epilepsie. Mgthodologie sttrtoraxique
et rPsultats thkrapeutiques. Neurochirurgie 20, Suppl I ,
1974
22. Wieser H G , 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 20:47-60, 1979
Engel et al: PCT and EEG Localization
537
Документ
Категория
Без категории
Просмотров
0
Размер файла
880 Кб
Теги
foci, eeg, localization, comparative, partial, pct, epilepsy
1/--страниц
Пожаловаться на содержимое документа