- Email: [email protected]
Regional brain glucose metabolism in patients with complex partial seizures investigated by intracranial EEG
Epilepsy Research, 12 (1992) 121-129 0920-1211/92/$05.00 © 1992 Elsevier Science Publishers B.V. All rights reserved
121
EPIRES 00484
Regional brain glucose metabolism in patients with complex partial seizures investigated by intracranial EEG
B. Sadzot a, R.M.C. Debets b, P. Maquet a, C.W.M. van Veelen c, E. Salmon a, W. van Emde Boas b, D.N. Velis b, A.C. van Huffelen c and G. Franck a aUniversity of Liege, Departement of Neurology and Cyclotron Research Center, Liege (Belgium), blnstituut voor Epilepsiebestrijding 'Meer and Bosch - De Cruquiushoeve', Heemstede and CUniversity Hospital Utrecht, Department of Neurosurgery and Clinical Neurophysiology, Utrecht (Netherlands) (Accepted 10 April 1992)
Key words: Positron emission tomography; ~SF-Fluorodeoxyglucose; Partial epilepsy; Intracranial EEG recordings
We performed interictal lSF-2-fluoro-2-deoxy-D-glucose positron emission tomography (~SFDG-PET) studies in 57 patients with complex partial epilepsy (CPE), not controlled by medical treatment and considered for surgical resection of their epileptic focus. A precise localization of the epileptic focus was obtained in 37 of these patients with a combination of subdural and depth electrodes. We visually inspected the metabolic images; we also measured glucose consumption in a number of brain regions and compared the values with those obtained in 17 normal controls. Eighty-two percent of the 57 patients had an area of glucose hypometabolism on the l SFDGPET images. Six patients had a frontal epileptic focus, 3 of them had a frontal lobe hypometabolism. Twenty-six patients had a unilateral temporal lobe focus and all of them displayed a temporal lobe hypometabolism. The asymmetry was more pronounced in the lateral temporal cortex ( - 2 0 % ) than in the mesial part of the temporal lobe ( - 9 . 6 % ) . In each cortical brain region on the side of the epileptic focus (except the sensorimotor cortex), glucose consumption rate was lower than in the contralateral region or than in controls. No differences could be found between patients with a seizure onset restricted to the hippocampus and patients with a seizure onset involving the hippocampus and the adjacent neocortex. Divergent metabolic patterns were obtained in 5 patients with bilateral temporal seizure loci. Combined with other non invasive techniques (EEG, neuroradiology), PET contributes increasingly to the selection of patients with CPE who could benefit from surgical treatment. It could, in some cases, eliminate the need for IcEEG recordings, while in other cases it could help target the area to be explored.
Introduction
Correspondence to: Bernard Sadzot, Department of Neurology, University Hospital of Liege, B.35, 4000 Liege, Belgium. Abbreviations: PET, positron emission tomography; IcEEG, intracranial electroencephalography; CMRGIu, cerebral metabolic rate for glucose; Ctx, cortex; L, left; R, right; FS, focus side; NFS, non focus side; 18FDG, [18F]2-fluoro-2-deoxy-Dglucose; CPE, complex partial epilepsy; CPS, complex partial seizures; MRI, magnetic resonance imaging.
Some patients with medically refractory complex partial seizures may benefit from a neurosurgical resection of their epileptic focus. The presurgical evaluation process usually includes a period of non invasive neurophysiological monitoring during which the patient's usual seizures are recorded. Surface EEG recordings obtained during complex partial seizures can help to pinpoint the brain
122
regions from which the seizures are likely to originate. For a number of reasons, however, the accuracy and the reliability of non invasive recordings must be questioned, because of sampling errors due to the propagation of the epileptiform discharges 3'21'24. This explains why intracranial EEG recordings (IcEEG) are usually carried out before a surgical treatment is decided. IcEEG is still considered the most reliable technique, as far as the precise localization of the epileptic focus is concerned 5'23. These IcEEG recordings, however, are complex and time consuming, and they are not devoid of risks. They represent the most important limitation to the number of patients that can be treated by surgery. Thus, there is a real need for a technique that could confirm surface EEG findings and thereby avoid IcEEG before surgery in selected cases. It has been shown with PET and 18FDG, that between seizures, local cerebral metabolic rate for glucose (CMRGIu) is decreased in the region of the epileptic focus in 70~80% of patients with drug resistant complex partial epilepsy TM. The reliability of ~SFDG-PET for confirming the side of the epileptic focus is now well established. We wished to determine whether visual inspection of the metabolic images and quantitative analysis of the regional values of glucose consumption could be used to better localize the epileptogenic focus within the affected cerebral hemisphere. We present here our tSFDG-PET findings in 57 patients with drug resistant CPE, 37 of whom were later studied by IcEEG and operated. Methods P E T studies
PET was carried out in controls and in patients during the interictal state. We used the ~8FDG method 18 according to a standardized protocol developed in our laboratory 15. The subjects were asked to lie quietly in supine position on the table of the tomograph with eyes closed and ears free. Ambient light and noise were reduced as much as possible. A catheter was inserted in the humeral artery under local anesthesia. A bolus of 5-10 mCi tSFDG was given intravenously. Arterial blood samples (_+ 25 times 2 ml) were withdrawn at a de-
creasing rate during the study. Scan data collection was started 45 rain after injection. Six to 15 planes (8 mm apart) were acquired with a NeuroEcat PET scanner (EG&G, Ortec), in the medium resolution mode (12 mm in plane resolution, FWHM). The imaging planes were parallel to the orbitomeatal plane. Attenuation correction was calculated using an operator-positioned skull fitting ellipse algorithm and a uniform absorption coefficient of 0.088 cm 1 Cerebral metabolic rates for glucose were calculated pixel by pixel, using the operational equation of Phelps et al. 18, as well as the rate and lumped constant proposed by these authors for the adult man. Parametric images were then obtained and regional values of C M R G l u were collected, following a standard template of regions of interest (2.232.69 cm2) 15. A control population of 17 subjects (mean age: 24 years old) was studied to define reference values and confidence intervals. Thereafter, 57 epileptic patients were studied in the same conditions. The analysis of the PET data was performed in a single blind manner and in two steps. First, we carried out a visual assessment of the metabolic images for the symmetry of glucose consumption between homologous regions of the 2 hemispheres. A PET study was considered abnormal if an area of relative decrease in glucose metabolism could be seen on at least 2 contiguous PET slices. The brain side with the hypometabolic area was later referred to as the focus side (FS), while the contralateral side was called the non focus side (NFS). Subsequently, the quantitative data were analyzed. In each region of the controls, a 95% prediction interval for the regional C M R G l u values was calculated according to the formula m + k s where m is the mean, s is the standard deviation of the mean and k is x/(l + l/n) Qt(0.975; n - 1 ) . Qt(0.975; n - 1 ) is the 95% quantile of the Student's t-test distribution at n - 1 degrees of freedom ~3. Comparisons between regions of the 2 hemispheres were carried out using a paired t-test; regional values in epileptics were compared to the values obtained in controls and considered as reference values for a t-test. All the results were considered significant at the P~<0.05 level. Finally, a percentage difference between homologous regions of the 2 hemispheres was calculated
123
following the formula [ ( L - R ) / R ] x 100 in controis and [ ( F S - N F S ) / N F S ] x 100 in patients. Intracranial EEG IcEEG monitoring was performed with a combination of subdural and intracerebral electrodes 28. Miniaturized flexible subdural multicontact electrodes, combined into bundles or reeds, were introduced through two symmetrical fronto-central trephine holes (diameter 2.5 cm) and manipulated under fluoroscopy in order to cover most of the cerebral convexity. The number of individual electrodes varied from 5 to 12 per patient (mean = 9). In addition and using the same trephine holes, 2 4 depth electrodes were stereotactically implanted in the mesial temporal and/or mesial frontal structures. Each electrode had 7 recording leads. An extensive recording area and a detailed spatial delineation of the epileptogenic zone can thus be obtained from 100 recording contacts with minimal risks to the brain. Antiepileptic medication was gradually reduced during the EEG monitoring which continued for 5-12 days until a sufficient number of the patient's usual seizures had been recorded. The EEG seizure onsets were considered focal when ictal changes were initially confined to mesiolimbic electrodes. Initial changes simultaneously involving more than 4 adjacent contacts on one electrode or more than one subdural or 2 ipsilateral mesiolimbic electrodes were classified as widespread.
Results Over the past 5 years, 57 patients (33 men, 24 women; mean age: 32 years) with resistant complex partial epilepsy were included in the presurgical evaluation protocol. On X-ray CT scan, these patients had no gross brain lesion that would have required surgery by itself. Magnetic resonance imaging (MRI) was normal in 56% of the patients while it revealed either a localized change in signal, or widespread and non specific (but lateralized) structural changes (i.e., ventricular horn asymmetry) in 44%. The most frequent finding was a localized increase in signal on T2 weighted images (12 patients). The parametric images of the 18FDGPET studies were reviewed independently by two of us (B.S. and P.M.) for the symmetry of glucose metabolism between the 2 sides of the brain. The PET studies were rated abnormal in 82% of the patients whose PET images revealed a more or less widespread zone of hypometabolism on one side of the brain relative to the other. Out of the 57 patients, 37 underwent chronic IcEEG recordings, weeks to months after the PET study; the other 20 patients either will have an IcEEG in the near future, or, for various reasons, were excluded from the procedure. The decision was never based on the PET result alone. Patients with IcEEG were sorted out according to the site of seizure onset during the IcEEG. The results are given in Table I. We therefore compared the location and the extent of the area of glucose hypometabolism to the electrophysiologic location of the
TABLE I IcEEG results
Patients were divided according to the site of seizure onset during the IcEEG. Several seizures were recorded in each patient. Localization of the seizure onset (IcEEG)
n
Unilateral
Frontal lobe Amygdalo-hippocampal region (focal) Amygdalo-hippocampal region and/or adjacent neocortex (widespread)
5 17 9
13.5 46.0 24.3
Bilateral
Frontal Amygdalo-hippocampal region
1
5
2.7 13.5
37
100.0
Total
%
124
7
Fig. 1. 30 year old w o m a n with CPE since the age o f 12. A regional onset of abnormal electric activity at the beginning of the seizures could not be determined on surface EEG. MRI was normal. The PET study with ]8FDG revealed a decrease in glucose metabolism in the lateral cortex and in the mesial part of the left temporal lobe (on each image, the left is on the left). In this patient, PET was the only non invasive technique to indicate a focalization. She later underwent IcEEG monitoring indicating that the seizure onsets involved the left amygdalo-hippocampal complex and the adjacent neocortex.
seizure focus. In cases of frontal lobe epilepsy (n = 6), 3 patients had an area of glucose hypometabolism limited to the frontal lobe where the epileptic focus was later localized by IcEEG; in 2 patients, the hypometabolism involved the frontal and the temporal lobe as well, and it was not possible to determine on the PET images which lobe was the most involved; and finally, PET was normal in a patient in whom the ictal activity seemed to start in both frontal lobes. All patients with a unilateral temporal lobe seizure focus (n = 26) had an area of glucose hypometabolism, either on the same side of the brain as the epileptic focus (25 patients), or on the opposite side (1 patient). Two typical examples are shown in Fig. 1 and in Fig. 2. The hypometabolism was usually maximal in the temporal lobe but it is noteworthy that, in most patients, the hypometabolism was widespread, involving not only all the temporal cortex, but also adjacent cortical areas (i.e., frontal and parietal lobe), although to a lesser degree. In some patients, the hypometabolism also involved the homolateral basal ganglia and thalamus and the contralateral cerebellar lobe. For example, the homolateral thalamus appeared relatively hypome-
Fig. 2. 20 year old woman with a story of febrile convulsions at age 2. Since then, she experienced typical CPS. At the time of her presurgical evaluation, she had as m a n y as 3 seizures per day despite appropriate medical treatment. MRI disclosed an area of increased signal intensity on T2 weighted images at the level of the left hippocampus (left). On the tSFDG-PET images (right), a vast area of glucose hypometabolism was observed, involving all the left temporal lobe; note that the homolateral thalamus is also hypometabolic as compared to the right thalamus, icEEG monitoring indicated that her seizures originated from the left amygdalo-hippocampal complex.
125
TABLE II Glucose metabolic rates in selected regions of interest
For the epileptic patients, the percentage of asymmetry between homologous regions of the 2 hemispheres, calculated as [(FS - NFS)/NFS] x 100, is also indicated. The CMRGIu values presented for the controls were used as reference values. The regional 95% prediction intervals in controls are also indicated. CMRGlu (mg/100 g/min)
Anterior temporal cortex Mid temporal cortex Mesiotemporal area Insula Orbito-frontal cortex Frontal cortex Parietal cortex Sensorimotor cortex Thalamus Cerebellum
Controls (n = 17)
Patients (n = 31)
CMRGIu Prediction Interval (Mean) (L + R)/2
Focus side CMRGlu
5.27 5.93 4.27 5.49 6.30 5.91 5.29 5.40 6.38 5.09
3.19 4.30 2.78 3.87 4.12 4.00 3.74 3.68 4.63 2.58
6.98 9.38 6.00 8.49 8.62 9.10 8.09 8.28 9.76 7.54
Non focus side CMRGlu
Asymmetry
Mean
_+ SD
Mean
_+ SD
Mean (%) _+ SD
3.81ab 4.68ab 3.72ab 4.62~b 4.91ab
(_+1.07) (_+ 1.17) (_+0.92) (_+ 1.33) (_+ 1.42) (+ 1.41) (_+ 1.31) (+ 1.37) (_+ 1.33) (+ 1.43)
4.82 5.55 4.09 5.15 5.48a 5.31~ 5.07 5.20 5.83 4.86
(_+ 1.22) (_+ 1.23) (_+0.92) (_+ 1.36) (_ 1.43) (_+ 1.44) (_+ 1.37) (_+ 1.50) (___1.49) (_+ 1.43)
-21.5 ac - 15.7~c - 9.6a¢ - 9.7a¢ - 10.8ac - 3.4ac - 7.2a¢ - 4.1~c - 4.9ac 3.8ac
5 . 1 0 ab
4.70ab 4.98b 5 . 5 1 ab 5 . 0 4 ab
(+11.1) (_+ 10.0) (+ 8.4) (_+ 11.5) (_+ 7.8) (_+ 7.2) (_+ 11.4) (_+ 10.6) (+ 7.1) (_+ 5.5)
a Significantly different from controls (t-test). b Significantly different from the non focus side (paired Student's t-test). ¢ Significantly different from zero. The level of significance is always P ~< 0.05.
t a b o l i c in 7 5 % o f t h e p a t i e n t s w i t h a cortical h y p o metabolism. The quantitative data ( C M R G I u values) obtained in selected r e g i o n s o f i n t e r e s t f r o m p a t i e n t s w i t h a u n i l a t e r a l t e m p o r a l l o b e epileptic focus are g i v e n in T a b l e II. I n each s u b t e n t o r i a l r e g i o n , g l u c o s e m e t a b o l i s m was s i g n i f i c a n t l y l o w e r o n the side o f the epileptic focus t h a n o n the c o n t r a l a t e r a l ( n o r m a l ) side, a n d t h a n the n o r m a t i v e v a l u e s o b t a i n e d in c o n t r o l s . I n 6 8 % o f the p a t i e n t s , the greatest (or m a x i m a l ) p e r c e n t a g e o f a s y m m e t r y was b e t w e e n - 15 a n d - 3 5 % (Fig. 3). O n average, the h y p o m e t a b o l i s m was d e e p e r in the l a t e r a l c o r t e x o f the temporal lobe (-21.5% a s y m m e t r y ) t h a n i n the m e s i o t e m p o r a l a r e a ( - 9 . 6 % ) w h e r e the epileptic focus was located. T h e a v e r a g e C M R G l u v a l u e s in epileptics, e v e n o n t h e focus side, were well w i t h i n the 9 5 % c o n f i d e n c e i n t e r v a l d e f i n e d in c o n t r o l s ; m o r e o v e r , l o o k i n g at i n d i v i d u a l C M R G I u values, it was e x c e p t i o n a l w h e n it fell o u t s i d e this interval. It was n o t p o s s i b l e to s e p a r a t e p a t i e n t s w i t h a m e s i o t e m p o r a l seizure o n s e t f r o m those w i t h a m o r e w i d e s p r e a d ictal o n s e t o n the basis o f the v i s u a l
i n s p e c t i o n o f the P E T images; similarly, we c o u l d n o t find a q u a n t i t a t i v e i n d e x o r p a r a m e t e r s e p a r a t i n g t h e s e 2 c a t e g o r i e s o f p a t i e n t s ( T a b l e III). W e also a v e r a g e d the C M R G I u v a l u e s o f all the cortical areas, o n the focus side a n d o n the n o n focus
12
t-t~
"= t~
8
O_
"6 E
-i Z
4
0
Fig. 3. Distribution of the patients according to the degree of hypometabolism in the most hypometabolic area (largest percentage of asymmetry).
126
Discussion
TABLE II1
Quantitative parameters calculated in controls and in two subgroups of epileptic patients defined by the IcEEG results Controls (n = 17)
Maximum asymmetry (% ± SD) YAsC (% _+ SD) £AsC/nROI (% ± SD) Ant and Mid Temp. Ctx (FS) (CMRGlu ± SD)
Epileptics IcEEG
- 9.5 ± 2.9 - 5. l _+48.6 - 0.2 _+ 2.3 5.75
Focal ( n - 14)
Widespread (n=9)
- 29.8* ± 13.1 -166.8" ± 75.2 8.1" ± 3.2 4.13" ± 0.75
- 27.4* _+ 10.5 -193.5" ± 49.7 8.9* ± 2.4 4.80* ± 0.6t
Focal: seizure onset limited to the amygdalohippocampal complex; widespread: seizure onset involving the amygdalohippocampal complex and/or the adjacent neocortex. 5~AsC: sum of all the regional cortical asymmetries; nRoi: number of regions interest. * Significantly different from the value calculated in controls (P < 0.05, Student's t-test).
side (epileptic patients), or on the left and on the right side (normal controls) (Table IV). Glucose metabolism on the focus side was significantly lower than on the non focus side, or than the values in the controls; on the non focus side, glucose metabolism was also lower than in controls, but without reaching the level of significance. Finally, in 5 patients, the seizures appeared to start in both hippocampi. The PET images were falsely 'lateralized' in 2 of these patients, normal in 2, while in an additional patient, a relative glucose hypometabolism was seen on both sides, depending on the brain level examined.
TABLE IV
Average glucose metabolic rate (mg/lO0 g/rain) calculated from all the cortical areas
CMRGIu ± SD
Epileptics (n = 31 )
Controls (n = 17)
FS
NFS
L
R
5.02* _+1.26
5.48 _+1.28
5.88 ±l.10
5.89 _+1.22
* Significantly different from the non focus side and from the values calculated in normal controls (P < 0.05).
Our results in patients with complex partial seizures agree with and extend those reported by other investigators. The originality of our work is related to the large number of patients in whom the epileptic focus was precisely localized by a combination of subdural and depth electrodes, and to the quantification of the PET studies. In 1980, Kuhl et al. 14 showed that focal depression of glucose metabolism can be detected by l SFDG-PET on the side of the epileptic focus in patients with therapy resistant complex partial epilepsy. These findings have been confirmed by a number of investigators 1,4,7,19,26,27 including us 9"1°'2°. PET is now widely considered to be an important step in the presurgical evaluation of patients with CPE. Surface EEG is associated with sampling errors due to the propagation of ictal discharges and PET has become an independent confirmatory test of functional deficit. With the availability of the new non invasive neuroimaging modal±ties (PET, MRI), the detection and the localization of the structurally or functionally lesioned area corresponding or related to the epileptic focus has become easier. Therefore, epilepsy surgery no longer remains a desperate therapy of last resort. If adequate medical treatment with major ant±epileptic drugs fails to control the seizures, the surgical option should be considered as soon as 3 years following the onset of the disease< Some authors <8 have suggested that, depending on the surgical technique, surface EEG and PET, when congruent, can be sufficient, in selected cases, for directly considering epilepsy surgery, therefore avoiding lcEEG. All our patients with a unilateral temporal lobe epileptic focus proven by IcEEG had an area of glucose hypometabolism. In all the published series, the vast majority of which are based on surface EEG, the rate of positive findings is also high ( > 70%). Our rate is even higher but a bias in population selection could explain this maximal sensitivity. Indeed, the hypometabolism itself was considered in the process of deciding to further explore our patients by IcEEG. With the group of 57 patients as a whole, some of whom will not have an IcEEG, the sensitivity is 82%, which is similar to
127
that found in other series 7'2°'27. The hypometabolism was always observed on the same side as the epileptic focus, except for one patient whose hypometabolic area was actually observed on the opposite side. He had 2 PET studies with the same results. He had a temporal lobectomy on the side of the epileptic focus defined by IcEEG and he remains seizure free at more than 6 months after the intervention. A false lateralization is exceptional 6 and PET can be considered an extremely reliable test as far as the lateralization of the epileptic focus is concerned. The PET findings cannot be considered in isolation and should always be correlated with the results of other tests (MRI, video monitoring, neuropsychological evaluation): indeed, PET can be normal despite the existence of a well localized epileptic focus or it can reveal a lateralized hypometabolism despite bilateral independent seizure loci. Like others L4, we have been impressed by the extent of the area of glucose metabolism which, in most patients, involved not only all the temporal cortex but also adjacent cortical areas. Clearly, the hypometabolic area is much larger than the epileptic focus. The most hypometabolic area is usually, but not always, found in the lobe corresponding to the epileptic focus. The pathophysiology of the regional glucose hypometabolism found in epileptic patients remains unknown. The presence of a decrease in glucose metabolism in homolateral basal ganglia or thalamus and contralateral cerebellar lobe (thus far remote from the epileptic focus) in some patients suggests that a 'de-activation' phenomenon could be involved (i.e., a loss of activating influences). This could be a valid explanation for why the hypometabolism is so vast. There is no consensus about the best way to analyze the PET data collected in epileptic patients. We consider that, for individual cases, visual interpretation is the best approach because, as with other imaging modalities, the patient serves as his or her own control, and the eyes integrate far more information than is provided by a few regions of interest. It is important however to quantify these studies, first to obtain objective confirmation of what is seen, and second, because it is the only way to extract information from the patients as a group. For
example, we show that even on the non focus side, cortical metabolism tends to be lower than in controis, a finding that could reflect the effect of antiepileptic drugs 25. We demonstrated that glucose metabolism is significantly lower in all cortical regions, except for the sensorimotor cortex, on the FS as compared to the NFS. This indicates that the spatially limited epileptic focus influences or even disturbs the function of the whole hemisphere. This might explain the frequency of the cognitive and psychiatric or affective disturbances observed in these patients. The fact that the degree of glucose hypometabolism is greater in the lateral cortex of the temporal lobe than in its mesial part, where the epileptic focus is located, deserves some comment. The amygdalo-hippocampal complex is a small structure which is therefore not optimally imaged by our current tomograph. Only a limited number of brain slices can be obtained with the NeuroEcat, and one can never be sure that the imaging plane passes right through that structure; moreover, the spatial resolution is limited. Therefore, both imaging and quantification are affected by the partial volume effect 17. It is expected that more accurate information will be obtained with the newest tomographs. These multislice cameras are characterized by a better spatial resolution and spatial sampling, allowing tridimensional reconstructions. These technological improvements should result in a better knowledge of the relationships between the epileptic focus on the one hand and the extent and the degree of glucose metabolism on the other hand. It might therefore be possible to better localize the epileptic focus non invasively by PET. Other new interesting information is also expected from the use of other neurochemical markers such as ligands for opiate 12'16 or benzodiazepine 11'22 receptors. The experience acquired with PET in epilepsy illustrates the fact that PET is not only useful for clinical purposes but it is also an exceptional tool to investigate in vivo physiological and biochemical processes, yielding unique information on the pathophysiology of brain diseases.
128
C h e m i s t r y s t a f f f o r the ~ 8 F D G p r o d u c t i o n s , J. H o -
Acknowledgements
diaumont
for e x c e l l e n t t e c h n i c a l a s s i s t a n c e d u r i n g
T h i s w o r k was s u p p o r t e d by t h e N a t i o n a l F o u n -
the P E T studies, C. D e g u e l d r e f o r the m a i n t e n a n c e
d a t i o n f o r Scientific R e s e a r c h a n d b y t h e Q u e e n
a n d the c a l i b r a t i o n o f the t o m o g r a p h , a n d f o r h e l p
Elisabeth Medical Foundation
a n d a d v i c e in d a t a analysis, D. L a m o t t e
(Belgium). T h e au-
t h o r s w o u l d like to t h a n k all the C y c l o t r o n a n d
and D.
C o m a r f o r their s u p p o r t .
References 1 Abou-Khalil, B.W., Siegel, G.J., Sackellares, J.C., Gilman,
2
3
4
5 6
7
8
S., Hichwa, R. and Marshall, R., Positron emission tomography studies of cerebral glucose metabolism in chronic partial epilepsy, Ann. Neurol., 22 (1987) 480-486. Debets, R.M.C., van Veelen, C.W.M., van Huffelen, A.V. and van Emde Boas, W., Presurgical evaluation of patients with intractable partial epilepsy: the Dutch epilepsy surgery program, Acta Neurol. Belg., 91 (1991) 125 140. Driver, M.B. and MacGillivray, B.B., Electroencephalography. In: J. Laidlaw and A. Richens (Eds.), A Text Book of Epilepsy, Churchill Livingstone, London, 1976, pp. 109-144. Engel, J. Jr., Brown, W.J., Kuhl, D.E., Phelps, M.E., Mazziotta, J.C. and Crandall, P.H., Pathological findings underlying focal temporal lobe hypometabolism in partial epilepsy, Ann. Neurol., 12 (1982) 518-528. Engel, J. Jr., Functional localization of epileptogenic lesions, Trends Neurosci., 6 (1983) 6ff65. Engel, J. Jr., Henry, T.R., Risinger, M.W., Mazziotta, J.C., Sutherling, W.W., Levesque, M.F. and Phelps, M.E., Presurgical evaluation for partial epilepsy: relative contribution of chronic depth-electrode recordings versus FDGPET and scalp-sphenoidal ictal EEG, Neurology, 40 (1990) 1670--1677. Engel, J. Jr., Kuhl, D.E., Phelps, M.E. and Crandall, P.H., Comparative localization of epileptic foci in partial epilepsy by PCT and EEG, Ann. Neurol., 12 (1982) 529-537. Engel, J. Jr., Sutherling, W.W., Cahan, L., Crandall, P.H., Kuhl, D.E. and Phelps, M.E., The role of positron emission tomography in the surgical therapy of epilepsy. In: R.J. Porter, R.H. Manson, A.A. Ward and M. Dam (Eds.), Ad-
12
13 14
15
16
17
18
vances in Epileptology." XVth Epilepsy International Symposium, Raven Press, New York, NY, 1984, pp. 427
432. 9 Franck, G., Sadzot, B., Salmon, E., Depresseux, J.C., Grisar, Th., Peters, J.M., Guillaume, M., Quaglia, L., Delfiore, G. and Lamotte, D., Regional cerebral blood flow and metabolic rates in human focal epilepsy and status epilepticus. In: A.V. Delgado-Escueta, A.A. Ward, D.M. Woodbury and R.J. Porter (Eds.), Advances in Neurology, Vol. 44: Basic mechanisms of the epilepsies, Raven Press, New York, NY, 1986, pp. 935-948. 10 Franck, G., Salmon, E., Sadzot, B. and Maquet, P., Epilepsy: the use of oxygen-15-1abeled gases, Semin. Neurol., 9 (1989) 307-316. 11 Frey, K.A., Henry, T.R., Holthoff, V.A., Koeppe, R.A., Sackellares, J.C., Brunberg, J.A., Berent, S., Gilman, S. and Kuhl, D.E., Comparison of regional glucose metabo-
19
20
21
lism and in vivo benzodiazepine receptor binding in refractory temporal lobe epilepsy, J. Nucl. Med., 32 (1991) 912. Frost, J.J., Mayberg, H.S., Fisher, R.S., Douglass, K.H., Dannals, R.F., Links, J.M., Wilson, A.A., Ravert, H.T., Rosenbaum, A.E., Snyder, S.H. and Wagner, H.N. Jr., Mu-opiate receptors measured by positron emission tomography are increased in temporal lobe epilepsy, Ann. Neurol., 23 (1988) 231 237. Guttman, I., Statistical Tolerance Regions, Griffin, London, 1970. Kuhl, D.E., Engel, J., Phelps, M.E. and Selin, C., Epileptic patterns of local cerebral metabolism and perfusion in humans determined by emission computed tomography of 18FDG and 13NH3, Ann. Neurol., 8 (1980) 348 360. Maquet, P., Dive, D., Salmon, E., Von Frenckel, R. and Franck, G., Reproducibility of cerebral glucose utilization measured by positron emission tomography and the [18F]2-fluoro-2-deoxy-d-glucose method in resting, healthy human subjects, Eur. J. Nucl. Med., 16 (1990) 267-273. Mayberg, H.S., Sadzot, B., Cidis Meltzer, C.C., Fisher, R.S., Lesser, R.P., Dannals, R.F., Lever, J.R., Wilson, A.A., Ravert, H.T., Wagner, H.N., Jr., Bryan, R.N., Cromwell, C.C. and Frost, J.J., Quantification ofmu and non-mu opiate receptors in temporal lobe epilepsy using positron emission tomography, Ann. Neurol., 30 (1991) 3-11. Mazziotta, J.C., Phelps, M.E., Plummer, D. and Kuhl, D.E., Quantitation in positron emission computed tomography: 5. Physical-anatomical effects, J. Comp. Assist. Tomogr., 5 (1981) 734-743. Phelps, M.E., Huang, S.C., Hoffman, E.J., Selin, C., Sokoloff, L. and Kuhl, D.E., Tomographic measurements of local cerebral glucose metabolic rate in humans with [FI8]-2fluoro-2-deoxy-D-glucose: validation of method, Ann. Neurol., 6 (1979) 371 388. Sackellares, J.C., Siegel, G.J., Abou-Khalil, B.W., Hood, T.W., Gilman, S., McKeever, P.E., Hichwa, R.D. and Hutchins, G.D., Differences between lateral and mesial temporal metabolism interictally in epilepsy of mesial temporal origin, Neurology, 40 (1990) 1420-1426. Sadzot, B., Salmon, E., Maquet, P., Debets, R., Grisar, Th., Lemaire, Ch., Degueldre, Ch. and Franck, G., Assessment of the interest of PET and MRI in complex partial epilepsy, J. Cereb. Blood Flow Metab., 7 (Suppl. 1) (1987) $432. Sammaritano, M., de Lotbini~re, A., Andermann, F., Olivier, A., Gloor, P. and Quesney, L.F., False lateralization by surface EEG of seizure onset in patients with temporal lobe epilepsy and gross focal cerebral lesion, Ann. Neurol., 21
129
(1987) 361-369. 22 Savic, I., Roland, P., Sedvall, G., Persson, A., Pauli, S. and Widen, L., In vivo demonstration of reduced benzodiazepine receptor binding in human epileptic foci, Lancet, 2 (1988) 863-866. 23 So, N., Gloor, P., Quesney, L.F., Jones-Gotman, M., Olivier, A. and Andermann, F., Depth electrode investigations in patients with bitemporal epileptiform abnormalities, Ann. Neurol., 25 (1989) 423-431. 24 Spencer, S., Williamson, P.D., Bridgers, S.L., Mattson, R.H., Cicehetti, D.V. and Spencer, D.D., Reliability and accuracy of localization by scalp ictal EEG, Neurology, 35 (1985) 1567-1575. 25 Theodore, W.H., Antiepileptic drugs and cerebral glucose metabolism, Epilepsia, 29 (Suppl. 2) (1988) $48 $55.
26 Theodore, W.H., Fishbein, D. and Dubinsky, R., Patterns of cerebral glucose metabolism in patients with partial seizures, Neurology, 38 (1988) 1201-1206. 27 Theodore, W.H., Newmark, M.E., Sato, S., Brooks, R., Patronas, N., De La Paz, R., Di Chiro, G., Kessler, R.M., Margolin, R., Manning, R.G., Channing, M. and Porter, R.J., [18F]Fluorodeoxyglucose positron emission tomography in refractory complex partial seizures, Ann. Neurol., 14 (1983) 429-437. 28 van Veelen, C.W.M., Debets, R.M.C., van Huffelen, A.C., van Emde Boas, W., Binnie, C.D., Storm van Leeuwen, W., Velis, D.N. and van Dieren, A., Combined use of subdural and intracerebral electrodes in preoperative evaluation of epilepsy, Neurosurgery, 26 (1990) 93-101.
121
EPIRES 00484
Regional brain glucose metabolism in patients with complex partial seizures investigated by intracranial EEG
B. Sadzot a, R.M.C. Debets b, P. Maquet a, C.W.M. van Veelen c, E. Salmon a, W. van Emde Boas b, D.N. Velis b, A.C. van Huffelen c and G. Franck a aUniversity of Liege, Departement of Neurology and Cyclotron Research Center, Liege (Belgium), blnstituut voor Epilepsiebestrijding 'Meer and Bosch - De Cruquiushoeve', Heemstede and CUniversity Hospital Utrecht, Department of Neurosurgery and Clinical Neurophysiology, Utrecht (Netherlands) (Accepted 10 April 1992)
Key words: Positron emission tomography; ~SF-Fluorodeoxyglucose; Partial epilepsy; Intracranial EEG recordings
We performed interictal lSF-2-fluoro-2-deoxy-D-glucose positron emission tomography (~SFDG-PET) studies in 57 patients with complex partial epilepsy (CPE), not controlled by medical treatment and considered for surgical resection of their epileptic focus. A precise localization of the epileptic focus was obtained in 37 of these patients with a combination of subdural and depth electrodes. We visually inspected the metabolic images; we also measured glucose consumption in a number of brain regions and compared the values with those obtained in 17 normal controls. Eighty-two percent of the 57 patients had an area of glucose hypometabolism on the l SFDGPET images. Six patients had a frontal epileptic focus, 3 of them had a frontal lobe hypometabolism. Twenty-six patients had a unilateral temporal lobe focus and all of them displayed a temporal lobe hypometabolism. The asymmetry was more pronounced in the lateral temporal cortex ( - 2 0 % ) than in the mesial part of the temporal lobe ( - 9 . 6 % ) . In each cortical brain region on the side of the epileptic focus (except the sensorimotor cortex), glucose consumption rate was lower than in the contralateral region or than in controls. No differences could be found between patients with a seizure onset restricted to the hippocampus and patients with a seizure onset involving the hippocampus and the adjacent neocortex. Divergent metabolic patterns were obtained in 5 patients with bilateral temporal seizure loci. Combined with other non invasive techniques (EEG, neuroradiology), PET contributes increasingly to the selection of patients with CPE who could benefit from surgical treatment. It could, in some cases, eliminate the need for IcEEG recordings, while in other cases it could help target the area to be explored.
Introduction
Correspondence to: Bernard Sadzot, Department of Neurology, University Hospital of Liege, B.35, 4000 Liege, Belgium. Abbreviations: PET, positron emission tomography; IcEEG, intracranial electroencephalography; CMRGIu, cerebral metabolic rate for glucose; Ctx, cortex; L, left; R, right; FS, focus side; NFS, non focus side; 18FDG, [18F]2-fluoro-2-deoxy-Dglucose; CPE, complex partial epilepsy; CPS, complex partial seizures; MRI, magnetic resonance imaging.
Some patients with medically refractory complex partial seizures may benefit from a neurosurgical resection of their epileptic focus. The presurgical evaluation process usually includes a period of non invasive neurophysiological monitoring during which the patient's usual seizures are recorded. Surface EEG recordings obtained during complex partial seizures can help to pinpoint the brain
122
regions from which the seizures are likely to originate. For a number of reasons, however, the accuracy and the reliability of non invasive recordings must be questioned, because of sampling errors due to the propagation of the epileptiform discharges 3'21'24. This explains why intracranial EEG recordings (IcEEG) are usually carried out before a surgical treatment is decided. IcEEG is still considered the most reliable technique, as far as the precise localization of the epileptic focus is concerned 5'23. These IcEEG recordings, however, are complex and time consuming, and they are not devoid of risks. They represent the most important limitation to the number of patients that can be treated by surgery. Thus, there is a real need for a technique that could confirm surface EEG findings and thereby avoid IcEEG before surgery in selected cases. It has been shown with PET and 18FDG, that between seizures, local cerebral metabolic rate for glucose (CMRGIu) is decreased in the region of the epileptic focus in 70~80% of patients with drug resistant complex partial epilepsy TM. The reliability of ~SFDG-PET for confirming the side of the epileptic focus is now well established. We wished to determine whether visual inspection of the metabolic images and quantitative analysis of the regional values of glucose consumption could be used to better localize the epileptogenic focus within the affected cerebral hemisphere. We present here our tSFDG-PET findings in 57 patients with drug resistant CPE, 37 of whom were later studied by IcEEG and operated. Methods P E T studies
PET was carried out in controls and in patients during the interictal state. We used the ~8FDG method 18 according to a standardized protocol developed in our laboratory 15. The subjects were asked to lie quietly in supine position on the table of the tomograph with eyes closed and ears free. Ambient light and noise were reduced as much as possible. A catheter was inserted in the humeral artery under local anesthesia. A bolus of 5-10 mCi tSFDG was given intravenously. Arterial blood samples (_+ 25 times 2 ml) were withdrawn at a de-
creasing rate during the study. Scan data collection was started 45 rain after injection. Six to 15 planes (8 mm apart) were acquired with a NeuroEcat PET scanner (EG&G, Ortec), in the medium resolution mode (12 mm in plane resolution, FWHM). The imaging planes were parallel to the orbitomeatal plane. Attenuation correction was calculated using an operator-positioned skull fitting ellipse algorithm and a uniform absorption coefficient of 0.088 cm 1 Cerebral metabolic rates for glucose were calculated pixel by pixel, using the operational equation of Phelps et al. 18, as well as the rate and lumped constant proposed by these authors for the adult man. Parametric images were then obtained and regional values of C M R G l u were collected, following a standard template of regions of interest (2.232.69 cm2) 15. A control population of 17 subjects (mean age: 24 years old) was studied to define reference values and confidence intervals. Thereafter, 57 epileptic patients were studied in the same conditions. The analysis of the PET data was performed in a single blind manner and in two steps. First, we carried out a visual assessment of the metabolic images for the symmetry of glucose consumption between homologous regions of the 2 hemispheres. A PET study was considered abnormal if an area of relative decrease in glucose metabolism could be seen on at least 2 contiguous PET slices. The brain side with the hypometabolic area was later referred to as the focus side (FS), while the contralateral side was called the non focus side (NFS). Subsequently, the quantitative data were analyzed. In each region of the controls, a 95% prediction interval for the regional C M R G l u values was calculated according to the formula m + k s where m is the mean, s is the standard deviation of the mean and k is x/(l + l/n) Qt(0.975; n - 1 ) . Qt(0.975; n - 1 ) is the 95% quantile of the Student's t-test distribution at n - 1 degrees of freedom ~3. Comparisons between regions of the 2 hemispheres were carried out using a paired t-test; regional values in epileptics were compared to the values obtained in controls and considered as reference values for a t-test. All the results were considered significant at the P~<0.05 level. Finally, a percentage difference between homologous regions of the 2 hemispheres was calculated
123
following the formula [ ( L - R ) / R ] x 100 in controis and [ ( F S - N F S ) / N F S ] x 100 in patients. Intracranial EEG IcEEG monitoring was performed with a combination of subdural and intracerebral electrodes 28. Miniaturized flexible subdural multicontact electrodes, combined into bundles or reeds, were introduced through two symmetrical fronto-central trephine holes (diameter 2.5 cm) and manipulated under fluoroscopy in order to cover most of the cerebral convexity. The number of individual electrodes varied from 5 to 12 per patient (mean = 9). In addition and using the same trephine holes, 2 4 depth electrodes were stereotactically implanted in the mesial temporal and/or mesial frontal structures. Each electrode had 7 recording leads. An extensive recording area and a detailed spatial delineation of the epileptogenic zone can thus be obtained from 100 recording contacts with minimal risks to the brain. Antiepileptic medication was gradually reduced during the EEG monitoring which continued for 5-12 days until a sufficient number of the patient's usual seizures had been recorded. The EEG seizure onsets were considered focal when ictal changes were initially confined to mesiolimbic electrodes. Initial changes simultaneously involving more than 4 adjacent contacts on one electrode or more than one subdural or 2 ipsilateral mesiolimbic electrodes were classified as widespread.
Results Over the past 5 years, 57 patients (33 men, 24 women; mean age: 32 years) with resistant complex partial epilepsy were included in the presurgical evaluation protocol. On X-ray CT scan, these patients had no gross brain lesion that would have required surgery by itself. Magnetic resonance imaging (MRI) was normal in 56% of the patients while it revealed either a localized change in signal, or widespread and non specific (but lateralized) structural changes (i.e., ventricular horn asymmetry) in 44%. The most frequent finding was a localized increase in signal on T2 weighted images (12 patients). The parametric images of the 18FDGPET studies were reviewed independently by two of us (B.S. and P.M.) for the symmetry of glucose metabolism between the 2 sides of the brain. The PET studies were rated abnormal in 82% of the patients whose PET images revealed a more or less widespread zone of hypometabolism on one side of the brain relative to the other. Out of the 57 patients, 37 underwent chronic IcEEG recordings, weeks to months after the PET study; the other 20 patients either will have an IcEEG in the near future, or, for various reasons, were excluded from the procedure. The decision was never based on the PET result alone. Patients with IcEEG were sorted out according to the site of seizure onset during the IcEEG. The results are given in Table I. We therefore compared the location and the extent of the area of glucose hypometabolism to the electrophysiologic location of the
TABLE I IcEEG results
Patients were divided according to the site of seizure onset during the IcEEG. Several seizures were recorded in each patient. Localization of the seizure onset (IcEEG)
n
Unilateral
Frontal lobe Amygdalo-hippocampal region (focal) Amygdalo-hippocampal region and/or adjacent neocortex (widespread)
5 17 9
13.5 46.0 24.3
Bilateral
Frontal Amygdalo-hippocampal region
1
5
2.7 13.5
37
100.0
Total
%
124
7
Fig. 1. 30 year old w o m a n with CPE since the age o f 12. A regional onset of abnormal electric activity at the beginning of the seizures could not be determined on surface EEG. MRI was normal. The PET study with ]8FDG revealed a decrease in glucose metabolism in the lateral cortex and in the mesial part of the left temporal lobe (on each image, the left is on the left). In this patient, PET was the only non invasive technique to indicate a focalization. She later underwent IcEEG monitoring indicating that the seizure onsets involved the left amygdalo-hippocampal complex and the adjacent neocortex.
seizure focus. In cases of frontal lobe epilepsy (n = 6), 3 patients had an area of glucose hypometabolism limited to the frontal lobe where the epileptic focus was later localized by IcEEG; in 2 patients, the hypometabolism involved the frontal and the temporal lobe as well, and it was not possible to determine on the PET images which lobe was the most involved; and finally, PET was normal in a patient in whom the ictal activity seemed to start in both frontal lobes. All patients with a unilateral temporal lobe seizure focus (n = 26) had an area of glucose hypometabolism, either on the same side of the brain as the epileptic focus (25 patients), or on the opposite side (1 patient). Two typical examples are shown in Fig. 1 and in Fig. 2. The hypometabolism was usually maximal in the temporal lobe but it is noteworthy that, in most patients, the hypometabolism was widespread, involving not only all the temporal cortex, but also adjacent cortical areas (i.e., frontal and parietal lobe), although to a lesser degree. In some patients, the hypometabolism also involved the homolateral basal ganglia and thalamus and the contralateral cerebellar lobe. For example, the homolateral thalamus appeared relatively hypome-
Fig. 2. 20 year old woman with a story of febrile convulsions at age 2. Since then, she experienced typical CPS. At the time of her presurgical evaluation, she had as m a n y as 3 seizures per day despite appropriate medical treatment. MRI disclosed an area of increased signal intensity on T2 weighted images at the level of the left hippocampus (left). On the tSFDG-PET images (right), a vast area of glucose hypometabolism was observed, involving all the left temporal lobe; note that the homolateral thalamus is also hypometabolic as compared to the right thalamus, icEEG monitoring indicated that her seizures originated from the left amygdalo-hippocampal complex.
125
TABLE II Glucose metabolic rates in selected regions of interest
For the epileptic patients, the percentage of asymmetry between homologous regions of the 2 hemispheres, calculated as [(FS - NFS)/NFS] x 100, is also indicated. The CMRGIu values presented for the controls were used as reference values. The regional 95% prediction intervals in controls are also indicated. CMRGlu (mg/100 g/min)
Anterior temporal cortex Mid temporal cortex Mesiotemporal area Insula Orbito-frontal cortex Frontal cortex Parietal cortex Sensorimotor cortex Thalamus Cerebellum
Controls (n = 17)
Patients (n = 31)
CMRGIu Prediction Interval (Mean) (L + R)/2
Focus side CMRGlu
5.27 5.93 4.27 5.49 6.30 5.91 5.29 5.40 6.38 5.09
3.19 4.30 2.78 3.87 4.12 4.00 3.74 3.68 4.63 2.58
6.98 9.38 6.00 8.49 8.62 9.10 8.09 8.28 9.76 7.54
Non focus side CMRGlu
Asymmetry
Mean
_+ SD
Mean
_+ SD
Mean (%) _+ SD
3.81ab 4.68ab 3.72ab 4.62~b 4.91ab
(_+1.07) (_+ 1.17) (_+0.92) (_+ 1.33) (_+ 1.42) (+ 1.41) (_+ 1.31) (+ 1.37) (_+ 1.33) (+ 1.43)
4.82 5.55 4.09 5.15 5.48a 5.31~ 5.07 5.20 5.83 4.86
(_+ 1.22) (_+ 1.23) (_+0.92) (_+ 1.36) (_ 1.43) (_+ 1.44) (_+ 1.37) (_+ 1.50) (___1.49) (_+ 1.43)
-21.5 ac - 15.7~c - 9.6a¢ - 9.7a¢ - 10.8ac - 3.4ac - 7.2a¢ - 4.1~c - 4.9ac 3.8ac
5 . 1 0 ab
4.70ab 4.98b 5 . 5 1 ab 5 . 0 4 ab
(+11.1) (_+ 10.0) (+ 8.4) (_+ 11.5) (_+ 7.8) (_+ 7.2) (_+ 11.4) (_+ 10.6) (+ 7.1) (_+ 5.5)
a Significantly different from controls (t-test). b Significantly different from the non focus side (paired Student's t-test). ¢ Significantly different from zero. The level of significance is always P ~< 0.05.
t a b o l i c in 7 5 % o f t h e p a t i e n t s w i t h a cortical h y p o metabolism. The quantitative data ( C M R G I u values) obtained in selected r e g i o n s o f i n t e r e s t f r o m p a t i e n t s w i t h a u n i l a t e r a l t e m p o r a l l o b e epileptic focus are g i v e n in T a b l e II. I n each s u b t e n t o r i a l r e g i o n , g l u c o s e m e t a b o l i s m was s i g n i f i c a n t l y l o w e r o n the side o f the epileptic focus t h a n o n the c o n t r a l a t e r a l ( n o r m a l ) side, a n d t h a n the n o r m a t i v e v a l u e s o b t a i n e d in c o n t r o l s . I n 6 8 % o f the p a t i e n t s , the greatest (or m a x i m a l ) p e r c e n t a g e o f a s y m m e t r y was b e t w e e n - 15 a n d - 3 5 % (Fig. 3). O n average, the h y p o m e t a b o l i s m was d e e p e r in the l a t e r a l c o r t e x o f the temporal lobe (-21.5% a s y m m e t r y ) t h a n i n the m e s i o t e m p o r a l a r e a ( - 9 . 6 % ) w h e r e the epileptic focus was located. T h e a v e r a g e C M R G l u v a l u e s in epileptics, e v e n o n t h e focus side, were well w i t h i n the 9 5 % c o n f i d e n c e i n t e r v a l d e f i n e d in c o n t r o l s ; m o r e o v e r , l o o k i n g at i n d i v i d u a l C M R G I u values, it was e x c e p t i o n a l w h e n it fell o u t s i d e this interval. It was n o t p o s s i b l e to s e p a r a t e p a t i e n t s w i t h a m e s i o t e m p o r a l seizure o n s e t f r o m those w i t h a m o r e w i d e s p r e a d ictal o n s e t o n the basis o f the v i s u a l
i n s p e c t i o n o f the P E T images; similarly, we c o u l d n o t find a q u a n t i t a t i v e i n d e x o r p a r a m e t e r s e p a r a t i n g t h e s e 2 c a t e g o r i e s o f p a t i e n t s ( T a b l e III). W e also a v e r a g e d the C M R G I u v a l u e s o f all the cortical areas, o n the focus side a n d o n the n o n focus
12
t-t~
"= t~
8
O_
"6 E
-i Z
4
0
Fig. 3. Distribution of the patients according to the degree of hypometabolism in the most hypometabolic area (largest percentage of asymmetry).
126
Discussion
TABLE II1
Quantitative parameters calculated in controls and in two subgroups of epileptic patients defined by the IcEEG results Controls (n = 17)
Maximum asymmetry (% ± SD) YAsC (% _+ SD) £AsC/nROI (% ± SD) Ant and Mid Temp. Ctx (FS) (CMRGlu ± SD)
Epileptics IcEEG
- 9.5 ± 2.9 - 5. l _+48.6 - 0.2 _+ 2.3 5.75
Focal ( n - 14)
Widespread (n=9)
- 29.8* ± 13.1 -166.8" ± 75.2 8.1" ± 3.2 4.13" ± 0.75
- 27.4* _+ 10.5 -193.5" ± 49.7 8.9* ± 2.4 4.80* ± 0.6t
Focal: seizure onset limited to the amygdalohippocampal complex; widespread: seizure onset involving the amygdalohippocampal complex and/or the adjacent neocortex. 5~AsC: sum of all the regional cortical asymmetries; nRoi: number of regions interest. * Significantly different from the value calculated in controls (P < 0.05, Student's t-test).
side (epileptic patients), or on the left and on the right side (normal controls) (Table IV). Glucose metabolism on the focus side was significantly lower than on the non focus side, or than the values in the controls; on the non focus side, glucose metabolism was also lower than in controls, but without reaching the level of significance. Finally, in 5 patients, the seizures appeared to start in both hippocampi. The PET images were falsely 'lateralized' in 2 of these patients, normal in 2, while in an additional patient, a relative glucose hypometabolism was seen on both sides, depending on the brain level examined.
TABLE IV
Average glucose metabolic rate (mg/lO0 g/rain) calculated from all the cortical areas
CMRGIu ± SD
Epileptics (n = 31 )
Controls (n = 17)
FS
NFS
L
R
5.02* _+1.26
5.48 _+1.28
5.88 ±l.10
5.89 _+1.22
* Significantly different from the non focus side and from the values calculated in normal controls (P < 0.05).
Our results in patients with complex partial seizures agree with and extend those reported by other investigators. The originality of our work is related to the large number of patients in whom the epileptic focus was precisely localized by a combination of subdural and depth electrodes, and to the quantification of the PET studies. In 1980, Kuhl et al. 14 showed that focal depression of glucose metabolism can be detected by l SFDG-PET on the side of the epileptic focus in patients with therapy resistant complex partial epilepsy. These findings have been confirmed by a number of investigators 1,4,7,19,26,27 including us 9"1°'2°. PET is now widely considered to be an important step in the presurgical evaluation of patients with CPE. Surface EEG is associated with sampling errors due to the propagation of ictal discharges and PET has become an independent confirmatory test of functional deficit. With the availability of the new non invasive neuroimaging modal±ties (PET, MRI), the detection and the localization of the structurally or functionally lesioned area corresponding or related to the epileptic focus has become easier. Therefore, epilepsy surgery no longer remains a desperate therapy of last resort. If adequate medical treatment with major ant±epileptic drugs fails to control the seizures, the surgical option should be considered as soon as 3 years following the onset of the disease< Some authors <8 have suggested that, depending on the surgical technique, surface EEG and PET, when congruent, can be sufficient, in selected cases, for directly considering epilepsy surgery, therefore avoiding lcEEG. All our patients with a unilateral temporal lobe epileptic focus proven by IcEEG had an area of glucose hypometabolism. In all the published series, the vast majority of which are based on surface EEG, the rate of positive findings is also high ( > 70%). Our rate is even higher but a bias in population selection could explain this maximal sensitivity. Indeed, the hypometabolism itself was considered in the process of deciding to further explore our patients by IcEEG. With the group of 57 patients as a whole, some of whom will not have an IcEEG, the sensitivity is 82%, which is similar to
127
that found in other series 7'2°'27. The hypometabolism was always observed on the same side as the epileptic focus, except for one patient whose hypometabolic area was actually observed on the opposite side. He had 2 PET studies with the same results. He had a temporal lobectomy on the side of the epileptic focus defined by IcEEG and he remains seizure free at more than 6 months after the intervention. A false lateralization is exceptional 6 and PET can be considered an extremely reliable test as far as the lateralization of the epileptic focus is concerned. The PET findings cannot be considered in isolation and should always be correlated with the results of other tests (MRI, video monitoring, neuropsychological evaluation): indeed, PET can be normal despite the existence of a well localized epileptic focus or it can reveal a lateralized hypometabolism despite bilateral independent seizure loci. Like others L4, we have been impressed by the extent of the area of glucose metabolism which, in most patients, involved not only all the temporal cortex but also adjacent cortical areas. Clearly, the hypometabolic area is much larger than the epileptic focus. The most hypometabolic area is usually, but not always, found in the lobe corresponding to the epileptic focus. The pathophysiology of the regional glucose hypometabolism found in epileptic patients remains unknown. The presence of a decrease in glucose metabolism in homolateral basal ganglia or thalamus and contralateral cerebellar lobe (thus far remote from the epileptic focus) in some patients suggests that a 'de-activation' phenomenon could be involved (i.e., a loss of activating influences). This could be a valid explanation for why the hypometabolism is so vast. There is no consensus about the best way to analyze the PET data collected in epileptic patients. We consider that, for individual cases, visual interpretation is the best approach because, as with other imaging modalities, the patient serves as his or her own control, and the eyes integrate far more information than is provided by a few regions of interest. It is important however to quantify these studies, first to obtain objective confirmation of what is seen, and second, because it is the only way to extract information from the patients as a group. For
example, we show that even on the non focus side, cortical metabolism tends to be lower than in controis, a finding that could reflect the effect of antiepileptic drugs 25. We demonstrated that glucose metabolism is significantly lower in all cortical regions, except for the sensorimotor cortex, on the FS as compared to the NFS. This indicates that the spatially limited epileptic focus influences or even disturbs the function of the whole hemisphere. This might explain the frequency of the cognitive and psychiatric or affective disturbances observed in these patients. The fact that the degree of glucose hypometabolism is greater in the lateral cortex of the temporal lobe than in its mesial part, where the epileptic focus is located, deserves some comment. The amygdalo-hippocampal complex is a small structure which is therefore not optimally imaged by our current tomograph. Only a limited number of brain slices can be obtained with the NeuroEcat, and one can never be sure that the imaging plane passes right through that structure; moreover, the spatial resolution is limited. Therefore, both imaging and quantification are affected by the partial volume effect 17. It is expected that more accurate information will be obtained with the newest tomographs. These multislice cameras are characterized by a better spatial resolution and spatial sampling, allowing tridimensional reconstructions. These technological improvements should result in a better knowledge of the relationships between the epileptic focus on the one hand and the extent and the degree of glucose metabolism on the other hand. It might therefore be possible to better localize the epileptic focus non invasively by PET. Other new interesting information is also expected from the use of other neurochemical markers such as ligands for opiate 12'16 or benzodiazepine 11'22 receptors. The experience acquired with PET in epilepsy illustrates the fact that PET is not only useful for clinical purposes but it is also an exceptional tool to investigate in vivo physiological and biochemical processes, yielding unique information on the pathophysiology of brain diseases.
128
C h e m i s t r y s t a f f f o r the ~ 8 F D G p r o d u c t i o n s , J. H o -
Acknowledgements
diaumont
for e x c e l l e n t t e c h n i c a l a s s i s t a n c e d u r i n g
T h i s w o r k was s u p p o r t e d by t h e N a t i o n a l F o u n -
the P E T studies, C. D e g u e l d r e f o r the m a i n t e n a n c e
d a t i o n f o r Scientific R e s e a r c h a n d b y t h e Q u e e n
a n d the c a l i b r a t i o n o f the t o m o g r a p h , a n d f o r h e l p
Elisabeth Medical Foundation
a n d a d v i c e in d a t a analysis, D. L a m o t t e
(Belgium). T h e au-
t h o r s w o u l d like to t h a n k all the C y c l o t r o n a n d
and D.
C o m a r f o r their s u p p o r t .
References 1 Abou-Khalil, B.W., Siegel, G.J., Sackellares, J.C., Gilman,
2
3
4
5 6
7
8
S., Hichwa, R. and Marshall, R., Positron emission tomography studies of cerebral glucose metabolism in chronic partial epilepsy, Ann. Neurol., 22 (1987) 480-486. Debets, R.M.C., van Veelen, C.W.M., van Huffelen, A.V. and van Emde Boas, W., Presurgical evaluation of patients with intractable partial epilepsy: the Dutch epilepsy surgery program, Acta Neurol. Belg., 91 (1991) 125 140. Driver, M.B. and MacGillivray, B.B., Electroencephalography. In: J. Laidlaw and A. Richens (Eds.), A Text Book of Epilepsy, Churchill Livingstone, London, 1976, pp. 109-144. Engel, J. Jr., Brown, W.J., Kuhl, D.E., Phelps, M.E., Mazziotta, J.C. and Crandall, P.H., Pathological findings underlying focal temporal lobe hypometabolism in partial epilepsy, Ann. Neurol., 12 (1982) 518-528. Engel, J. Jr., Functional localization of epileptogenic lesions, Trends Neurosci., 6 (1983) 6ff65. Engel, J. Jr., Henry, T.R., Risinger, M.W., Mazziotta, J.C., Sutherling, W.W., Levesque, M.F. and Phelps, M.E., Presurgical evaluation for partial epilepsy: relative contribution of chronic depth-electrode recordings versus FDGPET and scalp-sphenoidal ictal EEG, Neurology, 40 (1990) 1670--1677. Engel, J. Jr., Kuhl, D.E., Phelps, M.E. and Crandall, P.H., Comparative localization of epileptic foci in partial epilepsy by PCT and EEG, Ann. Neurol., 12 (1982) 529-537. Engel, J. Jr., Sutherling, W.W., Cahan, L., Crandall, P.H., Kuhl, D.E. and Phelps, M.E., The role of positron emission tomography in the surgical therapy of epilepsy. In: R.J. Porter, R.H. Manson, A.A. Ward and M. Dam (Eds.), Ad-
12
13 14
15
16
17
18
vances in Epileptology." XVth Epilepsy International Symposium, Raven Press, New York, NY, 1984, pp. 427
432. 9 Franck, G., Sadzot, B., Salmon, E., Depresseux, J.C., Grisar, Th., Peters, J.M., Guillaume, M., Quaglia, L., Delfiore, G. and Lamotte, D., Regional cerebral blood flow and metabolic rates in human focal epilepsy and status epilepticus. In: A.V. Delgado-Escueta, A.A. Ward, D.M. Woodbury and R.J. Porter (Eds.), Advances in Neurology, Vol. 44: Basic mechanisms of the epilepsies, Raven Press, New York, NY, 1986, pp. 935-948. 10 Franck, G., Salmon, E., Sadzot, B. and Maquet, P., Epilepsy: the use of oxygen-15-1abeled gases, Semin. Neurol., 9 (1989) 307-316. 11 Frey, K.A., Henry, T.R., Holthoff, V.A., Koeppe, R.A., Sackellares, J.C., Brunberg, J.A., Berent, S., Gilman, S. and Kuhl, D.E., Comparison of regional glucose metabo-
19
20
21
lism and in vivo benzodiazepine receptor binding in refractory temporal lobe epilepsy, J. Nucl. Med., 32 (1991) 912. Frost, J.J., Mayberg, H.S., Fisher, R.S., Douglass, K.H., Dannals, R.F., Links, J.M., Wilson, A.A., Ravert, H.T., Rosenbaum, A.E., Snyder, S.H. and Wagner, H.N. Jr., Mu-opiate receptors measured by positron emission tomography are increased in temporal lobe epilepsy, Ann. Neurol., 23 (1988) 231 237. Guttman, I., Statistical Tolerance Regions, Griffin, London, 1970. Kuhl, D.E., Engel, J., Phelps, M.E. and Selin, C., Epileptic patterns of local cerebral metabolism and perfusion in humans determined by emission computed tomography of 18FDG and 13NH3, Ann. Neurol., 8 (1980) 348 360. Maquet, P., Dive, D., Salmon, E., Von Frenckel, R. and Franck, G., Reproducibility of cerebral glucose utilization measured by positron emission tomography and the [18F]2-fluoro-2-deoxy-d-glucose method in resting, healthy human subjects, Eur. J. Nucl. Med., 16 (1990) 267-273. Mayberg, H.S., Sadzot, B., Cidis Meltzer, C.C., Fisher, R.S., Lesser, R.P., Dannals, R.F., Lever, J.R., Wilson, A.A., Ravert, H.T., Wagner, H.N., Jr., Bryan, R.N., Cromwell, C.C. and Frost, J.J., Quantification ofmu and non-mu opiate receptors in temporal lobe epilepsy using positron emission tomography, Ann. Neurol., 30 (1991) 3-11. Mazziotta, J.C., Phelps, M.E., Plummer, D. and Kuhl, D.E., Quantitation in positron emission computed tomography: 5. Physical-anatomical effects, J. Comp. Assist. Tomogr., 5 (1981) 734-743. Phelps, M.E., Huang, S.C., Hoffman, E.J., Selin, C., Sokoloff, L. and Kuhl, D.E., Tomographic measurements of local cerebral glucose metabolic rate in humans with [FI8]-2fluoro-2-deoxy-D-glucose: validation of method, Ann. Neurol., 6 (1979) 371 388. Sackellares, J.C., Siegel, G.J., Abou-Khalil, B.W., Hood, T.W., Gilman, S., McKeever, P.E., Hichwa, R.D. and Hutchins, G.D., Differences between lateral and mesial temporal metabolism interictally in epilepsy of mesial temporal origin, Neurology, 40 (1990) 1420-1426. Sadzot, B., Salmon, E., Maquet, P., Debets, R., Grisar, Th., Lemaire, Ch., Degueldre, Ch. and Franck, G., Assessment of the interest of PET and MRI in complex partial epilepsy, J. Cereb. Blood Flow Metab., 7 (Suppl. 1) (1987) $432. Sammaritano, M., de Lotbini~re, A., Andermann, F., Olivier, A., Gloor, P. and Quesney, L.F., False lateralization by surface EEG of seizure onset in patients with temporal lobe epilepsy and gross focal cerebral lesion, Ann. Neurol., 21
129
(1987) 361-369. 22 Savic, I., Roland, P., Sedvall, G., Persson, A., Pauli, S. and Widen, L., In vivo demonstration of reduced benzodiazepine receptor binding in human epileptic foci, Lancet, 2 (1988) 863-866. 23 So, N., Gloor, P., Quesney, L.F., Jones-Gotman, M., Olivier, A. and Andermann, F., Depth electrode investigations in patients with bitemporal epileptiform abnormalities, Ann. Neurol., 25 (1989) 423-431. 24 Spencer, S., Williamson, P.D., Bridgers, S.L., Mattson, R.H., Cicehetti, D.V. and Spencer, D.D., Reliability and accuracy of localization by scalp ictal EEG, Neurology, 35 (1985) 1567-1575. 25 Theodore, W.H., Antiepileptic drugs and cerebral glucose metabolism, Epilepsia, 29 (Suppl. 2) (1988) $48 $55.
26 Theodore, W.H., Fishbein, D. and Dubinsky, R., Patterns of cerebral glucose metabolism in patients with partial seizures, Neurology, 38 (1988) 1201-1206. 27 Theodore, W.H., Newmark, M.E., Sato, S., Brooks, R., Patronas, N., De La Paz, R., Di Chiro, G., Kessler, R.M., Margolin, R., Manning, R.G., Channing, M. and Porter, R.J., [18F]Fluorodeoxyglucose positron emission tomography in refractory complex partial seizures, Ann. Neurol., 14 (1983) 429-437. 28 van Veelen, C.W.M., Debets, R.M.C., van Huffelen, A.C., van Emde Boas, W., Binnie, C.D., Storm van Leeuwen, W., Velis, D.N. and van Dieren, A., Combined use of subdural and intracerebral electrodes in preoperative evaluation of epilepsy, Neurosurgery, 26 (1990) 93-101.