EEG changes, recorded with subdural electrodes, after thiopental injections in patients with temporal lobe seizures

EEG changes, recorded with subdural electrodes, after thiopental injections in patients with temporal lobe seizures

J Epilepsy 1990;3:81-90 © 1990 Demos Publications EEG Changes, Recorded with Subdural Electrodes, After Thiopental Injections in Patients with Tempor...

761KB Sizes 0 Downloads 4 Views

J Epilepsy 1990;3:81-90 © 1990 Demos Publications

EEG Changes, Recorded with Subdural Electrodes, After Thiopental Injections in Patients with Temporal Lobe Seizures 'Roland Flink, 'Sigfrid Blom, 2Greta Engberg, 3Per Hilton-Brown, and 4Bo Sp/innare

Nineteen patients with temporal lobe epilepsy, examined with intracranial subdural electrodes for seizure monitoring, were subjected to intravenous thiopental injections to the level of deep narcosis. The induced EEG changes were recorded, analyzed, and correlated to the localization of the seizure-generating focus. In 14 patients with unilateral seizure onset, the thiopental injections evoked a focal spike activation on the focus side. The first suppression period in the EEG with a duration exceeding 1 s was recorded from the strip electrode overlying the focus. Only in six patients did asymmetries in the barbiturate-induced beta activity in the background EEG show lateralizing signs in concordance with the side of seizure onset. In five patients with bilateral seizure-generating foci, no significant lateralization in induced beta activity, focal spike activation, or focal suppression pattern could be found. The results are discussed, comparing similar studies performed with surface, sphenoidal, and depth electrodes. Key Words: Epilepsy surgery-- EEG--Subdural--Barbiturate--Thiopental.

In m a n y patients considered for surgery of epilepsy, it is necessary to p e r f o r m long-time EEG recordings and monitor interictal epileptiform activity and ictal onset with intracranial electrodes in order to localize the focus. After surgery, s o m e patients, notably those in w h o m no pathological changes are found in the r e m o v e d tissue, continue to have seizures. This might in part be due to false localization of seizure onsets (1). Therefore, any m e t h o d that provides additional information on the site of the seizuregenerating focus will be of value for the decision whether to operate. Various pharmacological tests have b e e n used in this context.

From the Departments of lClinical Neurophysiology, 2Anesthesiology, 3Neurology, and 4Neurosurgery, University Hospital, Uppsala, Sweden. Address correspondence and reprint requests to Dr. R. Flink at Department of Clinical Neurophysiology, University Hospital, S-751 85 Uppsala, Sweden.

Barbiturates can be used to induce or enhance EEG asymmetries in the beta frequency band in order to localize cortical lesions in surface recordings. This was first described by Pampiglione (2) in patients with large cerebral lesions involving one lobe or an entire hemisphere. Since then, EEG asymmetries induced by barbiturates have b e e n investigated in several studies (3-8). Similar effects after benzodiazepine injections have also been recognized (9-11). Activating effects on interictal epileptiform discharges have been described for short-acting barbiturates (methohexital and amobarbital) (12,13). This effect has also been used during acute electrocorticography in epilepsy surgery (14). The well-known EEG pattern of burst-suppression during d e e p barbiturate narcosis (15) is another EEG p a r a m e t e r that could be of interest to study. Our investigation was u n d e r t a k e n to decide whether the a b o v e - m e n t i o n e d EEG changes induced by intravenously administered thiopental (Pentothal) J EPILEPSY, VOL. 3, NO. 2, 1990

81

R. FLINK ET AL. a n d r e c o r d e d with chronically i m p l a n t e d s u b d u r a l electrodes m i g h t be of value to localize seizure-generating foci. The specific questions p o s e d were: (a) Is there also an a s y m m e t r y in fast activity i n d u c e d by barbiturates w h e n r e c o r d i n g with s u b d u r a l electrodes? (b) D o s u b d u r a l electrodes record spike activation like intracerebral electrodes? (c) Is there a difference in the i n d u c e d b u r s t - s u p p r e s s i o n pattern b e t w e e n the s u b t e m p o r a l region and t e m p o r a l neocortex?

Methods Patients N i n e t e e n patients (Table 1) with intractable epileptic seizures, referred to the Epilepsy G r o u p in Uppsala with the q u e s t i o n of operability, w e r e included in the study. The case histories as well as previous EEG r e c o r d i n g s with s p h e n o i d a l electrodes indicated a possible anterior t e m p o r a l focus. The patients w e r e subjected to a p r e o p e r a t i v e evaluation according to the protocol described earlier (16). Subdural strip electrodes w e r e implanted bilaterally, r e c o r d i n g from the t e m p o r a l lateral convexity and the s u b t e m p o r a l cortex, including the medial parts of

Table 1.

g y r u s parahippocampalis. The patients w e r e m o n i tored by m e a n s of a closed circuit television system, and their anticonvulsive d r u g s w e r e gradually withdrawn. The patients w e r e i n f o r m e d of the thiopental injection p r o c e d u r e , and t h e y gave their c o n s e n t to participate in the study.

Thiopental Injection The thiopental injection p r o c e d u r e was p e r f o r m e d within 2 weeks after the electrode implantation. The test was p e r f o r m e d in the a f t e r n o o n with the patient awake and alert. At least 24 h had passed since the last r e c o r d e d seizure. The EEG was r e c o r d e d on a 17c h a n n e l N i h o n K o d e n EEG m a c h i n e with an amplification of 3 0 - 5 0 B V / m m (time constant, 0.3; high freq u e n c y filter, 70 Hz) and a p a p e r s p e e d of 15 m m / s . The EEG signals w e r e also r e c o r d e d on an analog tape r e c o r d e r (Sangamo) for further f r e q u e n c y analysis. A 5-min EEG r e c o r d i n g was m a d e to obtain baseline conditions. The thiopental was then administered i n t r a v e n o u s l y in r e p e a t e d doses of 100 m g every 60 s until a burst-suppression pattern d e v e l o p e d in the EEG (Fig. 1) or a total dose of 1,000 m g had b e e n reached. The EEG r e c o r d i n g was c o n t i n u e d d u r i n g a

Summary of patient data

Patient no.

Interictal activity~

No. of seizures

Onset of seizures b

Thiopental dose

10 48 52 53 58 64 68 76 93 94 105 114 133 185 5 8

B (right)~ left) B (right = left) B (left > right) B (left > right) B (left)~ right) B (right > left) B (right) left) B (left ~ right) U (left) U (right) B (right)~ left) U (left) B (right ~ left) B (right ~ left) B (right > left) B (left ~ right)

2 5 1 8 4 3 10 7 1 5 5 5 4 2 13 11

ST (right) ST (right) ST (left) ST (left) ST (left) ST (right) ST (right) ST (left) ST (left) ST (right) ST (right) ST (left) ST (right) ST (right) 10 right/3 left 10 left/1 right

500 mg 700 mg 800 mg 700 mg 1,000 mg 1,000 mg 1,000 mg 800 mg 600 mg 500 mg 800 mg 400 mg 800 mg 700 mg 1,000 mg 1,000 mg

14 17 59

B (right ~ left) B (right ~ left) B (right)* left)

15 21 4

13 right/2 left 20 right/1 left 3 right/1 left

900 mg 1,000 mg 700 mg

Surgery/ pathology~ ATLR right/glios ATLR right/glios STA Failed ICAT ATLR left/glios ATLRright/glios ATLR right/glios ATLR left/glios ATLR left/glios ATLR right/glios ATLR right/glios Failed ICAT ATLR right/glios ATLR right/glios Failed ICAT ATLR left/glios + MN scar Planned STA Failed ICAT Planned ICAT

aB, bilateral; U, unilateral. bOnset, location of seizure onset; ST, subtemporal. CATLR,anterior temporal lobe resection; STA, stereotactic amygdalectomy; MN, meningocerebral; ICAT, intracarotid Amytal test. 82

J EPILEPSY, VOL. 3, NO. 2, 1990

THIOPENTAL-INDUCED EEG CHANGES

LST

RST

1-2 ~

~

LLTC 2 -3 ~,-,-"'~-'~

~¢e~-~-~-~-~

,~-"~

~

'

3-4 ~ 1-2

~

RLTC 2-3 3-4 "~'~'~-~'~'~

-

. ~

,

,

.

~

t

~

~-~-300JJV

Sodiumthiopental 700mg

ls

Figure 1. A generalized burst-suppression pattern recorded from all electrode surfaces after an accumulated dose of 700 mg thiopental. Abbreviations in this and subsequent figures: LST, left subtemporal; RST, right subtemporal; LLTC, left lateral temporal convexity; RLTC, right lateral temporal convexity. recovery period of 15-20 rain. The injections were performed under all necessary precautions by a trained neuroanesthetist.

EEG Analysis Drug-Induced Beta Activity The EEG was digitized with a sampling rate of 256 samples/s using an A/D converter (AD 100-64, Telefactor Corp.). A fast Fourier transform was computed for one lead from each recording subdural strip electrode (subtemporal bilateral and temporal convexity bilateral) on a Tandy 3000HL computer. The EEG was sampled for frequency analysis during 4 min, starting I min before the first injection and continuing during the first 3 min of the test. Sequential frequency power (~V2) spectra for periods of 4 s were calculated and plotted as compressed spectral array from 0 to 32 Hz using the Rhythm software programs (Stellate Systems, Inc.) comparing focus and nonfocus sides (Fig. 2).

Spike Activation The spike activity was quantified by counting all spikes with an amplitude exceeding the background amplitude with 100% and with a duration less than 100 ms. The background of the potentials was also observed, omitting high-amplitude beta rhythms. The spike activity was measured as number of spikes/ min. A baseline value was calculated from the 5-min recording preceding the first thiopental injection. The number of spikes/min was then calculated from the first injection to the first suppression period. Student's t test was used to compare focus versus nonfocus side and convexity versus subtemporal region, respectively.

Burst-Suppression Activity The time from the onset of the first injection to the first-appearing suppression period with a duration exceeding 1 s (Fig. 3), a parameter referred to as the "silent second" (17), was identified for each lead. A comparison was made between focus and nonfocus sides as well as subtemporal and convexity leads within the same temporal lobe. J EPILEPSY, VOL. 3, NO. 2, 1990

83

R, FLINK ET AL.

Figure 2. Compressed spectral array of frequency power spectra comparing the different regions and sides. The lowest line in the spectra represents time for first injection and the subsequent lines correspond to following 4-s periods. There is an asymmetry in evoked beta activity, most pronounced on the convexities, with a lack of fast activity on the focus side. Patient No. 94, unilateral seizure onset from the subtemporal region on the right side.

I

I

I

[

I

I

I

32 0

0 Subtemporal focus

I

l

32 0

Subtemporal non-focus

32 Hz

32 0 Convexity focus

Convexity non- focus

Results

tion en bloc including hippocampus. There was clear gliosis in the resected tissue in all cases operated on. A stereotactic lesion in the amygdala and anterior part of pes hippocampus was performed in one case. Two patients are awaiting further investigation, and the remaining four patients have been rejected for

In 14 cases, seizure monitoring with subdural electrodes showed a unilateral seizure onset in the subtemporal strip electrode. Five patients had bilateral seizure-generating foci. Twelve patients have so far been operated on with anterior temporal lobe resec-

f~JIIII1|11~11111111BI

1111 tl.L~JII|111"IILFTT~'~11111[TUTm~TU]I11111|L~1]In1~LIIIHIIIlUIIIIIIIIU

UTT|1~If~-UI~1~1~1~]~jLf~T~T~1~111~H~1~J~11~1~u

.~]]~

1-2

LST 2-3 3-4

1-2 RST 2-3 3-4 1-2

LLTC 2-3 3-4 1-2

RLTC 2"3 3-4 Sodiumthiopental

500mg

~

~

1300JJV ls

Figure 3. Example offocal suppression appearing in the right subtemporal lead (channels 4-6) after 4 rain and an accumulated dose of 500 mg thiopental. Patient No. 64, unilateral seizure onset from the subtemporal region on the right side. Abbreviations as in Fig. 1. 84 J EPILEPSY, VOL. 3, NO. 2, 1990

THIOPENTAL-INDUCED EEG CHANGES 1-2 LST

~ ~ # 4 ~ ~ W r ' h , ~ A ~ ,

2-3

i

~

v

J

~

~

~

1-2 RST

2-3

3-4 ~

'

,

~-~-

3-4

- - ~ . . . . a".-,.L-=u,..,#=l~,~u.l,a..a,e~kj,.J,=,,jat.~u~.,,,,a~ RLTC

2-3 3 - 4 ..... ~---

~=~* . . . . . . . . . .

r-

'r~ r'

$ / 30 sec after injection ot 100 mg sodiumthiopentel

1,50pV q

ls

Figure 4. Bilateral beta activation on the convexities 30 s aflerfirst injection of 100 mg thiopental. There is no side asymmetry. Patient No. 93, unilateral seizure onset from the subtemporal region on the left side. Abbreviations as in Fig. 1.

surgical resection due to memory problems revealed by the intracarotid Amytal test (Table 1). No patient experienced any inconvenience from the injections. Some patients required assistance to keep airways free during the recovery period of 10-15 min.

Drug-Induced Beta Activation In all cases, the analysis showed a beta frequency activation within 30 s after the first injection (Fig. 4). Three minutes after the first injection, the background activity was usually dominated by low-frequency components, indicating the start of the depression period soon changing to the burst-suppression pattern. The expected reduction in beta activity in the focus region relative to the nonfocus side was only found in 6 of the 14 unilateral cases (Fig. 2). No side difference in beta activation could be detected in another six patients (Fig. 5). In the remaining two patients there was a side difference, although with increased fast activity on the seizure-generating side. Four of the five patients with alternating bilateral seizure onset did not show any asymmetries. One patient (No. 8) had increased fast activity on the side where the majority of his seizures started.

Spike Activation In all 14 cases with unilateral seizure onset, the spike counting showed a significant (p < 0.01, Student's paired t test) increase in the number of epileptiform spikes on the focus side as compared with the nonfocus side (Table 2). The spike activation usually appeared within the first or second minute of the test, i.e., after an accumulated dose of 100-200 mg thiopental. In no case did the injection evoke any seizure activity. The spike frequency increased also on the nonfocus side as compared with baseline values in most cases, but this increase was lower than on the focus side. In the subtemporal leads, the frequency increased from a baseline value of 5 to 70 spikes/min on the focus side. The corresponding increase on the nonfocus side was from 1 to 19 spikes/min (for details see Table 2). When comparing all the leads, the spike activation on the focus side in these patients showed two different spatial patterns. In eight patients, the activation was restricted to the subtemporal leads and virtually no activation appeared in the lateral neocortex on the focus side (Fig. 6). In six cases, however, there was a simultaneous activation of spikes both in the subtemporal and neocortical leads J EPILEPSY, VOL. 3, NO. 2, 1990

85

K FLINK ET AL.

HD

Figure 5. Compressed spectral array of frequency power spectra in Patient No. 133. The patient had unilateral seizure onset from the subtemporal region on the right side, but the evoked beta activity showed no side asymmetry. First line in spectra corresponds to the time for first injection. The subsequent lines represent the following 4-s periods.

I

I

I

I

I

I

Subtemporal focus

32 0

32 0

32 0

0

Subtemporal non- focus

Convexity focus

on the focus side (Fig. 7). Three of the five patients with bilateral, i n d e p e n d e n t seizure-generating foci showed a simultaneous spike activation in the subtemporal recordings on both sides. The other two patients s h o w e d a clear a s y m m e t r y with increased spiking in one subtemporal region without any corresponding contralateral changes (Table 3). One of

Table 2. Spike activation for the different regions in patients with unilateral seizure-generating focus a Subtemporal leads

32 Hz Convexity non- focus

these two patients (No. 14) had, during EEG monitoring, 15 episodes of seizure activity, 13 with a rightsided onset and 2 with a left-sided onset. The latter were considered subclinical, and the interictal subtemporal spike activity dominated on the right side. The other patient (No. 59) had four seizures with subtemporal onset, three with a right-sided start, and one with a left-sided start, and the interictal epileptiform activity was p r e d o m i n a n t in the right temporal lobe.

Convexity leads

Patient no.

Focus

Nonfocus

Focus

Nonfocus

10 48 52 53 58 64 68 76 93 94 105 114 133 185

0/109 4/105 0/16 8/20 8/39 19/184 14/117 3/26 0/65 2/21 2/78 3/81 0/22 3/92

0/10 0/52 0/2 2/18 1/4 4/28 2/5 1/6 0/6 0/1 0/42 0/3 0/10 2/75

0/0 0/2 0/2 1/8 2/12 4/60 7/99 2/21 0/0 2/9 0/63 0/5 0/1 0/240

0/3 0/2 0/0 0/2 0/1 3/47 0/5 0/4 0/0 0/5 0/20 0/0 0/0 0/40

Mean +_1 SD

5/70 +6/+50

1/19 +1/+23

1/37 +2/+66

0/9 +1/+15

Suppression-Burst Onset

aNumber of spikes/min for the different regions on the focus and the nonfocus side; baseline value/maximal spike activation. 86 J EPILEPSY, VOL. 3, NO. 2, 1990

When the time was m e a s u r e d from the first thiopental injection to the a p p e a r a n c e of the first suppression period with a duration exceeding I s, a significant difference (p < 0.01, Student's paired t test) between focus and nonfocus sides was found. The EEG activity was first s u p p r e s s e d on the focus side (Fig. 3), indicating that the epileptogenic area was m o r e susceptible to the drug. The first suppression period appeared in the 14 unilateral cases after 4.5 min in the subtemporal lead on the focus side and after almost 6 min on the nonfocus side. C o r r e s p o n d i n g times for the convexity leads were 5 min on the focus side and 5.5 min on the nonfocus side (for details see Table 4). During the recovery period, the suppression disappeared at first on the nonfocus side but continued for a longer period on the seizure-generating side. In the cases with bilateral seizure-generating foci, neither the subtemporal nor the convexity leads s h o w e d any differences concerning the time for appearance of focal suppression (Table 4).

THIOPENTAL-INDUCED EEG CHANGES JllllllllUllllllll

UIIIItlllUIIIIIIIIUIIIIIIIIUIIIIIIIIUI|ilIIIIUIIII

It [IU

LJllilllll~lllliiilLIITll~n~JIIIIIt IIUIIIIIIIIIJIIliilIlulIIIIIIIUIIIIIIllUlIIIIHIU

UIIIIIHIUIIIlIIII~IIIIIIITUIIIIIIII

d IIIIIIII1~1

1-2 L S T 2-3

3-4 1-2 RST 2-3

3-4 1-2 LLTC 2-3 3-4

RkTC 2-3 3-4

Sodiumthiopental

200mg

1300,uV ls

Figure 6. Unilateral spike activation restricted to the subtemporal region (channels 4-5) on the right side, with no corresponding activation of spikes in the ipsilateral convexity, after an accumulated dose of 200 mg thiopental. Patient No. 64, unilateral seizure onset from the subtemporal region on the right side. Abbreviations as in Fig. 1.

Discussion It has been considered that the absence of beta activity is due to a loss of cells in the focus area. There are, however, diverging opinions in the literature concerning the correlation between absence of beta activation and medial temporal lobe sclerosis (3,8). We found a beta asymmetry corresponding to the seizure-generating focus in 6 of 14 patients (43%). This is very similar to the results in the study of Lieb et al. (18), who, with depth electrodes and computer evaluation of thiopental-induced beta activity, found asymmetries in 6 of 13 cases (46%). Engel et al. (6), using scalp and sphenoidal recordings and visual evaluation of EEG abnormalities, reported beta asymmetries with thiopental in 11 of 30 cases. Subdural electrodes, therefore, seem to be as sensitive as depth electrodes in this respect. The use of thiopental-induced sleep to evoke interictal epileptiform activity in temporal lobe epilepsies (19) focused the interest on the barbiturate effect on the EEG in focal epilepsies. The method was later used by Lombroso and Erba (5), who, in a study of 82 patients with generalized or focal epilepsy, concluded

that thiopental injections disclosed focal epileptiform discharges. Such activating effects have later been shown to exist for other short-acting barbiturates such as methohexital (Brietal) (12,13) and amobarbital (Amytal) (13). The use of methohexital for activation during acute electrocorticography in epilepsy surgery is well known (14). The reason for the excitatory effect of barbiturates on the central nervous system (CNS) has been discussed, and several explanations have been brought forward. The exact mechanism of the barbiturates is unclear. There is, however, evidence that the gamma-aminobutyric acid (GABA) receptor and its chloride-ion channel is only a part of a protein complex containing receptor sites for GABA, benzodiazepines, and picrotoxin/barbiturates (20, 21). The depressive or convulsive action of barbiturates may thus involve modulation of CNS inhibitory synaptic transmission at the level of this postsynaptic receptor-ionophore. Our study showed a significant correlation between the focal spike activation and seizure-generating side. Whether that is due to a direct excitation of the epileptic neurons or a release of inhibition exerted j EPILEPSY, VOL. 3, NO. 2, 1990

87

R. FLINK ET AL. I11 [ I ] l l ] , J I I I I H I l U l l l l l l l

I u II I I I I I I U

OiiiiilllLiitlOltltJIiiiiiiiUiitlllH

UtlllliltUlltlllllLiiil[llllUitlllltlUllllllllLI

U3 I I 1 1 1 1 1 U i i i i t

i ii Ulllt IIIIUIIIIIIII

UIHIDII~T~

1-2 LST

-w

3-4 1-2

RST

2-3

~

3-4

1-2

~!'i

LLTC

~

3-4 1-2

RLTC 2-3

" ~ ~ ~

~ ~ "

~.~ ~ , ~ ~ , t

3-4 Sodiumthiopental 300 mg

]300.uV ls

Figure 7. Unilateral spike activation both in subtemporal {channels 1-2) and convexity leads {channels 7-8) on the left side after an accumulated dose of 3 O0 mg thiopental. Patient No. 58, unilateral seizure onset in subtemporal region on the left side. Abbreviations as in Fig. 1.

f r o m t e m p o r a l n e o c o r t i c a l areas, a m e c h a n i s m sugg e s t e d b y C o c e a n i et al. (22), is still u n c l e a r . It s e e m s , h o w e v e r , to b e m o r e fruitful to u s e t h e d e g r e e of s p i k e a c t i v a t i o n t h a n t h e b e t a a s y m m e t r y as a s u p p l e m e n tary test in e s t a b l i s h i n g t h e s e i z u r e - g e n e r a t i n g side.

Table 3. Spike activation in the different regions in patients with bilateral seizure-generating focP Subtemporal leads Patient no.

Right

5 8 14 17 59

16/175 2/62 20/105 0/48 2/50

Mean +1 SD

8/88 -+9/+54

Left

Convexity leads Right

Left

1/123 1/85 1/1 2/30 0/2

2/54 0/16 2/2 2/10 2/3

0/85 0/25 0/0 5/15 0/0

1/48 +1/_+54

2/17 _+1/+21

1/25 +2/-+35

=Spikes/min for the different regions in right and left sides; baseline value/maximal spike activation. 88

J EPILEPSY, VOL. 3, NO. 2, 1990

This w a s also r e p o r t e d b y Engel et al. (1), w h o f o u n d the activation of interictal s p i k e d i s c h a r g e s m o r e s e n sitive t h a n t h e i n d u c e d fast b a c k g r o u n d activity w h e n

Table 4. Seizure onset (no. patients) Unilateral (14) Mean Focus Nonfocus +1 SD Focus Nonfocus Bilateral (5) Mean Right Left +1 SD Right Left

Time to first suppression period Subtemporal leads

Convexity leads

4 min, 27 s 5 min, 42 s

5 min, 10 s 5 min, 23 s

+ 2 min, 28 s -+2 min, 48 s

+ 2 min, 43 s +2 min, 29 s

4 min, 34 s 4 min, 24 s

5 min, 46 s 5 min, 54 s

+1 min, 25 s _+1 min, 46 s

_+2 min, 59 s + 2 min, 37 s

THIOPENTAL-INDUCED EEG CHANGES

using thiopental and depth electrodes. In this connection it is important to emphasize that the recording electrodes should cover mesial temporal spike-generating areas, since our results clearly show that the convexity leads are considerably more unreliable in this respect. The burst-suppression pattern in EEG was first described in deeply barbiturate-narcotized dogs (15). Electrocorticography during lobotomy surgery in humans showed that this pattern was confined to surgically isolated cerebral cortex and not seen in adjacent regions (23). The same pattern, although located in a generalized m a n n e r over both hemispheres, was also reported u n d e r deep anesthesia (24). We found that the suppression periods defined as the first "silent second" started subtemporally on the focus side in the cases with unilateral seizure onset, whereas their appearance on the temporal convexity did not differ significantly in time w h e n comparing focus and nonfocus sides. In patients with bilateral seizure onset, we could not find any side difference with respect to suppression periods. It has been shown in animal experiments (25) that the accumulated dose of barbiturate required to induce the suppression is correlated to the concentration of barbiturate in the brain tissue. An u n e v e n distribution of the thiopental in the brain with accumulation in the epileptic lesion could explain the early appearance of suppression. This is, however, less likely considering the focally reduced blood flow during interictal state in epileptic lesions (26). It might be that the epileptic lesion is more susceptible to barbiturates. Whether this is due to gliosis and neuronal loss or a changed barbiturate sensitivity mediated through the picrotoxin/barbiturate binding site of the GABA-receptor complex can only be speculated upon. Similar conclusions were drawn by Sperling et al. (27), who, with depth electrodes, recorded focal burst-suppression induced by thiopental in five patients with temporal lobe epilepsy. The area with focal burst-suppression correlated in their material to a focal lesion (gliosis or sclerosis). In conclusion, we think that the use of thiopental injections to induce EEG changes can be of value in predicting the seizure-generating side in temporal lobe epilepsy. Subdural electrodes are just as sensitive as depth electrodes in recording the asymmetries. This procedure will require a subtemporal location of the subdural electrodes, though, since the spike activation is less prominent on the convexity. For the suppression periods, the subtemporal location of the strips is crucial, since we could not show any clear side differences w h e n recording from the temporal convexities. The evoked asymmetric beta-frequency

background activity seemed less useful for the purpose of locating the seizure-generating side. Although this procedure will not replace the monitoring of spontaneous seizures, it will give additional information in the preoperative evaluation of patients with intractable epilepsy. Acknowledgment: This work was supported by the Margareth~ Foundation for Epilepsy and the Astrid Karlsson Donation Fund.

References 1. 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 1981;9:215-24. 2. Pampiglione G. Induced fast activity in the EEG as an aid in the location of cerebral lesions. Electroencephalogr Clin Neurophysiol 1952;1:79-82. 3. Kennedy WA, Hill D. The surgical prognostic significance of the electroencephalographic prediction of Ammon's horn sclerosis in epileptics. J Neurol Neurosurg Psychiatry 1958;21:24-30. 4. Brazier MAB. Prenarcotic doses of barbiturates as an aid in localizing diseased brain tissue. Anesthesiology 1969;31:78-83. 5. Lombroso CT, Erba G. Primary and secondary bilateral synchrony in epilepsy. Arch Neurol 1970;22:321-34. 6. Engel J Jr, Driver MV, Falconer MA. Electrophysiological correlates of pathology and surgical results in temporal lobe epilepsy. Brain 1975;98:129-56. 7. Duffy FH, Jensen F, Erba G, Burchfiel JL, Lombroso CT. Extraction of clinical information from electroencephalographic background activity: the combined use of brain electrical activity mapping and intravenous sodium thiopental. Ann Neurol 1984;15:22-30. 8. Lieb JP, Babb TL, Engel J Jr. Quantitative comparison of cell loss and thiopental-induced EEG changes in human epileptic hippocampus. Epilepsia 1989;30:14756. 9. Brazier MAB, Crandall PH, Walsh GO. Enhancement of EEG lateralizing signs in temporal lobe epilepsy: a trial of diazepam. Exp Neurol 1976;51:241-58. 10. Lieb J, Sclabassi R, Crandall P, Buchness R. Comparison of the action of diazepam and phenobarbital using EEG-derived power spectra obtained from temporal lobe epileptics. Neuropharmacology 1974;13:769-83. 11. Gotman J, Gloor P, Quesney LF, Olivier A. Correlations between EEG changes induced by diazepam and the localization of epileptic spikes and seizures. Electroencephalogr Clin Neurophysiol 1982;54:614-21. 12. Pampiglione G. Very short acting barbiturate (Methohexital) in the detection of cortical lesions. Electroencephalogr Clin Neurophysiol 1965;19:P314. 13. Aasly J, Blom S, Silfvenius H, Zetterlund B. Effects of amobarbital and methohexital on epileptic activity in mesial temporal structures in epileptic patients. An EEG study with depth electrodes. Acta Neurol Scand 1984;70:423-31. 14. Paul R, Harris R. A comparison of methohexitone and thiopentone in electrocorticography. J Neurol Neurosurg Psychiatry 1970;33:100-4. J EPILEPSY, VOL. 3, NO. 2, 1990

89

IL FLINK ET AL. 15. Swank RL, Watson CW. Effects of barbiturates and ether on spontaneous electrical activity of dog brain. J Neurophysiol 1949;12:137-60. 16. Blom S, Flink R, Hetta J, Hilton-Brown P, Hoffstedt C, Osterman PO, Sp/innare B. lnterictal and ictal activity recorded with subdural electrodes during preoperative evaluation for surgical treatment of epilepsy. J Epilepsy 1989;2:9-20. 17. Wahlstr6m G. Estimation of brain sensitivity of hexobarbitone (Enhexymal NFN) in rats by an EEG threshold. Acta Pharmacol Toxicol 1966;24:404-18. 18. Lieb JP, Sperling MR, Mendius JR, Skomer CE, Engel J Jr. Visual versus computer evaluation of thiopentalinduced EEG changes in temporal lobe epilepsy. Electroencephalogr Clin Neurophysiol 1986;63:395-407. 19. Fuster B, Gibbs EL, Gibbs FA. Pentothal sleep as an aid in the diagnosis and localization of seizure discharges of the psychomotor type. Dis Nerv Syst 1948;9:199202. 20. Olsen RW. GABA-benzodiazepine-barbiturate receptor interactions. J Neurochem 1981;37:1-13. 21. Olsen RW. Drug interactions at the GABA receptor-

90

J EPILEPSY, VOL. 3, NO. 2, 1990

22.

23. 24.

25. 26.

27.

ionophore complex. Ann Rev Pharmacol Toxico11982;22: 245-77. Coceani F, Libman I, Gloor P. The effect of intracarotid amobarbital injections upon experimentally induced epileptiform activity. Electroencephalogr Clin Neurophysiol 1966;20:542-58. Henry CE, Scoville WB. Suppression-burst activity from isolated cerebral cortex in man. Electroencephalogr Clin Neurophysiol 1952;4:1-22. Bickford RG, Faulconer A, Sem-Jacobsen CW, Petersen MC, Dodge HW, Schnugg FJ. Some effects of barbiturate anesthesia on the depth electrogram. Proc Mayo Clin 1953;28:162-75. Bolander HG. Evaluation of central nervous effects of some lipid-soluble barbiturates. Ume~, Sweden: Ume~ University Medical, Dissertations, 1983;No. 109. 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 18FDG and 13NH3. Ann Neurol 1980;8:348-60. Sperling MR, Jann Brown W, Crandall PH. Focal burstsuppression induced by thiopental. Electroencephalogr Clin Neurophysiol 1986;63:203-8.