вход по аккаунту


Electroencephalographic spiking activity drug levels and seizure occurence in epileptic patients.

код для вставкиСкачать
in Epileptic Patients
J. Gotman, PhD, and M. G. Marciani, MD'
We investigated the relationships among the electroencephalographic spiking rate, drug levels, and seizure occurrence
in 44 patients with focal epilepsy. Seizure occurrence was continuously monitored by personnel or videorecording and
spiking rate was quantified by an automatic detection method. Results indicate that drug levels do not influence
spiking rate, and spiking rate does not change before seizures but increases markedly after them, particularly secondarily generalized seizures. This increase can last several days and is observed during wakefulness and sleep. High or low
spiking rates do not influence the Occurrenceof seizures. We suggest that interictal spikes may passively reflect damage
to the brain, a damage which is worsened by further seizures. Spikes may not be directly related to seizure generation.
Gotman J, Marciani MG: Electroencephalographicspiking activity, drug levels, and seizure occurrence in
epileptic patients. Ann Neurol 17:597-603, 1985
The rate and spatial extent of interictal spiking can vary
considerably from one moment to the next or one day
to the next. Apart from the established fluctuations
brought about by various stages of sleep, the reasons
for these changes in interictal activity are not known.
The effect of anticonvulsant drugs on interictal spiking
is unclear: Ludwig and Ajmone-Marsan { 10) describe
activation of spiking after withdrawal of medication,
whereas other studies show no relationship between
interictal spiking and plasma levels of several anticonvulsants 116, 20, 21). The acute intravenous injection
of phenytoin is, however, followed by decreased interictal spiking [l 11.
The temporal relationship between changes in interictal spiking and seizure occurrence is also unclear.
Human studies found no obvious change before seizures 1161 and no relationship between spiking level
and the probability of seizure occurrence {16,20,21).
A tendency for interictal activity to become more synchronous over wide areas was observed during the
hour preceding a seizure 181. In the kindling model of
epilepsy, some authors have observed an increase in
spiking before the occurrence of spontaneous seizures
{14, 191. Long-term electroencephalographic (EEG)
monitoring and quantification techniques in kindled
cats 151showed no change in spiking rate before spontaneous seizures, but a long-lasting increase was observed in the days after seizures. In the alumina cream
model, a positive correlation between the rate of interictal spiking and the number of seizures was found
during evaluation of anticonvulsant drugs, whereas a
negative correlation was found after cerebellar stimulation 191.
The aims of this investigation were to clarify the
reasons for the variations in spiking rate, in particular
two factors, the effects of anticonvulsant drugs on interictal activity, and the temporal relationships between the variations in interictal activity and the occurrence of seizures. This study was performed on
epileptic patients who were candidates for surgical
treatment; medication was decreased to facilitate the
occurrence of seizures and presumably activate interictal activity. The methodology differed from that of
other studies because interictal spiking was quantified
almost continuously throughout the investigation.
From the Montreal Neurological Institute and Department of Neurology and Neurosurgery, McGill University, Montreal, Quebec,
"Present address: Clinica Neurologica 11, UniversitL di Roma,
Rome, Italy.
Received Aug 15, 1984, and in revised form Nov 20. Accepted for
publication Nov 24, 1984.
Materials and Methods
Forty-four patients with medically refractory epilepsy were
investigated for possible surgical therapy; ages ranged from
19 to 45 years. All but 2 suffered from temporal lobe
epilepsy with complex partial seizures, evolving in some
cases to secondarily generalized tonic-clonic seizures; 2 patients had widespread epileptic activity with predominance in
one hemisphere. Thirty-three patients were subsequently
operated upon and 7 probably will be. Four were not suitable
candidates for operation. Twenty-five patients were inves-
Address reprint requests to Dr Gotman, 3801 University St,
Montreal, Quebec, Canada H3A 2B4.
tigated with scalp and chronic sphenoidal electrodes; the remaining 19, having mainly a bitemporal epileptic disturbance
and sometimes different types of epileptic seizures, were
studied with multicontact bilateral and symmetrical chronic
intracerebral electrodes.
Monitoring and Spike Quantification
The study took place while patients were under intensive
EEG and video monitoring. The mean duration of this period was 7 days (range, 4 to 14 days) for the patients with
scalp and sphenoidal electrodes, and 1 3 days (range, 6 to 30
days) for the patients with intracerebral electrodes. Patients
were monitored approximately 7 5 to 80% of the day (usually
from 1500 to 1000 hours the next day); when the patients
with intracerebral electrodes were not being monitored by
telemetry, their EEG was continuously recorded in the
laboratory under supervision.
The EEG was transmitted from the patient"s room to the
laboratory through a 16-channel cable telemetry system. It
was continuously analyzed by computer, but the EEG was
recorded only when specific events occurred: seizures,
spikes, or scheduled samples. The interictal spikes were automatically recognized by a computer program that we have
been using routinely for several years and that has been
previously validated {6, 7). It allowed us to display the temporal and spatial distribution of the interictal activity during
any subset of a monitoring session. In patients with scalp
electrodes, quantification was performed only during sleeping hours (2300 to 700 hours) to avoid the frequent artifacts
found during wakefulness. In patients with intracerebral recordings, spike quantification was performed continuously,
as there was little artifact. Short sections of EEG were recorded on paper when spikes were detected; it was therefore
easy to assess whether false detections had taken place because of physiological transients or artifacts. To obtain
quantified values that did not include false detections, interactive editing was performed to eliminate them. Seizures
were recorded by an automatic seizure recognition program
[4] or in response to a push button activated when the patient felt a seizure coming or an observer witnessed a seizure.
The automatic seizure recognition program made frequent
false detections; they were simply rejected upon visual examination of the paper tracing. From direct observation or from
review of videotapes, clinical seizures were divided into partial seizures and secondarily generalized seizures.
Patients with intracerebral electrodes provided more complete information than patients with scalp electrodes. Because of longer monitoring sessions, generally closer observations, and the artifact-free nature of their EEG, they were
the initial subjects of this investigation and were studied in
greater detail. Results from patients with scalp electrodes
largely confirmed those obtained from intracerebral recordings, although they were obtained with slightly less certainty.
Pharmacological Assessment
The study was started while some patients were on full medication and continued during reduction of medication; in
other patients the telemetry was started after the medication
had already been reduced or stopped. In order to take these
differences into consideration, the patients were divided into
598 Annals of Neurology Vol 17 NO 6 June 1985
three groups: Group 1: in 21 patients (8 with scalp and 13
with depth electrodes) the study was performed throughout
the period of changing medication (mean duration of monitoring, 9 days); Group 2: in 12 patients (9 with scalp and 3
with depth electrodes) the study took place when the medication was at a low steady level (mean duration, 5 days); and
Group 3: the last 11 patients (8 with scalp and 3 with depth
electrodes) were without any medication (mean duration, 12
days). In the 33 patients of Groups 1 and 2 the drugs taken
at the beginning of the study were as follows. Fourteen patients were receiving a single drug: 9, carbamazepine; 2,
diphenylhydantoin; 1, valproic acid; 1, primidone; and 1,
phenobarbital. Nineteen patients were receiving two drugs:
15, carbamazepine and diphenylhydantoin; 2, diphenylhydantoin and primidone; 1, diphenylhydantoin and valproic
acid; and 1, diphenylhydantoin and phenobarbital. The withdrawal schedule was adapted to each patient. Plasma levels
were regularly determined in accordance with standard
Interictal Activity and Drug Withdrawal
Drug withdrawals were often followed by an increase
in seizure frequency. To avoid the possible confounding effect of seizure occurrence, the effect of the drug
reduction on spiking activity had to be observed in the
few patients in whom the withdrawal was not followed
by any seizure, and in patients who had seizures for
only a short interval and a long seizure-free period. In
7 of the 21 patients whose medication was progressively decreased, interictal activity was recorded for 5
to 18 days before seizures occurred; at that time anticonvulsant drugs were stopped or reduced to a low
steady level. In all cases, the rate of spiking activity did
not change and was unaffected by drug withdrawal.
Figure 1A illustrates a patient with depth electrodes in
whom carbamazepine and diphenylhydantoin dosages
were gradually reduced and withdrawn. No seizure
took place during the 21 days of the telemetry; the
interictal activity in the right temporal region remained
at a low but constant level.
Although not directly related to medication withdrawal, in patients receiving no medication spiking activity also remained stable when no seizure took place.
This could be observed in 4 of the 11 patients without
medication during the monitoring session as they had a
relatively long seizure-free period (6 to 7 days).
Interictal Activity and Seizure Occurrence
To eliminate the influence of drug withdrawal on relationships between interictal spikes and seizures, we
examined monitoring periods with stable medication.
In a few patients in whom the medication was not
stable, it could still be appreciated that the interictal
activity did not vary as a consequence of variations in
medication. No changes in spiking rate were observed
in the hours or days preceding either partial or secon-
Rlght temporal
Lett temporal
Lett temporal
Rlght temporal
: &
, ,I,
16118) i2N71
Generalized seizure
Partial seizure
684 --
Righl lemporal
Lett temporal
, ,
, , ,
Fig 1. Variations in interictal electroencephalographicspiking.
(A) Patient with intracerebral electrodes in whom carbamazepine
(CBZ) and diphenylhydantoin (DPH) were gradually withdrawn. No seizure took pbce during the telemetry session. The
abscissa is days of the month. The upper graph shows oral dosage
(lines) and plasma levels of anticonvulsant medication (dots).
Lower graph shows average number of spikes per minute during
the 21 days of telemetvy. Each point represents the average rate
during 7 hours of sleep. In the absence of seizures the spiking
rate remained stable despite the withdrawal of medication. (B)
Patient with intracerebral electrodes. Allseizures had a right
temporal onset. Spiking data are given for right and ldt temporal regions during 7 hours of sleep. One group of seizures followed withdrawal ofprimidone (PRM) and a second onefollowed withdrawal of PRM and DPH. lnterictal spiking
appears to have increased afrer seizures and not in accordance
with drug levels: compare days 15, 18, and 22. O n the last 3
days of monitoring, spiking activity decreased when seizures
stopped occurring with no change in medication. Seizures occuwed when spiking was high or low; they were not preceded by
a change in spiking rate. Changes in right and left temporal regions are parallel. (C) Patient with intracerebral electrodes.
Drugs were withdrawn and plasma levels were 0 on day 1 1 . All
seizures had a right temporal onset. Day data were obtained between 1800 and 2200 hours and night data between 0 and
600 hours. There was one partialseizure on day I 1 anda large
cluster on days 14 and IS (numbers in parentheses indicate
number of seizures);there were also many electrographically recorded seizures (21 on day I I and 7 on day 16) without clinical
manifestations. When the seizures stopped, spiking decreased and
remainedstable during several days. Spiking during waking and
sleeping hours followed similar patterns. (D) Patient with scalp
electrodes in the absence of medication. There were wide juctuations in spiking rate in the right temporal Fegion, where the secondarily generalized seizures started, whereas spiking in the contralateral temporal lobe remained stable. Increases in spiking
followed the occurrence of seizures; it could be argued that there
was a decrease in spiking before seizures but this is not confirmed
Sy the last decrease (night 9-1 O),following which there was no
generalized seizure for at least 2 days.
darily generalized seizures. Figure 2 illustrates typical
examples with no change in interictal activity in the
hours or even the minutes preceding seizures. Figure
1B illustrates that there was no change in interictal
activity in the days preceding seizures; it also illustrates
that seizures could occur when the spiking activity was
low as well as when it was considerably higher. In none
of our patients was there an influence of the level of
spilung activity on the probability of seizure occurrence. Seizures were, however, frequently followed by
a clear increase in interictal activity. To illustrate these
results, we separated the patients into those having
partial seizures only, those having generalized seizures
only, and those having both types of seizures.
the 44 patients ( 5 of 25 with surface electrodes and 10
of 19 with intracerebral electrodes) had only partial
seizures during the monitoring period (events in which
electrographic seizure activity was not accompanied by
clinical manifestations were not tabulated as partial seizures). In 13 of these 15 patients the withdrawal was
Gotman and Marciani: EEG Spikes, Drugs, and Seizures
16 18 20 22 0
8 10 12 14 16 18 20 22 0
10 12 14 16 18 20 22 0
followed by a relatively long-lasting increase in partial
seizure frequency. In 2 patients the withdrawal had no
effect on seizure frequency.
Results are summarized in the Table. In 8 of the 10
patients with depth electrodes who had partial seizures
only, the interictal activity was enhanced after most
seizures. Figure 2A illustrates a case of increase in
spiking activity immediately after a partial seizure; Figure 2B illustrates a case in which the first partial seizure was followed by an increase in spiking rate; a
subsequent partial seizure was not followed by any
change. Figure 1C illustrates a case in which the patient
was receiving no medication and in whom spiking activity increased moderately after a cluster of partial
seizures and then decreased and remained stable during the following days. In 2 of the 10 patients with
depth electrodes who had partial seizures only, there
was no change in interictal activity after seizures. In the
5 patients with scalp electrodes who had partial seizures only, 2 had an increase in spiking activity after
seizures; in 2 no change was observed, and in 1 we
observed a decrease not related to the time of seizure
9 11 13 1 5 17 19 21 23
Five of the 44 patients had only secondarily generalized seizures during the period of reduced medication;
seizures either occurred in clusters or were isolated.
Interictal activity sharply increased in the hours and
days after a generalized seizure (Table); an example is
I 10 spikeslminule
4u.AAdAd 4 l w 4 d h k
13 15 17 19 21 23
Fig 2. Temporal distribution of spike rate, minute Sy minute,
before and after seizures in patients with intracerebral electrodes.
(A)Patient showing very rare spikes during wakefulness and cyclic activation at night, most likely during light slow-wave sleep;
a partial seizure occuwed at 14 hours and was followed by a
clear increase in spiking (the patient remained awake and
showed no postictal clinical signs). (B) Patient having high
spiking rate day and night, with possible cyclic decreases during
sleeping hours. The first partial seizure was followed by an increase in spiking lasting several hours. The second partial seizure
was not followed 6y any change. Neither seizure was preceded by
an increase or decrease in spiking. (C) Secondarily generalized
seizure follwed I.y an increase in spiking extending over waking
and sleeping hours; there was no change before the seizure.
given in Figure 2C. Figure 1D shows successive increases and decreases in a patient with three isolated
seizures; in this patient changes were apparent only in
the region of seizure onset, and not on the contralateral side. This was not always the case, as increases
were most often widespread. In Figure ID, it could be
argued that there was actually a decrease before seizures rather than an increase after seizures; this is not
confirmed, however, by the last decrease, which was
not followed by a generalized seizure for 2 days (medication was resumed after these 2 days).
PATIENTS WITH PARTIAL AND GENERALIZED SEIZURES. The most frequent effect of drug reduction
or discontinuation was to enhance the rate of both
lnterictal Activity Changes afer Seizure Occurrence (44 Patients)
after Seizures
Seizures Only
( 5 Patients)
Seizures Only
(15 Patients)
Both Types
of Seizures
(20 Patients)
No Seizures
(4 Patients)
Scalp (25 patients)
(19 patients)
Not related to seizures
Not related to seizures
600 Annals of Neurology
Vol 17 No 6 June 1985
partial and secondarily generalized seizures. Eight of
the 19 patients with depth electrodes had both partial
and generalized seizures (Table). In 5 of these, the
interictal activity increased after both types of seizures;
in 3, it increased only after generalized seizures. Figure
1B illustrates a case in which the increase was more
marked after a group of partial seizures and a generalized seizure than after a group of partial seizures only;
the changes in spiking were not directly related to
changes in medication, since in the last 2 days of monitoring after the last seizure, spiking decreased despite
the absence of medication. This example also illustrates a frequent finding: the level of spiking activity
did not influence the probability of seizure occurrence,
as seizures could occur when the level of spiking was
high or low. Twelve patients with scalp electrodes had
both partial and generalized seizures. In 8, the spiking
rate increased only after generalized attacks, and was
unaffected by partial seizures. In 4 patients the changes
in interictal activity could not be correlated to seizure
occurrence or to medication levels.
An increase in spike activity after partial seizures
was never observed in a patient in whom the interictal
activity was unaffected by generalized seizures. In patients with depth electrodes, because of longer monitoring sessions, it was possible to define approximately
the temporal profile of the increase in spiking after
generalized attacks (partial seizures could not be considered because they rarely occurred in isolation). The
maximum spiking activity was reached 24 to 48 hours
after the seizure; in 2 patients it took as long as 3 to 5
days. The return to baseline spike rate took 1 to 4 days.
Interictal Activity Changes and Sleep
In the majority of patients interictal activity was increased by slow-wave sleep (Fig 2A, C). In some the
interictal activity did not clearly change during sleep
(Fig 2B). In scalp recordings the interictal activity was
only measured during sleep, since recordings during
wakefulness were too contaminated with artifacts. In
depth recordings we could measure spiking activity
during wakefulness and sleep. Changes in interictal activity were parallel in both states. Fig 2C shows how
the increase after a generalized seizure took place during wakefulness and sleep. Fig 1C illustrates this situation for partial seizures. The observation that the level
of spiking activity did not appear to influence the OCcurrence of seizures was further supported by the fact
that seizures rarely took place when the spiking rate
was clearly the highest, during slow wave sleep (partial
seizures could be recorded during sleep because of the
automatic seizure detection program).
Results indicate that variations in intensity of EEG
spiking may not be random and unpredictable. These
variations do not appear, however, to be related to the
two factors one could expect to influence spiking rate:
lowering plasma levels of anticonvulsant medication
did not cause any increase in interictal activity, and
seizures were not preceded by changes in interictal
spiking rate. The only factor that caused a change in
this rate was the occurrence of a seizure: it caused an
increase in spiking activity that could last several days,
outlasting the clinical postictal period.
The rate of interictal spiking was quantified by a
computer method [G, 71; this has the advantages of
consistently using the same criteria and allowing many
hours of EEG to be quantified every day. Because
these recordings included several seizures as well as
long periods before and after them, we could identify
relationships between three intimately linked phenomena: drug levels, seizures, and spike rates. Without
such a procedure it has been impossible in the past to
study systematically the temporal relationship between
spikes and seizure occurrence.
It is often assumed that spikes and seizures share
common pathophysiological mechanisms: indeed in
the most commonly studied models of focal epilepsy,
topical penicillin application and in vitro hippocampal
slices, the characteristics of the “epileptic” neuron are
determined at the time of interictal spikes 1131. In
these models, seizures start with a spike and the transition between spike and seizure has also been investigated f13}. In epileptic patients the coupling between
spikes and seizures is not as clear: there is rarely a
perfect spatial correspondence between the “spike
focus” and the “seizure focus”; seizures rarely start with
a spike, but rather with a desynchronization of the
EEG f31. The results of the present study support the
view that spikes and seizures are not so closely related
in patients. Four main conclusions can be drawn from
our results:
1. Interictal spiking did not increase in the hours or
days before seizure occurrence. Previous studies have
not found a clear change in interictal activity before
seizures. In the penicillin model of focal epilepsy, several changes in the interictal activity have been observed: abrupt changes in spike frequency 111, change
in spike morphology Il51, or complex changes variable
from animal to animal but consistent for the same
animal [18]. In the kindling model, an increase in interictal spiking was observed in the days preceding
frequent spontaneous seizures [14, 191; because these
studies used only short daily EEG recordings, investigators could not determine with precision the relationship of increased spiking to seizure occurrence.
Using continuous monitoring and spike quantification,
no change was found before spontaneous seizures in
kindled cats [5}. Very few studies have approached
this problem in humans: no change was found in a few
Gotman and Marciani: EEG Spikes, Drugs, and Seizures 601
patients who happened to have a seizure during a 12hour monitoring session 1161; a moderate decrease in
focal spiking activity, as well as an increase in coupling
between various regions, was observed 20 minutes before seizures @]. The present study and others reported in the literature support the view that, if pathophysiological changes take place in the days, hours, or
minutes preceding a seizure, they do not, or only marginally, affect interictal activity.
2 . Seizures could occur when spiking activity was at
its lowest for a given patient, or when it was at its
highest; we also observed that the spiking rate during
slow-wave sleep could become much higher than during wakefulness without a corresponding increase in
seizure occurrence. Therefore our results indicate that
a high spiking rate did not increase the likelihood of
seizures and a low rate did not decrease it. This is in
agreement with results found with spontaneous seizures in kindled cats { S ] . Our finding does not support
the hypothesis that a high spiking rate reduces seizure
susceptibility { 2 } .
3. Spiking activity increased in the hours and days
after many seizures, particularly secondarily generalized seizures. Increases after partial seizures were observed in several recordings from depth electrodes,
but not in scalp recordings. Seizures were not different
in the two patient groups and the difference in spiking
could be caused by the different recording properties
of the electrodes. If one postulates that the increase in
spiking after partial seizures is limited to a small brain
volume, then it could be recorded by depth electrodes
close to that volume, but would be missed by scalp
electrodes, which can only record widespread fields;
the change after generalized seizures could be more
extensive and consequently would be apparent in both
types of recordings. Such a postictal increase, lasting
much longer than the clinically apparent postictal state,
has not been reported before in patients, except as a
casual observation {l6]. A similar increase was found
after triggered and spontaneous seizures in kindled
cats {5). It indicates that seizures have an effect on the
brain longer than is apparent from behavior.
4. Our results indicate that decreases in plasma
levels of anticonvulsant medication did not cause an
increase in interictal spiking. This contradicts one
study reporting interictal activation {lo] as well as the
experience of many clinicians who have observed this
activation. The contradiction can be explained as follows: the reduction in medication usually causes an
increase in seizure occurrence and this increase, in
turn, results in increased interictal spiking. When observing a group of patients first on high doses and later
on low doses of medication, there would be a tendency
602 Annals of Neurology
Vol 17 No 6 June 1985
to see an increase in spiking activity, as the patients
would be more likely to have had seizures in the hours
or days before the recording on low medication. There
is a contradiction more difficult to resolve between the
present results and the decrease in interictal spiking
seen after acute administration of anticonvulsants 1111.
Perhaps different modes of action operate for chronically and acutely administered drugs.
In conclusion, this study, as well as the one in kindled cats using a similar methodology [SJ, points to a
dissociation between spikes and seizures. It appears
that spikes are not “abortive” seizures and may originate in different generating mechanisms having different biochemical characteristics. The facts that seizures
cause an increase in interictal spiking and that spiking
activity has no influence on seizure occurrence lead us
to suggest that “interictal” spikes could indeed be
caused by seizures and therefore could really be
“postictal” spikes (the patictal state possibly lasting
forever). Some other evidence supports this unconventional suggestion in that interictal spikes were
found to run down over a period of several months
after complete medical control of seizures was
achieved 117). Our findings were observed over a period of days or weeks and their relationship with the
appearance of secondary epileptic foci in humans after
years of focal epilepsy 112) remains uncertain.
Supported by grant no. MA7204 of the Medical Research Council
of Canada. J. Gotman was a Scholar of the Medical Research Council. We appreciate the collaboration of Drs A. Olivier, F. Andermann, P. Gloor, and F. Quesney, who cared for the patients during
the study. We are thankful to P. Bergsma for technical assistance.
1. Angeleri F, Gianquinto S, Marchesi G F Temporal distribution
of interictal and ictal discharges from penicillin foci in cats. In
Petsche H, Brazier MAB (eds): Synchronization of EEG Activity in Epilepsies. New York, Springer, 1972, pp 221-234
2. Engel J, Ackerman RF: Interictal EEG spikes correlate with
decreased, rather than increased, epileptogenicity in amygdaloid
kindled rats. Brain Res 190:543-548, 1980
3. Gloor P: Contributions of electroencephalography and electrocorticography to the neurosurgical treatment of the epilepsies. In Purpura DP, Penry JK, Walter R D (eds): Neurosurgical
Management of the Epilepsies. Advances in Neurology, Vol 8.
New York, Raven, 1975, pp 59-105
4. Gotman J: Automatic recognition of epileptic seizures in the
EEG. Electroencephalogr Clin Neurophysiol 54:530-540,
5. Gotman J: Relationships between triggered seizures, spontaneous seizures and interictal spiking in the kindling model of
epilepsy. Exp Neurol 84:259-273, 1984
6. Gotman J, Gloor P, Schaul N: Comparison of traditional reading of the EEG and automatic recognition of interictal epileptic
activity. Electroencephalogr CLin Neurophysiol44:48-60, 1978
7. Gotman J, Ives JR, Gloor P: Automatic recognition of interictal
epileptic activity in prolonged EEG recordings. Elecrroencephalogr Clin Neurophysiol46:5 10-520, 1979
8. Lange HH, Lieb JP, Engel J Jr, Crandall PH: Temporo-spatial
patterns of pre-ictal spike activity in human temporal lobe
epilepsy. Electroencephalogr Clin Neurophysiol 56:543-555,
9. Lockard JS: A primate moded of clinical epilepsy: mechanisms
of action through quantification of therapeutic effects. In Lockard JS, Ward AA Jr (eds): Epilepqy: A Window to Brain Mechanisms. New York, Raven, 1980, pp 11-49
10. Ludwig LI, Ajmone-Marsan C: EEG changes after withdrawal of
medication in epileptic patients. Electroencephalogr Clin
Neurophysiol 39:173-181, 1975
11. Milligan N, Oxley J, Richens A: Acute effects of intravenous
phenytoin on the frequency of inter-ictal spikes in man. Br J
Clin Pharmacol 16:285-289, 1983
12. Morrell F: Secondary epileptogenesis in man. Arch Neurol (in
13. Pedley TA: The pathophysiology of focal epilepsy: neurophysiological considerations. Ann NeuroI 3:2-9, 1978
14. Pinel JPJ, Rovner LI: Electrode placement and kindling-induced
experimental epilepsy. Exp NeuroI 58:335-346, 1978
15. Ralston BL The mechanism of transition of interictal spiking
foci into ictal seizure discharges. Electroencephalogr Clin
Neurophysiol 10:2 17-232, 1958
16. Rodin EA, Rim CS, Rennick PM: The effects of carbamazepine
on patients with psychomotor epilepsy. Epilepsia 15:547-561,
17. Seino M, Kakegawa N: Therapeutic concentration of antiepileptic drugs and the severity of epileptogenesis. In Buser
PA, Cobb WA, Okuma T (ed): Kyoto Symposia (EEG Suppl
No 36). Amsterdam, Elsevier Biomedical, 1982, pp 438-446
18. Sherwin I: Interictal-ictal transition in the feline penicillin
epileptogenic focus. Electroencephalogr Clin Neurophysiol
45:525-534, 1978
19. Wada JA, Sat0 M, Corcoran ME: Persistent seizure susceptibility and recurrent spontaneous seizures in kindled cats. Epilepsia
15:465-478, 1974
20. Wilkus RJ, Dodrill CB, Troupin AS: Cabamazepine and the
electroencephalogram of epileptics: a double blind study in
comparison to phenytoin. Epileppia 19:283-291, 1978
2 1. Wilkus RJ, Green J R Electroencephalographic investigations
during evaluation of the antiepileptic agent Sulthiame. Epilepsia
15113-25, 1974
I n the paper “Decreased Axon Caliber and Neurofilaments
in Hereditary Motor and Sensory Neuropathy, Type I” by
rJukada and Dyck (16238-241, 1984),footnote a in Tables
I and 2 should read: “Not significantly different from control
group ( p > 0.05).”Footnotes b and c in Table 2 should read:
“Significantly different from control group: bO.OO1 < p <
0.005; ‘ p < 0.001.’’ We apologize for this error.
Gotman and Marciani: EEG Spikes, Drugs, and Seizures
Без категории
Размер файла
705 Кб
electroencephalographic, drug, level, patients, epileptic, spiking, activity, seizure, occurence
Пожаловаться на содержимое документа