Epilepsy Res., 5 (1990) 49-60

49

Elsevier EPIRES 00276

A progressive sequence of electroencephalographic changes during generalized convulsive status epilepticus* David. M. Treiman a'b, Nancy Y. Walton a'b and Carol Kendrick a aNeurology and Research Services, VA West Los Angeles Medical Center, and bDepartmentof Neurology, UCLA School of Medicine, Los Angeles, CA (U.S.A.) (Received 5 March 1989; revision received 16 March 1989; accepted 22 March 1989)

Key words: Status epilepticus; EEG; Monitoring; Experimental models

Review of 60 electroencephalograms recorded during episodes of generalized convulsive status epilepticus suggested that there are 5 identifiable EEG patterns which occur in a predictable sequence during the course of generalized convulsive status epilepticus in man: (1) discrete seizures; (2) merging seizures with waxing and waning amplitude and frequency of EEG rhythms; (3) continuous ictal activity; (4) continuous ictal activity punctuated by low voltage 'flat periods'; and (5) periodic epileptiform discharges on a 'flat' background. We confirmed our hypothesis that this sequence represents the natural history of electroencephalographic changes in untreated generalized convulsive status epilepticus by observing the same sequence in the EEGs of rats in which status epilepticus had been induced by 3 different methods: (1) systemic administration of kainic acid, (2) injection of homocysteine thiolactone to cobalt-lesioned rats; and (3) injection of lithium chloride followed 24 h later by injection of pilocarpine.

INTRODUCTION The EEG patterns of discrete seizures which occur early in the course of generalized convulsive status epilepticus have been well described 12. Most commonly such seizures begin with fast lowvoltage spikes which gradually increase in amplitude and decrease in frequency. When the clonic phase of the seizure is reached the spikes become intermittent and are separated by slow waves. The * Presented in part at the 39th Annual Meeting of the American Academy of Neurology, New York, NY, April 198715.

Correspondence to: Dr. D.M. Treiman, Department of Neurology, Reed Neurological Research Center, UCLA School of Medicine, 710 Westwood Plaza, Los Angeles, CA 90024, U.S.A.

last clonic contraction is followed by low-voltage slow waves until the onset of the next discrete seizure. There have been no reports of the EEG patterns which may occur during prolonged episodes of generalized convulsive status epilepticus. We recorded 60 EEGs during episodes of status epilepticus as part of a comparison of lorazepam and phenytoin in the initial treatment of generalized convulsive status epilepticus in adults 13. A review of these records suggested to us that 5 identifiable EEG patterns occur in a predictable sequence during the course of generalized convulsive status epilepticus in man. This paper describes the 5 EEG patterns observed during generalized convulsive status epilepticus in humans and in 3 experimental models, and the sequence in which they occurred in the 3

0920-1211/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)

50 models. We suggest that this sequence of EEG changes represents the natural history of EEG evolution of untreated generalized convulsive status epilepticus.

METHODS

Human studies Subjects and diagnosis. EEGs were recorded in 60 males (ages 30-87 years) who were in generalized convulsive status epilepticus. In each case the EEG was recorded during the episode of status and was used to confirm the effectiveness of treatment. Generalized convulsive status epilepticus was defined as the occurrence of either: (1) 2 or more generalized convulsions without full recovery of consciousness between the seizures, or (2) bilateral ictal discharges on the EEG associated with marked depression of consciousness and at least some convulsive activity, lasting for at least 10 min and still present when treatment was initiated. The etiology of status epilepticus in the 60 cases is detailed in Table I. Fourteen of the patients TABLE I

Etiology of status epilepticus in 60 cases Etiology

No. of cases

%

Structural lesion Toxic-metabolic AED withdrawal CNS infection Ethanol withdrawal AED + ethanol withdrawal Multiple causes Other Unknown Insufficient data

5 9 5 5 2 3 14 1 5 11

8 15 8 8 3 5 23 2 8 18

Total

60

(23%) had a known history of epilepsy. The most common single causes of status were presence of a structural CNS lesion, toxic-metabolic encephalopathy, CNS infection and anti-epileptic drug withdrawal. Multiple causes accounted for 23% of the episodes of status. In 8%, an exact etiology could not be determined. EEG recording. All of the recordings were done using a standard montage consisting of 4 anterior-posterior bipolar electrode chains, 2 temporal and 2 parasagittal (AEEGS montage LB16.3) a. An Electrocap ® (ECI) and conducting gel were used for rapid electrode placement. All EEGs were recorded on a Grass Model 8 EEG machine, with either 16 or 18 channels. Recordings were done with the low-frequency filter set at 1 Hz and the high-frequency filter set at 70 Hz. EEGs were recorded as soon as possible after the status team was notified of the case. All of the ictal recordings illustrated in this paper were obtained prior to specific treatment of the ~pisode of status. In some cases the patient was already receiving anti-epileptic drugs on a chronic basis. EEG recordings were carried out until all electrical evidence of ictal activity had disappeared or until it was determined that the patient would not respond to any available treatment.

Rat studies Animals. A total of 20 adult male SpragueDawley rats were used in these experiments. Rats were housed individually in clear Plexiglas cages and provided with food and water ad libitum. A 24-h diurnal light cycle was maintained, with lights on from 07.00 to 19.00 h each day. Implantation of recording electrodes. Rats were anesthetized by intraperitoneal (i.p.) injection of 87 mg/kg ketamine plus 13 mg/kg xylazine. They were then mounted in a stereotaxic frame and the skull exposed. Epidural recording electrodes made from no. 0-80 x 1/8 in. stainless steel screws

Fig. 1. A: discrete generalized tonic-clonic seizures with interictal slowing, recorded prior to treatment in a 39-year-old man. Example shows ,:nd of clonic phase of the seizure and the appearance of post-ictal slowing. B: the top recording shows the end of a discrete seizure in rat no. 06, 26 min after i.p. injection of 10 mg/kg kainic acid. The middle recording shows the same phenomenon in rat no. 206, 30 min after i.p. injection of 5.5 mmoi/kg D,L-homocysteine thiolactone. The rat had had a cobalt lesion created in the left frontal cortex, as described in the text. The third recording illustrates the end of a discrete seizure in rat no. 325, 21 rain after i.p. injection of 25 mg/kg pilocarpine. The rat had received an i.p. injection of 3 mmol/kg LiCI, 24 h before pilocarpine was injected.

51

A

[~O0~V ,

T6-02'

:

II

I

Isec

~...._._/

'

k(' f~-It 1 l[ [ [ /'( if! f

---~,-j

-~.-.,.,'~,~--...~ ~

~i

o, .'~! F,,,-,>,,"H

.

,-..__

,,,~

~i1"" . ,.~. ~Jr%r~'Pt~'d#'rIA"Ar"~~ '~"~

,~-

-

v

.

.

.

.

,~.

.

~

~

-

~

,

~

-

~

.

.

/

-

-

-

-

-

-

-

.,

i

FPI-F3~l~y~ P~s- Ot

F~" ~

- ~ , ' I

........

,,,,

,~,,

,

i~"1

,

h,~

~

""~"''"''""'~'---'''"

I IO0~V

,4. ,,,,, ,;,.I Nt,,'.,,,,i;.k',b,.;.~,b,,~!~.%t!',i,,'.~".ff.j', ~.; 'lt~' .~li , ,' ~

,,

,~,t ....

'~ ~ 1 , " ' , , , . . . .

, , , ,

.~-~,

,,,,,

~-u-,~

!i.'i, .~.1~,!,r, fftl~,,rf,q~,1',:lt,?t,,.rT, .i,; K,._ '~' ,,. ,~'~, ,.' ::,.,

,~.li ~l'~'v

"I' ,"

.~ •

. ''.~ "~1 l , , ~

,~,b~l.]', ' ,~,,~ I'

~,, ~ .I', :, I , ~ ,, ~,.,,-.f~,,

~, ~..

'

',' .v ".,, ~

. . . . . . . . . . .

,.

:~vV]I~IV' ~"' v~W':

v~~

'~Y ~ '~lf

|~I

l

~ ~.

:~

i

~

~

~

52 were placed at the following stereotaxic coordinates (referenced to bregma)" A - P +2.0 mm, lateral +3.0 mm, and A - P - 4 . 0 mm, lateral +3.0 mm. For animals being used to study the cobalt/ homocysteine model 17, powdered cobalt (25 rag) was placed onto the dura at the site of the left anterior screw prior to its being implanted. Electrodes were secured with dental acrylic, the skin sutured and the animals allowed to recover 24 h before being handled again. Status epilepticus was induced 5-14 days after implantation of the epidural electroOes. Recording of EEG. EEGs were recorded in freely moving, unrestrained rats. Rats were placed in the recording cage and connected to a Grass Model 6 or Model 8 EEG machine by a flexible cable suspended over the top of the cage. Seizure behavior was simultaneously recorded by video camera. All recordings were done with the low-frequency filter set at 1 Hz and the high-frequency filter set at 70 Hz. The 60 Hz notch filter was used. EEGs were recorded at a paper speed of 30 mm/sec. Induction of status. Experimental status epilepticus was induced in rats by 1 of 3 techniques: (1) Kainic acid model 4. Five to 10 days after implantation of screw electrodes 10 mg/kg kainic acid (Sigma) was injected i.p. Rats were then observed and EEG recorded. (2) Cobalt/homocysteine model 17. Status was induced by i.p. injection of 5.5 mmol/kg homocysteine thiolactone (Sigma) when motor seizures or very frequent epileptiform activity were seen arising from the cobalt focus. (3) Lithium/pilocarpine model TM. Five to 7 days after implantation of screw electrodes 3 mmol/kg LiCI (Sigma) was injected i.p. Twenty-four hours later 25 mg/kg pilocarpine was injected i.p. to induce status epilepticus. EEG and behavior were menitored continuously throughout the experiment.

RESULTS

Status epilepticus in humans Forty-six EEGs were recorded prior to specific treatment of status epilepticus, in 14 episodes treatment was begun before the EEG recording was started. Five ictal EEG patterns could be identified and are described in detail below. Table II provides the number of cases in which each of these patterns was observed as the initial EEG pattern recorded before treatment was started and also provides the number of cases in which a pattern was observed as the initial EEG pattern when the recording was started after treatment had been initiated.

Discrete seizures with interictal slowing (Fig. 1A). This pattern typically started with focal lowvoltage fast activity which increased in amplitude, spread across midline, and then slowed in the frequency of discharges. Because this pattern usually was associated with overt generalized, although frequently markedly asymmetric, tonic-clonic seizures, there was often a period of generalized musTABLE II Pattern seen when EEG recording started Pattern

No. of cases in which pattern was seen first on EEG EEG sta;'ted before treatment

Discrete seizures 10 Merging seizures 9 Continuous ictal activity 18 Continuous with fiat periods 5 Periodic epileptiform discharges 4 Total

46

Treatment started before EEG

2 1 6 3 2 14

Fig. 2. A: merging of discrete seizures, recorded prior to treatment in a 64-year-old man. Ictal discharges are continuous, but with waxing and waning of frequency and amplitude. An increase in frequency and amplitude can be seen beginning on the right side of the recording. Waxing and waning patterns are best appreciated when at least 4-6 EEG pages can be viewed simultaneously. B: the top recording illustrates waxing and waning of amplitude and frequency of ictal discharges in rat no. 06, 75 min after kainic acid injection. The middle recording illustrates the same phenomenon in rat no. 206, 37 min after homocysteine thiolactone injection. The lower recording illustrates the same phenomenon in rat no. 325, 24 min after pilocarpine injection.

53 A

~,-

F~

l,.~j,~,:.,: :',.:.

,

F7" T3

75,'vl T5- OI

'~,:'~,:"':~,

'""" ~I "~':.

i'": ,' ".""'.'" .',

I

"/

C3-P3 P3" OI

B

"

'

II ,"~

'.ll'i",

. . . . . .

,~

'

. . . . . .

' ' ' "

....

,'

"

i,,,

,k,,' ' ,'' , ,' ' ,~,¢ ';'.' ', ,~ ,~# ,;'-',,',' , .... a,..,: .... ,,, : v , . , , , , , , . . , , . , , , , . , , , ,

:t~i'~'..i ~t_~,~.

,,;4

.,,, ,~ ~ ~,~

~,

,.

¢* I~ :1~

P4-P3 °It~f,lil~ ',~1~

,itl,!~,(? I I:~,, ,, ~t~ ?/,~ '/I,~,,

I

;

,, ,,,

,'

•

,,,

50~,V I

I

.

~

s

li

IrlF

I

,1"

r

¢"

t

'~1

°

if

~1

. -

,,.,, ,,;,,.::~,~/~,s¢~.,,:,.#,~ ~.~,,j@.,~¢~~.jJ,,,~,./J,

I

~ ~

i

/

I

~

:

•

.~

:

' ....

. . . . . . .

HH,~..~,¢,,~,,:#, ::, :,,:. :...::.;:.. ,.,;...... .... ........

3oo~vI I SlIC

i

F3"P3 '~ :f(~

Ps-,4 I

I

l~

'-,

• Rt,

P4-F4

F4-F3 , .~1~ I

54 cle artifact associated with the tonic seizure followed by periodic muscle artifacts as the tonic activity converted to clonic jerks. With the end of the last clonic jerk, low-voltage slow activity appeared and continued until the development of the next discrete seizure.

Merging seizures with waxing and waning ictal discharges (Fig. 2A). In this pattern rhythmic but frequently asymmetric sharp wave or spike/wave patterns were seen with recurrent build-up and then slowing of frequency and waxing and waning of amplitude. There was no intervening low-voltage slow activity, as was seen in the discrete seizure pattern. The rhythmic build-up was sometimes associated with overt generalized seizures. More commonly focal intermittent tonic and/or clonic convulsive activity was seen. Continuous ictal discharges (Fig. 3A). This pattern consisted of rhythmic, relatively constant, sharp wave or spike/wave discharges which were frequently asymmetric, as seen in the figure. Such discharges were associated with either continuous generalized clonic jerks, or only subtle clonic movements. Two patients had only diffuse, continuous, rhythmic slowing during status. The continuous ictal slowing responded promptly to intravenous lorazepam, with normalization of the EEG and recovery of consciousness.

Continuous ictal discharges with 'flat periods' (Fig. 4,4). The continuous ictal discharges just described sometimes were punctuated by brief (0.5-8 see) episodes of generalized flattening on the EEG. Although the epileptiform discharges sometimes were asymmetric, the flat periods were always generalized. This EEG pattern could be associated wit~ overt ~,, subtle focal clonic move ments or no :uotor symptoms at all.

Periodic epileptiform discharges on a 'flat' background (Fig. 5A). In this pattern, bitaterai, sometimes asy::;metric, high-voltage, monomorphic, repetitive sharp waves wt~:~: superimposed on a

relatively flat background. This was the initial pattern in only 4 patients in whom the EEG was started before treatment. Duration of status was known in only 2 of these 4 patients and was more than 4 h in one and more than 19 h in the other. PEDs were also seen in some partially treated patients before status was completely controlled. These 5 patterns, and the order in which they occur, were delineated by observation of fragments of the proposed sequence in individual patients. Because of the obligation to treat human patients in status epilepticus as quickly as possible, we could not observe the entire sequence in any one patient. In order to verify that during generalized convulsive status epilepticus EEG changes occur in the order suggested by the human records, we induced status epilepticus in rats, using the 3 experimental models of status epilepticus described above.

Experimental status in the rat Kainic acid and lithium plus pilocarpine induced limbic seizure activity which, in its most severe form, involved bilateral forelimb clonus, rearing and falling. Initial EEG and behavioral seizure activity appeared 20-30 min after kainate injection, then progressively worsened in severity and frequency over a period of hours. Status epilepticus was rarely fatal in this model, but EEG spiking sometimes continued up to 48 h. Onset of EEG and behavioral seizure activity following lithium plus pilocarpine also occurred 20-30 min after injection, but the initial seizure activity was much more intense. Status epilepticus was almost always fatal within a few hours in this model. In contrast, administration of homocysteine thiolactone to cobalt-lesioned rats caused an activation of the epileptogenic focus and induced true generalized tonic-clonic seizures (GTCS). These GTCS began 20-30 min after injection of homocysteine and recurred every few minutes for 2-4 h.

Fig, 3. A: continuous ictal discharges recorded prior to treatment in a 68-year-old man. Examples are ! 6 min apart. Continuous ictal activity persisted 101 rain, stopping only after phenytoin infusion was completed and 4 min after the end of lorazepam infusion. B: the top recording illustrates continuous ictal discharges in rat no. 06, 103 min after kainic acid injection. The middle recording illustrates the same phenomenon in rat no. 206, 48 min after homocysteine injection. The bottom recording also exhibits continuous ictal discharges in rat no. 325, 28 min after pilocarpine injection.

A'

55

FPI- F7

'

'

' ~,~,~~~,~,.~~,~~,~

...... v , r ~ ~ , f ~ . ~ m f , ~ ~ ~ ~ , 14"P4

~h~"r~z~~__

,

j-"-,.¢~w"~f'w'.,

'~-°' r~ 9 " ~ ~ e ~ ' ~ % ' ~ ' ~ ,

~

w

~

,

v

J

,

. . . .

s"~

t~

"';" ~ ,"~' ,,..,,C ,' ,~ ,,,' • ' ,~' ~ ' x. r~,~ ,~' ,: ~'., ,', ,', '; .,p, .j;,, ~ .~.!~, ~r,~i ;~ v t~ ~I ~. ~ .~. dYk I '

,

•

l,

"

I

I'!i~¢

I

I

'

,

/-"-"

|

,

('-;'''

".

.

'""

, ..

,

~

,

.,

*

~

..

• ,~

,

',,

' ' ,',

,' ' ".

~1 ;,

t

,r ,".'.":"~1 ~, ~ .'-!,~ ~,,,:~ .~ ~.~ ~ ~ },~ ~.~ .,~, ~!

"-"

--~ . -

,

-

t

F3"P3

m , ~

~

,,~',,"I~,~:'i' ~i, ~:,~,..~;~l:,''it'd't~ll,l~l~:~(:~l;I ,I, i@i'

""'

-"

"/

~i'

' 1'

t,

~ .~.' ~~ ~I/j~ll I ,~ ~

~1, ~ ,'

II

)~l~ i

~;

'/t

'~

~J

'fi'

.~

,, ~'~

r ; i '~I

['l~

I'

~'~',,

I~ ' 1. ~ ' ~ "

I 'I

~t~'i

',

' '

.

,

/'~'i ~

~~i

t

,,,;,.~ ~I'" q ~, q ' ,~; ! '~ ~1~'"' '

P '

r t,

,,,'

~

I

,

',,! '.,"

,

'1

I

.

'~o ,~", /','~,'

"

150/,.V

. ...., P4"P3

/'

!':'"""

•

J~' ~I '~/,'.~r~,~

. , ;,,'. .. ', ~I ~,,~

t ,'.}

~

': '~, ,14..','.'/',

,

.

.: ": .~4 '"

'.L,

.

".

," . J'~.

' .

I",

,,.

~"

,:,I~ ...

,

....

,'~.,

, , .. ~ I~ ,~,-,,~.,L~,~,,,,~,,4~.,'~~.~,.,,,,,,,.. , , ' ! ~ ~ , " , I .'.1 ' , , ~ _ ,~ ~ :l ~ 'A,,~, ~ ~, ,t I !).

I~',~,,,,

,,

,,.,..,.,~,, ..... .,-!,,;,,,-,..,.,,..~ .....

'~

..,,.

I

'..

' '

~ ~

,1~

~,~

'~-'~

/,i,,,;.,.,Ii!~, ,,I,

,., . . . . I,,,[,i!l.,~,~.~,~i,l,,~,,~,,~ .~,~,.

~I,..........., . '.....~ , ,,. ~"I ~". ', ,~. .,',,~ ,,,, .; , I,l,. , ,. ,. ~ ,,. ,,, ~ ., , ; ,. ,.,.~, !,~,, ~,,. ,;~ iw,, ,,, .M. ..~ ~ ~'.. l d,,~,,,,,~Jl,~./,,,!l~,,~ ~ V, ~, ~,,,.,,,~,,,~,,,t.,~, ~., A..,~ ~ ..., .... ' A ; ,... .., ...-. ., ! I I ~ ' I" l ~ ~'/ I~ IVltll I J~,~ ~ jl,' I'lL '~ ,,

,~,

'3-"

,..

'""':r."!~'"'~'°""~/"!~':#"!~"

'

'

.,.,

~

"

, ,

,,

I

I

,

I

,

~

,

r

1, t

" !,;'.'l'~,,h~,~.#.~. ~' .~1~ ~k ;. ,,,,,, ,,..~.,,,,~i,I~tl I ~, ~.l,I~ ~.,~

..,

~

,

,,

~t.

,,,~"~','~,~", ~,I I~

,

.

.

.~

.,

,

,

~,t~'", ','~

,.,' ! ~ ~ ~' ','~. ,~, .i'~,.,,-~I I~~ ~,~ :.

I sec

w.q.l~|b

,

.

....

.,'

,,

,~ v,~,

'f'['"

i~!'~

~i I!'t!~

'

;:

.

';

,

r~],~l

rz'.,

.It

rlL

(~,,!'

I"t,I'l,,

",".,".

'

~'.'"

'

56 Status was fatal in about 1/4 of the animals in this model. When status epilepticus was induced in rats using kainic acid, lithium and pilocarpine, or homocysteine thiolactone in cobalt-lesioned rats, EEG patterns were observed which were markedly similar to the human patterns (Figs. 1B-5B). In each model, status epilepticus began with discrete electrographic seizures which began to merge after different periods of time for each model and eventually evolved into continuous, high amplitude spiking. Rats developed flat periods whenever a period of continuous spiking exceeded 2 or 3 min. Flat periods were even seen during unusually long discrete seizures. As can be seen from the figures, rats spike at a much faster rate than humans, and this may account for the more rapid appearance of flat periods in the EEGs of rats. After a period of continuous spiking, the flat periods which occurred became longer and appeared more frequently, until finally the pattern converted to PEDs. The time period over which this progression of EEG changes took place was different for each model, but the order in which the changes occurred was identical. The EEG patterns illustrated in Figs. 1B-5B for each of the 3 experimental models of status epilepticus were recorded sequentially from 1 rat for each of the 3 models. However, these EEG patterns and the sequence in which they occurred were seen consistently in all rats in which fully developed status epilepticus was induced, using any of the 3 experimental procedures. DISCUSSION On the basis of fragments of human EEGs recorded during status epilepticus, we hypothesized that there is a progressive sequence of EEG changes which will occur in untreated generalized convulsive status epilepticus. We have confirmed our hypothesis by the observation of this sequence

of patterns in 3 experimental models of generalized convulsive status epilepticus in the rat. We suggest that this sequence of EEG changes represents a universal electrical response to generalized convulsive status epilepticus and that the changes in the EEG morphology are a reflection of underlying pathophysiological responses during the course of status epilepticus. For instance, the 'flat periods' seen in the last 2 patterns may be caused by increased inhibitory drive in the brain in response to prolonged status epilepticus. We have preliminary data that GABA concentrations rise progressively in many areas of the brain during prolonged experimental status epilepticus 16. In humans, when the EEG exhibits discrete seizures or a merging pattern of waxing and waning ictal activity overt generalized convulsions are usually observed clinically. When the electrical pattern is one of continuous rhythmic ictal activity, with or without flat periods, or periodic epileptiform discharges on a flat background, overt generalized convulsions may or may not be seen. These patterns may be associated clinically with what we have called 'subtle' generalized convulsive status epilepticus ~4, or the patient may exhibit electrical status epilepticus only. Status epilepticus may be the result of excessive excitatory neurotransmission, reduced inhibitory neurotransmission or some combination of these 19. Two of the experimental models which we have used, lithium plus pilocarpine and kainic acid, are based on administration of an excitatory agent. The convulsant mechanism of homocysteine is less clear. Olney et al. 1~have recently presented evidence that homocysteic acid is the endogenous ligand for the excitatory NMDA receptor, while others have thought that homocysteine convulsions are a result of inhibition of GABA synthesis 7. It might therefore be argued that this sequence of electrical changes occurs only in status epilepticus induced by excessive excitation. However, we

Fig. 4. A: continuousictal dischargeswith flat periods recordedprior to treatment in a 68-year-oldman. The seizurefocusis clearlyin the lefthemispherebut spreadof ictalactivityto the righthemispherecan be seenas well. B: continuousictaldischargespunctuated by flat periods are illustrated in the top recording in rat no. 06, 148 rain after kainic acid injection. The middle recording also demonstrates continuous ictal dischargeswith flat periods in rat no. 206, 75 min after homocysteineinjection. Continuous ictal discharges punctuated by flat periodsare also seen in the bottomrecordingin rat no. 325,109min after pilocarpineinjection.

58 suggest that this sequence of patterns also occurs when status epilepticus is induced by other means. Figures and descriptions published by Blennow et al. 5 suggest that the same sequence of EEG changes may also occur when status epilepticus is induced by bicuculline, a postsynaptic GABA antagonist. These authors describe a sequence of EEG changes after injection of bicuculline which starts with a massive seizure discharge (our discrete seizures), changes quickly to a mixed pattern of irregular spike discharge and rhythmic spikeand-wave discharges (our waxing and waning ictal pattern) and then converts after about 5 rain to a pattern dominated by bursts of spikes separated by periods of very low voltage EEG (our continuous seizures with 'flat' periods). The pattern finally converts to one of single spike and slow wave complexes (our PEDs on a 'flat' background). In addition, in response to our preliminary presentation of this work is, Lothman et al. 8 recently presented data showing that this same sequence of patterns also occurs when status epilepticus is induced in rats by electrical stimulation of the hippocampus. There is convincing evidence from experimental animal studies that prolonged ictal discharges cause neuronal damage, even in the absence of generalized convulsive motor seizure activity and even when oxygenation and blood pressure are carefully controlled 9'1°. Aicardi and Chevrie 2 have reported a 34% incidence of newly acquired neurologic deficits in a group of children who experienced status epilepticus lasting more than 1 h. Aicardi and Baraton ~ observed ventricular enlargement following prolonged generalized status epi!epticus and Fugiwara et al. 6 found cerebral atrophy on the CAT scan following prolonged status epilepticus in children. This evidence, from experimental animal and human clinical studies, provides ample reason to believe that continuous seizure discharges pose a

significant threat to neuronal integrity. Thus it is important to identify electrographic evidence of continuing status epilepticus, even if an electromechanical dissociation has occurred which prevents the patient from having continuous or repeated generalized tonic-clonic seizures 14. Our data suggest that all of the 5 patterns (discrete seizures, waxing and waning ictal discharges, continuous ictal activity, continuous activity with 'fiat' periods, and periodic epileptiform discharges on a 'flat' background) are examples of continuous ictal epileptic activity. All 5 of these patterns can be stopped with the use of anticonvulsant drugs, with the EEG converting to a low-voltage slow post-ictal pattern. Thus, it is important to monitor treatment of status epilepticus with EEG. If a patient who presents with generalized tonic-clonic status epilepticus stops having overt seizures but continues to have subtle convulsive activity or remains unconscious, it is likely that the patient is having electrographic seizures which should be treated. Any patient who does not clearly respond to antiepileptic drug treatment of his status epilepticus with rapid improvement of consciousness should have an EEG. Identification of a predictable sequence of EEG changes during status epilepticus in human patients, and confirmation of the sequence in experimental models of status epilepticus in the rat argues strongly that this is the natural history of untreated status epilepticus in man. Use of experimental models to elucidate the neurochemical and pathophysiologic mechanisms responsible for the progressive changes in cortical electrographic activity may provide insights wihich will lead to new approaches to the treatment of human status. ACKNOWLEDGEMENTS We thank the following individuals who partici-

Fig, 5. A: periodic epileptiform discharges on a fiat background recorded prior to treatment in a 64-year-old man. He was given 1000 mg phenytoin and 5 mg lorazepam intravenously end the EEG converted to a typical pattern of low-voltage slow waves. B: periodic epileptiform discharges are seen in the top recording from rat no. 06, 5 h 46 rain after kalnic acid injection. At this point the rat is barely conscious and can be picked up and moved with minimal alerting behavior. The middle tracing also illustrates periodic epileptiform discharges on a flat background in rat no. 206, 2 h 12 min after homocysteine injection. The bottom recording exhibits very regular high-voltage periodic epileptiform discharges as the terminal pattern of status epilepticus in rat no. 325 recorded 2 h 19 min after pilocarpine injection.

59 A

"

j

.

.

734

.,

FP2- F8 F8- T4

T 4 - T6

/

T6" 02

.

~

/,, G,.~-~-.~

r

FPI" F 7

F?-T3

'

T3"T5

•

~

o

,

T S " OI

• !.,

t

~

,

~

/

1

.

t

J

I

t

'.

,

FP2 - F 4

F4"¢4

- - - . - . ~ ' ~ ~

-_ ~ , _ , t ~ - - - ~ - L . L - -

_.,,..-.~

~.,,,/.----~....,,

¢4 "P4

P4 "0;~

::

FPI" F 3

J.4/',.;/,f'----.,.,.-,:

!

,I

i

,

!

'

',

i

F3"C3

¢3"P3 P 3 " OI

B F3-P3

% •

~.

z

150~v

P3-P4

'.

r

,~.

_e'

~

A

~

,.

"

!~ ~

f

'

~_

~..

t zoo,vl ISllC

u

g

t,

v~-,, . ~ - - - ~ - - - ~ ~ - v ~ - v - - ~ - ¢ - - ~ - ~ - ; - ~ - ~ - - ~ P3-P4 ' •

P4 " F 4

F4"F3

~

~

-

-

'

~

'

~

'

~

' ~

~

-

"'-'~""-"~ •

rl~k.,.,~,...,..,...~

'

60 p a t e d on our status epilepticus t r e a t m e n t t e a m at various times during the course of this study: Kat e n O. B a r b e r , R . N . , Mary C. B a r e , R . N . , M . P . H . , Elinor B e n - M e n a c h e m , M . D . , Cynthia Chabay, M . D . , Christopher D e G i o r g i o , M . D . , David G h e r e t , M . D . , Steve G l y m a n , M . D . , Margaret G o o d m a n , M . D . , Jan Lee, R . N . , M . S . N . ,

L i n d a Norton, R . N . , L o w e l l Nelson, M . D . , Susan Salisbury, R . N . , M a r k Shultz, M . D . W i t h o u t their s u p p o r t and the excellent care they p r o v i d e d for o u r patients in status epilepticus this study could not have been carried out. This study was supp o r t e d in part by U S P H S Contract No. N O 1 - N S 80-2332.

REFERENCES

tus epilepticus in well-oxygenated rats causes neuronal necrosis, Ann. NeuroL, 18 (1985) 281-290. 11 Olney, J.W., Price, M.T., Shahid SaUes, K., Labruy~re, J., Ryerson, R., Mahan, K., Friedrich, F. and Samson, L., LHomocysteic acid: an endogenous excitotoxic ligand of the NMDA receptor, Brain Res. Bull., 19 (1987) 597-602. 12 Roger, J., Lob, H. and Tassinari, C.A., Status epilepticus. In: P.J. Vinken, G.W. Bruyn, O. Magnus and A.M. Lorentz de Haas (Eds.), Handbook of Clinical Neurology. Vol. 15. The Epilepsies, North-Holland, Amsterdam, 1974, pp. 145-188. 13 Treiman, D.M., DeGiorgio, C.M., Ben-Menachem, E., Gehret, D., Nelson, L., Salisbury, S.M., Barber, K.O. and Wickboldt, C.L., Lorazepam vs. phenytoin in the treatment of generalized convulsive status epilepticus, Neurology, 35, Suppl. 1 (1985) 284. 14 Treiman, D.M., DeGiorgio, C.M., Salisbury, S. and Wickboldt, C., Subtle generalized status epilepticus, Epilepsia, 25 (1984) 653. 15 Treiman, D.M., Walton, N.Y., Wickboldt, C. and DeGiorgio, C.M., Predictable sequence of EEG changes during generalized convulsive status epilepticus in man and three experimental models of status epilepticus in the rat, Neurology, 37, Suppl. 1 (1987) 244. 16 Walton, N.Y., Gunawan, S. and Treiman, D.M., Changes in brain amino acid levels in the rat during early, mid, and late status epilepticus induced by lithium and pilocarpine, Epilepsia, 29 (1988) 690. 17 Walton, N.Y. and Treiman, D.M., Experimental secondarily generalized convulsive status epilepticus induced by D,L-homocysteine thiolactone, Epilepsy Res., 2 (1988) 79-86. 18 Walton, N.Y. and Treiman, D.M., Response of status epilepticus induced by lithium and pilocarpine to treatment with diazepam, Exp. Neurol., 101 (1988) 267-275. 19 Woodbury, D.M., Experimental models of status epilepticus and mechanisms of drug action. In: A.V. Delgado-Escueta, C.G. Wasterlain, D.M. Treiman and R.J. Porter (Eds.), Advances in Neurology: Status Epilepticus, Raven Press, New York, 1983, pp. 149-160.

1 Aicardi, J. and Baraton, J., A pneumoencephalographic demonstration of brain atrophy following status epilepticus, Dev. Med. Child. Neurol., 13 (1971) 660-667. 2 Aicardi, J. and Chevrie, J.J., Convulsive status epilepticus in infants and children. A study of 239 cases, Epilepsia, 11 (1970) 187-197. 3 American Electroencephalographic Society. A proposal for standard montages to be used in clinical EEG, J. Clin. Neurophysiol., 3, Suppl. 1 (1986) 26-33. 4 Ben-Ari, Y., Tremblay, E., Riche, D., Ghilini, G. and Naquet, R., Electrographic, clinical and pathological alterations following systemic administration of kainic acid, bicuculline or pentetrazole:~.metabolic mapping using the deoxyglucose method with special reference to the pathology of epilepsy, Neuroscience, 6, Suppl. 7 (1981) 1361-1391. 5 Blennow, G., Folbergrova, J., Nilsson, B. and Siesj6, B.K., Effects of bicuculline-induced seizures on cerebral metabolism and circulation of rats rendered hypoglycemic by starvation, Ann. Neurol., 5 (1979) 139-151. 6 Fugiwara, T., Ishida, S., Miyakoshi, M., Sakuma, N., Moriyama, T., Seino, M. and Wada, T., Status epilepticus in childhood: a retrospective study of initial convulsive status epilepticus and subsequent epilepsies, Folia Psychiat. Neurol. Jpn., 33 (1979) 337-344. 7 Griffiths, R., Williams, D.C., O'Neill, C., Dewhurst, I.C., Ekuwem, C.E. and Sinclair, C.D., Synergistic inhibition of [3H]muscimol binding to calf-brain synaptic membranes in the presence of L-homocysteine and pyridoxal 5'-phosphate: a possible mechanism for homocysteine-induced seizures, Eur. J. Biochem., 137 (1983) 467-478. 8 Lothman, E.W., Bertram, E.H., Bekenstein, J.W. and Perlin, J.B., Self-sustaining iimbic status epilepticus induced by 'continuous' hippocampal stimulation: electrographic and behavioral characteristics, Epilepsy Res., 3 (1989) 107-119. 9 Mek,.um, B.S., Vigouroux, R.A. and Brierley, J.B., Systemic factors and epileptic brain damage, Arch. Neurol., 29 (!973) 82-87. 10 Nevander, G., Ingvar, M., Auer, R. and Siesj6, B.K., Sta-