Generalized epilepsies: a review1
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Generalized epilepsies: a review 1
Hans Luders, M.D., Ph.D. In this discussion of the pathophysiology of generalized seizures,
Ronald P. Lesser, M.D. the authors conclude that current evidence supports a primarily
Dudley S. Dinner, M.D. cortical origin, and that there is a continuum between strictly
Harold H. Morris III, M.D. partial seizures at one end of the spectrum and definite generalized
seizures at the other. Typical interictal electroencephalogram
(EEG) patterns seen in patients with generalized seizures and their
clinical correlations are discussed, and the ictal EEG patterns are
subdivided according to their relationship to interictal EEG pat-
terns. Special tests, such as sleep, photic stimulation, hyperventi-
lation, evoked potentials, computer analysis, and prolonged EEG
video-monitoring, which enhance diagnostic ability, are reviewed.
Index terms: Epilepsy • Seizures
Cleve Clin Q 51:205-226, Summer 1984
The International Federation of Societies for Electroen-
cephalography and Clinical Neurophysiology (IFSECN)1
classification of seizures distinguishes two main groups: (a)
generalized, and (b) partial.
In partial seizures, only a limited portion of the brain is
involved whereas in generalized seizures, the brain is af-
fected diffusely. In extreme cases, these two groups can be
clearly differentiated, e.g., focal motor seizures with no
1
Department of Neurology (H.L., R.P.L., secondary generalization vs. generalized tonic-clonic sei-
H.H.M.) and Plastic Surgery (D.S.D.), The Cleveland zures with no aura and apparently bilateral, bisynchronous
Clinic Foundation. Submitted for publication and ac- EEG onset. Most seizures, however, fall somewhere be-
cepted Oct 1983.
tween these extremes. Moreover, focal seizures tend to
0009-8787/84/02/0205/22/$5.95/0 trigger secondary generalized seizures, and differentiation
is particularly difficult when the focal component is of
Copyright © 1984 by the Cleveland Clinic Foundation short duration. In some patients, all clinical seizures are
205
Downloaded from www.ccjm.org on November 2, 2021. For personal use only. All other uses require permission.206 Cleveland Clinic Q u a r t e r l y Vol. 51, No. 2
apparently of generalized onset a n d their focal
FOCAL REGIONAL
n a t u r e can be detected only by EEG studies a n d /
A or videotape monitoring. In those with two epi-
leptogenic foci a n d a m a r k e d tendency to second-
ary generalization, the distinction between partial
and generalized is usually impossible. Multiple
(three or more) epileptogenic foci almost invari-
ably result in secondary generalization, a n d these
cases a r e usually classified as generalized seizures.
T w o conclusions can be drawn f r o m these obser-
vations:
BILATERAL (MAXIMUM RIGHT) GENERALIZED (a) It is frequently impossible to distinguish
clearly between partial and generalized sei-
zures. T h e r e appears to be a c o n t i n u u m
between clear-cut partial seizures at o n e end
of the spectrum a n d generalized seizures at
the o t h e r (Fig. 1), most falling s o m e w h e r e
in between.
(b) If we move along the spectrum f r o m partial
toward generalized seizures, we can con-
clude, by extrapolation, that "generalized"
Fig. 1. Continuum between partial seizures (upper left) and seizures are actually multifocal seizures with
generalized seizures (lower right), showing the distribution map of an e x t r e m e tendency f o r secondary gener-
interictal epileptiform discharges. Darkened area = 90%-100% of alization (Fig. 2). T h i s concept will be dis-
the maximum amplitude of spike; area within the darkest circle = cussed in detail later.
more than 80% of the maximum amplitude of the spike.
From the surgeon's point of view, however, a
clear line must be drawn between "partial" and
"generalized" seizures, i.e., will resection of a
PARTIAL VS GENERALIZED SEIZURES limited area of cortex abolish the seizures or not?
T h e affirmative implies that (a) all seizures are
triggered f r o m one focus, a n d (/;) t h e r e are no
o t h e r secondary foci capable of triggering sei-
zures a f t e r resection of the primary focus. T h e s e
NUMBER surgical criteria can occasionally be m e t even in
OF S P I K E the presence of a m a r k e d tendency for secondary
FOCI generalized seizures.
Influenced by these surgical criteria and the
concept of a c o n t i n u u m , we use the following
practical subdivision of seizures:
(a) Partial seizure disorder: Electroclinical in-
f o r m a t i o n indicates that resection of a lim-
ited area of cortex would abolish the sei-
zures, e.g., left mesial temporal focus.
TENDENCY OF
(b) Generalized seizure disorder: Electroclinical
SECONDARY
evidence indicates that resection of even an
GENERALIZATION extensive area of cortex would n o t abolish
the seizures.
Pathophysiology of generalized seizures
t
PARTIAL GENERALIZED
M o d e r n theories of the pathogenesis of gen-
SEIZURES SEIZURES eralized spike a n d wave complexes began with
Penfield a n d Jasper 2 (Fig. 3). T h e y assumed that
Fig. 2. Continuum between partial and generalized seizures,
depending on the number of spike foci and tendency toward primary "generalized" seizures (e.g., generalized
secondary generalization. absences) are essentially secondary to a focal ab-
Downloaded from www.ccjm.org on November 2, 2021. For personal use only. All other uses require permission.Summer 1984 Generalized epilepsies 207
normality in poorly defined, deep-seated midline maker mechanism, he was able to demonstrate
gray structures called the "centrencephalon." In that, in cats pretreated with penicillin I.M. (to
other words, the group most famous for pioneer- produce diffuse hyperexcitability), bursts of 3-Hz
ing surgical treatment of focal cortical epilepsy spike and wave complexes were triggered most
suggested that "generalized" epilepsy is an readily from the same deep-seated, nonspecific
expression of a fairly focal subcortical triggering midline gray structures that produced the re-
mechanism. In the words of Jasper and Kersh- cruiting response. He postulated, therefore, that
man: 3 primary generalized seizures are produced by an
abnormal interplay of both, namely, cortex and
The sudden loss of consciousness (in an absence) is a subcortical midline gray structures (reticulo-cor-
form of onset comparable to "tingling of one hand" in tical theory). He did not commit himself directly,
a patient with cortical focal seizures. In cases of focal but the obvious implication is that generalized
cortical onset, minor attacks may be noted as "whirling seizures result from hyper-reactivity of the cortex
lights" when the epileptic discharge is confined to the
occipital lobe. Minor attacks of the bilaterally synchron-
to afferent impulses from the midline subcortical
ous 3/sec wave and spike form are associated with loss gray structures, which act as trigger (or pace-
of consciousness when the discharge is confined to maker) mechanisms. In a recent article, 6 he
centers (or circuits) primarily concerned with this func- stated:
tion.
The concept (of reticulo-cortical epilepsy) can be sum-
The following observations constitute the basis marized as follows: There is a mild diffuse epileptogenic
state of the cortex. T h e cortex is hyperexcitable and
for the centrencephalic theory: responds by elaborating spike and wave discharges pref-
(a) T h e cortical epileptiform discharges seen in erentially, but not exclusively, to volleys that originate
primary generalized seizures tend to be gen- from the thalamus, particularly those that normally
eralized and almost perfectly bilaterally syn- produce spindles and recruiting response.
chronous from the beginning.
(b) Stimulation of deep-seated midline gray nu- This reticulo-cortical theory is currently the
clei can produce bilateral synchronization of most widely accepted.
the EEG,4 and under appropriate condi- Although Gloor's work in cats is certainly con-
tions, bursts of spike and slow wave com- vincing, there is no need to attribute an active
plexes. 5 pathogenetic role to the subcortical gray struc-
The most vocal arguments against the centren- tures, as the name "reticulo-cortical epilepsy" im-
cephalic theory came almost two decades later plies. We believe that the evidence primarily
from the same Montreal group, now under the suggests that generalized epilepsies are an expres-
leadership of Peter Gloor 6 (Fig. 4) and can be sion of a pathological degree of diffuse cortical
summarized as follows: epileptogenicity (Fig. 5). Subcortical gray struc-
(a) Intracarotid injection of convulsants (Metra- tures certainly participate in bilateral synchroniz-
zol) produces generalized 3-Hz spike and ing mechanisms and in diffuse modulation of
wave complexes and clinical absences. In- cortical activity. However, there is no evidence
travertebral injection of convulsants has no to suggest that these physiological cortico-sub-
effect or arrests seizures.6'7 cortical mutual interactions are pathological in
(b) Diffuse cortical application of penicillin pro- patients with generalized seizures.
duces generalized 3-Hz spike and wave T h e following points support the cortical the-
complexes in relatively low concentration ory:
whereas its injection into subcortical gray (a) Gloor's evidence 6 of diffuse cortical hyper-
structures does not. 8 excitability supports the cortical theory di-
From the latter observation, Gloor 6 concluded rectly (see above).
that diffuse "hyperexcitability" of the cortex is an (b) Focal cortical epileptogenic lesions can pro-
essential factor in the pathogenesis of generalized duce generalized spike and wave complexes
seizures. However, he did not totally discard his and generalized seizures without evidence
preceptor's concept of the centrencephalon. He of subcortical gray structure lesions, for ex-
was impressed by the sudden onset and abrupt ample, mesial frontal lesions.9,10 In these
termination of generalized bursts of 3-Hz spike cases, interictal focal spikes were recorded
and wave complexes that resemble the turning from the mesial frontal region, stimulation
on and off of a light switch. Looking for a pace- of the focus produced seizures, and resec-
Downloaded from www.ccjm.org on November 2, 2021. For personal use only. All other uses require permission.208 Cleveland Clinic Q u a r t e r l y Vol. 51, No. 2
CENTRENCEPHALIC THEORY CORTICO-RETICULAR THEORY
PENFIELD A N D JASPER GLOOR
CORTICAL THEORY
Fig. 3. Centrencephalon theory 2
Fig. 4. Cortico-reticular theory 6
Fig. 5. Cortical theory
tion of the focus abolished the clinical sei- sphere to another. Figures 9, 10, and 11 show
zures, none of which has been demonstrated the distribution of interictal discharges in a
in subcortical gray structures. Focal lesions patient with idiopathic generalized tonic-
of the subcortical gray structures are not clonic seizures. It illustrates a continuum
accompanied by epilepsy, between strictly focal discharges and gener-
(r) Patients with generalized seizures frequently alized spike and wave complexes involving
exhibit focal epileptiform discharges (Figs. progressively larger areas of the cortex. Bi-
6-8), mostly in the frontocentral regions and lateral discharges may begin earlier or have
shifting f r o m one hemisphere to another. a significantly higher amplitude on one side.
Occasionally, these discharges involve other On EEG, the picture of an individual dis-
regions, but always shift from one hemi- charge is indistinguishable f r o m a focal cor-
Downloaded from www.ccjm.org on November 2, 2021. For personal use only. All other uses require permission.Summer 1984 Generalized epilepsies 209
ABSENCE SEIZURES
RIGHT FRONTAL SPIKES
Ts-0| Ps -0|
T 6 - OÍ
TlOOiV I 1
-L ^ I second
Fig. 6. Patient with typical absence seizures, showing extremely focal spikes limited to
Fp2 and to a lesser degree F4.
tical epileptogenic lesion with marked tend- charges that can be detected from the scalp.
ency to secondary bilateral synchrony, ex- There is usually no evidence of subcortical
cept that in generalized epilepsies, the foci gray pathology in these cases, and resection
preceding the generalized discharges shift of the cortical focus frequently abolishes the
between the two hemispheres. seizures. A similar phenomenon can occur
(d) Detailed analysis of spike foci by electrocor- when an entire hemisphere is affected. Fig-
ticography usually reveals involvement of ures 16 and 17 illustrate the bridge between
an extensive cortical area (10 cm 2 or more) focal and generalized epilepsy. Both proba-
harboring innumerable spike foci firing in- bly consist of multifocal independent spike
dependently or in groups (Figs. 12-15). Oc- foci, discrete in partial epilepsy but covering
casionally, these small spike foci fire syn- the whole cortex in generalized epilepsy
chronously and are usually the only dis- (Fig. 18). Penicillin-induced spike foci in cats
ABSENCE SEIZURES
INDEPENDENT LEFT A N D RIGHT FRONTAL SPIKES
I second
Fig. 7. Patient with typical absence seizures, showing independent left and right
frontal spikes. The short burst on the left side clearly originates at F7 and then becomes
bilateral. The burst on the right side is limited to F8 thoughout.
Downloaded from www.ccjm.org on November 2, 2021. For personal use only. All other uses require permission.210 Cleveland Clinic Q u a r t e r l y Vol. 51, No. 2
GENERALIZED 3Hz SPIKE-AND-WAVE COMPLEXES
ASYMMETRIC DISTRIBUTION
Ps-0|
Fp2 - F4 tvAA/W/vy'
F4-C4
T100.1V 1 1
I Itacond
Fig. 8. Patient with typical absence seizures, showing a burst of generalized
spike and wave complexes which is clearly lateralized to the left side. At other
times, the patient had generalized bursts which were symmetrically distributed.
will fire independently if relatively weak and m a t u r a t i o n , the EEC may evolve f r o m mul-
far apart, but synchronously if stronger a n d / tifocal i n d e p e n d e n t spikes to generalized
or closer t o g e t h e r . " spike a n d wave complexes. For example,
(?) Similar clinical syndromes may show either patients with minor m o t o r seizures exhibit
bilateral multifocal i n d e p e n d e n t spikes or generalized slow spike and wave complexes,
generalized spike and wave complexes. With but many also show bilateral multifocal in-
INTERICTAL EPILEPTIFORM DISCHARGES INTERICTAL EPILEPTIFORM DISCHARGES
Fig. 9. Distribution map of an interictal epileptiform discharge Fig. 10. Distribution map of an interictal epileptiform dis-
ili a patient with idiopathic generalized tonic-clonic seizures. Dark- charge in a patient with idiopathic generalized tonic-clonic seizures,
ened area = 100%; first circle > 9 0 % ; second circle >80%. Darkened area = 100%; first circle >90%; second circle >80%.
Downloaded from www.ccjm.org on November 2, 2021. For personal use only. All other uses require permission.S u m m e r 1984 Generalized epilepsies 211
d e p e n d e n t spikes, some evolving f r o m this
p a t t e r n to generalized slow spike a n d wave INTERICTAL E P I L E P T I F O R M D I S C H A R G E S
18 YEAR OLD MAN WITH GENERALIZED TONIC-CLONIC SEIZURES
complexes on maturation.
We conclude, t h e r e f o r e , that generalized epi-
lepsy is an expression of diffuse cortical epilep-
togenicity that is exhibited on the EEG as multi-
focal i n d e p e n d e n t spikes with a m a r k e d tendency
to secondary generalization. As this tendency
increases, the multifocal spikes evolve into gen-
eralized spike a n d wave complexes. T h e r e is no
need to assume pathology in the subcortical gray
structures. T h i s does not imply that the normal
functions of the gray structures d o not influence
the cortical discharges in patients with general-
ized seizures j u s t as they d o in those with focal
epilepsy. Increase in spike f r e q u e n c y d u r i n g sleep
and triggering of spikes by stimulation of sub-
cortical gray structures is seen as frequently in
focal as in generalized seizures. O n e speaks of
"cortical epilepsy" in r e f e r e n c e to focal seizures
since the pathology is mainly cortical. T h i s is Fig. 11. Distribution map of an interictal epileptiform dis-
probably also t r u e for generalized epilepsies. charge in a patient with idiopathic generalized tonic-clonic seizures.
Darkened area = 100%; First circle >90%; second circle >80%.
Primary vs. secondary bilateral synchrony
T h e centrencephalic theory conceives primary seen. However, once the epileptiform n a t u r e of
bilateral synchrony (primary generalized epilep- the discharge has been established, the distribu-
sies) to be seizures triggered f r o m the "centren- tion of the discharge is m o r e i m p o r t a n t than the
cephalon," whereas secondary bilateral syn- waveform.
chrony would be a focal cortical epilepsy second-
arily triggering generalized spike a n d wave com- EPILEPTOGENIC AREA
plexes by firing into the centrencephalon (Fig. MULTIPLE INDEPENDENT SPIKE FOCI
19).
T h e cortical theory assumes that secondary
bilateral synchrony is a c o m m o n mechanism for
both "primary" generalized seizures a n d focal
seizures with secondary generalized spike a n d
wave complexes (Fig. 20). T h e d i f f e r e n c e be-
tween the two is that (a) in generalized seizures,
multiple bilateral cortical foci tend to trigger
secondarily generalized bursts, whereas (b) in
focal seizures with secondary diffuse spike a n d
slow wave complexes, a single cortical focus trig-
gers the bursts.
Interictal epileptiform discharges
I. Specific abnormality: generalized epileptiform
discharges
T h e only EEG abnormality which supports the Fig. 12. Diagram traced directly from a lateral skull radio-
diagnosis of generalized seizures is generalized graph of a patient with intractable complex partial seizures, showing
the outline of the skull and six subdurally implanted flaps, each
epileptiform discharges, usually in the f o r m of containing four contacts. The two flaps in the anterior temporal
spike a n d wave or sharp a n d wave complexes. lobe were implanted to detect the main epileptogenic focus. The
Less frequently, polyspikes or polysharp waves, posterior four flaps were implanted to define the posterior edge of
paroxysmal beta or alpha, or o t h e r patterns are the focus prior to resection.
Downloaded from www.ccjm.org on November 2, 2021. For personal use only. All other uses require permission.212 Cleveland Clinic Quarterly Vol. 51, No. 2
13 EPILEPTOGENIC AREA 15 EPILEPTOGENIC AREA
MULTIPLE INDEPENDENT SPIKE FOCI MULTIPLE INDEPENDENT SPIKE FOCI
(46)1 (46)1
(46)2 ——>/-
4 (46)2
(46)3 (46)3 W V
(46)3 — (46)3
I
(48)1
(48)1
(48)2 ^ y ^
(48)2 li'-—
(48)3 •—
(48)3 1
(48)4
(48)4
(16)1
(16)1
(16)2
(16)2
(16)3 ~
(16)3 WV~ VWV
(16)4 ^ / y •VW
(16)4
j700>iV
1700/iV I sec
14 EPILEPTOGENIC A R E A
MULTIPLE INDEPENDENT SPIKE FOCI
(46)1
(46)2
Figs. 13-15. Recordings made
(46)3 via the two flaps in the anterior
temporal pole (#46 and 48) and one
(46)3 flap (the one with the black dot in
Fig. 12) of the more posteriorly
(48)1 implanted electrodes (#16); inde-
pendent spike foci from all elec-
trodes are marked in black. These
(48)2 foci stem from electrodes 46(1),
46(2), 46(3), 46(4), and 48(1) (Fig.
(48)3 13) electrodes 48(2), 48(3), 48(4),
and 16(4) (Fig. 14), and electrodes
(48)4 J 46(1), 46(2), 48(1), 48(2), 48(3),
and 48(4) [first column] and elec-
p-
trodes 48(1) 48(2), 48(3) and 48(4)
(16)1
[second column] (Fig. 15).
(16)2
(16)3
(16)4 -vA—J /V.
\aT—
j700/iV I sec
Downloaded from www.ccjm.org on November 2, 2021. For personal use only. All other uses require permission.S u m m e r 1984 Generalized epilepsies 213
MULTIFOCAL INDEPENDENT SPIKE FOCI
PARTIAL EPILEPSY
GENERALIZED EPILEPSY
Figure 16. Intractable partial seizures starting with visual hal-
lucinations and evolving into left-sided clonic seizures (there were
no generalized tonic-clonic seizures) in a patient with right hemia-
trophy and mild left hemiparesis. The awake recordings showed
almost continuous right occipital spikes at a repetition rate of
approximately I Hz. During sleep, the right occipital focus was
significantly less frequent and independent right frontal and right
temporal spikes were activated. [See Fig. 1 for explanation of
isopotential lines.]
In generalized seizures, discharges tend to be
diffuse but are not of equal amplitude in all
electrodes. Invariably, a clearly defined field is
seen, the m a x i m u m o c c u r r i n g usually at the su-
Fig. 18. Diagram showing the sim-
perior frontal or sometimes at the central elec-
ilarity between partial and generalized
trodes. A m a r k e d d r o p o f f is seen posteriorly and epilepsy. Both are expressions of mul-
tifocal independent spike foci with a
marked tendency toward synchrony,
MULTIFOCAL INDEPENDENT SPIKE FOCI the only difference being the cortical
WIDESPREAD DISCHARGE area involved.
into the temporal region. Ear electrodes, partic-
ularly the nasopharyngeal or sphenoidal, a r e in-
volved minimally or n o t at all. Sometimes this
d r o p o f f posteriorly a n d laterally is so accentuated
that the discharges resemble bifrontal or bicen-
tral spikes m o r e than generalized spike a n d wave
complexes. A n o t h e r i m p o r t a n t characteristic of
generalized seizures is the f r e q u e n t o c c u r r e n c e
of unilateral focal spikes (Figs. 6-11) in the f r o n -
tal or central region that shift f r o m side to side.
In cases with i n f r e q u e n t discharges, a sample of
2 0 - 3 0 minutes may, for example, show only iso-
lated right frontal spikes a n d no left frontal or
generalized discharges. Additional recordings
will be necessary to d i f f e r e n t i a t e between a focal
right frontal a n d a generalized seizure disorder
also had widespread right hemispheric discharges, with the three
foci illustrated in Figure 16 firing synchronously. with an interictal focal epileptiform manifesta-
Downloaded from www.ccjm.org on November 2, 2021. For personal use only. All other uses require permission.214 Cleveland Clinic Q u a r t e r l y Vol. 51, No. 2
BILATERAL SYNCHRONY (CENTRENCEPHALIC THEORY )
PRIMARY SECONDARY
Fig. 19. Centrencephalon theory. Primary bilateral synchrony is depicted on the left
and secondary bilateral synchrony on the right.
tion. In o t h e r words, even in a generalized sei- II. Neurophysiology of spike and slow wave complex
zure disorder, t h e EEG may exhibit considerable T h e most widely accepted hypothesis is that
focal manifestations. In o u r laboratory, these fo- the spike is an excitatory event correlating with
cal features a r e described in the body of o u r excitatory postsynaptic potentials (EPSPs) at the
reports, but n o t in the classification a n d impres- surface of the cortex, whereas the slow wave is
sion if consistent with a generalized seizure dis- inhibitory, being p r o d u c e d by inhibitory postsyn-
order. aptic potentials (IPSPs) in the d e e p cortical lay-
Generalized discharges are usually asymmetric, ers. 1 2 T h e superficial and d e e p location of the
but the m a x i m u m shifts f r o m side to side. All EPSPs a n d IPSPs explains why both events (spike
transition f o r m s along the following c o n t i n u u m a n d slow wave) are scalp-negative (Fig. 21).
can be seen: (a) strictly unilateral focal spikes;
(b) bifrontal m a x i m u m left or right hemisphere; III. Specificity of generalized spike and wave
(c) "generalized" m a x i m u m left or right hemi- complexes
sphere; and (d) "generalized" bilateral symmet- Specific epileptiform discharges are by defini-
rical (Fig. 1). tion w a v e f o r m s that are clearly differentiated
T h e s e focal features support the theory that f r o m physiological activity. Borderline wave-
generalized seizures are an expression of bilateral forms, however, are difficult to classify. For prac-
multiple cortical spike foci (Fig. 18). tical purposes, only waveforms of clearly patho-
BILATERAL SYNCHRONY (CORTICAL THEORY)
PRIMARY SECONDARY
Fig. 20. Cortical theory, showing primary bilateral synchrony on the left and
secondary bilateral synchrony on the right.
Downloaded from www.ccjm.org on November 2, 2021. For personal use only. All other uses require permission.S u m m e r 1984 Generalized epilepsies 215
logical significance should be classified as epilep- SPIKE SLOW WAVE
tiform. Conservative r e a d i n g will greatly increase
the specificity of the EEG. Note, however, that a
"normal" EEG by no means excludes epilepsy;
when d o c u m e n t a t i o n is crucial for diagnosis, ad-
ditional recordings frequently solve the problem.
K A*
T o increase the specificity of the EEG, gener-
alized epileptiform discharges should be differ-
entiated f r o m the following activity:
(a) Nonspecific burst of generalized sloiv waves.
Bursts of generalized slow waves not associ-
ated with spikes or s h a r p waves have abso-
lutely no epileptogenic significance. T h e y
are seen m o r e often in patients with gener-
alized epilepsy, 1 3 but also in a variety of
physiological circumstances (hyperventila-
tion, sleep) a n d in d i f f e r e n t diseases without
clinical epilepsy (e.g., metabolic encephalop-
athy).
(/;) Bursts of generalized slow waves with intermixed
sharp transients. T h e burst of generalized
slow waves seen in the hypnagogic-hypno- Fig. 21. Superficial EPSPs (spikes) and deep-seated IPSPs (slow
pompic state a n d d u r i n g hyperventilation is waves) produce the same current flow, and hence deflections have
occasionally accompanied by s h a r p tran- the same polarity when recorded from the surface.
sients, usually of relatively long duration
(sharp-wave-like). Less frequently, transients (d) Six-hertz phantom spike and wave complexes.
a r e of short duration (spike-like), but of rel- T h i s pattern consists of a burst of slow waves
atively low amplitude c o m p a r e d with the intermixed with low-amplitude sharp tran-
following slow-wave c o m p o n e n t . This is as- sients of short duration (spike-like) which
sociated only with hypnagogic hypersyn- repeat at 4 - 7 Hz (Fig. 22). Usually, the slow
chrony, t h e spike-like c o m p o n e n t limited to waves a r e much m o r e conspicuous than the
t h e beginning of the burst. W h e n bursts of sharp transients, a n d the whole burst rarely
generalized spike a n d wave complexes occur lasts m o r e than two seconds. However, fre-
only d u r i n g the hypnagogic-hypnopompic quency of spike repetition is only o n e of
state a n d / o r d u r i n g hyperventilation, the many characteristics of this p a t t e r n . Bursts
possibility of "over-reading" should be con- of spike a n d wave complexes with clearly
sidered. If the epileptiform n a t u r e of the outstanding, high-amplitude spikes a r e epi-
pattern is not absolutely clear, an additional leptogenic even if the spikes repeat at 6 Hz
stage II sleep a n d / o r resting awake record- (Fig. 23). Usually, the differentiation of epi-
ing should be obtained. leptogenic a n d nonepileptogenic 6-Hz spike
(r) Vertex sharp transients. Vertex sharp tran- and wave complexes is clearcut. In d o u b t f u l
sients of sleep f r e q u e n t l y are extremely cases, it is always safe to assume that one is
sharp a n d occasionally cannot be clearly dif- dealing with a nonepileptogenic p a t t e r n a n d
ferentiated f r o m epileptiform transients. additional recordings can be requested.
T h e typical prepositivity a n d e x t r e m e sharp- (e) Photoparoxysmal pattern. Photoparoxysmal
ness of the rising phase a n d the characteristic patterns consisting of sharp transients time-
spike-slow wave sequence help to identify locked to the stimulus, p r e d o m i n a t i n g in the
truly epileptogenic events. Persistence of posterior head regions have n o diagnostic
sharp transients d u r i n g the waking state will value, being f r e q u e n t l y seen in normal con-
also indicate its epileptogenic character. In trols and nonepileptic patients. T h e only
d o u b t f u l cases, however, it is wise to insist useful diagnostic p a t t e r n is a generalized,
on additional recordings b e f o r e making a self-sustained p h o t o p a r o x y s m not time-
decision. locked to the stimulus, with clearly d e f i n e d
Downloaded from www.ccjm.org on November 2, 2021. For personal use only. All other uses require permission.216 Cleveland Clinic Quarterly Vol. 51, No. 2
6 Hz P H A N T O M SPIKE-AND-WAVE FAST S P I K E - A N D - W A V E C O M P L E X
14 YRS OLD BOY WITH HEADACHES AND DIZZINESS 25 YRS OLD FEMALE WITH GENERALIZED TONIC-CLONIC SEIZURES
FP1-F3
Fpi - F 3
F3-C3
F3-C3
C3-P3 C3 -P3 v^'wwv'vwvvW'vwVWvAvyV^AjA^A^
FP2-F4
F4-C4
P4-O2
I200^ ' I second '
Fig. 23. Twenty-five-year-old woman with generalized tonic-
clonic seizures and generalized fast spike and wave complexes.
70/iV
I second epilepsy. Additional recordings will frequently
Fig. 22. Short burst of 6-Hz phantom spikes and clarify the diagnosis.
waves in a 14-yr-old boy with headaches and dizziness,
but no history of epilepsy IV. Normal EEG
In some types of epilepsy, normal EEG read-
spikes. A relatively high incidence of differ- ings are more likely than epileptiform manifes-
ent types of photoparoxysmal patterns has tations, e.g., in adults with infrequent generalized
been reported in normal children aged five tonic-clonic seizures.
to 15 years. 14 Nevertheless, a photoparox- T h e single most important factor which deter-
ysmal pattern meeting the above stringent mines the probability of recording epileptiform
criteria can be classified as epileptogenic, abnormalities is the frequency of seizures. This var-
even in young children. ies from patient to patient, but in individuals, the
False-positive tests will be rare if spike and wave correlation is excellent: Those with frequent sei-
complexes are rigidly defined and nonspecific zures almost invariably display EEG abnormali-
abnormalities excluded. However, spike and ties."' Conversely, those with infrequent seizures
wave complexes in apparently nonepileptic pa- will probably not show specific abnormalities on
tients do not always imply over-reading. The routine EEG.
EEG abnormality may be a subclinical manifes- A word of caution: Patients with one or more
tation of an epileptogenic tendency. In a study "spells" a day and normal EEGs may not be
by Zivin and Ajmone Marsan, 15 15% of nonepi- epileptics. These patients should undergo inten-
leptic patients with spike and wave complexes sive EEG monitoring and induction of these epi-
eventually developed seizures during follow-up sodes by suggestion to determine their origin.
studies.
Ictal EEG
False-positive tests can frequently lead to an
erroneous diagnosis of epilepsy or to an incorrect I. Ictal EEG patterns
classification of seizure type. A negative test (even T h e ictal EEG in patients with generalized
under-reading) should not confuse the astute cli- seizures can be divided according to the relation-
nician as a negative EEG by no means excludes ship between interictal and ictal patterns:
Downloaded from www.ccjm.org on November 2, 2021. For personal use only. All other uses require permission.Summer 1984 Generalized epilepsies 217
ABSENCE SEIZURES
2 YRS OtO GIRL
INTERICTAL ICTAL
Fs-A,A2 A A ^ A / ^ ^ - A A
TA
w firn W W
tyMr^
MJVvAV vw
W
"vYX/VW^
0|-A|A2 v hs. -Wv
02-A|A2 W vyVf
Fig. 24. Interictal and ictal patterns in a patient with generalized absence seizures
(a) Ictal pattern = prolonged interictal pattern. The tonic phase, but then evolve into spike and
ictal pattern consists of generalized spike wave complexes of progressively higher am-
and wave complexes (similar or identical to plitude but slower repetition rate. Clinically,
interictal spike and wave complexes) (Fig. this is accompanied by transition from the
24). In these cases, symptomatology is dom- tonic to the clonic phase. Occasionally, evo-
inated by alterations of consciousness (i.e., lution from one pattern to another is seen,
typical absences in patients with classical 3- the most typical from 3-Hz spike and wave
Hz spike and wave complexes and atypical complexes or polyspike and wave complexes
["bland"] absences in patients with slow spike to generalized tonic-clonic seizures (Fig. 26).
and wave complexes).
(b) Ictal pattern = intense interictal pattern. This II. Electroclinical correlations
is usually seen in myoclonic seizures. Low- In Part I on the neurophysiology of spike and
amplitude polyspike and wave complexes wave complexes, the spike was associated with
are usually asymptomatic whereas those of EPSPs and the following slow wave with IPSPs.
higher amplitude are accompanied by my- The spike component of the spike and wave
oclonic jerks. Their intensity is directly re- complex is asymptomatic when of relatively long
lated to amplitude and duration of the duration ("sharp wave") as in slow spike and wave
polyspikes. complexes. When the spike is of fast evolution
(c) Ictal pattern ^ interictal pattern. In these cases, (actual "spike"), it produces myoclonic jerks
the interictal patterns are isolated sharp whose intensity is related to the amplitude of the
waves, spikes, or spike and wave complexes, spike component. T h e typical example is the
but the ictus consists of or starts with an burst of 3-Hz spike and wave complexes accom-
electrodecremental and/or paroxysmal fast panied by myoclonic jerks of the eyelids at a
pattern (Fig. 25). Tonic seizures, akinetic similar repetition rate. With repetitive spikes oc-
seizures, and infantile spasms usually are as- curring in fast succession (up to 30-40 Hz), the
sociated with a five- to 10-second electro- intensity of the motor manifestation increases
decremental pattern with superimposed par- progessively. Short bursts of polyspikes (usually
oxysmal fast activity which not infrequently followed by prominent slow waves) tend to pro-
has progressively higher amplitude and duce violent generalized myoclonic jerks, e.g.,
slower frequencies. Generalized tonic-clonic bursts of polyspike and waves seen in myoclonic
seizures start in the same way during the epilepsy. Longer runs of spikes ("paroxysmal
Downloaded from www.ccjm.org on November 2, 2021. For personal use only. All other uses require permission.218 Cleveland Clinic Quarterly Vol. 51, No. 2
GENERALIZED TONIC-CLONIC SEIZURES
23 YRS OLD MAN
I NTERICTAL ICTAL
JzOO/iV h-p
Fig. 25. Twenty-three-year-old man with generalized tonic-clonic seizures and inter-
ictal epileptiform discharges on the left side. The middle tracing shows the start of a
seizure with an electrodecremental episode; the latter part of that tracing shows a
paroxysmal fast and/or muscle artifact which coincides clinically with the tonic phase.
The last trace on the right shows spike and wave complexes, associated clinically with the
clonic phase. Dz = between Fz and Cz (International System); Ez = between Cz and Pz.
ABSENCE AND GENERALIZED TONIC-CLONIC SEIZURE fast") tend to produce tonic seizures like the tonic
12 YRS HISTORY OF GENERALIZED TONIC-CLONIC SEIZURES
phase of generalized tonic-clonic seizures.
220 SECONDS AFTER START. PATIENT CONFUSED T h e slow wave of spike and wave complexes is
not accompanied by myoclonic or tonic motor
r,-C3 / Y l f si
r irti phenomena and actually tends to inhibit them.
P.-C Yf ÌY1YA/V In the evolution from generalized tonic to clonic
seizures, the spikes decrease progressively in rep-
etition rate and are followed by slow waves of
232 SECONDS AFTER START. PATIENT CONFUSED progressively longer duration and higher ampli-
FS
Fz " C z
fHVsTNfirJ ^ V K i r ^
VWhrV^VVI r\ V w
f\ tude. The motor manifestations strictly follow
the temporal evolution of the spike component
evolving from the tonic phase into a generalized
"tremor" when the repetition rate of the spikes
decreases to approximately 10 Hz, and then into
253 SECONDS AFTER START. GENERALIZED TONIC SEIZURE clonic seizures when spike and wave complexes
Fj-CJ .'"^»V^ V ^ W
at progressively slower repetition rate appear on
the EEG.
Fz-Fz T h e main clinical manifestation of bursts of
spike and wave complexes not accompanied by
309 SECONDS AFTER START. GENERALIZED CLONIC SEIZURE myoclonic phenomena (or in which the myoclonic
A— jerks are only a secondary symptom) is alteration
Fï-Cs
of consciousness, for example, the bursts of 3-Hz
-A spike and wave complexes typically associated
Fz-Cj
with absences. T h e degree of alteration of con-
sciousness is a complex function of type and
duration of burst and point in time within the
Fig. 26. Evolution from absence to generalized tonic-clonic burst. Three-hertz spike and wave complexes of
seizure. During the first 220 sec, the EEG showed extremely regular
generalized 2.5-Hz spike and wave completes; and the patient was
less than three seconds are usually not associated
extremely confused, but answered simple questions or commands. with clinical symptomatology. Bursts of more
At 232 sec, the spike and wave pattern became irregular, but no than five seconds, however, are usually mani-
clinical change was noticed. At 253 sec, the patient had a tonic fested clinically or subclinically. Reaction time is
seizure and the EEG showed an electrodecremental pattern with a prolonged in 43% of cases at the initiation of
superimposed paroxysmal fast and/or muscle artifact. At 309 sec,
the patient had generalized clonic seizures and the EEG showed
bursts of 3-Hz spike and wave complexes, in 80%
periodic sharp-wave transients. before 0.5 seconds, and in only 68% after four
Downloaded from www.ccjm.org on November 2, 2021. For personal use only. All other uses require permission.Summer 1984 Generalized epilepsies 219
seconds. 7 T h e progressive recovery of conscious- complexes. Bursts of similar morphology and du-
ness as the bursts become longer is clearly evident ration may show a variety of clinical manifesta-
in patients with absence status who, despite con- tions: (a) asymptomatic; (b) interference with
tinuous generalized spike and wave complexes, initiation of movement, but not with repetitive
usually show only mild alterations of conscious- movements; (c) interference with all voluntary
ness. Another example is the almost continuous movements; and/or (d) interference with mem-
slow spike and wave complexes (Lennox-Gastaut) ory. Bursts of longer duration, more widespread
associated with only very mild alteration of con- distribution, and higher amplitude are clearly
sciousness simulating mental retardation. Not in- correlated with the more severe clinical manifes-
frequently this is attributed to an underlying tation. For any given burst in any given patient,
static encephalopathy and only becomes evident however, no definite prediction can be made.
when the patient shows a clear recovery of intel-
lectual function following the disappearance of
the spike and wave complexes. Postictal EEG
Intimately related to the alteration of con- In generalized seizures associated with altera-
sciousness are automatisms. The occurrence of tions of consciousness only, the postictal EEG,
automatisms in absence seizures is directly related independent of the duration of the spell, is not
to the duration of the bursts of 3-Hz spike and changed significantly as compared with the preic-
wave complexes. Bursts of more than six seconds tal baseline EEG.
are accompanied by automatisms in 50% of these Generalized motor seizures are frequently as-
patients and bursts of more than 12 seconds in sociated with postictal EEG abnormalities. Both
90%. 18 T h e mechanism of generation of the au- degree and duration of these postictal changes
tomatisms is probably closely linked to the alter- are directly related to the duration and intensity
ation of consciousness and has no direct correla- of the motor manifestations. Isolated myoclonic
tion with myoclonic, clonic, or tonic motor man- jerks, even when violent, are not associated with
ifestations. Automatisms are well-coordinated, postictal abnormalities. Short, generalized, tonic
frequently repetitive motor manifestations for seizures are also frequently not followed by
which the patient is amnestic. They probably postictal changes. Generalized tonic-clonic sei-
represent a type of release phenomenon from zures, however, are almost always followed by
conscious control of motor activity and should be postictal abnormalities. Prolonged generalized
clearly differentiated from involuntary my- tonic-clonic seizures, particularly when accom-
oclonic, clonic, or tonic motor manifestations panied by severe motor manifestations, will be
produced directly by epileptogenic cortical activ- followed by marked postictal changes in the EEG-
ity. This differentiation is particularly important like diffuse flattening (less than 20 /uV delta EEG)
in patients with focal epilepsy in whom laterali- or burst-suppression pattern. Abnormal postictal
zation of focal myoclonic, clonic, or tonic motor EEG patterns occasionally persist for many hours.
activity is an extremely useful localizing sign. This is particularly true in patients who have had
Lateralization of an automatism is of absolutely a series of seizures or status epilepticus.
no localizing value. Postictal EEG changes show strict correlation
Alterations of consciousness in patients with 3- with clinical postictal symptomatology. Flat EEG,
Hz spike and wave complexes is produced, ac- burst-suppression pattern, and diffuse slow activ-
cording to the centrencephalic theory, by an ity in the delta range (replacing the alpha rhythm)
epileptic discharge in deep-seated subcortical are usually accompanied by clinical postictal
gray structures ("consciousness center"). Assum- coma. Diffuse theta rhythm (replacing the alpha
ing that generalized epilepsy is primarily a diffuse rhythm) will be associated with stupor. T h e
cortical disease, the best explanation for the al- postictal EEG pattern is very helpful in the dif-
teration of consciousness is the generalized cor- ferentiation of real and psychogenic seizures. Psy-
tical discharge which would interfere diffusely chogenic "spells" frequently consist of violent
with normal cortical functions. shaking of all extremities followed by unrespon-
T h e electroclinical correlations mentioned siveness. During the violent stage, the EEG is
above represent only general trends. T h e exact usually completely obscured by EMG and move-
clinical correlation of any given EEG pattern ment artifacts. During the following stage of "un-
cannot be predicted with certainty. T h e typical responsiveness," however, the EEG exhibits nor-
example is the burst of 3-Hz spike and wave mal alpha activity. In patients with real seizures,
Downloaded from www.ccjm.org on November 2, 2021. For personal use only. All other uses require permission.220 Cleveland Clinic Quarterly Vol. 51, No. 2
postictal unresponsiveness is always associated Table. Typical clinical correlates
with marked EEG slowing. EEC pattern Clinical correlate
The neurophysiological mechanism of postictal 3-Hz spike and wave complex Generalized absence
slowing has not been clearly elucidated. T h e most Slow spike and wave complex Minor motor seizures
probable explanation is neuronal "exhaustion" Hypsarrhythmia Infantile spasm
Multifocal independent spikes Minor motor seizures or infan-
resulting from imbalance of energy reserves, the tile spasms
demand for energy exceeding supply. This im- Polyspike and wave complex Myoclonic seizures
balance is the product of two factors:
(a) Neuronal activity is greatly increased with complexes in sleep. Morever, the repetition rate
a corresponding increase of energy de- of the so-called "3-Hz" spike and wave complexes
mand. only exceptionally is exactly 3 Hz. In the follow-
(b) Motor activity consumes a significant part ing section, the main characteristics of the pat-
of the energy supply and may interfere terns cited above are described in detail.
with respiration and thus restrict oxygen (a) Three-hertz spike and wave complexes. The
intake. definitive pattern consists of bursts of spike
T h e fact that postictal slow activity is seen and wave complexes lasting at least three
almost exclusively with generalized tonic-clonic seconds, at a repetition rate of approxi-
seizures is understandable if we consider the mately 3 Hz with a short spike component
above factors. Generalized tonic-clonic seizures (50-80 msec). Spike and wave complexes
exhibit the most intense epileptiform manifesta- tend to have a faster repetition rate at the
tions and the most violent motor activity. Both beginning (4-5 Hz), then stabilize at around
factors obviously contribute to the energy deficit. 3 Hz (always more than 2.5 Hz) during the
In patients with seizures not associated with mo- main part of the burst, finally slowing down
tor activity, the supply of energy is not affected. again at the end. T h e bursts have a clearly
T h e mechanism responsible for postictal slow defined onset and end abruptly. Frequently,
activity ("neuronal exhaustion"), however, is not shifting asymmetries occur at the beginning
necessarily responsible for terminating the sei- of the burst. T h e bursts occur most fre-
zure. Many seizures, e.g., absences associated quently in the waking and the REM sleep
with 3-Hz spike and wave complexes, do not stage and are activated by hyperventilation.
produce postictal slow activity, but tend to stop The most characteristic feature is the asso-
within 10-15 seconds. In these cases, an active ciation with clinical or subclinical unrespon-
inhibitory mechanism could be postulated. siveness. This is easily tested with a clicker
(see below). Bursts of spike and wave com-
Characteristic EEG patterns plexes with some atypical features should be
I. Typical clinical correlates thus classified as 3-Hz spike and wave com-
T h e following EEG patterns have definite clin- plexes only if associated with relative unre-
ical correlations: sponsiveness. Otherwise, they are simply
(a) 3-Hz spike and wave complex: generalized called generalized spike and wave com-
absence plexes. Conservative labeling of the 3-Hz
(b) Slow spike and wave complex: minor mo- spike and wave pattern will significantly in-
tor seizures crease its specific correlation with clinical
(c) Hypsarrhythmia: infantile spasm absences. It is important to remember that
(d) Multifocal independent spikes: minor mo- an EEG pattern classified as generalized
tor seizures or infantile spasms spike and wave complexes strongly supports
(e) Polyspike and wave complex: myoclonic the diagnosis of generalized epilepsy and
seizures. certainly is compatible with clinical absences
even if not specific for this type of epilepsy.
II. The concept of typical EEG patterns The classification of a record as "3 Hz
A real EEG pattern is obviously far more com- spike and wave complexes" depends on the
plex than the idealized waveforms described presence of the above described typical
above. For example, patients with 3-Hz spike and burst. Other types of epileptiform dis-
wave complexes may also show focal left and right charges, however, are seen almost always in
frontal spikes and bursts of polyspike and wave addition to the typical 3-Hz spike and wave
Downloaded from www.ccjm.org on November 2, 2021. For personal use only. All other uses require permission.Summer 1984 Generalized epilepsies 221
complexes (Fig. 27). This includes focal left ABSENCE SEIZURES
or right frontal spikes, atypical short bursts BURSTS OF POLYSPIKES DURING SLEEP
of spike and wave complexes, and irregular
bursts of polyspikes and waves. T h e last two
patterns are most frequent in slow-wave
sleep when the long bursts are replaced by
fVV^—
progressively more frequent bursts of
shorter duration, irregular repetition rate,
and the appearance of polyspikes. There are
exceptions to this rule, however, and occa-
sionally, bursts of 3-Hz spike and wave com-
plexes are seen exclusively during sleep.
With the epileptiform discharges, a change
in the stage of sleep, arousal, or apnea can
be observed, indicating that these bursts are
symptomatic. On the other hand, it is not
infrequent that a patient with absence sei-
zures may only show atypical bursts of gen-
eralized spike and wave complexes during
sleep. This often occurs when the patient is
clinically asymptomatic or exhibits only gen- 18 YRS, MALE
eralized tonic-clonic seizures at the time of
Fig. 27. Burst of polyspikes during sleep in a patient with
the test. generalized absence seizures
(b) Slow spike and wave complexes. T h e m o s t char-
acteristic and necessary pattern consists of Almost always, one sees other epilepti-
bursts of spike and wave complexes of at form discharges, such as generalized atypi-
least three seconds repeating at less than 2.5 cal spike and wave complexes, focal left or
Hz. The "spike" component of the spike and right frontal spikes, and occasionally, in
wave complexes is actually a sharp wave with more atypical locations, particularly in the
a relatively long duration of more than 80 occipital areas. Irregular, short bursts of
msec. Frequently, a significant proportion polyspike and wave complexes during sleep
of the record is occupied by slow sharp and are also noticed. A fairly characteristic fea-
wave complexes in bursts of variable dura- ture is the occurrence of electrodecremental
tion. Only rarely can one distinguish a dif- episodes, frequently with superimposed par-
ference in clinical behavior between the por- oxysmal fast activity. These episodes vary
tion of the EEG occupied by slow spike and between one and 10 seconds and are most
wave complexes and those free of epilepti- frequent during sleep. Longer episodes are
form discharges. T h e clicker test (see below) typically associated with either tonic, aki-
is usually unreliable because patients are netic, or atonic seizures or their combina-
uncooperative. Despite the apparent lack of tion.
clear correlation between bursts of slow (c) Hypsarrhythmia. T h e most characteristic and
spike and wave complexes and behavioral necessary pattern for this EEG classification
alterations, an impressive change in behav- consists of continuous multifocal independ-
ior is occasionally noticed on disappearance ent spikes superimposed on a slow back-
of the slow spike and wave pattern from the ground in the delta range which has an
EEG because of treatment or other reasons. average amplitude of at least 300 fx\. Dur-
Besides, in the occasional patient with infre- ing sleep, the epileptiform discharges are
quent but long bursts of slow spike and wave more prominent and tend to group in one-
complexes, a behavioral change can be cor- to 10-second bursts of slow waves with su-
related with each burst (relative unrespon- perimposed multifocal spikes. These bursts
siveness, complex automatisms). Frequently, are separated by relatively low amplitude
the background rhythm of patients with EEG segments. This pattern has been erro-
slow spike and wave complexes is relatively neously labelled "burst-suppression" (Fig.
slow for the age of the patient. 28), but the "suppression" consists of EEG
Downloaded from www.ccjm.org on November 2, 2021. For personal use only. All other uses require permission.222 Cleveland Clinic Quarterly Vol. 51, No. 2
"BURST-SUPPRESSION" AND INFANTILE SPASMS
dependent sharp waves of benign focal epi-
4 M O N T H S HISTORY O F INFANTILE SPASMS 6 M O N T H S , FEMALE
lepsy of childhood are as follows:
(1) T h e posterior background rhythm is
within normal limits for its age group.
(2) T h e epileptic discharges are sharp
waves (duration longer than 80 msec)
and not spikes.
(3) T h e sharp waves are followed by pre-
dominantly positive slow waves, with a
negative phase of usually lower ampli-
tude than the preceding sharp wave.
(4) T h e sharp waves are of extremely ster-
eotyped waveform, even if varied in
amplitude.
(5) Horizontal bipoles are characteristic,
even if not always present.
20CV V (6) T h e discharges are inhibited by mental
activation, hyperventilation, photic
Fig. 28. Patient with infantile spasms and a burst-suppression
pattern during sleep. Sleep spindles occurred preferentially during stimulation, and increase markedly
the so-called "suppression" phase. during sleep. Not infrequently, the pat-
tern is present only during sleep.
activity of relatively lower amplitude com- (7) They are not associated with atypical
pared with the high amplitude bursts. Ab- generalized spike and wave complexes.
solute measurement of amplitude fre- (8) Usually there are only three to four
quently reveals values between 20-70 ¿¿V, discrete foci which fire independently,
and one may see well-developed sleep spin- or one will trigger the other.
dles during these periods. Perhaps the (e) Polyspike and slow wave complex. Polyspike
French expression, trace paroxystique, better and slow wave complexes are frequently
describes this phenomenon. seen during sleep in association with any of
Infantile spasms are typically associated the above patterns. In addition, it is not
with five- to 10-second electrodecremental infrequent to see spike and wave complexes
patterns, which may be associated with low- with two or three spikes preceding the slow
amplitude paroxysmal beta or alpha waves, wave in patients who do not have myoclonic
particularly at the end. T h e pattern is fre- epilepsy. Therefore, this pattern is of very
quently inconspicuous and can easily be limited specificity and is of diagnostic value
missed. Careful clinical observation by the only in the awake tracing with a prominent
technologist and the corresponding nota- polyspike component.
tions on the record are extremely helpful
for correct identification of the ictal nature Special tests
of this pattern.
As in all other typical patterns, one may I. Sleep
see generalized spike and wave, or polyspike Sleep is an excellent activation procedure for
and wave complexes. interictal epileptiform discharges in all types of
(d) Multifocal independent spikes. This pattern generalized epilepsies. It is a natural activation
is very similar to hypsarrhythmia except that procedure with essentially no risk for the patient
the background is not in the delta range and no false positives. It is frequently said that
and/or is of less than 300 /¿V amplitude. hyperventilation is the best activator for 3-Hz
The ictal patterns are identical to slow spike spike and wave complexes. This is true if one is
and wave complexes, or hypsarrhythmia. It looking for typical 3-Hz spike and wave com-
is important to differentiate the multifocal plexes and/or absence seizures. In patients with
independent spike pattern from the multi- absence seizures for whom the routine awake
focal independent sharp waves frequently tracing including hyperventilation is negative,
seen in benign focal epilepsy of childhood. activation of atypical spike and wave complexes
T h e main characteristics of multifocal in- during sleep is not infrequent.
Downloaded from www.ccjm.org on November 2, 2021. For personal use only. All other uses require permission.Summer 1984 Generalized epilepsies 223
II. Photic Stimulation lapses of consciousness (Fig. 31). T h e tech-
nologist pushes a button which produces a
Photic stimulation is a relatively selective acti-
clearly audible click, and the patient is asked
vator of generalized epileptiform discharges. Fo-
to push another button as fast as possible
cal epileptiform discharges (including occipital)
when he hears the click. Both push buttons
are usually inhibited and rarely triggered by
are monitored on either the same or an
photic stimulation.
independent EEG channel. During hyper-
In photic stimulation, the most frequently ac-
ventilation, the responsiveness of the patient
tivated seizures are generalized myoclonic jerks,
is tested every 10-20 seconds. When a burst
clinical absences, or generalized tonic-clonic sei-
of spike and wave complexes occurs, the
zures. It is important to differentiate between
technologist pushes the button as soon as
photomyoclonic response (exaggerated physio-
possible. For short bursts, the patient can
logic myoclonus elicited by intermittent photic
also be asked to push the button continu-
stimulation) which has no diagnostic significance
ously. Symptomatic bursts are clearly iden-
and the photoparoxysmal patterns characteristic
tified because the patient will stop pushing
of generalized epilepsies.
the button. Some patients continue pushing
In a small subgroup of patients with photopa- the button repeatedly during bursts of spike
roxysmal responses, seizures will be triggered and wave complexes, but are unable to push
only by intermittent photic stimulation (photo- the button in response to random clicks pro-
sensitive patients). In this group, the use of lenses duced by the technologist. This is because
can prevent the seizures. Lenses can be tested in automatic activity tends to be less affected
the EEG lab by studying their effect on photo- than voluntary initiation of movement dur-
paroxysmal response. However, photoparoxys- ing bursts of 3-Hz spike and wave com-
mal responses may fluctuate considerably over plexes. Other patients will continue respond-
time, and therefore, reproducibility of results is ing to the clicker, but with a certain delay.
essential to reach any conclusion. Patients with slow spike and wave complexes
Some photosensitive patients are pattern-sen- frequently show no alteration of responsive-
sitive, i.e., photoparoxysmal EEG responses and ness in association with bursts of epilepti-
even seizures can be activated by a special geo- form activity (Fig. 32).
metric pattern. T o find it, the patient scans dif-
ferent patterns under strong illumination under It is very important to remember that
EEG monitoring, particularly from the occipital hyperventilation normally activates bursts of
region. slow waves which not infrequently have as-
sociated sharp transients. These bursts result
from respiratory alkalosis with secondary
III. Hyperventilation relative brain anoxia. Frequently, the pa-
Hyperventilation is an excellent activator of 3- tient also has paresthesia of the distal ex-
Hz spike and wave complexes, but may also be tremities and lips, dizziness, and occasion-
effective in other types of generalized epilepti- ally, slight alteration of consciousness with
form discharges. It is extremely useful to deter- unresponsiveness and amnesia for words he
mine whether bursts of spike and wave complexes had been asked to remember during the
are clinically symptomatic. There are two simple bursts of slow waves. T h e patient may stop
methods to objectively demonstrate whether a hyperventilating and become unresponsive
burst of spike and wave complexes is associated to clicks. This is greatly activated by relative
with absences: hypoglycemia and may disappear completely
(a) Respiration monitoring-. Most bursts of spike if sugar is given. This phenomenon is infre-
and wave complexes associated with unre- quent, but indicates that unresponsiveness
sponsiveness will be accompanied by a mo- during a burst activated during hyperventi-
mentary interruption of hyperventilation by lation is not sufficient proof of its epilepto-
a period of apnea. This can be easily dem- genic character.
onstrated by a respiratory monitor (Fig. 29);
however, spontaneous respiration is fre- IV. Evoked potentials
quently not altered during absence seizures Giant visual and somatosensory evoked poten-
(Fig. 30). tials have been described in patients with my-
(b) Clicker: This is a simple device to detect short oclonic epilepsy and/or photosensitivity. T h e am-
Downloaded from www.ccjm.org on November 2, 2021. For personal use only. All other uses require permission.You can also read
