вход по аккаунту


Precise characterization and quantification of infantile spasms.

код для вставкиСкачать
Precise Characterization and
Quantification of Infantile Spasms
Peter Kellaway, P h D , Richard A. Hrachovy, MD,
James D. Frost, Jr, MD, and Thomas Zion, MD
With the use of a time-synchronized video and polygraphic recording system, 5,042 infantile spasms were monitored and analyzed in 24 infants aged 1 to 43 months. Of these, 33.9% were flexor, 22.5% extensor, and 42.0% mixed
flexor-extensor. Sometimes t h e spasms were followed by a period of akinesia and diminished responsiveness lasting
up ro 90 seconds, and rarely (1.0%) this “arrest” effect constituted the entire seizure. More than one type of seizure
occurred in 2 1 of the 24 infants. I n t h e same number, 78.3% of t h e seizures occurred in clusters, and the intensity
and frequency of the spasms in each cluster often increased to a peak, then progressively decreased until they
stopped. Predominantly, the clusters occurred soon after arousal from sleep. T h e number of seizures occurring at
night (55.2%) was similar to the diurnal number (44.8%). T h e electroencephalographic seizure pattern was variable, but a marked generalized attenuation of electrical activity was a feature of 71.7% of the attacks. Attenuation
episodes of similar degree and duration occurred with no evidence of a seizure.
Kellawap P, Hrachovy RA, Frost JD Jr, et al: Precise characterization and quantlfication of infantile spasms
Ann Neurol 6:2 14-2 18, 1979
Infantile spasms, which are variously referred to in
the literature as massive spasms, flexion spasms,
jackknife seizures, infantile myoclonic seizures, and
in similar terms, have been recognized as epileptic
phenomena since they were first described by West
in 1841 181. Since that time a considerable literature
has accumulated worldwide 12-4, 71, but no attempt
has been made until recently to utilize modern techniques of prolonged polygraphic and video recording
t o make a precise characterization of the seizures
and associated pathophysiological changes. Precise
characterization and quantitative analysis of the
motor, autonomic, and behavioral phenomena associated with infantile spasms and their temporal and
dynamic relationships to changes in brain electrical
activity and to the sleepiwakefulness cycle require
concurrent recording, measurement, and correlation
of a wide range of physiological variables. Based o n
many years of clinical observation [ 5 , 61, w e have
developed an effective, time-synchronized monitoring
system for the study of this and other types of infantile seizures 111. W e report here the results of a study
of 24 infants with infantile spasms, using this system.
Twenty-four infants, aged 1 to 43 months, who were determined by clinical and electrographic study to be ex-
From r h e Departments of Neurology (Neurophysiology) and
Pediatrics, Bay lor College of Medicine, and the Neurophysiology
Service and the Blue Bird Clinic, The Methodist Hospital, Houston, TX.
periencing infantile spasms, were selected for the study.
Informed consent was obtained from each infanr’s parent
or guardian following a full explanation of the procedures.
Each infant was monitored continuously for 24 hours in
one ro eight different sessions. The monitoring system has
been described in detail elsewhere 111. Graphically recorded, concurrent, time-synchronized data included 6
channels of the electroencephalogram (EEG); 7 channels of
electromyograms (EMG) of flexor and extensor muscles of
the trunk, neck, and extremities; 4 triaxial vector accelerometer channels; and 1 channel each of respiration,
electrocardiogram, eye movement (electrooculogram), and
galvanic skin response. Preselected channels were recorded
on a 14-channel analog tape recorder, and all 2 1 data channels plus time information were recorded graphically using
a type R Beckman-Offner polygraph. A split-screen device
and twin-camera system permitted side-by-side videotaping
of the patient and 4 preselected channels of the graphic
record utilizing an NEC extended-time video cassette recorder. An audio record of the infant’s vocalizations was
made on the videotape. A continuous log was kept by a
trained observerhechnician who was continuously in attendance.
After each 24-hour session, the graphic record was
analyzed visually by at least two of us in reiation to the
observer’s log and review of the videotape. Unlimited replay of the videotape in real-time, slow-motion, o r
accelerated-time modes facilitated analysis and correlation
of the behavioral phenomena in terms of time-synchronized events in all data channels.
Accepted for publication Mar 12, 1979
d reprintd requests
~ to ~D~ ~ ~ l l~ ~~ p a~r t am e~of
n tNeural,
ogg (Neurophysiologg), Baylor College of Medicine, 1200 Moursund Ave, Houston, TX 77030.
0364-5 134/79/090214-05$01.25 @ 1078 by Peter Kellaway
A total of 5,042 seizures were monitored. The spasm
was a brief contraction involving muscles of the neck,
trunk, and extremities, usually bilaterally and symmetrically. The character of the seizure depended on
whether the extensor o r flexor muscles were predominantly affected and on the number and distribution of the muscle groups involved.
The muscle activity in an infantile spasm as revealed by the EMG consisted of two phases: an initial
phasic contraction lasting less than 2 seconds, followed by a less intense tonic contraction lasting from
2 to about 10 seconds. Sometimes the tonic phase
was not present, and the attack consisted of a phasic
muscular contraction lasting less than 0.5 second.
Seizure Types
The types of seizures recorded are summarized in
Table 1. Flexor spasms accounted for 33.9% of all
seizures observed. The full-fledged flexor spasm consisted of flexion of the neck, trunk, arms, and legs;
the contraction of abdominal muscles was sometimes
so massive as to cause the torso to jackknife at the
waist. Spasm of the muscles of the upper limbs resulted either in adduction of the arms in a selfhugging motion or in abduction of the arms to either
side of the head, with the arms flexed at the elbow.
The flexor spasm varied in the force of contraction
of the muscles involved. Also, the degree and extent
of muscle involvement in a given attack varied from
brief contraction of the abdominal rectus muscles to
total involvement of the axial muscles of the neck and
trunk and the flexors of the extremities. Some contractions were confined to the muscles of the neck,
causing only bobbing of the head. Similarly, some
spasms involved only the upper-limb girdle, so that
the seizure consisted of elevation and adduction of
the shoulders in a shruglike movement.
Extensor spasms accounted for 22.5% of the observed seizures. These were consequent to a predominance of extensor muscle contractions, producing an abrupt extension of the neck and trunk with
extensor abduction or adduction of the arms, legs, o r
Mixed flexor-extensor spasms included flexion of the
neck, trunk, and arms and extension of the legs, or,
less commonly, flexion of the legs and extension of
the arms with varying degrees of flexion of the neck
and trunk. Seizures of this type accounted for 42.0%
of the total and were the most common of all seizures
Asymmetrical infantile spasms were recorded in 1
infant and accounted for 0.6% of the total. This patient often assumed and maintained an abnormal
posture characterized as a “fencing” position. When
in this position, his seizures involved transient in-
Table I . Distribution of Infantile Spasms
by Type (N = 5,0421
No. of
Table 2. Distribution of 24 Patients
by Type of Seizures
Seizure TvDe
No. of
Mixed and flexor
Mixed and extensor
Mixed, flexor, and extensor
Flexor and extensor
Flexor, extensor, and arrest
Mixed, extensor, flexor, and
Mixed, flexor, extensor, and arrest
Mixed. extensor. and arrest
tensification of contraction of the flexor muscles already flexed and of the extensor muscles already extended.
Akinesia and attenuated responsiveness sometimes
followed the spasm and lasted as long as 90 seconds.
This phenomenon, which we have termed “arrest,”
also occurred independently, as a seizure, without a
preceding motor spasm; such seizures occurred in 3
patients and accounted for 1.0% of all seizures monitored.
Twenty-one of the 2 4 infants experienced more
than one seizure type (Table 2). The most common constellation consisted of flexor and mixed
extensor-flexor seizures and was seen in 9 of the 2 4
patients. Three patients experienced only mixed
flexor-extensor spasms.
Duration of Seizures
The precise duration of the seizures was difficult to
measure because in many instances the exact point of
termination could not be determined. Attacks consisting only of a phasic muscle contraction could be
measured precisely, but those involving akinesia and
diminished responsiveness did not have an exact
end-point. Clear persistence of the latter up to 90
seconds was documented, and persisting EEG
changes lasting up to 106 seconds were recorded.
Kellaway et al: Character of Infantile Spasms
Manner of Occurrence
In 2 1 of the 24 patients the spasms often occurred in
clusters of two or more in succession; 78.3% of all
seizures occurred in such clusters. The maximum
number of spasms in a cluster was 125; the maximum
spasm frequency was 1 3 per minute.
The intensity and frequency of the spasms in each
cluster often increased progressively to a peak, then
declined until the spasms ceased. Predominantly, the
clusters occurred soon after arousal from sleep. Increase in seizure incidence or clustering of seizures
was not a feature of falling asleep.
Only 2.5% of the seizures occurred while the infants were asleep, but, because of frequent awakenings, the number of seizures occurring at night
(55.2%) was similar to the diurnal number (44.8%).
Table 3. lctal E E G Patterns Associated
with Infantile Spasms ( N = 2,042)
No. of
Ictal Pattern
60 1
37 9
17 4
13 2
11 9
0 06
0 06
Precipitating Factors
Feeding, handling, and sudden stimuli such as a loud
sound have been reported to precipitate seizures, but
this was not evident in our patient series. Photic
stimulation did not precipitate seizures or produce
detectable EEG changes.
A t tenua t io n of Consciousness
Impaired consciousness is a difficult factor to assess
in infants and is impossible to evaluate in spasms of a
very brief duration. No objective measurements
were made in the longer seizures, but there was an
apparent attenuation of responsiveness as&ate;l
with arrest of activity.
Associated lctal Phenomena
Eye movements, consisting of deviation alone or
followed by rhythmic nystagmoid movements, accompanied the muscle contractions of the spasms in
2,767 (54.9%) of the seizures.
Changes in respiratory rhythm occurred in 2,987
(57.2%) of the seizures, and alteration of heart rate
in 3 3 spasms (0.6%). Although crying frequently
followed an attack, a cry or scream did not occur as an
ictal phenomenon in the 5,042 seizures monitored.
lctal EEG Patterns
Eleven different types of ictal EEG pattern were observed to accompany the clinical seizures (Table 3):
(1) a high-voltage, frontal-dominant, generalized
slow-wave transient followed by voltage attenuation
(electrodecremental episode), in 37.9% of all seizures; (2) a generalized sharp- and slow-wave complex, in 17.4% of the seizures; (3) a generalized
sharp- and slow-wave discharge followed by an attenuation period, in 13.2% of the spasms; (4) an attenuation episode only, in 11.9% of the seizures; (5)
a generalized slow-wave transient only, in 10.9% of
the seizures; (6) attenuation with superimposed fast
activity, in 6.9%; (7) a generalized slow-wave dis-
216 Annals of Neurology
Vol 6 No 3
= slow wave, Atten = reduction in ampllirude of background
activity, SSW = sharp and slow wave, Fast = i~ncreasedfast activity,
RS = rhythmic slow activity
charge followed by attenuation with superimposed
fast activity, in 1.3%; (8) attenuation and rhythmic
slow activity, in 0.2%; (9) fast activity only, in 0.2%;
(10) a sharp- and slow-wave complex followed by
attenuation and superimposed fast activity, in 0.06%;
and (11) attenuation with superimposed fast activity
followed by rhythmic slow activity, in 0.06%.
Although the character of the ictal EEG events did
not show a close correlation with the type of spasm
(Table 4), each of the three most common typesflexor, extensor, or mixed-was most frequently associated with a high-voltage slow-wave transient followed by attenuation of activity (SW/Atten). “Arrest”
attacks and asymmetrical seizures typically were accompanied by attenuation and increased fast activity
The duration of the ictal EEG pattern ranged from
0.5 to 106.0 seconds. T h e longer epidodes were
often associated with occurrence of the “arrest” phenomenon.
Episodes of generalized attenuation (electrodecremental episodes) might occur in the awake
or asleep state without evidence of a clinical seizure
or any recorded change in the EMG o r other polygraphic data channels.
Following the seizure event, both clinical and
electrographic, the EEG sometimes showed a variable period in which there was a marked decrease of
abnormal activity and the background activity more
closely approximated normal for age. No consistent
clinical changes were evident during these periods.
O n e of the most striking phenomena demonstrated
by this study was the broad spectrum of seizure intensity. Seizures ranged from a massive contraction
September 1979
Table 4 . Ictal EEG Patterns Associated with Seizure Types
Seizure Type (%)
(N = 1,708)
(N = 1,136)
- Pattern
5 WIAtten
S SWIAttenlFast
S WlAttenlFast
- tten/RS
SSW = sharp and slow wave; SW
rhythmic slow activity.
(N = 2,117)
(N = 29)
(N = 52)
slow wave; Atten = reduction in amplitude of background activity; Fast
of flexor and extensor muscles of the neck, trunk,
and extremities to an evanescent contraction confined to the rectus abdominis muscles. These lesser
seizure phenomena could not have been detected
rn ithout the combined video image, displayed in
real-time, slow-motion, and accelerated-time modes,
and synchronized data from the EMG, accelerometer, and EEG channels. Numerous replays (polygraphic or video or both) were often required to
establish the character of certain events. After reviewing many hundreds of ictal and nonictal events, we
found that our ability to distinguish fragmentary seiziires, and, indeed, seizures of a more fully developed nature, improved to such a degree that it
became necessary to review the video and polygraphic recordings of all the patients monitored earlier.
These two factors-the existence of fragmentary
or lesser seizures and the difficulties involved in distinguishing, by observation alone, between clinical
scizures and the random and often anarchic movements of infants with disordered brain functionrender it highly unlikely that reports by parents concerning the number or severity of seizures have any
clinical, much less research, utility.
The fact that the majority of seizures occur in
clusters that tend to begin on arousal from sleep constitutes another hazard for the assessment of seizure
ir cidence when using parents’ reports or, for that
matter, the reports of trained observers. Commonly,
o ir infants were observed to arouse from sleep, have
a cluster of a few to many seizures, then begin to cry.
Thus, by the time a parent might attend the child, the
seizures could be completed and not recur until
arousal following another period of sleep.
= increased
fast activity; RS =
It was our experience in this study that the reports
of parents concerning the number of seizures were
consistently discrepant with the monitored number,
and this discrepancy was usually in terms of several
hundred percent. Thus, it would appear that all reports in the literature concerning the effects of anticonvulsant drugs, corticotropin, or corticosteroid
treatment of infantile spasms must be regarded as
suspect until controlled monitoring studies are carried out.
The predominant age of onset of infantile spasms is
generally considered to be between 1 and 12 months,
with a peak incidence between 4 and 6 months. But
the problem of distinguishing seizures-particularly
the lesser or fragmentary types-from normal infant
behavior makes it possible that the age of onset in
many cases is earlier than suspected.
Because flexor spasms may be quite intense, it is
generally held that such attacks are the most common
type of infantile spasm [ 4 ] ;but, in our monitored
patients, mixed flexor-extensor spasms were more
common, accounting for 42.0% of all attacks and occurring in 2 2 of the 2 4 infants studied. It has been
suggested that if the attacks are not predominantly
flexor in type, there should be doubt that they are
true infantile spasms. While it is a fact that none of
our subjects had extensor seizures only, they accounted for 1,136, or 22.5%, of all seizures recorded. This type of infantile spasm can be distinguished from other types of tonic epileptic seizures,
and from nonepileptic brainstem release phenomena
seen in children with cortical depression, by the
coexistence of mixed or other types of infantile
spasms and by the characteristic ictal and interictal
EEG patterns.
Kellaway et al: Character of Infantile Spasms 217
Supported in part by Contract N01-NS-6-2342 and Grant NS
11535 from the National Institute of Neurological and Communicative Disorders and Stroke, National Institutes of Health,
US Public Health Service.
Frost JD Jr, Hrachovy RA, Kellaway P, et al: Quantitative analysis and characterization of infantile spasms. Epilepsia
19:273-282, 1978
Gastaut H , Soulayrol R, Roger J, et al: LEncephalopathie
myoclonique infantile avec hypsarythmie (syndrome de West).
Paris, Masson, 1964
218 Annals of Neurology Vol 6 No 3
3. Jeavons PM, Bower BD: Infantile Spasms: A Review of the
Literature and a Study of 112 Cases (Clinics in Developmental
Medicine No. 15). London, Spastics Society and Heinemann,
4. Jeavons PM, Bower BD: Infantile spasms, in Vinken PJ, Bruyn
G W (eds): Handbook of Clinical Neurology. New York,
American Elsevier, 1974, vol 15, pp 219-234
5. Kellaway P: Myoclonic phenomena in infants. Electroencephalogr Clin Neurophysiol 4:243, 1952
6. Kellaway P: Neurologic status of patients with hypsarhythmia,
in Gibbs FA (ed): Molecules and Mental Health. Philadelphia,
Lippincott, 1959, pp 134-149
7 . Lacy JR, Penry JK: Infantile Spasms. New York, Raven, 1976
8. West WJ: On a peculiar form of infantile convulsions. Lancet
1:724-725, 1841
September 1979
Без категории
Размер файла
421 Кб
spasm, precise, infantile, characterization, quantification
Пожаловаться на содержимое документа