Yawning
and seizures
The usual clinical symptoms of infantile
spasms have been well characterized in earlier
studies by analysis of
video-electroencephalographic (EEG) data.
Typical infantile spasms involve abrupt flexion
or extension movements of the extremities,
trunk, and neck. Contractions of the facial
muscles, eye opening or closure, upward eye
rolling, eyelid fluttering, or nystagmoid eye
movements may be associated.Large series and
review articles conceming infantile spasin
patients have mentioned that fragmentary
variants of these seizures may occur in some
infantile spasin patients. Previous authors have
emphasized that these seizures may be extremely
subtle and may be missed during history taking
and even during video-EEG evaluation, unless
specifically looked for. Infantile spasm
fragments that have been documented include head
nodding, shoulder shrugging, and abdominal
contractions. Infantile spasms
characteristically occur in a series that
crescendos and then decrescendos, and these
fragmentary forms may occur at the onset and
offset of the series.
Points that are important for the clinical
diagnosis of infantile spasms but that have not
been clarified in earlier reports include
whether fragmentary forms occur in a
series without typical infantile spasms and
whether fragmentary forms occur as the only form
of infantile spasms in some patients. Nodding
spasms have been reported to occur in series.
Holmes stated that eye movements could occur
independently from infantile spasms and precede
the development of infantile spasms by weeks.
Whether the occurrence of fragmentary forms has
any usefulness in predicting prognosis or
outcome of treatment has not been commented on
previously, except by Lombroso, who said that
their significance was uncertain.
In this report, we describe fragmentary
variants of infantile spasms that have not
previously been reported, we describe their
occurrence as series of seizures that did not
include typical infantile spasms, and we
describe their occurrence both in patients who
did and who did not also have typical infantile
spasms. We also discuss the physical
similarities and differences between fragmentary
and typical infantile spasms in patients who had
both, recount when during the history of
infantile spasms the variants were recorded, and
describe how the variants in some patients
appear to be modifications that result froin the
effects of time or treatment on typical
infantile spasms. [...]
Infantile Spasm Variant Found on Later
EEG
Four patients had typical infantile spasms
present on the first video-EEG evaluation and
the variants present on video-EEGs performed 1
day (patient 9) or 1 to 3 months later (patients
8, 10, and 11). Patient 8 had both typical and
variant infantile spasms present on the later
EEG. In this patient, the variants, which
consisted of a mouth opening or yawning
movement, did not appear to be fragments of
the patient's typical flexor infantfle spasms.
The ictal EEG changes also differed: the typical
infantile spasms were each associated with a
generalized slow-wave transient, but the
variants were each associated with a generalized
decrement. Both types of infantile spasms
responded to an increase in dose of ACTH.
Patient 9 had the variants (facial grimace) but
no longer had typical infantile spasms on a
video-EEG performed for the purpose of a
pyridoxine injection 1 day after the first EEG
and after two doses of ACTH; the parents of this
patient noted the grimace to be part of the
generalized infantile spasms, which had
disappeared. These episodes subsequently
stopped. Patient 10, who had a neonatal cerebral
infarct and asymmetrical hypsarrhythmia, had
asymmetrical extensor infantile spasms that
included a lateral head jerk and facial grimace
on the first video-EEG. Variants that consisted
of a lateral head jerk and grimace were recorded
on video-EEG 1 month later, after treatment with
valproic acid, and variants that consisted only
of the grimace were recorded on video-EEG 2
weeks after that (Figure 5). At this time,
clonazeparn was added and the seizures stopped.
13 Patient 11 had resolution'-of hypsarrhythmia
and typical infantile spasms documented on a
video-EEG performed 1 month after beginning ACTH
treatment. However, ACTH was stopped 6 weeks
later because of side effects. Two weeks later,
variants (eye opening) were recorded on videoEEG
at a tirne when the EEG still showed no
interictal epileptiform activity and the parents
had noticed no seizures. Treatment with valproic
acid was begun, but typical infantile spasms and
hypsarrhythmia were again recorded on an EEG 2
months later. Clonazeparn was added and the
seizures resolved.
Discussion
In most of our patients, the infantile spasm
variants appeared to be fragments or limited
expressions of more typical infantile spasms. In
some of these patients, the variants coexisted
with typical infantile spasms and probably
represented natural variability in physical
expression of infantile spasms. However, in some
patients, the variants may have resulted from a
change in level of consciousness, time, or
medication that modified the clinical expression
of the infantile spasms. The effect of level of
consciousness was seen in patient 6, who had
typical infantile spasms in the waking state but
had variants during sleep that consisted of only
part of the typical infantile spasm. Time
appeared to modify typical infantile spasms in
patient 4, who, by history, had generalized
infantile spasrns that resolved spontaneously 2
weeks before our video-EEG demonstrated the
variants, and by patient 5, who developed
typical infantile spasms 3 weeks after the
variants were recorded. These spontaneous
changes suggest that the variants were a stage
in the natural evolution of the infantile
spasms.
Of particular interest are the two patients
whose variants appeared to be modifications that
resulted from treatment of typical infantile
spasms. Increasingly restricted variant forms
were recorded in patient 10 on two occasions
during treatment before the seizures ceased. In
patient 11, the variants were recorded on a
video-EEG that was obtained after ACTH given for
typical infantile spasms was stopped because of
side effects. This was at a time when the
parents thought the seizures were gone and
before the interictal EEG showed a recurrence of
the hypsarrhythmia pattern. These variants may
have been an undetected residual type of
infantile spasms that persisted after treatment
with ACTH, or they may have heralded the
recurrence of infantile spasms after ACTH was
stopped. Patient 7 demonstrated a possible
difference in therapeutic response between
variants and typical infantile spasms: both
types of infantile spasms initially coexisted,
but onlv the variants persisted after treatment
with clonazepam. The variants then resolved
after treatment with ACTH.
In other patients, the infantile spasm
variants appeared to be separate types of
infantile spasms that were not fragments of
typical infantile spasms. In patient 8, the
infantile spasm variants appeared to be a type
of infantile spasms that was different both
clinically and electrographically from the saine
patient's typical infantile spasms, although
both types responded to the saine treatment. In
patients 1, 2, and 3, the infantile spasm
variants occurred with no previous or subsequent
history of typical infantile spasms;
unfortunately, due to death or lack of
follow-up, video-EEG documentation of the
subsequent course was not possible. Also, these
patients were treated for infantile spasms after
the videc>EEGs demonstrated the seizures to
be infantile spasm variants, and therefore the
natural historv was altered. The seizures in
these patients were diagnostic dilemmas because
the clinical descriptions of the seizures did
not fit recognized seizure types.
