Hyperventilation-induced
high-amplitude rhythmic slowing with altered
awareness: a video-EEG comparison with absence
seizures
Leanna M. Lum,
Mary B. Connolly, Kevin Farrell,
Peter K. H. Wong
Purpose: Hyperventilation-induced
high-amplitude rhythmic slowing (HIHARS) in
children may be associated with clinical
episodes of altered awareness. The presence of
automatisms has been proposed as a
distinguishing feature that helps to
differentiate absence seizures from nonepileptic
causes of decreased responsiveness. This
retrospective, controlled, video-EEG study
compared the clinical characteristics of
episodes of HIHARS with loss of awareness with
those of absence seizures.
Methods: The database of a tertiary
Children's Hospital was searched for patients
studied between April 1993 and April 1997 who
had at least one episode of HIHARS with loss of
awareness. The absence control group was
obtained by selecting the next patient, after an
HIHARS study subject, who met the following
criteria: (a) had at least one absence seizure
occurred during hyperventilation in the EEG
recording, and (b) had a diagnosis of idiopathic
generalized epilepsy. The video-EEG and medical
histories of all patients were reviewed and
summarized.
Results: We reviewed video-EEG
recordings of 77 episodes of HIHARS with loss of
awareness from 22 children and 107 absence
seizures during hyperventilation from 22
children. Eye opening and eyelid flutter were
seen more frequently in absence seizures,
whereas fidgeting, smiling, and yawning occurred
more frequently during HIHARS episodes. Arrest
of activity, staring, and oral and manual
automatisms were observed in both groups.
Conclusions: Automatisms are common
in both HIHARS and absence seizures. Yawning,
smiling, and particularly fidgeting occur more
commonly and eye opening and eyelid flutter less
commonly in HIHARS. However, episodes of HIHARS
with loss of awareness clinically mimic absence
seizures, and these conditions can be
distinguished reliably only by EEG.
DISCUSSION
Reduction in consciousness during voluntary
HV was first reported by Davis and Davis (10) in
1939, when failure to respond to commands was
observed with 2- to 3-Hz high-voltage sinusoidal
activity. In 1947, Engel (1) described loss of
consciousness and arrest of HV, with resumption
of HV as the level of consciousness returned in
six of nine patients. Consciousness was tested
during the last 30 s of a 3-min HV period by
presenting the patient with three objects to
remember and with a multiplication problem. The
reduction of consciousness correlated with the
degree of EEG slowing and was most marked when
frequency was < 5 Hz.
The first description of episodes of HIHARS
with loss of awareness was in an abstract
entitled "Pseudoabsences" (6). Three children
were described with "extremely unusual"
stereotyped attacks during HV, and the authors
considered that these were nonepileptic events.
Subsequent reports of episodic, altered
awareness during HIHARS speculated on whether
these episodes are epileptic seizures (3-5) or
nonepileptic phenomena (2,6,7). In a study of 12
children with the EEG criteria for HIHARS,
verbal recall and motor responsiveness to an
auditory click were normal at baseline and
during HV before slowing, but all children
exhibited impaired verbal recall, and eight of
12 failed to respond to auditory clicks during
an episode of HIHARS (2). The authors concluded
that responsiveness could be impaired in healthy
children without epilepsy. In that study,
automatisms were not observed during the
episodes of HIHARS, and the authors concluded
that the presence of automatisms helped to
differentiate absence seizures from nonepilepsy
causes of decreased responsiveness.
The present study is the first to use
video-EEG to analyze the clinical features of
the episodes of HIHARS and to compare the data
with those from a control population of children
with absence seizures. Unlike Epstein et al.
(2), we observed automatisms in all of our 22
patients. Automatism also were described in most
other reports of HIHARS and altered awareness
(3-7). Although automatisms were common in both
the HIHARS and the absence control group, eye
opening and eyelid flutter were seen more
frequently in absence seizures. In contrast,
fidgeting, smiling, and yawning occurred more
frequently during HIHARS episodes. Yawning (3,6)
and smiling (6) have been described previously
as automatisms in HIHARS patients. Indeed, in
his initial description of altered awareness
during 11V, Engel (1) described excessive
sighing and yawning. Arrest of activity,
staring, and automatisms, both oral and manual,
were observed in both groups. Although there are
some differences in the patterns of automatisms
between HIHARS and absence seizures, only
fidgeting occurred exclusively in the HIHARS
group, and the distinction between HIHARS and
absence seizures can be made reliably only by
using EEG.
The design of our study does not permit us
to determine whether HIHARS is an epileptic
seizure or a nonepileptic phenomenon. HIHARS has
been described largely in children with
neurologic symptoms or disease. A history of
previous seizures was obtained in 12 of our 22
children, and many of the others had headache,
school difficulties, or problems with
concentration. This is similar to results in
previous studies (2,7,15) and reflects very
probably an ascertainment bias. Epileptic
seizures are characterized by the co-occurrence
of ictal EEG activity and behavioral change. The
absence of spikes in episodes of HIHARS supports
the view that the EEG abnormality is not
epileptiform. However, rhythmic slow-wave
activity without spikes has been described as an
ictal EEG feature in nonconvulsive status (16).
HIHARS with loss of clinical awareness is
clearly associated with behavioral change, and
many of the stereotyped features, including the
automatisms, have been described in epileptic
seizures. However, there have been no systematic
studies of HV in normal children with video-EEG,
and altered awareness during HV appears to be
not uncommon in children (1). Despite careful
clinical testing, altered awareness was detected
in only 77 of the 233 episodes of HIHARS
recorded in 22 patients in this study. It is
likely that a much higher incidence of altered
awareness would have been detected if the
detailed testing described by Epstein et al. (2)
had been performed. The occurrence of
stereotyped behaviors, including automatisms,
has been reported to relate to the length of the
HIHARS discharge (3). We believe that the
episodes described by Epstein et al. (2)
represent one end of a
spectrum of severity of clinical change
associated with HIHARS. The patients described
in this study are those that could be detected
by clinical testing. The degree of delta
activity in HV in childhood is related to age,
and the most pronounced responses occur in
children aged 8 to 12 years (14). The age range
of our patients with HIHARS (median, 9 years)
was similar. This would be consistent with the
concept that MORS with altered awareness
represents one end of a spectrum of altered
responsiveness in patients undergoing HV.
We believe that the incidence of HIHARS with
altered awareness is affected by both awareness
of the phenomenon and the rigor of the technical
testing of awareness. Although it occurred in
only 2.2 per thousand recordings overall during
this study, the incidence increased with time or
experience or both and had increased to 5.5 per
thousand recordings in the last year of the
study; the incidence in our laboratory over the
past 12-month period is 7.4 per thousand
recordings. In our study, HIHARS with altered
awareness was observed only during HV. This is
similar to nearly all other reports, and we are
aware of only one description of altered
awareness associated with the EEG features of
HIHARS occurring outside HV (3). There is a
temporal distinction between episodes of HIHARS
with altered awareness and those of absence
seizures. The duration of HV before to the onset
of HIHARS with altered awareness was almost
double that before the onset of an absence
seizure in the control group, an observation
made also by one other group (7). True absence
seizures manifested early after the onset of HV,
whereas HIHARS with alterations of awareness was
never seen in the first 30 s of HV and appeared
with a mean latency of >2 min. There also is
a dissociation with the electrographic entity
and the clinical alteration of awareness.
Electrographically HIHARS appeared with a mean
latency of 73 s; however, clinical
manifestations did not appear until 139 s,
suggesting that unlike three per second spike
and wave, the clinical alterations in HIHARS are
related to a cumulative physiologic change
occurring during HV. Unlike that in one other
study (7), recurrence of HIHARS with altered
awareness in our study was uncommon, and only
one of the 22 children demonstrated this change
on a subsequent EEG. Thus whatever the nature of
the event, the natural history includes
resolution with or without treatment (7). It is
our practice to inform the family that the
episode is not epileptic, and we do not
recommend antiepileptic treatment.
This study demonstrates that automatisms are
observed commonly in children with HIHARS and
altered awareness. The similarity in the
clinical features and in the ages between those
with HIHARS and those with absence seizures
complicates the clinical diagnosis. Eye opening
and eyelid flutter occur less often than in
absence seizures, and yawning, smiling, and
particularly fidgeting occur more frequently
than in absence seizures.
