Seizure 2001; 10: 294–297

doi:10.1053/seiz.2000.0493, available online at http://www.idealibrary.com on

CASE REPORT

Area-selective stimulus-provoked seizures in
post-anoxic coma
†

†

‡

‡

‡

S. D. GATZONIS , CH. ZOURNAS , A. MICHALOPOULOS , S. PRAPAS , S. ARGENTOS & S.
‡
GEROULANOS
†

Department of Neurology, Athens Medical School, Eginition Hospital;
Center, Athens, Greece

‡ ICU, Onassis Cardiac Surgery

Correspondence to: S. D. Gatzonis, Eginition Hospital, 72 vas. Sofias av., 11528 Athens, Greece.
E-mail: sgatzon@cc.uoa.gr

We describe the case of a 70-year-old patient in whom hemiconvulsive seizures occurred during metabolic derangement, multiple stroke and post-anoxic coma following cardiac arrest.
We employed the methods of clinical and EEG evaluation and CT brain scan.
We found that hemiconvulsive seizures were provoked following a light tactile stimulus in the left-trigeminal area and occasionally a strong tapping in the right-trigeminal area.
We conclude that this type of stimulus-provoked seizure is extremely rare and could be explained by diffuse and severe brain
damage.
c 2001 BEA Trading Ltd
Key words: post-anoxic coma; hemiconvulsive seizures; stimulus-sensitive seizures; electroencephalography; encephalopathy.

INTRODUCTION
Stimulus-sensitive myoclonus has been observed in
encephalopathetic and post-anoxic comatose patients.
It is a well-established clinical syndrome associated
with a poor prognosis1, 2 . However, there are stimulussensitive epileptic phenomena other than myoclonus
which seem to be rare, poorly established clinical entities3 .
This paper deals with the case of a patient with
spontaneous left-sided clonic seizures experienced after cardiac arrest. Light tactile stimuli on the lefttrigeminal area also elicited the same kind of seizure
as the spontaneous one.

MATERIALS AND METHODS
A 70-year-old man with a history of diabetes mellitus,
chronic atrial fibrillation and chronic renal failure was
1059–1311/01/040294 + 04

$35.00/0

operated on to perform coronary artery bypass. During
his prolonged stay in the ICU (77 days) the patient developed nosocomial infections, severe sepsis, acute on
chronic renal failure and low cardiac output syndrome.
The patient was discharged from the ICU on the
77th POD. Five days later the patient sustained a cardiac arrest due to aspiration and was re-admitted to the
ICU. Ten hours after the cardiac arrest, spontaneous
hemiconvulsive seizures were noted with a frequency
of 4 to 5 seizures per hour. The neurological examination revealed coma, without spontaneous movements
and no reactions to painful stimuli. Both pupils were
dilated 2-mm with a slight light reflex. No corneal reflex or reaction to tracheal stimulation was observed.
Several slow movements of the jaw were noted. There
was diffuse hypotonia and muscle atrophy. Tendon reflexes were absent, and no response to the plantar stimulation was elicited bilaterally.
The CT scan revealed: (1) right-cerebellarhemisphere infarction; (2) smaller left-cerebellar inc 2001 BEA Trading Ltd

Area-selective stimulus-provoked seizures in post-anoxic coma

295

Fig. 1: (a) Seizure following a touch on the left side of the face. (b) Seizure following a tap on the right side of the face. (c) No
seizure after a touch on the right side of the face.

farcts; (3) a small infarct (5 mm) in the left pons;
(4) deep left-hemisphere infarction, involving the
corona radiata, the posterior portion of the external
capsule, the posterior limb of the internal capsule and
part of the lentiform nucleus; and (5) small focal infarctions in the right-lentiform nucleus, the posterior
limb of the internal capsule and part of the ventral
aspect of the thalamus (Fig. 2).
During a seizure, the patient displayed simultaneous repetitive clonic jerks of the left side of the face,
platysma, tongue, and both the left arm and left leg.
The seizures lasted 6–20 seconds. A portable EEG
was obtained. The clinical event was accompanied by
spike and polyspike bursts and high-voltage sharpslow-waves in the EEG, recorded bilaterally (Fig. 1).
There were prolonged periods of suppression between
seizures. While the electrodes were placed, it was
noted that a touch on the left-trigeminal area produced
a typical seizure similar to the spontaneous seizure.
A systematic application of stimuli was performed
as follows:
1. Tactile stimuli were applied to the left-

periorbital area 20, 30, 40, and 60 seconds after
seizure cessation, in two series.
2. Five tactile stimuli on the left-hemifacial area no
earlier than 60 seconds postictally (periorbital
area and cheek).
3. Five tactile and five strong stimuli on the left
and right side of the body, respectively (neck,
chest (2), arm, leg).
4. Five tactile and six strong stimuli on the rightperiorbital area.
The tactile stimulus was performed by a touch from
the examiner’s finger, without pressure. Strong stimulus was performed by tapping the paramedian finger
as strongly as the auscultatory percussion requires.
The results were as follows. In addition to the spontaneous seizures, a tactile stimulus on the left side of
the face also elicited a seizure. As shown in Fig. 1,
the seizure started 0.5–1.0 seconds after the stimulus
to the left-hemifacial area.
Stimuli applied earlier than 30 seconds postictally
produced no seizure or discharges. We noted eight

296

S. D. Gatzonis et al.

Fig. 2: CT brain scan after cardiac arrest.

seizures following eight light tactile stimuli on the
left-periorbital area, applied systematically at least
1 minute after the last seizure (Fig. 1(a)). No seizure
was recorded after five light tactile stimuli on the
right side of the face (Fig. 1(c)). However, a typical
seizure following strong finger tapping over the right
frontal area was observed four times in six attempts
(Fig.1(b)). In these cases a time delay of 2–2.5 seconds between stimulus and seizure was noted. Neither
auditory nor light stimuli provoked seizures.
An IV infusion of 4 mg of clonazepam resulted in a
decrease in the seizure frequency and duration of both
spontaneous and provoked seizures.
The patient died 28 hours after cardiac arrest.

DISCUSSION
Somatosensory stimuli as a triggered mechanism of
epileptic phenomena are very rare4 . A number of
papers describe a stimulus-sensitive myoclonus during post-anoxic coma. Wijdicks et al.1 referred to
40 patients with post-anoxic myoclonus in whom

touch, tracheal suctioning, loud sound, or placement of
central catheters triggered myoclonic jerks. Young et
al.5 reported six patients with post-anoxic myoclonus
triggered by touch or nail-bed pressure and one had
myoclonus triggered by auditory stimuli. Celesia et
al.2 reported that 11 out of 12 encephalopathetic
patients with myoclonus had a stimulus-triggered generalized status myoclonus. Recently, Van Cott et al.3
reported four patients with stimulus-sensitive seizures
following hypoxic-ischemic injury.
As far as we know there is only one case of selective stimulus-sensitive myoclonus in a post-anoxic
patient described by Niedermeyer et al.6 In this case,
isolated myoclonic jerks could be elicited by a light
touch to the right-trigeminal area. Our case seems to
be unique as a somatosensory stimulus in a restrictive
skin area, limited to the left trigeminal territory, provoked a hemiconvulsive seizure.
We suggest that a tactile stimulus causes an irritation of somatosensory trigeminal and ascending diffuse projection pathways. Because of the brain damage, concerning, among others, the inhibitatory connections of reticular formation, there is no appropriate

Area-selective stimulus-provoked seizures in post-anoxic coma

inhibition of this ascending signal. This signal seems
to be adequate—or appropriately enhanced—to activate the whole cortex or to produce synchronization,
or both, resulting in generalized epileptic discharges3 .
The epileptic discharges recorded in our case were
simultaneously bilateral. The affected parts started to
seize simultaneously. These facts imply a synchronous
whole-cortex activation, which in turn suggests an underlying mechanism as in generalized seizures.
In spite of the bilateral discharges only the left
side of the face and body were seized. A destruction
of cortico-pontine and cortico-spinal projections from
the left pro-Rolandic and Rolandic areas should be
considered. The CT scan findings—findings 3 and 4—
could represent such damage. The clinical manifestation of our patient may represent a fragment of generalized tonic–clonic epileptic seizure, the latter modified by brain damage, as Celesia et al.2 have proposed
for post-anoxic myoclonus.
Tactile stimuli applied to the neck and body of our
patient did not precipitate any seizure. This could be
the result of a bilateral destruction of the ascending
somatosensory fibres at a level lower than the pons,
but we have no evidence to support this point of
view. A meaningful explanation could be the relatively
lower ‘importance’ of tactile stimuli from spinal dermotomies in comparison with the tactile stimuli from
the trigeminal area, since the face has a greatly enlarged representation in the thalamus and cortex. Clinical data support this point of view; as in most cases of
touch-evoked seizures the effective stimulus is limited
to certain regions, particularly on the head7 . Seizures
evoked by hot water appear to be related to the habit
of repeatedly pouring hot water over the head and in
cases of tactically evoked myoclonic epilepsy of infancy, the face and head are the most usually precipitated areas8 .
In our patient, only few and strong stimuli such as
strong tapping on the right-periorbital area provoked
a seizure, with a time delay suggesting that the pathway from the right-trigeminal nucleus to the cortex
is not functionally deficient in comparison with the
left side. The time delay may represent a spreading
of the irritant from the right-trigeminal nucleus to the
left brain stem, through polysynaptic connections of
reticular formation, ascending to the higher structures
through the same pathway as the stimulus from the
left-hemifacial area.
Long periods of EEG suppression, lasting from

297

2–10 minutes were a permanent feature of the EEG
recording. Burst suppression patterns in comatose
patients have periods of suppression lasting from 2–
25 seconds3 . Only one patient reported in Van Cott et
al.3 had suppression periods longer than 25 seconds
and it was considered an indication of severe brain injury. In our case, the deep coma, the absence of brain
stem reflexes and ensuing death could also suggest that
our patient had a severely diffuse brain injury responsible for the long periods of suppression in the EEG.
A post-mortem examination was not performed.

CONCLUSION
In conclusion, widespread hypoxic brain damage was
responsible for the abnormal neuronal hyperexcitability or synchronization, or both.
This damage caused, through the ascending diffuse projection system, stimulus-evoked generalized
epileptiform discharges, which had a modified clinical
expression such as hemiconvulsive seizures. Our case
is in agreement with the literature since the seizures in
post-anoxic coma carry a poor prognosis.

REFERENCES
1. Wijdikcks, E. F. M., Parisi, J. E. and Sharbrough, F. W. Prognostic value of myoclonus status in comatose survivors of cardiac arrest. Annals of Neurology 1994; 35: 239–43.
2. Celesia, G. G., Grig, M. M. and Ross, E. Generalized
status myoclonus in acute anoxic and toxic-metabolic encephalopathies. Archives of Neurology 1988; 45: 781–4.
3. Van Cott, A. C., Blatt, I. and Brenner, R. P. Stimulus-sensitive
seizures in postanoxic coma. Epilepsia 1996; 37: 868–874.
4. Bickford, R. G. and Klass, D. W. Sensory precipitated and reflex mechanisms. In: Basic Mechanisms of the Epilepsies (Eds
H. H. Jasper, A. A. J. Ward and A. Pope). Boston, Little Brown
& Co, 1969: pp. 543–564.
5. Young, G. B., Gilbert, J. J. and Zochodne, D. W. The significance of myoclonic status epilepticus in postanoxic coma.
Neurology 1990; 40: 1843–8.
6. Niedermayer, E., Bauer, G., Burnite, R. and Reichenbach,
D. Selective stimulus-sensitive myoclonus in acute cerebral anoxia. Archives of Neurology 1977; 34: 365–368.
7. Gastaut, H. and Tassinari, C. A. Triggering mechanisms in
epilepsy. The electroclinical point of view. Epilepsia 1966; 7:
85–138.
8. Deonna, T. Reflex seizures with somatosensory precipitation.
In: Reflex Epilepsies and Reflex Seizures: Advances in Neurology Vol. 57 (Eds B. G. Zifkin, F. Andermann, A. Beaumanoir and A. J. Rowan). Philadelphia, Lippincott-Raven,
1988: pp. 193–206.