Neurophysiologie Clinique/Clinical Neurophysiology (2009) 39, 101—105

ORIGINAL ARTICLE / ARTICLE ORIGINAL

Prognostic role of somatosensory and auditory
evoked potentials in paediatric hypoxic-ischemic
encephalopathy managed with hypothermia:
An illustrative case
Un cas illustratif de la valeur pronostique des potentiels évoqués
somesthésiques et auditifs dans l’encéphalopathie anoxo-ischémique
traitée par hypothermie
A. Suppiej ∗, A. Cappellari , P.E. Cogo
Department of paediatrics, university of Padova, via Giustiniani, 3, 35100 Padova, Italy
Received 4 July 2008; accepted 22 February 2009
Available online 17 March 2009

KEYWORDS
Postanoxic coma;
Somatosensory
evoked potentials;
Mismatch negativity;
Children;
EEG;
Prognosis

Summary
Introduction. — The contribution of clinical neurophysiology in the neurological prognosis of
hypoxic-ischemic coma has been well established in adults: the bilateral absence of cortical somatosensory evoked potentials (SEP) is considered the single best indicator of adverse
outcome, while the presence of the auditory mismatch negativity (MMN) is thought to herald
arousal.
Study aim. — To use MMN combined with serial EEG recordings, somatosensory and brainstem
auditory evoked potentials (BAEP) in a paediatric case of postanoxic coma managed with
hypothermia, since they have not yet been described in children.
Methods. — We report the case of a nine-year-old boy with hypoxic-ischemic encephalopathy
due to cardiorespiratory arrest after accidental burial in sand, who was treated with therapeutic
hypothermia for 72 hours. Serial EEG recordings, evoked potentials, brain CT scan and brain MRI
were performed in the first few days after the event.
Results. — SEP to median nerve stimulation showed bilateral absence of the N20 component,
while the N13 and P14 peaks were preserved; BAEP showed normal I—V interpeak latency and
normal hearing threshold. At the same time, the MMN component of auditory event related
potentials, recorded in the classical oddball paradigm, was absent. Seventeen months after the
accident, the patient is alive in persistent vegetative state.

∗ Corresponding author.

E-mail address: suppiej@pediatria.unipd.it (A. Suppiej).
0987-7053/$ – see front matter © 2009 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.neucli.2009.02.003

102

A. Suppiej et al.
Conclusions. — This case illustrates the particular significance of SEP and MMN together with EEG
in gaining prognostic information, even in sedated and hypothermic patients, and encourages
systematic study of these prognostic tools in paediatric postanoxic coma.
© 2009 Elsevier Masson SAS. All rights reserved.

MOTS CLÉS
Coma postanoxique ;
Potentiels évoqués
somesthésiques ;
Négativité de
discordance ;
Enfants ;
EEG ;
Pronostic

Résumé
Introduction. — L’apport de la neurophysiologie clinique dans le pronostic neurologique du coma
postanoxique est bien établi chez l’adulte : l’absence bilatérale des composantes corticales
des potentiels évoqués somesthésiques (PES) constituent le meilleur indicateur individuel d’un
mauvais pronostic alors que la présence de la négativité de discordance (mismatch negativity
[MMN]) annonce un éveil imminent.
But de l’étude. — L’utilisation conjointe de la MMN, d’enregistrements EEG répétés, des PES et
des potentiels auditifs du tronc cérébral (PEAT) n’a jusqu’ici pas été décrite chez l’enfant.
Méthodes. — Nous rapportons le cas d’un enfant de neuf ans victime d’une encéphalopathie
hypoxique ischémique due à un enterrement accidentel dans le sable, traité par hypothermie
durant 72 heures. Des enregistrements répétés de l’EEG et des potentiels évoqués, un scanner
et une IRM cérébrale furent réalisés dès le premier jour après l’accident.
Résultats. — Les PES des nerfs médians montrent une absence bilatérale d’onde N20, les ondes
N13 et P14 étant conservées, les PEAT montrent un intervalle I—V et des seuils normaux. La
MMN, obtenue au moyen d’un paradigme oddball classique, est absente. Dix-sept mois après
l’accident, le patient est toujours en vie mais dans un état végétatif.
Conclusions. — Ce cas démontre que les PES et la MMN enregistrés conjointement à l’EEG
gardent une valeur pronostique, même chez les patients sous sédation et hypothermie. Il
constitue un encouragement à une étude systématique de ces instruments pronostiques dans
l’encéphalopathie hypoxique ischémique de l’enfant.
© 2009 Elsevier Masson SAS. Tous droits réservés.

Introduction
The clinical contribution of evoked potentials (EP) in the
neurological prognosis of coma has been well established
in adults: the bilateral absence of cortical somatosensory evoked potentials (SEP) is considered the single best
indicator of adverse outcome [1] while the presence of
the mismatch negativity (MMN) component of the auditory
event-related potentials (AERP) is thought to herald arousal
[2—4].
In Paediatric Intensive Care Units, EP are not wellestablished techniques and to the best of our knowledge,
there are no reports on the role of MMN in paediatric coma.
The aetiology of coma in children is heterogeneous, and
previous studies have mainly dealt with a small number
of cases, pooling together different ages and aetiologies.
Multimodal neurophysiological techniques (SEP, Brainstem
Auditory Evoked Potentials [BAEP], EEG monitoring) have
been described in children with coma from mixed aetiologies, and results show that neurophysiology is more accurate
than the Glasgow Scale (GCS), and that the combination of
different methods gives better results than a single method.
Few studies report on the role of SEP in post-traumatic
coma [5] and even fewer deal specifically with paediatric
postanoxic coma [6]. Taken together, these studies suggest
that the absence of cortical SEP result in death or persistent
vegetative state with a positive predictive value approaching 100%, while the role of BAEP and visual evoked potentials
is less clear [7].
Hypothermia has recently become an established treatment in adults and children in order to reduce the brain
damage related to a hypoxic-ischemic insult [8]. The effect
of hypothermia on EEG and EP and the role of EP in monitor-

ing the efficacy of treatment have not been systematically
studied in children.
The aim of this study is to describe the prognostic role
of neurophysiological techniques, particularly SEP and MMN,
combined with EEG, in a child affected by postanoxic coma
and treated with hypothermia.

Patient and methods
Approval for the retrospective case-note review was
obtained from our Institution Ethical Committee.
We report on a nine-year-old boy with hypoxic-ischemic
encephalopathy due to cardiorespiratory arrest after accidental burial in sand. It was assumed that the cardiac arrest
had lasted about 30 minutes. The child was found with a
GCS of 3 and with mydriatric pupils sluggishly reactive to
light. He was resuscitated, intubated, ventilated, and transferred immediately to the Paediatric Intensive Care Unit,
Department of Paediatrics, University of Padova, Italy.
On admission, his clinical parameters were the following: rectal temperature 34.5 ◦ C, arterial blood pressure
107/53 mmHg, heart rate 112 b/min, O2 saturation 100% on
synchronized intermittent mechanical ventilation with FiO2
31%. His pupils were miotic and reactive to light, and the
GCS was 5. The brain CT scan, performed two hours after
admission, did not show any signs of cerebral oedema. The
patient was initially sedated with fentanyl (1 mcg/kg per
hour) and midazolam (0.1 mg/kg per hour) and mannitol
therapy was given because of the severe asphyxia. It was
decided to monitor the intracranial pressure and to start
therapeutic hypothermia with a cooling blanket, with a core
temperature ranging from 32.5 to 34.5 ◦ C for 72 hours [8].

Prognostic role of evoked potentials in paediatric postanoxic coma

103

Figure 1 The EEG carried out on day 1, 2, 4 and in the first 13 days after the event. On Day 1, there is massive generalized
spiking with subtle but consistent bilateral central regions accentuation; there was no clinical motor correlate. During the course
of therapeutic hypothermia (days 2 and 4), the EEG was characterized by an alpha-theta coma pattern consisting of non-reactive,
diffuse, continuous rhythms in the 6—9 Hz frequency band, but by Day 6, although levels of antiepileptic drugs were within the
therapeutic range, a progressive slowing and flattening occurred. Fp1, Fp2, T3, T4, C3, C4, O1, O2: locations by International
10—20 System; PB: Phenobarbital; PHT: phenytoin.

Meanwhile the patient presented continuous generalized
tonic clonic seizures. At this time, the EEG was not available, and he was treated with lorazepam (0.1 mg/kg i.v.),
phenytoin (15 mg/kg i.v.) and phenobarbital (18 mg/kg i.v.)
with clinical remission. An 8-channel EEG record using silver
chloride disk electrodes placed at the standard scalp sites
according to the 10—20 International System, was started a
few hours later; it showed an ictal pattern characterized
by massive generalized spiking with subtle but consistent
bilateral central region accentuation, but there were no
clinical motor manifestations (Fig. 1, 1st day). Remission was
achieved following treatment with lorazepam (0.1 mg/kg
i.v.), phenytoin (20 mg/kg i.v), and phenobarbital (20 mg/kg
i.v). The antiepileptic treatment was then continued with
phenytoin (7 mg/kg per day) and phenobarbital (10 mg/kg
per day). The EEG was repeatedly performed in the course
of follow-up treatment. On Day 2, during the course of
therapeutic hypothermia, his state was characterized by
an alpha-theta coma pattern consisting of non-reactive,
diffuse, continuous rhythms in the 6-9 Hz frequency band

(Fig. 1, 2nd day). By Day 6, although the levels of antiepileptic drugs were within the therapeutic range, a progressive
slowing and flattening of the background activity occurred
(Fig. 1). SEP to median nerve stimulation, performed on
Day 7, showed bilateral absence of the N20 component
(Fig. 2). At that time, the patient was treated with phenytoin 7 mg/kg per day and phenobarbital 8 mg/kg per day.
BAEP, carried out on Day 12, showed normal I-V inter-peak
latency and normal hearing threshold. The MMN component
of AERP recorded in the classical oddball paradigm [9] was
absent on Day 13. The anti-epileptic drugs were stopped one
and two days before BAEP and MMN recordings, respectively;
the plasmatic level of phenobarbital was within the therapeutic range, and that of phenytoin was at the lower limits
of the range. The brain CT scan on Day 6 showed diffuse
cerebral oedema and hypodensity of the basal ganglia; the
magnetic resonance imaging demonstrated signs of severe
hypoxic-ischemic insult with hyperintense signal in the T2
sequences at a diffuse cortical level, thalamus, and basal
ganglia. On Day 8, a brain CT scan was repeated and the

104

A. Suppiej et al.

Figure 2 A. Normal BAEP to 75 and 25 dB nHL. B. SEP to median nerve stimulation showing bilateral absence of N20 cortical
component and presence of N13 (§) and P14 (#) peaks. C. Auditory ERP to standard 1000 Hz tones (80% probability) and deviant
2000 Hz tones (20% probability) delivered within the Odd-ball paradigm [8], showing absence of MMN component. Cv7: spinous
process of the 7th cervical vertebra; Fz, Cz, Pz, T3, T4: locations by International 10—20 System; Cc: centro-parietal scalp location
contralateral to stimulus, Ci: centro-parietal scalp location ipsilateral to stimulus; Mc: mastoid process contralateral to stimulus;
Mi: mastoid process ipsilateral to stimulus, M2: bilateral linked mastoids, EOG: electro-oculogram.

signs of severe hypoxic-ischemic encephalopathy were confirmed.
On Day 13, the patient was transferred to another hospital, still intubated and ventilated with a GCS of 8. The
neurological examination was characterized by the absence
of response to verbal stimulation, symmetrical miotic pupils
with a sluggish reaction to light; corneal reflex and cough
reflex were present. The patient had spontaneous eye opening and eyelid movements and he showed flexion withdrawal
of arms and legs in response to strong pain stimulation.
Generalized hypertonia, hyperreflexia and bilateral Babinski
sign were also observed.
Seventeen months after the accident, the patient is
alive, breathing spontaneously, unresponsive, tetraplegic
and affected by drug resistant clinical seizures and continuous slow spike-wave activity during sleep.

Discussion
The main issue raised by our study is the prognostic role
of EP in postanoxic coma occurring in childhood. Studies
concerning the role of SEP in the prognosis of comatose
children have been hampered not only by reports combining
various aetiologies and age groups, but also by the small
number of cases studied homogeneously. However, the
value of SEP, particularly the finding of bilateral absence of

the N20 cortical component in children suffering from coma
of various aetiologies, has now been reported in at least
700 patients less than 17 years of age. Where testing has
been done with appropriate timing, the positive predictive
values have been in the range of 92—100% [1,10].
A bilateral absence of cortical SEP has been reported
with good recovery only in cases of focal lesions disrupting
the afferent somatosensory pathway, subdural and subgaleal
effusions, in a rapid surgical correction of intracranial hypertension and after a lightning strike [11], but interestingly
loss of cortical SEP associated with good outcome has not
been reported in postanoxic aetiology. Thus, the bilateral
absence of cortical SEP shown by the patient reported in
this study correctly predicted the evolution to persistent
vegetative state and confirms previous findings.
We also examined auditory ERP (13 days after resuscitation) and we found no response. The role of auditory
ERP, particularly the MMN component, in heralding arousal
following postanoxic coma has been confirmed in adult
studies, where MMN was shown to be superior to N100
auditory response, SEP, middle latency auditory evoked
potentials and other clinical parameters [2]. We are not
aware of previous studies on the MMN component in paediatric hypoxic-ischemic coma. It is well known that the MMN
is always present in normal children of similar ages under
comparable recording conditions [9]. Cochlear impairment
due to asphyxia was excluded in our patient by BAEP testing

Prognostic role of evoked potentials in paediatric postanoxic coma
showing normal hearing (occurrence of wave V at near
threshold stimulation). Although research has not yet clarified the effect of slow wave sleep mechanisms and sedation
on SEP and MMN in children, the finding of non-reactive
background EEG activity and the subsequent unfavorable
neurological prognosis, even after hypothermia treatment,
suggests that our SEP and MMN findings could be considered
unfavorable features, rather than the effect of sedation or
sleep.
The importance for prognostication of the evolution of
EEG changes in the early hours and days after an hypoxic
insult has long been recognized, since the pioneering work
of Binnie et al. [12] and Prior [13].
The neurophysiological monitoring of our child suffering
from severe hypoxic-ischemic encephalopathy was useful to
confirm the role of the EEG in prompt identification of non
convulsive status epilepticus as well as in monitoring the
efficacy of antiepileptic treatment. Status epilepticus complicates postanoxic coma and, in adults, it has been shown
to represent an independent outcome predictor [14]. Status epilepticus has been estimated to occur in about one
third of unresponsive paediatric patients [15]; it can occur
in absence of clinical signs, as with the child reported in
this study. Its prompt recognition by means of EEG monitoring is important to limit the development of secondary brain
damage [13].
Following recovery from status epilepticus, during the
course of therapeutic hypothermia, the EEG of our patient
was characterized by an alpha-theta coma pattern consisting
of non-reactive, diffuse, continuous rhythms in the 6—9 Hz
frequency band. The alpha-theta coma pattern, when not
due to sedative drugs, is rare in children, 1% in the study by
Hosain et al. [15]. No information is available on the role of
hypothermia in association with the alpha-theta pattern in
the EEG of comatose children.
Since brainstem involvement and severe cortical dysfunction with functional disconnection of the hemispheres
have both been implicated in the pathogenesis of postanoxic
alpha-theta pattern coma, the finding of normal BAEP and
absent cortical SEP helped localize lesions of our patient in
the cerebral cortex. Research indicates a worse prognosis in
adults when the alpha pattern coma shows no reactivity to
stimulation [16]. However, its prognostic significance is less
clear in children [15]. SEP and AERP findings may also be used
in clarifying the prognostic significance of this EEG pattern.
In our patient, EP were performed when the EEG had already
changed to a slow low voltage pattern, yet the evolution of
EEG features was difficult to interpret because of the effect
of previous hypothermia and plasma levels of antiepileptic
drugs in a child. EP findings supported the suspicion of an
unfavorable evolution.

Conclusion
Our data confirm the role of EP together with EEG in the
management and prognosis of paediatric hypoxic-ischemic
encephalopathy. When sedation and hypothermia can affect
the interpretation of the EEG, the finding of normal BAEP will
suggest a residual brainstem function and exclude brainstem
death (even in the presence of a flat EEG), while the absence

105

of the MMN and cortical SEP and the lack of EEG reactivity
orient the prognosis to persistent vegetative state. It would
be interesting to study whether MMN is able to herald recovery of consciousness in paediatric patients with the presence
of cortical SEP.
Further studies are required to evaluate the role of integrated clinical neurophysiology, and in particular the role of
MMN in paediatric postanoxic coma.
Conflict of interests: none.

References
[1] Robinson LR, Micklesen PJ, Tirschwell DL, Lew LH. Predictive
value of somatosensory evoked potentials for awakening from
coma. Crit Care Med 2003;31:960—7.
[2] Daltrozzo J, Wioland N, Mutschler V, Kotchoubey B. Predicting
coma and other low responsive patients outcome using eventrelated brain potentials: a meta-analysis. Clin Neurophysiol
2007;118:606—14.
[3] Guérit JM, Amantini A, Amodio P, Andersen KV, Butler S, de
Weerd A, et al. Consensus on the use of neurophysiological
tests in the intensive care unit (ICU): Electroencephalogram
(EEG), Evoked Potentials (EP), and Electroneuromyography
(ENMG). NCCN Neurophysiol Clin Clin Neurophysiol 2009;
39.
[4] Fischer C, Luauté J, Adeleine P, Morlet D. Predictive value of
sensory and cognitive evoked potentials for awakening from
coma. Neurology 2004;63:669—73.
[5] Carter BG, Butt W. A prospective study of outcome predictors after severe brain injury in children. Intensive Care Med
2005;31:840—5.
[6] Mandel R, Martinot A, Delepoulle F, Lamblin MD, Laureau E,
Vallee L, et al. Prediction of outcome after hypoxic-ischemic
encephalopathy: a prospective clinical and electrophysiologic
study. J Pediatr 2002;141:45—50.
[7] Mewasingh LD, Christophe C, Fonteyne C, Dachy B, Ziereisen
F, Christiaens F, et al. Predictive value of electrophysiology in children with hypoxic coma. Pediatr Neurol 2003;28:
178—83.
[8] Hypothermia after cardiac arrest study group. Mild therapeutic
hypothermia to improve the neurologic outcome after cardiac
arrest. N Engl J Med 2002;346:549—56.
[9] Morr ML, Shafer VL, Kreuzer JA, Kurtzberg D. Maturation of mismatch negativity in typically developing infants and preschool
children. Ear & Hearing 2002;23:118—36.
[10] Abend NS, Licht DJ. Predicting outcome in children with
hypoxic ischemic encephalopathy. Pediatr Crit Care Med
2008;9:32—9.
[11] Beca J, Cox PN, Taylor MJ, Bohn D, Butt W, Logan WJ, et al.
Somatosensory evoked potentials for prediction of outcome in
acute severe brain injury. J Pediatr 1995;126:44—9.
[12] Binnie CD, Prior PF, Lloyd DSL, Scott DF, Margerison JH. Electroencephalographic prediction of fatal anoxic brain damage
after resuscitation from cardiac arrest. Br Med J 1970;4:265—8.
[13] Prior PF. The EEG in acute cerebral anoxia. Amsterdam:
Excerpta Medica; 1973.
[14] Rossetti AO, Logroscino G, Liauder L, Ruffieux C, Ribordy V,
Schaller MD, et al. Status epilepticus: An independent outcome
predictor after cerebral anoxia. Neurology 2007;69:255—60.
[15] Hosain SA, Solomon GE, Kobylarz EJ. Electroencephalographic
patterns in unresponsive pediatric patients. Pediatr Neurol
2005;32:162—5.
[16] Kaplan PW, Genoud D, Ho WT, Jallon P. Etiology, neurologic
correlations, and prognosis in alpha coma. Clin Neurophysiol
1999;110:205—13.