Predictive value of clinical and EEG features in the diagnosis of stroke and hypoxic ischemic encephalopathy in neonates with seizures.
ABSTRACT In neonates, the differentiation of stroke and hypoxic ischemic encephalopathy (HIE) is important. Neuroimaging presents technical challenges in unstable neonates, resulting in frequently delayed or missed diagnosis of stroke. Differentiating clinical and electroencephalographic (EEG) features would assist physicians in the timely diagnosis. We sought to determine, in neonates with seizures, clinical and EEG features that differentiate stroke and HIE.
Retrospective cohort study comparing clinical, seizure, and EEG features in term neonates with ischemic stroke or HIE and seizures within 7 days after birth, admitted at The Hospital for Sick Children. Putative clinical and EEG predictors of stroke were analyzed with univariate and multivariate methods.
Sixty-two newborns with stroke (n=27) or HIE (n=35) were studied. With univariate analysis, predictors of stroke included delayed seizure onset (>or=12-hours after birth) (P<0.0001; OR, 26.4; 95% CI, 6.8, 102.5), focal motor seizures (P=0.001; OR, 7.2; 95% CI, 2.0, 26.0) and pattern of neurological abnormalities (P<0.0001). With multivariate analysis, delayed seizure onset (P<0.0001; OR 39.7; 95% CI, 7.3, 217.0) and focal motor seizures (P=0.007; OR, 13.4; 95% CI, 2.1, 87.9) predicted stroke. Presence of both predictors had 100% positive predictive value and specificity, 61% negative predictive value and 37% sensitivity.
In neonates, onset of seizures beyond 12 hours of birth and clinically observed focal seizures are predictive of stroke. These preinvestigation indicators of stroke may facilitate earlier diagnosis and institution of specific management strategies.
Article: Pediatric arterial ischemic stroke.[Show abstract] [Hide abstract]
ABSTRACT: This article aims to provide a broad overview of pediatric arterial ischemic stroke, from recognition and diagnosis to the short-term and long-term management based on current available literature. Arterial ischemic stroke in children represents a significant disorder with a concerning high rate of adverse outcomes, including potentially preventable recurrent stroke. Although awareness of pediatric stroke is increasing, diagnosis is still commonly delayed or missed altogether, particularly in younger children. Current vascular imaging techniques have limitations in accurate diagnosis of arteriopathies that are now recognized as an important cause of childhood stroke. Significant variability exists in treatment of pediatric stroke. Management is based on published consensus guidelines; however, individual children require an individualized approach. As pediatric stroke specialists become increasingly available, the collaboration of such experts on individual management is crucial. Definitive evidence-based treatment for pediatric stroke awaits the development of randomized controlled trials.Continuum (Minneapolis, Minn.). 04/2014; 20(2 Cerebrovascular Disease):370-86.
- [Show abstract] [Hide abstract]
ABSTRACT: Stroke is the second most common cause of seizures in term neonates and is associated with abnormal long-term neurodevelopmental outcome in some cases.PLoS ONE 01/2014; 9(7):e100973. · 3.53 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: Perinatal arterial ischemic stroke (AIS) occurs in an estimated 17 to 93 per 100000 live births, yet the etiology is poorly understood. Although investigators have implicated hypoxia as a potential cause of AIS, the role of hypoxia in AIS remains controversial. The aim of this study was to estimate the association between perinatal hypoxia factors and perinatal arterial ischemic stroke through a meta-analysis of published observational studies. A systematic search of electronically available studies published through July 2013 was conducted. Publication bias and heterogeneity across studies were evaluated and summary odds ratios (ORs) and 95% confidence intervals (CIs) were calculated with fixed-effects or random-effects models. A total of 8 studies describing the association between perinatal hypoxia factors and neonatal arterial ischemic stroke (AIS) met inclusion criteria, and 550 newborns with AIS were enrolled. The associations were found for AIS: preeclampsia (OR 2.14; 95% CI, 1.25 to 3.66), ventouse delivery (OR 2.23; 95% CI, 1.26 to 3.97), fetal heart rate abnormalities (OR 6.30; 95% CI, 3.84 to 10.34), reduced fetal movement (OR 5.35; 95% CI, 2.17 to 13.23), meconium-stained liquor (OR 3.05; 95% CI, 2.02 to 4.60), low Apgar score (OR 5.77; 95% CI, 1.66 to 20.04) and resuscitation at birth (OR 4.59; 95% CI, 3.23 to 6.52). Our data did not show any significant change of the mean risk estimate for oxytocin induction (OR 1.33; 95% CI, 0.84 to 2.11) and low arterial umbilical cord ph (OR 4.63; 95% CI 2.14 to 9.98). There is a significant association between perinatal hypoxia factors and AIS. The result indicates that perinatal hypoxia maybe one of causes of AIS. Large scale prospective clinical studies are still warranted.PLoS ONE 01/2014; 9(2):e90106. · 3.53 Impact Factor
Woojin Yoon, Shafagh Fallah and Aideen M. Moore
Mubeen F. Rafay, Miguel A. Cortez, Gabrielle A. deVeber, Cherrie Tan-Dy, Amna Al-Futaisi,
Ischemic Encephalopathy in Neonates With Seizures
Predictive Value of Clinical and EEG Features in the Diagnosis of Stroke and Hypoxic
Print ISSN: 0039-2499. Online ISSN: 1524-4628
Copyright © 2009 American Heart Association, Inc. All rights reserved.
is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
2009;40:2402-2407; originally published online May 28, 2009;
World Wide Web at:
The online version of this article, along with updated information and services, is located on the
is online at:
Information about subscribing to
Information about reprints can be found online at:
document. Permissions and Rights Question and Answer process is available in the
Request Permissions in the middle column of the Web page under Services. Further information about this
Once the online version of the published article for which permission is being requested is located, click
can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office.
Requests for permissions to reproduce figures, tables, or portions of articles originally published
by guest on June 1, 2013 http://stroke.ahajournals.org/Downloaded from
Predictive Value of Clinical and EEG Features in the
Diagnosis of Stroke and Hypoxic Ischemic Encephalopathy
in Neonates With Seizures
Mubeen F. Rafay, MB, BS; Miguel A. Cortez, MD; Gabrielle A. deVeber, MD; Cherrie Tan-Dy, MD;
Amna Al-Futaisi, MD; Woojin Yoon, BSc; Shafagh Fallah, PhD; Aideen M. Moore, MD
Background and Purpose—In neonates, the differentiation of stroke and hypoxic ischemic encephalopathy (HIE) is
important. Neuroimaging presents technical challenges in unstable neonates, resulting in frequently delayed or missed
diagnosis of stroke. Differentiating clinical and electroencephalographic (EEG) features would assist physicians in the
timely diagnosis. We sought to determine, in neonates with seizures, clinical and EEG features that differentiate stroke
Methods—Retrospective cohort study comparing clinical, seizure, and EEG features in term neonates with ischemic stroke
or HIE and seizures within 7 days after birth, admitted at The Hospital for Sick Children. Putative clinical and EEG
predictors of stroke were analyzed with univariate and multivariate methods.
Results—Sixty-two newborns with stroke (n?27) or HIE (n?35) were studied. With univariate analysis, predictors of
stroke included delayed seizure onset (?12-hours after birth) (P?0.0001; OR, 26.4; 95% CI, 6.8, 102.5), focal motor
seizures (P?0.001; OR, 7.2; 95% CI, 2.0, 26.0) and pattern of neurological abnormalities (P?0.0001). With
multivariate analysis, delayed seizure onset (P?0.0001; OR 39.7; 95% CI, 7.3, 217.0) and focal motor seizures
(P?0.007; OR, 13.4; 95% CI, 2.1, 87.9) predicted stroke. Presence of both predictors had 100% positive predictive
value and specificity, 61% negative predictive value and 37% sensitivity.
Conclusions—In neonates, onset of seizures beyond 12 hours of birth and clinically observed focal seizures are predictive
of stroke. These preinvestigation indicators of stroke may facilitate earlier diagnosis and institution of specific
management strategies. (Stroke. 2009;40:2402-2407.)
Key Words: acute care ? EEG ? hypoxix ischemic encephelopathy ? predictors of diagnosis
? cereberal infarction ? neonates
etiologic diagnosis of neonatal seizures is necessary to permit
accurate decisions regarding management and prognosis. The
incidence of seizures in neonates ranges from 0.15% to 3.5%,
with most occurring within the first week of life.1–3Although
neonatal seizures can result from transient metabolic derange-
ments such as hypoglycemia or hypocalcaemia, hypoxic ische-
and ischemic stroke for 12% to 20%.5,11,12
Stroke is frequently misdiagnosed as HIE because of
overlapping clinical features.9,10In neonates with stroke,
seizures are the only manifestation in 70% to 91%1–5and
hemiparesis is uncommon (20%).13Generalized neurological
abnormalities including encephalopathy can be present in
both HIE and stroke. The prognosis and clinical management
eizures occurring in the neonatal period may be the
manifestation of a serious underlying brain insult. The
of neonatal stroke and HIE are very different.6–12Early
identification of these conditions maximizes opportunities for
disease specific management strategies. The current approach
for differentiating neonatal stroke and HIE relies on neuro-
imaging. Neuroimaging, however, is not straightforward in
neonates who, when unstable, can be difficult to safely
transport and image with CT (computed tomogram) and
preferably MRI.13,19In neonates, cranial ultrasound is a useful
noninvasive easily accessible intervention for the evaluation
of both HIE and focal brain lesions such as stroke. However,
it may not accurately diagnose stroke.15,16In addition, an early
CT and MRI without diffusion may be normal in neonates with
diagnostic tool used to evaluate seizures, howeve, the utility of
clinical seizure and EEG features in ascertaining the cause for
neonatal seizures is not well established. Few studies have
Received January 9, 2009; final revision received March 4, 2009; accepted March 19, 2009.
From the Section of Neurology, Department of Pediatrics and Child Health (M.F.R.), University of Manitoba, Winnipeg; the Program in Brain &
Behavior (M.A.C.), the Division of Neurology, Department of Pediatrics (M.A.C., G.A.d.V., A.A.-F.), the Population Health Sciences Program (G.A.d.V.,
W.Y., A.M.M.), and the Division of Neonatology, Department of Pediatrics (C.T.-D., S.F., A.M.M.), The Hospital for Sick Children, Toronto, Ontario,
Correspondence to Mubeen F. Rafay, Section of Neurology, Department of Pediatrics and Child Health, Winnipeg Children’s Hospital, Room AE 308,
820 Sherbrook Street, Winnipeg, MB, Canada R3A 1R9. E-mail firstname.lastname@example.org
© 2009 American Heart Association, Inc.
Stroke is available at http://stroke.ahajournals.orgDOI: 10.1161/STROKEAHA.109.547281
by guest on June 1, 2013http://stroke.ahajournals.org/Downloaded from
systematically examined the association between EEG find-
ings, clinical features, and specific brain lesions in neo-
nates.18,19Therefore, we conducted the current study to deter-
mine which clinical and EEG features differentiate HIE and
stroke in neonates.
Study Design and Methods
Term neonates with clinical seizures and a diagnosis of neonatal
stroke (January 1992 to December 2003) or HIE (from January 1996
to December 2001) admitted to The Hospital for Sick Children
comprised the study population. All infants were born in peripheral
hospitals in and around Toronto (total number of live births approx-
imately 65 000 per year).
Neonates with the diagnosis of ischemic stroke and seizures were
identified from the Canadian Pediatric Ischemic Stroke Registry
database, which has enrolled patients since January 1992. Registry
identifies patients based on the referral to the hospital’s Stroke
service supplemented by the International Classification of Diseases,
9th revision (ICD-9) discharge codes searches of the medical
records. Neonates with the diagnosis of HIE and seizures were
identified by the hospital’s neonatal intensive care unit database and
the ICD-9 medical records discharge codes searches. Neonates with
HIE were studied from January 1994 (because of the poor availabil-
ity of both EEG and neuroimaging for review before that period). All
cases with ICD-9 diagnostic codes for neonatal seizures, neonatal
stroke, neonatal HIE, and birth asphyxia were evaluated, so that all
potential cases were included. The health records of study patients
were reviewed and inclusion determined based on the criteria below.
Inclusion criteria consisted of: (1) gestational age ?36-weeks to
42-weeks, (2) seizure onset within 7-days of birth, (3) diagnosis of
ischemic stroke or HIE during the neonatal period (?28-days age),
and (4) availability of EEG and neuroimaging (CT or MRI) during
the neonatal period. The diagnosis of AIS required neuroimaging
findings of focal infarction confined to an established arterial
territory. Neonatal HIE is defined as an abnormal neurological state
occurring in the newborn after a significant perinatal hypoxic
ischemic insult.4For this study, the diagnosis of neonatal HIE was
based on the criteria proposed by the American College of Obstetrics
and Gynecology (ACOG) 2003 Task force Report for intrapartum
asphyxia.25Diagnosis of HIE required 3 or more of the following:
(1) evidence of metabolic acidosis on cord gas (pH ?7 and base
deficit ?12mmol per liter), (2) early onset neonatal encephalopathy,
(3) fetal bradycardia or absence of fetal heart rate variability in the
presence of persistent, late, or variable fetal heart rate decelerations,
(4) Apgar Scores of 0 to 3 beyond 5 minutes of age, (5) evidence of
multisystem involvement within 72 hours of birth, and in all cases no
evidence of acute focal cerebral abnormality on CT or MRI.20All
neuroimaging features were ascertained from radiology reports, or when
unclear, by review of neuroimaging films by the study neurologists.
Infants with cerebral sinovenous thrombosis (CSVT) were excluded,
as most had associated venous infarction. Newborns with cerebral
hemorrhage not associated with stroke, other structural central
nervous system disorders, or other identifiable causes for encepha-
lopathy such as genetic disorders, infections, and inborn errors of
metabolism were also excluded.
Clinical features assessed included: time of seizure onset, number,
duration, and type of seizure (unifocal or hemifocal versus other
seizure types such as generalized, multifocal, and subtle seizures)
according to the seizure classification proposed by Volpe.4Other
clinical features included neurological examination (diffuse or asym-
metrical neurological abnormalities), and requirement for antiepileptic
agents (single or multiple agents). Diffuse neurological abnormalities
included decreased responsiveness, diffuse tone abnormalities, and
diminished or absent primitive reflexes. Asymmetrical neurological
abnormalities included asymmetry in tone, limb, facial, or eye
EEG features were obtained from the EEG reports. All EEGs were
performed with neonatal montage with 16 channels for EEG and 5
additional channels for polygraphic recording for eye movements,
muscle activity, respirations, and ECG. The EEG features included:
background abnormalities, presence of unilateral or bilateral rolandic
sharp waves, lateralized and midline EEG findings, and electro-
graphic seizures. Background abnormalities were categorized into
background type (normal or abnormal) and background asymmetry.
The background activity was considered abnormal if the EEG
demonstrated low voltage, discontinuity, or immaturity of back-
ground for the conceptional age. The background asymmetry was
defined as more than 50% difference in amplitude between each
hemisphere. Lateralized EEG findings were defined as the presence
of positive or negative sharp or slow waves originating from one
hemisphere. The laterality of EEG features (right or left) was also
recorded. Midline EEG findings were defined as rhythmic or
nonrhythmic, high, low, or normal amplitude theta or delta activity
originating form midline hemispheric regions lasting for ?5 but ?10
seconds. The electrographic seizures were defined as rhythmic
activity in the alpha, theta, delta range or sharp waves lasting ?10
seconds with or without clinical signs.
The radiographic diagnosis of stroke on CT or MRI was the study
outcome. We also collected data in the neonates with stroke
regarding the parenchymal infarct characteristics. These were clas-
sified by the presence or absence of hemorrhagic conversion of the
infarct, infarct location (unilateral or bilateral), and infarct number
(single or multiple). One third of the HIE infants had basal ganglia
injury, one third had watershed injury, and one third had combined
basal ganglia/watershed lesions.
Both clinical and EEG putative predictors of stroke or no stroke were
selected a priori. For categorical and nominal variables, the ?2test or
Fisher exact test (when at least 1 cell had an expected value of less
than 5) were used. When there was a 0 cell in a 2?2 table, 0.5 was
added to that cell. The significance level was set at ?0.05. Standard
deviations of the mean and odds ratios (OR) and 95% confidence
interval (CI) were noted. With respect to the number of tests applied,
Bonferroni correction was performed. After univariate analysis, multi-
variate models of association between the significant predictors and the
outcome (stroke/no stroke) were constructed using a binary logistic
regression analysis. All statistical analyses were performed using the
statistical software package SAS version 8.2 (SAS Institute Inc).
During the study period, 102 neonates with seizures and the
clinical diagnosis of either AIS (n?49) or HIE (n?53) were
screened. We excluded 22 neonates with AIS and 18 with
HIE as study inclusion criteria were not met, primarily as
EEG and neuroimaging were unavailable (Figures 1 and 2).
The remaining 62 neonates formed the study cohort, 27 with
AIS and 35 with HIE.
The mean gestational age of the enrolled cohort was 39.4
weeks (range 36 to 42 weeks). There were 38 males (18 in
stroke, 20 in HIE) and 24 females (9 in stroke, 15 in HIE).
The mean seizure onset was 15.0 hours (27.8 hours ?23.3
hours in stroke and 5.1 hours ?5.2 hours in HIE). The mean
Rafay et alPredictors of Neonatal Stroke and HIE
by guest on June 1, 2013http://stroke.ahajournals.org/Downloaded from
time from seizure onset to the initial EEG recording was
2.2?2.5 days (median day 1 for HIE and day 2 for stroke).
The EEG was abnormal in all patients with HIE and 20 patients
with AIS. The clinical and EEG features are compared between
neonates with stroke and HIE in Tables 1 and 2.
In neonates with stroke, 6 had Apgar scores of ?3 at 1
minute, including 3 neonates with persistently low Apgar
scores at 5 and 10 minutes. Diffuse neurological abnormali-
ties were present in nearly one third and an asymmetrical
motor examination in 3 neonates (2 unilateral AIS and 1
bilateral AIS). All neonates with HIE had diffuse neurologi-
cal abnormalities and none had an asymmetrical neurological
examination. Focal motor seizures were observed in 11 of 22
neonates with unilateral AIS infarcts and 2 of 5 with bilateral
AIS infarcts. In neonates with unilateral infarcts, focal motor
seizures were contralateral and lateralized EEG findings were
ipsilateral to the infarct in all except 1 patient.
In univariate analysis the predictors significant for the
diagnosis of stroke were: seizure onset 12 hours or more after
birth (P?0.0001; OR, 26.4; 95% CI, 6.8, 102.5), focal motor
seizures (P?0.001; OR, 7.2; 95% CI, 2.0, 26.0), and pattern of
neurological abnormalities (P?0.0001; Table 1). Differences
were also identified in the duration of seizures (P?0.020; OR,
0.91; 95% CI, 0.84, 0.99), frequency of response to single
antiepileptic drug (P?0.03; OR, 12.1; 95% CI, 0.7, 225.4), and
lateralized EEG findings (P?0.02; OR, 5.7; 95% CI, 1.1, 28.7;
presence of rolandic sharp waves (either unilateral or bilateral),
midline EEG findings, and electrographic seizures, were not
significantly different between AIS or HIE group (Table 2).
Using multivariate binary logistic regression analysis, sig-
nificant predictors for stroke included delayed seizure onset
(12 hours or more after birth; P?0.0001; OR 39.7; 95% CI,
7.3, 217) and focal motor seizures (P?0.007; OR, 13.4; 95%
CI, 2.1, 87.9). The presence of both later onset and focal
motor seizures had a positive predictive value and specificity
of 100%, negative predictive value and sensitivity of 61% and
37%, respectively, and conferred a 2.6 times increased
likelihood of stroke (95% CI, 1.8 to 3.5).
Our study is the first to assess in a well-defined cohort of term
neonates the clinical and EEG characteristics differentiating
stroke from HIE. We found that delayed seizure onset and
Figure 1. Neonates with ischemic stroke
enrolled in the study: 49 neonates with
arterial ischemic stroke and seizures
from January 1992 to December 2003.
Figure 2. Neonates with HIE enrolled in
the study: 53 neonates with HIE and
neonatal seizures from January 1994 to
by guest on June 1, 2013http://stroke.ahajournals.org/Downloaded from
clinically observed focal seizures are predictors of stroke in
neonates with seizures. The differentiation of stroke and HIE
in neonates is important as management approaches differ in
the two conditions. These predictors will be helpful to
prioritize neonates who require rapid neuroimaging useful for
the diagnosis of stroke or the initiation of neuroprotective
therapies for HIE.
Asymmetrical neurological examination cannot differenti-
ate stroke from HIE, because with an immature brain hemi-
paresis is only seen in the minority of neonates with stroke.13
In our study, an asymmetrical neurological examination was
present in only 3 neonates with stroke. If stroke is suspected
additional vascular imaging including MR or CT angiogram
or venogram or DWI (diffusion weighted imaging) is indi-
cated. Once stroke is diagnosed, investigations to identify
cardiac sources for embolism underlying stroke and pro-
thrombotic disorders are performed.10,26,27Recommended treat-
ments for neonatal stroke include anticoagulant therapy in
specific stroke subtypes such as CSVT and cardiogenic AIS.17
For HIE, early hypothermia (starting within 6 hours of
delivery) has been proven to be effective, and this and other
neuroprotective strategies are increasingly being recognized
as standard treatment options.18,19,28For clinically suspected
neonatal HIE, neuroimaging, preferably MRI, is recom-
mended at or more than 48 hours after birth, however if other
conditions such as birth trauma are suspected, an earlier CT
or MRI may be required.20The prompt differentiation of
stroke and HIE is therefore crucial because these interven-
tions are maximally effective if instituted early or within a
narrow therapeutic window.19,28–31Early diagnosis also as-
sists in early and specific prognostication. Death and severe
neurological deficits are frequent in neonates with HIE
compared to stroke.34–40The nature of neurological deficits
typically also differs, with hemiparesis in stroke and cogni-
tive and bilateral motor deficits in HIE.9,10,33,34
In our study the seizure onset in neonates with stroke was
significantly delayed beyond 12 hours of life compared to
neonates with HIE. Traditional teaching has associated neo-
natal stroke with seizures occurring between 24 and 72 hours
and neonatal HIE with seizures occurring within 12 to 24
hours after birth.35Although in neonates with stroke, studies
have also reported early seizures,6,12,14our findings show that
in neonates with stroke, seizure onset after 12 hours is 5 times
more frequent than in neonates with HIE.
The second predictor of stroke was the occurrence of
clinically apparent focal seizures, which were present in half
of the neonates with stroke and only 11% of neonates with
HIE. Focal seizures are uncommon among neonatal seizures,
and have been previously associated with focal structural
abnormalities including stroke.5,6,13,36,37In these studies, focal
seizures were contralateral to the stroke in the majority, as we
observed in our patients. Other seizure types, including
automatisms and generalized seizures, were also seen which
have been previously reported in unilateral neonatal
stroke.6,12,27In our study, focal motor seizures, combined
with later seizure onset, reliably predicted stroke (all neonates
with both features had AIS). This combination when present
can be used to confidently predict stroke in the first few days
Table 1.Comparison of Clinical Characteristics of Neonates With Stroke and HIE
Clinical CharacteristicsStroke (n?27)HIE (n?35)OR (95% CI)
Sz. onset ?12 hours after birth
Duration of Sz (minutes?SD)
Focal motor Sz
Response to single AED*
26.4 (6.8, 102.5)
7.2 (2.0, 26.0)
12.1 (0.7, 225.4)
n indicates number of patients; SD, standard deviation; AED, antiepileptic drug; Sz, seizure; OR, odds ratio; CI, confidence interval.
*Fisher exact test reported. All P values and OR represent univariate analysis, †P value 0.08, OR: 0.1 (0.0, 2.0).
Table 2.Comparison of EEG Characteristics of Neonates With Stroke and HIE
EEG CharacteristicsStroke (n?27)HIE (n?35)OR (95% CI)
Unilateral rolandic PSW
Lateralized EEG findings
Midline EEG findings*
2.1 (0.74, 5.8)
2.4 (0.8, 7.3)
0.23 (0.02, 1.4)
8.25 (1.4, 83.6)
4.3 (0.31, 229.5)
0.59 (0.2, 1.8)
n indicates number of patients; PSW, positive sharp waves; OR, odds ratio; CI, confidence interval.
*Fisher exact test reported. All P values and OR represent univariate analysis.
Rafay et alPredictors of Neonatal Stroke and HIE
by guest on June 1, 2013http://stroke.ahajournals.org/Downloaded from
of life, however the absence of this combination cannot rule
out stroke, and neuroimaging is still necessary.
In our study EEG features were not significantly different
between neonates with stroke or HIE. We observed electro-
graphic seizures in approximately one third of neonates at
similar frequencies in both groups, emphasizing that clinical
observation alone underestimates seizure frequency in neo-
nates. A normal EEG background and lateralized EEG
findings tended to be more frequent in neonates with stroke.
A normal EEG background in neonates with stroke may
reflect a smaller localized distribution of cerebral damage
compared with the diffuse injury in HIE, and corresponds
with the observation that neonates with stroke are less likely
to be encephalopathic or obtunded between seizures.5,6,38In
one study, diffuse background abnormalities were frequently
seen with mutifocal lesions as compared to unifocal lesions.39
EEG background provides prognostic information for both
neonatal stroke and HIE. Moderate to markedly abnormal
background has been correlated with poor outcome in both
conditions.3,8,12,13,15,16,40–42Our study may have been under-
powered to detect normal background as a predictor of stroke.
This may also be related to the coexistence of hypoxic insult
in some neonates with stroke in our cohort.
We observed asymmetrical EEG abnormalities in one third
of neonates with stroke and infrequently in neonates with
HIE. None of our neonates had periodic lateralized epilepti-
form discharges. The presence of asymmetrical EEG abnor-
malities has been associated with focal structural brain
lesions such as stroke in children.5,6,23,24,43–45However, asym-
metrical EEG abnormalities with HIE23,44and transient asym-
metries of the background in healthy term neonates45,46have
also been reported.
Our study had several limitations related to the retrospec-
tive design including potential patient selection bias consist-
ing of inclusion of patients admitted to our tertiary care
hospital (may have resulted in studying more severely af-
fected neonates), small sample size with resultant wide 95%
confidence intervals, requirement of EEG and neuroimaging
(may have biased our inclusion to those with most severe
seizures requiring these investigations or exclusion of those
who did not have these tests performed in the time frame),
and reliance on reports for radiographic and EEG findings.
Because the description of clinical seizures was obtained by
retrospective chart review, the seizure occurrence and char-
acteristics were potentially subject to both under- and over-
reporting. In some children, considerable delay in EEG
recording and use of antiseizure medications before EEG may
have affected the EEG findings. Individual reader differences
in reporting EEG features and also may have been a potential
source of error.4,47However, these limitations would likely
have applied equally to both groups and therefore should not
account for the differences we observed.
In conclusion, in neonates with seizures, clinical features
including delayed seizure onset (12 hours or more after birth)
and focal motor seizures reliably differentiate stroke from
HIE among neonatal seizure etiologies, especially when both
features are present. These findings are highly relevant to
clinicians caring for neonates, because application of these
predictors can assist in the diagnosis of stroke even before
EEG, CT, or MRI. Earlier diagnosis will be helpful in
allowing more specific management strategies and prognos-
tication which are geared to reducing adverse outcomes in
neonates with stroke and HIE.
The authors thank Professor David L. Streiner, University of
Toronto, Canada, for his guidance and comments.
Sources of Funding
This study was funded by the Hospital for Sick Children Foundation.
1. Allan WC, Riviello JJ Jr. Perinatal cerebrovascular disease in the neonate.
Parenchymal ischemic lesions in term and preterm infants. Pediatr Clin
North Am. 1992;39:621–650.
2. Holden KR, Mellits ED, Freeman JM. Neonatal seizures. I. Correlation of
prenatal and perinatal events with outcomes. Pediatrics. 1982;70:
3. Lombroso CT. Prognosis in neonatal seizures. Adv Neurol. 1983;34:
4. Volpe JJ. Neonatal Seizures. Neurology of the Newborn. IV ed. Phila-
delphia: W.B. Saunders Company; 2001.
5. Clancy R, Malin S, Laraque D, Baumgart S, Younkin D. Focal motor
seizures heralding stroke in full-term neonates. Am J Dis Child. 1985;
6. Estan J, Hope P. Unilateral neonatal cerebral infarction in full term
infants. Arch Dis Child Fetal Neonatal Ed. 1997;76:F88–F93.
7. Levene MI, Trounce JQ. Cause of neonatal convulsions. Towards more
precise diagnosis. Arch Dis Child. 1986;61:78–79.
8. Ortibus EL, Sum JM, Hahn JS. Predictive value of EEG for outcome
and epilepsy following neonatal seizures. Electroencephalogr Clin
9. Ferriero DM. Neonatal brain injury. N Engl J Med. 2004;351:1985–1995.
10. Nelson KB, Lynch JK. Stroke in newborn infants. Lancet Neurol. 2004;
11. Ment LR, Duncan CC, Ehrenkranz RA. Perinatal cerebral infarction. Ann
12. Sreenan C, Bhargava R, Robertson CM. Cerebral infarction in the term
newborn: clinical presentation and long-term outcome. J Pediatr. 2000;
13. Mercuri E. Early diagnostic and prognostic indicators in full term infants
with neonatal cerebral infarction: an integrated clinical, neuroradiological
and EEG approach. Minerva Pediatr. 2001;53:305–311.
14. Lee J, Croen LA, Lindan C, Nash KB, Yoshida CK, Ferriero DM,
Barkovich AJ, Wu YW. Predictors of outcome in perinatal arterial stroke:
a population-based study. Ann Neurol. 2005;58:303–308.
15. Caravale B, Allemand F, Libenson MH. Factors predictive of seizures and
neurologic outcome in perinatal depression. Pediatr Neurol. 2003;29:
16. Selton D, Andre M. Prognosis of hypoxic-ischaemic encephalopathy
in full-term newborns–value of neonatal electroencephalography.
17. Monagle P, Chalmers E, Chan A, DeVeber G, Kirkham F, Massicotte P,
Michelson AD; American College of Chest Physicians. Antithrombotic
therapy in neonates and children: American college of chest physicians
Evidence-based clinical practice guidelines (8thEdition). Chest. 2008;
18. Battin MR, Penrice J, Gunn TR, Gunn AJ. Treatment of term infants with
head cooling and mild systemic hypothermia (35.0 degrees C and 34.5
degrees C) after perinatal asphyxia. Pediatrics. 2003;111:244–251.
19. Shankaran S, Laptook AR, Ehrenkranz RA, Tyson JE, McDonald SA,
Donovan EF, Fanaroff AA, Poole WK, Wright LL, Higgins RD, Finer
NN, Carlo WA, Duara S, Oh W, Cotten CM, Stevenson DK, Stoll BJ,
Lemons JA, Guillet R, Jobe AH. Whole-body hypothermia for neonates with
hypoxic-ischemic encephalopathy. N Engl J Med. 2005;353:1574–1584.
by guest on June 1, 2013 http://stroke.ahajournals.org/Downloaded from
20. Ment LR, Bada HS, Barnes P, Grant PE, Hirtz D, Papile LA, Pinto-Martin
J, Rivkin M, Slovis TL. Practice parameter: neuroimaging of the neonate:
report of the Quality Standards Subcommittee of the American Academy
of Neurology and the Practice Committee of the Child Neurology Society.
21. Cowan F, Mercuri E, Groenendaal F, Bassi L, Ricci D, Rutherford M, de
Vries L. Does cranial ultrasound imaging identify arterial cerebral infarction
in term neonates? Arch Dis Child Fetal Neonatal Ed. 2005;90:F252–F256.
22. Golomb MR, Dick PT, MacGregor DL, Armstrong DC, deVeber GA.
Cranial ultrasonography has a low sensitivity for detecting arterial ische-
mic stroke in term neonates. J Child Neurol. 2003;18:98–103.
23. Aso K, Scher MS, Barmada MA. Neonatal electroencephalography and
neuropathology. J Clin Neurophysiol. 1989;6:103–123.
24. Barabas RE, Barmada MA, Scher MS. Timing of brain insults in severe
neonatal encephalopathies with isoelectric EEG. Pediatr Neurol. 1993;9:
25. American College of Obstetricians and Gynecologists (ACOG). Neonatal
encephalopathy and cerebral palsy: executive summary. Obstet Gynecol.
26. Lynch JK, Hirtz DG, DeVeber G, Nelson KB. Report of the National
Institute of Neurological Disorders and Stroke workshop on perinatal and
childhood stroke. Pediatrics. 2002;109:116–123.
27. Gu ¨nther G, Junker R, Sträter R, Schobess R, Kurnik K, Heller C, Kosch
A, Nowak-Göttl U. Symptomatic ischemic stroke in full-term neonates:
role of acquired and genetic prothrombotic risk factors. Stroke. 2000;31:
28. Shah PS, Ohlsson A, Perlman M. Hypothermia to treat neonatal hypoxic
ischemic encephalopathy: systematic review. Arch Pediatr Adolesc Med.
29. The National Institute of Neurological Disorders and Stroke rt-PA Stroke
Study Group. Tissue plasminogen activator for acute ischemic stroke.
N Engl J Med. 1995;333:1581–1587.
30. Gunn AJ, Gunn TR. The ‘pharmacology’ of neuronal rescue with cerebral
hypothermia. Early Hum Dev. 1998;53:19–35.
31. Jacobs S, Hunt R, Tarnow-Mordi W, Inder T, Davis P. Cooling for
newborns with hypoxic ischaemic encephalopathy. Cochrane Database
Syst Rev. 2003;4:CD003311.
32. Golomb MR, MacGregor DL, Domi T, Armstrong DC, McCrindle BW,
Mayank S, deVeber GA. Presumed pre- or perinatal arterial ischemic
stroke: risk factors and outcomes. Ann Neurol. 2001;50:163–168.
33. deVeber GA, MacGregor D, Curtis R, Mayank S. Neurologic outcome in
survivors of childhood arterial ischemic stroke and sinovenous
thrombosis. J Child Neurol. 2000;15:316–324.
34. Badawi N, Felix JF, Kurinczuk JJ, Dixon G, Watson L, Keogh JM,
Valentine J, Stanley FJ. Cerebral palsy following term newborn enceph-
alopathy: a population-based study. Dev Med Child Neurol. 2005;47:
35. Fenichel GM. Paroxysmal Disorders. Clinical Pedaitric Neurology. III
ed. Philadelphia: W.B. Saunders Company; 1997:1–46.
36. Randò T, Ricci D, Mercuri E, Frisone MF, Luciano R, Tortorolo G, Guzzetta
F. Periodic lateralized epileptiform discharges (PLEDs) as early indicator of
stroke in full-term newborns. Neuropediatrics. 2000;31:202–205.
37. Selton D, André M, Hascoët JM. [EEG and ischemic stroke in full-term
newborns]. Neurophysiol Clin. 2003;33:120–129.
38. Lanska MJ, Lanska DJ, Horwitz SJ, Aram DM. Presentation, clinical course,
and outcome of childhood stroke. Pediatr Neurol. 1991;7:333–341.
39. Aso K, Abdab-Barmada M, Scher MS. EEG and the neuropathology in
premature neonates with intraventricular hemorrhage. J Clin Neurophysiol.
40. Sinclair DB, Campbell M, Byrne P, Prasertsom W, Robertson CM. EEG
and long-term outcome of term infants with neonatal hypoxic-ischemic
encephalopathy. Clin Neurophysiol. 1999;110:655–659.
41. Ekert P, Perlman M, Steinlin M, Hao Y. Predicting the outcome of
postasphyxial hypoxic-ischemic encephalopathy within 4 hours of birth.
J Pediatr. 1997;131:613–617.
42. Holmes GL, Lombroso CT. Prognostic value of background patterns in
the neonatal EEG. J Clin Neurophysiol. 1993;10:323–352.
43. Ramaswamy V, Miller SP, Barkovich AJ, Partridge JC, Ferriero DM.
Perinatal stroke in term infants with neonatal encephalopathy. Neurology.
44. Scher MS, Beggarly M. Clinical significance of focal periodic discharges
in neonates. J Child Neurol. 1989;4:175–185.
45. Oliveira AJ, Nunes ML, Haertel LM, Reis FM, da Costa JC. Duration of
rhythmic EEG patterns in neonates: new evidence for clinical and prog-
nostic significance of brief rhythmic discharges. Clin Neurophysiol. 2000;
46. O’Brien MJ, Lems YL, Prechtl HF. Transient flattenings in the EEG of
newborns–a benign variation. Electroencephalogr Clin Neurophysiol.
47. Mizrahi EM, Kellaway P. Characterization and classification of neonatal
seizures. Neurology. 1987;37:1837–1844.
Rafay et al Predictors of Neonatal Stroke and HIE
by guest on June 1, 2013http://stroke.ahajournals.org/ Downloaded from