REVIEW ARTICLE

Prenatal, perinatal and neonatal risk factors for perinatal arterial

ischaemic stroke: a systematic review and meta-analysis
C. Lia,*, J. K. Miaob,*, Y. Xuc, Y. Y. Huaa, Q. Maa, L. L. Zhoua, H. J. Liua and Q. X. Chena

a
Department of Neonatology, Ministry of Education Key Laboratory of Child Development and Disorders, Children’s Hospital of
Chongqing Medical University, Chongqing; bChongqing International Science and Technology Cooperation Center for Child Development
and Disorders, Children’s Hospital of Chongqing Medical University, Chongqing; and cDepartment of Neonatology, Chongqing Health
Center for Women and Children, Chongqing, China

EUROPEAN JOURNAL OF NEUROLOGY

Keywords:
meta-analysis, neonatal
cerebral infarction,
perinatal arterial
ischaemic stroke, risk
factors, systematic
review
Received 21 February 2017
Accepted 1 May 2017
European Journal of
Neurology 2017, 0: 1–10
doi:10.1111/ene.13337
The aim of the present study was to perform a meta-analysis of published
data to determine the significance of clinical factors and exposures to the risk
of perinatal arterial ischaemic stroke (PAIS) and provide guidance for clinical
diagnosis and treatment. A comprehensive literature search of the PubMed,
Embase, MEDLINE and Cochrane Library databases for relevant observa-
tional studies (cohort/case control) from March 1984 to March 2016 was
undertaken. Two review authors independently examined the full text records
to determine which studies met the inclusion criteria and evaluated risk factors
for PAIS. Risk ratios, odds ratios and 95% confidence intervals were esti-
mated. A total of 11 studies were included in the analyses. Intrapartum fever
>38°C, pre-eclampsia, oligohydramnios, primiparity, forceps delivery, vacuum
delivery, fetal heart rate abnormalities, abnormal cardiotocography tracing,
cord abnormalities, birth asphyxia, emergency caesarean section, tight nuchal
cord, meconium-stained amniotic fluid, umbilical arterial pH <7.10, Apgar
score at 5 min <7, resuscitation at birth, hypoglycaemia, male gender and
small for gestational age were identified as risk factors for PAIS. This systemic
review and meta-analysis provides a preliminary evidence-based assessment of
the risk factors for PAIS. Patients with any of the risk factors identified in this
analysis should be given careful consideration to ensure the prevention of
PAIS. Future studies focusing on the combined effects of multiple prenatal,
perinatal and neonatal risk factors for PAIS are warranted.

28th postnatal day, confirmed by neuroimaging or

Introduction
Perinatal ischaemic stroke is a cerebrovascular disease
that results in ischaemic and haemorrhagic injury
around focal or multifocal cerebral vessels [1]. In
2006, perinatal ischaemic stroke was defined as ‘a
group of heterogeneous conditions in which there is
focal disruption of cerebral blood flow secondary to
arterial or cerebral venous thrombosis or emboliza-
tion, between 20 weeks of fetal life through to the
Correspondence: Q. X. Chen, Department of Neonatology, Ministry
of Education Key Laboratory of Child Development and Disorders,
Children’s Hospital of Chongqing Medical University, Chongqing
400014, China (tel.: +86-131-93039866; fax: +86-236-3626904;
e-mail: chengqixiong@126.com).
*
The first two authors contributed equally to this work.
neuropathological studies’ [2].
Perinatal ischaemic stroke can be divided into peri-
natal arterial ischaemic stroke (PAIS) and cerebral
sinovenous thrombosis (CSVT). The incidence of
PAIS ranges from 1 in 2300 to 1 in 5000 births [3–
9], while the incidence of CSVT is much lower, rang-
ing from 0.6 to 12 per 100 000 live births [10]. PAIS
is associated with substantial morbidity, including
long-term neurological disabilities, such as cerebral
palsy and epilepsy, and cognitive abnormalities
[3,11–16]. Several studies have identified potential
risk factors for PAIS, including oligohydramnios,
chorioamnionitis, premature rupture of membranes,
pre-eclampsia and low Apgar score at 1 or 5 min
[17–20]. Other reports suggest the risk of PAIS may

© 2017 EAN 1
2 C. LI ET AL.
increase in the presence of multiple risk factors
[19,21].
Numerous studies have focused on specific
antepartum, intrapartum and neonatal exposures as
possible risk factors for PAIS. Many support the
hypothesis that prenatal [e.g. primiparity, pre-
eclampsia, intrauterine growth restriction (IUGR),
reduced fetal movement], perinatal (e.g. chorioam-
nionitis, prolonged rupture of membranes, fetal
heart rate abnormality, emergency caesarean deliv-
ery) and neonatal (e.g. Apgar score ≤7 at 5 min,
resuscitation at birth, umbilical arterial pH ≤7.10,
hypoglycaemia) exposures increase the risk of PAIS
[22–25].
The aetiology and pathogenesis of PAIS is consid-
ered complex and multifactorial and is currently not
well understood. There is an unmet need for a system-
atic review and meta-analysis investigating prenatal,
perinatal and neonatal risk factors for PAIS to pro-
vide guidance for clinical diagnosis and treatment.
The aim of the present study was to perform a meta-
analysis of published data to determine the signifi-
cance of clinical factors and exposures on the risk of
PAIS.
Methods
Search strategy and selection criteria
Two review authors (Chun Li, Jingkun Miao) inde-
pendently searched the PubMed, MEDLINE, Embase
and Cochrane Library databases from March 1984 to
March 2016. Search terms included ‘neonatal stroke’,
‘neonatal arterial ischemic stroke’, ‘perinatal stroke’,
‘perinatal ischemic stroke’, ‘neonatal ischemic
stroke’, ‘fetal stroke’, ‘presumed prenatal or perinatal
arterial ischemic stroke’, ‘perinatal arterial ischemic
stroke’, ‘neonatal cerebral infarction’ and ‘risk fac-
tors’. The search was not restricted by language.
Inclusion and exclusion criteria
Perinatal arterial ischaemic stroke was defined as
stroke that occurred in utero or up to 28 days after
birth [19]. Inclusion criteria were (i) case control
studies (defined as incidence, prevalence or nested
case control studies) or cohort studies; (ii) on
PAIS, neonatal stroke or neonatal cerebral infarc-
tion; (iii) describing risk factors in the perinatal or
neonatal period; and (iv) reporting odds ratios
(ORs) in case control studies or relative risks
(RRs) in cohort studies with 95% confidence inter-
vals (CIs); or, if 95% CIs were not available, the
reported data were sufficient to calculate them.
Exclusion criteria were (i) studies that were not
published as full reports, (ii) studies without control
subjects or (iii) reviews, case reports or studies on
other factors or outcomes.
Study selection
Titles and abstracts were independently examined by
two review authors (CL, JM) to select eligible studies.
The full text of potentially relevant studies was
retrieved. The full text records were independently
examined by two review authors (CL, JM) to deter-
mine which studies met the inclusion criteria. Dis-
agreements about study selection were resolved by
discussion with a third review author (YH) until con-
sensus.
Data extraction
Data from eligible studies, including first author’s last
name, publication year, country where the study was
performed, study period, participants’ sex and age,
sample size (cases and controls or cohort size), mea-
sure and range of exposure, variables adjusted for in
the analysis and outcome measures, were indepen-
dently extracted by two review authors (CL, JM).
Outcome measures were OR or RR estimates with
corresponding 95% CIs. The primary analysis was
focused on ORs and RRs that were adjusted for
potential confounders. Study authors were contacted
to retrieve missing information, when necessary.
Disagreements about data extraction were resolved
by discussion with a third review author (YH) until
consensus.
Quality assessment
Methodological quality was independently assessed by
two review authors (CL, JM) using the New-
castle Ottawa Scale for case control or cohort stud-
ies as appropriate [26]. The scale assesses potential
selection bias, comparability of cases and controls or
cohorts, and ascertainment of outcome (case control
studies) or exposure (cohort studies). Points (or
‘stars’) are awarded for high-quality elements. Stars
are summed and used to compare study quality in a
quantitative manner, as recommended by the
Cochrane Non-Randomized Studies Methods Work-
ing Group. For the purpose of this analysis, studies
with ≥5 stars were considered high quality and were
included; those with <4 stars were considered low
quality and excluded. Disagreements about quality
assessment were resolved by discussion with a third
review author (YH) until consensus.
© 2017 EAN

Data analysis
Statistical analysis was performed using RevMan 5.2.
Study outcomes were pooled using the Mantel–Haen-
szel formula (fixed-effects model) or the Dersimonian–
Laird formula (random-effects model). Heterogeneity
amongst studies was evaluated using the chi-squared
test and inconsistency index (I2) [27]. A random-effects
model was used to pool studies with significant
heterogeneity (I2 ≥ 50%, P < 0.1). A fixed-effects
model was used to pool studies with no evidence of
heterogeneity (I2 < 50%, P > 0.1).
Beggar’s regression test was used to investigate pub-
lication bias. Sensitivity analysis involved comparing
fixed-effects and random-effects models.
Results
Study identification
This meta-analysis was conducted in accordance with
PRISMA guidelines (Appendix S1). The searches iden-
tified 375 potentially relevant articles; two conference-
related publications were identified from other sources.
Titles and abstracts were screened, and 33 studies were
considered potentially eligible for inclusion. Full text
articles were retrieved, and 22studies were excluded. Of
these, 13 studies did not include a comparison group,
four studies did not provide complete data (missing
information included risk factors, adjusted hazard
ratios or ORs, or contingency tables), three studies
focused on CSVT, one study included presumed perina-
tal stroke, and one study focused on periventricu-
lar intraventricular haemorrhage. Eleven studies were
found to be eligible for inclusion according to our crite-
ria for considering studies in this review (Fig. 1).
Characteristics of included studies
The characteristics of the 11 included studies are shown
in Table 1. Of these, three were from the USA, two
were from the UK, two were from France, one each
were from Denmark, Netherlands, Iran, and one was
published by the International Pediatric Stroke Study .
The studies included 32 potential risk factors for
PAIS (Table 2). Breech presentation, vaginal blood
loss, congenital heart diseases and umbilical venous
catheterization were not considered because the stud-
ies were small or too few (<5 patients or <2 studies).
Prenatal risk factors
The meta-analysis demonstrated that pre-eclampsia
(defined as a physician diagnosis of either
© 2017 EAN
RISK FACTORS FOR PAIS 3
pre-eclampsia or pregnancy-induced hypertension)
[5,19,23,28,29] (PAIS, n = 229; control, n = 1231; OR
1.92, 95% CI 1.17–3.13; Z = 2.60; P = 0.009; hetero-
geneity P = 0.20, I2 = 34%; Fig. 2) and primiparity
[19,22–24,28,29] (PAIS, n = 302; control, n = 1348;
OR 2.03, 95% CI 1.55–2.65; Z = 5.18; P < 0.001;
heterogeneity P = 0.09, I2 = 48%; Fig. 2) were signifi-
cant risk factors for PAIS. IUGR [manifested as neo-
nates with low birthweight or small for gestational
age (SGA)] [5,19,24,29] (PAIS, n = 138; control,
n = 505; OR 1.53, 95% CI 0.85 2.75; Z = 1.43;
P = 0.15; heterogeneity P = 0.160, I2 = 42%), gesta-
tional diabetes [5,19,24,30] (PAIS, n = 121; control,
n = 582; OR 1.55, 95% CI 0.79–3.04; Z = 1.27;
P = 0.20; heterogeneity P = 0.16, I2 = 42%), oxytocin
induction [19,23,24,29] (PAIS, n = 151; control,
n = 454; OR 1.31, 95% CI 0.85–2.03; Z = 1.22;
P = 0.22; heterogeneity P = 0.27, I2 = 24%) and pre-
vious miscarriage [19,22,23,29] (PAIS, n = 167; con-
trol, n = 545; OR 1.07, 95% CI 0.69–1.66; Z = 0.29;
P = 0.77; heterogeneity P = 0.50, I2 = 0%) were not
associated with PAIS.
Based on a limited number of studies (n < 4), the
meta-analysis indicated that oligohydramnios [19,24]
and intrapartum fever >38°C [22,23] were significant
risk factors for PAIS. Decreased fetal movement
(based on a maternal report of decreased fetal move-
ment before labour or a non-stress test) [19,22,29] was
not associated with PAIS.
Perinatal complications
The meta-analysis demonstrated vacuum delivery
[5,19,22,23] (PAIS, n = 206; control, n = 720; OR
1.69, 95% CI 1.11–2.57; Z = 2.44; P = 0.01; hetero-
geneity P = 0.61, I2 = 0%; Fig. 2), fetal heart rate
abnormalities [19,23,24,29] (PAIS, n = 152; control,
n = 455; OR 5.21, 95% CI 3.43–7.90; Z = 7.77;
P < 0.001; heterogeneity P = 0.73, I2 = 0%; Fig. 2),
prolonged second stage of labour (defined as a second
stage of labour longer than 2 h) [19,22–24] (PAIS,
n = 171; control, n = 588; OR 4.13, 95% CI 2.40–
7.10; Z = 5.12; P < 0.001; heterogeneity P = 0.40,
I2 = 0%; Fig. 2), emergency caesarean section [be-
cause of abnormal cardiotocography (CTG) tracing or
obstructed labour] [5,19,22–24,28,29] (PAIS, n = 361;
control, n = 1466; OR 4.14, 95% CI 2.49–6.90;
Z = 5.46; P < 0.001; heterogeneity P = 0.01,
I2 = 62.8%; Fig. 2) and meconium-stained amniotic
fluid [5,17,19,22–24] (PAIS, n = 250; control, n = 833;
OR 2.82, 95% CI 1.64–4.85; Z = 3.74; P < 0.001;
heterogeneity P = 0.04, I2 = 56.1%; Fig. 2) were sig-
nificant risk factors for PAIS. Prolonged rupture of
membranes [5,19,22–24,29,31] (PAIS, n = 359; control,
4 C. LI ET AL.

Records identified through database Additional records identified through

searching (n = 373) other sources (n = 2)

Records after duplicates

Records excluded for anmials (n = 2)
removed (n = 217)

Records excluded based on screening of

Records screened (n = 215) abstracts (n = 182)

Full-text articles excluded,with reasons (n = 22)

No comparison group (n = 13)

No outcome data (n = 4)

Had CSVT (n = 3)

Had PIVH (n = 1)
Full-text articles assessed
for eligibility (n = 33) Included presumed perinatal stroke (n = 1)

Studies included in
qualitative synthesis (n = 11)

Studies included in
quantitative synthesis
(meta-analysis)(n = 11)
Figure 1 Flowchart of article selection.

n = 978; OR 2.04, 95% CI 0.88–4.73; Z = 1.66;
P = 0.10; heterogeneity P = 0.001; I2 = 73.0%; Fig. 2)
was not associated with PAIS.
Based on a limited number of studies (n < 4), the
meta-analysis also indicated forceps delivery [5,19,22],
chorioamnionitis (defined as inflammation of the
fetal membranes) [5,19,24], abnormal CTG tracing
(including persistent late or variable decelerations,
fetal bradycardia and/or reduced fetal heart variabil-
ity) [17,22], cord abnormalities (including cord entan-
glements, hypercoiling, true knots, strictures and
short cords) [19,24], birth asphyxia (defined as a
reduction of serum oxygen levels and nutrient supply
to the vital organs of the neonate) [5,19] and tight
nuchal cord [5,22] were significant risk factors for
PAIS.
Newborn characteristics
The meta-analysis demonstrated umbilical arterial pH
<7.10 [5,22,23,29] (PAIS, n = 123; control, n = 334;
OR 8.41, 95% CI 1.75–40.39; Z = 2.66; P = 0.008;
heterogeneity P = 0.012, I2 = 72.4%; Fig. 2), Apgar
score at 5 min <7 [5,19,23,28,29] (PAIS, n = 229; con-
trol, n = 1227; OR 6.76, 95% CI 2.30–19.83;
Z = 3.48; P < 0.001; heterogeneity P = 0.012,
I2 = 68.8%; Fig. 2) and resuscitation at birth (includ-
ing intubation for ventilation with or without cardiac
compressions and epinephrine) [5,19,22,31,32] (PAIS,
n = 514; control, n = 46 189; OR 5.31, 95% CI 3.74–
7.53; Z = 9.33; P < 0.001; heterogeneity P = 0.62,
I2 = 0%; Fig. 2) were significant risk factors for
PAIS. Male gender [5,17,19,22,23,28,29] (PAIS,
n = 320; control, n = 1491; OR 1.27, 95% CI 0.99–
1.62; Z = 1.86; P = 0.06; heterogeneity P = 0.07,
I2 = 49.0%; Fig. 2) was not associated with PAIS.
Based on a limited number of studies (n < 4), the
meta-analysis also indicated hypoglycaemia [defined
as blood glucose <45 mg/dl (2.6 mmol/l) within the
first 2 days after birth] [23,29] and SGA (defined as a
birthweight below the 10th percentile of the birth-
weight distribution according to gestational age)
[22,28] were significant risk factors for PAIS. Gesta-
tional age >42 weeks [22,28] was not associated with
PAIS.
Quality assessment and publication bias
All studies included in our meta-analyses were consid-
ered high quality (Table 1). In all analyses, visual
assessment of a funnel plot and Beggar’s test showed
no evidence of publication bias (Fig. 3).

© 2017 EAN
 Table 1 Characteristics of included studies

© 2017 EAN
Study period Quality
Study (first author) (years) Country Study type Study design Case Control Neonatal score Risk factors reported

Estan [17] 1987–1993 UK Retrospective Case control 12 24 Term 6 Lower Apgar score, resuscitation
Darmency- 2000–2007 France Retrospective Case control 32 96 Term 5–7 Gestational diabetes, fetal heart rate abnormalities,
Stamboul [24] meconium-stained amniotic fluid, smoking during
pregnancy
Tuckuviene [28] 1994–2006 Denmark Retrospective Nested case control 71 657 Both 8 GA ≥ 42 weeks, SGA (≤10th percentile), Apgar at 5
min <7, emergency caesarean section, primiparity
Lee [19] 1997–2002 USA Retrospective Nested case control 37 111 Both 6–8 Primiparity, pre-eclampsia, oligohydramnios, decreased fetal
movement, chorioamnionitis,
prolonged rupture of membranes, prolonged second stage
of labour, fetal heart rate abnormality, cord abnormality,
vacuum, emergency caesarean delivery, diagnosis of birth
asphyxia, Apgar score at 5 min <7, resuscitation at birth,
history of infertility
Benders [29] 1990–2005 USA Retrospective Case control 31 93 Preterm 6–8 TTTS, reduced fetal movement, fetal heart rate
abnormalities, hypoglycaemia (<2 mmol/l)
Chabrier [31] 2003–2006 France Prospective Cohort study 100 100 Both 6 Tobacco consumption, caesarean section, low Apgar score,
resuscitation at birth, premature rupture of membranes
Wu [4] 1991–1998 USA Retrospective Nested case control 38 218 Term 5–7 IUGR, pre-eclampsia, chorioamnionitis, emergency
caesarean section, Apgar score at 5 min <7, umbilical
arterial pH 7.0, birth asphyxia, resuscitation at birth
Martinez- 1992–2012 UK Prospective Case control 79 239 Term 6–8 Family history of seizures, family history of
Biarge [22] neurological diseases, autoimmune disease, gynaecological
problems, primiparity, viral infection, abdominal pain,
prolonged rupture of membranes, maternal fever >38°C,
shoulder dystocia, prolonged second stage, tight nuchal
cord, unassisted vaginal, elective prelabour caesarean,
emergency caesarean, low Apgar score, arterial cord pH
<7.10
Harteman [23] 2000–2010 Netherlands Retrospective Case control 52 156 Term 7 Primiparity, intrapartum fever >38°C, meconium-stained
amniotic fluid, fetal heart decelerations, emergency
caesarean section, low Apgar score, arterial umbilical
cord pH <7.10, hypoglycaemia (<2 mmol/l), early-onset
sepsis/meningitis
Kirton [9] 2003–2007 IPSS Prospective Case control 248 Population Both 5–6 Emergency caesarean delivery, resuscitation, low Apgar
score
Farhadi [30] 2012–2103 Iran Prospective Cross-sectional study 14 160 Both 4–5 Umbilical venous catheterization

GA, gestational age; TTTS, twin-to-twin transfusion syndrome; IPSS, International Pediatric Stroke Study.
RISK FACTORS FOR PAIS
5
6 C. LI ET AL.

Table 2 Prenatal, perinatal and neonatal risk factors for PAIS

Complications Number of studies Cases of PAIS OR (95% CI)

Prenatal complications
Previous miscarriage 4 32 1.07 (0.69, 1.66)
Intrapartum fever >38°C 2 16 6.00 (2.58, 13.97)
Decreased fetal movement 3 25 3.61 (0.98, 13.29)
Gestational diabetes 3 10 1.55 (0.79, 3.04)
Oligohydramnios 2 7 3.77 (1.23, 11.62)
Oxytocin induction 4 46 1.31 (0.85, 2.03)
Pre-eclampsia 5 28 1.92 (1.17, 3.13)
IUGR 4 19 1.53 (0.85, 2.75)
Primiparity 6 199 2.03 (1.55, 2.65)
Perinatal complications
Forceps delivery 3 15 2.51 (1.25,5.02)
Abnormal CTG tracing 2 45 6.87 (3.81, 12.38)
Breech presentation 3 3 NA
Chorioamnionitis 3 16 3.41 (1.68, 6.92)
Fetal heart rate abnormalities 4 72 5.21 (3.43, 7.90)
Vacuum delivery 4 39 1.69 (1.11,2.57)
Cord abnormalities 2 17 2.93 (1.43, 6.00)
Birth asphyxia 2 9 44.04 (5.31, 364.99)
Prolonged second stage of labour 4 41 4.13 (2.40, 7.10)
Prolonged rupture of membranes 7 50 2.04 (0.88, 4.73)
Vaginal blood loss 1 2 NA
Emergency caesarean section 7 122 4.14 (2.49, 6.90)
Tight nuchal cord 2 14 2.50 (1.22, 5.14)
Meconium-stained amniotic fluid 6 79 2.82 (1.64, 4.85)
Umbilical venous catheterization 1 7 NA
Newborn characteristics
Male 7 179 1.27 (0.99, 1.62)
Umbilical arterial pH <7.10 4 37 8.41 (1.75, 40.39)
Resuscitation at birth 6 142 5.31 (3.74, 7.53)
Apgar at 5 min <7 5 41 6.76 (2.30, 19.83)
Gestational age >42 weeks 2 14 4.48 (0.63, 31.58)
Hypoglycaemia 2 28 6.85 (1.53, 30.60)
Congenital heart diseases 1 2 NA
SGA 2 24 2.29 (1.35, 3.88)

the placental bed that can reduce uteroplacental

Discussion
To the authors’ knowledge, this is the first systematic
review and meta-analysis investigating prenatal, peri-
natal and neonatal factors as risk factors for PAIS.
Published data on risk factors for stroke in the neona-
tal period are sporadic and non-uniform, making it
difficult for the scientific and clinical community to
interpret. This meta-analysis consolidated the evidence
on PAIS to provide guidance for clinical diagnosis
and treatment.
Prenatal risk factors for PAIS were intrapartum
fever >38°C, pre-eclampsia, oligohydramnios and
primiparity. PAIS may occur due to changes in the
maternal environment. Intrapartum fever is sugges-
tive of intrauterine inflammation [19,33,34] and the
presence of inflammatory cytokines, which may dam-
age the central nervous system in the fetus, resulting
in PAIS [35]. Pre-eclampsia is a vascular defect in
blood flow and may cause fetal hypoxia and PAIS
[36–39]. Several studies prove that pre-eclampsia is
related to maternal prothrombotic disorders, a
maternal history of thromboembolism, as well as
thrombotic lesions in the placenta [5,19,23]. Evidence
suggests that pre-eclampsia is associated with a vari-
ety of adverse pregnancy outcomes including IUGR,
neonatal encephalopathy, neonatal sinovenous throm-
bosis and fetal death [36,40–43].The current meta-
analysis suggests that pre-eclampsia may also be
associated with PAIS. Primiparity was strongly
related to PAIS [19,22]. In our study, the association
of primiparity and PAIS was significant. However, as
primiparity may predispose some women to experi-
ence more intrapartum complications, such as pro-
longed second stage of labour, primiparity may be
identified as a predictor rather than a risk factor for
PAIS [19].

© 2017 EAN

Risk Factors
Pre-eclampsia
Primiparity
Vacuum delivery
Fetal heart rate abnormalities
Prolonged second stage of labor
Prolonged rupture of membranes
Emergency caesarean section
Meconium-stained amniotic fluid
Resuscitation at birth
Gender (male)
Umb.art.pH 7.10
Apgar at 5 min 7
–5
Figure 2 Associations between prenatal complications, perinatal complications and newborn characteristics for PAIS.
Other perinatal risk factors associated with PAIS
were forceps delivery, vacuum delivery, fetal heart rate
abnormalities, abnormal CTG tracing, cord abnor-
malities, birth asphyxia, emergency caesarean section,
tight nuchal cord and meconium-stained amniotic
fluid. These complications are all indicative of hypox-
ic-ischaemic events, which may have resulted in a
moderate degree of perinatal asphyxia and PAIS
[5,17,19,22,24,31,32,44]. In accordance with our data,
a previous meta-analysis found an association between
hypoxia and PAIS [45]. Chorioamnionitis was also a
perinatal risk factor for PAIS. Chorioamnionitis may
result in thromboembolism in the placenta and
increase the risk for emboli in the fetal brain [23].
Newborn characteristics associated with PAIS were
umbilical arterial pH <7.10, Apgar score at 5 min <7,
resuscitation at birth, hypoglycaemia, male gender
and SGA. Umbilical arterial pH is an objective mea-
surement of the physical condition of newborns. An
umbilical arterial pH <7.2 may indicate fetal asphyxia
[46], and an umbilical arterial pH <7.00 is considered
severe fetal acidosis, which increases the risk of subse-
quent neurological dysfunction [46,47]. Our findings
suggest that fetal acidosis is associated with PAIS.
Subsequent studies are warranted to determine
whether respiratory or metabolic acidosis is the
© 2017 EAN
RISK FACTORS FOR PAIS 7
OR (95% Cl)
1.92 (1.17, 3.13)
2.03 (1.55, 2.65)
1.69 (1.11, 2.57)
5.21 (3.42, 7.90)
4.13 (2.40, 7.10)
2.04 (0.88, 4.73)
4.14 (2.49, 6.90)
2.82 (1.64, 4.85)
5.31 (3.74, 7.53)
1.27 (0.99, 1.62)
8.41 (1.75, 40.39)
6.76 (2.30, 19.83)
.1 1 5 10
contributing factor. Low Apgar scores often result in
a clinical diagnosis of birth asphyxia. Birth asphyxia
can lead to congestion, endothelial injury and
intravascular coagulation, with subsequent hypoxia-
ischaemia and PAIS. Hypoglycaemia is a common
metabolic condition in human infants [48]. Previous
studies support out findings that hypoglycaemia is a
risk factor for PAIS [23,29]. In full-term infants,
symptomatic hypoglycaemia is associated with bilat-
eral occipital infarction, [49] and in neonates hypogly-
caemia is associated with diffuse cortical and
subcortical white matter damage [50]. Male fetuses are
more susceptible to placental dysfunction and birth
complications [31]. Hormonal status may partly
explain this gender disparity in neonatal infants [2].
Several animal studies suggest that oestrogen has neu-
roprotective effects [51,52]. In a rat stroke model,
testosterone increased and oestrogen decreased gluta-
mate toxicity [52].
Intrauterine growth restriction was amongst the fac-
tors not associated with PAIS in the current meta-
analysis. This is in contrast to other studies, which
indicate IGUR is an independent risk factor for PAIS
[5]. However, in our study, SGA was a risk factor for
PAIS. A large proportion of SGA pregnancies are
IUGR and the major proportion of IUGR
8 C. LI ET AL.

0 SE(log[OR]) 0 SE(log[OR]) 0 SE(log[OR])

0.2 0.2 0.2

0.4 0.4 0.4

0.6 0.6 0.6

0.8 0.8 0.8

OR OR OR
1 1 1
0.01 0.1 1 10 100 0.01 0.1 1 10 100 0.01 0.1 1 10 100
Apgar at 5 min < 7 Fetal heart rate abnormalities Emergency caesarean section

0 SE(log[OR]) 0 SE(log[OR]) 0 SE(log[OR])

0.2 0.2
0.5

0.4 0.4
1

0.6 0.6

1.5
0.8 0.8

2 OR OR OR
1 1
0.01 0.1 1 10 100 0.01 0.1 1 10 100 0.01 0.1 1 10 100
Pre-eclampsia Meconium-stained amniotic fluid Gender (male)
SE(log[OR]) SE(log[OR])
0 SE(log[OR]) 0 0

0.1 0.2
0.5

0.2 0.4

0.3 0.6

1.5
0.4 0.8

OR OR OR
0.5 1 2
0.01 0.1 1 10 100 0.01 0.1 1 10 100 0.01 0.1 1 10 100
Primiparity Prolonged rupture of membranes Prolonged second stage of labor

SE(log[OR]) 0 SE(log[OR])
0 0 SE(log[OR])

0.2 0.2
0.5

0.4 0.4

0.6 0.6

1.5
0.8 0.8

OR OR OR
1 2 1
0.01 0.1 1 10 100 0.01 0.1 1 10 100 0.01 0.1 1 10 100
Vacuum delivery Umb.art.pH <7.10 Resuscitation at birth

Figure 3 Funnel plots.

pregnancies are SGA [53]. A larger sample size may air pollution, socioeconomic status, and the level of
find a significant association between IUGR and PAIS diagnosis and treatment of PAIS were not considered.
in future studies. In the current study, decreased fetal
movement was not associated with PAIS. This is sur-
Conclusion
prising as a history of decreased fetal movements has
been associated with bilateral parasagittal infarction In conclusion, this meta-analysis revealed that intra-
[54]. partum fever >38°C, pre-eclampsia, oligohydramnios,
This study was associated with several limitations. forceps delivery, vacuum delivery, fetal heart rate
First, the sample sizes in many of the analyses were abnormalities, abnormal CTG tracing, cord abnor-
small. Secondly, most of the included studies were ret- malities, birth asphyxia, emergency caesarean section,
rospective. Thirdly, the possibility of information and tight nuchal cord, meconium-stained amniotic fluid,
selection biases and unidentified confounders cannot umbilical arterial pH <7.10, Apgar score at 5 min <7,
be completely excluded because all the included stud- resuscitation at birth, hypoglycaemia and SGA are
ies were observational. Lastly, all studies included in risk factors for PAIS. As the pathogenesis of PAIS is
our meta-analysis were from high-income countries; multifactorial, it is likely that a combination of prena-
therefore, the effects of factors such as environmental tal, perinatal and neonatal risk factors is involved in

© 2017 EAN

PAIS aetiology. It is suggested that patients with any
of the risk factors identified in this preliminary analy-
sis should be given careful consideration to ensure the
prevention of PAIS. In infants with risk factors and
clinical symptoms (such as seizures), magnetic reso-
nance imaging scanning should be used as the first
line imaging modality for detection of PAIS. In the
future, large scale prospective clinical studies are
required to better understand the pathophysiological
mechanisms of PAIS and to identify potential
therapies.
Acknowledgements
Our sincere thanks are expressed to all authors whose
literatures have been cited in our study. Appreciation
is also expressed to Medjaden Bioscience Limited for
editing and proofreading this paper.
This study was supported by the Project of
Chongqing Science and Technology Commission
(cstc2016shmszx130001) and Key Project of Chongq-
ing National Health and Family Planning Commis-
sion (2012-2-102, 2016ZDXM014).
Disclosure of conflicts of interest
The funders/sponsors had no role in the design and
conduct of the study; collection, management, analysis
or interpretation of the data; preparation, review or
approval of the manuscript; and decision to submit
the manuscript for publication. The authors declare
that they have no competing interests.
Supporting Information
Additional Supporting Information may be found in
the online version of this article:
Appendix S1. PRISMA checklist.
References
1. Aden U. Neonatal stroke is not a harmless condition.
Stroke 2009; 40: 1948–1949.
2. Raju TN, Nelson KB, Ferriero D, Lynch JK; NICHD-
NINDS Perinatal Stroke Workshop Participants.
Ischemic perinatal stroke: summary of a workshop
sponsored by the National Institute of Child Health and
Human Development and the National Institute of Neu-
rological Disorders and Stroke. Pediatrics 2007; 120:
609–616.
3. Nelson KB, Lynch JK. Stroke in newborn infants. Lan-
cet Neurol 2004; 3: 150–158.
4. Wu YW, Lynch JK, Nelson KB. Perinatal arterial
stroke: understanding mechanisms and outcomes. Semin
Neurol 2005; 25: 424–434.
© 2017 EAN
RISK FACTORS FOR PAIS 9
5. Wu YW, March WM, Croen LA, Grether JK, Escobar
GJ, Newman TB. Perinatal stroke in children with
motor impairment: a population-based study. Pediatrics
2004; 114: 612–619.
6. Wu YW, Linda CE, Henning LH, et al. Neuroimaging
abnormalities in infants with congenital hemiparesis.
Pediatr Neurol 2006; 35: 191–196.
7. Hunt RW, Inder TE. Perinatal and neonatal ischaemic
stroke: a review. Thromb Res 2006; 118: 39–48.
8. Perlman JM. Brain injury in the term infant. Semin
Perinatol 2004; 28: 415–424.
9. Kirton A, deVeber G. Cerebral palsy secondary to
perinatal ischemic stroke. Clin Perinatol 2006; 33: 367–
386.
10. Berfelo FJ, Kersbergen KJ, van Ommen CH, et al.
Neonatal cerebral sinovenous thrombosis from symptom
to outcome. Stroke 2010; 41: 1382–1388.
11. Mercuri E, Rutherford M, Cowan F, et al. Early prog-
nostic indicators of outcome in infants with neonatal
cerebral infarction: a clinical, electroencephalogram, and
magnetic resonance imaging study. Pediatrics 1999; 103:
39–46.
12. Golomb MR, MacGregor DL, Domi T, et al. Presumed
pre- or perinatal arterial ischemic stroke: risk factors
and outcomes. Ann Neurol 2001; 50: 163–168.
13. deVeber GA, MacGregor D, Curtis R, Mayank S. Neu-
rologic outcome in survivors of childhood arterial
ischemic stroke and sinovenous thrombosis. J Child
Neurol 2000; 15: 316–324.
14. de Vries LS, Groenendaal F, Eken P, van Haastert IC,
Rademaker KJ, Meiners LC. Infarcts in the vascular
distribution of the middle cerebral artery in preterm and
fullterm infants. Neuropediatrics 1997; 28: 88–96.
15. Humphreys P, Whiting S, Pham B. Hemiparetic cerebral
palsy: clinical pattern and imaging in prediction of out-
come. Can J Neurol Sci 2000; 27: 210–219.
16. Koelfen W, Freund M, Varnholt V. Neonatal stroke
involving the middle cerebral artery in term infants: clin-
ical presentation, EEG and imaging studies, and out-
come. Dev Med Child Neurol 1995; 37: 204–212.
17. Estan J, Hope P. Unilateral neonatal cerebral infarction
in full term infants. Arch Dis Child Fetal Neonatal Ed
1997; 76: F88–F93.
18. Laugesaar R, Kolk A, Tomberg T, et al. Acutely and
retrospectively diagnosed perinatal stroke: a population-
based study. Stroke 2007; 38: 2234–2240.
19. Lee J, Croen LA, Backstrand KH, et al. Maternal and
infant characteristics associated with perinatal arterial
stroke in the infant. JAMA 2005; 293: 723–729.
20. Cheong JL, Cowan FM. Neonatal arterial ischaemic
stroke: obstetric issues. Semin Fetal Neonatal Med 2009;
14: 267–271.
21. Grunt S, Mazenauer L, Buerki SE, et al. Incidence and
outcomes of symptomatic neonatal arterial ischemic
stroke. Pediatrics 2015; 135: e1220–e1228.
22. Martinez-Biarge M, Cheong JL, Diez-Sebastian J, Mer-
curi E, Dubowitz LM, Cowan FM. Risk factors for
neonatal arterial ischemic stroke: the importance of the
intrapartum period. J Pediatr 2016; 173: 62–68.
23. Harteman JC, Groenendaal F, Kwee A, Welsing PM,
Benders MJ, de Vries LS. Risk factors for perinatal arte-
rial ischaemic stroke in full-term infants: a case control
study. Arch Dis Child Fetal Neonatal Ed 2012; 97: F411–
F416.
10 C. LI ET AL.
24. Darmency-Stamboul V, Chantegret C, Ferdynu C, et al.
Antenatal factors associated with perinatal arterial
ischemic stroke. Stroke 2012; 43: 2307–2312.
25. Ecury-Goossen GM, Raets MM, Lequin M, Feijen-
Roon M, Govaert P, Dudink J. Risk factors, clinical
presentation, and neuroimaging findings of neonatal per-
forator stroke. Stroke 2013; 44: 2115–2120.
26. Wells GA, Shea B, O’Connell D, et al. The New-
castle Ottawa Scale (NOS) for assessing the quality of
non-randomized studies in meta-analyses. Ottawa: Ottawa
Hospital Research Institute, http://www.ohri.ca/programs/
clinical_epidemiology/oxford.asp (accessed 18/10/2016).
27. Higgins JP, Thompson SG. Quantifying heterogeneity in
a meta-analysis. Stat Med 2002; 21: 1539–1558.
28. Tuckuviene R, Christensen AL, Helgested J, Hundborg
HH, Kristensen SR, Johnsen SP. Infant, obstetrical and
maternal characteristics associated with thromboem-
bolism in infancy: a nationwide population-based
case control study. Arch Dis Child Fetal Neonatal Ed
2012; 97: F417–F422.
29. Benders MJ, Groenendaal F, Uiterwaal CS, et al.
Maternal and infant characteristics associated with peri-
natal arterial stroke in the preterm infant. Stroke 2007;
38: 1759–1765.
30. Farhadi R, Alaee A, Alipour Z, Abbaskhanian A,
Nakhshab M, Derakhshanfar H. Prevalence of stroke in
neonates who admitted with seizures in neonatal inten-
sive care unit. Iran J Child Neurol 2015; 9: 41–47.
31. Chabrier S, Saliba E, Nguyen The Tich S, et al. Obstet-
rical and neonatal characteristics vary with birthweight
in a cohort of 100 term newborns with symptomatic
arterial ischemic stroke. Eur J Paediatr Neurol 2010; 14:
206–213.
32. Kirton A, Armstrong-Wells J, Chang T, et al. Symp-
tomatic neonatal arterial ischemic stroke: the Interna-
tional Pediatric Stroke Study. Pediatrics 2011; 128:
e1402–e1410.
33. Dueck CC, Grynspan D, Eisenstat DD, Caces R, Rafay
MF. Ischemic perinatal stroke secondary to chorioam-
nionitis: a histopathological case presentation. J Child
Neurol 2009; 24: 1557–1560.
34. Redline RW, Sagar P, King ME, Krishnamoorthy KS,
Grabowski EF, Roberts DJ. Case records of the Mas-
sachusetts General Hospital. Case 12-2008. A newborn
infant with intermittent apnea and seizures. N Engl J
Med 2008; 358: 1713–1723.
35. Dammann O, Leviton A. Infection remote from the
brain, neonatal white matter damage, and cerebral palsy
in the preterm infant. Semin Pediatr Neurol 1998; 5:
190–201.
36. Shah DM. Perinatal implications of maternal hyperten-
sion. Semin Pediatr Neurol 2001; 8: 108–119.
37. Roberts JM, Cooper DW. Pathogenesis and genetics of
pre-eclampsia. Lancet 2001; 357: 53–56.
38. Walker ID. Prothrombotic genotypes and pre-eclampsia.
Thromb Haemost 2002; 87: 777–778.
39. Kim YJ. Pathogenesis and promising non-invasive
markers for preeclampsia. Obstet Gynecol Sci 2013; 56:
2–7.
40. Badawi N, Kurinczuk JJ, Keogh JM, et al. Antepartum
risk factors for newborn encephalopathy: the Western
Australian case control study. BMJ 1998; 317: 1549–
1553.
41. Hunt RW, Badawi N, Laing S, Lam A. Pre-eclampsia: a
predisposing factor for neonatal venous sinus thrombo-
sis? Pediatr Neurol 2001; 25: 242–246.
42. Hnat MD, Sibai BM, Caritis S, et al. Perinatal outcome
in women with recurrent preeclampsia compared with
women who develop preeclampsia as nulliparas. Am J
Obstet Gynecol 2002; 186: 422–426.
43. Buchbinder A, Sibai BM, Caritis S, et al. Adverse peri-
natal outcomes are significantly higher in severe gesta-
tional hypertension than in mild preeclampsia. Am J
Obstet Gynecol 2002; 186: 66–71.
44. Laugesaar R, Kolk A, Tomberg T, et al. Acutely and
retrospectively diagnosed perinatal stroke: a population-
based study. Stroke 2007; 38: 2234–2240.
45. Luo L, Chen D, Qu Y, Wu J, Li X, Mu D. Association
between hypoxia and perinatal arterial ischemic stroke:
a meta-analysis. PLoS One 2014; 9: e90106.
46. Goldaber KG, Gilstrap LC 3rd, Leveno KJ, Dax JS,
McIntire DD. Pathologic fetal acidemia. Obstet Gynecol
1991; 78: 1103–1107.
47. Sehdev HM, Stamilio DM, Macones GA, Graham E,
Morgan MA. Predictive factors for neonatal morbidity
in neonates with an umbilical arterial cord pH less than
7.00. Am J Obstet Gynecol 1997; 177: 1030–1034.
48. Ennis K, Tran PV, Seaquist ER, Rao R. Postnatal age
influences hypoglycemia-induced neuronal injury in the
rat brain. Brain Res 2008; 1224: 119–126.
49. Filan PM, Inder TE, Cameron FJ, Kean MJ, Hunt RW.
Neonatal hypoglycemia and occipital cerebral injury.
J Pediatr 2006; 148: 552–555.
50. Yalnizoglu D, Haliloglu G, Turanli G, Cila A, Topcu
M. Neurologic outcome in patients with MRI pattern of
damage typical for neonatal hypoglycemia. Brain Dev
2007; 29: 285–292.
51. Toung TJ, Traystman RJ, Hurn PD. Estrogen-mediated
neuroprotection after experimental stroke in male rats.
Stroke 1998; 29: 1666–1670.
52. Yang SH, Perez E, Cutright J, et al. Testosterone
increases neurotoxicity of glutamate in vitro and ische-
mia-reperfusion injury in an animal model. J Appl Phys-
iol (1985) 2002; 92: 195–201.
53. Mars al K, Persson PH. Ultrasonic measurement of fetal
blood velocity waveform as a secondary diagnostic test
in screening for intrauterine growth retardation. J Clin
Ultrasound 1988; 16: 239–244.
54. Rutherford MA, Ramenghi LA, Cowan FM. Neonatal
stroke. Arch Dis Child Fetal Neonatal Ed 2012; 97:
F377–F384.
© 2017 EAN