Articles

Epilepsy in pregnancy and reproductive outcomes:

a systematic review and meta-analysis
Luz Viale, John Allotey, Fiona Cheong-See, David Arroyo-Manzano, Dougall Mccorry, Manny Bagary, Luciano Mignini, Khalid S Khan,
Javier Zamora, Shakila Thangaratinam, for EBM CONNECT Collaboration*

Summary
Background Antenatal care of women with epilepsy is varied. The association of epilepsy and antiepileptic drug Lancet 2015; 386: 1845–52
exposure with pregnancy outcomes needs to be quantified to guide management. We did a systematic review and Published Online
meta-analysis to investigate the association between epilepsy and reproductive outcomes, with or without exposure to August 26, 2015
http://dx.doi.org/10.1016/
antiepileptic drugs.
S0140-6736(15)00045-8
See Comment page 1804
Methods We searched MEDLINE, Embase, Cochrane, AMED, and CINAHL between Jan 1, 1990, and Jan 21, 2015, with
*Members listed at the end of
no language or regional restrictions, for observational studies of pregnant women with epilepsy, which assessed the risk the paper
of obstetric complications in the antenatal, intrapartum, or postnatal period, and any neonatal complications. We used Centro Rosarino de Estudios
the Newcastle-Ottawa Scale to assess the methodological quality of the included studies, risk of bias in the selection and Perinatales, Rosario, Argentina
comparability of cohorts, and outcome. We assessed the odds of maternal and fetal complications (excluding congenital (L Viale MD, L Mignini MD);
malformations) by comparing pregnant women with and without epilepsy and undertook subgroup analysis based on Women’s Health Research Unit,
Barts and the London School of
antiepileptic drug exposure in women with epilepsy. We summarised the association as odds ratio (OR; 95% CI) using Medicine and Dentistry, Queen
random effects meta-analysis. The PROSPERO ID of this Systematic Review’s protocol is CRD42014007547. Mary University of London,
London, UK (J Allotey MSc,
Findings Of 7050 citations identified, 38 studies from low-income and high-income countries met our inclusion F Cheong-See MBBS,
Prof K S Khan MSc,
criteria (39 articles including 2 837 325 pregnancies). Women with epilepsy versus those without (2 809 984 pregnancies) Prof J Zamora PhD,
had increased odds of spontaneous miscarriage (OR 1·54, 95% CI 1·02–2·32; I²=67%), antepartum haemorrhage Prof S Thangaratinam PhD);
(1·49, 1·01–2·20; I²=37%), post-partum haemorrhage (1·29, 1·13–1·49; I²=41%), hypertensive disorders (1·37, Hospital Ramón y Cajal,
1·21–1·55; I²=23%), induction of labour (1·67, 1·31–2·11; I²=64%), caesarean section (1·40, 1·23–1·58; I²=66%), any Madrid, Spain
(D Arroyo-Manzano MSc,
preterm birth (<37 weeks of gestation; 1·16, 1·01–1·34; I²=64%), and fetal growth restriction (1·26, 1·20–1·33; Prof J Zamora); Queen Elizabeth
I²=1%). The odds of early preterm birth, gestational diabetes, fetal death or stillbirth, perinatal death, or admission to Hospital Birmingham,
neonatal intensive care unit did not differ between women with epilepsy and those without the disorder. Birmingham, UK
(D Mccorry MD); Birmingham
and Solihull Mental Health
Interpretation A small but significant association of epilepsy, exposure to antiepileptic drugs, and adverse outcomes Foundation Trust,
exists in pregnancy. This increased risk should be taken into account when counselling women with epilepsy. Birmingham, UK
(M Bagary PhD);
Multidisciplinary Evidence
Funding EBM CONNECT Collaboration.
Synthesis Hub, Queen Mary
University of London, London,
Introduction antepartum and post-partum bleeding, caesarean section, UK (Prof K S Khan,
Epilepsy is one of the common chronic disorders affecting fetal growth restriction, and admission to neonatal Prof S Thangaratinam); and
CIBER Epidemiology and Public
women of reproductive age.1,2 Maternal mortality is intensive care unit (NICU).7–10 Existing guidelines on
Health, Madrid, Spain
ten-times higher in women with epilepsy than in those women with epilepsy have suggested a possible association (Prof J Zamora)
without the disorder.3 Care of women with epilepsy is between epilepsy and pregnancy complications on the Correspondence to:
fragmented, with few units providing joint obstetric– basis of a few small studies.11–13 Prof Javier Zamora, Hospital
epilepsy care.4 The Confidential Enquiries into Maternal We did a systematic review of medical literature by Universitario Ramón y Cajal,
28034 Madrid, Spain
and Child Health in the UK emphasised that epilepsy was collating the available evidence to generate precise
javier.zamora@hrc.es
not always perceived as a high-risk disorder in pregnancy.5 estimates of the association between epilepsy in
Adequate engagement with women with epilepsy is pregnancy, with or without exposure to antiepileptic
needed during preconception and pregnancy to plan drugs, and reproductive outcomes.
appropriate management.5
Quantification of the risks associated with pregnancy Data collection
in women with epilepsy is essential for appropriate Search strategy and study selection criteria
counselling and provision of care. Evidence tends to be We followed the present methods recommendations and
focused on fetal harm from in-utero exposure to used a prospective protocol14 to enable compliance with
antiepileptic drugs or on severity of maternal seizures, the Preferred Reporting Items for Systematic Reviews
with less emphasis on other pregnancy outcomes.6 and Meta-Analyses (PRISMA) guidelines for reporting
Individual studies provide varied and imprecise estimates this systematic review.14
of the association between epilepsy and pregnancy We searched MEDLINE, Embase, Cochrane, AMED,
complications such as miscarriage, preterm delivery, and CINAHL from Jan 1, 1990, to Jan 21, 2015, without

www.thelancet.com Vol 386 November 7, 2015 1845

 Articles
language restrictions with the following search terms:
“pregnan*”, “obstetric*” or “wom*” and combined with
“anti-epilep*”, “AED”, “anticonvuls*”, or “epilep*” to
represent the population, and combined them with
terms relevant to the outcomes such as “pregnan*
outcome*”, “adverse pregnan* outcome*”, “obstetric*
complicat*”, “pregnan* disorder*”, or “obstetric*
See Online for appendix outcome*” (appendix). We supplemented the searches
with manual searches of the reference lists of all included
primary studies and relevant review articles.
We selected studies for inclusion in a two-stage process.
First, two independent reviewers (LV and JA) screened the
titles and abstracts to identify eligible studies; this
screening was followed by retrieval and assessment of the
full texts of the relevant citations. Any disagreements were
resolved after discussion with a third reviewer (FC-S).
We included studies of pregnant women with epilepsy,
which assessed the risk of obstetric complications in
the antenatal, intrapartum, or postnatal period, and
any neonatal complications. The following maternal
outcomes were assessed: spontaneous miscarriage,
induction of labour, caesarean section, antepartum
haemorrhage, post-partum haemorrhage, any preterm
birth (<37 weeks of gestation) or early preterm birth
(<34 weeks of gestation), hypertensive disorders, and
gestational diabetes. The following fetal and neonatal
outcomes were assessed: fetal death, fetal growth
restriction that included both low birthweight and small
for gestational age babies, admission to the NICU, and
perinatal death. For all outcomes, we used the definitions
provided by the investigators. Low birthweight was
defined as birthweight of less than 2·5 kg in the studies.
When data were provided for both low birthweight and
small for gestational age fetus, we chose to include data
for small for gestational age fetus. We excluded studies
that only assessed the risk of congenital abnormalities
in the fetus, seizure deterioration in pregnancy, or
long-term adverse maternal and child outcomes, without
providing any data on reproductive outcomes. We
excluded abstracts, case reports, case series, in-vitro
studies, and animal studies.
Quality assessment and data extraction
We used the Newcastle-Ottawa Scale to assess the
methodological quality of the included studies, risk of
bias in the selection and comparability of cohorts, and
outcome.15 Two independent reviewers (LV and JA)
undertook quality assessment and allocated stars for
adherence to the prespecified criteria. Studies that scored
four stars for selection, two stars for comparability, and
three stars for ascertainment of the outcome were
regarded to have a low risk of bias. Studies with two or
three stars for selection, one for comparability, and two
for outcome ascertainment were deemed to have a
medium risk of bias. Any study with a score of one for
selection or outcome ascertainment, or zero for any of the
three domains, was deemed to have a high risk of bias.
1846
Data were extracted in duplicate (LV and JA) with
predesigned data extraction forms. Dichotomous data
were extracted as 2 × 2 tables and continuous data as
means and SDs. We contacted authors of potentially
eligible manuscripts by email for relevant data. When the
same cohort was reported twice, we extracted data from
the most recent study with the largest sample size for
relevant outcomes.
Outcomes
The main outcomes were maternal, fetal, and neonatal
complications.
Statistical analysis
We compared the odds of maternal and perinatal
complications in pregnant women in the following
groups: women with epilepsy versus women without
epilepsy, antiepileptic drug treatment versus no
treatment in women with epilepsy, and antiepileptic
drug polytherapy versus monotherapy in women with
epilepsy. Pregnancy was the unit of analysis. We reported
the results obtained after pooling individual study
estimates with a random effects meta-analysis as odds
ratio (OR) with 95% CIs. Heterogeneity was assessed
with the I² statistic.
We undertook subgroup analyses planned a priori for
any differences in the association of epilepsy and
pregnancy outcomes based on the following factors:
economic status of the country (developed or developing
countries according to International Monetary Fund),
year of publication (before 2000 or in 2000 and after),
study quality (high or low), and proportion of women
with epilepsy who were exposed to antiepileptic drugs
(<30% or ≥30%). We created funnel plots by plotting the
natural logarithm of the ORs against the inverse of the
standard error to assess publication and related bias. We
statistically checked for the asymmetry of the funnel plot
by using Egger’s method. All analyses were done using
Revman statistical software and Stata 13.0.16,17 The
PROSPERO ID of this Systematic Review’s protocol is
CRD42014007547.
Role of the funding source
The funders of this study had no role in study design,
data collection, data analysis, data interpretation, or
writing of the report. The corresponding author had full
access to all the data in the study and had final
responsibility for the decision to submit for publication.
Results
Of the 7050 citations identified, we selected 89 abstracts
for detailed assessment (figure 1). 38 studies published
in 39 articles,7–10,18–52 including 2 837 325 pregnancies,
met our inclusion criteria.
31 studies (2 809 984 pregnancies)7–10,18–20,22–25,27–29,31–33,36–47,50–52
provided rates of adverse outcomes for women with
epilepsy compared with those without epilepsy.
www.thelancet.com Vol 386 November 7, 2015
 Articles

11 studies (934 443 pregnancies)7,9,10,19,20,27,30,33,44,46,51,52 pro-

vided relevant outcome data for pregnant women with 7050 citations identified
7031 electronic databases
epilepsy exposed to antiepileptic drugs compared with 19 reference lists
those not exposed to antiepileptic drugs, and
eight studies (839 380 pregnancies)7,10,20,26,30,31,44,46 provided
data for women with epilepsy exposed to antiepileptic 6961 articles excluded
711 duplicates
drug polytherapy compared with those exposed to 6250 exclusion criteria
antiepileptic drug monotherapy. Of the 38 primary
studies, 13 studies21,26,27,29,31,34–36,42–44,48,49 were prospective
and 25 were retrospective.7–10,18,19,22–25,28,30,32,33,37–41,45–47,50–52 Data 89 articles retrieved for full assessment
were obtained from population-based cohorts in
21 studies8,9,19–21,28,30–32,34,36,37,40–45,47–49 and from registry data in
50 articles excluded
17 studies.7,10,18,22–27,29,33,35,38,39,46,50–52 6 reviews
The definition of epilepsy and the type of exposure 16 data cannot be extracted
2 inappropriate outcome
to antiepileptic drugs varied between the studies. 8 inappropriate population
32 (84%)7,9,10,18–21,23–30,32,34–38,40–46,48–52 of the 38 studies included 18 estimates of prevalence
women exposed to antiepileptic drugs such as carb-
amazepine, topiramate, lamotrigine, valproate, and
38 final studies included* (39 articles, 2 837 325 pregnancies)
phenytoin. Very few studies provided rates of maternal 31 epilepsy versus no epilepsy studies (2 809 984 pregnancies)
and fetal outcomes for the individual drugs. 11 antiepileptic drug in pregnancy versus no antiepileptic drug in
women with epilepsy studies (934 443 pregnancies)
29 (76%)7–10,19–22,24–36,39,44–48,50–52 of the 38 included studies were 8 antiepileptic drug monotherapy versus antiepileptic drug
published after 2000 and 28 (74%)7–10,19–22,24,26,27,29–35,37,38,41–43,46–49,51,52 polytherapy in women with epilepsy studies (839 380 pregnancies)
were done in a developed country setting (appendix).
Quality assessment by the Newcastle-Ottawa Scale Figure 1: Study selection process in the systematic review
showed that 23 (61%)7,9,19,20,22,25–28,30,33–36,38–40,43,45,49–52 of the *References 19 and 20 reported different outcomes from the same
38 included studies had low or medium risk of bias population-based cohort study.
(figure 2, appendix). 33 (87%)7–10,18–23,25–28,30,33–46,48–52 of the
38 studies had low risk of bias for study selection, one24 100 Risk of bias
had medium risk (3%), and four29,31,32,47 (11%) had high 2
Low
4
risk of bias. For comparability of the cohorts, 1 Medium
High
15 (40%) studies7,9,19,20,22,25,27,28,30,33,34,36,40,50–52 had low risk, 1
seven (18%)26,35,38,39,43,45,49 had medium risk, and
80 16
16 (42%)8,10,18,21,23,24,29,31,32,37,41,42,44,46–48 had high risk of bias. The
risk of bias for outcome assessment was low in 35 (92%)
studies,7–10,18–28,30–39,41–46,48–52 medium in one (3%) study,40 and
high in two (5%) studies.29,47
Proportion of studies (%)

The odds of the following maternal outcomes were 60

increased in pregnant women with epilepsy compared

with those without epilepsy: spontaneous miscarriage, 7
antepartum haemorrhage, post-partum haemorrhage, 35
33
hypertensive disorders, induction of labour, caesarean 40
section, and any preterm birth before 37 weeks of
gestation. No differences were reported between the
two groups for early preterm birth or gestational diabetes
(figure 3). 20 15
Fetal and neonatal outcomes were assessed in
21 studies.8–10,18,22–25,27,32,33,38–41,43,44,46,47,50,52 The odds of delivering
a baby with fetal growth restriction were increased in
women with epilepsy compared with women without
0
epilepsy (figure 3). Other infant outcomes such as fetal Selection Comparability Outcome
death, perinatal death, and admission to the NICU were
Figure 2: Quality assessment using the Newcastle-Ottawa Scale for risk of
not increased. bias of studies included in the systematic review
11 studies (934 443 pregnancies)7,9,10,19,20,27,30,33,44,46,51,52 The absolute numbers of studies are shown in boxes.
examined the association between antiepileptic drug
exposure in pregnant women with epilepsy and significantly higher in pregnant women with epilepsy
maternal and fetal outcomes (figure 4A). The odds of exposed to antiepileptic drugs compared with those not
post-partum haemorrhage and induction of labour were given drugs. No significant differences were reported

www.thelancet.com Vol 386 November 7, 2015 1847

 Articles

Studies (n) Events (n) Pregnancies (n) Odds ratio (95% CI) p value I2 (%)

Maternal outcomes
Spontaneous miscarriage 6 110 340 842 610 1·54 (1·02–2·32) 0·04 67
Antepartum haemorrhage 10 8090 670 100 1·49 (1·01–2·20) 0·04 37
Post-partum haemorrhage 7 107 905 952 477 1·29 (1·13–1·49) 0·0002 41
Induction of labour 9 67 474 473 006 1·67 (1·31–2·11) <0·0001 64
Caesarean section 20 162 937 1 126 069 1·40 (1·23–1·58) <0·00001 66
Any preterm birth (<37 weeks) 19 100 104 1 525 237 1·16 (1·01–1·34) 0·03 64
Early preterm birth (<34 weeks) 3 12 292 422 421 1·96 (0·97–3·99) 0·06 69
Hypertensive disorders 17 57 870 1 131 629 1·37 (1·21–1·55) <0·00001 23
Gestational diabetes 5 12 004 322 790 0·99 (0·53–1·83) 0·96 73

Fetal and neonatal outcomes

Fetal death or stillbirth 8 4473 569 231 1·27 (0·73–2·20) 0·39 52
Perinatal death 5 2633 261 345 1·83 (0·79–4·25) 0·16 39
Fetal growth restriction 18 123 386 1 508 306 1·26 (1·20–1·33) <0·00001 1
Admission to neonatal 3 37 810 411 375 2·06 (0·97–4·37) 0·06 88
intensive care unit
0·3 0·6 1 2 4 6 8 10

Favours women Favours women

with epilepsy without epilepsy

Figure 3: Association between maternal epilepsy diagnosis and adverse maternal and fetal outcomes

A Studies (n) Events (n) Pregnancies (n) Odds ratio (95% CI) p value I2 (%)

Maternal outcomes
Spontaneous miscarriage 3 1417 10 327 1·30 (0·61–2·79) 0·50 32
Antepartum haemorrhage 2 77 3262 1·21 (0·46–3·21) 0·7 70
Post-partum haemorrhage 2 949 5666 1·33 (1·16–1·54) <0·0001 0
Induction of labour 4 652 3604 1·40 (1·05–1·85) 0·02 28
Caesarean section 4 1244 6272 1·25 (0·96–1·61) 0·09 60
Any preterm birth (<37 weeks) 7 1044 12 779 1·12 (0·94–1·32) 0·20 12
Early preterm birth (<34 weeks) 2 123 3010 1·35 (0·93–1·94) 0·11 0
Hypertensive disorders 5 425 7043 1·03 (0·69–1·55) 0·88 63

Fetal and neonatal outcomes

Fetal death or stillbirth 4 39 3888 0·98 (0·37–2·64) 0·97 15
Fetal growth restriction 6 1103 9649 3·51 (1·23–10·01) 0·02 77
Admission to neonatal 1 364 2861 1·42 (1·13–1·78) 0·002 ··
intensive care unit
0·3 0·6 1 2 4 6 8 10

Favours women Favours women

with epilepsy on with epilepsy not
antiepileptic drugs on antiepileptic drugs

B Studies (n) Events (n) Pregnancies (n) Odds ratio (95% CI) p value I2 (%)

Maternal outcomes
Spontaneous miscarriage 3 44 1099 0·79 (0·40–1·56) 0·5 0
Antepartum haemorrhage 1 39 335 1·01 (0·46–2·22) 0·99 ··
Post-partum haemorrhage 2 206 1045 0·91 (0·57–1·46) 0·7 0
Induction of labour 3 319 1380 1·36 (0·97–1·90) 0·07 0
Caesarean section 4 343 1497 1·47 (1·07–2·02) 0·02 1
Any preterm birth (<37 weeks) 3 174 1511 1·47 (0·91–2·38) 0·11 20
Hypertensive disorders 2 39 756 0·76 (0·36–1·60) 0·47 0

Fetal and neonatal outcomes

Fetal death or stillbirth 3 15 1523 1·37 (0·34–5·48) 0·65 0
Fetal growth restriction 3 141 1224 1·44 (0·60–3·47) 0·41 54
Admission to neonatal 1 171 1023 1·45 (0·92–2·27) 0·11 ··
intensive care unit

0·3 0·6 1 2 4 6 8 10

Favours women Favours women

with epilepsy on with epilepsy on
antiepileptic drug antiepileptic drug
·· = not applicable polytherapy monotherapy

Figure 4: Association between exposure to antiepileptic drugs and pregnancy outcomes

(A) Exposure to antiepileptic drugs versus no exposure. (B) Exposure to antiepileptic drug polytherapy versus monotherapy.

1848 www.thelancet.com Vol 386 November 7, 2015

 Articles

Country status Year of publication Risk of bias Antiepileptic drug exposure

Developed Developing p value Published Published p value Low risk High risk p value ≥30% of <30% of p value
countries countries in 2000 or before 2000 women women
after
Maternal outcome
Spontaneous 1·27 2·08 0·37 1·51 1·69 0·77 1·20 1·97 0·08 1·86 1·11 0·002
miscarriage (0·88–1·44) (0·76–5·68) (0·87–2·61) (0·97–2·96) (0·77–1·87) (1·42–2·73) (1·35–2·57) (1·05–1·18)
Antepartum 1·65 1·44 0·9 1·67 1·93 0·76 2·04 1·52 0·62 1·73 0·65 0·34
haemorrhage (1·19–2·30) (0·18–11·71) (0·71–3·93) (1·24–3·01) (0·77–5·37) (0·83–2·80) (1·07–2·82) (0·09–4·62)
Post-partum 1·31 0·82 0·36 1·23 1·96 0·007 1·23 1·36 0·65 1·31 0·91 0·59
haemorrhage (1·14–1·51) (0·31–2·20) (1·14–1·32) (1·40–2·74) (1·11–1·36) (0·90–2·06) (1·12–1·53) (0·24–3·46)
Induction of labour 1·66 1·87 0·76 1·66 1·93 0·77 1·62 1·85 0·46 1·65 1·89 0·74
(1·28–2·14) (0·90–3·88) (1·29–2·13) (0·69–5·39) (1·24–2·13) (1·50–2·27) (1·28–2·15) (0·89–4·01)
Caesarean section 1·43 1·28 0·55 1·34 1·61 0·17 1·36 1·44 0·70 1·39 1·58 0·48
(1·28–1·61) (0·91–1·81) (1·15–1·54) (1·29–2·03) (1·10–1·69) (1·20–1·73) (1·22–1·59) (1·12–2·23)
Any preterm birth 1·23 0·81 0·06 1·18 1·01 0·51 1·36 1·02 0·01 1·13 1·27 0·47
(<37 weeks) (1·08–1·40) (0·53–1·24) (1·02–1·37) (0·66–1·56) (1·19–1·55) (0·86–1·22) (0·96–1·34) (0·98–1·64)
Early preterm birth 1·96 ·· NA 2·22 2·34 0·95 2·22 2·34 0·95 1·96 ·· NA
(<34 weeks) (0·97–3·99) (0·52–9·41) (1·15–4·74) (0·52–9·41) (1·15–4·74) (0·97–3·99)
Hypertensive disorders 1·32 1·47 0·8 1·35 1·54 0·36 1·32 1·41 0·7 1·42 1·15 0·34
(1·21–1·45) (0·68–3·15) (1·17–1·56) (1·21–1·97) (1·01–1·72) (1·19–1·66) (1·23–1·63) (0·86–1·53)
Gestational diabetes 1·40 0·60 0·06 0·99 ·· NA 0·98 0·96 0·98 0·84 1·31 0·48
(0·96–2·05) (0·26–1·35) (0·53–1·83) (0·31–3·04) (0·45–2·06) (0·31–2·30) (0·62–2·79)
Fetal and neonatal outcome
Fetal death or stillbirth 1·22 2·73 0·35 1·08 13·58 0·001 1·33 1·25 0·92 1·35 0·99 0·64
(0·88–1·69) (0·52–14·27) (0·82–1·44) (2·96–62·22) (0·54–3·29) (0·61–2·57) (0·72–2·56) (0·32–3·09)
Perinatal death 1·39 15·45 0·04 1·07 3·47 0·1 2·27 1·86 0·83 2·17 1·41 0·59
(0·72–2·67) (1·71–139·37) (0·48–2·40) (1·09–11·08) (0·75–6·87) (0·44–7·81) (0·65–7·27) (0·52–3·78)
Fetal growth restriction 1·27 1·03 0·36 1·27 1·02 0·24 1·26 1·25 0·94 1·27 1·22 0·65
(1·20–1·33) (0·68–1·58) (1·21–1·33) (0·71–1·47) (1·13–1·40) (1·13–1·39) (1·18–1·36) (1·06–1·41)
Admission to neonatal 2·12 1·74 0·85 1·41 3·33 <0·0001 3·23 1·41 <0·0001 2·12 1·74 0·85
intensive care unit (0·91–4·93) (0·28–10·83) (1·26–1·57) (2·23–4·98) (2·18–4·79) (1·26–1·57) (0·91–4·93) (0·28–10·83)

Data are odds ratio (OR) and ranges in brackets are 95% CIs, unless stated otherwise. NA=calculation of p value is not applicable.

Table: Subgroup analysis according to country status, year of publication, risk of bias, and antiepileptic drug exposure for maternal and fetal outcomes in women with epilepsy
compared with women without epilepsy

between the two groups for caesarean section,
antepartum haemorrhage, spontaneous miscarriage,
any preterm births before 37 weeks of gestation, and
hypertensive disorders.
No differences were noted between the two groups of
women with epilepsy, exposed or not exposed to
antiepileptic drugs, for fetal death or stillbirth
(figure 4A). Admission to the NICU was reported in one
study,46 which showed an increase in the odds in women
exposed to antiepileptic drugs. The odds of fetal growth
restriction were significantly higher in pregnant women
with epilepsy exposed to antiepileptic drugs than in
those not exposed.
Eight studies (839 380 pregnancies)7,10,20,26,30,31,44,46 assessed
the association between exposure to more than one
antiepileptic drug in pregnancy (polytherapy) and
maternal and fetal complications, compared with
monotherapy (figure 4B). Caesarean section was
increased in women given polytherapy compared with
those given monotherapy. No differences were noted
between the two groups in the odds of induction of
labour, any preterm births before 37 weeks of gestation,
antepartum haemorrhage, and post-partum haemorrhage.
Polytherapy was not associated with any fetal or
neonatal outcomes such as fetal death, admission to the
NICU, or fetal growth restriction (figure 4B).
Subgroup analysis based on country economic status,
publication year, risk of bias, and proportion of women
with epilepsy exposed to antiepileptic drugs did not
show significant differences between women with
epilepsy and those without epilepsy for caesarean
section, antepartum haemorrhage, induction of labour,
early preterm births before 34 weeks of gestation,
hypertensive disorders, gestational diabetes, and
fetal growth restriction (table). Significant differences
were reported between the groups based on the risk of
bias for admission to the NICU (p<0·0001) and any
preterm birth before 37 weeks of gestation (p=0·01),
publication year for post-partum haemorrhage
(p=0·007), admission to the NICU (p<0·0001), and
fetal death or stillbirth (p=0·001), the country status
for perinatal death (p=0·04), and exposure to
antiepileptic drugs for spontaneous miscarriage
(p=0·002).

www.thelancet.com Vol 386 November 7, 2015 1849

 Articles
We noted no differences in the odds of any adverse
outcome between the subgroups based on risk of bias,
country status, or year of publication for exposure to
antiepileptic drugs in pregnant women with epilepsy and
for polytherapy with antiepileptic drugs (table).
Funnel plot asymmetry was assessed for the outcomes
that had at least ten studies. We did not observe
evidence of small studies effect for any of the outcomes
and epilepsy, except for any preterm birth, which
showed a slightly asymmetrical funnel plot (p=0·039;
table).
Discussion
Our meta-analysis has provided precise quantitative
estimates of the magnitude of association between
epilepsy in pregnancy, exposure to antiepileptic drugs,
and various maternal and fetal outcomes. In pregnant
women, a diagnosis of epilepsy is associated with a small
but significant increase in adverse pregnancy outcomes
such as antepartum and post-partum haemorrhage,
spontaneous miscarriage, hypertensive disorders,
induction of labour, caesarean section, any preterm birth,
and fetal growth restriction. In women with epilepsy,
exposure to antiepileptic drugs is associated with an
increase in the odds of post-partum haemorrhage,
induction of labour, fetal growth restriction, and
admission to the NICU. Our review has provided much
needed information about epilepsy and obstetric
outcomes and will help in the counselling of women
with epilepsy and their partners, during the preconceptual
and antenatal period.
To our knowledge, our review is the first comprehensive
assessment of the association between epilepsy and
antiepileptic drug exposure and pregnancy outcomes by
meta-analyses. We did a detailed literature search without
language restrictions, thereby increasing our potential to
capture all relevant studies. The review was done with a
prospective protocol, and we prespecified the relevant
subgroups to explore the sources of heterogeneity. We
assessed study quality in detail and the effect of study
quality on the results. We were able to provide results
with high precision for epilepsy and outcomes owing to
the large sample size.
The studies varied in characteristics of the population;
cause of epilepsy; details of antiepileptic drugs therapy
such as type, dose, and compliance; and the definition
of outcomes. Because of the inconsistency and paucity of
reporting, we were unable to explore the effects of
seizures, parity, smoking status, congenital abnormalities
in the fetus, type and dose of antiepileptic drug, and
underlying medical disorders on the results. Fewer
studies were published about exposure to antiepileptic
drugs than epilepsy and pregnancy outcomes, which
contributed to a reduced precision in the findings for this
group. Evidence to provide estimates of adverse outcomes
for various individual antiepileptic drugs and their
dosage is insufficient.
1850
Recommendations for the care of mothers with
epilepsy are based on few observational studies.11–13 In
the UK, the National Institute for Health and Care
Excellence guidance reported on the increased risk of
preterm delivery, caesarean section, and induction of
labour, and no changes in the rates of perinatal deaths in
women with epilepsy based on the findings of three
individual studies.12 In the USA, the American Academy
of Neurology recommendations have suggested a
probable increase in the rates of caesarean section and
fetal growth restriction in women with epilepsy and in
those exposed to antiepileptic drugs.13 All guidelines
have consistently emphasised the absence of robust
evidence in this field.
A diagnosis of epilepsy and exposure to antiepileptic
drugs were significantly associated with increased risk
of induction of labour and caesarean section.7,20,29,46
The proportion of women with substantial seizure
deterioration needing delivery is too low to account for
this occurrence. Many other factors such as the
perception of epilepsy as a chronic disease, uncertainty
in management, an increase in rates of antepartum
haemorrhage, hypertensive disorders and fetal growth
restriction, and a low threshold for intervention might
have contributed to this rise.7,20 The observed increase
in hypertensive disorders in pregnant women with
epilepsy was consistent with the findings of the
individual studies that adjusted for maternal education,
smoking, diabetes, underlying medical disorder, and
maternal age.53
We did not note any association between the presence
of epilepsy or antiepileptic drug exposure with fetal and
neonatal mortality, despite including pregnancies with
congenitally malformed fetuses. The odds will probably
be even lower if these pregnancies with congenital
malformation of the fetus were excluded. The increased
complications such as stillbirth and NICU admission in
the pre-2000 published studies could be attributed to the
adverse effects of older antiepileptic drugs, improvements
in clinical care since then, or to the poor quality of
reporting in studies published before 2000.
On the basis of our findings, women with epilepsy,
and those given antiepileptic drugs, should be informed
that a small but significant risk of obstetric complications
can occur. Regular monitoring of pregnant women with
epilepsy in the antenatal period is essential for early
detection of hypertensive disorders and growth-
restricted fetuses.
Prospective studies and registries need to focus
on pregnancy outcomes in addition to congenital
malformations. Further assessment is needed on the
relation between seizure control in pregnancy, individual
antiepileptic drugs and their dosage, and pregnancy
complications.
Health-care professionals should incorporate the
estimates for various reproductive outcomes quantified
by our review, when counselling women with epilepsy.
www.thelancet.com Vol 386 November 7, 2015

Contributors
ST was involved in the conception of the research question and
designed the protocol. LV and JA did the literature search, study
selection, and data extraction, with the help of FC-S. JZ and DA-M
did the statistical analysis. The tables, figures, and appendices were
designed by JA and LV. The initial drafts of the manuscript were
prepared by ST, JA, and LV, with additional input from KSK, DM,
and MB. All authors contributed to the drafts and final version of
the manuscript.
Declaration of interests
MB reports grants and personal fees from Eisai and personal fees from
UCB, outside the submitted work. All other authors declare no
competing interests.
EBM CONNECT Collaboration
Luciano Mignini (Centro Rosarino de Estudios Perinatales, Argentina);
Ben W Mol (University of Adelaide, Australia); Peter von Dadelszen,
Laura Magee, Diane Sawchuck (University of British Columbia, Canada);
Ersheng Gao (Shanghai Institute of Planned Parenthood Research,
China); Katrien Oude Rengerink (Academic Medical Centre,
Netherlands); Javier Zamora (Ramón y Cajal, Spain); Caroline Fox,
Jane Daniels (University of Birmingham, UK); Catherine Meads
(Brunel University, UK); and Khalid S Khan, Shakila Thangaratinam,
Seema A Tirlapur (Barts and the London School of Medicine, Queen
Mary University of London, UK).
Acknowledgments
We received funding from the European Union made available to the
EBM CONNECT Collaboration through its Seventh Framework
Programme, Marie Curie Actions, International Staff Exchange Scheme
(proposal number 101377; grant agreement number 247613), which
contributed to LV’s travel to the UK. EBM CONNECT Canadian
Collaborators received funding from the Canadian Institutes of
Health Research.
References
1 Hauser WA, Annegers JF, Kurland LT. Incidence of epilepsy and
unprovoked seizures in Rochester, Minnesota: 1935–1984.
Epilepsia 1993; 34: 453–68.
2 Morrell M. Epilepsy in women. Am Fam Physician 2002;
66: 1489–94.
3 Edey S, Moran N, Nashef L. SUDEP and epilepsy-related mortality
in pregnancy. Epilepsia 2014; 55: e72–74.
4 Fairgrieve SD. Population based, prospective study of the care of
women with epilepsy in pregnancy. BMJ 2000; 321: 674–75.
5 Cantwell R, Clutton-Brock T, Cooper G, et al. Saving mothers’ lives:
reviewing maternal deaths to make motherhood safer: 2006–2008.
The eighth report of the confidential enquiries into maternal deaths
in the United Kingdom. BJOG 2011; 118 (suppl 1): 1–203.
6 Meador K, Reynolds MW, Crean S, Fahrbach K, Probst C.
Pregnancy outcomes in women with epilepsy: a systematic review
and meta-analysis of published pregnancy registries and cohorts.
Epilepsy Res 2008; 81: 1–13.
7 Borthen I, Eide MG, Daltveit AK, Gilhus NE. Obstetric outcome in
women with epilepsy: a hospital-based, retrospective study.
BJOG 2011; 118: 956–65.
8 Katz O, Levy A, Wiznitzer A, Sheiner E. Pregnancy and perinatal
outcome in epileptic women: a population-based study.
J Matern Fetal Neonatal Med 2006; 19: 21–25.
9 McPherson JA, Harper LM, Odibo AO, Roehl KA, Cahill AG.
Maternal seizure disorder and risk of adverse pregnancy outcomes.
Am Obstet Gynecol 2013; 208: 378.
10 Richmond JR, Krishnamoorthy P, Andermann E, Benjamin A.
Epilepsy and pregnancy: an obstetric perspective.
Am J Obstet Gynecol 2004; 190: 371–79.
11 Scottish Intercollegiate Guidelines Network. Diagnosis and
management of epilepsy in adults: a national clinical guideline.
Edinburgh: Scottish Intercollegiate Guidelines Network; 2003.
12 National Institute of Clinical Excellence. The epilepsies: the
diagnosis and management of the epilepsies in adults and children
in primary and secondary care. CG137. London: National Institute
for Health and Care Excellence; 2011.
www.thelancet.com Vol 386 November 7, 2015
Articles
13 Harden CL, Hopp J, Ting TY, et al. Practice parameter update:
management issues for women with epilepsy—focus on pregnancy
(an evidence-based review): obstetrical complications and change in
seizure frequency: report of the Quality Standards Subcommittee
and Therapeutics and Technology Assessment Subcommittee of the
American Academy of Neurology and American Epilepsy Society.
Neurology 2009; 73: 126–32.
14 Viale L, Allotey J, Fong F, Zamora J, Thangaratinam S. Risk of adverse
pregnancy outcomes in women with epilepsy: a systematic review.
PROSPERO: international prospective register of systematic reviews
2014: CRD42014007547. http://www.crd.york.ac.uk/PROSPERO/
display_record.asp?ID=CRD42014007547 (accessed Feb 17, 2015).
15 Wells G. Proceedings or the third symposium on systematic reviews
beyond the basics. SBOD. Improving quality and impact; The
Newcastle–Ottawa Scale (NOS) for assessing the quality of
non-randomised studies in meta-analysis. Oxford. July 3–5, 2000.
16 Review Manager. User guide. Version 52. Copenhagen: The Nordic
Cochrane Centre, The Cochrane Collaboration; 2012. http://tech.
cochrane.org/sites/tech.cochrane.org/files/uploads/documents/
revman/RevMan_5.2_User_Guide.pdf.
17 StataCorp. Stata Statistical Software: release 13. College Station, TX:
StataCorp LP; 2013.
18 Al Bunyan M, Abo-Talib Z. Outcome of pregnancies in epileptic
women: a study in Saudi Arabia. Seizure 1999; 8: 26–29.
19 Borthen I, Eide M, Daltveit AK, Gilhus NE. Complications during
pregnancy and labor for women with epilepsy. BJOG 2009;
116: 1736–42.
20 Borthen I, Eide MG, Daltveit AK, Gilhus NE. Delivery outcome of
women with epilepsy: a population-based cohort study. BJOG 2010;
117: 1537–43.
21 Cassina M, Dilaghi A, Di Gianantonio E, et al. Pregnancy outcome
in women exposed to antiepileptic drugs: teratogenic role of
maternal epilepsy and its pharmacologic treatment.
Reprod Toxicol 2013; 39: 50–57.
22 Chen YH, Chiou HY, Lin HC, Lin HL. Affect of seizures during
gestation on pregnancy outcomes in women with epilepsy.
Arch Neurol 2009; 66: 979–84.
23 Delmis J, Drazancic A, Tkalcevic T, Ivanisevic M. Epilepsy and
pregnancy. Jugosl Ginekol Perinatol 1991; 31: 23–26.
24 Endo S, Hagimoto H, Yamazawa H, et al. Statistics on deliveries of
mothers with epilepsy at Yokohama City University Hospital.
Epilepsia 2004; 45 (suppl 8): 42–47.
25 Goel P, Devi L, Saha P, Takkar N, Huria A, Dua D. Maternal and
perinatal outcome in pregnancy with epilepsy.
Internet J Gynecol Obstet 2005; 5.
26 Hunt S, Craig J, Russell A, et al. Levetiracetam in pregnancy:
preliminary experience from the UK Epilepsy and Pregnancy
Register. Neurology 2006; 67: 1876–79.
27 Hvas CL, Henriksen TB, Ostergaard JR, Dam M. Epilepsy and
pregnancy: effect of antiepileptic drugs and lifestyle on birthweight.
BJOG 2000; 107: 896–902.
28 Jadhav SV, Jadhav VK. Comparative study of obstetric outcome in
epileptic and non-epileptic pregnant women. Indian Medical
Gazette. 2013. http://medind.nic.in/ice/t13/i9/icet13i9p352.pdf
(accessed on Aug 2, 2015).
29 Klein A, McElrath TF, Keenan HA, et al. Obstetrical outcomes in
women with epilepsy enrolled in the North American antiepileptic
drug pregnancy registry (NAAPR). Epilepsy Curr 2013; 13: 429.
30 Kulaga S, Sheehy O, Zargarzadeh A, Berard A. Antiepileptic drug
use during pregnancy: perinatal outcomes with monotherapy,
polytherapy, or discontinuation of treatment.
Pharmacoepidemiol Drug Saf 2011; 19: S45–S6.
31 Laskowska M, Leszczynska-Gorzelak B, Oleszczuk J. Evaluation of
antiepileptic therapy during pregnancy. Ginekol Pol 2002; 73: 35–42.
32 Laskowska M, Leszczynska-Gorzelak B, Oleszczuk J. Pregnancy in
women with epilepsy. Gynecol Obstet Invest 2001; 51: 99–102.
33 Lin HL, Chen YH, Lin HC. No increase in adverse pregnancy
outcomes for women receiving antiepileptic drugs. J Neurol 2009;
256: 1742–49.
34 Mawer G, Briggs M, Baker GA, et al. Pregnancy with epilepsy:
obstetric and neonatal outcome of a controlled study. Seizure 2010;
19: 112–19.
1851
 Articles
35 Meador KJ, Baker GA, Finnell RH, et al. In utero antiepileptic drug
exposure: fetal death and malformations. Neurology 2006;
67: 407–12.
36 Najafi MR, Sonbolestan F, Sonbolestan SA, Zare M, Mehvari J,
Meshkati SN. The course and outcome of pregnancy and neonatal
situation in epileptic women. Adv Biomed Res 2012; 1: 4.
37 Olafsson E, Hallgrimsson JT, Hauser WA, Ludvigsson P,
Gudmundsson G. Pregnancies of women with epilepsy:
a population-based study in Iceland. Epilepsia 1998; 39: 887–92.
38 Sabers A, aRogvi-Hansen B, Dam M, et al. Pregnancy and epilepsy:
a retrospective study of 151 pregnancies. Acta Neurol Scand 1998;
97: 164–70.
39 Saleh AM, Abotalib ZM, Al-Ibrahim AA, Al-Sultan SM. Comparison
of maternal and fetal outcomes, in epileptic and non-epileptic
women. Saudi Med J 2008; 29: 261–66.
40 Sawhney H, Vasishta K, Suri V, Khunnu B, Goel P, Sawhney IM.
Pregnancy with epilepsy—a retrospective analysis.
Int J Gynaecol Obstet 1996; 54: 17–22.
41 Sonneveld SW, Correy JF. Outcome of pregnancies complicated by
epilepsy in Tasmania 1981–1988. Aust N Z J Obstet Gynaecol 1990;
30: 286–89.
42 Steegers-Theunissen RP, Renier WO, Borm GF, et al. Factors
influencing the risk of abnormal pregnancy outcome in epileptic
women: a multi-centre prospective study. Epilepsy Res 1994;
18: 261–69.
43 Tanganelli P, Regesta G. Epilepsy, pregnancy, and major birth
anomalies: an Italian prospective, controlled study. Neurology 1992;
42: 89–93.
44 Thomas SV, Sindhu K, Ajaykumar B, Sulekha Devi PB, Sujamol J.
Maternal and obstetric outcome of women with epilepsy.
Seizure 2009; 18: 163–66.
1852
45 Vanya M, Arva N, Szok D, Bartfai G. Risks of adverse pregnancy
and birth outcomes in women treated with antiepileptic drugs for
epilepsy: retrospective case-controlled study.
Eur J Contracept Reprod Health Care 2013; 18: S254–S55.
46 Veiby G, Daltveit AK, Engelsen BA, Gilhus NE. Pregnancy, delivery,
and outcome for the child in maternal epilepsy. Epilepsia 2009;
50: 2130–39.
47 Venturella A, Harper L, Odibo A, Roehl K, Cahill A. Maternal
seizure disorder is not associated with adverse pregnancy outcomes.
Am J Obstet Gynecol 2012; 206 (suppl): S152.
48 Viinikainen K, Heinonen S, Eriksson K, Kalviainen R.
Community-based, prospective, controlled study of obstetric and
neonatal outcome of 179 pregnancies in women with epilepsy.
Epilepsia 2006; 47: 186–92.
49 Waters CH, Belai Y, Gott PS, Shen P, De Giorgio CM. Outcomes of
pregnancy associated with antiepileptic drugs. Arch Neurol 1994;
51: 250–53.
50 Barroso FV, Araujo Junior E, Guazelli CA, et al. Perinatal outcomes
from the use of antiepileptic drugs during pregnancy: a case-control
study. J Matern Fetal Neonatal Med 2014; published online Sept 10.
DOI:10.3109/14767058.2014.955006.
51 Bech BH, Kjaersgaard MI, Pedersen HS, et al. Use of antiepileptic
drugs during pregnancy and risk of spontaneous abortion and
stillbirth: population based cohort study. BMJ 2014; 349: g5159.
52 Kilic D, Pedersen H, Kjaersgaard MI, et al. Birth outcomes after
prenatal exposure to antiepileptic drugs—a population-based study.
Epilepsia 2014; 55: 1714–21.
53 Borthen I, Gilhus NE. Pregnancy complications in patients with
epilepsy. Curr Opin Obstet Gynecol 2012; 24: 78–83.
www.thelancet.com Vol 386 November 7, 2015
Reproduced with permission of the copyright owner. Further reproduction prohibited without
permission.