Psychomotor
been estimated to be between 0.3 and 0.4% (Gaily et al. 1990,
development and
minor anomalies in
children exposed to
antiepileptic drugs in
utero: a prospective
population-based study
Katarina Wide* MD, Department of Paediatrics;
Birger Winbladh MD PhD, Department of Paediatrics;
Torbjörn Tomson MD PhD, Department of Clinical
Neuroscience;
Kerstin Sars-Zimmer MD, Department of Paediatrics;
Eva Berggren MD, Department of Paediatrics, Sachs’
Children’s Hospital Karolinska Institute, Stockholm,
Sweden.
*Correspondence to first author at Department of
Paediatrics, Sachs’ Children’s Hospital, PO Box 17912,
S-118 95 Stockholm, Sweden.
The aim of this study was to assess psychomotor development,
using Griffiths’ test, and the incidence of minor anomalies at
birth in children who had been exposed to antiepileptic drugs
(AEDs) in utero. The study sample comprised 100 children of
mothers who were treated with AEDs during pregnancy and
100 matched control children. Women with epilepsy were
recruited from a pregnant urban population (450 000
inhabitants). The lowest possible dose of the fewest AEDs to
maintain seizure control was used. Drug levels were
controlled on a monthly basis. The children were assessed at 9
months of age. The study children had a significant increase
in the number of minor anomalies (31 compared with 18
control children; odds ratio 2.4, CI 1.15 to 5.02, P=0.02
McNemars test), and an increased number of facial anomalies
after carbamazepine exposure (11 compared with six control
children). Drug exposure did not influence the Griffiths’ score
at 9 months of age. Even a meticulous AED treatment
strategy during pregnancy increases the number of minor
anomalies. However, treatment with AEDs does not
necessarily influence short-term psychomotor development.
The prevalence of epilepsy among pregnant women has
Dansky and Finnell 1991). Uncontrolled seizures may pose a
considerable risk to the mother as well as to the foetus
(Hiilesmaa 1997), therefore, most of these women are treat-
ed throughout pregnancy with antiepileptic drugs (AEDs),
which are potentially teratogenic. In addition to a two- or
threefold risk for major malformations (Shapiro et al. 1976,
Bertollini et al. 1985, Dansky and Finnell 1991), children
who are prenatally exposed to AEDs have been reported to
be at increased risk for minor anomalies, delayed pre- and
postnatal growth, and delayed psychomotor development
(Hanson et al. 1976; Shapiro et al. 1976; Gaily and
Granström 1987; Gaily et al. 1988a, b; Jones et al. 1989).
However, the literature disagrees on the incidence of minor
anomalies, delayed growth, and developmental disorders.
Differences in methodology and in study populations may
partly explain these conflicting results, but changes and dif-
ferences in therapeutic strategies may also contribute.
This study aimed to prospectively investigate growth para-
meters and the incidence of minor anomalies and delayed
psychomotor development in children exposed to AEDs in
utero in an unselected population of women with epilepsy
who were closely monitored during pregnancy. For this pur-
pose, a population-based study was initiated in the south-
east region of Stockholm in 1985. The aim was to ascertain all
women with epilepsy in early pregnancy who were treated
with AEDs from conception and to monitor the treatment of
their epilepsy throughout pregnancy in a specialized epilep-
sy clinic. The infants of these mothers were compared with
matched control children of mothers not treated with AEDs.
Both groups were examined blindly at birth, at 9 months of
age, and are currently being followed until school age. This
first report covers the neonatal period and follow-up at 9
months. The growth parameters will be reported separately,
as part of a larger population study.
Method
MATERNAL DATA
All children in this study were recruited from the depart-
ments of neurology and paediatrics at the Sodersjukhuset
Hospital in the south-east region of Stockholm. This hospi-
tal served the total population in the area. One hundred
consecutive children born to mothers with epilepsy
between 1985 and 1995 were prospectively studied. The
inclusion criteria were as follows: (1) the mothers were con-
tinuously treated with AEDs from conception throughout
pregnancy; (2) the mothers were identified during the first
trimester, from which time they attended the outpatient
clinic for pregnant women with epilepsy at the department
of neurology, and followed the protocol at this clinic, i.e. the
women were clinically assessed and drug plasma levels were
determined at monthly intervals; (3) the parents gave per-
mission for their children to participate in the follow-up; (4)
the children were born at one of the two delivery wards in
the departments of paediatrics and neurology at the South
Hospital; (5) one of the participating paediatricians (KW, BW,
KS-Z, EB) was available to organize the examination of the
newborn infant within 4 days of birth. No home deliveries
were included.
The treatment strategy during pregnancy was to use as
few AEDs as possible at the lowest dose to maintain seizure
Developmental Medicine & Child Neurology 2000, 42: 87–92 87
control. Doses were increased only to achieve acceptable
seizure control. Further details of the maternal protocol have
been described elsewhere (Tomson et al. 1994). During this
prospective study the antenatal clinics serving the area were
advised to refer all pregnant women with epilepsy to the out-
patient clinic at the department of neurology. Table I shows
the demographic characteristics of the study population.
Thirty-one of the potential subjects were lost to follow-up
at birth because none of the participating paediatricians was
available at this time. Five further infants were lost to follow-
up because of refusal to participate (three), miscarriage
(one), and intrauterine foetal death (one). We have no data
on the number of miscarriages and intrauterine deaths
before the women were referred to the neurology clinic.
Data on birthweight, birth length, and head circumference of
the 31 infants lost to follow-up were obtained from medical
records; data on the remaining five infants were not
obtained. There were no significant differences between
these data and those of the infants included in the study.
INFANT DATA
The control subjects were born in the same hospital within 2
days of the study subjects and matched for gestational age
and mode of delivery to enable blind evaluation of minor
malformations, and for sex to evaluate psychomotor devel-
opment, as described below. Gestational age was based on
the first day of the last menstrual period at the start of the
study in most infants and by ultrasonic determination of
foetal size (week 16 to 17) at the end of the study.
The following information was obtained at the maternity
ward by the paediatrician (KW, BW, KS-Z, EB) for all study and
control children within 4 days of birth: (1) the occurrence of
minor anomalies according to a standardized protocol; (2)
data on birthweight, birth length, and head circumference
from medical records; (3) foetal exposure to other drugs,
tobacco, and alcohol through maternal use during pregnan-
cy using a structured interview.
FOLLOW- UP AT 9 MONTHS AFTER BIRTH
A child psychologist performed the following: (1) Griffiths’
test, in the child’s home, to measure gross motor function,
personal and social behaviour, hearing and speech, eye and
hand coordination, and performance (Lindstam 1968); (2) a
parental interview about their child’s general health, their
living conditions, and the mother’s educational level. All
Table I: Demographic characteristics of the study population
(1985 to 1995)
Catchment area South-east region of
Stockholm, Sweden
Inhabitants 450 000
Total nr of births in area (1985–1995) 71 727
Total nr of women treated with AEDs 136
during pregnancy identified as
mothers of potential subjects
% of total eligible population of women 47–63%
treated with AEDs during pregnancy
(prevalence 0.3 to 0.4%a)
AED, antiepileptic drug.
a Gaily et al. 1990, Dansky and Finnell 1991.
88 Developmental Medicine & Child Neurology 2000, 42: 87–92
assessments were performed blind, with the investigator
unaware of whether the child belonged to the study group or
the control group. The study was approved by the local med-
ical ethics committee.
STATISTICAL ANALYSIS
Thirteen of the 100 study pairs were excluded from the pri-
mary statistical analysis, because the mothers of these infants
had more than one child participating in the study. The
minor anomalies were evaluated by comparing the number
of infants with and without concordance regarding any
anomaly within each study pair (McNemars test), taking into
account that the matching procedure might influence the
anomaly rate. Because of the matching procedure, condi-
tional logistic regression analyses were performed to calcu-
late odds ratios (OR) and 95% confidence intervals (CI)
(Hosmer and Lemeshow 1989). Griffiths’ test was analysed
by correcting the age of the child for gestational age with a
linear model; calculating the difference in the total Griffiths’
scores, and the scores from the five different subsets for each
pair; and performing a one-way analysis of variance (ANOVA)
with these differences as the dependent variable. The analy-
sis was performed on the whole group and the different AED
treatments.
Results
MATERNAL DATA
One hundred children of 87 women were followed up at
birth; 13 women were the mothers of more than one child.
The following results are based on the 87 mothers and their
first-born infants. The mean age of the mothers of the sub-
jects was 29 years (range 20 to 44 years), and of the control
infants, 31 years (range 20 to 43 years). Twenty mothers had
localization related epilepsy, 55 generalized idiopathic
epilepsy (five with absence epilepsy and 19 with juvenile
myoclonic epilepsy), and 12 undetermined epilepsy
(Commission on Classification and Terminology of the
International League Against Epilepsy 1989).
Seizure control deteriorated during pregnancy in 11
women, 32 had an unchanged seizure frequency, 35 were
seizure free, and nine had a decreased number of seizures
during pregnancy compared with the preceding 9 months.
AED treatment during pregnancy is reported in Table II.
Seventy-one women (83%) were receiving monotherapy:
carbamazepine (CBZ) and phenytoin (PHT) were most fre-
quently used. A systematic evaluation of plasma levels
(Table IV) was therefore confined to these two AEDs.
Plasma concentrations, in particular for PHT, were low
compared with the usually recommended therapeutic
ranges (Johannessen 1995). The dose of CBZ was increased
during pregnancy in six mothers, and the dose of PHT in
nine. Maternal use of tobacco, alcohol, and medication
other than AEDs during pregnancy did not differ signifi-
cantly between the groups.
INFANT DATA IN THE NEONATAL PERIOD
The study group comprised 42 boys and 45 girls (49 and 51
of 100). Sex-matched control infants were found for 69 of 87
(87 of 100) subjects, but the other matching criteria were ful-
filled for the total study group. The mean gestational age was
39.3 weeks for the subjects and 39.5 weeks for the control
infants (39.4 weeks and 39.6 weeks for the whole study). The
neonatal period was uneventful for all term subjects and the control group. The 31 potential subjects who were lost
control infants. Six subjects and five control infants were to follow-up did not differ from the other subjects in terms
preterm (gestational age <37 weeks). Two pairs were of mean gestational age 39 weeks (28 to 42 weeks), birth-
extremely preterm (gestational age 27 and 28 weeks). These weight, birth length, and head circumference. The growth
four children all had moderate infant respiratory distress and parameters were all within 2 SDs for gestational age. One of
needed treatment with continuous positive airway pressure, the 31 potential subjects had a diagnosis of talipes equino-
but none needed mechanical ventilation. There was a signifi- varus in the medical birth registry, but no other malforma-
cant increase in the total number of subjects with minor tions were recorded.
anomalies compared with the control infants when regard-
ing them as non-dependent observations (Table Va). A signif- INFANT DATA 9 MONTHS AFTER BIRTH
icant increase was found for CBZ but not for PHT (Table Va), Eighty-one of the 87 pairs were examined with Griffiths’ test
when dividing into the different treatment groups. The con- at 9 months of age. There were six drop-outs due to sudden
fidence intervals were, however, rather wide due to the com-
paratively low number of patients on PHT. Regarding the
observations as dependent, a significant increase was found
only for the total group (P=0.02) (McNemars test). Table Vb
shows the exact number and types of anomalies studied. Table II: Antiepileptic drugs used during pregnancy in the
Mean maternal CBZ dosage and serum levels during the women participating once or for the first time in the study
whole pregnancy were not significantly different in subjects (N=87)
with minor anomalies (prescribed dose 586 mg/day, Monotherapy Polytherapy
19 µmols/L, N=15) and those without (prescribed dose N=71 N=16
578 mg/day, 19 µmols/L, N=24, P=0.49) (Student’s t test).
The prescribed dose was increased in three women who Carbamazepine 39 9
gave birth to a child with an anomaly, but in five women with Phenytoin 22 6
children born without any anomaly. The same was true for Clonazepam 4 3
Primidone or phenobarbital 2 6
PHT, i.e. the mean doses and serum levels during the whole
Valproic acid 4 3
pregnancy were not significantly different in those with
Ethosuximide 0 1
anomalies (prescribed dose 325 mg/day, 27 µmols/L, N=5)
and those without anomalies (prescribed dose 255 mg/day,
17 µmols/L, N=17, P=0.12). The prescribed dose was
increased in two mothers with a child with an anomaly, and
in seven mothers with children without anomalies. Thus,
Table III: Combinations prescribed in the polytherapy group
there was no difference in plasma concentrations between
in the women participating once or for the first time in the
mothers of infants with and without minor anomalies.
study (N=16)
Nine subjects and five control infants had more than one
anomaly, and three subjects but no control infants had more Drug Nr of women
than two anomalies. One of the latter three subjects was
exposed to four AEDs (clonazepam, ethosuximide, PHT, and Carbamazepine + clonazepam 1
valproic acid) and had four minor anomalies. This child has Carbamazepine + ethosuximide 2
Carbamazepine + phenytoin 2
also developed nystagmus but with no anatomical abnormal-
Carbamazepine + primidone 2
ities in the eyes. The other two subjects with more than two
Carbamazepine + valproic acid 2
anomalies were exposed to CBZ in monotherapy, both with Ethosuximide + clonazepam 1
plasma levels well within the normal therapeutic range. Phenytoin + clonazepam 1
Among the 13 women who had two children participating Phenytoin + phenobarbital 3
in the study, two had children with one or more anomalies Valproic acid + phenobarbital 1
after the first pregnancy, three after the second, and three Valproic acid + phenytoin + clonazepam + ethosuximide 1
after both pregnancies. None of the siblings had anomalies
localized to the same area. The sibling of the child with four
anomalies after CBZ exposure had no minor anomalies. The
prescribed maternal dose of CBZ was decreased during the
second pregnancy from 600 mg/day to 400 mg/day (the cor- Table IV: Prescribed doses and plasma concentrations of
responding mean plasma levels were 19 µmols/L and carbamazepine and phenytoin in women receiving
10 µmols/L respectively). One infant in the PHT group had monotherapy participating once or for the first time in the
two anomalies, his older sibling had no anomalies. For the study (N=60) mean (range)
second child the mother’s dose was unchanged. Dose Plasma concentration (µmol/L)
The following major malformations were observed in the (mg/day) 1st 2nd 3rd
infants exposed to AEDs: one had a ventricular septal defect trimester trimester trimester
and a nail hypoplasia, and another had an isolated hypospa-
dia. In the control group one child had a ventricular septal Carbamazepine 559 21 20 20
defect and a simian crease. (N=39) (200–1600) (6–41) (10–35) (11–32)
No child was small for gestational age (weight less than Phenytoin 274 22 21 20
(N=22) (75–450) (<5–53) (<5–78) (5–78)
mean weight for gestational age minus 2 SDs) in the study or

Outcome of Children Exposed to Antiepileptic Drugs in Utero Katarina Wide et al. 89

infant death syndrome (one subject), refusals (one subject,
one control infant), and three who could not be traced (one
subject, two control infants). No statistically significant differ-
ence was found between the study and the control children,
for the whole group, the drug subgroups, or for the results
from the five subscales of Griffiths’ test (Table VI). These
results remained unchanged after including the 13 second-
born children as well as after excluding the children not
matched for sex.
Discussion
Numerous studies have shown that AED treatment during
pregnancy represents a risk for the foetus regarding major
and minor anomalies, delayed growth, and possibly psy-
chomotor development. The question is to what extent an
optimized treatment strategy can reduce these risks. The
incidence of delayed growth, minor anomalies, and
impaired psychomotor development varies greatly between
different studies. In addition to treatment strategy, this may

Table Va: Number of infants with minor anomalies exposed to antiepileptic drugs during pregnancy (subjects) (N=84) and the
control infants (N=83)a,b

Carbamazepine Phenytoin Others Polytherapy Total

Subjects Control Subjects Control Subjects Control Subjects Control Subjects Control
infants infants infants infants infants
(N=39) (N=37) (N=21) (N=19) (N=10) (N=11) (N=15) (N=16) (N=84) (N=83)

Anomaly present 15 5 5 6 3 2 8 5 31 18
No anomaly 24 32 16 13 7 9 7 11 53 65
OR 11.0 0.8 3.0 2.0 2.4
CI 1.42–85.2 0.22–2.98 0.31–28.8 0.5–8.00 1.15 –5.02
a Complete protocols on minor anomalies were obtained from 84 of 87 subjects who were born to mothers participating once or for first time

in the study and 83 of 87 of the control infants.

b Odds ratio (OR) and 95% confidence intervals (CI) were calculated as conditional logistic regression analyses (Hosmer and Lemeshow 1989,

Logistic Regression Examples using SAS System 1995).

Table Vb: Distribution of minor anomalies in infants exposed to antiepileptic drugs during
pregnancy (subjects) (N=84) and the control infants (N=83)a

Carbamazepine Phenytoin Others Polytherapy Total Total control

(N=39) (N=21) (N=9) (N=15) subjects infants
(N=84) (N=83)

Facial anomalies 11 2 0 4 17 6
Malformed ears 6 0 0 1 7 3
Epicanthal folds 0 1 0 1 2 1
Joined eyebrows 2 1 0 0 3 1
High palate 2 0 0 1 3 1
Micrognathia 1 0 0 0 1 0
Prognathia 0 0 0 1 1 0
Digital anomalies 5 1 2 5 13 8
Simian crease 1 0 0 1 2 1
Nail hypoplasia 4 0 0 1 5 3
Clinodactyly 0 1 1 2 4 4
Syndactyly 0 0 1 0 1 0
3rd toe longer than 2nd toe 0 0 0 1 1 0
Genital anomalies 1 0 1 0 2 0
Undescended testes 1 0 1 0 2 0
Hypospadia 0 0 0 0 0 0
Skin anomalies 4 4 0 3 11 9
Increased hair growth 2 2 0 2 6 3
Mongolian spot 2 2 0 1 5 2
Angioma in the midline 0 0 0 0 0 2
Other angiomas 0 0 0 0 0 2
Other anomalies 1 0 0 1 2 2
Two or >2 anomalies in 4 2 0 3 9 5
same individual
a Complete protocols on minor anomalies were obtained from 84 of the 87 subjects who were born to

mothers participating once or for first time in the study and 83 of the 87 control infants.

90 Developmental Medicine & Child Neurology 2000, 42: 87–92

depend on factors such as the study population, time of drug infant showed the patterns of minor anomalies associated
exposure and compliance, and study methodology (Robert with learning disabilities described for PHT and CBZ
et al. 1986, Mastroiacovo et al. 1988, Jones et al. 1989, (Hanson et al. 1976, Jones et al. 1989, Scolnik et al. 1994,
Gladstone et al. 1992, Scolnik et al. 1994, Steegers- Ornoy and Cohen 1996). The study population is too small
Theunissen 1994, Ornoy and Cohen 1996). Very high inci- to draw any conclusions about the three major malforma-
dences are found mainly in studies of case series, while the tions in the subjects, as compared with one in the control
lowest figures are from population-based studies (Hanson et infants. Both the subject and the control infant with congeni-
al. 1976, Gaily et al. 1988a, D’Souza et al. 1991, Battino et al. tal heart disease had one minor anomaly each, but this was
1992, Dravet et al. 1992, Waters et al. 1994). not the same.
The loss to follow-up at birth was mainly due to the absence Previous well controlled studies (Gaily et al. 1988b,
of the participating paediatrician who could arrange examina- Scolnik et al. 1994) have reported an association between a
tion within 4 days of birth. The data available for the group lost delayed psychomotor development and exposure to various
to follow-up and the study group do not differ. Further, the AEDs, maternal epilepsy per se, or seizure frequency during
way in which care is organized in the Stockholm area for peo- pregnancy. In these studies, estimations of psychomotor
ple who have epilepsy suggests that the group referred to the development were made at 18 to 30 months of age (Scolnik
outpatient clinic do not represent those who have less than et al. 1994), and at 5.5 years (Gaily et al.1988b), i.e. at an
average severe epilepsy. Thus, we have reason to believe that older age than in the present study. A major problem when
our study is representative of women with epilepsy in the investigating psychomotor development is that the sensitivi-
catchment area during the study period. However, on average ty for detection of deviations increases with increasing age in
this group probably has a less severe epilepsy than in other parallel with the increase in influence of environmental fac-
studies which are not population-based. tors. One important such factor is maternal education. In the
Difference in methodology is probably the main reason present report we found no difference in maternal educa-
for the variable incidence of minor anomalies between differ- tional level between the study and control groups according
ent studies. In addition to the selection of a population- to our later follow-up (Wide et al. forthcoming), but Griffiths’
based study group, the definition of minor anomalies, i.e. the test was performed at 9 to 10 months of age when the sensi-
use of a fixed protocol for evaluation and a blinded compari- tivity for detection of deviance is still rather low (Lindstam
son with a matched control group are necessary for an ade- 1968). The absence of a negative influence on psychomotor
quate estimation of the incidence of minor anomalies. Even development thus has to be interpreted with caution.
with such rigorous protocols, comparisons between differ- In addition to differences in methodology, our study dif-
ent studies are hampered by difficulties in the delineation of fers from many others with respect to the AED treatment dur-
minor anomalies from normal variations. The population- ing pregnancy. Most women were treated with monotherapy,
based study by Gaily et al. (1987) showed 38% with more preferably CBZ, the treatment was closely monitored, and
than two typical minor anomalies in the PHT exposed group,
but they attributed only hypertelorism and digital hypoplasia
to the PHT treatment. The frequency of minor anomalies is
more variable in studies based on women who attend special Table VI: Results of the Griffiths test in infants exposed to
clinics because of their own concern for the teratogenic antiepileptic drugs during pregnancy (N=81) and the
potential of the AEDs they are taking during pregnancy matched control infants (N=81)a,b
(Jones et al. 1989, Gladstone et al. 1992).
The study and control infants in our study were simulta- Antiepileptic Total score Mean differences
neously examined by the same investigator which increases drugs Subjects Control (Subjects–
infants control infants)
the reliability within each pair. However, at least four differ-
Mean (range) Mean (range) 95% CI
ent investigators (KW, BW, KS-Z, EB) were enrolled which
may have introduced a variation in scoring between differ- Carbamazepine 350 335 –0.59, 16.57
ent pairs in spite of the common protocol. The method, (N=35) (324–435) (307–396)
however, does not introduce bias for diagnosing more Phenytoin 346 344 –7.13, 11.34
cases than were actually found even though we found that (N=21) (307–385) (318–378)
37% of the study children and 22% of the control infants Others 340 355 –70.96, 22.60
had at least one minor anomaly. Two or more anomalies (N=9) (274–368) (322–477)
Polytherapy 344 352 –24.87, 8.95
were found in 11% of the study group and 6% of the control
(N=16) (289–381) (316–386)
group. The incidence of one anomaly in a non-selected
Total 349 350 –25.88, 14.86
newborn population is reported in several studies to be 14 (N=81) (274–435) (307–477)
to 15% (Marden et al. 1964, Méhes 1983), a figure rather
a The Griffiths test was performed in 81 of 87 pairs who were born to
close to our 22%. Our high levels of minor anomalies in the
control infants indicate the readiness of detecting minor mothers participating once or for the first time in the study. See text
anomalies when looking for them, and emphasizes the for explanation of the six non-participating pairs.
b Statistical method: one-way univariate ANOVA. The test was
importance of a blinded method.
performed for the different therapy groups separately (P=0.4).
As reported by Méhes (1983), minor anomalies in an
One-way univariate ANOVA was also used for the five subscales in
infant indicate an increased risk for a coexisting major mal-
the Griffiths test analysed separately for the therapy groups: gross
formation, the risk increasing with increasing numbers of motor function (P=0.77); personal and social behaviour (P=0.10);
minor anomalies. Only nine subjects and five control infants hearing and speech (P=0.11); eye and hand coordination (P=0.12);
in the present study had more than one anomaly and no performance (P=0.08); total (P=0.20).

Outcome of Children Exposed to Antiepileptic Drugs in Utero Katarina Wide et al. 91

the drug dosage and plasma concentrations were kept as low
as possible. Drug exposure as expressed by maternal serum
concentrations was lower in our study, in particular for PHT,
than in most previous studies. This may be a second impor-
tant reason for the limited foetal effects observed in the pre-
sent investigation, in particular in relation to PHT exposure.
The results show that even if there is a higher incidence of
minor anomalies in the children of mothers exposed to
AEDs, few children have more than one anomaly, and no spe-
cific pattern of anomalies is seen. Low doses of maternal AED
treatment in monotherapy during pregnancy may reduce the
negative effects observed at birth. However, the results of the
psychomotor testing should be interpreted with caution
awaiting the results of the 4-year follow-up.
Accepted for publication 2nd September 1999.
Acknowledgements
We are most grateful to Professor Bengt Källén, Lund, for valuable
criticism of the manuscript. This study was supported by research
grant no K97-17X-12225-01A from the Swedish Medical Research
Council, the Foundation of the First of May flower, The Foundation
of Samariten, the Orion-Pharma Research Foundation, and the
Holmia Insurance Company, Ltd.
References
Battino D, Binelli S, Caccamo ML, Canevini MP, Canger R, Como ML,
Croci D, de Giambattista M, Granata T, Pardi G et al. (1992)
Malformations in offspring of 305 epileptic women: a prospective
study. Acta Neurologica Scandinavia 85: 204–7.
Bertollini R, Mastroiacovo P, Segni G. (1985) Maternal epilepsy and
birth defects: a case-control study in the Italian multi-centric
registry of birth defects (IPIMC). European Journal of
Epidemiology 1: 67–72.
Commission on Classification and Terminology of the International
League Against Epilepsy. (1989) Proposal for revised classification
of epilepsies and epileptic syndromes. Epilepsia 30: 389–99.
D’Souza SW, Robertson IG, Donnai D, Mawer G. (1991) Fetal
phenytoin exposure,hypoplastic nails, and jitteriness. Archives of
Diseases in Childhood 66: 320–4.
Dansky LV, Finnell RH. (1991) Parental epilepsy, anticonvulsant
drugs, and reproductive outcome: epidemiological and
experimental findings spanning three decades. 2: Human
studies. Reproductive Toxicology 5: 301–35.
Dravet C, Julian C, Legras C, Magaudda A, Guerrini R, Genton P,
Soulayro S, Giraud N, Medsdjian E, Trentin G et al. (1992)
Epilepsy, antiepileptic drugs, and malformations in children of
women with epilepsy: a French prospective cohort study.
Neurology 42: (Suppl.) 75–82.
Gaily E, Granström M-L. (1987) Minor anomalies in epileptic mothers’
children. In: Wolf P, Dam M, Janz D, Dreifuss FE, editors. Advances
in Epileptology. Vol. 16. New York: Raven Press. p 567–71.
— — Hiilesmaa VK, Bardy A. (1988a) Minor anomalies in offspring
of epileptic mothers. Journal of Pediatrics 112: 520–9.
— Kantola-Sorsa E, Granström M-L. (1988b) Intelligence of children
of epileptic mothers. Journal of Pediatrics 113: 677–84
— — — (1990) Specific cognitive dysfunction in children with
epileptic mothers. Developmental Medicine & Child Neurology
32: 403–14.
Gladstone D, Bologa M, Maguire C, Pastuszak A, Koren G. (1992)
Course of pregnancy and fetal outcome following maternal
exposure to carbamazepine and phenytoin: a prospective study.
Reproductive Toxicology 6: 257–61.
Hanson JW, Myrianthopoulos NC, Sedgwick Harvey MA, Smith DW.
(1976) Risks to the offspring of women treated with hydantoin
anticonvulsants, with emphasis on the fetal hydantoin syndrome.
Journal of Pediatrics 89: 662–8.
Hiilesmaa V. (1997) Effects of maternal seizures on the foetus. In:
Tomson T, Gram L, Sillanpää M, Johannessen SI, editors. Epilepsy
and Pregnancy. Petersfield, UK: Wrightson Biomedical
Publishing Ltd. p 135–41.
92 Developmental Medicine & Child Neurology 2000, 42: 87–92
Hosmer DW, Lemeshow S. (1989) Applied Logistic Regression. New
York: John Wiley and Sons, Inc.
Johannessen S. (1995) General principles: laboratory monitoring of
antiepileptic drugs. In: Levy RH, Mattson RH, Meldrum BS,
editors. Antiepileptic Drugs, Fourth Edition. New York: Raven
press Ltd. p 179.
Jones KL, Lacro RV, Johnsson KA, Adams J. (1989) Pattern of
malformations in the children of mothers treated with
carbamazepine during pregnancy. The New England Journal of
Medicine 320: 1661–6.
Lindstam R. (1968) Griffiths’ Mental Development Scale. Svensk
Version 0-2 år. Stockholm: Skandinaviska testförlaget.
Marden PM, Smith DW, McDonald MJ. (1964) Congenital
malformations in the new born infant, including minor variants.
Journal of Pediatrics 64: 357–71.
Mastroiacovo P, Bertollini D, Licata D. (1988) Fetal growth in the
offspring of epileptic women: results of an Italian multicentric
cohort study. Acta Neurologica Scandinavia 78: 110–4.
Méhes K. (1983) Minor Malformations in the Neonate. Budapest:
Akadémiai Kiadó.
Ornoy A, Cohen E. (1996) Outcome of children born to epileptic
mothers treated with carbamazepine during pregnancy. Archives
of Diseases in Childhood 75: 517–20.
Robert E, Löfkvist E, Mauguière F, Robert JM. (1986) Evaluation of
drug therapy and teratogenic risk in a Rhône-Alpes district
population of pregnant epileptic women. European Journal of
Neurology 25: 436–43.
Scolnik D, Nulman I, Rovet J, Gladstone D, Czuchta D, Gardner HA,
Gladstone R, Ashby P, Weksberg R, Einarson T et al. (1994)
Neurodevelopment of children exposed in utero to phenytoin
and carbamazepine monotherapy. Journal of the American
Medical Association 10: 767–70.
Shapiro S, Slone D, Hartz SC, Siskind V, Mitchell AA, Slone D,
Rosenberg T, Monson RR, Heinonen OP. (1976) Anticonvulsants
and parental epilepsy in the development of birth defects. Lancet
i: 272–5.
Steegers-Theunissen RP, Renier WO, Borm GF, Thomas CH, Merkus
HI, Op de Coul DA, De Jong PA, van Geijn HP, Wouters M, Eskes
TK. (1994) Factors influencing the risk of abnormal pregnancy
outcome in epileptic women: a multi-centre prospective study.
Epilepsy Research 18: 261–9.
Tomson T, Lindbom U, Ekqvist B, Sundqvist A. (1994) Epilepsy and
pregnancy: a prospective study of seizure control in relation to
free and total plasma concentrations of carbamazepine and
phenytoin. Epilepsia 35: 122–30.
Waters C, Belai Y, Gott P, Shen P, De Giorgio C. (1994) Outcomes of
pregnancy associated with antiepileptic drugs. Archives of
Neurology 51: 250–3.