Pregnancy in women with epilepsy : preliminary results of Kerala registry of epilepsy and pregnancy.
Eighty-five women with epilepsy were followed up for reproductive functions under the registry of epilepsy and pregnancy. 32 of them had completed the pregnancy. Their mean age was 26 years and mean seizure frequency was 0.7 during current pregnancy. Nineteen of them (59.4%) had generalized epilepsy. Nine of them were not on any anti epileptic drugs (AED), 23 women were on various AEDs, 19 being on monotherapy. Only 40% of the women were taking folic acid during pregnancy. Pregnancy ended as spontaneous abortion in one patient. Nearly one third required cesarean section. Majority (87.5%) had term babies. Three (10.7%) babies had birth asphyxia. Six babies (21.4%) had low birth weight. Congenital malformations were detected in four cases (12.5%). Malformations included neural tube defects, talipes equinovarus and other minor anomalies. These babies were exposed to sodium valproate, carbamazepine or phenobarbitone. The risk of malformation was significantly greater (p<0.05) when the mother had generalized epilepsy. The odds ratio for risk of malformation was much higher with sodium valproate (6) than that with carbamazepine (1.2) or phenobarbitone (0.8). Majority of women with epilepsy had safe pregnancy and childbirth without any aggravation of epilepsy.
Epilepsy is one of the commonest neurological
disorders in our country. Some 30% of the 75,000
women with epilepsy (WWE) in Kerala come under
the reproductive age group. The social stigma and
negative attitude prevalent in the society make it hard
for WWE to integrate well into normal social life.
There is considerable concern regarding the safety of
pregnancy and health of the baby among these
couples. Most of the previous studies indicate that
WWE may have a slightly reduced fertility, when
compared to others. However, a recent community
based study revealed that fertility is not reduced
among WWE. They may also face increased risk of
complications of pregnancy such as hyperemesis
gravidarum, spontaneous abortions, premature labour
and assisted delivery or cesarean section. The risk of
congenital malformations is increased among the
infants born to WWE by two to three times (4%-6%).
Most of the data on epilepsy and pregnancy had been
compiled through the registries operating in the
developed countries. The special situations prevailing
in India and other developing countries are not
adequately addressed in these studies. The outcome of
pregnancy among WWE in developing countries
further depends on these factors also. The preliminary
results of the registry are presented here.
This study was carried out in the Kerala registry of
epilepsy and pregnancy. The protocol of registry is
described in Annexure I. Patients got enrolled in the
registry at the preconception period or during
pregnancy. Terminology according to International
League Against Epilepsy (ILAE) was followed in
classifying seizures and epileptic syndromes.,
Patients were followed up according to the consensus
protocol developed by a team of doctors consisting of
neurologists, gynaecologists, imageologists and
experts from genetics and biochemistry departments
during a workshop held for this purpose. Clinical and
laboratory investigations were carried out at
predetermined intervals as laid down in the protocol.
85 WWE were enrolled in the registry between June
1998 and November 1998. Twenty of them (23.5%)
reported in the preconception period and 65 (76.5%)
during pregnancy. 14 patients were in the first
trimester, 16 were in the second trimester and 35 in
the third trimester. Follow up was incomplete in four
of them and they were excluded from the analysis.
The results pertain to the 32 patients, who had
completed their current pregnancy. As per the policy
of the registry, obstetrical follow up and delivery are
conducted at a centre chosen by the patient. We
compiled data on deliveries conducted elsewhere.
Student's 't' test was performed to ascertain
significance of group means. Odds ratio was
calculated to estimate the risk of malformation with
Patient characteristics and AED therapy : Thirty-two
women had completed their pregnancy under this
protocol. Their age ranged from 19-38 years (mean 26
years), and parity ranged from 1-4 (mean 1.6).
Nineteen (59.38%) of them had generalized epilepsy,
11 (34.37%) had localisation related epilepsy and 2
(6.26%) had other types. The seizure frequency during
current pregnancy ranged from 0-8 (mean 0.7). Only
23 (71.9%) of them were taking AED, whereas nine
(28.1%) were not on any AED. Nineteen were on
monotherapy and 4 (12.5%) were on polytherapy.
Nine patients were on phenobarbitone, seven on
carbamazepine, six on sodium valproate and five on
phenytoin. Regular consumption of folic acid was
documented in only 40% patients although majority
of them were recommended nutritional supplements
including folic acid. None of them had any
preconception evaluation or counseling.
Pregnancy outcome : There was an increase in seizure
frequency in only 4 women (12.5%) whereas there
was no change in the seizure frequency in 25 (78.1%).
Seizure frequency had decreased among 3 (9.4%)
women. Over a quarter of these patients experienced
pregnancy induced hypertension (PIH) or preeclampsia.
Other complications observed during
pregnancy were intrauterine growth retardation
(IUGR), post term, placenta previa and hydramnios
[Table I]. There was one case of spontaneous abortion.
Nineteen (59.4%) women had normal vaginal
delivery. Nearly one-third (10) required cesarean
section [Table II]. The indication for cesarean section
was foetal distress, uterine inertia and failure of
induction. There was no episode of postpartum
Foetal outcome : There were 28 (87.5%) term babies,
2 (6%) were post term and 1 (3%) was pre-term. The
mean head circumference of the babies was 34.9 cm
(range 32-44 cm) and mean birth weight 2.84 kg
(range 1.8-4 kg). 6 (21.4%) of the term babies were of
low birth weight(<2.5 kg) and 3 (10.7%) had birth
asphyxia (apgar 1 mt/5 mts = 2-6/3-7).
Congenital malformations : Congenital malformations
were detected in 4 babies [Table III]. The
anomalies included hydrocephalus, meningomyelocele,
spina bifida, Arnold-Chiarri malformation,
intracerebral haemorrhage, talipes equinovarus and
minor anomalies like asymmetric ears, pectus
excavatus. Two of these babies were exposed to
sodium valproate, while others to carbamazepine and
phenobarbitone respectively. Two babies with
malformations died in the immediate neonatal period.
The baby with minor malformation Case 4, [Table III]
was delivered in a peripheral hospital. The baby was
transferred to neurology care when it developed
neonatal seizures. An ultrasound examination through
anterior fontenella showed intracerebral haemorrhage
that was confirmed by CT Scan. This baby was
managed with vitamin K injections, anti-convulsants
and cerebral anti-oedema measures but died on 30th
The clinical characteristics of pregnancies with and
without malformations in the babies are compared
[Table IV], even though the number of pregnancies with
foetal anomalie was too small, due to small sample.
There was no statistically significant difference in the
maternal age, between those who had babies with
malformations and those without malformations.
There was a statistically significant (p <0.05) increase
in frequency of major anomalies among babies
exposed to AEDs in general, when compared to those
not exposed to AEDs. The odds ratio for occurrence of
malformations with various AEDs was as follows:
sodium valproate (6), carbamazepine (1.2) and
phenobarbitone (0.8). There was no anomaly in
patients exposed to phenytoin.
The gender issues and medical aspects of epilepsy and
pregnancy are gaining more focussed attention.
Majority of WWE had normal outcome of pregnancy
in our study. Only one patient (3%) had a spontaneous
abortion which is less than that in the community. This
is much less than that observed in our earlier
retrospective analysis. Another cohort of WWE,
followed under a registry, had 9.4% rate of
spontaneous abortion. One of the reasons for this low
estimate of spontaneous abortion in our series may be
that nearly half of the patients were enrolled in the
registry only in the third trimester.
The complications of pregnancy, observed in our
series is similar to that reported earlier.9,10 However,
the frequency of pregnancy induced hypertension and
pre-eclampsia is much higher in our series which
could partly be due to the geographical
preponderance. PIH and PET are multifactorial
complications of pregnancy that show much
geographical variation in frequency and severity. It is
commoner among lower socio-economic group.
The head circumferences of the newborns were within
normal limits for the community. One fifth of the term
babies had low birth weight (<2.5 kg), which indicates
some degree of intrauterine growth retardation.
The most important finding in this study is the
occurrence of major malformations in a relatively
high proportion (12.5%) of pregnancies. This
frequency is comparable to that described in other
prospective studies.,, It is generally accepted that
the incidence of congenital malformations in
pregnancies among WWE is higher than that in
general population. The incidence of malformations is
comparatively higher in prospective studies (10 -
12%) than in retrospective studies were incidence of
2-6% has been reported.14 A retrospective study from
our center also had shown a lower (4%) frequency of
birth defects. It appears that risk of malformations in
WWE is more than what was previously estimated.
We had examined several maternal clinical
characteristics for possible association with foetal
anomalies. There was no statistically significant
difference in the maternal age between those who had
babies with malformations and those without
malformations. Other maternal characteristics such as
age, parity and frequency of seizures during
pregnancy did not have any significant correlation
with congenital anomalies in our cohort. This is in
contrast to the observation from Italian registry where
the mothers who had babies with malformations had
higher age. Frequent seizures during pregnancy had
been associated with adverse foetal outcome in earlier
studies. However, we could not establish any such
relationship. This may be because our patients had
very few seizures during pregnancy (mean seizure
frequency was 0.7) [Table IV]. In our study, the only
clinical factors associated with birth anomalies were
the type of maternal epilepsy and use of AED. There
was statistically significant increased in incidence of
malformations associated with generalized epilepsy.
statistically significant. Reports from Berlin also have
shown increased incidence of birth defects associated
with generalized epilepsy in mothers. In our study
there was a clear correlation between the risk of neural
tube defects and the exposure to sodium valproate or
carbamazepine. Two of the six (33.3%) infants
exposed to sodium valproate and one of the seven
(14.3%) exposed to carbamazepine showed neural
tube defects. The odds ratio was maximum (6) for
sodium valproate. These observations are in
agreement with other studies from elsewhere.,,,,
Polytherapy had been identified as one of the risk
factors for malformations in earlier studies.22,23 We
did not find such a correlation. This could be because
there were very few patients (12.5%) on polytherapy.
Another major observation in this study is the
inconsistency in the prescription and use of folic acid
during pregnancy. There was very poor compliance in
this aspect as only 40% patients were found to be
taking folic acid on a regular basis. The situation in
developed countries is better (64%-78%) although it
falls short of ideal situation. Medical Research
Council study in UK has shown that folic acid
supplementation of 4 mg before conception prevented
nearly thee quarters of neural tube defects in nonepileptic
women. There appears to be no uniform
prescription policy regarding folic acid during
pregnancy among WWE.
One of these babies with minor anomalies developed
intracerebral bleed and died on 30th day. This baby
was delivered in a peripheral center and was not
protected by vitamin K. There is much debate as to the
usefulness of administering vitamin K to the mother in
the immediate pre parturition period to prevent
hemorrhagic disease of newborn. The American
Academy of Neurology in its practice parameters have
recently recommended Vitamin K (10 mg per day) in
the last month of pregnancy or parenteral Vitamin K1
as soon as labor sets in.
The study has brought out many of the lacunae in the
care of WWE during pregnancy. Special situations
prevailing in India and other developing countries do
have a bearing on their outcome as observed in this
study. The consumption of folic acid during
pregnancy was less among WWE in our cohort. They
had increased frequency of PIH/PET, and low birth
weight babies. The frequency of cesarean sections was
more in our cohort. These preliminary results indicate
that major malformations occur in about 10% of cases
and the risk appears to be higher with sodium
valproate. The medical care of WWE in pregnancy
need an integrated multidisciplinary approach. In fact
the planning should begin much ahead of conception.
A detailed preconception counseling should address
all aspects of pharmacotherapy and obstetrical issues.
The AED therapy need to be carefully evaluated and
folic acid should be administered regularly. Pregnancy
should be followed up with regular review and
screening for serious congenital anomalies.
The authors gratefully acknowledge the financial
support from the Department of Science, Technology
and Environment, Government of Kerala to the Kerala
registry of epilepsy and pregnancy.