Discussions
In this study, we observed that approximately 60% of pregnancies in WWE
were unplanned, which was more common among WWE with a lower education
level. More importantly, we found that planned pregnancy was associated
with more optimized AEDs treatment patterns, better seizure control,
more appropriate folic acid supplementation and less adverse fetal
outcomes with pregnancy planning as independent risk factor. Our results
indicate that pregnancy planning may be a critical strategy for WWE to
fulfill satisfactory seizure control and improve their chances of having
a healthy baby.
Although pregnancy planning is critical for WWE, more than half of the
pregnancies in WWE were unplanned. The result from EBCR web-based survey
of 1,144 WWE showed that 65.0% of their pregnancies were
unintended.18 The data from
the Pregnancy Risk Assessment
Monitoring System (PRAMS), including 13 states in the United States from
2009 to 2014, demonstrated that 55% of WWE pregnancies were
unintended.22 They reported that unintended
pregnancies were more common in WWE with younger ages, lower incomes, or
who were members of racial minority groups or
Hispanic.22 The population in our study was Han
Chinese. We found that WWE in the unplanned-pregnancy group had a lower
level of education and tended to reside in areas that were more rural
compared with the planned-pregnancy group. These patients often have
limited access to medical advice.
The greatest concern for WWE is the teratogenicity of AEDs. Accumulating
evidence suggests that different AEDs and dosages have different
teratogenic risks,23-25 and that the newer generation
AEDs, such as LTG, LEV, and OXC, are not associated with significantly
increased risks of congenital malformations compared with no AED
exposure control.15, 16 In this study, we found that
more than 50% of pregnancies in the unplanned-pregnancy group were not
prescribed AEDs during pregnancy, which may be because those patients
feared drug teratogenicity and had rarely consulted epileptologist
before and during pregnancy for relevant knowledge of epilepsy and AED
compliance. This absence AED use may also explain the greater number of
GTCS attacks in the unplanned-pregnancy group.
The most commonly used AEDs as monotherapy in first trimester were LTG,
LEV, and OXC in the planned group, which is in accordance with current
guidelines and opinions.14, 26, 27 However, in the
unplanned group, LTG and VPA were among the medications most commonly
prescribed. In Shaanxi Province, in the last decade, we have been
implementing a program on rational drug selection for WWE. This program
could make many WWE in this study be prescribed with LTG at the onset of
illness. VPA is a broad-spectrum AED with potent antiepileptic effects
in WWE.23, 24 Additionally, it is the most teratogenic
AED, with a several-fold increase in MCMs risk.15, 16,
25 In the planned-pregnancy group, no WWE took VPA as monotherapy. This
finding could also be attributed to the promotion of our program.
Compared with planned-pregnancy group, the unplanned group had a
relatively high rate of induced abortions. This choice was likely made
because of the following reasons: 1) fear of the effect of frequent
seizures on the fetus; 2) fear of AEDs teratogenicity, or 3) abnormal
prenatal examination. In addition, the unplanned group tended to have a
greater rate of SFL, which could be also attributed to more GTCS attacks
in the unplanned-pregnancy group.
Adverse fetal outcomes of WWE mainly include MCMs, cognitive outcomes
such as impaired intellectual and verbal performance, autism spectrum
disorder, and growth restriction (i.e., being small for gestational age
or smaller than normal head circumference).23, 28-30Our study focused on MCMs. MCMs are defined as structural abnormalities
with significant interference with function and/or require surgery for
correction.31, 32 Results from Swedish registries
demonstrated that among the 1,429,652 included births, MCM prevalence
among AED-exposed offspring was 6.7% and was 4.7% in offspring of WWE
without AED treatment.4 Recent data from the European
and International Registry of Antiepileptic Drugs in Pregnancy (EURAP)
showed that prevalence of MCMs in offspring exposed prenatally to AED
monotherapy ranged from 2.8% to 10.3%, depending on the type of
AEDs.15 In the Maternal Outcomes and
Neurodevelopmental Effects of Antiepileptic Drugs study, adverse fetal
outcomes including fetal loss and MCMs were significantly more common in
WWE than in healthy pregnant women.33
In our study, the MCM rate was significantly lower in the
planned-pregnancy group (1.6%) than in the unplanned-pregnancy group
(7.5%). This decrease in MCMs rate could be partly because of the
relatively less use of VPA, PB, phenytoin (PHT), CBZ, and topiramate
(TPM) in the planned-pregnancy group than in the unplanned group. The
five aforementioned AEDs were reported in the EURAP registry to be
related to a high prevalence of MCMs.15 For types of
MCM, cardiac defect was most common in both the planned and unplanned
group, which was consistent with the finding of the EURAP
registry.15 However, in the unplanned-pregnancy group,
except for cardiac defect, there were cases of bifid spine, cleft
lip/palate, hypospadias, conjoined twins, and anotia. Encouragingly, the
planned-pregnancy group only had an MCM prevalence of 1.6%, which
indicates that optimized pregnancy management can significantly decrease
the risk of offspring adverse outcomes.
This study has several limitations.
First, our study assessed a relatively small sample and was conducted
at a single site in a university hospital setting. Thus, validation of
our findings in different areas and with a large sample would be
necessary to better define the importance of pregnancy planning. Second,
instead of all types of seizures, we recorded only the number of GTCS
during pregnancy, which has been reported to have significant negative
effects on the offspring of WWE. Notably, because of the nature of this
retrospective study, recording an accurate number of seizures other than
GTCS was difficult. Third, detailed information on neonates such as
birth weight and Apgar score was not available at the time of our study,
which prevented comprehensive analysis of the effect of pregnancy
planning on neonatal outcomes. Our study also did not record minor
congenital malformations such as ocular hypertelorism, lop ear,
hypoplastic fingernails, and ankyloglossia, etc. In addition, tobacco
smoking and alcohol consumption during pregnancy were not recorded.
These factors have also been reported to affect fetal
outcomes.34, 35