Pregnancy is a time of great expectations but also concern for the health of the future child. For mothers with epilepsy, this period involves additional considerations, especially regarding antiseizure medications. Some types of antiseizure medication have been associated with adverse neurodevelopment in the child. However, a new study suggests that the mother’s use of specific types of antiseizure medication during pregnancy is not linked to risk of epilepsy in the children. Although this is good news, some types of antiseizure medication are still not recommended for pregnant women because of other risks.
Pregnant women with epilepsy face a difficult choice: should they continue taking antiseizure medication during pregnancy or do without the antiseizure medicine, thereby facing a risk of seizures? Valproate is a type of antiseizure medication that is highly effective in epilepsy but is also associated with adverse fetal brain development, raising concerns about long-term effects for the child. This dilemma between ensuring the mother’s health and protecting the child from potential harm makes the decision-making process complex and burdensome. A new study adds new and important knowledge.
“A large study based on more than 38,000 children born to mothers with epilepsy showed that epilepsy is more frequent among the children of mothers using valproate during pregnancy compared with the children of mothers with epilepsy who did not use antiseizure medication during pregnancy or children of mothers who used other types of epilepsy medication. We nevertheless conclude that the exposure to valproate is probably not the explanation for the increased risk of epilepsy among the children, partly because the risk of epilepsy was the same among siblings for which the mothers had used valproate in one pregnancy but not in the other. Instead, we suspect that the higher incidence of epilepsy among the children of mothers with epilepsy using valproate results from the type of epilepsy typically treated with valproate being more hereditary than other types of epilepsy,” explains Julie Werenberg Dreier, lead author and Senior Researcher, National Centre for Register-based Research, Aarhus University, Denmark.
Additional complexity
The use of antiseizure medication during pregnancy has long been controversial, primarily because of concerns about the safety of the drugs and how they affect the development of the unborn child.
“In particular, valproate has been studied for association with neurodevelopmental birth defects and cognitive disturbances in the child’s early development. Epidemiological studies have shown that children exposed to valproate during pregnancy have a higher risk of autism spectrum disorder and lower IQ,” says co-author Jakob Christensen, Consultant and Clinical Professor, Aarhus University Hospital, Denmark.
The findings have led to strict restrictions on the use of valproate among women of childbearing age. One of the researchers, Rebecca L. Bromley from the University of Manchester, United Kingdom, works daily with children exposed to valproate during pregnancy and told the researchers from Denmark during a meeting that some parents had approached her because their child had now been diagnosed with epilepsy.
“These children were examined and monitored for developmental disorders, which are assumed to result from the mother’s use of valproate. When the children then also develop epilepsy, the parents naturally wonder whether the medication may also have contributed to this – but no study had examined exactly that, probably because you would be inclined think that the child’s epilepsy was instead related to the mother’s epilepsy rather than her use of antiseizure medication,” explains Julie Werenberg Dreier.
The phenomenon coined “the maternal effect” adds further complexity: the children of mothers with epilepsy have been shown to have a higher incidence of epilepsy than the children of fathers with epilepsy.
“This has reinforced the suspicion that the mother’s use of antiseizure medication during pregnancy could amplify the existing increased risk of epilepsy among the children of parents with epilepsy, so there was a significant need for detailed studies that could uncover these relationships more precisely,” adds Julie Werenberg Dreier.
Not possible when data sets are too small
The new study was based on a large Nordic population-based study in the project SCAN-AED (www.scanaed.org), including 38,663 children born to mothers with epilepsy from 1996 to 2017. The researchers used medical registries to collect information on the mothers’ medication use, epilepsy diagnoses and other relevant health conditions, with the primary aim of evaluating the association between the use of antiseizure medication during pregnancy and the risk of epilepsy.
“Prior to the study, doctors commented on the expected study of potential link with childhood epilepsy, ‘Well, you will see that the children whose mothers have taken valproate have a higher incidence of epilepsy.’ This is because the type of epilepsy typically treated with valproate is often a more hereditary type. Examining the data without doing anything further would therefore just show that the incidence would be highest in that group of children,” explains Julie Werenberg Dreier.
In addition to considering confounding factors such as the mothers’ age at birth and socioeconomic status, the researchers therefore also focused on differences between siblings to minimize the genetic and environmental confounding that can arise in observational studies. This enabled the effect of antiseizure medication to be isolated more precisely.
“It was really only when we obtained access to the large Nordic databases that we could perform some of the more sophisticated analyses, which also involve sibling data – which is often not possible when the data sets are too small,” notes Julie Werenberg Dreier.
So, although the initial analysis indicated that children of mothers using valproate during pregnancy had a two-fold higher risk of epilepsy compared to children of mothers with epilepsy who had not used antiseizure medication during pregnancy, subsequent analysis showed that there was good reason to believe that using valproate in pregnancy was not related to this risk. Remarkably, no dose-dependent relationship was found between valproate dose and the child’s risk of epilepsy.
“If valproate caused the children’s epilepsy, the risk would typically be expected to be greater for a higher dose of valproate. But the risk was the same regardless of whether the mother had taken a low, medium or high dose of valproate, indicating that the observed risk could result from other underlying factors,” says Julie Werenberg Dreier.
To be absolutely sure
As an extra check on the results, the researchers also decided to determine whether there was a dose–response relationship with some of the established risks associated with using valproate during pregnancy, such as malformations and autism.
“We found a higher risk of both malformations and autism with higher doses of valproate. So, the reason that we found no dose–response risk of epilepsy associated with valproate was probably not because our dose categorisation of valproate was poor,” elaborates Julie Werenberg Dreier.
Valproate is typically used for a certain type of epilepsy that is more hereditary than other types. The researchers therefore also compared the risk of epilepsy among the children of mothers who had used valproate during pregnancy with that of the children of mothers who had used valproate before but not during the pregnancy itself, with the hypothesis that these mothers probably have the same type of epilepsy.
“This analysis showed that the children’s risk of epilepsy no longer differed significantly, which again indicates that maternal valproate use in pregnancy does not increase the risk of epilepsy in the child,” adds Julie Werenberg Dreier.
Could be attributed to something else
Finally, the researchers examined the risk among siblings whose mother had taken valproate in one pregnancy but not in the other, and this did not indicate that valproate increases the risk of epilepsy.
“Siblings did not differ in the risk of developing epilepsy regardless of whether the mother used valproate in pregnancy. However, when the researchers studied autism, the risk was almost six times higher in the sibling where the mother used valproate during pregnancy compared to the risk in the sibling where the mother did not use valproate in pregnancy,” explains Julie Werenberg Dreier.
The results indicate that, although children exposed to valproate during pregnancy have a higher incidence of epilepsy, it is not directly caused by the medication itself.
Somewhat prominent effect
The new findings are another piece of evidence in our understanding of the complex dynamics between epilepsy, pregnancy and medication. The results emphasize the need for continued attention to safely and carefully assessing the use of medicine among women with epilepsy. Although this study could not link valproate to an increased risk of epilepsy among children, strong evidence indicates that valproate should not be used in pregnancy due to risk of congenital malformations and other developmental disorders.
“Pharmaceutical authorities have issued warnings against women of childbearing age using valproate. There must be documentation that the women need exactly valproate and are not better treated with other types of antiseizure medication. Both doctor and patient must sign this document once yearly confirming that no other types of antiseizure medication could be used instead,” says Julie Werenberg Dreier.
“The problem is that, for example, valproate is effective at treating epilepsy. So, the women are in the somewhat difficult situation of asking themselves whether they should try to switch to another type of antiseizure medication and risk seizures. And having epileptic seizures can be such a frightening experience that the women would sometimes choose not to have children rather than to change medicine. This is a significant result of having epilepsy,” explains Julie Werenberg Dreier.
Jakob Christensen thinks that it is important to remember that the probability that a child will not develop epilepsy – even if the mother has epilepsy – is about 90%, and that although it can be problematic to have a chronic disease as a child, epilepsy can often be treated with medicine without major problems.
“There is now sufficient knowledge that other types of antiseizure medication such as lamotrigine are both safe and effective. But not all women can be switched from valproate to lamotrigine without the risk of seizures,” concludes Jakob Christensen.
“Prenatal exposure to antiseizure medications and risk of epilepsy in children of mothers with epilepsy” has been published in JAMA Network Open. The research was supported by NordForsk, the Nordic initiative on Health and Welfare; Independent Research Fund Denmark; the Danish Epilepsy Association; the Central Denmark Region and the Novo Nordisk Foundation.
