Overview
A new Australian cohort study adds to growing evidence that some antiseizure medications taken during pregnancy may affect a child’s later cognitive development. In this study, children exposed in the womb to valproate, levetiracetam, topiramate, or carbamazepine performed worse in several cognitive domains than children who were not exposed to antiseizure medications. Children exposed to lamotrigine, however, performed similarly to unexposed peers.
This is an important topic because many women with epilepsy need to continue antiseizure medication during pregnancy to prevent seizures, which can also endanger both mother and baby. The challenge is finding a treatment that balances seizure control with fetal safety. Neurocognitive outcomes are especially important because they can affect learning, memory, attention, language, and school performance over time.
Why this study matters
Most research on antiseizure medication in pregnancy has focused on birth defects. But many families and clinicians are equally concerned about long-term development, including IQ, language, executive function, and academic achievement. Some older drugs, especially valproate, have already been linked to poorer neurodevelopmental outcomes. For newer medicines such as levetiracetam and topiramate, data have been more limited.
This study is valuable because it directly compared several common antiseizure medications in a single cohort and included a group of children with no prenatal exposure to antiseizure medication. The children were assessed by a neuropsychologist who did not know which medication they had been exposed to, which helps reduce bias.
Study design
This was a semiprospective cohort study based on the Raoul Wallenberg Australian Pregnancy Register of Antiepileptic Drugs. Researchers followed children from birth and later performed standardized neuropsychological testing when the children were between 3 and 18 years old.
A total of 110 children were assessed:
– 20 exposed to carbamazepine monotherapy
– 23 exposed to lamotrigine monotherapy
– 20 exposed to levetiracetam monotherapy
– 19 exposed to valproate monotherapy
– 11 exposed to topiramate monotherapy
– 19 unexposed to antiseizure medications
Children with major congenital anomalies were excluded. The main outcome was Full-Scale IQ (FSIQ). Researchers also looked at other cognitive domains, including processing speed, verbal comprehension, visuospatial reasoning, memory, language, executive function, and academic skills.
To make the comparison more accurate, the analysis adjusted for factors that can also influence child development, such as maternal IQ, epilepsy-related characteristics, and relevant pregnancy or birth factors.
Main findings
The results showed a clear pattern: children exposed in utero to valproate, levetiracetam, topiramate, or carbamazepine had lower scores in several cognitive areas than unexposed children.
The most affected domains included:
– Processing speed
– Verbal comprehension
– Visuospatial reasoning
– Academic skills
The differences were moderate to large, with Cohen d values ranging from about -0.8 to -1.0, which suggests a meaningful shift in performance rather than a trivial difference.
Two findings stood out most strongly:
– Topiramate exposure was associated with the largest reduction in FSIQ, with an estimated difference of -17.4 points compared with unexposed children.
– Levetiracetam exposure was also associated with a substantial reduction in FSIQ, with an estimated difference of -13.9 points.
Children exposed to lamotrigine performed similarly to children who were not exposed to antiseizure medications, suggesting a more favorable neurocognitive profile in this study.
Interestingly, the study did not find a clear dose-response relationship. In other words, higher doses were not clearly linked to worse outcomes in this cohort, likely because the number of children in each medication group was relatively small. That means the absence of a dose effect should not be interpreted as proof that dose does not matter.
How to interpret the results
The findings suggest that prenatal exposure to some antiseizure medications may be associated with specific learning and thinking difficulties later in childhood. This does not mean every exposed child will have problems, and it does not prove that the medication alone caused the outcome. Epilepsy itself, maternal health, genetics, seizure control, and pregnancy factors can all play a role.
Still, the pattern is clinically important. Valproate has long been recognized as the antiseizure medication with the greatest concern in pregnancy for fetal development. The fact that valproate showed an effect here is consistent with prior research, although the apparent impact was somewhat less pronounced than expected in this cohort.
The association with levetiracetam and topiramate is especially noteworthy because these medications are often considered when trying to avoid older, higher-risk drugs. Levetiracetam is widely used in pregnancy because earlier studies suggested it may be safer than valproate. Topiramate is also used for epilepsy and migraine prevention, but its pregnancy safety profile has become a major concern in recent years.
Carbamazepine also showed lower scores in some domains, although its risk is generally considered less severe than valproate. Lamotrigine, by contrast, continued to look comparatively reassuring.
Clinical implications
For women of childbearing age with epilepsy, treatment decisions should always be individualized. The key goals are:
– Good seizure control before and during pregnancy
– Use of the lowest effective dose when possible
– Preference for medications with better reproductive safety data when appropriate
– Preconception counseling whenever possible
– Folic acid supplementation and careful prenatal care
This study supports the idea that medication choice matters not only for birth defects, but also for a child’s later learning and cognitive development. In practical terms, it strengthens the case for avoiding valproate during pregnancy when reasonable alternatives exist. It also raises caution about topiramate and levetiracetam, especially when other suitable options are available.
At the same time, seizure control remains essential. Uncontrolled seizures can cause falls, trauma, hypoxia, and other complications that may also threaten fetal well-being. Therefore, no one should stop an antiseizure medication abruptly without medical guidance. Treatment changes should be planned well in advance, ideally before conception, and supervised by a neurologist or epilepsy specialist.
Why lamotrigine stands out
Lamotrigine is often considered one of the preferred antiseizure medications in pregnancy because it has generally shown a more favorable profile in previous studies. In this cohort, children exposed to lamotrigine had cognitive performance similar to unexposed children.
That does not guarantee that lamotrigine is risk-free, and every pregnancy is different. However, this result adds to the broader literature suggesting that lamotrigine may be a relatively safer option for neurodevelopmental outcomes compared with several other antiseizure medications.
Strengths and limitations
This study had several strengths:
– Children were assessed with standardized neuropsychological testing
– The examiner was blinded to prenatal exposure status
– Multiple cognitive domains were measured, not just IQ
– Researchers adjusted for important confounders such as maternal IQ and epilepsy characteristics
There were also limitations:
– The sample size was small, especially within each medication group
– The study may not fully separate medication effects from epilepsy-related effects
– The cohort came from a single national register, which may limit generalizability
– Dose-response analysis was underpowered
– Some developmental differences may emerge later or be influenced by environmental factors not fully captured in the study
Because of these limitations, the findings should be seen as informative rather than definitive. Larger studies with long-term follow-up are needed to better clarify the relative safety of individual antiseizure medications.
What families should know
If you are pregnant or planning pregnancy and take an antiseizure medication, do not stop the medicine on your own. Instead, speak with your neurologist, obstetrician, or epilepsy specialist. The safest plan depends on your seizure type, seizure history, and the medication that controls your epilepsy best.
Questions to discuss with your doctor may include:
– Is my current medication the best choice for pregnancy?
– Would switching to a different antiseizure medication reduce risk?
– What dose is needed to keep seizures controlled?
– Should I take folic acid, and at what dose?
– How will my medication levels be monitored during pregnancy?
– What developmental follow-up is recommended for my child?
Early developmental screening can be helpful for children with prenatal antiseizure medication exposure, especially if there are concerns about speech, attention, learning, or school readiness.
Bottom line
In this Australian cohort, prenatal exposure to valproate, levetiracetam, topiramate, and carbamazepine was linked to worse neurocognitive outcomes in children, while lamotrigine exposure was not. The study reinforces the importance of careful medication selection before and during pregnancy and highlights the need to balance seizure control with long-term child development.