- Design
- Prospective international registry cohort (EURAP)
- Population
- 15,893 babies exposed to antiseizure medication in pregnancy, 1999-2023
- Primary outcome
- Birthweight centile
- Effect
- Polytherapy SGA OR 1.48 (1.29 to 1.70); topiramate −11.9 centiles vs lamotrigine
A prospective cohort from the EURAP international registry in The Lancet Neurology (September 2026) followed 15,893 babies born to women with epilepsy on antiseizure medication between 1999 and 2023, adjusting for a wide range of clinical and demographic factors.
Against monotherapy, polytherapy was associated with a lower birthweight centile (−2.74, 95% CI −4.22 to −1.26) and about 50% higher odds of small for gestational age (OR 1.48, 1.29 to 1.70), severe SGA (1.49) and low birthweight (1.50). Centile fell with each added drug — by 14 points with four. Compared with lamotrigine, centiles were lower with topiramate (−11.9), phenobarbital (−8.1), oxcarbazepine (−5.1), carbamazepine (−3.2), valproate (−2.5) and levetiracetam (−2.5). Lamotrigine combined with levetiracetam or valproate showed no difference from lamotrigine alone.
This is observational, and women on polytherapy or older drugs may have had more severe epilepsy. Some groups — topiramate, four drugs — were small. But growth restriction now joins malformations and neurodevelopment as a harm to weigh.
In practice this strengthens the case for lamotrigine where it suits, for avoiding topiramate and phenobarbital in women who may conceive, and for rationalising polytherapy before pregnancy. Phenobarbital remains widely used in India, which makes that last point particularly relevant here.
- Prefer lamotrigine monotherapy where it suits the epilepsy in women who may become pregnant
- Avoid topiramate and phenobarbital where alternatives exist in women of childbearing potential
- Reduce to monotherapy before conception where seizure control allows
- Arrange growth scans in pregnancies exposed to polytherapy, topiramate or phenobarbital
- Do not switch drugs abruptly in a pregnancy already under way — seek specialist advice
Why it matters
It adds fetal growth to the harms that differ between antiseizure drugs, and puts topiramate and phenobarbital near the top.
Don't overread it
Observational data cannot fully separate drug effects from epilepsy severity, and some drug groups were small.
The statistics, in plain English
A coefficient of −11.9 for topiramate means babies' birthweight averaged about 12 centile points lower than with lamotrigine — for example, 38th rather than 50th centile. An odds ratio of 1.48 for polytherapy means about 50% higher odds of a baby below the 10th centile.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for neurology, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free