Anti-seizure medication exposure in pregnancy showed mixed signals for orofacial clefts
Maternal use of anti-seizure medications during pregnancy was associated with uneven patterns of risk for orofacial clefts in a large United Kingdom primary care cohort, with the clearest signal appearing for valproate and for first-trimester polytherapy. The study drew on CPRD GOLD records from 393,281 mothers and 518,050 babies, giving it a scale large enough to examine uncommon congenital outcomes without relying on small registry fragments or single-center experience.
The biological concern is familiar to obstetric neurologists and neurologists alike: embryologic fusion of the lip and palate occurs early, and drug exposure during that narrow developmental window can disrupt normal morphogenesis. Even so, the absolute number of exposed pregnancies that resulted in clefts remained small, and most children with orofacial clefts were not exposed to anti-seizure treatment at all.
Valproate carried the most consistent association, but confounding remained a concern
Prescription data identified 3,651 women exposed to anti-seizure medications during the preconception period and 3,199 during the first trimester, regardless of indication. Within that exposed population, valproate stood out. Prescription during both preconception and the first trimester was linked to higher odds of orofacial clefts, although the association weakened after adjustment for potential confounders.
The adjusted finding matters because epilepsy severity, comorbidity burden, and concurrent medication use can all cluster with teratogenic risk. A prescription record alone does not distinguish a woman taking a single low-dose drug for migraine from a woman with refractory epilepsy receiving multiple agents. The paper’s cautious phrasing reflects that reality: the signal was present, but not immune to confounding by indication or related clinical factors.
A single sentence captures the core message: valproate remained the most concerning exposure in the cohort.
Polytherapy produced the sharpest increase in odds
The strongest relative risk emerged with ASM polytherapy in the first trimester. Children exposed to more than one anti-seizure medication were six times as likely to have an orofacial cleft as children exposed to monotherapy. In teratology, combination treatment often marks greater underlying disease complexity, yet it also raises the possibility of additive embryotoxic effects during organogenesis.
This pattern fits clinical intuition. Multiple agents can mean higher cumulative pharmacologic burden, more difficult-to-control seizures, or both. Each of those pathways can complicate pregnancy care, and each can blur the boundary between medication effect and maternal disease effect. The finding does not isolate one culprit, but it does sharpen the attention given to regimen selection before conception and during early gestation.
Carbamazepine, pregabalin, levetiracetam and the drugs without a clear signal
Several commonly used agents produced weaker and less certain associations. Carbamazepine, pregabalin, and levetiracetam showed limited evidence for increased odds of orofacial clefts. In contrast, no evidence of an association appeared for gabapentin or lamotrigine.
No cases of orofacial cleft were identified among children born to mothers exposed to topiramate or phenytoin in this dataset. The absence of observed cases is informative but not definitive; when exposures are uncommon and the outcome is rare, a null count can reflect limited statistical opportunity as much as a true absence of risk. Still, the pattern adds to the broader message that not every anti-seizure medication carries the same developmental profile.
What the cohort design can and cannot prove
This was an observational study based on linked United Kingdom primary care data, so the analysis depended on prescription records rather than directly observed ingestion, and it captured pregnancies through routine healthcare systems rather than a prospectively enrolled teratology clinic. That design brings reach and real-world relevance, but it also leaves room for residual confounding, missing information on adherence, and incomplete capture of some clinical context. The authors adjusted for potential confounders, yet any pharmacoepidemiology study of congenital anomalies must still contend with the fact that treatment choice is rarely random.
The scale of the cohort still gives the findings weight. A study that includes more than half a million babies can identify patterns that smaller series miss, especially when the outcome is rare. It can also show where signals do not hold up under adjustment, which is just as useful for clinical decision-making as a positive association.
Bedside implications for women of reproductive age
For clinicians managing women who may become pregnant, the study reinforces the long-standing need for preconception medication review, especially when valproate or combination therapy is involved. Pregnancy exposure to anti-seizure drugs cannot be interpreted as uniformly dangerous, but neither can it be dismissed as clinically neutral. The data argue for careful regimen planning before conception whenever possible, with particular attention to the first trimester, when facial development is most vulnerable.
The practical consequence is not broad alarm. It is more precise than that. Most exposed pregnancies in this cohort did not end in an orofacial cleft, and most children with clefts had no anti-seizure medication exposure at all. For counseling, that dual reality matters: the absolute risk appears limited, yet certain treatments deserve heightened caution because the developmental stakes are concentrated early and the window for prevention is narrow.
The study closes one question and leaves several others open. Dose, adherence, seizure indication, and the clinical reasons behind switching or combining drugs were not fully resolved here, and those details will shape how future pregnancy safety work interprets the same medications in different women.
Full citation and study context
The article, Risk of orofacial clefts in offspring following maternal exposure to anti-seizure medications in a UK CPRD GOLD observational study, was published in Scientific Reports in 2026. It analyzed prescription exposure during the preconception period and first trimester and linked those exposures to orofacial cleft outcomes in offspring using routine UK healthcare data. The funding sources had no role in study design, data collection, analysis, interpretation, or publication decisions.