Topiramate

Pregnancy

Birth defects: a US Medicaid cohort (Hernández-Díaz et al., 2018) found oral clefts in 4.1 per 1,000 of 2,425 first-trimester-exposed infants versus 1.1 per 1,000 unexposed infants (RR 2.90, 95% CI 1.56 to 5.40). A French nationwide cohort (Blotière et al., 2019; 517 topiramate exposures) reported cleft lip (OR 6.8, 95% CI 1.4 to 20.0); its sensitivity analysis also found a hypospadias signal, but the supplemental estimate was not retrieved. A 2025 North American registry study of first-trimester monotherapy found major malformations in 5.1% (26/510) after topiramate versus 2.1% (52/2,461) after lamotrigine (RR 2.2, 95% CI 1.2 to 4.0) and a higher cleft-lip risk. A 2026 South Korean cohort of 2,494,958 pregnancies found no significant overall anomaly increase for topiramate but reported signals for tetralogy of Fallot and talipes equinovarus; the abstract did not provide topiramate-specific counts or effect estimates. Fetal growth: the 2014 North American registry found small-for-gestational-age birth in 17.9% of 347 topiramate-exposed newborns versus 6.8% after lamotrigine (RR 2.4, 95% CI 1.8 to 3.3). A 2024 Nordic cohort associated topiramate monotherapy with small-for-gestational-age birth (aOR 1.48, 95% CI 1.18 to 1.85), and a 2026 EURAP cohort found lower birthweight centiles with topiramate than lamotrigine (adjusted b -11.93, 95% CI -16.72 to -7.15; n=248). The US label also reports a higher prevalence of small-for-gestational-age birth with higher doses and when use continued later in pregnancy compared with use that stopped before the third trimester. This label comparison is not a trimester-specific effect estimate. Preterm birth: the label and MotherToBaby report a possible association, but the 2017 registry's preterm estimates covered antiseizure-medication users as a group rather than topiramate specifically, so a topiramate-specific estimate was not verified. Neurodevelopment: evidence is mixed. Bjørk et al. (2022) reported increased ASD and ID hazard ratios in a Nordic cohort. The 2024 US study found no ASD association after adjustment for indication (topiramate aHR 0.96, 95% CI 0.56 to 1.65; 1,030 exposed, with exposure from week 19 to delivery). A 2026 late-pregnancy Canadian analysis found an association with combined neurodevelopmental disorders (p-aHR 1.56, 95% CI 1.04 to 2.34; 69 exposed cases), while the BMJ 2026 cohort found no meaningful association across most outcomes, with possible ID and learning-difficulty signals based on small numbers. A 2026 Taiwanese cohort of 2,196,156 births found that population associations were largely attenuated in sibling comparisons, except for a sensitivity signal when exposure extended 90 days before conception. A 2026 review of 14 studies reported 7 positive, 3 null, and 4 inconclusive findings, with topiramate-exposed groups ranging from 2 to 1,000. An Australian 2026 cognitive cohort included only 11 topiramate-exposed children, so its lower Full-Scale IQ estimate is imprecise. These observational studies cannot fully separate medication effects from epilepsy, treatment indication, dose, or shared family factors.

Breastfeeding

LactMed reports that maternal doses up to 200 mg daily produce low infant serum levels, while occasional infant sedation and diarrhea have been reported. Most infants in the available reports tolerated exposure through milk. LactMed lists diarrhea, drowsiness, irritability, weight gain, and developmental milestones for monitoring, especially in younger exclusively breastfed infants or when other anticonvulsant or psychotropic medicines are used. A 2014 case report describes a partially breastfed infant who developed recurrent watery diarrhea at 40 days of age; symptoms stopped within two days after breastfeeding was suspended. The case report does not state that the infant had slower weight gain. MotherToBaby also notes limited human reports and lists possible diarrhea, drowsiness, irritability, or trouble gaining weight.

What was tested

Pregnancy evidence includes prospective medication registries, population-based insurance and health-record cohorts, and a national French claims cohort. Researchers studied oral clefts and other malformations, fetal growth, preterm birth, and child neurodevelopment. Most pregnancy studies are observational, so treatment indication, epilepsy severity, dose, and other family factors may affect results. Breastfeeding evidence includes small milk and infant blood-level reports, clinical follow-up, and a single-infant diarrhea case report. No randomized pregnancy trial of topiramate was identified in this review.