Gabapentin

Pregnancy

MotherToBaby summarizes earlier human studies as not showing an increased chance of miscarriage or major birth defects. In the 2025 North American Antiepileptic Drug Pregnancy Registry study, 4 of 270 first-trimester gabapentin monotherapy exposures had major malformations (1.5%); the authors reported that the results did not support a meaningful risk elevation for gabapentin compared with lamotrigine. The registry figure is verified, but it is not a comparison with an unexposed group. Patorno et al. studied 1,753,865 Medicaid-eligible pregnancies, including 4,642 first-trimester gabapentin exposures; the adjusted relative risk for major malformations overall was 1.07 (95% CI 0.94 to 1.21), while two or more dispensings were linked to a possible cardiac-defect association (RR 1.40, 95% CI 1.03 to 1.90). A 2024 Manitoba cohort of 289,227 pregnancies, including 870 gabapentin exposures, found no adjusted increase in major congenital malformations (aRR 1.00, 95% CI 0.73 to 1.36), but its pooled meta-analysis used unadjusted estimates and found higher malformation estimates. A 2026 European case-malformed-control study found a ventricular septal defect association for gabapentinoids combined (adjusted OR 2.5, 95% CI 1.1 to 5.3), based on 37 exposed registrants; it did not estimate gabapentin-specific risk and did not measure incidence among all pregnancies. A 2025 Scottish population cohort reported higher adjusted odds of its composite pregnancy-loss outcome with gabapentin monotherapy (aOR 1.33, 95% CI 1.21 to 1.46; 2,624 exposed pregnancies). The study defined pregnancy loss as any outcome other than live birth, including miscarriage, stillbirth, and termination, so the result does not isolate miscarriage. In analyses limited to pregnancies with recorded epilepsy, gabapentin was not among the medicines that remained associated with higher odds; the authors also reported that analyses separating spontaneous loss from termination were less certain. Prescription exposure does not prove that a medicine was taken, and illness, other medicines, and social factors may affect results. The DailyMed label separately reports animal developmental toxicity, including fetal abnormalities and embryofetal mortality, at doses similar to or below clinical doses; animal findings are not equivalent to human outcomes.

Breastfeeding

MotherToBaby reports small transfer into breast milk and no side effects in the limited published infant reports. LactMed, revised April 15, 2025, says maternal doses up to 2.1 grams daily produce relatively low infant serum levels and recommends monitoring for drowsiness, adequate weight gain, and developmental milestones, especially for younger or exclusively breastfed infants or when other medicines are used. LactMed calls gabapentin an acceptable choice for refractory restless legs syndrome during lactation, which is a limited use statement rather than a finding for every condition. A 2025 pharmacokinetic study enrolled 21 healthy pregnant people planning to breastfeed after cesarean birth; a 300 mg dose before surgery followed by 300 mg every 6 hours for 48 hours was estimated to result in a relative infant dose below 0.15% during that period. The study measured drug transfer, not long-term infant development or a large number of infant health outcomes. The NHS page reports very small milk transfer and advises infant monitoring; DailyMed states that effects on the breastfed infant and milk production are unknown. The evidence supports describing low transfer and limited reports, not a conclusion that infant effects cannot occur.

What was tested

The pregnancy evidence includes a North American pregnancy registry, a large US Medicaid cohort, a Manitoba population cohort, a Scottish linked-record cohort, a European case-malformed-control study, smaller prospective studies, and systematic reviews. Most pregnancy findings are observational, so prescription records may not show whether the medicine was taken and other conditions or medicines can affect results. The 2026 West Virginia study included 77,059 live singleton births but only 177 gabapentin-exposed infants, of whom 105 also had opioid exposure; it cannot address losses before live birth and its authors state that causal conclusions cannot be drawn. Breastfeeding evidence includes small milk-level studies, a 21-person short postpartum transfer study, a few infant serum measurements, and limited infant follow-up. Long-term infant outcomes from breastfeeding exposure have not been established in the sources reviewed.