Pantoprazole

Pregnancy

The NHS says pantoprazole is not usually recommended during pregnancy because it has very little pregnancy information; it notes that omeprazole has more information. The PROTONIX label reports that published observational studies did not demonstrate an association with major malformations or other adverse pregnancy outcomes, but says the observational data cannot establish or exclude a drug-related risk. Its label warning reads: "Based on animal data, may cause fetal harm." The label describes rat and rabbit organogenesis studies with no evidence of adverse developmental outcomes, plus a separate rat pre- and postnatal study. Pregnant rats received 5, 15, or 30 mg/kg/day from gestation day 6 through lactation day 21, and pups also received direct doses from postnatal day 4 through 21; bone morphology changes were reported in pups exposed before birth, through milk, and by direct dosing. The label also reports increased pup mortality or moribundity and lower body weight and weight gain during direct postnatal dosing at 5 mg/kg/day and higher doses, with bone changes reported at these dose levels. This design cannot isolate a milk-only contribution. Direct pantoprazole human data are limited. In a prospective controlled study, Diav-Citrin et al. followed 53 pantoprazole-exposed pregnancies, including 47 with first-trimester exposure; one major anomaly occurred among 48 evaluated pantoprazole pregnancies (2.1%), compared with 30 of 792 controls (3.8%). The small exposed group limits precision. In the Danish registry cohort by Pasternak and Hviid, 3,651 infants had first-trimester exposure to any PPI; the adjusted prevalence odds ratio for major birth defects was 1.10 (95% CI 0.91-1.34), and secondary analyses of individual PPIs did not show significant increases. The 2010 PubMed abstract does not report the pantoprazole-specific subgroup size that appears in later label review material. Class-wide research is mixed. Gill et al.'s 2009 meta-analysis included 1,530 PPI-exposed and 133,410 unexposed pregnancies and found no significant increase in major malformations (OR 1.12, 95% CI 0.86-1.45) or spontaneous abortion (OR 1.29, 95% CI 0.84-1.97). Li et al.'s 2020 meta-analysis of 26 observational studies reported an association with congenital malformations (OR 1.28, 95% CI 1.09-1.52), but not with abortion, stillbirth, neonatal death, preterm birth, or low birth weight; it could not assess individual PPI drugs. Peron et al.'s 2023 meta-analysis of 11 studies and 5,618 exposed pregnancies found no significant increase in overall or specific malformations (overall OR 1.10, 95% CI 0.95-1.26), while noting insufficient evidence to assess individual drugs. Choi et al.'s South Korean cohort included 2,696,216 pregnancies, with 40,540 exposed to any PPI in the first trimester. The main adjusted analysis found small increases for major malformations (RR 1.07, 95% CI 1.02-1.13) and congenital heart defects (RR 1.09, 95% CI 1.01-1.17), while sibling-controlled estimates were not statistically significant. This study was not pantoprazole-specific. Breddels et al.'s Swedish cohort included 1,089,514 singleton deliveries and associated PPI use with higher odds of preeclampsia, gestational diabetes, preterm birth, and small-for-gestational-age; second- and third-trimester PPI exposure was associated with preeclampsia, but the cohort also was not pantoprazole-specific and cannot establish cause. New evidence includes a 2025 FAERS disproportionality analysis with exploratory pantoprazole signals for premature labor, low birth weight, and morning sickness; the authors say the signals are hypothesis-generating and require confirmation. A 2025 chick-embryo experiment reported a possible higher-dose neural-tube trend, but the between-group result was not statistically significant (p=0.548), the sample was small, and the experiment does not establish a human effect. Overall, human evidence is mostly class-wide, direct pantoprazole pregnancy data remain limited, and major sources differ in how they frame the information gap and animal-label warning.

Breastfeeding

LactMed, revised June 15, 2026, says maternal pantoprazole 40 mg daily produces low milk levels. It summarizes one mother who took a single 40 mg dose at 10 months postpartum: milk concentrations were 36 mcg/L and 24 mcg/L at 2 and 4 hours, were below 10 mcg/L at 6, 8, and 24 hours, and the estimated infant dose was 0.14% of the maternal weight-adjusted dose. The mother reported no infant adverse events. LactMed reports that published infant-level and breastfed-infant outcome information was not found. Bor et al. (2026) measured pantoprazole in milk and plasma in 16 lactating women taking 40 mg daily after they had discontinued direct breastfeeding. Pantoprazole appeared in 26 of 149 milk samples (17.4%). The authors estimated a worst-case infant dose of about 0.011 mg/kg/day from milk concentration, but the expressed milk was discarded and no infants were exposed in the study. The full text says maternal body weight was not recorded, so a formal relative infant dose could not be calculated; it also reports no infant clinical outcome data. The study adds milk-transfer measurements, not evidence about outcomes in exposed infants. NHS SPS lists pantoprazole and omeprazole as preferred PPIs during breastfeeding and limits its recommendation to full-term, healthy infants. It also says the evidence for these recommendations is generally poor and limited. NHS patient guidance says milk transfer is tiny and advises monitoring for symptoms. The PROTONIX label describes one nursing mother, reports no infant effects noticed by the mother, and states that effects on milk production have not been studied. The label's separate rat-pup study combined prenatal, milk, and direct postnatal exposure, so it cannot identify a milk-only effect in humans.

What was tested

Pregnancy evidence includes a small prospective controlled study of pantoprazole-exposed pregnancies; registry cohorts of PPI-class exposure; meta-analyses of observational studies; a Swedish cohort of maternal and birth outcomes; a 2025 analysis of spontaneous adverse-event reports; a 2026 MarketScan maternal signal screen; and a chick-embryo experiment. These designs differ in drug specificity, exposure timing, outcomes, and ability to control for the condition being treated. The animal label study involved pregnant rats and direct dosing of pups as well as exposure before birth and through milk. Breastfeeding evidence includes a one-mother milk-level study with one mother-reported infant outcome and a 2026 milk-level study of 16 women who had already stopped direct breastfeeding. Neither milk-level study followed infants exposed during breastfeeding. No clinical study measuring infant blood levels or systematic infant outcomes was identified in LactMed as of June 15, 2026.