Methyldopa

Pregnancy

DailyMed reports no fetal harm in the mouse, rabbit, and rat reproduction studies it cites, while noting animal studies may not predict human response. Hoeltzenbein et al. (2017) prospectively followed 261 pregnancies with first-trimester methyldopa exposure and 526 comparison pregnancies without chronic hypertension. Major birth defects occurred in 3.7% of exposed pregnancies and 2.5% of comparison pregnancies; the adjusted odds ratio was 1.24 (95% confidence interval 0.4 to 4.0), which was not statistically significant. The authors noted a tendency toward more spontaneous abortions, more preterm births, lower birthweight, and smaller head circumference in exposed boys. In a sensitivity analysis comparing methyldopa monotherapy with metoprolol, the differences in early pregnancy loss, growth restriction, and head circumference were not found, but the preterm birth result remained. Ishikawa et al. (2023) studied 93 first-trimester methyldopa exposures in a Japanese claims database; the adjusted odds ratio for major malformations was 0.921 (95% confidence interval 0.331 to 2.564), and the authors said an increased risk was not suggested. The wide interval means a small increase or decrease cannot be ruled out. Hup et al. (2025) reviewed 23 randomized trials with 3,989 women. Methyldopa reduced severe hypertension compared with placebo or no treatment (relative risk 0.44, 95% confidence interval 0.20 to 0.99), while no significant differences were detected for most other outcomes; the review rated overall study quality low to moderate. Cockburn et al. (1982) followed 195 children from a trial to age 7 and one-half. The frequency of health, physical or mental handicap, sight, hearing, and behavior problems was similar in treated and untreated groups, and there were no significant differences on 14 ability tests. Chan et al. (2010) followed 25 children exposed to methyldopa and 42 children in a comparison group. The methyldopa group had lower mean full-scale IQ scores (105.24 versus 111.90, p=0.043) and performance IQ scores (98.80 versus 110.19, p=0.002). Maternal IQ predicted child full-scale IQ, so this small observational study cannot show that the medicine caused the difference. Alkhalaf et al. (2025) studied 179,024 children in Wales. Learning and communication difficulties were associated with methyldopa exposure and untreated maternal hypertension, suggesting the illness may explain some of the pattern; the abstract reported autism associations with antihypertensive medication overall and beta-blocker exposure, but did not give a methyldopa-specific autism estimate; it found no ADHD association with high blood pressure disorders in pregnancy or antihypertensive medication. A 2026 California Medicaid cohort included 2,281 singleton births, including 653 methyldopa initiators. Its composite complication rate was 19% among methyldopa initiators and 23% among labetalol initiators (adjusted risk ratio 0.82, 95% confidence interval 0.68 to 1.00). The authors cautioned that residual confounding may explain the apparent difference. A 2026 post-hoc analysis of early fetal-growth-restriction pregnancies included 1,453 women, 849 of whom received oral antihypertensives; exposure groups ranged from 14 to 230. It found no significant differences in birthweight, gestational age, or adverse perinatal outcomes between methyldopa, labetalol, and nifedipine, but the authors cautioned that some groups were small and could be biased by treatment selection.

Breastfeeding

LactMed's record, last revised November 15, 2024, summarizes small milk-level and infant reports. White et al. (1985) studied three breastfeeding women and their infants; after a 500 mg maternal dose, milk concentrations ranged from 0.02 to 1.14 micrograms per milliliter. The authors estimated that at a maternal daily dose of 1 gram, the infant's average daily exposure would be about 0.02% of the maternal dose. Two of three infant blood samples were below the detection limit at six hours, while one infant had a measurable level at ten hours. Hauser et al. (1985) followed one nursing infant for three months; the drug was found in the infant's urine, but no adverse clinical effects were noted. LactMed reports no acute or long-term adverse effects in 15 infants, ranging from less than one week to eight weeks old, across four reports. The reports are small. A 2026 study of colostrum collected 2 to 4 days after birth found lower TSG-6 and related microRNA levels in a group with gestational hypertension, but all hypertensive participants received methyldopa, so the study could not separate the medicine from the condition. The authors performed no functional tests and reported no infant clinical outcomes. A 2025 postpartum trial studied maternal blood-pressure control and early depression scores after continuing methyldopa or changing to captopril; it did not measure infant exposure through milk. A 2026 Danish registry cohort found no increased peripartum depression association, but the postpartum-only methyldopa estimate was based on 10 cases and had a wide confidence interval.

What was tested

Pregnancy evidence includes animal reproduction studies, first-trimester observational cohorts, claims data, randomized treatment trials, and follow-up studies of children's development. The published reports do not isolate every trimester, and observational comparisons can be affected by the reason a person needed treatment. Breastfeeding evidence consists mainly of small maternal milk and infant blood or urine studies. The postpartum randomized trial assessed maternal blood-pressure control and early depression symptoms, not infant exposure through milk. No large breastfeeding cohort was found in the sources reviewed.