Mirtazapine

Pregnancy

The evidence is observational and mixed. MotherToBaby reports that pregnancy studies and case reports have not found an increased chance of birth defects, but its nearly 5,000-pregnancy count is an aggregate across reports and was not verified as a unique, non-overlapping cohort total. In a Danish propensity-score-matched cohort, 895 pregnancies were exposed in the major-malformation analysis: 31 of 895 exposed children (3.5%) and 152 of 3,580 comparators (4.3%) had a major malformation (OR 0.81, 95% CI 0.55 to 1.20). The miscarriage analysis included 1,900 exposed pregnancies and 7,600 matched comparators; spontaneous abortion occurred in 237 exposed pregnancies (12.5%) and 931 comparators (12.3%) (HR 1.04, 95% CI 0.91 to 1.20). A 2015 prospective study compared 357 exposed pregnancies with 357 SSRI-exposed and 357 general-control pregnancies. Major malformations did not differ significantly from the SSRI group; a higher rate versus general controls appeared only in a restricted analysis, then was no longer significant when pregnancies without first-trimester exposure were excluded. A 2013 Danish population study found an unadjusted association between mirtazapine and miscarriage in a depression-diagnosis subgroup (RR 2.23, 95% CI 1.34 to 3.70), but the authors noted few exposed pregnancies and no data on differences in depression severity. In 2025, a systematic review included 41 observational studies and found conflicting results for miscarriage and malformations, with overall low-quality evidence and substantial study overlap. A 2026 UK, Norway, and Sweden cohort found mirtazapine use during pregnancy associated with higher risk of small for gestational age, defined as birth weight below the 10th percentile. The main text did not provide a drug-specific absolute risk estimate, and confounding by indication could not be ruled out. A 2024 review of postpartum hemorrhage included mirtazapine among eligible exposures but did not establish a mirtazapine-specific risk estimate; its overall conclusion was about late-pregnancy antidepressant exposure as a group.

Breastfeeding

LactMed summarizes limited reports of low milk levels and advises monitoring exclusively breastfed infants for behavioral effects and growth. In a primary milk-transfer study of eight breastfeeding women taking a median 38 mg daily, mean relative infant doses were 1.5% for mirtazapine and 0.4% for desmethylmirtazapine; no adverse effects were seen, and the drug was detected in one of four tested infants. A 56-infant case series reported no sleeping or feeding problems; among 54 infants with third-trimester exposure, 14 were diagnosed with poor neonatal adaptation. The rate was lower among infants who were breastfed than among exclusively formula-fed infants, but this observational comparison cannot establish that breastfeeding caused the difference. NHS Specialist Pharmacy Service advises caution and infant monitoring, while preferring SSRIs or tricyclic antidepressants. The 2025 systematic review found breastfeeding evidence too limited to support firm conclusions.

What was tested

Pregnancy evidence includes nationwide registry cohorts, a prospective multicenter cohort, case series, case reports, and a 2025 systematic review. The 2022 Danish cohort used propensity-score matching; the 2013 miscarriage analysis was unadjusted for differences in depression severity. The 2026 UK, Norway, and Sweden cohort used maternal exposure, sibling comparisons, and paternal negative controls, but the mirtazapine-specific small-for-gestational-age finding was not accompanied by an absolute risk estimate in the main text or a second-trimester-only estimate. The VOMIT randomized trial tested short-term treatment of hyperemesis gravidarum in 59 pregnant participants, including 21 assigned to mirtazapine, for 14 days; its primary outcome was nausea and vomiting severity, not fetal safety. Its PubMed abstract reports more mild-to-moderate adverse events with mirtazapine and few, evenly distributed serious adverse events. The publisher full text could not be read because the site returned a security verification page, so the abstract was used. Breastfeeding evidence includes a milk-transfer study of eight women, small case reports, and a 56-infant case series. There are no large controlled studies of long-term infant outcomes through breast milk.