Buprenorphine

Pregnancy

Most outcome research concerns opioid use disorder (OUD) treatment and compares buprenorphine with methadone, not with no treatment or across all pain formulations. Suarez et al. 2024 studied 13,360 Medicaid pregnancies with first-trimester treatment exposure, including 9,514 buprenorphine and 3,846 methadone pregnancies; the adjusted relative risk of any major malformation was 0.82 (95% CI 0.69-0.97) for buprenorphine versus methadone. The study had no unexposed comparison group, and secondary analyses found a higher risk of gastrointestinal malformations with buprenorphine than methadone. Suarez et al. 2022 examined 2,548,372 Medicaid pregnancies ending in live birth; among treatment in the 30 days before delivery, NOWS occurred in 52.0% of buprenorphine-exposed and 69.2% of methadone-exposed infants (adjusted RR 0.73). Early pregnancy buprenorphine was also associated with lower preterm birth, small-for-gestational-age, and low-birthweight risks than methadone, but these were treatment comparisons. Wilson et al. 2026 pooled 27 studies and found lower unadjusted risks of preterm birth, low birth weight, small for gestational age, and NOWS treatment with buprenorphine than methadone; the authors noted residual confounding, heterogeneous studies, and small sample sizes. A 140-person randomized trial published in 2026 compared weekly extended-release with sublingual buprenorphine; it found higher opioid abstinence during pregnancy with extended-release treatment, similar infant need for NOWS medication, and a borderline head-circumference difference at birth. That weekly formulation is not the monthly SUBLOCADE product. A 2026 US Medicaid cohort found no increased long-term neurodevelopmental risk with prenatal buprenorphine versus methadone overall (adjusted HR 0.81, 95% CI 0.70-0.94), but the association was not present for people starting treatment during pregnancy and no unexposed group was studied. A 2026 retrospective ultrasound study of 442 pregnancies found smaller third-trimester head-circumference percentiles in the traditional-dose group than controls (P=0.045); the very-low-dose group did not differ significantly from either group, and the study cannot establish cause. A separate 2026 cohort of 768 infants already diagnosed with NOWS reported less pharmacologic treatment and shorter hospital stays for buprenorphine-exposed infants than methadone-exposed infants, but its selected population limits generalization. MotherToBaby states that limited studies have not found higher miscarriage rates, but a primary miscarriage estimate was not verified in this review and that claim is omitted from the period summaries.

Breastfeeding

LactMed's May 2026 review reports low milk and infant levels across studies and supports breastfeeding for stable patients taking buprenorphine, with monitoring for infant drowsiness, breathing, weight gain, and development. ABM's 2023 protocol recommends breastfeeding with sublingual buprenorphine but stated that lactation data for monthly injectable extended-release products were lacking at that time. Nelson-Rodriguez et al. 2025 studied three people receiving monthly subcutaneous extended-release buprenorphine: mean buprenorphine concentration was 8.9 ng/mL in milk, the estimated relative infant dose was 1%, and buprenorphine was below quantitation in all six infant plasma samples. N-methyl-2-pyrrolidone (NMP), an excipient in that product, was found in milk in samples collected 1 to 5 hours after dosing, but was not detected in infant plasma; the authors said more data are needed before firm conclusions. SUBOXONE labeling says the available milk data are from buprenorphine treatment and that there are no data on the buprenorphine/naloxone combination. SUBLOCADE labeling calls for caution and weighing breastfeeding benefits and potential effects, while BUTRANS and BELBUCA labeling says breastfeeding is not recommended because of potential serious infant reactions, including sedation and respiratory depression. Most lactation evidence concerns OUD treatment, not analgesic patch or buccal products; long-term infant outcomes remain poorly studied.

What was tested

Pregnancy evidence includes large observational cohorts, a comparative meta-analysis, one randomized trial of weekly extended-release versus sublingual treatment, and smaller case series. Most human pregnancy studies concern OUD and compare one treatment with another; they do not provide a direct comparison with no exposure and do not establish equivalent risk for pain formulations. Breastfeeding evidence includes milk-level and infant-plasma studies, mostly in small samples, plus expert guidance. The 2025 monthly-injection study included three lactating participants. Product labels also summarize animal reproductive studies, which do not estimate human outcomes.