GLP-1 While Breastfeeding: The Real-Talk Safety Rundown

The answer I would bring to my clinician

Small human milk studies exist for semaglutide and tirzepatide, but they do not establish long-term safety for every breastfed infant. Advice differs by medicine and formulation. I would discuss the exact product, why I need treatment, my baby's age and feeding pattern with my prescriber and pediatrician before starting or continuing^1,3.

The one thing I'll say plainly: nothing I write here — and nothing anyone writes anywhere — can clear you to take a prescription medication while nursing. That's a real clinician looking at your specific situation.

I would bring the exact medication name and formulation to that appointment. My aim is to make the conversation more useful, especially when a search result offers a confident yes or no without naming the product.

What the semaglutide milk study measured

The published 2024 study collected milk from eight women using injected semaglutide. Semaglutide was not detected in the collected samples. The infants were 4–23 months old and had mixed feeding, so this was not a study establishing safety in exclusively breastfed newborns².

LactMed summarizes the same research. I would not count the LactMed entry and the published paper as two independent studies. A reference that explains a study is useful, but it does not double the number of mothers or babies studied¹.

For me, the reassuring part is the low measured transfer. The unanswered part includes longer-term infant outcomes, higher treatment doses, milk production and maternal nutrition during treatment. A small study can answer a narrow question without answering the whole decision.

Tirzepatide has human milk data too

The current Zepbound label describes 11 lactating adults who received one 5 mg dose. Tirzepatide was undetectable in 164 of 171 milk samples; the total detected in the remaining samples was less than 0.02% of the administered maternal dose. The label says there are no available data on effects on the breastfed infant or milk production⁴.

That is a single-dose study, which matters to my interpretation of a medicine usually taken repeatedly. I would not translate its milk measurement into a claim that a full course is proven safe for my baby.

LactMed advises caution with tirzepatide during breastfeeding, particularly with a newborn or preterm infant. I would bring that distinction into the conversation rather than treat all nursing situations as equivalent³.

The labels do not all say the same thing

The statement that every semaglutide and tirzepatide label prohibits breastfeeding is too broad. Zepbound's label calls for weighing breastfeeding benefits, maternal treatment need and potential infant effects. The Wegovy label separates the injection from the oral tablet^4,5.

For Wegovy tablets, breastfeeding is not recommended because the absorption enhancer SNAC and its metabolites can enter milk and could accumulate in an infant. The injection's lactation section instead calls for an individual benefit-risk assessment. That difference is why I would take a photograph of the exact prescription to the appointment⁵.

I would not switch from a pill to an injection on my own after reading that distinction. A formulation difference is a reason for a prescribing conversation, not an instruction to find a workaround.

The questions I would write down

I would ask what condition we are treating now and what alternatives fit breastfeeding. I would ask how my baby's age, prematurity, growth and reliance on breast milk affect the assessment. I would also ask whether waiting until after weaning would change the balance in my situation.

My own food and fluid intake belongs in that discussion. If treatment makes it difficult to eat or drink, I would contact the clinician rather than assume low milk transfer settles every concern. I would ask who should review changes in milk supply or the baby's feeding and growth.

For a telehealth intake, I would disclose breastfeeding accurately. A program's eligibility policy is not the same thing as a universal drug-label rule. If it declines treatment, I would take the question to my regular clinician rather than omit nursing on another form.

The decision needs a follow-up plan

Before any treatment decision, I would want to know who follows me, who follows the baby, and what changes should prompt a call. I would ask for the plan in plain language so I do not have to reconstruct it from a label later.

My postpartum guide covers the broader timing conversation. For breastfeeding specifically, I would keep the conclusion narrower: early milk-transfer evidence is useful, important questions remain, and the answer must fit the medicine, the parent and the infant together.