GLP-1s and Pregnancy: What to Know if You're Planning — or if You Conceive While on One

The short version

Two questions bring most mothers here, and they deserve different answers. If you are planning a pregnancy, the guidance is clear: GLP-1 medications are not used in pregnancy, so you stop before you conceive, on a planned washout¹. If you have already conceived while on one — which happens, because these drugs can quietly restore fertility² — the honest answer is more reassuring than the label alone suggests. The concern comes from animal studies, where GLP-1 exposure in pregnancy has been linked to reduced fetal growth and skeletal, visceral, and embryonic harm². But the human data, now pooled across tens of thousands of exposed pregnancies, have not shown a significant increase in major birth defects^3,4. The right move if you see a positive test is not panic — it is to stop the drug and call your OB promptly¹.

This is educational only, not medical advice.

Why GLP-1s are not used in pregnancy

The caution starts in the lab. Animal reproductive studies of GLP-1 receptor agonists have shown adverse offspring outcomes — decreased fetal growth, skeletal and visceral anomalies, and embryonic death — which is why prescribers plan a washout before conception². The Wegovy label puts the rest of the case in two sentences: weight loss offers no benefit to a pregnant patient and may cause fetal harm, and the drug should be discontinued in pregnant patients using it for weight reduction¹. Layered on top is a simple absence — there are no prospective human studies, so there is no body of trial safety data to lean on². When a drug shows a signal in animals and has thin human evidence, the conservative default in obstetrics is not to use it.

Novo Nordisk also runs a pregnancy exposure registry for semaglutide, and the label encourages exposed patients and their clinicians to enroll¹. If you are reading this because it already happened to you, joining is one concrete way the next mother gets a better answer than you did.

What the human data actually show

This is the part that should lower the temperature for anyone facing an unplanned exposure. Because inadvertent first-trimester exposure has become more common, researchers have now pooled the real-world evidence⁵:

• A 2026 systematic review and meta-analysis of periconceptional GLP-1 exposure — six studies, 286,599 women, 43,577 of them exposed — found no significant difference in the risk of major congenital malformations between exposed and unexposed pregnancies³.
• A separate 2026 meta-analysis of seven cohort studies covering more than 40,000 exposed pregnancies likewise found that exposure was not associated with a statistically significant increase in congenital malformations, nor in stillbirth, miscarriage, small-for-gestational-age, or preterm birth⁴.

Because the confidence interval is the whole story here, the numbers are worth seeing in columns rather than trusting a summary word like "reassuring":

Read the intervals, not only the point estimates. In the largest analysis the malformation figure sits almost exactly on 1.0 with tight bounds — that is a real null, not a shrug. The first-trimester and stillbirth intervals, by contrast, are wide enough to hide a genuine effect; that is what "underpowered" looks like in practice. The one interval that excludes 1.0 is the urinary-tract signal, and its own authors attribute it to residual confounding because it rests exclusively on unadjusted estimates⁴. The broader safety review adds that the evidence base overall remains sparse and is not a license to continue a GLP-1 into pregnancy⁵. Fair summary: the pooled human data are genuinely reassuring about an accidental early exposure, and still support the standard advice to stop the drug once pregnancy is known.

If you conceive while on a GLP-1: a calm plan

1. 1.Stop the medication. The guidance is to discontinue a GLP-1 once pregnancy is recognized¹. Do not take another dose while you sort out next steps.
2. 2.Call your OB or prescriber promptly. Tell them exactly which product and dose you were on and when your last dose was. This is routine for them, and the reassuring human data are worth hearing from your own clinician^3,4.
3. 3.Don't spiral. An early, inadvertent exposure is not known to meaningfully raise the risk of major birth defects based on the pooled human evidence^3,4. Standard early-pregnancy care — prenatal vitamins with folate, your normal first prenatal visit, and any monitoring your OB recommends — is the path forward.
4. 4.Ask about the weight-management plan for pregnancy. The goal shifts from weight loss to healthy gestational weight gain; your OB will help you reset expectations for the months ahead.

Planning ahead is still the better path

None of the above changes the ideal: if a pregnancy is on your horizon, plan the stop rather than discover it. The semaglutide labels are specific about the number — discontinue at least two months before a planned pregnancy, because of the drug's long half-life¹. That means contraception while you are still taking it, a deliberate washout, and only then trying to conceive. Our guide to trying to conceive on a GLP-1 lays out that sequence, and if you have PCOS, Ozempic for PCOS explains why fertility can return faster than you expect. For the postpartum side, see GLP-1s and breastfeeding. The tell of a serious prescribing program is that it raises pregnancy intentions before the first dose rather than treating reproduction as an afterthought.

How much to trust all of this

The two claims on this page do not deserve equal confidence, and it is worth separating them.

"Do not use a GLP-1 in pregnancy" is strong and settled. It rests on reproductive toxicity across three species, an explicit label instruction, and the plain logic that intentional weight loss has no benefit in pregnancy^1,2.

"An accidental early exposure probably did not harm your baby" is moderate, and improving. It rests on roughly 43,000 exposed pregnancies in one meta-analysis and more than 40,000 in another, with the malformation estimate sitting essentially at unity^3,4. That is a lot of women — but every one of those studies is observational, and both author groups say so themselves: one warns its findings "should not be interpreted as proof of safety" and calls for prospective cohorts and randomized trials³; the other rates the certainty of the evidence as low⁴. The maternal-fetal medicine review is blunter still, concluding that patients should be counselled there is not enough evidence to predict adverse effects of periconceptional exposure or the lack of them².

So: reassurance from quantity, not from design. What would upgrade it is prospective registry data — which is exactly why the exposure registry on the label exists, and why enrolling matters. What would downgrade it is a large adjusted analysis reproducing the urinary-tract signal. Neither has happened yet, and either could.