Skip to content
This GLP
← Research
Safety

Pregnancy: the exposure the drug itself makes likelier

Weight loss restores ovulation, so unplanned pregnancy is a foreseeable consequence of these drugs working. Across 36 studies no consistent signal of harm has appeared — which is not the same as evidence of safety.

Ruth Alvarez6 min read
Where the evidence is, and where it stopsaround conceptionand first trimester36 studiescontinued use throughthe whole pregnancylimitedOne pharmacokinetic study found no semaglutide in human milk.No pooled risk estimate has been published for any of it.

This is the question where the gap between what a buyer is told and what is known is widest. It is also the one where the drug working is what creates the situation — which makes the silence around it harder to excuse than most of the silences counted in what sellers will not tell you.

Why it comes up at all

Many women with irregular or absent periods who had been unable to conceive become pregnant unexpectedly after losing weight on these drugs [1]. The mechanism is ordinary reproductive physiology: sustained weight loss restores ovulation. Someone who has spent years being told conception is unlikely is the person least likely to be using contraception and least likely to test early.

There is a second route to the same place. Tirzepatide can reduce the absorption of oral contraceptives around dose changes, which we cover separately in tirzepatide and oral contraception. Two independent mechanisms both raise the chance of an exposure nobody planned.

What the human data shows

A 2026 systematic review searched four databases and included 36 human studies of exposure before conception, during pregnancy, or during lactation [2]. Across large observational cohorts, exposure around conception or in early pregnancy was not consistently associated with a higher risk of major congenital malformations in adjusted analyses, nor with fetal growth restriction, stillbirth, or neonatal death, compared with insulin-treated or disease-matched controls.

Maternal outcomes — gestational diabetes, hypertensive disorders, preterm birth, gestational weight gain — were described as heterogeneous, with no reproducible safety signal in either direction. Lactation data were sparse: one pharmacokinetic study reported no detectable semaglutide transfer into human milk.

The animal data, in proportion

Some small-animal studies of exposure during pregnancy reported decreased fetal growth, skeletal and visceral anomalies, and embryonic death. Those findings are why the label advises stopping before a planned pregnancy, and they are also why the human cohorts were assembled. So far the human data has not reproduced the pattern — which is reassuring without being conclusive, because the doses, species and exposure windows differ.

What to do with this

If pregnancy is possible, the practical answer is contraception while taking the drug and a conversation with a clinician about timing before stopping it. If an exposure has already happened, the useful thing to know is that the large cohorts have not found a pattern of anomalies — and that this is a conversation for an obstetric clinician rather than a storefront chat window.

That last part is the constraint this market makes hardest. Most sellers tracked here publish nothing about who reviews a case or how to reach them, which is listed seller by seller in the questions they leave open. A question that has to be answered in days is a bad fit for a service that answers in weeks, and the related question of what stopping does is in stopping and weight regain.

Frequently asked

Can these drugs make pregnancy more likely?
Yes, and it is a recognized pattern. Sustained weight loss restores ovulation in women whose cycles had stopped, so people previously told conception was unlikely become pregnant unexpectedly.
Is exposure in early pregnancy harmful?
Across 36 studies, exposure around conception or in early pregnancy has not been consistently associated with major congenital malformations, growth restriction, stillbirth or neonatal death. The reviews stop short of calling that a finding of safety.
What do the reviews actually recommend?
That patients be counseled there is not enough evidence to predict adverse effects or their absence, and that anyone taking these drugs use contraception to prevent unintended pregnancy.
What about breastfeeding?
Lactation data are sparse. One pharmacokinetic study reported no detectable semaglutide transfer into human milk, which is a single study rather than a settled answer.

Sources

  1. [1] Drummond RF, et al. (2025). Glucagon-like peptide-1 receptor agonist use in pregnancy: a review American Journal of Obstetrics and Gynecology. PMID 39181497
  2. [2] Ozbek L, et al. (2026). Safety of GLP-1 and Dual GLP-1/GIP Receptor Agonists in Preconception, Pregnancy, and Lactation: A Systematic Review of Maternal, Fetal, and Neonatal Outcomes Diabetes, Obesity and Metabolism. PMID 41885132

More in Safety