GLP-1 Drugs and Pregnancy: Fertility and Safety
GLP-1 Drugs for Teens
8 weeks washout
Current guidelines recommend stopping GLP-1 receptor agonists at least 8 weeks before planned conception, based on the drug's half-life and the precautionary principle given limited human pregnancy data.
Fotheringham et al., Drugs, 2026
Fotheringham et al., Drugs, 2026
View as imageTens of millions of reproductive-age women now take GLP-1 receptor agonists for diabetes or weight loss. The drugs are not approved during pregnancy, but pregnancies occur: sometimes planned after treatment, sometimes unplanned while still on the medication. The phenomenon of "Ozempic babies," unintended pregnancies in women who were previously subfertile and restored ovulatory cycles through GLP-1RA-mediated weight loss, has generated public attention that has outpaced the evidence. This article examines what the research actually shows about GLP-1 agonists and fertility effects, periconceptional exposure outcomes, and fetal safety. For how these drugs affect different populations, see our pillar article on GLP-1 drugs for teens.
Key Takeaways
- A systematic review of semaglutide exposure in pregnancy found no significant increase in major congenital malformations compared to obesity- and diabetes-matched controls (Mandal et al., EJOG, 2026)
- Unintentional periconceptional GLP-1RA exposure did not increase risks of adverse pregnancy outcomes in a cohort study, though sample sizes remain small (Chou et al., Diabetes, Obesity and Metabolism, 2026)
- Animal studies show skeletal anomalies at supratherapeutic GLP-1RA doses, but GLP-1RAs are not classified as direct teratogens (Fotheringham et al., Drugs, 2026)
- GLP-1RA prescribing trends show increasing use among reproductive-age women, with rising rates of periconceptional exposure (Lessard et al., Obstetrics and Gynecology, 2026)
- Current guidelines recommend stopping GLP-1RAs at least 8 weeks before planned conception and using effective contraception during treatment
- GLP-1RAs improve fertility in women with PCOS and obesity by restoring ovulatory cycles through weight loss and insulin sensitization (Voros et al., IJMS, 2026)
The fertility paradox: treatment restores what it cannot protect
GLP-1 receptor agonists create a reproductive paradox. The weight loss and insulin sensitization they produce can restore ovulatory cycles in women with obesity-related anovulation and PCOS, increasing fertility. But the drugs are contraindicated in pregnancy, meaning the very women who become more fertile while taking GLP-1RAs must stop treatment before conceiving.
Roberts et al. (2026) reviewed the relationship between obesity, GLP-1RAs, and female reproductive health. They noted that obesity impairs fertility through multiple mechanisms: anovulation, insulin resistance, chronic inflammation, and altered endometrial receptivity. GLP-1RA-mediated weight loss addresses several of these simultaneously, making conception more likely.[1]
Voros et al. (2026) specifically examined GLP-1RA effects on reproductive health, integrating IVF data and ovarian physiology. They found that GLP-1 receptors are expressed on ovarian granulosa cells and that preconception GLP-1RA treatment improved IVF outcomes in obese PCOS patients. The improved fertility was attributed to both weight-dependent and potentially weight-independent ovarian mechanisms.[2]
The clinical implication is that women on GLP-1RAs who are not planning pregnancy must use effective contraception, and those planning pregnancy must coordinate drug discontinuation with their reproductive timeline.
What happens when exposure occurs: the human data
Systematic review of semaglutide pregnancy exposure
Mandal et al. (2026) published a systematic review in the European Journal of Obstetrics and Gynecology examining the impact of semaglutide exposure on fetal and neonatal outcomes. They pooled data from available case reports, case series, and observational studies of women exposed to semaglutide during early pregnancy.[3]
The key finding: no significant increase in the rate of major congenital malformations was observed compared to background rates or obesity/diabetes-matched controls. Rates of spontaneous abortion and miscarriage were also comparable to expected rates in pregnancies complicated by obesity and diabetes. However, the total number of exposed pregnancies in the literature remains small (hundreds, not thousands), limiting the statistical power to detect rare events.
Unintentional periconceptional exposure cohort
Chou et al. (2026) reported on unintentional periconceptional exposure to GLP-1 receptor agonists in a cohort study published in Diabetes, Obesity and Metabolism. They examined pregnancy outcomes in women who were exposed to GLP-1RAs around the time of conception, typically because pregnancy was detected after several weeks of early gestation.[4]
The study found no statistically significant increase in adverse pregnancy outcomes including congenital anomalies, preterm birth, or neonatal complications compared to unexposed controls matched for BMI and diabetes status. The reassuring finding carries the same caveat: small sample size limits the ability to detect low-frequency events.
Maternal periconceptional exposure outcomes
Lee et al. (2026) published a study in the Journal of Obstetrics and Gynaecology Canada examining outcomes specifically in women with documented periconceptional GLP-1RA exposure. Their analysis found no evidence of increased teratogenic risk, with birth defect rates consistent with background population rates.[5]
Case reports and case series
Jin et al. (2026) published a case report and review of semaglutide exposure in early pregnancy, documenting a patient who became pregnant while taking semaglutide and was exposed through the first trimester. The infant was born without congenital anomalies. While case reports cannot establish safety, the accumulating series of normal outcomes following early exposure provides an evidence pattern consistent with the cohort data.[6]
Animal reproductive toxicology
The precautionary contraindication of GLP-1RAs in pregnancy is based primarily on animal data. Fotheringham et al. (2026) reviewed the pharmacological management of obesity in pregnancy and summarized the animal reproductive toxicology data for GLP-1RAs.[7]
Key findings from animal studies:
- Skeletal anomalies: observed in offspring of rats and rabbits exposed to semaglutide and liraglutide at doses several times the maximum recommended human dose (MRHD)
- Fetal growth restriction: reduced fetal weight observed at supratherapeutic doses in multiple species
- Embryo-fetal lethality: increased resorption and post-implantation loss at high doses
- No teratogenicity at clinical doses: at doses approximating human therapeutic exposure, no consistent pattern of birth defects was observed
The dose-response relationship is critical. Animal reproductive toxicity studies use multiples of the human dose to identify hazard signals. Effects observed at 2-10x the MRHD do not necessarily predict risk at clinical doses. GLP-1RAs have not been classified as direct teratogens, and the current contraindication is based on the precautionary principle rather than demonstrated human harm.
The "Ozempic babies" phenomenon
Lessard et al. (2026) documented prescribing trends for GLP-1 medications among pregnant and postpartum women in Obstetrics and Gynecology. Their analysis revealed increasing rates of GLP-1RA prescribing in reproductive-age women and correspondingly increasing rates of periconceptional exposure as the drug class has expanded from diabetes to obesity treatment.[8]
The "Ozempic babies" narrative reflects a real biological phenomenon: women with obesity-related anovulation who lose weight on GLP-1RAs can resume ovulation, sometimes unexpectedly. This is particularly common in women with PCOS, where weight loss of 5-10% can restore menstrual cycles. Women who had been told they were unlikely to conceive without fertility treatment may become pregnant naturally after several months on semaglutide or liraglutide.
The clinical gap is that many prescribers do not adequately counsel reproductive-age women about the fertility-restoring effects of GLP-1RA-mediated weight loss. The standard recommendation for contraception during treatment exists in prescribing information, but the specific risk of unexpected fertility restoration in previously subfertile women deserves more prominent clinical attention.
Preclinical insights: GLP-1RA and reproductive biology
Matiki et al. (2026) published research in the European Journal of Pharmacology showing that GLP-1RA treatment partially alleviated obesity-induced reproductive dysfunction in animal models. The mechanism involved modulation of the inflammatory microenvironment in reproductive tissues, suggesting GLP-1RAs may have direct effects on reproductive biology beyond the indirect benefits of weight loss.[9]
Deameh et al. (2026) examined the other side of the reproductive equation: GLP-1RA effects on male reproductive hormones, semen parameters, and sexual function. Published in the Journal of Sexual Medicine, their review found that GLP-1RAs may improve male fertility parameters through weight loss and metabolic improvement, though the evidence is more limited than for female reproduction.[10]
The evidence gap: what we still do not know
The current evidence on GLP-1RA pregnancy safety is built almost entirely on unintended exposures and retrospective analysis. No randomized controlled trial has intentionally exposed pregnant women to GLP-1RAs, and none is ethically feasible. The evidence we have comes from women who became pregnant while on treatment and were then followed.
This creates several structural limitations:
Exposure duration and timing. Most documented exposures occurred in the first trimester, with drug discontinuation upon pregnancy detection. The safety profile of second- or third-trimester exposure is essentially unknown because women stop the drug once pregnancy is discovered. The first trimester is the critical window for teratogenicity (weeks 3-8 of gestation), but organogenesis continues, and later exposure could theoretically affect fetal growth or metabolic programming.
Sample size. The total number of well-documented GLP-1RA pregnancy exposures in the literature is in the low hundreds. Rare adverse events (those occurring in 1 per 500-1000 pregnancies) would require thousands of exposed pregnancies to detect with statistical confidence. The absence of a signal in small cohorts is reassuring but not conclusive.
Confounders. Women exposed to GLP-1RAs are typically obese and/or diabetic, both of which independently increase risks of congenital anomalies, gestational diabetes, preeclampsia, macrosomia, and cesarean delivery. Separating a drug effect from the background risk associated with these conditions requires large, carefully matched cohort studies.
Long-term offspring outcomes. No study has examined childhood development, metabolic health, or other long-term outcomes in children exposed to GLP-1RAs in utero. Metabolic programming effects (alterations in offspring metabolism resulting from in utero exposure to metabolic drugs) are a theoretical concern that requires multi-year follow-up to assess.
Breastfeeding. Data on GLP-1RA excretion in breast milk and effects on nursing infants is almost nonexistent. Most guidelines recommend against GLP-1RA use during lactation, but the actual risk is unstudied.
Current clinical guidance
The consistent recommendation across guidelines and reviews:
- Stop GLP-1RAs at least 8 weeks before planned conception. This washout period accounts for the drug's half-life and allows complete elimination. Semaglutide has a 1-week half-life; 8 weeks provides approximately 8 half-lives for complete clearance.
- Use effective contraception during GLP-1RA treatment. This applies to all reproductive-age women, including those who believe they are subfertile. The fertility-restoring effects of weight loss can be rapid and unexpected.
- Do not panic over inadvertent exposure. Available human data shows no signal for increased teratogenicity. Women who discover pregnancy while on a GLP-1RA should stop the drug and proceed with standard prenatal care. The data, while limited, is reassuring.
- Plan preconception weight optimization. GLP-1RAs can be used to achieve target weight before conception, then discontinued with the understanding that some weight regain may occur during the washout period and pregnancy.
The contraception question: OCP interaction
A practical clinical question is whether GLP-1RAs interact with oral contraceptive pills (OCPs). GLP-1RAs slow gastric emptying, which could theoretically affect absorption of oral medications including OCPs. The clinical data on this interaction is limited, but prescribing information for semaglutide notes that delayed gastric emptying could reduce the rate of oral drug absorption. For women relying on OCPs as their primary contraception while taking GLP-1RAs, this potential interaction deserves attention. Long-acting reversible contraceptives (IUDs, implants) are not affected by gastric emptying and may be preferred in women on GLP-1RA therapy who require reliable contraception.
The broader reproductive planning challenge is that GLP-1RAs are most commonly prescribed to women in their 20s-40s, the exact demographic most likely to be planning or at risk for pregnancy. The intersection of metabolic treatment and reproductive planning requires coordinated care between endocrinologists, obesity medicine specialists, and obstetricians/gynecologists, a coordination that does not always happen in practice.
For how these drugs are used in other special populations, see GLP-1 agonists in older adults, GLP-1 drugs across racial and ethnic groups, and GLP-1 agonists after bariatric surgery.
The Bottom Line
GLP-1 receptor agonists are contraindicated during pregnancy based primarily on animal data showing skeletal anomalies at supratherapeutic doses, though they are not classified as direct teratogens. Human data from observational studies, case series, and a systematic review of semaglutide exposure show no signal for increased major congenital malformations, with outcomes comparable to obesity- and diabetes-matched controls. The drugs improve fertility by restoring ovulatory cycles in women with obesity and PCOS, creating a clinical paradox where treatment success increases the risk of inadvertent pregnancy. Current guidance recommends an 8-week washout before planned conception and effective contraception during treatment. The evidence base is reassuring but remains small, and larger prospective studies are needed to detect rare events.