New Nature Portfolio Study Supports Reassuring GLP-1 Safety Data Before Pregnancy

What the latest human data mean for fertility, IVF, endometriosis, and the controversial “washout” period

Commentary by Andrea Vidali, MD · internationalendo.com

GLP-1 receptor agonists like semaglutide (Ozempic®, Wegovy®) and liraglutide (Saxenda®) have transformed care for weight management and metabolic health. But for women planning a pregnancy—especially those navigating infertility, IVF, or endometriosis—one big question remains: Are GLP-1s safe?

A 2026 systematic review and meta-analysis in Scientific Reports analyzed more than 40,000 pregnancies exposed to GLP-1 receptor agonists. The results are cautiously reassuring—and they should change how fertility patients are counseled about accidental exposure and prolonged preconception washouts.

The Direct Answer

Current human evidence does not show a statistically significant increase in overall congenital malformations, major first-trimester birth defects, miscarriage, small-for-gestational-age birth, or preterm birth after maternal GLP-1 exposure. The data do not prove zero risk or support routine weight-loss treatment during pregnancy. They do support calm counseling after inadvertent exposure and an individualized—not automatic—approach to preconception washout.

Why This Question Matters Now

GLP-1-based medications—including semaglutide (Ozempic and Wegovy), liraglutide (Saxenda), dulaglutide (Trulicity), and the dual GIP/GLP-1 agonist tirzepatide (Mounjaro and Zepbound)—are increasingly used by women of reproductive age for obesity, insulin resistance, and type 2 diabetes.

The fertility issue is not theoretical. Weight loss and improved insulin sensitivity can restore ovulation in some patients, making conception more likely. At the same time, many women pursuing IVF are told to stop treatment one or two months before egg retrieval or embryo transfer, even when age, diminished ovarian reserve, endometriosis, or repeated treatment failure makes every month important.

The new publication is not an article in Nature magazine itself; it appeared in Scientific Reports, a Nature Portfolio journal. That distinction matters for accuracy, but the study remains important because it is the largest synthesis of human pregnancy outcome data in this area to date.

What the 2026 Meta-Analysis Studied

Investigators searched major medical databases through January 2026 and included seven cohort studies. Depending on the analysis, exposure was defined from 90 days before conception through pregnancy, or specifically during the first trimester. The studies included semaglutide, liraglutide, exenatide, dulaglutide, lixisenatide, albiglutide, and limited tirzepatide exposure.

The largest pooled analysis included 42,282 GLP-1-exposed pregnancies and 723,892 comparator pregnancies. The authors preferentially used adjusted estimates when available, but some large datasets supplied only crude estimates. This is why the conclusions are reassuring but not definitive: all included studies were observational, comparator groups differed, and the certainty of evidence was rated low or very low for most outcomes.

The Main Findings: No Clear Human Teratogenic Signal

The study did not identify a statistically significant increase in overall congenital malformations or major congenital malformations after first-trimester exposure. It also found no significant increase in cardiac defects, spontaneous abortion, small-for-gestational-age birth, or preterm birth. Case reports and other qualitative evidence did not reveal a consistent or recurrent pattern of fetal malformations.

Stillbirth did not reach statistical significance, but the estimate was imprecise (OR 2.17; 95% CI 0.95–4.91). The wide interval means a clinically important increase cannot yet be excluded, so this outcome requires larger, better-adjusted studies.

Pooled Pregnancy Outcomes Reported in the 2026 Meta-Analysis

Note: Odds Ratio (OR) values crossing 1.0 are not statistically significant. The urinary finding was the only nominally significant signal and was based strictly on unadjusted raw data.

The Urinary-Malformation Finding Needs Context

The analysis found a modest association with urinary malformations (OR 1.24; 95% CI 1.05–1.47), but this result came from only two studies and relied on crude, unadjusted estimates. The underlying studies did not adequately account for maternal diabetes, obesity, baseline glycemic control, or other confounding lifestyle factors that can independently affect congenital-malformation risk. They also did not identify a consistent urinary-tract phenotype.

The appropriate clinical interpretation is not “GLP-1 drugs cause urinary defects.” It is that this is a hypothesis-generating signal requiring better-adjusted future studies. It should be disclosed to patients, but it should not overshadow the absolute absence of a broader malformation pattern across more than 40,000 exposed pregnancies.

What This Means After Accidental Exposure

A patient who becomes pregnant while taking a GLP-1 medication should immediately contact her prescribing clinician and fertility or obstetric team, but she should not be counseled as though a major, catastrophic teratogenic exposure has occurred. The meta-analysis specifically concluded that inadvertent first-trimester exposure should not be considered an indication for pregnancy termination.

This does not mean the medication should be continued automatically. Weight-loss treatment is generally discontinued after pregnancy is recognized, and patients using a GLP-1 agent for type 2 diabetes need a pregnancy-compatible plan that protects glycemic control. Poorly controlled diabetes is itself a well-established maternal and fetal risk; simply stopping medication without an immediate replacement can be harmful.

What the Study Means for Fertility and IVF Strategy

The most useful clinical shift is to separate three distinct questions that are often incorrectly treated as one monolithic rule: reproductive toxicology, anesthesia safety, and the timing cost of delaying fertility treatment.

• Before Egg Retrieval: There is zero human evidence that a prolonged reproductive “washout” improves oocyte quality, embryo competence, blastocyst conversion, aneuploidy rates, or live-birth outcomes. Those specific endpoints remain understudied, so strong claims of benefit or harm are not justified.

• Anesthesia for Retrieval: GLP-1 medications can delay gastric emptying. This is a procedural aspiration-risk question for the anesthesiologist, not a fetal-exposure question. Current multi-society anesthesia guidance states that most stable patients can continue GLP-1 therapy before elective procedures, while patients in active dose escalation or with significant gastrointestinal symptoms may need a liquid diet, modified anesthesia plan, gastric ultrasound assessment, or procedural delay.

• Before Embryo Transfer or Natural Conception: The new human data are highly reassuring for periconceptional and early first-trimester exposure. They support a patient-specific stopping plan rather than an automatic multi-month delay for every single patient.

Why Endometriosis Patients Are Different

GLP-1 medications are not established treatments for endometriosis. They do not remove organic endometriosis lesions, they do not replace specialized laparoscopic excision surgery (LAPEX), and they do not have proven direct effects on implantation failure or endometriosis-related pain networks. Interesting anti-inflammatory and metabolic mechanisms discovered in pre-clinical settings should never be marketed to patients as clinical proof.

Their relevance is indirect but highly important: many endometriosis patients also face age-related fertility decline, diminished ovarian reserve, prior ovarian surgery, or a narrow reproductive treatment window.

For these patients, a blanket, non-individualized “wait another two months” instruction offers no demonstrated reproductive benefit while allowing rapid weight regain, metabolic rebound, or worsening metabolic control. For a patient in her late thirties or early forties, time is itself a critical fertility variable—and an un-nuanced delay is never neutral.

Is a Two-Month Washout Really Necessary?

A pharmacokinetic washout describes how long it takes a drug to physically clear the body. It does not automatically establish a clinically meaningful threshold for birth-defect or miscarriage risk. The 2026 meta-analysis did not compare specific washout intervals, so it cannot prove that no washout is needed. However, no human outcome study has ever demonstrated that waiting for complete drug elimination before conception improves congenital-malformation, miscarriage, or live-birth outcomes.

Current product labels are not uniform, reflecting a lack of a universal class-wide washout rule:

• The February 2026 FDA prescribing information for Wegovy advises stopping semaglutide at least two months before a planned pregnancy because of its long half-life.

• The 2026 Zepbound label advises discontinuing tirzepatide when pregnancy is recognized and does not state an equivalent mandatory two-month preconception interval.

ESSI’s clinical interpretation is that a prolonged washout should not be imposed automatically on every IVF or endometriosis patient. The decision must carefully balance the specific drug, why it is being used, maternal age, ovarian reserve, embryo status, timing of transfer, gastrointestinal symptoms, anesthesia plan, diabetes control, and the real biological consequences of delaying treatment.

A Practical ESSI Framework for GLP-1 Use in Fertility

1. Do Not Panic After Inadvertent Exposure: Current human data are cautiously reassuring, and early exposure alone is never an medical indication for pregnancy termination.

2. Do Not Use GLP-1s as an Endometriosis Treatment: Use them only for valid metabolic indications with a clearly defined fertility plan.

3. Avoid Unnecessary Delay: In selected IVF patients, stopping close to conception or embryo transfer may be reasonable after individualized counseling, even though semaglutide labeling remains more conservative.

4. Do Not Routinely Continue Weight-Loss Therapy Through Pregnancy: Once pregnancy is recognized, the medication is generally stopped and nutrition, glycemic control, and gestational weight gain are managed appropriately.

5. Protect Diabetes Control: Patients using a GLP-1 agent for type 2 diabetes require an alternative, pregnancy-compatible treatment plan to avoid the severe risks of uncontrolled maternal hyperglycemia.

6. Coordinate with Anesthesia Before Egg Retrieval: The relevant issue is delayed gastric emptying and aspiration risk—not fetal reproductive washout.

Commentary from Dr. Andrea Vidali

“The historical two-month rule was built primarily from pharmacokinetics, animal data, and the absence of human evidence. It was not built from a clinical trial showing that a woman who waits eight weeks has a healthier pregnancy than a woman who stops near embryo transfer or when pregnancy is recognized.”

“This meta-analysis does not prove zero risk. What it gives us is something we did not have before: data from more than 40,000 exposed pregnancies without a major human teratogenic pattern or a demonstrated increase in miscarriage.”

“In fertility medicine, delay has a definitive biological cost. A 40-year-old patient with endometriosis or diminished ovarian reserve cannot treat two months as biologically neutral. We must compare a theoretical exposure risk with the real, measurable risks of ovarian aging, weight regain, worsening insulin resistance, and losing a finite treatment opportunity.”

“My approach is not to prescribe GLP-1 medications throughout pregnancy, and it is not to ignore product labeling. It is to reject reflexive, one-size-fits-all washouts. The plan should be based on the medication, the indication, ovarian reserve, embryo status, transfer timing, gastrointestinal symptoms, anesthesia safety, and glycemic control. Most importantly, accidental exposure around conception should be met with calm, evidence-based counseling and appropriate follow-up—not panic.”

— Andrea Vidali, MD

Reproductive Surgeon, ESSI

Frequently Asked Questions

Are GLP-1 medications safe before pregnancy?

The best available human data are reassuring for exposure around conception and during the first trimester, with no statistically significant increase in overall or major congenital malformations. The evidence remains observational and low-certainty, so “reassuring” is a more accurate scientific description than “proven completely safe.”

Do I need to stop Ozempic or Wegovy two months before IVF?

The Wegovy label recommends stopping semaglutide two months before a planned pregnancy. However, no human study has shown that a fixed two-month washout improves IVF or pregnancy outcomes. For time-sensitive patients, the timing should be individualized with your prescribing clinician and reproductive endocrinologist.

What if I become pregnant while taking semaglutide or tirzepatide?

Contact your prescribing clinician and obstetric or fertility team immediately. Weight-loss treatment is generally discontinued when pregnancy is recognized. Current evidence does not support treating inadvertent early exposure as an indication for pregnancy termination.

Do GLP-1 medications increase miscarriage risk?

In the 2026 meta-analysis, maternal GLP-1 exposure was not associated with a statistically significant increase in spontaneous abortion (OR 0.95; 95% CI 0.42–2.16).

Can GLP-1 medications treat endometriosis?

No clinical evidence currently shows that GLP-1 medications eradicate endometriosis, replace excision surgery, or reliably treat endometriosis pain or implantation failure. Their role is metabolic, not disease-eradicating.

Can I take a GLP-1 medication during an egg-retrieval cycle?

The pregnancy-safety question and the anesthesia question are completely different. GLP-1 therapy may be continued for many patients, but delayed gastric emptying can affect sedation safety. The fertility clinic and anesthesia team should thoroughly assess dose escalation, gastrointestinal symptoms, gastroparesis risk, and procedural timing.

The Bottom Line

We can no longer say there are no meaningful human data. The 2026 meta-analysis provides cautiously reassuring evidence from more than 40,000 exposed pregnancies. It does not prove that GLP-1 medications are risk-free or justify routine use during pregnancy. It does, however, support less fear-based counseling after accidental exposure and a more rational approach to preconception planning.

For fertility and endometriosis patients, a prolonged washout should not be treated as an automatic rule. It should be a shared clinical decision that weighs uncertain theoretical risk against the measurable costs of delay, metabolic rebound, and lost reproductive time.

References

1. Uysal N, Horoz E, Gungor M, et al. Pregnancy outcomes following maternal GLP-1 receptor agonist exposure: a systematic review and meta-analysis. Scientific Reports. 2026. doi:10.1038/s41598-026-61582-8.

2. U.S. Food and Drug Administration. Wegovy (semaglutide) Prescribing Information. Revised February 2026.

3. U.S. Food and Drug Administration. Zepbound (tirzepatide) Prescribing Information. 2026.

4. American Society of Anesthesiologists and partner societies. Multi-society GLP-1 perioperative guidance. October 2024.

MEDICAL DISCLAIMER: This article is educational and does not replace individualized medical advice. GLP-1 medications are not approved for routine weight-loss treatment during pregnancy. Medication decisions should be coordinated among the prescribing clinician, reproductive endocrinologist, anesthesiologist, and obstetric team when appropriate.