Scientific deep-dive

Stopping a GLP-1 Before Pregnancy: Preeclampsia and Weight Gain Evidence

Does stopping a GLP-1 before pregnancy change preeclampsia risk? Pooled data show no clear answer, while extra weight gain in pregnancy appears more consistently. What the observational studies found and why they disagree.

By Eli Marsden · Founding Editor
Editorially reviewed & fact-checked against primary sources · How we verify contentLast reviewed
8 min read·20 citations

Whether stopping a GLP-1 drug before pregnancy changes preeclampsia risk is unresolved: one 2026 meta-analysis found no significant difference from three highly heterogeneous studies (odds ratio 1.05, 95% CI 0.60 to 1.84)[8], and individual cohorts point in both directions on hypertensive disorders, the broader category that includes preeclampsia. The signal for extra weight gain during pregnancy is more consistent, including a 3.3 kg difference in a matched JAMA cohort[1]. All of this evidence is observational.

About this article

This review is about pregnancy complications linked to the timing of use. Birth defect risk and the reasoning behind stop dates are covered in the GLP-1 washout guide. Blood sugar is covered in the gestational diabetes review.

What individual cohorts report

Cohorts of GLP-1 use before or early in pregnancy. Exposure windows differ widely, so the rows are not directly comparable.
StudyExposure window and sizePregnancy weight gainHigh blood pressure outcome
Maya, JAMA 2025Order from 3 years before to 90 days after conception; 448 exposed, 1,344 matched+3.3 kg (95% CI 2.3 to 4.2); excess gain 65% vs 49%Hypertensive disorders 46% vs 36%, RR 1.29 (1.12 to 1.49)
Yu, Obstet Gynecol 2026Semaglutide stopped before pregnancy, n=801; continued into pregnancy, n=429; nonusers, n=2,203Excess gain aOR 1.98 (1.56 to 2.51) stopped before; 2.88 (2.04 to 4.05) into pregnancyPregnancy-related hypertension was assessed; the abstract gives no estimate
Pondugula, Obstet Gynecol 2026Use up to 1 year before pregnancy; 243 exposed, 65.4% with evidence of use during pregnancyLower odds of gaining below recommendations in the weight-management group, aOR 0.37 (0.18 to 0.78)Hypertensive disorders aOR 0.83 (0.45 to 1.52) weight management; 0.72 (0.42 to 1.25) diabetes
Imbroane, Am J Obstet Gynecol 2025Prescription within 2 years before pregnancy; 4,267 per groupNot reported in abstractHypertensive disorders 19.9% with prescription vs 22.8% without, OR 0.84 (0.76 to 0.94)
Hviid, Hum Reprod Open 2026Danish registers; prescription within 8 weeks of last menstrual period; 529 exposed pregnanciesNot reported in abstractAfter matching, only preterm birth stayed elevated
Lemas, Obesity 2026Order within 12 or 24 months before delivery; 31 exposed, 29 matched; obesity and diabetesNot reported in abstractPreeclampsia more common (p = 0.023 at 12 months, p = 0.018 at 24 months); no effect size in abstract
Chukir, Int J Obes 2026Liraglutide; same women, pregnancies before and after exposure; 97 conceived after exposure; 53 with a pregnancy on each side8.0% before vs 9.2% after, p = 0.554Hypertensive disorders 9.2% before vs 0% after, p = 0.063

The JAMA cohort drew the most attention. Among 149,790 pregnancies at one academic health system, 448 with a GLP-1 order were matched to 1,344 without one. The exposed group gained a mean 13.7 kg versus 10.5 kg, and hypertensive disorders occurred in 46% versus 36% (RR 1.29, 95% CI 1.12 to 1.49)[1]. Mean prepregnancy BMI in the exposed group was 36.1, 84% had obesity and 23% had preexisting diabetes[1], so those high rates describe a high-risk population rather than typical pregnancies. For scale, a US task force estimates preeclampsia complicates about 4% of US pregnancies[18]. The JAMA abstract reports the combined category of hypertensive disorders, not preeclampsia alone[1].

A TriNetX study reached the opposite conclusion on blood pressure: hypertensive disorders were less common with a prescription in the two years before pregnancy (19.9% vs 22.8%, OR 0.84, 95% CI 0.76 to 0.94)[4]. Its abstract lists matching on age, race, ethnicity and comorbidities, without naming BMI[4]. A smaller cohort manually abstracted stop dates from the medical record and found no association with hypertensive disorders in either indication group[3]. It also found that 65.4% of people labeled as exposed in the year before pregnancy had evidence of use during pregnancy[3], a reminder of how seldom these studies isolate a clean stop before conception.

The pooled estimates disagree on preeclampsia

2026 meta-analyses. Almost entirely observational data (Hattler also included one randomized trial follow-up), and they include different studies.
Meta-analysisIncludedPreeclampsiaHypertensive disorders
Hattler, Obstet Gynecol8 studies, 186,598 pregnancies, 47,159 exposedOR 1.05 (0.60 to 1.84), from 3 studies, I² 85%OR 0.79 (0.34 to 1.83), from 3 studies, 9,015 pregnancies
D'Antonio, Med10 studies, 8,325 exposedOR 0.87 (0.78 to 0.98), from two datasetsNo clearly detectable increase
Pisani, Endocrine3 US retrospective cohorts, 10,880 pregnancies, 4,942 exposedNot reported separately in abstractOR 0.91 (0.57 to 1.47)
Hakim, Am J Obstet Gynecol22 studies, 10 in the meta-analysisNot reported separately in abstractNo significant association
Mavigök, Front Pharmacol14 data sources, searched through July 2026Not pooled; signals not independently replicatedNot reported in abstract; certainty very low for key outcomes

One analysis finds nothing, another finds a slightly lower odds of preeclampsia, and the most recent declined to pool it at all. The null estimate pools three studies with high heterogeneity[8]. The lower estimate rests on two datasets and its authors say to interpret it cautiously[9]. They add that their findings do not establish safety because of heterogeneity, residual confounding and limited datasets[9]. A narrative review in BJOG reaches a similar place: current data show no clear safety signal but cannot exclude clinically important risks[17]. The small study reporting more preeclampsia had only 29 matched exposed pregnancies[5]. Analyses limited to the broader category found no significant association with hypertensive disorders, including a pooled odds ratio of 0.91 (95% CI 0.57 to 1.47) from three US retrospective cohorts[10] and a 10-study meta-analysis[11].

Why weight regain is the leading suspect

The rebound idea starts with a pattern. In a national US records study, women who stopped semaglutide before pregnancy and those exposed into early pregnancy both had higher odds of excess weight gain than nonusers, and they did not differ from each other[2]. The authors read that as an effect of stopping rather than of exposure in the womb[2]. One author discloses advisory board roles with Eli Lilly, Novo Nordisk and Boehringer Ingelheim[2]. The same pattern would also fit a shared difference between users and nonusers, since both user groups are compared with the same nonusers.

Weight is a plausible route because it tracks these complications outside any drug study. In individual data from 265,270 births, higher prepregnancy BMI and higher gestational weight gain were each associated with more hypertensive disorders, and mothers with obesity and high gain had the highest odds of any pregnancy complication (OR 2.51, 95% CI 2.31 to 2.74, compared with normal-weight mothers with medium gain)[13]. A meta-analysis of 415,605 women linked weight gain between pregnancies with hypertensive disorders (OR 1.37, 95% CI 1.21 to 1.53) and weight loss with lower risk (OR 0.87, 0.75 to 0.99)[14]. Neither analysis tested GLP-1 drugs, and neither shows that regain after a GLP-1 behaves the same way.

The within-woman data complicate the picture. Among 53 women with a pregnancy both before and after liraglutide, weight gained during pregnancy was similar across the two pregnancies (gestational weight change 8.0% before vs 9.2% after, p = 0.554), and women who conceived after liraglutide had regained weight overall[7]. The study was small, the later pregnancies came about four years on with more metformin use (35.8% vs 11.3%), and all women who conceived had inconsistent obesity medication use[7].

Does the gap between stopping and conception matter?

The closest thing to a timing analysis concerns gestational diabetes, not preeclampsia. In a TriNetX cohort of women without preexisting diabetes, matched on BMI, those whose last prescription was more than 90 days before the pregnancy had a similar risk to nonusers (16.3% vs 16.8%, RR 0.967, 95% CI 0.719 to 1.315), while a prescription within 90 days was associated with higher risk (RR 1.536, 95% CI 1.020 to 1.955) and steeper BMI rebound[15]. The authors call for confirmation in prospective studies[15]. We did not find a published equivalent for preeclampsia, so extending this result to blood pressure would be a guess.

Reason for use may also matter. In the Danish registers, preterm birth was elevated only when the drug was used for diabetes (semaglutide aOR 1.84, 95% CI 1.24 to 2.71) and not for weight management (semaglutide aOR 0.71, 0.30 to 1.70), which the authors say is compatible with diabetes rather than the drug being responsible[6]. Several of those authors report Novo Nordisk Foundation funding or Novo Nordisk ties[6]. In women with polycystic ovary syndrome, a large TriNetX analysis tied preconception GLP-1 or GIP/GLP-1 therapy to a lower composite of pregnancy complications than metformin (HR 0.38, 95% CI 0.34 to 0.41), though associations varied across sensitivity analyses and the authors say the findings do not support use during gestation[16].

Why these studies disagree

  • Exposure means different things. It ranges from an order up to 3 years before conception[1] to a prescription within 8 weeks of the last menstrual period[6]. Those are different biological questions.
  • Prescriptions are not doses. The Danish authors note they had no data on whether redeemed prescriptions were taken[6]. Few studies can state when the last dose occurred.
  • Prepregnancy BMI can mislead. A woman recently on a GLP-1 may enter pregnancy lighter than her long-term weight, so matching on that BMI may pair her with someone metabolically different. In one sensitivity analysis matched on the pre-treatment window, treated women still carried a higher mean BMI (37.0 vs 35.0)[15].
  • Outcome labels vary. Most reports use the broad category of hypertensive disorders, which is why preeclampsia itself has so few pooled datasets[9][12].

What the labels say

Neither label below addresses preeclampsia after stopping. Wegovy, revised June 2026, says to “discontinue WEGOVY in patients at least 2 months before they plan to become pregnant” for fetal safety reasons and because of semaglutide’s long half-life[19]. Zepbound, revised August 2026, gives no planned-pregnancy window and says “to discontinue ZEPBOUND when a pregnancy is recognized”[20]. Those stop dates exist to limit fetal exposure, and the cohorts above cannot say whether they also change blood pressure risk.

What to raise with your clinicians

If you are planning a pregnancy on a GLP-1 drug, ask your prescriber about a stop date and contraception until then, and how to limit regain after stopping. Tell your obstetric team when you last used the drug, what you weighed before treatment and why you were prescribed it. Ask whether your risk profile makes you a candidate for low-dose aspirin: the US Preventive Services Task Force recommends 81 mg a day after 12 weeks of gestation for people at high risk of preeclampsia[18]. If you become pregnant while taking one of these drugs, contact your clinician promptly.

Frequently Asked Questions

The evidence does not settle this. One 2026 meta-analysis found no significant difference (odds ratio 1.05), another found slightly lower odds based on two datasets (odds ratio 0.87), and a third declined to pool preeclampsia at all. One small study of 29 matched pregnancies reported more preeclampsia. All of these are observational, and none establishes safety.
Some cohorts suggest more gain. A JAMA study found 3.3 kg more than matched pregnancies, and a national US study found higher odds of excess gain in women who stopped semaglutide before pregnancy. A small study of liraglutide users comparing their own pregnancies found no difference. Ask your obstetric team what weight gain range fits you.
That depends on the drug. The Wegovy label says to stop at least 2 months before a planned pregnancy. The Zepbound label gives no planned-pregnancy window and says to stop when a pregnancy is recognized. No study has shown which gap best protects against preeclampsia, so plan the date with your prescriber and obstetric clinician.
It has not been shown. Weight loss between pregnancies is associated with fewer hypertensive disorders in general studies, and one records study linked prior GLP-1 prescriptions to fewer such disorders, but another cohort found more. The drug must also be stopped before or at the start of pregnancy, so any regain has to be considered.

References

  1. 1.Maya J, Pant D, Fu Y, James K, Batlle C, et al. Gestational Weight Gain and Pregnancy Outcomes After GLP-1 Receptor Agonist Discontinuation. JAMA. 2025. PMID: 41284263.
  2. 2.Yu Y, Li X, Groth SW, Zaman A, Chao AM, et al. Gestational Weight Gain and Pregnancy Outcomes After Semaglutide Exposure. Obstet Gynecol. 2026. PMID: 42208070.
  3. 3.Pondugula N, Culhane JF, Lundsberg LS, Partridge C, Merriam AA, et al. Gestational Weight Gain and Hypertensive Disorders of Pregnancy With Prepregnancy and Early Pregnancy Glucagon-Like Peptide-1 Receptor Agonist Exposure. Obstet Gynecol. 2026. PMID: 40609095.
  4. 4.Imbroane MR, LeMoine F, Nau CT. Preconception glucagon-like peptide-1 receptor agonist use associated with decreased risk of adverse obstetrical outcomes. Am J Obstet Gynecol. 2025. PMID: 39848396.
  5. 5.Lemas DJ, Zapata R, Loy I, Mahadevan A, Wen T, et al. Prepregnancy GLP-1RA Exposure and Risk of Hypertensive Disorders of Pregnancy Among Women With Obesity and Diabetes. Obesity (Silver Spring). 2026. PMID: 42120704.
  6. 6.Hviid KVR, Banasik K, Mortensen LH, Madsbad S, Strandberg-Larsen K, et al. Periconceptional GLP-1 receptor agonist exposure and obstetric outcomes: a Danish nationwide cohort study. Hum Reprod Open. 2026. PMID: 41852577.
  7. 7.Chukir T, Haider H, Sarkar S, Nair S, Kokash R, et al. A single centre retrospective study of the long-term weight and pregnancy-related outcomes in women of reproductive age before and after liraglutide exposure. Int J Obes (Lond). 2026. PMID: 42174221.
  8. 8.Hattler E, Schluter H, Greene C, Adgent MA, Sundermann AC, et al. Glucagon-Like Peptide-1 Receptor Agonists and Risk of Adverse Maternal Pregnancy Outcomes: A Systematic Review and Meta-analysis. Obstet Gynecol. 2026. PMID: 42391628.
  9. 9.D'Antonio F, Flacco ME, Manzoli L, Samara A, Khalil A, et al. GLP-1 receptor agonist exposure in the periconceptional period and adverse obstetric outcomes: A systematic review and meta-analysis. Med. 2026. PMID: 42810337.
  10. 10.Pisani D, Lorenza D, Fordellone M, Cerillo A, Fatigati M, et al. Hypertensive disorders of pregnancy and GLP-1 receptor agonist timing: a systematic review and meta-analysis. Endocrine. 2026. PMID: 42536323.
  11. 11.Hakim J, Rajesh D, Tello JA. Neonatal and obstetric outcomes following periconceptional exposure to glucagon-like peptide-1 receptor agonists: a systematic review and meta-analysis. Am J Obstet Gynecol. 2026. PMID: 42061782.
  12. 12.Mavigök E, Mavigök AN. Pregnancy and perinatal outcomes following exposure to GLP-1 and dual GLP-1/GIP receptor agonists: a time-window-stratified systematic review and meta-analysis. Front Pharmacol. 2026. PMID: 42964193.
  13. 13.Santos S, Voerman E, Amiano P, Barros H, Beilin LJ, et al. Impact of maternal body mass index and gestational weight gain on pregnancy complications: an individual participant data meta-analysis of European, North American and Australian cohorts. BJOG. 2019. PMID: 30786138.
  14. 14.Martínez-Hortelano JA, Cavero-Redondo I, Álvarez-Bueno C, Sanabria-Martínez G, Poyatos-León R, et al. Interpregnancy Weight Change and Hypertension During Pregnancy: A Systematic Review and Meta-analysis. Obstet Gynecol. 2020. PMID: 31809428.
  15. 15.Banerjee M, Dutta S, Dasgupta S. Pre-Pregnancy GLP-1 Receptor Agonist or Tirzepatide Use and Gestational Diabetes Risk: Evaluating Pharmacodynamic Carry-Over Versus Post-Discontinuation Metabolic Rebound in a Multinational Federated Cohort. Diabetes Obes Metab. 2026. PMID: 42098901.
  16. 16.Henney AE, Holmes PJ, Riley D, Heague M, Hapangama DK, et al. Preconception incretin therapy reduces pregnancy and cardiometabolic outcomes in polycystic ovary syndrome. J Clin Endocrinol Metab. 2026. PMID: 42857939.
  17. 17.Lau KGY, Chopra A, Mullins E, Agha-Jaffar R, Tan T, et al. Glucagon-Like Peptide-1 Receptor Agonists in Pregnancy: Periconception Exposure and Perinatal Outcomes. BJOG. 2026. PMID: 42830218.
  18. 18.US Preventive Services Task Force, Davidson KW, Barry MJ, Mangione CM, Cabana M, et al. Aspirin Use to Prevent Preeclampsia and Related Morbidity and Mortality: US Preventive Services Task Force Recommendation Statement. JAMA. 2021. PMID: 34581729.
  19. 19.Novo Nordisk. WEGOVY (semaglutide) prescribing information, revised June 2026. DailyMed, US National Library of Medicine. 2026. https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=ee06186f-2aa3-4990-a760-757579d8f77b
  20. 20.Eli Lilly and Company. ZEPBOUND (tirzepatide) prescribing information, revised August 2026. DailyMed, US National Library of Medicine. 2026. https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=487cd7e7-434c-4925-99fa-aa80b1cc776b

GLP-1 Exposure Around Conception: Birth Defect Meta-Analyses

Pooled observational data show no clear rise in major birth defects after GLP-1 exposure around conception, but certainty is very low. See what the 2026 meta-analyses found, and why one large cohort shapes the results.

9 min read

GLP-1 and IVF Egg Retrieval: Hold + Washout (2026)

GLP-1 and IVF involve two holds: a short anesthesia hold for egg retrieval and a longer pre-conception washout — semaglutide ≥2 months before pregnancy.

12 min read

GLP-1 Around Labor & C-Section: Washout + Safety (2026)

GLP-1 labels say stop semaglutide ≥2 months before pregnancy, so you should be off it by labor. If exposed late, here's how it fits C-section anesthesia.

12 min read

How Long Does Ozempic Stay in Your System? Semaglutide Half-Life and Clearance Timeline (2026)

Ozempic (semaglutide) has a roughly one-week half-life, so it takes about five weeks to clear after your last dose. See the week-by-week washout, why it matters for pregnancy and surgery, and how it compares to tirzepatide and Rybelsus.

8 min read

Ozempic and Pregnancy: Is Semaglutide Safe? The 2-Month Washout Rule (2026)

Semaglutide (Ozempic, Wegovy) is not recommended in pregnancy: stop it when pregnancy is recognized, and at least 2 months before a planned pregnancy because of its roughly one-week half-life. Also covers accidental conception, breastfeeding, and birth control.

9 min read

Gestational Diabetes and BMI: Causes, Weight Before Pregnancy and GLP-1 Timing

What causes gestational diabetes, how pre-pregnancy BMI and weight change between pregnancies affect risk, why treatment matters, the later type 2 diabetes risk and how to lower it, and what is known about stopping GLP-1 drugs before pregnancy.

12 min read