Weight-loss drugs linked to preterm birth risk in women with pre-existing diabetes

Weight loss drugs have been linked to an increased risk of premature births among women who took them inadvertently just before or during early pregnancy to treat pre-existing diabetes.

However, a large study of over 750,000 pregnancies found that there was no link to preterm births or other obstetric complications if the medication was being used to lose weight.

The authors of the study, published today (Wednesday) in Human Reproduction Open, one of the world's leading reproductive medicine journals, say their findings suggest that it is diabetes rather than the drugs that may be contributing to the increased risk of preterm births. 

Weight loss drugs such as semaglutide (brand names: Ozempic, Wegovy) and liraglutide (Saxenda) belong to a class of medications called GLP-1 receptor agonists (GLP-1 RAs), which were developed originally to treat type 2 diabetes. They work, primarily, by suppressing appetite and were quickly found to be useful in helping people to lose weight as well. However, there is no evidence to suggest they are safe to take during pregnancy.

Professor Henriette Svarre Nielsen, from the Department of Gynaecology and Obstetrics at Copenhagen University Hospital Hvidovre, Hvidovre, Denmark, who led the study, said: "GLP-1 RA treatment has, within the last few years, become the medication with the sharpest increase in prescription worldwide. Current guidance suggests the treatment should be stopped eight weeks prior to planning a pregnancy. However, this is based on early-phase model organism studies, and not real-world evidence. Nonetheless, because of its widespread adoption, inadvertent exposure in early pregnancy is inevitable, and there is hardly any evidence to guide clinical counselling when it happens."

The researchers decided to investigate the association between women inadvertently taking GLP-1 RA treatment during the periconceptional period and the risk of complications such as preterm birth (birth before 37 weeks), pre-eclampsia, gestational diabetes, giving birth to a child large for gestational age, still birth, and problems with the placenta.

They analyzed data from the Danish nationwide health registries for 756,636 singleton pregnancies among 480,231 women that resulted in deliveries between 1 October 2009 and 31 December 2023. If women had redeemed a prescription for liraglutide or semaglutide within eight weeks before or after the date of their last menstruation, this was defined as inadvertent exposure to the medications. This 16-week window captured the period when the mother might not have known she was pregnant and the early stages of organ growth in the embryo. A total of 529 pregnancies were exposed to GLP-1 RAs during the periconceptional period.

The researchers adjusted their results to take account of the mothers' age, body mass index (BMI), smoking status, geographic region, education, pre-existing diabetes, and the month and year of pregnancy in order to control for seasonal or other trends.

The first author of the paper, Dr Kathrine Hviid, a PhD student in the same department, said: "We made some extremely important findings that have implications for future studies, such as randomised controlled trials of GLP-1 RA usage in pregnancy, and for clinical counselling.

"We found that these medications were associated with an increased risk of preterm birth, but the risk was only present when the medication was used for diabetes treatment, and not for weight management. This suggests that the underlying condition of diabetes, rather than the medication, may be driving this association."

The researcher found that women who took GLP-1 RAs had higher rates of several obstetric complications, but after adjustment for the factors that could affect the results, they found only a greater risk of preterm birth in women who had taken either liraglutide or semaglutide for diabetes treatment. Compared to women who did not take GLP-1 RAs, the increased risk was 70% higher for liraglutide and 84% higher for semaglutide. This means that among women with pre-existing taking semaglutide, the drug was associated with a higher risk of preterm birth of approximately 11%. Liraglutide showed a 9% increased risk.

Future studies must take account of the reason a woman has been prescribed these medications, as the risks differ between women using GLP-1 RAs for diabetes versus weight management. This is one of the first studies to examine GLP-1 RA exposure in early pregnancy according to the reason for prescription. As more evidence accumulates, these findings will guide clinical counselling for women inadvertently exposed to GLP-1 RAs in early pregnancy. In Denmark, about 70% of people who use weight loss medications are women and so it is inevitable that some may take it without realising they are pregnant."

Professor Henriette Svarre Nielsen, Department of Gynaecology and Obstetrics, Copenhagen University Hospital Hvidovre

However, she said further studies were needed and it was too soon to change the recommendation to stop GLP-1 RAs prior to pregnancy, or to implement counselling, regardless of the reason why women were taking them.

The strength of the study is its large size. Limitations include the fact that the results cannot show that GLP-1 RAs cause preterm birth, only that they are associated with it. In addition, there were no data showing the women had actually taken the medications after they redeemed their prescriptions. However, in Denmark women need to pay for an injection of GLP-1 RAs, although there is a state subsidy if it is for treating diabetes. For instance, an injection of 1mg of Wegovy or Ozempic (semaglutide) costs around €180 and €114 respectively, including the subsidy for diabetes treatment.

Dr Hviid said: "Because of the high cost, we assume that compliance is very high among the women prescribed GLP-1 RAs."

In an invited commentary accompanying the paper, Drs Yeyi Zhu and Monique Hedderson, from Kaiser Permanente Northern California and the Center for Upstream Prevention of Adiposity and Diabetes Mellitus, in Pleasanton, California, USA, write that the findings of the study "complement and extend an emerging evidence base on the reproductive safety of GLP-1 RAs".

They continue: "For clinicians, the study by Hviid et al (2026) may inform preconception and early pregnancy counselling. The observation that elevated preterm birth risk was confined to women treated for diabetes, and not to those using GLP-1 receptor agonists for weight management, supports more balanced, individualized discussions with patients who may have experienced inadvertent periconceptional exposure. For patients with diabetes, the study reinforces the importance of recognizing diabetes as an important risk factor for obstetric complications and prioritizing metabolic health and glycemic management before and during pregnancy."