GLP-1 medicines in the first trimester: a study of 168 pregnancies

Why this study matters

When someone becomes pregnant while using a GLP-1 medicine, they and their care team want to know what is known about risks to the baby. Teratology information services are specialist centres that counsel people about medicine use in pregnancy and record what happens. This study pooled records from six such services across several countries. It is one of the few studies that followed pregnancies forward in time rather than looking back through insurance records, and it included people using the medicines for weight management as well as for diabetes.

Who was in the study

The study included 168 pregnancies in which the person was exposed to a GLP-1 receptor agonist during the first trimester, either for diabetes or for obesity. They were compared with two reference groups: 156 pregnancies in people with diabetes who used at least one non-GLP-1 diabetes medicine in the first trimester, and 163 pregnancies in people with overweight or obesity who did not have diabetes. Pregnancies were recorded between 2009 and 2022. The summary we reviewed does not state which countries took part or the ages of participants.

What the researchers did

This was a multicentre, observational, prospective cohort study. Observational means nobody was assigned a treatment; the researchers recorded what people were already taking. Prospective means people were enrolled during pregnancy and followed to the outcome, rather than being identified after the fact. Cohort means groups were followed and compared.

The main outcomes were major birth defects, live births, pregnancy losses and pregnancy terminations. The researchers used statistical adjustment to account for differences between the groups. According to the extracted data, missing maternal age was filled in with the median value, and missing information on smoking or medical history was treated as absent.

What they found

Major birth defects occurred in 2.6% of GLP-1-exposed pregnancies (R32-C124). This compared with 2.3% in the diabetes group and 3.9% in the overweight or obesity group.

An odds ratio of 1.0 would mean no difference. A confidence interval is the range in which the true value most likely sits. Both intervals here are very wide, running from well below 1.0 to well above it, which means the study cannot say whether risk was higher, lower or the same.

The study also reported how pregnancies ended. In the GLP-1 group, 59% ended in a live birth, 23% in pregnancy loss, and 18% in termination. In the diabetes group the figures were 69%, 26% and 6%. In the overweight or obesity group they were 63%, 29% and 8%. Statistical models found no increased risk of pregnancy loss with GLP-1 exposure compared with either reference group, in both crude and adjusted analyses.

The authors say the study offers reassurance in cases of unintended exposure during the first trimester, but that larger studies are needed because of the small sample.

Side effects and people who stopped

This study measured pregnancy outcomes, not side effects in the pregnant person. The extracted data do not report discontinuation of the medicine, and the summary we reviewed does not describe when in pregnancy the medicine was stopped.

What this study does not tell you

• With only 168 exposed pregnancies, the study is small. The confidence intervals are so wide that a real increase in birth defects cannot be ruled out.
• People who contact a teratology information service may differ from the general population, so the sample may not represent everyone who uses these medicines.
• The summary we reviewed does not say which GLP-1 medicines were used or at what doses.
• The higher rate of terminations in the GLP-1 group (18%) is reported but not explained in the abstract, and the reasons are unknown from the summary we reviewed.
• This is not proof that GLP-1 medicines are safe in pregnancy. It is one small piece of observational evidence.
• The summary we reviewed does not state the funding source.

What it means in Canada

This study does not include Canadian data and does not tell anyone what to do about their medicine before or during pregnancy. For information about specific medicines, see semaglutide in Canada and tirzepatide in Canada. For what provincial plans pay for, see public drug coverage by province.

Related reading

Other studies in this library that bear on the same question.

Source