Although extensive evidence provides cautious reassurance after inadvertent GLP-1 exposure during pregnancy, routine use is still not considered safe due to unresolved risks and less robust data.
Research: Pregnancy outcomes after maternal GLP-1 receptor agonist exposure: a systematic review and meta-analysis. Image credit: While At My Desk / Shutterstock
In a recent study published in the journal scientific reporta group of researchers evaluated whether maternal exposure to glucagon-like peptide-1 (GLP-1) receptor agonists during the preconception period or during pregnancy is associated with congenital malformations or other adverse pregnancy outcomes.
background
As the use of weight loss and diabetes drugs continues to increase around the world, more women of reproductive age are becoming pregnant while on these treatments without even knowing they are pregnant. GLP-1 receptor agonists are increasingly used to treat type 2 diabetes and obesity.
Weight loss and improved metabolic control during treatment may restore ovulatory function, especially in women with obesity-related reproductive dysfunction, increasing the likelihood of unintended pregnancy. However, although the use of these drugs is expanding, evidence regarding their safety in humans remains limited, making it still difficult for medical professionals to give patients reliable advice about their use during pregnancy.
About research
The researchers conducted a systematic review and meta-analysis according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines, and study protocols were registered with the International Register of Systematic Reviews (PROSPERO) prior to data collection.
The electronic search targeted numerous databases, including PubMed, MEDLINE, Embase, Web of Science, and Reprotox, and covered the period from database inception to January 2026. Additionally, we examined the references of included studies and related reviews to further identify potentially relevant publications.
Eligible studies were cohort or case-control studies that assessed maternal exposure to GLP-1 receptor agonists from 90 days before pregnancy until the end of pregnancy and included an appropriate unexposed comparison group.
When pooling data for scientific analysis, studies based solely on reviews, editorials, animal studies, and voluntary pharmacovigilance reports were excluded. Risk of bias was assessed using the Risk of Bias in Non-Randomized Intervention Studies (ROBINS-I) tool and methodological quality was assessed using the Newcastle-Ottawa Scale. The Grading, Development, and Evaluation of Recommendations Framework (GRADE) system was applied to assess the certainty of evidence for each outcome. This study also includes qualitative evaluation of data from case reports, case series, pharmacovigilance analyses, and other studies not suitable for quantitative pooling.
Research results
Seven cohort studies involving more than 40,000 pregnancies exposed to GLP-1 receptor agonists met the criteria for quantitative evaluation. The timing of exposure varied between studies, with two analyzing exposure during pregnancy, while the other five focused on exposure during the first trimester.
The most frequently reported drugs were semaglutide and liraglutide, followed by exenatide and dulaglutide. Tirzepatide was reported in only one study, and each drug could not be analyzed separately. Quality assessment showed that the majority of studies had moderate to high methodological quality.
Studies reporting adjusted analyzes had a low to moderate risk of bias, whereas studies reporting only rough estimates had a severe risk of bias, mainly due to confounding. Overall, the certainty of the evidence for the assessed results was rated as low or very low due to study limitations and imprecise estimates.
The analysis of congenital malformations included 7 studies with 42,282 exposure pregnancies and 723,892 comparison pregnancies. Maternal exposure to GLP-1 receptor agonists during pregnancy was not associated with a statistically significant increase in overall risk of birth defects, with an odds ratio (OR) of 1.11 and a 95% confidence interval (CI) of 0.82 to 1.51.
Five studies, including 825 exposure pregnancies and 592,714 comparison pregnancies, evaluated first-trimester exposure and major congenital malformations. This analysis also found no statistically significant increased risk (OR 1.39; 95% CI, 0.73-2.65).
A sensitivity analysis focusing only on 4 studies reporting adjusted effect estimates, including 793 exposed pregnancies and 488,871 comparison pregnancies, yielded similar results (OR 1.40; 95% CI, 0.63-3.12).
Three studies, including 41,984 exposure pregnancies and 136,485 comparison pregnancies, found no statistically significant association with cardiac congenital malformations (OR 0.92; 95% CI, 0.67-1.27).
In contrast, two studies including 41,046 exposure pregnancies and 131,407 comparison pregnancies identified a statistically significant association with urinary tract congenital malformations (OR 1.24; 95% CI, 1.05-1.47). However, these findings are based only on unadjusted estimates. The results were driven primarily by one large cohort, and neither study reported specific urinary tract malformation subtypes, limiting causal interpretation.
None of the other pooled adverse pregnancy outcomes reached statistical significance, although some estimates remain imprecise.
Three studies with a total of 366 exposure pregnancies and 873 comparison pregnancies found that stillbirth estimates did not reach statistical significance (OR 2.17; 95% CI, 0.95-4.91). However, the elevated point estimate and wide confidence interval meant that a clinically significant increase could not be ruled out.
Two other studies, including 206 exposure pregnancies and 267 comparison pregnancies, showed no evidence of an association with spontaneous abortion (OR 0.95; 95% CI, 0.42-2.16).
Three studies with 295 exposure pregnancies and 100,239 comparison pregnancies found no significant increase in early age delivery for gestational age (OR 0.84; 95% CI 0.43-1.64).
Four studies including 41,225 exposure pregnancies and 232,786 comparison pregnancies also reported no statistically significant increase in preterm birth (OR 1.17; 95% CI, 0.79-1.72). However, this estimate exhibits considerable uncertainty as there is considerable heterogeneity across studies. A qualitative review found no consistent recurrence or biologically plausible pattern of congenital anomalies across case reports and case series.
conclusion
The authors concluded that maternal exposure to GLP-1 receptor agonists was not significantly associated with major congenital malformations, stillbirth, spontaneous abortion, small-for-gestational age miscarriage, or preterm birth. Although an association with congenital anomalies of the urinary system was identified, this is based on limited unadjusted data and may reflect confounding by maternal diabetes or obesity rather than a true treatment effect.
Overall, the results of this study provide cautious and uncertain reassurance, particularly regarding inadvertent maternal exposure during early pregnancy, but do not establish that these drugs are safe for routine use throughout pregnancy. Comparison groups and definitions of exposure also varied between studies, and prescription records did not necessarily confirm that drugs were taken. Nevertheless, the low or very low certainty of evidence highlights the need for larger population studies with standardized exposure definitions, detailed outcome classification, and comprehensive adjustment for confounding. Further research is needed to strengthen the available evidence regarding pregnancy safety.
Reference magazines:
- N. Uysal, E. Holoz, M. Gungor, I. Timarsi, MK Sozmen, B. Karadas, YC Kaplan (2026). Pregnancy outcomes after maternal GLP-1 receptor agonist exposure: A systematic review and meta-analysis. scientific report. DOI: 10.1038/s41598-026-61582-8, https://www.nature.com/articles/s41598-026-61582-8
