GLP-1 agonist exposure around conception: no clear maternal harm signal

Design

Systematic review and meta-analysis of 8 observational and trial datasets; random-effects odds ratios

Population

186,598 pregnancies, 47,159 with peri-conception GLP-1 receptor agonist exposure

Primary outcome

Maternal obstetric complications

Effect

No significant difference across gestational diabetes, preterm birth, pre-eclampsia and hypertensive disorders; all intervals crossed 1

Eight studies covering 186,598 pregnancies, 47,159 of them exposed to a glucagon-like peptide-1 receptor agonist before or in early gestation, were pooled for maternal obstetric outcomes. No significant difference emerged for gestational diabetes, preterm birth, pre-eclampsia or hypertensive disorders of pregnancy; every confidence interval crossed 1.

A leave-one-out sensitivity analysis raised the possibility that exposure lowers gestational diabetes odds, but the authors flag this as exploratory. Study quality varied and heterogeneity was high, reflecting differing exposure and outcome definitions.

The practical value is in the booking conversation. As more reproductive-age women conceive while taking these drugs for weight or diabetes, the first-visit question is what the early exposure means. For maternal outcomes, these data support reassurance — while the advice to stop the agent once pregnancy is confirmed is unchanged.

• No significant difference in gestational diabetes (odds ratio 0.99, 95% CI 0.61–1.61), preterm birth (1.01, 0.76–1.33), pre-eclampsia (1.05, 0.60–1.84) or hypertensive disorders (0.79, 0.34–1.83).
• A leave-one-out analysis hinted at lower gestational diabetes odds (0.81, 0.67–0.98), but this is exploratory.
• Reassure a woman who conceived on a GLP-1 agonist, while still advising she stop the drug once pregnancy is confirmed.
• Heterogeneity was high and the data observational — read the null as absence of a strong signal, not proof of safety.

Why it matters

More reproductive-age women are conceiving on GLP-1 agonists, and these pooled data answer what the early exposure means for the mother.

Don't overread it

The analysis covered maternal outcomes only and cannot speak to fetal or longer-term child outcomes.

The statistics, in plain English

An odds ratio near 1 with a confidence interval spanning 1.0 means no detectable difference. High heterogeneity means the studies were measuring somewhat different things, so the pooled estimate is less firm than the numbers alone suggest.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.