ESPE Yearbook of Paediatric Endocrinology

Diabetes Obes Metab. 2025;27(2):976–986. doi: 10.1111/dom.16115

Brief Summary: This population-based retrospective cohort study investigated whether initiating metformin instead of insulin after 20 weeks of gestation affects adverse pregnancy outcomes among women with gestational diabetes mellitus (GDM). The study used data from the UK Clinical Practice Research Datalink and Hospital Episode Statistics from 1998 to 2018, including 2,192 singleton pregnancies (1,297 on metformin; 895 on insulin).The primary outcome was a composite of large for gestational age (LGA) and macrosomia.

Key Findings include: 1) Compared to insulin initiation, metformin was associated with a lower risk of LGA infants and macrosomia (HR 0.64; 95% CI 0.49–0.78). 2) Metformin was also linked to reduced risk of caesarean delivery (HR 0.83; 95% CI 0.69–0.98) and a trend to lower risk of preterm birth (HR 0.83; 95% CI 0.58–1.08). 3) No difference was found in hypertensive disorders of pregnancy (HR 0.92; 95% CI 0.57–1.27). 4) Risk of SGA infants appeared higher with metformin, although not statistically robust (HR 1.33; 95% CI 0.67–2.00). 5) Sensitivity analyses, including per-protocol, high-dimensional propensity score adjustment, and bias models, confirmed the robustness of primary findings.

The study provides real-world evidence supporting metformin as a potentially safer, effective alternative to insulin in the treatment of GDM, especially in cases where insulin adherence is poor. However, the possible increased risk of SGA births as well as the previously summarised study suggesting that oral hypoglycemic agents do not significantly reduce the risk of LGA birthcalls for further investigation. Individualized risk benefit assessment remains essential when considering metformin in pregnancy.