Publications

Abstract

OBJECTIVE: To assess the efficacy of metformin and insulin in the treatment of pregnant women with gestational diabetes mellitus (GDM). METHODS: A meta-analysis was conducted by including randomized controlled trials comparing metformin and insulin in GDM. An electronic search was conducted to identify relevant studies. Data were synthesized by a random effects meta-analysis model. A Bayesian analysis was also performed to account for uncertainties in the treatment efficacy. RESULTS: Eight clinical trials involving 1712 individuals were included in the final analysis. The pooled estimates of metformin-insulin differences were very small and statistically non-significant in fasting plasma glucose, postprandial plasma glucose and HbA1c, measured at 36-37 weeks of gestation. Notably, 14-46% of those receiving metformin required additional insulin. Compared with the insulin group, metformin treatment was associated with a lower incidence of neonatal hypoglycemia (relative risk, RR 0.74; 95% CI 0.58-0.93; P=0.01) and of neonatal intensive care admission (RR 0.76; 95% CI 0.59-0.97; P=0.03). Bayesian analysis revealed that the efficacy of metformin was consistently higher than insulin with a probability of over 98% on these two neonatal complications. Other outcomes were not significantly different between the two treatment groups. CONCLUSION: In women with gestational diabetes, metformin use and insulin therapy have comparable glycemic control profile, but metformin use was associated with lower risk of neonatal hypoglycemia.