RESUMO
AIMS: We aimed to assess the comparative efficiency and safety of the use of glyburide, metformin, and insulin in gestational diabetes mellitus (GDM). METHODS: We searched for randomized controlled trials that compared glyburide, metformin, and insulin in GDM. Data regarding glycemic control and neonatal safety were collected and analyzed in pairwise and network meta-analyses. RESULTS: A total of 4533 individuals from 23 trials were included. Compared with glyburide, metformin reduced 2-h postprandial blood glucose (2HPG) to a greater extent (standard mean difference (SMD) 0.18; 95% credible interval (CI) 0.01, 0.34). There were significantly lower prevalence of neonatal hypoglycemia (risk difference (RD) - 0.07; 95%CI - 0.11, - 0.02) and preeclampsia (RD - 0.03; 95%CI - 0.06, 0) in the metformin group than in the insulin group. The metformin group had significantly lower birth weight (SMD - 0.17; 95%CI - 0.25, - 0.08) and maternal weight gain (SMD - 0.61; 95%CI - 0.86,- 0.35) compared with the insulin group. Network meta-analysis suggested that metformin had the highest probability of successfully controlling glycemia and preventing neonatal complications. CONCLUSIONS: The present meta-analysis suggests that metformin may be as effective as insulin for glycemic control and is the most promising drug for the prevention of neonatal and maternal complications.