Abstract
AIMS/INTRODUCTION: Despite there being several meta-analyses on the effects of antidiabetic agents in patients with gestational diabetes mellitus, the reliability of their findings is a concern, mainly due to undetermined methodological quality of these studies. This study aimed to assess the methodological quality of available meta-analyses and provide a summary estimation of the effectiveness of treatments modalities. MATERIALS AND METHODS: PubMed, Web of Science and Scopus databases were comprehensively searched for retrieving relevant meta-analyses published in English up to May 2020. A Measurement Tool to Assess Systematic Reviews (AMSTAR-2) was applied to evaluate methodological quality of eligible meta-analyses. A network meta-analysis was used to calculate the pooled odds ratio of maternal and neonatal outcomes in gestational diabetes mellitus patients treated with metformin or glyburide compared with those treated with insulin. The rank network analysis was carried out for ranking of the treatments and reporting the most efficient treatment. RESULTS: A total of 27 and 17 studies were included for qualitative and quantitative syntheses, respectively; of these, just four studies were classified as high quality. The results showed that metformin had the highest probability of being the best treatment, compared with insulin and glyburide, for the majority of adverse neonatal outcomes, whereas glyburide was the best treatment in reducing the risk of adverse maternal outcomes. The results were not significantly changed after excluding low-quality studies. CONCLUSIONS: This review study of available literature shows that metformin can be a superior option in most neonatal and maternal adverse pregnancy outcomes in women with gestational diabetes mellitus; the results need to be further updated by including future more qualified studies.