In Reply We thank Mr Meng and colleagues and Dr Park and colleagues for their comments about the EMERGE trial. We agree with using caution in the interpretation of the trial, as the primary outcome did not reach the prespecified level of statistical significance. However, the constituents of the primary outcome (insulin initiation or a mean fasting glucose level of ≥5.1 mmol/L at weeks 32 and 38) were significantly improved in those randomized to metformin. In addition, secondary outcomes of maternal insulin dose, weight gain, and satisfaction with treatment also favored metformin. The rates of large for gestational age and macrosomia were lower in metformin-exposed pregnancies, without an increase in neonatal or maternal morbidities. Unlike the MiG trial, gestational age at delivery and preterm births were similar between groups in the EMERGE trial; our results are consistent with meta-analyses.
