OPTIMAL management of Type 2 diabetes in pregnancy is unknown. A research team led by Kim Boggess, University of North Carolina at Chapel Hill, USA, aimed to estimate the effect of metformin added to insulin for the treatment of maternal Type 2 diabetes.
The multicentre trial randomly allocated patients with pre-gestational Type 2 diabetes, or diabetes diagnosed after 22 weeks gestation, into two treatment groups. The first received insulin and metformin, while the second received insulin and a placebo until delivery. Composite outcomes recorded encompassed miscarriage, stillbirth, birth trauma, neonatal death, neonatal hypoglycaemia that required treatment, pre-term birth, large for gestational age, and hyperbilirubinemia requiring phototherapy. Maternal hypoglycaemia and neonatal fat mass at birth were also recorded.
Each group contained 397 participants. The metformin group had a mean age of 32.8, where 76% had had a previous live birth at >20 weeks’ gestation and 93% had used metformin in the past. The placebo group had an average age of 33.1, where 86% had had a previous live birth at >20 weeks’ gestation and 97% had used metformin in the past.
Importantly, the rates of composite neonatal outcome (70% versus 73%) and maternal hypoglycaemia (22% versus 21%) occurred at similar rates in the metformin and placebo group, respectively; however, infant fat mass did not differ. A subgroup analysis of females with a BMI greater than 30 kg/m2 or pre-gestational diabetes also showed no difference in composite neonatal outcomes. However, the rates of large for gestational age was higher in the metformin group (adjusted odds ratio: 0.66; 95% confidence interval: 0.50–0.92) without affecting the mode of delivery.
Therefore, adding metformin to insulin to treat Type 2 diabetes during pregnancy does not reduce the risk of composite adverse neonatal outcomes, maternal hypoglycaemia, or infant fat mass. However, the reduction in the rate of large for gestational age in the metformin group suggests additional studies are needed to measure long term metabolic outcomes in infants. Boggess concluded: “For populations comparable to our study group, the use of metformin should be discouraged.”
