Obesity and metformin in pregnancy
Obesity is the most common antenatal comorbidity, affecting one in five of the antenatal population in the UK. It is associated with adverse outcomes for mother and baby in both the short and long term. Increasing data suggest that maternal obesity may programme offspring later life obesity and premature mortality, with high birth weight being a marker for increased risk. The mechanism by which maternal obesity causes excessive neonatal birth weight is incompletely understood but considerable evidence implicates insulin resistance and/or hyperglycaemia. There are currently no effective interventions to mitigate the effects of obesity during pregnancy. In this thesis, we present the findings from a randomised, double blind, placebo controlled trial designed to examine the efficacy of metformin, an insulin-sensitising agent, in obese pregnant women. The aim of the trial was to determine whether giving metformin to obese pregnant women from between 12 and 16 weeks’ gestation until birth, would improve maternal and fetal outcomes. The primary outcome measure was birth weight of the baby, using this as a surrogate marker for the future life risk of the child developing obesity. Nested within this large clinical trial were a series of mechanistic sub-studies. To examine the effect of metformin on maternal insulin resistance at 36 weeks’ gestation, we used the hyperinsulinaemic euglycaemic clamp with concomitant use of stable isotope tracers. This enabled us to characterise in greater detail insulin sensitivity, endogenous glucose production and lipolysis. To determine the effect of metformin on maternal and fetal body composition we used magnetic resonance imaging and spectroscopy. This allowed us to quantify subcutaneous and intra-abdominal adipose tissue deposition and hepatic and skeletal muscle ectopic lipid deposition in the mother; and to measure subcutaneous adipose tissue deposition, hepatic lipid and hepatic volume in the fetus. To determine the effect of metformin on maternal endothelial function, we measured endothelium-dependent flow-mediated dilatation at the beginning and end of pregnancy. Change in diameter of the brachial artery in response to a flow stimulus created by arterial occlusion was measured using ultrasound imaging. We found no significant effect of metformin on birth weight. Mean birth weight was 3463 g (SD 660) in the placebo group and 3462 g (SD 548) in the metformin group (adjusted mean difference in z score –0·029, 95% CI –0·217 to 0·158; p=0·7597). Subjects taking metformin did demonstrate increased insulin sensitivity (M/I difference between means during high dose insulin of 0.02 [95% CI 0.001 to 0.03] milligrams per kilogram fat free mass per minute per pmol/L, p=0.04) but also enhanced endogenous glucose production (difference between means 0.54 [95% CI 0.08 to 1.00] milligrams per kilogram fat free mass per minute, p=0.02), compared with those taking placebo. We did not demonstrate any differences between treatment groups in maternal subcutaneous and intra-abdominal adipose tissue, or ectopic lipid deposition, or in fetal body fat distribution and liver volume. Participants in both treatment groups demonstrated a decline in endothelium-dependent flow-mediated dilatation between early and late pregnancy but there were no differences in the magnitude of that decline between the treatment groups. In conclusion, metformin, administered to obese, non-diabetic pregnant women, does not have any significant effect on birth weight of the baby. Our clamp studies demonstrated that subjects taking metformin were indeed more insulin-sensitive than those taking placebo, but the higher endogenous glucose production in this group suggests a reduced ability to suppress hepatic glucose production in response to insulin. This increased glucose flux may in part explain the lack of effect of metformin on fetal nutrition and growth. We can conclude that metformin, should not be used as an intervention in obese pregnant women to prevent excess birth weight. The global obesity epidemic is one of the greatest public health challenges we face and the cycle of disadvantage continues to be perpetuated to the next generation. The lack of any effective interventions for this high-risk group remains a significant concern and an important area for further research.