Metformin is widely used in pregnancy, despite lack of long-term safety for children. We hypothesised that metformin exposure in utero is associated with increased cardiovascular risk. We tested this hypothesis in a follow-up study of children born to obese mothers who had participated in a randomised controlled trial of metformin versus placebo in pregnancy (EMPOWaR). We measured body composition, peripheral blood pressure (BP), arterial pulse wave velocity and central haemodynamics (central BP and augmentation index) using an oscillometric device in 40 children of mean (SD) age 5.78 (0.93) years, exposed to metformin (n = 19) or placebo (n = 21) in utero. There were no differences in any of the anthropometric or vascular measures between metformin and placebo-exposed groups in univariate analyses, or after adjustment for potential confounders including the child’s behaviour, diet and activity levels. Post-hoc sample size calculation indicated we would have detected large clinically significant differences between the groups but would need an unfeasible large number to detect possible subtle differences in key cardiovascular risk parameters in children at this age of follow-up. Our findings suggest no evidence of increased cardiovascular risk in children born to obese mothers who took metformin in pregnancy and increase available knowledge of the long-term safety of metformin on childhood outcomes.