This study aimed to assess the perinatal outcome, especially foetal growth, following the continuation of metformin during the first trimester of pregnancy. All women with polycystic ovary syndrome (PCOS) treated with metformin in the first trimester and who delivered a baby weighing 500 g or more between 2003 and 2005 were studied. Subjects were matched for age and parity with randomly selected controls. The perinatal outcomes studied were: growth parameters, gestational age, congenital defects, hypoglycaemia and neonatal unit admission. Sixty-six pregnancies were compared with 66 controls; all had singleton deliveries. There was no difference in mean birth weight between the metformin and the control groups (p=0.84). The percentage of small (<10th centile) and large (>90th centile) for gestational age babies was lower in the metformin group. In the metformin group, there were no major congenital malformations and 24% of the babies were admitted to the neonatal intensive care unit (NICU) compared with 27% of the babies in the control group (non-significant). Neonatal hypoglycaemia was less common in the metformin group (18.5% vs. 24.5%) and fewer babies required intravenous glucose therapy (6.3% vs. 12%). We found no evidence that the continuation of metformin in the first trimester of pregnancy was associated with an adverse foetal outcome.