Our aim was to re-evaluate whether metformin may reduce late miscarriage/preterm delivery, pre-eclampsia and gestational diabetes in women with polycystic ovary syndrome (PCOS). We performed an epi-analysis of two randomized controlled trials. The participants were 313 women aged 18–42 years with PCOS who had singleton pregnancies. They were randomized to metformin or placebo treatment from first trimester until delivery. We analysed the prevalence of late miscarriage/preterm delivery, pre-eclampsia and gestational diabetes according to both the intention-to-treat principle and per protocol analysis. The metformin-treated patients had less late miscarriage/preterm delivery; five (3%) vs. 18 (11%) in the placebo group (p < 0.01). There was no difference in the prevalence of gestational diabetes and pre-eclampsia between the metformin and the placebo group. In this epi-analysis, metformin treatment during pregnancy seems to reduce early delivery in women with PCOS. We believe that further randomized studies should be performed before firm conclusions can be drawn.