ObjectiveTo assess the success rate of vaginal delivery among women with twin pregnancies; the Twin Birth Study has shown that vaginal delivery and caesarean section are equally safe for twin delivery but >40% of the planned vaginal delivery group delivered by caesarean section. DesignA retrospective cohort study. SettingA tertiary medical centre. PopulationA total of 2194 women with twin pregnancies not complicated with very low birthweight. MethodsPlanned mode of delivery was documented in the woman's electronic record upon entering the delivery room. Information regarding maternal age at delivery, parity, gestational age, presentation, previous history of caesarean delivery, birthweight and Apgar score was collected from the obstetric electronic charts. Main outcome measuredRate of vaginal twin delivery. ResultsOf the 2194 women included, 1311 twin pregnancies had planned caesarean delivery and 883 underwent a trial of labour. Of the 883 women who underwent a trial of labour, the rate of vaginal delivery was 86.9%, whereas the rates of caesarean delivery and combined vaginal-caesarean delivery were 11.1% and 2.0%, respectively. Presentation of second twin, gestational age and maternal age did affect the chances of success. Nulliparity [odds ratio (OR) 2.38, 95% confidence interval (95% CI) 1.4-4.05], Foley induction of labour (OR 2.33, 95% CI 1.38-3.91) and body mass index >30kg/m(2) (OR 1.76, 95% CI 1.03-3) were independent risk factors for caesarean delivery. ConclusionsThe rate of vaginal delivery among women with twin pregnancies who undergo labour can be high, especially in women who laboured spontaneously and have delivered before. Tweetable abstractThe rate of vaginal delivery of twins can be high, especially in women who have delivered before. Tweetable abstract The rate of vaginal delivery of twins can be high, especially in women who have delivered before.