Twin pregnancies are becoming more common because of the increasing use of assisted reproduction technologies.1 They are associated with an eight to tenfold increase in the perinatal mortality rate,2 mainly because 40-70% of twins are born preterm.3 However, the safety of term vaginal delivery for twins has long been of concern; some large epidemiological studies have suggested that the second twin is at especially high risk of death.4 Complications associated with the second twin include the longer second stage, compound presentation leading to trauma during delivery, cord prolapse, and premature separation of the placenta.5 However, the only randomised controlled trial identified by a Cochrane review6 of caesarean section compared with vaginal delivery in twins7 and other small retrospective studies8 have not confirmed clinicians' subjective impressions of poor outcome in the second twin.

The study by Smith and colleagues in this week's BMJ is welcome for the light …