BACKGROUND Induction of labour after prelabour rupture of membranes may reduce the risk of neonatal infection. However an expectant approach may be less likely to result in caesarean section. OBJECTIVE The objective of this review was to assess the effects of induction of labour with prostaglandins versus expectant management for prelabour rupture of membranes at or near term. METHODS We searched the Cochrane Pregnancy and Childbirth Group trials register. METHODS Randomised and quasi-randomised trials comparing early use of prostaglandins (with or without oxytocin) with no early use of prostaglandins in women with spontaneous rupture of membranes before labour, and 34 weeks or more of gestation. METHODS Trials were assessed for quality and data were abstracted. RESULTS Fifteen trials were included. Most were of moderate to good quality. Different forms of prostaglandin preparations were used in these trials and it may be inappropriate to combine their results. Induction of labour by prostaglandins was associated with a decreased risk of chorioamnionitis (odds ratio 0.77, 95% confidence interval 0.61 to 0.97) based on eight trials and admission to neonatal intensive care (odds ratio 0.79, 95% confidence interval 0.66 to 0.94) based on seven trials. No difference was detected for rate of caesarean section, although induction by prostaglandins was associated with a more frequent maternal diarrhoea and use of anaesthesia and/or analgesia. Based on one trial, women were more likely to view their care positively if labour was induced with prostaglandins,. CONCLUSIONS Induction of labour with prostaglandins appears to decrease the risk of maternal infection (chorioamnionitis) and admission to neonatal intensive care. Induction of labour with prostaglandins does not appear to increase the rate of caesarean section, although it is associated with more frequent maternal diarrhoea and pain relief.