Abstract
Summary We conducted a meta-analysis of the mortality and morbidity of the breech presentation according to the method of delivery. Using the term 'breech', we used the Medline data base to search the English and French-language literature from 1978 to July 1995. Twenty-two studies (7239 deliveries) were found and analysed. The meta-analysis concerning perinatal mortality did not reveal a significant increased risk, when vaginal deliveries were compared with those delivered by a caesarean section (odds ratio: 1.90; confidence interval: 0.59-8.22) or when vaginal deliveries were compared with those done by planned caesarean section (odds ratio: 4.95; confidence interval: 0.44-80.06). The neonatal morbidity showed an increased risk with vaginal delivery. It is concluded that the practice of resorting to caesarean section for every breech presentation at term does not seem defensible.