ObjectiveThe objective was to review all emergency peripartum hysterectomies performed at a tertiary hospital in London, UK, and to identify the risk factors for emergency peripartum hysterectomy.MethodA retrospective case control study. The cases consisted of all women who had emergency peripartum hysterectomy between 1 January 1993 and 31 December 2003. Controls were women who delivered immediately before and after the indexed case. Demographic data, medical and surgical histories, pregnancy, intrapartum and postpartum data were collected. Differences between cases and controls were compared with χ2, Fisher exact and Student t tests. Multiple logistic regression analysis was performed to identify independent risk factors for emergency peripartum hysterectomy.ResultsThere were 15 cases of emergency peripartum hysterectomy in 31,079 deliveries, giving a rate of 0.48 per 1,000. Women who had emergency peripartum hysterectomy were significantly older (mean age 37 years vs. 29 years, P<0.001) and multiparous (P=0.02). More of the cases had a history of uterine surgery (67 vs. 30%, P=0.01), placenta praevia (60 vs. 3%, P<0.0001) and were delivered by caesarean section (86.7 vs. 30%, P=0.003). Eighty percent of the hysterectomies were performed in the daytime and all were done by consultants. Haemorrhage due to placenta praevia was the main indication for emergency peripartum hysterectomy (47%). Independent risk factors for emergency peripartum hysterectomy were older age (odds ratios [OR] 1.2, 95% confidence interval [95% CI] 1.2–1.6), multiparity (OR 2.6, 95% CI 1.3–10.2), history of previous caesarean section (OR 13.5, 95% CI 2.7–65.4), caesarean delivery in index pregnancy (OR 11.6, 95% CI 2.1–68.6) and caesarean delivery in index pregnancy for placenta praevia (OR 18, 95% CI 3.6–69).Conclusion Caesarean deliveries, especially repeat caesareans in women with placenta praevia, significantly increase the risk of emergency peripartum hysterectomy.