Abstract
Objective. To estimate the incidence of peripartum hysterectomy in an Italian Region (Emilia-Romagna) and investigate its association with cesarean delivery. Design. Population-based retrospective study using hospital discharge records. Setting. All public and private hospitals in Emilia-Romagna region, Italy. Population. A total of 151,494 women delivering between 2003 and 2006, 131 of whom had a peripartum hysterectomy. Methods. Peripartum hysterectomy was defined as a hysterectomy performed at the time of delivery or afterwards during the same hospitalization. Incidence rates were calculated by type of delivery. Odds ratios (ORs) with 95% confidence intervals (CIs) were estimated with logistic regression to evaluate the association between peripartum hysterectomy and delivery type. Main outcome measures. Incidence rates of peripartum hysterectomy by type of delivery; odds of peripartum hysterectomy after primary or repeat cesarean compared with vaginal delivery without previous cesarean. Results. A total of 131 peripartum hysterectomies were performed among 151,494 deliveries (0.86/1,000 deliveries; 95% CI 0.72–1.03) with 20.7% primary and 9.6% repeat cesarean deliveries. Women undergoing a primary caesarean delivery were more likely to have a peripartum hysterectomy than women having a vaginal delivery who had never had a cesarean delivery (OR 6.48; 95% CI 4.16–10.07). Women undergoing a repeat caesarean delivery were also at increased risk (OR 3.69; 95% CI 2.11–6.46). Conclusions. In this population, primary and repeat cesarean deliveries are associated with an increased risk of peripartum hysterectomy. These results are of particular concern given the steady increase in the cesarean delivery rate in many countries. The pathological mechanism of this association requires further investigation.
Acknowledgements
MK is funded by the National Institute for Health Research National Coordinating Centre for Research Capacity Development. All authors are independent of all funders. We greatly appreciated comments provided by Paola Rucci in the preparation of this manuscript.
Declaration of interest: The authors have no potential conflict of interest to disclose.