Predictive model for postpartum hemorrhage requiring hysterectomy in a minority ethnic region

Abstract

BACKGROUND Postpartum hemorrhage (PPH) is a leading cause of maternal mortality, and hysterectomy is an important intervention for managing intractable PPH. Accurately predicting the need for hysterectomy and taking proactive emergency measures is crucial for reducing mortality rates. AIM To develop a risk prediction model for PPH requiring hysterectomy in the ethnic minority regions of Qiandongnan, China, to help guide clinical decision-making. METHODS The study included 23490 patients, with 1050 having experienced PPH and 74 who underwent hysterectomies. The independent risk factors closely associated with the necessity for hysterectomy were analyzed to construct a risk prediction model, and its predictive efficacy was subsequently evaluated. RESULTS The proportion of hysterectomies among the included patients was 0.32% (74/23490), representing 7.05% (74/1050) of PPH cases. The number of deliveries, history of cesarean section, placenta previa, uterine atony, and placenta accreta were identified in this population as independent risk factors for requiring a hysterectomy. Receiver operating characteristic curve analysis of the prediction model showed an area under the curve of 0.953 (95% confidence interval: 0.928-0.978) with a sensitivity of 90.50% and a specificity of 90.70%. CONCLUSION The model demonstrates excellent predictive power and is effective in guiding clinical decisions regarding PPH in the ethnic minority regions of Qiandongnan, China.