Predictive factors for failure of pelvic arterial embolization for postpartum hemorrhage

Objective: To assess the efficacy of pelvic embolization in women with postpartum hemorrhage (PPH) and to determine factors associated with embolization failure.

Methods: In a retrospective observational study, data were analyzed from 98 consecutive women who underwent pelvic embolization for intractable PPH between January 2007 and November 2009 at Beaujon Hospital, Clichy, France. Women with persistent PPH despite pelvic embolization were compared with women who had cessation of hemorrhage and attained hemodynamic stability.

Results: Among the 98 women, 66 (67.3%) had been transferred from another hospital after delivery. Pelvic arterial embolization failed to control PPH in 8 (8.2%) women. On univariate analysis, factors significantly associated with embolization failure were the presence of placenta accreta (P<0.005), hemoglobin level (P<0.05), prothrombin time (P<0.04), fibrinogen level (P<0.03), red blood transfusion (P<0.02), number of packed red blood cell units transfused (P<0.05), and fresh-frozen plasma transfusion (P<0.02). Hospital-to-hospital transfer with a notable time interval between delivery and embolization was not associated with increased risk of failure.

Conclusion: Predictive factors significantly associated with failed pelvic arterial embolization were the presence of placenta accreta, biologic factors, and transfusional factors. Delay due to inter-hospital transfer did not affect the outcome of embolization.