In France, the maternal mortality ratio due to postpartum hemorrhage (PPH) is higher than in other high resources countries. This situation remains partially unexplained. The objectives of this PhD thesis were: 1/ to compare PPH epidemiological characteristics between France and Canada, 2/ to describe anesthesia and intensive care practices in PPH and to compare them with guidelines, first in cases of maternal deaths from PPH and secondly in a large population of women with PPH. The data sources were: the Discharge Abstract Database from the Canadian Institute for Health Information, the Pithagore6 trial and the French Confidential Enquiry into Maternal Deaths. The comparison of PPH epidemiological profiles between France and Canada shows that PPH incidence is not higher in France. Second-line treatments in PPH management are more frequently performed, suggesting a higher incidence of severe PPH in France. Among maternal deaths from PPH, some practices in anesthesia and critical care management appear to be inadequate: clinical monitoring and laboratory assessment, protocol for general anesthesia and transfusion strategy. The description on transfusion practices in a large population of women with PPH shows that the use of blood products is not sufficient. Conversely, transfusion strategies with unproved efficacy are frequently used. These results suggest a higher rate of severe PPH in France, that may result in the higher maternal mortality due to PPH. Inadequate practices in anesthesia and critical care could be involved in this severity. But the association between specific components of anesthesia and critical care management and PPH severity remains to be explored
