Critical care and transfusion management in maternal deaths from postpartum haemorrhage

Abstract

Objectives In postpartum haemorrhage (PPH), as for other causes of acute haemorrhage, management can have a major impact on patient outcomes. The aim of this study was to describe critical care management, particularly transfusion practices, in cases of maternal deaths from PPH. Study design This retrospective study provided a descriptive analysis of all cases of maternal death from PPH in France identified through the systematic French Confidential Enquiry into Maternal Death in 2000–2003. Results Thirty-eight cases of maternal death from PPH were analysed. Twenty-six women (68%) had a caesarean section [21 (55%) emergency, five (13%) elective]. Uterine atony was the most common cause of PPH ( n = 13, 34%). Women received a median of 9 (range 2–64) units of red blood cells (RBCs) and 9 (range 2–67) units of fresh frozen plasma (FFP). The median delay in starting blood transfusion was 82 (range 0–320) min. RBC and FFP transfusions peaked 2–4 h and 12–24 h after PPH diagnosis, respectively. The median FFP:RBC ratio was 0.6 (range 0–2). Fibrinogen concentrates and platelets were administered to 18 (47%) and 16 (42%) women, respectively. Three women received no blood products. Coagulation tests were performed in 20 women. The haemoglobin concentration was only measured once in seven of the 22 women who survived for more than 6 h. Twenty-four women received vasopressors, a central venous access was placed in 11 women, and an invasive blood pressure device was placed in two women. General anaesthesia was administered in 37 cases, with five patients being extubated during active PPH. Conclusions This descriptive analysis of maternal deaths from PPH suggests that there may be room for improvement of specific aspects of critical care management, including: transfusion procedures, especially administration delays and FFP:RBC ratio; repeated laboratory assessments of haemostasis and haemoglobin concentration; invasive haemodynamic monitoring; and protocols for general anaesthesia.