Objective:  The objective of this study was to assess the association between transfusion, per cent drop in haemoglobin (Hb), and estimated blood loss during the delivery and the first postoperative week following caesarean delivery for placenta praevia. Clinical data predictive of an objective laboratory test for risk of haemorrhage and the need for transfusion were investigated. Transfusions outside national Guidelines were noted.
Design:  Retrospective observational study of patients with placenta praevia, who were delivered consecutively by caesarean section at Royal Brisbane and Women’s Hospital from 1999 to 2005.
Setting:  University‐affiliated tertiary hospital. All caesareans were performed by one or more consultant obstetricians, gynaecology oncology surgeons and registrar assistants.
Results:  Seventy‐one (28.9%) of 246 patients with placenta praevia were transfused, with 45 of these receiving three or more red cell units. The antenatal Hb fell by a mean of 20.2% (SD 13.5). The average operative haemorrhage was estimated as 1225 mL (SD 996). No patient or surgical factors were significantly associated with changes in Hb. There was a significant association between per cent fall in antenatal Hb and both transfusion P < 0.001 and estimated loss P = 0.002. After transfusion, the Hb of 19 patients was higher than that recommended by Guidelines. Conclusions:  Whether transfusion is necessary, but not the number of red cell units, can be planned by the effect of haemorrhage on antenatal Hb during delivery by caesarean section complicated by placenta praevia.