Transfusion and Haemostasis
Background and Goal of Study: No recent data are available on blood management in elective knee or hip arthroplasty in Belgium. The primary objectives of this retrospective study were to determine prevalence of pre‐operative anaemia and rates of transfusion in this population. Secondary objectives included blood loss, evolution of haemoglobin (Hb) and variability among centres.
Materials and Methods: Data were collected in consecutive patients (pts) undergoing elective surgery for hip or knee prosthesis in Belgium between Jan 1st and March 31st 2008. Pts were excluded if age <18, emergency surgery, major medical event (e.g. trauma, haemorrhage) in the 30 days before surgery or inclusion in an interventional study. Hb levels were collected pre‐ and postoperatively, and prior to transfusion. Blood loss was calculated as described (1). Transfusion rate was determined for the whole population and per centre. Descriptive statistics were calculated using SAS (version 9.1.3).
Results and Discussion: 2449 pts from 23 centres were included (median 100, range 50–202 per centre). Mean age was 66, 61% were women and 51% were hip surgeries. There were no bilateral surgeries, and 9% were revisions. Blood sparing techniques were used in 4%. Pre‐op Hb was determined ≤7 days pre‐op in 37% of pts, 8–21 days pre‐op in 33%, and >21 days pre‐op in 30%. Mean pre‐op Hb was 13.9 g/dl (range of mean Hb per centre 13.4 ‐ 14.3). Hb was <13g/dl in 24% of patients. 17% of pts were transfused (range per centre 2–53%). Timing of transfusions: day of surgery: 16%, post‐op day 1: 16%, day 2–4: 40%, day 5+: 28%. Mean decrease in Hb from pre‐op to last measured was 3.2 g/dl (range per centre 2–4.5). Mean pre‐transfusion Hb was 8.2 g/ dl (range per centre 7.1–9.3). Figure 1 shows the variability in blood loss and transfusion rates. Note: blood loss could not be calculated where volume of transfusions was not known.
Conclusion(s): In Belgium there is a remarkable variability in transfusion rates among centres, possibly in part related to differences in blood loss and transfusion trigger. Clearly, where improvement of blood management is needed this will require a tailored approach per centre.
1 Samama CM et al. Anesth Analg 2002;95:287-93.
All RetroOrtho investigators.