EFFECTIVENESS OF PERIOPERATIVE RECOMBINANT-HUMAN-ERYTHROPOIETIN IN ELECTIVE HIP-REPLACEMENT

Concern about the risk of transmission of viral infection has led to a ttempts to reduce transfusion requirements in patients undergoing surg ery. To determine whether recombinant human erythropoietin decreases b lood transfusion requirements in patients undergoing elective hip arth roplasty, a multicentre double-blind, randomised, placebo-controlled t rial was conducted. 208 patients undergoing elective primary or revisi on hip arthroplasty were randomised to 3 groups. All received daily su bcutaneous injections of either erythropoietin or placebo starting 10 days before surgery. Group 1 (78 patients) received 14 days of placebo , group 2 (77 patients) received 14 days of erythropoietin (300 units/ kg to a maximum of 30 000 units), and group 3 (53 patients) received p lacebo for days 10 to 6 before surgery and erythropoietin for the next 9 days. A primary outcome event (any transfusion or a haemoglobin con centration < 80 g/L) occurred in 46% of patients in group 1, 23% in gr oup 2, and 32% in group 3 (p = 0.003). The mean number of transfusions was 1.14 in group 1, 0.52 in group 2, 0.70 in group 3. The mean retic ulocyte count the day before surgery was 72 x 10(9)/L in group 1, 327 in group 2, and 170 in group 3. Deep venous thrombi were detected in 5 patients in group 1, 8 patients in group 2, and 8 patients in group 3 . Patients who had a haemoglobin concentration before randomisation of < 135 g/L benefited most from erythropoietin. Thus erythropoietin giv en for 14 days perioperatively decreases the need for transfusion in p atients undergoing elective hip arthroplasty.