Objectives: Prospectively assess autologous blood transfusion for prog
rammed orthopedic procedures. Methods: From January 1 to December 31 1
993, 307 patients underwent programmed orthopedic procedures: total hi
p replacement (n = 191), total knee replacement (n = 83) and osteotomy
(n = 33). General (94%) or spinal anesthesia (6%) was used. The anest
hesist explained transfusion techniques and patients gave informed con
sent for inclusion in an autologous transfusion protocol including dif
fered autologous transfusion, intentional normovolemic hemodilution an
d intraoperative transfusion of shed blood. Results: A total of 269 au
tologous transfusions were performed among the 307 patients (87.6%). T
his was sufficient in 242 cases (78.8%) and in 65 (21.2%) homologous t
ransfusion was required. Among the 269 patients given autologous trans
fusions, differed transfusion was used in 145 (53.9%), intentional nor
movolemic hemodilution in 124 (46%) and intraoperative transfusion of
shed blood in 222 (82.5%). Among the patients given a differed autolog
ous transfusion, 9 (6.2%) required a homologous transfusion and among
the 40 patients in which all 3 techniques were used, only 2 (5%) recei
ved homologous blood, both due to secondary complications. Conclusion:
These findings show that when differed autologous transfusion is incl
uded in the transfusion strategy, less than 10% of the patients requir
e homologous blood. In addition, when the 3 autologous transfusion tec
hniques are used, the rate of homologous blood transfusion approaches
zero.