Intraoperative autotransfusion in hip arthroplasty.

Two groups of patients having primary or revision total hip replacement were studied during a period of 11 months. The first group of 40 consecutive patients consented to have whole blood collected coincident with acute normovolemic hemodilution. An inventory of autologous red blood cells, plasma, and platelets was prepared in the operating room as a preliminary to surgery. Subsequently, the same supplies and equipment were used for autotransfusion intraoperatively. In a case-control study, another 40 patients having total hip replacements were matched for age, gender, and weight. The second group of patients refused hemodilution and received autotransfusion alone. The perioperative transfusion requirements of the two groups were compared. The total blood product donor exposure rate of the first group was 1/4 of the controls (0.6 and 2.4 donor units per patient, respectively). The average length of stay in the hospital after surgery for the first group was shortened significantly (6.2 versus 8.4 days), possibly from less immunogenic insult associated with increased transfusions of autologous blood products. Hemodilution, followed by autotransfusion, was cost effective in primary and revision total hip replacements, autotransfusion alone was cost effective only in revision arthroplasty.