Perioperative blood loss associated with 89 cases of major orthopedic surgery was compared with that of a control group of 89 to determine the effectiveness of intraoperative autologous transfusion. Volume of banked blood transfused and hematocrit change were used to determine total blood loss. The orthopedic cases consisted of cemented "virgin" total hip replacement, cemented virgin tricompartmental knee replacement, and spine fusion. Use of an autotransfusion device (Cell Saver) intraoperatively was associated with significantly smaller volumes of transfused banked blood and significantly smaller hematocrit drops in the groups of patients who underwent total hip replacement or spine fusion, but not in the group of patients who underwent total knee replacement. One potential source of bias in the study stems from the fact that four days were allotted for equilibrium from perioperative blood loss in the hip and knee replacement groups, while, for reasons of data availability, equilibrium time in the spine fusion groups was two days.