Objectives: We aimed to evaluate the effectiveness of postoperative autotransfusion method on prevention of the need of allogeneic blood transfusion in hip and knee arthroplasty.Methods: Seventy-four patients who underwent 77 hip and knee arthroplasty operations were randomized into control and study groups, and evaluated prospectively. In the knee group (39 patients; 30 females, 9 males; mean age 66.6 years), cemented, cruciate retaining, and bicompartmental arthroplasty was performed under tourniquet control; whereas in the hip group (35 patients; 24 females, 11 males; mean age 59.3 years) cementless arthroplasty with posterolateral approach was performed. None of the patients received preoperative and intraoperative allogeneic blood transfusion. The collected blood in the surgical area was transfused with autotransfusion system to the patients in the study groups at the end of the fourth hour postoperatively. The mean amounts of autotransfused blood in hip and knee groups were 413 mL and 480 mL, respectively. Allogeneic blood transfusion was applied to the patients with hemoglobin level below 8 g/dL, hematocrit level below 25%, and clinical symptoms of anemia.Results: Preoperative and postoperative hemoglobin-hematocrit levels did not differ significantly between study and control groups. Allogeneic blood transfusion was applied to one patient(5%) in study and 8 patients (38%) in control groups during knee arthroplasty (p=0.01); whereas 9 patients (53%) in study and 15 patients (79%) in control groups received allogeneic blood transfusion during hip arthroplasty (p=0.044). The amount of allogeneic blood transfusion in study groups was significantly lower than that in control groups (p=0.008 for knee arthroplasty, p=0.048 for hip arthroplasty).Conclusion: The need and amount of allogeneic transfusion were reduced with postoperative autotransfusion in both knee and hip arthroplasty groups with greater extent in knee arthroplasty.