Aims: The purpose of this study was to assess whether auto Ðtransfusion of drainage blood in Total Knee Replacement is cost effective. Methods: This was a prospective study and the subjects were elective patients who underwent primary total knee arthroplasty between Dec 1999- Dec 2000. The subjects were randomly selected and identiþed from the theatre lists. Blood haemoglobin levels were obtained from the hospital pathology system. All the patients received an astra bellovac auto transfusion system re infusion drain. There was a protocol established for using the auto- transfusion system. Results: 27 patients were identiþed from which two had to be excluded and from the remaining 25, there were 10 men and 15 women. The mean post operative haemoglobin in the study was 12.3g/dl (m) and 10.8g/dl (f). 23 patients had blood auto-transfused and two did not have blood auto transfused due to problems with the transfusing blood. An average of 315 ml was re transfused. 22 patients did not require any allogenous blood transfusion post operatively. The requirement for homologous blood when using auto transfusion system was 0.12 unit, while an internal audit carried by the hospital haematology department found the requirement for homologous blood when standard drains were used to be 0.87 unit. The amount saved when re infusion drain used was £35.04 per patient which represents a 38% reduction in transfusion costs. As per the Hospital Episodes Statistics the average number of total knee replacements performed in the NHS (UK) during 2000/01 was 30,000. The cost savings to the National Health Service would be £1,051,200/yr if autotransfusion of drained blood was used. Conclusions: Auto transfusion of drainage blood in primary total knee arthroplasty is safe and cost effective. It signiþcantly reduces the requirement for homologous blood transfusion.