Methods	A prospective, randomised, clinical trial was undertaken to determine whether postoperative blood salvage in patients undergoing total hip or knee arthroplasty decreased the need for transfusion with banked blood. The groups included all patients undergoing hip or knee arthroplasty at the University of Arizona Medical Centre between August 1, 1988 and June 1, 1989. Method of randomisation and allocation concealment were not described
Participants	109 patients undergoing total hip or knee arthroplasty were randomly assigned to one of two groups: Group 1 (Autotransfusion group): n=51 Group 2 (Control group): n=51 NB: The average age of the patients was 70 years; the study and the control groups were evenly matched for age and sex. Demographic data for each study group were not reported. Of the 109 patients who entered the study seven were excluded
Interventions	Group 1: Autotransfusion group (Hemolite cell saver) had the wound drainage tubes connected in the operating room to a sterile reservoir which contained 200mls of a heparin saline solution. The reservoir was connected to wall suction (120mmHg) in the operating suite. Collection was continued in the post-anaesthetic care unit and the surgical ward. At the end of the 4 hour period the collected wound drainage was processed. Under sterile conditions the blood was washed with 2 litres of saline and processed in the Hemolite cell washer to remove heparin, cellular debris, platelets and clotting factors. After processing, the wound drainage consisted of concentrated red blood cells suspended in saline. This product was transfused back to the patient. After wound drainage collection was completed, the drains were attached to standard Hemovac suction and output was monitored at 8 hour intervals until the drains were disconnected at 48 hours. Group 2: Control group received a Hemovac standard drainage system. At the termination of the operative procedure the control group underwent Hemovac wound drainage only. Autotransfusion was not performed.
Outcomes	Outcomes reported: amount of blood collected by the cell saver, amount of blood retransfused from the cell saver, amount of allogeneic blood transfused, number of patients transfused allogeneic blood, adverse events, coagulopathy, blood loss, costs
Notes	Transfusion threshold: transfusion protocol was not reported. Patients who were transfused only one unit of blood received only pre-banked autologous blood
Risk of bias
Bias	Authors’ judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Method used to generate allocation sequences was not described
Allocation concealment (selection bias)	Unclear risk	Method used to conceal treatment allocation was unclear.
Blinding (performance bias and detection bias) All outcomes	High risk