Johnson 1992.

Methods	Randomised controlled trial
Participants	39 autologous blood donors undergoing elective myocardial revascularisation were randomized to 1 of 2 groups: Liberal group: n = 18; M/F = 16/2; mean (SD) age = 60.5 (6.9) years Restrictive group: n = 20; M = 20; mean (SD) age = 58.2 (7.5) years
Interventions	The liberal group received blood to achieve a Hct value of 32%. The restrictive (conservative) group received transfusions for a Hct value less than 25%. NB: operative management included sequestration of 1 or more units of fresh autologous blood in participants with a Hct value greater than 35% who were haemodynamically stable after anaesthetic induction. Red cell conservation was practised through the salvage of oxygenator contents and reinfusion of postoperatively shed mediastinal blood. On the 5th postoperative day, all participants were asked to complete an exercise treadmill test. A second test was performed the following day.
Outcomes	Cardiac events, complications, postoperative blood loss, blood use (total units), allogeneic blood use (units), autologous blood use (units), all product blood use (units), number of participants receiving transfusions, mean cardiac index, mean systemic resistance, exercise capacity, Hct levels, length of ICU stay, length of hospital stay
Notes	‐
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	High risk	A table of random numbers and an odd‐even designation randomized participants.
Allocation concealment (selection bias)	Unclear risk	It was unclear if assignment was concealed prior to randomisation.
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Surgeons and anaesthesiologists were blinded as to the group of randomisation until the participant reached the intensive care unit (ICU).
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Blinding of outcome assessment was not addressed.
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	A small number of exclusions were reported.
Selective reporting (reporting bias)	Low risk	No reporting bias was apparent.
Other bias	Low risk	No other biases were apparent.