Harig 1999 (B).

Study characteristics
Methods	Prospective, randomised study Run‐in period: not specified Registration date: not available Number of study centres and locations: unspecified, authors from Erlangen, Germany
Participants	Thirteen participants were included in part B of the analysis, including the participants from group B. Forty patients with CAD undergoing elective CABG were prospectively randomised to four groups of 10 – Group A: prednisolone preoperatively and postoperatively (2 × 250 mg); Group B: aprotinin perioperatively (6 million KIU); Group C: heparin‐coated circuits ('Bioline' by Jostra); Group D: no special measures were taken (controls). Mean age: 62 years Sex (female/male ratio): 66% No high‐risk patients (CABG) Inclusion criteria: normal LV function (i.e. LVEDP < 14 mmHg and EF > 55); all underwent elective CABG. Exclusion criteria: renal or hepatic insufficiency and known anaphylaxis to aprotinin
Interventions	Group A: prednisolone 250 mg pre‐ and postoperatively Group B: aprotinin as follows – a loading dose of 2,000,000 KIU over a 15‐minute period at the start of the operation was followed by a continuous infusion of 500,000 KIU/h throughout the operation. Another 2,000,000 KIU were added to the prime volume of the heart‐lung machine. Group C: heparin‐coated circuits modified by the Bioline procedure Group D: the control group, only standard anaesthetic techniques and operative procedures
Outcomes	Cytokine release (IL‐6, IL‐8, and IL‐10) In each group, blood samples were taken from the radial artery at the following times: after induction of anaesthesia and before the infusion of aprotinin or prednisolone preoperatively and 30 minutes after the start of CPB, at the end of CPB, and 30 minutes, 2 hours, 4 hours, 6 hours, 8 hours, and 12 hours after the end of CPB.
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	No randomisation method description is given; however, no significant differences were noted between the four groups: "Cohorts of 10 patients were randomized independently".
Allocation concealment (selection bias)	Unclear risk	No allocation concealment information
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Not specified; likely for participants, unlikely for study personnel
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Not specified, unlikely
Incomplete outcome data (attrition bias) All outcomes	Low risk	No trial group changes, no withdrawals, and no losses to follow‐up were reported, but no intention‐to‐treat analysis. Data from all the participants were included in the final analysis.
Selective reporting (reporting bias)	Low risk	All expected outcomes were reported properly.
Other bias	Unclear risk	No funding was disclosed, although no supporting statement was provided.