Gulielmos 2000 (B).

Study characteristics
Methods	Prospective, randomised trial Run‐in period: not specified Received: 14 December 1999 Received in revised form: 14 July 2000 Accepted: 15 August 2000 Number of study centres and locations: single centre, Cardiovascular Institute, University Hospital Dresden, Fetscherstrasse 76, 01307 Dresden, Germany
Participants	Twelve participants in part B of this study Four surgical techniques were compared: group 1, median sternotomy with CPB in 10 patients; group 2, median sternotomy and off‐pump in 10 patients; group 3, minithoracotomy with CPB in 10 patients; group 4, minithoracotomy and off‐pump in 10 patients. Mean age: 60.8 years Sex (female/male ratio): 30% No high‐risk patients Inclusion criteria: single LIMA bypass to the LAD, because of CAD or patients suffering from coronary artery double‐vessel or multivessel disease, with only the LAD being amenable for surgery Exclusion criteria: impaired LVEF (< 30%), with impaired lung and renal function, unstable angina, major calcification of the ascending aorta, and obesity (BMI ≥30)
Interventions	Intervention group: minithoracotomy with CPB. The LIMA was harvested via a minithoracotomy. The right atrium was cannulated transfemoral percutaneously for venous drainage for CPB. The ascending aorta was cannulated and dissected from the pulmonary trunk on pump. A conventional aortic clamp was used for external cross‐clamping, and cardioplegia was applied via the ascending aorta. During cardioplegic arrest, the relaxed heart was rotated for exposure of the LAD. Anastomosis of the LIMA to the LAD was performed. Control group: median sternotomy with CPB
Outcomes	IL‐1, IL‐6, and myocardial enzyme (troponin T and CK‐MB); clinical data (major complications, operative time, postoperative ventilation, CPB time [if used], cross‐clamping time, and coronary occlusion time) Venous blood samples were drawn at the following times: at 5 minutes before induction of anaesthesia; on induction of CPB or occlusion of the LAD; at 15 minutes after induction of CPB or occlusion of the LAD; at 1 hour after induction of CPB or occlusion of the LAD; at 2 hours after induction of CPB or occlusion of the LAD; at 6 hours after induction of CPB or occlusion of the LAD; at 12 hours after induction of CPB or occlusion of the LAD; at 24 hours after induction of CPB or occlusion of the LAD; and at 48 hours after induction of CPB or occlusion of the LAD.
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Between the four groups, there were no differences in terms of age, BMI, or incidence of previous infarction. No randomisation method description. One out of four techniques was prospectively chosen for each patient at random, but no other description about how the allocation was made is given.
Allocation concealment (selection bias)	Unclear risk	No clear description of allocation concealment
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Unspecified and unlikely because of the nature of the intervention
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Not specified
Incomplete outcome data (attrition bias) All outcomes	Low risk	All procedures were completed as they were planned. 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)	Unclear risk	Clinical data were not reported properly as the other outcomes.
Other bias	Low risk	No funding was disclosed.