Maravić‐Stojković 2011.

Methods	Randomised, single‐centre, open‐label, controlled clinical trial at a 200‐bed academic tertiary care hospital in Belgrade, Serbia
Participants	Inclusion criteria: People scheduled to undergo open heart surgery on cardio‐pulmonal bypass. We assessed people who were selected for elective cardiac surgery at the 200‐bed academic tertiary care hospital. The criterion for inclusion in the study was the type of operation: coronary artery bypass grafting (CABG) surgery, valve reconstruction, combined CABG and valve procedures. Entry criteria included stable and unstable angina pectoris, valve insufficiency, left ventricle ejection fraction (LVEF) above 30%, and epidemiological status with saprophyte bacteria. Exclusion criteria: People selected for redo cardiac operations, thoracic aortic surgery, as well as people having active endocarditis and people with LVEF < 30%. People with preoperative signs of infection (leukocyte count > 12,000/L; body temperature > 38°C) were also excluded. Included in this study: 205/205 included participants finished for follow‐up (102 PCT group/103 standard group).
Interventions	Guiding antibiotic decisions in patients after cardiac surgery Algorithm used in this study: Antibiotic prophylaxis was performed in all participants. The participants were divided at the time of surgery into the standard group and the PCT group. In the standard group, the antibiotic use was applied according to the criteria based on the laboratory and clinical signs; no antibiotic therapy was administrated routinely in the absence of clinical signs of infection or a bacteriologic positive sample. In the PCT group, the use of antibiotics was encouraged or discouraged on the basis of serum PCT concentrations. A serum PCT concentration of 0.5 ng/mL or less indicated the absence of bacterial infection, at which point the use of antibiotics was discouraged.
Outcomes	proportion of participants treated with antibiotics overall cost of antibiotics per 1 patient total cost of antibiotics per 1 hospital day after operation ICU stays and hospital stay rehospitalisation incidence of infections severe non‐infection complications mortality rate
Notes	Funding: No funding is mentioned in the main article. Follow‐up: 1 year Trial registration: No trial registration found.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer‐generated randomisation scheme
Allocation concealment (selection bias)	Low risk	The participants were divided at the time of surgery into the standard group and the PCT group by centralised randomisation.
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	No blinding of physicians due to the study design
Blinding of outcome assessment (detection bias) All outcomes	High risk	No blinding of outcome assessment mentioned in the study.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Follow‐up for mortality: 205/205 (100%)
Selective reporting (reporting bias)	Unclear risk	No trial registration found.
Other bias	Unclear risk	No information about adherence