Saleh 2017.

Study characteristics
Methods	Study design: parallel RCT Study duration: December 2011 to November 2016 Duration of follow‐up: 45 days
Participants	Country: Egypt, Saudi Arabia Setting: multicentre (3 sites) Patients on regular HD using long‐term catheters with suspected CRBSI Number: catheter exchange group (339) catheter removal group (339) Mean age ± SD (years): catheter exchange group (56.7 ± 9); catheter removal group (61.2 ± 10) Sex (female): catheter exchange group/catheter removal group (207/239) Diabetes: catheter exchange group/catheter removal group (237/236) Hypertension: catheter exchange group/catheter removal group (288/278) Coronary artery disease: catheter exchange group/catheter removal group (149/132) Congestive heart failure: catheter exchange group/catheter removal group (27/34) Exclusion criteria: septic shock (CRBSI + haemodynamic instability); metastatic infection; sources of infection other than CRBSI; without catheter replacement after 10 days (because of AVF maturation or persistent fever or death); negative catheter tip culture result
Interventions	Catheter exchange group Catheter replacement over a guidewire 2 or 3 days after initiating systemic antibiotic therapy Catheter removal group Catheter removal and new catheter insertion when negative blood cultures were obtained (approximately 3 to 7 days after the removal) HD was performed through a short‐term femoral catheter after primary catheter removal Both groups Empiric systemic antibiotics (vancomycin and gentamicin) during dialysis through the catheter for 3 weeks (tailored according to culture if needed)
Outcomes	Catheter infection‐free survival time Cure: asymptomatic for at least 45 days after completion of the antibiotic course in addition to negative blood culture Treatment failure: any bacteraemia involving the original organism within 45 days after antibiotic completion in addition to clinical manifestation Indetermined results: unrelated death, catheter removal for unrelated reason or infection with a different organism
Notes	Funding source: not reported Contact with study authors: twice, no response received
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Prospective randomisation was conducted by sealed envelope randomisation service
Allocation concealment (selection bias)	Low risk	The randomisation service also provided allocation concealment
Blinding of participants and personnel (performance bias) All outcomes	High risk	It was impossible to blind the personnel and patients
Blinding of outcome assessment (detection bias) All outcomes	Low risk	The published trial protocol reported single masking (care provider)
Incomplete outcome data (attrition bias) All outcomes	High risk	Primary outcomes were not fully reported
Selective reporting (reporting bias)	High risk	The trial protocol (NCT03054714) was published retrospectively. Additionally, did not mention excluded or withdrawn patients, deaths, treatment failures and absolute numbers for cure
Other bias	High risk	Among the exclusion criteria were patients in whom the catheter was not replaced within 10 days because of death. This exclusion criterion could be concealing an important adverse event for this specific intervention

AVF ‐ arteriovenous fistula; CRB ‐ catheter‐related bacteraemia; CRBSI ‐ catheter‐related bloodstream infection; HD ‐ haemodialysis; RCT ‐ randomised controlled trial; SD ‐ standard deviation; TPA ‐ tissue plasminogen activator