O'Kelly 1995.

Methods	RCT or quasi‐RCT: RCT Setting: UK
Participants	Population: patients undergoing abdominal surgery Inclusion criteria: patients undergoing abdominal surgery and a full‐length abdominal incision Comment: It is not clear whether patients with initial positive urine cultures were excluded Exclusion criteria: not reported Age (median, range): A: 65 (42 ‐ 81) years; B: 68 (35 ‐ 79) years Number of participants: Eligible: 68 Randomised: 62 Reported: 57 Dropouts (n of participants + reasons): 5 participants:2 died shortly after operation and in 3 participants no catheter was inserted. Authors did not report from which group participants were lost
Interventions	A (n = 29): Indwelling urethral catheterisation (14 French) before operation B (n = 28): Suprapubic catheterisation (14 French) after the abdomen was opened Duration of intervention: Catheters were removed when this was appropriate on clinical grounds, but it was recommended that suprapubic catheters should remain in place until the 5th postoperative day. Before removal the SPC was clamped and was withdrawn when the participant could micturate spontaneously and the residual urine was < 100 ml
Outcomes	Primary outcome (symptomatic UTI): not reported Bacteriuria: A: 3/29; B: 3/28 Definition of bacteriuria: Significant bacteriuria was defined as > 105 cfu/ml. Specimens were daily aspirated from the catheter and a final midstream was obtained 2 days after the catheter was removed Recatheterisation: A: 2/29; B: 1/28 Duration of catheterisation (median): A: 4 days (range 2 ‐ 11); B: 5 days (range 4 ‐ 10) Number of participants with pain: A: 13/29; B: 2/28 Number of catheter days with pain: A: 37/126; B: 6/142
Sponsorship/Funding	Not reported
Notes	Each participant kept a daily record of catheter‐related pain, scoring discomfort on a standard VAS (no pain = 0; worst possible pain = 10) Not reported whether prophylactic antibiotics were used
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	"62 patients were randomized into two groups" ‐ no information given on method of randomisation
Allocation concealment (selection bias)	Unclear risk	No method of allocation concealment given
Blinding of participants	High risk	No information given but as suprapubic vs urethral can assume blinding of participants did not occur
Blinding of personnel	High risk	"All catheters were inserted by the surgeon in the operating theatre" ‐ based on this statement, can assume surgeons not blinded "patients were reviewed each day by one of the authors (T.J.O.K. or A. Mathew)" ‐ based on this statement, can assume authors not blinded
Blinding of microbiological outcome assessment	Low risk	Symptomatic UTI assessed by microbiologist who would not know the route of catheter the participant had inserted
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	No information given
Incomplete outcome data (attrition bias) All outcomes	Low risk	"Two patients died shortly after operation and in three cases a catheter was not inserted" ‐ reason unlikely to be related to true outcome
Selective reporting (reporting bias)	Low risk	Methods gives some details of what data will be measured, for which there are results. Unable to access protocol, so some uncertainty about selective reporting
Other bias	Low risk	Appears to be free from other sources of bias