Perrin 1997.

Methods	RCT or quasi‐RCT: RCT Setting: Australia
Participants	Population: patients undergoing rectal surgery Inclusion criteria:patients undergoing rectal surgery Exclusion criteria: If the rectum was not mobilised; if a urethral catheter was required as part of the operation technique; already having an indwelling catheter; infected urine at time of operation Age ("average age"): A: 62 years; B: 64 years Number of participants: Eligible: 137 Randomised: 108 Reported: 108 Dropouts (n of participants + reason): 28 patients were excluded: 17 according to exclusion criteria, 6 because of bacteriuria at baseline, 4 because of no initial sample had been taken and 1 because the SPC could not been inserted. Comment: 1) In the SPC group there were 49 participants, in the UC group 59 participants; 2) 137 participants reported randomised, but 28 participants were excluded and 108 patients were analysed (= 136 participants)
Interventions	A (n = 59): Indwelling urethral catheterisation (16‐French; Foley) inserted following induction of anaesthesia B (n = 49): Suprapubic catheterisation (16‐French; Foley) inserted after the opening of the abdomen Duration of intervention: Until the 5th postoperative day or when the medical condition did not necessitate continuing the monitoring of urine output. The suprapubic catheters were clamped prior to removal to ensure satisfactory voiding (residual volume of < 100 ml). For the UC the residual volume of urine was only measured when voiding did not occur after 6 hours or the participant became uncomfortable
Outcomes	Primary outcome (symptomatic UTI): not reported Bacteriuria: Overall A:29/59; B:12/49 In subgroups: a)In males A:12/28; B:7/24 b)In females A:17/31; B:5/25 Definition of bacteriuria: Significant bacteriuria was defined as greater than 10 000 cfu/ml. Specimens were aspirated from the drainage tube at the time of catheterisation, if clinically indicated and immediately prior to removal. Recatheterisation: A: 4/59; B: 3/49 Number of participants catheterised > 5 days: Overall: A: 33/59; B: 44/49 In subgroups: a) men A: 15/28; B: 23/24 b) women A: 18/31; B: 21/25 Discomfort: Overall: A: 17/59; B: 6/49 In subgroups: a) men A: 9/28; B: 3/24 b) women A: 8/31; B: 3/25 Haemorrhage A: 0/59; B: 0/49 Blockage A: 0/59; B: 2/49
Sponsorship/Funding	Not reported
Notes	Duration of catheterisation: Prolonged in the suprapubic group (see outcomes) No definition was given for participants' discomfort All participants received a single dose of tinidazole and/or ticarcillin
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"Patients were randomly allocated (by computer generated random numbers" ‐ random component in sequence generation process
Allocation concealment (selection bias)	Unclear risk	No information given
Blinding of participants	High risk	No information given but as suprapubic vs urethral can assume participants not blinded
Blinding of personnel	High risk	No information given but as both catheters inserted during surgery can assume clinicians were not blinded
Blinding of microbiological outcome assessment	Low risk	Low risk of bacteriuria being affected by bias as based on culture
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	"All patients entered into the study were questioned specifically by the treating doctor in regard to catheter discomfort." ‐ no information given on blinding
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	"An additional 11 patients were excluded (6 had infected urine at time of operation, 4 had no initial sample taken, and in 1 patient a suprapubic catheter was unable to be inserted)" ‐ does not give reason for other 17 participants excluded "28 patients of total 137 who were initially entered in this trial were excluded. Of the remaining 108 patients..." 137 minus 28 equals 109, not 108
Selective reporting (reporting bias)	Unclear risk	Outcomes not detailed in Methods section but mentions data that will be collected. Unable to access protocol, so some uncertainty about selective reporting
Other bias	Low risk	Appears to be free of other sources of bias