Nine trials were included (1,364 patients). Few quality assessment results were reported. Two studies were reported as being double-blind. Withdrawal rates ranged from 0% (in four studies) to 17%. The authors reported that there were no significant baseline differences between groups. Follow-up ranged from one to six weeks (six trials) or was not reported (three trials).
There were no significant differences between groups in rate of fever (RR 0.39, 95% CI 0.07 to 2.36; Ι²=60%; four studies), incidence of positive urine culture (RR 0.77, 95% CI 0.54 to 1.11; Ι²=30%; nine comparisons, eight of which were for short-term results) and incidence of urinary tract infection (RR 0.54, 95% CI 0.29 to 1.01; Ι²=0%; seven comparisons, six of which reported short-term results).
For positive urine culture, a subgroup analysis of two studies with some patients with a preoperative ureteral catheter yielded no significant differences. Two of the three studies that reported symptoms found no statistically significant differences and one reported a statistically significant reduction with antibiotics in incidence of urinary frequency and nocturia.
There was no evidence of publication bias.