Introduction

Approximately 20-40% of patients after radical cystectomy (RC) suffer from perioperative urinary tract infection (UTI). Guidelines for antimicrobial prophylaxis, such as the AUA Best Practice Statement, are hampered by a lack of RC-specific research and generally based on colorectal surgery literature.

Methods

We analyzed RC registries between 2009-2015 from three different urological centers._x000D_ Baseline variables included i. a. ASA score, TNM-classification and type of urinary diversion. We recorded the presence of urologic devices before RC, neo-adjuvant chemotherapy, previous radiotherapy, prolonged steroids therapy before RC, microbiological urine and blood cultures. UTI was diagnosed according to EAU/ESIU definitions._x000D_ Primary objective was to analyze the type and length of antibiotic therapy (AB), the percentage and severity of UTIs after RC, the responsible bacteria and their AB sensitivity. We recorded the frequency of CT-scans and invasive procedures after RC and 30-days-readmission rate. _x000D_

Results

We analyzed 217 cases of RC. Median length of hospital stay was 13 days (IQR 11-20.5) with 9.2% of the patients still in hospital after 30 days. 30-days-readmission rate was 16.1%. The most frequent AB was a combination of metronidazole (98.2%) with a cephalosporin (89.9%). The median of days of AB administration after RC was 7 days (IQR 5-14). After cessation of the first AB therapy, additional antibiotics were used in 51.6% of the patients. _x000D_ The overall number of UTIs was 42 (19.4%): 9.7% pyelonephritis, 7.8% urosepsis, 1.8% uroseptic shock. Enterococcus spp. was the most frequently isolated bacteria in urine (25.7%) and in blood (42.9%). Enterococcus showed, as awaited, a cephalosporin (100%) and gentamicine (90%) resistance but also a 100% resistance to ciprofloxacin. _x000D_ In univariate logistic regression analysis, age (odds ratio (OR)=0.94; p=0.001), continent derivation (OR=4.36; p<0.001), neoadjuvant chemotherapy (OR=3.07; p=0.044) and the presence of any urologic device before surgery (OR=0.36; p=0.013) were correlated with UTI. In multivariate logistic regression analysis only continent derivation was associated to UTI after surgery (OR=3.16; p=0.010)._x000D_

Conclusions

The UTI rate after RC was 19.4% and in 9.6% of the cases UTI was urosepsis. Continent diversion was the only independent factor associated with a higher risk of UTI but perioperative AB therapy length was not. Enterococci spp. are involved in early infection and not routinely covered by the most common used AB prophylaxes.

Funding

none