- Email: [email protected]
Urinary Tract Infections After Urinary Diversion—Different Occurrence Patterns in Patients With Ileal Conduit and Orthotopic Neobladder
Accepted Manuscript Title: Urinary Tract Infections after Urinary Diversion – Different Occurrence Patterns in Patients with Ileal Conduit and Orthotopic Neobladder Author: Roy Mano, Hanan Goldberg, Yariv Stabholz, Danny Hazan, David Margel, Daniel Kedar, Jack Baniel, Ofer Yossepowitch PII: DOI: Reference:
S0090-4295(18)30299-1 https://doi.org/10.1016/j.urology.2018.03.042 URL 20983
To appear in:
Urology
Received date: Accepted date:
18-1-2018 22-3-2018
Please cite this article as: Roy Mano, Hanan Goldberg, Yariv Stabholz, Danny Hazan, David Margel, Daniel Kedar, Jack Baniel, Ofer Yossepowitch, Urinary Tract Infections after Urinary Diversion – Different Occurrence Patterns in Patients with Ileal Conduit and Orthotopic Neobladder, Urology (2018), https://doi.org/10.1016/j.urology.2018.03.042. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Urinary Tract Infections after Urinary Diversion – Different Occurrence Patterns in Patients with Ileal Conduit and Orthotopic Neobladder
Roy Manoa,c, Hanan Goldberga,c, Yariv Stabholza,c, Danny Hazana,c, David Margela,c, Daniel Kedara,c, Jack Baniela,c, Ofer Yossepowitchb,c a
Department of Urology, Rabin Medical Center, Beilinson Campus, Petah-Tikva, Israel
b
Department of Urology, Tel Aviv Sourasky Medical Center, Tel Aviv, Israel
c
Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel
Correspondence: Roy Mano, M.D. Department of Urology Rabin Medical Center Petach Tikva 4941492, Israel Fax: 972-3-937-6569 Tel: 972-3-937-6553 Email: [email protected]
Word count: Abstract – 231; Text – 2,534
Keywords: Urinary diversion; radical cystectomy; ileal conduit; orthotopic neobladder; surgical complications; urinary tract infection.
3 Page 1 of 23
Supporting Funds None.
Financial Disclosures All authors have nothing to disclose.
Abstract
Objectives To compare the incidence rate of urinary tract infections and associated pathogens between patients with an ileal conduit and an orthotopic neobladder urinary diversion.
Methods The medical records of 179 patients treated with radical cystectomy between 2006– 2011 were reviewed and data pertaining to postoperative UTI was collected. UTI
4 Page 2 of 23
incidence was reported at 3 months’ intervals and compared by diversion type. Preoperative predictors of UTI were evaluated with Cox regression analyses.
Results The study cohort included 130 ileal conduit and 49 orthotopic neobladder patients. Patients with a neobladder were younger (p<0.001). Median follow-up was 38 months (IQR 11-63). Median time from surgery to first infection was 1.5 months (IQR, 1–12.5) for patients with a neobladder and 11 months (IQR, 2.5–27) for patients with a conduit (p=0.04). During the first 3 months after surgery 29% of neobladder patients and 8% of ileal conduit patients had a UTI episode (p=0.001). Rates of UTI did not differ during subsequent follow-up. Diversion type was not associated with UTI on multivariable analysis. E. Coli was the most common pathogen in conduit patients (58%), and Klebsiella spp. in neobladder patients (29%).
Conclusion The risk of UTI is significantly higher in patients with a neobladder during the first 3 months after surgery, and comparable to ileal conduit during subsequent follow-up. These findings may facilitate preoperative counseling regarding the expected risk of UTI after urinary diversion.
5 Page 3 of 23
Introduction
Bladder cancer is the fourth most common cancer in men in the United States, estimated to account for 4% of cancer related death during 2017.1 Approximately 20% of patients have muscle invasive disease at presentation,2 and are generally treated by radical cystectomy (RC) and urinary diversion with or without the addition of systemic chemotherapy. The most common diversion types are ileal conduit (IC) and orthotopic neobladder (ONB).3
RC with urinary diversion is a complex and morbid procedure with high postoperative complication rates ranging from 49% to 78% using standard methodologies for reporting of complications.4-12 Infectious complications, particularly urinary tract infections (UTIs), are the most common.4-15 ONB provides patients with excellent functional outcomes translating into favorable quality of life, but this may often come at
6 Page 4 of 23
the expense of a higher rate of UTI.5, 7, 9, 14 We recently reported on the specific characteristics of UTI in patients with ONBs: almost half of the patients had at least one episode of symptomatic UTI during the first year of follow-up, most commonly within the first 3 months after surgery. Common pathogens included P. Aeruginosa and E. Coli.16 While few reports compared infectious complications in patients with IC and ONB, changes in UTI incidence over time and associated pathogens in both diversion types have not been well delineated. We sought to characterize the incidence of UTI and associated pathogens in patients undergoing RC and compare those with IC and ONB at different time intervals after surgery.
Patients and Methods
After obtaining local ethics committee approval, we queried our institutional RC database and retrieved the medical records of 179 consecutive patients who underwent radical cystoprostatectomy or anterior pelvic exenteration and pelvic lymph node dissection with urinary diversion by either IC or ONB between 2006 and 2011. The latter was performed using ileal or ileocolonic bowel segments. Patients with compromised renal function, inability to perform clean intermittent catheterization, severe urethral stricture and tumors involving the prostatic urethra in men or bladder neck in women were advised to undergo IC urinary diversion. Otherwise, the type of diversion was based on the patient’s preference following a comprehensive consultation with his/her treating urologist. The procedures were performed as previously described.17-19 7 Page 5 of 23
Perioperative management was standardized per diversion type. Preoperative urine cultures were obtained for all patients, and if positive, treatment was administered preoperatively according to bacterial strain and susceptibility. Bowel preparation with polyethyleneglycol solution and oral neomycin and metronidazole was performed in all patients scheduled for ONB reconstruction. Patients scheduled to undergo IC urinary diversion did not receive bowel preparation. All patients were treated with intravenous penicillin (replaced by clindamycin for documented sensitivity), an aminoglycoside (dose tailored by weight and renal function) and metronidazole one hour prior to surgery. Postoperatively, an intravenous second-generation cephalosporin was given to all patients for 72 hours. An additional dose of intravenous cephalosporin was given prior to removing the ureteral stents, after which the patients were discharged, usually at post-operative day 10. Those with ONB were given additional daily administration of oral low-dose cephalosporin or quinolone, and trained to perform self neobladder irrigation for as long as the catheter was retained (a total of 3 weeks in general). Following catheter removal, patients were instructed to perform timed voiding (every 2-3 hours) in a sitting position assisted by the use of Valsalva or Crede maneuver to facilitate complete emptying of their neobladder and the antibiotics were discontinued. For those with incomplete emptying (residual urine detected by ultrasound at routine office visit), self-clean intermittent catheterization was initiated. None of the patients with IC were given routine antibiotics after discharge.
8 Page 6 of 23
Follow-up was conducted every 3 months during the first postoperative year and semiannually thereafter. Urine cultures and antimicrobial susceptibility assays were obtained (clean catch midstream urine samples from patients with ONB and via stoma catheterization from those with IC) during each follow-up visit and upon clinical symptoms consistent with symptomatic UTI, defined as a positive urine culture in the presence of fever (temperature ≥ 38° C), with or without associated flank or abdominal pain. All patients with febrile UTI were admitted and treated with intravenous antibiotics. Patients with asymptomatic bacteriuria were not treated.
Clinical and pathological data were compared between diversion types. Data pertaining to febrile UTI after surgery were collected including the time from surgery to first infection, causative pathogens and their antimicrobial susceptibility. The primary study endpoint was the incidence of symptomatic UTI during follow-up.
Data were reported as median and interquartile range for continuous variables and number and percentage for categorical variables, and compared by diversion type using the Mann-Whitney U and Fisher exact tests, respectively. The number of UTI events divided by the number of patients available for follow-up were reported at 3 months’ intervals and compared by diversion type. Estimated probability of UTI was calculated using the Kaplan Meier method, and predictors of UTI evaluated using univariable and multivariable Cox regression models. All statistical analyses were 2sided. A P value of less than 0.05 was considered statistically significant. IBM SPSS 9 Page 7 of 23
Statistics for Windows, Version 20.0. (Armonk, NY: IBM Corp.) was used for all data analyses.
Results
The study cohort consisted of 130 IC and 49 ONB patients at a median age of 69 years (IQR, 61-76). Table 1 describes patient and tumor characteristics stratified by diversion type. Patients who underwent ONB urinary diversion were younger (median age 60 years vs. 72 years for patients with an IC, p<0.001). Additional patient and tumor characteristics did not differ significantly between the two groups.
Median follow-up for all patients was 38 months (IQR, 12-62). Follow-up length was longer for patients with an ONB (median 45 months; IQR, 29-61) compared to patients with an IC (median 31 months; IQR, 10-65, p=0.04). Median time from surgery to the first episode of UTI was 1.5 months (IQR, 1–13) for ONB patients and 11 months (IQR, 3–27) for IC patients, (p=0.04, log-rank test). During the first 3 months after surgery 14/48 (29%) of the patients with ONB had at least one episode of UTI compared to 10/120 (8%) of the IC patients, p=0.001. However, rates of infection did not differ significantly during the subsequent follow-up, ranging from 0%-7% for both 10 Page 8 of 23
diversion types (p≥0.331 across all time intervals, Figure 1). Among patients with at least one episode of UTI, 10/22 with ONB (45%) and 13/37 with IC (35%) had recurrent infections. Figure 1 compares the total number of infections per 100 patients at 3months intervals during the first 3 years after surgery for patients with IC and ONB urinary diversion.
Table 2 reports univariable and multivariable analyses of predictors of symptomatic UTI. On univariable analysis, patients with ONB were more likely to develop an infection compared to those with IC (HR=1.71, 95% CI 1.01-2.89, p=0.047). However, on multivariable analysis adjusting for age, Charlson comorbidity index score and presence of diabetes mellitus, diversion type was not significantly associated with the occurrence of UTI.
Figure 2 details the common pathogens involved in symptomatic UTI episodes stratified by diversion type. While E. Coli was the most common causative pathogen in patients with IC (58%), ONB patients had comparable rates of K. Pneumonia (29%), E. Coli (24%) and P. Aeruginosa (22%). Less common pathogens and their corresponding incidence were Enterobacter spp. (2% of IC infections and 5% of ONB), Staph aureus coagulase+ (0% of IC and 5% of ONB), Citrobacter (2% of IC and 0% of ONB), Morganella (2% of IC and 0% of ONB), Proteus (2% of IC and 0% of ONB), Acinetobacter (0% of IC and 2% of ONB), and Serratia marcescens (0% of IC and 2%
11 Page 9 of 23
of ONB). Multiple pathogens were isolated in 2/41 infectious episodes (5%) in patients with ONB and 6/52 episodes (12%) in patients with IC.
Antibiotic susceptibility data were available for 97/102 (96%) pathogens. For all antibiotics tested, susceptibility rates did not differ by diversion type. The highest susceptibility was to the drug Amikacin (81% for IC and 79% for ONB), followed by Gentamycin (74% for IC and 67% for ONB), Ceftazidime (64% for IC and 64% for ONB), Ceftriaxone (55% for IC and 38% for ONB), Ciprofloxacin (50% for IC and 38% for ONB), Trimethoprim Sulfamethoxazole (38% for IC and 41% for ONB), and Nitrofurantoin (52% for IC and 36% for ONB).
Comment
In the current study, we compared the incidence of symptomatic UTI in 130 patients with IC and 49 patients with ONB following RC for bladder cancer. Patients with an ONB had a shorter median time from surgery to the first episode of UTI (1.5 in ONB vs 11 months in IC, p=0.04) and a higher overall 90-days UTI rate (29% vs 8%, p=0.001). However, following the initial 3 months, rates of infection did not differ significantly between diversion types. E. Coli was the predominant causative pathogen in IC patients, accounting for 58% of cases, whereas in patients with ONB, K.
12 Page 10 of 23
Pneumonia, and P. Aeruginosa were as likely to be isolated. Antibiotic susceptibility did not differ by diversion type, the highest rate (80%) being achieved with Amikacin.
Complication rates following RC and urinary diversion were initially thought to be relatively low affecting 12%-22% of IC patients and 12%-21% of ONB patients.20, 21 Recent studies, however, using a more rigorous methodology for reporting of complications have challenged these numbers, revealing much higher early complication rates of 48%-72% for IC patients and 42%-76% for ONB patients, the latter likely representing the true incidence.6-12 Infectious complications are among the most common complications encountered after RC and urinary diversion, with most studies indicating a higher incidence in those with a continent urinary diversion.7, 9-11, 14 Early infection rates have been reported to range from 13%-31% for IC compared to 11%43% for ONB.7-15 While most investigators agree that early infection rates are substantially higher among patients with ONB compared to IC, the long-term variance between the two is less clear. In the current study, 29% of the patients with ONB had at least one episode of symptomatic UTI within the first 3 months from surgery compared to 8% in those with IC (p=0.001). However, during the subsequent follow-up, annual rates of UTI declined markedly in the former (<7%) rendering the difference between the two diversion types non-significant.
The decreased rate of UTI observed in patients with ONB overtime is consistent with previous reports,5, 13, 14, 16 however its etiology remains unknown. Functional and 13 Page 11 of 23
histologic changes that occur in the neobladder as it matures have been postulated to play a role in this process. Using serial urodynamic examinations in patients with neobladders, a significant increase in maximal flow rate and bladder capacity concomitant to a decrease in residual urine volume were documented over a 6 months interval from surgery, possibly contributing to the decrease in UTI events.22 This low residual volume remained unchanged or further decreased in other studies with longer follow-up,23, 24 explaining the sustained decrease in UTI episodes over time. In addition, atrophic changes in the intestinal mucosa lining the neobladder occurring 3-6 months after surgery may render the neobladder less susceptible to bacterial colonization and infection.25, 26 Finally, changes in expression patterns of pathogen recognition receptors or innate immune activation may contribute to the decline in the incidence of UTI with time after surgery.13
Few studies evaluated risk factors for infectious complications after urinary diversion. Abe et al. reported older age and continent diversion type (HR 1.97, 95% CI 1.35-2.88, p=0.0005) as independent risk factors for infection.10 Similarly, Parker et al. identified continent urinary diversion (OR 2.17; 95% CI 1.36-2.41; P=0.001), diabetes, perioperative blood transfusion and development of a urinary leak as risk factors for UTI.14 Contrary to these findings, Nazmy et al. failed to show an association between ONB diversion and UTI (HR= 2.56, 95% CI 0.71-9.15, p=0.1).9 Thulin et al. evaluated a large cohort of patients with ONB, continent reservoirs and non-continent urostomies and reported the presence of diabetes mellitus as a predictor of symptomatic UTI. 27 In a recent study, Clifford et al. evaluated 1,133 patients who underwent RC and urinary 14 Page 12 of 23
diversion, 72% of whom had an ONB. The 90-days postoperative UTI rate was 11%. The type of urinary diversion (IC or ONB) was not associated with the risk of UTI (OR= 1.13, 95% CI 0.7-1.79, p=1.106). Patients with a Charlson comorbidity index >2 were at a significantly higher risk of UTI (p=0.045).15 In the current study, the type of urinary diversion did not retain its role as an independent predictor of UTI after adjusting for age, presence of diabetes mellitus and comorbidity status, suggesting that poor patients characteristics (likely driving the choice to pursue an IC) and not the actual pathophysiological diversion-associated changes are the primary determinants of postoperative UTI. Different pathogens were identified as the cause of infection in patients with IC and ONB. The prolonged presence of a urinary catheter in patients with an ONB may alter the pathogens involved in UTI events. Initially S. epidermidis, E. Coli or Enterococcus faecalis are involved in catheter related infections; however, at longer indwelling times other species appear including P. aeruginosa and K. pneumonia, as can be seen in the current cohort.28 In addition, while omitting the use of pre-operative bowel preparation does not alter the rate of peri-operative infection,29 its selective use before ONB diversion may have modified the bacterial flora and subsequent infectious pathogens.30 Nevertheless, in both diversion types, bacteria were most susceptible to Amikacin, representing the susceptibility patterns at our institute. Thus, we use Amikacin as the initial empirical treatment for cases of symptomatic UTI in patients with urinary diversion when possible based on the patient’s renal function.
15 Page 13 of 23
The study limitations include the retrospective study design with an inherit selection bias. Additionally, in lack of a prospective protocol for culture collection, not all infectious events may have been captured within our database, thus the actual UTI rate may be higher than reported. Furthermore, the relatively small cohort size may have limited the significance of the findings on the multivariable analysis evaluating preoperative predictors of UTI. The use of neoadjuvant and adjuvant chemotherapy was not standardized in the current study possibly affecting the rates of infection. Similarly, we did not document the number of patient who required prolonged CIC, which may be a major contributing factor for UTI in patients with a ONB. Finally, common infectious pathogens and antibiotic susceptibilities described in the current study may not be generalizable as they are related to the local hospital bacterial flora. Nevertheless, the current study adds to previous publications by reporting the rates of UTI beyond the first 90-days from RC and may assist when counseling patients regarding the long-term risk of UTI after RC and urinary diversion.
Conclusions
The risk of febrile UTI during the initial 3 months after surgery is significantly higher in patients with an ONB compared to IC urinary diversion. UTI events during subsequent follow-up are comparable between IC and ONB. Common causative pathogens differ between the two diversion groups. These findings may facilitate preoperative counseling regarding the expected risk and pattern of UTI episodes after
16 Page 14 of 23
urinary diversion. Furthermore, considering the current findings, maintaining a low postvoid residual volume and using bladder irrigation as required in the initial 3 months after surgery may be warranted in patients with an ONB to lower the rate of UTI, and should be evaluated in future studies.
References 1.
Siegel RL, Miller KD, Jemal A. Cancer Statistics, 2017. CA Cancer J Clin. 2017;67:7-30.
2.
Nielsen ME, Smith AB, Meyer AM, et al. Trends in stage-specific incidence rates for urothelial carcinoma of the bladder in the United States: 1988 to 2006. Cancer. 2014;120:86-95.
3.
Lee RK, Abol-Enein H, Artibani W, et al. Urinary diversion after radical cystectomy for bladder cancer: options, patient selection, and outcomes. BJU Int. 2014;113:11-23.
4.
Novara G, De Marco V, Aragona M, et al. Complications and mortality after radical cystectomy for bladder transitional cell cancer. J Urol. 2009;182:914-921.
17 Page 15 of 23
5.
Shabsigh A, Korets R, Vora KC, et al. Defining early morbidity of radical cystectomy for patients with bladder cancer using a standardized reporting methodology. Eur Urol. 2009;55:164-174.
6.
van Hemelrijck M, Thorstenson A, Smith P, Adolfsson J, Akre O. Risk of inhospital complications after radical cystectomy for urinary bladder carcinoma: population-based follow-up study of 7608 patients. BJU Int. 2013;112:1113-1120.
7.
Nieuwenhuijzen JA, de Vries RR, Bex A, et al. Urinary diversions after cystectomy: the association of clinical factors, complications and functional results of four different diversions. Eur Urol. 2008;53:834-842; discussion 842834.
8.
Erber B, Schrader M, Miller K, et al. Morbidity and Quality of Life in Bladder Cancer Patients following Cystectomy and Urinary Diversion: A Single-Institution Comparison of Ileal Conduit versus Orthotopic Neobladder. ISRN urology. 2012;2012:342796.
9.
Nazmy M, Yuh B, Kawachi M, et al. Early and late complications of robotassisted radical cystectomy: a standardized analysis by urinary diversion type. J Urol. 2014;191:681-687.
10.
Abe T, Takada N, Shinohara N, et al. Comparison of 90-day complications between ileal conduit and neobladder reconstruction after radical cystectomy: a retrospective multi-institutional study in Japan. Int J Urol. 2014;21:554-559.
18 Page 16 of 23
11.
Roghmann F, Trinh QD, Braun K, et al. Standardized assessment of complications in a contemporary series of European patients undergoing radical cystectomy. Int J Urol. 2014;21:143-149.
12.
Antonelli A, Belotti S, Cristinelli L, De Luca V, Simeone C. Comparison of Perioperative Morbidity of Radical Cystectomy With Neobladder Versus Ileal Conduit: A Matched Pair Analysis of 170 Patients. Clinical genitourinary cancer. 2016;14:244-248.
13.
Kim KH, Yoon HS, Yoon H, Chung WS, Sim BS, Lee DH. Febrile Urinary Tract Infection after Radical Cystectomy and Ileal Neobladder in Patients with Bladder Cancer. Journal of Korean medical science. 2016;31:1100-1104.
14.
Parker WP, Toussi A, Tollefson MK, et al. Risk Factors and Microbial Distribution of Urinary Tract Infections Following Radical Cystectomy. Urology. 2016;94:96101.
15.
Clifford TG, Katebian B, Van Horn CM, et al. Urinary tract infections following radical cystectomy and urinary diversion: a review of 1133 patients. World journal of urology. 2018.
16.
Mano R, Baniel J, Goldberg H, Stabholz Y, Kedar D, Yossepowitch O. Urinary tract infections in patients with orthotopic neobladder. Urol Oncol. 2014;32:50 e59-14.
17.
Studer UE, Varol C, Danuser H. Orthotopic ileal neobladder. BJU Int. 2004;93:183-193.
18.
Baniel J, Tal R. The "B-bladder"-an ileocolonic neobladder with a chimney: surgical technique and long-term results. Eur Urol. 2004;45:794-798.
19 Page 17 of 23
19.
Colombo R, Naspro R. Ileal conduit as the standard for urinary diversion after radical cystectomy for bladder cancer. European Urology Supplements. 2010;9:736-744.
20.
Gburek BM, Lieber MM, Blute ML. Comparison of studer ileal neobladder and ileal conduit urinary diversion with respect to perioperative outcome and late complications. J Urol. 1998;160:721-723.
21.
Parekh DJ, Gilbert WB, Koch MO, Smith JA, Jr. Continent urinary reconstruction versus ileal conduit: a contemporary single-institution comparison of perioperative morbidity and mortality. Urology. 2000;55:852-855.
22.
Wang D, Li LJ, Liu J, Qiu MX. Long-term urodynamic evaluation of laparoscopic radical cystectomy with orthotopic ileal neobladder for bladder cancer. Oncol Lett. 2014;8:1031-1034.
23.
Burkhard FC, Kessler TM, Springer J, Studer UE. Early and late urodynamic assessment of ileal orthotopic bladder substitutes combined with an afferent tubular segment. J Urol. 2006;175:2155-2160; discussion 2160-2151.
24.
Ferriero M, Simone G, Rocchegiani A, et al. Early and late urodynamic assessment of Padua ileal bladder. Urology. 2009;73:1357-1362.
25.
Di Tonno F, Siracusano S, Ciciliato S, Visalli F, Lampropoulou N, Lavelli D. Morphological changes on the intestinal mucosa in orthotopic neobladder. Urol Int. 2012;89:67-70.
26.
Senkul T, Yildirim S, Iseri C, Karademir K, Erden D, Baykal K. Histopathologic changes in the mucosa of ileal orthotopic neobladder--findings in 24 patients followed up for 5 years. Scand J Urol Nephrol. 2003;37:202-204.
20 Page 18 of 23
27.
Thulin H, Steineck G, Kreicbergs U, et al. Hygiene and urinary tract infections after cystectomy in 452 Swedish survivors of bladder cancer. BJU Int. 2010;105:1107-1117.
28.
Stickler DJ. Bacterial biofilms in patients with indwelling urinary catheters. Nature clinical practice. Urology. 2008;5:598-608.
29.
Large MC, Kiriluk KJ, DeCastro GJ, et al. The impact of mechanical bowel preparation on postoperative complications for patients undergoing cystectomy and urinary diversion. J Urol. 2012;188:1801-1805.
30.
Drago L, Toscano M, De Grandi R, Casini V, Pace F. Persisting changes of intestinal microbiota after bowel lavage and colonoscopy. European journal of gastroenterology & hepatology. 2016;28:532-537.
Legends to Figures Figure 1 – Rates of urinary tract infection during the first 3 years after surgery compared by diversion type Figure 2 – Common pathogens (>10%) cultured during urinary tract infections compared by diversion type
21 Page 19 of 23
Table 1 – Patient and tumor characteristics of the study cohort (n=179) stratified by diversion type IC (n=130) 72 (65, 78)
ONB (n=49) 60 (53, 65)
p-value <0.001
Median age in years (IQR) Sex Male 112 (86%) 43 (88%) 1 Female 18 (14%) 6 (12%) CCI ≤3 87 (67%) 38 (78%) 0.2 >3 43 (33%) 11 (22%) Presence of DM Yes 34 (26%) 8 (16%) 0.24 No 96 (74%) 41 (84%) Positive preoperative culture Yes 18 (14%) 5 (10%) 0.62 No 112 (86%) 44 (90%) Pathological T-stage T0 20 (15%) 11 (22%) 0.28 Ta/is 15 (12%) 1 (2%) T1 19 (15%) 8 (16%) T2 20 (15%) 10 (20%) T3 34 (26%) 15 (31%) T4 22 (17%) 4 (8%) Lymph node involvement Negative 95 (73%) 41 (84%) 0.17 Positive 35 (27%) 8 (16%) IC = ileal conduit; ONB = orthotopic neobladder; IQR = interquartile range; CCI = Charlson comorbidity index; DM = diabetes mellitus
3 Page 20 of 23
Table 2 - Cox regression analyses for the association between pre-operative characteristics and the probability of acquiring a urinary tract infection Variable Age Sex Male Female Diabetes Mellitus Absent Present CCI ≤3 >3 Diversion Ileal Conduit Neobladder
Univariable analysis HR (95% CI) p-value 0.98 (0.96-1.01) 0.119
Multivariable analysis HR (95% CI) p-value 0.99 (0.96-1.02) 0.455
Ref. 0.93 (0.44-1.96)
0.851
Ref. 1.46 (0.82-2.59)
0.202
Ref. 1.27 (0.65-2.47)
0.487
Ref. 1.49 (0.86-2.6)
0.157
Ref. 1.51 (0.79-2.89)
0.212
Ref. 1.71 (1.01-2.89)
0.047
Ref. 1.59 (0.84-3.03)
0.152
HR = Hazards ratio; CI = Confidence interval; CCI = Charlson comorbidity index
3 Page 21 of 23
Figure 1 01-16-18.tif
4 Page 22 of 23
Figure 2 01-16-18.tif
5 Page 23 of 23
S0090-4295(18)30299-1 https://doi.org/10.1016/j.urology.2018.03.042 URL 20983
To appear in:
Urology
Received date: Accepted date:
18-1-2018 22-3-2018
Please cite this article as: Roy Mano, Hanan Goldberg, Yariv Stabholz, Danny Hazan, David Margel, Daniel Kedar, Jack Baniel, Ofer Yossepowitch, Urinary Tract Infections after Urinary Diversion – Different Occurrence Patterns in Patients with Ileal Conduit and Orthotopic Neobladder, Urology (2018), https://doi.org/10.1016/j.urology.2018.03.042. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Urinary Tract Infections after Urinary Diversion – Different Occurrence Patterns in Patients with Ileal Conduit and Orthotopic Neobladder
Roy Manoa,c, Hanan Goldberga,c, Yariv Stabholza,c, Danny Hazana,c, David Margela,c, Daniel Kedara,c, Jack Baniela,c, Ofer Yossepowitchb,c a
Department of Urology, Rabin Medical Center, Beilinson Campus, Petah-Tikva, Israel
b
Department of Urology, Tel Aviv Sourasky Medical Center, Tel Aviv, Israel
c
Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel
Correspondence: Roy Mano, M.D. Department of Urology Rabin Medical Center Petach Tikva 4941492, Israel Fax: 972-3-937-6569 Tel: 972-3-937-6553 Email: [email protected]
Word count: Abstract – 231; Text – 2,534
Keywords: Urinary diversion; radical cystectomy; ileal conduit; orthotopic neobladder; surgical complications; urinary tract infection.
3 Page 1 of 23
Supporting Funds None.
Financial Disclosures All authors have nothing to disclose.
Abstract
Objectives To compare the incidence rate of urinary tract infections and associated pathogens between patients with an ileal conduit and an orthotopic neobladder urinary diversion.
Methods The medical records of 179 patients treated with radical cystectomy between 2006– 2011 were reviewed and data pertaining to postoperative UTI was collected. UTI
4 Page 2 of 23
incidence was reported at 3 months’ intervals and compared by diversion type. Preoperative predictors of UTI were evaluated with Cox regression analyses.
Results The study cohort included 130 ileal conduit and 49 orthotopic neobladder patients. Patients with a neobladder were younger (p<0.001). Median follow-up was 38 months (IQR 11-63). Median time from surgery to first infection was 1.5 months (IQR, 1–12.5) for patients with a neobladder and 11 months (IQR, 2.5–27) for patients with a conduit (p=0.04). During the first 3 months after surgery 29% of neobladder patients and 8% of ileal conduit patients had a UTI episode (p=0.001). Rates of UTI did not differ during subsequent follow-up. Diversion type was not associated with UTI on multivariable analysis. E. Coli was the most common pathogen in conduit patients (58%), and Klebsiella spp. in neobladder patients (29%).
Conclusion The risk of UTI is significantly higher in patients with a neobladder during the first 3 months after surgery, and comparable to ileal conduit during subsequent follow-up. These findings may facilitate preoperative counseling regarding the expected risk of UTI after urinary diversion.
5 Page 3 of 23
Introduction
Bladder cancer is the fourth most common cancer in men in the United States, estimated to account for 4% of cancer related death during 2017.1 Approximately 20% of patients have muscle invasive disease at presentation,2 and are generally treated by radical cystectomy (RC) and urinary diversion with or without the addition of systemic chemotherapy. The most common diversion types are ileal conduit (IC) and orthotopic neobladder (ONB).3
RC with urinary diversion is a complex and morbid procedure with high postoperative complication rates ranging from 49% to 78% using standard methodologies for reporting of complications.4-12 Infectious complications, particularly urinary tract infections (UTIs), are the most common.4-15 ONB provides patients with excellent functional outcomes translating into favorable quality of life, but this may often come at
6 Page 4 of 23
the expense of a higher rate of UTI.5, 7, 9, 14 We recently reported on the specific characteristics of UTI in patients with ONBs: almost half of the patients had at least one episode of symptomatic UTI during the first year of follow-up, most commonly within the first 3 months after surgery. Common pathogens included P. Aeruginosa and E. Coli.16 While few reports compared infectious complications in patients with IC and ONB, changes in UTI incidence over time and associated pathogens in both diversion types have not been well delineated. We sought to characterize the incidence of UTI and associated pathogens in patients undergoing RC and compare those with IC and ONB at different time intervals after surgery.
Patients and Methods
After obtaining local ethics committee approval, we queried our institutional RC database and retrieved the medical records of 179 consecutive patients who underwent radical cystoprostatectomy or anterior pelvic exenteration and pelvic lymph node dissection with urinary diversion by either IC or ONB between 2006 and 2011. The latter was performed using ileal or ileocolonic bowel segments. Patients with compromised renal function, inability to perform clean intermittent catheterization, severe urethral stricture and tumors involving the prostatic urethra in men or bladder neck in women were advised to undergo IC urinary diversion. Otherwise, the type of diversion was based on the patient’s preference following a comprehensive consultation with his/her treating urologist. The procedures were performed as previously described.17-19 7 Page 5 of 23
Perioperative management was standardized per diversion type. Preoperative urine cultures were obtained for all patients, and if positive, treatment was administered preoperatively according to bacterial strain and susceptibility. Bowel preparation with polyethyleneglycol solution and oral neomycin and metronidazole was performed in all patients scheduled for ONB reconstruction. Patients scheduled to undergo IC urinary diversion did not receive bowel preparation. All patients were treated with intravenous penicillin (replaced by clindamycin for documented sensitivity), an aminoglycoside (dose tailored by weight and renal function) and metronidazole one hour prior to surgery. Postoperatively, an intravenous second-generation cephalosporin was given to all patients for 72 hours. An additional dose of intravenous cephalosporin was given prior to removing the ureteral stents, after which the patients were discharged, usually at post-operative day 10. Those with ONB were given additional daily administration of oral low-dose cephalosporin or quinolone, and trained to perform self neobladder irrigation for as long as the catheter was retained (a total of 3 weeks in general). Following catheter removal, patients were instructed to perform timed voiding (every 2-3 hours) in a sitting position assisted by the use of Valsalva or Crede maneuver to facilitate complete emptying of their neobladder and the antibiotics were discontinued. For those with incomplete emptying (residual urine detected by ultrasound at routine office visit), self-clean intermittent catheterization was initiated. None of the patients with IC were given routine antibiotics after discharge.
8 Page 6 of 23
Follow-up was conducted every 3 months during the first postoperative year and semiannually thereafter. Urine cultures and antimicrobial susceptibility assays were obtained (clean catch midstream urine samples from patients with ONB and via stoma catheterization from those with IC) during each follow-up visit and upon clinical symptoms consistent with symptomatic UTI, defined as a positive urine culture in the presence of fever (temperature ≥ 38° C), with or without associated flank or abdominal pain. All patients with febrile UTI were admitted and treated with intravenous antibiotics. Patients with asymptomatic bacteriuria were not treated.
Clinical and pathological data were compared between diversion types. Data pertaining to febrile UTI after surgery were collected including the time from surgery to first infection, causative pathogens and their antimicrobial susceptibility. The primary study endpoint was the incidence of symptomatic UTI during follow-up.
Data were reported as median and interquartile range for continuous variables and number and percentage for categorical variables, and compared by diversion type using the Mann-Whitney U and Fisher exact tests, respectively. The number of UTI events divided by the number of patients available for follow-up were reported at 3 months’ intervals and compared by diversion type. Estimated probability of UTI was calculated using the Kaplan Meier method, and predictors of UTI evaluated using univariable and multivariable Cox regression models. All statistical analyses were 2sided. A P value of less than 0.05 was considered statistically significant. IBM SPSS 9 Page 7 of 23
Statistics for Windows, Version 20.0. (Armonk, NY: IBM Corp.) was used for all data analyses.
Results
The study cohort consisted of 130 IC and 49 ONB patients at a median age of 69 years (IQR, 61-76). Table 1 describes patient and tumor characteristics stratified by diversion type. Patients who underwent ONB urinary diversion were younger (median age 60 years vs. 72 years for patients with an IC, p<0.001). Additional patient and tumor characteristics did not differ significantly between the two groups.
Median follow-up for all patients was 38 months (IQR, 12-62). Follow-up length was longer for patients with an ONB (median 45 months; IQR, 29-61) compared to patients with an IC (median 31 months; IQR, 10-65, p=0.04). Median time from surgery to the first episode of UTI was 1.5 months (IQR, 1–13) for ONB patients and 11 months (IQR, 3–27) for IC patients, (p=0.04, log-rank test). During the first 3 months after surgery 14/48 (29%) of the patients with ONB had at least one episode of UTI compared to 10/120 (8%) of the IC patients, p=0.001. However, rates of infection did not differ significantly during the subsequent follow-up, ranging from 0%-7% for both 10 Page 8 of 23
diversion types (p≥0.331 across all time intervals, Figure 1). Among patients with at least one episode of UTI, 10/22 with ONB (45%) and 13/37 with IC (35%) had recurrent infections. Figure 1 compares the total number of infections per 100 patients at 3months intervals during the first 3 years after surgery for patients with IC and ONB urinary diversion.
Table 2 reports univariable and multivariable analyses of predictors of symptomatic UTI. On univariable analysis, patients with ONB were more likely to develop an infection compared to those with IC (HR=1.71, 95% CI 1.01-2.89, p=0.047). However, on multivariable analysis adjusting for age, Charlson comorbidity index score and presence of diabetes mellitus, diversion type was not significantly associated with the occurrence of UTI.
Figure 2 details the common pathogens involved in symptomatic UTI episodes stratified by diversion type. While E. Coli was the most common causative pathogen in patients with IC (58%), ONB patients had comparable rates of K. Pneumonia (29%), E. Coli (24%) and P. Aeruginosa (22%). Less common pathogens and their corresponding incidence were Enterobacter spp. (2% of IC infections and 5% of ONB), Staph aureus coagulase+ (0% of IC and 5% of ONB), Citrobacter (2% of IC and 0% of ONB), Morganella (2% of IC and 0% of ONB), Proteus (2% of IC and 0% of ONB), Acinetobacter (0% of IC and 2% of ONB), and Serratia marcescens (0% of IC and 2%
11 Page 9 of 23
of ONB). Multiple pathogens were isolated in 2/41 infectious episodes (5%) in patients with ONB and 6/52 episodes (12%) in patients with IC.
Antibiotic susceptibility data were available for 97/102 (96%) pathogens. For all antibiotics tested, susceptibility rates did not differ by diversion type. The highest susceptibility was to the drug Amikacin (81% for IC and 79% for ONB), followed by Gentamycin (74% for IC and 67% for ONB), Ceftazidime (64% for IC and 64% for ONB), Ceftriaxone (55% for IC and 38% for ONB), Ciprofloxacin (50% for IC and 38% for ONB), Trimethoprim Sulfamethoxazole (38% for IC and 41% for ONB), and Nitrofurantoin (52% for IC and 36% for ONB).
Comment
In the current study, we compared the incidence of symptomatic UTI in 130 patients with IC and 49 patients with ONB following RC for bladder cancer. Patients with an ONB had a shorter median time from surgery to the first episode of UTI (1.5 in ONB vs 11 months in IC, p=0.04) and a higher overall 90-days UTI rate (29% vs 8%, p=0.001). However, following the initial 3 months, rates of infection did not differ significantly between diversion types. E. Coli was the predominant causative pathogen in IC patients, accounting for 58% of cases, whereas in patients with ONB, K.
12 Page 10 of 23
Pneumonia, and P. Aeruginosa were as likely to be isolated. Antibiotic susceptibility did not differ by diversion type, the highest rate (80%) being achieved with Amikacin.
Complication rates following RC and urinary diversion were initially thought to be relatively low affecting 12%-22% of IC patients and 12%-21% of ONB patients.20, 21 Recent studies, however, using a more rigorous methodology for reporting of complications have challenged these numbers, revealing much higher early complication rates of 48%-72% for IC patients and 42%-76% for ONB patients, the latter likely representing the true incidence.6-12 Infectious complications are among the most common complications encountered after RC and urinary diversion, with most studies indicating a higher incidence in those with a continent urinary diversion.7, 9-11, 14 Early infection rates have been reported to range from 13%-31% for IC compared to 11%43% for ONB.7-15 While most investigators agree that early infection rates are substantially higher among patients with ONB compared to IC, the long-term variance between the two is less clear. In the current study, 29% of the patients with ONB had at least one episode of symptomatic UTI within the first 3 months from surgery compared to 8% in those with IC (p=0.001). However, during the subsequent follow-up, annual rates of UTI declined markedly in the former (<7%) rendering the difference between the two diversion types non-significant.
The decreased rate of UTI observed in patients with ONB overtime is consistent with previous reports,5, 13, 14, 16 however its etiology remains unknown. Functional and 13 Page 11 of 23
histologic changes that occur in the neobladder as it matures have been postulated to play a role in this process. Using serial urodynamic examinations in patients with neobladders, a significant increase in maximal flow rate and bladder capacity concomitant to a decrease in residual urine volume were documented over a 6 months interval from surgery, possibly contributing to the decrease in UTI events.22 This low residual volume remained unchanged or further decreased in other studies with longer follow-up,23, 24 explaining the sustained decrease in UTI episodes over time. In addition, atrophic changes in the intestinal mucosa lining the neobladder occurring 3-6 months after surgery may render the neobladder less susceptible to bacterial colonization and infection.25, 26 Finally, changes in expression patterns of pathogen recognition receptors or innate immune activation may contribute to the decline in the incidence of UTI with time after surgery.13
Few studies evaluated risk factors for infectious complications after urinary diversion. Abe et al. reported older age and continent diversion type (HR 1.97, 95% CI 1.35-2.88, p=0.0005) as independent risk factors for infection.10 Similarly, Parker et al. identified continent urinary diversion (OR 2.17; 95% CI 1.36-2.41; P=0.001), diabetes, perioperative blood transfusion and development of a urinary leak as risk factors for UTI.14 Contrary to these findings, Nazmy et al. failed to show an association between ONB diversion and UTI (HR= 2.56, 95% CI 0.71-9.15, p=0.1).9 Thulin et al. evaluated a large cohort of patients with ONB, continent reservoirs and non-continent urostomies and reported the presence of diabetes mellitus as a predictor of symptomatic UTI. 27 In a recent study, Clifford et al. evaluated 1,133 patients who underwent RC and urinary 14 Page 12 of 23
diversion, 72% of whom had an ONB. The 90-days postoperative UTI rate was 11%. The type of urinary diversion (IC or ONB) was not associated with the risk of UTI (OR= 1.13, 95% CI 0.7-1.79, p=1.106). Patients with a Charlson comorbidity index >2 were at a significantly higher risk of UTI (p=0.045).15 In the current study, the type of urinary diversion did not retain its role as an independent predictor of UTI after adjusting for age, presence of diabetes mellitus and comorbidity status, suggesting that poor patients characteristics (likely driving the choice to pursue an IC) and not the actual pathophysiological diversion-associated changes are the primary determinants of postoperative UTI. Different pathogens were identified as the cause of infection in patients with IC and ONB. The prolonged presence of a urinary catheter in patients with an ONB may alter the pathogens involved in UTI events. Initially S. epidermidis, E. Coli or Enterococcus faecalis are involved in catheter related infections; however, at longer indwelling times other species appear including P. aeruginosa and K. pneumonia, as can be seen in the current cohort.28 In addition, while omitting the use of pre-operative bowel preparation does not alter the rate of peri-operative infection,29 its selective use before ONB diversion may have modified the bacterial flora and subsequent infectious pathogens.30 Nevertheless, in both diversion types, bacteria were most susceptible to Amikacin, representing the susceptibility patterns at our institute. Thus, we use Amikacin as the initial empirical treatment for cases of symptomatic UTI in patients with urinary diversion when possible based on the patient’s renal function.
15 Page 13 of 23
The study limitations include the retrospective study design with an inherit selection bias. Additionally, in lack of a prospective protocol for culture collection, not all infectious events may have been captured within our database, thus the actual UTI rate may be higher than reported. Furthermore, the relatively small cohort size may have limited the significance of the findings on the multivariable analysis evaluating preoperative predictors of UTI. The use of neoadjuvant and adjuvant chemotherapy was not standardized in the current study possibly affecting the rates of infection. Similarly, we did not document the number of patient who required prolonged CIC, which may be a major contributing factor for UTI in patients with a ONB. Finally, common infectious pathogens and antibiotic susceptibilities described in the current study may not be generalizable as they are related to the local hospital bacterial flora. Nevertheless, the current study adds to previous publications by reporting the rates of UTI beyond the first 90-days from RC and may assist when counseling patients regarding the long-term risk of UTI after RC and urinary diversion.
Conclusions
The risk of febrile UTI during the initial 3 months after surgery is significantly higher in patients with an ONB compared to IC urinary diversion. UTI events during subsequent follow-up are comparable between IC and ONB. Common causative pathogens differ between the two diversion groups. These findings may facilitate preoperative counseling regarding the expected risk and pattern of UTI episodes after
16 Page 14 of 23
urinary diversion. Furthermore, considering the current findings, maintaining a low postvoid residual volume and using bladder irrigation as required in the initial 3 months after surgery may be warranted in patients with an ONB to lower the rate of UTI, and should be evaluated in future studies.
References 1.
Siegel RL, Miller KD, Jemal A. Cancer Statistics, 2017. CA Cancer J Clin. 2017;67:7-30.
2.
Nielsen ME, Smith AB, Meyer AM, et al. Trends in stage-specific incidence rates for urothelial carcinoma of the bladder in the United States: 1988 to 2006. Cancer. 2014;120:86-95.
3.
Lee RK, Abol-Enein H, Artibani W, et al. Urinary diversion after radical cystectomy for bladder cancer: options, patient selection, and outcomes. BJU Int. 2014;113:11-23.
4.
Novara G, De Marco V, Aragona M, et al. Complications and mortality after radical cystectomy for bladder transitional cell cancer. J Urol. 2009;182:914-921.
17 Page 15 of 23
5.
Shabsigh A, Korets R, Vora KC, et al. Defining early morbidity of radical cystectomy for patients with bladder cancer using a standardized reporting methodology. Eur Urol. 2009;55:164-174.
6.
van Hemelrijck M, Thorstenson A, Smith P, Adolfsson J, Akre O. Risk of inhospital complications after radical cystectomy for urinary bladder carcinoma: population-based follow-up study of 7608 patients. BJU Int. 2013;112:1113-1120.
7.
Nieuwenhuijzen JA, de Vries RR, Bex A, et al. Urinary diversions after cystectomy: the association of clinical factors, complications and functional results of four different diversions. Eur Urol. 2008;53:834-842; discussion 842834.
8.
Erber B, Schrader M, Miller K, et al. Morbidity and Quality of Life in Bladder Cancer Patients following Cystectomy and Urinary Diversion: A Single-Institution Comparison of Ileal Conduit versus Orthotopic Neobladder. ISRN urology. 2012;2012:342796.
9.
Nazmy M, Yuh B, Kawachi M, et al. Early and late complications of robotassisted radical cystectomy: a standardized analysis by urinary diversion type. J Urol. 2014;191:681-687.
10.
Abe T, Takada N, Shinohara N, et al. Comparison of 90-day complications between ileal conduit and neobladder reconstruction after radical cystectomy: a retrospective multi-institutional study in Japan. Int J Urol. 2014;21:554-559.
18 Page 16 of 23
11.
Roghmann F, Trinh QD, Braun K, et al. Standardized assessment of complications in a contemporary series of European patients undergoing radical cystectomy. Int J Urol. 2014;21:143-149.
12.
Antonelli A, Belotti S, Cristinelli L, De Luca V, Simeone C. Comparison of Perioperative Morbidity of Radical Cystectomy With Neobladder Versus Ileal Conduit: A Matched Pair Analysis of 170 Patients. Clinical genitourinary cancer. 2016;14:244-248.
13.
Kim KH, Yoon HS, Yoon H, Chung WS, Sim BS, Lee DH. Febrile Urinary Tract Infection after Radical Cystectomy and Ileal Neobladder in Patients with Bladder Cancer. Journal of Korean medical science. 2016;31:1100-1104.
14.
Parker WP, Toussi A, Tollefson MK, et al. Risk Factors and Microbial Distribution of Urinary Tract Infections Following Radical Cystectomy. Urology. 2016;94:96101.
15.
Clifford TG, Katebian B, Van Horn CM, et al. Urinary tract infections following radical cystectomy and urinary diversion: a review of 1133 patients. World journal of urology. 2018.
16.
Mano R, Baniel J, Goldberg H, Stabholz Y, Kedar D, Yossepowitch O. Urinary tract infections in patients with orthotopic neobladder. Urol Oncol. 2014;32:50 e59-14.
17.
Studer UE, Varol C, Danuser H. Orthotopic ileal neobladder. BJU Int. 2004;93:183-193.
18.
Baniel J, Tal R. The "B-bladder"-an ileocolonic neobladder with a chimney: surgical technique and long-term results. Eur Urol. 2004;45:794-798.
19 Page 17 of 23
19.
Colombo R, Naspro R. Ileal conduit as the standard for urinary diversion after radical cystectomy for bladder cancer. European Urology Supplements. 2010;9:736-744.
20.
Gburek BM, Lieber MM, Blute ML. Comparison of studer ileal neobladder and ileal conduit urinary diversion with respect to perioperative outcome and late complications. J Urol. 1998;160:721-723.
21.
Parekh DJ, Gilbert WB, Koch MO, Smith JA, Jr. Continent urinary reconstruction versus ileal conduit: a contemporary single-institution comparison of perioperative morbidity and mortality. Urology. 2000;55:852-855.
22.
Wang D, Li LJ, Liu J, Qiu MX. Long-term urodynamic evaluation of laparoscopic radical cystectomy with orthotopic ileal neobladder for bladder cancer. Oncol Lett. 2014;8:1031-1034.
23.
Burkhard FC, Kessler TM, Springer J, Studer UE. Early and late urodynamic assessment of ileal orthotopic bladder substitutes combined with an afferent tubular segment. J Urol. 2006;175:2155-2160; discussion 2160-2151.
24.
Ferriero M, Simone G, Rocchegiani A, et al. Early and late urodynamic assessment of Padua ileal bladder. Urology. 2009;73:1357-1362.
25.
Di Tonno F, Siracusano S, Ciciliato S, Visalli F, Lampropoulou N, Lavelli D. Morphological changes on the intestinal mucosa in orthotopic neobladder. Urol Int. 2012;89:67-70.
26.
Senkul T, Yildirim S, Iseri C, Karademir K, Erden D, Baykal K. Histopathologic changes in the mucosa of ileal orthotopic neobladder--findings in 24 patients followed up for 5 years. Scand J Urol Nephrol. 2003;37:202-204.
20 Page 18 of 23
27.
Thulin H, Steineck G, Kreicbergs U, et al. Hygiene and urinary tract infections after cystectomy in 452 Swedish survivors of bladder cancer. BJU Int. 2010;105:1107-1117.
28.
Stickler DJ. Bacterial biofilms in patients with indwelling urinary catheters. Nature clinical practice. Urology. 2008;5:598-608.
29.
Large MC, Kiriluk KJ, DeCastro GJ, et al. The impact of mechanical bowel preparation on postoperative complications for patients undergoing cystectomy and urinary diversion. J Urol. 2012;188:1801-1805.
30.
Drago L, Toscano M, De Grandi R, Casini V, Pace F. Persisting changes of intestinal microbiota after bowel lavage and colonoscopy. European journal of gastroenterology & hepatology. 2016;28:532-537.
Legends to Figures Figure 1 – Rates of urinary tract infection during the first 3 years after surgery compared by diversion type Figure 2 – Common pathogens (>10%) cultured during urinary tract infections compared by diversion type
21 Page 19 of 23
Table 1 – Patient and tumor characteristics of the study cohort (n=179) stratified by diversion type IC (n=130) 72 (65, 78)
ONB (n=49) 60 (53, 65)
p-value <0.001
Median age in years (IQR) Sex Male 112 (86%) 43 (88%) 1 Female 18 (14%) 6 (12%) CCI ≤3 87 (67%) 38 (78%) 0.2 >3 43 (33%) 11 (22%) Presence of DM Yes 34 (26%) 8 (16%) 0.24 No 96 (74%) 41 (84%) Positive preoperative culture Yes 18 (14%) 5 (10%) 0.62 No 112 (86%) 44 (90%) Pathological T-stage T0 20 (15%) 11 (22%) 0.28 Ta/is 15 (12%) 1 (2%) T1 19 (15%) 8 (16%) T2 20 (15%) 10 (20%) T3 34 (26%) 15 (31%) T4 22 (17%) 4 (8%) Lymph node involvement Negative 95 (73%) 41 (84%) 0.17 Positive 35 (27%) 8 (16%) IC = ileal conduit; ONB = orthotopic neobladder; IQR = interquartile range; CCI = Charlson comorbidity index; DM = diabetes mellitus
3 Page 20 of 23
Table 2 - Cox regression analyses for the association between pre-operative characteristics and the probability of acquiring a urinary tract infection Variable Age Sex Male Female Diabetes Mellitus Absent Present CCI ≤3 >3 Diversion Ileal Conduit Neobladder
Univariable analysis HR (95% CI) p-value 0.98 (0.96-1.01) 0.119
Multivariable analysis HR (95% CI) p-value 0.99 (0.96-1.02) 0.455
Ref. 0.93 (0.44-1.96)
0.851
Ref. 1.46 (0.82-2.59)
0.202
Ref. 1.27 (0.65-2.47)
0.487
Ref. 1.49 (0.86-2.6)
0.157
Ref. 1.51 (0.79-2.89)
0.212
Ref. 1.71 (1.01-2.89)
0.047
Ref. 1.59 (0.84-3.03)
0.152
HR = Hazards ratio; CI = Confidence interval; CCI = Charlson comorbidity index
3 Page 21 of 23
Figure 1 01-16-18.tif
4 Page 22 of 23
Figure 2 01-16-18.tif
5 Page 23 of 23