Review Article. Upper Urinary Tract Recurrence Following Radical Cystectomy for Bladder Cancer: A Meta-Analysis on 13,185 Patients

Transcription

1 Review Article Upper Urinary Tract Recurrence Following Radical Cystectomy for Bladder Cancer: A Meta-Analysis on 13,185 Patients Stefano Picozzi,* Cristian Ricci, Maddalena Gaeta, Dario Ratti, Alberto Macchi, Stefano Casellato, Giorgio Bozzini and Luca Carmignani From the Urology Department (SP, DR, AM, SC, GB, LC) and Biometry and Clinical Epidemiology Unit (CR), IRCCS Policlinico San Donato, University of Milan, Milan, and the Hygiene Unit, Department of Preventive and Occupational Medicine, University of Pavia, Pavia (MG), Italy Abbreviations and Acronyms CIS carcinoma in situ TCC transitional cell cancer UUT upper urinary tract Submitted for publication December 8, Supplementary material can be obtained at Nothing to disclose. * Correspondence: Urology Department, IRCCS Policlinico San Donato, Via Morandi 30, San Donato Milanese, Milan Italy (telephone: ; FAX: ; stepico@tin.it). For other articles on related topics see pages 2377, 2384, 2391 and Editor s Note: This article is the first of 5 published in this issue for which category 1 CME credits can be earned. Instructions for obtaining credits are given with the questions on pages 2446 and Purpose: Patients who undergo radical cystectomy for urothelial cancer are at risk for upper urinary tract disease in the remnant transitional tissue. Previous studies have identified several risk factors for upper urinary tract recurrence but the predictive value of each factor remains controversial. Furthermore, the schedule for surveillance of the upper urinary tract with imaging techniques and cytology has not been established. International guidelines do not address these topics and refer only to isolated works with a large case based analysis. We performed this meta-analysis to evaluate the effective incidence of upper urinary tract recurrence after cystectomy for bladder cancer, to analyze the risk factors so we can create subgroups of patients at high risk for recurrence and to investigate the real role of screening in the detection of upper tract lesions at an early stage. Materials and Methods: A bibliographic search covering the period from January 1970 to July 2010 was conducted using PubMed, MEDLINE and EMBASE. This analysis is based on the 27 studies that fulfilled the predefined inclusion criteria. Data were analyzed using a fixed effect logistic regression approach and classic meta-analysis. Results: A total of 13,185 participants were included in the analysis. Followup was described in 22 studies and ranged from 0.36 to months. The overall prevalence of upper tract transitional cell cancer after cystectomy ranged from 0.75% to 6.4%. Recurrence appeared at a range of 2.4 to 164 months, and in an advanced (64.6%) or metastatic state (35.6%) as reflected in poor survival rates. Patients with low grade vs high grade lesions at cystectomy showed as strong a significant difference in incidence as those with carcinoma in situ and superficial cancer vs invasive cancers and as strong as in those without lymph node involvement, with multifocal disease, with a history of multiple urothelial recurrences, with positive ureteral margins, with positive urethral margins, with urethral involvement and a history of upper urinary tract urothelial cancer. Data do not support a statistically significant difference in recurrence among patients with a history of carcinoma in situ, solitary lesion and among various types of urinary diversion adopted. In 24 studies the followup schedule included periodic radiological assessment of the upper urinary tract and in 20 it included urinary cytology. In 14 studies in 63 of 166 patients (38%) upper urinary tract recurrence was diagnosed by followup investigation whereas in the remaining 62% diagnosis was based on symptoms. When urine cytology was used in surveillance the rate of primary detection was 7% and with upper urinary tract imaging it was 29.6%. Of 5,537 patients who underwent routine cytological examination, recurrence was diagnosed in 1.8/1,000 and of those who underwent upper urinary tract imaging recurrence was diagnosed in 7.6/1, /12/ /0 THE JOURNAL OF UROLOGY Vol. 188, , December by AMERICAN UROLOGICAL ASSOCIATION EDUCATION AND RESEARCH, INC. Printed in U.S.A.

2 UPPER URINARY TRACT RECURRENCE FOLLOWING RADICAL CYSTECTOMY 2047 Conclusions: The recurrence values could appear low when considering the pan-urothelial field defect theory, but these values reflect, in part, the mortality associated with the initial bladder cancer. Based on anamnesis and pathological examination of cystectomy specimens, a group of patients is at high risk. Extensive regular followup with cytology, urography and loopgraphy yields insufficient benefits. Periodic computerized tomography with urography combines the ability to study the upper urinary tract oncologically and functionally, and the identification of any parenchymal, osseous or lymph node secondary lesion. Key Words: urinary bladder neoplasms; ureteral neoplasms; carcinoma, transitional cell; recurrence; cystectomy TRANSITIONAL cell carcinoma is often multifocal, and while it most frequently involves the bladder mucosa, the urothelial lining of the renal pelvis, the ureters and the urethra are also at risk. In fact, urothelial carcinoma is thought to be associated with a pan-urothelial field defect characterized by frequent, multifocal metachronous tumors due to the transformation of epithelial cells at different sites, and the intraluminal seeding and implantation of tumor cells derived from an initial clone. 1 Today radical cystectomy is the standard treatment for patients with invasive bladder cancer and for those with superficial disease at high risk for progression. 2,3 Patients who undergo radical cystectomy for urothelial cancer are at risk for UUT disease in the remnant transitional tissue and clearly not all patients have an equal propensity for a UUT tumor. Previous studies identified several risk factors but the predictive value of each factor remains controversial. In addition, the optimal schedule for surveillance, and the role of cytology and imaging techniques have not yet been established. International guidelines do not cover these issues, and refer only to some work with a large case based analysis. 2,3 MATERIALS AND METHODS In this systematic review and meta-analysis we evaluate the effective incidence of UUT recurrence after cystectomy for bladder cancer, analyze the risk factors so we can create subgroups of patients at high risk for recurrence, and study the real role of screening in the detection of upper tract lesions at an early stage. Search Strategy Studies were identified by searching electronic databases and scanning reference lists of articles. A bibliographic search covering January 1970 to July 2010 was conducted using PubMed, MEDLINE and EMBASE. Additional hand searches were performed of the reference lists of included studies, reviews, meta-analyses and guidelines on UUT recurrence after cystectomy. Several search terms were used for each, including bladder cancer, upper urinary tract cancer, cystectomy, recurrence, population based, incidence, treatment, urinary cytology, urography, tomography and followup. The searches were restricted to publications in English. Study Selection Studies were excluded from analysis if they were case reports, meeting abstracts and conference proceedings. Identified studies were reviewed and selected if they reported data related to UUT recurrence after radical cystectomy. The inclusion or exclusion of studies was performed hierarchically based first on the title of the report, then on the abstract and finally on the contents of the full text. A study was accepted for inclusion on the basis of the agreement of 2 investigators (SP and CR). Any disagreement on study inclusion was resolved by consulting a third investigator (LC). Database searches yielded 143 references. Exclusion of irrelevant references left 29 references describing studies. We excluded 2 further references because they were not in English. Thus, this analysis is based on the 27 studies that fulfilled the predefined inclusion criteria. Data Extraction, Quality of Comparative Study and Level of Evidence One author (SP) extracted the data and a second author (CR) checked the extracted data to ensure data quality. Disagreements were resolved by discussion between the 2 review authors, and if no agreement could be reached it was planned that a third author (LC) would decide. The quality of studies was scored using the methods of the U.S. Preventive Services Task Force. 4,5 PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines were applied in the preparation of this report. 6 All the studies included in the meta-analysis were classified as good, 7 32 except that of Akkad et al, 23 which was fair due to the moderate sample size. Statistical Analysis Data were analyzed in 2 ways. If the survey reported incomplete data on exposure (no events reported for a given level), a fixed effect logistic regression approach was applied to all reported outcomes. To correct for overdispersion, the covariance matrix was multiplied by the estimate of the dispersion parameter using Williams or deviance criterion where appropriate. 33 The effects of factors of interest were evaluated by odds ratio estimate and 95% confidence limits. A second approach of meta-analysis was applied to surveys reporting complete data about exposures (the outcomes rate for all factors of interest by study). For this analysis the random or fixed effect model analysis was used according to the heterogeneity among studies (random effect model was used when I 2 was greater than 50%). 34

3 2048 UPPER URINARY TRACT RECURRENCE FOLLOWING RADICAL CYSTECTOMY Evaluation of small study effect and publication bias was performed with funnel plot visual inspection for all of the comparisons and with Harbord-Egger tests where applicable (number of included studies greater than 5). 35 Incomplete data meta-analysis was performed using SAS version and classic meta-analysis was performed using RevMan software version 5.0 (Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2008). RESULTS Analysis of Database Upper urinary tract recurrence is defined as any documented radiographic, endoscopic or pathologically proven recurrence in the renal collecting system or ureter after cystectomy. A total of 13,185 participants were included in the 27 studies examined in this meta-analysis. The number of participants in each survey ranged from 85 to 1,420 (mean 488, median 382). Of the 27 studies 1 was published in the 1970s, 1 in the 1980s, 9 in the 1990s and the remaining 16 after There were 14 studies from the United States, 3 from Japan, 2 from Spain, and Switzerland, Germany, Finland, Korea, Great Britain and Austria provided 1 study each. The interval in which patients were recruited in these studies varied between 5 and 22 years. Patient Characteristics Of the 27 studies 15 reported patient gender, with 7,972 male subjects and 1,638 female for a ratio of 5:1. Thirteen studies reported participant age, with a mean of 64.8 years (median 65, range 23 to 93). Followup was described in 22 studies, and ranged between 0.36 and months. Overall Analysis of Outcomes The overall prevalence of UUT-TCC after cystectomy for muscle invasive or refractory nonmuscle invasive TCC ranged from 0.75% to 6.4%. There were no cases of recurrence due to other types of cancer. This range of recurrence could appear low with respect to the pan-urothelial field defect theory, but this in part reflects the mortality associated with the initial bladder cancer. In addition, cancerous substances eliminated in the urine remain stored and in contact with the bladder mucosa for a consistent time, compared to the pelvic and ureteral mucosa acting in this site for a longer time, thereby promoting cellular malignant transformations. Small Study and Publication Bias Assessment Funnel plot visual inspection did not show publication or small study bias for the comparisons performed as shown in figures 1 and 2, with the exception of history of concomitant CIS vs absent CIS (fig. 2, 9). For this comparison the Egger test did not generate a statistically significant result (p 0.398), whereas the Harbord test result was significant (p ). For the comparison of positive vs negative ureteral margins the absence of publication bias by funnel plot visual inspection was confirmed by the Egger test (p ) but not by the Harbord test (p , fig. 2, 12). Tumor Grading at Cystectomy Bladder cancer was classified as grade 1 (well differentiated), grade 2 (moderately differentiated) and grade 3 (poorly differentiated) according to 1973 WHO grading. We compared the different grades of cancerous lesions using the 2 statistical methods of logistic regression with fixed effects and classic meta-analysis. The comparison of UUT cancer recurrence in patients with G1 vs G2 tumors using logistic regression with fixed effects showed an OR of (p ). This result was also confirmed by the second analysis, classic meta-analysis (OR 7.25, p 0.03, fig. 1, 1). These data reveal a statistically significant difference in recurrence for G1 vs G2. When comparing G1 and G3, there was a tendency in the difference between the incidence of G1 and G3 (fig. 1, 2). The comparison between patients with G2 and G3 bladder cancer was not significant (fig. 1, 3). The difference in G1-G2 vs G3 was also not significant (fig. 1, 4). Ultimately the comparison of G1 vs G2 associated to G3 showed a strong significant difference (fig. 1, 5). These data show that patients treated with radical cystectomy for low grade tumors have a greater propensity for a secondary lesion to develop in the UUT during followup. The relationship between low grade disease and late upper tract recurrence can be explained in 2 ways. 1) These patients are often affected by plurifocal and plurirecurrent TCC (not under control by simple transurethral resection). 2) These lesions are at high risk for extravesical urothelial involvement, and patients with low grade tumors have a potential long-term survival and followup. Disease Staging In this meta-analysis bladder cancer staging was as CIS, and superficial and muscle invasive tumors. In the comparison of UUT cancer recurrence in patients with superficial vs invasive tumors using logistic regression with fixed effects, the data indicated a statistically significant difference in recurrence in superficial vs invasive disease (fig. 1, 6). When comparing the recurrence rate in patients who underwent radical cystectomy for CIS vs superficial and invasive lesions, the data showed a statistically significant difference in recurrence for superficial vs invasive lesions and a strong significant difference in recurrence for CIS vs invasive lesions (fig. 1, 7 and fig. 2, 8).

4 UPPER URINARY TRACT RECURRENCE FOLLOWING RADICAL CYSTECTOMY 2049 Figure 1. Forrest and funnel plot of meta-analysis. Fixed or random effect was applied according to heterogeneity. In funnel plot vertical line represents OR estimate and oblique lines define region free from publication bias.

5 2050 UPPER URINARY TRACT RECURRENCE FOLLOWING RADICAL CYSTECTOMY Figure 2. Forrest and funnel plot of meta-analysis. Fixed or random effect was applied according to heterogeneity. In funnel plot vertical line represents OR estimate and oblique lines define region free from publication bias. UUTC, upper urinary tract cancer.

6 UPPER URINARY TRACT RECURRENCE FOLLOWING RADICAL CYSTECTOMY 2051 Patients who underwent cystectomy for CIS and superficial tumors had a high risk of UUT recurrence during followup compared to those with invasive lesions. The relationship between low stage and late upper tract recurrence can be explained in 2 ways. 1) Patients with CIS or superficial tumors are those with potential long-term followup. 2) These patients with recurrent multifocal superficial tumors or associated bladder CIS are at high risk for extravesical urothelial involvement. History or Concomitant Presence of CIS The classification of concomitant CIS or history of CIS includes foci of CIS concomitant to invasive or superficial lesions at cystectomy or history of a CIS lesion at prior transurethral resection. In the comparison of UUT cancer relapse in patients with a history of or with concomitant CIS vs the absence of CIS, the data do not support a statistically significant difference in recurrence (fig. 2, 9). Node Staging The comparison of N0 vs N (including all patients with lymph node involvement by the urothelial lesion) shows a strong significant difference in incidence (fig. 2, 10). Patients with nodal involvement have potential short-term followup related to their high mortality rate. Recurrent, Solitary or Multifocal Disease Data show a statistically significant difference in recurrence in patients with a history of multiple urothelial recurrences or with multifocal disease compared to those with a solitary lesion at precystectomy transurethral resection (fig. 2, 11). Ureteral Margin Status The comparison of UUT cancer recurrence in patients with positive vs negative ureteral margins using logistic regression with fixed effects showed an OR of (p ). This result was also confirmed but underestimated using classic meta-analysis (OR 3.28, p , fig. 2, 12). These data show a strong statistically significant difference in recurrence in patients with positive ureteral margins at cystectomy. Intraoperative analysis of ureteral margins via frozen sections at cystectomy before urinary tract reconstruction has long been accepted as a standard practice. The apparent rationale for such a procedure is that carcinoma, particularly carcinoma in situ, may involve the distal ureteral margin. Using frozen section analysis, surgeons can resect additional ureteral tissue to obtain an uninvolved ureteral margin, hopefully improving cancer control with complete tumor excision and decreasing the likelihood of disease recurrence. The actual usefulness of routine frozen section analysis before reconstruction of the ureterointestinal anastomosis has not been clearly proven. Doubt has been based on factors such as the limited ability of frozen section analysis to accurately identify ureteral malignancy, the fact that sequential resection does not definitively eliminate malignancy and the questionable effect of a malignant ureteral anastomotic margin on oncologic recurrence. With these data of strong statistical difference, we can delineate a risk class of patients who need accurate followup related to UUT recurrence, which constitutes a point in favor of this practice. Urethral Involvement Unfortunately, most studies do not differentiate between cases with direct extension of the prostate stroma by bladder tumor (pt4) and noncontiguous urothelial cancer involving the prostatic urethra. Considering these 2 possibilities as a single entity in the interpretation of urethral margin status, in the comparison of UUT recurrence with involvement of the urethral margins vs negative cases, the data show a strong statistically significant difference in recurrence (fig. 2, 13). The true incidence of eventual UUT tumor in those patients with urothelial tract cancer involving the prostatic urethra is unknown, with only 2 reports showing a statistically significant association. 18,22 Taking into consideration these 2 studies, in the comparison of UUT cancer recurrence using classic meta-analysis we found an OR of 7.19 with p These data show a stronger statistically significant difference in recurrence in patients with TCC of the male prostatic urethra or female urethra at cystectomy. Sved et al noted that superficial prostatic transitional cell disease was strongly associated with an increased risk of recurrence, whereas prostatic stromal invasion was not. 21 Sanderson et al identified superficial prostatic TCC as the only pathological variable in men significantly associated with an increased rate of UUT recurrence, and their data indicated that at 10 years men with prostatic urethral involvement were at a fourfold increased cumulative risk. 25 The lower rate of UUT recurrence in men with invasive prostatic pt4 disease is likely the result of significantly decreased survival and, thus, a shorter time at risk for recurrence. Supporting this hypothesis is the observation of Sved et al that the median survival of men with invasive prostatic TCC was significantly shorter than that of men with superficial prostatic TCC, and shorter than the median time for UUT to recur. 21 Urinary Diversion Type Urinary diversion was categorized as neobladder, continent stoma, ileal stoma, ureterosigmoidostomy and ureterocutaneostomy. No differences were

7 2052 UPPER URINARY TRACT RECURRENCE FOLLOWING RADICAL CYSTECTOMY found by logistic regression with fixed effects and classic meta-analysis. The cohort of patients treated with a neobladder had a favorable prognosis compared to those treated with other diversions. This finding was reflected in the better overall survival, which was mainly related to age and underlying comorbidities. Likewise, due to a higher proportion of organ confined tumors, disease specific survival was also superior in the neobladder group. Such inevitable selection bias in these methods of diversion must be considered when drawing conclusions related to cancer control, but we did not observe an increased risk of UUT recurrence in the various cohorts. History of UUT Urothelial Cancer The comparison of UUT recurrence in patients with vs without a history of UUT-TCC using classic metaanalysis revealed an OR of 7.25 with p 0.03 (fig. 2, 14). These data show a statistically significant difference in patients already treated for UUT cancer. Evaluation of Recurrence by Followup Method Due to clinical improvements in the detection and treatment of organ confined bladder cancer, in recent years many patients have had a favorable longterm prognosis. However, the optimum surveillance of UUT recurrence after cystectomy is controversial. The followup protocols vary in terms of timing and investigations. Three studies did not specify the type of followup. 7,15,32 In the other studies the followup schedule included radiological assessment of the UUT with loopgraphy, urography or computerized tomography. 1,8 14,16 31 In 18 of the 25 studies a periodic evaluation of urinary cytology was performed, 1,8,11,12,14,16,19 27,29 31 while in 2 it was performed only on the finding of radiological abnormalities due to hematuria and irritative symptoms. 9,13 One of the aims of regular followup is to increase the likelihood of the early detection of any subsequent upper tract tumor, enabling treatment at a curable stage. From the literature analysis 13 studies reported that recurrence was identified during the followup investigations or in response to symptoms suggestive of recurrence. 1,10,12 14,16 18,21,23,25,26,31 In 58 of 161 patients (38%) UUT recurrence was diagnosed by followup investigation and the majority of recurrence was diagnosed by symptoms in the remaining 62% of the cases. In 9 studies including 100 cases of recurrence (28 of which were identified by followup investigation), 66 described symptoms indicating a UUT lesion. 1,10,13,14,17,21,23,25,31 These included 46 cases with hematuria (60.5%), 19 with back pain and hydronephrosis (25%), 2 with fever (2.6%), 2 with weight loss (2.6%) and 1 with pyelonephritis (1.3%). In the remaining 4 cases the symptoms were unknown and in 2 the diagnosis of recurrence was incidental. Cytology In 9 studies 112 patients with recurrence, of whom 38 were identified at followup, positive urine cytology suggested the presence of an upper tract lesion in 10. 1,12,14,16,21,23,25,26,31 When urine cytology was used in the surveillance of patients after radical cystectomy, the rate of primary detection was 8.9%. There are many points of view on the usefulness of urinary cytology. It is an easy and noninvasive examination method for detecting tumors of the upper and/or lower urinary tract including the urethra. The incidence of UUT recurrence after radical cystectomy was extremely low, and it was somewhat difficult to interpret the urine cytology obtained from the diverted urinary tract. The urine excreted through the intestinal urinary diversion is different from normally voided urine, and it contains many degenerating desquamated intestinal epithelial cells that could obscure or be confused with cancer cells. Furthermore, the urine voided through a continent reservoir or an orthotopic neobladder usually stagnates for a few hours in the isolated intestine or the colon, which also allows the urinary cells to degenerate. In addition, urine cytology itself depends to some extent on cytopathologist ability/experience. Nevertheless, some authors argue that cytology is an inexpensive way to monitor the urinary tract for TCC and, since most UUT cancers after cystectomy are reported to be high grade, the sensitivity of cytology should accordingly be high. Based on this analysis the incidence of UUT-TCC was extremely low and the interpretation of urine cytology from the diverted urinary tract was somewhat difficult. Therefore, frequent routine urine cytology examinations might yield insufficient benefit if not restricted to high risk patients. In fact, in 9 studies 4,616 patients underwent routine cytological examination (2 to 4 times in the first years) to identify 10 of 112 patients with recurrence (2.1/1,000). 1,12,14,16,21,23,25,26,31 Thus, approximately 2,000 urinary cytology examinations were performed to identify 1 patient with UUT recurrence. UUT Imaging In 13 studies on 161 patients of whom 58 were identified at followup, radiological findings suggested the presence of an upper tract lesion in 40 patients. 1,9,10,12 14,16 18,21,23,25,26,31 When UUT imaging was used in the surveillance of patients after radical cystectomy, the rate of primary detection was 24.8%. The value of urography in the detection of UUT recurrence after radical cystectomy is controversial. Some groups consider these regular investigations a necessity, while others maintain the opposite based

8 UPPER URINARY TRACT RECURRENCE FOLLOWING RADICAL CYSTECTOMY 2053 on the low yield of UUT imaging, the fact that the outcome does not appear to improve and that this examination lost information regarding the status of secondary lesions in parenchymal organs and bones. This renders regular urography examinations for followup in all patients after cystectomy hardly defensible. It has been estimated that 5,885 patients underwent routine UUT imaging to identify 40 of 161 patients with recurrence (6.7/1,000). Thus, approximately 800 radiological examinations were performed to identify 1 patient with UUT recurrence. 1,9,11 14,16 18,21,23,25,26,31 It is impossible to define the sensitivity and specificity of a single radiological procedure. We can emphasize that the use of computerized tomography with urography combines the ability to study the upper urinary tract oncologically and functionally for the identification of any parenchymal, osseous or lymph node secondary lesion. Evaluation of Treatment In 16 articles the possibility of undergoing surgical intervention was reported, and from these data of 218 patients 175 underwent surgery, 38 presented with characteristics of inoperability and 5 refused therapy. 8 10,12,14,15,18 21,24 26,28,29,31 In several studies the operability of the patients was reported along with the surgical procedure. 9,10,12,14 26,28,29,31 Of 193 patients 165 underwent nephroureterectomy, 23 tumorectomy (1 of whom then underwent nephroureterectomy) and 5 ureterotomy (bilateral in 1). Analysis and Outcome of Recurrence Recurrence varied during followup with a range of 2.4 to 174 months. 8,10,12 14,16 18,21 27,29 31 In analyzing the recurrences, most tumors were only detected in an advanced (41 noninvasive lesions compared to 93 negative nodes in 13 studies) or metastatic state (16 lymph node metastases compared to 29 negative nodes in 13 studies), which is reflected in the poor survival rates (74 survived vs 127 deceased in 17 studies during the 1 to 120-month interval of discovery of UUT recurrence to death). 8 10,12,14,15,18 21,24 26,28,29,31 We could not detect a relevant survival advantage in patients in whom the tumor was diagnosed by followup studies compared to those in whom symptoms led to the diagnosis because the data are poor. Potential Flaws Potentials flaws of these meta-analyses involve the retrospective nature of some of the included studies, the different protocols of followup adopted with their varying ability to reveal UUT-TCC recurrence, the progress in the surgical and imaging studies that was made since the 1970s and, in the end, the different adjuvant protocols adopted. CONCLUSIONS After reading this article most investigators would agree that there is a subset of patients at higher risk for recurrence who might benefit from more diligent upper tract surveillance after cystectomy. Early detection of recurrent disease results in the timely administration of appropriate therapy and potentially improves patient outcomes. Based on cystectomy specimens and patient urological history, we can delineate risk factors for UUT recurrence and weigh each of these in every single patient. Patients with a G1 tumor have an 8 times greater probability of a UUT-TCC lesion compared to those with G2 cancer, and a probability 10 times higher than those with G3 disease. Patients with CIS have a probability of a UUT-TCC lesion twice as high as those with superficial cancer, and a probability 4 times higher than those with invasive disease. Patients with superficial cancer have a probability of a UUT-TCC lesion twice as high as those with invasive disease. Multifocality increases the risk of recurrence of 3 times and recurrence from 2 to 4 times. Positive ureteral or urethral margins increase the risk sevenfold. The same increase in risk is present in patients with a history of UUT urothelial cancer. Patients with a negative node status have a risk increased eightfold. The extensive followup for the detection of upper tract recurrence based on cytology and excretory urography is ineffective and expensive. Cytology would be eliminated in the routine followup and urography would need to be substituted by computerized tomography to unite the oncologic study of the upper urinary tract, the abdominal organs and lymph nodes in the functional study of the new modified urinary diversion. REFERENCES 1. Volkmer BG, Schnoeller T, Kuefer R et al: Upper urinary tract recurrence after radical cystectomy for bladder cancer who is at risk? J Urol 2009; 182: AUA Bladder Cancer Clinical Guideline Update Panel: Guideline for the Management of Nonmuscle Invasive Bladder Cancer: (Stages Ta, T1, and Tis): 2007 Update. Available at clinical-practice-guidelines/clinical-guidelines.cfm? sub bc. Accessed August 3, EAU Guidelines 2011: Bladder Cancer Muscleinvasive and Metastatic. 4. Harris RP, Helfand M, Woolf SH et al: Current methods of the US Preventive Services Task Force: a review of the process. Am J Prev Med 2001; 20: Harris R, Atkins D, Berg AO et al: US Preventive Services Task Force Procedure Manual. Rockville: Agency for Healthcare Research and Quality 2001.

9 2054 UPPER URINARY TRACT RECURRENCE FOLLOWING RADICAL CYSTECTOMY 6. Moher D, Liberati A, Tetzlaff J et al: Preferred reporting items for systematic reviews and metaanalyses: the PRISMA statement. PLoS Med 2009; 6: e Schellhammer PF and Whitmore WF Jr: Transitional cell carcinoma of the urethra in men having cystectomy for bladder cancer. J Urol 1976; 115: Zincke H, Garbeff PJ and Beahrs JR: Upper urinary tract transitional cell cancer after radical cystectomy for bladder cancer. J Urol 1984; 131: Malkowicz SB and Skinner DG: Development of upper tract carcinoma after cystectomy for bladder carcinoma. Urology 1990; 36: Hastie KJ, Hamdy FC, Collins MC et al: Upper tract tumours following cystectomy for bladder cancer. Is routine intravenous urography worthwhile? Br J Urol 1991; 67: Schwartz CB, Bekirov H and Melman A: Urothelial tumors of upper tract following treatment of primary bladder transitional cell carcinoma. Urology 1992; 40: Kenworthy P, Tanguay S and Dinney CP: The risk of upper tract recurrence following cystectomy in patients with transitional cell carcinoma involving the distal ureter. J Urol 1996; 155: Tsuji Y, Nakamura H and Ariyoshi A: Upper urinary tract involvement after cystectomy and ileal conduit diversion for primary bladder carcinoma. Eur Urol 1996; 29: Silver DA, Stroumbakis N, Russo P et al: Ureteral carcinoma in situ at radical cystectomy: does the margin matter? J Urol 1997; 158: Studer UE and Zingg EJ: Ileal orthotopic bladder substitutes. What we have learned from 12 years experience with 200 patients. Urol Clin North Am 1997; 24: Balaji KC, McGuire M, Grotas J et al: Upper tract recurrences following radical cystectomy: an analysis of prognostic factors, recurrence pattern and stage at presentation. J Urol 1999; 162: Slaton JW, Swanson DA, Grossman HB et al: A stage specific approach to tumor surveillance after radical cystectomy for transitional cell carcinoma of the bladder. J Urol 1999; 162: Huguet-Pérez J, Palou J, Millán-Rodríguez F et al: Upper tract transitional cell carcinoma following cystectomy for bladder cancer. Eur Urol 2001; 40: Solsona E, Iborra I, Rubio J et al: Late oncological occurrences following radical cystectomy in patients with bladder cancer. Eur Urol 2003; 43: Yossepowitch O, Dalbagni G, Golijanin D et al: Orthotopic urinary diversion after cystectomy for bladder cancer: implications for cancer control and patterns of disease recurrence. J Urol 2003; 169: Sved PD, Gomez P, Nieder AM et al: Upper tract tumour after radical cystectomy for transitional cell carcinoma of the bladder: incidence and risk factors. BJU Int 2004; 94: Lee SE, Byun SS, Hong SK et al: Significance of cancer involvement at the ureteral margin detected on routine frozen section analysis during radical cystectomy. Urol Int 2006; 77: Akkad T, Gozzi C, Deibl M et al: Tumor recurrence in the remnant urothelium of females undergoing radical cystectomy for transitional cell carcinoma of the bladder: long-term results from a single center. J Urol 2006; 175: Furukawa J, Miyake H, Hara I et al: Upper urinary tract recurrence following radical cystectomy for bladder cancer. Int J Urol 2007; 14: Sanderson KM, Cai J, Miranda G et al: Upper tract urothelial recurrence following radical cystectomy for transitional cell carcinoma of the bladder: an analysis of 1,069 patients with 10-year followup. J Urol 2007; 177: Meissner C, Giannarini G, Schumacher MC et al: The efficiency of excretory urography to detect upper urinary tract tumors after cystectomy for urothelial cancer. J Urol 2007; 178: Tran W, Serio AM, Raj GV et al: Longitudinal risk of upper tract recurrence following radical cystectomy for urothelial cancer and the potential implications for long-term surveillance. J Urol 2008; 179: Boström PJ, Mirtti T, Kössi J et al: Twenty-year experience of radical cystectomy for bladder cancer in a medium-volume centre. Scand J Urol Nephrol 2009; 43: Tollefson MK, Blute ML, Farmer SA et al: Significance of distal ureteral margin at radical cystectomy for urothelial carcinoma. J Urol 2010; 183: Umbreit EC, Crispen PL, Shimko MS et al: Multifactorial, site-specific recurrence model after radical cystectomy for urothelial carcinoma. Cancer 2010; 116: Yoshimine S, Kikuchi E, Matsumoto K et al: The clinical significance of urine cytology after a radical cystectomy for urothelial cancer. Int J Urol 2010; 17: Gakis G, Schilling D, Perner S et al: Sequential resection of malignant ureteral margins at radical cystectomy: a critical assessment of the value of frozen section analysis. World J Urol 2011; 29: Lindsey JK: On the use of corrections for overdispersion. J R Stat Soc Ser C Appl Stat 1999; 48: DerSimonian R and Laird N: Meta-analysis in clinical trials. Control Clin Trials 1986; 7: Harbord RM, Egger M and Sterne JA: A modified test for small-study effects in meta-analyses of controlled trials with binary endpoints. Stat Med 2006; 25: 3443.