Hydronephrosis as a Prognostic Marker in Bladder Cancer in a Cystectomy-Only Series

Transcription

1 european urology 51 (2007) available at journal homepage: Bladder Cancer Hydronephrosis as a Prognostic Marker in Bladder Cancer in a Cystectomy-Only Series Georg C. Bartsch *, Rainer Kuefer, Juergen E. Gschwend, Robert de Petriconi, Richard E. Hautmann, Bjoern G. Volkmer Department of Urology, Faculty of Medicine, University of Ulm, Ulm, Germany Article info Article history: Accepted July 11, 2006 Published online ahead of print on July 28, 2006 Keywords: Bladder carcinoma Cystectomy Hydronephrosis Transitional cell carcinoma Abstract Objectives: Hydronephrosis in patients with bladder cancer is caused by tumour at the ureteral orifice, secondary ureteral tumours, intramural or extravesical tumour infiltration, or compression of the ureter. This study investigated the prognostic impact of hydronephrosis in bladder cancer. Methods: A series of 788 patients were treated with radical cystectomy with curative intent for transitional cell carcinoma of the bladder without neoadjuvant/adjuvant radiotherapy/chemotherapy between January 1986 and September All patients had a complete follow-up until death or until the study s end date. Survival rates were calculated using the Kaplan-Meier method. A multivariate analysis with a Cox regression model was performed with respect to potential influencing factors. Results: A total of 108 patients (13.7%) had unilateral and 25 patients (3.2%) had bilateral hydronephrosis. The rate of organ-confined tumours was significantly higher in patients without hydronephrosis (67.9% vs. 37.6%; p < 0.001). Forty-three (32.3%) of the 133 hydronephrotic patients had a tumour involving the ureteral orifice. In this group the rate of organ-confined tumours was significantly higher than in the other patients with hydronephrosis (53.5% vs. 30.0%; p = 0.009). In the multivariate analysis, preoperative hydronephrosis was determined as an independent prognostic marker for recurrence-free survival besides the pt classification and lymph node status ( p = ). The etiology of hydronephrosis did not affect the tumour-specific survival. Conclusions: Hydronephrosis at the time of diagnosis of bladder cancer is associated with a high probability of advanced tumours. It is an independent prognostic factor for recurrence-free survival. # 2006 European Association of Urology. Published by Elsevier B.V. All rights reserved. * Corresponding author. Department of Urology, Faculty of Medicine, University of Ulm, Prittwitzstrasse 43, Ulm, Germany. Tel ; Fax: address: georg.bartsch@uniklinik-ulm.de (G.C. Bartsch) /$ see back matter # 2006 European Association of Urology. Published by Elsevier B.V. All rights reserved. doi: /j.eururo

2 european urology 51 (2007) Introduction Hydronephrosis is a common finding in patients with bladder carcinoma. The incidence of hydronephrosis associated with bladder carcinoma ranges from 7.2% to 54.1% [1 3]. Its prognostic value in patients with bladder carcinoma is controversial [4,5]. Three different etiologies are recognised for hydronephrosis in patients with a malignant intravesical process [1]: Intramural or extravesical tumour extension leading to ureteral compression or infiltration Tumour involving the ureteral orifice Simultaneous ureteral tumour Information on whether these three entities influence the prognosis of bladder cancer is lacking. Therefore, the aim of this study was to analyse the prognostic impact of hydronephrosis at the time of diagnosis of transitional cell carcinoma (TCC) in the bladder in a large homogeneous, single-centre, radical cystectomy-only series. In particular, the following topics have been investigated: Is hydronephrosis at the time of diagnosis of bladder cancer a valid marker for advanced disease? Is hydronephrosis at the time of diagnosis an independent prognostic factor for recurrence-free survival of patients with TCC? Does the etiology of the tumour-associated hydronephrosis matter? 2. Patients and methods This study is based on the complete data set of 1131 patients who underwent radical cystectomy at our institution between January 1986 and September We defined the following inclusion criteria: Transitional cell carcinoma (TCC) only Radical cystectomy performed with curative intent with at least standard bilateral pelvic lymph node dissection (PLND). An extended-field PLND was offered from January 2003 (n = 43). Exclusion of distant metastases by ultrasound, computed tomography (CT) scan, and bone scan No neoadjuvant/adjuvant radiotherapy or chemotherapy administered We included 788 patients. The excluded patients had benign diseases (n = 30), malignancies other than TCC (n = 142), or TCC of the bladder with neoadjuvant or adjuvant treatment (n = 171). All patients had a complete follow-up until September The median patient age at time of cystectomy was 65 yr (range: yr), the median follow-up was 35 mo (range: mo; Table 1). The majority of patients with pta tumours underwent cystectomy for a dysfunctional bladder despite oncologic reasons. Almost all T1 tumours were highgrade tumours. In these patients early cystectomy is our preferred treatment. All patients had a complete preoperative work-up: physical examination, blood count, blood urea nitrogen, creatinine, electrolyte analysis, chest x-ray, abdominal ultrasound, Table 1 Principal characteristics of 788 patients Characteristics All patients pn0 pn+ No. % No. % No. % Pathologic staging pta/ptis pt pt pt pt Total Grading Grade 1 TCC 12 2 Grade 2 TCC Grade 3 TCC Grade 4 TCC 18 2 Age, yr (median/mean SD) 65/63 10 (range: 23 91) Men (n = 652) 63/63 9 (range: 23 91) Women (n = 136) 65/65 10 (range: 31 87) Median/mean follow-up, mo 35/54 (range: ) TCC = transitional cell carcinoma; SD = standard deviation.

3 692 european urology 51 (2007) intravenous pyelogram (IVP), and bone scan. CT scanning of the abdomen was performed in most cases but was considered mandatory only in cases with hints for metastatic disease, due to its limited predictive value for local tumour extension or lymph node metastases. In patients with reduced renal function magnetic resonance imaging instead of an IVP/CT scan was performed. Hydronephrosis was defined as dilatation of the renal pelvis and calyces with or without secondary changes of the renal parenchyma or renal function. Hydronephrosis was diagnosed by renal ultrasound, CT scan, or IVP. All ultrasound images were reviewed by an expert urologist. CT scans and x-rays were reviewed by a uroradiologist. In all patients with signs of reduced renal function an isotope nephrogram was obtained. Criteria for the preservation of the hydronephrotic kidney were renal function >20% of the total renal function in case of normal renal function or renal function >15% of the overall renal function in case of reduced renal function. An ileal neobladder was performed whenever possible with a serum creatinine cut-off level of 2.0 mg/dl. Pathologic staging was classified according to the 2002 TNM classification. Pathologic subgroups were defined as organconfined (<pt3a, pn0), non organ-confined (>pt2b, pn0), and lymph node positive (any pt, pn+). In case of a pt0 situation (152 of 788), the tumour was classified according to the maximum tumour stage in the transurethral resection of bladder tumours (TURBTs) [6]. Patients were examined every 3 mo for the first 2 yr and thereafter at 6-mo intervals according to Bochner et al., but with closer follow-up intervals [7]. In all cases the following data were acquired: The presence and localisation of hydronephrosis at the time of bladder cancer diagnosis. Because some patients with recurrent superficial bladder cancer required cystectomy many years after the first onset of TCC, we defined the time of diagnosis as the time of the TCC diagnosis leading to cystectomy. The manifestation site of the TCC within the urinary tract (endoscopic finding) at the time of diagnosis (ureteral orifice, other regions of the urinary bladder, simultaneous ureteral tumour). The histologic reports of all TURBT procedures and cystectomy specimens, including both distal ureters, and of the nephroureterectomy specimens, if applicable to define an intramural or extravesical extension of the tumour and simultaneous ureteral tumours. The survival rates were calculated using the Kaplan-Meier method and compared using the log-rank test. The end point of recurrence-free survival was the date of disease recurrence or in case of a nondiagnosed disease recurrence the date of tumour-related death. All cases of death unrelated to the tumour were censored. To compare the tumour stages (organ-confined [pt < 3a pn0 M0], non organ-confined [pt > 2b pn0 M0], and lymph node positive [ptx pn+]), a x 2 test was performed. To determine independent prognostic factors, we performed a multivariate analysis using a Cox regression model with respect to the following potential influencing factors: pt stage, pn stage, grade, hydronephrosis, age at time of cystectomy, sex, and interval between tumour diagnosis and cystectomy. Significance was defined as p < Results Hydronephrosis was evident in 133 of the selected 788 patients (16.9%), of whom 108 (13.7%) had unilateral hydronephrosis (left kidney in 53; right kidney in 55 patients) and 25 patients (3.2%) presented with bilateral hydronephrosis. Fortythree patients had a tumour at the ureteral orifice. In six of these cases an additional intramural/ extravesical tumour extension with ureteral compression/infiltration was observed. One patient had an additional ureteral tumour. Seventy-eight patients had hydronephrosis from intramural/ extravesical tumour extension with ureteral compression/infiltration. Nineteen patients had ureteral tumours simultaneously with the bladder cancer. In the preoperative setting, the urologist only knows about the presence of a tumour at the ureteral orifice, whereas the true etiology of hydronephrosis can only be determined from the cystectomy specimen. We analysed two different settings: The prediction of the postoperative course after cystectomy depending on the true etiology of hydronephrosis. For this analysis the most advanced tumour extent had to be considered. Cases with tumour both at the ureteral orifice and intramural/extravesical tumour extension with ureteral compression/infiltration were rated as intramural/extravesical tumour extension with ureteral compression/infiltration. We compared 36 patients (27.1%) with tumour at the ureteral orifice alone with 78 patients (58.6%) who had ureteral infiltration/compression by intramural/ extravesical tumour extension and with 19 patients (14.3%) who had additional ureteral tumours. The prediction of the tumour extension from the TURBT findings. In this case the 43 patients (32.3%) who had a tumour at the ureteral orifice (irrespective of the further tumour extent) were compared with 90 patients (67.7%) who had tumour-free orifices. Comparing the distribution of tumour stages in hydronephrotic and non-hydronephrotic patients, we observed a significantly higher rate of non organ-confined, lymph node-negative tumours (27.8% vs. 17.3%) and lymph node-positive tumours

4 european urology 51 (2007) Table 2 Distribution of tumour stages in patients with or without hydronephrosis Without hydronephrosis With hydronephrosis No. % No. % Organ-confined (pt < 3a pn0) Non organ-confined (pt > 2b pn0) Lymph node metastasis (pn+) Total x 2 test: ; p < (34.6% vs. 14.8%) in hydronephrotic patients (x 2 test: ; p < 0.001; Table 2). In patients with bilateral hydronephrosis, 12 (48%) patients showed an organ- confined and 4 (16%) a non organ-confined tumour growth, and 9 (36%) presented with positive lymph nodes. The majority of these patients had large papillary tumours occluding both ureteral orifices. Organ-confined tumour stages were significantly more frequent in patients with a tumour involving the ureteral orifice (58.3%) than in those with an intramural or extravesical ureteral compression or infiltration (29.5%; x 2 test: 9.517; p = 0.009; Table 3). No significant difference in tumour distribution was evident in patients with a simultaneous ureteral tumour compared to patients with an intramural or extravesical ureteral compression or infiltration (x 2 test: 2.403; p = 0.301; Table 3). A urologist s decision for bladder cancer therapy is usually based on the T stage of the TURBT specimen, its grading, and the presence of metastases. In hydronephrotic patients the urologist usually knows by endoscopy whether the tumour is located at the ureteral orifice or not. Because patients with a tumour at the ureteral orifice show a higher probability of organ-confined disease, we compared two groups that may be distinguished prior to cystectomy: hydronephrosis without tumour at the ureteral orifice versus no hydronephrosis or hydronephrosis with tumour at the ureteral orifice. This stratification increased the prognostic impact of hydronephrosis (x 2 : ; p < 0.001). These data suggest that a hydronephrotic patient with a histologically proven bladder TCC without tumour at the orifice has a 70% risk of having non organ-confined disease and a 40% risk of having lymph node metastases. In patients with tumour at the ureteral orifice, these risks are significantly lower at 41.7% and 19.4%, respectively. The risk is lowest in patients without any hydronephrosis (32.1% and 14.8%, respectively). A multivariate analysis determined hydronephrosis to be an independent prognostic factor for the recurrence-free survival for patients with bladder cancer besides the pt and the pn classification (x 2 : 10.13; p = ; Table 4). The recurrence-free survival rate was significantly higher for patients without hydronephrosis (69.3%) than for hydronephrotic patients (42.3%; p < 0.001; Fig. 1). Furthermore, we observed a distinct but not significant recurrencefree survival benefit for patients without hydronephrosis for all different tumour stages (Table 5). Five years after cystectomy no significant stagedependent difference was detectable in the recurrence-free survival rates of patients with a tumour with ureteral involvement, with a tumour with an Table 3 Distribution of tumour stages based on etiology of hydronephrosis Ureteral tumour Tumour at the ureteral orifice Intramural/extravesical infiltration/compression of the ureter No. % No. % No. % Organ-confined (pt < 3a pn0 M0) Non organ-confined (pt > 2b pn0 M0) Lymph node status (pn+) Total Tumour at the ureteral orifice versus intramural/extravesical infiltration/compression of the ureter: x 2 : 9.517; p =

5 694 european urology 51 (2007) Table 4 Multivariate analysis with a Cox regression model of potential influencing factors on recurrence-free survival Influencing factor x 2 p Maximum pt stage (pta/is/1 vs. pt2a/b vs. pt3a/b vs. pt4a/b) < Lymph node status (pn0 vs. pn+) < Preoperative hydronephrosis Age at cystectomy (<60 yr vs yr vs. >69 yr) ns ns Sex ns ns Interval from primary diagnosis to cystectomy (<90 d vs d vs. >365 d) ns ns Year of cystectomy ( vs vs ) ns ns Cystectomy for primary vs. recurrent tumour ns ns Grading (1 vs. 2 vs. 3 vs. 4) ns ns Fig. 1 Recurrence-free survival rates for the overall population. intramural/extravesical ureteral compression, or in patients with a secondary ureteral tumour causing hydronephrosis (Figs. 2 and 3). In this study 32 renal units (4.1%) were removed due to hydronephrosis. In two patients with a bladder-sparing attempt, nephroureterectomy was performed between the first onset of bladder cancer and radical cystectomy. In 29 cases a nephroureterectomy was performed simultaneously with the radical cystectomy. In all cases the renal function of the removed kidney was <15% of the overall renal function (0 13%). One patient developed a secondary ureteral carcinoma requiring nephroureterectomy 53 mo after cystectomy. In all other hydronephrotic cases the follow-up examinations revealed preserved renal function without hydronephrosis. An ileal neobladder as urinary diversion was performed in 54.9% (73 of 133) of the hydronephrotic patients and in 79.7% (522 of 655) of the patients without hydronephrosis. Table 5 Recurrence-free survival rates at 5 yr following radical cystectomy by log-rank test Without hydronephrosis With hydronephrosis p pta/is/1 pn0 89.4% 74.3% ns pt2a/b pn0 76.8% 69.7% ns pt3a/b pn0 65.7% 36.6% ns pt4a/b pn0 47.6% 25.0% ns ptx pn+ 23.1% 14.0% ns Total 72.1% 45.9% <0.001 ns = not significant.

6 european urology 51 (2007) Fig. 2 Tumour-specific survival according to the etiology of hydronephrosis in patients with organ-confined tumours (pt < 3a pn0 M0). Fig. 3 Tumour-specific survival according to the etiology of hydronephrosis in patients with non organ-confined tumours (pt < 2b pn0 M0). 4. Discussion To analyse the prognostic impact of hydronephrosis in patients with bladder cancer, it is important to define a homogeneous population with one concise treatment regimen and with a precise pathologic staging. To date, no published study on this topic fulfills these criteria. Two major studies on the prognostic value of hydronephrosis in bladder cancer patients were published from the University of Southern California and the University of Tel Aviv [1,5]. Haleblian et al. analysed a series of 415 patients with TCC who underwent radical cystectomy with a well-defined preoperative and postoperative workup regimen [5]. A small percentage of patients with advanced stages received adjuvant chemotherapy. In this population 22.7% of patients presented with unilateral and 5.3% with bilateral hydronephrosis at time of diagnosis. The rates of hydronephrosis (unilateral and bilateral) in our series (13.7% and 3.2%) were lower than those described by Haleblian et al. Similar to our study, Haleblian reported lower rates of organ-confined tumours in patients with hydronephrotic kidneys. Nevertheless, their distribution of tumour stages showed a distinct shift towards advanced disease compared to our series for patients with and without hydronephrosis. As the major end point for the study, Haleblian et al. defined overall survival, although overall survival is influenced by many factors such

7 696 european urology 51 (2007) as age, comorbidity, and secondary malignancies. Because no information is provided on these factors, their impact cannot be estimated. Cancer-specific survival would have been a more appropriate parameter because it excludes this influence. Recurrence-free survival even excludes the influence of different treatment modalities for tumour recurrence. Therefore the value of this study is limited. Haleblian et al. found a significant correlation for bilateral hydronephrosis and advanced disease, and further, for bilateral hydronephrosis and overall survival [5]. These findings were not reproducible in our study. The rate of organ-confined tumours among patients with bilateral hydronephrosis was 9% in the study by Haleblian but 48% in our series. Although the number of patients with bilateral hydronephrosis in our study is comparable with that of Haleblian, we consider it too small to perform a reliable subgroup analysis. Leibovitch et al. analysed 122 patients with invasive bladder TCC; 82% underwent radical cystectomy [1]. In 9% bladder-sparing surgery was performed. The remaining 9% were managed conservatively. Only half the patients had a PLND. Therefore, tumour staging was based on a complete histopathologic work-up in only a limited number of cases. Due to these different modalities this study investigated an nonhomogenous population. To investigate the prognostic impact on tumour stage the patients in the study of Leibovitch et al. were subgrouped into different pathologic tumour stages and different tumour grades. The lymph node status was not investigated. In the series of Leibovitch et al. ureteral obstruction was evident in 54.1% of patients [1]. This rate is far higher than in any other study published [2,3,8,9]. In this study bilateral hydronephrosis was a frequent finding, with no rate provided. It did not confer a significant prognostic impact compared to unilateral hydronephrosis. This finding was supported by our data. A significant cause-specific survival benefit was observed for patients without hydronephrosis compared to those with hydronephrosis: 65.9% versus 32.2% at 5 yr. The authors also found a significant difference of serum creatinine levels between patients with and without hydronephrotic kidneys (2.4 mg/dl vs. 1.1 mg/dl), and performed a simultaneous nephrectomy in 13.9% during surgical procedures. In our study, only 3% of patients with hydronephrosis had serum creatinine levels higher than 2.0 mg/dl and the rate of nephrectomy for hydronephrosis was 4.1%. This may be attributable to different diagnostic regimens. Leibovitch et al. investigated the impact of the tumour localisation in hydronephrotic patients on the tumour stage [1]. In all hydronephrotic cases without involvement of the ureteral orifices the tumour extended to the external layer of the detrusor muscle (>pt2a). An infiltration of adjacent organs in pt4 tumours was evident in 58%. In most cases with intravesical involvement of the ureteral orifices they found low-stage disease. This finding is supported by our data. A new aspect of our study is the additional group of patients with simultaneous ureteral tumours. Several studies analysed prognostic factors, including hydronephrosis, for survival of patients with bladder cancer [10 17]. In a series of 531 patients having cystectomy for TCC, Thrasher et al. used univariate analysis and identified preoperative hydronephrosis as a significant prognostic factor for cancer-specific survival in patients with muscle-invasive TCC [10]. Of a total of 217 patients, 156 underwent neoadjuvant radiation therapy and 61 curative radiation therapy for other malignancies. The total rate of hydronephrosis was not mentioned. In a multivariate approach hydronephrosis was not determined as an independent prognostic marker for cancer-specific survival. Because radiation therapy of the true pelvis may cause fibrotic changes of the ureteral and periureteral tissue, this may have caused a significant bias of the hydronephrosis rate. To clarify the prognostic impact of hydronephrosis, we therefore recommend the analysisofpatientswithoutanyneoadjuvant therapy. Yang et al. included 310 patients with a locally advanced urothelial carcinoma of the urinary bladder [11]. Of these, 159 patients (51.3%) underwent radical cystectomy with PLND and 151 (48.7%) were treated with an organ-preserving approach. In a univariate analysis, hydronephrosis was 1 of 11 statistically significant markers for disease-specific survival. Again in a multivariate analysis hydronephrosis was determined not to be an independent marker. Scrimger et al. reported data on a series of 184 patients with muscle-invasive TCC, who underwent curative treatment, including radiotherapy, surgery, and radiotherapy combined with surgery. In 40% of patients a neoadjuvant or adjuvant chemotherapy regimen was administered [12]. Patients treated with radiation therapy were more likely to have hydronephrosis. Univariate and multivariate analyses showed hydronephrosis to be an independent prognostic factor for overall survival. The value of these studies for the evaluation of the impact of hydronephrosis is limited. These studies analysed patient populations with very nonhomogenous treatment modalities. Because

8 european urology 51 (2007) the indication for different therapeutic options may have been influenced by the presence of hydronephrosis, this bias must not be underestimated. Urologists search for easy-to-assess and highly reproducible prognostic markers for choosing the appropriate therapy in patients with bladder cancer. Our data show that hydronephrosis is a parameter fulfilling these requirements, especially when cases with tumours at the ureteral orifice are excluded. 5. Conclusions From the data of our homogenous cystectomy-only series in patients with TCC, we conclude that hydronephrosis is a significant marker for advanced disease, as long as the tumour does not involve the ureteral orifice. Hydronephrosis is an independent prognostic marker for recurrence-free survival and tumour-specific survival. The presence of hydronephrosis without involvement of the ureteral orifices may be a helpful marker for the decision-making process regarding neoadjuvant and adjuvant therapy. References [1] Leibovitch I, Ben-Chaim J, Ramon J, Madjar I, Engelberg IS, Goldwasser B. The significance of ureteral obstruction in invasive transitional cell carcinoma of the urinary bladder. J Surg Oncol 1993;52:31 5. [2] Bowles WT, Silber I. Carcinoma of the bladder: a computer analysis of 516 patients. J Urol 1972;107: [3] Greiner R, Skaleric C, Veraguth P. The prognostic significance of ureteral obstruction in carcinoma of the bladder. Int J Radiat Oncol Biol Phys 1977;2: [4] Denkhaus H, Crone-Münzenbrock W, Huland H. Noninvasive ultrasound in detecting and staging bladder carcinoma. Urol Radiol 1985;7: [5] Haleblian GE, Skinner EC, Dickinson MG, Lieskovsky G, Boyd SD, Skinner DG. Hydronephrosis as a prognostic indicator in bladder cancer patients. J Urol 1998;160: [6] Volkmer BG, Kuefer R, Bartsch Jr G, et al. Effect of a pt0 cystectomy specimen without neoadjuvant therapy on survival. Cancer 2005;104: [7] Bochner BH, Nichols PW, Skinner DG. Overstaging of transitional cell carcinoma: clinical significance of lamina propria fat within the urinary bladder. Urology 1995;45: [8] Lang EK. The roentgenographic assessment of bladder tumors: a comparison of the diagnostic accuracy of roentgenographic techniques. Cancer 1969;23: [9] Hatch TR, Barry JM. The value of excretory urography in staging bladder cancer. J Urol 1986;135:49. [10] Trasher JB, Frazier HA, Robertson JE, Dodge RK, Paulson DF. Clinical variables which serve as predictors of cancerspecific survival among patients treated with radical cystectomy for transitional cell carcinoma of the bladder and prostate. Cancer 1994;73: [11] Yang MH, Yen CC, Chen PM, et al. Prognostic-factorsbased risk-stratification model for invasive urothelial carcinoma of the urinary bladder in Taiwan. Urology 2002;59: [12] Scrimger RA, Murtha AD, Parliament MB, et al. Muscleinvasive transitional cell carcinoma of the urinary bladder: a population-based study of patterns of care and prognostic factors. Int J Radiat Oncol Biol Phys 2001;51: [13] Takashi M, Murase T, Mizuno S, Hamajima N, Ohno Y. Multivariate evaluation of prognostic determinants in bladder cancer patients. Urol Int 1987;42: [14] Ficarra V, Dalpiaz O, Alrabi N, Novara G, Galfano A, Artibani W. Correlation between clinical and pathological staging in a series of radical cystectomies for bladder carcinoma. BJU Int 2005;95: [15] Matos T, Cufer T, Cervek J, Bornstnar S, Kragelj B, Zumer- Pregelj M. Prognostic factors in invasive bladder carcinoma treated by combined modality protocol (organsparing approach). Int J Radiat Oncol Biol Phys 2000;46: [16] Sengelov L, Kamby C, Schou G, von der Maase H. Prognostic factors and significance of chemotherapy in patients with recurrent or metastatic transitional cell cancer of the urinary tract. Cancer 1994;74: [17] Golding RP, van Zanten TE, Tierie AH, Batterman JJ, Hart G. Intravenous urography as a prognostic indicator in vesical carcinoma. Cancer 1987;60: Editorial Comment Wiking Mansson, Department of Urology, University Hospital Lund, Lund, Sweden wiking.mansson@urokir.lu.se Muscle-invasive bladder cancer is a killer and speedy work-up is necessary to obtain the maximum amount of information before vigorous countermeasures are instituted. This is most often cystectomy with lymphadenectomy, the extent of which should be more rather than less. Indications for adjunct treatment, that is, neoadjuvant and adjuvant chemotherapy, rest on the prognostic information obtained from tumour stage and lymph node status. However, these are insufficient and additional prognostic markers are needed. Although transcription-mediated amplification (TMA) and cdna microarray analyses provide promising means of obtaining prognostic markers, they are not of clinical use as of today. In this carefully analysed large single-centre series on the incidence of hydronephrosis in patients

9 698 european urology 51 (2007) undergoing cystectomy, findings from previous studies are clearly corroborated: hydronephrosis is more common in non organ-confined disease, including N+ disease. Furthermore, it is an independent prognostic factor for survival. The authors show that the most serious situation is the presence of hydronephrosis in the absence of visible tumour growth at the ureteric orifice. In such cases the risk of node-positive disease was calculated at 40%. In the series of 788 patients analysed, 583 had muscle-invasive disease, among whom 140 had positive nodes, that is 24%. The vast majority of these patients underwent a standard lymphadenectomy, that is, to the iliac bifurcation. A more extensive node dissection combined with better staging measure, that is, identification of sentinel nodes [1], would most likely increase this figure, and thus substantially increase the risk of nodal disease, should hydronephrosis be present. The importance of hydronephrosis is further reinforced by the finding of accompanying significantly lower recurrence-free survival in this study. Remembering that clinical understaging in invasive bladder cancer is common [2], the finding of hydronephrosis indicates advanced tumour stage with ominous outcome. These patients should be considered for extensive lymphadenectomy at cystectomy, preceded or followed by chemotherapy. References [1] Liedberg F, Chebil G, Davidsson T, Gudjonsson S, Månsson W. Intraoperative sentinel node detection improves nodal staging in invasive bladder cancer. J Urol 2006;175: [2] Ficarra V, Dalpiaz O, Alrabi N, Novara G, Galfano A, Artibani W. Correlation between clinical and pathologic staging in a series of radical cystectomies for bladder carcinoma. BJU Int 2005;95: