Ethical consideration: The study was carried out under an institutional review board-approved protocol and all patients signed written informed consent before surgery.
Michael Rink M.D., Department of Urology, University Medical Centre Hamburg-Eppendorf, Martinistraße 52, D-20246 Hamburg, Germany. Email: firstname.lastname@example.org
Objectives: Radical cystectomy (RC) remains a complex procedure in older patients. Perioperative morbidity can be significant and it can represent a limitation for its indication in this population. The aim of the present study was to evaluate the outcomes of RC in elderly patients from a large single-center cohort.
Methods: A total of 447 patients who underwent RC between 1996 and 2009 at our institution were considered. Patients were stratified by age (≤70 vs >70 years). Logistic regression analyses were carried out comparing both groups regarding clinical, perioperative and histopathological findings, as well as complications according to the modified Clavien system and survival.
Results: Data of 390 patients were available for the analysis. Of these, 265 (67.9%) versus 125 (32.1%) patients were <70 versus ≥70 years-of-age. The median age was 61 and 75 years, respectively. In the elderly, ASA score (P < 0.001), delay between transurethral resection of the bladder (TURBT) and RC (P = 0.004), and number of perioperative blood transfusions (P = 0.002) were significantly higher. Additionally, a clear trend towards higher stages (pT3–4) was observed (P = 0.04). However, complications, and overall and cancer-specific mortality were not increased in older patients. Finally, age was identified as a significant risk factor for upstaging (P = 0.04). Upstaging between TURBT and final histopathology in patients <70 versus ≥70 years occurred in 45% versus 58%, respectively (P = 0.03).
Conclusions: RC is equally feasible in older patients without increasing morbidity or mortality. On the contrary, older patients have a higher risk of significant upstaging and advanced stages at final histopathology. These findings suggest that RC should neither be delayed in nor withheld from elderly patients.
Urothelial carcinoma (UC) of the bladder is one of the most common cancer entities and causes of cancer-specific death in the Western world. Paralleling the increase in life expectancy, the number of bladder cancer cases is steadily increasing, particularly in elderly patients.1,2
However, medical decision making (MDM) and treatment of elderly patients is hampered twofold:
First, MDM in elderly patients can be aggravated, as clinical understaging seems to be more prevalent compared with younger patients.3 As clinical outcome after radical cystectomy (RC) is mainly driven by pathological stage, stage discrepancy after extirpative surgery for bladder cancer remains an unsolved problem.4,5
Second, perioperative complications as a result of pre-existing comorbidities, as well as procedure-inherent risks, are not negligible. In particular, older people patients have been reported to be at greater risk of suffering from perioperative complications and morbidity.6,7 However, various contemporary RC series reported acceptable complication rates even in older patients.8,9
According to the current European Association of Urology guidelines on muscle-invasive bladder cancer, controversy still persists regarding advanced age and the carrying out of radical cystectomy. A recommendation regarding this issue is currently missing. Because of this insecurity, clinicians today are still reluctant to recommend RC in elderly patients with muscle invasive and recurrent, high-grade, superficial bladder cancer.
To address this controversy, we investigated patients <70 and ≥70 years-of-age regarding their clinical and histopathological findings, as well as perioperative complications and survival.
Between April 1996 and May 2009, clinical and pathological data were prospectively collected in 449 consecutive patients. All patients had transurethral resection of the bladder (TURBT) confirmed, recurrent advanced or muscle-invasive UC and underwent radical cystectomy with urinary diversion. Of these, 59 patients were excluded because of missing TURBT (n = 7), postoperative complications (n = 42) and perioperative data (n = 10). None of the investigated patients had received neo-adjuvant chemotherapy. At the time of cystectomy, no patient showed evidence of distant metastatic disease, based on a staging including computed tomography of the thorax, abdomen and pelvis. Radical cystectomies and perioperative management were based on a standard protocol comparable to protocols previously described by others.10
For the remaining patients, data on sex, American Society of Anesthesiologists physical status classification score (ASA score), delay in days between TURBT and RC (n = 297), TURBT histology, final pathological tumor stage and grade, status of lymph node involvement, operative time, blood loss, units of blood transfusion, perioperative complications, and time to death from date of diagnosis were available for our analyses. Complications were defined and graded according to the established modified Clavien system.11 Postoperative complications were recorded within 90 days of surgery. The highest grade experienced by each patient was used for analysis. In the case of no complications, the grade is 0. Minor complications are defined as grade 1 and 2 complications, major complications as grade 3–5 (see Table 2).
Table 2. Summary of complication grades and severity level after radical cystectomy according to the modified Clavien classification stratified by age
Patients <70 years, n (%)
Patients ≥70 years, n (%)
Grade 0, no event observed; grade 1, use of oral medications or bedside intervention; grade 2, use of intravenous medications, total parenteral nutrition, enteral nutrition or blood transfusion; grade 3, interventional radiology, therapeutic endoscopy, intubation, angiography or operation; grade 4, residual and lasting disability requiring major rehabilitation or organ resection; grade 5, death of patient.
Modified Clavien grade
P = 0.15
Minor complication (grade 1–2)
P = 0.93
Major complication (grade 3–5)
Survival data were retrospectively gathered and available in 301 of the 390 remaining patients.
All pathological samples were examined by a dedicated uropathologist at University Hospital Hamburg-Eppendorf, Hamburg, Germany. Bladder specimens were classified according to the WHO/ISUP classification for pathological grading12 and the tumor, lymph nodes and metastasis (TNM) system.13
Patients were categorized into pathologically downstaged (lower pathological than clinical stage and negative lymph nodes), same stage (same clinical and pathological stage and negative lymph nodes) and pathologically upstaged (higher pathological than clinical stage or positive lymph nodes) according to previously used classifications.14
Frequently, patients >65 years are considered “elderly”.15 For our statistical analysis, we stratified our cohort according to <70 versus ≥70 years. This cut-off was chosen on the basis of previous studies to ensure sufficient comparability. Furthermore, we carried out a subgroup analysis of patients <70 versus ≥75 years to address the most contemporaneous aging demographic development of our society.
Descriptive analyses consisted of χ2-test and Fisher's exact test for categorical variables. Differences in variables with a continuous distribution across dichotomous or ranked categories were assessed using the Mann–Whitney U-test. Univariable and multivariable logistic regression analyses to predict upstaging were carried out. Probabilities of cancer-specific and overall survival were calculated using the Kaplan–Meier method. Differences among groups were assessed using the pairwise log–rank test. All tests were two-sided and a P-value <0.05 was considered to be statistically significant. All statistical analyses were carried out using spss Software (Version 13; SPSS, Chicago, IL, USA).
Characteristics of all RC patients are shown in Table 1. Overall, 265 patients (67.9%) were <70 years and 125 patients (32.1%) ≥70 years, with a median age of 61 versus 75 years. Sex and year of surgery (not shown) were homogenously distributed. ASA score was significantly higher in older patients (P < 0.001), representing patients with higher anesthesiological risks. Furthermore, the delay between TURBT and RC was significantly longer in older patients (median 30 vs 50 days; P = 0.004). Other clinical and perioperative data, such as operation time, blood loss and time of hospitalization, were not significantly different. Conversely, the number of transfused blood units per patient was higher in older patients compared with the younger patients (P = 0.002), and incontinent urinary diversion was more often used in this group (P < 0.001).
Table 1. Characteristics of 390 radical cystectomy patients stratified by age (<70 vs≥70 years)
Clinical or histopathological parameter (%)
Age <70 years
Age ≥70 years
(n = 265)
(n = 125)
ASA, American Society of Anesthesiologists physical status classification; RC, radical cystectomy; SD, standard deviation; TURBT, transurethral resection of bladder tumor.
Mean ± SD
59.5 ± 7.5
75.5 ± 4.4
Delay between TURBT and RC (in days)
(297 patients in analysis)
n = 204
n = 93
Mean ± SD
34.8 ± 21.1
64.9 ± 63.1
Final histopathological result after RC
Localized stages (pTa–pT2)
Advanced stages (pT3–pT4)
Lymph node status
Operation time (min)
Mean ± SD
392 ± 94
383 ± 91
Blood loss (mL)
Mean ± SD
1387 ± 827
1380 ± 811
Blood transfusion (units per patient)
Mean ± SD
3.2 ± 2.5
3.9 ± 3.7
Time of hospitalization (days)
Mean ± SD
22 ± 8.9
22 ± 11.8
Continent (neobladder, Mainz pouch I)
Incontinent (conduit, ureterocutaneostomia)
Interestingly, overall TURBT histology, final RC pathological stage and lymph node status did not show statistically significant differences between both groups (Table 1). However, stratification of final RC specimens in localized stages (pTa–pT2) versus advanced stages (pT3–pT4) showed that advanced histopathological findings were significantly more frequent in patients ≥70 years (P = 0.04).
Overall, early complications after surgery occurred in 225 (57.7%) cases, consisting of 168 (74.7%) minor and 57 (25.3%) major complications, respectively. Complication grading (P = 0.15) and severity level (P = 0.93) were equally distributed between patients <70 and ≥70 years (Table 2). Most common were grade 1 and 2 complications as a result of infectious diseases (urinary tract or pulmonary), prolonged bowel stasis (subileus) and lymphoceles. Overall, four patients (1%) died postoperatively. These patients were younger than 70 years and three of the four had severe comorbidities (ASA 3 and 4). Two of these patients died as a result of acute heart failure, one patient with severe systemic diseases (status post-heart transplantation, immunosuppressant therapy, multiple sclerosis, obstructive lung disease and renal insufficiency) died as a result of septic pneumonia and one patient died as a result of arteria basilaris thrombosis.
Overall, up- and downstaging was found in 51.3 and 17.7%, respectively (Table 1). Stratified to age groups, older patients showed significantly more upstaging (59.2 vs 47.5%; P = 0.03). In univariable analysis, age, delay between TURBT and RC, and TURBT histology were significant risk factors of upstaging (all P ≤ 0.05). In multivariable logistic regression analysis predicting significant upstaging, only age and TURBT histology remained independent risk factors (Table 3).
Table 3. Univariable and multivariate logistic regression models to predict upstaging between clinical stage and final histopathological result in 390 patients treated with radical cystectomy for bladder cancer treatment
Table 4 shows the results of a subgroup analysis of patients <70 vs >75 years. Using this different age threshold, our previous findings regarding patient characteristics were confirmed. Furthermore, younger patients showed significantly more frequent downstaging (19.6 vs 7.6%, P = 0.02).
Table 4. Subgroup analysis of patients <70 versus >75 years receiving a radical cystectomy for bladder cancer treatment
Clinical or histopathological parameter (%)
Age <70 years
Age >75 years
(n = 265)
(n = 66)
Grade 0, no event observed; grade 1, use of oral medications or bedside intervention; grade 2, use of intravenous medications, total parenteral nutrition, enteral nutrition or blood transfusion; grade 3, interventional radiology, therapeutic endoscopy, intubation, angiography or operation; grade 4, residual and lasting disability requiring major rehabilitation or organ resection; grade 5, death of patient. ASA, American Society of Anesthesiologists physical status classification; RC, radical cystectomy; SD, standard deviation; TURBT, transurethral resection of bladder tumor.
Follow-up data on survival after RC were available in 301 patients (70.4% <70 years vs 29.6% ≥70 years). Mean follow-up time was 42 months. Overall, 129 men died within the follow up. In Kaplan–Meier analysis addressing overall survival, no statistically significant difference between both age groups was recorded (P = 0.16; Fig. 1a). Similarly, no statistically significant difference between both age groups was recorded when cancer-specific survival was assessed (P = 0.22, Fig. 1b). Figure 2 shows Kaplan–Meier analysis comparing age groups <70 versus >75 years regarding overall und cancer-specific survival. Unlike the results in the entire cohort, cancer-specific survival was significantly shorter in the subgroup of patients >75 years (P = 0.02).
Advanced age and incidence of bladder cancer are closely related. Specifically, age is widely accepted as one of the most important single risk factors for developing bladder cancer.2 The present study is the largest contemporary European single center report that evaluated the association of advanced age (≥70 vs <70 years) and adverse histopathological outcome after radical cystectomy. Furthermore, clinical and perioperative risk factors were analysed, which might influence the preoperative decision planning of radical cystectomy in elderly patients. Finally, early complications and intermediate term survival outcomes were recorded.
Our data show a worse outcome regarding final pathological results in elderly compared with younger patients. Elderly patients are significantly more often diagnosed with advanced pathological stage (pT3–pT4, P = 0.04). One in two patients ≥70 years had extravesical disease (51.2%). In fact, our subgroup analysis using a second age threshold (<70 vs >75 years.) confirmed this observation. Of the patients aged >75 years, 56.1% showed advanced disease. Our observation is in line with others who also reported that advanced age was associated with advanced pathological stage, higher pathological tumor grade and a higher likelihood for extravesical disease and lymphovascular invasion.3,16 An intuitive explanation for these results might reside in a reluctance to carry out radical surgery in older comorbid patients. Indeed, in our analysis, older patients had significantly higher ASA scores compared with younger patients (P < 0.001). An increased ASA score does not only translate into a higher anesthesiological risk,17 but also represents a proxy for the presence of more comorbidities. Therefore, as a result of optimization of perioperative risk factors, a longer duration between diagnosis and RC can often be necessary in these patients. However, a longer delay between diagnosis and definitive treatment might result in disease progression and worse survival, as different groups previously reported.18,19 The present findings show a significantly prolonged median delay between TURBT and indication for RC and definitive surgery in elderly patients of 7 weeks (P = 0.004). Despite this delay of less than 12 weeks18,19 in older patients, advanced stages were significantly more often recorded. Therefore, we feel that stage progression is not only attributable to delaying surgery. In fact, another reason might be the change in the biological potential of the tumor or a decrease in the host's defence mechanisms associated with advancing age, leading to a more aggressive tumor growth.2 A combination of delay and biology might play a key role in progression to advanced stages with increasing age.
Our findings further show that one in two patients (51.3%) is upstaged at radical cystectomy. These results are in line with previous reports. Stage divergences were already reported in the early 1970s,20 and range in different studies between 27% and 62%.5 Even in contemporary series, upstaging occurs in as many as 50%.4,5,14 In fact, stage divergence remains an unsolved staging and prognostic problem, because it has been reported to be associated with worse overall and cancer-specific survival.4,5 Several reasons can be advanced to explain persistent stage discrepancy, including TUR sampling error,21 delay in the interval from TUR to radical cystectomy18,19 and poor sensitivity of preoperative staging tools.22 Furthermore, the present findings show that, specifically, older patients are at a significantly higher risk of being upstaged compared with younger patients (P = 0.04). This effect is even more pronounced in patients aged >75 years, where two-thirds of patients (67%) were upstaged. Our multivariable analysis confirmed age (P = 0.04) as a risk factor for upstaging. However, our findings are in contradistinction to Hollenbeck et al., who did not find significant stage divergences stratified by age.4 Conversely, Nielsen et al.3 confirmed age as a significant risk factor for upstaging, and furthermore found that advanced age was associated with extravesical disease, and all-cause and cancer-specific mortality in a large cohort of 888 patients. As an explanation for this age effect, a prolonged delay between diagnosis and radical cystectomy has been discussed. Specifically, it has been reported that a time interval of >12 weeks is related to a worse outcome and advanced pathological stages.18,19 In the present analysis, we observed a significantly prolonged delay between TURBT and RC in older patients (P = 0.004). However, the delay between TURBT and RC did not remain independently associated with upstaging within our multivariable risk analysis. On the contrary, we confirmed that age and TURBT histology are independently associated with upstaging (Table 3).
Radical cystectomy not only remains the gold standard treatment for advanced bladder cancer,23 but also remains a burdening surgical approach with significant rates of perioperative complications and morbidity.6,7,11 In reports of contemporary RC cohorts, mean operation time, blood loss and postoperative hospital stay ranged from 287 to 370 min, 750 to 1200 mL and 7 to 24 days, respectively.8,24,10 In their report of radical cystectomy in the elderly, Tilki et al. compared patients of an age <75 and ≥75 years. Interestingly, they found a reduced operation time in elderly compared with younger patients. They did not find significant differences in blood loss. However, the blood transfusion rate was significantly higher in the elderly (12.9 vs 5.7%).10 The findings in our present study are in line with this study (Table 1). We also recorded a shorter operation time in older patients, most likely attributable to application of incontinent urinary diversion. Our blood transfusion rate was significantly higher (P = 0.002) in the elderly, although intraoperative blood loss was equal to younger patients. In this context, we hypothesize that blood transfusions in older patients are more commonly used to protect these patients from adverse cardiac events; for example, cardiopulmonal decompensation or myocardial infarction.
Because of increased pre-existing comorbidities in the elderly, several studies reported a significant relationship between age at RC and perioperative complication rates (22–64%).6,7,11,25 We recorded a high overall early postoperative complication rate (57%) defined according to the modified Clavien system. Interestingly, when stratified to age groups, no statistically significant difference was detected for neither minor (grades 1–2) nor major (grades 3–5) complications within 90 days after RC (Table 2). These results are in contradistinction to Shabsigh et al., who identified advanced age as a risk factor for major complications in their multivariate analysis.11 On the contrary, a more recent report by Tilki et al. corroborated our findings. Similarly, they reported no differences in complications comparing younger and older patients (21 vs 22%; P = 0.8).10 Despite true differences, discordant findings might be related to the time period of RC and RC technique, or, most importantly, varying complication records where standardized recording systems are highly recommended.11
Taken together, it is important to note that the influence of age on RC perioperative and prognostic outcomes is an ongoing debate.3,9,26–28 In the present study, we neither identified any difference in overall (P = 0.16) nor in cancer-specific survival (P = 0.22). Therefore, we feel that in case of non-metastatic muscle invasive bladder cancer, RC should not be withheld to healthy, elderly patients.
This report is not devoid of limitations. A lot of patients for cystectomy are referred to our center from outside. Therefore, the time interval between first TURBT and final surgery was not available in several cases, introducing a potential bias. Furthermore, the investigated time period of more than a decade is long, where significant technical and perioperative management improvements have been experienced for which we did not correct our analyses. Furthermore, all analyses used the most recent TNM staging system at diagnosis, respectively. A central pathological re-review would have been desirable. Finally, follow-up information was limited and not available in all cases.
In conclusion, radical cystectomy represents a demanding surgical procedure with a non-negligible number of perioperative complications. However, our findings show that RC is feasible in elderly patients without a higher risk of perioperative mortality and morbidity. Specifically, older patients suffer from an increased risk to be understaged and have advanced histopathological stages at the time of RC. Therefore, we suggest that radical cystectomy should not be withheld to fit-enough and older patients, because reluctance in decision-making can translate into a devastating adverse prognostic outcome.