EFFECT OF ROUTINE REPEAT TRANSURETHRAL RESECTION FORSUPERFICIAL BLADDER CANCER: A LONG-TERM OBSERVATIONAL

STUDY

MARC-OLIVER GRIMM,* CHRISTINE STEINHOFF, XENIA SIMON, PHILIPP SPIEGELHALDER,ROLF ACKERMANN AND THOMAS ALEXANDER VOGELI

From the Department of Urology, Heinrich-Heine University (M-OG, CS, XS, PS, RA, TAV), Dusseldorf and Department of ComputationalMolecular Biology, Max Planck Institute for Molecular Genetics (CS), Berlin, Germany

ABSTRACT

Purpose: We determined the long-term outcome in patients with superficial bladder cancer (Taand T1) undergoing routine second transurethral bladder tumor resection (ReTURB) in regard torecurrence and progression.

Materials and Methods: We performed an inception cohort study of 124 consecutive patientswith superficial bladder cancer undergoing transurethral resection and routine ReTURB (83)between November 1993 and October 1995 at a German university hospital. Immediately aftertransurethral resection all lesions were documented on a designed bladder map. ReTURB of thescar from initial resection and other suspicious lesions was performed at a mean of 7 weeks.Patients were followed until recurrence or death, or a minimum of 5 years.

Results: Residual tumor was found in 33% of all ReTURB cases, including 27% of Ta and 53%of T1 disease, and in 81% at the initial resection site. Five of the 83 patients underwent radicalcystectomy due to ReTURB findings. The estimated risk of recurrence after years 1 to 3 was 18%,29% and 32%, respectively. After 5 years 63% of the patients undergoing ReTURB were stilldisease-free (mean recurrence-free survival 62 months, median 87). Progression to muscleinvasive disease was observed in only 2 patients (3%) after a mean observation of 61 months.

Conclusions: These data suggest a favorable outcome regarding recurrence and progression inpatients with superficial bladder cancer who undergo ReTURB. ReTURB is suggested at least inthose at high risk when bladder preservation is intended.

KEY WORDS: bladder, bladder neoplasms, urethra, transurethral resection, outcome, recurrence, progression

Transurethral bladder tumor resection (TURB) is the ini-tial step in the treatment of bladder cancer. Complete tumorresection is mandatory for adequate staging and it serves asdefinitive therapy, at least for superficial tumors. However,there is growing evidence that TURB is incomplete in asignificant number of cases.1�4 It may contribute to the highnumber of recurrences observed in up to 50% to 80% ofpatients, of which most occur during year 1 after TURB.5

Furthermore, staging error due to incomplete resectionmight be related to the progression rate of 14% to 53% insuperficial tumors.4, 6–8

To improve treatment results repeat TURB (ReTURB) 2 to8 weeks after the initial operation has been recommended.1, 4

Retrospective studies support this approach, describing afrequency of up to 75% tumor detection by ReTURB for Taand T1 bladder cancers.4 However, the frequency of residualtumor after TURB and the ability of ReTURB to control forcomplete resection has been widely neglected by the urolog-ical community. Although a negative impact of residual tu-mor after TURB has been anticipated by several groups,ReTURB remains unaccepted since to our knowledge a (fa-vorable) long-term outcome has never been reported. To beconsidered beneficial to the patient ReTURB should result inprolonged time to recurrence as well as decreased recurrenceand eventually a decreased progression rate.5

We evaluated the outcome in a cohort of patients followedfor at least 5 years after treatment with routine ReTURB ina prospective study for quality control of TURB for bladder

cancer. We defined the frequency of residual tumor and stag-ing error after TURB prospectively and evaluated the rate ofrecurrence and progression after ReTURB in a long-termobservational study.

PATIENTS AND METHODS

Treatment strategy. Between November 1, 1993 and Octo-ber 31, 1995 patients undergoing TURB at the department ofurology, Heinrich-Heine University were enrolled in a pro-spective study for quality control by routine ReTURB todefine the frequency and location of residual tumor aftertransurethral resection. In general, resection was performedas a differential transurethral resection with separate sam-pling of the tumor as well as of the base and the mucosaadjacent to the tumor.9 After TURB the surgeon documentedthe location of the tumor on a designed bladder map.

ReTURB of the scar of the initial resection and other sus-picious lesions was performed at a mean of 7 weeks. Allpatients with histologically confirmed superficial bladdercancer underwent ReTURB except those with small, singlefocus, low grade papillary tumors (TaGI) and those with poorperformance status. If tumor was detected during ReTURB,another ReTURB was recommended given the same exclu-sion criteria. Tumor stage, grade, focality, volume, concomi-tant carcinoma in situ, the result of postoperative cytologyand adjuvant intravesical instillation (performed at the dis-cretion of the treating urologist) were recorded. Informedconsent for the treatment strategy was obtained from eachpatient.

Patient and tumor characteristics. Overall 214 patients

Accepted for publication March 7, 2003.* Corresponding author: Department of Urology, Heinrich-Heine

University, Moorenstrasse 5, 40225 Dusseldorf, Germany.

0022-5347/03/1702-0433/0 Vol. 170, 433–437, August 2003THE JOURNAL OF UROLOGY® Printed in U.S.A.Copyright © 2003 by AMERICAN UROLOGICAL ASSOCIATION DOI: 10.1097/01.ju.0000070437.14275.e0

433

(155 men and 59 women) with a mean age of 71 years (range23 to 98) underwent a total of 257 TURBs during the studyperiod. Table 1 lists the stage and grade of 194 histopatho-logically confirmed bladder cancers. Multifocal tumors and asolitary lesion were observed in 62% and 38% of the patients,respectively. There were 62% of cases with newly diagnosedbladder cancer, while 15% had a first, 9% had a second and14% had multiple recurrences. Transitional cell carcinomarepresented 94%, squamous cell carcinoma represented 3%and adenocarcinoma represented 1% of tumors. The distri-bution of tumor characteristics within different subgroups ofsuperficial bladder cancers analyzed (T2 or greater and car-cinoma in situ excluded) closely matched the overall studypopulation.

Followup. Patients were followed every 3 to 6 months withcystoscopy. All patients undergoing ReTURB were followeduntil recurrence or death, or a minimum of 5 years. Only 1patient was lost to followup 12 months after TURB. Meanobservation, including all recurrences, was 61 months.

The primary end points of this study were the time tohistologically confirmed bladder cancer recurrence and timeto progression to muscle invasive disease. Actuarialrecurrence-free survival rates and estimated times to recur-rence reported were considered beginning from the lastTURB. Recurrence rates yearly were not considered becausetreatment after the first recurrence was performed on anindividual basis and was not part of the study protocol. Sincethe protocol did not distinguish between primary and recur-rent superficial bladder cancers, recurrent disease during thestudy period was considered a separate case.

Statistical evaluation. All statistical evaluations were per-formed using commercially available software. Standard sta-tistical procedures were used, such as the Pearson chi-squareand Mann-Whitney U tests. Categorical variables with apossible correlation with tumor detection by ReTURB accord-ing to standard univariate tests were subjected to logisticregression analysis. The disease free-interval, defined as thetime until first recurrence since the last transurethral resec-tion, was evaluated to construct Kaplan-Meier survivalcurves. Patients who died of other causes or were withoutrecurrence at the last followup were censored. The log ranktest was used to stratify patients with respect to prognosticfactors, and for comparison between ReTURB and the cohortof patients that did not undergo repeat resection. All reportedp values are 2-sided.

RESULTS

Overall 214 patients underwent a total of 257 TURBsduring the study period. Smooth muscle was obtained in97.4% of primary transurethral resections. In 63 cases pre-malignant or nonmalignant findings were observed afterTURB, for example dysplastic lesions, chronic cystitis (fre-quently in patients with a history of bladder cancer andintravesical instillation), bilharziosis or nephogenic ade-noma. Bladder cancer was histopathologically confirmed in194 cases, including carcinoma in situ in 2, Ta in 90, T1 in 34and muscle invasive (T2 or greater) in 68 (table 1).

Figure 1 shows the treatment scheme for the 124 cases ofsuperficial bladder cancer (Ta and T1), which form the basisof this analysis. ReTURB was performed for 63 of 90 Tatumors and for 20 of 34 T1 tumors. According to the studyprotocol 12 patients with small unifocal TaG1 tumors and 8

with poor performance status did not undergo ReTURB. Ofthe patients scheduled for ReTURB 20 refused a secondintervention. One patient underwent immediate cystectomydue to extensive superficial disease considered to be incur-able by TURB (final histology T3a). TURB was considered tobe incomplete by the surgeon in 9 of 124 cases (7%).

Tumor detection rate by ReTURB after first transurethralresection. Excluding incomplete resection, residual tumorwas found in 33% of all ReTURBs, including 27% of Ta and53% of T1 tumors (table 2). Residual tumor was located at theprimary site only in 46% of cases and at another site only in19%. In 35% of cases the scar after TURB and at least 1 othersite were affected, most frequently in T1 tumors.

Univariate analysis identified tumor stage (Pearson chi-square p � 0.04) and grade (p �0.04) as predicting factors forresidual tumor. Other factors evaluated include focality ormultifocality, carcinoma in situ, tumor volume, postoperativecytology and adjuvant instillation, although none of thesefactors correlated significantly with tumor detection by Re-TURB. Using residual tumor at ReTURB as a dependentvariable in a logistic regression model with the variablestumor stage and/or grade the only significant variable iden-tified was tumor grade (Wald statistic p �0.02).

Progression in stage or grade was noted in 4 and 2 cases(8%), respectively. Muscle invasive disease was discovered in3 (4%) of these patients. In an additional case urothelialcarcinoma invasive into the prostate gland was detected byReTURB.

Long-term outcome after ReTURB. Of 83 ReTURBs per-

TABLE 1. Distribution of stage and grade of 192 bladder cancers*

Stage No. G1 No. G2 No. G3

Ta 34 50 6T1 20 14T2 or greater 8 60

* In 2 additional cases carcinoma in situ (Tis) was the only lesion detected.

FIG. 1. Treatment scheme for 124 cases of superficial bladdercancer. ReTURB was performed in 83 patients at mean of 7 weeks(ReTURB and eventually repeat ReTURB at 8.1 weeks).

REPEAT TRANSURETHRAL RESECTION FOR SUPERFICIAL BLADDER CANCER434

formed 5 were excluded from followup evaluation. Two of the3 patients with stage migration to muscle invasive diseaseunderwent cystectomy, while 1 received no immediate fur-ther treatment due to advanced age. In addition, the patientwith concomitant invasive urothelial cancer of the prostateand one with persistent T1G3 carcinoma detected by repeatReTURB were treated with radical cystectomy.

Tables 3 and 4 show the characteristics of the remaining 78cases. Except 1 patient who was lost to followup after 12months all who underwent ReTURB were followed until re-currence or death, or a minimum of 5 years. Recurrence wasobserved in 30 of the 78 patients (38%) treated with ReTURBat a mean followup of 46 months. Followup of patients with-out recurrence was 59 months. The estimated risk of recur-rence after years 1 to 3 was 18%, 29% and 32%, respectively.After 5 years 63% of patients undergoing ReTURB were stilldisease-free, equivalent to an actuarial mean recurrence-freesurvival of 62 months (fig. 2).

Stage and grade were predictors of time to recurrence.Only 2 of the 17 patients with a stage T1 tumor had recur-

rence, resulting in longer disease-free survival than in thosewith Ta tumors. Seven patients (9%) had stage or gradeprogression, while progression to muscle invasive diseasewas observed in 1 patient (1%) only at the first recurrence.

Recurrence-free survival of patients undergoing TURBonly. ReTURB was not performed in 41 patients, including 12with small unifocal TaG1 disease, 8 with poor performancestatus, 1 who underwent immediate cystectomy and 20 whorefused ReTURB. Four patients with incomplete initialTURB and 1 with a history of muscle invasive bladder cancerwere excluded from followup analysis. Table 3 lists the char-acteristics of these 36 patients.

Overall 19 of the 36 patients (53%) had recurrence. Theestimated risk of recurrence after years 1 to 3 was 21%, 57%and 61% compared with 18%, 29% and 32% in the ReTURBgroup, respectively. The difference in recurrence-free sur-vival between patients undergoing ReTURB versus TURBonly was statistically significant (log rank test p �0.03, fig.2).

DISCUSSION

Retrospective studies consistently show a high frequencyof residual tumor in 30% to 75% of cases when ReTURB isperformed for superficial bladder cancer.1, 4 This retrospec-tive data has been criticized since only a few studies distin-guish between incomplete and complete primary TURB, and

TABLE 2. Tumor stage, grade and frequency of residual tumor on routine ReTURB in 78 cases of superficial bladder cancer*

Stage G1 G2 G3 No. Residual Tumor/TotalNo. (%)

No. Ta 2/18 13/35 1/6 16/59 (27)No. T1 4/12 6/7 10/19 (53)No. residual tumor/total No. (%) 2/18 (11) 17/47 (36) 7/13 (54) 26/78 (33)

* Incomplete primary resections excluded.

TABLE 3. Characteristics by treatment group of 114 patientsconsidered for followup analysis

No. ReTURB(%)

No. No ReTURB(%)

Total pts 78 36Age:*

Younger than 50 19 (24) 3 (8)50–59 28 (36) 7 (19)60–69 23 (30) 15 (42)70 or Older 8 (10) 11 (31)

Gender:Male 64 (82) 25 (69)Female 14 (18) 11 (31)

Bladder Ca history:†No (primary) 51 (65) 14 (39)Yes (recurrence) 27 (35) 22 (61)

Stage:‡Ta 61 (78) 25 (69)T1 17 (22) 11 (31)

Grade:‡G1 19 (24) 15 (42)G2 47 (60) 15 (42)G3 12 (16) 6 (16)

Additional Ca in situ§ 3 (4) 2 (6)Tumors:‡

Single 49 (63) 23 (64)Multiple 29 (37) 13 (36)

Wt (gm):�2 or Less 27 (35) 14 (39)2–10 44 (57) 21 (58)Greater than 10 6 (8) 1 (3)

Tumor stage, grade, weight, number of tumors (single or multiple) andfrequency of concomittant carcinoma in situ did not differ significantly be-tween the groups.

* Mean age for patients undergoing ReTURB or observation was 67 and 75years, respectively (t test p � 0.001).

† Mean recurrence-free interval before study entry was 22 months in eachgroup and recurrence within 3 months was noted in 5 and 4 cases of theReTURB and no ReTUR groups, respectively (Pearson chi-square, p � 0.01).

‡ According to the study protocol the no ReTURB group included 12 smallunifocal TaG1 tumors resulting in a significant higher proportion of TaG1 andunifocal G1 tumors (Pearson chi-square p � 0.02) as well as a trend of overallmore G1 tumors (p � 0.06) compared to patients undergoing ReTURB.

§ Adjuvant instillation therapy was administered in a similar proportion ofpatients in each group (30% and 31%).

� Mean weight of the resected tumor of patients undergoing ReTURB was 4.7gm compared to 2.4 gm in the no ReTURB group (median 2.0 gm each, notsignificant).

TABLE 4. Tumor characteristics in 78 patients followed afterReTURB

Stage No. G1(%)

No. G2(%)

No. G3(%)

Total No.(%)

Ta:Unifocal 7 (9) 25 (32) 5 (6) 37 (47)Multifocal 12 (15) 11 (14) 1 (1) 24 (30)

T1:Unifocal 10 (13) 2 (3) 12 (16)Multifocal 1 (1) 4 (5) 5 (6)

Totals 19 (24) 47 (60) 12 (15) 78

FIG. 2. Kaplan-Meier recurrence-free survival in patients under-going ReTURB vs no ReTURB.

REPEAT TRANSURETHRAL RESECTION FOR SUPERFICIAL BLADDER CANCER 435

the accumulation of patients at risk, for example with highstage, grade or multifocal disease, undergoing ReTURB hasbeen assumed.

In this prospective study, ReTURB was performed rou-tinely except in patients with small, high grade, solitary,mucosa confined tumors (TaG1). Patients with TURBs re-ported by the surgeon to be incomplete were excluded fromanalysis. Consistent with previous retrospective data,1�4 tu-mor was detected by ReTURB in about a third of the patientswith superficial Ta or T1 bladder cancer. Residual tumor wasmost frequently located at the site of initial TURB (81%),strongly suggesting incomplete initial surgery.

It seems reasonable to assume that the high frequency ofresidual tumors after intended complete TURB for superfi-cial bladder cancer is responsible for a significant number ofrecurrences. It is in accordance with molecular biologicalstudies showing that most bladder tumors have commongenetic alterations suggesting a single progenitor cell.10, 11

However, the observation of the clonal development of blad-der cancer has been mainly attributed to intraluminal seed-ing, neglecting the possibility of incomplete surgery. Accord-ingly intravesical immunotherapy and chemotherapy havebeen extensively studied as adjuvant treatment to preventbladder cancer recurrence and progression.5, 6

To our knowledge the impact of routine second transure-thral resection on the long-term outcome of patients in re-spect to the recurrence rate, time to recurrence and progres-sion as well as the need for cystectomy has not beeninvestigated up to date. Based on the current literature about48% and 40% of patients who undergo TURB only are ex-pected to be recurrence-free by 3 and 5 years, respectively.12

In contrast, estimated recurrence-free survival of patientsundergoing ReTURB was 62 months (median 87), equivalentto a 3 and 5-year recurrence-free survival rate of 68% and63%, respectively.

Assumption of a favorable impact on the outcome of super-ficial bladder cancer by ReTURB treatment is further sup-ported by significant prolongation of recurrence-free survivalcompared with patients undergoing TURB only in this study(fig. 2). This observation must be considered with cautionsince this study was not randomized and patient character-istics in the group designated TURB only were similar butnot balanced compared with the ReTURB group. Indeed,except for patient age and a higher proportion of recurrencesthe TURB only group included several tumors with compa-rably favorable prognostic factors, for example a third soli-tary TaG1 tumors (table 3).

Studies addressing recurrence rates after intravesical che-motherapy, focusing mainly on patients at low risk, showworse or at best similar outcomes compared with the cohortundergoing ReTURB in this report.6, 13–16 Notably some in-stillation protocols included followup cystoscopy after 4 to 6weeks to ensure complete initial resection, leading to Re-TURB in up to 33% of cases.13, 16, 17 However, a correlation ofcystoscopic and histological findings of ReTURB showed that40% of neoplastic lesions are invisible endoscopically in theTURB scar.1

Another major aim in the treatment of superficial bladdercancer is the prevention of disease progression and a de-creased need for radical cystectomy.5 The progression ratevaries widely in different studies due to patient selection. Forexample, in a European Organization for Research andTreatment of Cancer Genitourinary Group study 14% of pa-tients mainly at low risk had progression,5, 18, 19 while Cook-son et al focused on those at high risk and found a 53%progression rate and a 34% rate of bladder cancer death inthe long term.7 In this study progression to muscle invasivedisease was observed in only 1 patient (1%) undergoing Re-TURB who was followed until the first recurrence. Includinglater recurrences only 1 additional patient had muscle inva-sive disease during an observation of 61 months. Tumor

characteristics may explain the low number of patients af-fected, although the low number at risk was due to betterpatient selection by ReTURB. Based on ReTURB histologicalresults 5 patients (6%) would have been treated with cystec-tomy (1 was not eligible). Notably 2 of these patients as wellas those who had recurrence with muscle invasive diseaseduring followup had stage Ta G1-G2 disease at initial TURB.

ReTURB probably does not prevent progression but it al-lows early detection of patients at risk and muscle invasivedisease that is masked by incomplete resection. It seemsreasonable to conclude that ReTURB leads to better stratifi-cation since none of the patients at high risk (T1) who wererendered tumor-free by ReTURB had progression to muscleinvasive disease. It is further supported by the retrospectivestudy of Brauers et al, in which none of the patients withprimary T1G2 to 3 bladder cancer without residual tumorduring ReTURB (36%) had progression to muscle invasivedisease at 60 months of followup.20 According to these data itmay be assumed that cystectomy for stage T1 bladder canceris often performed too early, neglecting attempts at completeresection.20 However, given the fact of a 50% residual cancerrate after TURB for this high risk disease, ReTURB is sug-gested when bladder preservation is intended.

CONCLUSIONS

The results of this study suggest that a significant propor-tion of superficial bladder cancer recurrences might be due toincomplete surgery. ReTURB performed 4 to 8 weeks afterTURB eradicates this cause of recurrence and provides ex-cellent long-term recurrence-free survival. Furthermore, Re-TURB allows better stratification of risk for progression andit may decrease the need for radical cystectomy. Therefore,ReTURB is suggested, at least in patients at high risk, whenbladder preservation is intended. The positive long-term out-come in patients undergoing ReTURB in this study must beconfirmed in a prospective, randomized trial.

REFERENCES

1. Klan, R., Loy, V. and Huland, H.: Residual tumor discovered inroutine second transurethral resection in patients with stageT1 transitional cell carcinoma of the bladder. J Urol, 146: 316,1991

2. Kohrmann, K. U., Woeste, M., Kappes, J. and Alken, P.: DerWert der transurethralen Nachresektion beim oberflachlichenHarnblasenkarzinom. Akt Urol, 25: 208, 1994

3. Mersdorf, A., Brauers, A., Wolff, J. M., Schneider, V. and Jakse,G.: 2nd TUR for superficial bladder cancer: a must? J Urol,suppl., 159: 143, abstract 542, 1998

4. Herr, H. W.: The value of a second transurethral resection inevaluating patients with bladder tumors. J Urol, 162: 74, 1999

5. Kurth, K. H., Bouffioux, C., Sylvester, R., van der Meijden, A. P.,Oosterlinck, W. and Brausi, M.: Treatment of superficial blad-der tumors: achievements and needs. The EORTC Genitouri-nary Group. Eur Urol, 37: 1, 2000

6. Bouffioux, Ch., Kurth, K. H., Bono, A., Oosterlinck, W., BoekenKruger, C., De Pauw, M. et al: Intravesical adjuvant chemo-therapy for superficial transitional cell bladder carcinoma:results of 2 European Organization for Research and Treat-ment of Cancer randomized trials with mitomycin C and doxo-rubicin comparing early versus delayed instillations andshort-term versus long-term treatment. J Urol, 153: 934, 1995

7. Cookson, M. S., Herr, H. W., Zhang, Z.-F., Soloway, S., Sogani,P. C. and Fair, W. R.: The treated natural history of high risksuperficial bladder cancer: 15-year outcome. J Urol, 158: 62,1997

8. Schwaibold, H. E., Treiber, U., Kuebler, H., Leyh, H. andHartung, R.: Second transurethral resection detects histo-pathological changes worsening the prognosis in 25% ofpatients with T1 bladder cancer. J Urol, suppl., 163: 153,abstract 680, 2000

9. Bressel, M., Kemper, K. and Stadtler, F.: Requirements andtechnique of transurethral electroresection of bladder carci-noma. Urologe, 8: 73, 1969

REPEAT TRANSURETHRAL RESECTION FOR SUPERFICIAL BLADDER CANCER436

10. Sidransky, D., Frost, P., von Eschenbach, A., Oyasu, R.,Preisinger, A. C. and Vogelstein, B.: Clonal origin bladdercancer. N Engl J Med, 326: 737, 1992

11. Habuchi, T., Takahashi, R., Yamada, H., Kakehi, Y., Sugiyama,T. and Yoshida, O.: Metachronous multifocal development ofurothelial cancers by intraluminal seeding. Lancet, 342: 1087,1993

12. Millan-Rodriguez, F., Chechile-Toniolo, G., Salvador-Bayarri, J.,Palou, J. and Vicente-Rodriguez, J.: Multivariate analysis ofthe prognostic factors of primary superficial bladder cancer.J Urol, 163: 73, 2000

13. Oosterlinck, W., Kurth, K. H., Schroder, F., Bultinck, J.,Hammond, B., Sylvester, R. et al: A prospective EuropeanOrganization for Research and Treatment of Cancer Genito-urinary Group randomized trial comparing transurethral re-section followed by a single intravesical instillation of epiru-bicin or water in single stage Ta, T1 papillary carcinoma of thebladder. J Urol, 149: 749, 1993

14. The effect of intravesical thiotepa on tumour recurrence afterendoscopic treatment of newly diagnosed superficial bladdercancer. A further report with long-term follow-up of a MedicalResearch Council randomized trial. Medical Research CouncilWorking Party on Urological Cancer, Subgroup on SuperficialBladder Cancer. Br J Urol, 73: 632, 1994

15. Tolley, D. A., Parmar, M. K. B., Grigor, K. M., Lallemand, G. andthe Medical Research Council Superficial Bladder CancerWorking Party: The effect of intravesical mitomycin C onrecurrence of newly diagnosed superficial bladder cancer: a

further report with 7 years of followup. J Urol, 155: 1233, 199616. Krege, S., Giani, G., Meyer, R., Otto, T., Rubben, H. and Partic-

ipating Clinics: A randomized multicenter trial of adjuvanttherapy in superficial bladder cancer: transurethral resectiononly versus transurethral resection plus mitomycin C versustransurethral resection plus bacillus Calmette-Guerin. J Urol,156: 962, 1996

17. Flamm, J., Donner, G., Oberleitner, S., Hausmann, R. andHavelec, L.: Adjuvant intravesical mitoxantrone after trans-urethral resection of primary superficial transitional cell car-cinoma of the bladder. A prospective randomised study. Eur JCancer, 31: 143, 1995

18. Pawinski, A., Sylvester, R., Kurth, K. H., Bouffioux, C., Van DerMeijden, A., Parmar, M. K. B. et al: A combined analysis ofEuropean Organization for Research and Treatment of Can-cer, and Medical Research Council randomized clinical trialsfor the prophylactic treatment of stage TaT1 bladder cancer.J Urol, 156: 1934, 1996

19. Kurth, K., Tunn, U., Ay, R., Schroder, F. H., Pavone-Macaluso,M., Debruyne, F. et al: Adjuvant chemotherapy for superficialtransitional cell bladder carcinoma: long-term results of aEuropean Organization for Research and Treatment of Cancerrandomized trial comparing doxorubicin, ethoglucid andtransurethral resection alone. J Urol, 158: 378, 1997

20. Brauers, A., Buettner, R. and Jakse, G.: Second resection andprognosis of primary high risk superficial bladder cancer: iscystectomy often too early? J Urol, 165: 808, 2001

REPEAT TRANSURETHRAL RESECTION FOR SUPERFICIAL BLADDER CANCER 437