N38 How accurate we really are in predicting final stage of non-invasive TCC of the bladder when performing cystoscopy?

      Introduction and Objectives

      Approximately 75–85% of patients with bladder cancer present with disease confined to the mucosa (stage Ta-Cis) or submucosa (stage T1). The distinction, whether the bladder carcinoma is non-muscle invasive (NMIBC) or muscle invasive (MIBC) has a cardinal influence on further treatment. The pathological examination of post-TURBT specimen plays the key role in this context. The management of NMIBC became more complex with regard to initial investigation, treatment strategy, intravesical therapy and follow-up. The ability to estimate the tumor stage and grade accurately would be beneficial for patients. Therefore, there is a need to define, if cystoscopy alone can reliably identify tumor stage and grade. The aim was to assess the accuracy of visual staging (by stage and grade) of bladder cancer during cystoscopy. Thereafter, we evaluated the differences in predictability of more and less experienced urologists and analyzed the most common errors in tumor stage and grade identification.

      Material and Methods

      The records of 189 NMIBC-TURBT procedures performed in 164 patients (aged 29–99, av. 68) from 2007 to 2009 were collected. In all cases stage (T) and grade (G) were assessed by the treating surgeon and documented in operation protocol. Cystoscopic appearance of the tumor was digitally recorded. All data were blindly reevaluated by another two urologists. All clinical results were compared with final pathological examination. Intraobserver and interobserver variations were also noticed.


      Urologist predicted correctly both T and G in 60 out of 189 tumors – accuracy of only 31.7%. The accuracy in different pT and G stages were as follows: TaG1 – 25%, TaG2 – 0%, T1G1 – 6%, T1G2 – 66%. The predictibility of T was higher than G (53% vs.47%). The overdiagnosis (between TCC and T0) was noticed in 10 out of 11 patients. Overstaging and understaging between Ta and T1 were noticed in 63% and 24% of cases, respectively. The predictability differed between more and less experienced urologists and the accuracy was as follows: 50% and 27%, respectively.


      Our study revealed the lack of appropriate knowledge in the intraoperative assessment of tumor stage. Nowadays, the ability of an urologist to predict T and G depends on the clinical experience level. Therefore, a professional training process and teaching programme are necessary.