frozen-section examination


radical cystectomy


upper urinary tract (tumour)


carcinoma in situ.

The value of ureteric frozen-section examination (FSE) at the time of radical cystectomy (RC) for urothelial carcinoma remains a topic of debate. In patients undergoing RC for TCC of the bladder, upper urinary tract tumours (UUTTs) develop in 2–8%[1–9]. Several factors are associated with increased risk of developing UUTTs after RC for TCC of the bladder including high grade, multifocality of the bladder tumour, presence of carcinoma in situ (CIS) of the bladder, tumour involvement of the prostatic urethra, and intramural tunnel or juxtavesical ureteric involvement with CIS, or the presence of a distal ureter tumour [1–9]. To minimize the risk of recurrences at the uretero-intestinal anastomosis and subsequently in the remaining UUT, FSE of the ureteric margins at RC is advocated although its true value remains highly controversial. Proponents argue that a negative ureteric FSE helps to avoid recurrence at the uretero-intestinal anastomosis and individuals at risk can be followed closely [4,5]. Others argue that progressive proximal resection of the ureter until FSE is negative does not reduce the risk of UUT recurrence. Furthermore, the incidence of tumour recurrence at the uretero-intestinal anastomosis is low (≈1%) and most patients with UUTTs die of systemic disease rather than local failure [3,5,6].

The prevalence of ureteric CIS at RC has been reported to be 2–35%[3,5,6,10,11]. Herr and Whitmore [10] reviewed a series of 105 patients treated for TCC plus CIS of the bladder. The surgical RC specimens of 37 patients had evidence of CIS involving some or all portions of the intramural and resected distal ureter (≈10 cm). Step-sectioning techniques in these 37 patients revealed CIS in the intramural ureter in 81%, in the juxtavesical ureter in 68%, at the FSE site in 43%, and in the most proximal portion of the ureter resected before completion of the uretero-intestinal anastomosis in 14%. Contiguous ureteric mucosal involvement by CIS was seen in most of the cases, while only 17% presented with pagetoid tumour spread [10]. Data on the incidence of ureteric CIS in patients undergoing RC for TCC of the bladder without concomitant CIS is scant. Raj et al.[3] reported ureteric involvement in the RC specimen of 30% of patients with bladder CIS vs only 9% of patients without.

In more recent series the overall incidence of ureteric CIS was <8%[3,5,6]. Silver et al.[6] reported ureteric CIS in 7.7% of 401 patients, identified in 6.2% by FSE in the ureteric segment just proximal to the juxtavesical ureter and in the remaining 1.5% on permanent section in the redundant ureter segment resected before completion of the uretero-intestinal anastomosis. Raj et al.[3] found CIS on histopathological examination in 174 of 2579 distal ureters (6.7%) at the initial FSE site. We reported similar results in 805 patients treated with RC for TCC of the bladder with distal ureteric CIS being found on FSE in 3.7%, and in the corresponding permanent sections in 2.7%[5]. CIS was detected in the redundant ureter segment resected above the common iliac vessels on permanent section in only 0.1%[5]. It appears therefore that the overall incidence of ureteric CIS in patients undergoing RC is relatively low (<8%), occurring most often in the distal ureter and less often in the proximal ureter [3,5,6,10].

The prognostic relevance of concomitant CIS of the bladder as a risk factor for recurrent UUTTs in patients undergoing RC has been extensively investigated [1–5,10]. Current RC series report that ≈60% of patients were found to have CIS in the bladder [3,4,9]. Solsona et al.[12] reported a significantly higher recurrence rate of UUTTs in patients with CIS of the bladder (21.2%) vs patients without (2.3%; P < 0.001). In their patients UUT recurrences were located mainly in the distal ureter (88.2%) and seldom in the pyelocalyceal system (11.8%). A multivariate analysis by Raj et al.[3] showed that bladder CIS was a significant predictor of ureteric involvement on permanent section (odds ratio 3.99 [95% CI 1.87–8.51]; P < 0.001). Tran et al.[9] reported that 82% of patients with ureteric involvement at RC had CIS of the bladder and that the 5-year cumulative incidence for UUTT recurrence was 16% for patients with ureteric involvement compared with 5% for those without. These findings are in accord with our results showing recurrent UUTTs in 17% of patients with CIS of the ureter at FSE vs 3% in patients with negative FSE [5]. In contrast, Sanderson et al.[4] found no correlation between bladder CIS and UUTTs using multivariate analysis in 1069 RC patients followed for >10 years. According to their results, patients with tumour involvement of the urethra are at highest risk for recurrent UUTTs.

Time to tumour relapse after RC can be termed ‘early’, recurring after <3 years of disease-free survival or ‘late’, recurring after >3 years. Early relapse accounts for >80% of all tumour recurrences after RC and are located mainly in the viscera and bony pelvis, i.e. outside the remnant urothelium [2,4,6–8]. By contrast, late relapses typically affect the UUT and are usually detected by tumour-related symptoms such as gross haematuria and flank pain despite routine surveillance [1,2,4,7–9]. No correlation exists between the occurrence of symptoms and advanced UUTT stage or decreased survival [2,4,8,12]. There is also no evidence that complete extirpation of malignant ureteric disease, i.e. of CIS, by FSE during RC eliminates the risk for recurrent UUTTs [3–6,8,10]. The median overall survival has not differed in our patients with CIS of the ureters at FSE compared with those with negative FSE [5]. The prognosis for ureteric CIS depends therefore mainly on the prognosis for the primary bladder tumour.

In conclusion, ureteric CIS at RC is common, especially if CIS is present in the bladder, and is found mostly in the distal ureters. FSE of the ureters is recommended because of the high incidence of urothelial abnormalities if the ureters are divided distal to the cross of the common iliac artery at RC. This also reduces the risk of missing patients with unknown CIS of the bladder (60%) who are at increased risk for ureteric CIS and therefore also for UUTT [3–5,9]. Only if the ureters are divided at the level where they cross the common iliac arteries the probability of missing CIS at the resection margin is <1%[5]. Thus, FSE would seem to be of little benefit to the patient.


  1. Top of page

None declared.


  1. Top of page