Quality control in transurethral resection of bladder tumours

Authors


Harry W. Herr, 1275 York Avenue, New York, NY 10021, USA. e-mail: herrh@mskcc.org

Abbreviations
TUR(BT)

transurethral resection (of bladder tumour)

CIS

carcinoma in situ.

INTRODUCTION

Transurethral resection (TUR) is the essential surgical procedure used to diagnose, stage and treat primary and recurrent non-muscle-invasive bladder tumours. TUR of bladder tumours (TURBT) is both a diagnostic and a therapeutic operation. The initial TURBT has three main goals: to provide pathological material to determine the histological type and grade of bladder tumours, to determine the presence, depth, and type of tumour invasion (broad front or tentacular), and to remove all visible and microscopic superficial and invasive tumours. The findings are used to direct additional therapy, dictate follow-up schedules, and indicate prognosis.

Although ideally the initial TURBT should be thorough and complete, many factors confound the adequacy of resection. These include the multiplicity, extent and location of tumours, skill and perseverance of the urological surgeon, quality of specimens provided, and pathological analysis. Each tumour should be visibly and completely resected, if possible, and delivered to the pathology laboratory as one specimen, orientated so longitudinal histological sections through the tumour show intact overlying and surrounding mucosa contiguous with underlying lamina propria, and at least superficial muscle. This might not be possible for larger neoplasms, and many tumours are submitted to pathology in pieces, where poor orientation of specimens, crush and cautery artefacts, and presence of inflammation, can hamper adequate histological evaluation, especially of invasive tumours. The urologist (not the pathologist) is responsible for assigning primary tumour stage, based on an assessment of the tumour during cystoscopy, and providing a visibly complete resection, verified by the pathological findings. Despite the inherent pitfalls in assessing TUR specimens, an accurate pathological evaluation depends on the quality of tumour and tissue specimens provided by the urologist, and patient outcome is largely determined by the quality of resection.

Although TURBT is one of the most common urological procedures, it is a difficult operation to do well, even using modern video-endoscopy, especially for multiple papillary or sessile tumours associated with carcinoma in situ (CIS). Nonetheless, all overt, suspicious and reddened areas in the bladder should be removed and submitted as separate specimens for pathology. It is not necessary to include muscle deep to obvious low-grade papillary tumours, nor completely resect muscle-invasive neoplasms in cases where cystectomy is advised.

Although recent studies stress the importance of the quality of cystectomy and pelvic lymph node dissection on the outcome of advanced bladder cancer, there is a paucity of information about quality control of TURBT. For example, of the 2410 patients with papillary tumours entered into seven randomized trials by the European Organisation for Research and Treatment of Cancer, tumour recurrence rates at the first follow-up cystoscopy varied from 7% to 45% among different institutions. Such wide variability in the frequency of tumour recurrence was unexplained by other factors assessed, such as tumour type or treatment, and investigators concluded that the quality of TUR done by individual surgeons was most likely responsible [1]. We discuss how the quality of TURBT can be evaluated and improved.

HOW TO ASSESS THE QUALITY OF TURBT?

We could measure the quality of TURBT by asking three questions: was the resection complete? (staging accuracy); was deep muscle present in the specimen and identified by the pathologist?; and how often do tumours recur at the site of a previous TURBT? As a corollary, we suggest two methods of how TURBT might be improved; by an initial fractionated wide resection of all tumours, and a contemporary re-staging (second) TUR.

STAGING ACCURACY OF THE FIRST TUR

Incomplete resection and understaging of bladder tumours is well known because of the stochastic nature of TURBT. Table 1 shows the results of a second TURBT 2–6 weeks after an initial TURBT in 150 consecutive patients evaluated for localized bladder tumours [2]. A significant proportion (76%) was found to have residual tumour. Of 96 patients with superficial (stage pTa, pT1, CIS) tumours, only 25% had no tumour left; 31% had residual noninvasive tumour; 15% had persistent lamina propria invasion; and 29% were upstaged to muscle invasion. There was an incomplete initial TUR in 49% of stage pT1 tumours having no muscle submitted in the first TUR specimen, compared with 14% when muscle was identified. Table 1 also shows reported series of repeat TUR of stage pT1 bladder tumours [3–9]. Residual invasive cancer was present in 15–53% of cases; another 4–29% were upstaged to muscle invasion.

Table 1. 
Comparison of bladder tumour stage after a first and second TUR, and reported values after a second TUR of pT1 tumours
TUR or referenceN patientsStage at second TUR, %
pT0pTa/CISpT1pT2-3
Stage at first TUR
CIS2030402010
pTa182839285
pT15822262428
muscle3526312914
no muscle2317171749
pT2542213659
Totals150114 (76)   
pT1 tumours
[3]46 15262
[4]60 17245
[5]4235172424
[6]28 185329
[7]766711158
[8]52 16174
[9]19 374319

We updated our results with routine repeat TUR in 1312 patients with papillary tumours, according to tumour stage and grade (unpublished data; Table 2). Although half the patients with low-grade tumours had residual disease, only 5% were high grade, and none was invasive. However, most patients staged on the initial TUR as having high-grade noninvasive papillary tumours had persistent disease after their second TUR, and 15% were invasive. Of 701 patients with pT1 tumours, 25% had residual pT1 cancer left behind after the first TUR. Upstaging to muscularis propria invasion varied directly with whether or not muscle was present in the first TUR specimen. The conclusion of these data is that an initial TUR often fails to remove all tumours in the bladder.

Table 2.  Re-staging pTa and pT1 bladder tumours
Tumour typeN patients% pathology on re-TUR
pT0pTa LGpTa HG/CISpT1pT2
  1. LG, low-grade; HG; high-grade.

pTa LG2154946500
pTa HG39635050105
pT1701220232530
muscle421250312915
no muscle280200152045

Bimanual examination is also an integral component of accurate staging. This should be done both before and after resection. Any mass or induration palpable after TUR indicates deep muscle or extravesical tumour spread, and thus unlikely to have been completely removed.

IS MUSCLE PRESENT IN TUR SPECIMENS?

The pathologist should record whether deep muscle is present in submitted specimens, considering that fat occurs within muscularis mucosa traversing the lamina propria. A recent pathology review found that muscularis propria was missing in up to 51% of TUR specimens submitted by urologists [10]. Many of these were low-grade papillary tumours and the absence of muscle can be justified, but in 26% of invasive tumours, muscle specimens were not submitted or identified. In our series, 40% of T1 tumours had no muscle included in the initial TUR specimen. Further, when a second TUR specimen contained muscularis propria, we showed that understaging of T1 tumours (based on cystectomy specimens) is negligible [11]. The presence of negative muscle is an important quality control measure indicating a complete resection.

HOW OFTEN DO TUMOURS RECUR AT INITIAL TUR SITES?

Even after optimal surgical treatment by experienced urologists, immediate mitomycin C treatment, and further intravesical therapy, bladder tumours commonly recur, usually in the first year [12]. In half the cases, tumours are found at the same sites of resection during the first follow-up cystoscopy [13]. In one series, the repeat TUR was positive in 33% of cases; 18% recurred within a year, and in 81% tumour was present at initial TUR sites [9]. In another study of patients evaluated after starting BCG therapy after an initial TUR, same-site recurrences were found at the first follow-up cystoscopy and TUR biopsy in 54% of high-grade Ta and 67% of T1 tumours [14].

The most likely source of early recurrences is tumour present at the peripheral margins or base of original TUR sites. Evidence for this is illustrated by a study in which 35% of 462 TURs had residual tumour on deep resection of the tumour base and at least 2 cm lateral to visible tumours [15]. The extended TUR found incomplete initial resections in 13% of pTa, 35% of pT1 and 56% of pT2 tumours. Original-site recurrences provide further compelling evidence of incomplete removal of tumours during initial TUR, suggesting that the quality of a TUR might be improved by wide local resection of primary tumours followed by a re-staging (second) contemporary TUR of high-grade tumours.

IMPROVING THE QUALITY OF TUR

Local control and accurate assessment of pathological extent of disease (stage) is only possible if each tumour is resected completely. Sessile or nodular tumours are more likely than papillary tumours to be high-grade and invasive, but exceptions are common. To remove and accurately stage bladder tumours, they must be resected widely and deeply, at least for measurable lesions. How wide and how deep has practical implications for tumour staging and treatment, and requires considerable judgement, experience and skill. Figure 1 shows a sessile tumour resected at its peripheral margin A. The resection is then extended laterally into normal mucosa for 2 cm to margin B, to ensure complete wide excision of the lesion and to detect adjacent CIS. Figure 2 shows how different patterns of pT1 tumours affect margins of resection. Tumour A has broad-front invasion and tumour B has tentacular invasion. Resection of tumour A around the A margins results in complete resection of all visible and invasive components of the tumour. Resecting out to margin B verifies complete resection by obtaining microscopically negative margins. For tumour B, resection from margin A to A, or margin B, leaves tumour behind because of finger-like submucosal invasion extending laterally beyond visible margins of the tumour. Complete resection of tumour B requires extending the resection another 1 or 2 cm out to margin C. A complete TUR removes not only the visible epithelial portion of tumours, but also the invisible microscopic mucosal or invasive tumour, including deep muscle. Muscle should be clearly identified during the resection (Fig. 2).

Figure 1.

TURBT; the margins of resection.

Figure 2.

TUR of pT1 bladder tumours.

In cases of multiple papillary tumours, the larger tumours are resected completely as described above for histology, but smaller tumours (<0.5 cm) and adjacent mucosa that might harbour CIS can be fulgurated using a roller-ball electrode. Experienced urologists can usually distinguish low-grade papillary tumours from high-grade or invasive tumours to permit fulguration of smaller lesions and aggressive resection of high-risk tumours [16].

Another method to improve TUR quality is to classify the resection as R0, R1 or R2, based on histological evidence of tumour from each resection margin (periphery and depth of tumour) submitted separately from the tumour, where R0 indicates microscopic negative margins, R1 microscopic positive tumour residual, and R2 macroscopic tumour left behind. Only R0 is considered a complete TUR; R1 and R2 are incomplete resections. Urologists do not routinely record TUR in this manner, except for those participating in bladder-sparing protocols for muscle-invasive tumours, where adequacy of the initial resection is probably the important variable predicting the response to chemotherapy [17]. We suggest urologists adopt the R-method of reporting as an internal control to measure the quality of their TURs.

For all suspected invasive carcinomas, and in cases of muscle-invasive tumours being considered for bladder-sparing, a maximum, aggressive TUR is required. Deep resection of the detrusor is commenced only after the exophytic tumour has been excised, and tumour invading the bladder wall is resected in layers, avoiding deep excavation is one place, to maintain a clear view of the whole resection area. Continuous-flow irrigation facilitates the resection of large lesions. The TUR is considered complete when normal glistening yellow fat is seen between deep muscle fibres or perivesical tissues. Invasive tumour is usually firm and easy to resect, whereas uninvolved ‘normal’ fat is difficult to cut with the resectoscope. This is a useful finding indicating a complete tumour resection when corroborated by a negative histological examination of separate deep muscle specimens.

VALUE OF A RE-STAGING TUR

Although the first TUR of a small, single and well-defined tumour usually suffices, many tumours are multiple, poorly defined at the margins, and associated with CIS, features which become blurred because of tissue swelling and bleeding as the TUR proceeds, attempting to remove extensive disease. Here lies the value of a second, or re-staging TUR, performed a few weeks later when oedema created by the first resection has had a chance to subside. We have already documented that a re-staging TUR improves staging accuracy, but another TUR is also likely to provide better therapy of non-muscle-invasive disease if all macroscopically visible residual and suspicious tumour is re-resected. Recent level 1 and 2 evidence shows that compared to the initial TUR, a re-staging TUR reduced the 3-year recurrence rate by 42–58% and improved the response to BCG therapy by 24–35%[18]. A randomized trial comparing one vs two TURs in which all patients received adjuvant mitomycin C therapy showed a 38% recurrence-free survival advantage among patients after the second vs only one TUR [12]. High recurrence rates in patients who did not have a repeat TUR were due to residual tumour left behind after the initial TUR, and as both groups received mitomycin C, the authors concluded that intravesical chemotherapy could not compensate for inadequate TUR.

Even patients treated by experienced urologists can have a high percentage of persistent carcinoma after an initial TUR. Of 214 patients undergoing re-staging TUR for pTa and pT1 tumours, carcinoma was still present in 37% treated by senior urologists and in 26% treated by urologists in training [19].

The pathological findings of a re-staging TUR might also identify patients with superficial bladder cancer who are at high risk of early tumour progression [20]. Among 710 patients with papillary neoplasms, including 352 with pT1 tumours, 76% who had residual pT1 tumours on re-staging TUR progressed to muscle invasion, vs 14% having no cancer or noninvasive (pTa, CIS) tumour. Further, most patients with persistent pT1 tumour on a second TUR, even if they responded initially to BCG therapy, eventually developed worse disease. We suggest that residual invasive carcinoma (pT1) on a re-staging TUR might help to identify patients who need immediate radical cystectomy [21].

COMPLICATIONS OF TURBT

TURBT is associated with a low (5%) overall complication rate [22]. The frequency of complications is higher with large (>5 cm) tumours, multiple tumours, and tumours in the dome of the bladder. The most common complication, occurring in 2.3–3.5% of cases, is haemorrhage [23,24]. Delayed bleeding can largely be avoided by meticulous haemostasis, insisting on clear urine from the catheter after the procedure, and reducing bladder spasms. (A caveat: if brisk bleeding develops in the recovery room, and blocks the catheter, return the patient to the operating room to evacuate clots and fulgurate bleeders. The patient can be discharged on the same day, avoiding spending days in hospital receiving painful continuous bladder irrigation).

The most serious complication, occurring in 1.3–3.5% of cases, is bladder perforation [25,26]. Most are small, extravesical perforations, easily managed by several days of catheter drainage. Intraperitoneal perforations are rare, but require open surgical repair if they are gaping and associated with significant extravasation of fluid and urine despite catheter drainage.

Although tumour seeding is a concern when the bladder is perforated, documented cases of extravesical pelvic disease after TURBT are rare. Clinically ‘silent’ extravasation probably occurs during many procedures. One study reported that 58% of patients had contrast medium extravasation after a TUR, based on a comparison of cystograms before and after surgery, but none had evidence of extravesical tumour seeding over 2 years of follow-up [25]. TUR of invasive bladder tumours does not appear to disseminate urothelial tumour cells in the peripheral circulation, nor is there evidence that the procedure upstages (i.e. positive nodes) the disease at cystectomy [26]. In a recent review of 1589 cases of radical cystectomy done after a re-staging TUR, there was local recurrence in 3.5% (all in patients with advanced extravesical tumours and none among patients with organ-confined disease) [27].

CONCLUSIONS

TUR is both diagnostic and therapeutic for non-muscle-invasive bladder tumours. A re-staging TUR improves staging accuracy and achieves better local control of superficial and minimally invasive bladder tumours. Although TURBT is an essential procedure familiar to urologists, it is a difficult operation to do well. However, a properly performed and aggressive TUR is one of the most successful and powerful procedures available to the urologist, if the information provided is appropriately applied with full knowledge of its inherent limitations. Its success or failure, as well as that of subsequent treatments of noninvasive and invasive tumours localized to the bladder, directly depend on the information provided by well-executed TURs providing quality tumour specimens to permit an accurate pathological examination.

CONFLICT OF INTEREST

None declared.

Ancillary