Carcinoma, bladder

Cancer of the bladder has a high incidence in patients who smoke, consume artificial sweeteners, coffee or aromatic amines and who are exposed to cyclophosphamides. The diagnosis of bladder tumours includes clinical history, physical examination, urinalysis, cystoscopy, biopsy and careful bimanual examination under anaesthesia. The evaluation of bladder tumours includes an analysis of the size and site, and pattern of tumour growth. Tumours are classified into four growth patterns: in situ (noninvasive); papillary; infiltrating; and combined papillary and infiltrating.

For malignant bladder tumours, the staging system most commonly used today is TNM classification. The most important diagnostic technique for examination of the bladder has been and remains cystoscopy. Cross-sectional imaging, ultrasound (US) (Fig.1), computed tomography (CT), or magnetic resonance imaging (MRI) examinations are indicated only after clinical, conventional radiographic and endoscopic studies have provided a provisional diagnosis. Approximately 70% of patients with bladder carcinoma will present with superficial disease. At the time of diagnosis if the tumour is superficial, transurethral resection and multiple cystoscopic biopsies are sufficient not only for diagnosis but also for staging and definitive therapy. When deeply invasive tumours are detected, anatomical staging with US, CT or MRI is appropriate.

Imaging

Evaluation of anatomical extent of disease must include determination of the site, number, and growth pattern of individual bladder tumours, as well as depth of muscular and perivesical fat invasion and presence or absence of lymph node extension. Conventional studies, intravenous urography and cystograms may detect bladder cancer but are not used for staging (Fig.2) (Fig.3) (Fig.4). Enthusiastic reports as to the value of US evaluation of the depth of bladder invasion can be found, but it is generally agreed that suprapubic US is of only limited value in staging bladder neoplasms. With transurethral and transrectal US, higher frequency transducers can be used, allowing better spatial resolution. If the tumour is in the body of the bladder, staging accuracies of 75-95% have been reported on transurethral US. The evaluation of perivesical tumour spread is limited, and US does not allow lymph node assessment. In common with other radiographic modalities, ultrasound lacks the specificity to allow confident differentiation between the neoplastic and inflammatory components of such a mass. On CT, bladder neoplasms appear as sessile or pedunculated soft tissue masses projecting into the bladder lumen (Fig.5). These tumours have a similar density to the bladder wall on enhanced scans, and occasionally the intraluminal surface is encrusted wit perivesical fat. CT is useful in detecting lymph node metastases, involvement being judged by node size. Nodes greater than 10 mm are considered malignant.

On MRI, the papillary component of bladder carcinoma is best seen on a T1-weighted image, where it demonstrates a higher signal intensity than the surrounding urine. On the T2-weighted sequence, both the tumour and the urine increase in signal intensity, diminishing the tissue contrast between them (Fig.6) (Fig.7). Tumour depiction, especially for small multiple lesions, is facilitated by the use of contrast media. To evaluate the infiltrating component of tumour, T2-weighted or contrast-enhanced T1-weighted sequences are needed. This spin-echo sequence demonstrates the tumour with higher signal intensity, in contrast to the lower signal intensity of the uninvolved bladder wall. Surface coil imaging (e.g. the use of an endorectal coil) improves the visualization of the layers of the bladder wall and thus improves staging of T2 to T3b lesions. The criteria used for MR tumour staging also follow the TNM classification system. Visualization of the low-intensity line (on T2-weighted images) representing the uninvolved bladder wall between the tumour and perivesical fat is an important feature in differentiating stage T3a from stage T3b. Disruption of the low-signal-intensity muscle layer of the bladder, irregularity of the outer bladder wall and perivesical stranding indicate stage T3b disease. Inflammatory postbiopsy changes in the bladder wall or in the perivesical fat, however, cannot be differentiated from extravesical tumour extension and are a source of staging error at both MR and CT.

Depiction of direct tumour invasion into adjacent organs is facilitated by multiplanar imaging. The sagittal plane of imaging is needed for the evaluation of uterine or vaginal invasion. Invasion of the seminal vesicles is also facilitated by the sagittal plane and demonstrated by an increase in size, decrease in signal intensity on a T2-weighted image and obliteration of the angle between the seminal vesicle and the posterior bladder wall. Invasion of the prostate and rectum is seen as direct tumour extension with an increase in signal intensity on T2-weighted images. The criteria for lymph node metastases, as with CT, are based on size; lymph nodes larger than 10 mm in diameter are considered abnormal.

CT or MRI are employed in the evaluation of patients who cannot be appropriately treated with transurethral tumour resection. Imaging can be used to determine the feasibility of segmental cystectomy, the desirability of pelvic lymphadenectomy, or the possibility of radical cystectomy. Imaging can also be used for determination of disease extent prior to treatment with pelvic radiation and to provide evidence of measurable disease before and during treatment with systemic chemotherapy. Staging accuracy with either CT or MRI varies from approximately 65% to 85% and staging accuracy is generally higher when study populations have included patients with advanced disease. Reports demonstrating the tumours of 5 mm in size or less are available. In most cases, tumours smaller than 1 cm are difficult to detect. Furthermore, imaging is not capable of distinguishing a superficial from deeply invasive lesion nor can it document microscopic invasion of perivesical tissue. In the latter, overstaging due to adjacent inflammatory reaction often occurs.

While the advantage of CT over MRI is a subject of ongoing controversy, it is generally agreed that when tumours involve the bladder dome, base, or bladder neck, multiplanar MR images may allow better anatomic delineation than CT scans. Furthermore, the excellent tissue contrast combined with multiplanar MR imaging are helpful in patients with scar tissue anterior to urinary bladder due to previous surgery or placement of suprapubic bladder catheters. In those patients, CT is limited in differentiating between scar tissue and direct tumour invasion. On T2-weighted MR scans, scar tissue will demonstrate low signal intensity while direct tumour extension will demonstrate medium or high signal intensity. However, while such a differentiation is possible in the majority of cases, false-positive interpretations may be caused by the similar appearance of inflammatory response in the scar tissue.

Imaging can also be used in the evaluation of tumour recurrence following either segmental or radical cystectomy. Following segmental cystectomy, while large tumours can be detected, small sessile plaque-like recurrent lesions are difficult to evaluate by imaging and cystoscopy remains the optimal diagnostic tool. Following radical cystectomy, the multiplanar and excellent soft tissue imaging of MRI are considered superior to CT scans. Published data comparing CT and MR for tumour staging reveals staging accuracies ranging from 40–92% for CT and 60–96% for MRI. MR is, however, slightly better than CT for the evaluation of tumours at the bladder base of the dome, and for differentiation between deep muscular (T3a) and perivesical (T3b) tumour invasion, and for demonstration of stage T4 disease.

HH