CME JOURNAL OF MAGNETIC RESONANCE IMAGING 38:1299–1309 (2013)
Review Article
MR Imaging of Urinary Bladder Cancer for T-Staging: A Review and a Pictorial Essay of Diffusion-Weighted Imaging Mitsuru Takeuchi, MD, PhD,1* Shigeru Sasaki, MD, PhD,2 Taku Naiki, MD, PhD,3 Noriyasu Kawai, MD, PhD,3 Kenjiro Kohri, MD, PhD,3 Masaki Hara, MD, PhD,1 and Yuta Shibamoto, MD, PhD1 Treatment decisions and the prognosis of bladder cancer patients are based on the depth of bladder wall invasion by the tumor, degree of differentiation of the tumor, and presence or absence of metastatic disease. In particular, bladder muscle invasion is important because bladder cancer not invading the muscle is treated with transurethral resection (TUR) and muscle-invasive bladder cancer is treated with radical cystectomy. MR imaging of the pelvis is usually performed for T (tumor) staging once bladder cancer has been diagnosed. The usefulness of diffusion-weighted imaging (DWI) to detect bladder cancer and the advantage of determining the T-stage and its range of extension have been reported (2–6); however, there is no educational, detailed and practical review paper describing how to use DWI for bladder cancer staging; therefore, this issue is discussed here. Three hundred nineteen pelvic MR data were used to emphasize and reinforce the role of DWI for bladder cancer imaging.
Treatment decisions for bladder cancer patients are mainly based on the depth of bladder wall invasion by the tumor. In this article, we review the conventional MRI and exhibit a recently emerged diffusion-weighted imaging (DWI) of urinary bladder cancer for T-staging. We discuss limitations of conventional MRI, scanning protocols of DWI, normal pelvic findings on DWI, determination of Tstage using DWI, and pitfalls of DWI. DWI provides high contrast between bladder cancer and background tissue because the cancer shows markedly high SI. DWI has high sensitivity for detecting the stalk seen in stage Ta or T1. An inflammatory change or fibrosis surrounding the tumor mimics the invasion of bladder cancer on T2weighted imaging or enhanced MRI and could lead to over-staging, but DWI could differentiate them clearly because these benign changes do not show high SI on DWI. DWI is also useful for detecting ureteral, urethral, and prostatic extension by means of the urethra. DWI provides more accurate information on the extent of bladder cancer and contributes to determination of the treatment strategy. Key Words: bladder cancer; staging; diffusion-weighted imaging J. Magn. Reson. Imaging 2013;38:1299–1309. C 2013 Wiley Periodicals, Inc. V
T-Stage Classification of Bladder Cancer
IT IS ESTIMATED that 72,570 men and women (54,610 men and 17,960 women) will be diagnosed with and 15,210 men and women will die of cancer of the urinary bladder in the United States in 2013 (1). 1 Nagoya City University, Graduate School of Medical Sciences and Medical school, Department of Radiology, Mizuho-ku, Nagoya, Japan. 2 Nagoya City West Medical Center, Department of Radiology, Nagoya, Japan. 3 Nagoya City University, Graduate School of Medical Sciences and Medical school Department of Nephro-Urology, Mizuho-ku, Nagoya, Japan. *Address reprint requests to: M.T., Nagoya City University, Graduate School of Medical Sciences and Medical school Departments of Radiology, 1 Kawasumi Mizuho-cho, Mizuho-ku, Nagoya, 467-8601, Japan. E-mail: [email protected] Received January 18, 2013; Accepted April 26, 2013. DOI 10.1002/jmri.24227 View this article online at wileyonlinelibrary.com. C 2013 Wiley Periodicals, Inc. V
Bladder cancer is staged using the TNM (tumor-nodemetastasis) staging system (Table 1) (7). In this system, T-stage is based on the degree of invasion into the bladder wall. TUR is usually chosen as an initial treatment for early-stage urinary bladder carcinoma. Tumors of stage T2 or greater are treated with partial or total cystectomy or with adjuvant therapies because TUR for invasive tumors often results in local tumor recurrence. Therefore, preoperative differentiation between stage T1 tumors and stage T2 or greater tumors is crucial for the appropriate treatment of patients with bladder cancer. Conventional MR Imaging and Its Limitations T2-Weighted Imaging On T2-weighted images, the urine has high SI and the bladder wall has low SI. Bladder cancer shows slightly higher SI than that of the muscle layer. In patients with muscle-invasive tumors, the low SI of the muscle layer is interrupted by the slightly high SI of the
1299
1300
Takeuchi et al.
Table 1 T-Stage Classification of Bladder Cancer Tis Ta T1 T2a T2b T3a T3b T4a T4b
Carcinoma in situ Papillary noninvasive tumor Tumor invades subepithelial connective tissue Tumor invades superficial muscle Tumor invades deep muscle Tumor invades perivesical tissue microscopically Tumor invades perivesical tissue macroscopically Tumor invades prostate, uterus, or vagina Tumor invades pelvic wall or abdominal wall
tumor. The overall staging accuracy of T2WI has been reported to be 40–67% in reports published in the past 10 years (3,5,6,8). Tekes et al (8) reported that 81% of bladder tumors showed SI similar to that of muscle on T2WI, and over-staging was the most common error when evaluating T-stage. It has been reported that the SI of the tumor stalk seen in stage Ta or T1 varies on T2WI (9). According to our data of 116 Ta or T1 tumors, the SI of the tumor stalk is lower than that of muscle in 14%, higher than muscle SI in 56% and similar to muscle SI in 30% on T2WI (Fig. 1a–c). If the stalk shows iso SI to the tumor on T2WI, the tumor seems like a sessile tumor and could cause over-staging. In addition, a low SI area similar to a tumor reflecting inflammatory change or fibrosis, which are sometimes seen beneath the tumor or perivesical tissue, could cause over-staging (Fig. 2a). Dynamic Contrast Enhanced Imaging (DCEI) The bladder tumor, mucosa, and submucosa are enhanced early after contrast material administration, but the muscle layer maintains its hypointensity and is enhanced late (8,10,11). Hayashi et al (12) reported the usefulness of submucosal linear enhancement (SLE), which is seen immediately after the injection of contrast media, and found that the SI of the muscle layer remained low. Intact SLE adjacent to a tumor was regarded as indicating stage T1 or lower stage;
however, Takeuchi et al (3) reported that the tumor and submucosa had similar SI in 60% (31/52) of dynamic contrast-enhanced studies, and submucosal linear enhancement was difficult to recognize in these cases. In addition, Tekes et al (8) reported that improvement of the spatial resolution of DCEI was able to detect perivesical stranding, which is considered to be a reactive or inflammatory change rather than tumor invasion. These findings can cause false positivity of perivesical invasion of bladder cancer. The overall staging accuracy using DCEI has been reported to be 52–85% (3,8,10,11). DWI DWI is constructed by quantifying the diffusion of water molecules in tissues (13). DWI is increasingly being used for cancer assessment throughout the body and has shown great promise in the detection and characterization of urinary malignancies (2–6). Scanning Protocol Susceptibility artifacts from bowel peristalsis which could degrade the image quality of DWI can be minimized by administering an intramuscular antispasmodic agent. To moderately distend the bladder, patients are recommended not to urinate for approximately 1 h before the examination. We use a 1.5 Tesla (T) MR scanner (Gyroscan Intera [maximum amplitude of the gradient, 33 mT/m; maximum slew rate, 180 T/m/s]; Philips Medical Systems, Best, The Netherlands) for bladder cancer staging. Because the bladder is a pouched-form organ and cancer can arise anywhere in the bladder wall, at least axial and sagittal planes should be scanned. Although an oblique plane is optimal for bladder cancer staging, DWI scanned in the plane cannot maintain a sufficient signal-to-noise ratio (14). We use chemical shift selective pulse (CHESS) as fat suppression. We do not use short inversion time
Figure 1. Signal intensities of the tumor stalk on T2-weighted image (T2WI) and diffusion-weighted image (DWI). a: Sagittal T2WI and DWI of bladder cancer on the dome in a 66-year-old man. The signal intensity (SI) of the stalk is high on T2WI (arrow) and low on DWI (arrowhead). b: Axial T2WI and DWI of bladder cancer on the posterior wall in a 49-year-old man. The signal intensity (SI) of the stalk is low on T2WI (arrow) and DWI (arrowhead). c: Axial T2WI and DWI of bladder cancer on the right side wall in a 78-year-old woman. The stalk could not be seen on T2WI because the SI of the stalk is iso-intense to the tumor (arrow) but the stalk, which shows low SI (arrowhead), could be seen on DWI.
DWI of Urinary Bladder Cancer Staging
1301
Figure 2. An 82-year-old woman in whom diffusion-weighted image could rule out extravesical invasion of bladder cancer. a: Axial T2-weighted image (T2WI) shows a nonpapillary large mass on the right-side bladder wall (arrow). The right-side bladder wall is disrupted and the signal intensity (SI) of the extravesical fat tissue is decreased, indicating extravesical location (arrowheads). b: Axial integrated diffusion-weighted image and T2-weighted image show a nonpapillary mass with high SI on the right-side bladder wall (arrow); however, there is no signal increase in the perivesical fat tissue (arrowheads). Thus, extravesical invasion could be ruled out.
inversion recovery (STIR) because DWI with STIR cannot maintain a good signal-to-noise ratio (15). A high b-value (800–1000 s/mm2) is needed to visualize bladder cancer with high contrast to surrounding tissues (3–5). A large field-of-view and large number of excitations are recommended to build up the signalto-noise ratio. An example of the scan parameters is shown in Table 2. The optimal protocol of DWI using a 3T scanner has not been reported. Normal Anatomy of the Pelvis on DWI The bladder wall consists of the urothelium, lamina submucosa, muscularis propia and serosa (16,17). There is fat tissue outside of the bladder. On DWI, normal urothelium, submucosa, and serosa are not visualized because these layers are very thin. The muscularis propia shows intermediate SI and extravesical fat tissue shows low SI on DWI with an appropriate b-factor (800–1000 s/mm2) (3). Radiologists should know that some organs show high SI physiologically on DWI and should not diagnose them as metastasis or invasion of bladder cancer. The testis and endometrium show markedly high SI. The corpus cavernosum and corpus spongiosum of the penis and lymph node show intermediate to high SI. The seminal vesicle and bowel wall show intermediate SI. The SI of the bone marrow changes depending on the status of the marrow. Hematopoietic bone marrow shows intermediate to high SI and fatty bone marrow shows low SI (4). Bladder Cancer Staging Using DWI It has been reported that the ADC value in the carcinoma was 1.18 10 3 mm2/s, which was significantly lower than that of urine (3.28 10 3 mm2/s)
and a normal bladder wall (2.27 10 3 mm2/s) (2). The SI of background tissues is well suppressed and bladder cancer shows very high SI on DWI, and the area showing high SI correlates with the extent of the tumor very well. On the other hand, the anatomical structures are difficult to recognize. We strongly recommend comparing T2WI and DWI to identify the location of highlighted lesions. If integrated viewing is available, it is useful because both DWI and T2WI information can be seen at the same time. The area of high SI on DWI depends on the setting of the viewing window width and center level. An inappropriate window setting causes misdiagnosis of the number of tumor(s), tumor size, and stage (Fig. 3a,b). It is important not to display the tumor too brightly when interpreting DWI.
Table 2 Example of Pelvic DWI Protocol for Bladder Cancer Staging With a 1.5 T Scanner Coil Technique B value Fat suppression EPI factor SENSE factor No. of excitations Plane Flip angle Repetition time Echo time Matrix Field of view Slice thickness and gap No. of slices Acquisition time
5 channel phased array Single shot, echo-planar imaging 1000 s/mm2 CHESS 53 2 12 Axial and sagittal 90 degree 2790–4560 ms (set at shortest) 88 ms 129 x 109 33 cm 4 mm and 0.4 mm 21 136–151 s
1302
Takeuchi et al.
Figure 3. A 67-year-old woman with multiple stage T1 bladder cancer in the same plane on diffusion-weighted image at different viewing settings. a: In an inappropriate viewing setting (too narrow window width and too low level), the bladder cancer on the posterior wall seems to be one sessile tumor (arrow). b: Three papillary tumors are seen with the appropriate viewing setting (arrows).
Figure 4. Stage Ta papillary bladder cancer in an 85-year-old woman. a: Transverse DWI shows a U-shaped high-intensity area (arrow) with a lowsignal stalk (white arrowhead) connecting to the anterior bladder wall (black arrowheads). b: Photomicrograph of the specimen shows a papillary cancer (star) with a stalk (asterisk) consisting of submucosa (hematoxylin-eosin stain).
Figure 5. Stage T1 sessile bladder cancer in an 80-year-old man. a: Axial T2-weighted image shows a sessile mass on the posterior bladder wall (arrow). The signal intensity (SI) of the bladder wall beneath the tumor is elevated, indicating muscle invasion (arrowhead). b: On axial diffusion-weighted image, the surface part of the tumor shows very high SI, indicating the cancer component (arrows), but SI of the deep part of the tumor is low, indicating an inflammatory or fibrosing reaction (arrowhead). This finding is an “inch worm sign”. Thus, the tumor was staged correctly as T1, though the tumor was sessile at cystoscopy.
DWI of Urinary Bladder Cancer Staging
1303
Figure 6. Stage T2 bladder cancer in a 65-year-old man. a: Sagittal T2-weighted image shows a hypointense oval mass on the dome of the bladder protruding into extravesical fat tissue (arrow). b: Sagittal diffusion-weighted image shows a hyperintense oval mass with a smooth contour (arrow).
Figure 7. Stage T3 bladder cancer in a 70-year-old man. a: Axial T2-weighted image shows a nonpapillary mass on the right-side bladder wall (arrow). The signal intensity (SI) of the bladder wall beneath the tumor is elevated, indicating muscle invasion (arrowhead). b: Axial diffusion-weighted image of a nonpapillary mass on the rightside bladder wall (arrow) with an irregular margin spreading toward the surrounding fat tissue (arrowhead).
Figure 8. Stage T3 bladder cancer in a 56-year-old woman. a: Axial T2-weighted image (T2WI) shows a large bladder mass protruding into the bladder lumen (arrow). The bladder wall beneath the mass is markedly thickened and the signal is very low, indicating fibrosis (arrowhead). b: In contrast to the prior case, the thickened bladder wall (arrowhead), which shows very low signal intensity (SI) on T2WI as well as the mass protruding into the bladder lumen (arrow), shows high SI on axial diffusion-weighted image. Thus, the tumor is staged correctly as T3.
1304
Takeuchi et al.
Figure 9. Stage T4 bladder cancer with prostatic invasion in a 72-year-old man. a: Sagittal T2-weighted image shows a nonpapillary lesion on the neck of the bladder (arrow). Although it seems that the tumor extends toward the urethra and prostate, the accurate extent of the tumor is unclear. b: Sagittal integrated diffusion-weighted image and T2-weighted image show that cancer on the neck of the bladder (arrow) clearly invades the prostate (arrowheads).
Tis Stage Tis cancer is invisible on MRI including DWI and it should be detected and staged by cystoscopy and TUR. Ta–T1 versus T2 It is most important to differentiate T2 from T1 or a lower stage in bladder cancer staging. Stage Ta or T1 cancer shows a papillary or sessile form on MRI. On DWI, the stalk of a papillary tumor, which consists of submucosa, generally shows low SI in contrast to the various SI on T2WI, and the tumor covering the stalk shows very high SI (Fig. 1a–c). It has been reported that the stalk consists of submucosa with edema and fibrosis (Fig. 4a,b) (3). Takeuchi et al (3) reported this unique finding as the inch worm sign. According to our unpublished data, DWI could almost always
(94%) visualize the stalk as low SI compared with T2WI; therefore, DWI could significantly reduce such false positivity for diagnosing muscle invasion compared with T2WI alone. DWI is also useful for staging sessile tumors. DWI could differentiate thickened submucosa from inflammatory change or fibrosis occurring beneath the tumor, which mimics muscle invasion on T2WI or DCEI. Inflammatory change or fibrosis beneath the tumor shows lower SI than that of the tumor and the intermediate SI of the muscle could be differentiated on DWI (Fig. 5a,b). In such a case, the area of high SI on DWI seems to be more important than the T2WI and DCEI findings and it could reduce false positivity for diagnosing muscle invasion. If a sessile lesion is seen and the SI of the muscle beneath the tumor is elevated but the area with high SI does not extend to the perivesical space and the
Figure 10. Stage T4 bladder cancer with uterine invasion in a 37-year-old woman. a: Sagittal T2-weighted image (T2WI) shows a large mass occupying the caudal part of the bladder (arrow). The signal intensity (SI) of the vaginal portion of the cervix is partially decreased (white arrowhead). Liquid accumulates in the vagina, indicating vesicovaginal fistula (black arrowhead). b: Sagittal integrated diffusion-weighted image and T2WI show that both bladder tumor (arrow) and lesion in the vaginal portion of cervix (arrowhead) are hyperintense.
DWI of Urinary Bladder Cancer Staging
1305
Figure 11. Bladder cancer extending into the prostatic urethra in a 71-year-old man. a: Sagittal T2-weighted image shows multiple papillary lesions on the neck and trigone of the bladder (arrows), one of which extends into the prostatic urethra (arrowhead). b: Sagittal integrated diffusionweighted image and T2weighted image show bladder cancer (arrows) extending into the prostatic urethra (arrowhead).
tumor margin is smooth on DWI, the stage is T2 (Fig. 6a,b). The TNM staging system subclassifies pT2 cancer into two categories: cancer invading less than half of the depth of the muscular propria (pT2a) and cancer invading greater than half of the muscle wall (7); however, MRI including DWI could not differentiate them accurately. The clinical utility of pT2 tumor substaging has been questioned, and, therefore,
Figure 12. Bladder cancer invades the prostate by means of the prostatic urethra in a 64-year-old man. a: The signal intensity of prostatic right peripheral zone is slightly decreased on axial T2-weighted image (arrow). There is no evident prostatic urethral lesion. b: Axial diffusion-weighted image shows signal increase of the prostatic urethra (arrow) and right lobe of the prostate (arrowhead). c: Photomicrograph of the specimen shows cancer invasion into the prostate (arrowheads) by means of the prostatic urethra (arrow) (hematoxylineosin stain).
radiologists need not be fixated on differentiating these sub-stages in our opinion (18). T2 versus T3 The efficacy of DWI for diagnosing extravesical invasion has been uncertain (3); however, we have experienced some cases which were diagnosed accurately
1306
Takeuchi et al.
Figure 13. Bladder cancer extending into the right ureter in a 68-year-old man. a: Axial T2-weighted image shows a large tumor in the bladder (arrow). The right ureter is dilated and the signal intensity of the lumen is decreased (arrowhead). b: Axial diffusion-weighted image shows bladder cancer (arrow) and a right ureteral lesion (arrowhead), which are clearly visualized as marked high intensity.
using DWI added to T2WI or DCEI. On DWI, if the contour of the mass invading the muscle is irregular or has a ridge protruding into the perivesical space, the T-stage is T3 (Fig. 7a,b). In differentiating T2 from T3, inflammatory change or fibrosis surrounding the tumor could mislead the T-staging using T2WI or DCEI, but DWI could differentiate them clearly because these changes do not show high SI on DWI (Figs. 2a,b, 8a,b). Differentiating T2b from T3a is quite difficult because of the limited spatial resolution; however, it is not a serious problem because the treatment procedure and prognosis are the same for these T-stages (19). T3 versus T4 Stage pT4 bladder cancer is defined by tumor invasion into an adjacent organ, including the uterus, vagina, prostate, pelvic wall, or abdominal wall. Stage T4a includes invasion to the prostate, uterus, or vagina, and stage T4b indicates pelvic or abdominal wall invasion. If the extent of high SI reflecting cancer invasion is seen in an adjacent organ on DWI, the stage is T4. The efficacy of DWI added to conventional
imaging for diagnosing stage T4 cancer is uncertain, but tumor invasion into adjacent organs is clearly seen on DWI compared with conventional imaging because the tumor is highlighted (Figs. 9a,b, 10a,b). Integrated DWI and T2WI would be useful to evaluate the invasion to these organs. Donat et al (20) identified three pathways for prostatic stromal invasion by urothelial carcinoma. These include extravesical, intraurethral, and bladder neck invasion. The 5-year overall survival rate was 21% for the patients with extravesical and bladder neck invasion and 55% for the patients with intraurethral invasion (21,22). Bladder cancer invading the prostate by means of the bladder neck or through the perivesical fat tissue is staged as T4a but bladder cancer extending into the prostate by means of the urethra is not staged as T4a in the existing TNM classification.
Supplement Detecting prostatic extension by means of the urethra and ureteral extension of bladder cancer is important because the extent of resection changes depending on
Figure 14. A case of pyuria in a 79-year-old man. a: On diffusion-weighted image, there is a markedly high signal intensity (SI) area in the posterior part of the bladder lumen with a fluid-fluid level (arrow). b: On T2-weighted image, the SI of the urine in the posterior part of the bladder is decreased and a fluidfluid level is seen (arrow).
DWI of Urinary Bladder Cancer Staging
1307
Figure 15. A 72-year-old man with T1 bladder cancer with intravesical hemorrhage. a: On diffusion-weighted image, there are a high signal intensity (SI) mass with a low SI stalk on the right-side bladder wall (arrow) and a markedly high SI mass at the left side of the bladder lumen (arrowhead). b: On T1weighted image, the rightside mass shows intermediate SI, indicating tumor (arrow) and the left-side mass shows high SI, indicating hemorrhage (arrowhead).
the extent of bladder cancer. On DWI, normal prostatic urethra shows low SI and prostatic urethra with cancer shows high SI along the urethra (Fig. 11a,b) (4). If areas of high SI are seen in the prostatic urethra and prostate simultaneously, it is a finding of prostatic invasion by means of the urethra (Fig. 12a–c) (4);
Figure 16. A susceptibility artifact from rectal gas in a 76-year-old man. a: There is bow-shaped high signal intensity (SI) at the posterior bladder wall (arrow) on diffusion-weighted image (DWI). The area surrounding the rectum (white arrowhead) and uterus (black arrowhead) is distorted. b: On T2weighted image, there is no tumor corresponding to the high SI area on DWI (arrow).
Figure 17. Prostatic abscess in a 70-year-old man. a: Axial diffusion-weighted image shows a markedly high signal intensity (SI) nodule in the left peripheral zone of the prostate (arrow). b: Axial T2-weighted image shows a relatively high SI nodule in the left peripheral zone of the prostate (arrow).
however, incidental prostatic cancer also shows high SI on DWI, and a prostatic biopsy is needed to differentiate them. Bladder cancer could extend into the ureter along the urothelial epithelium. If ureteral extension is present, a nephro-ureterectomy is performed. If the
1308
Takeuchi et al.
Figure 18. Granulomatous prostatitis induced by intravesical Bacille-Calmette-Guerin therapy in a 72-year-old man. a: Axial diffusionweighted image shows diffuse signal elevation of the prostate (arrows). b: The prostate is enlarged and replaced with the lesion which shows intermediate signal intensity on T2-weighted image and the zonal anatomy of the prostate gland is indistinct (arrows).
ureteral orifice is covered with bladder cancer, the ureteral extension cannot be observed by cystoscopy; therefore, radiologists need to notify urologists of this finding. Cancer extending into the ureter shows high SI on DWI (Fig. 13a,b).
Pitfalls Some benign lesions and artifacts show high SI and mimic bladder cancer, which could lead to an incorrect T-stage of bladder cancer. It is important to compare DWI with T2WI and DCE to diagnose falsepositive lesions accurately. Purulent urine shows very high SI on DWI (Fig. 14a). Purulent urine settles out in the bladder or diverticulum. A relatively high SI on T2WI and a fluid–fluid level seen between normal and purulent urine are diagnostic keys to purulent urine (Fig. 14b). Clots that sometimes accompany bladder cancer also show high SI on DWI (Fig. 15a). They show high SI on T1-weighted images and are not enhanced (Fig. 15b). Susceptibility artifact from gases in the colon and rectum causes distortion of the images and could result in a false-positive lesion (Fig. 16a,b). This false-positive lesion is not seen on images other than DWI. Air in the bladder is also seen after cystoscopy or insertion of a catheter; therefore, MRI is not recommended after these procedures. Prostatic inflammation including abscess and granulomatous prostatitis induced by intravesical Bacille-CalmetteGuerin (BCG) therapy also shows high SI (Figs. 17a,b, 18a,b) (4). Clinical symptoms and treatment history should be examined to exclude these inflammatory diseases. In conclusion, DWI provides more accurate information on the extent of bladder cancer and contributes to the determination of treatment strategy. We recommend that DWI is added to conventional bladder MR examination. REFERENCES 1. Siegel R, Naishadham D, Jemal A. Cancer statistics, 2013. CA Cancer J Clin 2013;63:11–30.
2. Matsuki M, Inada Y, Tatsugami F, Tanikake M, Narabayashi I, Katsuoka Y. Diffusion-weighted MR imaging for urinary bladder carcinoma: initial results. Eur Radiol 2007;17:201–204. 3. Takeuchi M, Sasaki S, Ito M, et al. Urinary bladder cancer: diffusion-weighted MR imaging--accuracy for diagnosing T stage and estimating histologic grade. Radiology 2009;251: 112–121. 4. Takeuchi M, Suzuki T, Sasaki S, et al. Clinicopathologic significance of high signal intensity on diffusion-weighted MR imaging in the ureter, urethra, prostate and bone of patients with bladder cancer. Acad Radiol 2012;19:827–833. 5. Watanabe H, Kanematsu M, Kondo H, et al. Preoperative T staging of urinary bladder cancer: does diffusion-weighted MRI have supplementary value? AJR Am J Roentgenol 2009;192:1361– 1366. 6. El-Assmy A, Abou-El-Ghar ME, Mosbah A, et al. Bladder tumour staging: comparison of diffusion- and T2-weighted MR imaging. Eur Radiol 2009;19:1575–1581. 7. Urinary bladder. In: Edge SB, Byrd DR, Compton CC, et al, editors. AJCC Cancer Staging Manual. 7th edition. New York, NY: Springer, 2010. p 497–505. 8. Tekes A, Kamel I, Imam K, et al. Dynamic MRI of bladder cancer: evaluation of staging accuracy. AJR Am J Roentgenol 2005;184:121–127. 9. Saito W, Amanuma M, Tanaka J, Heshiki A. Histopathological analysis of a bladder cancer stalk observed on MRI. Magn Reson Imaging 2000;18:411–415. 10. Kim B, Semelka RC, Ascher SM, Chalpin DB, Carroll PR, Hricak H. Bladder tumor staging: comparison of contrast-enhanced CT, T1- and T2-weighted MR imaging, dynamic gadolinium-enhanced imaging, and late gadolinium-enhanced imaging. Radiology 1994; 193:239–245. 11. Tanimoto A, Yuasa Y, Imai Y, et al. Bladder tumor staging: comparison of conventional and gadolinium-enhanced dynamic MR imaging and CT. Radiology 1992;185:741–747. 12. Hayashi N, Tochigi H, Shiraishi T, Takeda K, Kawamura J. A new staging criterion for bladder carcinoma using gadoliniumenhanced magnetic resonance imaging with an endorectal surface coil: a comparison with ultrasonography. BJU Int 2000;85: 32–36. 13. Koh DM, Collins DJ. Diffusion-weighted MRI in the body: applications and challenges in oncology. AJR Am J Roentgenol 2007;188:1622–1635. 14. Narumi Y, Kadota T, Inoue E, et al. Bladder tumors: staging with gadolinium-enhanced oblique MR imaging. Radiology 1993;187: 145–150. 15. Kazama T, Nasu K, Kuroki Y, Nawano S, Ito H. Comparison of diffusion-weighted images using short inversion time inversion recovery or chemical shift selective pulse as fat suppression in patients with breast cancer. Jpn J Radiol 2009;27:163–167.
DWI of Urinary Bladder Cancer Staging 16. Kirkali Z, Chan T, Manoharan M, et al. Bladder cancer: epidemiology, staging and grading, and diagnosis. Urology 2005;66(Suppl 1):4–34. 17. Wong-You-Cheong JJ, Woodward PJ, Manning MA, Sesterhenn IA. Neoplasms of the urinary bladder: radiologic-pathologic correlation. Radiographics 2006;26:553–580. 18. Cheng L, Neumann RM, Weaver AL, et al. Grading and staging of bladder carcinoma in transurethral resection specimens. Correlation with 105 matched cystectomy specimens. Am J Clin Pathol 2000;113:275–279. 19. Quek ML, Stein JP, Clark PE, et al. Microscopic and gross extravesical extension in pathological staging of bladder cancer. J Urol 2004;17:640–645.
1309 20. Donat SM, Genega EM, Herr HW, Reuter VE. Mechanisms of prostatic stromal invasion in patients with bladder cancer: clinical significance. J Urol 2001;165:1117–1120. 21. Esrig D, Freeman JA, Elmajian DA, et al. Transitional cell carcinoma involving the prostate with a proposed staging classification for stromal invasion. J Urol 1996;156:1071– 1076. 22. Pagano F, Bassi P, Ferrante GL, et al. Is stage pT4a (D1) reliable in assessing transitional cell carcinoma involvement of the prostate in patients with a concurrent bladder cancer? A necessary distinction for contiguous or noncontiguous involvement. J Urol 1996;155:244–247.
Review Article
MR Imaging of Urinary Bladder Cancer for T-Staging: A Review and a Pictorial Essay of Diffusion-Weighted Imaging Mitsuru Takeuchi, MD, PhD,1* Shigeru Sasaki, MD, PhD,2 Taku Naiki, MD, PhD,3 Noriyasu Kawai, MD, PhD,3 Kenjiro Kohri, MD, PhD,3 Masaki Hara, MD, PhD,1 and Yuta Shibamoto, MD, PhD1 Treatment decisions and the prognosis of bladder cancer patients are based on the depth of bladder wall invasion by the tumor, degree of differentiation of the tumor, and presence or absence of metastatic disease. In particular, bladder muscle invasion is important because bladder cancer not invading the muscle is treated with transurethral resection (TUR) and muscle-invasive bladder cancer is treated with radical cystectomy. MR imaging of the pelvis is usually performed for T (tumor) staging once bladder cancer has been diagnosed. The usefulness of diffusion-weighted imaging (DWI) to detect bladder cancer and the advantage of determining the T-stage and its range of extension have been reported (2–6); however, there is no educational, detailed and practical review paper describing how to use DWI for bladder cancer staging; therefore, this issue is discussed here. Three hundred nineteen pelvic MR data were used to emphasize and reinforce the role of DWI for bladder cancer imaging.
Treatment decisions for bladder cancer patients are mainly based on the depth of bladder wall invasion by the tumor. In this article, we review the conventional MRI and exhibit a recently emerged diffusion-weighted imaging (DWI) of urinary bladder cancer for T-staging. We discuss limitations of conventional MRI, scanning protocols of DWI, normal pelvic findings on DWI, determination of Tstage using DWI, and pitfalls of DWI. DWI provides high contrast between bladder cancer and background tissue because the cancer shows markedly high SI. DWI has high sensitivity for detecting the stalk seen in stage Ta or T1. An inflammatory change or fibrosis surrounding the tumor mimics the invasion of bladder cancer on T2weighted imaging or enhanced MRI and could lead to over-staging, but DWI could differentiate them clearly because these benign changes do not show high SI on DWI. DWI is also useful for detecting ureteral, urethral, and prostatic extension by means of the urethra. DWI provides more accurate information on the extent of bladder cancer and contributes to determination of the treatment strategy. Key Words: bladder cancer; staging; diffusion-weighted imaging J. Magn. Reson. Imaging 2013;38:1299–1309. C 2013 Wiley Periodicals, Inc. V
T-Stage Classification of Bladder Cancer
IT IS ESTIMATED that 72,570 men and women (54,610 men and 17,960 women) will be diagnosed with and 15,210 men and women will die of cancer of the urinary bladder in the United States in 2013 (1). 1 Nagoya City University, Graduate School of Medical Sciences and Medical school, Department of Radiology, Mizuho-ku, Nagoya, Japan. 2 Nagoya City West Medical Center, Department of Radiology, Nagoya, Japan. 3 Nagoya City University, Graduate School of Medical Sciences and Medical school Department of Nephro-Urology, Mizuho-ku, Nagoya, Japan. *Address reprint requests to: M.T., Nagoya City University, Graduate School of Medical Sciences and Medical school Departments of Radiology, 1 Kawasumi Mizuho-cho, Mizuho-ku, Nagoya, 467-8601, Japan. E-mail: [email protected] Received January 18, 2013; Accepted April 26, 2013. DOI 10.1002/jmri.24227 View this article online at wileyonlinelibrary.com. C 2013 Wiley Periodicals, Inc. V
Bladder cancer is staged using the TNM (tumor-nodemetastasis) staging system (Table 1) (7). In this system, T-stage is based on the degree of invasion into the bladder wall. TUR is usually chosen as an initial treatment for early-stage urinary bladder carcinoma. Tumors of stage T2 or greater are treated with partial or total cystectomy or with adjuvant therapies because TUR for invasive tumors often results in local tumor recurrence. Therefore, preoperative differentiation between stage T1 tumors and stage T2 or greater tumors is crucial for the appropriate treatment of patients with bladder cancer. Conventional MR Imaging and Its Limitations T2-Weighted Imaging On T2-weighted images, the urine has high SI and the bladder wall has low SI. Bladder cancer shows slightly higher SI than that of the muscle layer. In patients with muscle-invasive tumors, the low SI of the muscle layer is interrupted by the slightly high SI of the
1299
1300
Takeuchi et al.
Table 1 T-Stage Classification of Bladder Cancer Tis Ta T1 T2a T2b T3a T3b T4a T4b
Carcinoma in situ Papillary noninvasive tumor Tumor invades subepithelial connective tissue Tumor invades superficial muscle Tumor invades deep muscle Tumor invades perivesical tissue microscopically Tumor invades perivesical tissue macroscopically Tumor invades prostate, uterus, or vagina Tumor invades pelvic wall or abdominal wall
tumor. The overall staging accuracy of T2WI has been reported to be 40–67% in reports published in the past 10 years (3,5,6,8). Tekes et al (8) reported that 81% of bladder tumors showed SI similar to that of muscle on T2WI, and over-staging was the most common error when evaluating T-stage. It has been reported that the SI of the tumor stalk seen in stage Ta or T1 varies on T2WI (9). According to our data of 116 Ta or T1 tumors, the SI of the tumor stalk is lower than that of muscle in 14%, higher than muscle SI in 56% and similar to muscle SI in 30% on T2WI (Fig. 1a–c). If the stalk shows iso SI to the tumor on T2WI, the tumor seems like a sessile tumor and could cause over-staging. In addition, a low SI area similar to a tumor reflecting inflammatory change or fibrosis, which are sometimes seen beneath the tumor or perivesical tissue, could cause over-staging (Fig. 2a). Dynamic Contrast Enhanced Imaging (DCEI) The bladder tumor, mucosa, and submucosa are enhanced early after contrast material administration, but the muscle layer maintains its hypointensity and is enhanced late (8,10,11). Hayashi et al (12) reported the usefulness of submucosal linear enhancement (SLE), which is seen immediately after the injection of contrast media, and found that the SI of the muscle layer remained low. Intact SLE adjacent to a tumor was regarded as indicating stage T1 or lower stage;
however, Takeuchi et al (3) reported that the tumor and submucosa had similar SI in 60% (31/52) of dynamic contrast-enhanced studies, and submucosal linear enhancement was difficult to recognize in these cases. In addition, Tekes et al (8) reported that improvement of the spatial resolution of DCEI was able to detect perivesical stranding, which is considered to be a reactive or inflammatory change rather than tumor invasion. These findings can cause false positivity of perivesical invasion of bladder cancer. The overall staging accuracy using DCEI has been reported to be 52–85% (3,8,10,11). DWI DWI is constructed by quantifying the diffusion of water molecules in tissues (13). DWI is increasingly being used for cancer assessment throughout the body and has shown great promise in the detection and characterization of urinary malignancies (2–6). Scanning Protocol Susceptibility artifacts from bowel peristalsis which could degrade the image quality of DWI can be minimized by administering an intramuscular antispasmodic agent. To moderately distend the bladder, patients are recommended not to urinate for approximately 1 h before the examination. We use a 1.5 Tesla (T) MR scanner (Gyroscan Intera [maximum amplitude of the gradient, 33 mT/m; maximum slew rate, 180 T/m/s]; Philips Medical Systems, Best, The Netherlands) for bladder cancer staging. Because the bladder is a pouched-form organ and cancer can arise anywhere in the bladder wall, at least axial and sagittal planes should be scanned. Although an oblique plane is optimal for bladder cancer staging, DWI scanned in the plane cannot maintain a sufficient signal-to-noise ratio (14). We use chemical shift selective pulse (CHESS) as fat suppression. We do not use short inversion time
Figure 1. Signal intensities of the tumor stalk on T2-weighted image (T2WI) and diffusion-weighted image (DWI). a: Sagittal T2WI and DWI of bladder cancer on the dome in a 66-year-old man. The signal intensity (SI) of the stalk is high on T2WI (arrow) and low on DWI (arrowhead). b: Axial T2WI and DWI of bladder cancer on the posterior wall in a 49-year-old man. The signal intensity (SI) of the stalk is low on T2WI (arrow) and DWI (arrowhead). c: Axial T2WI and DWI of bladder cancer on the right side wall in a 78-year-old woman. The stalk could not be seen on T2WI because the SI of the stalk is iso-intense to the tumor (arrow) but the stalk, which shows low SI (arrowhead), could be seen on DWI.
DWI of Urinary Bladder Cancer Staging
1301
Figure 2. An 82-year-old woman in whom diffusion-weighted image could rule out extravesical invasion of bladder cancer. a: Axial T2-weighted image (T2WI) shows a nonpapillary large mass on the right-side bladder wall (arrow). The right-side bladder wall is disrupted and the signal intensity (SI) of the extravesical fat tissue is decreased, indicating extravesical location (arrowheads). b: Axial integrated diffusion-weighted image and T2-weighted image show a nonpapillary mass with high SI on the right-side bladder wall (arrow); however, there is no signal increase in the perivesical fat tissue (arrowheads). Thus, extravesical invasion could be ruled out.
inversion recovery (STIR) because DWI with STIR cannot maintain a good signal-to-noise ratio (15). A high b-value (800–1000 s/mm2) is needed to visualize bladder cancer with high contrast to surrounding tissues (3–5). A large field-of-view and large number of excitations are recommended to build up the signalto-noise ratio. An example of the scan parameters is shown in Table 2. The optimal protocol of DWI using a 3T scanner has not been reported. Normal Anatomy of the Pelvis on DWI The bladder wall consists of the urothelium, lamina submucosa, muscularis propia and serosa (16,17). There is fat tissue outside of the bladder. On DWI, normal urothelium, submucosa, and serosa are not visualized because these layers are very thin. The muscularis propia shows intermediate SI and extravesical fat tissue shows low SI on DWI with an appropriate b-factor (800–1000 s/mm2) (3). Radiologists should know that some organs show high SI physiologically on DWI and should not diagnose them as metastasis or invasion of bladder cancer. The testis and endometrium show markedly high SI. The corpus cavernosum and corpus spongiosum of the penis and lymph node show intermediate to high SI. The seminal vesicle and bowel wall show intermediate SI. The SI of the bone marrow changes depending on the status of the marrow. Hematopoietic bone marrow shows intermediate to high SI and fatty bone marrow shows low SI (4). Bladder Cancer Staging Using DWI It has been reported that the ADC value in the carcinoma was 1.18 10 3 mm2/s, which was significantly lower than that of urine (3.28 10 3 mm2/s)
and a normal bladder wall (2.27 10 3 mm2/s) (2). The SI of background tissues is well suppressed and bladder cancer shows very high SI on DWI, and the area showing high SI correlates with the extent of the tumor very well. On the other hand, the anatomical structures are difficult to recognize. We strongly recommend comparing T2WI and DWI to identify the location of highlighted lesions. If integrated viewing is available, it is useful because both DWI and T2WI information can be seen at the same time. The area of high SI on DWI depends on the setting of the viewing window width and center level. An inappropriate window setting causes misdiagnosis of the number of tumor(s), tumor size, and stage (Fig. 3a,b). It is important not to display the tumor too brightly when interpreting DWI.
Table 2 Example of Pelvic DWI Protocol for Bladder Cancer Staging With a 1.5 T Scanner Coil Technique B value Fat suppression EPI factor SENSE factor No. of excitations Plane Flip angle Repetition time Echo time Matrix Field of view Slice thickness and gap No. of slices Acquisition time
5 channel phased array Single shot, echo-planar imaging 1000 s/mm2 CHESS 53 2 12 Axial and sagittal 90 degree 2790–4560 ms (set at shortest) 88 ms 129 x 109 33 cm 4 mm and 0.4 mm 21 136–151 s
1302
Takeuchi et al.
Figure 3. A 67-year-old woman with multiple stage T1 bladder cancer in the same plane on diffusion-weighted image at different viewing settings. a: In an inappropriate viewing setting (too narrow window width and too low level), the bladder cancer on the posterior wall seems to be one sessile tumor (arrow). b: Three papillary tumors are seen with the appropriate viewing setting (arrows).
Figure 4. Stage Ta papillary bladder cancer in an 85-year-old woman. a: Transverse DWI shows a U-shaped high-intensity area (arrow) with a lowsignal stalk (white arrowhead) connecting to the anterior bladder wall (black arrowheads). b: Photomicrograph of the specimen shows a papillary cancer (star) with a stalk (asterisk) consisting of submucosa (hematoxylin-eosin stain).
Figure 5. Stage T1 sessile bladder cancer in an 80-year-old man. a: Axial T2-weighted image shows a sessile mass on the posterior bladder wall (arrow). The signal intensity (SI) of the bladder wall beneath the tumor is elevated, indicating muscle invasion (arrowhead). b: On axial diffusion-weighted image, the surface part of the tumor shows very high SI, indicating the cancer component (arrows), but SI of the deep part of the tumor is low, indicating an inflammatory or fibrosing reaction (arrowhead). This finding is an “inch worm sign”. Thus, the tumor was staged correctly as T1, though the tumor was sessile at cystoscopy.
DWI of Urinary Bladder Cancer Staging
1303
Figure 6. Stage T2 bladder cancer in a 65-year-old man. a: Sagittal T2-weighted image shows a hypointense oval mass on the dome of the bladder protruding into extravesical fat tissue (arrow). b: Sagittal diffusion-weighted image shows a hyperintense oval mass with a smooth contour (arrow).
Figure 7. Stage T3 bladder cancer in a 70-year-old man. a: Axial T2-weighted image shows a nonpapillary mass on the right-side bladder wall (arrow). The signal intensity (SI) of the bladder wall beneath the tumor is elevated, indicating muscle invasion (arrowhead). b: Axial diffusion-weighted image of a nonpapillary mass on the rightside bladder wall (arrow) with an irregular margin spreading toward the surrounding fat tissue (arrowhead).
Figure 8. Stage T3 bladder cancer in a 56-year-old woman. a: Axial T2-weighted image (T2WI) shows a large bladder mass protruding into the bladder lumen (arrow). The bladder wall beneath the mass is markedly thickened and the signal is very low, indicating fibrosis (arrowhead). b: In contrast to the prior case, the thickened bladder wall (arrowhead), which shows very low signal intensity (SI) on T2WI as well as the mass protruding into the bladder lumen (arrow), shows high SI on axial diffusion-weighted image. Thus, the tumor is staged correctly as T3.
1304
Takeuchi et al.
Figure 9. Stage T4 bladder cancer with prostatic invasion in a 72-year-old man. a: Sagittal T2-weighted image shows a nonpapillary lesion on the neck of the bladder (arrow). Although it seems that the tumor extends toward the urethra and prostate, the accurate extent of the tumor is unclear. b: Sagittal integrated diffusion-weighted image and T2-weighted image show that cancer on the neck of the bladder (arrow) clearly invades the prostate (arrowheads).
Tis Stage Tis cancer is invisible on MRI including DWI and it should be detected and staged by cystoscopy and TUR. Ta–T1 versus T2 It is most important to differentiate T2 from T1 or a lower stage in bladder cancer staging. Stage Ta or T1 cancer shows a papillary or sessile form on MRI. On DWI, the stalk of a papillary tumor, which consists of submucosa, generally shows low SI in contrast to the various SI on T2WI, and the tumor covering the stalk shows very high SI (Fig. 1a–c). It has been reported that the stalk consists of submucosa with edema and fibrosis (Fig. 4a,b) (3). Takeuchi et al (3) reported this unique finding as the inch worm sign. According to our unpublished data, DWI could almost always
(94%) visualize the stalk as low SI compared with T2WI; therefore, DWI could significantly reduce such false positivity for diagnosing muscle invasion compared with T2WI alone. DWI is also useful for staging sessile tumors. DWI could differentiate thickened submucosa from inflammatory change or fibrosis occurring beneath the tumor, which mimics muscle invasion on T2WI or DCEI. Inflammatory change or fibrosis beneath the tumor shows lower SI than that of the tumor and the intermediate SI of the muscle could be differentiated on DWI (Fig. 5a,b). In such a case, the area of high SI on DWI seems to be more important than the T2WI and DCEI findings and it could reduce false positivity for diagnosing muscle invasion. If a sessile lesion is seen and the SI of the muscle beneath the tumor is elevated but the area with high SI does not extend to the perivesical space and the
Figure 10. Stage T4 bladder cancer with uterine invasion in a 37-year-old woman. a: Sagittal T2-weighted image (T2WI) shows a large mass occupying the caudal part of the bladder (arrow). The signal intensity (SI) of the vaginal portion of the cervix is partially decreased (white arrowhead). Liquid accumulates in the vagina, indicating vesicovaginal fistula (black arrowhead). b: Sagittal integrated diffusion-weighted image and T2WI show that both bladder tumor (arrow) and lesion in the vaginal portion of cervix (arrowhead) are hyperintense.
DWI of Urinary Bladder Cancer Staging
1305
Figure 11. Bladder cancer extending into the prostatic urethra in a 71-year-old man. a: Sagittal T2-weighted image shows multiple papillary lesions on the neck and trigone of the bladder (arrows), one of which extends into the prostatic urethra (arrowhead). b: Sagittal integrated diffusionweighted image and T2weighted image show bladder cancer (arrows) extending into the prostatic urethra (arrowhead).
tumor margin is smooth on DWI, the stage is T2 (Fig. 6a,b). The TNM staging system subclassifies pT2 cancer into two categories: cancer invading less than half of the depth of the muscular propria (pT2a) and cancer invading greater than half of the muscle wall (7); however, MRI including DWI could not differentiate them accurately. The clinical utility of pT2 tumor substaging has been questioned, and, therefore,
Figure 12. Bladder cancer invades the prostate by means of the prostatic urethra in a 64-year-old man. a: The signal intensity of prostatic right peripheral zone is slightly decreased on axial T2-weighted image (arrow). There is no evident prostatic urethral lesion. b: Axial diffusion-weighted image shows signal increase of the prostatic urethra (arrow) and right lobe of the prostate (arrowhead). c: Photomicrograph of the specimen shows cancer invasion into the prostate (arrowheads) by means of the prostatic urethra (arrow) (hematoxylineosin stain).
radiologists need not be fixated on differentiating these sub-stages in our opinion (18). T2 versus T3 The efficacy of DWI for diagnosing extravesical invasion has been uncertain (3); however, we have experienced some cases which were diagnosed accurately
1306
Takeuchi et al.
Figure 13. Bladder cancer extending into the right ureter in a 68-year-old man. a: Axial T2-weighted image shows a large tumor in the bladder (arrow). The right ureter is dilated and the signal intensity of the lumen is decreased (arrowhead). b: Axial diffusion-weighted image shows bladder cancer (arrow) and a right ureteral lesion (arrowhead), which are clearly visualized as marked high intensity.
using DWI added to T2WI or DCEI. On DWI, if the contour of the mass invading the muscle is irregular or has a ridge protruding into the perivesical space, the T-stage is T3 (Fig. 7a,b). In differentiating T2 from T3, inflammatory change or fibrosis surrounding the tumor could mislead the T-staging using T2WI or DCEI, but DWI could differentiate them clearly because these changes do not show high SI on DWI (Figs. 2a,b, 8a,b). Differentiating T2b from T3a is quite difficult because of the limited spatial resolution; however, it is not a serious problem because the treatment procedure and prognosis are the same for these T-stages (19). T3 versus T4 Stage pT4 bladder cancer is defined by tumor invasion into an adjacent organ, including the uterus, vagina, prostate, pelvic wall, or abdominal wall. Stage T4a includes invasion to the prostate, uterus, or vagina, and stage T4b indicates pelvic or abdominal wall invasion. If the extent of high SI reflecting cancer invasion is seen in an adjacent organ on DWI, the stage is T4. The efficacy of DWI added to conventional
imaging for diagnosing stage T4 cancer is uncertain, but tumor invasion into adjacent organs is clearly seen on DWI compared with conventional imaging because the tumor is highlighted (Figs. 9a,b, 10a,b). Integrated DWI and T2WI would be useful to evaluate the invasion to these organs. Donat et al (20) identified three pathways for prostatic stromal invasion by urothelial carcinoma. These include extravesical, intraurethral, and bladder neck invasion. The 5-year overall survival rate was 21% for the patients with extravesical and bladder neck invasion and 55% for the patients with intraurethral invasion (21,22). Bladder cancer invading the prostate by means of the bladder neck or through the perivesical fat tissue is staged as T4a but bladder cancer extending into the prostate by means of the urethra is not staged as T4a in the existing TNM classification.
Supplement Detecting prostatic extension by means of the urethra and ureteral extension of bladder cancer is important because the extent of resection changes depending on
Figure 14. A case of pyuria in a 79-year-old man. a: On diffusion-weighted image, there is a markedly high signal intensity (SI) area in the posterior part of the bladder lumen with a fluid-fluid level (arrow). b: On T2-weighted image, the SI of the urine in the posterior part of the bladder is decreased and a fluidfluid level is seen (arrow).
DWI of Urinary Bladder Cancer Staging
1307
Figure 15. A 72-year-old man with T1 bladder cancer with intravesical hemorrhage. a: On diffusion-weighted image, there are a high signal intensity (SI) mass with a low SI stalk on the right-side bladder wall (arrow) and a markedly high SI mass at the left side of the bladder lumen (arrowhead). b: On T1weighted image, the rightside mass shows intermediate SI, indicating tumor (arrow) and the left-side mass shows high SI, indicating hemorrhage (arrowhead).
the extent of bladder cancer. On DWI, normal prostatic urethra shows low SI and prostatic urethra with cancer shows high SI along the urethra (Fig. 11a,b) (4). If areas of high SI are seen in the prostatic urethra and prostate simultaneously, it is a finding of prostatic invasion by means of the urethra (Fig. 12a–c) (4);
Figure 16. A susceptibility artifact from rectal gas in a 76-year-old man. a: There is bow-shaped high signal intensity (SI) at the posterior bladder wall (arrow) on diffusion-weighted image (DWI). The area surrounding the rectum (white arrowhead) and uterus (black arrowhead) is distorted. b: On T2weighted image, there is no tumor corresponding to the high SI area on DWI (arrow).
Figure 17. Prostatic abscess in a 70-year-old man. a: Axial diffusion-weighted image shows a markedly high signal intensity (SI) nodule in the left peripheral zone of the prostate (arrow). b: Axial T2-weighted image shows a relatively high SI nodule in the left peripheral zone of the prostate (arrow).
however, incidental prostatic cancer also shows high SI on DWI, and a prostatic biopsy is needed to differentiate them. Bladder cancer could extend into the ureter along the urothelial epithelium. If ureteral extension is present, a nephro-ureterectomy is performed. If the
1308
Takeuchi et al.
Figure 18. Granulomatous prostatitis induced by intravesical Bacille-Calmette-Guerin therapy in a 72-year-old man. a: Axial diffusionweighted image shows diffuse signal elevation of the prostate (arrows). b: The prostate is enlarged and replaced with the lesion which shows intermediate signal intensity on T2-weighted image and the zonal anatomy of the prostate gland is indistinct (arrows).
ureteral orifice is covered with bladder cancer, the ureteral extension cannot be observed by cystoscopy; therefore, radiologists need to notify urologists of this finding. Cancer extending into the ureter shows high SI on DWI (Fig. 13a,b).
Pitfalls Some benign lesions and artifacts show high SI and mimic bladder cancer, which could lead to an incorrect T-stage of bladder cancer. It is important to compare DWI with T2WI and DCE to diagnose falsepositive lesions accurately. Purulent urine shows very high SI on DWI (Fig. 14a). Purulent urine settles out in the bladder or diverticulum. A relatively high SI on T2WI and a fluid–fluid level seen between normal and purulent urine are diagnostic keys to purulent urine (Fig. 14b). Clots that sometimes accompany bladder cancer also show high SI on DWI (Fig. 15a). They show high SI on T1-weighted images and are not enhanced (Fig. 15b). Susceptibility artifact from gases in the colon and rectum causes distortion of the images and could result in a false-positive lesion (Fig. 16a,b). This false-positive lesion is not seen on images other than DWI. Air in the bladder is also seen after cystoscopy or insertion of a catheter; therefore, MRI is not recommended after these procedures. Prostatic inflammation including abscess and granulomatous prostatitis induced by intravesical Bacille-CalmetteGuerin (BCG) therapy also shows high SI (Figs. 17a,b, 18a,b) (4). Clinical symptoms and treatment history should be examined to exclude these inflammatory diseases. In conclusion, DWI provides more accurate information on the extent of bladder cancer and contributes to the determination of treatment strategy. We recommend that DWI is added to conventional bladder MR examination. REFERENCES 1. Siegel R, Naishadham D, Jemal A. Cancer statistics, 2013. CA Cancer J Clin 2013;63:11–30.
2. Matsuki M, Inada Y, Tatsugami F, Tanikake M, Narabayashi I, Katsuoka Y. Diffusion-weighted MR imaging for urinary bladder carcinoma: initial results. Eur Radiol 2007;17:201–204. 3. Takeuchi M, Sasaki S, Ito M, et al. Urinary bladder cancer: diffusion-weighted MR imaging--accuracy for diagnosing T stage and estimating histologic grade. Radiology 2009;251: 112–121. 4. Takeuchi M, Suzuki T, Sasaki S, et al. Clinicopathologic significance of high signal intensity on diffusion-weighted MR imaging in the ureter, urethra, prostate and bone of patients with bladder cancer. Acad Radiol 2012;19:827–833. 5. Watanabe H, Kanematsu M, Kondo H, et al. Preoperative T staging of urinary bladder cancer: does diffusion-weighted MRI have supplementary value? AJR Am J Roentgenol 2009;192:1361– 1366. 6. El-Assmy A, Abou-El-Ghar ME, Mosbah A, et al. Bladder tumour staging: comparison of diffusion- and T2-weighted MR imaging. Eur Radiol 2009;19:1575–1581. 7. Urinary bladder. In: Edge SB, Byrd DR, Compton CC, et al, editors. AJCC Cancer Staging Manual. 7th edition. New York, NY: Springer, 2010. p 497–505. 8. Tekes A, Kamel I, Imam K, et al. Dynamic MRI of bladder cancer: evaluation of staging accuracy. AJR Am J Roentgenol 2005;184:121–127. 9. Saito W, Amanuma M, Tanaka J, Heshiki A. Histopathological analysis of a bladder cancer stalk observed on MRI. Magn Reson Imaging 2000;18:411–415. 10. Kim B, Semelka RC, Ascher SM, Chalpin DB, Carroll PR, Hricak H. Bladder tumor staging: comparison of contrast-enhanced CT, T1- and T2-weighted MR imaging, dynamic gadolinium-enhanced imaging, and late gadolinium-enhanced imaging. Radiology 1994; 193:239–245. 11. Tanimoto A, Yuasa Y, Imai Y, et al. Bladder tumor staging: comparison of conventional and gadolinium-enhanced dynamic MR imaging and CT. Radiology 1992;185:741–747. 12. Hayashi N, Tochigi H, Shiraishi T, Takeda K, Kawamura J. A new staging criterion for bladder carcinoma using gadoliniumenhanced magnetic resonance imaging with an endorectal surface coil: a comparison with ultrasonography. BJU Int 2000;85: 32–36. 13. Koh DM, Collins DJ. Diffusion-weighted MRI in the body: applications and challenges in oncology. AJR Am J Roentgenol 2007;188:1622–1635. 14. Narumi Y, Kadota T, Inoue E, et al. Bladder tumors: staging with gadolinium-enhanced oblique MR imaging. Radiology 1993;187: 145–150. 15. Kazama T, Nasu K, Kuroki Y, Nawano S, Ito H. Comparison of diffusion-weighted images using short inversion time inversion recovery or chemical shift selective pulse as fat suppression in patients with breast cancer. Jpn J Radiol 2009;27:163–167.
DWI of Urinary Bladder Cancer Staging 16. Kirkali Z, Chan T, Manoharan M, et al. Bladder cancer: epidemiology, staging and grading, and diagnosis. Urology 2005;66(Suppl 1):4–34. 17. Wong-You-Cheong JJ, Woodward PJ, Manning MA, Sesterhenn IA. Neoplasms of the urinary bladder: radiologic-pathologic correlation. Radiographics 2006;26:553–580. 18. Cheng L, Neumann RM, Weaver AL, et al. Grading and staging of bladder carcinoma in transurethral resection specimens. Correlation with 105 matched cystectomy specimens. Am J Clin Pathol 2000;113:275–279. 19. Quek ML, Stein JP, Clark PE, et al. Microscopic and gross extravesical extension in pathological staging of bladder cancer. J Urol 2004;17:640–645.
1309 20. Donat SM, Genega EM, Herr HW, Reuter VE. Mechanisms of prostatic stromal invasion in patients with bladder cancer: clinical significance. J Urol 2001;165:1117–1120. 21. Esrig D, Freeman JA, Elmajian DA, et al. Transitional cell carcinoma involving the prostate with a proposed staging classification for stromal invasion. J Urol 1996;156:1071– 1076. 22. Pagano F, Bassi P, Ferrante GL, et al. Is stage pT4a (D1) reliable in assessing transitional cell carcinoma involvement of the prostate in patients with a concurrent bladder cancer? A necessary distinction for contiguous or noncontiguous involvement. J Urol 1996;155:244–247.
