- Email: [email protected]
Ultrasonographic Assessment of Bladder Tumors. II. Clinical Staging
0022-5347/81/1261-0034$02.00/0 Vol. 126, July Printed in U.S.A.
THE JOURNAL OF UROLOGY
Copyright© 1981 by The Williams & Wilkins Co.
ULTRASONOGRAPHIC ASSESSMENT OF BLADDER TUMORS. II. CLINICAL STAGING DOV SINGER, YACOV ITZCHAK
AND
YACOV FISCHELOVITCH
From the Department of Diagnostic Ultrasound, Chaim Sheba Medical Center, Tel Aviv University, Tel-Hashomer, Department of Urology, Kaplan Hospital, Rehovot and Medical School of Hebrew University and Hadassah, Jerusalem, Israel
ABSTRACT
Uniformity of the echo return from the bladder wall has been demonstrated in experimental and patient studies. An infiltrating bladder tumor disrupts this uniformity. The studies also showed a sharp bladder wall outline that was less echogenic, probably representing the perivesical fat. Thus, the ultrasonographic estimation of tumor staging was related to the depth of infiltration of the bladder wall. A staging accuracy of nearly 100 per cent was found for deep tumors of the bladder wall or those penetrating extravesical structures. On the other hand, many superficial tumors were overstaged, their staging accuracy being only slightly higher than 55 per cent. However, in view of its accuracy in staging infiltrating tumors, in addition to its other already known advantages, ultrasound seems to be the method of choice in preoperative staging of tumors. We believe that it should be included in the routine protocol and should follow diagnostic cystoscopy in all cases of tumors larger than 0.5 cm. were obtained using frequencies of 2.5, 3.5 and 5 MHz. The bladders then were sectioned along a sagittal axis, and the anterior and posterior walls as well as the trigone were scanned separately (fig. 1, A). Patient studies. Normal anatomy: A retrospective study of 20 ultrasonic scans of full bladders from patients with normal pelvic anatomy was done to define accurately the outline of the normal bladder wall (fig. 1, B). Technique: A standard abdominal scanning technique was used. Patients were examined with full and partially empty bladders using a 3.5 MHz. frequency. The equipment consisted
The conventional staging of bladder neoplasms consists essentially of a transurethral biopsy and careful bimanual palpation under anesthesia. Radiological procedures that have been used include excretory urography, fractionated cystography, lymphangiography and pelvic arteriography. 1 More recently, the development of computerized tomography and the advent of gray scale ultrasonography have provided new means for the non-invasive staging of bladder tumors. After encouraging previous reports on ultrasonographic staging using a transrectal probe,1· 2 an attempt is made herein to evaluate the accuracy of the abdominal approach. According to
FIG. 1. Normal bladder wall. A, in vitro ultrasound scan of sectioned posterior wall 0.5 cm. thick (arrows). B, longitudinal scan of normal bladder shows echo-lucent strip delineating outer limit of posterior wall (arrowheads). Note thickness of wall in trigone area (open arrows). Each dot represents 1 cm.
these results and to the conclusions drawn in part I of this report3 we attempted to determine the place of ultrasound examination in the diagnostic study protocol of bladder tumors.
of an SOL Echoview System gray scale scanner.* Study group: The study group was the same as in part I of our report and consisted of 31 male and 6 female patients, with a total of 71 separate tumors. Two patients with 3 large tumors refused any operation, even palliative. At operation 10 tumors were considered small and superficial enough to undergo fulguration. Since these 10 tumors belonged to patients with multiple tumors they were not biopsied before fulguration. The 58 remaining tumors were resected and were examined
MATERIALS AND METHODS
Experimental studies. We first studied 5 cadaver bladders without any macroscopic pathology in an attempt to characterize ultrasonically the various layers of the normal bladder wall. Every bladder was submerged in water and scanned when filled and partially empty. Longitudinal, transverse and oblique scans
* Picker Corporation, 12 Clintonville Rd., Northford, Connecticut 06472.
Accepted for publication September 19, 1980.
34
ULTRASONOGRAPHIC ASSESSMENT OF BLADDER TUMORS
35
FIG. 2. Transverse scan at 2 cm. above symphysis pubis shows tumor invading part of wall. Note differences in echo reflection from involved segment of wall (arrows).
pathologically. Ultimately, pathological staging was obtained in 27 tumors and compared to the clinical staging as established by ultrasonography. In the other 31 tumors histological differentiation grade was reported but the degree of infiltration could not be established. RESULTS
The experimental studies showed the whole thickness of the bladder wall to have a uniform echo pattern, making the distinction of the muscle layer from the mucosa or the submucosa impossible. Resolution did not improve by using magnification, focusing or various frequencies. With a 5 MHz. transducer good visualization of the anterior wall was achieved but the optimal scanning frequency was found to be 3.5 MHz. This uniformity of the bladder wall also was found in the retrospective study of the normal bladder scans. However, on these scans we were able to define a sharp bladder outline around the posterior and lateral walls (fig. 1, B). This outer limit of the bladder wall is a thin strip less echogenic than the bladder wall itself that was seen on longitudinal and transverse scans and was thought to represent the perivesical fat. Other landmarks helpful in defining the thickness of the posterior bladder wall are the prostate and the uterus. The outline of the anterior wall could not always be distinguished from the subcutaneous tissue, nor could the dome area be delineated clearly from intraperitoneal contents. An infiltrating tumor causes disruption in the uniformity of the echoes returning from the bladder wall, thus, making it possible to assess the depth of penetration as well as the extent of the superficial spread on the mucosal surface (fig. 2). After the outer limit of the bladder wall was delineated exactly the ultrasonographic estimation of tumor staging was related to the depth of infiltration of this wall. Accordingly, we divided all tumors into 3 groups using the Jewett-StrongMarshaH classification: stages O and Bl-superficial or no tumor invasion (fig. 3, A), stage B2-deep bladder wall invasion (fig. 3, B), and stages C and D-extravesical penetration (fig. 3, C). The third group includes pelvic tumors infiltrating into the bladder wall as well. A staging accuracy of nearly 100 per cent was found in groups 2 and 3, while only 55 per cent of the superficial tumors were staged accurately (see table). DISCUSSION AND CONCLUSIONS
Accurate staging of a bladder tumor is of crucial importance for establishing treatment and predicting prognosis. Therefore, a whole series of diagnostic procedures presently is available for this purpose. Not only are some of them relatively invasive and discomforting4 but Schmidt and Weinstein found that all of them were associated with an understaging error of 25 to 50
FIG. 3. A, ultrasonic longitudinal scan of bladder 3 cm. to right of midline reveals small non-invasive tumors on posterior wall. Note normal echo reflections from bladder wall behind tumor and normal outer limit of wall (arrowheads). B, oblique scan at 3.5 cm. to right of midline shows tumor penetrating deep into wall, interrupting normal echo reflections. Continuity of outer limit of wall is preserved (arrowheads). C, transverse scan at 0.5 cm. above symphysis pubis shows large tumor on left lateral wall protruding beyond bladder wall. Note interruption of normal wall in affected area (arrow).
per cent; the higher the stage the greater the rate of error." In our study the superficial tumors had a relatively low staging accuracy, which increased as the tumor infiltrated deeper in the bladder wall. However, it should be noted that in no case were the tumors understaged. The overstaging observed in this study was mostly caused by difficulty in distinguishing submucosa from the muscularis owing to contiguous normal extravesical structures and because of thickening of the bladder wall after previous surgery or radiation. These findings are in correlation with those of others regardless of the scanning method (abdominal" or transrectal 1). Of the 10 small, clearly superficial tumors that were fulgur-
36
SINGER, ITZCHAK AND FISCHELOVITCH
Accuracy of ultrasonic tumor staging according to pathological findings No. Tumors
Superficial or no invasion Deep bladder.wall invasion Beyond bladder wall invasion or penetration into bladder from pelvic tumors Totals
20 2 5
27
Accurate Staging No.(%) 11 (55) 2 (100)
~
18
ated without biopsy 8 were staged correctly as such by the ultrasonographic examination. Thus, the actual accuracy in staging superficial tumors was higher than the 55 per cent shown in the table. Although still subject to discussion most investigators agree that superficial tumors are amenable to a less aggressive management on one hand and that metastatic disease is to be treated palliatively only on the other. The treatment of choice for deeply infiltrating tumors should be an extensive operation with radiation therapy. 4 Therefore, it seems that accurate staging of a deeply invasive or metastatic tumor, as obtained in our study, clinically is more important to the urologist in terms of surgical decisions and prognosis than the consequences of an erroneous staging of a superficial tumor. Thus, it becomes obvious that ultrasound, while being only of secondary importance in tumor diagnosis as shown in part I of this report,3 is the method of choice in preoperative tumor staging. It should be included in the routine protocol and should follow diagnostic cystoscopy in all cases of tumors that are >0.5 cm., <0.5 cm. but with high malignancy grade and located on the posterior or lateral walls, flat and solid with large mucosal involvement, otherwise hard to outline accurately, and suspected as protruding into the bladder from adjacent tissues. In addition to accuracy, there are other advantages in the ultrasonographic staging of bladder tumors. The technique offers in 1 examination a reliable preoperative evaluation of node involvement or distant metastases, which is of major clinical importance. Since no prior patient preparation or postprocedure observation is necessary the examination can be done on an outpatient basis.
Although there were no tumors on the anterior wall in our study their accurate staging should be difficult. Accurate staging of these tumors probably awaits a better delineation of this wall from subcutaneous tissue, possibly by using short focus transducers or a water bag placed on the abdomen during examination. There have been reports of similar results and benefits in tumor staging by computerized tomography. Since this technique is associated with some radiation exposure and the everpresent risk of acute reaction to intravenous administration of contrast material ultrasound remains the only absolutely noninvasive staging method. Therefore, it appears to us that the ease, wide accessibility and relatively low cost of ultrasound should make it the preferred modality following cystoscopy in the diagnostic studies of a bladder tumor. Prof. M. Manny, Department of Urology, Chaim Sheba Medical Center, Tel-Hashomer, and Dr. M. Meltzer, Department of Urology, Assaf Harofeh Hospital, Zerifin, Israel referred their patients to us. REFERENCES 1. Resnick, M. I., Willard, J. W. and Boyce, W. H.: Recent progress in ultrasonography of the bladder and prostate. J. Urol., 117: 444, 1977. 2. Boyce, W. H., McKinney, W. M., Resnick, M. I. and Willard, J. W.: Ultrasonography as an aid in the diagnosis and management of surgical diseases of the pelvis: special emphasis on the genitourinary system. Ann. Surg., 184: 477, 1976. 3. Itzchak, Y., Singer, D. and Fischelovitch, Y.: Ultrasonographic assessment of bladder tumors. I. Tumor detection. J. Urol., 126: 31, 1981. 4. Gittes, R. F.: Tumors of the bladder. In: Campbell's Urology, 4th ed. Edited by J. H. Harrison, R. F. Gittes, A. D. Perlmutter, T. A. Stamey and P. C. Walsh. Philadelphia: W. B. Saunders Co., vol. 2, chapt. 29, p. 1048, 1979. 5. Schmidt, J. D. and Weinstein, S. H.: Pitfalls in clinical staging of bladder tumors. Urol. Clin. N. Amer., 3: 107, 1976. 6. Morley, P.: Clinical staging of epithelial bladder tumours by echotomography. In: Ultrasound in Tumor Diagnosis. Edited by C. R. Hill, V. R. McCready and D. 0. Cosgrove. London: Pitman Publishing, chapt. 8, p. 145, 1979.
THE JOURNAL OF UROLOGY
Copyright© 1981 by The Williams & Wilkins Co.
ULTRASONOGRAPHIC ASSESSMENT OF BLADDER TUMORS. II. CLINICAL STAGING DOV SINGER, YACOV ITZCHAK
AND
YACOV FISCHELOVITCH
From the Department of Diagnostic Ultrasound, Chaim Sheba Medical Center, Tel Aviv University, Tel-Hashomer, Department of Urology, Kaplan Hospital, Rehovot and Medical School of Hebrew University and Hadassah, Jerusalem, Israel
ABSTRACT
Uniformity of the echo return from the bladder wall has been demonstrated in experimental and patient studies. An infiltrating bladder tumor disrupts this uniformity. The studies also showed a sharp bladder wall outline that was less echogenic, probably representing the perivesical fat. Thus, the ultrasonographic estimation of tumor staging was related to the depth of infiltration of the bladder wall. A staging accuracy of nearly 100 per cent was found for deep tumors of the bladder wall or those penetrating extravesical structures. On the other hand, many superficial tumors were overstaged, their staging accuracy being only slightly higher than 55 per cent. However, in view of its accuracy in staging infiltrating tumors, in addition to its other already known advantages, ultrasound seems to be the method of choice in preoperative staging of tumors. We believe that it should be included in the routine protocol and should follow diagnostic cystoscopy in all cases of tumors larger than 0.5 cm. were obtained using frequencies of 2.5, 3.5 and 5 MHz. The bladders then were sectioned along a sagittal axis, and the anterior and posterior walls as well as the trigone were scanned separately (fig. 1, A). Patient studies. Normal anatomy: A retrospective study of 20 ultrasonic scans of full bladders from patients with normal pelvic anatomy was done to define accurately the outline of the normal bladder wall (fig. 1, B). Technique: A standard abdominal scanning technique was used. Patients were examined with full and partially empty bladders using a 3.5 MHz. frequency. The equipment consisted
The conventional staging of bladder neoplasms consists essentially of a transurethral biopsy and careful bimanual palpation under anesthesia. Radiological procedures that have been used include excretory urography, fractionated cystography, lymphangiography and pelvic arteriography. 1 More recently, the development of computerized tomography and the advent of gray scale ultrasonography have provided new means for the non-invasive staging of bladder tumors. After encouraging previous reports on ultrasonographic staging using a transrectal probe,1· 2 an attempt is made herein to evaluate the accuracy of the abdominal approach. According to
FIG. 1. Normal bladder wall. A, in vitro ultrasound scan of sectioned posterior wall 0.5 cm. thick (arrows). B, longitudinal scan of normal bladder shows echo-lucent strip delineating outer limit of posterior wall (arrowheads). Note thickness of wall in trigone area (open arrows). Each dot represents 1 cm.
these results and to the conclusions drawn in part I of this report3 we attempted to determine the place of ultrasound examination in the diagnostic study protocol of bladder tumors.
of an SOL Echoview System gray scale scanner.* Study group: The study group was the same as in part I of our report and consisted of 31 male and 6 female patients, with a total of 71 separate tumors. Two patients with 3 large tumors refused any operation, even palliative. At operation 10 tumors were considered small and superficial enough to undergo fulguration. Since these 10 tumors belonged to patients with multiple tumors they were not biopsied before fulguration. The 58 remaining tumors were resected and were examined
MATERIALS AND METHODS
Experimental studies. We first studied 5 cadaver bladders without any macroscopic pathology in an attempt to characterize ultrasonically the various layers of the normal bladder wall. Every bladder was submerged in water and scanned when filled and partially empty. Longitudinal, transverse and oblique scans
* Picker Corporation, 12 Clintonville Rd., Northford, Connecticut 06472.
Accepted for publication September 19, 1980.
34
ULTRASONOGRAPHIC ASSESSMENT OF BLADDER TUMORS
35
FIG. 2. Transverse scan at 2 cm. above symphysis pubis shows tumor invading part of wall. Note differences in echo reflection from involved segment of wall (arrows).
pathologically. Ultimately, pathological staging was obtained in 27 tumors and compared to the clinical staging as established by ultrasonography. In the other 31 tumors histological differentiation grade was reported but the degree of infiltration could not be established. RESULTS
The experimental studies showed the whole thickness of the bladder wall to have a uniform echo pattern, making the distinction of the muscle layer from the mucosa or the submucosa impossible. Resolution did not improve by using magnification, focusing or various frequencies. With a 5 MHz. transducer good visualization of the anterior wall was achieved but the optimal scanning frequency was found to be 3.5 MHz. This uniformity of the bladder wall also was found in the retrospective study of the normal bladder scans. However, on these scans we were able to define a sharp bladder outline around the posterior and lateral walls (fig. 1, B). This outer limit of the bladder wall is a thin strip less echogenic than the bladder wall itself that was seen on longitudinal and transverse scans and was thought to represent the perivesical fat. Other landmarks helpful in defining the thickness of the posterior bladder wall are the prostate and the uterus. The outline of the anterior wall could not always be distinguished from the subcutaneous tissue, nor could the dome area be delineated clearly from intraperitoneal contents. An infiltrating tumor causes disruption in the uniformity of the echoes returning from the bladder wall, thus, making it possible to assess the depth of penetration as well as the extent of the superficial spread on the mucosal surface (fig. 2). After the outer limit of the bladder wall was delineated exactly the ultrasonographic estimation of tumor staging was related to the depth of infiltration of this wall. Accordingly, we divided all tumors into 3 groups using the Jewett-StrongMarshaH classification: stages O and Bl-superficial or no tumor invasion (fig. 3, A), stage B2-deep bladder wall invasion (fig. 3, B), and stages C and D-extravesical penetration (fig. 3, C). The third group includes pelvic tumors infiltrating into the bladder wall as well. A staging accuracy of nearly 100 per cent was found in groups 2 and 3, while only 55 per cent of the superficial tumors were staged accurately (see table). DISCUSSION AND CONCLUSIONS
Accurate staging of a bladder tumor is of crucial importance for establishing treatment and predicting prognosis. Therefore, a whole series of diagnostic procedures presently is available for this purpose. Not only are some of them relatively invasive and discomforting4 but Schmidt and Weinstein found that all of them were associated with an understaging error of 25 to 50
FIG. 3. A, ultrasonic longitudinal scan of bladder 3 cm. to right of midline reveals small non-invasive tumors on posterior wall. Note normal echo reflections from bladder wall behind tumor and normal outer limit of wall (arrowheads). B, oblique scan at 3.5 cm. to right of midline shows tumor penetrating deep into wall, interrupting normal echo reflections. Continuity of outer limit of wall is preserved (arrowheads). C, transverse scan at 0.5 cm. above symphysis pubis shows large tumor on left lateral wall protruding beyond bladder wall. Note interruption of normal wall in affected area (arrow).
per cent; the higher the stage the greater the rate of error." In our study the superficial tumors had a relatively low staging accuracy, which increased as the tumor infiltrated deeper in the bladder wall. However, it should be noted that in no case were the tumors understaged. The overstaging observed in this study was mostly caused by difficulty in distinguishing submucosa from the muscularis owing to contiguous normal extravesical structures and because of thickening of the bladder wall after previous surgery or radiation. These findings are in correlation with those of others regardless of the scanning method (abdominal" or transrectal 1). Of the 10 small, clearly superficial tumors that were fulgur-
36
SINGER, ITZCHAK AND FISCHELOVITCH
Accuracy of ultrasonic tumor staging according to pathological findings No. Tumors
Superficial or no invasion Deep bladder.wall invasion Beyond bladder wall invasion or penetration into bladder from pelvic tumors Totals
20 2 5
27
Accurate Staging No.(%) 11 (55) 2 (100)
~
18
ated without biopsy 8 were staged correctly as such by the ultrasonographic examination. Thus, the actual accuracy in staging superficial tumors was higher than the 55 per cent shown in the table. Although still subject to discussion most investigators agree that superficial tumors are amenable to a less aggressive management on one hand and that metastatic disease is to be treated palliatively only on the other. The treatment of choice for deeply infiltrating tumors should be an extensive operation with radiation therapy. 4 Therefore, it seems that accurate staging of a deeply invasive or metastatic tumor, as obtained in our study, clinically is more important to the urologist in terms of surgical decisions and prognosis than the consequences of an erroneous staging of a superficial tumor. Thus, it becomes obvious that ultrasound, while being only of secondary importance in tumor diagnosis as shown in part I of this report,3 is the method of choice in preoperative tumor staging. It should be included in the routine protocol and should follow diagnostic cystoscopy in all cases of tumors that are >0.5 cm., <0.5 cm. but with high malignancy grade and located on the posterior or lateral walls, flat and solid with large mucosal involvement, otherwise hard to outline accurately, and suspected as protruding into the bladder from adjacent tissues. In addition to accuracy, there are other advantages in the ultrasonographic staging of bladder tumors. The technique offers in 1 examination a reliable preoperative evaluation of node involvement or distant metastases, which is of major clinical importance. Since no prior patient preparation or postprocedure observation is necessary the examination can be done on an outpatient basis.
Although there were no tumors on the anterior wall in our study their accurate staging should be difficult. Accurate staging of these tumors probably awaits a better delineation of this wall from subcutaneous tissue, possibly by using short focus transducers or a water bag placed on the abdomen during examination. There have been reports of similar results and benefits in tumor staging by computerized tomography. Since this technique is associated with some radiation exposure and the everpresent risk of acute reaction to intravenous administration of contrast material ultrasound remains the only absolutely noninvasive staging method. Therefore, it appears to us that the ease, wide accessibility and relatively low cost of ultrasound should make it the preferred modality following cystoscopy in the diagnostic studies of a bladder tumor. Prof. M. Manny, Department of Urology, Chaim Sheba Medical Center, Tel-Hashomer, and Dr. M. Meltzer, Department of Urology, Assaf Harofeh Hospital, Zerifin, Israel referred their patients to us. REFERENCES 1. Resnick, M. I., Willard, J. W. and Boyce, W. H.: Recent progress in ultrasonography of the bladder and prostate. J. Urol., 117: 444, 1977. 2. Boyce, W. H., McKinney, W. M., Resnick, M. I. and Willard, J. W.: Ultrasonography as an aid in the diagnosis and management of surgical diseases of the pelvis: special emphasis on the genitourinary system. Ann. Surg., 184: 477, 1976. 3. Itzchak, Y., Singer, D. and Fischelovitch, Y.: Ultrasonographic assessment of bladder tumors. I. Tumor detection. J. Urol., 126: 31, 1981. 4. Gittes, R. F.: Tumors of the bladder. In: Campbell's Urology, 4th ed. Edited by J. H. Harrison, R. F. Gittes, A. D. Perlmutter, T. A. Stamey and P. C. Walsh. Philadelphia: W. B. Saunders Co., vol. 2, chapt. 29, p. 1048, 1979. 5. Schmidt, J. D. and Weinstein, S. H.: Pitfalls in clinical staging of bladder tumors. Urol. Clin. N. Amer., 3: 107, 1976. 6. Morley, P.: Clinical staging of epithelial bladder tumours by echotomography. In: Ultrasound in Tumor Diagnosis. Edited by C. R. Hill, V. R. McCready and D. 0. Cosgrove. London: Pitman Publishing, chapt. 8, p. 145, 1979.