- Email: [email protected]
Eight Years of Experience with Preoperative Angiographic and Lymphographic Staging of Bladder Cancer
CJ
0022-534 7!78/1192-0208$02. 00/0 THE JOURNAL OF UROLOGY Copyright © 1978 by The Williams & Wilkins Co.
Vol. 119, February Printed in U.S.A.
EIGHT YEARS OF EXPERIENCE WITH PREOPERATIVE ANGIOGRAPHIC AND LYMPHOGRAPHIC STAGING OF BLADDER CANCER A. R. WINTERBERGER, Z. WAJSMAN, C. MERRIN
AND
G. P. MURPHY*
From the Departments of Diagnostic Radiology and Urologic Oncology, Roswell Park Memorial Institute, New York State Department of Health, Buffalo, New York
ABSTRACT
Our experience with the preoperative staging of bladder cancer by bilateral selective hypogastric arteriography has been accumulated since 1968. More than 150 patients have been studied by selective angiography before radical cystectomy. Our latest series of 52 patients (1972 to 1976) compares to our previous experience demonstrating angiographic staging accuracy to detect bladder invasion and occult metastases at a rate exceeding that of clinical staging alone. Arteriographic staging ofD lesions, when supplemented with lymphography, approaches 100 per cent accuracy. Falsely negative lymphograms currently are extremely uncommon (1.9 per cent). In several illustrated instances angiographic staging was proved to be even more accurate than the pathologic staging of a limited cystectoiny specimen. The over-all angiographic and lymphangiographic staging accuracy in our most recent series of cystectomy patients was 78.8 per cent. The techniques and reliability of the data are discussed in detail, including the factors that interfere with the exact arteriographic staging of bladder cancer. However, these factors are more troublesome in early stage lesions. These studies demonstrate the role and value as well as areas of limitation of preoperative arteriography and lymphography in the evaluation of invasive bladder cancer. Our data with 60 of the first 100 sequential bladder cancer angiographic staging. The specific details for 112 of the 150 patients undergoing selective arteriography with subsequent patients are shown in tables 1 and 2. The information is positive pathologic correlation were published initially in 1972 available in the remaining 38 patients but will not be used to (table 1). 1 Our experience at that time suggested that the confirm the accuracy of tumor staging since followup is less addition of arteriographic staging in bladder cancer, particu- than 1 year. larly in suspected clinically invasive lesions, improved overTECHNIQUES all accuracy in preoperative staging. After additional clinical studies in 1973 and 19742 we concluded that the addition of The Marshall modification of the Jewett and Strong classifipedal lymphography to the radiographic staging further in- cation of bladder cancer was used generally by those particicreased preoperative staging accuracy. The clinical value of pating in this study. 5 Clinical stages of bladder cancer in this lymphography in the staging of bladder cancer also has been system are as follows: stage O neoplasms are in situ lesions described in 1975 by Wajsman3 and in 1976 by Merrin 4 and limited to the mucosa, stage A tumors have infiltrated the their associates. Herein we present the data on the total submucosa but not the muscularis, stage Bl tumors have series of patients from our institution to substantiate the infiltrated the superficial musculature, stage B2 lesions inusefulness of combined preoperative angiography and lym- volve the deep musculature, stage C tumors extend to the phangiography in staging preoperative bladder tumors. perivesical and fat, stage Dl lesions have invaded the pelvic wall structures, pelvic lymph nodes or adjacent organs and METHODS stage D2 patients have metastatic involvement beyond the The study includes 150 patients with a proved diagnosis of pelvis. 5 bladder cancer who were candidates for radical cystectomy TABLE 1. Preoperative staging of bladder tumors (60 cases with between 1968 and October 1976. All patients underwent roudefinitive pathological diagnosis): 1968 to 1972 tine clinical staging, including excretory urography (IVP), bimanual examination under general anesthesia and cystosStage Totals copy with transurethral biopsy. After this clinical staging No.(%) A B C D bilateral selective hypogastric arteriography and bilateral peNo. (%) No. (%) No. (%) No. (%) dal lymphography were performed. After the independent in- Staged exactly arterio- 2/7 (29) 10/19 (53) 8/9 (89) 22/24 (92) 42/60* (72) terpretation of the angiographic studies the majority of the graphically patients were subjected to radical cystectomy and pelvic lymphadenectomy. The gross pathological features, as well as Staged exactly clinically 4/7 (57) 10/19 (53) 1/9 (11) 17/24 (71) 32/60* (53) the histology with numerous slides of bladder step sections and Pathology proved by cys46*/60* (77) 10 7 19 9 lymph node preparations from the surgical specimens, were tectomy later correlated with the preoperative clinical staging and the Pathology proved by surgical exploration
Accepted for publication April 7, 1977. Supported in part by United States Public Health Service Grant RR-05648-10 of the National Institutes of Health. * Requests for reprints: Roswell Park Memorial Institute, 666 Elm St., Buffalo, New York 14263.
Total pts.
7
19
9
None of these patients had lymphography. * One patient had no tumor at the time of operation.
208
14
14/60* (23)
24
60*
BLADDER CANCER ANGIOGRAPHY TABLE
2, Preoperative staging of bladder tumors (52 cases with
pathologic correlation from cystectomy specimens): 1972 to 1976 Stage A
B
C
D
No, (%)
No, (%)
No, (%)
No, (%)
Totals No,(%)
Staged exactly 7/9 (77.8) 5/11 (45,5) 8/10 (80.0) 19/22 (85.6) 39/52 (75.5) arteriographica!ly 7/9 (77,8)
Stage changed by lymphography
0
6/9 (66,7)
Tota! pts,
* In
9
5/11 (45,5) 8/10 (80,0) 21/22 (95,5) 41/52 (78,8)
0
0
4/11 (36.4) 3/10 (30,0)
11
10
that have infiltrated the bladder walL In such tumors when lymphography is added to the approach 100 per cent. As with our arteriographic staging accuracy in than when the tumor has ,,,,,n,,J-r<>tc>rl the wall of the bladder. The precise preoperative arteriographic staging tumors subsequently proved to penetrate through the wall (C and D) was 80.0 and 85,6 per cent, rP<:nP,·tn'P 2), These percentages compared to the concurrent nr,P"'""''"t' clinical staging in these lesions of 30,0 and 22.7 per cent. In clinical staging biopsies may not reveal tion of the bladder musculature layer a reasons. rn"u"1µe1cceu metastatic lesions also can be out of reach for
6*
5/22 (22,7) 18/52 (34,6)
22
52
patients the stage was changed from Dl to D2 by lymphography,
we have demonstrated that the selective injections of the hypogastric arteries an additional accurate method of bladder u~,,., ..w,,,,u 1, 2 The selective 1111cu,Ja,,,w11 of both arteries after retrograde catheterization of the femoral arteries, 1• 2 The bladder is inflated with air and contrast material is simultaneously through both cathe10 cc per catheter per second, The contrast iothalamate in a total volume of is obtained rapid film ·-uSAHfSLUF> sequencing for all A relatively is used with high examples of bilatcontrast studies are shown in not all patients for other adeconsidered candidates for this 150 ,,~-"·~·'""° do not represent cy,,teictomv cases that were done 1968 to October 1976. cystectornies performed during this interval cent were included in this study, Contraindicathese n-,,annor~ti procedures are medical m nature and determined on an individual care basis suitable consultation with v.o,_uo,w, knowledgeable of medical conditions. LWoLH_,~,u
RESULTS
Data for our most recent series of 52 patients are table 2, because of the referral nature of our advanced bladder cancer early stage lesions are treated-----,-------_, at local cystoscopy and transurethral resection and are referred to us frequently at the time Of recurrence, 'hc,rPtm'I'> OUr need to nr,Prn~p,•ci all of our bladder cancer cases for stimulus for these studies. with multiple prior treatments routine clinical staging more difficult less comparable to the accuracy of the routine clinical procedures used ., ...•. -...J in the case, In our rn.ost recent group 32 of the 52 patients (6L5 per cent) had C or D while 33 of the first 60 patients (55,0 per cent) studied also had C or D tumors, In our latest group of 52 there were 42 men and 10 women, ranging in from 54 to 79 years, All of these patients were studied before radical cys .. Precise as presented in such cases exstaging accuracy in tumors
209
FIG. L Typical example of double hypogastric arteriography. 1, scout film. hypogastric arteries. B, bladder phase, H, hypogastric arteries distended with air. 3, venous phase,
210
WINTERBERGER AND ASSOCIATES
The addition of lymphography improved arteriographic staging in D lesions from 85.6 to 95.5 per cent. The lymphogram was also valuable in detecting extension of tumor beyond the pelvis in unsuspected metastatic foci. In 4 patients staging was extended from Dl to D2 on the basis of the lymphogram. There were 4 falsely positive lymphograms (7. 7 per cent). These falsely positive studies were all owing to reactive hyperplasia and/or lipoid granulomatosis secondary to reaction of the lymph nodes to contrast material. In the 38 cases not included in tables 1 and 2, we have observed occasionally a falsely negative lymphogram wherein microscopic tumor foci were found but not seen on x-ray. Our current falsely negative lymphogram rate is 1.9 per cent. In the most
FIG. 2. Example of bilateral selective hypogastric arteriography study with uterus superimposed on bladder. 1, injection phase. A, hypogastric arteries injected. L, contrast-laden lymph nodes. B, bladder distended with air. U, uterus with contraceptive device. 2, capillary phase. U, uterus with characteristic vascularity T, tumor along base of bladder. 3, venous phase. U, uterus. T, tumor.
palpation and so forth. Therefore, C and D lesions m:...y be understaged by clinical methods in patients as we have described. We also have demonstrated the angiographic study to be even more reliable than pathologic staging at times. This situation can occur since the cystectomy specimen given to the pathologist may not include all tumor or biopsy material. Figure 4 is an additional example of the value of these studies in that a metastatic bone lesion is demonstrated angiographically in a patient clinically staged as having a C tumor. The tumor metastatic foci were not suspected on routine roentgenographic films. This bone metastasis was not detectable on routine radioisotope bone scan owing to the high concentration of radioactivity in the bladder. We have observed similar cases, proving that this is not a rare condition. The lymphogram is a valuable adjunct in the staging process, particularly in detecting unsuspected metastatic foci. Although falsely positive studies owing to reactive inflammatory hyperplasia, lipoid granulomatosis or other benign processes present some difficulties, there is an extremely low incidence of falsely negative studies. These results also improve with time and exposure in reading the studies. A typical positive lymphogram in bladder cancer is demonstrated in figure 5.
FIG. 3. Example of multiple contrast technique. I, scout film. A, catheters selectively in hypogastric arteries. B, bladder distended with air. N, contrast-laden lymph nodes. 2, arterial phase. A, hypogastric arteries selectively opacified. U, normal uterus. N, contrast-laden lymph nodes. Tumor in capillary phase in base of bladder (arrows). 3, venous phase. Tumor in base of bladder extending into urethra (arrows).
BLADDER CANCER ANGIOGRAPHY
211
recent 52 patients (table 2) lymphangiograms were considered positive in 34.6 per cent. Our arteriographic and lymphographic techniques have not resulted in any patient morbidity, mortality, anaphylactic reactions, arterial bleeding, emergency operation, severe hematomas or pulmonary insufficiency. 1-4 None was observed in the present series described herein. DISCUSSION
The accuracy of angiographic staging of bladder cancer generally improves with advancing stage of the neoplasm, particularly after the tumor has penetrated the bladder wall. We have demonstrated the angiographic study to be even more
Fm. 5. Example of lymphography improving arteriographic staging. 1, arterial phase. A, hypogastric arteries injected. B, bladder distended with air. T, tumor along base and left wall of bladder. 2, venous phase. A, hypogastric arteries injected. B, bladder distended with air. Tumor extending only into perivesical fat layer (arrows). 3, lymphogram. Extensive lymph node involvement in pelvic nodes extending presumed arteriographic stage from C to D (arrows).
Fm. 4. Example of arteriographic staging detecting occult bone lesion (L) not seen on bone scan.1, injection phase. A, hypogastric arteries. B, bladder distended with air. T, tumor along left bladder wall. 2, capillary phase. B, bladder distended with air. T, tumor along left bladder wall appears to be stage C. L, lesion in pubic ramis changes stage to D. 3, venous phase. T, tumor along left bladder wall. L, lesion in pubis.
reliable than the pathologic staging at times, depending on a number of factors such as the extent of the operation. In our previous experience, arteriographic staging accuracy was influenced somewhat by previous radiation therapy, prior pelvic operation or the presence of other pelvic tumors, for example an enlarged uterus in the female patient superimposed on the bladder (fig. 2). 1• 2 These factors also presented some difficulty in the present study, particularly in 4 cases of previous radiation therapy with severe post-radiation cystitis. Again, the diagnostic limitation occurs primarily with the lower stage neoplasms. As with our previous experience, aortic injections rather than bilateral selective injections are less accurate in the arteriographic staging of all bladder neoplasms and should not be substituted routinely. Chronic inflammatory disease of the bladder, tumors in the dome of the bladder, concurrent large prostatic tumors and the presence of multiple tumors
212
WINTERBERGER AND ASSOCIATES
may also limit the arteriographic staging accuracy in the lower stage neoplasms. Cellular tumor grading is based on the degree of anaplasia and undifferentiated tumors probably do not have as favorable a prognosis as differentiated lesions.' However, the depth of infiltration (stage) may affect the prognosis more than any other feature of bladder cancer. The angiographic staging has been refined to a high level ofaccuracy and reliability. We also have studied many patients with ultrasonic scanning but did not further improve preoperative staging accuracy. 2 We did find ultrasound a valuable adjunct in cases in which the arteriogram was not indicated clinically. 2 With the recent availability of computerized tomographic total body scanners, we anticipate adding this valuable modality to our preoperative evaluation of bladder cancer patients. Adequate or comparable data as we have described are not, however, yet available using this technique.
REFERENCES
1. Winterberger, A. R., Kenny, G. M., Choi, S. H. and Murphy, G.
2.
3. 4.
5.
P.: Correlation of selective arteriography in the staging of bladder tumors. Cancer, 29: 332, 1972. Winterberger, A. R. and Murphy, G. P.: Correlation of B-scan ultrasonic laminography with bilateral selective hypogastric arteriography and lymphography in bladder tumors. Vase. Surg., 8: 169, 1974. Wajsman, Z., Baumgartner, G., Murphy, G. P. and Merrin, C.: Evaluation of lymphangiography for clinical staging of bladder tumors. J. Urol., 114: 712, 1975. Merrin, C., Wajsman, Z., Baumgartner, G. and Jennings, E.: The clinical value of lymphangiography: are the nodes surrounding the obturator nerve visualized? J. Urol., 117: 762, 1977. Marshall, V. F.: The relation of the preoperative estimate to the pathologic demonstration of the extent ofvesical neoplasms. J. Urol., 68: 714, 1952.
0022-534 7!78/1192-0208$02. 00/0 THE JOURNAL OF UROLOGY Copyright © 1978 by The Williams & Wilkins Co.
Vol. 119, February Printed in U.S.A.
EIGHT YEARS OF EXPERIENCE WITH PREOPERATIVE ANGIOGRAPHIC AND LYMPHOGRAPHIC STAGING OF BLADDER CANCER A. R. WINTERBERGER, Z. WAJSMAN, C. MERRIN
AND
G. P. MURPHY*
From the Departments of Diagnostic Radiology and Urologic Oncology, Roswell Park Memorial Institute, New York State Department of Health, Buffalo, New York
ABSTRACT
Our experience with the preoperative staging of bladder cancer by bilateral selective hypogastric arteriography has been accumulated since 1968. More than 150 patients have been studied by selective angiography before radical cystectomy. Our latest series of 52 patients (1972 to 1976) compares to our previous experience demonstrating angiographic staging accuracy to detect bladder invasion and occult metastases at a rate exceeding that of clinical staging alone. Arteriographic staging ofD lesions, when supplemented with lymphography, approaches 100 per cent accuracy. Falsely negative lymphograms currently are extremely uncommon (1.9 per cent). In several illustrated instances angiographic staging was proved to be even more accurate than the pathologic staging of a limited cystectoiny specimen. The over-all angiographic and lymphangiographic staging accuracy in our most recent series of cystectomy patients was 78.8 per cent. The techniques and reliability of the data are discussed in detail, including the factors that interfere with the exact arteriographic staging of bladder cancer. However, these factors are more troublesome in early stage lesions. These studies demonstrate the role and value as well as areas of limitation of preoperative arteriography and lymphography in the evaluation of invasive bladder cancer. Our data with 60 of the first 100 sequential bladder cancer angiographic staging. The specific details for 112 of the 150 patients undergoing selective arteriography with subsequent patients are shown in tables 1 and 2. The information is positive pathologic correlation were published initially in 1972 available in the remaining 38 patients but will not be used to (table 1). 1 Our experience at that time suggested that the confirm the accuracy of tumor staging since followup is less addition of arteriographic staging in bladder cancer, particu- than 1 year. larly in suspected clinically invasive lesions, improved overTECHNIQUES all accuracy in preoperative staging. After additional clinical studies in 1973 and 19742 we concluded that the addition of The Marshall modification of the Jewett and Strong classifipedal lymphography to the radiographic staging further in- cation of bladder cancer was used generally by those particicreased preoperative staging accuracy. The clinical value of pating in this study. 5 Clinical stages of bladder cancer in this lymphography in the staging of bladder cancer also has been system are as follows: stage O neoplasms are in situ lesions described in 1975 by Wajsman3 and in 1976 by Merrin 4 and limited to the mucosa, stage A tumors have infiltrated the their associates. Herein we present the data on the total submucosa but not the muscularis, stage Bl tumors have series of patients from our institution to substantiate the infiltrated the superficial musculature, stage B2 lesions inusefulness of combined preoperative angiography and lym- volve the deep musculature, stage C tumors extend to the phangiography in staging preoperative bladder tumors. perivesical and fat, stage Dl lesions have invaded the pelvic wall structures, pelvic lymph nodes or adjacent organs and METHODS stage D2 patients have metastatic involvement beyond the The study includes 150 patients with a proved diagnosis of pelvis. 5 bladder cancer who were candidates for radical cystectomy TABLE 1. Preoperative staging of bladder tumors (60 cases with between 1968 and October 1976. All patients underwent roudefinitive pathological diagnosis): 1968 to 1972 tine clinical staging, including excretory urography (IVP), bimanual examination under general anesthesia and cystosStage Totals copy with transurethral biopsy. After this clinical staging No.(%) A B C D bilateral selective hypogastric arteriography and bilateral peNo. (%) No. (%) No. (%) No. (%) dal lymphography were performed. After the independent in- Staged exactly arterio- 2/7 (29) 10/19 (53) 8/9 (89) 22/24 (92) 42/60* (72) terpretation of the angiographic studies the majority of the graphically patients were subjected to radical cystectomy and pelvic lymphadenectomy. The gross pathological features, as well as Staged exactly clinically 4/7 (57) 10/19 (53) 1/9 (11) 17/24 (71) 32/60* (53) the histology with numerous slides of bladder step sections and Pathology proved by cys46*/60* (77) 10 7 19 9 lymph node preparations from the surgical specimens, were tectomy later correlated with the preoperative clinical staging and the Pathology proved by surgical exploration
Accepted for publication April 7, 1977. Supported in part by United States Public Health Service Grant RR-05648-10 of the National Institutes of Health. * Requests for reprints: Roswell Park Memorial Institute, 666 Elm St., Buffalo, New York 14263.
Total pts.
7
19
9
None of these patients had lymphography. * One patient had no tumor at the time of operation.
208
14
14/60* (23)
24
60*
BLADDER CANCER ANGIOGRAPHY TABLE
2, Preoperative staging of bladder tumors (52 cases with
pathologic correlation from cystectomy specimens): 1972 to 1976 Stage A
B
C
D
No, (%)
No, (%)
No, (%)
No, (%)
Totals No,(%)
Staged exactly 7/9 (77.8) 5/11 (45,5) 8/10 (80.0) 19/22 (85.6) 39/52 (75.5) arteriographica!ly 7/9 (77,8)
Stage changed by lymphography
0
6/9 (66,7)
Tota! pts,
* In
9
5/11 (45,5) 8/10 (80,0) 21/22 (95,5) 41/52 (78,8)
0
0
4/11 (36.4) 3/10 (30,0)
11
10
that have infiltrated the bladder walL In such tumors when lymphography is added to the approach 100 per cent. As with our arteriographic staging accuracy in than when the tumor has ,,,,,n,,J-r<>tc>rl the wall of the bladder. The precise preoperative arteriographic staging tumors subsequently proved to penetrate through the wall (C and D) was 80.0 and 85,6 per cent, rP<:nP,·tn'P 2), These percentages compared to the concurrent nr,P"'""''"t' clinical staging in these lesions of 30,0 and 22.7 per cent. In clinical staging biopsies may not reveal tion of the bladder musculature layer a reasons. rn"u"1µe1cceu metastatic lesions also can be out of reach for
6*
5/22 (22,7) 18/52 (34,6)
22
52
patients the stage was changed from Dl to D2 by lymphography,
we have demonstrated that the selective injections of the hypogastric arteries an additional accurate method of bladder u~,,., ..w,,,,u 1, 2 The selective 1111cu,Ja,,,w11 of both arteries after retrograde catheterization of the femoral arteries, 1• 2 The bladder is inflated with air and contrast material is simultaneously through both cathe10 cc per catheter per second, The contrast iothalamate in a total volume of is obtained rapid film ·-uSAHfSLUF> sequencing for all A relatively is used with high examples of bilatcontrast studies are shown in not all patients for other adeconsidered candidates for this 150 ,,~-"·~·'""° do not represent cy,,teictomv cases that were done 1968 to October 1976. cystectornies performed during this interval cent were included in this study, Contraindicathese n-,,annor~ti procedures are medical m nature and determined on an individual care basis suitable consultation with v.o,_uo,w, knowledgeable of medical conditions. LWoLH_,~,u
RESULTS
Data for our most recent series of 52 patients are table 2, because of the referral nature of our advanced bladder cancer early stage lesions are treated-----,-------_, at local cystoscopy and transurethral resection and are referred to us frequently at the time Of recurrence, 'hc,rPtm'I'> OUr need to nr,Prn~p,•ci all of our bladder cancer cases for stimulus for these studies. with multiple prior treatments routine clinical staging more difficult less comparable to the accuracy of the routine clinical procedures used ., ...•. -...J in the case, In our rn.ost recent group 32 of the 52 patients (6L5 per cent) had C or D while 33 of the first 60 patients (55,0 per cent) studied also had C or D tumors, In our latest group of 52 there were 42 men and 10 women, ranging in from 54 to 79 years, All of these patients were studied before radical cys .. Precise as presented in such cases exstaging accuracy in tumors
209
FIG. L Typical example of double hypogastric arteriography. 1, scout film. hypogastric arteries. B, bladder phase, H, hypogastric arteries distended with air. 3, venous phase,
210
WINTERBERGER AND ASSOCIATES
The addition of lymphography improved arteriographic staging in D lesions from 85.6 to 95.5 per cent. The lymphogram was also valuable in detecting extension of tumor beyond the pelvis in unsuspected metastatic foci. In 4 patients staging was extended from Dl to D2 on the basis of the lymphogram. There were 4 falsely positive lymphograms (7. 7 per cent). These falsely positive studies were all owing to reactive hyperplasia and/or lipoid granulomatosis secondary to reaction of the lymph nodes to contrast material. In the 38 cases not included in tables 1 and 2, we have observed occasionally a falsely negative lymphogram wherein microscopic tumor foci were found but not seen on x-ray. Our current falsely negative lymphogram rate is 1.9 per cent. In the most
FIG. 2. Example of bilateral selective hypogastric arteriography study with uterus superimposed on bladder. 1, injection phase. A, hypogastric arteries injected. L, contrast-laden lymph nodes. B, bladder distended with air. U, uterus with contraceptive device. 2, capillary phase. U, uterus with characteristic vascularity T, tumor along base of bladder. 3, venous phase. U, uterus. T, tumor.
palpation and so forth. Therefore, C and D lesions m:...y be understaged by clinical methods in patients as we have described. We also have demonstrated the angiographic study to be even more reliable than pathologic staging at times. This situation can occur since the cystectomy specimen given to the pathologist may not include all tumor or biopsy material. Figure 4 is an additional example of the value of these studies in that a metastatic bone lesion is demonstrated angiographically in a patient clinically staged as having a C tumor. The tumor metastatic foci were not suspected on routine roentgenographic films. This bone metastasis was not detectable on routine radioisotope bone scan owing to the high concentration of radioactivity in the bladder. We have observed similar cases, proving that this is not a rare condition. The lymphogram is a valuable adjunct in the staging process, particularly in detecting unsuspected metastatic foci. Although falsely positive studies owing to reactive inflammatory hyperplasia, lipoid granulomatosis or other benign processes present some difficulties, there is an extremely low incidence of falsely negative studies. These results also improve with time and exposure in reading the studies. A typical positive lymphogram in bladder cancer is demonstrated in figure 5.
FIG. 3. Example of multiple contrast technique. I, scout film. A, catheters selectively in hypogastric arteries. B, bladder distended with air. N, contrast-laden lymph nodes. 2, arterial phase. A, hypogastric arteries selectively opacified. U, normal uterus. N, contrast-laden lymph nodes. Tumor in capillary phase in base of bladder (arrows). 3, venous phase. Tumor in base of bladder extending into urethra (arrows).
BLADDER CANCER ANGIOGRAPHY
211
recent 52 patients (table 2) lymphangiograms were considered positive in 34.6 per cent. Our arteriographic and lymphographic techniques have not resulted in any patient morbidity, mortality, anaphylactic reactions, arterial bleeding, emergency operation, severe hematomas or pulmonary insufficiency. 1-4 None was observed in the present series described herein. DISCUSSION
The accuracy of angiographic staging of bladder cancer generally improves with advancing stage of the neoplasm, particularly after the tumor has penetrated the bladder wall. We have demonstrated the angiographic study to be even more
Fm. 5. Example of lymphography improving arteriographic staging. 1, arterial phase. A, hypogastric arteries injected. B, bladder distended with air. T, tumor along base and left wall of bladder. 2, venous phase. A, hypogastric arteries injected. B, bladder distended with air. Tumor extending only into perivesical fat layer (arrows). 3, lymphogram. Extensive lymph node involvement in pelvic nodes extending presumed arteriographic stage from C to D (arrows).
Fm. 4. Example of arteriographic staging detecting occult bone lesion (L) not seen on bone scan.1, injection phase. A, hypogastric arteries. B, bladder distended with air. T, tumor along left bladder wall. 2, capillary phase. B, bladder distended with air. T, tumor along left bladder wall appears to be stage C. L, lesion in pubic ramis changes stage to D. 3, venous phase. T, tumor along left bladder wall. L, lesion in pubis.
reliable than the pathologic staging at times, depending on a number of factors such as the extent of the operation. In our previous experience, arteriographic staging accuracy was influenced somewhat by previous radiation therapy, prior pelvic operation or the presence of other pelvic tumors, for example an enlarged uterus in the female patient superimposed on the bladder (fig. 2). 1• 2 These factors also presented some difficulty in the present study, particularly in 4 cases of previous radiation therapy with severe post-radiation cystitis. Again, the diagnostic limitation occurs primarily with the lower stage neoplasms. As with our previous experience, aortic injections rather than bilateral selective injections are less accurate in the arteriographic staging of all bladder neoplasms and should not be substituted routinely. Chronic inflammatory disease of the bladder, tumors in the dome of the bladder, concurrent large prostatic tumors and the presence of multiple tumors
212
WINTERBERGER AND ASSOCIATES
may also limit the arteriographic staging accuracy in the lower stage neoplasms. Cellular tumor grading is based on the degree of anaplasia and undifferentiated tumors probably do not have as favorable a prognosis as differentiated lesions.' However, the depth of infiltration (stage) may affect the prognosis more than any other feature of bladder cancer. The angiographic staging has been refined to a high level ofaccuracy and reliability. We also have studied many patients with ultrasonic scanning but did not further improve preoperative staging accuracy. 2 We did find ultrasound a valuable adjunct in cases in which the arteriogram was not indicated clinically. 2 With the recent availability of computerized tomographic total body scanners, we anticipate adding this valuable modality to our preoperative evaluation of bladder cancer patients. Adequate or comparable data as we have described are not, however, yet available using this technique.
REFERENCES
1. Winterberger, A. R., Kenny, G. M., Choi, S. H. and Murphy, G.
2.
3. 4.
5.
P.: Correlation of selective arteriography in the staging of bladder tumors. Cancer, 29: 332, 1972. Winterberger, A. R. and Murphy, G. P.: Correlation of B-scan ultrasonic laminography with bilateral selective hypogastric arteriography and lymphography in bladder tumors. Vase. Surg., 8: 169, 1974. Wajsman, Z., Baumgartner, G., Murphy, G. P. and Merrin, C.: Evaluation of lymphangiography for clinical staging of bladder tumors. J. Urol., 114: 712, 1975. Merrin, C., Wajsman, Z., Baumgartner, G. and Jennings, E.: The clinical value of lymphangiography: are the nodes surrounding the obturator nerve visualized? J. Urol., 117: 762, 1977. Marshall, V. F.: The relation of the preoperative estimate to the pathologic demonstration of the extent ofvesical neoplasms. J. Urol., 68: 714, 1952.