COMPUTERIZED TOMOGRAPHIC URETEROSCOPY FOR DIAGNOSING URETERAL TUMORS

COMPUTERIZED TOMOGRAPHIC URETEROSCOPY FOR DIAGNOSING URETERAL TUMORS

0022-5347/00/1631-0042/0 THE JOURNAL OF UROLOGY® Copyright © 2000 by AMERICAN UROLOGICAL ASSOCIATION, INC.® Vol. 163, 42– 46, January 2000 Printed in...

474KB Sizes 0 Downloads 69 Views

0022-5347/00/1631-0042/0 THE JOURNAL OF UROLOGY® Copyright © 2000 by AMERICAN UROLOGICAL ASSOCIATION, INC.®

Vol. 163, 42– 46, January 2000 Printed in U.S.A.

COMPUTERIZED TOMOGRAPHIC URETEROSCOPY FOR DIAGNOSING URETERAL TUMORS SHIGEO TAKEBAYASHI,* MASAHIKO HOSAKA, YOSHINOBU KUBOTA, KAZUMI NOGUCHI, MOMOKUNI FUKUDA, YOSHIO ISHIBASHI, TAKESHI TOMODA AND SHO MATSUBARA From the Departments of Radiology and Urology, Yokohama City University Hospital, Department of Urology, Yokosuka Kyosai Hospital and Department of Urology, Ohguch Hospital, Yokohama, Japan

ABSTRACT

Purpose: Helical computerized tomography (CT) image acquisition has led to the availability of improved data sets for CT endoscopic imaging that represent virtual endoscopy using CT. We assessed the usefulness of CT ureteroscopic imaging for diagnosing ureteral tumors. Materials and Methods: A total of 16 patients with ureteral stenosis underwent surface rendering CT ureteroscopy after the intravenous administration of contrast material and furosemide. To distinguish ureteral tumors from ureteral strictures 2 observers blinded to other patient history and evaluation data independently and prospectively evaluated CT ureteroscopy with reformatted CT ureterograms in these 16 patients. CT ureteroscopic images were then correlated with surgical and pathological findings, which served as the gold standard. Results: Surgical and pathological findings in the 16 patients revealed 16 ureteral tumors, including carcinoma in 10 (carcinoma in situ in 1, fibroepithelial polyps in 2 and hyperplastic polypoids in 4), inflammatory intrinsic stricture in 2 and extrinsic stricture in 4 caused by retroperitoneal fibrosis in 2 and lymph node metastasis in 2. CT ureteroscopy correctly detected all lesions except 1 carcinoma in situ, 1 polypoid carcinoma and 1 hyperplastic polypoid. The sensitivity and specificity of CT ureteroscopy for detecting ureteral tumors and carcinoma were 81% and 100%, and 80% and 75%, respectively, when tumors without stalks were considered carcinoma. Conclusions: CT ureteroscopy is useful for visualizing the complex morphology of ureteral tumors and distinguishing tumor from ureteral stricture. KEY WORDS: ureter; x-ray computerized tomography; ureteroscopy; neoplasms; diagnosis, differential

virtual ureteroscopy for diagnosing ureteral tumors and stricture.4

Common diagnostic indications for ureteroscopic examination include upper urinary tract obstruction, unilateral hematuria, positive urinary cytology and lateralized cytology.1 Ureteroscopic procedures have also been performed in the treatment of ureteral stenosis, benign tumors and low grade transitional cell carcinoma.2 Ureteroscopic complications include ureteral and occasionally vesical perforation, urinoma, hematoma, ureteral obstruction and fistula.1 Furthermore, a number of factors may make ureteral access difficult, such as an enlarged prostate or urethral stricture.3 However, contemporary endoscopic techniques using small diameter and flexible ureteroscopes allow a less invasive outpatient procedure with easier ureteral access, and enables visualization and histological characterization of upper tract urothelial malignancies with a high degree of accuracy.4 Helical computerized tomography (CT) with its continuous acquisition of volumetric data enables the presentation of acquired data in 3-dimensional (D) images. CT urography is composed of reformatted images that display the collecting system in a planar format similar to that of excretory urography (IVP).5 CT or virtual endoscopy using surface rendering techniques enables imaging of the interior of a target organ by extracting CT data from only the boundary regions between the organ walls and contrast material.6 This processing technique is used for analyzing the ureteral interior by extracting CT attenuations for contrast material in a reformatted CT urogram. We evaluate the usefulness of CT

MATERIALS AND METHODS

Between October 1996 and April 1999 helical CT for CT ureteroscopy was performed in 10 men and 6 women 44 to 74 years old (mean age 55). All patients underwent urinary cytology, IVP and retrograde ureterography before helical CT for CT ureteroscopy. Urinary cytology findings were class V in 3 patients, class IV in 2, class III in 3, class II in 4 and class I in 4. In the 16 patients serum creatinine ranged from 0.8 to 1.5 mg./dl. (mean 1.1, normal 0.6 to 1.2). IVP revealed bilateral hydronephrosis in 3 patients and unilateral hydronephrosis in 13. In those with severe hydronephrosis delayed 20 to 30-minute images showed opacification of the ureters. Stenotic lesions were observed in the proximal ureter in 4 patients, mid ureter in 6 and distal ureter in 6. Retrograde ureterography demonstrated smooth filling defects in 2 patients, irregular filling defects in 5, equivocally irregular filling defects in 4, a filling defect and stricture in 1, and strictures only in 4. Urine cytology obtained by retrograde catheterization into the ureters with stenosis was class V in 3 patients, class IV in 3, class III in 2, class II in 5 and class I in 3. None of the patients underwent fiberoptic ureteroscopy preoperatively. All patients underwent surgical exploration and 9 in whom ureteral carcinoma was suspected on imaging underwent complete nephroureterectomy with a bladder cuff. One patient in whom ureteral carcinoma was suspected due to positive cytology only underwent distal ureterectomy with a tumor-free proximal margin and a bladder cuff. Two patients

Accepted for publication August 20, 1999. * Requests for reprints: Department of Radiology, Yokohama City University Hospital, 3– 46, Urafune-cho, Minami-ku, Yokohama, 232, Japan. 42

43

COMPUTERIZED TOMOGRAPHIC URETEROSCOPY FOR DIAGNOSING URETERAL TUMORS

freely to display the interiors of the organs of interest. Images were recorded as paper copies and as dynamic images of 12 fly-through movies. All intraluminal navigation was performed by physicians familiar with operating the navigator system. Average evaluation time was 8 minutes per series of ureteroscopic images. A complete CT ureteroscopic examination, including the acquisition and interpretation of images, required approximately 40 minutes. A urologist and a radiologist blinded to patient history and examination data independently and prospectively evaluated the CT ureteroscopy studies with reformatted CT ureterograms. These physicians are experts on urological endoscopy and CT, respectively. When disagreements occurred, the observers discussed the findings and reached a consensus. CT ureteroscopic images were evaluated to localize ureteral lesions and detect ureteral tumors, and intrinsic and extrinsic strictures. Intrinsic stricture on the CT ureteroscopy was defined as concentric stenosis without masses. Extrinsic stricture was defined as ureteral stenosis by extrinsic compression. Surface morphology of the tumors was classified as sessile, sessile with ulceration, polypoid and polyp. Sensitivity and specificity of CT ureteroscopy for revealing ureteral tumors were determined by comparing surgical and pathological findings, which were used as the gold standard. Sensitivity and specificity to reveal carcinoma were also obtained when tumors showed no stalks and, therefore, were considered carcinoma. Ureteral carcinoma stage was assigned based on CT ureterogram findings as stage T2 or less—a ureteral tumor with complete preservation of the surrounding fat plane, stage T3—a tumor extending into the periureteral fat and stage T4 —a tumor extending into an adjacent structure. Furthermore, ureteral carcinoma stages on the CT ureterogram were compared with those on pathological evaluation.

in whom benign polyps were suspected on imaging and urinary cytology underwent partial ureterectomy with end-toend anastomosis. Ureteral carcinoma was pathologically classified using a modified TNM staging system of renal pelvic carcinoma7 as stage Tis— carcinoma in situ, stage Ta—papillary noninvasive carcinoma, stage T1—submucosal invasion only, stage T2—invasion to lamina propria, stage T3—periureteral fat invasion and stage T4 —tumor extension to an adjacent structure. All patients underwent helical CT with a HiSpeed Advantage RP* scanner. Opacification of the ureters was done by the same method that we used previously in renal collecting systems for CT nephroscopy.8 Before intravenous administration of 100 ml. of the nonionic low osmolarity contrast agent iohexol 300, 6 to 7 mg. furosemide at 0.1 mg./kg. were given intravenously for diuresis to distend the ureters, enabling homogeneous, dense opacification by the contrast material. In patients with normal functioning kidneys helical enhanced images were obtained 8 to 10 minutes after starting the administration of contrast material. In those with deteriorated renal function images were obtained after 20 to 30 minutes of scan delay. Patients did not breathe for 30 to 40 seconds per scan. Scan parameters were 200 to 250 mA. at 120 kVp. The helical reconstruction algorithm of the scanner was developed based on the 180-degree linear interpolation algorithm. Images were acquired using 3 mm. collimation and a table speed of 3 mm. per second with a helical pitch of 1. Overlapping helical images were reconstructed retrospectively every 1 mm. In the 7 most recent patients the target ureters were also imaged using a 1.5 mm. per second table speed and 1 mm. collimation with a helical pitch of 1.5 to decrease partial volume averaging. Reconstruction time for the 3-D images was approximately 7 minutes in each series of CT ureteroscopy, which was generated from 80 to 90, 2-D axial source images. Using a CT endoscopic navigator system at a computer workstation a threshold surface rendering technique was used with the 3-D endoscopic mode that classifies voxels with 2 numbers specifying the upper and lower limits (thresholds) in a binary method to display the surface of specified anatomical parts as 3-D images. We used 100 to 150 HU of the lower threshold and the upper threshold maximum to generate CT ureteroscopic images. The navigator system automatically provided standard sagittal, coronal and, as necessary, oblique reference images of CT ureterograms that indicated observer position. CT urograms were derived and reformatted from the original CT data set at 3 mm. intervals. Using the trackball real-time angles and cut planes were chosen

RESULTS

Surgical and pathological findings. Gross examination of the 12 resected ureters revealed 15 tumors (0.4 to 2.0 cm., mean 1.1) and 2 intrinsic strictures. Microscopic examination of the 15 tumors showed that 9 were transitional cell carcinomas, 2 were fibroepithelial polyps and 4 were hyperplastic polypoids. Of the 9 carcinomas 4 were sessile, 4 were sessile with ulceration and 1 was polypoid. Of the 4 sessile carcinomas 2 were stage T1 and 2 were stage T2, and of the 4 sessile carcinomas with ulcerations 3 were stage T2 and 1 was stage T3. The polypoid carcinoma was stage Ta. Microscopic examination of the ureter with an intrinsic stricture also revealed 1 transitional cell carcinoma in situ. A fibroepithelial polyp was associated with 3 hyperplastic polypoids and an inflammatory intrinsic stricture, while another polyp was associated with 1 hyperplastic polypoid.

* General Electric Medical Systems, Milwaukee, Wisconsin.

CT ureteroscopic and pathological findings in 22 lesions in 16 patients with ureteral stenosis CT Findings Pathological Findings

Transitional cell Ca: Sessile Ulcerated sessile Polypoid In situ Fibroepithelial polyp Hyperplastic polypoid Inflammatory intrinsic stricture Extrinsic stricture: Retroperitoneal fibrosis Lymph node metastasis Totals

Tumor

No. Pts.

Stricture

Sessile

Ulcerated Sessile

Polypoid

Polyp

Intrinsic

Extrinsic

Not Detected

4 4 1 1 2 4 2

4 0 0 0 0 0 0

0 4 0 0 0 0 0

0 0 0 0 0 3 0

0 0 0 0 2 0 0

0 0 0 0 0 0 2

0 0 0 0 0 0 0

0 0 1 1 0 1 0

2 2 22

0 0 4

0 0 4

0 0 3

0 0 2

2 0 4

0 2 2

0 0 3

44

COMPUTERIZED TOMOGRAPHIC URETEROSCOPY FOR DIAGNOSING URETERAL TUMORS

The remaining 4 patients, in whom ureterectomy was not performed, underwent biopsy of the periureteral lesions. Surgical and pathological findings revealed ureteral strictures caused by lymphadenopathy which compressed the ureter in 2 patients, and idiopathic retroperitoneal fibrosis in 2. Of the 22 ureteral lesions identified by surgical and pathological findings 16 were ureteral tumors, 2 were inflammatory intrinsic strictures and 4 were extrinsic ureteral strictures (see table). CT ureteroscopic findings. The high attenuation values of 214 to 250 HU in the ureters enabled us to obtain clear ureteroscopic images. CT ureteroscopy may reveal fine details of the anatomy of the interior of normal ureter as well as ureteropelvic and ureterovesical junctions (fig. 1). CT ureteroscopy correctly detected 13 tumors, including 4 sessile lesions, 4 sessile lesions with ulcerations (fig. 2, A), 2 polyps and 3 polypoids (fig. 3, A). The appearance of tumors on CT ureteroscopy generally coincided with that of the gross specimens. However, CT ureteroscopy missed 1 carcinoma in situ, 1 polypoid shaped carcinoma and 1 hyperplastic polypoid. There was no false-positive tumor on CT ureteroscopy. Sensitivity and specificity of CT ureteroscopy for detecting ureteral tumors were 81% (13 of 16 tumors) and 100%, respectively. Sensitivity and specificity for detecting carcinoma were 80% (8 of 10 carcinomas) and 75% (9 of 12 benign lesions), respectively, when tumors without stalks were considered carcinoma. CT ureteroscopy enabled observation of extrinsic compression by lymphadenopathy and the differentiation of lesions from urothelial tumors. The imaging technique showed an intrinsic stricture as a dark ring or pinhole on the image (fig. 3, B). However, this modality failed to distinguish idiopathic retroperitoneal fibrosis encasing the ureter from inflammatory intrinsic stricture. CT ureterography findings. CT ureterography clearly revealed stenosis. In all patients this study enabled proximal and distal evaluation of the ureter to the stenotic lesion (figs. 2, B and 3, C). The ureterogram did not show 1 stage Tis and 1 stage Ta carcinomas, which were also not shown by CT ureteroscopy. The observers correctly interpreted the CT ureterogram as demonstrating no evidence of invasion to periureteral fat in 6 of 7 stage T1 or T2 carcinomas and as periureteral fat invasion in stage T3 carcinoma (fig. 2, B). However, the CT urogram revealed periureteral invasion or over staging in the remaining stage T2 carcinomas. Overall

FIG. 1. Ureteroscopy of hydroureter shows normal ureter interior that appears smaller as distance from observers increases. Inset of reference axial CT indicates viewpoint of CT ureteroscopy.

staging accuracy of CT ureterography was 70% (7 of 10 tumors). DISCUSSION

Primary ureteral tumors are relatively rare, accounting for 1/1,000 to 1/3,600 of the general population.9 Most primary neoplasms involve the distal third of the ureter and are discovered by IVP or retrograde urography.10 Distinguishing ureteral tumors is an important and often critical part of the assessment of ureteral lesions because ideal surgical candidates for ureterectomy or endourological resection are identified. A ureteral tumor demonstrates irregular narrowing of the ureteral lumen with sharp margins when the lesion infiltrates the submucosa. However, distingishing these features cannot invariably be accomplished using fluoroscopically monitored retrograde ureterography. Furthermore, in some cases the lesion may resemble a benign ureteral stricture.10 CT endoscopy using a surface rendering technique enables imaging of the interiors of organs by extracting CT attenuation only from the boundary regions between the organ walls and contrast material.5 A great difference in CT attenuation between the interior and lumen of an organ is necessary to generate clear CT endoscopic images. Diuresis caused by intravenously administered low osmolarity contrast material and furosemide distends the ureter, resulting in dense, homogeneous opacification. Surface rendering techniques specify a preprocessing stage that identifies isodensity surfaces within the volume. With surface rendering contiguous CT images are manipulated using a marching cubes algorithm to create a wire frame model.11 Volume rendering has inherent advantages over the surface rendering technique because it uses information from all voxels within the volume.12 However, a volume rendering technique requires considerably longer time for post-processing than a surface rendering technique.12, 13 Thinner collimation, such as 1 mm. collimation, improves the spatial resolution of surface rendering CT endoscopy and decreases partial volume averaging. Computer workstations for CT endoscopic navigator system cost from $8,000 to $200,000.14 However, the equipment for surface rendering CT endoscopic navigator system that we used in our study is less expensive than that for volume rendering. Another aspect that must be addressed is the physician time needed to make the diagnosis and achieve a finished product.14 A physician who is unfamilar with the navigator system requires approximately 20 to 30 minutes of navigator system operating time. However, as the operating system is learned, this time decreases to less than 10 minutes. Of the approximately 40 minutes overall needed for the complete CT ureteroscopic evaluation in our study approximately 20 represented physician time, including intravenous administration of contrast material, monitoring CT images and analyzing the results. A major advantage of CT ureteroscopy over conventional ureteroscopy is that the former is a noninvasive technique that provides visualization of the proximal and distal portions of the stenotic ureter. CT ureteroscopy enables the distinction of ureteral tumors from intrinsic or extrinsic ureteral stricture. However, this imaging technique does not distinguish extrinsic stricture caused by retroperitoneal fibrosis from inflammatory intrinsic stricture. Ureteral carcinomas vary in gross appearance. Approximately 40% of ureteral tumors are nonpapillary, of which most are solid ulcerated tumors.9 Identification of a stalk is most helpful for distinguishing fibroepithelial or hyperplastic polyps from ureteral carcinoma, although Davides and King reported a case of well differentiated transitional cell carcinoma at the tip of a fibroepithelial polyp.15 CT ureteroscopy enables the characterization of polyps as well-defined, rounded or oval, smooth intraluminal projections with stalks. However, some-

COMPUTERIZED TOMOGRAPHIC URETEROSCOPY FOR DIAGNOSING URETERAL TUMORS

45

FIG. 2. A, CT ureteroscopy in 53-year-old man reveals sessile tumor (straight arrow) with ulceration (curved arrow). B, reformatted oblique CT ureterogram shows filling defect (thin arrows) with preserved periureteral fat and stage T2 or less disease. Short arrow indicates viewpoint of CT ureteroscopy in A. C, gross specimen of resected ureter demonstrates 1.5 cm. sessile tumor (straight arrow) with ulceration (curved arrow), corresponding to A. Microscopic examination of tumor revealed grade 2 stage pT2 transitional cell carcinoma.

FIG. 3. A, CT ureteroscopy of distal ureter in 63-year-old man reveals smooth surfaced polyp (arrows) and 3 adjacent polypoid lesions (arrowheads). B, CT ureteroscopy of mid ureter shows concentric circles with pinpoint (arrow) representing intrinsic stenosis. C, reformatted oblique CT ureterography demonstrates stenosis (arrowhead) and smooth filling defect (arrow) in distal ureter. D, gross specimen of resected ureter reveals inflammatory intrinsic stricture (long arrow), fibroepithelial polyp moved out of lumen (short arrow) and hyperplastic polypoids (arrowheads).

times CT endoscopy may fail to distinguish a stalk occupying the ureteral lumen. Characterization of the surface morphology of a urothelial tumor is important because it has some correlation with tumor grade.7, 16, 17 Low grade transitional cell carcinoma is usually polypoid, pedunculated or exophytic. High grade lesions are mostly flat-solid infiltrating tumors. CT endoscopy may evaluate the surface morphology of ureteral tumors. However, it shows only the inner surface of hollow organs and provides no information on anatomy within or beyond the wall. This limitation prevents evaluation of the transmural extent of tumors and limits the ability to localize the lesion relative to surrounding anatomical structures. How-

ever, the recently developed workstations for CT endoscopic images produce reformatted CT ureterograms that may be used as a map for CT ureteroscopy. CT ureterography enables evaluation of extraluminal extension of ureteral carcinoma. A reformatted CT ureterogram from CT data obtained during abdominal compression results in reliable opacification of the collecting system and ureters.5 As an alternative to abdominal compression, diuresis induced by intravenously administered furosemide also results in a reliable CT ureterogram. As in CT nephroscopy,8 CT ureteroscopy has some inherent disadvantages, including a lack of information about the color and texture of the lesion. The technique is also limited

46

COMPUTERIZED TOMOGRAPHIC URETEROSCOPY FOR DIAGNOSING URETERAL TUMORS

because the physician is unable to biopsy the tumor. Furthermore, CT ureteroscopic images cannot be obtained in patients in whom the excretion of contrast material into the upper urinary tract is impeded by renal dysfunction or high grade tumor obstruction. Nevertheless, CT ureteroscopy allows rapid evaluation of the often complex morphology of hollow organs in a more intuitive fashion. The technique also enables rapid analysis of the large number of slices generated.18 CT ureteroscopy combined with referenced CT ureterograms may greatly assist urologists who may be less familiar with cross-sectional imaging than radiologists. Currently with the increasing use of fiberoptic ureteroscopy ureteral tumors are certain to be recognized preoperatively and properly managed by conservative endoscopic techniques. CT imaging with reformatted ureterography and CT ureteroscopy allows correct staging of ureteral tumors and identifies invasive lesions requiring surgical excisional therapy of ureteral stenosis. The results of CT ureteroscopy provide useful information to urologists and help to limit the need for fiberoptic ureteroscopic examinations, which are invasive and technically limited. CONCLUSIONS

Although our clinical data set is small, noninvasive CT ureteroscopy is apparently useful for visualizing the complex morphology of ureteral tumors. CT ureteroscopy is a practical and acceptable technique that is promising for the detection of ureteral tumors and distinction of lesions from ureteral stricture. REFERENCES

1. Horrow, M. M., Tuncali, K. and Kirby, C. L.: Imaging of ureteroscopic complications. AJR Am J Roentgenol, 168: 633, 1997. 2. Albala, D. M. and Grasso, M., III: Therapeutic upper urinary tract endoscopy. In: Color Atlas of Endourology. Philadelphia: Lippincott-Raven, p. 63, 1999. 3. Harmon, W. J., Sershon, P. D., Blute M. L. et al: Ureteroscopy: current practice and long-term complications. J Urol, 157: 28, 1997.

4. Grasso, M. and Bagley, D.: Small diameter, actively deflectable, flexible ureteropyeloscopy. J Urol, 160: 1648, 1998. 5. McNicholas, M. M., Raptopoulos, V. D., Schwarz, R. K. et al: Excretory phase CT urography for opacification of the urinary collecting system. AJR Am J Roentgenol, 170: 1261, 1998. 6. Kimura, F., Shen, Y., Date, S. et al: Thoracic aortic aneurysm and aortic dissection: new endoscopic mode for threedimensional CT display of aorta. Radiology, 198: 573, 1996. 7. Buckley, J. A., Urban, B. A., Soyer P. et al: Transitional cell carcinoma of the renal pelvis: a retrospective look at CT staging with pathologic correlation. Radiology, 201: 194, 1996. 8. Takebayashi, S., Hosaka, M., Takase, K. et al: CT nephroscopic images of renal pelvic carcinomas. J Urol, 162: 315, 1999. 9. Goldman, S. M. and Gatewood, O. M. B.: Neoplasms of the renal collecting system, pelvis, and ureters. In: Clinical Urography. Edited by H. M. Pollack. Philadelphia: W. B. Saunders Co., chapt. 45, pp. 1292–1352, 1990. 10. Dunnick N,. R., McCallum, R. W. and Sandler, C. M.: The ureter. In: Textbook of Uroradiology. Baltimore: Williams & Wilkins, chapt. 14, pp. 287–318, 1991. 11. Fenlon, H. M. and Ferrucci, J. T.: Virtual colonoscopy: what will the issues be? AJR Am J Roentgenol, 169: 453, 1997. 12. Rubin, G. D., Beaulieu, C. F., Argiro, V. et al: Perspective volume rendering of CT and MR images; applications for endoscopic imaging. Radiology, 199: 321, 1996. 13. Royster, A. P., Fenlon, H. M., Clarke, P. D. et al: CT colonoscopy of colorectal neoplasms: two-dimensional and threedimensional virtual-reality techniques with colonoscopic correlation. AJR Am J Roentgenol, 169: 1237, 1997. 14. Mezwa, D. G.: Colorectal lesions: evaluation with CT colography: invited commentary. RadioGraphics, 17: 1167, 1997. 15. Davides, K. C. and King, L. M.: Fibrous polyps of the ureter. J Urol, 115: 651, 1976. 16. Narumi, Y., Kumatani, T., Sawai, Y. et al: The bladder and bladder tumors: imaging with three-dimensional display of helical CT data. AJR Am J Roentgenol, 167: 1134, 1996. 17. Bennington, J. L. and Beckwith, J. B.: Tumors of the renal pelvis and ureter. In: Atlas of Tumor Pathology: Tumors of the Kidney, Renal Pelvis, and Ureter. Washington, D. C.: Armed Forces Institute of Pathology, p. 243, 1975. 18. Vining, D. J., Zagoria, R. J., Liu K. et al: CT cystoscopy: an innovation in bladder imaging. AJR Am J Roentgenol, 166: 409, 1996.