Ureteropelvic Junction Obstruction: An Innovative Approach Combining Metallic Stenting and Virtual Endoscopy

0022-5347/02/1686-2383/0 THE JOURNAL OF UROLOGY® Copyright © 2002 by AMERICAN UROLOGICAL ASSOCIATION, INC.®

Vol. 168, 2383–2386, December 2002 Printed in U.S.A.

DOI: 10.1097/01.ju.0000032172.60874.6d

URETEROPELVIC JUNCTION OBSTRUCTION: AN INNOVATIVE APPROACH COMBINING METALLIC STENTING AND VIRTUAL ENDOSCOPY GEORGE A. BARBALIAS, EVANGELOS N. LIATSIKOS, GEORGE C. KAGADIS, DIMITRIOS KARNABATIDIS, CHRISTINA KALOGEROPOULOU, GEORGE NIKIFORIDIS AND DIMITRIOS SIABLIS From the Departments of Urology, Radiology and Medical Physics, University of Patras, School of Medicine, Patras, Greece

ABSTRACT

Purpose: We report our experience with auto-expandable metallic stents for treating ureteropelvic junction obstruction. Materials and Methods: We treated 4 patients with a mean age of 45 years who had ureteropelvic junction obstruction with placement of a self-expandable intraureteral metallic stent (Wallstent, Schneider, Zurich, Switzerland). All patients presented with recurrent ureteropelvic junction obstruction after open pyeloplasty. Excretory urography and 3-dimensional reconstruction computerized tomography were performed 1 and 6 months after stent insertion. Virtual endoscopy images were obtained at followup due to the need to define ureteral patency. Results: Mean followup was 16 months (range 9 to 24). Wallstent placement was successful and immediate patency was achieved in all cases. During followup 3 patients required no further intervention and the stented ureteropelvic junction remained patent. In the remaining patient stricture recurred 2 months after initial stent insertion due to the ingrowth of scar tissue through the prosthesis. Additional intervention was deemed necessary after placing a longer 6 cm., completely coaxial overlapping metal stent. Virtual endoscopy and excretory urography findings concurred. Virtual endoscopy allows visualization of the stented ureteropelvic junction lumen cephalad and caudal to the prosthesis. It also enables easy navigation within the stent at different angles of view. Conclusions: The concept of applying metallic stents for ureteropelvic junction obstruction and adjacent adynamic ureteral segments combined with virtual endoscopy is strengthened by the results of this study. KEY WORDS: ureter; stents; ureteral obstruction; endoscopy; imaging, three-dimensional

Ureteropelvic junction obstruction is caused by various obstructive processes that result from multiple etiological factors. The histological hallmark of ureteropelvic junction obstruction is increased collagen deposition in the extracellular matrix.1–3 In the past when narrowing between the renal pelvis and ureter caused back pressure and obstruction, standard operative technique involved resection of a segment of renal pelvis, including the ureteropelvic junction, followed by pyeloplasty. Thus, ureteropelvic junction obstruction was traditionally repaired surgically via open pyeloplasty with a reported success rate of 72% to 98%.1, 2 In 1943 Davis devised an alternative technique, namely intubated ureterotomy, that involved incision into the stenotic segment of the ureter, which remained stented.4 With time the ureter regenerated and bridged the defect. Since the advent of percutaneous renal surgical techniques in the 1980s, endoscopic incisional approaches to ureteropelvic junction obstruction have become well established in the urological armamentarium. Treating ureteropelvic junction obstruction has evolved with time and endopyelotomy has become a well established treatment option with a success rate of 75% to 89%. This technique is based on regeneration of the mucosal and smooth muscle layers, provided that there is tissue continuity and a sufficient blood supply. Approximately 11% to 25% of endopyelotomies result in recurrent stenosis and to date the etiology of these failures remains an enigma.5–9 Metal stents are routinely used in the cardiovascular and Accepted for publication July 5, 2002.

biliary systems, and they also have various applications in the urinary tract.10 –12 In previous series we have reported our encouraging experience with metallic stents for treating malignant ureteral obstruction as well as benign anastomotic strictures that developed after ureteroileal diversion.13–15 A comprehensive search of the literature on metal stents used for ureteropelvic junction obstruction revealed no previous reports. We report our experience with auto-expandable metallic stents for ureteropelvic junction obstruction. Furthermore, virtual endoscopy provides new versatility in verifying stent position as well as potential narrowing of the ureteral lumen.16, 17 In this study we performed virtual endoscopy during followup, applying the described principles. MATERIALS AND METHODS

We treated 4 patients 35 to 58 years old (mean age 45) who had ureteropelvic junction obstruction with placement of a self-expandable intraureteral metallic Wallstent. The applied protocol was approved by the hospital ethics committee. Informed consent was obtained from all patients. The 4 patients presented with recurrent ureteropelvic junction obstruction 10 and 14 months, 2 and 4 years, respectively, after open pyeloplasty. Dismembered AndersonHynes pyeloplasty was performed in all cases. All patients presented with adynamic ureteropelvic junction segments involving not only the previously formed ureteropelvic junction, but also a distal ureteral segment of about an average of

2383

2384

URETEROPELVIC JUNCTION OBSTRUCTION AND METALLIC STENTS

FIG. 1. A, antegrade nephrostogram shows left ureteropelvic junction obstruction. B, nephrostogram reveals stenotic segment crossed by 0.035-inch hydrophilic guide wire and dilated by balloon catheter. After ureteral dilation Wallstent was inserted. C, IVP 1 month after stent insertion demonstrates patent stented segment.

2.5 cm. that caused functional obstruction, as shown on preoperative furosemide washout renal scan and/or cine study of ureteral peristalsis. In addition, the lack of organic obstruction was documented by retrograde catheterization of this area, which was performed easily. Obstruction was on the right side in 2 cases and on the left side in 2. A single patient presented with a coexisting lower caliceal stone. The Wallstent is a self-expandable device of elastic mesh woven from stainless cobalt based alloy filaments and mounted on a 7Fr catheter. Its design and a detailed methodology of stent insertion have been previously described.14 The stents were 4 cm. long, fully expanded to a diameter of 7 mm. and were placed so that the upper end extended cephalad to the obstruction and protruded into the renal pelvis. When the stents did not expand to the desired diameter after insertion, supplementary balloon dilation was performed. The procedure was done with the patient under intravenous sedation (fig. 1). After stent position was finalized the nephrostomy tube remained indwelling for gravity drainage. A double pigtail ureteral stent was not inserted. Ureteral lumen patency was controlled by a nephrostogram performed 24 and 48 hours after the initial procedure, respectively. The percutaneous nephrostomy tube was removed in the absence of obstruction and the patients were discharged home. Perioperative antibiotics were administered in all cases. Followup examination consisted of creatinine measurement, urinalysis, urine culture and transabdominal Doppler ultrasonography 1, 2, 4 and 6 months after stent placement and every 3 months thereafter. Excretory urography (IVP) was also performed at 1 and 6 months of followup. At the same time 3-dimensional (D) computerized tomography (CT) reconstruction was also done (fig. 2). Virtual endoscopy images were obtained as a followup tool due to the need to define ureteral patency. For processing virtual endoscopy 3-D reconstruction images data acquisition was performed with a Somatom Plus 4 Power (Siemens, Erlangen, Germany) during patient deep

FIG. 2. Stented ureteropelvic caliceal unit on 3-D CT reconstruction with contrast medium. A, posterior view. B, anterior view.

inspiration with and without contrast medium. Imaging was done at 140 kV. and 171 mA. with 3 mm. collimation and a 3 mm. per second feed. Reconstruction was performed with a soft reconstruction kernel at 3 mm. The first acquisition was done without contrast medium. The second acquisition consisted initially of the administration of 100 ml. nonionic Visipaque contrast material (Nycomed, Princeton, New Jersey). After 5 minutes data acquisition was concluded. Acquired data were then transferred to a server through an asynchronous transfer mode based local area network. The medical physicist responsible for image processing provided a selection of studies from the server at request. Image processing was performed using software. The technique consisted of several stages. At stage 1 acquired transverse sections from the CT scanner were reconstructed, thus, creating a 3-D volume. Several consequent threshold values were then applied to the data to remove any information not relative to the data under investigation (calices, pelvis, ureter and stent). Applied perspective volume rendering techniques were then applied, producing virtual endoscopic images.

RESULTS

All patients required parenteral analgesics on day 1 after stent insertion but no analgesics were used after hospital discharge. All 4 patients remained in the hospital 3 days, including the preoperative day. No sepsis or other complications were observed. Objective evaluation of successful treatment was based on improved drainage of the collecting system and visualization of all ureteral segments on IVP. Subjective improvement was based on relief of the patient discomfort experienced before stent placement. Mean followup was 16 months (range 9 to 24). Wallstent placement was successful and immediate patency was achieved in all cases. During followup 3 patients required no further intervention since the stented ureteropelvic junction remained patent and without stent dislocation. During the whole followup 2 of these patients appeared symptom-free and there was distinct improvement on excretory urography. In the other patient a lower caliceal stone entered the pelvis adjacent to the stented segment but did not cause any further problems. The stone was successfully treated with extracorporeal shock wave lithotripsy and the patient remained asymptomatic thereafter. In the remaining case stricture recurred 2 months after initial stent insertion due to the ingrowth of scar tissue through the prosthesis. Additional intervention was considered necessary, namely placement of a longer 6 cm., completely coaxial, overlapping metal stent. This patient was asymptomatic on followup excretory urography. Virtual endoscopy and IVP findings concurred. Virtual endoscopy allowed visualization of the stented ureteropelvic junction and proximal ureter cephalad and caudal to the

URETEROPELVIC JUNCTION OBSTRUCTION AND METALLIC STENTS

2385

FIG. 3. Virtual endoscopy. A, proximal end of stented ureteropelvic junction segment. B, proceeding distally within lumen, patency is documented throughout whole length of prosthesis. C, distal end of stented ureteral segment.

prosthesis. In addition, it provided easily performed navigation within the stent at different angles of view (fig. 3). DISCUSSION

For many years open surgery has been considered the gold standard for ureteropelvic junction obstruction. Current endourological instrumentation and techniques have changed the management of various urological conditions. Urologists achieved endoscopic skills that finally resulted in minimally invasive techniques, such as endopyelotomy, that gradually replaced time honored open surgical techniques, such as pyeloplasty. This evolution was due to refinement in technique and it has been validated by the high long-term success rates reported in the literature.6 – 8 Various minimally invasive options are available for ureteropelvic junction obstruction, including balloon dilation, laparoscopic pyeloplasty and antegrade or retrograde endopyelotomy.7, 18 The basic principle of endopyelotomy is fullthickness incision of the narrow segment, followed by prolonged stenting and drainage to allow regeneration of an adequate caliber ureter around the stent.5– 8 Laparoscopic pyeloplasty was further developed in an attempt to duplicate the high success rate achieved with open pyeloplasty, while providing the advantages of minimally invasive technique.18 The goal of all alternative procedures for ureteropelvic junction obstruction should be a pain-free or at least significantly improved patient with improvement in objective parameters on imaging, such as excretory urography and/or furosemide washout renal scan. In earlier studies we and others used metal stents for treating malignant ureteral obstruction.13–15, 19 Our experience was encouraging and we are in the process of evaluating different stent materials and techniques. Nevertheless, the application of metal stents for benign strictures has been cautiously addressed in the literature.15, 20 We have begun to use them and we present our experience in cases of ureteropelvic junction obstruction based on our positive experience with stent application for benign anastomotic stricture after ureteroileal diversion. The presented technique has been useful for managing ureteropelvic junction obstruction and avoiding repeat operation, which would cause the risk of increased problems with adynamic ureteral segments or the ureteropelvic junction. Notably in these cases we addressed a problem not only with ureteropelvic junction obstruction, but also with an adynamic ureteral segment. Therefore, it was not thought worthwhile to perform reoperation using an open procedure or another method, that is endopyelotomy. Based on the current data this method is recommended in cases of ureteral adynamic segments since complicated reoperation may be avoided that could even further complicate this problem. Endopyelotomy and Acucise (Applied Medical Resources, Laguna Hills, California) endopyelotomy are established techniques for managing ureteropelvic junction obstruction. In our study the reoperation rate was 25%, which is in fact comparable

with that of the former techniques. Further research and clinical evaluation must be performed with metal stents to establish the longevity of patency and absence of complications. Notably the central end of the metallic stent must lie free within the renal pelvis, ideally for about 0.5 cm., and away from any caliceal ostia. The latter technical recommendation can easily be met in those dilated renal pelves. We have not yet observed any encrustation. This observation is certainly limited by the length of the current followup. However, if we can reverse our experience with metal stents protruding through the ureteral orifice into the bladder, we should not expect any problems with encrustation from the upper end of the ureter, especially because of free urine flow through the stent. If such encrustation is observed in the future, this phenomenon may be ameliorated by embedding the stent into various anti-inflammatory agents.21 Also, over dilation of the ureter must be avoided since in the long term it may cause further fibrosis and ischemia of the ureteral wall. The position of the proximal end of the stent can also be assessed by virtual endoscopy in relation to the central ureteral orifice and its correlation with the renal pelvis. Virtual endoscopy represents a noninvasive technique that amplifies image perception in 3-D space. In this respect virtual endoscopy takes advantage of information in axial CT slices and further provides 3-D relationships of pathological regions. When virtual endoscopy is compared with endoscopy, it is obvious that it can provide visualization via the stenosis to any tubular structure beyond the narrowed segment. It has also been shown that virtual endoscopy provides intraluminal as well as extraluminal information on a given tubular structure in all possible directions. Thus, virtual endoscopy enables the exploration of body regions that are unapproachable with conventional endoscopic procedures. Nevertheless, its main disadvantage is the inability to differentiate structures with similar absorbing characteristics in the photon quantum energies used. Thus, virtual endoscopy does not differentiate fine detail in the epithelial lining of anatomical structures, which can be visualized with conventional endoscopic procedures. Further research is required. Study has been done by our group to improve the specificity of this technique, which we consider represents a diagnostic innovation for investigating hollow and tubular structures.16, 17 CONCLUSIONS

The concept of applying metallic stents for managing ureteropelvic junction obstruction and adjacent adynamic ureteral segments combined with virtual endoscopy is strengthened by the results of this study. Further research and longer followup in more patients would validate these conclusions. REFERENCES

1. Park, J. M. and Bloom, D. A.: The pathophysiology of UPJ obstruction. Current concepts. Urol Clin North Am, 25: 161, 1998 2. Streem, S. B.: Ureteropelvic junction obstruction. Open opera-

2386

URETEROPELVIC JUNCTION OBSTRUCTION AND METALLIC STENTS

tive intervention. Urol Clin North Am, 25: 331, 1998 3. Hoenig, D. M., Shalhav, A. L., Elbahnasy, A. M., McDougall, E. M., Smith, D. and Clayman, R. V.: Impact of etiology of secondary ureteropelvic junction obstruction on outcome of endopyelotomy. J Endourol, 12: 131, 1998 4. Davis, D. M.: Intubated ureterotomy: a new operation for ureteral and ureteropelvic stricture. Surg Gynecol Obstet, 76: 513, 1943 5. Goldfischer, E. R., Jabbour, M. E., Stravodimos, K. G., Klima, W. J. and Smith, A. D.: Techniques of endopyelotomy. Br J Urol, 82: 1, 1998 6. Bernardo, N. O. and Smith, A. D.: Percutaneous endopyelotomy. Urology, 56: 322, 2000 7. Shalhav, A. L., Giusti, G., Elbahnasy, A. M., Hoenig, D. M., McDougall, E. M., Smith, D. S. et al: Adult endopyelotomy: impact of etiology and antegrade versus retrograde approach on outcome. J Urol, 160: 685, 1998 8. Goldfischer, E. R. and Smith, A. D.: Endopyelotomy revisited. Urology, 51: 855, 1998 9. Motola, J. A., Fried, R., Badlani, G. H. and Smith, A. D.: Failed endopyelotomy: implications for future surgery on the ureteropelvic junction. J Urol, 150: 821, 1993 10. Sigwart, U., Puel, J., Mirkovitch, B., Joffre, F. and Kappenberger, L.: Intravascular stents to prevent occlusion and restenosis after transluminal angioplasty. N Engl J Med, 316: 701, 1987 11. Coons, H. G.: Self-expanding stainless steel biliary stents. Radiology, 170: 979, 1989 12. Kirby, R. S., Heard, S. R., Miller, P., Eardley, I., Holmes, S., Vale, J. et al: Use of the ASI titanium stent in the management of bladder outflow obstruction due to benign prostatic hyperplasia. J Urol, 148: 1195, 1992 13. Barbalias, G. A., Siablis, D., Liatsikos, E. N., Karnabatidis, D., Yarmenitis, S., Bouropoulos, K. et al: Metal stents: a new treatment of malignant ureteral obstruction. J Urol, 158: 54, 1997 14. Barbalias, G. A., Liatsikos, E. N., Kalogeropoulou, C., Karnabatidis, D. and Siablis, D.: Metallic stents in gynecologic cancer: an approach to treat extrinsic ureteral obstruction. Eur Urol, 38: 35, 2000 15. Barbalias, G. A., Liatsikos, E. N., Karnabatidis, D., Yarmenitis, S. and Siablis, D.: Ureteroileal anastomotic strictures: an innovative approach with metallic stents. J Urol, 160: 1270, 1998 16. Kagadis, G. C., Patrinou, V., Kalogeropoulou, C. P., Karnabatidis, D., Petsas, T., Nikiforidis, G. C. et al: Virtual endoscopy in the diagnosis of an adult double tracheal bronchi case. Eur J Radiol, 40: 50, 2001 17. Robb, R. A.: Virtual endoscopy: development and evaluation using the Visible Human datasets. Comput Med Imaging Graph, 24: 133, 2000 18. Janetschek, G., Peschel, R., Frauscher, F. and Franscher, F.: Laparoscopic pyeloplasty. Urol Clin North Am, 27: 695, 2000 19. Lo´ pez-Martı´nez, R. A., Singireddy, S. and Lang, E. K.: The use of metallic stents to bypass ureteral strictures secondary to metastatic prostate cancer: experience with 8 patients. J Urol, 158: 50, 1997 20. Pauer, W. and Eckerstorfer, G. M.: Use of self-expanding permanent endoluminal stents for benign ureteral strictures: midterm results. J Urol, 162: 319, 1999 21. Antimisiaris, S. G., Siablis, D., Liatsikos, E., Kalogeropoulou, C., Tsota, I., Tsotas, V. et al: Liposome-coated metal stents: an in vitro evaluation of controlled-release modality in the ureter. J Endourol, 14: 743, 2000

EDITORIAL COMMENTS These authors report the use of metal stent for benign ureteropelvic junction obstruction as an alternative to standard pyeloplasty and a plethora of minimally invasive procedures, such as percutaneous and ureteroscopic endopyelotomy, Acucise and laparoscopic repair. This small experience does not provide a clear advantage over these techniques. There is a need for improved stenting on a longterm basis in long ureteral strictures of the upper ureter, malignant strictures and ureterointestinal strictures. Unlike the successful use of UroLume (American Medical Systems, Minnetonka, Minnesota) for bladder outflow obstruction, in which exposure to urine only occurs during voiding, a wide lumen precludes obstruction secondary to hyperplastic tissue response. The use of metal stents in the ureter on a long-term basis has been plagued with these problems. It would be interesting to see if a Wallstent in the upper ureter would be better than in the lower ureter and/or different from other stent designs at longer term followup. Virtual endoscopy provides excellent images of the stent and lumen. However, it is unable to provide key information, such as epithelialization and function across the stent. Gopal Badlani Department of Urology Long Island Jewish Medical Center New Hyde Park, New York This innovative article describes the first use of a permanent metal stent to treat recurrent ureteropelvic junction obstruction after failed open dismembered pyeloplasty in 4 patients. The followup assessments combine standard imaging with 3-D CT reconstructions of the region of interest, further enhanced by virtual endoscopy with “fly through” sequences within the stent. This new technique joins the ever expanding list of alternative treatments available for ureteropelvic junction obstruction. This fact in itself perhaps reflects our failure to understand the pathophysiology of the original condition and more importantly why the so called gold standard treatment of open dismembered pyeloplasty fails. Use of the Wallstent throughout the genitourinary tract has met with mixed success, and the problems with encrustation and exuberant urothelial ingrowth within the interstices and the ends of the stent have limited its uptake. Of the 4 stents reported 3 remain patent but 1 failed at 2 months. One wonders whether this technique is effectively a temporizing method and not a permanent solution for the patients. 3-D CT reconstruction and virtual endoscopy hold great promise. The images in the article are remarkable and are of a quality that we could only dream about until recently. The speed of advances in CT data acquisition and manipulation is extraordinary and it is likely that these types of images will be available from most machines in the near future. The exact role of this technology in the urological imaging armamentarium will need to be discussed at length, particularly with the concerns about the amount of radiation exposure required. Presently 3-D reconstructive techniques are used extensively for preoperative planning of open and laparoscopic renal parenchymal surgery. Virtual endoscopy is likely to be just as valuable as a preoperative planning tool for renoscopy and perhaps will form the basis of the surgical simulators of the near future. Ken Anson and Roger Kirby Department of Urology St George’s Hospital London, United Kingdom