Fistulas of the Upper Urinary Tract: Percutaneous Management

0022-534 7/87 /1386-1382$02.00/0 Vol. 138, December Printed in U.S. A.

THE JOURNAL OF UROLOGY

Copyright© 1987 by The Williams & Wilkins Co.

FISTULAS OF THE UPPER URINARY TRACT: PERCUTANEOUS MANAGEMENT P. J. MAILLET,* D. PELLE-FRANCOZ, A. LERICHE, R. LECLERCQ AND C. DEMIAUX From the Services de Radiologie d'Urgence and d'Uroradiologie, Hopital Edouard Herriot, Lyon and Clinique d'Urologie, Centre Hospitalier Lyon-Sud, Pierre-Benite, France

ABSTRACT

We treated 40 patients with urinary fistulas by interventional radiology. The antegrade percutaneous route, catheterization of the ureter and bypassing of the fistula enabled ureteral stenting in 36 patients (90 per cent). Criteria for successful treatment were healing of the fistula, normal renal function (evaluated by excretory urography and radionuclide studies) and absence of secondary stenosis at 6 months. Of the patients 28 (70 per cent) were treated successfully. The number of nephrectomies after failure of percutaneous techniques (5 of 40, or 12.5 per cent) seems lower than in the case of surgery. The results were excellent for fistulas occurring after endourology (all 9 successful) or after ureterointestinal anastomoses (7 of 8). On the other hand, the results appear disappointing in patients with fistulas in transplanted kidneys (3 of 4 failures). (J. Urol., 138: 13821385, 1987) Fistulas of the upper urinary tract most often occur after an operation and they are difficult to treat in critically ill patients. Surgical repair of these lesions leads to nephrectomy in about 20 per cent of the cases and to a mortality rate of 10 per cent. 1 Catheterization of fistulas long has been used by urologists but the classical retrograde approach has a high failure rate when a fistula must be crossed. Goldin 2 and Lang3 first attempted antegrade percutaneous treatment of such fistulas. We present our experience with this method in 40 fistulas, together with a review of the literature (approximately 100 cases reported to date). MATERIAL AND METHODS

Of 40 patients referred to us for the management of a urinary fistula 16 were women and 24 were men from 15 to 77 years old (mean age 51 years). In 30 patients the fistulas were subsequent to an operation, including renal transplantation in 4, gynecological intervention for neoplasm in 7 (a Wertheim procedure in 5 and pelvectomy in 2, 3 of which followed radiotherapy), nonmalignant urological operations in 11 (ureterotomy for lithiasis in 8 and ureterovesical reimplantation in 3) and ureterointestinal anastomoses in 8 (Bricker, Camey, Coffey and Kock pouch procedures in 5, 1, 1 and 1, respectively, 2 of which followed radiotherapy). In 9 patients the fistulas occurred after interventional radiology or endourology: ureteroscopy in 4, percutaneous lithotripsy in 2, explosion of the balloon during dilation of a ureter in 1 and percutaneous removal of a ureteral calculus with a Dormia basket in 2. Finally, in 1 patient the fistula occurred spontaneously above an obstructive ureteral uric acid calculus. Among the 4 transplant patients the site of the fistula was the renal pelvis in 1 and the distal ureter in 3. In the remaining 36 patients the site was the kidney in 2, renal pelvis in 7, and upper, middle and lower ureter in 7, 7 and 13, respectively. These fistulas were accompanied by urinomas or neighboring abscesses in 8 patients, all of whom required associated percutaneous drainage. The fistulas appeared an average of 10 days postoperatively. The delay between occurrence of the fistula and referral to the radiologist was less than 1 week in 22 cases and more than 1 month in 6. Before referral 11 patients had undergone an Accepted for publication April 21, 1987. *Requests for reprints: Service de Radiologie, Institut Arnault Tzanck, 06700 Saint Laurent du Var, France.

attempt at retrograde catheterization that failed in 10 and 8 already had undergone an unsuccessful surgical repair. The primary aim of percutaneous techniques is to achieve drying of the fistula, and percutaneous nephrostomy generally is sufficient therapy. However, we prefer to insert a ureteral stent percutaneously, which not only allows for drying but also closure of the fistula in the best conditions. The catheter is passed down through the ureter, crosses the fistulous opening and descends into the bladder. Bypassing the fistula presents widely varying difficulties related to its size, site and cause, and to changes in the surrounding tissues, as well as frequently associated stenosis of the neighboring ureter. The catheter, usually a 9F polyethylene tube, is provided with multiple drainage holes except at the level of the fistula. It is obturated at the cutaneous emergence to provide internal drainage. Such access to the catheter makes possible periodic control of its position and functioning, as well as closure of the fistula. The catheter is left in place for a more or less lengthy period, depending on the cause and severity of the fistula (an average of 65 days in our series). Its presence allows for immediate drying of the fistula so that treatment is done on an outpatient basis as often as possible. After the indwelling period the stent is exchanged for 2 to 3 days for a simple nephrostomy tube. If the fistula has healed, the lumen is patent and the mucosal ureteral edema owing to the local irritation has resolved, the nephrostomy tube is removed definitively. RESULTS

Results of the technique. The percutaneous approach to the collecting system was successful in all 40 patients but placement of a stent across the fistula site was successful in only 36 (90 per cent). The 4 failures were owing to the presence of a surgical clip on the ureter in 1, extensive necrosis of the ureter of a transplanted kidney in 1 and total severance of the ureter associated with an infected urinoma in 2. The former 2 patients underwent nephrectomy. Of the latter 2 patients 1 underwent autotransplantation and 1 underwent ureterovesical reimplantation. No serious complication was noted during percutaneous procedures nor during the stenting period. Results of the method. As criteria of success we used those proposed by Lang: closure of the dehiscence and healing of the fistula, normal renal function at 6 months as assessed by excretory urography (IVP), biological and radionuclide studies, and absence of secondary stenosis of the urinary tract. 3 As

1382

138S might be expected the ultimate lower than the success rate for µ-~H·-··..,·~ 1). Of our 40 patients 28 per cent) had a ing renal unit 6 months after the salvage ~"""'"r''"'" Of the failures 2 were deaths owing to while the fistulas appeared to be cured, and 6 were the method itself: 2 unclosed fistulas despite efficient in the transplanted kidneys and 4 secondary ureteral stenoses. For the 4 patients who presented with secondary stenosis we chose an operation rather than balloon catheter dilation by the percutaneous route. These 6 failures were treated nephrectomy in 3 (including 2 transplanted kidneys), autotransplantation in 1, ureterovesical reimplantation in 1 and a psoas bladder hitch procedure in 1. No patient required emergency surgical intervention and none died during percutaneous treatment. Apart from regular radiological surveillance it appeared to be of great importance to perform systematically urinary bacteriology studies and to prescribe appropriate treatmento We noted only 1 episode of acute pyelonephritis that was caused obstruction of the catheter by purulent incrustations. DISCUSSION

Surgical repair of a ureteral fistula is a long and difficult task of reconstruction in weak patients who already have been operated on recently and whose hospitalization will be lengthened. Conversely, percutaneous treatment often can be perrehosformed on an outpatient basis, the patient simply pitalized for a few days for removal of the stenL The ~"···~"~··of conservative treatment by means of a stent in the is not new. However, the success rate of retrograde catheterization is only 10 to 40 per cent,4 which explains why the anteg:rade percutaneous approach, which is able to bypass the fistula site in 90 per cent of the cases, is accepted as the method of choice to date. Several advantages seem to justify antegrade stenting of the ureter. Pyelography allows for initial evaluation of the lesion in excellent conditions. A catheter is used that is designed specially for the individual patient and the site of the fistula. Subsequent followup is done regularly, approximately every 20 days. Local treatment may be given if required. The stent may TABLE 2.

Correlation between therapeutic delay and success rate of the percutaneous procedure No. Pts./No. Successes Total No. Pts.

Transplanted kidneys Gynecological surgery Urological surgery Ureterointestinal anastomoses Endourology Spontaneous Totals(%)

4 7 11 8

9

tirr1e 'Hhenever n0c-

a double balloon

of meteral stenting as described by

of the urothelium to close over the trE1Q1-1er,t ureteral ischemia. 9 The catheter up, contributes to healing and molds reducing the risk of secondary stenosis. Independent of the ischemic cmnpromise of the involved ureteral segment, another factor that may be considered for the failures is the interval between occurrence of a fistula and initiation of percutaneous treatment. In table 2 we arbitrarily chose as critical time the end of week l and the end of month 1: the success rate reached 82 per cent with the early treatment but decreased to 33 per cent when the efficient treatment was initiated after more than 1 month. Such remarkable results should prompt one to make an immediate h,orn,,io,rri" decision as soon as the fistula appears. Our results (70 per cent success at 6 months) confirm the 60 to 75 per cent success rate found in the literature.'3, 5 - 7, 10 , 11 Failures of the percutaneous methods do not compromise the future, since an operation remains possible once the critical has and local conditions have uose(1miffc to failure of the ""''-'""1 4 c1c~ was 12,5 per cent (5 of 40 n,ir1Pnr
Dsiay


(1 wk.-1 mo.)

>1 Mo.

4/2 6/5 4/3 8/8

nAnrr,.~r,Ar>

::i, 5, 7, 14

22/18 (82)

UCHh>vcrn.,c,

usuof the anastomosis rather than to

TABLE l

Attempted Stentings Transplanted kidneys Gynecological surgery

Successful Sten tings

Normal Kidney Function at 6 Mos.

4

3

7

5

1 3

11

10

7

Ureterointestinal anastomoses

8

8

7

Endourology Spontaneous Total Nos.(%)

9

9 1

9 1

Urological surgery

1

40

36 (90)

28 (70)

Cause of Failure 2 nonhealing l secondary stenosis, 1 death of Ca before 6 mos. 3 secondary stenoses

1 death of Ca before 6 mos.

Surgical Treatment of Failures 2 nephrectomies 1 ureterovesical reirnplantation 1 nephrectomy, 1 autotransplantation, 1 psoas bladder hitch procedure

1384

MAILLET AND ASSOCIATES

ischemia. 15' 16 Of 8 patients treated we obtained 7 successful outcomes. The remaining patient eventually died of cancer before month 6. The literature reports 9 successful conservative measures in 9 patients. 5 ' 6 ' 10• 16- 18 These results are all the more remarkable, since these patients are critically ill and the technical difficulties are great. Antegrade catheterization of the ileal loop sometimes necessitates retrograde fibroscopy, and the custom-made catheter inserted in the back has to exit at the stoma. In our opinion fistulas subsequent to endourology or interventional radiology appear to be excellent indications for im-

FIG. 1. Persistent leakage after 3 weeks of ureteral stenting. Transplant kidney was removed.

mediate repair by the percutaneous route (9 of 9 successful cases) (fig. 3). This remarkable result undoubtedly is related to the immediate recognition of the lesion. In these cases the absence of surgical trauma of the collecting system and the retroperitoneum reduces the risks of ischemia and infection. Finally, other percutaneous methods may be helpful. Simple drainage by percutaneous nephrostomy is used in the case of

FIG. 3. Perforation of ureter occurred during retrograde ureteroscopy for lithotripsy. Urinoma (1.5 1.) was drained percutaneously and ureteral stent was left for 4 weeks with success.

FIG. 2. External and internal radiotherapy followed by surgery for cancer of cervix. Ureterovaginal fistula occurred 12 days after Wertheim's operation. A, antegrade pyelography reveals vaginal leakage from right lower ureter. B, catheter bypassed fistula and ureteral stent was left for 7 weeks. C, followup IVP 7 months later shows success.

FKSTULAS OF UPPER URINARY TRACT

stent failure. This procedure allows for efficient drying of the fistula but it frequently causes secondary stenosis. In the 6 renal units on which we attempted this treatment alone (not included in this series) 3 secondary nephrectomies were necessary.18 In the case of terminally ill patients with neoplastic pelvic infiltration and an impassable fistula only palliative treatment could be done. The association of a definitive percutaneous nephrostomy and embolization of the ureter with a detachable balloon sometimes is useful.1 9 • 20 In the 2 patients in whom it was used we found this method to be satisfactory. CONCLUSION

Percutaneous treatment of ureteral fistulas appears to be a rewarding and highly efficacious method. It is noteworthy that the value lies in crossing the fistula and in molding the ureter, since it is thus that the best success rate may be achieved. 3• 6 • 7 Our results show that more often surgeons faced with a fistula first quickly resort to interventional radiology. Absence of a cutaneous scar, no general anesthesia, short hospitalization, reduced cost and relatively limited invasion are the most appealing arguments of this method.

6. 7. 8. 9. 10. 11. 12. 13. 14.

Profs. J. M. Dubernard, J. P. Archimbaud and A. Leriche, Department of Urology, and Prof. D. Dargent, Department of Gynecology, referred their patients to us. REFERENCES

15.

16.

1. Higgins, C. C.: Ureteral injuries during surgery: a review of 87

cases. J.A.M.A., 199: 82, 1967. 2. Goldin, A. R.: Percutaneous ureteral splinting. Urology, 10: 165, 1977. 3. Lang, E. K.: Antegrade ureteral stenting for dehiscence, strictures, and fistulae. Amer. J. Roentgen., 143: 795, 1984. 4. Benoit, G., Boccon-Gibod, L., Teyssier, P. and Steg, A.: Traumatismes iatrogenes de l'uretere. Analyse de 47 cas. Chirurgie, 109: 160, 1983, 5. Andriole, G. L., Bettmann, M. A., Garnick, M. B. and Richie, J. P.: Indwelling Double-J ureteral stents for temporary and per-

17. 18. 19. 20.

manent urinary drainage: experience with 87 patients. J. UroL, 131: 239, 1984. Druy, E. M., Gharib, M. and Finder, C. A.: Percutaneous nephroureteral drainage and stenting for postsurgical ureteral leaks. Amer. J. Roentgen., 141: 389, 1983. Mitty, H. A., Train, J. S. and Dan, S. J.: Antegrade ureteral stenting in the management of fistulas, strictures and calculi. Radiology, 149: 433, 1983. Van Arsdalen, K. N., Pollack, H. M. and Wein, A. J.: Ureteral stenting. Sem. Urol., 2: 180, 1984. Borkowski, A.: Regeneration of the ureter and the renal pelvis from transversely transected and anastomosed strip of the ureteral wall. IL Clinical Eur. Urol., 6: 21, 1980. Bettmann, M.A., Murray, P. D., Perlmutt, L. M., Whitmore, W. F., III and Richie, J. P.: Ureteroileal anastomotic leaks: percutaneous treatment. Radiology, 148: 95, 1983. Elyaderani, M. K., Gabriele, 0. F., Kandzari, S. J. and Belis, J. A.: Percutaneous nephrostomy and antegrade ureteral stent insertion. Urology, 20: 650, 1982. Benoit, G., Benarbia, S., Bellamy, J., Charpentier, B. and Fries, D.: Complications urologiques de la transplantation renale. Role de la longeur de l'uretere. Presse Med., 15: 101, 1986. Goldstein, I., Cho, S. I. and Olsson, C. A.: Nephrostomy drainage for renal transplant complications. J. Urol., 126: 159, 1981. Goodwin, W. E. and Scardino, P. T.: Vesicovaginal and ureterovaginal fistulas: a summary of 25 years of experience. J. Urol., 123: 370, 1980. Hensle, T. W., Bredin, H. C. and Dretler, S. P.: Diagnosis and treatment of a urinary leak after ureteroileal conduit for diversion. J. Urol., llS: 29, 1976. Kaufman, J. J. and Ehrlich, R. M.: Prophylactic and therapeutic use of abdominal drains and percutaneous nephrostomies and stents for treatment of precarious ureteroileal anastomosis. Urology, 20: 118, 1982. Nieh, P. T.: Ureteroileal fistulas. J. Urol., 130: 555, 1983. Persky, L., Hampel, N. and Kedia, K.: Percutaneous nephrostomy and ureteral injury. J. Urol., 125: 298, 1981. Gunther, R., Klose, K. and Alken, P.: Transrenal ureteral occlusion with a detachable balloon. Radiology, 142: 521, 1982. Papanicolaou, N., Pfister, R. C. and Yoder, I. C.: Percutaneous occlusion of ureteral leaks and fistulae using nondetachable balloons. Urol. Rad., 7: 28, 1985.