The Use of Bowel for Ureteral Replacement for Complex Ureteral Reconstruction: Long-Term Results

The Use of Bowel for Ureteral Replacement for Complex Ureteral Reconstruction: Long-Term Results Benjamin I. Chung,* Karim J. Hamawy, Leonard N. Zinman and John A. Libertino From the Department of Urology, Lahey Clinic Medical Center, Burlington, Massachusetts

Purpose: Ileal and intestinal ureteral replacement remains a useful procedure for complex ureteral reconstruction. We examined the long-term safety and efficacy of this procedure, especially in regard to maintaining preoperative renal function and the avoidance of major complications. Materials and Methods: A total of 56 patients underwent intestinal ureteral substitution at our institution between 1979 and 2003, including 52 with an ileal ureteral replacement, 2 with colonic replacement alone and 2 with bilateral ureteral replacement, necessitating ileum and colon for 1 ureter each. The factors reviewed were indications for surgery, type of ureteral replacement, and the presence and type of complications. Followup data included excretory urogram or equivalent imaging results, and measurement of serum chloride, bicarbonate and creatinine before and after the procedure. Results: Overall the complication rate remained low. Mean followup was 6.04 years (median 3.2). Most postoperative complications, which occurred in 10 patients (17.9%), were minor in nature, including pyelonephritis, fever of unknown origin, neuroma, hernia, recurrent urolithiasis and deep venous thrombosis. Major complications occurred in 6 patients (10.5%), including anastomotic stricture, ileal graft obstruction, wound dehiscence and chronic renal failure. Overall patients did not experience worsening renal function after the procedure with equivalent median creatinine before and after the procedure (1.0 mg/dl). Conclusions: During long-term followup major complications are rare and renal function remains preserved. Ileal and intestinal ureteral substitution remains a safe and efficacious procedure in patients with complex and difficult ureteral issues not amenable to more conservative measures. Key Words: ureter, ileum, ureteral stricture, intestine

and the avoidance of major surgical and metabolic complications.

leal or intestinal ureteral replacement has become an accepted reconstructive procedure since Goodwin et al pioneered its usefulness.1 Despite its complexity as a procedure and greater potential for complications it remains a useful procedure for complex ureteral reconstruction when more conservative approaches are not feasible. Common indications are ureteral stricture disease secondary to retroperitoneal fibrosis, iatrogenically induced ureteral injury, recurrent ureteropelvic junction obstruction, intractable stone disease and ureteral carcinoma. Contraindications are azotemia and inflammatory bowel disease, necessitating careful patient selection. Various concerns and questions remain as to the long-term viability of the procedure. Is a refluxing or nonrefluxing anastomosis preferred? How efficacious is the procedure for maintaining renal function? How prevalent are metabolic derangements postoperatively and how can they be avoided? How safe is the procedure overall? Most groups would agree that intestinal ureteral replacement is an acceptable form of treating complex ureteral problems. Therefore, we examined our extensive series and the long-term results of the procedure, especially with respect to the preservation of renal function,

I

MATERIALS AND METHODS A total of 56 patients underwent ureteral replacement with intestinal segments at our institution between 1979 and 2003. These patients were identified by examining operating room records for these dates. Institutional Review Board approval was obtained in standard fashion for the study. All patients who underwent the procedure were included and none were excluded. The majority of the procedures were performed by 2 surgeons (JAL and LNZ). Medical records were retrospectively reviewed to examine the indications for surgery, type of intestinal segment used for ureteral replacement, the presence and type of complications, and the date of last followup. Followup data included the results of IVP or equivalent upper tract imaging. Preoperative and postoperative serum chloride, creatinine and bicarbonate were recorded and the data were then examined and analyzed for significance. Data were compared using SAS statistical software (SAS Institute, Cary, North Carolina) using the paired t test to compare means with ⬍0.05 considered statistically significant. The technique of ileal ureteral replacement at our institution has been described previously.2 Briefly, preoperative bowel preparation is administered in the form of oral polyethylene glycol solution and oral antibiotics. An approxi-

Submitted for publication March 27, 2005. Study received institutional review board approval. * Correspondence: Department of Urology, Lahey Clinic Medical Center, 41 Mall Rd., Burlington, Massachusetts 01805 (telephone: 781-744-2511; FAX: 781-744-5429; e-mail: Benjamin_Chung@ hotmail.com).

0022-5347/06/1751-0179/0 THE JOURNAL OF UROLOGY® Copyright © 2006 by AMERICAN UROLOGICAL ASSOCIATION

179

Vol. 175, 179-184, January 2006 Printed in U.S.A. DOI:10.1016/S0022-5347(05)00061-3

180

BOWEL FOR URETERAL REPLACEMENT FOR COMPLEX URETERAL RECONSTRUCTION

FIG. 1. Ileal segment is anastomosed to renal pelvis in end-to-end fashion.

mately 30 cm length of ileum is inspected to ensure adequate mesenteric length for mobilization for the procedure. The segment is marked with sutures, placed in an isoperistaltic direction and the diseased ureter is resected or ligated. The renal pelvis is then widely spatulated at a 45-degree angle and anastomosed in 2 layers to the ileum end to end (fig. 1). Prior to the completion of the anastomosis a 7Fr 90 cm single J ureteral stent is placed into the renal pelvis, down the ileal segment and externalized through the vesical and abdominal walls. The distal anastomosis is begun by first opening the bladder vertically along the anterior surface. Vesicostomy is created in the dome and refluxing anastomosis is constructed via a seromuscular and mucosal 2-layer closure between the distal ileal segment and bladder (fig. 2). RESULTS A total of 52 patients received an ileal ureteral replacement, 2 received a colonic ureteral replacement alone and 2 required bilateral ureteral replacement, necessitating ileum and colon for 1 ureter each (table 1). More conservative procedures including ureterolysis were considered but subsequently ruled out prior to the procedure. Mean followup was 6.04 years (median 3.2, range 2 months to 21 years). Mean patient age at surgery was 48.6 years (range 19 to 71). The most common indication for surgery was iatrogenically induced ureteral stricture (23 cases). Various iatrogenic causes included operative trauma, fibrosis secondary to operative procedures in the vicinity of the ureter and strictures secondary to external beam radiation therapy. Other indications were recurrent ureteropelvic junction obstruction in 8 cases, retroperitoneal fibrosis in 8 and recurrent urolithiasis resulting in ureteral stricture or intractable stone issues in 6. Four patients had a solitary kidney (table 2). Table 3 lists all complications. Minor complications occurred in 10 patients (17.9%), including pyelonephritis, fe-

FIG. 2. Retroperitoneal ileal segment is anastomosed to bladder via vertical cystotomy.

ver of unknown origin, neuroma, hernia, recurrent urolithiasis and deep venous thrombosis. Major complications occurred in 6 patients (10.5%), including anastomotic stricture, ileal graft obstruction, wound dehiscence and chronic renal failure. The 1 instance of anastomotic stricture developed at the proximal anastomosis and was amenable to endoscopic dilation. The 1 case of ileal graft obstruction developed approximately 3 years after the initial procedure after assessment for chronic flank pain revealed a redundancy in the graft that intermittently kinked. The graft was taken down, shortened and re-implanted. Three patients (3.6%) experienced worsening renal function postoperatively. All 3 patients had preexisting azotemia with serum creatinine preoperatively greater than 2.0 mg/dl. One of these patients subsequently required conversion to an ileal

TABLE 1. Procedural subtype Procedure

No. Pts

Ileal ureteral replacement Sigmoid colon ureteral replacement Bilat replacement, 1 side colon, 1 side ileum

52 2 2

BOWEL FOR URETERAL REPLACEMENT FOR COMPLEX URETERAL RECONSTRUCTION

181

TABLE 2. Indications Indication

No. Pts

Iatrogenic ureteral stricture Recurrent calculi leading to stricture Retroperitoneal fibrosis Recurrent ureteropelvic junction obstruction Stone chute Stricture secondary to external beam radiation Ureteral transitional cell Ca Other

23 8 7 7 4 3 2 2

loop to prevent worsening renal function, presumably due to reflux. Another patient had multifactorial reasons for renal failure, including hypertension and diabetes mellitus, but reflux nephropathy could not be ruled out as a cause. There were also 3 patients who underwent the procedure with initial serum creatinine greater than 2.0 mg/dl in whom worsening underlying renal function did not develop. Four patients had metabolic acidosis, of whom 3 had preexisting azotemia, which predisposed them to the condition. The fourth patient did not have a clear etiology for the condition. No patients with colonic ureteral replacement had metabolic acidosis. Overall most patients tolerated the procedure quite well. On the last followup radiological upper tract study there was no evidence of obstruction in any patient during a median followup of 821 days (range 15 to 8,700). Figures 3 and 4 show preoperative and postoperative IVP results, respectively. Table 4 lists preoperative and postoperative creatinine, bicarbonate and chloride values. Patients in whom metabolic derangements developed were included in the analysis. These preoperative and postoperative values were not statistically different when compared using paired t test statistical analysis (table 4). There were no postoperative deaths. A successful outcome of the procedure was defined as the absence of death, major complications, worsening baseline renal function, metabolic derangements and obstruction of the intestinal segment used to replace the ureter. Using these criteria 46 of 56 patients (82%) had a successful outcome. DISCUSSION The reconstruction of complex ureteral injuries is one that requires careful planning and decision making. ReplaceTABLE 3. Complications No. Pts

Comments

Pyelonephritis

Type

4

Chronic renal failure

3

Incisonal hernia Fever of unknown origin

2 1

Neuroma Deep venous thrombosis Recurrent urolithiasis

1 1 1

Ileal segment obstruction

1

Pyeloileal anastomotic stricture Wound dehiscence

1

Hospitalized for intravenous antibiotics (2) All pts had serum creatinine ⱖ2 Required surgical repair Pt hospitalized but no source ever found, fever spontaneously resolved Treated conservatively Treated conservatively Required extracorponeal shock wave lithotripsy on ileal ureteral replacment side Caused by kinking due to redundancy in segment, subsequently shortened Treated conservatively with balloon dilation Closed in operating room

1

FIG. 3. Preoperative IVP reveals nonexcretion of contrast material secondary to long obliterating stricture of left upper tract.

ment of the ureter with ileum or other intestinal segments is performed as a means of ureteral reconstruction after more conservative methods have failed or been ruled out. Other indications are intractable, recurrent renal calculi, for which the large caliber of an intestinal ureter acts as a stone chute to facilitate the passage of calculi out of the upper tracts. Often this is done in patients who have concomitant native ureteral abnormalities.3 In the past there have been concerns as to the long-term safety of the procedure. To predict a successful outcome in these patients they must be chosen carefully. The procedure is reported to have a higher complication rate than other, more conservative methods4 and it can be time-consuming and extremely difficult.5 Contraindications are preexisting azotemia, hepatic dysfunction, irradiated bowel, inflammatory bowel disease and an inadequate length of usable bowel. Dysfunctional voiding is a relative contraindication to avoid excessive high intravesical pressures, which could lead to reflux nephropathy or increase the likelihood of long-term exposure of the bowel mucosa to urine, which could lead to metabolic derangements. In this situation even in patients with normal voiding habits some groups recommend frequent voiding to minimize the degree of vesicoileal reflux.6 In our series there were patients who underwent ureteral replacement with large bowel. This was done in patients who had already undergone replacement with ileum on 1 side or to avoid 2 bowel anastomoses in a patient who already had an end sigmoid colostomy. These patients have done well in followup with no metabolic derangements. The use of large bowel is feasible in patients unable to undergo ileal replacement.

182

BOWEL FOR URETERAL REPLACEMENT FOR COMPLEX URETERAL RECONSTRUCTION

FIG. 4. Postoperative IVP demonstrates widely patent proximal anastomosis at renal pelvis with contrast material opacifying ileal segment.

Patients with pre-procedure azotemia and serum creatinine greater than 2.0 mg/dl should undergo ureteral replacement with caution.5 Post-procedure metabolic derangements has not been a major issue in our patient cohort. Much of the success can be attributable to careful patient selection. In our series 3 patients (5.3%) had worsening azotemia and all 3 had preprocedure serum creatinine greater than 2.0 mg/dl. However, there were 3 patients who also had serum creatinine greater than 2.0 mg/dl who did not have worsening renal function. In these complicated instances it is not always possible to exclude patients due to preexisting renal insufficiency because the intestinal replacement procedure may be the only option. Nonetheless, caution must be exercised in these patients and close followup is warranted.5 Some groups have theorized that in patients with normal renal function any potential issues can be averted due to the normal metabolic compensation that occurs.3 In patients with preexisting azotemia performing segmental interposition using much less bowel may prevent postoperative metabolic problems.2 Other issues involve a refluxing anastomosis and the potential for upper tract damage. Tanagho reported on 5 patients who experienced renal deterioration and electrolyte derangements after persistent reflux caused progressive ileal graft dilatation.7 He believed that the incorporation of a bowel segment in the closed urinary system was at issue, leading to intractable metabolic problems.7 However, in examining his series preoperative serum creatinine is not available but all patients had a significant degree of preexisting renal failure. His patients are the type that would not

do well with the procedure due to azotemia, highlighting the need for careful patient selection. Some groups have advocated adding an intussuscepted nipple to prevent reflux with avoidance of the resultant problems.8 Others have recommended creating the antireflux system at the proximal ureteroileal segment anastomosis.9 However, some found problems in implementing nonrefluxing anastomoses. Shokeir et al found that using a distal nipple valve to prevent reflux led to significant problems, including detussusception of the nipple mechanism and stone formation at the staple line of the intussuscepted segment.8 Because of such issues, others have deemed an antireflux procedure unnecessary.5 In a series of Boxer et al only 1 of 40 patients with a freely refluxing anastomosis had worsening azotemia secondary to presumed reflux nephropathy.5 Also, some groups have thought that the natural peristaltic antegrade waves of the intestinal segment prevent significant reflux from reaching the kidney.3 Olsson believed that this peristalsis provided a dampening effect to any potential retrograde reflux that could reach the kidney.3 Animal studies have shown that a antirefluxing anastomosis in dogs did not confer any advantage in maintaining renal morphology and function.10 Also, tapering the ileal segment did not decrease the prevalence of hyperchloremic metabolic acidosis.10 In our patient cohort we used freely refluxing techniques and renal function was preserved with equivalent creatinine before and after the procedure. However, the issue of close followup in all of these patients remains germane for troubleshooting potential problems that may arise in terms of the upper tracts. The issue of urinary tract infections after the implementation of this procedure is well documented. Olsson noted that urinary tract infection can be expected in about 75% of patients who have had bowel interposed into the urinary tract.3 In our series urinary tract infection was a minor issue. Two patients required inpatient hospitalization to

TABLE 4. Preoperative and postoperative serum electrolytes Values

p Value

Short-term follow-up Creatinine (ml/dl): Median preop (range) Median postop (range) Normal range Bicarbonate (mEq/l): Mean preop (range) Mean postop (range) Normal range Chloride (mEq/l): Mean preop (range) Mean postop (range) Normal range Creatinine (mg/dl): Mean preop (range) Mean postop (range) Normal range Bicarbonate (mEq/l): Mean preop (range) Mean postop (range) Normal range Chloride (mEq/l): Mean preop (range) Mean postop (range) Normal range

1 (0.7–8.2) 1 (0.7–3.3) 0.4–1.3 26 (17–33) 25.5 (17–31) 24–32

0.5638

0.2803

104 (98–115) 107 (98–114) 98–110 Long-term followup

0.1121

1.2 (0.6–2.4) 1.1 (0.6–7.9) 0.4–1.3

0.4514

27 (14–33) 27 (14–33) 24–32

0.8428

104 (100–112) 106 (98–118) 98–110

0.1216

Short-term followup less than 1,000 days (median 241, range 11 to 950) and long-term followup greater than 1,000 days (median 2,884, range 1,005 to 8,931).

BOWEL FOR URETERAL REPLACEMENT FOR COMPLEX URETERAL RECONSTRUCTION treat infections but these episodes were short and without major sequelae. One patient had chronic pyelonephritis and subsequent worsening renal function but that patient also had other significant comorbidities, including diabetes and hypertension, which could have led to worsening renal insufficiency. If the issue of urinary tract infection becomes chronic, these patients should be treated with chronic administration of prophylactic antibiotics.3 It should also be noted that most patients who underwent the procedure at our institution came with complex and difficult reconstructive issues. In many patients a number of previous procedures had failed, which emphasizes that our success rate was excellent considering the state in which most of these patients initially presented. Despite this fact in our series we noted that there was no evidence of any obstruction at final radiological upper tract followup study. Our findings are corroborated by Fritzsche et al, who retrospectively examined 34 patients who had undergone an ileal ureter procedure and found that only 3 had radiographic evidence of increasing upper tract dilatation during longterm followup.11 This study gives credence to the safety of the freely refluxing anastomosis in this procedure. New variations of the previously described technique hold promise. Ali-El-Dein and Ghoneim described an ileal ureteral model based on the Yang-Monti principle, which uses a shorter piece of bowel that is incised along its antimesenteric border and tubularized around a 16Fr tube.12 The rationale behind using a shorter piece of bowel was to decrease the potential absorptive area and potentially decrease the incidence of metabolic derangements. Also, using less bowel produces less mucus. Gill et al reported the creation of an ileal ureter with a laparoscopic approach with good overall results.13 The indication for their procedure was proximal ureteral transitional cell carcinoma in a solitary kidney. Total ureterectomy was performed with laparoscopic interposition of an ileal segment via a transperitoneal approach. If such technical advances are shown to be efficacious in the long term, they will continue to improve a technique that we observed has good overall longterm results.

CONCLUSIONS Replacement of the ureter with ileum or other intestinal material remains a procedure reserved for complex ureteral reconstruction. The majority of our patients had a successful outcome of the procedure, defined as no evidence of metabolic derangements, no worsening baseline renal function and no evidence of obstruction. Although it may be intrinsically more difficult than more conservative procedures, we noted a low rate of major complications. There were minor complications, of which some were significant and required intervention, but these instances were infrequent. In our series progressive renal deterioration has not been an issue even during long-term followup with equivalent creatinine preoperatively and postoperatively. Likewise metabolic derangements have not been an issue except in the few patients with preexisting renal insufficiency. We recommend using this procedure with careful patient selection and close attention to surgical technique to effect the best outcomes.

183

Abbreviations and Acronyms IVP ⫽ excretory urogram

REFERENCES 1. Goodwin, W. E., Winter, C. C. and Turner, R. D.: Replacement of the ureter by small intestine: clinical application and results of the ileal ureter. J Urol, 81: 406, 1959 2. Smith, J. J., III and Libertino, J. A.: Ileal ureter. In: Reconstructive Urologic Surgery. Edited by J. A. Libertino. St. Louis: Mosby-Year Book Publishers, pp. 227–233, 1998 3. Olsson, C. A.: Ileal ureter and renal autotransplantation. Urol Clin North Am, 10: 685, 1983 4. Koch, M. O. and McDougal, W. S.: Ureteral substitution. In: Difficult Problems in Urologic Surgery. Edited by W. S. McDougal. Chicago: Yearbook Medical Publishers, pp. 76 –100, 1988 5. Boxer, R. J., Fritzsche, P., Skinner, D. G., Kaufman, J. J., Belt, E., Smith, R. B. et al: Replacement of the ureter by small intestine: clinical application and results of the ileal ureter in 89 patients. J Urol, 121: 728, 1979 6. Monnig, J. A., Dale, G. and Bicknell, S. L.: The ileal ureter in recurrent urolithiasis. J Urol, 116: 699, 1976 7. Tanagho, E. A.: A case against incorporation of bowel segments into the closed urinary system. J Urol, 113: 796, 1975 8. Shokeir, A. A. and Ghoneim, M. A.: Further experience with the modified ileal ureter. J Urol, 154: 45, 1995 9. Kato, H., Abol-Enein, H., Igawa, Y., Nishizawa, O. and Ghoneim, M. A.: A case of ileal ureter with proximal antireflux system. Int J Urol, 6: 320, 1999 10. Vatandaslar, F., Reid, R. E., Freed, S. Z., Smey, P., Kogan, S. J., Goldsmith, D. et al: Ileal segment replacement of ureter. I. Effects on kidney of refluxing vs nonrefluxing ileovesical anastomosis. Urology, 23: 549, 1984 11. Fritzsche, P., Skinner, D. G., Craven, J. D., Cahill, P. and Goodwin, W. E.: Long-term radiographic changes of the kidney following the ileal ureter operation. J Urol, 114: 843, 1975 12. Ali-El-Dein, B. and Ghoneim, M. A.: Bridging long ureteral defects using the Yang-Monti principle. J Urol, 169: 1074, 2003 13. Gill, I. S., Savage, S. J., Senagore, A. J. and Sung, G. T.: Laparoscopic ileal ureter. J Urol, 163: 1199, 2000

EDITORIAL COMMENT The authors present long-term results observed in 52 patients with ileal ureters and another 4 with colonic ureters. Their study honors Goodwin et al, who popularized the ileal ureter almost 50 years ago (reference 1 in article). The most intriguing aspect of this study is probably the use of an open, refluxing ileovesical anastomosis. While isotope studies assessing renal function are missing, no gross change in renal function based on serum creatinine could be observed with time, except in 3 patients who had preexisting azotemia. Whether a decrease in renal function in these 3 patients was indeed due to possible reflux or to progressive renal impairment after preoperative renal damage remains an open question. Expectedly also chronic acidosis was almost found only in patients with preexisting impaired renal function, thus, supporting the author recommendation for restrictive patient selection if renal function is impaired.

184

BOWEL FOR URETERAL REPLACEMENT FOR COMPLEX URETERAL RECONSTRUCTION

The absence of overt renal damage despite an open refluxing system in the presence of a contractile bladder is particularly noteworthy. This honors the studies of Mann and Bollman, who first pointed out the antirefluxive properties of isoperistaltic Ileum 75 years ago,1 and of Hinman and Oppenheimer, who 50 years ago noted that the longer the ileal segment, the better the antireflux protection.2 Nevertheless, the favorable results are most likely also supported by the author efforts to maintain sterile urine in these patients. Damage from reflux is more likely if urine is infected or, as shown by McGuire et al almost 25 years ago,3 if voiding pressure exceeds 40 cm H2O. This may become an issue in the future in men with bladder outlet obstruction by due to benign prostatic hyperplasia.

Urs E. Studer Uralagische Universitätsklinik Inspitel, Anna Seiler-Hans Bern, Switzerland 1. Mann, F. C. and Bollman, J. L.: A method for making a satisfactory fistula at any level of the gastrointestinal tract. Ann Surg, 93: 794, 1931 2. Hinman, F., Jr. and Oppenheimer, R.: Functional characteristics of the ileal segment as a valve. J Urol, 80: 448, 1958 3. McGuire, E. J., Woodside, J. R., Borden, T. A. and Weiss, R. M.: Prognostic value of urodynamic testing in myelodysplastic patients. J Urol, 126: 205, 1981