Long-term outcome of ureterosigmoidostomy: an analysis of patients with >10 years of follow-up


Igor Frank, Department of Urology, Mayo Clinic, 200 First Street Southwest, Rochester, Minnesota 55905, USA.
e-mail: frank.igor@mayo.edu


Study Type – Therapy (case series)

Level of Evidence 4


To examine our long-term experience with ureterosigmoidostomy (USS) to evaluate its potential applicability in the treatment of benign and malignant conditions of the urinary bladder, as USS has been largely disregarded recently, secondary to concerns of long-term complications, but has had a resurgence of interest due to its potential applicability to newer minimally invasive surgical techniques.


We identified 51 patients who had USS from 1956 to 2006 at our institution and with >10 years of follow-up. The patients were followed retrospectively by a chart review. Patient data were analysed in a multifaceted fashion, paying particular attention to metabolic abnormalities, early (≤30 days) and late (>30 days) complication rates, continence rates, imaging changes, and the rate of repeat surgical intervention.


The median (range) follow-up was 15.7 (10.0–45.4) years and the median age at surgery was 58.8 (0.4–79.0) years; 40 (79%) patients had the procedure for malignancy and 11 (22%) for benign disease. Six patients (12%) had at least one early complication, including one wound dehiscence and one pulmonary embolus. In all, 22 patients (43%) had at least one late complication, with anastomotic stricture being the most common (11/51, 22%). This was followed by recurrent pyelonephritis in eight patients (16%), stones in five (10%), chronic renal insufficiency in three (6%) and severe intractable acidosis in two (4%). A repeat surgical intervention was required in 19 (37%) patients. In all, 94% (48) reported complete continence. No patient developed colonic malignancy during the course of this study.


USS is associated with long-term complications. While this complication rate might not be acceptable for all patients, some might be willing to undergo the procedure as the primary method of urinary diversion. When designing newer minimally invasive techniques for the treatment of benign and malignant conditions of the bladder, consideration could be given to USS as a form of urinary diversion in highly selected patients.




Despite being the most common form of urinary diversion for many decades, ureterosigmoidostomy (USS) has fallen out of favour due to concerns of infection, metabolic abnormalities, incontinence and secondary malignancy. Furthermore, the development of orthotopic urinary diversion has provided patients with an acceptable body image and the preservation of volitional voiding. Nonetheless, due to the technical simplicity of USS, particularly in patients for whom the use of external appliances and catheters is culturally or personally unacceptable, interest in USS has persisted. The development of minimally invasive surgical techniques has encouraged some to revisit the procedure.

Simon [1] first described the diversion of urine into the rectum in 1852, and despite the fact that the patient did not survive, it remained the primary method of urinary diversion until Bricker [2] described the ileal conduit nearly a century later. USS developed over the initial century from a procedure that was essentially the creation of a needle-induced fistula from the ureter to the rectum, to a more refined anastomosis. Goodwin et al.[3] and Leadbetter [4] described an antirefluxing submucosal tunnelled anastomosis that represented improvements to the previous operation. These revisions, among others, significantly reduced the incidence of obstruction and ascending pyelonephritis. USS remained the procedure of choice until the late 1950s, when concerns of secondary malignancy at the uretero-intestinal anastomosis began to be recognized. This, coupled with the advances in enterostomal appliances developed in the 1950s, permitted the ileal conduit to become the predominant form of urinary diversion in the USA for several decades.

The last three decades have seen an unprecedented development of techniques to reconstruct the lower urinary tract, with the goals of maintaining control over urine storage and elimination, preserving renal function and minimizing the impact on the quality of life of patients requiring urinary diversion. Continent cutaneous diversions have been promoted by some, but have not become widespread due to problems in duplicating a durable continence mechanism. It is really the orthotopic neobladder that has revolutionized the field and, arguably, represents the current ‘gold standard’ urinary diversion.

Nonetheless, interest has been revived in USS due to the widespread growth in minimally invasive surgery. Technically, it is a much simpler diversion to construct in a laparoscopic or robotic setting. Furthermore, USS precludes the need for significant bowel manipulation and its associated complications. Proponents of the procedure argue that minimization of stomal and bowel-related complications might attract some patients and their surgeons to revisit the procedure. However, few studies have been reported on the long-term outcome of these patients within the last several decades, to assess their likelihood of further complications. Therefore, we reviewed our experience in the long term follow-up of patients who had undergone USS for benign and malignant conditions of the urinary bladder.


We identified 245 patients who had a cystectomy and USS for benign or malignant disease at our institution from 1956 to 2006; 51 of these patients survived their initial disease and had >10 years of follow-up and were included in the study. Data were then obtained by a retrospective analysis of their medical records. The criteria for creating a USS were competence of the external anal sphincter, a serum creatinine level of <2.0 mg/dL, no previous or planned pelvic radiotherapy and no sigmoid diverticulosis, polyposis or other sigmoid pathology. All patients had instillation of a 300–400 mL enema for preoperative evaluation of the external anal sphincter.

The standard operative technique developed over the course of the study, but most patients had a standard cystectomy. Ureters were then implanted in an antimesenteric fashion along the taenia of the sigmoid colon, typically with an unstented, tunnelled ureterosigmoid anastomosis.

Patient data were analysed in a multifaceted fashion, paying particular attention to metabolic abnormalities, early (≤30 days) and late (>30 days) complication rates, continence rates, incidence of pyelonephritis, and the rate of repeat surgical intervention. Continence was defined as being totally dry (or using no pads) at the time of the last interview with a physician; incontinence was further divided into daytime, night-time and total incontinence.


The median (range) age at time of surgery in the 51 patients included in the study was 58.8 (0.4–79.9) years; 31 (61%) were male (Table 1). The median follow-up interval was 15.7 (10.0–45.4) years. Forty (78%) patients had USS for malignant disease; 30 had urothelial carcinoma, five had squamous cell carcinoma of the bladder, four had colorectal tumours invading the bladder, one had an adenocarcinoma of the bladder, and 11 (22%) had USS for benign disease (five with bladder exstrophy/epispadias, four with interstitial cystitis, and two with severe vesicovaginal fistulae).

Table 1.  The patient characteristics and long-term complications
CharacteristicMedian (range) or n (%)
Age at surgery, years58.8 (0.4–79.7)
 Male31 (61)
 Female20 (39.2)
Follow-up, years15.7 (10.0–45.4)
Indication for cystectomy 
 Malignant tumour40 (78)
 Benign disease 11 (22)
Alive at last follow-up 
 Yes17 (33)
 No34 (67)
Late complications (>30 days)22 (43)
 Anastomotic stricture 11 (22)
 Recurrent pyelonephritis 8 (16)
 Calculi 5 (10)
 Chronic renal insufficiency 3 (6)
 Acidosis refractory to oral medications 2 (4)
 Conversion to ileal conduit 3 (6)
Repeat surgical intervention 
 Yes19 (37)
 No32 (63)
Treatment for metabolic acidosis 
 Oral agents32 (63)
 None19 (37)
 Total47 (92)
 Nocturnal incontinence 4 (8)
 Daytime incontinence 2 (4)
Endoscopic surveillance 
 Yes36 (71)
 No15 (29)
Time from cystectomy, years15.6
 Adenoma (low or moderate dysplasia) 11 (31)
 Invasive colon adenocarcinoma 0

Early surgical complications (≤30 days) were identified in six patients; this included two who developed acute renal failure, requiring short-term dialysis, one who had a pulmonary embolus, one a wound dehiscence and two who developed superficial wound infections. Late surgical complications are also listed in Table 1; 19 (37%) patients had a second surgical procedure during the follow-up, but only three (6%) were converted to an alternative form of urinary diversion. Most (32/51, 63%) were prescribed oral medications for managing metabolic acidosis; two were refractory to this treatment and were converted to an ileal conduit.

Overall, patient continence was excellent, with 92% of patients reporting total continence, 8% reporting some night-time incontinence and 4% reporting day-time incontinence.

Endoscopic tumour surveillance was used in 36 (71%) of the patients; of the remaining 15, 10 had serious comorbid conditions and the treating physicians felt that colorectal screening was unnecessary. Five patients died in the early to mid-1970s, before screening protocols were widely used at our institution. In the 36 patients who were screened, they underwent the procedure at a median of 13.6 years after USS. No patients developed an invasive colorectal carcinoma; 11 developed tubular adenomas with low or moderate degrees of dysplasia. Of these 11 tubular adenomas, only three were within the sigmoid colon. The remainder of the adenomas were within either the ascending, transverse or descending colon proximal to the sigmoid anastomosis.


USS as a method of urinary diversion was largely abandoned by modern urological surgeons. However, as novel methods to divert urine become more commonplace, it is important to understand the history of the science of urinary diversion, to understand its future. USS is a durable technique that provides some patients with excellent continence and function while eliminating the need for external appliances or intermittent catheterization. Clearly, most patients would be better served with orthotopic bladder reconstruction. However, there might be a selected group of patients who are adequately served with a USS. Anecdotal evidence suggests that most of these patients are very satisfied with their method of diversion and are extremely reluctant to undergo conversion to alternative methods.

As expected, these patients had a higher rate of some complications than series of similar patients undergoing ileal conduit diversion. The incidence of uretero-intestinal stricture, at 22%, was high. The incidence of stricture formation in ileal conduit series over a similar period was typically 2–18%[5–8]. The higher rate of stricture is not completely unexpected, as most of these anastomoses were constructed with an antirefluxing mechanism to prevent large-volume reflux of enteric bacteria. This is clearly not as important in diversions that do not directly merge with the fecal stream. However, reflux of infected urine did not appear to be a large problem, as the incidence of clinically significant pyelonephritis was similar to those in previous studies in patients undergoing ileal conduit diversions. Furthermore, despite the length of follow-up, relatively few patients (three, 6%) developed chronic renal insufficiency.

A large proportion of the present patients (10%) developed a significant stone burden either within the rectal vault or in the kidney. This incidence is higher than the 3–5% incidence in patients undergoing ileal conduit. While a full stone-composition analysis was not available for these stones, presumably these stones formed as a result of chronic bacteriuria and metabolic acidosis, both known factors that contribute to stone disease.

Metabolic complications represent another major area of concern after urinary diversion, particularly in patients with a continent urinary diversion, as prolonged storage times can lead to increased absorption of harmful metabolites. Acidosis develops from an excess reabsorption of ammonium chloride across the intestinal mucosa. In the present series, most patients (32/51, 63%) were found to have a mild acidosis and were treated with oral agents. Only two of the 51 patients were refractory to this treatment and had an alternative form of urinary diversion for managing their metabolic abnormality. Severe acidosis typically presents in the setting of severe dehydration and should be treated with vigorous hydration, alkalinization and repletion of potassium, as the total-body potassium is often depleted in these patients and acute hydration without potassium can lead to life-threatening hypokalaemia.

While some of the aforementioned complications were more common, importantly, no patients in this series had bowel complications due to their diversion. Specifically, due to the method of diversion, no patient developed stomal stenosis, parastomal hernias or intestinal obstruction. These bowel-related complications are a significant morbidity in long-term series of patients undergoing other methods of diversion.

Continence is a critically important measure after USS, as incontinence can be particularly devastating. Fortunately, the vast majority of patients reported complete continence. Similar to other continent urinary diversions, night-time continence was slightly lower than day-time continence and required the use of pads or diapers at night in three patients. Modifications to the standard USS have been reported (such as the Mainz II) and they might serve to decrease pressure in the rectal vault and improve continence [9–11].

Tumour formation at the site of uretero-intestinal anastomosis is the most feared complication after USS. There is a general consensus that patients with a USS are at increased risk for the subsequent development of tumours within the colon. It is estimated that these patients are 100–200 times more likely to develop sigmoid tumours [12], with an overall incidence of ≈24% at 20 years [13,14]. Most of these tumours are adenocarcinomas, with the remainder consisting of adenomas and other benign lesions [15]. The earliest reported case of carcinogenesis after USS occurred 10 years after diversion [16]. The mean latent period after USS for the development of adenomas is 19.8 years and the mean time to the development of adenocarcinoma is up to 26 years [17], suggesting that the time of degeneration from adenoma to adenocarcinoma takes ≈6 years in these patients.

Notably, no patients in the study developed invasive adenocarcinoma, and although 30% of patients who had endoscopy had adenomas, only three adenomas were within the sigmoid colon. The reason that no patient in our series developed invasive disease is unclear, but might be related to the length of follow-up as well as the patient population studied. Most of the present patients had USS for malignancy and only 8% had USS in childhood for bladder exstrophy/epispadias complex. Furthermore, because of the long delay in development, a low rate of secondary carcinogenesis might be acceptable for many patients undergoing cystectomy for malignant disease, as they could already have a limited life-expectancy. Nonetheless, we recommend colonoscopic surveillance in patients with a USS, beginning 10 years after the diversion and continuing every 2 years thereafter.

Importantly, the current study has limitations; the study population was a group of single-institutional, retrospective patients who survived for >10 years after cystectomy and, as such, selected for patients with satisfactory functional and oncological outcomes. However, these studies remain important, as most patients strive to survive their initial disease and have long-term expectations. Also, the present patients were operated upon over the course of nearly 50 years by several different surgeons. While operative technique and perioperative care have developed over this interval, we feel that the diversity in surgical technique represents the durability of the procedure in different hands.

In summary, USS is a definitive form of urinary diversion. In a long-term follow-up, it has a complication profile similar to that reported for other methods of urinary diversion over the same interval. While the incidence of anastomotic stricture and urinary calculi might be somewhat greater, there have been no bowel- or stoma-related complications. We feel that urologists should be familiar with USS, as it might be amenable to minimally invasive surgical techniques in highly selected patients who are appropriately informed of the inherent risks.

In conclusion, USS is associated with long-term surgical complications that are comparable with the complication rates of other forms of urinary diversion with similar follow-up intervals. Furthermore, the vast majority of patients are continent after USS, even when performed without the Mainz modification, and are very satisfied with their form of diversion. Finally, USS is a relatively straightforward procedure that might be amenable to minimally invasive techniques in highly selected patients who are informed of the higher risk of complications.


None declared.