Introduction

After the failure of all conservative treatment options in patients with neurogenic bladders, surgical intervention should be considered. The options for urinary diversion are discussed in part I. For patients with chronic renal failure, severely dilated upper urinary tracts with deterioration of renal function and those who are unable to perform self-catheterization of a continent stoma, the conduit diversion is the therapy of choice.

In 1937 Seifert described an incontinent urinary diversion performed by a jejunum conduit, and Bricker popularized the use of the ileal conduit in 1950 [1, 2]. Early results were promising, and the procedure was also commonly used as supravesical urinary diversion in children. However, the number of reported complications increased with the duration of follow-up. Deterioration of the upper urinary tract and renal function, calculi formation and stoma stenosis resulted in a late complication rate between 19–86% [3, 4, 5, 6, 7]. In the 1970s, Hendren and Middleton concluded that this operation should be abandoned in children or any patient with potential longevity [5, 8]. The alternative form of incontinent urinary diversion is the colonic conduit, introduced by Übelhör in 1952 and popularized by Mogg in the early 1960s [9, 10]. Elder and co-workers suggested that it had the same fate as the ileal conduit in the long run [11]. However, they used the conduit anisoperistaltically and refluxively. Since 1968, the isoperistaltic antirefluxive colonic conduit has been performed at our institution with good long-term results [12].

In this third part of a retrospective study, we analyzed whether the use of the colonic conduit was safe in the long run in this difficult group of patients with neurogenic bladder.

Materials and methods

Between 1968 and 2002, urinary diversion was performed in 170 children and adolescents with a neurogenic bladder under the age of 20 years. Up to 1983 the colonic conduit (Fig. 1) was the standard form of diversion [12] and was later used in patients with chronic renal failure with severely dilated upper urinary tracts with deterioration of the renal function and those unable to perform a CISC through a continent stoma. It was performed in a total of 88/170 patients, of whom 11 were later converted to a continent diversion, so that these patients are no longer included in the follow-up of colonic conduit patients. Of the remaining 77 patients, 1 was a conversion from an ileal conduit, 3 from ureterocutaneostomy and 1 from loop-ureterocutaneostomy; all were performed in other institutions.

Fig. 1
figure 1

For construction of the colon conduit a sigmoid segment of 10 to 12 cm length was isolated (a) and the ureters were reimplanted into that segment. After closure of the conduit and re-establishment of bowel continuity, the cranial part of the conduit was retroperitonealised by closure of the peritoneal incision. The stoma was placed in a laterocolic position in the lower left abdominal quadrant (b)

Full size image

The data from the follow-up studies were obtained from the charts in 22 of the 56 patients. All these patients were seen annually at our institutions. Follow-up visits included detailed histories concerning complications of the conduit diversion. Blood chemistry included serum creatinine, the electrolytes and a capillary blood gas analysis. An ultrasound of the urinary tract was performed. In the case of dilatation of the upper urinary tract, a MAG-III clearance was performed. In the remaining 34 patients, data were obtained by telephone interview with the patient or the parents and the outside urologist, who performs the ultrasound and the laboratory investigations at least on an annually basis.

The latest intravenous urography or a renal ultrasound were compared to the preoperative study, and dilatation was classified into three degrees: none, mild to moderate and severe. The renal function was assessed by serum creatinine levels (normal for adolescents: >120 µmol/l). The acid-base balance was determined from capillary blood gas analysis. At base excess (BE) below −2.5 mmol/l, alkali-substitution (e.g., sodium/potassium citrate) was recommended (normal range: between +2.5 and −2.5 mmol/l).

Results

Of the 77 patients with colonic conduit diversion, 21 patients were not available for follow-up: 11 were deceased and 10 were lost to follow-up. Three of the deaths were related to nephrological complications in patients who already had an impaired renal function before conduit diversion. The mean age at the time of operation was 10.1 years (0.8–20; median 9.4). An average follow-up of 21.8 years (2–32.7; median 23.8 years) was available in 56 patients with 99 RUs (6 solitary kidneys and 7 nephrectomies during the follow-up period). A total of 21 patients (37.5%) developed 36 complications requiring re-operation in 18 patients (32%).

Surgical complications

Five early complications were encountered in four patients. Two developed adhesive ileus, and one required nephrectomy because of pyonephrosis after dislocation of the ureteral splint postoperatively. One further patient developed adhesive ileus 5 days after surgical revision of an early stenosis of the ureterocolic anastomosis. One woman developed adhesive ileus 3 years postoperatively.

Upper urinary tract

At the latest follow-up, 34/99 RUs had an improved upper tract drainage compared to preoperatively, 63 were unchanged and 2 preoperatively normal RUs had moderate dilatation. Stenosis of the uretero-intestinal implantation developed in six RUs (6%) of five patients after a mean of 6.6 years (0.25–11). Four RUs were successfully reimplanted, and balloon dilatation was carried out in two. Nephrectomy of seven RUs became necessary in six patients. In two patients nephrectomy was performed because of pyonephrosis/renal abscess (postoperative n =1, after stone surgery n =1), and in four patients five non-functioning kidneys were removed after recurrent pyelonephritis. One girl is now on haemodialysis. Recurrent pyelonephritis during follow-up was encountered in an additional six patients who already had experienced pyelonephritis before conduit diversion. One patient required an Anderson-Hynes pyeloplasty due to a secondary UPJ obstruction 5 years postoperatively.

Calculi developed in eight RUs (8%) after a mean of 7.6 years (range, 3–14). Open stone surgery was performed in five patients before the era of extracorporal shock wave lithotripsy (ESWL), while two others underwent ESWL. One of the patients developed a renal abscess after open stone surgery requiring nephrectomy. One calculus was removed endoscopically through conduitoscopy.

Stoma complications

Conduit revision was necessary in nine patients (16%) after a mean follow-up of 11.9 years (range, 5–20). Conduit elongation was corrected in two and a stoma stenosis in seven patients, in one of them twice.

Renal function

Serum creatinine was elevated preoperatively in 2 of the 56 patients. In both of them, the renal function deteriorated gradually requiring haemodialysis; in one of them, both kidneys were removed after recurrent pyelonephritis. No patient with a preoperative normal serum creatinine developed renal failure.

Acid-base balance

During follow-up the base excess dropped temporarily below −2.5 mmol/l in 6 of the 56 patients followed. None of them developed clinical acidosis or a decrease in blood pH and serum electrolyte disturbances. Of the patients with colonic conduit, 4/56 used prophylactic alkali substitution at the latest follow-up.

Secondary malignancy

None of the 56 patients developed a bowel neoplasm.

Discussion

The primary goal of urinary diversion is to protect the upper urinary tract over an average life span and to enable the patient to lead a normal life. This is especially important in children. In 1965, Mogg published promising results of the colonic conduit [10]. However, after an average follow-up of 13.2 years, Elder et al. reported, “of the 26 surviving children there were only 7 who could be considered to have a good long-term result from the operation”. They concluded that in children the use of colon showed no advantage compared to the ileal conduit [11]. Mogg used the colon segment anisoperistaltic and performed a refluxing ureter implantation. Using an isopersitaltic colonic conduit with an antirefluxive ureter implantation, we reported a good result after an average follow-up period of 16.3 years (range: 5–26 years) [12].

Richie and co-workers compared the classic refluxing ileal conduit with the non-refluxive colonic conduit in a series of dogs [13]. An ileal loop was created to divert one ureter and a colon loop to divert the other. Reflux was present in all but one ileal diversion, and in none of the colonic conduits. Emptying times of the two conduits were the same, as were resting pressures within the bowel segments. Peristalsis in the ileal conduit occurred six times per minute, whereas in the colon segment peristalsis was measured at one wave per minute with lower pressures. When the animals were investigated 3 months postoperatively, pyelonephritic changes were found in 83% of the kidneys diverted by ileal conduit in contrast to only 7% in those diverted by non-refluxing colonic conduit. These experimental findings were supported by clinical results. Pyelonephritis was observed in 7–10% of the antirefluxive colonic conduits and in 29% of the refluxive colonic conduits [3, 11]. In children with the refluxing ileal conduit acute pyelonephritis was observed in 16–22% [5, 6]. Until now, there has been no randomized study comparing ileal und colonic conduits in children. However, the indication for conduit diversion in this age is rare. The results cannot be compared to those of adults, because the length of the used bowel segment in children is increasing; the problems of conduit elongation are clearly shown in the literature [5, 6]. Due to the better results obtained with the colonic conduit in children, the authors think that a randomized study is not justified.

Recently, the necessity of antirefluxive ureteral implantation has been discussed [14, 15, 16, 17]. Clearly, there is a need for a prospective randomized study studying the impact of reflux from a potentially contaminated reservoir. Until evidence is obtained that even a low pressure reflux does not harm the kidneys in the long run, in young patients with a long life expectancy, the kidneys should be protected from developing renal scars by an appropriate antireflux technique.

The incidence of ureteroileal stenosis in children with ileal loop urinary diversion rises to up to 22.3% of the cases [6]. The rate of ureterocolic stenosis in 70 patients with non-refluxing colonic conduits reported by the Massachusetts General Hospital was 8.5%, although 75% of the patients were followed for less than 3 years [3]. In a series with 25 children, 7 of 50 preoperative normal ureters (14%) developed a stenosis at the ureterocolic anastomosis. The anastomosis was performed as described by Leadbetter and Clark [18]. After a follow-up of 21.8 years, we observed in this group of patients a stenosis of the ureterocolic implantation site in 8 of 99 ureters after 5.7 years (0.25–11). However, it is difficult to compare ileal and colonic conduits because of the various types of bowel tissue and ureteral anastomosis used.

Stenosis of the stoma, one of the late complications, occurred in children with ileal conduit in up to 42% [5, 6, 7]. We encountered stoma stenosis in 16% of the patients 11.9 years (5–20) postoperatively.

In puberty, body image and freedom of lifestyle become important issues and eliminating the external collection device represents a considerable advantage. After the recovery of renal or upper urinary tract function, young adults often desire continent diversion. In 1975, Hendren reported nine patients with a non-refluxing colonic conduit, satisfactory upper tracts and normal rectal control in whom the conduit was later joined to the colon [8]. In 11 of our 88 patients conversion was performed. In three patients with intact urethral sphincter, conversion to a bladder substitution was performed and in eight patients the colonic conduit was integrated into an ileocecal pouch (Fig. 2) (parts I and II). The main indication for urinary conversion is the patient’s wish for continence. However, every new surgical intervention is accompanied by a high risk of complications and failures. Consequently, the risks and advantages have to be realistically explained to the patients. It should be kept in mind that the conduit allows for the setting up of an own family and thus 3 of the 27 women in our collective who were older than 18 years of age delivered 6 children without any complications.

Fig. 2
figure 2

For conversion to continent cutaneous diversion, 8–10 cm of the ascending colon including the ileocecal valve and 8–10 cm of the terminal ileum were isolated (a). The conduit was isolated and opened from the distal end and integrated into the pouch, leaving the ureters in place (b)

Full size image

Hyperchloremic metabolic acidosis, the most common electrolyte abnormality seen in intestinal urinary diversion [19], was not observed in our series of patients due to prophylactic correction of a negative base excess (e.g., BE <−2.5 mmol/l). Secondary malignancies represent a further important late complication of intestinal urinary diversion [20], for which screening of all patients after more than 5 years of diversion is recommended. Up to now, none were found in our series of 56 patients with a colonic conduit after a median follow-up of 23.8 years.

Conclusion

In patients with neurogenic bladders and failure of conservative treatment who are unable to perform a CISC through the urethra or a continent stoma, or those who have a severe deterioration of the upper urinary tract or chronic renal failure, the colonic conduit offers a reliable urinary diversion with preservation of the upper urinary tract in the long run. Our results are clearly superior to those long-term results of the ileal conduit diversion in children reported in the literature. Whenever incontinent urinary diversion is required in children and adolescents, a colonic conduit seems to be preferable to an ileal conduit in our opinion.