Introduction

The ultimate goal of urologic management of patients with neurogenic bladder dysfunction is not only preservation of the upper urinary tract and renal function, but also prevention of urologic and general complications.1 The introduction and widespread acceptance of clean self intermittent catheterization (CSIC) have simplified long-term management.2 However, there are always some patients unable or unwilling to perform self-catheterization. Quadriplegia, limited dexterity, devastating cognitive impairment or lack of ancillary support are the main causes for failure of intermittent catheterization in patients of both sexes. Women are even more difficult to manage due to lack of available external drainage systems.

One challenge is the establishment of low-pressure urinary drainage without the use of an indwelling urethral or suprapubic catheter. Such drainage may be accomplished by cutaneous diversion. Two major procedures have gained popularity: ileocystostomy, first described by Cordonnier in 1957,3 and ileal conduit (ureteroileostomy) or supravesical diversion, described by Bricker in 19504 and 1956.5 Ileocystostomy is conservative: urine is collected in the bladder. Bladder neck closure must be performed in selected patients, but there is no need for ureteral dissection. The Bricker procedure diverts urine from the bladder regardless of whether or not cystectomy is performed.

Ileocystostomy has been widely used and the functional results are satisfactory. The ileal conduit has been largely reported after pelvic cancer surgery,6,7 but previous reports concerning neurogenic bladder management have not always been enthusiastic because of general complications.8,9,10,11 However, more recently, Bennett12 stressed that the Bricker procedure remained a good procedure in female spinal cord injury patients when necessary.

The objective of our prospective follow-up study was to evaluate outcome and assess early and late complications of cutaneous ileo-ureterostomy in the management of neurogenic bladder for patients who failed CSIC and other methods of bladder drainage.

Materials and methods

Thirty-three patients with neurogenic bladder dysfunction unable to use CSIC, either because of their neurologic disease or lack of motivation, underwent an incontinent cutaneous urinary diversion between March 1979 and March 1999. The technique has been stable throughout the follow-up period and has always been a non-continent ileo-ureterostomy. Neurogenic bladder was due to spinal cord injury (SCI) in 21 patients (14 between C1 and T10, six from T11 to L1, one below L2), multiple sclerosis in four, sacral agenesis in one, and other diseases in four (including various forms of myelitis and cerebral palsy) (Table 1). Mean age of patients at operation was 40.6±15.2 years. Neurourologic disease was 16.1±10.7 years when the surgical decision was made. The mean age of onset of neurological disease was 24.5±1.2 years.

Table 1 Sex and neurological handicap by disease category (complete handicap: confined to a wheelchair or confined to bed; incomplete handicap: patient able to move by himself and walk with or without assistance)

Indications

All patients were incontinent. Surgery was decided for social or functional requests, upper urinary tract protection, or severe perineum alteration (Table 2). Ten men (10 out of 14, 71.5%) had severe refractory perineal problems with skin ulcers and four of these had urethrocutaneous fistula with indwelling catheters. Thirty-seven per cent (7 out of 19) of female patients had an isolated request for cutaneous diversion because self-intermittent catheterization was either very difficult to perform (multiple sclerosis and cerebellar syndrome) or impossible due to wheel chair position for orthopedic reasons (previous hip resection) or tetraplegia. One low-level SCI tetraplegic female patient (C8–D1) also requested diversion as she wished to become pregnant and wanted to avoid recurrent acute pyelonephritis (three to six episodes per year). A total number of 17 patients had previous upper urinary tract dilatation.

Table 2 Indications for incontinent urinary diversion

Twenty patients had urologic complications before diversion was decided. Two patients lost one kidney due to neurogenic bladder. Seven had vesico-ureteral reflux with previous failed surgery and recurrent pyelonephritis. Two had previously undergone ileoenterocystoplasty (11 and 13 years before) and were unable to perform regular SCIC (recurrent fever, infections and hospitalizations). Three men (SCI) had had previous multiple endoscopic sphincterotomies and/or urethrotomies, which failed to eliminate infections and related autonomic dysreflexia (AD). Multiple procedures had been performed in other centers over many years for the four men with recurrent urethrocutaneous fistula either of the perineum or scrotum leading to complete failure and watering pot perineum. Two patients had a previous permanent colostomy for the same reasons as the urinary diversion (perineal skin ulcers). Fifty per cent of patients (16 out of 33) had either indwelling or suprapubic catheters. Condom catheters were used by one male patient (Table 3).

Table 3 Pre-operative status for bladder drainage

Simultaneous cystectomy was decided with the patient and no bladder was removed without the patient's consent. The final choices and the procedures performed are listed in Table 4. Cystectomies were performed at the same time as cutaneous urinary diversion in 14 patients (14 out of 33, 42.4%). Two of them (women) were conversions of diversion ileoenterocystoplasties. One woman had complete abdominal wall repair at the same time using a synthetic prosthesis. She was an obese SCI paraplegic patient and colostomy had been performed for the same indication 8 years previously. One other woman requested to keep her bladder despite a tetraplegic situation related to cervical spinal astrocytoma (neurologically stable for 4 years). At the time of surgery, she was incontinent with intermittent catheterization and had good bladder neck competence. Open cervicotomy was performed to prevent retention of bladder secretions.

Table 4 Procedures performed in the 33 patients and sex distribution

Pre-operative assessment included intravenous urography (IVU), serum creatinine (70.44±47.27 μmol/l, two patients had levels greater than 150 μmol/l with creatinine clearance >50 ml/min) and cystoscopy (to detect stone and tumors). Urine cultures and past urinary tract infections were documented. Antibiotics were prescribed according to culture results. All patients had bacteriuria before the operation and received antibiotics for at least two full days prior to surgery.

Procedure

The cutaneous diversion was performed via a midline incision (after retroperitoneal cystectomy, when requested) and according to Bricker's description.4,5 Both ureters were prepared for a Wallace 1 or 213,14 anastomosis according to presentation and length. Sutures were performed with absorbable suture material. Catheters intubating the ureters were fixed to the new distal ureteral plate and passed through the ileal lumen.

Long-term patient satisfaction was assessed by means of a visual analog scale (0=worse situation than before and high level of regret to 10=high level of satisfaction in terms of quality of life and wishes according to choice of urinary diversion).

Results

All patients were reviewed, with follow-up ranging from 12 to 240 months (mean: 48 months; 14 patients operated on after or during year 1998, seven between 1996 and 1997, 12 before or during year 1995). No patient died, either from surgery or disease. There were no cases of deterioration of renal function, and serum creatinine at last follow-up was 65±53 μmol/l.

All patients were dry after diversion, and perineal skin ulcers were all cured either spontaneously or with specific surgical excision and coverage. All patients (100%) were satisfied with stomal appliances and had a well functioning cutaneous diversion. There were no cases of leakage around the stoma. Two patients (6%) described local problems without any leakage or appliance adaptation problems. One had ulceration around the stoma, which resolved spontaneously (3 years after surgery) and another had cutaneous retraction of the stoma without necrosis of the ileal loop or local problems (obese patient).

There were no cases of ureteral anastomosis or stoma cutaneous stenosis requiring re-operation, and no cases of bowel obstruction or incisional hernia. The overall incidence of non-urological complications was low in this high-risk population with limited treatment options. Among the 17 patients who had a pre-operative hydronephrosis, 10 showed either a decrease or disappearance of dilatation.

A total of 12 patients (12 out of 33, 36%) developed one or more, minor or major, early or late, complications during follow-up (Table 5).

Table 5 Complications of ileal conduit procedure in 33 patients. (18 complications in 12 patients, some patients presented more than one complication)

Early complications included severe life-threatening perioperative infection and septic shock, leading to prolonged hospital stay in the intensive care unit (2 months) and severe pneumonia. All resolved with medical treatment and without renal impairment. One leaking uretero-ileal anastomosis was treated by prolonged ureteral drainage (21 days), without stenosis on IVU 8 years later. The ileal loop was shortened (after 6 years) in one patient because of urine stagnation causing urinary tract infection.

Long-term complications occurred in six out of 33 patients (18.2%). Among the 19 patients in whom the bladder was preserved at the time of the diversion procedure, four (four out of 19, 21%) (three men and one woman) had recurrent pyocystitis leading to three secondary cystectomies. The woman had undergone unsuccessful cervicotomy and requested secondary cystectomy 2 years after diversion. Two men (SCI, paraplegic) also had secondary cystectomies, 6 and 56 months after the initial procedure. Only one had cystectomy without prostatectomy because of local technical reasons. A suprapubic collection (genital secretions) appeared 1 year later, and punctures had to be performed twice a year as an outpatient procedure. Two other men who had undergone primary cystectomy with preservation of the prostate and seminal vesicles (intentionally because of difficult dissection of chronic infected bladder and prostate) complained of chronic urethral leakage. However, erections were preserved and patients were able to maintain sexual intercourse. At latest follow-up, 48% of patients (16 out of 33) have been able to keep their bladder without any related complications.

Among other long-term complications, acute pyelonephritis (four out of 33) had no major consequences, and IVU of the upper urinary tract and ileal loop were normal. Patient satisfaction was evaluated on a visual analog scale: 9.1±2.8. No patient regretted his or her choice as they were all dry, skin ulcers had resolved and catheters or pads were no longer required. Patients with pyocystitis were less satisfied because of recurrent problems or re-operations.

Discussion

Preservation of renal function is the main long-term goal of management of neurogenic bladder dysfunction. Low-pressure urinary storage and appropriate drainage system are the two key-points. Over the last 20 years, drug therapy2 and surgical improvements15 have been able to decrease bladder pressure and increase bladder capacity and compliance in neurogenic bladders. However because of striated sphincter dyssynergia, emptying must be ensured by clean self-intermittent catheterization.

Unfortunately, a few patients are unable or unwilling to perform self-catheterization. Complications include urethral erosions, bladder neck dilatation, urethro-cutaneous fistula, stones, vesicoureteral reflux, infection and sepsis, and impaired renal function.1,2 When wet perineum or decubitus skin ulcers are present, impairment of quality of life, and general or life-threatening complications can occur. For most of these patients, urinary diversion is still the final solution in order to achieve perineal dryness and stable upper urinary tract function.

We reviewed our experience with these types of neurogenic patients and the use of a non-continent diversion procedure (ileo-ureterostomy). In view of the literature, and including all minor and major urologic and general complications, we achieved a low complication rate (36%) for this population. No early re-operation for complications related to surgery was performed, and perfect local stomal management and perineal dryness were obtained in 100% of cases. We achieved the main objective of renal function protection and patient satisfaction. It has now been clearly established that long-term indwelling urethral or suprapubic catheters lead to serious morbidity (16 out of 33 in this series) and should be avoided.2,16,17 Moreover, long-term indwelling catheter can lead to bladder cancer.18 Catheters should be removed before the onset of irreversible complications. Some patients do not agree for SCIC or sphincterotomy (men) despite continuous improvements of appliances. They must be clearly informed about the risks of high pressure bladders and the consequences of long-term indwelling catheter. Another group of patients is unable to perform SCIC: it is the medical and especially the urology team's work to inform them about available choices for urinary tract protection and improvement of everyday life. Once major complications have occurred, there is no longer any room for discussion and a majority of these patients will require incontinent urinary diversion.

According to our data, recently published long-term results of such diversions are very satisfactory in terms of renal function preservation.1,10,19,20,21 Past reports on the Bricker procedure were not so enthusiastic,11,22 but good renal function protection was masked by high complication rates due to the poorer quality of peri-operative and post-operative care and to high-risk patients with severe pre-operative renal impairment. Mänsson,23 in his study on renal function after urinary diversion, showed that the functional outcome was neither related to the method of diversion nor to the type of ureteral anastomosis (anti-reflux or not) for patients with cutaneous incontinent diversion. We recommend urinary diversion early before irreversible renal complications in patients in whom no other modality of incontinence management is available. Information, strict follow-up, specific multidisciplinary neuro-urological care units must strive to achieve this objective. Regarding these results, the rate of patients using this diversion is increasing with years in our neuro-urology unit. Regardless of level of spinal cord injury or type of neurologic disease, it can be explained both by age of onset of neurological disease and by a high request of quality of life for these patients.

By diverting urine from the bladder, supravesical diversion can achieve complete perineal protection (as in our series). Ileovesicostomy1,19,21 avoids bladder management problems, but still requires bladder neck and urethral closure. Pubovaginal slings have been used1,19,24 and chosen because of the documented incidence of post-operative fistula information after transabdominal or transvaginal bladder neck closure. More recently, Moreno et al.25 and Andrews26 stated that one of the most challenging problems in the management of women with neurogenic vesical dysfunction is destroyed urethra secondary to chronic urethral catheterization.

The issue of whether cystectomy should be performed at the same time as urinary diversion is still controversial. The risk of significant pyocystitis can cause serious future complications. During follow-up, only 37% of women (seven out of 19) required cystectomies (either primary or secondary) versus 71.4% of men (10 out of 14). The decision to perform cystectomy at the time of diversion is a difficult one. After careful information of the patients, we selected patients for cystectomy with consideration of bladder risks (retention, carcinoma), quality of life, and nursing and perioperative risks (bleeding, duration of surgery). We made the wrong decision in three cases due to the patient's initial request to preserve the bladder. However, about 50% (16 out of 33) of our patients preserved their bladder without any problem.

This Bricker-like procedure has had a poor reputation on the basis of reports published more than 25 years ago, mostly concerning children operated on for urologic malformations and upper urinary tract dilatation. The analysis by Shapiro et al.,9 based on 90 children, reported good results in terms of renal preservation, but a significant number of long-term complications. However, patients in this series had very poor renal function condition. Kambouris,10 at the same time, was more optimistic, despite a high complication rate related to end-stage patients. Our experience led us to believe that the adult ileal conduit is a useful technique in bladder management of neurogenic bladder patients. In 1995, Bennett12 reported more encouraging results on women in the same indication. Avoiding pyocystitis by means of good bladder management led to no major sepsis being observed during follow-up. Stomal stenosis did not occur and the rate of this complication should not be different from that of other ileal diversions. Stomal stenosis rates as high as 15%19 are probably more specific to ileovesicostomy (fascial stenosis). A wide-mouthed vesical anastomosis, a pre-marked stomal placement, and an adequate diameter of the fascial window are the main points of local success.21

There were no instances of electrolyte disturbances, either in the immediate post-operative or late follow-up. This is probably due to the short segment used as a conduit and to the wide stoma that prevents urinary stasis and re-absorption. The shortest segment was always used, although length must be adapted to the quality of the blood supply to avoid very short segment ischemia. We did not observe any case of necrosis of the ileal segment, as has been reported when a very short segment of ileum was used.7

Conclusion

We conclude that the ileo-ureterostomy procedure, based on Bricker's report, is still27 a safe and well-tolerated procedure in neurologically impaired patients. This procedure is suitable for most of these end-stage patients with poor lower urinary tract conditions or as a definitive choice of patients to manage their neurogenic bladder dysfunction. This technique must still be part of armentarium.