The Department of Urology in Mansoura has a well-known experience in, among many things, urinary tract reconstruction in patients with bladder cancer. They review their results in 338 male patients who had a radical cystectomy and Kock pouch. They found good functional and oncological outcomes in properly selected patients. However, they also drew attention to several valve-related complications.
To evaluate our experience with men who underwent radical cystectomy and urethral Kock pouch construction between January 1986 and January 1996.
PATIENTS AND METHODS
Complications were classified as early (within the first 3 months after surgery) or late. Continence was assessed by interviewing the patient; they were considered continent if they were completely dry with no need of protection by pads, condom catheter or medication. The patients were followed oncologically and Kaplan-Meier survival curves constructed. Urodynamic studies were used to define the possible causes of enuresis.
Three patients died after surgery from pulmonary embolism. There were 67 early complications in 63 patients. The mean (sd) follow-up was 87.8 (49.1) months. There were 111 treatment failures from cancer; of these, four men only had an isolated local recurrence in the urethra. Late complications included 72 pouch stones in 55 patients, and 36 deteriorated renal units caused by reflux (17), uretero-ileal stricture (11), nipple valve eversion (four) or stenosis (four). Interestingly, 65 renal units that were dilated before surgery improved significantly afterward. Ileo-urethral strictures occurred in seven men and anterior urethral strictures in six. Nine patients were totally incontinent and two had chronic urinary retention. Daytime continence was complete in 94% of men, with nocturnal enuresis in 55; the latter had significantly more residual urine, and a higher amplitude and duration of phasic contractions.
Orthotopic bladder substitution after cystectomy for cancer is feasible, with good functional and oncological outcomes in properly selected patients. Nevertheless, the use of a hemi-Kock pouch is associated with many valve-related complications.
Radical cystectomy and orthotopic diversion are becoming the standard treatment for muscle-infiltrating bladder carcinoma. One of the earliest techniques used for urethrally controlled bladder substitutes was the hemi-Kock pouch [1,2]. The procedure relies on creating a low-pressure bladder substitute and including an intussuscepted nipple valve to prevent reflux. Our preliminary experience and critical analysis of the initial results were the subject of several reports [1,3–5]. In the present study the long-term outcome is addressed from both oncological and functional perspectives.
PATIENTS AND METHODS
Between January 1986 and January 1996, 1569 patients underwent radical cystectomy for bladder cancer at our institution. The different methods of urinary diversion used are shown in Table 1. Of these, 338 men and 15 women were provided with a urethral hemi-Kock pouch (mean age 48.7 years, sd 8.9, range 28–71). Generally, patients were fit or rendered fit enough for prolonged surgery. Synthetic liver function was adequate, with a plasma albumin of > 30 g/L and prothrombin concentration of > 75%. Serum creatinine levels were required to be in the normal range; in patients with high serum creatinine levels, a creatinine clearance of < 50 mL/min was considered a contraindication. Men with prostatic urethral involvement were excluded. The type of primary bladder cancer had no effect on patient selection for orthotopic bladder substitution. Women with no tumour extension to the bladder neck region or anterior vaginal wall were suitable candidates for the procedure.
|Diversion type||N (%)|
|Ileal loop conduit||523 (33.3)|
|Kock pouch||353 (22.5)|
|Continent cutaneous diversion||19 (1.2)|
|Rectal diversion||421 (26.7)|
|Early complications (urethral Kock pouch)|
|Ileo-urethral urine leak||26 (7.7)|
|Uretero-ileal urine leak||8 (2.4)|
|Prolonged bowel ileus||4 (1.2)|
|Upper gastrointestinal bleeding||7 (2.1)|
|Wound sepsis||2 (0.6)|
|Deep venous thrombosis||4 (1.2)|
|Pelvic collections||10 (3.0)|
|Distribution of treatment failures [% of failures]|
|local recurrence||59  (17.5)|
|Distant metastasis||38  (11.2)|
|Local and distant metastasis||10  (3.0)|
|Isolated urethral recurrence||4  (1.2)|
|Late complications, N|
|Nipple valve stenosis||2|
The surgery comprised a standard radical cystoprostatectomy, including pelvic lymph node dissection. The final stages of the operation were adjusted to prepare an adequate urethral stump. Two patients had nephroureterectomy in conjunction with cystectomy. Two more patients had a solitary kidney. A frozen-section biopsy of the distal prostatic urethra was necessary for proper patient selection in the presence of carcinoma in situ and/or multifocal tumours.
The urethral Kock pouch was constructed according to the technique previously described . A 45–50 cm long intestinal segment was isolated from the distal ileum ≈ 20 cm proximal to the ileocaecal valve. The distal two-thirds were used to construct the reservoir by detubularization and double-folding. The proximal third of the isolated segment was preserved to construct the antireflux nipple valve. A stented end-to-side anastomosis was made between the spatulated ends of the ureters and the inlet of the pouch. Two tube drains were placed in the pelvic cavity; gravity drainage only was used. Prophylactic antibiotics were administered routinely. The tube drains were removed once fluid drainage had ceased and the urethral catheter retained for 3 weeks.
Patients were followed up every 3 months during the first year, and 6-monthly thereafter. The follow-up included a history, physical examination, urine culture and blood chemistry studies. The radiological evaluation consisted of renal ultrasonography, IVU and an ascending study. CT and bone scans were used as clinically indicated. Continence was assessed by interviewing the patients at each follow-up visit; they were considered to be continent if they were completely dry during the day and night, with no need for protection by pads, condom catheter or medication.
A urodynamic investigation was undertaken in 88 patients in an attempt to define the cause(s) of enuresis, comprising uroflowmetry, medium-fill pouchometry (20 mL/min) and urethral pressure profilometry. Enterocystometry was undertaken with the patient supine, using normal saline at room temperature (UD 5500, Dantec, Denmark). The standards and terminology of the ICS were adopted  and the detailed methods used described previously .
Student's t-test was used to compare parametric data, the Mann–Whitney U-test for nonparametric variables and the chi-squared test for categorical values. The disease-free survival was estimated by the Kaplan-Meier method . The period of disease-free survival was defined as the time between the date of cystectomy and death from cancer or the development of local recurrence or distant metastasis. Death from unknown causes was considered death from cancer.
There were 67 early complications in 63 patients (19%) (Table 1); three patients died from pulmonary embolism 7–10 days after surgery, giving a mortality rate of 0.9%. Prolonged urinary leakage was the commonest complication, either from the ileo-urethral or the uretero-ileal anastomosis. The former was treated by prolonged urethral catheter drainage, and the latter by percutaneous nephrostomy (Fig. 1).
Pelvic collections were found in 10 cases; they required percutaneous tube drainage in nine and open surgery in one. There was prolonged bowel ileus in four patients and upper gastrointestinal bleeding from stress ulcers occurred in seven patients. All were treated successfully by conservative measures.
The mean (sd) follow-up was 87.8 (49.1) months, the 5- and 10-year disease-free survival rates were 60.6 (2.7)% and 56.2 (3.0)%, respectively (Fig. 2). Most of the deaths from cancer were during the first 2 years after surgery. During the follow-up 105 patients died from cancer, six were living with disease, 12 died from unrelated cause and seven were not followed (205 evaluable patients).
The relative distribution of causes of treatment failure is given in Table 1. Four patients had recurrences in the retained urethra (1.2%). Diagnosis was confirmed by urethroscopy and biopsy. Urethral washout cytology was not used for routine follow up. Urethrectomy and conversion to continent cutaneous diversion were necessary.
The most common late complication was the development of 72 pouch stones on staples in 55 patients (Fig. 3). The interval to stone detection was 5–132 months. All stones but six were treated endoscopically. Open surgery was indicated because of the large stone volume and/or need for revisional surgery for the nipple valve.
There was ileo-urethral stenosis at the anastomotic site in seven patients presenting with difficulty in passing urine. These patients had required dilatation of the stenosed segment in three and incision in four. Stenosis recurred in three patients which necessitated re-treatment. Six patients had anterior urethral strictures and were treated by visual internal urethrotomy. The nipple valve was revised in 11 patients with failed valves; nipple valve stenosis was treated by endoscopic dilatation.
The configuration of the upper urinary tract as evaluated by excretory urography (Fig. 4) showed that 36 (5.3%) renal units had evidence of deterioration, 17 caused by a failing nipple valve with reflux that occurred as an early event after surgery, four by valve eversion with no reflux, four by valve stenosis (Fig. 5), and 11 by uretero-ileal anastomotic strictures. Notably, 65 renal units which were dilated before surgery had considerably improved afterward. Evidence from ascending studies showed that the nipple valve can provide an efficient antireflux mechanism in most cases. Reflux caused by valve eversion was seen in only 30 patients (9%).
The continence status was assessed in 205 evaluable patients; nine were totally incontinent, two patients had chronic urinary retention and required clean intermittent self-catheterization. During the day, 192 patients (93.7%) were completely continent and two had stress incontinence.
Fifty-five patients (26.8%) had nocturnal enuresis. The urodynamic findings of 55 enuretic and 33 continent patients are outlined in Table 2 ; enuretics had a larger volume of residual urine, and the frequency, amplitude and duration of phasic contractions were significantly higher among the enuretics.
|Mean (sd) variable||Continent||Enuretic||P|
|Max. flow rate, mL/min||19.5 (4.4)||16.2 (7.3)||0.03|
|Functional urethral length, mm||33 (6.8)||29 (9.2)||0.08|
|Max. urethral closure pressure, cmH2O||85.9 (23)||67.2 (23.3)||0.005|
|Max. resting urethral pressure, cmH2O||90.4 (23.3)||71.9 (23.8)||0.001|
|Residual urine, mL||13.9 (29.3)||59.1 (86.1)||0.02|
|Max. enterocystometric capacity, mL||594 (133.4)||587 (180.8)||0.59|
|Pressure at max. capacity, cmH2O||20.3 (9.7)||26.1 (16.7)||0.07|
|frequency (in last 5 min of filling)||2.1 (1.2)||2.9 (1.5)||0.009|
|max. amplitude, cmH2O||26.4 (13.2)||37.9 (21.1)||0.008|
|duration, s||28.9 (15.4)||37.3 (16.8)||0.005|
The use of a hemi-Kock pouch for orthotopic bladder substitution after cystectomy for cancer was reported by our group in 1987 . The operation is based on two principles, i.e. reconfiguration of an ileal segment to accommodate a large volume at low pressure, and provision of an antireflux mechanism by an intussuscepted nipple valve. While the first principle is universally accepted by all investigators [1,2,9–11], there is controversy about the optimal method of uretero-intestinal anastomosis, and whether an antireflux mechanism should be incorporated with orthotopic neobladders.
Proponents for using a simple direct anastomosis argue that an antireflux mechanism is unnecessary in low-pressure reservoirs [12–14]. Furthermore, the potential benefits from an antireflux operation may be outweighed if the technique has a high risk of stricture formation. On the other hand, a need to incorporate an antireflux mechanism is indirectly supported by some experimental evidence  and clinical findings [16,17]. Asymptomatic persistent bacteriuria is a frequent finding in patients with orthotopic bladder substitutes [7,18]. In the present study, a third of the patients had persistent bacteriuria. While it is true that the pressure within orthotopic bladder substitutes is low during the storage phase, Gotoh et al. provided evidence that voiding is achieved by increasing the intra-abdominal pressure. As a result, the pressure within the reservoir is markedly increased to a mean of 77.3 cmH2O. Therefore, the micro-organisms might spread to the upper urinary tract if an antireflux technique is not used.
Initial concerns were expressed about the potential hazards of developing a local recurrence in the retained urethra. This risk seems to be minimal in patients with squamous cell tumours . With TCC this possibility can be minimized if the high-risk subpopulation of patients is excluded, i.e. those with tumour near the bladder neck or infiltrating the prostate, and/or the presence of carcinoma in situ in the prostatic urethra [21–23]. In the present series selected by these criteria, only four patients had a local recurrence in the urethra during the follow-up, an incidence of 1.2%. Freeman et al. reported that the incidence of urethral recurrence in patients with TCC and who had Kock orthotopic urinary diversion was significantly less than that after cutaneous diversion. These investigators proposed that ileum connected to the urethra in some way protects the urethra from recurrent TCC, or that a defunctionalized urethra is at greater risk of recurrence.
Imaging of the upper urinary tract provided evidence that the operation can give an unobstructed unidirectional flow of urine in most patients. There was deterioration in 36 renal units (5.3%) from various causes, but in only 11 units (1.7%) was it a result of uretero-ileal anastomotic strictures, because a wide spatulated mucosa-to-mucosa anastomosis was used. These data are similar to those published by others using the same technique [25,26] and compare favourably with the 3% incidence reported by Studer et al. when a long afferent loop was used.
A critical assessment of the causes of chronic complications indicates that many were valve-related, as noted by others [25,26]. Stones developed on metallic staples in 55 patients (16.3%), most of which could be treated endoscopically. This risk could be considerably reduced by using absorbable staples . Valves failed in 9% of patients; in 11, open revision of the valve was needed because of the resulting back-pressure changes of the upper urinary tract. Nipple valve stenosis was first reported by Stein et al., presumably caused by ischaemia resulting from applying four lines of staples. This complication occurred in two patients. Repeated endoscopic dilatation was needed to stabilize the upper urinary tract.
The rate of continence after orthotopic bladder substitution varies in different studies, particularly for night-time continence, largely because there is no uniform definition. In the present study, continence was defined as dryness with no need for condom catheter, pads or medication; of the present patients, 93.7% were continent during the day and 67.9% during the night. These data are consistent with those in contemporary studies [9,26,30,31]. Continence in orthotopic diversion depends on the balance of pressure in the reservoir and urethral resistance factors. The maximum urethral closure pressure was lower in the enuretic than in continent patients. Similarly, Koraitim et al. noted a positive correlation of maximum urethral closure pressure with diurnal and nocturnal continence. Nevertheless, the resistance offered by the urethra sustained diurnal continence in both groups. However, these variables were measured during the day and significant changes may occur during sleep. Jackobsen et al. reported a decrease of ≈ 20% in urethral closure pressure during the deep stage of rapid eye movement sleep. There was no difference in measured pouch capacity in the continent and enuretic patients. Nevertheless, the reservoir pressure at maximum capacity was higher, and the frequency, amplitude and duration of phasic contractions increased in patients with enuresis. The presence of these contractions is surprising as all the pouches were detubularized and reconfigured. To date the cause of these contractions is unknown.
El-Bahnasawy et al. reported that the incidence of bacteriuria was significantly higher among enuretics, but they did not assess the influence of treating such infections on the level of continence. The influence of food intake, which may trigger motor activity in intestinal reservoirs, could also be blamed . Steven and Poulsen  reported that night-time continence is reduced in patients aged > 70 years. All these possibilities must be assessed and explored in an attempt to improve night-time continence after orthotopic bladder substitution.
Thus orthotopic bladder substitution after cystectomy for cancer is feasible, with good functional and oncological outcomes in properly selected patients. Nevertheless, the use of a hemi-Kock pouch is associated with several important valve-related complications. This led our group to investigate a procedure in which a nipple valve was not used nor staples applied. There was experimental and clinical evidence that this could be achieved by embedding the ureters in serous-lined extramural tunnels [35,36]. The excellent functional outcome reported initially was confirmed by a second report with a longer follow-up of more patients . The basic technique has two limitations, i.e. dilated ureters and/or a concomitant pathological condition in the distal ureters, necessitating their excision. A solution was proposed that relied on the same principle of embedding a tubular structure in a serous-lined trough . A separate 6–8 cm segment of ileum is tapered and embedded in a serous-lined trough. The ureters are then anastomosed to the inlet of the embedded segment using the mucosa-to-mucosa technique. This principle was reported by Stein et al., who used it with the Kock neobladder as an alternative to an intussuscepted nipple valve.
In conclusion, the use of the hemi-Kock pouch paved the way for orthotopic bladder substitution; the valve-related complications stimulated the introduction of refinements and improvements. Whilst the principles of detubularization and reconfiguration were maintained, the creation of an intussusception nipple valve to prevent reflux was abandoned. Although the need for an antireflux mechanism with orthotopic reservoirs remains controversial, alternative techniques for reflux prevention are successful.