Further Experience with the Modified Rectal Bladder (The Augmented and Valved Rectum) for Urine Diversion

0022-534 7/92/14 75-1252$03.00/0 Vol. 147, 1252-1255, May 1992

THE JOURNAL OF UROLOGY

Copyright© 1992 by AMERICAN UROLOGICAL ASSOCIATION, INC.

Printed in U.S.A.

FURTHER EXPERIENCE WITH THE MODIFIED RECTAL BLADDER (THE AUGMENTED AND VALVED RECTUM) FOR URINE DIVERSION MOHAMED A. GHONEIM,* ALBAIR K. ASHAMALLAH, MOHAMED R. MAHRAN NILS G. KOCK

AND

From the Department of Urology, Urology and Nephrology Center Mansoura, Egypt

ABSTRACT

Continent diversion with the modified rectal bladder was done in 83 patients and 65 are currently evaluable with followup ranging from 6 to 36 months. There was no postoperative mortality and the morbidity rate was acceptable. Renal function and configuration were maintained in most patients (91 %). Dessusception of the colorectal valve was observed in 7.6% of the patients. Evidence was provided that this valve is effective in prevention of regurgitation of the rectal contents to the proximal colon. All patients were continent during the day. Enuresis was noted in 6 patients and all of them responded to imipramine hydrochloride therapy. The results support earlier observations that the procedure offers distinct advantages over ureterosigmoidostomy and the simple rectal bladder. KEY WORDS:

urinary diversion, bladder, rectum

The modified rectal bladder (the augmented and valved rectum) is a new surgical procedure that incorporates 3 principles: 1) the rectum is functionally isolated by an isoperistaltic intussusception valve at the colorectal junction, 2) the ureters are implanted into this isolated rectal bladder by an antireflux mechanism and 3) the urodynamic qualities of the neo-reservoir are improved by patching with an open sheet of ileum. These concepts were initially tested in a series of animal experiments (dogs). 1 On the basis of the reported favorable outcome a clinical trial was done in 19 patients. The preliminary results were the subject of a previous report. 2 We report on a larger group of patients with longer followup. Accepted for publication September 13, 1991. for reprints: Department of Urology, Urology and Nephrology Center, Mansoura, Egypt.

* Requests

MATERIALS AND METHODS

Patients. From December 1986 to December 1990, this operation has been done in 83 patients: 57 women, 20 men and 6 children. Median age for the adults was 53 years (range 22 to 67 years). The indications for diversion were bladder cancer in 70 patients, irreparable urinary fistulas in 5, bladder exstrophy in 4, contracted bladder in 2 and bladder sarcoma in 2. Operation. The surgical technique is essentially similar to what has been previously described in detail. 2 A 10 cm. long incision is made in the anterior wall of the lower rectum. A colorectal intussusception valve is created and its position is maintained by 3 to 4 rows of metallic staples. The ureters are then anastomosed to the thus isolated rectal pouch by an antireflux system. Initially, a technique was performed in which the colorectal valve was used to provide the antireflux mecha-

FIG. 1. Antegrade barium study (follow through) via defunctioning colostomy shows free flow of barium through colorectal valve

1252

l\lWDIFIED RECTAL BLADDER FOR URINE DIVERSION

Ureterorectal anastomosis No. Renal Units Totals Through the sigmoid valve Through a submucous tunnel Through an ilea! valve Totals

Rt.

Lt.

30 41 12 83

26 43 11

80

56 84 23 163

nism. The ureters were brought between the 2 layers of the intussusceptum, a button hole was created at the tip of the nipple valve and a stented mucosa-to-mucosa anastomosis was made between the spatulated end of the ureter and the sigmoid mucosa. Alternatively, an independent reimplantation was used with a submucosal tunnel as described by Goodwin et al. 3 For rectal augmentation a 20 cm. long segment of the distal ileum was isolated, and its antimesenteric border was opened and folded into a U-shaped plate. The intestinal sheet was then patched to the anterior wall of the opened rectum. If the ureters are grossly dilated the aforementioned 2 techniques are not feasible. In such circumstances a separate antireflux valve was created from the ileum in connection with the augmentation procedure. The ureters were then anastomosed by an end-to-side technique to the afferent limb of that valve. The different methods used for reflux prevention are summarized in the table. Finally, a temporary transverse colostomy was made, which was closed after 6 to 8 weeks. The rectal pouch was drained by a tube for 3 weeks. The ureteral stents were removed after 10 days. Evaluation. Patients were examined and evaluated at discharge from the hospital, and then regularly every 3 months for 1 year and every 6 months thereafter. Followup ranged from 6 to 36 months. Evaluation included symptom analysis for daytime continence, nocturnal enuresis and frequency of voiding. Renal function and configuration were assessed by excretory urography (IVP) and radioisotope renograms. Ascending proctograms were performed to detect the presence of reflux and determine the efficiency of the colorectal valve in providing functional isolation. Slow-fill (10 ml. per minute) water proctometry was recorded to measure the volume and pressure relationship in the neobladder. Voiding pressures and flow rates were also obtained. RESULTS

There was no postoperative mortality. Of the patients 9 had 14 early complications, including prolonged ileus in 3, sepsis in

1253

5, deep venous thrombosis in 2, intestinal obstruction requiring surgical intervention in 1, internal hemorrhage requiring surgery in 1 and a rectovaginal fistula in 2 (1 required surgery and 1 was treated by prolonged rectal tube drainage). All but 3 patients were treated conservatively. During the observation period 18 patients died of cancer, while 65 are currently fully evaluable. All patients are continent during the daytime with an emptying frequency of 2 to 5 times. Similarly, all but 6 patients are dry at night with a voiding frequency of O to 2 times. It must be noted that of the 6 enuretics 2 were children and all of them responded favorably to treatment with imipramine hydrochloride given orally before bedtime. None of the patients complained of constipation or abdominal distention following closure of the colostomy. In the initial phase of the study an antegrade barium study through the temporary colostomy was done to ensure unobstructed flow of material through the colorectal valve (fig. 1). Thereafter, this investigation was judged to be unnecessary and was abandoned. IVP and renographic studies documented an improvement or stabilization of the upper tract in 115 of 127 renal units (91 %). Deterioration was observed in 12 units due to an anastomotic stricture in 4 and reflux in 8. An example case with a longitudinal followup of 3 years is demonstrated in figure 2. The possibility to incorporate a second ileal intussusception valve to provide an antireflux mechanism for dilated ureters unsuitable for submucosal reimplantation is shown in figure 3. Endourological treatment of anastomotic strictures was successful in 2 patients. This involved percutaneous nephrostomy, guided antegrade balloon dilation and stenting with a DoubleJ* catheter. In the remaining 2 patients open surgical correction was necessary. Ascending proctography provided evidence that the colorectal intussusception valve was efficient in preventing regurgitation of the rectal contents to the proximal colon in all but 5 patients (fig. 4). In 4 of these patients, and due to the incorporation of the ureters in the colorectal valve, reflux was observed to the upper tract as well. Surgical correction of these failing valves was done before closure of the transverse colostomy. Revision was successful in 3 patients, while in 1 the attempt was abandoned in view of detection of extensive local recurrence at exploration. Filling proctometry revealed an adequate capacity of the reservoir, with a range of 480 to 800 ml. 6 months postopera* Medical Engineering Corp., New York, New York.

FIG. 2. A, preoperative IVP. B, early postoperative IVP shows ureters implanted by Goodwin's technique. Note mild pooling of contrast medium presumably from residual edema at anastomotic site. C, IVP 36 months postoperatively demonstrates perfect function and configuration of upper tract.

1254

GHONEIM AND ASSOCIATES

FIG. 3. A, preoperative IVP shows bilateral hydroureters unsuitable for submucosal tunnel implantation. Antireflux was achieved by second intussusception ilea! valve. B, postoperative IVP reveals significant decompression on both sides. C, ascending proctogram with air and contrast medium to delineate 2 intussusception valves: colorectal valve on left side and ilea! valve on right side. Note absence of reflux.

FIG. 4. Ascending proctography and micturition study. A, plain x-ray of pelvis before filling. Note staples seen as parallel rows. B, partially filled reservoir. C, reservoir full to capacity. Note larger capacity without reflux to ureters or regurgitation to proximal colon. D, post-voiding film shows complete emptying with minimum residual urine.

tively. Filling was achieved at a low pressure without any segmental contractions. The pressure at maximum capacity did not exceed 30 cm. water. Figure 5 illustrates a simultaneous tracing of the subtracted intrarectal pressure during voiding as well as uroflowmetry. This study demonstrates that the driving force of voiding in this system is the function of a voluntary increase in the intra-abdominal pressure, rather than due to intrinsic contractions of the rectal musculature. Furthermore, it shows that peak flows of greater than 20 ml. per second are usually achieved. DISCUSSION

The modified rectal bladder provides several significant improvements over the simple rectal bladder on one hand and ureterosigmoidostomy on the other hand. There is no need for

a terminal left iliac colostomy. Meanwhile, regurgitation of the rectal contents to the proximal colon is prevented by the colorectal intussusception valve. The feasibility of creation of such a valve was initially proposed by Knox and Jago in 1975 in an experimental model using piglets. 4 Its efficiency to achieve this goal and consequently decrease the surface area of the colonic mucosa exposed for reabsorption of urine has been established experimentally5 and clinically. 6 The procedure allows several possibilities for an antireflux ureteral reimplantation. Although the functional results and efficiency of reflux prevention are similar whether the ureters are implanted through the nipple valve or through an independent submucosal tunnel, the latter procedure currently is preferred because dessusception of the colorectal valve would certainly lead as well to failure of the antireflux system if the ureters were

MODIFIED RECTAL BLADDER FOR URINE DIVERSION

7.

:~1

u

8.

30

V

en 20

" E

10

o

9.

20t~ O

Fl.OW RA'fF

10.

10

:2

11.

0

5

10

12.

20 SUBTRACTED INTRARECTAL PRESSURE

FIG. 5. Combined recording of urine flow rate and subtracted intrarectal pressure. Note negative slope of voiding pressure, which indicates that voiding in this system is achieved by increased intra-abdominal pressure.

incorporated within this valve. Furthermore, the urodynamic qualities of the rectal bladder have been improved by patching with ileum. The volume capacity and compliance are increased. Filling is achieved at a low pressure without any segmental contractions, which explains continence during the day and night. It is noteworthy that the favorable clinical outcome after this operation had been reproduced by other investigators. 7 •8 Nevertheless, 2 questions remain unanswered: 1) what will be the incidence of pyelonephritis and 2) how much could it affect the renal function during the long term? We know that the reported incidence of pyelonephritis after ureterosigmoidostomy in which an antireflux procedure has been used is approximately 30%. 9 - 11 However, it could be postulated that the projected incidence of pyelonephritis after the modified rectal bladder procedure could be significantly decreased, since we are dealing with a low pressure system. What will be the incidence of tumors in this system when stools and urine are mixed? It may also be argued that in view of the experimental observations of Crissey et al1 2 some protection is offered by the temporary transverse colostomy. The suture line is allowed to heal without exposure to the fecal stream. In any case, the importance of longitudinal followup and accurate reporting does not need emphasis. The balance of all of these factors will determine the exact role of this method among other techniques used to achieve a continent diversion. REFERENCES 1. Kock, N. G., Berglund, B., Ghoneim, M.A., Lindholm, E., Lycke, K. G. and Virseda, J.: Urinary diversion to the augmented and

2.

3. 4. 5. 6.

valved rectum. An experimental study in dogs. Scand. J. Urol. Nephrol., 22: 227, 1988. Kock, N. G., Ghoneim, M.A., Lycke, K. G. and Mahran, M. R.: Urinary diversion to the augmented and valved rectum. Preliminary results with a novel surgical procedure. J. Urol., 140: 1375, 1988. Goodwin, W. E., Harris, A. P., Kaufman, J. J. and Beal, J. M.: Open transcolonic ureterointestinal anastomosis, a new approach. Surg., Gynec. & Obst., 97: 295, 1953. Knox, A. J. J. and Jago, R. H.: The value of intestinal valves in urinary diversion-an experimental study. Brit. J. Urol., 47: 391, 1975. Miller, K., Matsui, U. and Hautmann, R.: The functional rectal bladder-prevention of hyperchloremic acidosis following vesicosigmoidostomy in dogs. J. Urol., 144: 375, 1990. El-Mekresh, M. M., Shehab El-Din, A. B., Fayed, S. M., Brevinge, H., Kock, N. G. and Ghoneim, M.A.: Bladder substitutes con-

1255

trolled by the anal sphincter: a comparison of the different absorption potentials. J. Urol., 146: 970, 1991. Skinner, D. G., Lieskovsky, G. and Boyd, S.: Continent urinary diversion. J. Urol., 141: 1323, 1989. Miller, K., Matsui, U. and Hautmann, R.: The augmented functional rectal bladder. First clinical experience J. Urol., part 2, 143: 398A, abstract 840, 1990. Williams, D. F., Burkholder, G. V. and Goodwin, W. E.: Ureterosigmoidostomy: a 15-year experience. J. Urol., 101: 168, 1969. Wear, J. B. and Barquin, 0. P.: Ureterosigmoidostomy: long-term results. Urology, 1: 192, 1973. Zincke, H. and Segura, J. W.: Ureterosigmoidostomy: critical review of 173 cases. J. Urol., 113: 324, 1975. Crissey, M. M., Steele, G. D. and Gittes, R. F.: Rat model for carcinogenesis in ureterosigmoidostomy. Science, 207: 1079, 1980. EDITORIAL COMMENT

The authors are to be congratulated on their continued innovative contributions to continent urinary diversion. We have performed 15 of these procedures since 1986 and have not seen pyelonephritis or renal deterioration in any. Several points warrant consideration and emphasis. The longitudinal rectotomy must be as low as possible near the anus to prevent angulation or poor emptying of the reservoir. The authors reported an incidence of ureterocolonic obstruction or reflux of 10% (12 of 127 renal units) with relatively short followup. This again raises the question of the efficacy of a tunneled ureterocolonic anastomosis without even addressing the possible late development of carcinogenesis. We strongly believe that incorporation of an antireflux nipple valve into construction of the ilea! patch with direct end-to-side ureteroileal anastomosis, similar to what the authors describe for their 12 patients with dilated ureters, is preferable to any tunneled anastomosis. Our long-term experience with ureteroileal anastomoses reveals a stenosis rate of only 3% with a long-term malfunction rate of the afferent ilea! nipple of only 1%. In addition, it moves the ureteral anastomosis away from the fecal stream and greatly facilitates revision should it be required. The issue of nighttime incontinence cannot be minimized, since the stench and social consequences of fecaluria are far worse than simple urinary incontinence. Although the authors report improvement in children with use of imipramine, we have not seen similar success in adults. In our experience the patients who are ideal candidates for this operation are former urete•osigmoidostomy patients in whom ureterocolonic difficulties have developed and yet who are otherwise satisfied with the excretory function of ureterosigmoidostomy. These patients do extremely well with conversion to an ileal-anal reservoir and in fact they find it far superior to the former ureterosigmoidostomy. They have less electrolyte abnormality, better capacity, less urgency, less frequency and far better nighttime control. However, we have found the operation unsatisfactory in older women with somewhat lax anal sphincter tone and prefer continent cutaneous diversion with intermittent catheterization. As the authors suggest, longer followup and more experience will determine the ultimate role of the operation in managing patients with pelvic malignancy. Donald G. Skinner Division of Urology University of Southern Cali/ornia Los Angeles, California REPLY BY AUTHORS We agree with Doctor Skinner's comment that the best and safest method for ureteral reimplantation is a wide, spatulated end-to-side mucosa-to-mucosa anastomosis to an independent ilea! nipple valve. It has been our experience that this technique results in superior function. This finding is supported by the reports of other investigators who use a second ilea! nipple value exclusively for reimplantation.1 This method is indeed unsuitable for patients with a diseased colon or a weak anal sphincter. Nevertheless, with regard to patching the reservoir our patients have less frequency, less urgency and far better nighttime control compared to patients with ureterosigmoidostomy. 1. Miller, K., Matsui, U. and Hautmann, R.: Functional, augmented rectal bladder: early clinical experience. Eur. Urol., 19: 269, 1991.