SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References

Twenty-two women with primary and secondary (five patients) vesicovaginal fistula attending a tertiary level urological unit in India were treated by repair of the fistula using bladder mucosal autografts. The fistula was approached transabdominally or via a combined abdominal and vaginal approach (for those involving the trigone). After closure of the vaginal layer, bladder mucosa was harvested from the dome of the bladder and laid over the fistula with sutures at each corner to fix it in place. Patients were catheterised for 12–14 days. At follow up after 3 to 12 months, 20 out of 22 patients were continent, with no other symptoms. The two failures had undergone two previous repairs each. This series is the first from India, and demonstrates the efficacy of bladder mucosal autografts for managing large fistulae, those where a previous repair has failed and fistulae adjacent to the ureteric orifice without the need for uretero-neocystostomy.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References

Vesicovaginal fistulae are an uncommon cause of urinary incontinence. There is a large difference in the incidence and aetiology of vesicovaginal fistulae around the world. In the developed world, vesicovaginal fistulae usually occur after elective hysterectomy, with an incidence of 0.5–2%.1 In developing countries, prolonged obstructed labour remains the most common cause of vesicovaginal fistula. There have been no population-based studies to estimate the incidence of vesicovaginal fistulae, but hospital series have quoted rates up to 0.5% after vaginal delivery. Vesicovaginal fistulae that develop after obstructed labour usually involve the urethra or trigone. Poor hygiene and inaccessibility of health care facilities by rural population contribute to the magnitude of the problem.

Surgical repair remains the cornerstone for the management of vesicovaginal fistulae. Refinement in techniques, development of finer suture materials and use of broad-spectrum antibiotics have improved the rate of a successful outcome. Vesicovaginal fistulae involving the trigone often involve the ureteric orifices as well. Repair of these cases usually requires ureteric stenting and reimplantation, and often an abdominal approach is necessary, increasing the associated morbidity and cost to the patient (financial, physical and mental). The use of free bladder mucosal grafts for the repair of iatrogenically created vesicovaginal fistulas has been reported in dogs.2 Since this pioneering work, several authors have reported excellent results by using this technique.3–5 This paper reports our experience of bladder mucosal grafts for managing vesicovaginal fistulae in the Indian subcontinent.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References

Between January 2003 and January 2004, patients with vesicovaginal fistulae were treated using bladder mucosal autografts. All patients were operated upon after obtaining written consent and explaining the nature of the technique. The diagnosis of the fistula in all cases was made by the history, physical examination (including a speculum examination) cystoscopy and vaginoscopy. An intravenous urogram was done if clinically indicated. Vaginal douching with povidone–iodine was started one night before the procedure. All patients received antibiotic prophylaxis with third generation cephalosporins and aminoglycosides. A transabdominal approach was used for vesicovaginal fistulae above the trigone and a combined abdominal and vaginal approach for vesicovaginal fistulae involving the trigone. The bladder was exposed using an extraperitoneal infraumbilical midline incision. Any scar tissue from previous repairs was removed. After the bladder was opened the ureteric orifices were identified and cannulated with 5 Fr open-ended ureteric catheters. The fistula was circumscribed with sharp dissection making no attempt to separate the vaginal and vesical layers. The vaginal layer of the fistula was closed with interrupted 4-0 polyglactin sutures with the knots facing the vaginal lumen.

A free mucosal graft approximately 50% larger than the fistula was taken from a healthy area in the dome of the bladder. Haemostasis was achieved at the graft harvest site. The grafts were laid over the repaired vaginal layer of the fistula with the mucosal surface facing into the bladder. The graft was secured to the margins of the fistula by three or four interrupted 4-0 polyglactin sutures at the corners of the graft. The bladder was drained by an 18 Fr suprapubic cystostomy tube and a urethral catheter. The urethral catheter was inflated with just 3–5 mL of saline when the fistula involved the trigone. A perivesical drain was placed and the bladder closed in two layers using 2-0 polyglactin sutures.

Post-operatively, the patient was kept on antibiotics for 7–10 days. Drains were removed in all patients within 48 hours. The urethral catheter was removed between the 12th and 14th post-operative day, and the suprapubic cystostomy tube was removed 24–48 hours later. In cases where the repair was leaking after removal of the urethral catheter, the patient was kept on suprapubic drainage for a period of four weeks. All the patients were kept on anticholinergic medication until the catheters were removed to prevent bladder spasms that might disrupt the graft site and to prevent pericatheter leaks. The patients were discharged with advice to avoid sexual intercourse for a period of three months, which is routine practice after vesicovaginal fistulae repair at our hospital.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References

Details of the 22 patients operated upon and the outcomes are shown in Table 1. Five patients had previously undergone a failed repair. In three of these we used a combined transabdominal and transvaginal route and in two cases closure of the fistula was attempted transabdominally only. The average operating time in the present series was 100 minutes.

Table 1.  Details of patients.
Clinical featuren (%) or median [range]
Age (years)36 [22–50]
 
Type of fistula
Obstetric15 (68)
After abdominal hysterectomy5 (23)
After vaginal hysterectomy2 (9)
 
Trigone involved12 (55)
Diameter of fistula (mm)15 [7–33]
Duration of admission13 [12–14]
Urinary tract infection3 (14)
Continent at follow up (3 to 12 months)20 (91)
Stress incontinence at follow up0

Two patients who continued to leak after initial clamping of the cystostomy tube remained incontinent after four weeks of free drainage. These two patients had complex trigonal vesicovaginal fistulae and both had been operated twice previously by transabdominal approaches. Repeat cystoscopy showed a decrease in size of fistulas to 3 and 5 mm from pre-operative dimensions of 13 and 15 mm, respectively, suggesting partial uptake of the graft.

In four patients, we noted that the margin of the fistula lay close to the ureteric orifice. We have previously used layered closure of the fistula with uretero-neocystostomy to manage such cases. By using bladder mucosal autografts as overlays, we were able to avoid uretero-neocystostomies that would have been required. These four patients had normal renal function and intravenous urogram's post-operatively.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References

Davis and Mirands6 described successful closure of vesicovaginal fistulas by bladder drainage alone, although it generally true that some form of surgical repair is necessary for successful closure of vesicovaginal fistulae. The repair can be performed either vaginally7 or via an abdominal approach.8

The use of free mucosal grafts was first described in dogs.2 A postmortem examination at eight weeks showed good healing at the fistula sites with resultant fibrosis. The reported success rate from a Brazilian study of 80 women with vesicovaginal fistulae repaired with bladder mucosal grafts over a period of 15 years was 96.3% with no late failures.3 These promising figures are supported by two smaller studies with success rates of 100% and 93%.4,5 Our cure rate of 91% compares favourably with these. It remains true that the plan of repair should be tailored for each patient depending on the size, position and number of fistulae and the state of surrounding tissues that may be used in the repair. The present study has confirmed the role of mucosal autografts in managing peri-ureteral fistulas that would otherwise require a ureteric reimplantation.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References

Accepted 17 May 2004