In developed countries, common causes of vesicovaginal fistulae are gynecologic and urologic procedures and pelvic radiation therapy. Whereas fistulae occurring after a gynecologic procedure usually are evident 10 days to 2 weeks after surgery, radiation-induced fistulae can develop many years remote from the initial therapy. Although patients from the procedure-induced and radiation-induced fistulae groups are distressed about the urinary leakage, those experiencing continuous urinary leakage after an iatrogenic procedure are especially devastated by this unexpected occurrence. A common risk factor for the development of a fistula is a technically difficult hysterectomy, but vesicovaginal fistula may also occur after “routine” hysterectomies.
Because vesicovaginal fistulae cause significant mental and physical anguish, all attempts must be made to provide patients with reliable repairs that can be performed in an expedient fashion, with brief hospital stays and recovery times. Ideally, the chosen repair procedure should have low perioperative morbidity and minimal de novo postoperative problems. Various surgical techniques for posthysterectomy fistula repair have been published, with successful outcomes achieved in 82 to 100% of the cases. 1–3 There does not appear to be consensus regarding approach (transvaginal versus transabdominal), timing of repair, use of interposition tissue, or whether removal of the fistulous tract is necessary. Also lacking in these studies is discussion of postoperative issues such as voiding dysfunction, sexual function, and vaginal pain.
Our objective was to evaluate postoperative sexual function, voiding dysfunction, and pain-related symptoms, and to evaluate the success rate of posthysterectomy fistula closure by a transvaginal entire-cuff excision with removal of the fistulous tract.
Materials and Methods
Thirty-four women underwent transvaginal cuff excision for repair of their vesicovaginal fistula between January 1997 and October 1999. Patients were excluded from the study if they had received pelvic radiation therapy or had a ureterovaginal or urethrovaginal fistula. All patients underwent a pelvic examination, urine culture, excretory urogram, cystoscopy, and vaginoscopy. We recorded the number and type of previous attempts at fistula repair, time between hysterectomy or last repair, and location and cystoscopic dimension of the fistula.
Concern for postoperative issues, such as voiding dysfunction, sexual function, and vaginal pain, prompted the use of pre- and postoperative questionnaires. These issues were addressed for the last 14 patients. The questionnaires included inquiries concerning quality of life, sexual activity, pelvic pain, and voiding symptoms. 4 Also, the 14 patients who completed questionnaires also underwent pre- and postoperative pelvic evaluation, which included measuring total vaginal length according to the pelvic organ prolapse staging. 5
We used a Bookwalter vaginal retractor for all patients and found that it provided excellent exposure of the vaginal cuff and required no intraoperative adjustment (Fig. 1). Our surgical technique began with catheterization of the fistulous tract and outlining of the entire vaginal-cuff scar (Fig. 2). The vaginal-cuff scar and fistula tract were then excised (Fig. 3), leaving clean bladder and vaginal margins (Fig. 4). The bladder wall was closed with interrupted 4.0 vicryl. The pubocervical fascia was approximated to the rectovaginal fascia using similar interrupted sutures, with an attempt not to overlap closure lines. The vaginal closure completed the repair.
After the procedure, a suprapubic catheter was placed, and patients were admitted for a 24-hour observational period. Upon discharge, catheters were left for drainage for the next 3 weeks and patients were given an anticholinergic agent, for use if required for suppression of detrusor activity that could compromise healing, and an antimicrobial prophylactic agent. At their initial 3-week appointment, if patients were dry at the vaginal incision without signs of infection, the suprapubic catheter was removed. Cystography was only performed before catheter removal if there is any doubt about the integrity of the repair. Use of anticholinergic and antimicrobial agents was discontinued, and patients were instructed not to have vaginal intercourse for 3 months but that they could continue other activities.
The average age of patients was 42 years (range, 29–57 years). One patient had undergone a radical hysterectomy for cervical cancer; all others had undergone an abdominal hysterectomy for benign conditions. Twelve of the 34 women (35%) had previous attempts at closure of the fistula. Seven had undergone an abdominal procedure, three had previous vaginal attempts, and two underwent attempted fulguration of the fistula. Patients presented at an average of 24.6 weeks (range, 4–52 weeks) after initial injury or previous attempt. The earliest that a vaginal-cuff excision repair was performed was 4 weeks after injury.
All 14 patients who completed the pre- and postoperative questionnaires reported normal sexual activity before their hysterectomy, but had refrained from vaginal intercourse since their fistula development because of fear, discomfort, and frustration. The average preoperative pad usage was 7.0 pads/d (range, 4–8 pads/d). No patient reported urinary urge nor stress incontinence before her hysterectomy. The average endoscopic size of the fistula was 7.1 mm (range, 3–12 mm). The fistulae were all at the vaginal cuff. The mean total preoperative vaginal length was 9.5 ± 1.68 cm.
All patients were dry at their initial 3-week appointment and the 12 French soft suprapubic catheter was removed. At the 4-month evaluation, all patients continued to be dry and had regained all normal activity. Nine percent (3/34) of the patients had minor urinary urgency and frequency symptoms. Patients had significant improvement in their quality of life with respect to urinary incontinence. The mean preoperative Incontinence Impact score was 285 ± 17.3; the mean postoperative score was 0 (P = 0.01). Ninety-three percent (13/14) of the patients resumed sexual activity after fistula repair. One patient divorced shortly after the surgery and was not sexually active at the time of evaluation. The other thirteen patients reported no decrease in satisfaction or development of dyspareunia, which was supported by the fact that vaginal length changed minimally, mean preoperative vaginal length was 9.5 ± 1.68 cm, and mean postoperative vaginal length was 8.7 ± 1.45 cm. (Table 1).
The success rate for uncomplicated vesicovaginal fistula repair is high, whether the repair is transabdominal or transvaginal, whatever technique used. However, failures do occur; in our series, 35% had at least one previous attempt at repair. Besides recurrent breakdown at the fistula area, another reason for possible failure is that more than one fistula may actually be present along the cuff. That any unidentified fistula will be removed is partly why removal of the entire cuff is so successful.
Our technique adhered to the principles of adequate operative exposure and a tension-free, well-vascularized, scar-free closure. Other surgeons have reported on successful fistula closure without excision of the fistulous track. 6 The motivation of these surgeons was that fistulous-track removal enlarged the bladder hiatus and reduce the possibility that the closure would cause severe bladder spasms, increasing the risk of reopening. Fortunately, individual iatrogenic posthysterectomy fistulae rarely are larger than 1 cm, and removal of the track minimally increases the size of the hiatus to close. Also important to the successful outcome of fistula repair is appropriate postoperative management of continuous urinary drainage and avoidance of bladder hyperactivity through use of anticholinergic therapy. 7
The optimal timing for repair has been frequently debated. The earliest repair that we performed was at 4 weeks, by which time the tract was epithelialized and cuff inflammation was resolved. The timing of our repairs was determined largely by the time of referral to us. Our philosophy is that intervention should occur as soon as possible after diagnosis. However, if there is considerable cuff inflammation and induration at initial diagnosis (usually 7–10 days after hysterectomy), the patient is seen and reexamined weekly until the site appears optimal.
A previous attempt at fistula repair did not affect the fistula-closure success rate nor complicate the surgery itself. This was also reported by Lee et al in a series of 303 fistula repairs. 8 Use of interpositional tissue has been advocated by some surgeons, 3 but its value in uncomplicated fistulae repair is unclear and similar cure rates are reported whether or not interposition flaps were routinely used. We did not evaluate patients with urodynamics preoperatively because no patient complained of significant voiding dysfunction symptoms, stress urinary incontinence, or urge incontinence symptoms. However, a report by Hilton suggests that urodynamic abnormalities may be present in patients with vaginal-vault vesicovaginal fistulae. 9 Like Hilton, we did not find that de novo urinary incontinence or voiding dysfunction occurred after vault vesicovaginal fistula repair.
Transvaginal vaginal-cuff excision fistula repair is a successful procedure for posthysterectomy fistula closure. It is minimally invasive, preserves vaginal function, and is not associated with significant urinary complaints.
1. Leng WW, Amundsen CL, McGuire EJ. Management of female genitourinary fistulas: transvesical or transvaginal approach? J Urol 160: 1995, 1998.
2. Cruikshank SH. Early closure of posthysterectomy vesicovaginal fistulas. South Med J 81: 1525, 1988.
3. Blaivas JG, Heritz DM, Romanzi LJ. Early versus late repair of vesicovaginal fistulas: vaginal and abdominal approaches. J Urol 153: 1110, 1995.
4. Uebersax JS, Wyman JF, Shumaker SA, et al. Short forms to assess life quality and symptom distress for urinary incontinence in women: the incontinence impact questionnaire and the urogenital distress inventory. Neurourol Urodyn 14: 131, 1995.
5. Bump RC, Mattiasson A, Bo K, et al. The standardization of terminology of female pelvic organ prolapse and pelvic floor dysfunction. Am J Obstet Gynecol 175: 10, 1996.
6. Raz S, Bregg KL, Nitti VW, et al. Transvaginal repair of vesicovaginal fistula using a peritoneal flap. J Urol 150: 56, 1993.
7. Carr LK, Webster GD. Abdominal repair of vesicovaginal fistula [editorial]. Urology 48: 10, 1996.
8. Lee RA, Symmonds RE, Williams TJ. Current status of genitourinary fistula. Obstet Gynecol 72: 313, 1988.
© 2001 Lippincott Williams & Wilkins, Inc.
9. Hilton P. Urodynamic findings in patients with urogenital fistulae. Br J Urol 81: 539, 1998.