The initial diagnosis was ulcerative colitis in 65 patients, familial adenomatous polyposis in 4 patients, and Crohn’s disease in 1 patient. Among the 65 patients with a diagnosis of ulcerative colitis, 13 (18.5%) of them later had their diagnosis changed to Crohn’s disease. Sixty-four percent (9/14) of the patients with a final diagnosis of Crohn’s disease had successful closure of the fistula. Five patients with Crohn’s disease had a history of biologic use within 3 months of surgery.
Postoperative complications after the intended closure were seen in 13 patients. These were atrial fibrillation in 1 patient, urinary retention in 1 patient, urinary tract infection in 1 patient, postoperative ileus in 1 patient, local surgical site infection in 3 patients, perineal wound dehiscence in 2 patients, and postoperative bleeding in 2 patients. The first patient with postoperative bleeding was managed with transfusion, and the second patient underwent reoperation. In the operative report there was no active bleeding. Two patients had multiple complications. One patient had postoperative ileus and hematoma, and a second patient had urinary tract infection, postoperative ileus, and pulmonary embolism. No patients were found to have an anastomotic leak after the pouch reconstruction and the mortality rate was 0.
Overall, 58 (82%) of 70 patients had undergone a previous surgery for fistula control or closure; 45 of those procedures were performed with an intent to close the fistula. Successful closure rate on the last procedure of those who underwent any previous surgery for fistula control or closure was 52% (30/58). At the time of surgery, 52 of 70 patients had a previous stoma or had a new stoma constructed. Twenty-seven patients had a previous stoma, 10 had revision/reconstruction of stoma, and 15 patients had a new stoma constructed during the surgery to close the PVF. The overall successful closure rate in patients with a stoma before/at the time of the definitive repair was 65.3% (34/52), and this rate was 27.7% (5/18) in patients without a stoma.
Mean age at the time of the planned definitive fistula closure surgery was 41.1 ± 12.4 years. Redo IPAA was the most common surgical treatment to close the fistula and was performed in 26 of 70 patients. The second most performed surgery was ileal pouch anal advancement flap in 23 of 70 patients. The overall successful closure rate was 56% in our study group. Overall successful closure was 69% for redo IPAA and 61% for ileal pouch anal advancement flap. Procedure-specific closure rates are given in Table 2.
Fistula closure was analyzed based on the localization, that is, those arising from the IPAA versus those from the anal transitional zone or anal canal. Ileal pouch advancement flap had the highest success rate for fistulas above/at the IPAA, and pouch reconstruction was the most successful procedure for fistulas distal to the anastomosis (ie, in the anal transitional zone; Table 2).
PVFs, although not life threatening, may severely impact daily activities and quality of life. Previously different approaches for the surgical treatment have been reported with varying success rates.2–5,7 A previous report of our earlier surgical experience found healing rates of 57.7%, and results of the current study are similar to our previous experience, with a healing rate of 56.0%.5
When the fistula opening was arising from the IPAA as in 44% of the patients in this study, the procedure with the highest success rate was ileal pouch advancement flap. In contrast, the results of our previous series found the fistula opening was arising from the IPAA in 52% of the patients and the success rate with this approach was 66% when it was performed as a second procedure.5 Successful closure with this procedure can reach 61%; however, the reported series have small sample sizes.7,9–11 Based on this we currently consider ileal pouch advancement in distal fistulas arising from the anal transitional zone and distal.
A simultaneous transabdominal and transperineal approach with pouch reconstruction and fistula closure is a technically demanding procedure compared with local repairs. Redo IPAA was the procedure with the overall highest successful closure rate of 69%. Additional analysis showed that this procedure had the overall highest rate of successful closure at 71% for PVFs located distal to the anastomosis (ie, the fistula arose from the rectum, anal transitional zone, or dentate line; this distal tissue was totally removed and a mucosectomy performed with handsewn anastomosis). In another previous study from our institution, which included >500 patients, redo IPAA for ulcerative colitis was the preferred procedure for a failed IPAA, which included 85 patients with PVFs. We have noted at our institution that the number of redo IPAAs is increasing, which probably reflects a dedicated small number of surgeons treating this group of patients. This then allows for more concentrated experience and more familiarity with such a complex operation. We believe that this could explain the higher success rate of redo IPAA in the current study.12
Previous studies suggest that fistulas distal to the anastomosis should be managed with local procedures including ileal pouch advancement flap, and fistulas located above/at the anastomosis should be managed with a transabdominal approach.6,10,13 Results of the current study contradict these reports. Our increased comfort with redo IPAA may possibly explain this discrepancy. Also, when deciding on the surgical approach, many factors must be considered, not just location of the internal opening of the fistula. It also may reflect that if the tissue causing the problem is the retained rectum/anal transition zone, removal should be considered as in a redo pouch for a successful outcome.
In this study, 18% had their diagnosis changed to Crohn’s disease, and the successful closure rate was 64% in this group. Our previous study by Mallick et al5 reported higher rates of pouch failure in patients with Crohn’s disease at 52.7%. Tekkis et al14 identified Crohn’s disease as a risk factor in the occurrence of pouch-related fistulas in >1900 patients, which included 44 patients with PVFs. It is unclear why the rate has improved, but as our experience with pouches having Crohn’s disease increases, we may be excluding those who have a higher potential to fail.
Multiple repairs are common when surgically treating women with PVF, and recurrence is a frequently anticipated problem. In our study, 82% of the patients underwent previous surgeries including seton placement, ileal pouch advancement flap, and redo IPAA. Others have reported tissue interposition procedures as an effective and successful alternative for recurrent fistula, suggesting that they should be preferred as the first line of treatment; however, these studies have a small number of patients and require additional investigation.15,16
Patients with a stoma had twice the chance of having a successful closure compared with those where no stoma was used (65.3% vs 27.7%). This has led us to favor diversion of all patients when operating to close a fistula from a pouch to the vagina.
Our study has certain limitations. First, this is a retrospective review. We have no definitive algorithm for deciding on the surgical procedure, and therefore the choice is left up to the discretion of the surgeon. In addition, our measure of success is resolution of symptoms, and we perform an examination under anesthesia in select patients where there is a question of healing. Another limitation is the small number of patients included in the study. Although it is one of the largest in the literature, this problem is rare and reflects concentration of repair at a high-volume referral center with extensive experience. Another limitation is that many of the index pouch construction operations were performed at other institutions. Nuances of a technical factor during pouch construction are not known and may have some effect on the local tissue conditions and success rate of a closure procedure. We do not have the information regarding stapled/handsewn anastomosis for all of the index IPAAs, because some were performed decades before the final operation. We believe that this may be an important factor; however, because most of these patients underwent multiple surgeries after their index operation, formation of PVF and achieving success was likely multifactorial.
The best procedure to close a PVF remains to be elucidated. Although certain algorithms can be formulated to aid in choosing the surgical approach, each patient must be evaluated to determine an individual treatment plan. Management of these fistulas should be individualized, and the surgical approach to close the fistula should be selected based on patient expectations and health, localization of the fistula, the viability of the tissue, and expertise of the surgeon. Redo IPAA and ileal pouch advancement flap result in a relatively higher closure success rate.
1. Lolohea S, Lynch AC, Robertson GB, Frizelle FA. Ileal pouch-anal anastomosis-vaginal fistula: a review. Dis Colon Rectum. 2005;48:1802–1810.
2. Johnson PM, O’Connor BI, Cohen Z, McLeod RS. Pouch-vaginal fistula
after ileal pouch-anal anastomosis: treatment and outcomes. Dis Colon Rectum. 2005;48:1249–1253.
3. Wexner SD, Ruiz DE, Genua J, Nogueras JJ, Weiss EG, Zmora O. Gracilis muscle interposition for the treatment of rectourethral, rectovaginal, and pouch-vaginal fistulas: results in 53 patients. Ann Surg. 2008;248:39–43.
4. Burke D, van Laarhoven CJ, Herbst F, Nicholls RJ. Transvaginal repair of pouch-vaginal fistula
. Br J Surg. 2001;88:241–245.
5. Mallick IH, Hull TL, Remzi FH, Kiran RP. Management and outcome of pouch-vaginal fistulas after IPAA surgery. Dis Colon Rectum. 2014;57:490–496.
6. Maslekar S, Sagar PM, Harji D, Bruce C, Griffiths B. The challenge of pouch-vaginal fistulas: a systematic review. Tech Coloproctol. 2012;16:405–414.
7. Shah NS, Remzi F, Massmann A, Baixauli J, Fazio VW. Management and treatment outcome of pouch-vaginal fistulas following restorative proctocolectomy. Dis Colon Rectum. 2003;46:911–917.
8. Gonsalves S, Sagar P, Lengyel J, Morrison C, Dunham R. Assessment of the efficacy of the rectovaginal button fistula plug for the treatment of ileal pouch-vaginal and rectovaginal fistulas. Dis Colon Rectum. 2009;52:1877–1881.
9. Ozuner G, Hull T, Lee P, Fazio VW. What happens to a pelvic pouch when a fistula develops? Dis Colon Rectum. 1997;40:543–547.
10. Tsujinaka S, Ruiz D, Wexner SD, et al. Surgical management of pouch-vaginal fistula
after restorative proctocolectomy. J Am Coll Surg. 2006;202:912–918.
11. Lee PY, Fazio VW, Church JM, Hull TL, Eu KW, Lavery IC. Vaginal fistula following restorative proctocolectomy. Dis Colon Rectum. 1997;40:752–759.
12. Remzi FH, Aytac E, Ashburn J, et al. Transabdominal redo ileal pouch surgery for failed restorative proctocolectomy: lessons learned over 500 patients. Ann Surg. 2015;262:675–682.
13. Zinicola R, Wilkinson KH, Nicholls RJ. Ileal pouch-vaginal fistula
treated by abdominoanal advancement of the ileal pouch. Br J Surg. 2003;90:1434–1435.
14. Tekkis PP, Fazio VW, Remzi F, Heriot AG, Manilich E, Strong SA. Risk factors associated with ileal pouch-related fistula following restorative proctocolectomy. Br J Surg. 2005;92:1270–1276.
15. Troja A, Käse P, El-Sourani N, Raab HR, Antolovic D. Treatment of recurrent rectovaginal/pouch-vaginal fistulas by gracilis muscle transposition: a single center experience. J Visc Surg. 2013;150:379–382.
16. Rottoli M, Vallicelli C, Boschi L, Cipriani R, Poggioli G. Gracilis muscle transposition for the treatment of recurrent rectovaginal and pouch-vaginal fistula
: is Crohn’s disease a risk factor for failure? A prospective cohort study. Updates Surg. 2018;70:485–490.
Fistula repair; Pouch-vaginal fistula; Redo pouch
Supplemental Digital Content
© 2019 The American Society of Colon and Rectal Surgeons