Outcomes of primary repair of anorectal and rectovaginal fistulas using the endorectal advancement flap

PURPOSE: The endorectal advancement flap is a surgical procedure used in the treatment of anorectal and rectovaginal fistulas. There is a wide range of success rates published in the literature. This study was undertaken to examine the success rate of primary endorectal advancement flap in our own institution. We attempted to identify factors that influence the rate of healing. METHODS: A retrospective review was performed on 105 patients (43 males) who underwent their first endorectal advancement flap at our institution between January 1, 1994, and June 30, 1999. Ninety-nine patients were available for follow-up. Sixty-two patients had anorectal and 37 had rectovaginal fistulas. The causes of fistula included cryptoglandular (48 patients), Crohn's disease (44), obstetric injury (5), trauma (1), and other (1). RESULTS: The median follow-up was 17.1 (range, 0.4-66.9) months. The median age was 42 (range, 16-78) years. Recurrence was seen in 36 patients (36.4 percent); thus, the primary rate of healing was 63.6 percent. Factors that were associated with higher rates of success were increased age (P = 0.011), greater body surface area (P = 0.012), history of incision and drainage of a perianal abscess preceding advancement flap (P = 0.010), previous placement of a seton drain (P = 0.025), and short duration of fistula (P = 0.003). Factors that negatively influenced the healing rate of the flap were the diagnoses of Crohn's disease (P = 0.027) and rectovaginal fistula (P = 0.002). Length of hospitalization, discharge on oral antibiotics, and the presence of a diverting stoma did not influence the rate of healing. Prednisone was associated with a distinct trend toward failure, with none of the patients on high-dose prednisone (greater than 20 mg/day) having achieved long-term healing. No fistulas recurred after a period of 15 months. CONCLUSION: The endorectal advancement flap is an effective method of repair for both anorectal and rectovaginal fistulas, even though the success rate may not be as optimistic as in some other published studies. Patient selection is imperative, realizing that a higher rate of failure may be present in Crohn's disease and rectovaginal fistulas. Control of sepsis before endorectal advancement flap with drainage of a perianal abscess and/or seton placement, whenever possible, is indicated.