Purpose: The many options for treatment of anal fistulas indicate that no one option has consistent results. The aim of this study was to analyze the efficacy of the Cook Surgisis® AFP™ anal fistula plug for the management of complex anal fistulas in our institution.
Methods: A retrospective review of all patients who underwent treatment for complex anal fistulas using Cook Surgisis® AFP™ anal fistula plug between October 2005 - 2007 was undertaken. Patient's demographics, fistula etiology and success rates were recorded. The plug was placed in accordance with the company’s guidelines. Success was defined as closure of all external openings, absence of drainage without further intervention, and absence of any inflammation process.
Results: Thirty three patients underwent 49 plug insertions. The median age of the patients was 44.35 (+/-13.18) years; 18 females and 15 males. The fistula etiology was cryptoglandular in 20 (61%) patients and Crohn's disease-associated in the 13(39%) patients (23 high transphincteric; 8 low anal and 2 rectovaginal). Two patients had concomitant treatments; one underwent concomitant anal advancement flap at the time of plug placement and failed, the other underwent concomitant fibrin glue in multiple fistulas and healed. There were 11 smokers, 4 patients with diabetes and 7 patients on steroids. The mean follow up was 251.12 days (+/- 163.06). One patient was lost to follow up. Twenty one patients had previously undergone failed surgical therapy to cure their fistula (anal advancement flap in 8, fistulotomy 10 and fibrin glue in 3). Twenty nine patients had draining setons in-situ at the time of plug placement. The mean time the seton was in place before insertion of plug was 55 days (range 31-143 days).The overall success rate was 11/48 (41%). The success rate after first attempt was 8/33(24%), 2/14(14%) after second attempt and 1/2(50%) after third attempt. Two of the 22 (15%) Crohn's disease-associated fistulas healed and 9 out of 26(12.5%) procedures resulted in healing of cryptoglandular fistulas. The reasons for failure were sepsis in 31 procedures (87.0%), plug dislodgement in 6(13.0%) procedures. Other predictor factors were race (African-American patient's healing rates were higher p=0.009), low fistula (internal os below dentate line) did better than high fistula, also presence of seton had a better outcome (p=0.05).
Conclusions: Anal fistula plug is associated with a low success rate than previously reported. African-Americans and presence of seton at time of plug placement were predictors of good outcome.