Is the outcome of transanal advancement flap repair affected by the complexity of high transsphincteric fistulas?

Background: Transanal advancement flap repair for the treatment of high transsphincteric fistulas fails in 1 of every 3 patients. Until now no definite risk factors for failure have been identified. The question is whether the more complex fistulas, such as those with horseshoe extensions and associated abscesses, have a less favorable outcome.
Objective: Aim of the present study was to indentify whether more complex fistulas have a less favorable outcome.
Design: This study is a retrospective case series review.
Patients: Between 1995 and 2007 a series of 162 patients underwent endoanal MR imaging before transanal advancement flap repair. Two investigators, without prior knowledge of the surgical findings, reviewed all MR images.
Results: Lateral fistulas were identified in 5 patients. Because of the small number, these patients were excluded from further analysis. Posterior fistulas were identified in 119 patients (76%). These fistulas had 3 types of extensions: a direct course (36%), a classic horseshoe extension (23%), or an intersphincteric horseshoe extension (41%). The corresponding healing rates were 37%, 81%, and 73%. Anterior fistulas were observed in 23% of the patients. These fistulas had 2 types of extensions: a direct course (61%) or a classic horseshoe extension (39%). The corresponding healing rates were 60% and 52%. The healing rate of fistulas with a direct course was significantly lower than the healing rate of fistulas with a classic or intersphincteric horseshoe extension. Associated abscesses were found in 47% of the posterior fistulas and 5% of the anterior fistulas. Once adequately drained, these abscesses did not affect the outcome of transanal advancement flap repair.
Conclusion: The complexity of high transsphincteric fistulas does not affect the outcome of transanal advancement flap repair.