The goal of surgical treatment for anal fistula is eradication of the fistula tract without compromising sphincter function. Fistulas can be classified as “simple” or “complex.” “Simple” fistulas are of cryptoglandular infection, are usually distal intersphincteric or distal transsphincteric and can be treated by lay open fistulotomy with reported success rates reported over 90 % [39]. Those “simple” fistulas that involve proximal (high) intersphincteric or transsphincteric tracks are more difficult to manage. Fistulotomy of such proximal anal fistulas is associated with lower healing rates and higher rates of incontinence. “Complex” fistulas including those arising from noncryptoglandular origin such as those associated with perianal Crohn’s disease, those persisting or recurring despite prior surgical interventions, and those of cryptoglandular origin that crosses > 30–50 % of the external sphincter, are anterior in a female, are associated with multiple tracts, develop in an individual with some degree of existing fecal incontinence, or occur in previously irradiated tissue (Table 43.3) [40]. Reported rates of incontinence after surgery for such “complex” anal fistulas vary from 0 to 25 % for flatus, up to 26 % for major fecal leakage, and as high as 63 % for minor and/or passive incontinence [41]. Female gender, type of surgery, prior fistula surgery, posterior internal opening, and horizontal extension have been variables associated with postoperative incontinence. Reported fistula recurrence rates range from 0 to 30 % and have been associated with a horseshoe tract, missed tracts, failure to identify the internal opening, prior surgery, and surgeon experience [41]. For distal (low) fistulas, it is generally accepted that the risk of incontinence is minimal and fistulotomy is advocated if less than one-third of the external sphincter is crossed by the fistula [42–44].

Several surgical techniques have been described to address high or complex anal fistula. The anatomy of these tracts can be defined in the operating room with fistula probes and/or with the aid of dyes or hydrogen peroxide. Alternatively, radiographic evaluation with either endoanal ultrasonography (with or without hydrogen peroxide injection) (Fig. 43.3) or magnetic resonance imaging (MRI) (Fig. 43.4) may prove helpful to identify the fistula and help quantify the amount of IAS and EAS involved by the tract and at potential risk for division [45–47].

Both draining and cutting setons may be used for high or complex fistulas. After defining the fistula tract with a probe, the surgeon may elect to use a combination of a seton placed through the tract and a partial sphincterotomy. A draining seton is used to assure complete resolution of associated abscesses and to induce fibrosis along the tract. When the inflammatory process has resolved, the seton may either be removed in hope that the fistula will go on to heal without further division of muscle or it can be removed and additional fistulotomy performed. Sometimes the draining seton is converted to a cutting seton or in the absence of significant associated abscess and inflammation, a cutting seton may be used instead of a draining seton in the first procedure. The cutting seton is gradually tightened to slowly divide the remaining involved muscle in the fistula tract. This theoretically allows scar to form as the seton is slowly “walked through the sphincter,” thus keeping the sphincter muscle intact and avoiding a wide gap as is noted when muscle is divided in one procedure.

Eradication of the fistula is reported to be 60–78 % with recurrence rates between 2 and 9 %. Although the cutting seton at one time was thought to preserve continence in comparison with direct division, reports have not confirmed this reputed advantage. Minor disturbance of continence occurs in 34–63 % of patients along with impaired anal manometry and postoperative deformity of the anal canal [41, 48]. Additionally, cutting setons are not well tolerated because of the discomfort associated with frequent tightening of the cutting seton. The two-stage seton fistulotomy results in similar rates of incontinence as the cutting seton with minor incontinence ranging from 54 to 66 % and major incontinence ranging from 4 to 26 % [49, 50]. Injection of fibrin glue or a collagen plug results in varying success rates ranging from 33 to 88 % with minimal associated morbidity or alteration of continence [51–53].

Endorectal advancement flaps have been used to obliterate the internal fistula opening without division of the sphincter complex in an attempt to preserve continence. Following debridement of the chronic fistula tract(s), a flap of mucosa and submucosa with or without a portion of internal sphincter muscle is mobilized beginning distal to the internal opening of the fistula. The flap is mobilized proximally increasing its width to maintain good vascularity. The proximal dissection proceeds until the mobilized flap can be advanced distally over the internal opening of the fistula and a tension-free repair of the flap to the anorectum distal to the internal opening can be achieved (Fig. 43.5). Long-term studies on advancement flaps report recurrence rates as high as 33 % in cryptoglandular disease and up to 57 % in Crohn’s associated fistula. Prior attempts at repair of the fistula have been associated with increased incontinence following advancement flaps [41]. This may be due to inadvertent sphincter injury with retractors, inelastic tissue secondary to scarring, and direct injury to the internal sphincter with mobilization. Identified key steps for successful flaps include correct identification of the fistula tract and internal opening. Sepsis must be resolved and the tract should be dry and fibrotic. Draining setons should be used liberally as a first-stage procedure to ready the operative field for an advancement flap. The external opening should be enlarged to prevent premature closure of the external opening, which could lead to a postoperative track abscess which may necessitate through the repair [54].

The Ligation of the Intersphincteric Fistula Tract (LIFT) procedure is another sphincter-sparing technique that involves identification and ligation of the fistula tract in the intersphincteric groove. Success rates range from 57 to 94 % [55, 56]. Vergara-Fernandez et al. (2013) performed a review of the current LIFT literature where the primary outcomes included fistula healing rates, mean healing time, and patient satisfaction. Eighteen studies were included in the review with an N of 592. The mean healing rate was 74.6 %. Several risk factors for failure were identified and included obesity, smoking, multiple previous operations, and the long fistula tracts. Mean healing time was 5.5 weeks with a mean follow-up of 42.3 weeks. No de novo incontinence developed secondary to the LIFT procedure and patient satisfaction ranged from 72 to 100 % [57]. Currently, there is not enough evidence to assess the alleged improvement of LIFT variants.