The success of fibrin sealant, even if modest, to treat anal fistulas established the concept of treating anal fistula by insertion of a material into the fistula in order to promote healing. Fibrin glue was used to provide a biologic scaffolding, which promoted fibroblast ingrowth and scar formation. This treatment strategy became the basis for application of newer materials, such as anal fistula plugs. The first commercially available product was the Biodesign^® anal fistula plug, formerly Surgisis^® anal fistula plug (Cook Medical, Bloomington, IN). The Biodesign^® material is made from porcine small intestine submucosa. The manufacturing process removes the cells from the material, leaving a collagen rich extracellular matrix which facilitates ingrowth of vascularized tissue. Biodesign^® had been used for abdominal wall reconstruction in the setting of contaminated wounds. It appeared to be relatively resistant to infection, which suggested that it would be a suitable choice for obviously contaminated anal fistula. The plug is simply inserted into the fistula, secured at the internal opening, and it then functions as a biologic scaffolding for scar ingrowth. The Biodesign^® is replaced by scar tissue in the months following implantation.

The successful application of the Biodesign^® anal fistula plug confirmed the viability of the fistula plug strategy. However, its success was not uniform. Surgeons believed that some of the failures were due to early plug extrusions, mismatch between the size of the plug and the diameter of the fistula, and performance of the Biodesign^® material in the setting of heavy bacterial contamination. In order to address these specific concerns, a second anal fistula plug was developed. The GORE^® BIO-A^® Fistula Plug (W.L Gore & Associates, Newark, DE) is made from a non-woven web of polyglycolic acid: trimethylene carbonate fibers that form a 3D matrix of open, highly interconnected pores. This entirely synthetic material is hydrolyzed within 3–6 months following implantation. After the material is hydrolyzed, no prosthetic material remains in the fistula. In addition, rather than a simple conical design, the BIO-A^® fistula plug was designed with two significant changes. First, a disk of the material was placed at the internal opening. This facilitated a more secure anchoring of the internal aspect of the plug to the internal sphincter. Second, the body of the plug is composed of six tubes of material. This allows the surgeon to remove a number of tubes so as to approximate the overall girth of the plug to the diameter of the fistula tract.

Fistulas treated with plugs are most often drained with a loose seton for 6–12 weeks preoperatively. The fistula is managed as with fibrin sealant injection: careful probing, gentle debridement, and flushing. The Biodesign^® plug is dragged with a suture from the internal opening to the external opening. The body is conical, and so gentle traction is applied until the plug fits snugly within the fistula. It is secured with absorbable sutures at the internal sphincter. The BIO-A^® plug is also pulled through the internal opening with a suture. The disk of material at the internal aspect can either be secured with simple suturing, or buried in a small submucosal pocket. Patients are typically discharged home postoperatively with instructions for light activity. Sitz baths avoided so as not to promote hydrolysis of the BIO-A^® material.

As the fistula plug procedures do not involve division of sphincter fibers, the risks of incontinence should be minimal. In fact, there is little evidence that either type of fistula plug causes postoperative incontinence. As with fibrin sealant, most publications do not document formal evaluation with validated survey instruments or physiologic testing, but the number of patients with postoperative incontinence is minimal. O’Riordan et al. identified 0 patients with worsening continence in their review of 8 trials including 196 patients. Narang [47] performed a systematic review of 6 trials including 221 patients treated with BIO-A^® fistula plug and identified worsening continence in 5.8% of patients. Interestingly, Stamos [45] et al. evaluated patients with the Wexner Score, and found that postoperative scores improved after treatment with the BIO-A^® plug in the overall population. However, 11% of patients documented a worsening score. The authors speculate that patients may not be capable of accurately differentiating true incontinence from fistula drainage.

Fistula plugs have been associated with postoperative abscess. The external fistula opening is specifically left open to drain; in fact, should be further opened to promote drainage of the contaminated tract. Trials included in Table 3.2 report abscesses to occur 5–14% [30, 35, 39, 42, 45, 46] Garg [48] conducted a systematic review of 25 publications including 317 patients, which identified 10% rate of abscess in trials reporting complications.

Both types of anal fistula plugs are at risk for early postoperative extrusion of the plug itself. Part or the entire plug may expel prior to adequate tissue ingrowth. The exact time required for the plugs to remain in place is unknown, however patients presenting with part or the entire plug within the first 2 weeks postoperatively seem very unlikely to obtain the full benefit of the operation. In the case of partial plug fallout, it is unclear if the benefit of the operation persists if the plug remains at the internal opening. This is likely to be the pathologic part of the fistula, the abnormality that keeps the fistula patent, however, so few such cases are reported that no conclusions may be drawn. Early fallout may be related to inadequate techniques of securing the plug at the internal opening. A consensus conference was convened in order to standardize best practices related to the Biodesign^® plug [49]. The BIO-A^® plug may be directly sutured or buried in a submucosal pocket. Inadequate evidence exists to recommend one technique over the other. Most of the trials represented in Table 3.2 suggest a 5–10% rate of early plug extrusion. Van Koperen [34] reports a fallout rate of 41%, however this was in a small number of patients. The trial was prematurely closed due to unacceptable results. The systematic reviews of patients treated with Biodesign^® document an extrusion rate of 19% in 432 patients [48] and 8.7% in 530 patients [46]. The systematic review of patients treated with BIO-A^® reveals an extrusion rate of 5% in 221 patients [47].

Some patients experience a significant inflammatory response to the plug material. This is a risk factor related to any foreign material implantation. This may be an allergy in the case of porcine tissue, or a simple inflammatory response to the BIO-A^® material. This has not been formally characterized enough to understand the relationship to healing rates.

Similar to fibrin sealant, the insertion of an anal fistula plug is a low risk procedure. Typically, if the plug fails, the wound simply continues to drain. There is some degree of inflammatory response to either material, and this may be variable between patients, but there is not significant disruption of the sphincter muscle fibers or the nerves. There is minimal disfigurement of the normal anatomy of the anus. Therefore, insertion of anal fistula plugs may be considered a first line therapy, provided the surgeon appropriately counsels the patient regarding the expected healing rate. It is also an excellent first line therapy in patients with marginal or impaired continence, chronic diarrhea, or Crohn’s disease.

Flap coverage of a chronic wound is a commonly employed wound closure strategy. This principle of transfer of normal tissue to an abnormal, nonhealing tissue bed has also been applied to anal fistula. The internal fistula opening is believed to be the pathological site. If healthy tissue can be transferred onto and subsequently heal the internal opening, then the remainder of the tract should eventually heal. This may be accomplished by transfer of tissue from within the rectum, as endorectal advancement flap, or from the perianal skin, as a dermal flap. Flaps are technically more challenging to perform than fibrin sealant or fistula plugs. Thorough knowledge of the relevant anatomy, and precise dissection of the appropriate tissue plans are critical to perform a safe procedure. This can be particularly difficult in patients that have undergone previous operative procedures. Fibrosis surrounding the fistula can make dissection of the flap quite challenging.

Endorectal advancement flaps were the initial flaps described to treat anal fistula, and have been widely adopted. Clinical experience, and therefore published data, are extensive. The endorectal advancement flap is designed as a broad based flap originating well proximal to the internal fistula opening. The endorectal advancement flap is often called a mucosal flap, which is somewhat of a misnomer, as the flap is generally thicker than mucosa alone. A purely mucosal flap is at high risk to become ischemic and retract. The flap should in fact incorporate the submucosa and even a small amount of muscle. Injection of an epinephrine containing solution into the submucosal space can facilitate this dissection and aid hemostasis. A broad based flap is prepared and generously mobilized. The internal opening should be debrided, and sutured closed with absorbable sutures. The distal aspect of the flap is debrided and finally secured just distal to the internal opening.

The dermal flap (also called dermal island flap, island flap, or anocutaneous flap), described by Nelson et al. [73], also transfers normal tissue to the internal opening. This flap however utilizes perianal skin. The fistula is flushed, and the internal opening closed as with endorectal advancement flaps. Either a “U” shaped, or teardrop-shaped incision is then created at the perianal skin in order to create a broad based flap of skin and subcutaneous fat, including the fistula tract itself. This tissue is then advanced into the anal canal, and sutured overlying the internal opening. The external wound may be closed, or left open to avoid tension on the flap.