Since the classic paper published by Crohn and colleagues in 1932 describing the chronic inflammatory process of the bowel there have been multiple articles published on the complications of this illness. The description of perianal fistula was followed 6 years later with the incorrect concept that the inflammatory process extended from the bowel down to the perianal area.
There are multiple problems that can affect the perianal area, including simple skin tags, fissures, hemorrhoids, high and low fistulas, strictures, rectovaginal fistulas, and, finally, neoplasia. Severe perianal skin excoriation can result in significant discomfort and impaired quality of life. The main purpose of this article is to review the perianal complications with a focus on management of fistulas.
Incidence and Pathogenesis
There has been great variation in the reported incidence of fistula in patients with Crohn’s disease ranging from a low of 17% up to almost 50%. There have been several population-based studies reporting that the overall incidence of fistulas was 35% in patients with Crohn’s disease with perianal fistulas occurring in 20%. A cumulative incidence of fistulizing Crohn’s disease appeared in 33% of patients after 10 years and 50% of patients after 20 years. Perianal fistulas were much more common in patients with colonic disease (41%) versus those with ileal disease (12%). The highest incidence occurred in those patients with Crohn’s disease involving the colon and rectum. It is interesting to note in this population-based study that recurrent fistulas occurred less frequently in patients who were placed on maintenance therapy with an immunosuppressive agent.
Often confusing to the practicing physician is the development of a perianal fistula in a patient who demonstrates no involvement of their bowel with Crohn’s disease. In 10% of patients the fistula may precede the onset by several years. Another confusing diagnostic problem is the female patient who develops an abscess in the labial area and a diagnosis of a Bartholin cyst is incorrectly made. Gynecologists should be fully aware that patients with Crohn’s disease may develop fistulas in this area.
Regarding pathogenesis, it is now quite clear that fistulas occurring in the perianal area do not arise from the small intestine or the sigmoid colon but rather develop locally, starting either as a deep penetrating ulcer in the anus or rectum or secondary to an anal gland abscess. There are various classification schemes for describing perianal fistulas, with some surgeons classifying them as either low or high, depending upon whether they are above or below the dentate line. What has been accepted recently as more precise is the Parks classification, which uses the external sphincter as the point of reference. This classification describes five types of fistula, which include the following:
- superficial (low),
- intersphincteric (low or high),
- transphincteric (low or high),
- suprasphincteric (high),
- extrasphincteric (high).
Following this classification, fistulas can be more specifically identified as simple or complex. Surgical management will depend on the site and classification of the perianal fistula.
Diagnostic Modalities
Medical Management
General
As noted, an adequate history and physical examination is essential. If the patient has an abscess, it should be adequately drained before proceeding with medications. Not all patients require a colorectal surgeon, that is if the abscess has either drained spontaneously or been drained by incision. On the other hand, if there is persistent rectal pain and/or tenderness, or there are multiple draining sites, then a colorectal surgeon should be consulted to be certain that all pus has been adequately drained. In this situation, an examination under anesthesia and/or MRI is indicated.
Although there have been multiple trials conducted throughout the world using sulfasalazine or mesalamine in the treatment of Crohn’s disease, there have never been any studies designed to study or reports indicating that these agents are efficacious in the treatment of perianal fistula. In the vast majority of cases, mesalamine or sulfasalazine will be maintained as part of the management of the active bowel disease, but again there is no reason to institute these agents if the major problem is perianal fistula.
In looking at the control trials using corticosteroids in the treatment of Crohn’s disease, neither the National Cooperative Crohn’s Disease Study, nor the European Cooperative Crohn’s Disease Study randomized patients for fistula, and there is, therefore, no data available for this subgroup of patients. In these two steroid placebo controlled trials the only deaths occurred in patients who were receiving steroids and who had internal fistula with the subsequent development of an abscess and overwhelming sepsis. Multiple controlled trials have been performed evaluating a newer steroid, budesonide, in the treatment of Crohn’s disease.
Efficacy has been demonstrated. However, all patients with a fistula were excluded from these placebo controlled trials. There is, therefore, no control data suggesting that steroids should be instituted in patients who developed perianal complications and fistula. I have experienced multiple patients with longstanding disease going on to develop fistulas and abscesses for the first time after steroids are introduced.
I have also seen a significant lack of healing when patients are taking steroids and attempts are made for closure with immunomodulatory agents. It is my personal opinion that steroids are “contraindicated” when trying to manage perianal fistulas, and if the bowel symptoms allow, I quickly withdraw them from the therapeutic regimen.
Effective Therapeutic Agents
Miscellaneous Medical Therapies
Through the years there have been several reviews evaluating the benefit of various therapies in Crohn’s disease patients with fistulization. A commonly used agent is total parenteral nutrition with bowel rest. In my experience this is never effective if there is active Crohn’s disease in the bowel, whereas it maybe effective in postoperative fistula management. Other agents used have been elemental diet, thalidomide, and hyperbaric oxygenation. Control trials are required before any of these agents are considered effective therapy and we should use those agents that have shown efficacy.
Summary
In the management of perianal Crohn’s disease, when a fistula develops appropriate diagnosis should be made by history, physical, colonoscopy, and small bowel series. If the fistula is simple, it should be drained and treated with antibiotics. If there is healing, it is my opinion that one or two antibiotics should be maintained for at least 6 to 12 months. I would use low doses of metronidazole 250 mg 3 times daily or ciprofloxacin 250 to 500 mg twice daily. If there is recurrence of the fistula and it drains freely, I would then add 6-MP/AZA to the regimen.
If pain persists or the fistula occurs in a new site, I prefer to obtain an examination by a colorectal surgeon with an examination under anesthesia. If the surgeon is not certain of the status, then an MM should be added to the regimen. Setons should be placed to allow time for the 6-MP/AZA to promote healing. If the fistula heals with 6-MP/AZA, I would maintain this agent for 5 years or longer. If healing is not induced or maintained with 6-MP/AZA, a 3-course infusion of infliximab is indicated.
I do not continue infliximab every 8 weeks, but rather maximize all therapy by giving adequate doses of antibiotics and 6-MP, then treat only after recurrence. It is my experience that not all patients will relapse and require infliximab every 8 weeks. If infliximab fails, a 7- to 10-day course of IV cyclosporin in a dose of 4 mg/kg is indicated.
Failure to respond to all of the above medical therapies would dictate proctectomy and a total colectomy if the Crohn’s disease is active in the remaining colon.
TABLE. Perianal Fistulas—Crohn’s Disease
Effective for closure |
B-MP/Azathioprine |
Methotrexate |
Cyclosporin |
Infliximab |
Effective for clinical response |
Ciprofloxacin |
Metronidazole |
Not effective |
Mesalamine |
Steroids |