Key points

Introduction

Perianal fistulas are common in the Crohn’s disease population and can be disabling to patients. Knowledge of fistulizing Crohn’s disease has grown immensely over the past 75 years and therapies have improved greatly. This article reviews fistulizing Crohn’s disease and examines the current strategies of management including medications, endoscopy, and surgical care.

Introduction

Perianal fistulas are common in the Crohn’s disease population and can be disabling to patients. Knowledge of fistulizing Crohn’s disease has grown immensely over the past 75 years and therapies have improved greatly. This article reviews fistulizing Crohn’s disease and examines the current strategies of management including medications, endoscopy, and surgical care.

Anatomy

Before embarking on a discussion of perianal fistulizing disease, it is important to understand the anatomy of the area. Perianal anatomy is complex and involves the pelvic floor musculature and the gastrointestinal tract. As seen in Fig. 1 , the anal canal is composed of epithelial lining, subepithelium, supporting tissues with intertwining neuronal networks, and specialized musculature including the pelvic floor and anal sphincter complex. Within the lumen, the upper anal canal is composed of the transitional and columnar epithelium of the rectum. This changes to the squamous anal epithelium at the dentate line. At the dentate line, there are anal columns and crypts. The bases of crypts may contain anal glands that then may penetrate the supporting tissues including the intersphincteric space. The anal sphincter complex is composed of the internal anal sphincter and the external anal sphincter. The internal anal sphincter is the thickened terminal extension of the circular muscle of the rectum and the external anal sphincter is a tube of striated muscle extending from the puborectalis muscle. Fig. 2 shows the interplay of the sphincter complex and the rest of the pelvic floor.

Anatomy of the anal canal.

( From Standring S. Lower intestine. In: Standring S, editor. Gray’s anatomy, vol. 41. Philadelphia: Elsevier; 2016. p. 1136–59; with permission.)

Diagram of the pelvic floor and anal sphincter complex.

( Modified from Parks AG, Gordon PH, Hardcastle JD. A classification of fistula-in-ano. Br J Surg 1976;63(1):1–12; with permission.)

Epidemiology

The first record of a granulomatous fistula was made by Gabriel in 1921 during a series of histologic investigations of 75 patients with rectal fistulae. In setting out to examine fistulae related to tuberculosis, he discovered that there were some fistulae with granulomas without evidence of tubercle bacilli. In 1932, Crohn’s and Ginzburg’s original paper describing regional ileitis did not mention perianal fistulas. Bissell in 1934 and Penner and Crohn’s in 1938 were the first to describe cases of patients with the clinical entity of Crohn’s disease and concomitant perianal fistulae. Morson and Lockhart-Mummery then described the histologic noncaseating giant-cell lesions in perianal fistulae to be similar to those seen in luminal Crohn’s disease in 1959. After this, further case series in the 1960s and 1970s firmly established that perianal fistulae could be a manifestation of Crohn’s disease.

Since the original descriptions of perianal Crohn’s disease, multiple population-based and referral-center-based studies have shown that perianal disease is common. In population-based studies from the United States, Canada, and Sweden, the rates of perianal fistulas in patients with Crohn’s disease range from 21% to 28%. Referral-center-based studies cite frequencies of perianal fistulas ranging from 22% to 40% in patients with Crohn’s disease. Perianal disease has been found to be more common in patients with distal luminal Crohn’s disease, such as colonic or ileocolonic disease, versus isolated ileal disease. The risk of developing a perianal fistula increases over time after diagnosis of Crohn’s disease with a 21% cumulative risk after 10 years and a 26% cumulative risk after 20 years as seen in Fig. 3 . It is also important to note that perianal disease can be the only manifestation of Crohn’s disease in about 5% of patients.

Cumulative incidence of overall fistulas ( solid line ) and perianal fistulas ( dashed line ) among 176 Olmsted County, Minnesota residents diagnosed with Crohn’s disease from 1970 to 1993.

( Modified from Schwartz DA, Loftus EV, Tremaine WJ, et al. The natural history of fistulizing Crohn’s disease in Olmsted County, Minnesota. Gastroenterology 2002;122(4):875–80.)

This population of patients has also been found to have slightly different characteristics than other patients with Crohn’s disease. Patients with perianal disease have more disabling Crohn’s disease after 5 years of diagnosis, especially if they also require steroids and are younger than the age of 40 at diagnosis. A recent cross-sectional analysis of 333 patients with Crohn’s disease on infliximab found that patients with perianal fistulas were almost three times more likely to be African American or Hispanic than white. Patients with perianal fistula are more likely to have extraintestinal manifestations of their inflammatory bowel disease, such as arthritis, oral ulcerations, or skin manifestations. They are more likely to develop penetrating or structuring complications including a five-fold increased risk of developing a luminal fistula and a two-fold increased risk of requiring surgery. Finally, patients are 2.3 times more likely to be steroid resistant if they have perianal disease.

Pathogenesis

The cause of perianal fistulas in Crohn’s disease is not entirely clear; however, multiple theories exist. One of the oldest theories of fistula pathogenesis involves a mucosal defect, such as a persistent infection or ulceration that then penetrates through the wall of the anal canal. After the track is established, it is then thought that the pressure of the fecal stream maintains this opening. Although the initial mucosal defect from this theory is thought to initiate fistula formation, the mechanical forces of the fecal material is what has been thought to perpetuate the fistula itself. Another theory proposes that infection in the anal glands themselves is a cause of fistula. Anatomically, the glands have ducts that can penetrate the internal anal sphincter and promote spread of infection into the intersphincteric space, the external anal sphincter, or even to the skin. On a cellular level, it is thought that the epithelial-to-mesenchymal transition (EMT) plays a role in fistula formation. EMT is a process whereby an epithelial cell loses its defining properties including certain cell contacts and polarity and takes on mesenchymal cell properties including motility and migration. EMT occurs physiologically during organ development and wound healing and pathologically in tumor growth and fibrosis. In Crohn’s disease fistula tissue studies, cytokine profiles (increased transforming growth factor-β, interleukin-13) and increased matrix remodeling enzyme concentration support the idea that EMT plays a role.

Classification

Multiple classification systems have been used to define and describe the extent of a patient’s perianal disease. It is important to know classification and apply this to patients because treatment algorithms depend on correct classification. It also allows for a common language between gastroenterologists and surgeons in describing fistula anatomy. The simplest and oldest classification includes defining the fistula as high or low based on the dentate line. A fistula that enters the rectum above the dentate line is considered a high fistula, whereas one that opens below the dentate line is a low fistula. Through further characterization of the anal canal and fistulas, additional more complex classification systems were developed. Two classifications are currently widely used, the Parks classification and the American Gastroenterological Association proposed classification system. The Parks classification was developed by Parks and colleagues in 1976 and uses the external anal sphincter as the main point of reference for five main groups of fistulae. Fistulas are sorted into intersphincteric, transsphincteric, suprasphincteric, extrasphincteric, and superficial. Intersphincteric fistulas do not involve the external anal sphincter and also fit with the previously described low fistula. Transsphincteric fistulas pass through the external anal sphincter. Suprasphincteric fistulas pass over the external anal sphincter and through the pelvic floor muscles (puborectalis, levator ani). Finally, extrasphincteric fistulas are outside the external sphincter complex altogether and pass through the rectal wall, pelvic floor muscles, and ischiorectal fossa. Superficial fistulas do not involve either the internal or external sphincter. In the 2003 American Gastroenterological Association technical review of perianal fistulas, an additional classification was proposed combining physical examination and endoscopic evaluation to categorize fistulas as simple or complex. Here, a simple fistula would be one that occurs low in the anal canal with a single external opening and no evidence of perianal abscess, rectovaginal fistula, or anorectal stricture. A complex fistula has a high origin above the dentate line; involves a significant portion of the external anal sphincter; and may have multiple external openings, pain, evidence of abscess, rectovagnial fistula, and stricture; and may have active rectal luminal disease on endoscopy.

Evolution of the role of the gastroenterologist

Throughout the history of perianal Crohn’s disease, the gastroenterologist’s role has evolved from bystander to complex caregiver. In the 1970s, the management of perianal fistulas involved either observation or surgery. Fistulas were observed if asymptomatic. If treated, fistulas were either surgically laid-open or drained. Rectal excision was also performed in certain cases. In the 1980s, it was recognized that perianal lesions may heal if luminal Crohn’s disease was successfully treated and that antibiotics may be beneficial; however, direct surgical management was still the mainstay of treatment, especially if there was evidence of an abscess. During this time, diversion of the fecal stream with either an elemental diet or a loop ileostomy was controversial but becoming popular. By the 1990s, surgical therapy had advanced to include drainage, fistulotomy, fistulectomy, seton placement, mucosal flaps, fecal diversion, and proctectomy. In a review of 224 patients with anorectal complications of Crohn’s disease from 1984 to 1999, a total of 200 patients underwent surgical procedures for management of their disease and patients with active rectal disease were noted to have a higher rate of proctectomy. With the advent of these more advanced surgical procedures, and knowledge of disease, medications, and detailed noninvasive tools, the gastroenterologist’s role has changed. Currently, the gastroenterologist participates in diagnosis, investigation, and classification of disease through endoscopic assessment of luminal disease, endoscopic ultrasound (EUS) assessment of the perianal area, and coordination with radiologists for other imaging modalities. The gastroenterologist also directly manages medical therapy including biologic therapy and antibiotics. Finally, the gastroenterologist participates in coordination of care with surgical colleagues.

Diagnosis and investigation of perianal disease

Accurate diagnosis of the perianal process allows the gastroenterologist to classify fistulas appropriately and apply evidence-based care to patients. Diagnosis should involve a detailed examination, imaging, clinical assessment, and assessment of rectal inflammation.

The gold standard for diagnosis of perianal fistula has been the examination under anesthesia (EUA). Performed by a surgeon, the examination involves a digital rectal examination in combination with probing of the perianal area to define fistula tracts. For many years, management decisions regarding perianal Crohn’s disease relied solely on the EUA. However, correct identification of all present abnormalities during EUA is difficult. A study by Van Beers and colleagues in 1994 showed that accuracy of a digital rectal examination to define fistulas was only around 62%. Radiologic advances have allowed further investigation into fistulous networks and identification of more complex fistulas or perianal abscesses. Therefore, imaging is now an integral part of the diagnosis of fistulizing disease. Imaging modalities used in perianal Crohn’s disease include fistulography, computed tomography, MRI, and EUS. Fistulography was the first imaging modality used. This involves injecting a small amount of radiopaque contrast into the fistula tract via a catheter that is inserted into the external fistula opening. This modality, however, has not been found to be extremely helpful, with only 16% to 50% accuracy when compared with operative findings; therefore its use has fallen out of favor. Computed tomography has also proven to be unreliable for baseline evaluation of fistulous networks with an accuracy of 24% to 60% because of poor spatial resolution in the pelvis and inflammation that may cloud the image. Computed tomography is, however, useful to identify secondary complications of perianal Crohn’s disease, such as abscesses.

MRI and EUS have been shown to accurately diagnose fistulas and therefore have become the imaging modalities of choice in fistulizing Crohn’s disease. Both modalities have similar accuracies ranging from 75% to 100% in multiple studies. In a study of 34 patients with perianal fistulas comparing EUS, MRI, and EUA, the accuracy of all three modalities included EUS at 91%, MRI at 87%, and EUA at 91%. For the patient, any of these imaging modalities is sufficient and the decision on which to use depends on the local expertise of the gastroenterologist. If high-quality EUS is available, it can be performed at the time of standard endoscopic evaluation of luminal disease activity. Using an imaging modality, such as MRI or EUS, has been shown to improve outcomes for patients. Combining two modalities of diagnosis (EUS, EUA, or MRI) led to an accuracy of 100%. In a randomized study, patients who underwent EUA with EUS guidance had improved healing of disease versus patients who had EUA alone. In addition to their usefulness in diagnosis, EUS and MRI are also integral to surveillance of fistulizing disease. A single-center experience reported in 2012 noted that follow-up EUS influenced patient management in 86% of its patients with perianal Crohn’s disease.

Objective scores of clinical activity can be gathered using the Perianal Disease Activity Index, which uses a Likert scale to rate specific metrics of quality of life and perianal disease severity. Luminal activity should be assessed with standard endoscopy, such as flexible sigmoidoscopy. This is important because active inflammation may affect any available surgical options for treatment.

Our suggestion for the gastroenterologist in investigation of patients with symptoms of fistulizing Crohn’s disease is as follows: (1) perform a detailed history and physical examination, (2) perform endoscopy to assess disease activity, (3) perform either EUS or MRI as an imaging study to map out the fistulous process, and (4) refer for EUA for complete diagnosis and possible treatments.