Perianal fistula, defined as an abnormal communication between the anal canal or lower rectum and the perianal or perineal skin, is among the more morbid manifestations of Crohn disease (CD). The development of a perianal fistula is usually accompanied by pain, fever, and purulent drainage, and may even be associated with fecal incontinence. The exact etiology of perianal fistulae in CD remains unclear, but it signifies a more aggressive and refractory disease phenotype. As a result, patients with fistulizing CD generally experience a lower quality of life than CD patients without perianal involvement.
Nearly one quarter of all patients with CD develop a perianal fistula, with fistulae being more common in patients with involvement of the rectum. Before the introduction of biologic agents, most fistulae required some surgical intervention, with more than one third of patients developing recurrent fistulae.
The introduction of antitumor necrosis factor (TNF)-α antibodies has given clinicians the most efficacious medication to date for treating perianal fistulae. Induction studies using infliximab at weeks 0, 2, and 6 for active CD perianal fistulae resulted in cessation of drainage of all the fistulae present in 55% of patients who were randomized to the 5 mg/kg dose compared with only 13% of those who received placebo ( P = .001). However, fistulae usually start to drain again if the anti–TNF-α medication is discontinued. Studies looking at maintaining cessation of fistula drainage utilizing the 3 anti–TNF-α agents currently available (infliximab, adalimumab, and certolizumab) for CD have yielded similar results with about 36% to 39% of patients able to maintain cessation of drainage over a 26- to 54-week study.
Several factors likely contribute to the low maintenance rates of fistula healing. Perhaps the 3 most significant factors contributing to the high fistula recurrence rates after initiating anti–TNF-α medications are (1) not utilizing surgical intervention to maximize the effect of the medications, (2) failure to initially identify all of the fistulae or abscesses present and thus the lack of control of fistula healing when using anti–TNF-α agents, and (3) the premature removal of setons before a fistula is completely healed.
The body’s natural tendency is to try to close the external cutaneous opening of the fistula. This process is accelerated when utilizing anti–TNF-α agents. When this occurs, an abscess and/or secondary or tertiary fistula branch can develop. Indeed, the Present and ACCENT 2 studies with infliximab showed that the rate of abscess formation was high with use of the anti–TNF-α agents (11% and 15%, respectively).
Two retrospective studies showed that the durable fistula healing rate could be improved by establishing drainage and controlling fistula healing before beginning medical treatment. In the study by Regueiro and colleagues, patients in whom an examination under anesthesia (EUA) with seton placement and abscess drainage was performed were significantly less likely to have a recurrence of their fistula compared with those who never had surgical drainage established (44% vs. 79%). Similarly, in the series reported by Topstad and co-authors, the authors demonstrated a 69% complete fistula healing rate in a small number of patients treated with combination surgical and medical therapy.
Correctly identifying all of the fistulae or abscesses present can be problematic with digital rectal examination or even during EUA in these patients because of the induration and scarring that can be present in association with the perianal disease. In 1 study, the accuracy of digital rectal examination in defining fistula anatomy when done by an experienced surgeon was estimated to be only 62%. Similarly, around 10% of fistulae are misclassified by EUA alone, resulting in the need for repeat surgical intervention in those patients in whom the fistula was incorrectly assessed.
Last, although the fistulae usually stop draining within 6 to 12 weeks of initiating anti–TNF-α therapy, inflammatory activity persists for weeks to months within the middle portion of the fistula tract. Relying on physical examination alone to determine fistula activity inaccurately assesses the patient’s progress, resulting in premature removal of setons or inappropriate changes in medical therapy. Studies using both ultrasonography and magnetic resonance imaging (MRI) have demonstrated persistent fistula activity in nearly all patients even after the third dose in the infliximab induction sequence.
Studies have demonstrated that missing occult tracts can result in recurrent fistulae, abscesses, and/or convert a simple fistula into a complex fistulizing process. If the fistulizing process becomes complex, the chance for healing is greatly reduced. To prevent development of a complex fistula and increase the chance of closure, it is important to optimize the tools available to assess the perianal pathology. Therefore, treatment should begin with correctly assessing disease activity and perianal anatomy as well as establishing drainage and providing control of the fistula, even before starting medical treatment. Ideally this can be accomplished through the use of imaging and by working closely with surgical colleagues ( Fig. 1 ) . The purpose of this article is to review the different imaging modalities available for the assessment of perianal CD as well as the various operative techniques and treatment that can be utilized in these patients. In the future, larger (and thus more adequately powered), multicenter trials are needed to help guide the optimal management strategies of these complex patients.
Imaging Modalities for Initial Evaluation
Fistulography
Fistulography involves inserting a small catheter into the external fistula opening and injecting a small amount of radiopaque contrast material into the tract and imaging the fistula using fluoroscopy. There are several major limitations to this modality. First and most important, fistulograms are not able to directly visualize the sphincter complex; thus, one has to infer the fistula’s anatomy in relation to the pelvic musculature. In addition, fistula tracts can be missed owing to the inability of contrast to fill extensions from the primary track, either from inadequate filling or from debris within the track. Fistulography can also be very uncomfortable for the patient and can be difficult technically for the radiologist to even access the tracts at times.
Studies assessing the use of fistulography have found its accuracy to range from 16% to 50%. In 1 representative study, the results from fistulography were compared with operative findings in 25 patients. Accuracy was 16%, with a false-positive rate of 12%. This study also showed the false-positive results resulted in unnecessary complications in some patients.
Because of these factors, fistulography is not widely used in assessment of CD perianal fistulae, with the exception of times when a connection to the bladder or vagina is suspected. In these clinical scenarios, extravasation of contrast into the other organs may be more easily demonstrated using fistulography.
Axial Computed Tomography
Computed tomography (CT) has been widely used for the evaluation of perianal fistulae. The main factor limiting CT’s accuracy for perianal fistulae is its poor spatial resolution in the pelvis, although it is better than that of fistulography. In addition, because the tissue characteristics are very similar, it can be difficult to differentiate between a fistula tract and inflammation using CT imaging. In a small, prospective study of 25 patients, CT was compared with rectal endosonography using a 5-MHz radial probe. The gold standard in this study was either operative findings or clinical course. They found endosonography to be more accurate than CT (82% vs. 24%). CT is primarily used for the assessment of abscesses, fluid collections, or other pathology higher in the pelvis or abdomen than for assessment of the perianal tracts themselves.
Endoscopic Ultrasonography
Rectal endoscopic ultrasonography (EUS) involves inserting either a rigid or flexible radial probe into the distal rectum and anal canal while the patient lies in the left lateral position. Using this modality, fistulae appear as hypoechoic round or oval structures, but can be internally hyperechoic if there is air or gas in the fistula ( Fig. 2 ). An abscess usually appears as an anechoic or hypoechoic mass in the perianal tissues. Some clinicians also inject hydrogen peroxide into the cutaneous fistula opening to enhance the identification of the fistula tract on EUS. The hydrogen peroxide creates bubbles that appear hyperechoic and thus make the fistula easier to identify. In Vanderbilt University Medical Center’s experience (D.S.’s personal experience) with imaging of CD perianal fistulae (usually with a 7.5-MHz radial scanning ultrasound probe), EUS can usually demonstrate air within a fistula without the addition of hydrogen peroxide. If air is not seen initially, the identification of the tract can be facilitated by applying gentle pressure to visualize the air bubbles moving in the tract, thus making hydrogen peroxide instillation rarely necessary.
The accuracy of ultrasonography in the evaluation of perianal disease has been demonstrated in several studies. Most of these studies have utilized a blind, rigid, transrectal probe. There have been 3 prospective, blinded studies using a flexible echoendoscope in the evaluation of perianal disease. One of these studies compared EUS to CT in 25 patients with suspected perianal CD. EUS was conducted using a 5-MHz radial scanning scope. Results were compared with findings at surgery and/or clinical course. EUS was found to be more accurate than CT in the evaluation of perianal fistulae (82% vs. 24%).
The other 2 studies compared EUS to MRI in a group of patients with perianal CD. In a pilot study by Orsoni and associates, rectal EUS, pelvic MRI, and EUA were compared in 22 patients with CD perianal fistulae. Rectal EUS was found to be the most sensitive modality for imaging CD perianal fistulae. The agreement for fistulae with rectal EUS and pelvic MRI when compared with the surgical findings was 82% and 50%, respectively. Rectal EUS in this study was performed with only a 7-MHz linear scanning probe. A similar study by Schwartz and colleagues (see below) comparing EUS to MRI found both to be equally accurate in the assessment of CD perianal fistulae (91% vs. 87%).
Transperineal Ultrasonography
Transperineal ultrasonography is performed by placing the ultrasound transducer directly on the perineum with the probe directly outside of the anus. To examine the fistula and gain a detailed image, the transducer can be placed above any external fistula openings and moved following any fistula up to their internal origin. Several small studies have shown this imaging may be comparable to pelvic MRI and rectal EUS, and that it can provide an accurate assessment of the perianal anatomy with a sensitivity of more than 85%. This may be particularly useful in patients with significant anal stenosis preventing passage of an EUS probe or endoanal coil for MRI.
MRI
MRI and EUS have become the imaging modalities of choice for the assessment of CD perianal fistulae. Most centers now use a dedicated external pelvic phased-array coil to achieve the best images in these patients. Most studies in the literature have utilized a 1.0- to 1.5-Tesla coil with accuracies for assessing the pretreatment fistula anatomy of around 80% to 90%.
Several studies have compared the accuracy of MRI with EUS in the initial evaluation of CD perianal fistulae. There has been a wide range of differences in outcomes, largely secondary to variations in the equipment utilized in these studies, patient selection, and operator experience. In a representative study of 34 patients with CD perianal fistulae undergoing EUS, MRI and EUA for initial fistula evaluation, the accuracy for EUS, MRI, and EUA was 91%, 87%, and 91%, respectively. In addition, when any 2 modalities were combined, accuracy increased to 100%. Therefore, both imaging modalities (MRI or EUS) have been shown to be accurate in the initial assessment of patients with CD perianal fistulae. The choice of which modality to use largely depends on local institutional expertise.
The use of MRI has been shown to impact patient care related to perianal fistulae. In a representative, prospective study of 56 patients with perianal fistulae (CD and non-CD), the results of the preoperative MRI was withheld from the surgeon initially and then provided once the procedure was finished, allowing the surgeon to revise their initial operation with the new information. The preoperative MRI provided additional information that led to changes in the surgical plan in 21% of the patients. However, if a patient had CD-related fistulae, then the benefit of MRI increased to 40%. Similarly, in study of 71 patients with recurrent perianal fistulae, utilizing the MRI findings to guide surgical treatment reduced the risk of further fistula recurrence by 75% ( P = .008). In cases where the surgical findings disagreed with the results of MRI, the rate of fistula recurrence was 52%, and the MRI predicted the site of recurrence in all of the cases.
MRI or EUS to Guide Therapy and Monitor Response to Therapy
Some imaging modalities available for initial assessment of perianal fistulae in CD can also be used to assess the longitudinal response to anti–TNF-α agents or other therapeutic interventions. Several studies utilizing EUS or MRI have shown persistent fistula inflammatory activity even after the fistula stops draining or is no longer symptomatic. This inflammation on these imaging modalities suggests that making treatment decisions based on physical examination or evidence of fistula drainage alone may lead to a higher risk of recurrence or other bad outcomes.
Several investigators have looked at using either EUS or MRI to monitor fistula healing and/or guide treatment. In 1 retrospective study of 21 patients with CD perianal fistulae, the patients were treated according to a treatment protocol of serial EUS examinations. Surgical and medical therapy was tailored to the results of the EUS findings with seton placement and incision and drainage procedures performed when appropriate. Follow-up EUS examinations guided when to remove setons and/or when to stop infliximab or antibiotics. The median follow-up was 68 weeks (range, 35–101). No abscesses developed in any patient. Eighteen of 21 patients (86%) had complete drainage cessation initially, and 16 of 21 (76%) had long-term cessation of drainage. Eleven (52%) had no persistent fistula activity on EUS, and 7 maintained fistula closure after stopping infliximab and antibiotics; the other 4 continued infliximab for their mucosal disease. This study showed that EUS-guided surgical and medical treatment with the use of infliximab had high short- and long-term fistula response rates.
A small, randomized, prospective study of the benefit of EUS monitoring of fistula healing showed a benefit to the use of EUS guidance in these patients as well. One of 5 (20%) in the control group and 4 of 5 (80%) in the EUS group had complete cessation of drainage. In this small study, EUS guidance for combination medical and surgical therapy in perianal CD seemed to improve outcomes.
Similarly, MRI monitoring of fistula healing has also been shown to be beneficial. In a prospective study of 34 patients with CD perianal fistulae, MRI was done at 6, 12, and 18 months to monitor fistula healing while the patients were on therapy. Those with persistent fistula activity on MRI had their medical therapy increased (e.g., increasing adalimumab to 40 mg weekly). This converted all of the nonresponders to responders by week 52. Therefore, longitudinal surveillance of perianal fistulae with either MRI or EUS seems to improve outcomes for patients with perianal CD by impacting their treatment regimen(s).
MRI or EUS to Guide Therapy and Monitor Response to Therapy
Some imaging modalities available for initial assessment of perianal fistulae in CD can also be used to assess the longitudinal response to anti–TNF-α agents or other therapeutic interventions. Several studies utilizing EUS or MRI have shown persistent fistula inflammatory activity even after the fistula stops draining or is no longer symptomatic. This inflammation on these imaging modalities suggests that making treatment decisions based on physical examination or evidence of fistula drainage alone may lead to a higher risk of recurrence or other bad outcomes.
Several investigators have looked at using either EUS or MRI to monitor fistula healing and/or guide treatment. In 1 retrospective study of 21 patients with CD perianal fistulae, the patients were treated according to a treatment protocol of serial EUS examinations. Surgical and medical therapy was tailored to the results of the EUS findings with seton placement and incision and drainage procedures performed when appropriate. Follow-up EUS examinations guided when to remove setons and/or when to stop infliximab or antibiotics. The median follow-up was 68 weeks (range, 35–101). No abscesses developed in any patient. Eighteen of 21 patients (86%) had complete drainage cessation initially, and 16 of 21 (76%) had long-term cessation of drainage. Eleven (52%) had no persistent fistula activity on EUS, and 7 maintained fistula closure after stopping infliximab and antibiotics; the other 4 continued infliximab for their mucosal disease. This study showed that EUS-guided surgical and medical treatment with the use of infliximab had high short- and long-term fistula response rates.
A small, randomized, prospective study of the benefit of EUS monitoring of fistula healing showed a benefit to the use of EUS guidance in these patients as well. One of 5 (20%) in the control group and 4 of 5 (80%) in the EUS group had complete cessation of drainage. In this small study, EUS guidance for combination medical and surgical therapy in perianal CD seemed to improve outcomes.
Similarly, MRI monitoring of fistula healing has also been shown to be beneficial. In a prospective study of 34 patients with CD perianal fistulae, MRI was done at 6, 12, and 18 months to monitor fistula healing while the patients were on therapy. Those with persistent fistula activity on MRI had their medical therapy increased (e.g., increasing adalimumab to 40 mg weekly). This converted all of the nonresponders to responders by week 52. Therefore, longitudinal surveillance of perianal fistulae with either MRI or EUS seems to improve outcomes for patients with perianal CD by impacting their treatment regimen(s).
Related posts:
The Role of Capsule Endoscopy in Evaluating Inflammatory Bowel Disease
Health Maintenance in the Inflammatory Bowel Disease Patient
Disability in Inflammatory Bowel Disease
Chromoendoscopy in Inflammatory Bowel Disease
Patient-Specific Approach to Combination Versus Monotherapy with the Use of Antitumor Necrosis Factor α Agents for Inflammatory Bowel Disease
Imaging for Luminal Disease and Complications: CT Enterography, MR Enterography, Small-Bowel Follow-Through, and Ultrasound
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
