| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Invited Commentary |
1 Department of Radiology, Medical College of Wisconsin and Froedtert Memorial Lutheran Hospital, Milwaukee, Wisconsin
Anyone who has performed sinography on a patient with a perianal fistula will be happy to read the preceding article by Morris and colleagues . Although imaging the perianal fistula as a road map for surgery is not a common task in most radiologic practices, this problem can be very frustrating to the radiologist, surgeon, and patient.
A perianal fistula can occur in patients with Crohn disease, tuberculosis, or human immunodeficiency viral infection or in any other immunocompromised patient. However, it usually occurs in an otherwise healthy patient, typically a middle-aged man. Most authors believe that perianal fistulas develop in this latter group of patients as the result of anal gland obstruction, secondary abscess formation, and subsequent external decompression through one of several fairly predictable routes. The internal origin of the fistula usually begins from the middle of the anal canal, at the dentate line. The fistulous track usually courses between the internal and external sphincters (intersphincteric) or crosses the sphincter complex (trans-sphincteric) to exit the skin around the anus. A fistula with a suprasphincteric route ascends the intersphincteric space before it exits laterally and relatively high through the striated muscle near the transition of the levator ani into the external sphincter. A supralevator fistula usually arises from a focus of inflammation originating in the pelvis above the levator ani muscle; its track crosses the ischiorectal fossa to exit in the perianal area.
The radiologist is involved only rarely in the evaluation of perianal fistulas. In the relatively early stages of this inflammatory process, the localized perianal abscess can be successfully drained without imaging guidance. A perianal fistula is the chronic phase of a perirectal abscess. If a surgeon can comfortably define the fistulous track from the external to the internal openings with a probe, the fistula can be cured by "unroofing" it internally. Thus, a fistulotomy cures this malady with only minimal sphincteric damage, which preserves continence.
However, a certain number of perianal fistulas become complicated and require a road map before the proper surgical approach can be devised. The routes of such fistulas can vary greatly, and they may even follow a radial course before exiting the skin. When an imaging evaluation of perianal fistulas is required, the goals are twofold: (a) to define the presence and course of any secondary tracks or any abscess formation, which could potentially cause a relapse if only simple fistulotomy were performed, and (b) to gauge the extent of sphincteric involvement by the fistula to best plan surgery for preservation of sphincteric function.
Imaging techniques in the past have included fistulography, computed tomography (CT), endoscopic ultrasonography (US), and, more recently, MR imaging. Direct injection of contrast material to obtain a sinogram has many drawbacks. Working in the perianal area is difficult for both the patient and the radiologist. First, the radiologist must find and then cannulate the cutaneous exit site. Providing an adequate seal at this opening to obtain the necessary filling pressure can be vexing. Even with a proper injection of contrast material, not all the track may be displayed if a tight strictured segment or track detritus obstructs flow. The entire fistula may be delineated, but it can be very difficult to correlate the track route with the local anatomic musculature and spaces necessary for preoperative planning.
CT can clearly depict a local perianal abscess. With thin-section helical CT, coronal images can be reconstructed to orient this anatomy to better advantage. But the critical definition of the fistulous track and its surrounding anatomy will still not be adequately displayed.
Endorectal US has been used to image perianal fistulas, and, particularly when a 10-MHz transducer is used, US can demonstrate the sphincteric anatomy and adjacent soft tissue. However, US does not provide an adequate deep and global display of all adjacent pelvic and perineal spaces that may be involved. In addition, from the surgeon's perspective, the modality does not provide a suitable reproduction of the anatomy that can be viewed in preparation for surgery and its accuracy is too dependent on operator technique.
As illustrated by Morris et al, MR imaging has become the method of choice for imaging perianal fistulas. Its orthogonal display of the perineum and lower pelvis, along with its superior contrast resolution, allows faithful reproduction of the anatomy and pathologic tracks. At present, MR imaging technology offers three options for imaging this area. MR imaging performed with an endorectal coil provides excellent delineation of the sphincteric components. It can also depict the internal origin of the fistulous track, which some authors believe is vital for the surgical road map (1). This mode of MR imaging suffers from the same drawbacks as endorectal US; that is, lack of adequate depth of perirectal anatomic coverage and the pain or possible harm inflicted on the patient by placing an almost 2-cm-diameter rectal probe through inflamed tissue.
The article by Morris et al supports the use of a body coil for imaging perianal fistulas. Although with this technique the internal origin of the fistula must be inferred and the normal intersphincteric space cannot be defined, body coil MR imaging does allow visualization of abnormalities related to the fistula. The protocol is also very efficient and noninvasive.
In addition, one should not overlook the authors' mention of local coil application to improve anatomic display. Use of a dedicated pelvic or torso coil or even the Helmholtz coil configuration should provide greater clarity without risk to the patient. Use of these coils to obtain a 20-cm field of view with a 256 x 256 or 256 x 192 matrix for both T1-and T2-weighted images will provide much improved spatial resolution of this area. Such improvement should enable radiologists to appreciate more subtleties or, at the very least, have more confidence in their interpretation of the images.
At present, numerous MR imaging sequences are available. Use of T1-weighted pulse sequences combined with T2-weighted sequences in the axial and coronal planes is widely accepted. A T1-weighted image without fat saturation will depict the dark fistulous track within the bright fat. A fat-saturated T2-weighted image will demonstrate the increased signal intensity tracks as well as other inflammatory foci. When needed, a series of T1-weighted images obtained with fat saturation before and after an injection of gadolinium will optimally depict enhancing, inflamed tracks or cause them to have a ringed appearance to help differentiate an abscess from a phlegmon. The ability to use fat saturation after gadolinium administration is probably of more value than a dynamic gradient-echo sequence if, as in the authors' case, a fat-saturated gradient-echo sequence is not available.
So be sure to clip out the article by Morris et al to share with your surgical colleagues when the next request surfaces for imaging of a perianal fistula. It would also be wise to keep the article as a reference when you interpret the MR imaging examination.
References
Department of Clinical Radiology, St James's University Hospital, Leeds, England
We appreciate the comments of Professor Taylor. We agree with his view that the great majority of patients with perianal sepsis will be adequately treated by incision and drainage, will not develop a fistula, and will therefore not require any form of imaging. In hospitals specializing in the treatment of perianal fistulas, however, MR imaging has assumed an important role in surgical planning. However, there are several unresolved issues. Professor Taylor highlights the many different techniques that have been used for imaging perianal fistulas. All too often in medicine, the existence of many solutions or explanations for a problem indicates that none is satisfactory. In the case of MR imaging of perianal fistulas, most of the techniques described work well. Our view is that the key to success is a clear understanding of the surgical anatomy of perianal fistulas and treatment options developed through close collaboration with surgical colleagues. Quite simply, if we can recognize that a fistula transgresses the full thickness of the sphincter complex, we can warn surgical colleagues that the condition (or its treatment) threatens sphincteric function.
In our view, the biggest unanswered questions relate not to optimization of the imaging technique but to its expanding clinical utility. We are no longer asked to simply provide surgical road maps. We are referred increasing numbers of cases to exclude disease in the perianal region and to characterize undiagnosed perianal and perineal masses. MR imaging of the perianal region has undergone a fairly rigorous proving phase and is now maturing and diffusing into clinical practice. As radiologists, we must be prepared for these new challenges.
Related Article
RadioGraphics 2000 20: 623-635.
This article has been cited by other articles:
|
|
 |
|
  P. J. Pickhardt, S. Bhalla, and D. M. Balfe Acquired Gastrointestinal Fistulas: Classification, Etiologies, and Imaging Evaluation Radiology, July 1, 2002; 224(1): 9 - 23. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOGRAPHICS | RADIOLOGY | RSNA JOURNALS ONLINE |
