The role of imaging tests in the evaluation of anal abscesses and fistulas

Authors:
David A Schwartz, MD
Maurits J Wiersema, MD
Section Editor:
J Thomas Lamont, MD
Deputy Editor:
Shilpa Grover, MD, MPH, AGAF

Contributor Disclosures

All topics are updated as new evidence becomes available and our peer review
process is complete.
Literature review current through: Jan 2018. | This topic last updated: Sep 14, 2017.

INTRODUCTION — Perianal fistulas and abscesses are among the most serious
manifestations of Crohn disease and non-Crohn related anorectal disease (picture 1
and picture 2). Complications can lead to difficulties with recurrent or non-healing
fistulas or abscesses. In addition, these patients are at risk of incontinence as a result
of the destructive nature of the fistulizing process and/or inadvertent damage to the
anal sphincters during surgical exploration.

The lifetime risk for developing a fistula in patients with Crohn disease is 20 to 40
percent [1-4]. The frequency of perianal fistulas/abscesses in patients without Crohn
disease has not been well established, but in a telephone survey of 102 randomly
selected individuals, 20 percent of the individuals contacted had perianal symptoms
(hemorrhoids, fistulas, etc) [5]. Despite the significant prevalence of perianal disease,
the evaluation of this problem was, in the past, largely limited to digital rectal
examination.

The inability of the clinician to directly visualize the fistula or abscess makes it
difficult to assess the lesions. The clinician must essentially discern the perianal
anatomy by touch. This task is made even more problematic by the induration and
inflammation that is usually present in these patients. Even surgical evaluation is only
35 to 85 percent accurate when compared with the results of other diagnostic tests and
clinical evaluation [6-9].

The importance of accurately characterizing the perianal process prior to embarking

on therapy cannot be overemphasized. The risk of incomplete healing, a recurrent
fistula, or even inadvertent sphincter injury is increased if fistula anatomy is
incorrectly delineated or an occult abscess missed. An imaging modality should
ideally provide a virtual road map that the clinician can use to plan therapy. This is
especially true with fistulas that involve a significant portion of the anal sphincter
complex. Such patients are at the greatest risk of developing incontinence from the
destructive fistulizing process or from overly aggressive surgical treatment.

Thus, patients with simple fistulas that only involve a small portion or none of the
external anal sphincter generally do well with either medical or surgical treatment.
Imaging of the fistula is helpful in determining the type of fistula to guide treatment
but is not always needed. By contrast, for patients with a complex fistula (ie, one that
involves a significant portion of the sphincter complex), preoperative imaging is
mandatory.

Several imaging modalities are available to evaluate perianal fistulas and abscesses.
These include fistulography, computed tomography, magnetic resonance imaging, and
ultrasonography (both transrectal and endoscopic). The efficacy of each modality
(with emphasis on endoscopic ultrasonography) will be reviewed here.

FISTULOGRAPHY — Fistulography has traditionally been the primary method of

evaluating perianal fistulas. It involves insertion of a small caliber catheter into the
external opening of a fistula and injection of radiographic contrast material directly
into the fistula track. Radiographs from different angles are then obtained.

Fistulography has several drawbacks. The crucial determination of the fistula's course
in relation to the sphincter complex must be inferred because the musculature of the
anorectum cannot be visualized. In addition, instillation of the contrast material can be
painful and can lead to the theoretical dissemination of septic fistulous contents.

Furthermore, the accuracy of fistulography has been questioned. In one retrospective

study of 25 patients, the findings of fistulography were compared with the operative
findings [10]. Fistulograms were correct in only four patients (16 percent). In
addition, false-positive results that could have led to unnecessary complications
occurred in three patients (12 percent).

As a result of these limitations, fistulography should generally be reserved for select

patients in whom there is concern about a fistulous connection between the rectum
and an adjacent organ such as the bladder.

COMPUTED TOMOGRAPHY — Computed tomography (CT) permits structures

outside of the bowel lumen to be visualized, an advantage compared with
fistulography. It is valuable for evaluating suspected perianal abscesses and
inflammation [11-15]. In one study, for example, CT (with intravenous and, when
possible, rectal contrast) was useful in differentiating perirectal abscesses from severe
perirectal cellulitis and correctly identified 13 surgical proven abscesses in 10 patients
[16].

The utility of CT for perianal fistulas is less clear. Fistulas are identified on CT when
either a linear track containing air or contrast material is demonstrated extending from
the bowel. The limited resolution of CT makes it difficult to differentiate between
inflammatory soft tissue streaking and a fistula tract [15]. One study of 25 patients
with suspected perianal Crohn disease compared the efficacy of endoscopic
ultrasound (EUS) and CT [17]. EUS was conducted using a 5 MHz radial scanning
scope. CT was performed using both intravenous and rectal contrast. 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 fistulas (82 versus 24 percent).

Another limitation of CT is that the classification of fistulas can be difficult because

CT scans in the transverse plane make the identification of the levator ani unreliable
[16]. Because of these reasons, and because CT exposes the patient to ionizing
radiation, the use of CT in these patients is limited.
MAGNETIC RESONANCE IMAGING — Because of the limitations of
fistulography and computed tomography (CT), attention has turned to the newer
technologies, such as magnetic resonance imaging (MRI) (image 1 and image 2) and
ultrasound, as a better means of imaging perianal fistulas. Several prospective studies
have looked at MRI in the evaluation of fistula-in-ano in patients without Crohn
disease. Most of these studies used examination under anesthesia as the gold standard
[8,18,19]. Two of the largest studies had widely differing results. In one, the results of
MRI were compared with surgical evaluation in 35 patients [20]. The overall
concordance between MRI and surgery was 85 percent. In contrast, using similar
methodology, the group from University Hospital Nottingham reported accuracy of
MRI in 33 patients to be only 42 percent [18].

However, at least three studies have questioned the use of surgical evaluation as a
gold standard [7,9,21]. Patients were followed postoperatively to allow time for a
missed fistula or abscess to declare itself, thereby permitting determination of the
predictive value of MRI and examination under anesthesia. In one study of 42 patients
MRI was found to be more sensitive than examination under anesthesia (95 versus 76
percent) [21]. Another series of 40 patients who were followed for 14 months
postoperatively found a sensitivity and specificity of 89 and 69 percent, respectively,
for MRI compared with 73 and 47 percent, respectively, for surgical exploration [7].
An updated report of 52 patients from the same group found MRI to be slightly more
sensitive than surgical assessment in determining disease severity (81 versus 77
percent), although these results were not statistically significant [9].

Only a few studies have looked at MRI utility specifically in patients with Crohn
disease with perianal fistulas:

●In a pilot study, MRI was able to identify eight out of nine (89 percent) perianal
fistulas [22]. However, this study used unreliable imaging modalities such as CT and
fistulography as the gold standard. Five years later, the same group found MRI to be
86 percent sensitive in delineating perianal fistula anatomy in 34 Crohn patients using
surgical findings as the gold standard [23]. MRI seemed to have more difficulty
demonstrating the shorter, more superficial tracks.

●One of the largest studies included 54 patients with suspected perianal Crohn disease
[24]. The authors reviewed the proctological, MRI, and intraoperative findings to
determine a consensus gold standard that they used as their benchmark. A total of 90
fistulas and 83 abscesses were found in these patients. MRI was 82 percent accurate
for determining fistula anatomy, and, as in the previous report [23], tended to miss the
short or superficial fistula tracks.

●Another report focused on 18 patients who were studied before and after treatment
with infliximab [25]. A fistula track with signs of active inflammation was visualized
with MRI in all patients prior to therapy. The fistula track remained visible in 8 of 11
patients who responded clinically to infliximab. After long-term treatment (46 weeks),
MRI signs of active track inflammation had resolved in three of six patients. These
findings suggest that despite closure of draining external orifices following infliximab
therapy, fistula tracks can persist with varying degrees of inflammation. Similar
findings have been described by others using both MRI and endosonography [26-29].
ENDOSONOGRAPHY — Endosonography, both blind transrectal (TRUS) and
endoscopic ultrasound (EUS), has also been used to evaluate perianal fistulas. Similar
to the magnetic resonance imaging (MRI) literature, the vast majority of the
ultrasound studies have focused on patients without Crohn disease.

Technical aspects — Two discrete rings of tissue can be seen when using a radial
scanning echoendoscope to examine the anorectum (image 3). The inner hypoechoic
ring of tissue represents the internal anal sphincter, which is formed by the thickened
continuation of the circular smooth muscle of the rectum. It is usually approximately
3 cm in length. The outer hyperechoic ring of tissue represents the external anal
sphincter, which is formed by the downward extension of the skeletal muscle of the
puborectalis. It is generally 4 cm in length.

The initial endosonographic studies used 7 MHz radial probes that were placed
blindly into the rectum. These early pilot studies yielded promising results with the
sensitivity for visualizing anal fistulas greater than 90 percent [30,31]. Several centers
have tried to increase the sensitivity of ultrasound by instilling hydrogen peroxide into
the fistula tracks [32,33]. Hydrogen peroxide acts as a contrast medium for
ultrasound, creating echo-rich bubbles within the fistula track. This method is limited
to fistulas with cutaneous openings. A limitation of this approach is that hydrogen
peroxide can cause acoustic shadowing that may lead to misinterpretation of the
fistula track. In our experience, a 7 MHz linear scanning ultrasound probe is able to
clearly demonstrate the air within a fistula tract, thus making instillation of hydrogen
peroxide unnecessary (image 4). Frequently, by applying gentle pressure to the fistula
tract with the linear probe we can clearly visualize the air bubble moving within the
tract itself (movie 1).

Accuracy — In one of the more commonly quoted TRUS studies in the literature, the
group from St. Mark's reported disappointing results using TRUS to evaluate fistula-
in-ano [6]. In this prospective study of 38 patients with suspected fistula-in-ano,
digital rectal examination by an experienced consultant was compared with TRUS.
Surgical findings were considered the gold standard. Digital rectal exam (DRE) was
found to be more accurate than ultrasound in determining the course of the primary
fistula track (85 percent versus 72 percent).

However, these results must be interpreted with caution for several reasons. The
difference between DRE and TRUS did not meet statistical significance. In addition,
as the authors readily admit, the rigid nature of the probe prevented good acoustic
coupling higher in the rectum, thus preventing the interpretation of higher fistula
tracks. Furthermore, the focal length of the probe utilized for this study was only 3
cm, which limited scanning to no deeper than the external anal sphincter (EAS). To
prevent these problems, we use an inflatable balloon probe on a flexible
echoendoscope to permit better acoustic coupling throughout the rectum. In addition,
we evaluate the fistulas with both radial and linear scanning instruments in order to
achieve a greater depth of imaging and to more thoroughly characterize the fistulas.

The destructive and recurrent nature of perianal Crohn disease makes accurate
imaging more difficult than with simple fistula-in-ano. However, several non-blinded
studies have shown ultrasound to be a viable modality for examining the perianal
manifestations of Crohn disease [34-39]. A prospective blinded study compared EUS
with computed tomography (CT) in 25 Crohn patients with suspected perianal
involvement [17]. A 5 MHz radial scanning probe was used to conduct the ultrasound
examinations. Surgery or fistulography was used as the gold standard. EUS was found
to be superior to CT with a sensitivity of 82 versus 24 percent, respectively.

A later, randomized prospective study randomly assigned 10 patients with EUS with
examination under anesthesia, or examination under anesthesia alone [40]. Patients
were managed medically or surgically based upon the finding. At the end of one year,
patients randomized to the EUS arm were more likely to have complete cessation of
drainage, suggesting that the EUS may have improved care.

At least four prospective studies have compared MRI with endosonography in the
evaluation of perianal fistulas [20,41-43]. Two of these reports concentrated
exclusively on patients with Crohn disease [42,43]. The two studies focusing
primarily on non-Crohn patients using surgery as the gold standard found MRI to be
superior to TRUS in imaging fistulas [20,41]. Both of these studies used 7 MHz rigid
radial scanning ultrasound probes. The limitation of this equipment (see 'Technical
aspects' above) most likely contributed to the lower ultrasound sensitivity seen in
these studies.

In contrast, one of the prospective studies comparing MRI and EUS for Crohn
perianal fistulas found endosonography to be the most sensitive modality for imaging
fistulas [42]. In this pilot study of 22 patients, surgical evaluation was used as the gold
standard. The agreement for fistulas with the surgical findings for endosonography
and MRI was 82 percent and 50 percent, respectively. Ultrasound was performed with
a 7 MHz linear scanning probe. Although not used in this study, we also use a radial
scanning probe to provide complimentary information. This can reveal fistulas not
apparent with the linear probe (image 5).

The poor sensitivity of MRI in this study (50 percent) may be secondary to the use of
a body coil for imaging instead of a pelvic phased array coil. The pelvic phased array
coil is a receive-only coil and provides better spatial resolution than is available with
the body coil. A study using the phased array suggested that it was highly accurate in
detecting primary tracks and secondary extensions and provided important additional
information in 12 out of 56 patients (21 percent) enrolled in the study [44]. The
benefit of MRI was most obvious in patients with fistulas related to Crohn disease and
in patients with complex fistulas associated with a recurrence.

Similar conclusions were reached in another study in 34 patients with Crohn disease
who were suspected of having perianal fistulas [43]. Patients underwent EUS and
MRI within the same week followed by surgical examination under anesthesia (EUA).
The gold standard anatomy was defined after reviewing data from all three modalities.
All three methods demonstrated good agreement with the gold standard (EUS 91
percent, MRI 87 percent, and EUA 91 percent). The accuracy increased to 100
percent when EUA was combined with either EUS or MRI.

Several studies have evaluated EUS and MRI in monitoring the course of fistula
healing in patients with perianal Crohn disease [25,45-48]. Two reports also
suggested that treatment based upon EUS findings was helpful in determining the
optimal time to remove a seton [40,47]. In addition, one prospective study utilizing
MRI to monitor fistula healing in patients with Crohn perianal fistulas on anti-TNF
treatment suggested that imaging may be useful in identifying patients who need
continued medical therapy [49].

INFORMATION FOR PATIENTS — UpToDate offers two types of patient

education materials, "The Basics" and "Beyond the Basics." The Basics patient
education pieces are written in plain language, at the 5th to 6th grade reading level, and
they answer the four or five key questions a patient might have about a given
condition. These articles are best for patients who want a general overview and who
prefer short, easy-to-read materials. Beyond the Basics patient education pieces are
longer, more sophisticated, and more detailed. These articles are written at the 10 th to
12th grade reading level and are best for patients who want in-depth information and
are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage
you to print or e-mail these topics to your patients. (You can also locate patient
education articles on a variety of subjects by searching on "patient info" and the
keyword(s) of interest.)

●Basics topics (see "Patient education: Anal abscess and fistula (The Basics)")

SUMMARY AND RECOMMENDATIONS

●Perianal fistulas and abscesses are a common problem; proper therapy depends upon
the accurate assessment of the perianal anatomy. (See 'Introduction' above.)

●Fistulography is not an accurate means of evaluating perianal disease. (See

'Fistulography' above.)

●Computed tomography (CT) may be helpful when looking for large intrapelvic
abscesses. CT is inaccurate for the detection and classification of perianal fistula
tracks and small perianal abscess collections. (See 'Computed tomography' above.)

●Magnetic resonance imaging (MRI) and endosonography are an accurate means of

evaluating perianal disease, with an accuracy of approximately 80 to 90 percent
compared with surgical exam under anesthesia. Accuracy of MRI depends upon
availability of MRI imaging with dedicated pelvic coils. The dedicated pelvic coil
provides better spatial resolution in the pelvis. Similarly, the accuracy of ultrasound
depends upon achievement of good acoustic coupling by using water-filled inflatable
balloons. (See 'Magnetic resonance imaging' above and 'Endosonography' above.)

Use of UpToDate is subject to the Subscription and License Agreement.

REFERENCES
1. Hellers G, Bergstrand O, Ewerth S, Holmström B. Occurrence and outcome
after primary treatment of anal fistulae in Crohn's disease. Gut 1980; 21:525.
2. Farmer RG, Hawk WA, Turnbull RB Jr. Clinical patterns in Crohn's disease: a
statistical study of 615 cases. Gastroenterology 1975; 68:627.
3. Rankin GB, Watts HD, Melnyk CS, Kelley ML Jr. National Cooperative
Crohn's Disease Study: extraintestinal manifestations and perianal
complications. Gastroenterology 1979; 77:914.
4. Schwartz D, Loftus E, Tremaine W, et al. The natural history of fistulizing
Crohn's disease: a population based study. Gastroenterology 2000; 118:A337.
5. Nelson RL, Abcarian H, Davis FG, Persky V. Prevalence of benign anorectal
disease in a randomly selected population. Dis Colon Rectum 1995; 38:341.
6. Choen S, Burnett S, Bartram CI, Nicholls RJ. Comparison between anal
endosonography and digital examination in the evaluation of anal fistulae. Br J
Surg 1991; 78:445.
7. Spencer JA, Chapple K, Wilson D, et al. Outcome after surgery for perianal
fistula: predictive value of MR imaging. AJR Am J Roentgenol 1998;
171:403.
8. Van Beers B, Grandin C, Kartheuser A, et al. MRI of complicated anal
fistulae: comparison with digital examination. J Comput Assist Tomogr 1994;
18:87.
9. Chapple KS, Spencer JA, Windsor AC, et al. Prognostic value of magnetic
resonance imaging in the management of fistula-in-ano. Dis Colon Rectum
2000; 43:511.
10. Kuijpers HC, Schulpen T. Fistulography for fistula-in-ano. Is it useful? Dis
Colon Rectum 1985; 28:103.
11. Fishman EK, Wolf EJ, Jones B, et al. CT evaluation of Crohn's disease: effect
on patient management. AJR Am J Roentgenol 1987; 148:537.
12. Goldberg HI, Gore RM, Margulis AR, et al. Computed tomography in the
evaluation of Crohn disease. AJR Am J Roentgenol 1983; 140:277.
13. Berliner L, Redmond P, Purow E, et al. Computed tomography in Crohn's
disease. Am J Gastroenterol 1982; 77:548.
14. Kerber GW, Greenberg M, Rubin JM. Computed tomography evaluation of
local and extraintestinal complications of Crohn's disease. Gastrointest Radiol
1984; 9:143.
15. Yousem DM, Fishman EK, Jones B. Crohn disease: perirectal and perianal
findings at CT. Radiology 1988; 167:331.
16. Guillaumin E, Jeffrey RB Jr, Shea WJ, et al. Perirectal inflammatory disease:
CT findings. Radiology 1986; 161:153.
17. Schratter-Sehn AU, Lochs H, Vogelsang H, et al. Endoscopic ultrasonography
versus computed tomography in the differential diagnosis of perianorectal
complications in Crohn's disease. Endoscopy 1993; 25:582.
18. Scholefield JH, Berry DP, Armitage NC, Wastie ML. Magnetic resonance
imaging in the management of fistula in ano. Int J Colorectal Dis 1997;
12:276.
19. Buchanan GN, Halligan S, Williams AB, et al. Magnetic resonance imaging
for primary fistula in ano. Br J Surg 2003; 90:877.
20. Lunniss PJ, Barker PG, Sultan AH, et al. Magnetic resonance imaging of
fistula-in-ano. Dis Colon Rectum 1994; 37:708.
21. Beckingham IJ, Spencer JA, Ward J, et al. Prospective evaluation of dynamic
contrast enhanced magnetic resonance imaging in the evaluation of fistula in
ano. Br J Surg 1996; 83:1396.
22. Koelbel G, Schmiedl U, Majer MC, et al. Diagnosis of fistulae and sinus tracts
in patients with Crohn disease: value of MR imaging. AJR Am J Roentgenol
1989; 152:999.
23. Skalej M, Makowiec F, Weinlich M, et al. [Magnetic resonance imaging in
perianal Crohn's disease]. Dtsch Med Wochenschr 1993; 118:1791.
24. Makowiec F, Laniado M, Jehle EC, et al. Magnetic resonance imaging in
perianal Crohn's disease. Inflamm Bowel Dis 1995; 1:256.
25. Van Assche G, Vanbeckevoort D, Bielen D, et al. Magnetic resonance
imaging of the effects of infliximab on perianal fistulizing Crohn's disease.
Am J Gastroenterol 2003; 98:332.
26. Bell SJ, Halligan S, Windsor AC, et al. Response of fistulating Crohn's disease
to infliximab treatment assessed by magnetic resonance imaging. Aliment
Pharmacol Ther 2003; 17:387.
27. Karmiris K, Bielen D, Vanbeckevoort D, et al. Long-term monitoring of
infliximab therapy for perianal fistulizing Crohn's disease by using magnetic
resonance imaging. Clin Gastroenterol Hepatol 2011; 9:130.
28. Rasul I, Wilson SR, MacRae H, et al. Clinical and radiological responses after
infliximab treatment for perianal fistulizing Crohn's disease. Am J
Gastroenterol 2004; 99:82.
29. Ziech ML, Lavini C, Bipat S, et al. Dynamic contrast-enhanced MRI in
determining disease activity in perianal fistulizing Crohn disease: a pilot
study. AJR Am J Roentgenol 2013; 200:W170.
30. Deen KI, Williams JG, Hutchinson R, et al. Fistulas in ano: endoanal
ultrasonographic assessment assists decision making for surgery. Gut 1994;
35:391.
31. Law PJ, Talbot RW, Bartram CI, Northover JM. Anal endosonography in the
evaluation of perianal sepsis and fistula in ano. Br J Surg 1989; 76:752.
32. Cheong DM, Nogueras JJ, Wexner SD, Jagelman DG. Anal endosonography
for recurrent anal fistulas: image enhancement with hydrogen peroxide. Dis
Colon Rectum 1993; 36:1158.
33. Poen AC, Felt-Bersma RJ, Eijsbouts QA, et al. Hydrogen peroxide-enhanced
transanal ultrasound in the assessment of fistula-in-ano. Dis Colon Rectum
1998; 41:1147.
34. el Mouaaouy A, Tolksdorf A, Starlinger M, Becker HD. [Endoscopic
sonography of the anorectum in inflammatory rectal diseases]. Z Gastroenterol
1992; 30:486.
35. Tio TL, Mulder CJ, Wijers OB, et al. Endosonography of peri-anal and peri-
colorectal fistula and/or abscess in Crohn's disease. Gastrointest Endosc 1990;
36:331.
36. Van Outryve MJ, Pelckmans PA, Michielsen PP, Van Maercke YM. Value of
transrectal ultrasonography in Crohn's disease. Gastroenterology 1991;
101:1171.
37. Solomon MJ. Fistulae and abscesses in symptomatic perianal Crohn's disease.
Int J Colorectal Dis 1996; 11:222.
38. Mulder C, Tio T, Tytgat G. Transrectal ultrasonography in the assessment of
perianal fistula and/or abscess in Crohn's disease. Gastroenterology 1988;
94:A313.
39. Schratter-Sehn AU, Lochs H, Handl-Zeller L, et al. Endosonographic features
of the lower pelvic region in Crohn's disease. Am J Gastroenterol 1993;
88:1054.
40. Spradlin NM, Wise PE, Herline AJ, et al. A randomized prospective trial of
endoscopic ultrasound to guide combination medical and surgical treatment
for Crohn's perianal fistulas. Am J Gastroenterol 2008; 103:2527.
41. Hussain SM, Stoker J, Schouten WR, et al. Fistula in ano: endoanal
sonography versus endoanal MR imaging in classification. Radiology 1996;
200:475.
42. Orsoni P, Barthet M, Portier F, et al. Prospective comparison of
endosonography, magnetic resonance imaging and surgical findings in
anorectal fistula and abscess complicating Crohn's disease. Br J Surg 1999;
86:360.
43. Schwartz DA, Wiersema MJ, Dudiak KM, et al. A comparison of endoscopic
ultrasound, magnetic resonance imaging, and exam under anesthesia for
evaluation of Crohn's perianal fistulas. Gastroenterology 2001; 121:1064.
44. Beets-Tan RG, Beets GL, van der Hoop AG, et al. Preoperative MR imaging
of anal fistulas: Does it really help the surgeon? Radiology 2001; 218:75.
45. Ardizzone S, Maconi G, Colombo E, et al. Perianal fistulae following
infliximab treatment: clinical and endosonographic outcome. Inflamm Bowel
Dis 2004; 10:91.
46. van Bodegraven AA, Sloots CE, Felt-Bersma RJ, Meuwissen SG.
Endosonographic evidence of persistence of Crohn's disease-associated
fistulas after infliximab treatment, irrespective of clinical response. Dis Colon
Rectum 2002; 45:39.
47. Schwartz DA, White CM, Wise PE, Herline AJ. Use of endoscopic ultrasound
to guide combination medical and surgical therapy for patients with Crohn's
perianal fistulas. Inflamm Bowel Dis 2005; 11:727.
48. Tozer P, Ng SC, Siddiqui MR, et al. Long-term MRI-guided combined anti-
TNF-α and thiopurine therapy for Crohn's perianal fistulas. Inflamm Bowel
Dis 2012; 18:1825.
49. Ng SC, Plamondon S, Gupta A, et al. Prospective evaluation of anti-tumor
necrosis factor therapy guided by magnetic resonance imaging for Crohn's
perineal fistulas. Am J Gastroenterol 2009; 104:2973.