Keywords
fistula-in-ano - MRI - recurrent - anal sphincter - incontinence
Introduction
Fistula-in-ano is an abnormal tract or cavity between the anal canal and the perianal
skin.[1] The majority of fistula-in-ano is a sequela of poorly managed cryptoglandular infection,
which starts in the intersphincteric space and then spreads.[2] Parks classification was the earliest classification of fistula-in-ano, and was
based on the location of the fistula tract with respect to the anal sphincter complex
as observed during the surgical treatment.[3] With the availability of magnetic resonance imaging (MRI) scan, better evaluation
of fistula was possible. The subsequent imaging classification proposed by Morris
et al was based on the location of the primary track, presence of secondary ramifications,
and associated abscesses.[4] More recently, Standard Practice Task Force (SPTF) classified fistula-in-ano into
simple or complex depending on the risk of fecal incontinence after fistulotomy.[5]
The goals of treatment of fistula-in-ano include resolving acute-on-chronic inflammatory
process, maintaining continence, and preventing future recurrence.[6]
[7] Failure to excise the primary tract with its secondary extensions or incomplete
drainage of septic focus may eventually result in persistence or recurrence of fistula-in-ano.[6] Despite treatment advances, the recurrence rates are astoundingly high ranging between
7 and 50%, and often need multiple surgeries.[8]
[9]
[10] The rates of fecal incontinence after different types of treatments for fistula-in-ano
have been reported to be as high as 64%.[11]
Preoperative MRI is the investigation of choice for fistula-in-ano.[12] MRI is useful for accurately mapping the fistula tract and for identifying features
known to be associated with recurrence. Thus, MRI positively affects the outcome of
fistula treatment. While interpretation of fistula-in-ano in treatment-naïve patients
is straightforward, MRI interpretation in a setting of recurrent fistula-in-ano can
be challenging. This is because of complex nature of the residual/recurrent fistula-in-ano,
disturbed anatomy, and scarring from previous treatment attempts. Though there are
numerous studies that emphasize the value of MRI in fistula-in-ano in general, the
number of studies on recurrent fistula-in-ano and studies assessing anal sphincter
morphology on these patients are far and few. One of the studies on recurrent fistula-in-ano
showed that surgery performed with the guidance of MRI findings reduced the chance
of further recurrence by 75%.[13] Moreover, the association between characteristics of recurrent fistula-in-ano, the
degree of anal sphincter scarring, and fecal incontinence is complex and less well
understood.
The two objectives of this study were to compare the MRI findings of patients in recurrent
and treatment-naïve fistula-in-ano and to correlate the presence and degree of anal
sphincter scarring seen on high resolution MRI with the fistula characteristics of
patients with recurrent fistula-in-ano.
Materials and Methods
Setting and Patients
This is an institution review board (IRB min no. = 13193,22.7.2020) approved retrospective
study conducted by radiology department and a dedicated colorectal surgery department
of a tertiary care teaching hospital. Consecutive adult patients with clinically suspected
fistula-in-ano who underwent MRI for its evaluation between January 2018 and December
2018 were identified from picture archiving and communication system. Patients with
fistula-in-ano on MRI were included, while those with alternative diagnosis on MRI
like pilonidal sinus, fissure-in-ano, and neoplasms of anorectal, presacral, and ischiorectal
fossa regions were excluded ([Fig. 1]). Similarly, poor-quality studies with artefacts and imaging performed elsewhere
were excluded.
Fig. 1 Consort statement of patients included in the study. MRI, magnetic resonance imaging.
MRI Protocol and Image Interpretation
All MRI studies were performed in one of the following two MRI scanners: Magnetom
Avanto fit, 1.5T (Siemens Healthcare Erlangen, Germany) or Intera 22 Achieva 3.0T
(Philips Healthcare, Best, Netherlands). MRI pelvis was performed with patients in
supine position using 16 channel external phased-array body coil. MRI protocol included
T2 fast spin echo and T2 spectral attenuated inversion recovery (TR/TE of 4500-6500/60-70
milliseconds, large field of view axial images of the pelvis, high resolution which
were acquired as small field of view with in plane resolution ≤0.7 mm) T2-weighted
and T2 short tau inversion recovery (STIR) images in sagittal, oblique axial, and
oblique coronal planes obtained perpendicular and parallel to the anal canal. We did
not routinely perform gadolinium-enhanced MRI or diffusion-weighted imaging for patients
with fistula-in-ano. Two radiologists blinded to clinical and laboratory findings
reread the MRI scans in consensus.
The type of fistula according to Parks classification and St James classification,
the site and the side of fistula-in-ano, the number of internal and external openings,
the distance between the anal verge and the internal and the external openings, the
length of the primary tract, secondary tracts, supralevator extension, presence and
the location of abscess or collections, activity of the tract, and the morphology
of the anal sphincter complex were documented for every patient.
For those with multiple internal or external openings, the longest tract was assumed
to be the primary tract and the other communicating tracts as secondary tracts. Abscess/collection
was defined as a localized widening of the primary or the secondary tract for more
than 1 cm ([Fig. 2]). Activity of the tract was assessed based on the signal intensity of the fistula
tract on T2 and STIR images. An active tract was defined as a tract that was completely
hyperintense on T2 and STIR images. Those tracts that were partly active (hyperintense
on STIR and T2) and partly fibrotic (hypo intense on T2 and STIR images) and the tracts
that were hyperintense on STIR but isointense or hypointense on T2-weighted images
were defined as healing tracts. On the other hand, a healed tract was completely hypointense
on both T2 and STIR images ([Fig. 3]).
Fig. 2 (a) High-resolution T2-weighted coronal, (b) axial, and (c) short tau inversion recovery magnetic resonance images demonstrate an intersphincteric
fistula-in-ano with supralevator extension. A portion of intersphincteric and supralevator
component of the tract is widened more than 10 mm representing an abscess.
Fig. 3 T2 coronal images and corresponding short tau inversion recovery (STIR) coronal images
demonstrate examples of the activity of fistula. (a, b) Active tract (asterisk) is completely hyperintense on both T2 and STIR images. (c, d) A healing tract (solid arrow) is mostly T2 hypointense with faint STIR hyperintensity.
(e, f) A healed tract (dashed arrow) is completely hypointense on both T2 and STIR images.
Anal sphincter complex was assessed for scarring and defects on T2 high-resolution
oblique axial and oblique coronal images according to definitions in previously published
work.[14] Scarring was defined as T2 markedly hypointense tissue replacing the internal and/or
external sphincter. Sphincter defect was defined as focal thinning or discontinuity
of the sphincter other than the fistula tract ([Fig. 4]). The degree of sphincter abnormality was documented in both length and circumference.
Fig. 4 (a, b) Normal anal sphincter complex on T2 coronal and axial images show T2 hypointense
external anal sphincter continuous with the levator ani muscle, T2 mildly hyperintense
internal anal sphincter continuous with the muscularis propria of the rectum and there
is T2 hyperintense intersphincteric plane. (c, d) Anal sphincter scarring (arrow) is seen as thick T2 hypointense tissue replacing
and distorting the left posterior aspect of the anal sphincter complex. Note the focal
defect in the internal anal sphincter at 4 O'clock.
Patients' details including clinical symptoms, laboratory test findings, and treatment
history were obtained from electronic medical records. For this study, MRI done at
our center during the study period was used as the index MRI. Patients were stratified
as recurrent and treatment-naïve fistula-in-ano based on the date of the index MRI
and past treatment history available on electronic medical records.
Statistical Analysis
For continuous data, the descriptive statistics were reported in terms of mean and
standard deviation (SD) and for non-normally distributed data, median values (interquartile
range) were reported. Number of patients and percentage were presented for categorical
data. To understand the difference in the fistula characteristics between the treatment
naïve and recurrent fistula-in-ano, imaging features of both these groups of patients
were tabulated and compared using one of the following tests: independent sample t-test, nonparametric Mann–Whitney U test, Pearson chi-squared test, or Spearman's
correlation coefficient. We also studied the anal sphincter morphology among patients
with recurrent fistula-in-ano and correlated anal sphincter abnormalities with imaging
features seen in patients with recurrent fistula-in-ano. A p-value of less than 0.05 was considered statistically significant. Statistical analysis
was done using SPSS v.22 software (SPSS Inc., Chicago, Illinois, United States).
Results
Demographic Data
[Fig. 1] shows the flowchart of patients included in the study. A total of 209 patients (187
males, 22 females) with a mean age of 40.6 (SD: 12.2) years and range of 18–73 years
were included for final analysis. Of the 209 patients, 106 patients had treatment-naïve
fistula-in-ano and 103 patients had recurrent fistula-in-ano. The majority (71.3%)
of patients were between 30 and 59 years, 21.1% patients were below 30 years, and
the rest (7.7%) were 60 years or above. Transsphincteric, intersphincteric, extrasphincteric,
and suprasphincteric fistula-in-ano were seen in 63.6, 33, 2.9, and 0.5%, respectively.
The most common location of the internal (n = 96, 45.9.%) and the external opening (n = 102, 48.8%) was posterior, between 5 and 7 O'clock. MRI showed no internal opening
in 25 (12%) patients and no external opening in 16 (7.7%) patients. A small proportion
(n = 13; 6.3%) had more than one internal opening with two internal openings in 11 patients
and 3 internal openings in two patients. The mean length of primary tract was 6.1 ± 3.4 cm.
Secondary tracts were seen in nearly half of the patients (49.3%) and supralevator
extension was seen in 12.9%. Underlying cause for fistula-in-ano was found in 31 patients
(14.8%): inflammatory bowel disease (n = 24), tuberculosis (n = 4), trauma (n = 3). The majority (80.4%) of patients had one or more active tracts, 15.8% patients
had partly healed fistula tracts, and 3.8% patients had completely healed tracts.
Demographic data is summarized in the [Supplementary Table 1], online only.
Comparison between Treatment-Naïve and Recurrent Fistula on MRI
[Table 1] shows the comparison between patient and fistula characteristics of treatment-naïve
and recurrent fistula-in-ano. We found no significant difference in the age and gender
distribution between the two groups. There was no difference in the fistula types,
length of the primary tract, location and the number of internal openings, secondary
tracts, supralevator extension, or collections seen in patients with treatment-naïve
and recurrent fistula-in-ano. There were significantly fewer external openings among
patients with recurrent fistula-in-ano, p = 0.005. The proportion of patients with active tracts (83 vs. 77.7%) and secondary
causes for fistula-in-ano (17.9 vs. 11.7%) was higher among the treatment-naïve group
compared to recurrent fistula-in-ano, but this was not statistically significant.
Table 1
Comparison between treatment-naïve and recurrent fistula on MRI
|
Treatment-naïve fistula (n = 106)
|
Recurrent fistula
(n = 103)
|
p-Value
|
|
Mean age
|
43.1 ±11.7
|
40.6 ± 12.2
|
0.133
|
|
Sex (M:F)
|
97:9
|
90:13
|
0.331
|
|
Type of fistula
|
|
|
|
|
− Intersphincteric
|
38 (35.8%)
|
31 (30.1%)
|
0.471
|
|
− Transsphincteric
|
64 (60.4%)
|
69 (67%)
|
|
− Supra and extrasphincteric
|
4 (3.8%)
|
3 (1.9%)
|
|
Location of internal opening
|
|
|
|
|
− Anterior (11–1 O'clock)
|
20 (21.3%)
|
23 (25.6%)
|
0.532
|
|
− Left (2–4 O'clock)
|
17 (18.1%)
|
10 (11.1%)
|
|
|
− Posterior (5–7 O'clock)
|
49 (52.1%)
|
47 (52.2%)
|
|
|
− Right (8–10 O'clock)
|
8 (8.5%)
|
10 (11.1%)
|
|
Location of external opening
|
|
|
|
|
− Anterior (11–1O'clock)
|
24 (25.3%)
|
19 (19.4%)
|
0.795
|
|
− Left (2–4 O'clock)
|
16 (16.8%)
|
19 (19.4%)
|
|
− Posterior (5–7 O'clock)
|
49 (51.6%)
|
53 (54.1%)
|
|
− Right (8–10 O'clock)
|
6 (6.3%)
|
7 (7.1%)
|
|
No. of internal opening
|
|
|
|
|
− None identified
|
12 (11.3%)
|
13 (12.6%)
|
0.777
|
|
− One
|
89 (84%)
|
82 (79.6%)
|
|
− Two or more
|
5 (4.7%)
|
8 (7.8%)
|
|
No. of external opening
|
|
|
|
|
− None identified
|
11 (10.4%)
|
5 (4.9%)
|
0.005
|
|
− One
|
68 (64.2%)
|
87 (84.5%)
|
|
− Two or more
|
27 (25.5%)
|
11 (10.7%)
|
|
Primary tract length (cm)
|
6.2 ± 3.3
|
6.0 ± 3.6
|
0.768
|
|
Mean distance (internal opening to anal verge [cm])
|
2.5 ± 1.5
|
2.6 ± 1.5
|
0.444
|
|
Mean distance (external opening to anal verge [cm])
|
3.9 ± 2.9
|
3.4 ± 2.6
|
0.244
|
|
Secondary tracts
|
49 (46.2%)
|
54 (52.4%)
|
0.370
|
|
Supralevator extension
|
13(12.3%)
|
14 (13.6%)
|
0.775
|
|
Collections
|
44 (41.5%)
|
41 (39.8%)
|
0.802
|
|
− Intersphincteric
|
26 (24.5%)
|
22 (21.4%)
|
0.352
|
|
− Ischiorectal and peri-anal
|
21 (19.8%)
|
23 (22.3%)
|
0.655
|
|
− Supralevator
|
12 (11.3%)
|
16 (15.3%)
|
0.371
|
|
− Others: abdominal wall/retroperitoneum
|
11 (10.4%)
|
14 (13.6%)
|
0.464
|
|
Secondary cause
|
19 (17.9%)
|
12 (11.7%)
|
0.202
|
|
Activity of tract
|
|
|
|
|
− Active tract
|
88 (83%)
|
80 (77.7%)
|
0.574
|
|
− Healing tract
|
15 (14.2%)
|
18 (17.5%)
|
|
|
− Healed tract
|
3 (2.8%)
|
5 (4.9%)
|
|
Abbreviation: MRI, magnetic resonance imaging.
MRI-Identified Anal Sphincter Scarring in Recurrent Fistula-in-Ano
Out of 103 patients with recurrent fistula-in-ano, 55 (53.4%) patients had features
of anal sphincter defect or scarring on MRI. Among them, there was sphincter abnormality
involving more than a third of the sphincter circumference in 41 (74.5%) patients
and more than a third of anal canal length in 33 (61.1%) patients. While external
sphincter was scarred in nearly all these patients (n = 53, 96.4%), internal sphincter scarring and defect were seen in 33 (60%) and 31
(56.4%) patients, respectively. There was severe thinning of the external sphincter
in 16 (29%) patients.
Association between Imaging Findings and Sphincter Scarring in Recurrent Fistula-in-Ano
[Table 2] compares the imaging findings between patients with and without sphincter abnormality
on MRI. The side of fistula, the length of the primary tract, the distance of internal
and external opening from the anal verge, and the number of internal and external
openings had no association with sphincter abnormality (p > 0.05). Location of internal opening had a significant association with sphincter
abnormality, p = 0.031. We found 58.8% of patients with sphincter abnormality on MRI had posteriorly
located internal opening, most commonly at 6 O'clock. Those without sphincter abnormality
most commonly (41%) had anteriorly located internal opening. There was significant
association between the location of fistula-in-ano and the extent of anal sphincter
abnormality, both in terms of the circumference (p = 0.037) and the length (p = 0.011) of anal sphincter abnormality. We found that 67.6% of patients with scarring
of more than a third of the sphincter circumference and 70% of patients with scarring
of more than a third of the sphincter length had internal opening located at 6 O'clock.
Similarly, majority (72.4%) of patients with internal sphincter defect had posterior
fistula-in-ano with internal opening at 6 O'clock.
Table 2
Comparison of fistula characteristics of patients with and without sphincter scarring
|
Sphincter scarring among recurrent fistula-in-ano (n = 104)
|
Present (n = 55)
|
Absent (n = 48)
|
p-Value
|
|
Mean age
|
41.9 ± 11.4
|
39.1 ± 13.1
|
0.259
|
|
Sex (M:F)
|
50:5
|
40:8
|
0.131
|
|
Type of fistula
|
|
|
|
|
Intersphincteric
|
14 (25.5%)
|
17 (35.4%)
|
0.573
|
|
Transsphincteric
|
39 (70.9%)
|
30 (62.5%)
|
|
Supra- and extrasphincteric
|
2 (3.6%)
|
1 (2.1%)
|
|
Location of internal opening
|
|
|
|
|
Anterior
|
7 (13.7%)
|
16 (41%)
|
0.031
|
|
Posterior
|
30 (58.8%)
|
17 (43.6%)
|
|
Primary tract length (cm)
|
5.7 ± 2.6
|
6.5 ± 4.6
|
0.329
|
|
Mean distance of internal opening from anal verge (cm)
|
2.6 ± 1.1
|
2.7 ± 1.9
|
0.891
|
|
Mean distance of external opening from anal verge (cm)
|
3.2 ± 2.4
|
3.7 ± 2.9
|
0.330
|
|
Secondary tracts
|
35 (63.6%)
|
19 (39.6%)
|
0.015
|
|
Supralevator extension/collections
|
26 (47.3%)
|
11 (22.9%)
|
0.010
|
|
Secondary cause
|
3 (5.5%)
|
9 (18.8%)
|
0.036
|
|
Activity of tract
|
|
|
|
|
– Active tract
|
42 (76.4%)
|
38 (79.2%)
|
0.083
|
|
– Healing/healing
|
13 (23.6%)
|
10 (20.8%)
|
There was significant association between the presence of sphincter abnormality on
MRI and the St James classification for fistula-in-ano. Higher grades of fistula-in-ano,
which had collections or supralevator/ translevator extensions, were associated with
the presence of anal sphincter abnormality (p = 0.010) and its severity. About 63.9% of patients with St James grade 2, 4, or 5
had associated sphincter defect or scarring. About 73.2% of patients with involvement
of more than a third of sphincter circumference (p = 0.008) and 72.7% of patients with involvement of more than a third of sphincter
length (p = 0.006) had St James grade 2, 4, or 5 fistula-in-ano. There was significant association
between the presence of sphincter abnormality and secondary tracts with higher incidence
among those with sphincter abnormality (63.6%) versus 39.6% among those without sphincter
abnormality, p = 0.015. There was a significant inverse relationship between secondary causes of
fistula-in-ano and sphincter abnormality, p = 0.036.
Discussion
Our attempts to identify if there were differences in the imaging features of fistula-in-ano
between the recurrent and treatment-naïve groups showed that there was no difference
in the type, location, and the extent of fistula-in-ano between the two groups. However,
those with recurrent fistula-in-ano had significantly fewer external openings. This
was probably due to healing and fibrosis of one or more of previously active tracts
and external openings.
Our study subjects underwent MRI with 16-channel external phased array coil. We could
adequately evaluate the anal sphincter morphology on high-resolution T2-weighted MRI
images in all the subjects irrespective of magnet strength (1.5 T or 3.0T). Endoanal
ultrasonography and endoanal MRI are the main imaging modalities used for anatomical
assessment of anal sphincter complex. However, previous studies have demonstrated
comparable performance of MRI with external phased-array coil and endoanal coil for
detecting clinically significant anal sphincter abnormalities.[15]
[16] Our study findings support the findings of these studies and establish the usefulness
of high-resolution T2 MRI with external phased-array coil for delineating morphology
of anal sphincter complex.
We found high incidence of MRI identified anal sphincter abnormality (53.4%) among
patients with recurrent fistula-in-ano. Among these patients, those with posteriorly
located fistula-in-ano, collections, supra, or translevator extension and secondary
tracts have higher incidence and worse extent of anal sphincter scarring/ defect.
These findings are in line with the classification system proposed by the SPTF, which
classifies fistula-in-ano as simple and complex based on the risk of fecal incontinence
after fistulotomy.[5] According to the SPTF classification, complex fistula-in-ano are tracks that cross
30 to 50% of external anal sphincter; for example, high trans-sphincteric, supra-sphincteric,
and extra-sphincteric types of fistula-in-ano, anterior fistula in females, multiple
tracks, and recurrent fistula-in-ano. Fistula-in-ano in patients with preexisting
incontinence, past pelvic irradiation, or Crohn's disease was also considered complex
according to SPTF classification.
The incidence of fecal incontinence after any form fistula-in-ano treatment ranged
from 0 to 64% in various studies with an average of 40%, though the majority of them
were minor incontinence.[6]
[11]
[17]
[18]
[19] Quality of life study done on patients with fistula-in-ano showed no significant
difference in the continence index between the primary and recurrence fistula-in-ano.
However, nearly double the number of patients with recurrent fistula-in-ano (36.3%)
experienced fecal urgency compared to primary fistula-in-ano (19.4%) leading to decreased
overall quality of life among those with recurrent fistula-in-ano.[20] Given the high incidence of incontinence, the decreased quality of life experienced
by patients with recurrent fistula-in-ano, and the paucity of literature available
on MRI morphology of anal sphincter complex in this group of patients, the findings
of our study are very relevant. Prior studies on outcomes of fistula surgery showed
patients with high trans-sphincteric and supra-sphincteric type of fistula-in-ano
had high postoperative incontinence.[6]
[21] The high incidence (64%) and worse MRI-identified anal sphincter scarring/defect
among patients with St. James 2, 4, and 5 type of fistula in our cohort provide an
indirect morphological correlate to prior workers findings. Unlike prior study, where
the incidence of fecal incontinence was 2.8 times higher among patients older than
45 years, there was no age- and gender-based differences in the incidence of anal
sphincter abnormalities in patients with recurrent fistula-in-ano.[6] This could be due to small sample size and alternatively, because of the population
structure in our country, where the majority of patients are young.
Anal sphincter abnormalities in patients with recurrent fistula-in-ano could be either
be due to sphincter defects from sphincter cutting surgical procedures or sphincter
scarring from surgical procedure itself or due to healing of the infection.[19]
[21]
[22] These abnormalities can cause a range of symptoms that include incontinence, fecal
impaction, and chronic anal pain. A small proportion (3.8%, n = 8) of symptomatic study subjects showed completely healed fistula tracks on MRI
and no other abnormalities. The persistent anal or perianal pain in these patients
was attributed to scarring associated with healed fistula. Thus, MRI was valuable
not only for excluding recurrent fistula-in-ano but also for identifying a cause for
patient's symptoms. A previous study assessing the value of MRI in chronic perianal
pain showed painful perianal scarring in 11%.[23]
Secondary causes were more common among treatment-naïve group and there was an inverse
relationship between secondary causes and anal sphincter abnormality. This can be
explained by the fact that recurrence and sphincter scarring might be less likely
when many of the secondary causes such inflammatory bowel disease, tuberculosis, and
trauma are adequately treated or brought under good control. We saw higher incidence
of secondary cause for fistula-in-ano in our cohort (14.8%) compared to 10% in other
published literature. This probably reflects the pattern of referral to our center,
which is a tertiary care teaching center in the region.[24]
[25]
There were few limitations, other than that posed by the retrospective study design.
We did not try to further analyze the recurrent fistula-in-ano based on its etiology
such as treatment failure, persistence of fistula-in-ano, or de-novo fistula-in-ano
and this may have affected the incidence and the degree of anal sphincter abnormality.
Since this study was conducted in a tertiary care referral center and the cohort was
from a dedicated colorectal unit, our results may be prone to a degree of referral
bias. This was indeed seen in the high rates of recurrent fistula-in-ano and secondary
causes of fistula-in-ano in our practice. Though this can affect the generalizability
of the results, we cannot disregard the fact that the results were in line with previous
clinical studies. T2 signal differences between internal anal sphincter and scar tissue
helped in delineating internal sphincter defects. However, due to somewhat similar
signal intensity of both the scar tissue and external sphincter, it was not possible
to differentiate external sphincter defects from scarring. Moreover, in patients with
recurrent fistula-in-ano, both sphincter scarring and sphincter defects can be seen
to a varying degree. Thus, further studies are needed to correlate anal sphincter
abnormalities seen in fistula-in-ano patients and with the continence score and the
quality of life. Lastly, this was mainly an imaging study which focused on the morphological
abnormalities seen on MRI. Thus, correlation with clinical continence score and follow-up
details were beyond the scope of the current study.
In conclusion, imaging features of recurrent and treatment-naïve fistula-in-ano were
mostly similar except for fewer external openings and fewer secondary causes for fistula
among the recurrent fistula-in-ano group. High-resolution T2-weighted MRI using external
phased array coil was effective in identifying sphincter abnormalities among patients
with recurrent fistula-in ano. There was high incidence (53.4%) of MRI identified
anal sphincter abnormality among recurrent fistula-in-ano, which was significantly
associated with posterior fistula, collections, supra- or translevator extension,
and secondary tracts. Though further studies are needed to correlate morphological
abnormalities of anal sphincter with fecal incontinence score, incorporating statements
on the morphology of anal sphincter complex and making specific mention of imaging
features that highly correlate with sphincter abnormality in radiology reports might
be of significant clinical value. We have provided a comprehensive list of review
areas while reporting fistula-in-ano in [Table 3].
Table 3
Review areas while reporting MRI of fistula-in-ano
|
Imaging features
|
Description
|
|
Type of fistula
|
Simple—Single internal and external opening
Complex—Multiple internal and/or external openings; multiple tracts
|
|
Low or high
|
Low—Lower third of anal sphincter
High—Upper two thirds of anal sphincter
-Puborectalis marks the ano-rectal junction
-The length of the anal canal is measured between the anorectal junction and the anal
verge
|
|
Park's classification
|
Intersphincteric
Transsphincteric
Suprasphincteric
Extrasphincteric
|
|
Internal openings of each track
|
Location in terms of clock position and distance from anal verge
|
|
External openings of each track
|
Anatomical location and distance from anal verge
|
|
Secondary tracts or ramifications
|
Branching of a primary tract
|
|
Collection
|
Presence of fluid signal collection greater than 10 mm wide and its location (intersphincteric/ischiorectal/mesorectal/
presacral/ extramesorectal pelvic space/ others such as abdominal wall/retroperitoneum,
etc.)
|
|
Supralevator extension
|
Extension of the track/ collection above the levator ani
|
|
Activity of the tract
|
Active tract—Hyperintense (fluid signal) on both T2 and T2 SPAIR
Healing tract—Partly active and partly fibrotic tracts or those that appear T2 hypointense
and T2 SPAIR hyperintense.
Fibrotic tract—Hypointense on both T2 and T2 SPAIR
|
|
Sphincter complex
|
Presence of scarring or thinning or defect involving the internal or external anal
sphincter
Does the sphincter abnormality involve greater than or less than a third of sphincter
circumference and/ or its length?
Presence of thinning or defect of levator ani
|
|
Secondary causes
|
Present/ absent, if present what cause?
|
Abbreviations: MRI, magnetic resonance imaging; SPAIR, spectral attenuated inversion
recovery.
