Long‐term Outcomes after Seton Placement for Perianal Fistulae with and without Crohn’s Disease

Abstract

Aim
Perianal sepsis in Crohn’s fistulae is managed with antibiotics and surgical drainage; a non-cutting seton is used for an identified trans-sphincteric fistula tract. Optimal management following seton placement for initial control of perianal sepsis remains to be determined. Our main objective was to assess success rates of curative surgery, seton removal, or long-term indwelling seton in patients with and without Crohn’s.

Methods
This was a retrospective cohort of consecutive patients with a perianal fistula treated with a non-cutting seton between 2010-2019, including 83 Crohn’s patients and 94 patients without Crohn’s. Initial control of symptomatic perianal infection with seton and subsequent healing and reintervention rates were compared between the three post-seton management strategies.

Results
177 patients, 61% male and 83.1% with complex fistulae were followed for a median of 23 months [interquartile range 11-40]. Immunomodulatory treatment was used in 90.4% of Crohn’s patients after seton placement. Good initial control of perianal infection with seton was achieved in Crohn’s and non-Crohn’s patients, 92.9% and 96.7%, respectively (p=0.11). Overall fistula healing or control for Crohn’s and non-Crohn’s patients, was 64% and 86% (p=0.1) after curative surgery, 49% and 71% after seton removal (p=0.21), and 58% and 50% with long-term seton placement (p=0.72). Overall reintervention for recurrence was 83% in Crohn’s versus 53.1% in non-Crohn’s patients during the follow-up period (p=0.002).

Conclusion
Definitive surgery was possible only in a minority of Crohn’s patients. Long-term seton management was an effective option in patients with Crohn’s disease with acceptable improvement and recurrence rates.

Full text

Colorectal Disease. 2021;23:2407–2415.
|
2407wileyonlinelibrary.com/journal/codi
INTRODUCTION
Perianal fistulas have a prevalence of approximately 18 per 100,000
in the general population, based on a recent UK estimate [1]. While
simple fistula- in- ano is a benign pathology with favourable cure
rates, complex fistulas are challenging to control [2]. Crohn's disease
(CD) contributes to up to a quarter of complex fistula cases [1]. Up to
one- third of patients with CD may experience penetrating perianal
problems associated with a longer duration of disease and inflamma-
tion in the rectum [3– 5].
Surgical management of simple fistulas most commonly involves
a fistulotomy. However, in cases where a significant portion of the
sphincter muscle is involved or the fistula has a complex tract, a
seton may be used to resolve sepsis and preser ve continence. Once
Received: 1 March 2021
|
Revised: 7 April 2021
|
Accepted: 19 May 2021
DOI : 10.1111 /cod i.15771
ORIGINAL ARTICLE
Long- term outcomes after seton placement for perianal fistulas
with and without Crohn’s disease
Mohammad Ali K. Motamedi1 | Sara Serahati2 | Luckshi Rajendran3 | Carl J. Brown1 |
Manoj J. Raval1 | Ahmer Karimuddin1 | Amandeep Ghuman1 | Paul T. Phang1
© 2021 Association of Coloproctology of Gre at Britain and Ireland
1Colorectal Surgery Division, Department
of Surger y, St Paul’s Hospital, Vancouver,
BC, Ca nada
2Department of Public Health, University
of Saskatchewan, Saskatoon, SK, Canada
3Divisio n of Gener al Surge ry, University of
Toronto, Toronto, ON, Canada
Correspondence
Paul T. Phang, 1081 Burrard Street, St
Paul’s Hospital, Vancouver, BC, Canada.
Email: tphang@providencehealth.bc.ca
Abstract
Aim: Perianal sepsis in Crohn's disease (CD) fistulas is managed with antibiotics and surgi-
cal drainage; a noncutting seton is used for an identified transsphincteric fistula tract. The
optimal management following seton placement for initial control of perianal sepsis re-
mains to be determined. Our main aim was to assess the success rates of curative surgery,
seton removal or long- term indwelling seton in patients with and without CD.
Method: This was a retrospective cohort of consecutive patients with a perianal fistula
treated with a noncutting seton between 2010 and 2019, including 83 CD patients and
94 patients without CD. Initial control of symptomatic perianal infection with a seton and
subsequent healing and reintervention rates were compared between the three postse-
ton management strategies.
Results: A total of 177 patients, 61% male and 83.1% with complex fistulas, were fol-
lowed for a median of 23 months (interquartile range 11– 40 months). Immunomodulatory
treatment was used in 90.4% of CD patients after seton placement. Good initial control
of perianal infection was achieved with a seton in CD and non- CD patients, at 92.9% and
96.7%, respectively (p = 0.11). Overall fistula healing or control for CD and non- CD pa-
tients was, respectively, 64% and 86% (p = 0.1) after curative surgery, 49% and 71% after
seton removal (p = 0.21) and 58% and 50% with long- term seton placement (p = 0.72).
Overall reintervention for recurrence was 83% in CD versus 53.1% in non- CD patients
during the follow- up period (p = 0.002).
Conclusion: Definitive surgery was possible in only a minority of CD patients. Long- term
seton management was an effective option in patients with CD with acceptable improve-
ment and recurrence rates.

2408
|
MOTAMED I ET Al.
sepsis is resolved, either the seton is left in situ to provide adequate
drainage or curative surgery is performed, using techniques such as
endorectal advancement flap [6], ligation of intersphinc teric fistula
tra ct (LIF T) [7], fibri n glu e [8] or fis tula plug [9], based on the individ-
ual surgeon's recommendation. In some patients, ongoing perianal
sepsis may require a diverting stoma or proctectomy.
Success rates are consistently low with all surgical options in
patients with CD compared with those without [10]. Historical co-
horts report healing rates of approximately 50% in CD, with many
patients having compromised continence or progressing towards a
proctectomy [11– 13]. The use of antitumour necrosis factor- alpha
(anti- TNF), immunosuppressive and newer biological medications
may promote healing by decreasing mucosal inflammation [9, 14, 15].
Optimal management of CD fistulas remains uncertain due to a lack
of well- designed randomized studies with adequate follow- up [16].
In this study, we reviewed the outcomes of all our patients with
perianal fistulas, with or without CD, af ter initial seton placement
during a 10- year period. We evaluated the time to resolution of peri-
anal infection with initial seton placement and subsequent reinter-
vention and healing rates, and looked at outcomes of three possible
postseton management options: curative surgical attempt, seton
removal and long- term indwelling seton.
METHOD
Study design and participants
This is a retrospective study of consecutive patients with or without
CD who presented with perianal fistulas and were treated using a
seton from January 2010 to April 2019 at a tertiary colorectal aca-
demic hospital. The first author, with the help of a research coor-
dinator, compiled a list of patients based on procedure billing code
for examination under anaesthesia (n = 712) using institutional elec-
tronic medical records. All patients were screened to determine
seton placement for a perianal fistula. All medical records were re-
viewed until the last available visit. We excluded patient s with an
existing seton in situ, other forms of perianal disease including non-
healing wounds, fissures and hidradenitis, as well as patient s with
anal stricture, a proximal diverting stoma, c ancer, history of radiation
or incomplete follow- up records.
Institutional review board approval was obtained before study
initiation (H18- 01025).
Surgical procedures
Initial surgery was intended to provide adequate drainage for peri-
anal fistulas and associated collections or cavities. Fistulotomy of
fistula tract s not involving the external anal sphincter muscle was
performed and underlying abscess cavities were deroofed. The
sphincteric por tion of the fistula tract was treated with a loose non-
cutting seton to preserve continence. The choice of seton material
was either a double- 0 Ethibond suture or a plastic vessel loop, at the
surgeon's discretion. For large cavities or supralevator extension of
the fistula/abscess, a Penrose drain or Malecot catheter was placed.
Gauze packing was used to prevent premature skin closure and pro-
mote healing by secondary intention. Subsequently, the decision for
any of the three approaches depended on resolution of the fistula
and the presence of active CD. Curative surgery was attempted in
suitable candidates with a reasonable chance of success (minimal fis-
tula drainage and no active CD), based on the fistula anatomy (using
pelvic MRI) and the individual surgeon's experience and recommen-
dations. The options included advancement flap, the LIFT proce-
dure, fistula plug or simple fistulotomy. Patients had a direct role in
their treatment decision.
Study variables and outcomes
Baseline data, including primary or recurring disease, perianal dis-
ease duration, fistula complexity (i.e. the presence of multiple tracts,
supralevator extension or a horseshoe configuration), the number
of internal and external openings, loc ation of the fistula and history
of diabetes mellitus, were collected. For CD patients, additional in-
formation on the duration of CD, the pattern of disease, the pres-
ence of active rectal inflammation/ulceration based on endoscopic
evaluation within the previous 6 months and medication histor y,
including the start date and duration of anti- TNFs or other monoclo-
nal antibodies (MAb) and/or immunosuppressive medications, was
also collected. All postoperative events including seton removal (in-
tentional or inadver tent), recurrence of symptoms, reinter ventions
requiring anaesthesia, curative surgical attempts, stoma creation,
proctocolectomy/abdominoperineal resection (APR), presentation
to the emergency department and repeat MRI for fistula evaluation
were documented.
The time until initial symptomatic control of perianal infection
with the seton was determined by the subjective improvement of
pain and discomfort, absence of fever, decreased purulent discharge
and no dependence on antibiotics for symptom control. Subsequent
outcomes included healing and reintervention rates after curative
surgery, simple removal of the seton and long- term indwelling seton
in situ. The healing outcome for each patient was categorized into
What does this paper add to the literature?
We have demonstrated that an indwelling seton has ac-
ceptable healing and recurrence rates for the treatment
of Crohn's disease (CD) perianal fistulas in the long- term.
Corrective surgery is not always suitable in CD because of
luminal disease and may lead to recurrence, despite multi-
modal therapy. CD patients continue to have inferior out-
comes for complex fistulas compared with patients who do
not have CD.

|
2409
MOTAMEDI ET Al .
one of the following four possibilities: (a) healing of the fistula, with
closure of the external opening (not possible with a long- term seton),
(b) controlled, minimally symptomatic fistula with little to no drain-
age or pain upon gentle compression, (c) ongoing symptomatic drain-
age and pain or (d) persistent infection requiring a diverting stoma
or resection. For those managed with a long- term indwelling seton,
outcomes were graded as (b), (c) or (d), as above. Using a patient-
centred perspective, we reported healing if it was achieved during
a patient's follow- up after their treatment approach. Reinter vention
was considered independently and reported as cumulative probabil-
ity and rates after each approach.
Statistical analysis
Data are summarized as count (percentage) or median [interquartile
range (IQR)]. Categorical variables were compared using chi- square
or Fisher 's exact test. For statis tical compariso n of healing outcomes ,
healed and minimally symptomatic fistulas were grouped and com-
pared with ongoing and persistently symptomatic fistulas. Healing
and reintervention outcomes were recorded as time- to- event data
from the time of seton placement. Life tables and Kaplan– Meier
curves with the log- rank (Mantel– Cox) test were used to compare
CD and non- CD groups for cumulative probabilities of outcomes
during the follow- up period until outcome or censoring. Hazard ra-
tios (HRs) and 95% confidence intervals (CIs) were calculated from
Cox proportional hazard models adjusted for baseline imbalances
in age, sex and fistula characteristics. Stat a (StataCorp LLC, College
Station, TX, USA) was used for statistical analysis. A two- sided p-
value of <0.05 was considered significant.
RESULTS
Baseline characteristics
A total of 177 patients with fistulas undergoing seton placement were
in clu d ed: 83 wi th CD, aged 40 . 2 ± 12 . 6 ye a rs an d 53% mal e , we r e co m-
pared with 94 non- CD patients, aged 46.7 ± 12.2 years and 69% male
(p < 0.001 and p < 0.03, respectively). Of thes e, 78.0% had a recurr ing
fistula and 83.1% had complex fistula configurations. Baseline patient
charac teristics are summarized in Table 1. Patients were followed for a
median of 23 (IQR 11– 40) months until healing or censoring.
In the CD group, 81.9% had established perineal disease be-
fore the presentation, 61% had colonic and 54.2% had ileal in-
vol vement; 12.0% of CD patient s had only pe ri anal manife statio ns
whereas 19.3% had perianal, colonic and ileal involvement. Rectal
inflammation seen on a recent scope was present in 37.3% of CD
patients at bas el in e. At pre se nt ation, 53 % we re on MAb tr ea tm ent
and 19.3% were on immunosuppressives. Postseton placement,
90.4% of CD patients received biological and/or immunosuppres-
sive medications.
Initial symptomatic control of perianal infection
with seton
Perianal sepsis was adequately controlled with initial seton place-
ment in 72 CD and 88 non- CD patients. This corresponded to
probabilities of 92.9% (95% CI 85.1– 97.2) in CD and 96.7% (95% CI
91.4– 99.1) in non- CD patients (log- rank χ2 = 2.520, p = 0.11). The
median times to symptomatic perianal infection control with a seton
was 3 (IQR 2– 6) months in CD and 2 (IQR 1– 5) months in non- CD
patients (p = 0.09).
TAB LE 1 Baseline characteristics of patients with perianal fistula
based on the presence of Crohn's disease
Variable
Crohn's
(n = 83)
Non- Crohn's
(n=94) p
Age (years) 40.2 ± 12 .6 46.7 ± 12.2 0.001
Male sex 44 (53.0) 65 (69.1) 0.03
Diabetes mellitus 6 (7.2) 4 (4.3) 0. 51
Fistula presentation 0.85
Primary 19 (22.9) 20 (21.3)
Recurring fistula 64 (77.1) 74 (78.7)
Simple 10 (12 .0) 20 (21.3) 0.11
Complexa 73 (88.0) 74 (78.7)
Abscess present 65 (78.3) 71 (75.5) 0.72
Single tract 12 (14.5) 22 (23.4) 0.044
Multiple tracts 30 (36.1) 19 (20.2)
Horseshoe
configuration
39 (47.0) 46 (48.9)
Supralevator
extension
2 (2.4) 7 ( 7.4)
Fistula type 0.003
Intersphincteric 18 (21.7) 7 (7.4 )
Transsphincteric 62 (74.7) 80 (85.1)
Suprasphincteric 3 (3.6) 7 ( 7.4)
Number of internal
openings
0.002
165 (78.3) 90 (95.7)
2 or more 18 (21.7) 4 (4.3)
Number of external
openings
0.19
136 (43.4) 49 (52.1)
2 or more 47 (56.6) 45 (47.9)
Perianal disease
duration (months)
31 [11.5– 82] 32 [9– 73] 0.79
Follow- up (months) 31 [15– 54] 18 [10– 3 3. 7] 0.002
Note: Italics indicate significant values (p ≤ 0.05).
Values are number (percentage), mean ± st andard deviation, or median
[interquartile range].
aPresence of multiple trac ts/openings, high fistula, supralevator
extension, horseshoeing or rectovaginal component.

2410
|
MOTAMED I ET Al.
Postseton management: curative surgery, seton
removal or long- term indwelling seton
Subsequent to initial seton management, 84 patients underwent cu-
rative surgery, 57 had their setons removed and 36 were managed
with a long- term indwelling seton (Table 2). Overall, for all postse-
ton management strategies, healing or symptomatic control of the
fistula was achieved in 54.2% of CD and 79.8% of non- CD patients
(p < 0.001). The overall probabilit y of undergoing reintervention for
recurrence was 83.0% (95% CI 72.5– 90.1) in CD and 53.1% (95%
CI 41.5– 65.6) in non- CD patients (log- rank χ2=9. 418, p < 0.002;
Figure 1).
Curative surgery was attempted in 14 (16%) CD patients com-
pared with 70 (74%) non- CD patient s. Healing or symptomatic con-
trol of the fistula was achieved in 64% of CD and 86% of non- CD
patients after curative surgery (p = 0.1). Nine of 14 CD (64%) and 18
of 70 non- CD (25%) patients experienced recurrence requiring rein-
tervention after an attempt at corrective surgery (p = 0.01; Figure 2).
Twelve of these 14 CD patient s received biologic al medication post-
surgery. No difference was observed between fistulotomy (n = 32),
advancement flap (n = 38), LIFT (n = 8) or fistula plug (n = 6) in terms
of fistula healing or reintervention rates in either group (Table 3).
In the 57 patients with simple seton removal, the median time to
removal of the seton was 10 (IQR 6– 17) months. Healing or symp-
tomatic control of fistulas was not statistically different between
those with and without CD, at 49% and 71%, respectively (p = 0.2).
In this cohort, 68% of CD patients required reintervention for recur-
rence compared with 31% of non- CD patients ( p=0.049).
In the 36 patients managed with a long- term indwelling seton, the
median duration of having a seton in situ (i.e. until the last follow- up)
Variable Crohn's (n = 83)
Non- Crohn's
(n=94) p
Fistula outcome by management strategy <0.0 01
Long- term seton (n = 36)
Healed or controlled fistulaa 15 (58) 5 (50) 0.72
Ongoing or persisting fistula 11 (42) 5 (50)
Seton removed (n = 57)
Healed or controlled fistula 21 (49) 10 (71) 0.21
Ongoing or persisting fistula 22 (51) 4 (29)
Curative surgery (n = 84)
Healed or controlled fistula 9 (64) 60 (86)b 0.11
Ongoing or persisting fistula 5 (36) 10 (14)
Tot al
Healed fistula 8 (9.6) 26 (27.7) <0.0 01
Controlled fistula 37 (44.6) 49 (52.1)
Ongoing symptomatic fistula 26 (31.3) 17 (18.1)
Persisting fistula 12 (14.5) 2 (2.1)
Initial control of perianal infection with setonc 92.9% (95% CI
8 5 . 1 – 9 7 . 2 )
96.7% (95% CI
9 1 . 4 – 9 9 . 1 )
0.11
Recurrence requiring reinterventionb,c 83.0% (95% CI
72.5– 90.1)
53.1% (95% CI
4 1 . 5 – 6 5 . 6 )
0.002
Diverting stoma 9 (10.8) 1 (1.1) 0.0 07
Proctocolec tomy 1 (1. 2) 00.56
Abdominoperineal resection 1 (1. 2) 1 (1.1)
Emergency department visit 4 3 (51. 8) 25 (26.6) 0.001
Repeat imaging (MRI or CT) 55 (66.3) 38 (40.4) 0.001
Notes: Italics indicate significant values (p ≤ 0.05).
Values are n (%) unless stated otherwise.
Abbreviations: CI, confidence interval; CT, computed tomography; MRI, magnetic resonance
imaging.
aHealed: fistula closed with no drainage. Controlled: minimal drainage or pain. Ongoing: ongoing
drainage and pain. Persisting infec tion: requiring ostomy or resection.
bAmong management options, curative surger y had the highest healing/symptomatic control rate
in non- Crohn’s patients (p = 0.022).
cCumulative probabilities based on Kaplan– Meier analysis.
TAB LE 2 Fistula management
outcomes and control of symptomatic
infection post- seton placement in patients
with and without Crohn's disease

|
2411
MOTAMEDI ET Al .
was 31.5 (IQR 15– 48.5) months. Control of fistula symptoms was
maintained in 58% of the CD group and 50% of the non- CD group
(p = 0.7). In this cohort, 65% of CD and 50% of non- CD patients un-
derwent reintervention for recurring infection during the follow- up
period (p = 1.0).
In patient s with CD, the three approaches of curative surgery
(n = 14), seton removal (n = 43) and long- term seton (n = 26) had sim-
ilar rates of healing or fist ula control, at 6 4%, 49% and 58% (p = 0.5),
respectively. Reintervention rates were also similar after curative
surgery (64%), seton removal (68%) or an indwelling seton (65%,
p = 0.6). In patients without CD, curative surgery had the highest
fistula healing or control rate (86%, p = 0.02) and the lowest reinter-
vention rate (25%, p = 0.041).
A diverting stoma was required in nine CD patients and one
non- CD patient (p = 0.0 07). Two CD patients and one non- CD patient
underwent proctocolectomy or APR. CD patients required more emer-
gency department visits (51.8 vs. 26.6%, p = 0.001) and repeat imaging
(66.3 vs. 40.4%, p = 0.001) during their follow- up period.
Prognostic factors and associations
In multivariate analysis, initial symptomatic control of perianal infec-
tion with a seton was inversely associated with recurring fistula (HR
0.55, 95% CI 0.34– 0.88, p = 0.013) and having two or more inter-
nal fistula openings (HR 0.50, 95% CI 0.25– 0.99, p = 0.049, Table 4).
Reintervention for recurrence had significant associations with CD
(HR 1.62, 95% CI 1.0 09– 2.62, p = 0.046), diabetes mellitus (HR 3.19,
95% CI 1.51– 6.74, p = 0.002) and supralevator extension of the
fistula (HR 3.17, 95% CI 1.13– 8.88, p = 0.028). No significant dif fer-
ences in healing or reintervention rates were obser ved between CD
patients on or off immunomodulatory medications after any of the
three management approaches. There was also no difference be-
tween infliximab (n = 27), adalimumab (n = 30), vedolizumab (n = 9)
or ustekinumab (n = 7). Time to seton removal or curative surgery
was not significantly associated with healing or reintervention rates
in either CD or non- CD patients.
DISCUSSION AND CONCLUSIONS
Initial symptomatic perianal infection was well controlled with seton
placement in patients with and without CD. After initial control, cu-
rative surgery had the highest success rate in non- CD patients. In
the few patients with CD in whom curative surgery was attempted,
more than half experienced recurrence. Use of a long- term indwell-
ing seton had comparable improvement and recurrence rates in CD
patents. The presence of CD was associated with more reinterven-
tions for recurrence despite multimodal immune- modulating treat-
ment in the majority of patients.
This study is among the few published reports to directly com-
pare fistula outcomes in CD and non- CD patients. Our results reflect
the ongoing challenges of complex perianal fistula management in
CD patients. This study also reveals a long- term seton as an option
for managing fistulas in CD, and these were utilized in about one-
third of our patients. Our low rate of attempting curative surger y in
CD patients may relate to unresolved gross or microscopic presence
of anorectal and colonic inflammation that is a contraindication to
FIGURE 1 Cumulative probability of
reintervention for recurrence in patients
with and without Crohn's disease
No. at risk
Crohn’s83
10
0
20
30
40
50
60
70
80
90
100
0122436486072
43 18 942
Non-Crohn’s943820105 4
log rank χ2=9.418, p=0.002
Time after seton (months)
Cumulative probability %
Reintervention for recurrence
Crohn's
Non-Crohn's

2412
|
MOTAMED I ET Al.
attempting curative surger y, as well as surgeon and patient prefer-
ence for conservative management, considering the high eventual
recurrence rates.
Limitations of this study include the small cohort size, the lack of
protocolized strategy for curative surgery as well as the lack of pro-
spective blinded healing scoring [17]. A validated objective evalua-
tion of functional status and quality of life that could better reflec t
patient- centred outcomes was not performed. Time- to- event data in
our study may partly reflect waiting times at our centre rather than ac-
curate time estimates for healing outcomes. Additionally, 10 patients
in our study presented with severe perianal fistulizing disease with-
out any other manifestation of CD and received a clinical diagnosis of
CD. Three of these patients had received biological treatment and/or
immunosuppressives at baseline, and an additional six were started
on biologicals postoperatively. These patients may have different out-
comes compared with t ypical CD patients with biopsy- proven luminal
involvement. Finally, considering the lack of treatment randomization
and the individual approach to selecting surgical options, we cannot
recommend a preferred surgic al procedure, especially in CD patients
in whom corrective surgery was performed in a minority.
FIGURE 2 Postseton outcomes based
on different management approaches
in patient s with and without Crohn's
disease. Data in the last column boxes
are n (%) of Crohn's and non- Crohn's
patients. Significant differences in each
box are signified by an asterisk. aFisher's
exact test p = 0.049. bFisher's exact test
p = 0.01
Initial seton
Long-term seton in situ (n=36)
No reintervention
9 (35%)3 (30%)
Reintervention
17 (65%)7 (70%)
Seton removed (n=57)
No reintervention
12 (32%)9 (69%)
Reintervention a
25 (68%)* 4 (31%)
Curative surgery (n=84)
No reintervention
5 (36%) 52 (74%)
Reintervention b
9 (64%)* 18 (26%)
Reintervention for recurrence
Crohn’sNon-Crohn’s
TAB LE 3 Healing and reinter vention outcomes af ter curative surgical attempts in patients with and without Crohn's disease (CD)
Procedure N
Healed or controlled fistulaa Required reinterventiona
CD (n = 14) Non- CD (n = 70) pCD (n = 14) Non- CD (n = 70) p
Fistulotomy 32 4 (of 6) 24 (of 26) 0.15 2 (of 6) 2 (of 26) 0.31
Advancement flap 38 4 (of 6) 28 (of 32) 0.23 6 (of 6) 10 (of 32) 0.06
LIFT 8 – 6 (of 8) - – 4 (of 8) –
Fistula plug 61 (of 2) 2 (of 4) - 1 (of 2) 2 (of 4) –
Tot alb 84 510 918
Abbreviation: LIFT, Ligation of intersphincteric fistula tract.
aHealed: fistula closed with no drainage. Controlled: minimal drainage or pain.
bThere was no significant intragroup difference in outcomes between the procedures.

|
2413
MOTAMEDI ET Al .
Healing rates for loose seton management are generally favour-
able when viewed as a less invasive option. Healing was achieved
in all of a cohort of 200 patients, 23% with CD, managed with se-
quential seton changes with only a 6% recurrence rate; only 1%
could not tolerate the presence of a long- term seton [18]. However,
healing rates decline over time, and many patients require repeat
interventions or definitive surgeries [19, 20]. A recent report of 23
patients with CD managed with a long- term seton reported clinical
response in only 40% of patients [21]. Similarly, the randomized PISA
trial terminated early due to the long- term seton arm having a higher
reintervention rate (10 of 15 patients), a finding not observed in its
parallel nonrandomized registry [22]. These rates, however, may
only partly guide the treatment decisions as patient preferences and
expectations are central to treatment decisions. The overall prob-
ability of recurrence requiring reintervention after the initial seton
placement was 83% in CD patients in our study, irrespective of the
subsequent treatment strategy. In this group, the decision to remove
the seton in those experiencing reasonable improvement and hope
for healing on multimodal immune- modulating medications remains
a challenge. The presence of anorectal and colonic inflammation
contraindicates an attempt at curative surgery or seton removal. Our
data showed higher reintervention rates after seton removal in CD
versus non- CD patients. In contrast, long- term seton management
in CD had a similar reintervention rate to that in non- CD patients. A
limitation of long- term seton placement is discomfort, possibly re-
lated to the t ype of suture or knot [23]. The decision to remove the
seton is preferred by some patient s who have improved, can no lon-
ger tolerate the seton and do not wish to have an attempt at curative
surgery that has suboptimal chance of success.
Of the 14 CD patients undergoing curative surgery in our study,
nine (64%) experienced healing or good control of their fistula
symptoms postoperatively. However, four of these patients even-
tually experienced recurrence and required reintervention later.
Advancement flaps achieved healing or symptomatic control in four
of the six (66.7%) CD patients, all of whom required reinter vention
during the follow- up period. In non- CD patients, advancement flaps
initially resulted in healing or good symptomatic control in 28 of the
32 patients (87%). However, 10 (31%) required a reintervention for re-
currence during follow- up. In other series with CD patients, advance-
ment flaps have had an overall success rate of up to 64% [24] and the
LIFT procedure between 69% and 76% [25, 26]. The LIFT procedure
was not attempted in any of our CD patient s. Some authors have also
recommended against a second- stage procedure in CD patients due
to compromised healing, which may lead to disappointing results
[27]. CD patients commonly receive concomitant medical therapy to
achieve mucosal healing, and optimal doses of anti- TNF agents remain
pivotal in the induction and maintenance of fistula control [28– 31]. As
such, CD patients should be counselled that curative surgery is associ-
ated with significant rates of recurrence requiring reintervention. This
may explain the low numb er of attempts at cur ative surgery in patients
with CD in our cohort. In some, reluctance to try another surgery also
plays a role in t his decision since many patients have had long- standing
recurrent disease with not much improvement with repeated inter-
ventions. Newer surgical techniques such as mesenchymal stem cell
injections, video- assisted anal fistula treatment and laser ablation of
the fistula trac t, may in the future serve as more successful surgical
options in patients with CD [9].
A thoro ug h evalu at ion of pe rianal CD us i n g MR I , CT sc an or endo-
anal ultrasound is recommended to ensure optimal surgical drainage
[32]. Recurring fistulas and those with complex configurations with
two or more internal openings were associated in our cohort with
worse initial control with a seton. The presence of CD, supralevator
Baseline variable
Univariate Multivariate
Hazardratio(95%
CI) p
Hazardratio(95%
CI) p
Crohn's disease 1.91 (1. 24– 2.94) 0.003 1.62 (1.009– 2. 62) 0.049
Age 0.98 (0.96– 0.99) 0.018 0.98 (0.96– 1.001) 0.07
Female sex 1.11 (0.73– 1.69) 0.6
Diabetes mellitus 2.28 (1.14– 4.56) 0.019 3.19 (1.51– 6.74) 0.0 02
Recurring fistula 1.36 (0.79– 2.35) 0.25 -
Perianal disease duration 1.00 (0.998– 1.003) 0.4 -
Two or more internal
openings
2.14 (1.27– 3.60) 0.004 1.70 (0.95– 3.05) 0.070
Two or more exter nal
openings
1.34 (0.88– 2.04) 0.16 1.15 (0.73– 1.81) 0.5
Fistula type (vs. single)
Multiple tracts 1.75 (0.92– 3.35) 0.08 1.16 (0.51– 2.66) 0.7
Horseshoe 1.32 (0.72– 2.43) 0.3 1.01 (0.51– 1.99) 0.9
Supralevator 2.74 (1.10– 6.86) 0.030 3.17 (1.13– 8.8 8) 0.028
Abscess 1.08 (0.65– 1.80) 0.7 -
Notes: Italics indicate significant values (p ≤ 0.05).
TAB LE 4 Association of baseline
patient and fistula characteristics with
reintervention for recurrence

2414
|
MOTAMED I ET Al.
fistula extension and diabetes mellitus were also prognostic factors
for requiring reintervention for recurrence in our cohort.
In conclusion, good initial control of symptomatic perianal infec-
tion was achieved using drainage, placement of a noncutting seton
and antibiotics in patients with and without CD. Post seton outcomes
of healing and recurrence were not significantly dif ferent in the long
term between indwelling seton, seton removal and curative surgery
in patients with CD. Despite multimodal immune- modulating ther-
apy, CD fistulas were associated with more frequent recurrence and
reinterventions. In view of high long- term failure rates of curative
surgery in CD patients with complex fistulas, an indwelling seton re-
mains a reasonable conservative management option. The presence
of anorectal and colonic inflammation and ongoing fistula complexi-
ties were in our experience viewed as contraindications to attempt-
ing curative surger y or removing the seton. Future research should
prospec tively evaluate healing, functional status and quality of life
to better determine the optimal approach in this patient population.
ACKNOWLEDGEMENT
The authors would like to thank Ms Kathleen Tom for her contribu-
tion to the project.
CONFLICTS OF INTEREST
None declared.
ETHICAL APPROVAL
Institutional review board approval was obtained before study initia-
tion (H18- 01025).
DATA AVAIL A BIL ITY S TATE M EN T
The data that support the findings of this study are available from
the corresponding author upon reasonable request.
ORCID
Mohammad Ali K. Motamedi https://orcid.
org/0000-0003-4205-2848
REFERENCES
1. Hokkanen SR, Boxall N, Khalid JM, Bennett D, Patel H. Prevalence
of anal fistula in the United Kingdom. World J Clin Cases.
2019;7(14):1795– 8 04.
2. Sandborn WJ, Fazio VW, Feagan BG, Hanauer SB . AGA tech-
nical review on perianal Crohn’s disease. Gastroenterology.
2003;125(5):1508– 30.
3. Schwartz DA, Loftus E V Jr, Tremaine WJ, Panaccione R, Harmsen
WS, Zinsmeis ter AR, et al. The natural history of fistulizing
Crohn's disease in Olmsted County, Minnesota. Gastroenterology.
2002;122(4):875– 80.
4. Scharl M, Rogler G, Biedermann L. Fistulizing Crohn's disease. Clin
Transl Gastroenterol. 2017;8(7):e106- e.
5. Panés J, Rimola J. Perianal fistulizing Crohn's disease: patho-
genesis, diagnosis and therapy. Nat Rev Gastroenterol Hepatol.
2017;14(11):652– 6 4.
6. Aguilar PS, Plasencia G, Hardy TG Jr, Hartmann RF, Stewar t WR.
Mucosal a dvancement in the treatment of anal fistula. Dis Colon
Rectum . 1985;28(7):496– 8.
7. Rojanasakul A. LIFT procedure: a simplified technique for fistula- in-
ano. Tech Coloproctol. 2009;13(3):237– 40.
8. Abel ME, Chiu YS, Russell TR, Volpe PA. Autologous fibrin glue
in the treatment of re ctovaginal and complex fistulas. Dis Colon
Rectum . 1993;36(5):447– 9.
9. Kotze PG, Shen B, Lightner A, Yamamoto T, Spinelli A, Ghosh S,
et al. Modern management of perianal fistulas in Crohn's disease:
future directions. Gut . 2018;67(6):1181– 94.
10. Bemelman WA, Warusavitarne J, Sampietro GM, Serclova Z, Zmora
O, Luglio G, et al. ECCO- ESCP consensus on surgery for Crohn's
disease. J Crohns Colitis. 2018;12(1):1– 16.
11. Williams JG, Rothenberger DA, Nemer FD, Goldberg SM. Fistula-
in- ano in Crohn's disease. Results of ag gressive surgical treatment.
Dis Colon Rectum. 1991;34(5):378– 84.
12. Scott HJ, Northover JM. Evaluation of surgery for perianal Crohn's
fistulas. Dis Colon Rectum. 1996;39(9):1039– 43.
13. Hellers G , Bergstr and O, Ewerth S, Holmström B. Occurrence and
outcome af ter primary treatment of anal fis tulae in Crohn's dise ase.
Gut. 1980;21(6):525– 7.
14. Regueiro M, Mardini H. Treatment of perianal fistulizing Crohn's
disease with infliximab alone or as an adjunct to exam under anes-
thesia with seton placement. Inflamm Bowel Dis. 2003;9(2):98– 103.
15. Kotze PG, Albuquerque IC, da Luz MA, Tonini WB, Olandoski M,
Coy CS. Perianal complete remission with combined therapy (seton
placement and anti- TNF agents) in Crohn's disease: a Brazilian mul-
ticenter observational study. Arq Gastroenterol. 2014;51(4):284– 9.
16. de Groof EJ, Sahami S, Lucas C, Ponsioen CY, Bemelman WA ,
Buskens C J. Treatment of perianal fistula in Crohn's disease: a
systematic review and meta- analysis comparing seton drain-
age and anti- tumour necrosis factor treatment. Colorectal Dis.
2016;18(7):667– 75.
17. Present DH , Rutgeerts P, Targan S, Hanauer SB, Mayer L, van
Hogezand R A, et al. Infliximab for the treatment of fistulas in pa-
tients with Crohn's disease. N Engl J Med. 1999;340(18):1398– 405.
18. Kelly ME, Heneghan HM, McDermott FD, Nason GJ, Freeman C,
Martin ST, et al. The role of loose seton in the management of
anal fistula: a multicenter s tudy of 20 0 patients. Tech Coloproctol.
2 0 1 4 ; 1 8 ( 1 0 ) : 9 1 5 – 9 .
19. Buchanan GN, Owen HA, Torkington J, Lunniss PJ, Nicholls RJ,
Cohen CR. Long- term outcome following loose- seton technique for
external sphinc ter preservation in complex anal fis tula. Br J Surg.
2 0 0 4 ; 9 1 ( 4 ) : 4 7 6 – 8 0 .
20. Galis- Rozen E, Tulchinsky H, Rosen A, Eldar S, Rabau M, Stepanski
A, et al. Long- term outcome of loose seton for complex anal fistula:
a two- centre study of patients with and without Crohn's disease.
Colorectal Dis. 2010;12(4):358– 62.
21. Mujukian A, Zaghiyan K, Banayan E , Fleshner P. Outcomes of de-
finitive draining seton placement for complex anal fistula in Crohn’s
disease. Am Surg. 2020;86(10):1368– 72.
22. Wasmann KA, de Groof EJ, Stellingwer f ME, D’Haens GR, Ponsioen CY,
Ge cse KB, et al . Tre at me nt of peri an al fist ul as in Cr oh n’s dis ease, set on
versus anti- TNF versus surgical closure following anti- TNF [PISA]: a
randomised controlled trial. J Crohns Colitis. 2020;14(8):1049– 56.
23. Kristo I, Stift A , Staud C , Kainz A, Bachleitner- Hofmann T,
Chitsabesan P, et al. The type of loose seton for complex anal
fistula is essential to improve perianal comfor t and quality of life.
Colorectal Dis. 2016;18(6):O194– 8.
24. Soltani A, Kaiser AM. Endorectal advancement flap for cryp-
toglandular or Crohn's fistula- in- ano. Dis Colon Rectum.
2010;53(4):486– 95.
25. Stellingwerf ME, van Praag EM, Tozer PJ, Bemelman WA, Buskens
CJ. Systematic review and met a- analysis of endorectal advance-
ment flap and ligation of the intersphincteric fistula tract for
cryptoglandular and Crohn's high perianal fistulas. BJS Open.
2019;3(3):231– 41.

|
2415
MOTAMEDI ET Al .
26. Hong KD, Kang S, Kalaskar S, Wexner SD. Ligation of intersphinc-
teric fistula tract (LIFT) to treat anal fistula: systematic review and
meta- analysis. Tech Coloproc tol. 2014;18(8):685– 91.
27. van Koperen PJ, Safiruddin F, Bemelman WA, Slors JF. Outcome of
surgical treatment for fistula in ano in Crohn's disease. Br J Surg.
2009;96(6):675– 9.
28. Fichera A , Zoccali M. Guidelines for the surgical treatment of
Crohn's perianal fistulas . Inflamm Bowel Dis. 2015;21(4):753– 8.
29. Lee MJ, Parker CE, Taylor SR, Guizzet ti L, Feagan BG, Lobo AJ,
et al. Efficacy of medical therapies for fistulizing Crohn's disease:
systematic review and meta- analysis. Clin Gastroenterol Hepatol.
2018;16(12):1879– 92.
30. Strik A, Löwenberg M, Ponsioen C, Gecse K, Buskens C,
Bemelman W, et al. Higher anti- TNF serum levels are associated
with perianal fistula closure in Crohn’s disease patients. Scand J
Gastroenterol. 2019;54(4):453– 8. https://doi.org/10.1080/00365
521.2019.1600014
31. Lichtenstein GR, Yan S, Bala M, Blank M, Sands BE. Infliximab mainte-
nance treatment reduces hospitalizations, surgeries, and procedures
in fistulizing Crohn’s disease. Gastroenterology. 2005;128(4):862– 9.
32. Vogel JD, Johnson EK, Morris AM, Pa quette IM, Saclarides TJ,
Feingold DL, et al. Clinical practice guideline for the management
of anorec tal abscess, fistula- in- ano, and rectovaginal fistula. Dis
Colon Rectum. 2016;59(12):1117– 33.
How to cite this article: Motamedi MAK, Serahati S, Rajendran
L, et al. Long- term outcomes after seton placement for perianal
fistulas with and without Crohn’s disease. Colorec tal Dis.
2021;23:2407– 2415. htt ps://doi. org/10 .1111/co di .157 71