#830-Grande Galley - 02

General Surgery
SURGICAL TECHNOLOGY INTERNATIONAL Volume 30

Surgical Treatment of Rectovaginal

Fistula in Crohn’s Disease:
A Tertiary Center Experience
GIOVANNI MILITO, MD, PHD MICHELA CAMPANELLI, MD
ASSOCIATE PROFESSOR RESIDENT
DEPARTMENT OF SURGERY DEPARTMENT OF SURGERY
UNIVERSITY HOSPITAL OF TOR VERGATA “PTV” UNIVERSITY HOSPITAL OF MODENA
ROME, ITALY MODENA, ITALY

GIORGIO LISI, MD ELENA ARONADIO, MD

RESIDENT INTERN
DEPARTMENT OF SURGERY DEPARTMENT OF SURGERY
UNIVERSITY HOSPITAL OF BORGO ROMA UNIVERSITY HOSPITAL OF TOR VERGATA “PTV”
VERONA, ITALY ROME, ITALY

DARIO VENDITTI, MD MICHELE GRANDE, MD

ASSISTANT PROFESSOR ASSISTANT PROFESSOR
DEPARTMENT OF SURGERY DEPARTMENT OF SURGERY
UNIVERSITY HOSPITAL OF TOR VERGATA “PTV” UNIVERSITY HOSPITAL OF TOR VERGATA “PTV”
ROME, ITALY ROME, ITALY
ABSTRACT
ackground: Rectovaginal fistula (RVF) is a disastrous complication of Crohn’s disease (CD) that is

B exceedingly difficult to treat. It is a disabling condition that negatively impacts a woman’s quality of

life. Current treatment algorithms range from observation to medical management to the need for sur-

gical intervention. A wide variety of success rates have been reported for all management options. The

choice of surgical repair methods depends on various fistula and patient characteristics, and its published

success rates vary with initial success being around 50% rising to 80% with repeated surgery. Several surgical

and sphincter sparing approaches have been described for the management of rectovaginal fistula, aimed to

minimize the recurrence and to preserve the continence.

Materials and Methods: A retrospective study was performed for RVF repair between 2008 and 2014 in our

tertiary centre at the University Hospital of Tor Vergata, Italy. All the patients were affected by Crohn’s

disease and underwent surgery for an RVF under the same senior surgeon. All patients were prospectively

evaluated.

-1-
 #830-Grande Galley - 02

Surgical Treatment of Rectovaginal Fistula in Crohn’s Disease: A Tertiary Center Experience

MILITO/LISI/VENDITTI/CAMPANELLI/ARONADIO/GRANDE

Results: All 43 patients that underwent surgery for RVF were affected by Crohn’s disease. The median age was
43 years (range 21–53). Four different surgical approaches were performed: drainage and seton, rectal
advacenment flap (RAF), vaginal advancement flap (VAF), transperineal approach using porcine dermal
matrix (PDM), and martius flap (MF). The median time to success was six months (range 2–11). None of the
patients were lost during the 18 months of follow-up. The failure group rate was 19% in contrast with the
healing rate group that was 81%. No demographic of disease-related factors were found to influence healing.
Conclusion: The case series of this study supports the dogma that “there are no absolute rules when treating
Crohn’s fistula”. There is no gold standard technique; however, it is mandatory to minimize the recurrence
with a sphincter saving technique. Randomized trials are needed to find a standard surgical approach.

reported in 17–50% of patients with from minimally invasive laparoscopic

INTRODUCTION
Fistula-in-ano is the most common
perianal manifestation of Crohn’s disease
(CD) and was first reported by Gabriel
et al.1 in 1921, nine years before Crohn
et al.2 identified regional enteritis as a cli-
nical entity. These fistulae are classified
by their relationship to the sphincter
complex as either high (supra or extra-
sphincteric vs. low [inter or trans-sphinc-
teric]). Low fistulae that transverse the
anal sphincter are more appropriately
named anovaginal fistulae, but by con-
vention, all such fistulae are termed rec-
tovaginal fistula (RVF).3 According to
Nielsen et al.,4 fistula formation has been
Table I
CD, and the incidence of RVF is reported
in 9% affected by CD. After obstetrical
trauma, CD is the most common etiolo-
gical factor for RVF and will occur in up
to 10% of women with CD.5 Rectovagi-
nal fistulae, secondary to CD, are associa-
ted with significant morbidity and carry
an increased risk for proctectomy.6 It is a
devastating and disabling condition, it is a
source of considerable social embarras-
sment, and it has a significant negative
impact on quality of life. Furthermore,
CD associated RVF are an extremely
challenging dilemma for the clinician and
present unique, and often frustrating,
management challenges. Presently, many
surgical treatments are available, ranging
techniques to muscle transpositions and
even rectal resections. There is currently
no surgical intervention that is widely
accepted to be the first and best choice
for the treatment of RVF. The choice for
a surgical procedure may also depend on
the cause of the RVF, making it even
more difficult to define the optimal
treatment.7 According to the literature,
the most common surgical options with
a wide success rate are: drainage and
seton, rectal advancement flap (RAF),
vaginal advancement flap (VAF), transpe-
rineal approach using porcine dermal
matrix (PDM), and martius flap (MF);
however, none has been accepted as the
procedure of choice. 8 In this clinical
study, we looked at the most common
and the safest surgical approaches to find
the best standard of care.

Materials and AND

Methods
Inclusion and exclusion criteria
MATERIALS METHODS
Inclusion Criteria Exclusion Criteria

In our tertiary center at the Univer-

sity Hospital of Tor Vergata, Italy,
Crohn’s disease Ongoing perianal sepsis

patients affected by Crohn’s Disease

Rectovaginal fistula Anal incontinence

undergoing surgery for rectovaginal

Urinary incontinence

fistula under the same senior surgeon

Psichiatric disease

between 2008 and 2014 were prospecti-

Immunosuppressives, infliximab

vely evaluated and their clinical notes

were reviewed in our database. We deci-
Table II ded to used the most common surgical
procedures: DS, RAF, VAF, transperineal
approach using PDM, and MF because
Surgical Techniques
Surgical Technique Operative Treatment (n = 43) these approaches are easy to reproduce
and are well described in the literature.
Inclusion and exclusion criteria are
Seton drainage 46.6 (20)

shown in Table I. Medical treatment

Rectal advancement flap 23.2 (10)

combined with surgery is not a contrain-

Vaginal advancement flap 11.6 (5)

dication, but its advantage is still deba-

Transperineal repair with biomesh 11.6 (5)

ted. According to the literature, the

Martius flap 6.9 (3)
Values are n (%). healing rate associated with preoperative

-2-
 #830-Grande Galley - 02

General Surgery
SURGICAL TECHNOLOGY INTERNATIONAL Volume 30

infliximab or immunosuppressive the-

rapy is not significant,9,10 so we decided
to exclude from the study all patients
who had undergone infliximab therapy
six months before surgery.
Demographic information including
age, body mass index (BMI), comorbidi-
ties, smoking status, menopausal status,
and surgical history was collected. Addi-
tionally, data specific to Crohn’s disease
at the time of surgery were collected,
including years since diagnosis, disease
condition, medical therapy, previous
diversion procedure, and previous surgi-
cal treatment. All the patients were stu-
died with US endoscopy and pelvic MRI.
The results were discussed at our interdi-
sciplinary board with a gastroenterologi-
st and radiologist to identify the best
treatment.
Results
RESULTS
Surgical approach
Fifty consecutive patients with CD—
mean age 39 years (range, 20–68)—
underwent surgery for a symptomatic
RVF. The mean duration of CD was 10
years (range 0.5–24), and the most com-
mon initial site of CD was perianal
(33%). The mean duration of perianal
CD was 3.9 years (range 0.5–9). Ten
(20%) of 50 patients had active proctitis
at the presentation, so they underwent
medical treatment to correct for this and
then went to surgery. Twenty-three
patients had a complex RVF. Seventeen
patients had surgery with infliximab the-
rapy due to perianal sepsis, and a fecal
diversion was performed in all of these
patients. They were excluded from
the follow-up. Forty-three patients
underwent five procedures, which inclu-
ded seton drainage (n = 20), RAF (n =
10), VAF (n = 5), transperineal repair
with biomesh (n = 5), and bulbocaver-
nosus MF (n = 5) (Table II). None of
these patients underwent fecal diversion.
The mean duration of the seton insertion
was nine weeks (range, 4–16). The ove-
Figure 1. Healing rate with a mean follow-up of 38.6 months. DS: drainage and seton; RAF: rectal advance-
ment flap; VAF: vaginal advancement flap; TPB: transperineal repair with biomesh; MF: martius flap.
rative treatment. Healing occurred in
85% of patients who underwent a minor
surgical procedure (DS) compared with
78% of those who underwent a more
extensive operation: transperineal repair,
RAF, VAF, or MF (Fig. 1). Transperineal
repair was the operative approach with
the highest healing rate regardless of all
procedures (Table III). Of the eight
patients who did not heal, six underwent
seton insertion for six months then to
immunosuppressive therapy; they are
still under outpatient medication and a
proctocolectomy will be considered
because of the failure of medical and sur-
gical therapy. The last two patients requi-
red fecal diversion, perianal drainage,
and corticosteroids therapy.
DiscussionDISCUSSION
RVF in CD continues to be a challen-
ging problem. A multistep approach has
been recommended to treat patients
with Crohn’s-related RVF, in which
medical treatment and drainage of local
sepsis are the first initial steps to allevia-
Table III
Healing rate compared with failure rate with a
mean follow-up of 18 months
Type of Operation
te their discomfort before definitive
surgical intervention is attempted.11,12
Many reports believed that the presence
of acute perianal sepsis needed to be
separately addressed prior to any
attempts to repair RVF. This may requi-
re surgical drainage accompanied by
placement of a loose draining seton.
The results show that over three-
quarters of rectovaginal fistulae opera-
ted on by various techniques are
successfully closed. With failure occur-
ring in a quarter, however, there is
uncer tainty in the outcome, and
patients with minimal symptoms should
be counselled not to have surgery.
However, it is likely that few will heed
this advice, as the presence of the fistula
is abhorrent to most.
We decided to exclude from the fol-
low-up, patients that had undergone
fecal diversion to evaluate the feasibility
of the perianal technique alone. Howe-
ver it would be interesting to make a
comparative study with fecal diversion
in the future.
Different surgical strategies are nee-
ded according to the physical features of
the fistula, including its level and the
N° Pts Healing Failure

rall mean period of follow-up was 38.6

months (range 3–204). None of the
patients were lost during follow-up. Sur-
Drainage and seton 20 17 3

gical techniques are shown in Table II.

Rectal advancement flap 10 7 3
Vaginal advancement flap 5 3 2

Fistula healing
Transperineal approach with biomesh 5 5 0

Thirty-five (81%) of 43 patients hea-

Martius operation 3 3 0

led completely with operative treatment

alone. The average duration to healing
DS: drainage and seton; RAF: rectal advancement flap; VAF: vaginal advance-

was 6.1 months (range 1–12) after ope-

ment flap; TPB: transperineal repair with biomesh; MF: martius flap.

-3-

Surgical Treatment of Rectovaginal Fistula in Crohn’s Disease: A Tertiary Center Experience
MILITO/LISI/VENDITTI/CAMPANELLI/ARONADIO/GRANDE
state of the anal sphincter, and the situa-
tion in which the fistula actually lies at
the dentate line (as this may be the case
with an ileo-anal pouch in IBD) which
can make things much more difficult. In
such cases, a seton may be the best
option, par ticularly when there is
poorly draining sepsis. According to the
literature,13 this option may guarantee
an easy sphincter saving technique with
a safe drainage of the sepsis and a good
outcome—as is shown in the present
study with a healing rate of 85%.
Rectal advancement flaps are best sui-
ted for Crohn’s RVF when the fistula is
low, the rectum is relatively spared, and
there is no significant anal stenosis.
However, this technique is contraindica-
ted in patients with extensive ulceration
or stricturing of the anal canal and transi-
tional zone.14,15 Unfortunately, most of
the RVF in our series were high and had
local sepsis. Our healing rate was extre-
mely interesting, with a 70% success
rate, but in only 10 cases. Alternative or
additional surgical procedures include a
vaginal advancement flap—this approach
removes all diseased tissue in the anal
canal and allows “normal” rectal tissue to
be sutured to the neodentate line. Com-
pared with a rectal flap, a vaginal flap is
easier to mobilize. Sher et al.16 treated 14
patients using a transvaginal flap with
excellent results, with complete healing
in 13 patients. The authors believe that
the success of this technique is to be
attributed to the fact that they used
healthy tissue from the vagina. Unfortu-
nately, we used this technique in five
patients with a high failure rate and no
other authors have published their expe-
rience with this interesting technique.
Therefore, further conclusions are
impossible to draw.
High RVF are uncommon and are
difficult therapeutic problems. Local
repair and flap advancement techniques
have a high incidence of recurrence
with poor functional outcomes. Tran-
sperineal repair with anal sphincter
reconstruction, when indicated, and
placement of an MF result in improved
rates of repair and better functional
outcomes. A consecutive series of
patients were retrospectively reviewed
from a prospective database between
2002 and 2006 in McNevin’s study.17
They concluded that selected complex
RVF can be reliably repaired with good
functional outcomes using the MF with
anal sphincter reconstruction. Persi-
#830-Grande Galley - 02
stent or recurrent fecal incontinence
and dyspareunia are common sequela of
the underlying perineal injury and
repair. However, the number of patients
is small. Prospective randomized trials
are lacking. Therefore, it remains que-
stionable whether the addition of an MF
improves outcome after rectovaginal
fistula repair.
Recently, promising results have be
shown with acellular dermal matrix
(ADM) in RVF using a transperineal or
transvaginal approach. Gottgens et al.18
operated on 12 patients with a healing
rate of 75% and one year without
recurrence. Unfortunately, as he wrote
in his paper, there were some biases in
the study. In our case series, we treated
only five patients and no recurrence has
occured during the follow-up. Our
study included a small number of
patients and we did not record any
information about quality of life, sexua-
lity, and pain in our database. This made
it difficult to assess the morbidity asso-
ciated with this procedure; therefore,
we focused the study only on the hea-
ling and failure rate.
Conclusion
CONCLUSION
Surgical skill, adherence to principles
of hemostasis, gentle tissue handling,
and complete debridement of diseased
tissue are imperative to success. The
technique of repair will vary depending
on the location of the fistula and extent
of local and distant disease activity. STI
Authors’ Disclosures
AUTHORS’ DISCLOSURES
The authors have no conflicts of
interest to disclose.
References
REFERENCES
1. Gabriel WB. Results of an experimental
and histological investigation into seventy-
five cases of rectal fistulae. Proc R Soc Med
1921;14:156-61.
2. Crohn BB, Ginzburg L, Oppenheimer
GD. Landmark article Oct 15, 1932.
Regional ileitis. A pathological and clinical
entity. JAMA 1984;251:73-9.
3. Michael A Valente. Contemporary surgi-
cal management of rectovaginal fistula in
-4-
Crohn's disease World J Gastrointest Patho-
physiol 2014;5(4):487-95.
4. Nielsen OH, Rogler G, Hahnloser D.
Diagnosis and management of fistulizing
Crohn's disease. Nat Clin Pract Gastroen-
terol Hepatol 2009;6(2):92-106.
5. Hannaway CD, Hull TL. Current consid-
erations in the management of rectovaginal
fistula from Crohn's disease. Colorectal Dis
2008;10:747-55.
6. El-Gazzaz G, Hull T, Mignanelli E, et al.
Analysis of function and predictors of failure
in women undergoing repair of Crohn's
related rectovaginal fistula. J Gastrointest
Surg 2010;14:824-9.
7. Gottgens K, Smeets RR, Stassen LP, et
al. The disappointing quality of published
studies on operative techniques for recto-
vaginal fistulas: a blueprint for a prospective
multi-institutional study. Dis Colon Rectum
2014;57:888-98.
8. Andreani SM, Dang HH, Grondona P, et
al. Rectovaginal fistula in Crohn's disease.
Dis Colon Rectum 2007;50:2215-22.
9. Liberati A, Altman DG, Tetzlaff J, et al.
The PRISMA statement for reporting sys-
tematic reviews and meta-analyses of studies
that evaluate health care interventions:
explanation and elaboration. J Clin Epidemi-
ol 2009;62:e1-34.
10.Tanaka S, Matsuo K, Sasaki T, et al. Clin-
ical advantages of combined seton placement
and infliximab maintenance therapy for peri-
anal fistulizing Crohn's disease: when and
how were the seton drains removed?
Hepatogastroenterology 2010;57:3-7.
11.Taxonera C, Schwartz DA, García-Olmo
D. Emerging treatments for complex peri-
anal fistula in Crohn's disease. World J Gas-
troenterol 2009;15:4263-72.
12.Agaev BA, Dzhavadov EA, Abbasova GA.
Surgical Treatment of Rectovaginal Fistulae.
Khirurgiia (Mosk). 2010;44-6.
13.Tozer PJ, Balmforth D, Kayani B, et al.
Phillips. Surgical management of rectovagi-
nal fistula in a tertiary referral centre: many
techniques are needed. Colorectal Disease
2013;15:871-7.
14.McNevin MS, Lee PY, Bax TW. Martius
flap: an adjunct for repair of complex, low
rectovaginal fistula. Am J Surg 2007;
193:597-9.
15.Gosselink MP, Oom DM, Zimmerman
DD, et al. Martius flap: an adjunct for repair
of complex, low rectovaginal fistula. Am J
Surg 2009;197:833-4.
16.Sher ME, Bauer JJ, Gelernt I. Surgical
repair of rectovaginal fistulas in patients with
Crohn's disease: transvaginal approach. Dis
Colon Rectum 1991;34:641-8.
17.McNevin MS, Lee PY, Bax TW. Martius
flap: an adjunct for repair of complex, low
rectovaginal fistula. Am J Surg 2007;193:
597-9.
18.Gottgens J, Heemskerk W, van Gemert
R, et al. Rectovaginal fistula: a new tech-
nique and preliminary results using collagen
matrix biomesh. Tech Coloproctol 2014;
18:817-82.