Introduction
Benign anastomotic stenosis after colorectal resection is not an unusual event and
occurs in 3 – 30 % of patients [1 ]. The exact pathophysiology of such stricture is unknown, but use of a stapling device,
postoperative anastomotic ischemia, anastomotic leak, pelvic infection, or postoperative
radiation are believed to be contributing factors [1 ]. Traditionally, postoperative colonic stenosis has been managed surgically [1 ], but endoscopic treatment is now preferred over traditional surgery. Endoscopic
balloon dilation (EBD) is presently the simplest therapeutic approach and has a good
short-term success rate; however, in some patients, a fibrotic stricture is so inelastic
that the balloon technique is ineffective or requires more than 1 dilation. In these
patients, repeat procedures are necessary to achieve and maintain clinical success,
but can markedly reduce quality of life and add considerably to the overall cost of
treatment [1 ]
[2 ]. The complications of EBD include bleeding and perforation, which occur in up to
5 % of patients [1 ]
[2 ].
The endoscopic radial incision and cutting (RIC) technique was recently developed
to treat refractory stenosis and has been effective for treating refractory benign
esophageal stricture [3 ]
[4 ]. To the best of our knowledge, there have been few reports on the application of
RIC at sites other than the esophagus. Furthermore, few studies have evaluated the
effectiveness and safety of RIC for severe benign anastomotic stenosis after surgery
for colorectal carcinoma. Here, we evaluated the effectiveness and safety of endoscopic
RIC therapy in patients with symptomatic severe anastomotic colonic stenosis.
Case series
3 consecutive patients were treated by RIC for severe benign anastomotic stenosis
after surgery for colorectal carcinoma at Hiroshima City Asa Citizens Hospital between
May 2014 and December 2016. Severe anastomotic stenosis was defined as (1) failure
to pass a colonoscope measuring 11.7 mm in diameter (PCF Q260AZI; Olympus Medical
Systems, Tokyo, Japan) through the stricture and (2) dyschezia-related problems such
as constipation and/or abdominal distension. The stenosis was assessed radiologically
as necessary. Written informed consent was obtained from all patients before RIC was
performed.
The RIC procedure was performed as previously described ([Fig. 1 ]) [3 ]
[4 ] using the ITknife nano Electrosurgical Knife (Olympus Medical Systems). Briefly,
the blade of the ITknife nano was first inserted into the stenosed area. 4 or more
incisions were then made at the site using the knife. Next, the flaps formed by the
incisions were removed. Finally, the scar tissue was excised in an arc from the incision
along the lumen.
Fig. 1 Schema showing the radial incision and cutting (RIC) method. a 4 or more incisions are made into the stenosed site using the ITknife nano. b The flaps formed by the incisions are removed using the blade. c The scar tissue is excised in an arc from the incision along the lumen. The staples
serve as good landmarks for determining the depth of the cutting line.
The patients received 7.5 – 15 mg of pentazocine and 2 – 3 mg of midazolam before
RIC to minimize discomfort. The RIC procedure was performed by an endoscopist (SN)
who had completed approximately 500 colorectal endoscopic submucosal dissection procedures
between December 2006 and December 2016. The electrosurgical unit was set to Dry Cut
mode (Effect 2, 30 W, VIO 300; Erbe Elektromedizin GmbH, Tübingen, Germany) for making
the incisions at the stricture site and to Swift coagulation mode (Effect 2, 30 W,
VIO 300) for cutting the scar tissue. We considered treatment to have been successful
if the anastomosis could be passed by the 11.7-mm diameter colonoscope immediately
after dilation.
Patient demographic and clinical characteristics are summarized in [Table 1 ]. All 3 patients were male, had a median age of 72 years, and had undergone laparoscopically
assisted transverse colectomy or anterior resection as additional surgery after endoscopic
resection for T1 carcinoma and sigmoid colectomy for advanced colorectal carcinoma.
In all patients, the main clinical manifestation of the anastomotic stenosis before
RIC was severe dyschezia. 1 of the patients also experienced abdominal pain.
Table 1
Patient demographic and clinical characteristics.
Patient
Age, years/sex
Comorbidity
Tumor location
Treatment approach
Surgical procedure
pTNM
Postoperative complications
Adjuvant therapy
1
68 /Male
AAA
T/C
EMR → operation
LAC
pT1bN0M0
Ileus
None
2
65 /Male
DM, HTN
S/C
Operation
Sigmoid colectomy
pT4aN1M0
None
Chemotherapy
3
76 /Male
HTN
S/C
ESD → operation
LAR
pT1bN0M0
None
None
AAA, abdominal aortic aneurysm; DM, diabetes mellitus; EMR, endoscopic mucosal resection;
ESD, endoscopic submucosal dissection; HTN, hypertension; LAC, laparoscopy-assisted
colectomy; LAR, laparoscopic anterior resection; S/C, sigmoid colon; T1b, submucosal
invasion depth ≥ 1000 μm; T/C, transverse colon.
The results of RIC are shown in [Table 2 ]. 1 patient had undergone 3 EBD procedures before RIC. The median interval from surgery
to RIC was 21 months (range 9 – 29 months) and the median follow-up after RIC was
27 months (range 8 – 37 months). RIC was successful in all 3 patients and reduced
the severity of dyschezia postoperatively according to patients’ subjective verbal
reports ([Fig. 2 ] and [Fig. 3 ]; [Video 1 ]). 2 patients experienced improvement after a single RIC session and the third experienced
improvement after 6 sessions. All 8 RIC sessions were performed on an inpatient basis
with a median stay of 4 days (range 2 – 5 days). The median RIC procedure time was
22 minutes (range 15 – 25 minutes). No severe adverse events, such as perforation,
severe bleeding, high fever, or severe pain, were observed. No re-stenosis occurred
and there was no need for additional dilation in any of the patients during follow-up.
Table 2
Clinical outcomes of RIC.
Patient
Clinical manifestations
Pretreatment
Interval from surgery to RIC, months
RIC sessions (n)
Operation time, min
Major complications
Hospital stay, days
Follow-up after RIC, months
1
Abdominal pain, severe dyschezia
BD
29
6
15 – 25
None
3 – 5
34
2
Severe dyschezia
None
9
1
20
None
2
5
3
Severe dyschezia
None
13
1
15
None
2
16
Median
21
3.5
22
4
25
BD, balloon dilation; RIC, radial incision and cutting.
Fig. 2 Details of the radial incision and cutting (RIC) procedure performed in patient 2.
a, b The blade of the ITknife nano is inserted into the stenosed area. c At least 4 incisions are made into the stenosed site using the knife. d, e The flaps formed by the incisions are removed using the blade of the knife, and the
scar tissue excised in an arc from the incision along the lumen. f Treatment is considered successful because the colonoscope could easily pass through
the anastomosis.
Fig. 3 Representative images from patient 2. a Colonoscopy performed before the radial incision and cutting (RIC) procedure shows
severe anastomotic stenosis 6 months after sigmoid colectomy for colorectal carcinoma.
b A barium enema shows the stricture at the level of the anastomosis in the sigmoid
colon. c 5 months after RIC, there is a remarkable improvement in the stenosis and dyschezia.
Video 1 Details of the radial incision and cutting procedure performed in patient 2. First,
the blade of the ITknife nano is inserted into the stenosed area. Next, at least 4
incisions are made in the stenosed site using the knife. The flaps formed by the incisions
are then removed using the blade of the knife, and the scar tissue excised in an arc
from the incision along the lumen. The staples serve as good landmarks for determining
the depth of the cutting line. If the staples are exposed, it is important not to
cut the scar tissue any further. Finally, treatment is considered successful when
the colonoscope could easily pass through the anastomosis.
Discussion
In this study, the severity of dyschezia caused by severe benign anastomotic stricture
after surgery for colorectal carcinoma was dramatically reduced by RIC in all 3 patients
and no severe adverse events occurred. EBD is the simplest therapeutic option for
benign anastomotic stenosis, but it is associated with a high recurrence rate and
with refractoriness in more than 20 % of patients.
In contrast, RIC is less time-consuming and less expensive. EBD can open the lumen
of the colon or rectum, but patency is not stably maintained because EBD tears the
existing scar tissue and repeated traumatic dilations can increase mucosal scarring
[5 ]. The scar tissue, which is a by-product of repeated EBD, remains after the dilation
procedure and results in re-stenosis after wound healing. Therefore, the superiority
of the RIC method may lie in its usefulness for treating refractory stenosis by directly
removing the fibrotic tissue caused by repeated balloon dilations. This ability to
remove the fibrotic tissue directly may contribute to the effectiveness of RIC and
resolve the refractory anastomotic stricture. Of note, we used the ITknife nano to
perform all RIC procedures in this series; the insulated tip avoids inadvertent injury
to the muscle layer, and the smaller insulated tip and small disc blade have the advantage
that fixing of the tip to the scar tissue allows for stable dissection of the fibrotic
tissues.
Few studies have evaluated the efficacy and safety of RIC for severe benign anastomotic
stenosis after surgery for colorectal carcinoma. Osera et al. [6 ] reported that RIC was successful in 5 of 7 patients with such stenosis after surgery
for lower rectal carcinoma. In the 2 patients for whom RIC was unsuccessful, the procedure
was deemed a failure after 4 RIC sessions in 1 patient and 6 in the other. Therefore,
16 RIC sessions in total were performed in the 7 patients. Harada et al. [7 ] reported that 4 sessions of RIC were effective in 3 patients with rectal anastomotic
strictures refractory to conventional EBD and repeated mechanical or finger-bougie
dilation. In a case report by Kawaguti et al. [8 ], RIC was suggested to be feasible, effective, and safe for the treatment of severe
anastomotic stenosis after surgery for lower rectal carcinoma. In our patients, the
stenoses were located at various sites in the colon, and RIC provided relief with
a low complication rate, thereby avoiding the need for reoperation.
1 of our 3 patients experienced improvement after 6 RIC sessions. This patient was
admitted to our hospital twice because of ileus after laparoscopic colectomy. Postoperative
anastomotic ischemia and leakage were considered to have contributed to the admission
for ileus, and the refractoriness of the anastomotic stenosis was attributed to technical
difficulties associated with the RIC procedure.
To our knowledge, this is the first report on the use of RIC for treating severe benign
anastomotic stenosis at various sites in the colon. There were no severe adverse events
and no reoperations were needed.
There are no clear criteria with regard to the indications for RIC, and individual
patients are presently accepted after consultation between the gastroenterologists,
surgeons, and pathologists. The contraindications are as follows: the blade of an
ITknife nano cannot pass through the stenosis because it is too narrow; the scar tissue
associated with the stenosis is so firm after electrocautery that an incision using
an ITknife nano cannot be performed; the axial direction of the intestine is not the
same between the anal side and the oral side; and the stenosis is too long for the
procedure to be technically feasible.
The main difficulty encountered when using an ITknife nano to excise scar tissue from
an anastomotic site is that the scar adheres tightly to the intestinal wall. Furthermore,
it is impossible to inject material, including sodium hyaluronate, into an anastomotic
stricture site when severe fibrosis is present. Therefore, RIC needs to be performed
carefully without any injection in the treatment region.
Several precautions are necessary to prevent perforation: the oral side of the intestine
should be assessed radiologically if necessary before an incision is made using the
ITknife nano; the patient should not be sedated too deeply (so that any adverse events
that occur can be detected early); the blade of the ITknife nano should be positioned
parallel to the lumen directly over the scar tissue; and incisions should be made
carefully with the knife perpendicular to the lumen.
It is very important that the amount of scar tissue incised does not exceed the length
of the stenosed segment on the oral side because the risk of perforation increases
when the stenosis is cut deeply in 1 session. Staples serve as good landmarks for
determining the depth of the cutting line. If the staples are exposed, it is important
not to cut the scar tissue any further.
Although several studies have demonstrated the feasibility of stent insertion for
refractory anastomotic colorectal stricture, this procedure is technically difficult
because of stent migration, bleeding, or even perforation. A stent might also need
to be removed after placement because of foreign body reaction [9 ]
[10 ]. Moreover, stent insertion for refractory anastomotic colorectal stenosis is not
covered by the Japanese health insurance system.
Our findings indicate that dilation by RIC is feasible, effective, and safe for the
treatment of severe anastomotic stenosis after surgery for colorectal carcinoma. The
effectiveness of RIC combined with steroid injection is still controversial and needs
further study. RIC is performed at only a few institutions at present, so limited
clinical data are available. Large-scale, multicenter, prospective investigations
are warranted to evaluate the long-term outcomes of RIC.
