ORIGINAL ARTICLE: Clinical Endoscopy

Self-expandable metal stents for relieving malignant colorectal

obstruction: short-term safety and efficacy within 30 days of stent
procedure in 447 patients
Søren Meisner, MD,1 Ferran González-Huix, MD,2 Jo G. Vandervoort, MD,3 Paul Goldberg, MD,4
Juan A. Casellas, MD,5 Oscar Roncero, MD,6 Karl E. Grund, MD,7 Alberto Alvarez, MD,8
Jesús García-Cano, MD, PhD,9 Enrique Vázquez-Astray, MD,10 Javier Jiménez-Pérez, MD,11 The WallFlex
Colonic Registry Group
Copenhagen, Denmark

Background: The self-expandable metal stent (SEMS) can alleviate malignant colonic obstruction and avoid
emergency decompressive surgery.
Objective: To document performance, safety, and effectiveness of colorectal stents used per local standards of
practice in patients with malignant large-bowel obstruction to avoid palliative stoma surgery in incurable patients
(PAL) and facilitate bowel decompression as a bridge to surgery for curable patients (BTS).
Design: Prospective clinical cohort study.
Setting: Two global registries with 39 academic and community centers.
Patients: This study involved 447 patients with malignant colonic obstruction who received stents (255 PAL, 182 BTS,
10 no indication specified).
Intervention: Colorectal through-the-scope SEMS placement.
Main Outcome Measurements: The primary endpoint was clinical success at 30 days, defined as the patient’s
ability to maintain bowel function without adverse events related to the procedure or stent. Secondary endpoints were
procedural success, defined as successful stent placement in the correct position, symptoms of persistent or recurrent
colonic obstruction, and complications.
Results: The procedural success rate was 94.8% (439/463), and the clinical success rates were 90.5% (313/346) as
assessed on a per protocol basis and 71.6% (313/437) as assessed on an intent-to-treat basis. Complications included
15 (3.9%) perforations, 3 resulting in death, 7 (1.8%) migrations, 7 (1.8%) cases of pain, and 2 (0.5%) cases of bleeding.
Limitations: No control group. No primary endpoint analysis data for 25% of patients.
Conclusion: This largest multicenter, prospective study of colonic SEMS placement demonstrates that colonic SEMSs
are safe and highly effective for the short-term treatment of malignant colorectal obstruction, allowing most curable
patients to have 1-step resection without stoma and providing most incurable patients minimally invasive palliation
instead of surgery. The risk of complications, including perforation, was low. (Gastrointest Endosc 2011;74:876-84.)

Abbreviations: ASA, American Society of Anesthesiologists Physical Status 0016-5107/$36.00

Classification System; BTS, bridge to surgery (curable patients); IRB, doi:10.1016/j.gie.2011.06.019
institutional review board; PAL, palliation (incurable patients); SEMS,
self-expandable metal stent. Received September 17, 2010. Accepted June 15, 2011.

DISCLOSURE: S. Meisner is a consultant for Boston Scientific and Colo-
plast Denmark, and F. González-Huix is a consultant for Shering
Plough. The WallFlex-eR colonic international and Spanish registries
were sponsored by Boston Scientific Corporation. No other financial
relationships relevant to this publication were disclosed
Copyright © 2011 by the American Society for Gastrointestinal Endoscopy
Current affiliations: Bispebjerg Hospital (1), Copenhagen, Denmark; Hospi-
tal Doctor Josep Trueta (2), Girona, Spain; Onze Lieve Vrouw Ziekenhuis
(3), Aalst, Belgium; Groote Schuur Hospital (4), Cape Town, South Africa;
Hospital General Universitario de Alicante (5), Alicante, Spain; Hospital La
Mancha
(footnotes continued on last page of article)

876 GASTROINTESTINAL ENDOSCOPY Volume 74, No. 4 : 2011 www.giejournal.org

Meisner et al
Colorectal cancer is among the most common malig-
nant diseases, with a global incidence of 1 million new
cases per year, ranking fourth in frequency in men and
third in women and with mortality of about 529,000
deaths.1 Obstruction has been reported in 7% to 29% of
patients with colorectal cancer.2 Patients with malignant
large-bowel obstruction tend to have advanced disease
and be poor surgical candidates. Surgical colostomy is
effective, but mortality can be high. In addition, a colos-
tomy diminishes quality of life, and its management poses
difficulties for elderly and frail patients.3 Patients present
either with acute total obstruction or subacute bowel ob-
struction. Severity of symptoms may vary. Conventional
therapies for relieving malignant colorectal obstruction in-
clude surgical resection (potentially curative) or palliative
colostomy. Resection is ideally carried out as a single-stage
procedure, with anastomosis to restore bowel continuity, but
multistage procedures may also be undertaken, first a resec-
tion and stoma formation, followed by restoration of conti-
nuity. However, a significant proportion of patients (up to
50%) receiving a staged procedure never undergo reversal of
the colostomy.4 In the emergency setting, surgery carries a
high mortality (15%-20%) and high morbidity (45%-50%) risk,
with increased prevalence of intensive care stay, infections,
and complications related to stomas.1,2
Endoscopic placement of a self-expandable metal stent
(SEMS) to relieve colonic obstruction was introduced more
than 10 years ago. It may be used for palliation to elimi-
nate the need for stomas in incurable patients or to
“bridge” patients to elective single-stage surgery, most
often without a stoma, thereby significantly reducing the
mortality and morbidity.5,6
Although effectiveness of the use of colonic SEMSs in a
palliative or bridge-to-surgery indication is broadly recog-
nized, some have questioned the safety of colonic stent-
ing, particularly as it pertains to colonic perforation.7,8 This
article reports on the largest prospective series of treat-
ment with colonic SEMSs per local standard of practice.
The focus of analysis is on broadness of applicability of the
method and patient safety.
PATIENTS AND METHODS
Patient population and study centers
Patients with acute or symptomatic colonic obstruction
were enrolled in two prospective, single-arm, multicenter
studies, the WallFlex-eR Colonic International Registry and
the WallFlex-eR Colonic Spanish Registry. Both were iden-
tically designed and conducted by using the same registry
platform over the course of 2 years and 9 months (inter-
national registry) and 1 year and 11 months (Spanish
registry). Institutional review board (IRB) approval was
required and obtained for the enrollment of 113 patients,
including those in the coordinating centers. Per guidance
from their local IRB departments, some centers did not
require IRB approval (300 patients), citing the registry
www.giejournal.org
SEMSs for malignant colorectal obstruction
Take-home Message
● Colonic self-expandable metal stents provided a safe and
highly successful short-term treatment of malignant
colorectal obstruction.
● Most of the curable patients can have 1-step resections
without stomas after decompression by self-expandable
metal stents. Risk of complications, including perforation,
is low.
nature of the studies, the use of a noninvestigational (ap-
proved) device, and treatment that is standard of care for
each patient included in the registries. IRB approval infor-
mation is unknown for 34 patients.
This study was conducted as a registry in which data
were collected for procedures performed per standard of
practice. Documentation of patient consent to the proce-
dure as well as to the participation in the study, or IRB
waiver of the need of such consenting to study participa-
tion, is available for 263 patients (group 1). Stratified anal-
ysis of procedural success, clinical success, and overall
complication rates was performed to compare group 1 to
the remaining 184 patients (group 2). When using the
Fisher exact test, there was no statistically significant dif-
ference in the rate of procedural success, clinical success,
and complications between the two groups.
The following data reflect the outcomes for the total
group of patients enrolled. Each patient was treated with
an uncovered WallFlex Enteral Colonic Stent (Boston Sci-
entific Corporation, Natick, Mass) according to the product
directions for use and per the usual treatment practice at
the enrolling center. Included were patients with colorec-
tal strictures secondary to malignant neoplasms, requiring
either palliative or bridge-to-surgery treatment with a
stent. The colonic obstruction was evaluated by imaging
and the presence of symptoms. Exclusion criteria were
placement of a previous colonic stent, enteral ischemia,
suspected or impending perforation, intra-abdominal
abscess/perforation, contraindication to endoscopic treat-
ment, and any use of the stent other than those specifically
outlined under indications of use. Participating centers
(39) consisted of 22 academic and 17 community hospitals
in 13 countries. Stent procedures were performed by en-
doscopists (87.5%) or surgeons (12.5%).
Patient assessments
Data were collected at stent placement and during
follow-up visits. Incurable patients undergoing palliative
treatment (PAL) were followed until death, and curable
patients (BTS) undergoing bridge to surgery were fol-
lowed until surgery or for 12 months, whichever came
first. All follow-up was done per each center’s usual med-
ical practices.
The focus of the present report is safety of the use of
colorectal stenting in both PAL and BTS patients up to 30
Volume 74, No. 4 : 2011 GASTROINTESTINAL ENDOSCOPY 877
SEMSs for malignant colorectal obstruction
days after stent placement. These two populations are
seemingly different in terms of long-term management but
can be appropriately grouped for the short-term analysis
of SEMS safety and efficacy because they share a nearly
identical clinical experience during the first few weeks
after stenting.
SEMS placement
Stents were placed by the aid of fluoroscopy and under
direct visualization with an endoscope. Sedation was used
during the majority of procedures or 75.7% (333) of cases,
general anesthesia was used in 16 patients, and no
sedation was used in 91 patients. A 0.035 inch guidewire
was passed across the site of the stricture. The size of
the stricture was measured fluoroscopically (58.0%),
endoscopically (14.8%), or by both methods (20.5%),
and the length and number of stents needed to cross the
stricture was determined. In a small group of patients
(14), the stricture was dilated before stent placement by
using either a balloon or bougie dilator.
In accordance with the instructions for use, the Wall-
Flex Enteral Colonic Stent delivery system was passed
over the guidewire, through the endoscope working
channel, and to the site of the stricture until the post-
deployment marker band was at the outermost point of
the proximal stricture end. The exterior tube marker
band was used to position the stent at the distal end of
the stricture. To begin stent deployment, we immobi-
lized the hub handle in one hand and grasped the
exterior tube handle with the other hand, after which
the exterior tube handle was gently slid back along the
stainless steel tube toward the hub handle. Stent posi-
tion was then assessed and the stent repositioned if
necessary. To complete stent deployment, the hub han-
dle was immobilized with one hand and the exterior
tube handle was grasped with the other hand and was
gently slid along the stainless steel tube toward the hub
handle until full deployment. After the stent was cor-
rectly positioned and fully deployed, we closed the
delivery system by pushing the exterior tube handle
forward and removing the delivery system (Fig. 1).
Outcomes
Procedural success. Procedural success was defined
as successful endoscopic placement of the stent in the
correct position.
Clinical success. The primary effectiveness measure
was defined as having excellent, good, or fair passage of
stool at 30 days after the procedure, no device-related or
procedure-related complications, and no emergency sur-
gery because of complications in the BTS cohort.
Surgery outcomes in BTS cohort. The incidence of
elective and emergency surgeries and time to surgery in
BTS patients were recorded.
Safety profiles. Complications were reported cumula-
tively up to 6 hours and 30 days after the procedures.
878 GASTROINTESTINAL ENDOSCOPY Volume 74, No. 4 : 2011
Meisner et al
Figure 1. Fully deployed WallFlex Enteral Colonic Stent.
Statistical analysis
Summary statistics were computed for either the en-
rolled or the treated population, depending on the mea-
sure being summarized. For categorical measures at each
visit, summary statistics consisted of frequency and per-
cent of responses in each category. The denominator of a
percentage is the number of patients with non-missing
values, based on available follow-up data (evaluable pa-
tients). For continuous measures at each visit, summary
statistics included sample size, mean, median, standard
deviation, minimum, and maximum. The sample size is
the number of non-missing values, based on evaluable
patients, unless noted otherwise.
RESULTS
Baseline characteristics
We enrolled 463 patients for a palliative or bridge-
to-surgery indication. The WallFlex-eR Colonic Interna-
tional Registry enrolled 216 patients (141 PAL, 61 BTS,
14 with no specified indication), and the Spanish regis-
try enrolled 247 patients (116 PAL, 121 BTS, 10 with no
specified indication). A stent could not be placed in 16
patients, mainly because it was impossible to pass a
guidewire through the tumor stricture (8) or to engage/
visualize the tumor site (5) (Fig. 2). The remaining 447
patients who received a WallFlex Enteral Colonic Stent
were 255 (57%) PAL, 182 (41%) BTS, and 10 patients
with no indication specified at enrollment.
Baseline demographics were reported on all enrolled
patients (463). There were 277 (59.8%) men, and the
mean (SD) patient age was 72.1 ⫾ 12.4 years. The
average (SD) body mass index was 25.2 ⫾ 3.9, and
patients were reported as having American Society of
Anesthesiologists Physical Status Classification System
www.giejournal.org
Meisner et al
Figure 2. Patient analysis flowchart. ITT, intent to treat; PAL, palliation (incurable patients); BTS, bridge to surgery (curable patients).
(ASA) scores of I in 61 cases (13.2%), II in 214 cases
(46.2%), III in 136 cases (29.4%), IV in 32 cases (6.9%),
and V in 2 cases (0.4%). Symptoms of colonic obstruc-
tion included nausea in 165 patients (38.3%), vomiting
in 129 patients (30%), constipation in 339 patients
(79.2%), diarrhea in 89 patients (20.7%), abdominal
pain/cramps in 329 patients (76.3%), and bloating in 255
patients (59.2%). At baseline, previous treatments for
colorectal cancer were recorded in 41 patients (9.4%),
with the majority (27) having undergone chemotherapy,
surgical resection (17), and/or radiation (6). Eighty-
eight patients (20.1%) had a history of lower abdominal
and/or pelvic surgery.
Tumor characteristics are described in Table 1. The ma-
jority of patients (49.8%) had localized colorectal cancer with-
out metastases. An intrinsic tumor was the source of the
www.giejournal.org
SEMSs for malignant colorectal obstruction
malignant colorectal obstruction in 93.1% of patients and
resulted from extrinsic compression in only 6.9% of patients.
The incidence of left-sided (rectosigmoid junction, sigmoid
and descending colon, splenic flexure) colonic strictures was
significantly higher (77.8%) than that of rectal tumors (15.8%)
and proximal colon (transverse colon, hepatic flexure, and
ascending colon) tumors (7.8%). The degree of intestinal
tortuosity was medium in 195 patients (44.3%), high in 103
patients (23.4%), and low in 142 patients (32.3%), of a total of
440 patients having tortuosity.
The stent size most commonly used was 9 cm in length
with 25/30 mm body/flare diameter (197 patients) and 12
cm in length with 25/30 mm body/flare diameter (91
patients). The majority of patients (421) required only 1
stent for bridging their stenoses; a second stent was
needed in 18 patients; 1 patient received 3 stents.
Volume 74, No. 4 : 2011 GASTROINTESTINAL ENDOSCOPY 879
SEMSs for malignant colorectal obstruction Meisner et al

TABLE 1. Baseline tumor characteristics and diagnostic TABLE 2. Procedural success and SEMS placement data
information in 447 treated patients
Procedural success* % (No.)
Characteristic %
Yes 94.8 (439/463)
Underlying disease*
No
Only local cancer 49.8
Poor position, too oral 0.6 (3/463)
Liver metastasis 57.2
Poor position, too anal 0.4 (2/463)
Lung metastasis 21.7
Inability of stent to deploy 0.2 (1/463)
Peritoneal carcinosis 28.1 enough through the stricture

Multiple metastasis 38.0 Perforation of the peritoneum 0.2 (1/463)

Type of tumor Colonic perforation 0.2 (1/463)

Intrinsic 93.1 No stent could be placed 3.5 (16/463)

Extrinsic 6.9 SEMS placement described by VAS†

Localization of stenosis/tumor† Correct stent placing 93.0 (409/440)

Rectum 15.8 Poor expansion 5.5 (24/440)

Left-sided colon 77.8 Impacted ends 2.5 (11/440)

Proximal colon 7.8 Proximal stricture 0.0 (0/440)

Diagnostic procedure to evaluate the obstruction‡ Migration 1.1 (5/440)

Radiograph thorax 82.5 Incomplete stenting 0.7 (3/440)

Abdominal US 40.5 Long stricture 1.6 (7/440)

CT scan 81.1 Fecal obstruction 0.2 (1/440)

Plain radiograph abdomen 83.4 SEMS, Self-expandable metal stent; VAS, visual analogue score.
*Intention-to-treat cohort, n ⫽ 463.
Large-bowel enema 22.6 †Treated cohort, n ⫽ 440.

*A patient may have more than one underlying disease.

†A patient may have more than one tumor location.
‡A patient may have more than one diagnostic procedure.
Procedural success
Procedural success was assessed on all intent-to-treat
patients (463). Successful endoscopic procedures were
reported in 94.8% (439/463) of patients (Table 2). Proce-
dural failures (24/463) were reported in 16 patients who
did not receive a stent as described earlier and 8 stented
patients (4 BTS, 4 PAL) because of poor position (5),
inability of the stent to deploy enough through the stric-
ture (1), and perforation (2).
Procedural success data were stratified by ASA scores
by analyzing patients with ASA scores of 3 or higher
(sicker patients) versus patients with ASA scores of 1 or 2
(healthier patients). A Fisher exact test showed no signif-
icant difference in procedural success outcomes between
the two groups (P ⫽ 1.000); specifically, rates were 98.1%
(158/161) in the sicker patients group versus 98.1% (256/
261) in the healthier patients group.
Fourteen (3.2%) patients underwent before-placement
dilation with a balloon (7), bougie (5), and both methods (2).
Additional SEMS placement data are described in Table 2.
Clinical success
As illustrated in Figure 2, 101 patients (25%) were not
eligible for clinical success evaluation. This was because 73
patients were missing 30-day visit or procedure data and
therefore were considered lost to follow-up (49 PAL, 14 BTS,
10 indication not specified) and 9 patients had unknown
passage of stool (2 BTS, 7 PAL). The 2 BTS patients with
unknown passage of stool at 30 days were reported to have
undergone late surgery; among the 7 PAL patients, 3 were
lost to follow-up after the 30-day visit (2) or the 6-month visit
(1), 1 patient had a successful operation (lavage and suture of
a perforation at day 20), 1 died before the 30-day visit (death
was colorectal cancer related), 1 patient had a successful
Hartmanns procedure performed because of stent dysfunc-
tion (no decompression because of shortening of the stent)
and exited the study, and 1 had an attempted but failed
curative resection and died because of a complication from
anastomotic leakage. The remaining 19 patients were not
eligible for primary outcome measure analysis for the follow-
ing reasons: BTS patients no longer being surgical candi-
dates because of progression of their colorectal disease
(9), procedural failure (8), and BTS patients with surgery
after 30 days (2).

880 GASTROINTESTINAL ENDOSCOPY Volume 74, No. 4 : 2011 www.giejournal.org

Meisner et al SEMSs for malignant colorectal obstruction

TABLE 3. Per protocol analysis of clinical success (30-day efficacy)

All patients (n ⴝ 437) % (no.) BTS patients (n ⴝ 182) % (no.) PAL patients (n ⴝ 255) % (no.)
Ability to pass stool*

Excellent/good 83.4 (311/373) 90.4 (151/167) 77.7 (160/206)

Fair 7.5 (28/373) 3.0 (5/167) 11.2 (23/206)

Poor/very poor 6.2 (23/373) 4.2 (7/167) 7.8 (16/206)

Unknown 2.9 (11/373) 2.4 (4/167) 3.4 (7/206)

Lost follow-up/death 14.6 (64/437) 8.2 (15/182) 19.2 (49/255)

Complication/adverse event†

Yes 9.8 (43/437) 7.1 (13/182) 11.8 (30/255)

No 90.2 (394/437) 92.9 (169/182) 88.2 (225/255)

Stoma done N/A 6.6 (12/182) N/A

Clinical success 90.5 (313/346) 94.0 (141/150) 87.8 (172/196)

BTS, Bridge to surgery; PAL, palliative; N/A, not applicable.
*At 30 days.
†From stent placement up to 30 days after procedure.

The rate of clinical success can be assessed in two

different ways, namely on an intent-to-treat basis (n ⫽ 437
enrolled patients with an indication of PAL or BTS), or on
a per protocol basis (n ⫽ 346 patients with available data
for primary endpoint analysis). The intent-to-treat rate of
clinical success was 71.6% (313/437), which includes the
16 patients who did not receive a stent.
The per protocol clinical success rate (Table 3), as-
sessed on the subgroup of patients who were eligible for
primary endpoint analysis according to its definition per
protocol was 90.5% (313/346). In the BTS group, per
protocol clinical success was achieved in 94.0% (141/150)
of cases. Clinical failures (9 BTS patients) resulted from a
complication related to the device or procedure in 4 pa-
tients, poor passage of stool in addition to a complication
in 3 patients, and poor passage of stool in 2 patients. In the
PAL group, per protocol clinical success was achieved in
87.8% (172/196) of cases. Clinical failures (24 PAL pa-
tients) were because of complications related to the device
or procedure in 10 patients, poor bowel function in 9
patients, and poor passage of stool in addition to a com-
plication in 5 patients. Figure 3. Tissue specimen from a bridge-to-surgery patient. A, After a
We found no significant difference in the clinical suc- laparoscopic resection and primary anastomosis. B, After dissection
cess outcome within 30 days between sicker patients (ASA showing colonic self-expandable metal stent.
score 3 or more) and healthier patients (ASA scores 1 and
2). Rates of per protocol clinical success were 87.4% (104/
119) versus 92.1% (198/215) tested with a Pearson chi- were lost to follow-up or died, and 10 did not undergo
square test (P ⫽ .1624). surgery. Figure 3 (A, B) shows a specimen after a BTS
procedure with laparoscopic resection and primary anas-
Surgery outcomes in BTS cohort tomosis. The overall mean (SD) time from stent placement
In the BTS group (n ⫽ 182), 150 patients (89.8%) un- to surgery was 16.2 ⫾ 12.5 days. A stoma was created in 9
derwent elective surgery, 7 had emergency surgery, 15 of the 150 patients (6%) because of surgical difficulties in

www.giejournal.org Volume 74, No. 4 : 2011 GASTROINTESTINAL ENDOSCOPY 881

SEMSs for malignant colorectal obstruction Meisner et al

was a result of fecal impaction in 6 patients, mucosal/

TABLE 4. Safety profile within 6 hours of procedure bowel impaction into the oral part of the stent in 2 pa-
and within 30 days tients, and a second colonic obstruction in 1 patient with a
patent stent.
% (No.)
Overall complication rates were stratified by ASA scores
Deaths 8.9 (40/447) and tested by a Pearson chi-square test. Sicker patients
Colorectal cancer related 47.5 (19/40) (ASA scores 3 or higher) had a cumulative complication
rate of 11.5% (15/131) versus 11.4% (27/237) in healthier
Not colorectal cancer related 32.5 (13/40)
patients (ASA scores 1 and 2), with no significant differ-
Other 20.0 (8/40) ence in safety outcome (P ⫽ .9866).
Adverse events within 6 h (procedural) A stratified analysis on patients with high intestinal
tortuosity versus patients with medium or low tortuosity
Perforation 1.3 (6/447)
showed evidence of difference between the two groups in
Bleeding 0.4 (2/447) one of the complication categories, reobstruction because
Stent migration 0.2 (1/447)
of fecal impaction. Patients with high tortuosity had a 4.4%
(4/90) rate of fecal impaction compared with the rest of
Pain 1.1 (5/447) patients, who had a rate of 0.7% (2/292), which was
Persistent obstruction 0.7 (3/447) statistically significant (P ⫽ .0294) based on a 2-sided
Fisher exact test.
Adverse events within 30 d (cumulative)
Mortality. Forty (8.9%) patients died (Table 4) during
Reobstruction due to fecal impaction 1.6 (6/382) the 30-day follow-up period, 3 of whom had perforations
Reobstruction due to mucosal/bowel impaction 0.5 (2/382) at 6, 24, and 34 days after stent placement, which could be
into stent oral related to the stent. The remaining 37 deaths were attrib-
uted to the colorectal cancer (16) or other causes unrelated
Stent patent but second colonic obstruction 0.3 (1/382)
to the colorectal cancer (21), including cardiovascular or
Bleeding 0.5 (2/382) pulmonary insufficiency, aspiration, infarction, and pro-
Perforation 3.9 (15/382) gression of other carcinomas.
Stent migration 1.8 (7/382)
DISCUSSION
Pain 1.8 (7/382)

Persistent obstruction 0.8 (3/382)

performing primary anastomosis in 2 patients with rectal
tumors and for unspecified reasons in 7 patients.
Safety profile
Safety is reported in Table 4.
Within 6 hours (procedural). The majority of pa-
tients who received a stent did not experience any com-
plications up to 6 hours after the endoscopic procedure
(96.2%). Six patients had a perforation at the time of stent
placement, of which 4 had no dilation performed before
the procedure. Additional procedural complications in-
cluded 2 cases of bleeding, 1 stent migration (patient
received a second stent and recovered), 5 cases of pain,
and 3 persistent obstructions.
Within 30 days. The authors chose to report cumula-
tive complication rates in a more conservative way,
namely by calculating rates from a denominator of 382
patients with a 30-day follow-up visit or a safety event
leading to discontinuation of follow-up before 30 days,
instead of a denominator of 447 (all treated patients).
Postprocedural complications included 9 perforations,
6 stent migrations, and 2 reports of pain. Reobstruction
The management of symptomatic colonic obstruction,
especially in acute form, is still a challenge. The various
surgical options continue to pose a high mortality and
morbidity risk to the patient. The single procedure that
makes the biggest difference is the placement of a colonic
stent, either in a palliative setting or as a bridge to surgery.
Previously, 3 systematic reviews by Khot et al,9 Sebas-
tian et al,10 and Watt et al11 evaluated and presented data
on the efficacy and safety of colorectal stents in 598, 1198,
and 1785 patients, respectively. They reported rates of
technical success of 92%, 93.2%, and 96.2% (median) and
clinical success of 88%, 88.6%, and 92% (median). Safety
data indicated 1% and 0.58% mortality (Watt et al did not
report stent-related mortality); 4%, 3.76%, and 4.5% (me-
dian) perforation rates; 10%, 11.81%, and 11% (median)
stent migration rates, respectively. A prospective study by
Repici et al12 evaluated the effectiveness and safety of the
WallFlex Colonic Stent for the treatment of malignant co-
lonic obstruction in 42 patients (23 PAL, 19 BTS). Reported
were high rates of technical success (93%), initial clinical
success (95%), and low complication rates (1 migration
[2.4 %] and 1 perforation [2.4%]). All patients with operable
tumors were successfully bridged to elective 1-stage
surgery.
Similarly, our study shows technical success of 94.8%,
per protocol clinical success of 90.5%, and migration rate

882 GASTROINTESTINAL ENDOSCOPY Volume 74, No. 4 : 2011 www.giejournal.org

Meisner et al
of 1.8%. Perforations (15) were reported by 8 tertiary-care
referral centers (academic) and 6 community hospitals.
There is no evidence that the rates of complications are
different between non-referral and tertiary-care centers.
The authors recognize the following limitations to this
study: non-randomized, single-arm study design, drop-out
rate of 25% for primary endpoint analysis, and lack of
ethics committee approval of the protocol at some centers.
During the last several years, discussions on safety of
colorectal SEMS placement have often focused on the
incidence of perforation. In 2006, van Hooft et al13 re-
ported that the Dutch Stent-in I study, a prospective, ran-
domized, controlled trial of the WallFlex Colonic Stent,
was prematurely terminated because of a higher than
expected incidence of late perforations in the stent treat-
ment arm. Four perforations (4/11) were cited, complica-
tions in 3 of which resulted in death. In a subsequent
publication, van Hooft et al14 indicated that among the 9
patients with successful stent placement, 2 had perfora-
tions at day 12, and 4 experienced late-onset perforations.
The majority of perforations were in patients who under-
went chemotherapy. In a more recent publication, Baron8
commented on some limitations of the Dutch Stent-in I
study. Indeed, Baron noted that only patients with a left-
sided obstruction and those with impending but incom-
plete colonic obstruction were enrolled. Whether the de-
layed perforations were related to the specific stent type
used in the study is unknown; however, the high rate of
delayed complications with the same stent has not been
duplicated by other groups.
Furthermore, our study, which provides the largest
prospective series to date, did not confirm the findings
of the Dutch Stent-in I study. Our clinical experience
with the WallFlex Enteral Colonic Stent resulted in a
3.9% overall perforation rate up to 30 days of stent
placement, which was comparable to the earlier-cited
perforation rates ranging from 2.4% to 4.5%.9-12 In ad-
dition, the rate of procedure-related perforations was
1.3% and seemed to be highly related to dilation before
stent placement and unrelated to the stent itself. In a
post hoc analysis, our data suggest that if a patient had
dilation before the stent procedure, the odds of having
a perforation were 9.41 times higher than without dila-
tion, with a significant P value (.0017). This finding is in
line with the 3 systematic reviews by Khot et al,9 Sebas-
tian et al,10 and Watt et al.11
More information regarding the long-term outcome of
colorectal stenting is still needed, especially in the group
of curable BTS patients. There is no evidence that SEMSs
in the bridge-to-surgery setting worsen the oncological
long-term prognosis. Questions about preoperative stent
placement remain, however. In one retrospective study by
Kim et al,15 35 patients with left-sided obstruction who
underwent surgical resection after SEMS placement were
matched to 350 patients who underwent elective surgery
for nonobstructive left-sided colon cancer. There were no
www.giejournal.org
SEMSs for malignant colorectal obstruction
significant differences in clinicopathologic variables be-
tween the two groups. Patients in the SEMS group had a
significantly lower 5-year overall survival rate (38.4% vs
65.6%, respectively) and 5-year disease-free survival rate
(48.3% vs 75.5%, respectively) than did the elective sur-
gery patients. However, it is not known whether these
outcomes are related to the underlying obstruction rather
than to the stenting itself. In a retrospective comparison of
40 emergency operations and 44 expandable metal stent
insertions as bridge to surgery, Saida et al16 could not
show any differences in the long-term prognosis of plac-
ing a stent in preoperative patients as well.
Randomized, controlled trials would be needed to re-
veal the true findings—several ongoing randomized clin-
ical trials are recruiting patients but at a very low and slow
inclusion rate.
The evidence from the published reviews and this study
indicates that SEMS placement is both a safe and effective
technique for relieving malignant colorectal obstructions
and should be considered in every case, whether for
palliation or as a bridge to surgery. For surgically noncur-
able patients, placement of SEMSs to relieve colonic ob-
struction should replace surgery whenever possible be-
cause it avoids the additional burden of surgery with a
stoma. The conversion of a surgical emergency procedure
to an elective procedure in the bridge-to-surgery group is
favorable because of the reduced mortality and morbidity
rate of SEMSs as well as shortened hospital and intensive
care unit stay and the reduced need for both temporary
and permanent stomas.
CONCLUSION
In a prospective study of through-the-scope SEMSs
used per local standard-of-practice, the WallFlex Colonic
SEMS provided a safe and highly successful short-term
treatment of malignant colorectal obstruction, allowing
most of the curable patients to have 1-step resection with-
out stomas after decompression by SEMSs and providing
most of the incurable patients minimally invasive pallia-
tion instead of surgery. The risk of perforation was very
low.
ACKNOWLEDGMENTS
The authors gratefully acknowledge the help of Bos-
ton Scientific Corporation employees Terry Liao, PhD,
John Evans, PhD, Amy Hu, and Brian Johnson for sta-
tistical analyses and Montserrat Agusti, Eduardo Sessa,
and Erika Alfieri for planning and execution of the
registries.
REFERENCES
1. Parkin DM, Bray F, Ferlay J, et al. Global Cancer Statistics, 2002. CA Can-
cer J Clin 2005;55:74-108.
Volume 74, No. 4 : 2011 GASTROINTESTINAL ENDOSCOPY 883
SEMSs for malignant colorectal obstruction
2. Deans GT, Krukowski ZH, Irwin ST. Malignant obstructions of the left
colon. Br J Surg 1994;81:1270-6.
3. Neugent KP, Daniels P, Stewart B, et al. Quality of life in stoma patients.
Dis Colon Rectum 1999;42:1569-74.
4. Wong R, Rappaport W, Witze D, et al. Factors influencing the safety of
colostomy closure in the elderly. J Surg Res 1994;57:289-92.
5. Leitman IM, Sullivan JD, Brams D, et al. Multivariate analysis of the mor-
bidity and mortality from initial surgical management of obstructing
carcinoma of the colon. Surg Gynaecol Obstet 1992;174:513-8.
6. Mulcahy HE, Skelly MM, Hussain A, et al. Long-term outcome following
curative surgery for malignant large bowel obstruction. Br J Surg 1996;
83:707-10.
7. Dohmoto M. New method: endoscopic implantation of rectal stent in
palliative treatment of malignant stenosis. Endosc Dig 1991;3:1507-
12.
8. Baron TH. Colonic stenting: a palliative measure only or a bridge to sur-
gery? Endoscopy 2010;42:163-8.
9. Khot UP, Lang AW, Murali K, et al. Systematic review of the efficacy and
safety of colorectal stents. Br J Surg 2002;89:1096-102.
10. Sebastian S, Johnston S, Geoghegan T, et al. Pooled analysis of the effi-
cacy and safety of self-expanding metal stenting in malignant colorectal
obstruction. Am J Gastroenterol 2004;99:2051-7.
11. Watt AM, Faragher IG, Griffin TT, et al. Self-expanding metallic stents for
relieving malignant colorectal obstruction: a systematic review [re-
view]. Ann Surg 2007;246:24-30.
12. Repici A, De Caro G, Luigiano C, et al. WallFlex Colonic stent placement
for management of malignant colonic obstruction: a prospective study
at two centers. Gastrointest Endosc 2008;67:77-84.
13. van Hooft JE, Fockens P, Marinelli AW, et al; Dutch Stent-in I study group.
Premature closure of the Dutch Stent-in I study. Lancet 2006;368:
1573-4.
14. van Hooft JE, Fockens P, Marinelli AW, et al; Dutch Colorectal Stent
Group. Early closure of a multicenter randomized clinical trial of endo-
scopic stenting versus surgery for stage IV left-sided colorectal cancer.
Endoscopy 2008;40:184-91.
15. Kim JS, Hur H, Min BS, et al. Oncologic outcomes of self-expanding me-
tallic stent insertion as a bridge to surgery in the management of left-
sided colon cancer obstruction: comparison with nonobstructing elec-
tive surgery. World J Surg 2009;33:1281-6.
16. Saida Y, Sumiyama Y, Nagao J, et al. Long-term prognosis of preopera-
tive “bridge to surgery” expandable metallic stent insertion for obstruc-
tive colorectal cancer: comparison with emergency operation. Dis Co-
lon Rectum 2003;46(suppl):S44-9.
Centro (6), Alcazar de San Juan, Spain; University Hospital Tuebingen, De-
partment of General, Visceral, and Transplantation Surgery (7), Tuebin-
gen, Germany; Hospital Universitario de Salamanca (8), Salamanca,
Spain; Hospital Virgen de La Luz (9), Cuenca, Spain; Complejo Hospital-
884 GASTROINTESTINAL ENDOSCOPY Volume 74, No. 4 : 2011
Meisner et al
ario Pontevedra (10), Pontevedra, Spain; Hospital de Navarra (11), Pam-
plona, Spain.
Presented at Digestive Disease Week, May 17-22, 2008, San Diego,
California (Gastrointest Endosc 2008;67:AB304-AB305 and AB307).
Presented at United European Gastroenterology Week, October 18-22,
2008, Vienna, Austria (Endoscopy 2008;40[suppl 1]:A6-A7).
Reprint requests: Søren Meisner, MD, Bispebjerg Hospital, Bispebjerg Bakke
23, entr. 7B, DK-2400 Copenhagen NV, Denmark.
The WallFlex Colonic Registry Group: Søren Meisner, MD, Bispebjerg Hospi-
tal, Copenhagen, Denmark, Ferran González-Huix, MD, Hospital Doctor Jo-
sep Trueta, Girona, Spain, Jo G. Vandervoort, MD, Onze Lieve Vrouw Zieken-
huis, Aalst, Belgium, Paul Goldberg, MD, Groote Schuur Hospital, Cape
Town, South Africa, Juan A. Casellas, MD, Hospital General Universitario de
Alicante, Alicante, Spain, Oscar Roncero, MD, Hospital La Mancha Centro,
Alcazar de San Juan, Spain, Karl E. Grund, MD, University Hospital Tuebin-
gen, Department of General, Visceral, and Transplantation Surgery, Tuebin-
gen, Germany, Alberto Alvarez, MD, Hospital Universitario de Salamanca,
Salamanca, Spain, Jesus García-Cano, MD, PhD, Hospital Virgen de La Luz,
Cuenca, Spain, Enrique Vázquez-Astray, Complejo Hospitalario Pontevedra,
Pontevedra, Spain, Javier Jiménez-Pérez, MD, Hospital de Navarra, Pam-
plona, Spain, Javier Barcenilla, MD, Hospital Rio Carrion, Palencia, Spain,
Alessandro Repici, MD, Istituto Clinico Humanitas, Rozzano (Mi), Italy, Dimi-
trios Xinopoulos, MD, Saint Savas Hospital, Athens, Greece, Leopoldo Lopez-
Roses, MD, Complejo Hospitalario Xeral-Calde, Lugo, Spain, Ovidio Belda,
MD, Hospital Virgen del Rocio, Sevilla, Spain, Marc Giovannini, MD, Institut
Paoli–Calmettes, Marseille, France, Jorge Espinós, MD, Fundacio Mutua de
Terrassa, Terrassa, Spain, Ivan Garcia-Tercero, MD, Hospital Clinico San Ceci-
lio, Granada, Spain, Guillermo Alcain, MD, Hospital Virgen de La Victoria,
Malaga, Spain, Vicente Sanchiz, MD, Hospital Clinico de Valencia, Valencia,
Spain, Luis Yuguero, MD, Hospital General Yague, Burgos, Spain, Antonio M.
Pueyo, MD, Hospital Virgen del Camino, Pamplona, Spain, Paul Kortan, MD,
St. Michael, Toronto, Canada, Truls Hauge, MD, Ullevål University Hospital,
Oslo, Norway, António J. Marques, MD, De Santa Maria Epe, Lisboa, Portugal,
Santiago Rodriguez, MD, Hospital Virgen de La Concha, Zamora, Spain, Fran-
cesc Vida, MD, Hospital de Manresa-Althaia, Manresa, Spain, Leopoldo
Martin-Herrera, MD, Hospital Puerta del Mar, Cadiz, Spain, Antonio Naranjo,
MD, Hospital Universitario Reina Sofia, Cordoba, Spain, Michael Hünerbein,
MD, Helios Robert-Rössle-Klinik, Berlin, Germany, Maria Gloria Fernandez,
Hospital Clinic I Provincial de Barcelona, Barcelona, Spain, Antonio Lopez-
Serrano, MD, Hospital Universitario Dr Peset, Valencia, Spain, Angelo Ferrari,
MD, Instituto Paulista de Gastroenterología, São Paulo, Brazil, Blanca Fer-
reiro, MD, Hospital Carlos Haya, Malaga, Spain, Victor Orive, MD, Hospital de
Basurto, Bilbao, Spain, Vicente Sanchiz, MD, Complejo Hospitalario de
Orense, Orense, Spain, Prem Premchand, MD, Oldchurch Hospital (The New
Queens Hospital), Romford, United Kingdom, Juan A. Gonzalez, Hospital
Ramon Y Cajal, Majadahonda, Spain, Joyce Peetermans, PhD, Lina Ginnetti,
MA, Robert Walsh, MD, Matthew Rousseau, Boston Scientific, Natick, Massa-
chusetts, USA
www.giejournal.org