ORIGINAL CONTRIBUTION

Improving Quality of Life for People with Incurable

Large-Bowel Obstruction: Randomized Control
Trial of Colonic Stent Insertion
Christopher J. Young, M.B.B.S., M.S.1,2 • Katie J. De-loyde, M.Sc.3
Jane M. Young, Ph.D.2,3 • Michael J. Solomon, M.B.B.Ch. (Hons.), M.Sc.1,2,3
Emily H. Chew, M.P.H.3 • Chris M. Byrne, M.B.B.S., M.S.1,2
Glenn Salkeld, B.Bus., G.Dip.Hlth.Econ., M.P.H., Ph.D.3,4
Ian G. Faragher, M.B.B.S.5
1 Department of Colorectal Surgery, Royal Prince Alfred Hospital, Sydney, New South Wales, Australia
2 University of Sydney, Sydney, New South Wales, Australia
3 Surgical Outcomes Research Centre (SOuRCe), Royal Prince Alfred Hospital, Sydney, New South Wales, Australia
4 Sydney School of Public Health, University of Sydney, Sydney, New South Wales, Australia
5 Department of Colorectal Surgery, Western Hospital, Melbourne, Victoria, Australia

BACKGROUND:  Surgery remains the dominant treatment
for large-bowel obstruction, with emerging data on self-
expanding metallic stents.
OBJECTIVE:  The aim of this study was to assess whether
stent insertion improves quality of life and survival in
comparison with surgical decompression.
DESIGN:  This study reports on a randomized control
trial (registry number ACTRN012606000199516).
SETTING:  This study was conducted at Royal Prince
Alfred Hospital, Sydney, and Western Hospital,
Melbourne.
Funding/Support: This study was entirely funded by the National
Health and Medical Rearch Council (NHMRC), Grant 457369, via Can-
cer Australia and the Priority-driven Collaborative Cancer Research
Scheme.
Financial Disclosure: None reported.
The trial is registered with the Australian and New Zealand Clinical Trial
Registry (ACTR), registration number: ACTRN01260600019951.
Poster presentation at the meeting of The American Society of Colon
and Rectal Surgeons, Phoenix, AZ, April 27 to May 1, 2013 and at the
meeting of the Clinical Oncology Society of Australia (COSA), Adelaide,
Australia, November 12 to 14, 2013.
Correspondence: Christopher Young, M.B.B.S., M.S., SOuRCe, PO
Box M157, Royal Prince Alfred Hospital, Sydney, NSW 2050, Australia.
E-mail: cyoungnsw@aol.com
Dis Colon Rectum 2015; 58: 838–849
DOI: 10.1097/DCR.0000000000000431
© The ASCRS 2015
PATIENTS AND INTERVENTION:  Patients with malignant
incurable large-bowel obstruction were randomly
assigned to surgical decompression or stent insertion.
MAIN OUTCOME MEASURES:  The primary end point was
differences in EuroQOL EQ-5D quality of life. Secondary
end points included overall survival, 30-day mortality,
stoma rates, postoperative recovery, complications, and
readmissions.
RESULTS:  Fifty-two patients of 58 needed to reach the
calculated sample size were evaluated. Stent insertion was
successful in 19 of 26 (73%) patients. The remaining 7
patients required a stoma compared with 24 of 26 (92%)
surgery group patients (p < 0.001). There were no stent-
related perforations or deaths. The surgery group had
significantly reduced quality of life compared with the stent
group from baseline to 1 and 2 weeks (p = 0.001 and
p = 0.012), and from baseline to 12 months (p = 0.01) in
favor of the stent group, whereas both reported reduced
quality of life. The stent group had an 8% 30-day mortality
compared with 15% for the surgery group (p = 0.668).
Median survival was 5.2 and 5.5 months for the groups
(p = 0.613). The stent group had significantly reduced
procedure time (p = 0.014), postprocedure stay (p = 0.027),
days nothing by mouth (p = 0.002), and days before free
access to solids (p = 0.022).
LIMITATIONS:  This study was limited by the lack of an
EQ-5D Australian-based population set.
CONCLUSIONS:  Stent use in patients with incurable
large-bowel obstruction has a number of advantages
with faster return to diet, decreased stoma rates, reduced
postprocedure stay, and some quality-of-life benefits.

838 Diseases of the Colon & Rectum Volume 58: 9 (2015)

Copyright © The American Society of Colon & Rectal Surgeons, Inc. Unauthorized reproduction of this article is prohibited.
 Diseases of the Colon & Rectum Volume 58: 9 (2015)
KEYWORDS:  Colonic stent insertion; Quality of life;
Large-bowel obstruction; Surgery.
L
arge-bowel obstruction (LBO) occurs in 8% to 29%
of patients with large-bowel cancer.1 Obstruction
can also occur owing to cancer of the ovary, uterus,
stomach, breast, bladder, and kidney. Treatment has tra-
ditionally been surgical, potentially involving resection of
part of the large bowel and possible stoma formation.2,3
Patients with LBO are more likely to experience complica-
tions from such surgery than patients without an obstruc-
tion.4,5 In patients with a noncurable obstruction, where
surgical intervention will only remedy the obstruction and
not the underlying disease, avoidance of surgery may be
of benefit to the patient in terms of physical functioning,
pain, and other dimensions of quality of life (QoL).6,7
The majority of obstructing colorectal cancers are
left-sided, potentially amenable to colonoscopic interven-
tion.1,8 There are data to suggest that the placement of a
self-expanding metallic stent can treat the obstruction
in such patients, with a high probability of technical and
clinical success, decreased length of stay, and similar mor-
bidity and survival to surgical decompression.7,9 However,
previous studies have not assessed whether stenting con-
fers improved QoL for patients with malignant LBO com-
pared with surgical intervention.
Colonic stenting is an expensive technology that may
prove cost-effective if the costs offset extended postopera-
tive stay and morbidity in patients undergoing conven-
tional surgery.10,11
The primary aim of this randomized control trial
(RCT) was to assess whether palliative stent insertion im-
proves QoL compared with surgical intervention in malig-
nant noncurable LBO. Secondary aims compared survival,
stoma rates, length of hospital stay, and complication and
readmission rates.
MATERIALS AND METHODS
Study Design
This prospective RCT was approved by the Sydney
South West Area Health Service Ethics Review Commit-
tee (Royal Prince Alfred Hospital [RPAH]). Five centers
were initially enrolled, but only 2 (RPAH in Sydney and
Western Hospital in Melbourne) recruited patients to the
study. All patients ≥18 years who presented between Sep-
tember 2006 and November 2011 with a malignant LBO,
deemed not curable by surgical intervention (assessed in
a multidisciplinary team meeting where possible because
of the emergency nature of cases), were invited by their
surgeon to participate. Patients were excluded if they were
ASA grade IV or V,12 required urgent laparotomy because
of perforation or ischemia of the bowel, had evidence of
synchronous and separate sites of small and LBO, or were
Copyright © The American Society of Colon & Rectal Surgeons, Inc. Unauthorized reproduction of this article is prohibited.
839
cognitively impaired or unable to give informed consent.
We did not record total numbers of patients with LBO
who did not meet inclusion criteria.
After informed consent was given patients were ran-
domly allocated to either a stent insertion group or a
surgical decompression group (control group) by using
a computer-generated permuted block randomization
schedule,13 completed by the study coordinator. It was not
possible to blind surgeons and patients to the procedure;
however, all subjective outcome assessments were per-
formed by a blinded investigator.
Treatment
At presentation, a plain abdominal and erect chest x-rays
were performed as well as CT and/or other imaging to es-
tablish the malignant and noncurable nature of the pa-
thology. Criteria for incurable malignancy included the
presence of distant metastases that were nonresectable
and/or a nonresectable primary, and were not related to
comorbidity. All participating surgeons were experienced
in colorectal surgery and the placement of self-expanding
metallic stents.8,14
Patients allocated to the stent insertion group received
a self-expanding metallic stent placed through the ob-
structing lesion by the use of a combined endoscopic and
fluoroscopic approach. All stents inserted were Boston Sci-
entific (Natick, MA) uncovered stents. Patients who were
not successfully stented underwent surgical intervention
deemed appropriate by the operating surgeon. Data for
these patients were analyzed in the stent group according
to intention-to-treat principles.
Patients allocated to the surgical decompression
group had surgery to decompress their obstruction by a
technique determined appropriate by the operating sur-
geon and the pathology encountered. Although it was
expected that the vast majority of patients undergoing
surgery would require a stoma, a stoma was not enforced
as the only option. This was to ensure that the control
group reflected what the surgery would truly be, whether
with stoma, resection, or anastomosis, when stent inser-
tion was not an option.
Data Collection and Follow-up
Patient-reported outcomes were collected using the Euro-
QOL EQ-5D and the European Organisation for Research
and Treatment of Cancer QLQ CR-29. The EQ-5D is a vali-
dated 2-part QoL measure consisting of a 5-part question-
naire (recorded as no, some, and extreme problems) used
to derive an EQ-5D index on a scale of 0 to 1 (based on
US population weights), where 1 is the best health imagin-
able,15,16 and a visual analog score (VAS) allowing patients
to rate their health on a scale of 0 to 100, where 100 is the
best health imaginable. As per ­recommendations, when
 840
calculating scores, patients who were deceased were coded
as 0.15,16
The previously validated QLQ CR-29 includes 29
items that evaluate 4 functional and 18 symptom scales.17
Questionnaires were scored by converting the original val-
ues to a 0 to 100 range: for functional scales, 100 represents
a better level of functioning, and for symptom scales, 100
represents more symptoms. Because there was no differ-
ence in the rates of death between the 2 treatment groups,
and no available guidelines on coding deceased patients,
QLQ CR-29 data were analyzed on a complete case basis.
Demographic information, disease characteristics,
and QoL (EQ-5D) were collected by a blinded research
nurse at baseline (the QLQ CR-29 was not used at base-
line). Procedure and postprocedure information was col-
lected by the treating surgeon. Follow-up was by telephone,
or face-to-face if the patient was still hospitalized, at 1, 2,
and 4 weeks, and 3, 6, and 12 months postprocedure by
a blinded investigator. The investigator administered the
EQ-5D and QLQ CR-29 questionnaires and recorded any
complications and/or hospital readmissions, as well.
Outcomes
The primary end point was QoL measured by differences
between groups in EQ-5D index change scores.18,19 The
secondary end points included overall survival defined as
survival at 12 months postprocedure, 30-day mortality de-
fined as death from any cause up to 30 days after the pro-
cedure, rates of permanent stoma formation, procedure
time, anesthetic time, postprocedure stay, days spent in the
intensive care unit and high dependency unit, time to first
flatus and first bowel movement, time to start of a normal
diet, early postprocedure complication rate, 12-month
complication rate, length of stay, disease-related readmis-
sion, and differences in QLQ CR-29 scales.
Statistical Analysis
All data were analyzed on an intention-to-treat basis.
Statistical analysis was conducted using SPSS 20.0 (IBM,
Armonk, NY). The level of significance for all tests was
p < 0.05. Continuous data were analyzed by using an in-
dependent T test or nonparametric tests where appropri-
ate. EQ-5D index change scores and QLQ CR29 data were
compared between treatment groups. Categorical data
were analyzed using the χ2 and Fisher exact tests (FET).
Mean and medians are reported alongside the SD, inter-
quartile range, or 95% CIs, where appropriate.
Kaplan-Meier analysis was used to describe time-to-
event data. Overall survival was measured from the date
of surgery or stent procedure to the date of last follow-up,
or the date of death. The log-rank test was used to deter-
mine statistical significance between survival curves. Me-
dian survival and 6- and 12-month survival are reported
alongside a SE.
Copyright © The American Society of Colon & Rectal Surgeons, Inc. Unauthorized reproduction of this article is prohibited.
Young et al: RCT of Colonic Stent Insertion
Sample Size
To detect a clinically statistically significant difference in
QoL of 0.3 on the EQ-5D index between the 2 treatment
groups at 6 months, with α = 0.05, power = 80%, and σ of
0.4, a sample size of 29 patients per group (58 in total) was
calculated. The decision to use a difference of 0.3 or 30%
in sample size calculations was chosen in 2005 by group
consensus, following review of what data were available at
that stage.20,21 A 30% difference was thought by the inves-
tigators to be clinically significant.
RESULTS
Sixty-one patients presented between September 2006 to
November 2011, 5 of whom were excluded after being as-
sessed as ASA grade IV; 56 patients consented to join the
study. All patients presented with complete LBO. Of these
56 patients, 3 patients were found to be ineligible (all 3
had intra-abdominal metastatic disease but were found to
have pseudo-obstruction intraoperatively) and 1 patient
withdrew from the study after being randomly assigned
to surgery (Fig. 1). Thus, 52 patients were included in the
final analysis (Fig.1).
Patient demographics and disease characteristics are
shown in Table 1. The majority of the patients were men
(67%). The mean age was 66 years (SD = 13). All patients
had unresectable metastatic spread.
Treatment
There were 26 patients in each treatment arm (Fig. 1).
Treatment and postprocedure details are presented in
Table 2, and early postprocedure complications are listed
in Table 3. A stent was successfully inserted in 19 of 26
(73%) patients. One patient (4%) had stent reinsertion
at 5 months postprocedure because of stent occlusion by
tumor ingrowth. All 19 patients who were successfully
stented had patent bowel at 48 hours postprocedure. The
remaining 7 patients all proceeded to surgery. All 7 patients
not stented could not receive stents because of the inability
to successfully pass a guide wire across the obstruction. An
on-table lavage was performed in 8 of 26 (31%) patients
from the surgery group and in 2 of 7 (29%) patients in the
stent arm who proceeded to surgery. There were no stent-
related perforations or deaths.
Primary Outcome
EQ-5D Scores
EQ-5D index change scores are shown in Figure 2. There
was a significant difference between the 2 treatment
groups from baseline to 1 and 2 weeks (p = 0.001 and
p = 0.01) in favor of the stent group (Fig. 2). There was
also a significant difference between groups from baseline
to 12 months (p = 0.01) in favor of the stent group; how-
ever, both groups reported a reduced QoL (Fig. 2).
 Diseases of the Colon & Rectum Volume 58: 9 (2015) 841

Assessed for eligibility (n = 61)

Excluded (n = 5)
Patient did not meet inclusion criteria (n = 5)
Declined to participate (n = 0)

Randomized (n = 56)

Excluded after randomization Excluded after randomization

stent group (n = 2) surgery group (n = 2)
Patient did not meet inclusion criteria (n = 2) Withdrew before surgery (n = 1)
Patient did not meet inclusion criteria (n = 1)

For analysis (n = 52)

Allocated to stent group (n = 26) Allocated to surgery group (n = 26)

Received allocated intervention (n = 19) Received allocated intervention (n = 26)
Did not receive allocated intervention (stent insertion
was unsuccessful) (n = 7)

Complete data (n = 25) Follow-up 1 week Complete data (n = 25)

Deceased (n = 0) Deceased (n = 1)
QoL questionnaires not completed (n = 1) QoL questionnaires not completed (n = 0)
Withdrawn from study (n = 0) Withdrawn from study (n = 0)

Complete data (n = 22) Follow-up 2 weeks Complete data (n = 22)

Deceased (n = 0) Deceased (n = 2)
QoL questionnaires not completed (n = 4) QoL questionnaires not completed (n = 2)
Withdrawn from study (n = 0) Withdrawn from study (n = 0)

Complete data (n = 22) Follow-up 4 weeks Complete data (n = 20)

Deceased (n = 2) Deceased (n = 4)
QoL questionnaires not completed (n = 2) QoL questionnaires not completed (n = 2)
Withdrawn from study (n = 0) Withdrawn from study (n = 0)

Complete data (n = 13) Follow-up 3 months Completed data (n = 15)

Deceased (n = 11) Deceased (n = 8)
QoL questionnaires not completed (n = 1) QoL questionnaires not completed (n = 2)
Withdrawn from study (n = 1) Withdrawn from study (n = 1)

Completed data (n = 11) Follow-up 6 months Completed data (n = 9)

Deceased (n = 13) Deceased (n = 15)
QoL questionnaires not completed (n = 1) QoL questionnaires not completed (n = 0)
Withdrawn from study (n = 1) Withdrawn from study (n = 2)

Completed data (n = 7) Follow-up 12 months Completed data (n = 4)

Deceased (n = 17) Deceased (n = 19)
QoL questionnaires not completed (n = 1) QoL questionnaires not completed (n = 1)
Withdrawn from study (n = 1) Withdrawn from study (n = 2)

Analyzed (n = 26) Analysis Analyzed (n = 26)

FIGURE 1.  Study flow chart. QoL = quality of life.

Copyright © The American Society of Colon & Rectal Surgeons, Inc. Unauthorized reproduction of this article is prohibited.
 842 Young et al: RCT of Colonic Stent Insertion

TABLE 1.   Patient demographics and disease characteristics between treatment groups
Stent group, n (%) Surgery group, n (%) p value between groups
Patient demographics
 Mean age in years (SD), range 66 (11), 41–83 67 (14), 35–86 0.667
 Sex 0.768
  Male 17 (65) 18 (69)
  Female 9 (35) 8 (31)
 English language 0.337
  English 18 (69) 21 (81)
  Non-English 8 (31) 5 (19)
 Education 0.304
  Primary 9 (35) 4 (15)
  Secondary 12 (46) 13 (50)
  Tertiary 3 (12) 5 (19)
 Employment 0.002
  Paid (full/part time work) 12 (46) 2 (8)
  Unpaid (retired/unemployed/home 14 (54) 24 (92)
duties/pension)
 Social status 0.021
  Living alone 2 (8) 10 (39)
  Not living alone 24 (92) 16 (62)
 Smoking 0.034
  Current smoker/ex-smoker 10 (38) 18 (69)
  Never smoked 16 (62) 7 (28)
Disease characteristics
 Pathology
  Primary colorectal cancer 19 (73) 20 (77)
  Recurrent colorectal cancer 1 (4) 0
  Primary noncolorectal cancer 3 (12) 2 (8)
  Recurrent noncolorectal cancer 3 (12) 4 (15)
 ASA grade 0.079
  I / II 17 (65) 11 (42)
  III 7 (27) 14 (54)
 Site of obstruction
  Rectum 5 (19) 6 (23)
  Rectosigmoid 9 (35) 5 (19)
  Sigmoid 8 (31) 12 (46)
  Descending colon 2 (8) 1 (4)
  Splenic flexure 1 (4) 1 (4)
  Transverse colon 0 0
  Hepatic flexure 1 (4) 0
  Ascending colon 0 1 (4)
 Obstruction type 1.000
  Intrinsic 20 (77) 20 (77)
  Extrinsic 6 (23) 6 (23)
 Metastasis
  Liver 19 (73) 21 (81)
  Lung 7 (27) 8 (31)
  Peritoneal 8 (31) 11 (42)
  Retroperitoneal 1 (4) 1 (4)
  Bone 0 1 (4)
  Brain 1 (4) 0
Where % ≠ 100 data are missing.

Figure 3 shows the median EQ-5D VAS scores between
treatment groups. There was no significant difference in
VAS scores at any time point between treatment groups.
Overall, 15 of 26 (58%) stent group patients were re-
corded as having an increased QoL from baseline to 1 week
in comparison with 7 of 26 (27%) of the surgery group
(χ² [1] = 5.7, p = 0.02). There were no significant d
­ ifferences
in whether the patient had an increased or decreased QoL
at any other time point.
Patients who reported extreme/some problems for
any of the 5 descriptive QoL questions are presented in
Table 4. Of the patients who were alive and completed a
QoL questionnaire at 12 months postprocedure (n = 11),
the majority (n = 8) reported no problems.

Copyright © The American Society of Colon & Rectal Surgeons, Inc. Unauthorized reproduction of this article is prohibited.
 Diseases of the Colon & Rectum Volume 58: 9 (2015) 843

TABLE 2.   Procedure and postprocedure details between treatment groups

Stent group, n (%) Surgery group, n (%) p value between groups
Stent data
 Median length of obstructing stricture, cm (IQR) 5 (3–5)
 Median distance of obstructing stricture to anal 20 (12–40)
verge, cm (IQR)
Surgery data
 Bowel resection 2 (29) 11 (42) 0.68, FET
 Bowel resection
  Right sided 0 2 (18) 0.71, FET
  Left sided 2 (100) 9 (82)
 Anastomosis 0 2 (8) 0.61, FET
Procedure data
 Median procedure time a (95% CI) 49 (35–128) 117 (70–145) 0.01
 Median anesthetic time b (95% CI) 82 (50–123) 162 (115–185) 0.02
Postprocedure
 Stoma 7 (27) 24 (92) χ² [1] = 23.1, p < 0.001
 Repeat stent insertion 1 (4) -
 Early postprocedure complication c 10 (38) 14 (54) χ² [1] = 0.70, p = 0.40
 Discharge to
  Hospital / other d 4 (16) 11 (44) χ² [1] = 4.7, p = 0.03
  Home 21 (84) 14 (56)
 Median length of postprocedure stay e (95% CI) 7 (3–12) 11 (8–17) 0.03
 Median days in ICU (95% CI) 0 (0-0) 0 (0-0) 0.32
 Median days in HDU (95% CI) 0 (0-0) 0 (0-0) 0.32
 Median days nothing by mouth (95% CI) 0 (0–1) 2 (1–4) 0.002
 Median days before free access fluids (95% CI) 1 (0–1) 2 (1–4) 0.001
 Median days before free access solids (95% CI) 2 (1–3) 4 (2–5) 0.02
 Median days first flatus (95% CI) 1 (0–1) 2 (1–2) 0.001
 Median days first bowel movement (95% CI) 1 (0–1) 2 (1–3) 0.002
Readmission
 Readmission f 18 (69) 13 (50) χ² [1] = 2.0, p = 0.16
30-day mortality 2 (8) 4 (15) 0.67, FET
FET = Fisher exact tests; HDU = high dependency unit; ICU = intensive care unit; IQR = interquartile range.
a
Procedure time: assessed in minutes from entry to anesthetic bay until entry to recovery.
b
Anesthetic time: assessed in minutes from induction or sedation commencement to time of leaving the procedure room.
c
Early postprocedure complication rates where a complication is defined as any procedure-related complication.
d
Other includes palliative care, community home, high-level care, and rehabilitation.
e
Postprocedure stay assessed as the number of days spent in the hospital for the procedure.
f
Readmission rate defined as a readmission for any reason up to 12 months postprocedure.

Secondary End Points
Thirty-Day Mortality and Survival
Thirty-day mortality for the stent group was 8% (95% CI,
2.1–24.1) compared with 15% (95% CI, 6.6–33.5) for the
surgery group (p = 0.67) (Table 2). These 6 patients died
of metastatic disease (n = 3), cardiac (n = 2), or respiratory
(n = 1) causes. All of these patients were ASA grade III.
Thirty- six patients had died by the end of the study (12
months postprocedure): 17 from the stent group and 19
from the surgery group. All causes of death were assessed,
and none were stent related, either in the short or long
term. Median survival for the stent group was 5.2 months
(SE, 3.1; 95% CI, 0.0–11.5), and 5.5 months (SE, 0.6; 95%
CI, 4.2–6.7) for the surgery group. There was no significant
difference in survival curves between treatment groups
(p = 0.61) (Fig. 4). All patients who died by 12 months
were ASA grade II or III, noting that patients who were
ASA grade IV were excluded from the study. When death
at 12 months was assessed, in the surgery group, 6 of 11
(45%) patients ≤65 years of age had died compared with
14 of 15 (93%) patients >65 years of age (p = 0.21, FET),
whereas patients with higher ASA grade were not more
likely to die (χ2 = 2.91, df = 2, p = 0.23), contrasting with
the stent group where 8 of 14 (57%) patients ≤65 years of
age had died compared with 9 of 12 (75%) >65 years of age
(p = 0.43, FET), whereas patients with higher ASA grade
were more likely to die (χ2 = 8.44, df = 2, p = 0.01).
QLQ-CR29 Scores
There was no significant difference in any QLQ-CR29 func-
tional scale between treatment groups (Table 5). Similarly
there were no significant differences in any symptom scale
at 3, 6, and 12 months between treatment groups. However,
patients from the stent group reported higher abdomi-
nal pain at 2 and 4 weeks compared with surgery patients
(T [41] = 2.0, p = 0.03 and T [40] = 2.4, p = 0.02) (Table 4).
Patients in the stent group were also significantly more like-
ly to report being embarrassed by their bowel movements

Copyright © The American Society of Colon & Rectal Surgeons, Inc. Unauthorized reproduction of this article is prohibited.
 844 Young et al: RCT of Colonic Stent Insertion

TABLE 3.   Early postprocedure complications

Complication Stent (n = 26) Surgery (n = 26)
n (%) n (%)
Early postprocedure complication (total) a
10 (38) 14 (54)
Pyrexia 2 (8) 6 (23)
Recurrent postoperative ileus 0 1 (4)
Pulmonary embolus/deep vein thrombosis 0 0
Urinary retention 3 (12) 2 (8)
Urinary tract infection 1 (4) 1 (4)
Wound infection 0 1 (4)
Anastomotic leakage 0 0
Other medical complication 2 (8) b 7 (17) d
Other complication 3 (12) c 4 (15) e
a
Some patients had more than one complication.
b
Includes: rapid atrial fibrillation, hypotension, pulmonary edema, anemia requiring blood transfusion.
c
Includes stoma site infection with Pseudomonas aeruginosa; superficial wound dehiscence and drainage of collection; weakness of ankle dorsiflexion, unable to walk.
d
Includes confusion days 1 and 2 postprocedure; pulmonary edema, atrial fibrillation; respiratory arrest; malignant pleural effusion; acute renal failure (prerenal); hospital-
acquired pneumonia; anemia requiring blood transfusion.
e
Includes small 1- to 2-mL wound hematoma that was evacuated; bleeding per stoma; high stomal output; electrolyte imbalance; leakage from wound requiring drainage of
collection.

and an increased stool frequency at 4 weeks than surgery
patients (T [40] = 2.9, p = 0.006 and T [40] = 2.8, p = 0.01)
(Table 5). Surgery patients were significantly more likely
to report having sore skin at 2 weeks than patients who re-
ceived stents (T [40] = 2.0, p = 0.048) (Table 5).
For patients in the stent group, stent insertion failed
in 4 of 6 (67%) patients with extrinsic bowel obstruc-
tion compared with 3 of 20 (15%) patients with intrin-
sic ­bowel obstructions and required stoma formation
(p = 0.03, FET).

Stoma Formation Procedure and Postprocedure Details

None of the 19 patients in the stent group who were suc-
cessfully stented required a stoma fashioned postpro-
cedure. The remaining 7 patients from this group who
proceeded to surgery all had a stoma fashioned compared
with 24 patients in the surgery group (p < 0.001) (Table 2).
The stent group had significantly reduced procedure and
anesthetic time compared with the surgery group (Ta-
ble 2). The stent group also had significantly reduced post-
procedure stay compared with the surgery group (Table 2)
(p = 0.03). Similarly, the stent group had significantly

Mean EQ-5D change score

0.4

(p = 0.01, 0.22)
0.2 (p = 0.19, 0.14)
(p = 0.001, 0.25)
(p = 0.26, 0.14)
(p = 0.09, 0.21) (p = 0.01, 0.24)
Baseline to 1 week Baseline to 2 weeks Baseline to 4 weeks Baseline to 3 months Baseline to 6 months Baseline to 12 months

-0.2

-0.4

-0.6

Stent group Surgery group

-0.8

FIGURE 2.  EQ-5D index change scores from baseline to follow-up. A negative score indicates a reduction in QoL. Error bars represent 95% CI
(p value, absolute difference).

Copyright © The American Society of Colon & Rectal Surgeons, Inc. Unauthorized reproduction of this article is prohibited.
 Diseases of the Colon & Rectum Volume 58: 9 (2015) 845

Median EQ-5D VAS score

70

60

50

Stent
40

Surgery
30

20

10

p = 0.06 p = 0.88 p = 0.86 p = 0.37 p = 0.37 p = 0.91 p = 0.88

Baseline 1 week 2 weeks 4 weeks 3 months 6 months 12 months

Time point

FIGURE 3.  Median EQ-5D VAS scores across time points. Median EQ-5D VAS is a scale of 0 to 100, where 100 is the best health imaginable. p
values shown are between treatment groups at each time point. VAS = visual analog score.

r­educed days regarding nothing by mouth (p = 0.002),
before free access fluid (p = 0.001) and solids (p = 0.02),
to first flatus (p = 0.001), and to first bowel movement
(p = 0.002), compared with those in the surgery group
(Table 2).Patientsfromthestent groupweresignificantly more
likely to be discharged home rather than to another health
care facility compared with patients from the surgery group
(p = 0.03) (Table 2).
Chemotherapy Postprocedure
Chemotherapy of any type was given postprocedure to 11
(42%) patients in the stent group and to 11 (42%) patients

TABLE 4.   Proportion of patients reporting “extreme” or “some” problems on the EQ-5D descriptive system across time points by group
Extreme / some problems, n (%)
Treatment group (n) Morbidity Self-care Usual activities Pain/discomfort Anxiety/depression
Baseline
 Stent (26) 16 (62) 8 (31) 17 (65) 22 (85) 18 (69)
 Surgery (26) 11(42) 8 (31) 12 (46) 16 (62) 11 (42)
1 week
 Stent (25) 12 (48) 9 (36) 22 (88) 20 (80) 18 (72)
 Surgery (25) 14 (56) 14 (56) 20 (80) 16 (64) 11 (44)
2 weeks
 Stent (22) 12 (55) 8 (36) 17 (77) 17 (77) 16 (73)
 Surgery (22) 12 (55) 10 (45) 19 (86) 12 (55) 10 (45)
4 weeks
 Stent (22) 10 (45) 9 (41) 16 (73) 14 (64) 14 (64)
 Surgery (20) 8 (40) 6 (30) 14 (70) 9 (45) 6 (30)
3 months
 Stent (13) 3 (92) 0 7 (54) 9 (69) 4 (31)
 Surgery (15) 8 (53) 5 (33) 11 (73) 8 (53) 7 (47)
6 months
 Stent (11) 2 (18) 3 (27) 7 (64) 8 (73) 4 (36)
 Surgery (9) 3 (33) 1 (11) 5 (56) 5 (56) 4 (44)
12 months
 Stent (7) 2 (29) 0 2 (29) 1 (14) 2 (29)
 Surgery (4) 0 0 0 0 1 (25)

Copyright © The American Society of Colon & Rectal Surgeons, Inc. Unauthorized reproduction of this article is prohibited.
 846
Survival (%)
100
80
60
40
20
Treatment No. Survival % (SE)
group events 6 month 12 month
Stent 17 / 26 49.5 (9.9) 31.5 (9.5)
Surgery 19 / 26 44.3 (9.9) 24.4 (8.6)
0 2 4 6 8 10
Months
FIGURE 4.  Overall survival between treatment groups.
in the surgery group (p = 0.61, FET). For the patients
with primary colorectal cancer, 7 of 17 (41%) patients in
the stent group and 8 of 18 (44%) patients in the surgery
group, who were alive after more than 30 days postpro-
cedure, had any chemotherapy postprocedure (p = 0.56,
FET). Chemotherapy for patients with primary colorec-
tal cancer in the stent group included oxaliplatin based
(n = 2), Camptosar based (n = 3), and capecitabine based
(n = 2). Four of these 7 also had bevacizumab added to
their chemotherapy. Chemotherapy for patients with
primary colorectal cancer in the surgery group includ-
ed oxaliplatin based (n = 5), Camptosar based (n = 1),
capecitabine based (n = 2), and none had bevacizumab
added to their chemotherapy.
Complications and readmissions
Ten (38%) patients from the stent group and 14 (54%)
from the surgery group developed early postprocedure
complications (p = 0.40) (Tables 2 and 3). Overall, 33
patients had at least one complication over the first 12
months after their procedure: 15 (58%) patients from the
stent group and 18 (70%) patients from the surgery group
(p = 0.56).
There was no significant difference in the total
­readmission rate in the 12 months postprocedure be-
tween groups (Table 2). There was also no significant
difference in the mean total readmission length of stay
(14.1 days (SD = 13.5) for the stent group and 8.2 days
(SD = 11.6) for the surgery group) between treatment
groups (T [50] = 1.7, p = 0.09).
Copyright © The American Society of Colon & Rectal Surgeons, Inc. Unauthorized reproduction of this article is prohibited.
Young et al: RCT of Colonic Stent Insertion
DISCUSSION
Study group This RCT demonstrated significantly less reduction in
Stent QoL for patients with an incurable LBO following stent
Surgery insertion compared with surgery, from baseline to 1 and
Stent-censored
Surgery-censored 2 weeks, as well as at 12 months postprocedure. The QoL
measures did not reach the predicted difference set to es-
tablish the sample size calculation. The patients in the stent
group also had significantly lower permanent stoma rates,
reductions in postprocedure stay and procedure times, as
well as an earlier return of bowel function, and they were
more likely to be discharged earlier.
Many patients with incurable LBO are not suitable
for resection and anastomosis, with stenting being an al-
ternate tool available to clinicians attempting to avoid a
stoma. Stenting benefits over traditional surgery, with
comparable long-term outcomes, reinforce the finding
that stent insertion at least appears to be safe. The 30-
day mortality rate of 8% for the stent group and 15% for
the surgery group, although not significantly different,
12 are slightly higher than reported in the meta-analyses of
Sagar7 from 2011 and Cirocchi et al9 from 2013, noting,
however, that all our patients had incurable disease and
that deaths occurred in patients with ASA grade III. Al-
though the complication rates of 27% for the stent group
and 46% for the surgery group appear clinically signifi-
cant, they were not statistically significant. These results
were similar to complication rates reported by Sagar7 in
2011 and Cirocchi et al9 in 2013.
The key issue for patients with incurable disease is
the length and quality of remaining life.22 The similar
median survivals of 5.2 months and 5.5 months for the
surgery and stent groups are slightly longer but similar
to the 4-month median survival reported by Young et al8
in an uncontrolled cohort of 100 patients who had an at-
tempted self-expanding metallic stent insertion, 89% with
palliative intent. That 12 of 56 (21.4%) of our patients had
metastatic noncolorectal malignancy, and 22 of 56 (42%)
had any postprocedure chemotherapy may be associated
with our survival times.
This study found a mixed picture in the QoL data,
some of which we cannot explain, and others which we
believe are areas for attention in the future. This study
found similar EQ-5D QoL VAS data in both groups at
all time points postprocedure. The stent group had sig-
nificantly higher EQ-5D change scores compared with the
surgery group, however, this did not reach clinical signifi-
cance. The QLQ-CR29 data also revealed more abdominal
pain and embarrassment with bowel movements as well
as higher stool frequency for stent group patients, whereas
surgery patients had more sore skin. This demonstrates
the complementary nature of both the EQ-5D and QLQ-
CR29. Presumably this is because the EQ-5D reveals an
overall picture of QoL, whereas the QLQCR-29 is more
specific regarding colorectal cancer sequelae.15–17,23 No
 Diseases of the Colon & Rectum Volume 58: 9 (2015) 847

TABLE 5.   QLQ-CR29 responses by study groups

Mean score (SD)
2 weeks 4 weeks 3 months 6 months 12 months
(n = 44) (n = 42) (n = 28) (n = 20) (n = 11)
Functional scales (scale of 0–100 where 100 represents a better level of functioning)
 Body image Stent 74 (26) 80 (21) 85 (18) 83 (14) 84 (22)
Surgery 80 (22) 84 (26) 80 (27) 77 (32) 89 (22)
 Anxiety Stent 43 (34) 50 (38) 64 (38) 63 (40) 62 (30)
Surgery 64 (37) 67 (39) 60 (38) 67 (29) 83 (19)
 Weight Stent 79 (27) 68 (33) 79 (28) 83 (24) 71 (36)
Surgery 68 (30) 87 (20) 73 (29) 67 (33) 92 (17)
 Sexual interest: men Stent 8 (24) (7%)a 15 (27) (21%) a 17 (25) (33%) a 27 (28) (50%) a 13 (30) (20%) a
Surgery 5 (13) (8%) a 4 (12) (7%) a 13 (18) (20%) a 0 (0) (0%) a 0 (0) (0%) a
 Sexual interest: women Stent 8 (17) (17%) a 0 (0) (0%) a 0 (0) (0%) a 0 (0) (0%) a 0 (0) (0%) a
Surgery 0 (0) (0%) a 0 (0) (0%) a 0 (0) (0%) a 0 (0) (0%) a -
Symptom scales (scale of 0–100 where 100 represents more symptoms)
 Urinary frequency scale Stent 42 (34) 26 (28) 27 (26) 33 (28) 38 (28)
Surgery 33 (26) 19 (22) 36 (18) 24 (22) 13 (16)
 Blood / mucus in stool scale Stent 6 (11) 7 (10) 5 (8) 3 (7) 0 (0)
Surgery 8 (23) 12 (23) 7 (11) 9 (15) 0 (0)
 Stool frequency b Stent 72 (30) 28 (15–41)* 23 (19) 12 (22) 24 (27)
Surgery 86 (18 8 (2–13)* 16 (28) 9 (12) 13 (25)
 Urinary incontinence Stent 11 (26) 8 (23) 10 (16) 10 (23) 5 (13)
Surgery 10 (24) 5 (12) 11 (24) 11 (24) 8 (17)
 Dysuria Stent 7 (18) 13 (27) 7 (19) 13 (17) 0 (0)
Surgery 10 (22) 5 (12) 4 (12) 0 (0) 8 (17)
 Abdominal pain Stent 38 (27–49)* 27 (16–39)* 29 (34) 43 (27) 10 (16)
Surgery 23 (11–34)* 10 (1–19)* 16 (28) 26 (32) 0 (0)
 Buttock pain Stent 19 (27) 18 (27) 14 (22) 17 (24) 10 (25)
Surgery 14 (25) 10 (19) 20 (30) 0 (0) 0 (0)
 Bloated feeling Stent 24 (24) 17 (25) 17 (31) 30 (33) 0 (0)
Surgery 21 (22) 15 (25) 22 (33) 30 (35) 0 (0)
 Dry mouth Stent 37 (28) 26(29) 41 (27) 50 (36) 38 (36)
Surgery 29 (28) 22 (29) 29 (31) 33 (33) 0 (0)
 Hair loss Stent 12 (25) 9 (18) 21 (28) 23 (32) 29 (41)
Surgery 4 (11) 0 (0) 10 (29) 2244) 17 (33)
 Trouble with taste Stent 27 (39) 29 (33) 31 (42) 23 (27) 5 (13)
Surgery 21 (30) 14 (20) 29 (35) 33 (33) 17 (19)
 Flatulence Stent 18 (27) 9 (19) 21 (31) 30 (25) 19 (18)
Surgery 25 (21) 18 (25) 27 (32) 15 (24) 8 (17)
 Fecal incontinence b Stent 13 (20) 8 (14) 17 (29) 13 (23) 0 (0)
Surgery 23 (30) 18 (25) 13 (17) 22 (17) 0 (0)
 Sore skin Stent 8 (0–17)* 17 (22) 17 (25) 17 (24) 10 (16)
Surgery 16 (8–24)* 13 (23) 18 (28) 7 (15) 8 (17)
 Embarrassed by bowel movement Stent 27 (28) 33 (8–59)* 33 (41) 44 (19) 67 (0)
Surgery 2 (27) 9 (2–17)* 10 (16) 13 (25) 8 (17)
 Stoma care problems Stent 33 (41) 5 (13) 0 (0) 11 (19) 0 (0)
Surgery 33 (38) 22 (40) 8 (28) 8 (15) 0 (0)
 Impotence Stent 39 (49) 38 (49) 25 (50) 8 (17) 0 (0)
Surgery 0 (0) 0 (0) 0 (0) 50 (71) 0 (0)
 Dyspareunia Stent 0 (0) 0 (0) – – 0 (0)
Surgery - – – – –
* represents a significant difference between groups at that time point (95% CI). – indicates no data
a
% of patients who completed the questionnaire who were “interested in sex” (a little, quite a bit, or very much).
b
These questions differed slightly for stoma and nonstoma patients as per QLQ-CR29 responses.

QLQ-CR29 symptoms or functionality outcomes that pre-
dicted a decrease in EQ-5D QoL were found. This study
reveals further QoL data in addition to the nonrandom-
ized QoL data reported by Nagula et al.24
The stent-related perforations reported in the study
by van Hooft et al25 have certainly been a cause for con-
cern in regard to the role of stents in LBO. We did not
have any stent-related perforations or deaths in any pa-
tients. We cannot explain the difference in perforation
rates, but note that the proceduralists in our RCT had
performed more than 100 stent insertions before the
RCT commencement.8

Copyright © The American Society of Colon & Rectal Surgeons, Inc. Unauthorized reproduction of this article is prohibited.
 848
The clinical success rate of stent insertion in our study
was 79%, compared with 78.05% reported by Sagar7 and
52.5% reported by Cirocchi et al.9 We note our higher suc-
cess rate of 85% in patients with intrinsic bowel obstruc-
tion in comparison with 33% in patients with extrinsic
bowel obstruction. We note the overall lower rate of stent
insertion in this RCT than reported in our prospective se-
ries of 89%, which we assume is due to the elimination of
selection bias.8
The strengths of this study are the RCT design, and
the very high compliance rate of adherence to the study
design, especially when it is difficult to randomly assign
patients when there is a large difference in treatments
in each arm.26 In this study, there was a large perceived
difference between stent and surgery, and patient pref-
erences made recruitment difficult, but, because of the
nonbiased manner in which we believe the recruiters
informed and consented the patients, all 56 patients
eligible initially enrolled. The limitations of this study
include the sample size of 56 (4 excluded), which fell
just short of the calculated sample size of 58. Having at-
tempted stent insertion in 79 patients with LBO with
ASA grade I, II, or III at RPAH alone between August
1999 and December 2005 (65 months), over the study re-
cruitment period of September 2006 to November 2011
(63 months), we expected, with 2 centers recruiting, to
reach the sample size.8 It is disappointing not to reach
the calculated sample size, but we had already extended
the trial for 2 years longer than planned, and only had
2 of 5 centers enrolled recruiting patients, with very low
recruitment numbers in these extra 2 years. We presume
this was because of the rapid uptake of stenting off trial
outside of the 2 recruiting centers, meaning that some
centers that had previously referred patients to us for
potential stenting now had the capabilities themselves.
The researchers felt justified in ending recruitment so
that further knowledge in this area could be published
promptly.
Other limitations of this study include the US pop-
ulation-based EQ-5D (no Australian set currently exists)
that was used. There was also potential for inaccuracy ow-
ing to the small sample size in EQ-5D QoL results because
the majority of patients died. Another limitation was the
exclusion of ASA grade IV patients from the study. It is in
this high-risk group that potential reductions in mortality
and morbidity from a procedure are paramount, but, at
the time of study design, not all investigators were pre-
pared to randomly assign ASA grade IV patients into the
study.
CONCLUSION
The use of metallic stents compared with surgical decom-
pression in patients with an incurable LBO has some QoL
Copyright © The American Society of Colon & Rectal Surgeons, Inc. Unauthorized reproduction of this article is prohibited.
Young et al: RCT of Colonic Stent Insertion
benefits, although it shows no survival benefits, and it has
a number of other advantages with faster return to diet,
lower permanent stoma rates, reduced procedure times,
and reduced postprocedure stay.
ACKNOWLEDGMENTS
The authors thank the patients who participated in this study
and the clinical and administrative staff who assisted with pa-
tient recruitment and data collection at each site.
REFERENCES
1. Deans GT, Krukowski ZH, Irwin ST. Malignant obstruction of
the left colon. Br J Surg. 1994;81:1270–1276.
2. Lee YM, Law WL, Chu KW, Poon RT. Emergency surgery
for obstructing colorectal cancers: a comparison be-
tween right-sided and left-sided lesions. J Am Coll Surg.
2001;192:719–725.
3. Single-stage treatment for malignant left-sided colonic ob-
struction: a prospective randomized clinical trial comparing
subtotal colectomy with segmental resection following intra-
operative irrigation. The SCOTIA Study Group. Subtotal Col-
ectomy versus On-table Irrigation and Anastomosis. Br J Surg.
1995;82:1622–1627.
4. McArdle CS, Hole DJ. Emergency presentation of colorec-
tal cancer is associated with poor 5-year survival. Br J Surg.
2004;91:605–609.
5. Biondo S, Martí-Ragué J, Kreisler E, et al. A prospective study of
outcomes of emergency and elective surgeries for complicated
colonic cancer. Am J Surg. 2005;189:377–383.
6. Temple LK, Hsieh L, Wong WD, Saltz L, Schrag D. Use of sur-
gery among elderly patients with stage IV colorectal cancer.
J Clin Oncol. 2004;22:3475–3484.
7. Sagar J. Colorectal stents for the management of malig-
nant colonic obstructions. Cochrane Database Syst Rev.
2011:CD007378..
8. Young CJ, Suen MK, Young J, Solomon MJ. Stenting large bowel
obstruction avoids a stoma: consecutive series of 100 patients.
Colorectal Dis. 2011;13:1138–1141.
9. Cirocchi R, Farinella E, Trastulli S, et al. Safety and efficacy of
endoscopic colonic stenting as a bridge to surgery in the man-
agement of intestinal obstruction due to left colon and rectal
cancer: a systematic review and meta-analysis. Surg Oncol.
2013;22:14–21.
10. Xinopoulos D, Dimitroulopoulos D, Theodosopoulos T, et al.
Stenting or stoma creation for patients with inoperable malig-
nant colonic obstructions? Results of a study and cost-effective-
ness analysis. Surg Endosc. 2004;18:421–426.
11. Singh H, Latosinsky S, Spiegel BM, Targownik LE. The cost-
effectiveness of colonic stenting as a bridge to curative sur-
gery in patients with acute left-sided malignant colonic
obstruction: a Canadian perspective. Can J Gastroenterol.
2006;20:779–785.
12. Prause G, Ratzenhofer-Comenda B, Pierer G, Smolle-Jüttner F,
Glanzer H, Smolle J. Can ASA grade or Goldman’s cardiac risk
index predict peri-operative mortality? A study of 16,227 pa-
tients. Anaesthesia. 1997;52:203–206.
 Diseases of the Colon & Rectum Volume 58: 9 (2015)
13. Beller EM, Gebski V, Keech AC. Randomisation in clinical trials.
Med J Aust. 2002;177:565–567.
14. Faragher IG, Chaitowitz IM, Stupart DA. Long-term results of
palliative stenting or surgery for incurable obstructing colon
cancer. Colorectal Dis. 2008;10:668–672.
15. Shaw JW, Johnson JA, Coons SJ. US valuation of the EQ-5D
health states: development and testing of the D1 valuation
model. Med Care. 2005;43:203–220.
16. Kind P, Dolan P, Gudex C, Williams A. Variations in population
health status: results from a United Kingdom national ques-
tionnaire survey. BMJ. 1998;316:736–741.
17. Whistance RN, Conroy T, Chie W, et al; European Organisation
for the Research and Treatment of Cancer Quality of Life Group.
Clinical and psychometric validation of the EORTC QLQ-CR29
questionnaire module to assess health-related quality of life in
patients with colorectal cancer. Eur J Cancer. 2009;45:3017–3026.
18. Sanz MA, Aledort L, Mathias SD, Wang X, Isitt JJ. Analysis of
EQ-5D scores from two phase 3 clinical trials of romiplostim
in the treatment of immune thrombocytopenia (ITP). Value
Health. 2011;14:90–96.
19. Bennett L, Zhao Z, Barber B, et al. Health-related quality of
life in patients with metastatic colorectal cancer treated with
panitumumab in first- or second-line treatment. Br J Cancer.
2011;105:1495–1502.
Copyright © The American Society of Colon & Rectal Surgeons, Inc. Unauthorized reproduction of this article is prohibited.
849
20. Drummond MF, O’Brien BJ, Stoddart GL, Torrance GW. Meth-
ods for the Economic Evaluation of Health Care Programmes, 2nd
ed. Oxford: Oxford University Press; 2003.
21. Dolan P, Gudex C, Kind P, Williams A. A Social Tariff for Euro-
Qol; Results From a UK General Population Survey. Discussion
paper 138, Centre for Health Economics. York, United King-
dom: University of York; 1995.
22. McCahill LE, Krouse R, Chu D, et al. Indications and use of pal-
liative surgery-results of Society of Surgical Oncology survey.
Ann Surg Oncol. 2002;9:104–112.
23. Hurst NP, Kind P, Ruta D, Hunter M, Stubbings A. Measuring
health-related quality of life in rheumatoid arthritis: validity,
responsiveness and reliability of EuroQol (EQ-5D). Br J Rheu-
matol. 1997;36:551–559.
24. Nagula S, Ishill N, Nash C, et al. Quality of life and symptom
control after stent placement or surgical palliation of malignant
colorectal obstruction. J Am Coll Surg. 2010;210:45–53.
25. van Hooft JE, Bemelman WA, Oldenburg B, et al; collaborative
Dutch Stent-In study group. Colonic stenting versus emergen-
cy surgery for acute left-sided malignant colonic obstruction:
a multicentre randomised trial. Lancet Oncol. 2011;12:344–352.
26. Harrison JD, Solomon MJ, Young JM, et al. Surgical and on-
cology trials for rectal cancer: who will participate? Surgery.
2007;142:94–101.