World J Surg
DOI 10.1007/s00268-014-2595-y
Trends in Demographics and Management of Obstructing
Colorectal Cancer
Zaheer Moolla • Thandinkosi E. Madiba
Ó Société Internationale de Chirurgie 2014
Abstract
Introduction Obstructing colorectal cancer (CRC) has an
aggressive clinical course and poorer prognosis. With the
increasing incidence and differing clinical and pathologic
spectrum of CRC among Black patients, as well as a paucity
of African studies, regional analysis is required. Our aim was
to describe the demographics and management of obstructing
CRC among the different racial groups in South Africa and to
compare these parameters with international standards.
Patients and methods Patients referred to Inkosi Albert
Luthuli Central Hospital, Durban, South Africa, with CRC
between 2000 and 2012 were followed prospectively.
Demographic information, site of obstruction, and manage-
ment of patients who underwent emergency surgery for
malignant large bowel obstruction were analyzed separately.
Results CRC was diagnosed in 1,425 patients. A total of 203
three patients (14.3 %) required emergent treatment for acute
large bowel obstruction. The mean age at presentation with
obstructing CRC was 59 years. Black patients presented sig-
nificantly younger (50 years) than White (64), Indian (60), or
Colored (61) patients (p \ 0.001). The most common sites of
obstruction were the sigmoid colon and rectum. A total of 58
patients (29 %) had concomitant metastatic disease. No dif-
ference was found between race, sex, and sex per race in
patients with concurrent metastatic disease (p = 0.227,
p = 0.415, p = 0.798, respectively). Of the 203 patients, 128
(63 %) were managed by resection, 37 (18 %) by colonic
stenting, 35 (17 %) by colostomy, and 3 (2 %) by colonic
bypass. Stenting was unsuccessful in six patients.
Z. Moolla (&)
T. E. Madiba
Department of Surgery, Nelson R. Mandela School of Medicine,
University of KwaZulu-Natal, Private Bag X7, Congella,
Durban, KwaZulu-Natal 4013, South Africa
e-mail: drzmoolla@hotmail.co.za
Conclusion Tumor location of patients presenting with
obstruction is comparable to that cited in international lit-
erature; however, the age of presentation among Black
patients is more than a decade earlier than in other ethnic
groups. Surgical management should be individualized.
Stenting remains a reliable alternative in select cases.
Introduction
Colorectal cancer (CRC) is the most common cause of
large bowel obstruction worldwide [1]. Patients requiring
emergency surgery for an obstructing CRC have increased
peri-operative morbidity and mortality. [2] Several factors,
such as old age, comorbid medical conditions, and dehy-
dration, have also been associated with poorer outcomes.
[2] Furthermore, the 5-year survival of patients with
obstructing CRC after curative resection is lower than
patients with non-obstructing carcinomas. [3]
Reported statistics of the frequency of patients pre-
senting with obstructing CRC in first-world countries range
between 8 and 21 % [4–6]. Inevitable differences in the
definition of obstruction as well as selection bias in first-
world referral centres makes interpretation of the true
population-based frequency of obstructing CRC more dif-
ficult. The incidence of CRC in South Africa is increasing;
however, details specific to patients with obstructing
lesions are deficient. It is well established that CRC among
Black patients has a variable clinicopathologic spectrum
and an earlier age of presentation/diagnosis when com-
pared with the Indian, White, and Colored population of
Southern Africa [7].
With the diverse ethnic population in South Africa,
details specific to obstructing CRC in the various ethnic
groups, including their management, is lacking. A variety
123

of options are available in treating colonic obstruction;
however, the optimal procedure remains controversial due
to the lack of prospective randomized trials. Obstructing
lesions are occasionally managed by colonic stenting, but
the practicality and success of this approach in the South
African setting has not been measured.
Our aim was to determine the frequency, demographic
profile, site, stage at presentation, and management of
patients presenting with obstructing CRC in KwaZulu-
Natal (KZN).
Methods
We collected data prospectively on 1,425 consecutive
patients with CRC referred from regional hospitals in the
province of KZN to Inkosi Albert Luthuli Central Hospital
for oncologic and surgical opinion between January 2000
and December 2012. Patients referred after emergency
surgery for acute large bowel obstruction were identified,
and data regarding their age, sex, race, tumor distribution,
metastatic status, and operative management were extrac-
ted and reviewed retrospectively. Inkosi Albert Luthuli
Central Hospital is the only hospital in KZN to offer ter-
tiary and quaternary services to a population of approxi-
mately 10 million people.
The diagnosis of bowel obstruction was made by the
attending surgeon based on the pre-operative history,
clinical, hematologic, and radiologic findings. Surgery was
performed within 24 h of admission, and operative inter-
ventions were decided by the attending surgeon. Patients
with an obstructing CRC proximal to the splenic flexure
were recorded as right-sided obstruction and those with
lesions distal and including the splenic flexure were
recorded as left-sided obstruction. Patients who required
emergency surgery for reasons other than obstruction, such
as bleeding or perforation, were excluded.
Data analysis
Data were entered in MicrosoftÒ Excel and analyzed using
SPSS version 21(IBM, Armonk, NY, USA). [8] Descriptive
Table 1 Demographic data and Race All pts with
study population CRC
Black 489
White 297
CRC colorectal cancer, pts Indian 582
patients, SD standard deviation
a Coloredb 57
Mean ± SD years
b Overall/total 1,425
Mixed racial descent
123
World J Surg
statistics, such as frequencies, percentages, means, and
standard deviations, were used to summarize data. The
Pearson chi-squared test was used to test for association
between categorical variables. Multiple logistic regressions
were used to test for factors associated with resection,
diversion, and stent. The level of significance was set at
0.05.
Ethical approval for the database was obtained from the
Biomedical Research Ethics Committee (E198/04).
Results
A total of 203 patients with malignant obstruction from
CRC were identified from the database of 1,425 patients.
Demographic data are shown in Table 1. Black patients
presented at an earlier age than did other race groups
(Blacks 50 ± 17 vs. Whites 64 ± 13, Indian 60 ± 12, and
Colored 61 ± 12 [p \ 0.001]). No statistical difference in
age was identified among the non-Black population
(p [ 0.05). The presenting age of males and females did
not differ between races except in White patients, in whom
female patients presented earlier than male patients
(59 ± 15 vs. 67 ± 10 [p = 0.039]). A total of 58 (29 %)
patients with obstructing CRC had concurrent metastatic
disease at presentation. No difference was found between
race, sex, and sex per race in patients with concurrent
metastatic disease (p = 0.227, p = 0.415, p = 0.798,
respectively) (Table2). Left-sided obstruction was more
common than right-sided obstruction among all race
groups (p = 0.794) (Table 3).
The treatment of obstructing CRC is shown in Fig. 1. Of
the 203 patients, 166 (82 %) were managed operatively by
either resection, diversion, or bypass; 37 (18 %) were man-
aged with stent placement. Palliative stenting was performed
in 30 patients, 21 for metastatic disease and nine for irre-
sectable locally advanced tumors. One patient with an
obstructing rectal cancer was stented prior to neoadjuvant
therapy and subsequent elective surgery. Of the six unsuc-
cessful attempts at stent placement, five were due to technical
difficulties and one to colonic perforation. All six patients
who were not stented successfully were subsequently man-
aged operatively. One patient experienced stent migration
Pts with obstructive Percentage Agea
CRC (male/female) obstruction
Males Females Combined
68 (41/27) 13.9 49 ± 17 52 ± 16 50 ± 17
50 (28/22) 16.8 67 ± 10 59 ± 15 64 ± 13
72 (41/31) 12.4 60 ± 13 60 ± 10 60 ± 12
13 (6/7) 23 64 ± 15 59 ± 9 61 ± 12
203 (116/87) 14.3 58 ± 16 57 ± 14 59 ± 15
World J Surg

Table 2 Metastatic status at presentation urgent operative intervention. Aslar et al [9]. defined
Race Male Female Overall
obstruction as the combination of absolute absence of flatus
or stool for at least 24 h, abdominal distension, and the
Black 9 (22) 10 (37) 19/68 (28) presence of dilated colon on radiography. As patient
White 6 (21) 5 (23) 11/51 (22) information concerning symptoms at the time of presen-
Indian 15 (37) 13 (42) 28/74 (38) tation may be vague and the clinical signs subjective, we
Colored 2 (33) 1 (14) 3/13 (23) defined obstruction as colorectal disease requiring emer-
Total 32 [28] 29 [33] 58/203 (29) gency operative decompression within 24 h, thereby
Data are presented as n (%), where data within () represent % within a altering the conventional elective management of CRC.
specific race group and data within [] represent % within sex with The proportion of patients presenting with obstructing
obstructing colorectal cancer CRC differs worldwide. Accepted as a surgical emergency,
patients with complete large bowel obstruction are often
complications post-stenting. Multivariate analysis (Table 4) managed at district hospitals. Factors such as the hospital
identified that patients younger than 60 years (p = 0.043) location and the clinician’s definition of obstruction may in
and patients with right-sided tumors (p \ 0.001) were more part explain the wide range in incidence (8–21 %) reported
likely to be managed by resection. In contrast, left-sided in the literature. [3–6] The proportion of patients presenting
lesions (p = 0.003) and older patients were more likely to be with obstructing CRC in our study fell within the range of
managed by stent placement. published series and did not differ between races nor sexes.
Two large cohorts totaling more than 2,100 patients with
malignant bowel obstruction in the UK and France
Discussion revealed a mean age of presentation to be in the seventh
decade of life. [4, 10] In contrast, our study showed an
The definition of large bowel obstruction varies widely important difference in our geographic area by race. Black
across the published literature, and no universally accepted patients in our study from South Africa presented at a mean
description exists to specify the degree of obstruction that age of 50 years, nearly a full two decades earlier than
requires emergency surgery. Terms such as sub-acute patients from developed countries and a decade earlier than
obstruction, partial obstruction, and sub-occlusion further other local racial groups in South Africa. The earlier age of
add to misunderstanding as to the need for emergent or presentation of CRC in young Black patients has been

Table 3 Site of obstruction

Right-sided CRC Left-sided CRC
Cecum Ascending Hepatic Transverse Splenic Descending Sigmoid Rectum Multiple
colon flexure colon flexure colon colon primary

Black 5 (7) 4 (6) 3 (4) 4 (6) 5 (7) 10 (15) 19 (28) 16 (24) 2 (3)
White 2 (4) 6 (12) 4 (8) 2 (4) 0 2 (4) 15 (30) 19 (38)
Indian 10 (14) 6 (8) 1 (1) 3 (4) 2 (3) 4 (6) 20 (28) 26 (36)
Colored 1 (8) 0 1 (8) 0 0 0 6 (46) 5 (38)
Overall 18 [9] 16 [8] 9 [4] 9 [4] 7 [3] 16 [8] 60 [30] 66 [33] 2 [1]
Data are presented as n (%) or n [%], where data within [] represent % within total study cohort and data within () represent % within specific
race group
CRC colorectal cancer

Fig. 1 Treatment of obstructing N = 203

colorectal cancers

Surgery Stent
N = 166 N = 37

Resection Diversion Bypass Successful Unsuccessful

N = 128 N = 35 N=3 N = 31 N=6

123
 World J Surg

Table 4 Factors associated Resection Diversion Stent

with treatment of obstructing
colorectal cancers OR (95 % CI) p value OR (95 % CI) p value OR (95 % CI) p value

Age \60 ya 1.99 (1.02–3.88) 0.043 1.08 (0.46–2.50) 0.862 0.38 (0.17–0.83) 0.160
CI confidence interval, OR odds Male sexb 1.15 (0.60–2.18) 0.675 0.73 (0.32–1.65) 0.452 1.03 (0.49–2.18) 0.939
ratio Racec
a
Compared with age Black 0.50 (0.21–1.22) 0.126 2.11 (0.73–6.10) 0.167 0.52 (0.18–1.58) 0.254
C60 years Indian 0.66 (0.28–1.56) 0.349 0.74 (0.24–2.30) 0.607 1.55 (0.62–3.86) 0.346
b
Compared with females Colored 1.26 (0.28–5.62) 0.758 0.00 (0.00–0.00) 0.999 1.25 (0.27–5.75) 0.779
c
Compared with Whites Locationd
d
Compared with left-sided Right 11.6 (3.44–39.17) 0.001 0.00 (0.00–0.00) 0.997 0.11 (0.03–0.48) 0.003
cancers

identified to occur primarily through a hereditary cause due
to a mismatch repair (MMR) deficiency caused by pro-
moter methylation. [11] MMR deficiency in low-incidence
areas in South Africa was calculated to be approximately
three times higher than other countries with a high inci-
dence of CRC. [12] Furthermore, healthcare facilities have
become more accessible for all race groups in South
Africa, particularly Black patients in rural areas, allowing
for both young and elderly patients to be included in
healthcare databases.
Although some studies demonstrated no difference
between the incidence of right- and left-sided obstruction
from CRC, [13, 14] others document the left colon to be the
more common site of obstruction. [2, 15, 16] This latter
finding is in keeping with all race groups in our study.
Opinion suggesting that the splenic flexure is at greatest
risk of obstruction due to acute angulation[2] was not
confirmed in our study; the most common site of obstruc-
tion was the sigmoid colon (27 %), with only seven
patients obstructing at the splenic flexure.
Several prior studies suggested that obstructing CRC are
associated with a greater incidence of distant metastases,
[6, 17] and that metastatic disease on presentation is
associated with poorer outcomes. [18] The incidence of
distant metastases in our study was in keeping with other
studies [5, 19].
The optimal management of obstructing CRC remains
controversial due to the paucity of randomized controlled
studies. For obstructing colonic cancers, resection and
diversion has been the mainstay of treatment in the
unprepared colon, particularly for left-sided obstruction or
in patients with peritonitis. Recent studies have demon-
strated the potential safety of resection and primary anas-
tomosis for right-sided lesions in unprepared bowel,
although this approach remains controversial for left-sided
tumors. [20, 21] In contrast, obstructing rectal cancers are
often managed by initial creation of a diverting stoma in
the emergent setting followed by a later elective resection
if amenable. Self-expanding metallic stents (SEMS) were
initially described in case reports in the 1990s but have
been used much more commonly in recent times for the
initial relief of acute obstruction, either as definitive pal-
liation or as a bridge to elective resection. In a systematic
review of 88 studies of patients with CRC, 1,785 patients
were managed with SEMS in the acute setting, with a
median clinical success rate for relieving obstruction of
92 % and a perforation rate of 4.5 % [22]. Comparatively,
the overall success rate in our study was 84 %, with a
perforation rate of 3 %. Failure of placement of SEMS in
five of the six patients in this study was due to inability by
the endoscopic operator to pass the guidewire across the
stenosis.
Our study has several limitations. First, ours was a ret-
rospective observational study. The definition of bowel
obstruction is not universally agreed upon; however, all
patients in our study underwent operative intervention
within 24 h of presentation and thus we maintain that all
patients had a clinically relevant symptomatic mechanical
obstruction. Although our study describes the present-day
management of obstructing CRC in a non-controlled, real-
world setting in both urban and rural South Africa, clini-
cian preference in the operative management of patients
was unavoidable. Due to shortcomings in the medical
record keeping in South Africa, and the lack of organized
healthcare systems in some areas, we were unable to cap-
ture reliable post-operative and follow-up data regarding
operative complications, mortality, disease recurrence, and
survival.
Finally, the earlier presentation of obstructing CRC in
Black patients highlights the need for more research into this
disease in developing countries. In the absence of random-
ized controlled trials, intra-operative management will
remain at the discretion of the operating surgeon. Stenting at
centers with experience is an attractive alternative.
Acknowledgment The authors would like thank Professors Michael
G. Sarr and S.R. Thomson for their correspondence and valuable
contribution. No grant support was received for the research.

123
World J Surg
Conflicts of interest No potential or real conflicts of interest exist
References
1. Chen HS, Sheen-Chen SM (2000) Obstruction and perforation in
colorectal adenocarcinoma: an analysis of prognosis and current
trends. Surgery 127:370–376
2. Phillips R, Hittinger R, Fry J et al (1985) Malignant large bowel
obstruction. Br J Surg 72:296–302
3. Ohman U (1982) Prognosis in patients with obstructing colorectal
carcinoma. Am J Surg 143:742–747
4. Cheynel N, Cortet M, Lepage C et al (2007) Trends in the fre-
quency and management of obstructing colorectal cance in a
well-defined population. Dis Colon Rectum 50:1568–1575
5. Tentes AA, Mirelis CG, Kakoliris S et al (2009) Results of sur-
gery for colorectal carcinoma with obstruction. Langenbecks
Arch Surg 394(1):49–53
6. Umpleby HC, Williamson RC (1984) Survival in acute
obstructing colorectal carcinoma. Dis Colon Rectum 27:299–304
7. Zulu B, Madiba T (2011) Colorectal cancer in KwaZulu-Natal: an
established disease with a variable clinicopathological spectrum.
S Afr J Surg 49(2):84–87
8. PASW Statistics 21 (2009) Rel. 18.0.0. Chicago: SPSS Inc
9. Aslar AK, Ozdemir S, Mahmoudi H et al (2011) Analysis of 230
cases of emergent surgery for obstructing colon cancer—lessons
learned. J Gastrointest Surg 15:110–119
10. Tekkis PP, Kinsman R, Thompson MR et al (2004) The Asso-
ciation of Coloproctology of Great Britain and Ireland study of
large bowel obstruction caused by colorectal cancer. Ann Surg
240:76–81
11. Cronje L, Paterson AC, Becker PJ (2009) Colorectal cancer in
South Africa: a heritable cause suspected in many young black
patients. S Afr Med J 99(2):103–106
12. Vergouwe F, Boutall D, Stupart D et al (2013) Mismatch repair
deficiency in colorectal cancer patients in a low-incidence area.
S Afr J Surg 51(1):16–21
13. Serpell J, McDermott F, Katrivessis H et al (1989) Obstructing
carcinomas of the colon. Br J Surg 76:965–969
14. Lee Y, Law W, Chu K et al (2001) Emergency surgery for
obstructing colorectal cancers: a comparison between right-sided
and left-sided lesions. Am Coll Surg 192:719–725
15. Runkel N, Hinz U, Lehnert T et al (1998) Improvement outcome
after emergency surgery for cancer of the large intestine. Br J
Surg 85:1260–1265
16. Alvarez J, Baldonedo R, Bear I et al (2005) Obstructing colo-
rectal carcinoma: outcome and risk factors for morbidity and
mortality. Dig Surg 22:174–181
17. Smothers L, Hynan L, Fleming J et al (2003) Emergency surgery
for colon carcinoma. Dis Colon Rectum 46:24–30
18. Wang H, Lin J, Mou C et al (2004) Long-term prognosis of
patients with obstructing carcinoma of the right colon. Am J Surg
187:497–500
19. Waldron RP, Donovan IA (1986) Mortality in patients with
obstructing colorectal cancer. Ann R Coll Surg 68(4):219–221
20. Ram E, Sherman Y, Weil R et al (2005) Is mechanical bowel prep
mandatory for elective colon surgery? A prospective randomized
study. Arch Surg 140(3):285–288
21. Jung B, Pahlman L, Nystrom PO et al (2007) Multicentre ran-
domized clinical trial of mechanical bowel preparation in elective
colonic resection. Br J Surg 94(6):689–695
22. Watt AM, Faragher IG, Griffin TT et al (2007) Self-expanding
metallic stents for relieving malignant colorectal obstruction: a
systematic review. Ann Surg 246(1):24–30
123