AJR 2004; 182:905-909
© American Roentgen Ray Society
Frequency and CT Patterns of Bowel Wall Thickening Proximal to Cancer of the Colon
Lin Xiong1,
Kedar N. Chintapalli1,
Gerald D. Dodd, III1,
Shailendra Chopra1,
Joe A. Pastrano1,
Cheryl Hill2,
John R. Leyendecker1,
Robert M. Abbott3,
David Grayson3 and
John Feig4
1 Department of Radiology, University of Texas Health Science Center at San
Antonio, 7703 Floyd Curl Dr., San Antonio, TX 78229-3900.
2 Department of Pathology, University of Texas Health Science Center at San
Antonio, San Antonio, TX 78229-3900.
3 Department of Radiology, Wilford Hall Medical Center, 759 MDTS/MTR, 2200
Bergquist Dr., Ste. 1, Lackland AFB, TX 78236-5300.
4 Department of Pathology/MTLP, Wilford Hall Medical Center, Lackland AFB, TX,
78236.
Received July 23, 2003;
accepted after revision October 7, 2003.
Address correspondence to K. N. Chintapalli.
The opinions and assertions contained herein are the private views of the
authors and are not to be construed as official or as representing the views
of the Department of the Air Force.
Abstract
OBJECTIVE. This study was performed to determine the frequency and
patterns of colon wall thickening seen on CT of patients with adenocarcinoma
of the colon.
MATERIALS AND METHODS. Preoperative abdominal and pelvic CT scans of
185 patients with surgically proven adenocarcinoma of the colon were
retrospectively evaluated by three abdominal radiologists for the presence of
colon obstruction and colon wall thickening proximal to the colon
adenocarcinoma. The distributions and patterns of colon wall thickening were
categorized by consensus. CT findings were compared with pathologic findings.
Fisher's exact test was used to determine the statistical significance of any
associations.
RESULTS. Of 185 patients, CT findings of 20 (10.8%) showed colon
wall thickening. Of these, the adenocarcinoma obstructed the colon in 19
patients (p < 0.01). Colon obstruction was partial in 10 patients
(53%) and complete in nine (47%). Colon wall thickening was contiguous to the
tumor in 14 (70%) patients and noncontiguous in six (30%). Segmental and
pancolonic, patchy and diffuse, and dependent and nondependent colon wall
thickening was observed in 10 patients (50%) in each category. Associated
small-bowel wall thickening was shown in 10 (50%) of the 20 patients.
Pathologic examination showed colon wall thickening to be due to edema in all
cases.
CONCLUSION. Colon wall edema can occur proximal to colon
adenocarcinoma, is almost always associated with colon obstruction, and is
predominantly contiguous with the obstructing adenocarcinoma.
Introduction
Obstructive colitis, an ulceroinflammatory process of the colon proximal to
an obstructing lesion, is reported to occur in 1–7% of patients with
partial or complete obstruction of the colon due to carcinoma
[1–5].
It is thought to represent ischemia of colon segments or changes in the fecal
bacterial flora proximal to the point of obstruction
[1,
3]. Failure to recognize this
entity at the time of surgery may cause anastomotic complications because of
decreased viability of the colon. Therefore a preoperative diagnosis of
obstructive colitis may be helpful to the surgeon for technical planning
[4]. Ko et al.
[3] describe the CT findings in
20 patients with ischemic colitis; the most common CT finding of colitis was
nonspecific bowel wall thickening. To our knowledge, no other studies
specifically address the frequency and patterns of colon wall thickening in
patients with colon cancer.
Materials and Methods
In this retrospective study, we searched the surgical oncology databases of
four regional teaching facilities and identified 345 consecutive patients who
underwent surgery for adenocarcinoma of the colon between January 1996 and
October 2000. The institutional review board of our institution approved the
plan for this study without the necessity of informed consent. We reviewed the
medical records of these patients and identified 185 for whom preoperative
abdominal and pelvic CT studies were available. Of these patients, 111 (60%)
were men, and 74 (40%) were women (age range, 22–85 years; mean, 59
years). Before CT was performed, the gastrointestinal tract was opacified with
oral and rectal contrast material in 104 patients and oral contrast material
alone in 78 patients. Three patients did not receive any gastrointestinal
contrast material. IV contrast material was administered to 179 patients who
underwent CT. Abdominal CT was performed in the helical mode using a 7- to
8-mm section thickness at a pitch of 1. Pelvic CT was performed in the axial
mode with 5-mm-thick sections at 10-mm intervals.
Three board-certified abdominal radiologists reviewed the CT scans
retrospectively and rendered a consensus opinion on thickening of the colon
wall proximal to the primary tumor; the patterns, presence, and degree of
colon obstruction; primary tumor characteristics; and tumor spread. Any
reviewer conflict was resolved by majority rule.
Following the criteria used in previous studies, we diagnosed colon wall
thickening if the wall thickness was more than just perceptible in a gas- or
fluid-filled segment of colon
[6]. If thickening was present,
its pattern was then described on the basis of its contiguity to the primary
tumor, its distribution along the length of the colon, the completeness of
involvement of the bowel wall, and its distribution along the circumference of
the colon. Wall thickening was called "contiguous"
(Fig. 1F) if it extended up to
the primary tumor and "noncontiguous"
(Fig. 1E) if normal bowel wall
was present between the tumor and the distal margin of the wall thickening.
The distribution along the colon length was called "pancolonic"
(Fig. 1A) if bowel wall
thickening was seen in all segments of the proximal colon and
"segmental" (Fig.
1B) if all segments were not involved. The completeness of the
bowel wall thickening was described as "diffuse"
(Fig. 1C) if it was continuous
throughout the involved segments of the colon and "patchy"
(Fig. 1D) if the areas of
thickening were scattered within the involved segments. Additionally, the
reviewers documented whether pneumatosis was present and whether the bowel
wall thickening was more pronounced in the dependent or nondependent portions
of the bowel.
View larger version (25K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1F. —Drawings illustrate descriptive terms used to characterize
colon wall thickening proximal to colon cancer. "Contiguous"
indicates that wall thickening proximal to obstructing neoplasm reaches up to
neoplasm.
|
|
View larger version (23K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1E. —Drawings illustrate descriptive terms used to characterize
colon wall thickening proximal to colon cancer. "Noncontiguous"
indicates that wall thickening proximal to obstructing neoplasm stops short of
neoplasm.
|
|
View larger version (24K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1A. —Drawings illustrate descriptive terms used to characterize
colon wall thickening proximal to colon cancer. "Pancolonic"
indicates that colon wall thickening involves all segments proximal to
obstructing neoplasm.
|
|
View larger version (26K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1B. —Drawings illustrate descriptive terms used to characterize
colon wall thickening proximal to colon cancer. "Segmental"
indicates colon wall thickening that does not involve all segments proximal to
obstructing neoplasm.
|
|
"Bowel obstruction" was defined as dilatation of the colon
lumen proximal to the primary tumor. Obstruction was described as
"complete" if oral contrast or fecal material was present up to
the tumor but not distal to it or the rectal contrast material failed to cross
the tumor. Obstruction was described as "incomplete" if oral
contrast, rectal contrast, or fecal material was present both proximal and
distal to the tumor.
The primary tumor characteristics evaluated were location, length, and
configuration. Each primary tumor was described as involving the cecum,
ascending colon, transverse colon, descending colon, sigmoid colon, or rectum,
depending on its anatomic location. On the basis of its length, each primary
tumor was assigned to one of the three categories: less than 5 cm, 5–9.9
cm, or 10 cm or greater. The tumors were divided into four categories by
configuration: focal thickening of the colon wall, polypoid, masslike, and
annular.
Areas of potential tumor spread were evaluated for evidence of involvement:
regional lymph nodes and paraaortic lymph nodes larger than 1 cm, liver or
lung lesions characteristic of metastases, peritoneal tumor masses, or
ascites.
The surgical pathology report, slides, and specimens were evaluated for
correlative findings. In patients without adequate pathology data, followup CT
studies were reviewed to evaluate the temporal evolution of the colon
findings.
Frequency of colon wall thickening and its various patterns proximal to the
tumor were determined. We sought any association between the presence of colon
wall thickening and characteristics of primary tumor, presence and degree of
bowel obstruction, and spread of the tumor. Statistical significance of
associations was tested using the Fisher's exact test.
Results
Of 185 patients, CT scans of 20 (10.8%) showed colon wall thickening
proximal to the primary tumor. Wall thickening was contiguous to the tumor in
14 patients (70%) (Fig. 2A,
2B) and noncontiguous in six
(30%). Wall thickening was segmental and pancolonic in 10 patients (50%) each.
Wall thickening was patchy in 10 patients (50%) and diffuse in the remaining
10 (50%). Colon wall thickening was greater in the dependent portions than in
the nondependent portions of the colon in 10 patients (50%), and the dependent
and nondependent portions were equally involved in 10 patients (50%). Three
patients (15%) with colon wall thickening had pneumatosis.
View larger version (114K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2A. —47-year-old woman with obstructing transverse colon cancer.
Axial CT scan of abdomen shows annular neoplasm at distal transverse colon
(arrow). Note wall thickening in transverse colon proximal to tumor
(arrowheads).
|
|
View larger version (73K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2B. —47-year-old woman with obstructing transverse colon cancer.
Axial CT scan of abdomen obtained at lower level than A shows wall
thickening (long arrow) of ascending colon. Note that descending
colon, which is distal to neoplasm, is spared (short arrow).
|
|
Nineteen (95%) of the 20 patients with colon wall thickening and 13 (8%) of
the 165 patients without colon wall thickening had CT findings of colon
obstruction. This difference was statistically significant (p <
0.001). The degree of obstruction in patients with colon wall thickening was
partial in 10 (53%) patients and complete in nine (47%). Conversely, 32 (17%)
of 185 patients had CT findings of colon obstruction. Of these, 19 (59%) had
colon wall thickening, whereas 13 (41%) had normal colon wall thickness. In
addition, 10 (50%) of the 20 patients with colon wall thickening had
associated small-bowel dilatation. Three patients (15%) with colon wall
thickening had pneumatosis. Tumor characteristics in patients with and without
colon wall thickening proximal to colon cancer are summarized in
Table 1.
View this table:
[in this window]
[in a new window]
|
TABLE 1 Frequency of Tumor Characteristics in Patients With and Without Colon
Wall Thickening Proximal to Colon Cancer
|
|
Of the 20 patients with colon wall thickening, 16 (80%) had surgical and
histologic evidence of colon wall edema. Histologically, the edema involved
primarily the mucosa and submucosa. Of these patients, two had histologic
evidence of ischemia. The changes ranged from frank mucosal–submucosal
coagulation necrosis to more chronic ischemic changes including mucin-depleted
epithelium with fibrosis of the lamina propria. Of the remaining four cases
(20%), one had no evidence of ischemia or edema. In two cases, pathology
slides were unavailable for review, and in one case, the interpretation was
difficult because of poor fixation. However, findings of follow-up CT studies
after surgical resection of the primary tumor in those four patients showed
resolution of colon wall edema within 1 month (Fig.
3A,
3B,
3C).
Discussion
Colon wall thickening measuring greater than 3–5 mm is a nonspecific
CT finding. It is measured as the distance from the outer colon wall edge, as
defined by identifying the mesenteric fat–bowel wall interface to the
inner bowel wall edge, which is demarcated by noting the interface between
bowel wall and intestinal gas, feces, or, rarely, diluted contrast material
[7–9].
CT has become more sensitive in detecting bowel wall thickening with reports
of 69% correlation between bowel wall thickening on abdominal CT and findings
at colonoscopy [10]. Colon
wall thickening has a variety of CT appearances ranging from eccentric to
circumferential, with underlying causes including neoplasms and inflammatory
processes such as inflammatory bowel disease, bacterial colitis, and ischemic
colitis [7,
11–16].
The degree of colon wall thickening has been graded by Bharucha et al.
[14] as mild, moderate, and
severe, with colon wall thickness of 3–6 mm, 6–12 mm, and greater
than 12 mm, respectively.
Colon cancer is a common malignancy affecting men and women and is
associated with morbidity and mortality. Ischemia occurring proximal to a
colon adenocarcinoma has been described in the literature with an incidence of
1–7%, and the most common finding on CT is colon wall thickening
[1–5].
We describe the frequency and patterns of proximal colon wall thickening
seen on CT in a group of patients with known colon adenocarcinoma. In our
series, 20 of 185 patients with a colon adenocarcinoma had proximal colon wall
thickening. Moreover, only two of the 16 patients with histologic evidence of
colon wall edema had evidence of obstructive ischemia—mucosal
ulceration, submucosal edema, and hemorrhage. The preponderance of cases (70%)
had proximal colon wall thickening contiguous with the primary tumor.
Our results are somewhat similar to those of Ko et al.
[3] in the frequency of colon
wall thickening and the frequency of non-contiguity. The frequency of colon
wall thickening proximal to the colon cancer was nearly 11% in our study and
6% in the study of Ko et al. Similarly, colon wall thickening was contiguous
to the tumor in 70% and noncontiguous in 30% of our patients, which resembled
the findings of 90% and 10%, respectively, in the series by Ko et al. This
finding of normal bowel lying between the proximal margin of the tumor and the
ischemic segment has been reported
[3]. It has been explained on
the basis of the Laplace law
[4]. Because bowel wall tension
is proportional to both intraluminal pressure and wall diameter, the zone
adjacent to the tumor is somewhat protected from high tension because it is
not as dilated as the proximal segments. In our study and in that of Ko et al.
[3], the frequency of the
appearance of the skip zone is far less compared with that found in earlier
reports from the surgical literature
[2]. This difference may be due
to the degree of the severity of obstruction between series. Although colon
obstruction was present in 95% of patients in our series, it was incomplete in
almost half of these cases. The pressure dynamics of the partially obstructed
colon are likely to be different from those of the completely obstructed colon
and thus cause a relatively low frequency of the appearance of the skip zone.
Although colon wall thickening does not develop in all cases of colon
obstruction, the latter has a major role in its development.
In contrast to Ko et al.
[3], we found a much lower
incidence of true ischemic changes. Only two of 16 patients in whom pathology
slides were available showed pathologic evidence of ischemia. In the remaining
14 patients, only submucosal edema was seen. Among the remaining four
patients, although no histologic reports were available, colon wall thickening
was reversible, as shown by its resolution on CT scans obtained 1 month after
tumor resection. Bowel wall thickening in this setting may represent a
reversible, preischemic CT finding. Therefore the term "colon wall
thickening or edema" may better reflect the nature of this process.
Pneumatosis coli has been suggested in the literature as a correlative
finding with frank bowel ischemia and colon infarction
[3]. Of the three patients in
our study that did have pneumatosis coli, only one had gross and microscopic
histopathologic evidence of ischemia, most notably submucosal edema and
mucosal ulceration and erosion, whereas the other two had no histopathologic
evidence of ischemia.
One general limitation of this retrospective study is that the pathology
reports provided only descriptions of the proximal margins of the tumor in
most cases, while infrequently describing the histopathologic findings in more
proximal areas. Furthermore, because cases were reported from multiple
institutions, variations in CT parameters and pathology reporting are
unavoidable.
In conclusion, our data suggest that colon wall thickening in patients with
colon adenocarcinoma is relatively common, occurring in approximately 11% of
our patients, is closely associated with the presence and not the degree of
obstruction, is commonly contiguous, and may represent a reversible CT finding
heralding ischemia of the colon wall. These observations should be kept in
mind when planning surgery in patients with colon adenocarcinoma.
References
- Ganchrow MI, Clark JE, Benjamin HG. Ischemic colitis proximal to
obstruction carcinoma of the colon. Dis Colon Rectum1971; 14:38
–42[Medline]
- Toner M, Condell D, O'Briain DS. Obstructive colitis:
ulceroinflammatory lesions occurring proximal to colonic obstruction.
Am J Surg Pathol1990; 14:719
–728[Medline]
- Ko GY, Ha HK, Lee HJ, et al. Usefulness of CT in patients with
ischemic colitis proximal to colonic cancer. AJR1997; 168:951
–956[Abstract/Free Full Text]
- Tietjen GW, Markowitz AM. Colitis proximal to obstructing colonic
carcinoma. Arch Surg1975; 110:1133
–1138[Abstract]
- Feldman PS. Ulcerative disease of the colon proximal to partially
obstructive lesions: report of two cases and review of the literature.
Dis Colon Rectum1975; 18:601
–612[Medline]
- Karahan OI, Dodd GD 3rd, Chintapalli KN, Rhim H, Chopra S.
Gastrointestinal wall thickening in patients with cirrhosis: frequency and
patterns at contrast-enhanced CT. Radiology2000
215:103
–107[Abstract/Free Full Text]
- Fisher JK. Abnormal colonic wall thickening on computed tomography.
J Comput Tomogr1983; 7:90
–97
- Taourel PG, Deneuville M, Pradel JA, Regent D, Bruel JM. Acute
mesenteric ischemia: diagnosis with contrast-enhanced CT.
Radiology1996; 199:632
–636[Abstract/Free Full Text]
- Macari M, Balthazar EJ, Megibow AJ. The accordion sign at CT: a
nonspecific finding in patients with colonic edema.
Radiology1999; 211:743
–746[Abstract/Free Full Text]
- Rockey DC, Halvorsen RA, Higgins JL, Cello JP. Prospective
evaluation of patients with bowel wall thickening. Am J
Gastroenterol 1995;90:99
–103[Medline]
- Balthazar EJ, Yen BC, Gordon RB. Ischemic colitis: CT evaluation of
54 cases. Radiology1999; 211:381
–388[Abstract/Free Full Text]
- Rha SE, Ha HK, Lee SH, et al. CT and MR imaging findings of bowel
ischemia from various primary causes. RadioGraphics2000; 20:29
–42[Abstract/Free Full Text]
- Simon AM, Birnbaum BA, Jacobs JE. Isolated infarction of the cecum:
CT findings in two patients. Radiology2000; 214:513
–516[Abstract/Free Full Text]
- Bharucha AE, Tremaine WJ, Johnson CD, Batts KP. Ischemic
proctosigmoiditis. Am J Gastroenterol1996; 91:2305
–2309[Medline]
- Guingrich JA, Kuhlman JE. Colonic wall thickening in patients with
cirrhosis: CT findings and clinical implications. AJR1999; 172:919
–924[Abstract/Free Full Text]
- Gandhi SK, Hanson MM, Vernava AM, Kaminki DL, Longo WE. Ischemic
colitis. Dis Colon Rectum1996; 39:88
–100[Medline]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
|
|
 
V. Goh, S. Halligan, S. A. Taylor, D. Burling, P. Bassett, and C. I. Bartram
Differentiation between Diverticulitis and Colorectal Cancer: Quantitative CT Perfusion Measurements versus Morphologic Criteria--Initial Experience
Radiology,
February 1, 2007;
242(2):
456 - 462.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
