AJR 2005; 184:1194-1198
© American Roentgen Ray Society
Vascular Virtual Endoluminal Visualization of Invasive Colorectal Cancer on MDCT Colonography
Gen Iinuma1,
Noriyuki Moriyama1,
Mitsuo Satake2,
Kunihisa Miyakawa2,
Ukihide Tateishi2,
Nachiko Uchiyama1,
Takayuki Akasu3,
Takahiro Fujii4 and
Toshiaki Kobayashi5
1 Cancer Screening Division, Research Center for Cancer Prevention and
Screening, National Cancer Center, 5-1-1 Tsukiji, Chuo-ku, Tokyo 104-0045,
Japan.
2 Diagnostic Radiology Division, National Cancer Center Hospital, Chuo-ku,
Tokyo, Japan.
3 Colorectal Surgery Division, National Cancer Center Hospital, Chuo-ku, Tokyo,
Japan.
4 Endoscopy Division, National Cancer Center Hospital, Chuo-ku, Tokyo,
Japan.
5 Cancer Screening Technology Division, Research Center Cancer for Prevention
and Screening, National Cancer Center, Chuo-ku, Tokyo, Japan.
Received October 22, 2003;
accepted after revision August 10, 2004.
Supported in part by grants for Scientific Research Expenses for Health and
Welfare Programs and the Foundation for the Promotion of Cancer Research and
by the third-term Comprehensive 10-Year Strategy for Cancer Control from the
Ministry of Health, Labor and Welfare.
Address correspondence to G. Iinuma.
Abstract
OBJECTIVE. The purpose of this study was to assess the utility of
vascular views for visualization of invasive colorectal cancers on
contrast-enhanced MDCT colonography.
CONCLUSION. By means of Hounsfield-transparency settings, we
obtained virtual endoluminal images that show vascular structures and
delineate invasive cancers of the colorectal wall, and we call these images
"vascular views." Using this technique for contrast-enhanced MDCT
colonography, we found that the increase in flow and pooling of blood related
to angiogenesis of cancerous lesions is easy to identify and that this finding
is useful in the detection of invasive colorectal cancers.
Introduction
CT colonography, a technique for visualizing colorectal lesions
using 3D volumetric data generated by helical CT, has developed rapidly over
the past several years [1,
2]. This method has been
reported to be useful for improving the diagnosis of colonic polyps and is now
being considered for colorectal cancer screening in the United States
[3,
4]. This potential has been
markedly enhanced by the advent of MDCT, which allows acquisition of entire
images of the colorectum during a single breath-hold
[5]. A major merit of MDCT is
its high acquisition speed that can be used to cover large volumes with thin
collimation, resulting in good spatial resolution and reduction of the partial
volume effect artifact [6]. The
thinness of the reconstructed axial CT slices has allowed an increase in the
image quality of CT colonography to depict colonic tumors more accurately.
Furthermore, in contrast-enhanced studies with MDCT, the ability to scan
through the entire abdomen in 20 sec or less means that data for the whole
colon can be acquired within the time generally regarded as the
arterial-dominant phase.
Detection of lesions on CT depends on lesion size, slice thickness, and
contrast differentiation [7].
By means of Hounsfield-transparency settings, we obtained virtual endoluminal
images that show vascular structures and delineate invasive cancers of the
colorectal wall, called "vascular views," on contrast-enhanced
MDCT colonography. Using this technique, we found that the increase in flow
and pooling of blood related to angiogenesis of cancerous lesions is easy to
identify and that this is useful in the detection of invasive colorectal
cancers.
The purpose of this study was to assess the utility of vascular views for
the visualization of invasive colorectal cancers on contrast-enhanced MDCT
colonography.
Materials and Methods
From January to March 2002, 28 consecutive patients presenting with 30
invasive colorectal carcinomas underwent contrast-enhanced MDCT examinations
at our hospital for preoperative staging. The series included 15 men and 13
women, ranging in age from 37 to 77 years (median, 60 years). Of these
patients, 22 (78.6%) underwent MDCT after preoperative colonoscopic
examinations with standard bowel preparation of up to 3 L of a polyethylene
glycol-electrolyte solution, and the remaining six patients (21.4%) with
advanced colorectal carcinomas underwent MDCT without preparation. Patients
with rectal cancers underwent MDCT in the prone position, whereas a supine
position was used for those with colon cancers. Before treatment, patients
received muscular injection of anticholinergic drugs, and room-air
insufflation via the anus was performed just before each scan.
Pathologic diagnosis with endoscopic biopsy or surgically resected
specimens was confirmed in each case. All colonic tumors had been initially
diagnosed at colonoscopy, and the presence and site of the lesion were known
at the time of the CT examination.
CT colonography was performed on an MDCT scanner (Aquilion, Toshiba Medical
Systems). The scans were obtained through the abdomen and pelvis with the
following parameters: 120 kV, 250-350 mA with automatic exposure control
[8], 4 rows x 2-mm
collimation, and helical pitch of 5 (pitch factor, 1.25). All patients
received an IV bolus injection of 150 mL of iohexol 350 (Omnipaque, Daiichi
Pharmaceutical) with a power injector at a rate of 3 mL/sec through a 20-gauge
plastic IV catheter placed in an antecubital vein, and the whole abdomen was
scanned 50 sec after this introduction of contrast material during the
arterial phase. All images were reconstructed at a thickness of 1 mm, and the
slices were transferred to an image workstation (M900/Pegasus, AMIN) for
generation of 3D images of each patient.
We used virtual endoluminal images obtained with Hounsfield-transparency
settings in MDCT colonography to show a surface or vascular view of the
colorectal wall on a videotape monitor (Figs.
1A,
1B,
1C,
2A,
2B,
2C,
2D,
3A,
3B,
3C,
3D,
4A,
4B,
4C,
4D,
5A,
5B,
5C, and
5D). Hounsfield-transparency
settings are based on Hounsfield units, which are the CT attenuation values.
First, we adjusted the CT monitor's transparency and opacity setting to a
value of 1 to display only the contour of the lumen and the mucosa. Next, we
adjusted the transparency and opacity setting to a value of 2 to display only
the arterial-dominant blood with contrast medium. Third, we adjusted the
spatial parameters to display only to a depth of 3 mm surrounding the lumen
and the mucosa, which corresponds to the thickness of the intestinal wall.
Fourth, we overlaid the data displayed in steps one through three to produce a
surface and vascular view of the colorectal wall, and then we reduced the
surface opacity to produce an unobstructed vascular view.
View larger version (169K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1A. —Colonoscopic view and surface and vascular virtual
endoluminal images for representative case of advanced colorectal cancer in
60-year-old woman. Colonoscopic view shows advanced cancer in sigmoid
colon.
|
|
View larger version (165K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1B. —Colonoscopic view and surface and vascular virtual
endoluminal images for representative case of advanced colorectal cancer in
60-year-old woman. Surface virtual endoluminal image shows lesion.
|
|
View larger version (137K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1C. —Colonoscopic view and surface and vascular virtual
endoluminal images for representative case of advanced colorectal cancer in
60-year-old woman. Vascular virtual endoluminal image clearly shows blood
pooling of tumor and vessels (arrow) in colorectal wall.
|
|
View larger version (139K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3C. —50-year-old man with colorectal cancer who underwent MDCT
after colonoscopy. Vascular virtual endoluminal image clearly depicts blood
pooling and small vessels (arrows) in colorectal wall.
|
|
View larger version (139K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4C. —59-year-old man with colorectal cancer who underwent MDCT
without preparation. Vascular virtual endoluminal image successfully shows
lesion as mass having blood pooling in colorectal wall.
|
|
The workstation was also equipped with navigation software for virtual
colonoscopy, and the two types of virtual endoluminal images were displayed on
the monitor. Two radiologists retrospectively evaluated primary lesions using
the virtual endoluminal images with or without the Hounsfield-transparency
settings—first, with a conventional surface view and then with a
vascular view. Consensus interpretations were rated against all clinical
information, including the results of colonoscopy; pathologic findings from
biopsy and surgically removed specimens served as the gold standard.
Results
In the 28 patients, a total of 30 invasive carcinomas were confirmed by the
preoperative colonoscopic examinations. Of the 30 lesions, 18 were in the
rectum, five in the sigmoid colon, four in the transverse colon, and three in
the ascending colon. The number of lesions over 2 cm in diameter was 21
(70.0%). Of the total, 19 (63.3%) were well differentiated and 11 (36.7%) were
moderately differentiated on histologic diagnosis.
Lesions showing invasion limited to the submucosal layer were defined as
early invasive colorectal cancer, whereas invasion farther than the submucosal
layer was characterized as advanced colorectal cancer. Among the 30 lesions,
23 (76.7%) were advanced colorectal cancer lesions and seven (23.3%) were
early invasive colorectal cancer lesions. Invasive lesions larger than 2 cm
are generally of more advanced stage, but four (44.4%) of nine small lesions,
2 cm or smaller, were found to be advanced colorectal cancer.
Of the 30 confirmed cancerous lesions, 22 were revealed on conventional
surface virtual endoluminal images, whereas 28 could be identified with
vascular views (Table 1). The
respective figures for lesions 2 cm or smaller were 44.4% (4/9) and 77.8%
(7/9). Of lesions larger than 2 cm, three (14.3%) of 21 were missed on surface
virtual endoluminal images, but all could be visualized on the vascular views.
Invasive lesions larger than 2 cm are generally considered to have high
potential for malignancy. However, even with the small lesions (
2 cm),
almost half were advanced colorectal cancers, so the use of the vascular
approach allowed identification of most lesions that should be treated as a
high priority (Table 1).
View this table:
[in this window]
[in a new window]
|
TABLE 1 Detection of Colorectal Lesions Using Conventional Surface Versus
Vascular Views for Virtual Endoluminal Imaging with Lesions Categorized by
Size
|
|
Of the 30 lesions, three of the seven early invasive colorectal cancer
lesions were revealed on conventional surface virtual endoluminal images,
whereas five of seven could be identified with vascular imaging. All 17
advanced colorectal cancer lesions in cases with preparation could be
recognized on the surface and vascular virtual endoluminal images. This
finding is especially noteworthy because among six advanced colorectal cancer
lesions in patients without preparation, four (66.7%) were missed with the
conventional surface approach, but all could be visualized on the vascular
virtual endoluminal images (Table
2).
View this table:
[in this window]
[in a new window]
|
TABLE 2 Detection of Colorectal Lesions Using Conventional Surface Versus
Vascular Views for Virtual Endoluminal Imaging with Lesions Categorized by
Severity of Invasion
|
|
Discussion
Amin et al. [9] first
described the merits of dynamic contrast-enhanced CT study with the
air-insufflation technique for the detection of colorectal cancers.
Subsequently, the same group reported the value of contrast-enhanced CT
colonography for the improvement of colorectal polyp detection
[10]. With contrast-enhanced
CT studies, the advent of MDCT allows acquisition of images of the entire
abdomen during a single breath-hold, which is regarded as the
arterial-dominant phase. The resulting thinner-collimated transverse images
with blood flow information provide better-quality MDCT colonographic data
than conventional CT, and these data should further increase the ability to
detect not only colonic polyps but also invasive lesions more accurately. In
addition, we can manipulate the 3D volumetric data on an image workstation
with navigation software for virtual endoscopy or with various display modes
including Hounsfield-transparency settings, such as the vascular views, to
show information about the blood flow within and around the colorectal
wall.
With conventional surface virtual endoluminal images of CT colonography, a
surface is just that—a surface. However, as shown in this study, pooling
of blood related to angiogenesis of invasive cancers and small vessels of the
colorectal wall can be more clearly visualized with vascular views of within
the colorectal wall. With the introduction of 16-MDCT scanners, the image
quality of virtual endoluminal images is expected to improve even further;
therefore, vascular views are going to be more and more in demand not only by
radiologists and gastroenterologists, but also by patients who, we believe,
will be happy that vascular views require no preparatory fasting, because
vascular views are not confused by the absence or presence of stool.
Vascular views also have a great potential for using blood flow information
to detect small invasive cancers with computer-aided diagnosis, which is
expected to improve radiologists' and gastroenterologists' diagnostic
performance enormously [11,
12]. We therefore believe that
a focus on the blood supply with the vascular views should be used in
conjunction with conventional surface virtual endoluminal images whenever
diagnostic or screening contrast-enhanced MDCT is performed until safer
contrast media are developed.
References
- McFarland EG, Brink JA, Loh J, et al. Visualization of colorectal
polyps with spiral CT colography: evaluation of processing parameters with
perspective volume rendering. Radiology1997; 205:701
-707[Abstract/Free Full Text]
- Hara AK, Johnson CD, Reed JE, et al. Detection of colorectal polyps
with CT colography: initial assessment of sensitivity and specificity.
Radiology1997; 205:59
-65[Abstract/Free Full Text]
- Miao YM, Amin Z, Healy J, et al. A prospective single centre study
comparing computed tomography pneumocolon against colonoscopy in the detection
of colorectal neoplasms. Gut2000; 47:832
-837[Abstract/Free Full Text]
- Johnson CD, Dachman AH. CT colonography: the next colon screening
examination? Radiology2000; 216:331
-341[Abstract/Free Full Text]
- Berland LL, Smith JK. Multidetector-array CT: once again,
technology creates new opportunities. Radiology1998; 209:327
-329[Free Full Text]
- McCollough CH, Zink FE. Performance evaluation of a multi-slice CT
system. Med Phys1999; 26:2223
-2230[Medline]
- Plewes DB, Dean PB. Detectability of spherical objects by computed
tonography. Radiology1979; 133:785
-786[Abstract]
- Itoh S, Ikeda M, Mori Y, et al. Lung: feasibility of a method for
changing tube current during low-dose helical CT.
Radiology2002; 224:905
-912[Abstract/Free Full Text]
- Amin Z, Boulos PB, Lees WR. Technical report: spiral CT pneumocolon
for suspected colonic neoplasms. Clin Radiol1996; 51:56
-61[Medline]
- Harvey CJ, Amin Z, Hare CMB, et al. Helical CT pneumocolon to
assess colonic tumors: radiologic-pathologic correlation.
AJR 1998;170:1439
-1443[Abstract/Free Full Text]
- Yoshida H, Masutani Y, MacEneaney P, Rubin DT, Dachman AH.
Computerized detection of colonic polyps at CT colonography on the basis of
volumetric features: pilot study. Radiology2002; 222:327
-336[Abstract/Free Full Text]
- Summers RM, Johnson CD, Pusanik LM, Malley JD, Youssef AM, Reed JE.
Automated polyp detection at CT colonography: feasibility assessment in a
human population. Radiology2001; 219:51
-59[Abstract/Free Full Text]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Top
Abstract
Introduction
