Surg Clin N Am 86 (2006) 819–847

An Update on Imaging of Colorectal

Cancer
Christoph Wald, MD, PhD*,
Christopher D. Scheirey, MD,
Tai M. Tran, MD, Nazli Erbay, MD
Department of Diagnostic Radiology, Lahey Clinic Medical Center,
41 Mall Road, Burlington, MA 01805, USA

For colon cancer, the second leading cause of death from malignancy in
the United States, screening of asymptomatic average-risk patients for the
presence of this disease and early detection in precursor stages is of great in-
terest to the general population [1]. Comprehensive evaluation of symptom-
atic or high-risk patients represents another important clinical focus.
It has been demonstrated that timely recognition and removal of adeno-
matous polyps significantly decreases the risk of death from colorectal
cancer in affected patients [2]. However, compliance with current recom-
mendations of colon cancer screening in the general population remains
low [3]. One can only speculate that compliance rates may improve with
the advent of less onerous or less invasive examinations.
Despite the interest in colon cancer screening from a public health per-
spective, selection of the right time point and methodology for screening
is difficult, because the incidence of cancer in nonadenomatous and small
adenomatous colon polyps in elderly patients is high. The cost and risks
of complications associated with an examination designed to detect clini-
cally significant lesions need to be weighed against the derived survival
benefit.

Total colon examinationdavailable methods

Methods currently available for total colon examination include double-
contrast barium enema (DCBE), endoscopic examination (colonoscopy),

* Corresponding author.
E-mail address: [email protected] (C. Wald).

0039-6109/06/$ - see front matter Ó 2006 Elsevier Inc. All rights reserved.
doi:10.1016/j.suc.2006.06.001 surgical.theclinics.com
820 WALD et al

and more recently, cross-sectional imaging-based technology such as CT

Colonography (CTC), also known as virtual colonoscopy or virtual endos-
copy. The feasibility of magnetic resonance Colonography (MRC) for total
colon assessment is also being evaluated, and will be discussed below.

Double-contrast barium enema

Dating back to the 1920s [4], radiographic examination of the colon with
barium and air has been an important modality. Originally, it was the sole
comprehensive colon examination, which was improved over time. Cur-
rently, it is one of several options. History and technique of the air contrast
examination have been well described, and the reader is referred to a review
article by Rubesin and colleagues [5] for more detailed information. With
the advent and ever increasing use of endoscopic screening, the number of
double-contrast enemas performed in the United States began to decrease
steadily. Over time, this resulted in fewer and fewer radiologists capable
of performing a technically adequate examination. In fact, the authors sus-
pect that only a minority of radiology training programs in this country are
currently able to offer residents sufficient exposure to become proficient in
this interactive live fluoroscopic examination, the results of which are very
examiner dependent. Such a trend was already suspected by Gazelle and col-
leagues [6] in their review of the issues surrounding screening for colorectal
cancer in 2000.
Studies in the past have shown a wide range of diagnostic sensitivity and
accuracy for air-contrast enema. The results vary depending on study pop-
ulation and lesion size. We do not know of any prospective study that
looked at the accuracy of DCBE examination in a screening population.
However, in symptomatic patients the reported sensitivity of double-
contrast enema for the detection of cancer ranges from 85% to 90% [7–9].
Sensitivity for adenomas larger than 1 cm is reported to be 75% to 90%,
whereas sensitivity for the detection of smaller lesions is reported to be
only 50% to 80% [10,11]. Results observed in this particular subgroup of pa-
tients cannot be extrapolated to the screening population. Furthermore,
judging the utility of an examination technique in context with colon cancer
screening should take into account the low rate of progression of polyps
!1-cm diameter into colon cancer. The study by Winawer and colleagues
[12] and the ensuing discussion among experts [13] may serve as an example
of this complex and controversial issue.

Colonoscopy
Although colonoscopy is undoubtedly sensitive, one needs to remember
that its success is also examiner dependent; in 5% to 15% of cases, endo-
scopists may not be able to reach the cecum for technical reasons [14–17].
 AN UPDATE ON IMAGING OF COLORECTAL CANCER 821

Furthermore, there may be endoscopic blind spots due to haustral anatomy

and convoluted arrangement of redundant bowel [18,19]. Nevertheless, co-
lonoscopy offers the undisputed advantage of synchronous diagnostic and
therapeutic intervention. However, this technique requires sedation, and
carries a small but real risk of complications secondary to its invasiveness
[15,17,20,21].

CT Colonography in colon cancer screening

After initial feasibility studies on CTC, the development of multidetector
row CT scanners permitted radiologists to rapidly acquire volumetric data-
sets of patient anatomy at a high spatial resolution. Suitable computer soft-
ware allows reconstruction and viewing of both two-dimensional (2D) and
three-dimensional (3D) image representations of the colon and surrounding
anatomy. CTC is minimally invasive (air or CO2 insufflation per rectum is
required), requires no sedation, and is less time consuming than an endo-
scopic examination. Currently, most centers have their patients undergo ca-
thartic bowel preparation before both endoscopy and virtual colonoscopy.
Individual technical aspects are discussed below.

Trials and reviews

At the time of the writing of this article, there are only two prospective
multicenter trials on CTC in the average-risk asymptomatic screening pop-
ulation, which reported very different results.
In 2003, Pickhardt and colleagues [22] published results of a multicenter
trial intended to assess the performance of CTC for the detection of colorec-
tal neoplasia in an average-risk screening population. A total of 1233
asymptomatic adults (mean age, 57.8 years) underwent same-day CTC
and optical colonoscopy. The sensitivity of virtual colonoscopy for adeno-
matous polyps was 93.8% for polyps at least 10 mm in diameter, 93.9%
for polyps at least 8 mm in diameter, and 88.7% for polyps at least 6 mm
in diameter. The sensitivity of optical colonoscopy for adenomatous polyps
was 87.5%, 91.5%, and 92.3% for the three sizes of polyps, respectively.
The specificity of virtual colonoscopy for adenomatous polyps was 96.0%
for polyps at least 10 mm in diameter, 92.2% for polyps at least 8 mm in
diameter, and 79.6% for polyps at least 6 mm in diameter. Two polyps
were malignant; both were detected on virtual colonoscopy, and one of
them was missed on optical colonoscopy before the results on virtual colo-
noscopy were revealed. The authors concluded that CT virtual colonoscopy
with the use of a 3D approach is an accurate screening method for the de-
tection of colorectal neoplasia in asymptomatic average risk adults, and
compares favorably with optical colonoscopy in terms of the detection of
clinically relevant lesions.
In 2004, Cotton and colleagues [23] published results of a multicenter
trial in an asymptomatic screening population. One hundred four
822 WALD et al

participants had lesions sized at least 6 mm. CTC detected 55.0% (95% con-
fidence interval [CI], 39.9–70.0%) of lesions of at least 10 mm, compared
with 100% for conventional colonoscopy. The specificity of CTC for detect-
ing participants without any lesion was greater than 90%. CTC missed two
of eight cancers. The accuracy of CTC varied considerably between centers,
and did not improve as the study progressed. Authors concluded that CTC
was not yet ready for widespread clinical application, and that techniques
and training need to be improved.
In an attempt to clarify the situation, the American College of Radiology
Imaging Network initiated another large multicenter trial, the results of
which will not be available for another year or 2, as mentioned in a recent
article by Ferrucci [24].
Two other recently published large comparative trials performed on
symptomatic patients [25] and high-risk patients [26] yielded better sensitiv-
ity for colonoscopy than for double-contrast enema or CTC.
One may speculate that the observed differences in the performance of
CTC is related to the differences in sophistication in postprocessing and
interactive review (2D versus 3D review, different software products, use
of stool tagging) and examiner experience, or a combination thereof [25,27].
In further attempts to establish the role of CTC, several authors have
reviewed the literature on this technology. Most studies looked at perfor-
mance of CTC in symptomatic patient cohorts; thus, the results and recom-
mendations cannot simply be extrapolated to the asymptomatic screening
population. Retrospective literature reviews are hampered by the lack of
standards in performance and reporting of CTC trials. In a 2002 Lature
review, Dachman [28] reported a wide range of sensitivity (8–100%) for de-
tection of polyps larger than 1 cm. The author stated that because of wide
technical variation (e.g., different bowel preparations, CT scanners, and in-
terpretation software) used in the studies, meta-analysis of the results would
not provide reliable statistics. Recently, Halligan and colleagues [29]
reviewed a large number of studies performed between 1994 and 2004,
and also found that reporting was highly variable, suggesting minimum
reporting standards to overcome this problem. Based on a meta-analysis,
the authors, Halligan and colleagues, concluded that CTC seemed suffi-
ciently sensitive and specific in the detection of large and medium polyps,
and especially sensitive in the detection of symptomatic lesions.
A US-based large prospective multicenter CTC trial is currently
under way sponsored by ACRIN, the American College of Radiology Im-
aging Network, a National Cancer Institute-funded cooperative group,
the results of which are expected for late 2006 or early 2007. The stated
goal is the clinical validation of widespread use of CTC in a screening pop-
ulation for the detection of colorectal neoplasia in just under 2300 patients.
It is widely expected that the results of this trial will determine whether CTC
will become a reimbursable and recommended technique for colon cancer
screening.
 AN UPDATE ON IMAGING OF COLORECTAL CANCER 823

CT Colonography techniquedtoo many choices?

The controversy about the true achievable sensitivity of CTC in screening
asymptomatic and symptomatic patients results from differences in technical
parameters. Many studies performed in recent years have looked at some of
the variables and important ingredients of an optimal technique.
Retained fluid, liquid, and solid stool particles interfere directly with the
test, either by masking of polyps or by creating false positive findings. Ad-
equate bowel preparation is therefore vital for a highly sensitive study. How-
ever, it is well established that cathartic preparation presents a well-known
barrier to colorectal cancer screening compliance [30]. A study comparing
CTC after noncathartic preparation and with colonoscopy performed 3 to
7 days later yielded an average sensitivity of 95.5% (95% CI, 92.1–99%)
for the identification of colorectal polyps R8 mm [31]. Patients had to com-
ply with a low-residue diet and ingested Gastrografin with meals for fecal
tagging purposes. However, as Pickhardt pointed out in a commentary
[32], this approach is not suitable if same-day colonoscopy/polypectomy is
desired.
The use of laxatives for bowel preparation proved to be a major barrier
for patient compliance [33]. With regard to laxatives, preparations contain-
ing magnesium citrate or sodium phosphate seem to be preferred because
they result in less fluid retention than polyethylene solutions [34,35]. Low-
residue diet in the days before the study is also helpful, and may reduce
the need for aggressive laxative therapy [36]. Labeling stool with oral barium
or iodine-containing substances is an important adjunct technique in dealing
with the issue of fecal residue, avoiding too many false-positive results in
CTC [22,27,37]. The utility of stool tagging is demonstrated in Fig. 1A–D.
Adequate colonic insufflation is another important technical factor; ei-
ther room air or CO2 are applied to the patient. There is no consensus as
to which agent yields better distension [38]. In an article by Barish and col-
leagues [38], many of the above variables of performing CTC in clinical
practice are discussed, moreover, the authors believe that spasmolytics are
not routinely necessary to achieve adequate results. Most clinical practices
today will have access to multidetector row CT scanners that are indispens-
able for state-of-the-art CTC. These modern scanners achieve fast anatomic
coverage at superior resolution. Scanning is performed in supine and prone
positioning to help differentiate mobile fecal residue, and so on, from fixed
pathology, and to shift intraluminal fluid, allowing for inspection of other-
wise obscured portions of colon [39–43]. CTC can be performed at 40% to
50% dose reduction in modern scanners, using thin collimation acquisition
parameters [44].
Last, adequate image viewing with use of various 2D and 3D display
methods and fast 3D rendering on specialized offline workstations play an
important role in CTC. A combination of the various displays seems to
be most accurate. Faster computers and improved software allow expedi-
tious review in a 3D mode with 2D problem solving. Polyps are more
824 WALD et al

Fig. 1. (A) Volume-rendered intraluminal 3D view demonstrates a possible polyp at 6:00

o’clock. (B) 3D translucency rendering technique reveals high-attenuation internal density of
polypoid structure, compatible with barium-tagged stool. (C,D) Axial images with different win-
dow/level settings allow user to recognize high internal density of small polypoid lesion, unequiv-
ocally identifying it as tagged stool. (Images courtesy of Dr. Perry Pickhardt, Madison, WI.)

conspicuous when dedicated primary 3D viewing technique is employed.

CTC with primary 2D viewing can lead to excessive eye strain, especially
when radiologists are performing serial evaluations, and has yielded disap-
pointing results in multicenter trials of low-prevalence groups [23,25]. How-
ever, results in smaller studies with polyp-rich patient cohorts have been
promising [45,46] as Pickhardt pointed out [32].

Novel image postprocessing

Novel image postprocessing methods under investigation provide
‘‘unwrapping’’ of the circular colon for ease of inspection. So-called ‘‘trans-
lucency rendering’’ may allow incorporated contrast visible in small wall-
adherent feces, and help distinguish fecal material from polyps [47]. The
term ‘‘electronic cleansing’’ describes automated removal of labeled stool
by expert software before inspection by a radiologist [48–50]. However,
this technique can produce significant artifacts by removing some of the
polyps instead of feces and slow-down image analysis. Computer-aided de-
tection software has been used to label suspicious areas before final inspec-
tion by the radiologist, but requires further development before mainstream
 AN UPDATE ON IMAGING OF COLORECTAL CANCER 825

clinical use [51–53]. Improved 3D postprocessing and routine incorporation

of the latter methods into screening protocols may eventually improve the
sensitivity of CTC.

Trainingdaccreditation
There is a steep learning curve associated with the successful interpreta-
tion of CTC. Some authors have postulated that up to 50 cases should be
interpreted under supervision of an experienced reader before the radiologist
is competent to render high-quality interpretations [24,54]. Accreditation of
facilities and certification of readers, analogous to mammographic accredi-
tation, are being considered and the American College of Radiology is cur-
rently developing practice guidelines for CTC.

Summary
Currently, there is no consensus for the single best CTC technique. Sta-
tistically, the best results have been obtained with primary 3D viewing and
2D problem solving, after cathartic preparation and stool tagging. Ulti-
mately, this technique may be the optimum strategy for CTC. Prone and su-
pine imaging with adequate distention of the colon is crucial to yield the
high-quality datasets necessary to perform both 3D and 2D analysis.

CT Colonography in symptomatic patients

Although the role of CTC in the asymptomatic screening population is
a matter of ongoing investigation, there seems to be little disagreement
about its use in the examination of patients with incomplete colonoscopies,
detection of synchronous lesions in obstructing colon cancers, and superior
localization of lesions before surgery, allowing for evaluation of extralumi-
nal and remote findings [55–58]. Several studies found that between 10%
and 13% of patients have significant extracolonic findings, often requiring
further workup, additional or immediate therapy [59,60].

Magnetic resonance colonography

MRC has become more feasible secondary to technologic advancements
in phased array coil technology and the development of faster image acqui-
sition protocols. Potential uses include screening for, and staging of, colo-
rectal pathology. Analogous to CTC, MRC is less invasive than
conventional colonoscopy (CC), allows for evaluation of intra- as well as ex-
traluminal disease, and is an opportunity for synchronous evaluation of dis-
tant sites, particularly the liver, for spread of disease. Soft tissue contrast
resolution of MRC is superior to that of CTC, which aids in the assessment
of local staging. The absence of ionizing radiation may represent an impor-
tant advantage of MRC over CTC in the context of screening for colorectal
cancer, because repeat CTC every 5 years exposes individuals to potentially
significant cumulative radiation exposure.
826 WALD et al

Most MRC imaging protocols use a body coil for the abdomen and
phased array coil for the pelvis. Phased array coils are capable of acquiring
higher resolution datasets but allow only limited craniocaudal anatomic
coverage. Similar to CTC, patients are imaged prone and supine, allowing
for optimal distension while shifting residue and fluid in the colon.
Several imaging sequences have been developed, and can be classified into
bright- or dark-lumen techniques. Bright-lumen techniques, such as a bal-
anced steady state free-precession sequence (true FISP, FIESTA, balanced
FFE), show colorectal masses as dark filling defects/areas of low signal on
a background of bright-distended colon after a water enema. A dark-lumen
technique can demonstrate enhancing lesions on a background of a dark dis-
tended colon. Most clinical studies on MRC use the dark-lumen technique.
Lauenstein and colleagues [61] demonstrated better results with a dark-lu-
men compared with the bright-lumen technique in their study of 37 patients.
The dark lumen was able to identify all polyps O5 mm without any false
positive findings. The overall sensitivity of dark- compared with bright-lu-
men MRC was 79% versus 69%, respectively. However, the bright-lumen
technique (balanced steady state free-precession sequence) had better image
quality and was less susceptible to motion; this may be preferred in patients
unable to hold their breath or lay still sufficiently long.
As with CTC, distention of the colon is an essential prerequisite to cor-
rectly identify colorectal pathology. Multiple schemes have been developed
including water, water with paramagnetic contrast, carbon dioxide, and bar-
ium enema, depending upon desired dark- or bright-lumen techniques on
MRC. Recently, a study compared colonic distension with water versus car-
bon dioxide. The authors found similar accuracy for lesion detection; how-
ever, air provided better contrast-to-noise ratio and better distention [62].
Water combined with paramagnetic contrast and barium oral/rectal enema
has been used for stool tagging. The use of paramagnetic contrast for bright-
lumen MRC stool tagging is prohibitively expensive. However, oral and
rectal barium administration will result in a good dark-lumen MRC [63].
Satisfactory fecal tagging may allow for a less cathartic bowel preparation
and therefore enhance patient acceptance of MRC. This approach and its
limitations have been further discussed in context with CTC. Past attempts
to achieve stool tagging with diet modification have been unsuccessful [64].
An extensive meta-analysis comparing MRC and CC involving 563 pa-
tients demonstrated an overall MRC sensitivity and specificity of 75%
and 96%, respectively [65]. MRC was able to identify synchronous lesions
proximal to high-grade stenosing lesions that were not accessible by colonos-
copy. The results of this study are encouraging, although limited by the
inherent differences in employed techniques, as in the studies date from
1990 to 2004 [66]. Several studies have demonstrated high accuracy of
MRC in detecting lesions greater than 10 mm [67–70]. In a study of 122
patients MRC was nearly 90% sensitive for lesions greater than 5 mm
[67]. In a separate study of 100 patients, the sensitivity for lesions 6 to
 AN UPDATE ON IMAGING OF COLORECTAL CANCER 827

9 mm was 85% [71]. MRC routinely missed lesions less than 5 mm and flat
adenomas.
Most medical centers use CT rather than MRI for staging of colorectal
carcinoma in accordance with recommendations by the radiology diagnostic
oncology group II study [72]. Since its publication, technical advancements
in MRC have shown improved accuracy in differentiating clinically favor-
able intramural (T1/T2) from unfavorable higher local stage tumor
(T3/T4), especially in rectal cancer staging. Low and colleagues [73] were
able to appropriately identify the TNM stage in 21 of 27 colon cancer patients,
resulting in an overall accuracy of 78%, while the accuracy rate for rectal
cancer was 95%, correctly staging 20 out of 21 cancers. Therefore, MRC
was able to differentiate T1/2 from T3/4 stage cancers 95% of the time. Dif-
ferentiating nodal metastatic disease from reactively enlarged lymph nodes
has proven to be difficult solely based on size criteria. Accurate assessment
of nodal status should also take in account the morphology of the perinodal
tissue. Clinical research on lymph node imaging agents such as ultrasmall
iron–oxide particles, as described in the rectal staging portion of this article,
may improve accuracy of staging nodal involvement [74,75].
In summary, MRC may become an important tool in the screening for,
and assessment of, known colorectal cancer. It uses no ionizing radiation
and has powerful properties in determining local stage and evaluating for
the presence of distant metastases. Performance of MRC is technically chal-
lenging, expensive, and requires significant patient cooperation, all of which
represent barriers to broad implementation. Further refinement of the MRC
technique is necessary. In addition, more prospective evaluations in compar-
ison to existing modalities such as CC and CTC are required to fully under-
stand its role in the preoperative evaluation of symptomatic patients and
screening of the asymptomatic population.

Summarydwhat to choose?
Local practice pattern, available expertise in the involved medical/surgi-
cal subspecialties, and availability of the various imaging resources will have
a great impact on preferences for total colon imaging. In our opinion, the
sensitivity of an optimal CTC probably equals or exceeds that of DCBE
in all but a few practices where highly skilled fluoroscopists are still avail-
able. After a gradual decline in performance of DCBE in both clinical prac-
tice and training, and considering its high examiner dependence, CTC will
likely become the radiologic procedure of choice for colon cancer screening.
The sensitivity of a carefully executed CTC in the asymptomatic screening
population may be similar to that of Colonoscopy, although further inves-
tigation is needed. If equality can be demonstrated in rigorous trials, CTC
could be added to the current reimbursable options available for colorectal
carcinoma screening, which include fecal occult blood testing, sigmoidos-
copy, DCBE examination, and colonoscopy.
828 WALD et al

At this point, most colon examination techniques require similar patient

preparation. If cathartic preparation is not performed for patient comfort
reasons, same-day polyp removal will not be feasible. Patients should prob-
ably be given the choice after explaining risks and benefits of the cathartic
versus noncathartic prep.
To fully leverage the advantages of a less invasive technique (ie, de-
crease in number of complications per polyp detected), practices could at-
tempt to stratify patients into primary invasive and noninvasive groups
based on history and clinical parameters (age, prior studies, clinical risk
profile). If immediate interpretation of CTC is feasible and established
in a given practice, patients with positive findings who have been ade-
quately prepped can proceed to diagnostic/therapeutic colonoscopy on
the same day. Although this requires considerable interdisciplinary coordi-
nation, it would almost certainly result in high patient acceptance. Prac-
tices planning for new colon cancer screening facilities, or restructuring
existing services, should keep this in mind. A combined CTC/colonoscopy
approach may require a shift of resource allocation to free up dedicated
CT scanning time in Radiology departments, or perhaps necessitate the in-
stallation of dedicated equipment in a screening facility and the acquisition
of dedicated workstations. Close collaboration between colonoscopists and
radiologists would be inevitable under such circumstances, and would
likely result in higher quality patient care. Although most facilities have
adequate scanner hardware to obtain sufficient source images, there is
probably a lack of experienced interpreting radiologists in some areas.
This shortfall may be overcome in the age of digital images as teleradiol-
ogy technology makes remote interpretation by dedicated expert radiolo-
gists feasible.

Rectal cancer staging

Traditionally, the focus of preoperative radiologic evaluation of rectal
cancer was stratification of patients into those receiving primary curative
surgery alone and those who would benefit from preoperative neoadjuvant
chemotherapy/radiation. Based on imaging criteria established by Hilde-
brandt and Feifel [76], the primary concern for endorectal ultrasound
(EUS) became differentiating T2 from T3 disease. However, more recent ad-
vances in surgical techniques and neoadjuvant therapy for a subgroup of
rectal cancer patients have significantly reduced cancer-related morbidity
and mortality. Identifying optimum treatment parameters based on local
staging, and precise differentiation of both superficial and locally invasive
lesions, are of utmost importance in preoperative radiology evaluation.
Radical surgery for rectal cancer is associated with potential mortality as
well as morbidity of poor bowel or sexual function, and possibly a stoma.
Transanal excisions of superficial lesions (carcinoma in situ or ultrasound
 AN UPDATE ON IMAGING OF COLORECTAL CANCER 829

stage T1), and in particular, transanal endoscopic microsurgery, allow for

improved functional outcome. Although their value as a curative cancer
operation is seen as controversial by some [77], other studies found similar
recurrence rates to more radical surgeries [78,79]. Preoperative chemoradia-
tion followed by local excision of ultrasound stage T2, node-negative le-
sions, has also been advocated [78,80]. Thus, accurate discrimination of
T1 from T2 disease is becoming crucial for preoperative staging.
Recent surgical management for T3/T4 lesions has focused on total mes-
orectal excision (TME). This incorporates the radical en bloc resection of
tumor, local draining nodes, and surrounding mesorectal fat, including
the thin mesorectal fascia. The success of TME surgery in preventing local
recurrence has been linked to circumferential resection margin (CRM) and
nodal status rather than T-staging [81]. The spatial relationship between tu-
mor and mesorectal fascia is essential in determining the status of the CRM.
Tumor found within 1 mm of the thin mesorectal fascia is an ominous prog-
nostic indicator of a positive CRM and ultimately local recurrence.
Use of neoadjuvant radiation or chemotherapy within the past decade
has had a major impact on treatment. The Swedish Rectal Cancer Trial
showed that a short course of preoperative radiation therapy reduces the re-
currence rate from 27% to 11% [82]. The Dutch TME trial stratified the pa-
tient population of locally advanced T3 or T4 stage lesions in whom
neoadjuvant therapy is indicated [83]. Neoadjuvant radiation should be re-
served for tumors with extramural spread or nodal metastasis [84]. Intensive
and targeted preoperative therapy has rendered previously irresectable tu-
mor resectable.

Staging techniques
The role of endorectal ultrasound in local staging of rectal cancer
EUS is the mainstay of local staging in many institutions across the
United States. A metanalysis of the pertinent radiology literature comparing
US/MR/CT published in 2004 [85] suggested that EUS is currently the mo-
dality of choice for local staging of all rectal carcinomas. Most practices use
a 7.5 MHz or 10 MHz rigid US transducer with a saline-filled balloon tip,
providing a 360-degree field of view. Accurate depiction of the five layers
of the rectal wall is easily obtained. Tumor most commonly appears as a hy-
poechoic lesion invading or disrupting layers of the rectal wall. Those le-
sions invading just into the submucosa are ultrasound stage T1, those into
but not beyond the muscularis propria are considered ultrasound stage T2
(Fig. 2), those lesions invading into the adjacent perirectal fat are ultrasound
stage T3, and invasion into adjacent organs represent ultrasound stage T4.
Many consider EUS the imaging modality of choice for early T1/T2
rectal staging because of its superior depiction of tumor ingrowth into su-
perficial layers of the rectal wall [86], with T staging accuracy ranging
830 WALD et al

Fig. 2. EUS depicting an ultrasound Stage T2 lesion. Hypoechoic tumor (T) is invading
through the echogenic submucosa (S), and up to, but not beyond, the muscularis propria (*).

from 69% to 97% [87]. In comparison, pelvic phased array coil MRI has
been shown to be less accurate in differentiating T1 from T2 lesions [88].
It is important to note that EUS can only detect those lymph nodes within
depth of range of the transducer resulting in potential understaging of dis-
ease. In the metanalysis performed by Bipat and colleagues [85], there was
an overall sensitivity of 67% and specificity of 78% for lymph node involve-
ment. Also, over staging of T2 tumors as T3 lesions is relatively common
with EUS, because frequently peritumoral inflammatory changes have a sim-
ilar appearance as primary tumor [89]. This possible overstaging may result
in more radical therapy and expose patients to unnecessary chemoradiation.
Although some studies suggest that EUS is better suited for imaging of
superficial lesions, its accuracy in locally advanced disease has recently
been questioned, particularly in light of advancements in MRI [88]. In Bi-
pat’s meta-analysis, US had an overall sensitivity of 90% and specificity
of 75% for perirectal fat invasion, better than CT or MRI [85]. However,
it is important to note that their meta-analysis included studies performed
over a 16-year time frame, and during that time, little has changed in
EUS technique. During the same time interval significant technologic ad-
vancements have occurred in MRI rectal cancer staging, as discussed below.
EUS is limited in its assessment of advanced rectal cancer by its limited
acoustic window, depth of penetration, and small field of view. Tumors
beyond 13 cm superior to the dentate line are difficult to stage with the
fixed-length rigid probe. Additional disadvantages of EUS are operator de-
pendence, the need for a bowel prep, and occasionally proper positioning.
Some patients are unable to tolerate the examination due to pain associated
with highly stenotic lesions, which one may not be able to cross with the
 AN UPDATE ON IMAGING OF COLORECTAL CANCER 831

probe. In addition, recent studies have identified CRM, rather than tradi-
tional local T staging, as the more important preoperative indicator of local
recurrence after surgery. EUS cannot accurately depict the relationship of
the tumor to the mesorectal fascia.
3D endoluminal ultrasound and ultrasound miniprobe examinations are
examples of recent advancements in endoluminal ultrasound technology. 3D
endoluminal ultrasound allows the examiner to gain a different perspective
on the lesion itself, particularly in regard to longitudinal extent and relation-
ships to adjacent organs. Sequential transverse images are ‘‘stacked’’ upon
one another, and using a separate workstation, the lesion can then be de-
picted in any plane. However, the scanning technique is identical to tradi-
tional EUS, with the same limitations in terms of depth of penetration
and patient factors, as outlined previously. Miniprobe endoscopic ultra-
sound (m-EUS) uses an ulrathin probe (3–4 mm diameter) that can be
used for local staging during routine colonoscopies. The advantages are
a smaller and flexible probe, allowing for staging of higher rectal and colonic
lesions, and the potential for crossing stenotic lesions. The use of m-EUS in
the rectum may be limited, given that m-EUS has similar problems to tradi-
tional EUS in terms of depth of penetration, and in differentiating T3 from
T4 lesions, which may be crucial in stenotic lesions [90]. The accuracy of
m-EUS in determining lymph node status is approximately 80% [91].
Thus, its clinical use in the assessment of rectal lesions maybe limited. How-
ever, m-EUS does have a potential role in the local staging of colonic
neoplasms, particularly in selected groups of patients, or in confirming su-
perficial lesions in patients who may be poor candidates for an open surgical
procedure [90].

The role of MRI in local staging of rectal cancer

A growing body of literature is supporting the use of pelvic phased array
coil MRI (PA-MRI) over endorectal MRI, endoluminal ultrasound, CT,
and digital rectal exam (DRE) for staging locally advanced extramural tu-
mors. Early MRI studies were less encouraging because images were obtained
with low-resolution body coils. However, the advent of high-resolution
phased array coils in combination with improved imaging sequences has
produced superior image quality to accurately assess tumor extent. In
essence, a phased array coil has multiple surface coils that simultaneously
detect signal, allowing for higher signal-to-noise ratios (SNRs), better
spatial resolution, and faster imaging. High spatial resolution MRI
can identify tumor features associated with poorer outcome, namely degree
of extramural extension, venous invasion, nodal involvement, and per-
itoneal infiltration. State-of-the-art rectal MRI staging should be per-
formed on a magnet with 1 or1.5 T field strength, phased array surface
pelvic coil, and thin T2-weighted fast spin echo sequence (T2W-FSE) im-
ages. T2W-FSE provides superior discrimination between hyperintense
832 WALD et al

mesorectal fat, intermediate tumor, hypointense rectal wall, and mesorectal

fascia compared with other pulse sequences. Further technical details are
discussed in an article by Brown and colleagues, in the British Journal
of Radiology [92].
PA-MRI can be accurate in predicting T stage and CRM status. In 98
patients, Brown and colleagues [93] had 94% and 92% accuracy rate in pre-
dicting T-stage and CRM status, respectively. In addition, a study of 76 pa-
tients by Beets-Tan and colleagues [94] resulted in 83% accuracy rate in
predicting CRM when examined by an experienced observer 1. Linear re-
gression analysis of the study revealed that a distance from the tumor to
the mesorectal resection plane of at least 2 mm could be predicted with
97% CI. A histologic distance of at least 1 mm can accurately was predicted
with high confidence when the measured distance on MRI is at least 5 mm
[94].
In 2002, the MERCURY Study (Magnetic Resonance Imaging and Rec-
tal Cancer European Equivalence Study) was launched [95,96]. This study is
assessing the equivalence of PA-MRI and histopathology, while evaluating
the prediction of CRM positive tumors using MRI in 11 European centers
[96]. Results of this study may eventually validate an MR-based preopera-
tive staging system, stratifying patients into prognostic groups, and identify-
ing patients who may benefit from neoadjuvant therapy [95]. At the time of
writing of this article, the results have yet to be formally published, but the
results presented at the RSNA in 2004 demonstrate an 82% accuracy in pre-
dicting involvement of the CRM, compared with histology, by using a 1-mm
cutoff on MRI [96a].
Comparative analysis of modalities used for staging and histology of tu-
mor prognostic favorability resulted in accuracy of 94% (MRI), 69%
(EUS), and 65% (DRE) [88]. Clinical favorability was defined as T1, T2,
and T3a lesions without lymph nodes metastasis. Borderline T3a lesions
are considered prognostically favorable because they present with low risk
of surgical failure, and neoadjuvant therapy would not be beneficial [97].
All 98 patients were able to tolerate the MRI examination. However, only
54 and 74 patients were able to undergo EUS and DRE, respectively.
EUS and DRE were aborted for various reasons, including poor bowel
prep, pain, and high tumor location out of reach.
MRI is limited in differentiating benign reactive lymph nodes (LN) from
micrometastasis in small nodes. Of concern is the fact that greater than 50%
of nodes containing metastases from rectal cancer are less than 5 mm in size
[98]. Therefore, determination of LN involvement has been variable across
all imaging modalities. When nodes of greater than 5 mm maximum short
axis diameter are considered malignant, the nodal status accuracy rate of
MRI ranges from 59% [97] to 85% [93]. Multiple small comparative studies
between multiple modalities favor the use of PA-MRI, endorectal (ERC)-
MRI, and EUS, in order of greatest accuracy [98,99]. One study of lymph
nodes in 75 patients with rectal cancer demonstrated that nodes O4 mm
 AN UPDATE ON IMAGING OF COLORECTAL CANCER 833

had a higher rate of involvement, and nodes O8 mm were invariably af-

fected by metastases [100]. However, the use of morphologic criteria to de-
termine whether LNs are affected, such as spiculation, indistinctness,
lobulation, and roundness, are similar to the use of a 5-mm size criteria. Ad-
ditional characteristics that have a high correlation with LN involvement
include mottled heterogenous post contrast-enhancement appearance, asso-
ciated venous encasement, and dirty perirectal fat [100,101].
Early investigational studies of MRI using ultrasmall particles of iron
oxide (USPIO) for LN detection in patients with rectal cancer have
shown encouraging results. LNs are normally populated among other cells
by macrophages, which will take up USPIO and result in signal loss (appear
dark) on aT2* weighted sequence. Invasion of the lymph nodes with
malignant cells displaces the macrophages, resulting in higher signal on
USPIO-enhanced T2* imaging. Multiple uptake patterns can be visualized
including uniform low, central low, eccentric high, and uniform high
signal in order of increasing malignant potential. In a small study, Koh
and colleagues [75] demonstrated 96% of benign LNs have a pattern of
uniform or central low-signal intensity, while eccentric or uniform high signal
is seen in those nodes containing metastases. Nodes containing metastatic
foci smaller than 1 mm could not be recognized as containing tumor.
Multiple small studies have claimed that ERC-MRI and EUS have no
statistical significant difference in T-staging accuracy [99,102,103]. In a study
of 20 patients Akin and colleagues [104] claimed an accuracy of 85% for T
staging when using ERC-MRI. The largest study of 89 patients showed that
ERC-MRI and EUS had similar results, both of which were far superior to
CT in the preoperative evaluation of rectal wall and adjacent organ invasion
[105]. ERC-MRI uses an endorectal surface coil with balloon cover. Balloon
insufflation is required for optimal contact with the affected bowel wall to
maximize the inherent high SNRs; however, it may be impossible to cross
stenotic lesions. The higher SNR with small field of view (FOV) in ERC-
MR greatly increases image resolution, and rectal wall layers are clearly de-
lineated. The limitation of ERC-MRI is the small field of view that is pro-
vided, with a very rapid drop in signal intensity beyond its immediate
vicinity [95]. Therefore, adequate evaluation of CRM and mesorectal nodal
status is difficult. ERC-MRI and EUS have similar technical advantages and
disadvantages.

The role of CT in local staging of rectal cancer

Significant advances in CT technology allow high-resolution imaging
of more anatomy in less time than ever before. However, its clinical use in
local rectal cancer staging may be limited, particularly in light of the con-
temporaneous advancements in MR technology. A CT meta-analysis of
78 studies with 4897 patients revealed T-staging accuracy of only 73%
[106]. The metanalysis of Bipat and colleagues [85] also demonstrated that
834 WALD et al

EUS was at least as good as CT in terms of local staging and lymph node
status. A prospective blinded study published in 2002 demonstrated that
EUS was superior to CT in local staging of rectal lesions [107]. Technical
advancements in MRI, as previously described, have replaced conventional
CT in assessing locally advanced tumors owing largely to its superior soft
tissue resolution capability. Small comparative studies between MRI and
CT demonstrated superior accuracy for MRI in predicting invasion into
the bladder/uterus [108], pelvic wall, and subtle bone marrow involvement
[109]. Although CT may not be able to depict the CRM as well as MRI
due to its limitation in contrast resolution, it may represent an imaging al-
ternative for patients with contraindications to MRI (implanted cardiac
pacemaker, and so on).

Restaging after neoadjuvant therapy and detection of local recurrence

Despite improvements in therapy, a significant number of treated rectal
cancers will recur. As discussed previously, the advent of local therapies
has dramatically reduced the morbidity of those undergoing surgical treat-
ment for rectal neoplasms. However, this decrease in morbidity must be
weighed against a potential increased risk of local recurrence. There is a sig-
nificant recurrence rate after local resection of superficial lesions [110], but
those patients could undergo salvage surgery for cure if detected early. In
addition, preoperative therapy may result in downstaging, potentially mak-
ing local resection feasible in place of more radical therapy [80,111]. In this
subgroup, accurate preoperative assessment, after neoadjuvant therapy,
would be crucial to identify those patients who can undergo local excision
for cure. Thus, accurate radiographic evaluation for local recurrence of rec-
tal cancer, and accurate restaging after neoadjuvant therapy, both have
a crucial impact on morbidity and long-term survival of these patients.
EUS was found to be accurate in detection of local recurrence in those
patients who underwent local resection for superficial lesions [112]. In a stud-
y including 108 patients who underwent local excision for cure, 32 developed
local recurrence. Twenty-six of these 32 patients were asymptomatic, and 10
of those who developed local recurrence (31%) were detected only by EUS
[113]. These same authors, as well as others, believe that postoperative
surveillance using EUS in those undergoing radical surgeries is also useful
[113–117]. However, a normal postoperative EUS does not entirely exclude
residual disease [118], and close postoperative surveillance of these patients
is necessary.
Many studies exploring the role of EUS and MRI after neoadjuvant
chemoradiation demonstrate poor accuracy in detecting local recurrence
for both. EUS interpretation is limited by the inflammatory fibrotic reaction
seen after neoadjuvant therapy [119]. Although some believe that EUS is
valuable in evaluation of rectal lesions after chemoradiation [119], there
are many reports of a dramatic decrease in accuracy after therapy, and
 AN UPDATE ON IMAGING OF COLORECTAL CANCER 835

EUS alone in this clinical scenario may be of little use [120–122]. However,
EUS-guided biopsy in combination with routine EUS surveillance is useful
in the evaluation of suspicious postoperative lesions [123,124].
Recent results from restaging MRI in patients with locally advanced tu-
mor after neoadjuvant chemoradiation therapy have been poor in accurately
predicting both T and N stage [125,126]. MRI cannot reliably distinguish ra-
diation fibrosis or postsurgical scarring from residual tumor, resulting in
overstaging. In a study by Peschaud and colleagues [127], an MRI restaging
study encountered overestimation errors particularly with tumors located in
the low anterior rectum.
Fluoro-2-deoxy-D-glucose-positron emission tomography (FDG-PET),
combined with CT, may offer the greatest advantage in detection of local re-
currence. Although some regions of fibrosis can still have minor radiotracer
uptake, recent studies have demonstrated high accuracy in detection of local
recurrence after abdominal personal resections and anterior resections, even
after chemoradiation exposure [128–130]. Combining the images with CT al-
lows for simultaneous depiction of regional anatomy, decreasing the poten-
tial for false positive studies [128]. The study performed by Moore and
colleagues [130], also demonstrated an increase in accuracy when performed
12 months after radiation, possible secondary to a decrease in inflammatory
response related to radiation. FDG-PET also has the advantage
of visualizing the rest of the body, detecting hepatic metastases, and evaluat-
ing a rising carcinoembryonic antigen level after a curative rectal cancer
resection [131].

Summary
The advent of new surgical techniques and multimodality treatment par-
adigms aimed at improved functional outcomes in patients with rectal can-
cer has made preoperative staging of rectal carcinomas even more critical.
Endoscipic ultrasound is currently the mainstay of local preoperative stag-
ing in most institutions. Although superficial lesions are best depicted
with endoscopic ultrasound, there are limitations, and pelvic phase array
coil MRI appears to be more accurate in visualization and depiction of ad-
vanced lesions. There is increasing reliance on assessment of the integrity of
the CRM in regard to preoperative planning and use of neoadjuvant chemo-
radiation, and we suspect that more and more institutions will depend
heavily on pelvic phase array coil MRI in the preoperative assessment of
rectal cancer. The results of the MERCURY study could have a major
impact on the utility of MRI in the preoperative staging of rectal lesions.
Expect to see these results soon in the journal Radiology.
Endoscopic ultrasound, particularly when combined with guided biopsy,
is valuable in the detection of local recurrence after initial therapies for
rectal carcinoma. Although to date MRI has been disappointing in
836 WALD et al

distinguishing posttherapeutic changes from local recurrence, FDG-PET is

both sensitive and specific, especially when combined with CT, and perhaps
MRI, and offers the additional advantage of a whole-body metastatic
survey.

The role of positron emission tomography imaging in colorectal

malignancy
PET, using fluoro-2-deoxy-D-glucose (18FDG), is impacting preopera-
tive staging and evaluation for recurrence of colorectal cancer. It is the
most sensitive and specific technique for in vivo imaging of metabolism
and receptor ligand interactions in human tissue [132].
PET localization is based on abnormally increased tissue metabolism of
cancer cells. In vivo imaging to detect increased glucose metabolism in can-
cer cells can be linked to increased expression of epithelial glucose trans-
porter proteins and increased activity of the principle enzymes of the
glycolytic pathway [132]. There is resulting intracellular accumulation,
also known as ‘‘metabolic trapping,’’ of 18FDG molecules. This mechanism
is, in large part, responsible for the high sensitivity of FDG-PET, but also
explains its limited specificity. Tracer may accumulate in any cells with
hypermetabolism, such as leukocytes, macrophages, and other inflammatory
cells.
Typically, a whole-body PET scan is obtained 60 minutes after adminis-
tration of 10 mCi of 18FDG. The whole body is imaged in axial plane over
30 to 40 minutes with correction for variable soft tissue attenuation. Semi-
quantitative index assessment of the radioactive uptake is determined by
standardized uptake value (SUV). SUV is based on tumor radiotracer con-
centration (Q) normalized to the injected activity (Qinj) and to the body
weight (W). The (limited) spatial resolution of a current PET scanner is in
the vicinity of 5 to 8 mm.
The utility of PET scanning alone is limited by poor spatial resolution
and difficulty in exact anatomic localization of areas of abnormal radio-
tracer uptake. On the other hand, while much less sensitive, CT and MRI
have inherent high spatial resolution that aids anatomic localization of le-
sions. Hybrid technology such as PET-CT has evolved, combining func-
tional evaluation with anatomic localization and improved spatial
resolution.
PET imaging has been available for staging of colorectal cancer since the
1980s using 18FDG; however, recently it has had more of an impact on stag-
ing and evaluation for recurrence. Other imaging modalities such as CT,
MRI, and ultrasound (US) have traditionally been used for staging. How-
ever, the recurrence rate after initial treatment of colorectal malignancy
has been reported to be as high as 30% to 40%, often within the first 2 years
of treatment [133]. High recurrence rates so shortly after diagnosis suggest
limitations in accuracy of the traditional staging tools.
 AN UPDATE ON IMAGING OF COLORECTAL CANCER 837

The main applications of PET imaging in colorectal carcinoma are stag-

ing of disease at time of initial diagnosis, staging of recurrent disease, and
assessment of response to therapy. Based on a study by Meta and colleagues
[134], FDG-PET has a major impact on the management of colorectal can-
cer patients and contributed to changes in clinical stage and management
decisions in O40% of the patients. In this study, PET had the highest sen-
sitivity (95%) of all the modalities, which was confirmed by other studies.
The sensitivity of PET was also dependent on histology, limited to
58% for detection of mucinous carcinoma versus 92% for nonmucinous
carcinoma [135].
The sensitivity for detection of lymph node involvement was low (29%);
however, the specificity and accuracy were higher: 88% and 75%, respec-
tively. The sensitivity, accuracy, and specificity for detection of liver metas-
tasis were similar comparing FDG-PET and CT. The sensitivity for
detection of liver metastasis by PET was found to be 78% in the study by
Meta and colleagues [134]. Other studies have cited higher sensitivity for de-
tection of liver metastasis using PET [136–138]. However, study by Selzner
[139] showed that hybrid PET/CT provided similar information regarding
hepatic metastasis. PET/CT was superior to contrast-enhanced CT for de-
tection of recurrent intrahepatic metastatic disease, extrahepatic metastases,
and local recurrence at the initial surgical site [139] (Fig. 3).
One has to be aware of the information that can be obtained for different
imaging modalities and how it may affect overall management in terms of
initial staging and for evaluation of recurrent disease. Tzimas and colleagues
[140] showed that the performance of different imaging modalities used for
colon and rectal cancer staging including tumor depth, lymph node inva-
sion, and metastasis does vary. PET has sensitivity of 87% to 100% and
specificity of 43% to 100%. The positive predictive value (PPV) is relatively
high, with 90% to 93% compared with CT, which has overall sensitivity of
48% to 97% in initial colorectal cancer staging. The specificity and PPV for
CT was 57% to 100% and 100%, respectively. MRI had variable sensitivity
and specificity based on the type of coil used such as external body coil
or endorectal coil. The sensitivity was lower with external body coil
with 22% to 89% versus 81% to 83% with endorectal coil. Specificity for
staging was 43% to 100% for PET, 57% to 100% for CT, 71% to 100%
external body coil MRI, and 42% to 100% for endoluminal rectal coil
MRI, respectively [140].
The evaluation of the recurrent disease is limited by the postsurgical
changes and fibrosis. There are also changes such as inflammation or fibro-
sis from chemotherapy and radiation (Fig. 4). PET imaging has been useful
in differentiation of fibrotic or inflammatory changes from the recurrent dis-
ease. Thus, another major benefit of PET in evaluation of recurrent disease
has been avoidance of inappropriate local therapies that may carry signifi-
cant morbidity by documentation of widespread disease [141]. In the con-
text of elevated carcinoembryogenic antigen (CEA), PET imaging is
838 WALD et al

Fig. 3. (A) A focus of radiotracer uptake in the segment VIII of the right lobe of the liver cor-
responding to a hypodense lesion in the noncontrast-enhanced CT, probable focus of metasta-
sis. (B) Another focus of FDG-PET uptake in the segment VI of the right lobe of the liver with
corresponding hypodense lesion in the noncontrast-enhanced CT. (C) An 83-year-old female
with a recent diagnosis of hepatic flexure colon cancer with increased uptake in the pelvis
with a FDG-PET scan. This corresponds to the soft tissue mass noted in the right presacral
space most likely a metastatic lymph node. (Images courtesy of Yamin Dou M.D, Methuen,
MA.)

shown to have superior sensitivity in detection of recurrent disease. It has

been shown by Selzner [139] that contrast-enhanced CT is not as sensitive
as PET/CT for evaluation of local recurrence, with sensitivity of 20% versus
93%.
The use of PET has also been evaluated for recurrence of colorectal can-
cer and its effect on patient management. A meta-analysis by Huebner and
colleagues [142] determined that the overall sensitivity and specificity of
FDG-PET for detection of recurrent colorectal carcinoma is 97% and
76%, respectively. The overall change in the management of patient treat-
ment based on the PET findings was 29%. Meta and colleagues [134] eval-
uated the impact of the use of FDG-PET from a referring physician’s point
of view, in terms of change in patient management. Changes in the manage-
ment were classified as intramodality (eg, altered medical, surgical, or radi-
ation treatment) or intermodality (eg, change from surgical to medical,
 AN UPDATE ON IMAGING OF COLORECTAL CANCER 839

Fig. 4. (A) Initial full-body 18FDG-PET scan demonstrates four foci of increased uptake in the
liver. There is normal uptake seen in the renal collecting system, bladder, ureters, cardiac mus-
cle, with mild activity in the colon. (B) Postchemotherapy full-body PET scan performed 5
months after treatment shows excellent response to chemotherapy with complete resolution
of the uptake in liver lesions. However, there is still normal expected uptake in the kidneys
and bowel. Repeat CT at this time also demonstrates decrease in the size of the liver lesions.
(Images courtesy of Yamin Dou M.D, Methuen, MA.)

surgical to radiation, medical to no treatment). Based on this study, FDG-

PET had a significant impact on the management of patients and contrib-
uted to changes in treatment or staging in O40% of the patients. Based
on this study, clinicians report changes in the staging in 42% of the patients.
The colon cancer was upstaged in 80% and downstaged in 20% of the pa-
tients. The PET findings contributed to intermodality changes in 37%, intra-
modality changes in 18%, combination of management changes in 7%, and
no change in 32% of the patients [134]. There are additional limitations to
PET. One is the use of chemotherapy, which decreases the sensitivity of PET
scan in detecting tumors. It was observed that the FDG uptake increases if
imaging is performed at 4 to 5 weeks after treatment. It is postulated that
there is a ‘‘flare’’ phenomenon based on increased macrophage infiltration
with greater tumor cell kill [132]. FDG uptake is also limited in tumors
less than 5 mm. Foci of inflammation and infection can result in false pos-
itive scans. In addition, high blood glucose levels and chemotherapy within
1 month of a study significantly decrease sensitivity [135,139].
The utility of combining CTC and whole-body FDG-PET has also been
studied by Veit and colleagues [143]. Based on their study, although a small
number of patient population, the integrated protocol may have substantial
benefit in staging patients with suspected colorectal malignancy focusing on
patients with incomplete colonoscopy and synchronous lesions. All but one
of the lesions in the colon were detected. An additional polyp was identified
840 WALD et al

as malignant. It also proved accurate in lymph node staging and detecting

extracolonic tumor sites [143]. Combined contrast-enhanced CTC and
PET is almost equivalent to having a routine staging CT; however, it is
more sensitive in detection of small polyps in the colon, which may be other
synchronous lesions, but may not be detected otherwise due to false positive
uptake of FDG in the colon.
Based on the current evidence, FDG-PET has utility in the initial staging
of colorectal cancer, and for evaluation of recurrent disease. Contrast-
enhanced CT and FDG- PET imaging appear to increase the sensitivity,
specificity, and accuracy for staging. Contrast-enhanced CT increases the
spatial localization, while FDG-PET adds functional information. The fu-
ture of tumor imaging is growing by new innovations in molecular imaging
with tumor-specific agents for improved initial staging and evaluation for
recurrence of colorectal carcinoma.

Acknowledgments
The authors thank Dr. Gina Brown, Consultant Radiologist and Honor-
ary Senior Lecturer, Department of Radiology, The Royal Marsden Hospi-
tal NHS Trust, UK, for her insights into rectal cancer imaging, and, in
particular, PA-MRI. Furthermore, we owe thanks to Drs. Perry Pickhardt,
of Madison, WI, and Yamin Dou, of Methuen, MA, for some illustrative
case material.

References
[1] Jemal A, Murray T, Samuels A, et al. Cancer statistics, 2003. CA Cancer J Clin 2003;53:5.
[2] Winawer S, Zauber A, Ho M. Prevention of colorectal cancer by colonoscopic polypec-
tomy. The National Polyp Study Workgroup. N Engl J Med 1993;329:1977.
[3] Anderson LM, May DS. Has the use of cervical, breast, and colorectal cancer screening in-
creased in the United States? Am J Public Health 1995;85:840.
[4] Fischer A. Fruehdiagnose des Dickdarmkrebses, insbesondere seine Differentialdiagnose
gegen Tuberkulose mit Hilfe der kombinierten Luft- und Bariumfuellung des Dickdarms.
Deutsch Ges Med 1923;35:86.
[5] Rubesin SE, Levine MS, Laufer I, et al. Double-contrast barium enema examination tech-
nique. Radiology 2000;215:642.
[6] Gazelle GS, McMahon PM, Scholz FJ. Screening for colorectal cancer. Radiology 2000;
215:327.
[7] Fork FT, Lindstrom C, Ekelund G. Double contrast examination in carcinoma of the colon
and rectum. A prospective clinical series. Acta Radiol Diagn (Stockh) 1983;24:177.
[8] Johnson CD, Carlson HC, Taylor WF, et al. Barium enemas of carcinoma of the colon: sen-
sitivity of double- and single-contrast studies. AJR Am J Roentgenol 1983;140:1143.
[9] Rex DK, Rahmani EY, Haseman JH, et al. Relative sensitivity of colonoscopy and bar-
ium enema for detection of colorectal cancer in clinical practice. Gastroenterology 1997;
112:17.
[10] Fork FT. Double contrast enema and colonoscopy in polyp detection. Gut 1981;22:971.
 AN UPDATE ON IMAGING OF COLORECTAL CANCER 841

[11] Steine S, Stordahl A, Lunde OC, et al. Double-contrast barium enema versus colonoscopy
in the diagnosis of neoplastic disorders: aspects of decision-making in general practice. Fam
Pract 1993;10:288.
[12] Winawer SJ, Stewart ET, Zauber AG, et al. A comparison of colonoscopy and double-
contrast barium enema for surveillance after polypectomy. National Polyp Study Work
Group. N Engl J Med 2000;342:1766.
[13] Glick SN, Fibus T, Fister MR, et al. Comparison of colonoscopy and double-contrast bar-
ium enema. N Engl J Med 2000;343:1728.
[14] Anderson ML, Heigh RI, McCoy GA, et al. Accuracy of assessment of the extent of exam-
ination by experienced colonoscopists. Gastrointest Endosc 1992;38:560.
[15] Godreau CJ. Office-based colonoscopy in a family practice. Fam Pract Res J 1992;12:
313.
[16] Lieberman DA, Smith FW. Screening for colon malignancy with colonoscopy. Am J Gas-
troenterol 1991;86:946.
[17] Rex DK, Lehman GA, Hawes RH, et al. Screening colonoscopy in asymptomatic average-
risk persons with negative fecal occult blood tests. Gastroenterology 1991;100:64.
[18] Laufer I, Smith NC, Mullens JE. The radiological demonstraction of colorectal polyps un-
detected by endoscopy. Gastroenterology 1976;70:167.
[19] Miller RE, Lehman G. Polypoid colonic lesions undetected by endoscopy. Radiology 1978;
129:295.
[20] Jentschura D, Raute M, Winter J, et al. Complications in endoscopy of the lower gastroin-
testinal tract. Therapy and prognosis. Surg Endosc 1994;8:672.
[21] Waye JD, Lewis BS, Yessayan S. Colonoscopy: a prospective report of complications.
J Clin Gastroenterol 1992;15:347.
[22] Pickhardt PJ, Choi JR, Hwang I, et al. Computed tomographic virtual colonoscopy to
screen for colorectal neoplasia in asymptomatic adults. N Engl J Med 2003;349:2191.
[23] Cotton PB, Durkalski VL, Pineau BC, et al. Computed tomographic colonography (virtual
colonoscopy): a multicenter comparison with standard colonoscopy for detection of colo-
rectal neoplasia. JAMA 2004;291:1713.
[24] Ferrucci JT. Colonoscopy: virtual and opticaldanother look, another view. Radiology
2005;235:13.
[25] Rockey DC, Paulson E, Niedzwiecki D, et al. Analysis of air contrast barium enema, com-
puted tomographic colonography, and colonoscopy: prospective comparison. Lancet 2005;
365:305.
[26] Johnson CD, Harmsen WS, Wilson LA, et al. Prospective blinded evaluation of computed
tomographic colonography for screen detection of colorectal polyps. Gastroenterology
2003;125:311.
[27] Pickhardt PJ. Differential diagnosis of polypoid lesions seen at CT colonography (virtual
colonoscopy). Radiographics 2004;24:1535.
[28] Dachman AH. Diagnostic performance of virtual colonoscopy. Abdom Imaging 2002;27:
260.
[29] Halligan S, Altman DG, Taylor SA, et al. CT colonography in the detection of colorectal
polyps and cancer: systematic review, meta-analysis, and proposed minimum data set for
study level reporting. Radiology 2005;237:893.
[30] Ristvedt SL, McFarland EG, Weinstock LB, et al. Patient preferences for CT colonog-
raphy, conventional colonoscopy, and bowel preparation. Am J Gastroenterol 2003;98:
578.
[31] Iannaccone R, Laghi A, Catalano C, et al. Computed tomographic colonography without
cathartic preparation for the detection of colorectal polyps. Gastroenterology 2004;127:
1300.
[32] Pickhardt PJ. CT colonography without catharsis: the ultimate study or useful additional
option? Gastroenterology 2005;128:521.
842 WALD et al

[33] Gluecker TM, Johnson CD, Harmsen WS, et al. Colorectal cancer screening with CT co-
lonography, colonoscopy, and double-contrast barium enema examination: prospective as-
sessment of patient perceptions and preferences. Radiology 2003;227:378.
[34] Fletcher JG, Johnson CD, MacCarty RL, et al. CT colonography: potential pitfalls and
problem-solving techniques. AJR Am J Roentgenol 1999;172:1271.
[35] Macari M, Lavelle M, Pedrosa I, et al. Effect of different bowel preparations on residual
fluid at CT colonography. Radiology 2001;218:274.
[36] Lefere PA, Gryspeerdt SS, Dewyspelaere J, et al. Dietary fecal tagging as a cleansing
method before CT colonography: initial results polyp detection and patient acceptance. Ra-
diology 2002;224:393.
[37] Ji H, Rolnick JA, Haker S, et al. Multislice CT colonography: current status and limita-
tions. Eur J Radiol 2003;47:123.
[38] Barish MA, Soto JA, Ferrucci JT. Consensus on current clinical practice of virtual colono-
scopy. AJR Am J Roentgenol 2005;184:786.
[39] Chen SC, Lu DS, Hecht JR, et al. CT colonography: value of scanning in both the supine
and prone positions. AJR Am J Roentgenol 1999;172:595.
[40] Fletcher JG, Johnson CD, Welch TJ, et al. Optimization of CT colonography technique:
prospective trial in 180 patients. Radiology 2000;216:704.
[41] Fletcher RH. The end of barium enemas? N Engl J Med 2000;342:1823.
[42] Laks S, Macari M, Bini EJ. Positional change in colon polyps at CT colonography. Radi-
ology 2004;231:761.
[43] Morrin MM, Farrell RJ, Keogan MT, et al. CT colonography: colonic distention improved
by dual positioning but not intravenous glucagon. Eur Radiol 2002;12:525.
[44] Wessling J, Fischbach R, Meier N, et al. CT colonography: protocol optimization with
multi-detector row CTdstudy in an anthropomorphic colon phantom. Radiology 2003;
228:753.
[45] Fenlon HM, Nunes DP, Schroy PC 3rd, et al. A comparison of virtual and conventional
colonoscopy for the detection of colorectal polyps. N Engl J Med 1999;341:1496.
[46] Yee J, Akerkar GA, Hung RK, et al. Colorectal neoplasia: performance characteristics of
CT colonography for detection in 300 patients. Radiology 2001;219:685.
[47] Pickhardt PJ. Translucency rendering in 3D endoluminal CT colonography: a useful tool
for increasing polyp specificity and decreasing interpretation time. AJR Am J Roentgenol
2004;183:429.
[48] Pickhardt PJ, Choi JH. Electronic cleansing and stool tagging in CT colonography: advan-
tages and pitfalls with primary three-dimensional evaluation. AJR Am J Roentgenol 2003;
181:799.
[49] Zalis ME, Hahn PF. Digital subtraction bowel cleansing in CT colonography. AJR Am J
Roentgenol 2001;176:646.
[50] Zalis ME, Perumpillichira J, Del Frate C, et al. CT colonography: digital subtraction bowel
cleansing with mucosal reconstruction initial observations. Radiology 2003;226:911.
[51] Nicholson FB, Taylor S, Halligan S, et al. Recent developments in CT colonography. Clin
Radiol 2005;60:1.
[52] Summers RM, Yao J, Johnson CD. CT colonography with computer-aided detection:
automated recognition of ileocecal valve to reduce number of false-positive detections.
Radiology 2004;233:266.
[53] Yoshida H, Masutani Y, MacEneaney P, et al. Computerized detection of colonic polyps at
CT colonography on the basis of volumetric features: pilot study. Radiology 2002;222:327.
[54] Ferrucci J, Barish M, Choi R, et al. Virtual colonoscopy. JAMA 2004;292:431.
[55] Dachman AH, Kuniyoshi JK, Boyle CM, et al. CT colonography with three-dimensional
problem solving for detection of colonic polyps. AJR Am J Roentgenol 1998;171:989.
[56] Fenlon HM, McAneny DB, Nunes DP, et al. Occlusive colon carcinoma: virtual
colonoscopy in the preoperative evaluation of the proximal colon. Radiology 1999;
210:423.
 AN UPDATE ON IMAGING OF COLORECTAL CANCER 843

[57] Macari M, Berman P, Dicker M, et al. Usefulness of CT colonography in patients with in-
complete colonoscopy. AJR Am J Roentgenol 1999;173:561.
[58] Morrin MM, Farrell RJ, Raptopoulos V, et al. Role of virtual computed tomographic co-
lonography in patients with colorectal cancers and obstructing colorectal lesions. Dis Colon
Rectum 2000;43:303.
[59] Hara AK, Johnson CD, MacCarty RL, et al. Incidental extracolonic findings at CT colo-
nography. Radiology 2000;215:353.
[60] Morrin MM, Kruskal JB, Farrell RJ, et al. Endoluminal CT colonography after an incom-
plete endoscopic colonoscopy. AJR Am J Roentgenol 1999;172:913.
[61] Lauenstein TC, Ajaj W, Kuehle CA, et al. Magnetic resonance colonography: comparison
of contrast-enhanced three-dimensional vibe with two-dimensional FISP sequences: pre-
liminary experience. Invest Radiol 2005;40:89.
[62] Ajaj W, Lauenstein TC, Pelster G, et al. MR colonography: how does air compare to water
for colonic distention? J Magn Reson Imaging 2004;19:216.
[63] Lauenstein TC, Goehde SC, Debatin JF. Fecal tagging: MR colonography without colonic
cleansing. Abdom Imaging 2002;27:410.
[64] Goehde SC, Ajaj W, Lauenstein T, et al. Impact of diet on stool signal in dark lumen mag-
netic resonance colonography. J Magn Reson Imaging 2004;20:272.
[65] Purkayastha S, Tekkis PP, Athanasiou T, et al. Magnetic resonance colonography versus
colonoscopy as a diagnostic investigation for colorectal cancer: a meta-analysis. Clin Ra-
diol 2005;60:980.
[66] Villavicencio RT, Rex DK. Colonic adenomas: prevalence and incidence rates, growth
rates, and miss rates at colonoscopy. Semin Gastrointest Dis 2000;11:185.
[67] Ajaj W, Pelster G, Treichel U, et al. Dark lumen magnetic resonance colonography: com-
parison with conventional colonoscopy for the detection of colorectal pathology. Gut 2003;
52:1738.
[68] Lauenstein TC, Debatin JF. Magnetic resonance colonography for colorectal cancer
screening. Semin Ultrasound CT MR 2001;22:443.
[69] Luboldt W, Debatin JF. Virtual endoscopic colonography based on 3D MRI. Abdom Im-
aging 1998;23:568.
[70] Meier C, Wildermuth S. Feasibility and potential of MR-Colonography for evaluating co-
lorectal cancer. Swiss Surg 2002;8:21.
[71] Hartmann D, Bassler B, Schilling D, et al. Colorectal polyps: detection with dark-lumen
MR colonography versus conventional colonoscopy. Radiology 2005;238:143.
[72] Zerhouni EA, Rutter C, Hamilton SR, et al. CT and MR imaging in the staging of colorec-
tal carcinoma: report of the Radiology Diagnostic Oncology Group II. Radiology 1996;
200:443.
[73] Low RN, McCue M, Barone R, et al. MR staging of primary colorectal carcinoma: com-
parison with surgical and histopathologic findings. Abdom Imaging 2003;28:784.
[74] Harisinghani MG, Saini S, Hahn PF, et al. MR imaging of lymph nodes in patients with
primary abdominal and pelvic malignancies using ultrasmall superparamagnetic iron oxide
(Combidex). Acad Radiol 1998;5(Suppl 1):S167.
[75] Koh DM, Brown G, Temple L, et al. Rectal cancer: mesorectal lymph nodes at MR imaging
with USPIO versus histopathologic findingsdinitial observations. Radiology 2004;231:91.
[76] Hildebrandt U, Feifel G, Schwarz HP, et al. Endorectal ultrasound: instrumentation and
clinical aspects. Int J Colorectal Dis 1986;1:203.
[77] Paty PB, Nash GM, Baron P, et al. Long-term results of local excision for rectal cancer.
Ann Surg 2002;236:522.
[78] Stipa F, Burza A, Lucandri G, et al. Outcomes for early rectal cancer managed with trans-
analendoscopic microsurgery: a 5-year follow-up study. Surg Endosc 2006;20:541.
[79] Winde G, Nottberg H, Keller R, et al. Surgical cure for early rectal carcinomas (T1).
Transanal endoscopic microsurgery vs. anterior resection. Dis Colon Rectum 1996;39:
969.
844 WALD et al

[80] Kim CJ, Yeatman TJ, Coppola D, et al. Local excision of T2 and T3 rectal cancers after
downstaging chemoradiation. Ann Surg 2001;234:352.
[81] Nagtegaal ID, Marijnen CA, Kranenbarg EK, et al. Circumferential margin involvement is
still an important predictor of local recurrence in rectal carcinoma: not one millimeter but
two millimeters is the limit. Am J Surg Pathol 2002;26:350.
[82] Improved survival with preoperative radiotherapy in resectable rectal cancer. Swedish Rec-
tal Cancer Trial. N Engl J Med 1997;336:980.
[83] Kapiteijn E, Marijnen CA, Nagtegaal ID, et al. Preoperative radiotherapy combined with
total mesorectal excision for resectable rectal cancer. N Engl J Med 2001;345:638.
[84] Camma C, Giunta M, Fiorica F, et al. Preoperative radiotherapy for resectable rectal can-
cer: a meta-analysis. JAMA 2000;284:1008.
[85] Bipat S, Glas AS, Slors FJ, et al. Rectal cancer: local staging and assessment of lymph node
involvement with endoluminal US, CT, and MR imagingda meta-analysis. Radiology
2004;232:773.
[86] Akasu T, Kondo H, Moriya Y, et al. Endorectal ultrasonography and treatment of early
stage rectal cancer. World J Surg 2000;24:1061.
[87] Beets-Tan RG, Beets GL. Rectal cancer: review with emphasis on MR imaging. Radiology
2004;232:335.
[88] Brown G, Davies S, Williams GT, et al. Effectiveness of preoperative staging in rectal can-
cer: digital rectal examination, endoluminal ultrasound or magnetic resonance imaging? Br
J Cancer 2004;91:23.
[89] Heriot AG, Grundy A, Kumar D. Preoperative staging of rectal carcinoma. Br J Surg 1999;
86:17.
[90] Schulzke JD. Does miniprobe endoscopic ultrasound have a role in the diagnostic reper-
toire for colorectal cancer? Int J Colorectal Dis 2003;18:450.
[91] Stergiou N, Haji-Kermani N, Schneider C, et al. Staging of colonic neoplasms by colono-
scopic miniprobe ultrasonography. Int J Colorectal Dis 2003;18:445.
[92] Brown G, Daniels IR, Richardson C, et al. Techniques and trouble-shooting in high spatial
resolution thin slice MRI for rectal cancer. Br J Radiol 2005;78:245.
[93] Brown G, Radcliffe AG, Newcombe RG, et al. Preoperative assessment of prognostic fac-
tors in rectal cancer using high-resolution magnetic resonance imaging. Br J Surg 2003;90:
355.
[94] Beets-Tan RG, Beets GL, Vliegen RF, et al. Accuracy of magnetic resonance imaging in
prediction of tumour-free resection margin in rectal cancer surgery. Lancet 2001;357:
497.
[95] Brown G, Daniels IR. Preoperative staging of rectal cancer: the MERCURY research pro-
ject. Recent Results Cancer Res 2005;165:58.
[96] Strassburg J. Magnetic resonance imaging in rectal cancer: the MERCURY experience.
Tech Coloproctol 2004;8(Suppl 1):s16.
[96a] Brown G, Daniels I, Norman A. MRI predicts surgical resection margin status in patients
with rectal cancer: results from the Mercury Study Group. Abstract presented at the
Radiological Society of North America Meeting 2004.
[97] Ferri M, Laghi A, Mingazzini P, et al. Pre-operative assessment of extramural invasion and
sphincteral involvement in rectal cancer by magnetic resonance imaging with phased-array
coil. Colorectal Dis 2005;7:387.
[98] Kotanagi H, Fukuoka T, Shibata Y, et al. The size of regional lymph nodes does not cor-
relate with the presence or absence of metastasis in lymph nodes in rectal cancer. J Surg On-
col 1993;54:252.
[99] Maldjian C, Smith R, Kilger A, et al. Endorectal surface coil MR imaging as a staging tech-
nique for rectal carcinoma: a comparison study to rectal endosonography. Abdom Imaging
2000;25:75.
[100] Kim JH, Beets GL, Kim MJ, et al. High-resolution MR imaging for nodal staging in rectal
cancer: are there any criteria in addition to the size? Eur J Radiol 2004;52:78.
 AN UPDATE ON IMAGING OF COLORECTAL CANCER 845

[101] Oh YT, Kim MJ, Lim JS, et al. Assessment of the prognostic factors for a local recurrence of
rectal cancer: the utility of preoperative MR imaging. Korean J Radiol 2005;6:8.
[102] Gualdi GF, Casciani E, Guadalaxara A, et al. Local staging of rectal cancer with transrectal
ultrasound and endorectal magnetic resonance imaging: comparison with histologic find-
ings. Dis Colon Rectum 2000;43:338.
[103] Hunerbein M, Pegios W, Rau B, et al. Prospective comparison of endorectal ultrasound,
three-dimensional endorectal ultrasound, and endorectal MRI in the preoperative evalua-
tion of rectal tumors. Preliminary results. Surg Endosc 2000;14:1005.
[104] Akin O, Nessar G, Agildere AM, et al. Preoperative local staging of rectal cancer with en-
dorectal MR imaging: comparison with histopathologic findings. Clin Imaging 2004;28:
432.
[105] Kim NK, Kim MJ, Yun SH, et al. Comparative study of transrectal ultrasonography, pel-
vic computerized tomography, and magnetic resonance imaging in preoperative staging of
rectal cancer. Dis Colon Rectum 1999;42:770.
[106] Kwok H, Bissett IP, Hill GL. Preoperative staging of rectal cancer. Int J Colorectal Dis
2000;15:9.
[107] Harewood GC, Wiersema MJ, Nelson H, et al. A prospective, blinded assessment of the
impact of preoperative staging on the management of rectal cancer. Gastroenterology
2002;123:24.
[108] Blomqvist L, Holm T, Nyren S, et al. MR imaging and computed tomography in patients
with rectal tumours clinically judged as locally advanced. Clin Radiol 2002;57:211.
[109] Beets-Tan RG, Beets GL, Borstlap AC, et al. Preoperative assessment of local tumor extent
in advanced rectal cancer: CT or high-resolution MRI? Abdom Imaging 2000;25:533.
[110] Madbouly KM, Remzi FH, Erkek BA, et al. Recurrence after transanal excision of T1 rec-
tal cancer: should we be concerned? Dis Colon Rectum 2005;48:711.
[111] Mohiuddin M, Marks G, Bannon J. High-dose preoperative radiation and full thickness
local excision: a new option for selected T3 distal rectal cancers. Int J Radiat Oncol Biol
Phys 1994;30:845.
[112] Sengupta S, Tjandra JJ. Local excision of rectal cancer: what is the evidence? Dis Colon
Rectum 2001;44:1345.
[113] de Anda EH, Lee SH, Finne CO, et al. Endorectal ultrasound in the follow-up of rectal
cancer patients treated by local excision or radical surgery. Dis Colon Rectum 2004;47:
818.
[114] Beynon J, Mortensen NJ, Foy DM, et al. The detection and evaluation of locally recurrent
rectal cancer with rectal endosonography. Dis Colon Rectum 1989;32:509.
[115] Lohnert MS, Doniec JM, Henne-Bruns D. Effectiveness of endoluminal sonography in
the identification of occult local rectal cancer recurrences. Dis Colon Rectum 2000;43:
483.
[116] Makela JT, Laitinen SO, Kairaluoma MI. Five-year follow-up after radical surgery
for colorectal cancer. Results of a prospective randomized trial. Arch Surg 1995;130:
1062.
[117] Novell F, Pascual S, Viella P, et al. Endorectal ultrasonography in the follow-up of rec-
tal cancer. Is it a better way to detect early local recurrence? Int J Colorectal Dis 1997;
12:78.
[118] Garcia-Aguilar J, Hernandez de Anda E, Rothenberger DA, et al. Endorectal ultrasound
in the management of patients with malignant rectal polyps. Dis Colon Rectum 2005;
48:910.
[119] Barbaro B, Schulsinger A, Valentini V, et al. The accuracy of transrectal ultrasound in pre-
dicting the pathological stage of low-lying rectal cancer after preoperative chemoradiation
therapy. Int J Radiat Oncol Biol Phys 1999;43:1043.
[120] Fleshman JW, Myerson RJ, Fry RD, et al. Accuracy of transrectal ultrasound in predicting
pathologic stage of rectal cancer before and after preoperative radiation therapy. Dis Colon
Rectum 1992;35:823.
846 WALD et al

[121] Napoleon B, Pujol B, Berger F, et al. Accuracy of endosonography in the staging of rectal
cancer treated by radiotherapy. Br J Surg 1991;78:785.
[122] Rau B, Hunerbein M, Barth C, et al. Accuracy of endorectal ultrasound after preoperative
radiochemotherapy in locally advanced rectal cancer. Surg Endosc 1999;13:980.
[123] Hunerbein M, Totkas S, Moesta KT, et al. The role of transrectal ultrasound-guided
biopsy in the postoperative follow-up of patients with rectal cancer. Surgery 2001;129:164.
[124] Morken JJ, Baxter NN, Madoff RD, et al. Endorectal ultrasound-directed biopsy: a
useful technique to detect local recurrence of rectal cancer. Int J Colorectal Dis 2006;21:
258.
[125] Kuo LJ, Chern MC, Tsou MH, et al. Interpretation of magnetic resonance imaging for lo-
cally advanced rectal carcinoma after preoperative chemoradiation therapy. Dis Colon
Rectum 2005;48:23.
[126] Chen CC, Lee RC, Lin JK, et al. How accurate is magnetic resonance imaging in restaging
rectal cancer in patients receiving preoperative combined chemoradiotherapy? Dis Colon
Rectum 2005;48:722.
[127] Peschaud F, Cuenod CA, Benoist S, et al. Accuracy of magnetic resonance imaging in rectal
cancer depends on location of the tumor. Dis Colon Rectum 2005;48:1603.
[128] Even-Sapir E, Parag Y, Lerman H, et al. Detection of recurrence in patients with rectal can-
cer: PET/CT after abdominoperineal or anterior resection. Radiology 2004;232:815.
[129] Fukunaga H, Sekimoto M, Ikeda M, et al. Fusion image of positron emission tomography
and computed tomography for the diagnosis of local recurrence of rectal cancer. Ann Surg
Oncol 2005;12:561.
[130] Moore HG, Akhurst T, Larson SM, et al. A case-controlled study of 18-fluorodeoxyglucose
positron emission tomography in the detection of pelvic recurrence in previously irradiated
rectal cancer patients. J Am Coll Surg 2003;197:22.
[131] Chessin DB, Kiran RP, Akhurst T, et al. The emerging role of 18F-fluorodeoxyglucose pos-
itron emission tomography in the management of primary and recurrent rectal cancer. J Am
Coll Surg 2005;201:948.
[132] Flamen P. Positron emission tomography in colorectal cancer. Best Pract Res Clin Gastro-
enterol 2002;16:237.
[133] Arulampalam TH, Costa DC, Loizidou M, et al. Positron emission tomography and colo-
rectal cancer. Br J Surg 2001;88:176.
[134] Meta J, Seltzer M, Schiepers C, et al. Impact of 18F-FDG PET on managing
patients with colorectal cancer: the referring physician’s perspective. J Nucl Med 2001;
42:586.
[135] Whiteford MH, Whiteford HM, Yee LF, et al. Usefulness of FDG-PET scan in the assess-
ment of suspected metastatic or recurrent adenocarcinoma of the colon and rectum. Dis Co-
lon Rectum 2000;43:759.
[136] Boykin KN, Zibari GB, Lilien DL, et al. The use of FDG-positron emission tomography
for the evaluation of colorectal metastases of the liver. Am Surg 1999;65:1183.
[137] Topal B, Flamen P, Aerts R, et al. Clinical value of whole-body emission tomography in
potentially curable colorectal liver metastases. Eur J Surg Oncol 2001;27:175.
[138] Zhuang H, Sinha P, Pourdehnad M, et al. The role of positron emission tomography with
fluorine-18-deoxyglucose in identifying colorectal cancer metastases to liver. Nucl Med
Commun 2000;21:793.
[139] Selzner M, Hany TF, Wildbrett P, et al. Does the novel PET/CT imaging modality impact
on the treatment of patients with metastatic colorectal cancer of the liver? Ann Surg 2004;
240:1027.
[140] Tzimas GN, Koumanis DJ, Meterissian S. Positron emission tomography and colorectal
carcinoma: an update. J Am Coll Surg 2004;198:645.
[141] Kalff V, Hicks RJ, Ware RE, et al. The clinical impact of (18)F-FDG PET in patients with
suspected or confirmed recurrence of colorectal cancer: a prospective study. J Nucl Med
2002;43:492.
 AN UPDATE ON IMAGING OF COLORECTAL CANCER 847

[142] Huebner RH, Park KC, Shepherd JE, et al. A meta-analysis of the literature for whole-body
FDG PET detection of recurrent colorectal cancer. J Nucl Med 2000;41:1177.
[143] Veit P, Kuhle C, Beyer T, et al. Whole body positron emission tomography/computed to-
mography (PET/CT) tumour staging with integrated PET/CT colonography: technical fea-
sibility and first experiences in patients with colorectal cancer. Gut 2006;55:68.