Virtual colonoscopy has captured the
attention of physicians and patients as a potential alternative method for
colorectal screening in average-risk individuals. Computed tomography (CT)
colonography was the first virtual method introduced and continues to be
the most widely available. The examination is performed using a helical CT
scanner, and the quality of the examination depends on the cleanliness of
the patient's colon. The patient is required to perform a full bowel
preparation with oral phosphosoda. The patient's colon is then insufflated
with room air, and the scan is performed in supine and prone positions
during a single breath. Helical CT scanning is usually performed using
5-mm collimation with reconstruction intervals of 2 to 3 mm; however,
newer methods allow higher resolution.[22*] Analysis of the
images can be performed in two-dimensional or three-dimensional format;
the three-dimensional viewing format simulates the endoluminal view of
conventional colonoscopy. Magnetic resonance colonography uses a technique
similar to that of CT colonography; however, colonic distension is
achieved using a saline-gadolinium rectal enema. Because magnetic
resonance is a motion-sensitive modality, control of bowel peristalsis is
necessary and requires administration of an intravenous antispasmodic
medication. Although magnetic resonance avoids the radiation exposure
required for CT scanning, magnetic resonance methods have been found to
have significant disadvantages of cost and complexity(intravenous
medications, contrast enemas) and have not yet achieved use as widespread
as that of CT.[22*]
Initial studies suggest that virtual CT
colonography can detect polyps 10 mm in size or larger with a sensitivity
of 90%[23**] . In a head-to-head study of conventional
colonoscopy versus virtual colonoscopy in 100 consecutive individuals,
Fenlon et al.[23**] showed that CT colonography detected
all cancers (n = 3) in the population, with sensitivity for polyps
dependent on polyp size (91% for polyps > 10 mm, 82% for polyps 6 to 9 mm,
and 55% for polyps </=5 mm). Other studies in the United States have
corroborated sensitivity for CT colonography ranging from 75 to 90% for
polyps larger than 10 mm,[24, 25] however, an Italian study
conducted by Spinzi et al.[26] reported lower test
sensitivity-88% for cancers and 62% for polyps larger than 10 mm.
Some investigators have suggested that
screening virtual colonoscopy may become like screening mammography and
eventually replace conventional endoscopic tests as the first-line
screening test for colorectal cancer.[22*] In a survey of
primary care physicians and patients, Angtuaco et al.[27]
found that 76.6% of potential patients preferred the idea of virtual
colonoscopy over conventional colonoscopy, 82.3% of patients believed that
they would be more willing to comply with recommendations for colorectal
cancer screening, and 62% of physicians would refer more patients for
screening if it could be performed via virtual colonoscopy. Advocates of
virtual colonoscopy emphasize the absence of sedation (and, thus, no
requirement for recovery time) and the shorter length of time required to
perform the test as major advantages over conventional colonoscopy.
However, in a study of patients who actually underwent CT colonography
followed by conventional colonoscopy, Akerkar et al.[28*]
found that patients reported more discomfort after virtual colonoscopy and
preferred conventional colonoscopy overall.
At present, CT colonography continues to
have several limitations. First, it still requires that the patient
perform a bowel preparation, which is believed to be one of the major
deterrents for patient participation in endoscopic colorectal screening.
The virtual examination can be compromised by retained fluid or stool,
which can result in either false-negative or false-positive readings. In
addition, CT colonography has the potential for generating more procedures
for follow-up of incidental extracolonic findings such as ovarian, renal,
liver, and lung lesions, which have been reported in 15 to 41% of
patients.[29*] Current research in techniques of stool tagging
suggests there may be a way to avoid preprocedure bowel cleansing in the
future; however, this technology is not yet readily available in most
centers. The cost of virtual colonoscopy has not been standardized.
However, a cost-effectiveness analysis by Sonnenberg et al.[30*]
comparing virtual colonoscopy with conventional colonoscopy showed that,
in order for virtual colonoscopy to be cost-effective for colorectal
cancer screening, its cost must be 54% less than conventional colonoscopy,
or compliance rates must be 15 to 20% better than those for conventional
colonoscopy.
Virtual colonoscopy has limited ability to
detect small polyps, flat adenomas, or other mucosal lesions. Proponents
of virtual colonoscopy have questioned the utility of the current practice
of endoscopic removal of all polyps, regardless of size. Population
studies suggest that although adenomas occur in approximately 50% of
individuals older than 50 years, only 3% of all adenomas become malignant.
It has been proposed that targeting adenomas larger than 10 mm may be a
way of limiting polypectomy to high-risk lesions. However, new information
about the prevalence and significance of flat adenomas challenges this
assumption.
