Julia Draznin Maltzman, MD
The Abramson Cancer Center of the University of Pennsylvania
Last Modified: February 29, 2004
Colorectal cancer is both a common and serious disease. The risk of developing colon cancer is influenced by many factors including genetics, environment, and previous medical history. Approximately 148,000 new cases of colon cancer are diagnosed each year in the US. About 40% will die of the disease; accounting for about 10% of all cancer deaths. In fact, in the US, colon cancer is only second to lung cancer as a cause of cancer death. Despite these unfavorable statistics, this disease can be completely prevented if polyps that grow into tumors are caught early enough and are removed so that the tumor never develops. The National Cancer Institute and the American Cancer Society urge people over the age of 50 to be screened for polyps every year.
There are four standard screening exams: fecal occult blood test, flexible sigmoidoscopy, a double contrast barium enema, and a conventional colonoscopy. There is little consensus among oncologists, gastroenterologists, surgeons, and primary care specialists as which modality to use and at what frequency. In general, the fecal occult blood test only has about 10% sensitivity. Furthermore, it is a test which finds blood in the stool caused by an existing cancer and not used for detecting precancerous lesions such as polyps that may potentially turn into cancer. A flexible sigmoidoscopy has 50% sensitivity, as it is only able to examine part of the colon. Critics have compared this to getting a mammogram done of only one breast. A double contrast barium enema, has limited sensitivity – 50%, and has little interest among patients and physicians alike. It is an uncomfortable, time consuming test. Colonoscopy, although has high specificity (90%) and is an excellent test for evaluating symptomatic patients, it is not very practical as a screening tool for large populations. It is expensive, requires sedation, and many patients refuse screening with this procedure.
Virtual colonoscopy is a new technique using CT scan technology that is non-invasive, is preformed world-wide, and is just as accurate in its ability to detect "significant" polyps (over 1 cm), as is a conventional colonoscopy. Furthermore, it can image the entire colon despite possible, strictures, spasms, or anatomical switchbacks.
Virtual colonoscopy is a rapidly developing technique that holds much promise and may soon be available for prime time use. It is safer, more economical, and less intrusive than conventional diagnostic tests for colon cancer. This procedure allows the physician to look inside the body without having to insert a long tube into the colon as in conventional colonoscopy. Also, this procedure does not require the discomfort of filling the colon with liquid barium as with a barium enema.
Virtual colonoscopy combines the techniques of a helical CT Scan with sophisticated image processing computers along with skilled radiologists to recreate and evaluate the inner surface of the colon. The CT scan provides the x-ray images and the image processing computers create the 3-dimentional, also called volumetric, image for final interpretation. This test, like a colonoscopy looks for small polyps, which can be pre-cancerous lesions. To enable the radiologist to find small polyps in the colonic labyrinth, the colon must be free of all fecal material. Colon cleansing, also called a colon-prep, is achieved in the same way one prepares for a conventional colonoscopy.
On the day of the exam, the patient lies on the CT scan table and a thin tube about the size of a rectal thermometer is placed into the patient's rectum. This tube delivers either air or carbon dioxide into the colon that is necessary to distend the bowel and allow the polyps to stand out from the normal bowel surface. CT scan images are collected once as the patient lies down on his/her back and once on the stomach. The entire procedure takes about ten minutes. Because the colon is filled with air, some patients may experience mild bloating and cramping; most patients, however, find the examination to be quick and not painful.
Virtual colonoscopy takes much less time than does a conventional colonoscopy and does not require sedation. It does still require laxatives and the insertion of a probe to push air into the rectum to allow for better visualization. Virtual colonoscopy does have some distinct advantages. The ability to reconstruct a 2- and a 3-dimentional picture of the inside surface of the colon using CT scans and high powered computers, allows for image manipulation and better viewing angles and re-viewing even after the procedure is complete. Physicians can return and review images days, weeks, months or even years after the completion of the test. During a conventional colonoscopy, the physician must spot cancers or pre-malignant lesions during the procedure itself. Opponents may argue that smaller polyps (less than five millimeters wide) are too tiny and maybe missed by virtual colonoscopy where as they would be picked up by a direct visualization enabled by a conventional colonoscopy. However, new multi-detector technology has greatly improved resolution and detection ability. As technology evolves, the ability to increase CT scan resolution and lower the limits of detection will improve.
The colon can be tortuous and can bend onto itself a number of times. This anatomy makes finding a small polyp or a cancerous lesion in the colon very difficult by both colonoscopy techniques. Sophisticated software is being developed that could smooth these switchbacks and straighten the image of the colon making the hunt for polyps easier and more accurate.
Proponents of the conventional colonoscopy method argue that it offers the physician the opportunity to remove the suspicious polyp while he/she inspects the colon thus obviating the need for another procedure and another colon prep. A recent proposal has been to create a multi-disciplinary approach to colon cancer diagnosis. In this model a patient who has pathology found on a virtual colonoscopy would be directed to the gastroenterology suite immediately for a conventional colonoscopy with biopsy. If the two tests are done the same day, a second colon prep is not required. One of the strongest arguments against virtual colonoscopy is that flat lesions are more likely to be missed by this technique. Although, this is true, these lesions are less common and pose the same challenge of diagnosis in conventional colonoscopy.
Currently there are only two indications for a virtual colonoscopy – a failed conventional one due to an occlusion that does not allow the colonoscope to pass through, or a frail and elderly patient who is unable to withstand a conventional colonoscopy.
A recent study published in the New England Journal of Medicine directly compared a conventional colonoscopy with a virtual one. Participants included over one thousand people with no symptoms or previous colon cancer history. The mean age of the participants was 58 years. The vast majority of the participants were at average risk for colon cancer with only 32 patients having a significant family history. Following a bowel prep with laxatives and a 24-hour fast, everyone underwent both a virtual colonoscopy that produced both 2- and 3-dimentional images, and a conventional colonoscopy by experienced gastroenterologists. The gastroenterologists did not know the results of the virtual colonoscopy done earlier in the day. At the end of both procedures, the participants were asked to fill out a one-page questionnaire commenting on comfort, convenience, and preference of procedure.
A total of 1,310 polyps were found in 622 patients. Most polyps were 5 mm or less and were thought not to have much clinical significance. A smaller portion, 344 polyps, were 6 mm or larger, and 210 of these were adenomas with 2 malignancies.
Overall, the sensitivity of virtual colonoscopy for adenomatous polyps was 93.8% for polyps at least 10 mm in diameter and 88.7% for those at least 6 mm. The sensitivity for conventional colonoscopy was 87.5% and 92.3% for the two polyp sizes respectively. The difference between the two modes of colonoscopy was not deemed statistically significant. Two cancerous polyps were found and both were identified by the virtual method. Traditional colonoscopy detected only one of the two malignant polyps. The second cancerous polyp was seen only after its location was revealed by virtual colonoscopy.
Analyzing the participant questionnaires, 54.3% said that virtual colonoscopy was more uncomfortable, where as 38.1% noted that the conventional technique was less comfortable. (7.6% were undecided). By contrast, 68.3% found virtual colonoscopy to be more convenient versus only 24.1% thought the conventional method was more convenient.
As an interesting sideline, virtual colonoscopy also resulted in cancer findings outside of the colon that were potentially significant. Unsuspected extracolonic cancers were found in five patients: two lung cancers, one kidney cancer, one lymphoma, and one ovarian cancer.
In an editorial that accompanied the paper, note was made that the participants in this study were asymptomatic individuals. A more powerful comparison of these two techniques may include a comparison of the symptomatic patients.
There is currently no billing code category under which insurance providers will reimburse for a virtual colonoscopy. This procedure could cost the consumer several hundreds of dollars. A colonoscopy costs about $1,500 and is usually covered by third party insurers for indications such as bleeding.
Despite these promising results, virtual colonoscopy faces a number of challenges. Hardware and software that can perform the necessary 2- and 3-dimentional analysis must be more readily available. Furthermore, radiologists need to be trained in using the new methods and interpret them properly. Additionally, radiologists need to work closely with gastroenterologists to facilitate polyp biopsy and removal on the same day for patients who are found to have lesions by the virtual technique. This would obviate the need for an additional colon prep and therefore be much more pleasant to the patient. Finally, third party insurers must offer reimbursement for the procedure to make it accessible for the general public.