CT Colonography Gaining Favor

CT colonography stands at the crossroads between widespread adoption and not-yet-ready for prime time. Clinicians that use the technology and vendors that offer both the hardware and/or software components agree that the technology is an attractive alternative to optical colonoscopy that looks to spur increases in screening rates for colorectal cancer. But hurdles such as reimbursement, radiologist productivity in reading exams and proper bowel prep need to overcome first.

Nearly 150,000 Americans are diagnosed with colorectal cancer each year, making it the third most frequent cancer in the United States and the second most common cause of cancer deaths. While colorectal cancer has the highest mortality rate of all cancers, it can also be prevented if detected and treated early.

The American College of Radiology recommends that all 74 million Americans over the age of 50 be screened annually for colon cancer as well as younger, at-risk individuals with a strong family history. Early screening can detect and identify polyps, the most common precursor of cancer. Despite the advantages, the majority of the population avoids colorectal screening for its unpleasantness. The ACR estimates that only 10 to 12 million Americans are screened each year.

"Only 30 percent of the population that should get screened do," says Bob Beckett, global product manager, diagnostic oncology, GE Healthcare Technologies. "The reason for that is because [optical colonoscopy] is uncomfortable and invasive." For years, optical colonoscopy has been the gold standard for colorectal cancer screening at the price of being an undesirable procedure patients are reluctant to undergo.

An alternative diagnostic imaging procedure to optical colonoscopy that is gaining in clinical confidence is multislice computed tomography colonography (CTC). CTC is the process of combining multislice CT images and advanced visualization software to allow physicians to view, manipulate and examine the interior of the colon and detect polyps or tumors.

For patients, CTC is a quick, low-dose procedure that is non-invasive and does not require sedation or anesthesia. Bowel prep, however, is still necessary. It also eliminates the physical discomfort of conventional colonoscopy and associated risks, such as bleeding and perforation of the colon wall.


Of late, the push for CTC as a way to reduce the incidence of colorectal cancer has been receiving more press and interest. Recent studies have demonstrated to both the medical community and patient population that CTC is an accurate method for the early detection of colorectal polyps. A 2003 study in the New England Journal of Medicine concluded that CTC is just as effective, easier and more convenient than optical colonoscopy. The department of Defense study (of 1,233 asymptomatic patients at low risk for colorectal polyps) found that virtual colonoscopy's sensitivity was 88.7 percent for polyps 6mm and larger, 93.9 percent for polyps 8mm and larger and 93.8 percent for lesions 10mm and larger, compared with slightly lower percentages for conventional colonoscopy of 92.3 percent, 91.5 percent and 87.5 percent, respectively. The study led to the Food and Drug Administration approval of Viatronix Inc.'s V3D Colon visualization software.

But adoption is not yet widespread. "CT colonography is a pretty new technique," says Haw Loke-Gie, clinical marketing manager for Siemens Medical Solutions' CT division. "There is not much awareness in the general community in terms of the patient, referring physicians and radiologists. I think that it requires more general education and training to make people more and more aware of the technique."

Siemens got on the CTC ticket in July 2003 when it released syngo Colonography. The application is available on Siemens' Wizard dedicated CT console, as well as its Leonardo multimodality standalone workstation.

Vendors and clinicians agree that 4-slice CT scanners and above is suitable to perform CTC. Increasing the slices count equates to faster acquisition times and higher resolution images. While faster scanning times improves the overall patient experience, it does not reduce the reading time for physicians.

"With conventional methods, the patient needs to be premedicated and sedated," explains Haw. "If the physician wants to do a full colonoscopy, it takes 30 to 45 minutes. A sigmoidoscopy takes about 20 minutes and the physician is only examining the lower colon. CTC requires no sedation, five minutes to inflate the bowel, 20 seconds on each side to scan the patient, the data set is processed almost immediately and it can take up to 30 minutes to interpret."


Other multislice CT vendors that offer proprietary CTC applications include Phillips' Medical Systems and GE Healthcare. Since 2002, GE has offered its CT Colonography application for the LightSpeed and HiSpeed families of CT scanners. GE soon will release a new version of the software (not yet named) that will improve sensitivity and productivity by speeding the reading time to about 10 to 15 minutes. Beckett says that sales have already increased 20 percent this year, primarily due to more installations in mid-size facilities.

"Historically, it was mainly research and academic centers performing CTC, but more and more imaging centers are purchasing and adapting the application," notes Beckett.

Coupled with its line of Brilliance CT scanners, Phillips offers visualization software called CT Endoscopy. Clinical applications include screening for lesions using virtual colonoscopy, bronchoscopy and angioscopy. Toshiba America Medical Systems, another player in the multislice CT scanner market, has a long-term partnership to package its family of Aquilion scanners with Vital Images Inc.'s Vitrea 2 imaging software. The 3D, post-processing software runs on a workstation and includes a package for interpreting colonography studies.

Additional virtual colonoscopy software offerings include Voxar Inc.'s Colonscreen, Viatronix's V3D-Colon, ScImage Inc.'s NetraMD and TeraRecon Inc.'s virtual colonoscopy application.

Voxar recently announced that Colonscreen will be used to interpret virtual colonoscopy studies performed by the Special Interest Group in Gastrointestinal and Abdominal Radiology (SIGGAR1) in a clinical trial sponsored by the National Health Service (NHS) Health Technology Assessment (HTA) program in the United Kingdom. The clinical trial will compare CT colonography with two established alternatives, barium enema and optical colonoscopy, for diagnosis of colonic cancer in older symptomatic patients.


Most insurance companies do not cover CTC as a screening test for colonic polyps, making it more difficult to convince radiologists and gastroenterologists that the technology is ready for prime time. But the key is that it is an alternative, not a replacement, to colonoscopy.

"CTC is an alternative diagnostic procedure," says GE's Beckett. "Colonoscopy is a diagnostic tool and a treatment. CTC is not going to replace colonoscopy. As a matter of fact, it could increase [gastroenterologists'] business. If more patients get a CTC test and positive findings are a result, more patients will come in for the resection of the polyps."

And some good news may be on the horizon for reimbursement. In July, the American Medical Association created new category III CPT codes for CTC screening (0066T) and diagnostic CTC (00677) that went into effect. In April, a local health maintenance organization (Physician Plus Insurance Corp.) in Madison, Wis., became the first insurance company to reimburse for CTC. Two more Wisconsin HMOs are due to add screening virtual colonoscopy to their covered services this summer.

Reimbursement will, in turn, fuel the creation of more standards. Since the technology began rolling at medical centers on a larger scale in 2000, a lack of standards exist in both the acquiring and interpreting portions of the exam.

"Reimbursement is usually a driver for standardizing things," says Thorsten Fleiter, MD, assistant professor of radiology, University Hospitals of Cleveland in Ohio. "That will force people to standardize their exams a little more."

Fleiter believes that standards must be developed for the scanning and preparation technique as well as for the reading technique. "Training on the software is easy to do," explains Fleiter. "The main problem is to learn the whole procedure; to get an idea of what is a real finding and what isn't. Reading depends on the site and the way in which people prefer to read the exam. We found the best way to do it is looking at a combination of 3D views and multi-planar reconstructions (MPR) of the data set at the same time. The 3D view alone is limited."

Jong Kim, MD, director of radiology at Advanced Diagnostic Radiology in Cumberland, Md., agrees that CTC software is easy to use and does not require a tremendous amount of training. "However, actual interpretation involves a fair amount of learning curve," says Kim. "I do not believe too many radiologists are familiar or proficient in virtual colonoscopy. Therefore, if this procedure is to be widely adopted, it would require more training and familiarity among the radiology community. For this to happen, this procedure needs to be more time-efficient for interpreting physicians."


Reading CTC studies is time consuming for physicians. Once the data sets are acquired (anywhere from 800 to 1,600 slices depending on the scanner), they are transferred to a workstation for processing. Almost immediately, physicians can use the software tools to examine the 2D and 3D views of the colon.

"I use both 2D and 3D simultaneously," explains Kim. "As I am flying through the colon on 3D, when a potential abnormality is found, I always look at the 2D images to verify if it is real or an artifact. It is almost impossible and impractical to look at all the 2D images as a matter of routine. To do that, it would take hours per patient. However, I always read the actual CT source images as part of the examination. This allows us to look at all the other abdominal organs such as liver, spleen, kidneys and pancreas for any potential unsuspected abnormality."

Having installed Viatronix's V3D Colon six months ago, Kim says that he and his colleagues are proficient in using the software tools, but they still spend an average of 20 to 25 minutes reading each case. "During that time, I could probably read four to five MR examinations," opines Kim.

"From a patient's point of view, I am completely convinced that CTC is a winner," continues Kim. "It is better for the patient, more accurate and more patients would choose to undergo this procedure than optical colonoscopy."

The physician's point of view is a little different because CTC is a time-consuming procedure. Kim views the bowel prep as the major determinant to CTC's widespread acceptance and implementation by radiologists and gastroenterologists." The bowel prep that is currently recommended by companies is not adequate for a substantial portion of the patients," opines Kim.

Similar to optical colonoscopy, a patient undergoing a CTC exam must follow a restricted cleansing process to prepare the bowels. Since preps are essential to obtaining accurate results, insufficient methods account in longer examination reads and false-positives. "Most of your time is spent trying to distinguish fecal material from a real polyp," he says. "If the prep is excellent and the patient's colon is really clean, you can finish the examination in 8 to 10 minutes."


If physicians do find a polyp of a particular size, usually 10 mm or larger, the patient will be referred for a colonoscopy to have it removed. As a result, some question the value of CTC if patients still need another referral to undergo a conventional colonoscopy. Advocates of the technology point out that CTC is for the large fraction of at-risk screening patients. This majority typically avoids the exam due to its discomforts and associated risks. CTC is a safer, less intrusive alternative. It can offer piece of mind that nothing needs to be done, or indicate that polyps exist and must be removed.

"The majority of patients at our medical center for screening purposes would never go to a colonoscopy without an evaluation," says Cleveland's Fleiter. "If an alternative was not available, they would never do a screening for the colon. If you don't find anything in CTC, you don't have to get a colonoscopy, but you get the patient in at least. Another group of patients are those who have had several colonoscopies already and are looking for an alternative. So there are two groups. The goal of this application is to increase the number of screening tests in the population to make sure that there is a chance to find cancer."


The outlook is bright for CTC. The technology itself is getting more sophisticated and new upgrades are offering physicians with advanced visualization toolsets. "We are starting to see options now in the software where the physician can actually take the colon and lay it out in a straight line," says Bryan Westerman, PhD, manager, clinical sciences for CT at Toshiba. "But not only that, you can open it up instead of having to follow its natural curves and folds. This should help radiologists read more quickly."

Westerman also points to the integration of CAD (computer aided detection) as the wave of the future. Researchers from the University of Chicago presented a study at May's American Roentgen Ray Society meeting that showed CAD playing a potential role in the effort to speed up CTC reading times. Readers missed 42 percent of the polyps without CAD and later identified 75 percent of the missed polyps with the help of the CAD system. The addition of CAD will be another step for CTC in improving the rate of colorectal cancer.

"CTC is the future," says Advanced Diagnostic Radiology's Kim. "Colon cancer is one of the major public health issues. As with mammograms, not enough patients are getting screened. Part of the reason is that colonoscopy is more invasive and a complicated procedure. Whenever massive screening tests are performed, it is important to come up with a test that is the least intrusive and most convenient for the patient. And fiber optic traditional colonoscopy does not fall into that category."