In June 2016, the influential U.S. Preventive Services Task Force published final updated guidelines on screening for colorectal cancer. It expanded the list of screening options for people at average risk, with the goal to encourage more testing. In its previous guidelines (2008) and in its draft update in November 2015, the Task Force recommended colonoscopy, sigmoidoscopy, and two types of stool tests: fecal immunochemical test (FIT) and fecal occult blood test (FOBT). To the surprise of many experts, the final guidelines added two more tests—CT colonography and a FIT-DNA stool test (Cologuard)—for which the evidence about benefits and risks is more limited.
We discussed the unanswered questions about Cologuard last year. Here we’ll focus on CT colonography.
Many people are attracted to CT colonography (popularly called “virtual colonoscopy") because...
- Unlike colonoscopy, it doesn’t require the insertion of the long flexible scope through the rectum up into the colon (though a short tube is inserted into the rectum to inflate the colon). Instead, the colon is visualized by computed tomography (CT), a special type of X-ray imaging test.
- Sedation is typically not needed.
- Since the test is largely noninvasive, there’s very little risk of complications.
It may sound like a great alternative, but...
- While a number of studies on CT colonography have found that it is good at identifying cancers and larger polyps, it may miss smaller polyps. That’s why it is recommended that the test be repeated every five years, rather than the 10-year interval recommended for colonoscopy. (Colorectal cancer usually starts in adenomas and some other less common types of polyps, though the vast majority of these growths are benign and will never progress to cancer.)
- You have to clean out your colon before the test by fasting and taking a strong laxative, just as you would for a colonoscopy. In addition, inflation of the colon during the procedure may cause discomfort.
- Doctors can’t take a biopsy or remove polyps during CT colonography as they can during a colonoscopy. If polyps are detected, a colonoscopy needs to be done to remove them (usually on another day, requiring another bowel prep).
- Unlike colonoscopy, about half of all CT colonography exams detect suspicious growths and other possible abnormalities outside the colon. That’s often promoted as a plus, but it usually isn’t. The overwhelming majority of such findings are not cancer and present no threat but may require a biopsy (which is invasive and can have complications), followed by further testing and possibly unnecessary treatments. “Given the frequency with which these incidental findings occur, it is difficult to accurately understand the overall balance of benefits and harms of this screening test,” according to the Task Force’s final guidelines.
- Like any CT scan, CT colonography exposes people to radiation. That’s worrisome because the test has to be repeated periodically, and radiation exposure has a cumulative effect. Americans are already being exposed to increasing amounts of radiation from medical scans. As the final guidelines state, “radiation-induced cancer is a potential long-term concern with repeated use of CT colonography.”
Bottom line: The Task Force noted these disadvantages, but it still included CT colonography as an option in its final recommendations. In contrast, its earlier draft guidelines concluded that there was insufficient evidence to recommend CT colonography (or Cologuard) for routine screening, citing “greater uncertainty” about its “net benefits.” We think that was better advice.
It is laudable to encourage more people to be screened for colorectal cancer by giving them additional testing options. The key message is just to get screened, period. But while there is no “best” test, at the very least the Task Force should have made clear that, until further studies and more data are available, the established tests are preferable to CT colonography (and Cologuard).
For more about your options, see our article Do You Really Need a Colonoscopy?