Colonoscopy, CT colonography or none of the above

Although a study that appeared in Lancet Oncology comparing colonoscopy and CT colonography concluded that both techniques could be acceptable for population-based screening for colorectal cancer, it may be too soon to confirm the true benefit of either approach, according to a commentary published in the July 17 edition of the American College of Physicians (ACP) Journal Club.

“Compared with flexible sigmoidoscopy or fecal blood testing, colonoscopy is more expensive and hazardous (e.g., need for bowel preparation and sedation); in addition, the ability of colonoscopy to prevent proximal colon cancer is uncertain,” wrote Ronald L. Koretz, MD, of Olive View—UCLA Medical Center in Sylmar, Calif. “CT colonography is expensive, exposes [the patient] to radiation, requires colonoscopic follow-up for intestinal lesions, and has unknown potential for harm or benefit when identifying noncolonic incidental abnormalities.”

Koretz argued that a colon cancer screening test should prevent mortality and morbidity at an affordable cost. The study from Stoop et al (Lancet Oncol 2012;13:55-64) that compared colonoscopy and CT colonography did not feature a no-screening control group, and doesn’t provide any insight into the cost-effectiveness of either test, according to Koretz. Only guaiac-based fecal testing and flexible sigmoidoscopy have high-grade evidence comparing the screening method to a no-screening control.

“Given the economic limitations in health care, we should be wary of accepting either colonoscopy or CT colonography without definitive evidence of benefit compared with doing nothing,” wrote Koretz.

Stoop and colleagues found that participation rates for colorectal cancer screening were higher when patients were invited to undergo CT colonography rather than colonoscopy. Colonoscopy, however, had a higher diagnostic yield for advanced neoplasia per participant, which made the overall diagnostic yield per invitee similar between the two techniques.

The ACP Journal Club is a monthly feature of Annals of Internal Medicine that summarizes new evidence for internal medicine from more than 130 clinical journals.