Virtual Colonoscopy Misses Nearly One Third of Lesions

Charlene Laino

October 16, 2003

Oct. 16, 2003 (Baltimore) — Using current technologies, virtual colonoscopy is not adequate as a screening tool, say researchers whose study showed that the imaging technique missed 27% of colorectal lesions, both precancerous polyps and colon cancers.

The meta-analysis of data from 16 studies showed that virtual colonoscopy missed 18% of lesions larger than 1 cm, said Aaron A. Link, MD, a resident at the University of Michigan in Ann Arbor.

"That's almost one in five patients with large lesions, which is unacceptable," he told Medscape. "These are patients at high risk, and the screen could give them a false sense of security."

Dr. Link presented the findings here Wednesday at the 68th annual scientific meeting of the American College of Gastroenterology.

While conventional endoscopic colonoscopy is the gold standard for the detection of colorectal cancer and precancerous polyps, many patients prefer virtual colonoscopy, also called computed tomography (CT) colonography or magnetic resonance (MR) colonography, Dr. Link said.

To determine whether the technique has any viability as an option for colorectal screening, the researchers performed a systematic review of the literature from 1966 to March 2002. English language studies were selected for analysis if they evaluated at least 10 patients using CT or MR colonography with conventional colonoscopy as the reference standard for the detection of colon polyps and cancers.

Nearly 450 studies were identified by the literature search, of which 16 were appropriate for the analysis, Dr. Link said. Of these, 11 examined CT-based screening, and five used MR colonography.

On a per-patient basis, the mean sensitivity of virtual colonoscopy for detecting colorectal lesions was 73%, and the specificity was 82%, Dr. Link said.

For CT colonography, the mean sensitivity and specificity were 75% and 80%, respectively, while for MR colonography, the mean sensitivity and specificity were 83% and 84%.

On a per-lesion basis, sensitivity remained low at 69%, although it improved to 96% when lesions less than 1 cm were excluded, he said.

"We'd rather not miss anyone but if we get a false-positive that's not so bad," Dr. Link said. "What we don't want is to miss anyone with a brewing cancer."

Researchers who participated in a panel discussion of emerging technologies in gastrointestinal diagnosis and treatment said that both sensitivity and specificity are problematic and preclude the use of virtual colonoscopy as a screening tool.

Panel comoderator Girish Mishra, MD, a gastroenterologist at Wake Forest University School of Medicine in Winston-Salem, North Carolina, said, "Virtual colonoscopy is not ready for prime time right now."

Comoderator Beth Schorr-Lesnick MD, FACG, assistant clinical professor of medicine at Albert Einstein School of Medicine in the Bronx, New York, noted that many patients fear conventional colonoscopy. "They're afraid of the laxative and the pain, and they fear the findings," she told Medscape.

Douglas K. Rex, MD, FACG, professor of Medicine at Indiana University School of Medicine in Indianapolis and president-elect of the American College of Gastroenterology, put it bluntly: "They are afraid of a long tube being put up the rectum. Most people don't even want to think about their rectum," he said to laughter.

He noted that virtual colonoscopy still requires a laxative prep, something most patients don't realize. "We're trying to figure out how to do it without a laxative," he added.

Also a conventional colonoscopy still needs to be performed if the findings are positive, Dr. Rex said.

But the biggest problem with virtual colonoscopy, as shown by the meta-analysis, is that it is just not accurate. "A lot of studies show virtual colonoscopy doesn't work," Dr. Rex said.

ACG 68th Annual Scientific Meeting: Abstract 701. Presented Oct. 15, 2003.

Reviewed by Gary D. Vogin, MD


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: