Broader, and More Expensive, Test Needed to Spot Colon Cancers, Studies Say
By DENISE GRADY July 20, 2000 New York Times
Copyright New York Times 2000

The test most commonly recommended to screen healthy adults for colorectal cancer misses too many precancerous growths and should be replaced by a more extensive procedure that examines the entire colon, doctors are reporting today.

Both tests require inserting long tubes equipped with cameras into the rectum to search for abnormal tissue. The difference lies in how far the tubes go.

The more common test, sigmoidoscopy, reaches only about two feet into the colon and is generally used to screen people 50 and older with an average risk of colon cancer. The more thorough procedure, colonoscopy, probes the full length of the colon, 4 to 5 feet, and is usually reserved for people with a higher risk, like those with blood in their stool, a history of intestinal polyps or a family history of colon cancer.

Colorectal cancer kills 56,000 Americans a year, and is the second leading cause of cancer death, after lung cancer. Doctors say many of the deaths could be prevented by screenings. Unlike the tests for breast and prostate cancer, which find cancers that have already formed, procedures that examine the colon can actually detect abnormal tissue and remove it before it turns cancerous.

Sigmoidoscopy, which is cheaper and easier to perform, has been used for screening on the optimistic theory that if no abnormalities were seen in the lower colon, none were likely to be found higher up.

But that theory is contradicted by two studies being published today in The New England Journal of Medicine, which included a total of more than 5,000 healthy people screened by colonoscopy. One study, which involved more than 3,000 patients, is the largest study to date of the procedure. Both studies show that it is not safe to assume that the upper colon is healthy just because the lower third looks normal. The studies found that half the patients who had precancerous lesions in the upper colon had nothing abnormal lower down. If those patients had had only sigmoidoscopy, they would have mistakenly been given a clean bill of health and left with dangerous, undetected growths high in the colon.

The studies confirm many doctors' suspicions about sigmoidoscopy, said Dr. Daniel K. Podolsky, chief of gastroenterology at the Massachusetts General Hospital, who wrote an editorial that accompanied the two articles. Dr. Podolsky said many experts liken sigmoidoscopy to screening women for breast cancer by doing mammograms on only one breast. He urged that doctors recommend a colonoscopy to all patients 50 and older, and that insurers pay for the test. He also said that when the difference between the tests is explained to them, many patients prefer a colonoscopy.

Medicare and most insurers do not cover colonoscopy as a screening test, and pay for it only for high-risk patients. Sigmoidoscopy costs about $200, Colonoscopy $1,200 to $1,500.

An insurance industry spokesman, Dr. Donald Young, medical director of the Health Insurance Association of America, said that guidelines for coverage should be re-examined, but that it would be premature to change the rules until experts can assess all the information.

Dr. David A. Lieberman, director of one of the studies and chief of gastroenterology at Oregon Health Sciences University in Portland, said, "These data do support offering colonoscopy as a screening option."

Dr. Robert Smith, director of cancer screening for the American Cancer Society, said that patients should consider the findings, but added that "quite a lot more needs to be learned about implementing it at the community level in terms of cost, safety and acceptability before we could recommend it as routine."

Dr. Barnett Kramer, senior medical scientist in the division of cancer prevention at the National Cancer Institute, said, "This is important enough information from both of these studies that it should be incorporated into the discussion with people who are pursuing their options about screening."

But Dr. Kramer was also cautious about offering a blanket recommendation that everyone 50 and older have the test. For one thing, he said, the studies were not designed to determine whether colonoscopy actually prevented deaths from colorectal cancer. In addition, colonoscopy has a low but slightly higher risk than sigmoidoscopy of causing bleeding or injuring the colon, he said.

In Dr. Lieberman's study, 10 patients out of 3,121 (0.3 percent) had serious complications, including 6 cases of bleeding and 1 heart attack.

In addition, colonoscopy, unlike sigmoidoscopy, often requires sedation, which can have side effects and means that patients cannot drive themselves home after the procedure. Some people, given the nature of the test, prefer sedation.

Dr. Kramer also said that it would not be possible now to offer colonoscopy to everyone 50 and older, because the test has to be performed by a specialist in gastroenterology, and there are not enough specialists to go around. Dr. Smith agreed, and added that it was important to ensure that specialists were available to perform the test on high-risk patients who need it most. Unlike colonoscopy, sigmoidoscopy can be done by a family practitioner, internist or nurse practitioner.

Despite their potential to save lives, the screening tests for colon cancer are not popular: all adults 50 and older are candidates for a screening, but fewer than a third are tested.

In addition to the unpleasant aspects of the tests, most patients dread the preparation, which involves cleaning out the colon. Sigmoidoscopy requires enemas, and colonoscopy requires a liquid diet for 1 to 3 days until the night before the test, when patients must drink a powerful laxative solution that purges the colon.

"The statistics on compliance with colon cancer screening represent a tragedy," Dr. Lieberman said. "They are much lower than for breast and cervical cancer, and yet colon cancer kills far more people and is far more preventable. But we don't like conversations about our bowels. There's been a problem with patients requesting screening and with physicians offering it. A Gallup poll of the public last year said that 80 to 90 percent said their doctors had not talked to them about colon cancer screening."

Dr. Lieberman and his team conducted their study at 13 Veterans Affairs medical centers, performing colonoscopy on 3,121 patients ages 50 to 75, with an average age of 62.9 years; 96.8 percent were male, and 83.6 percent were white. One percent of the patients had cancer, and 10 percent had growths in the colon that Dr. Lieberman called "serious," meaning abnormalities that could turn cancerous. Patients with growths in the lower part of the colon -- the part reached by sigmoidoscopy -- were more likely than those with no growths in that region to have abnormalities higher up in the colon as well. But 66 with healthy tissue below had dangerous growths above, which would have gone undetected if those patients had been given only sigmoidoscopy.

Findings were similar in the second study, directed by Dr. Thomas F. Imperiale, of the Indiana University Medical Center in Indianapolis. That study included nearly 2,000 healthy patients screened by colonoscopy.
Copyright New York Times

Image Copyright New York Times