Blind Spots: New study suggests that hard-to-see, flat lesions on the colon wall -- rather than polyps -- present a greater cancer danger

By Stephanie Desmon

Mar. 20, 2008 (McClatchy-Tribune Regional News delivered by Newstex) --
New findings by California researchers may change the way doctors think about colonoscopy -- a popular screening test for colon and rectal cancer.

The research suggests that flat lesions growing on the colon wall are more common than previously thought -- and are five times more likely to be cancerous than the more well-known, protruding growths called polyps. Many doctors are not as familiar with the flat lesions, which are much more difficult to spot during colonoscopy, and may not know how dangerous they can be.

"It has been thought in the past that big polyps were the big players that turned into cancer," said Dr. Peter Darwin, director of gastrointestinal endoscopy at the University of Maryland Medical Center. This and future studies "may change the way we think about cancer."

The research offers possible solutions to one mystery: why colon cancer, in some rare cases, appears not long after colonoscopy -- even when doctors found no abnormalities.

Standard practice has assumed that when no polyps are found during a colonoscopy, the patient is free of cancer and doesn't need to be rescreened for a decade -- the amount of time it is believed to take for a polyp to grow and become cancerous. But if pre-cancerous growths -- such as flat lesions -- are missed or not completely removed, cancer can develop between screenings.

Darwin noted that colon cancer is one of the few cancers that can be prevented, by removing polyps found during colonoscopy. Though the vast majority of cancers begin as polyps, some might start as these subtle, flat or even depressed lesions.

The study, published this month in the Journal of the American Medical Association, underscores the need to have a colonoscopy performed by well-trained and experienced physicians, doctors said.

"It's been shown very clearly that colonoscopy, even in the best hands, is going to miss 5 to 10 percent of abnormalities," said Dr. Durado Brooks, director of colon cancer for the American Cancer Society. "It's [the public's] perception that colonoscopy is almost a 100 percent accurate tool, and that's

ColonColon[From Page 1C]

simply not the case. It's a very good tool, but it is not a perfect tool.

"Letting the perfect be the enemy of the good is a mistake. Colonoscopy is the best tool we have."

A colonoscopy is performed by a doctor who inserts a long, flexible tube with a camera on the tip into the rectum to scan the colon for polyps or cancer. The procedure is recommended for people over age 50. Doctors also recommend it more frequently for people with a strong family history of cancer or those who have had polyps found in prior colonoscopies.

If the test comes back normal, doctors typically say the patient need not get a repeat colonoscopy for a decade. But if symptoms such as a change in bowel habits or blood in the stool appear within a few years of a colonoscopy, it could be a sign that something was missed -- and Brooks said a patient may need another procedure.

The JAMA study is among the first to find evidence of flat lesions in Americans. But studies in Japan dating to the 1980s suggest that these lesions were "common and ominous," Dr. David Lieberman, a gastroenterologist at the Oregon Health & Science University in Portland, wrote in an editorial accompanying the study.

Similar studies in the U.S. did not find the same evidence until this one, involving 1,819 patients at the Veterans Affairs Palo Alto Health Care System, where about 10 percent had these flat lesions.

Some doctors said further studies are needed to see if the lesions found among the mostly white, mostly male population in California can be extrapolated to other groups.

The doctors who treated the patients in the study received training in colonoscopy procedures used in Japan, where flat lesions are better understood. They used equipment that sprayed a blue dye on suspicious patches and made them easier to decipher.

The study also emphasized the need for the bowel to be fully emptied before the procedure is done, so that flat lesions aren't obscured.

The day after the study appeared, the American Cancer Society and other health groups endorsed two new screening tests for colorectal cancer. One is the virtual colonoscopy, which uses a CT scan to search for growths.

It is unclear whether the latest findings will have any impact on those guidelines in the future, since the virtual test is "not going to be able to detect a flat adenoma," said Dr. Michael Choti, director of the Johns Hopkins Colon Cancer Center.

Nationwide, the proportion of people older than 50 who had a colonoscopy or a similar procedure over the previous 10 years rose from 45 percent in 2002 to 56 percent in 2006, according to a recent Centers for Disease Control and Prevention survey.

Doctors say those figures are smaller than the proportion of women who get mammograms for breast cancer or men who have the PSA blood test for prostate cancer.

Colorectal cancer is a leading cause of cancer death in the United States, with nearly 150,000 new cases expected to be diagnosed this year.

Brooks said it is important to ask questions of the doctor performing a colonoscopy to learn about his skill level. He suggests asking about the doctor's polyp detection rate (most find lesions in 15 percent to 25 percent of patients); how often his scope gets all the way to the juncture of the colon and small intestine (most make it about 95 percent of the time); and how long it takes for the doctor to withdraw the scope from the colon (faster than six minutes may be too fast).

Many doctors take time entering the colon, he said, but colonoscopy is a two-way procedure: Abnormalities can also be found on the way out as the twists, turns and folds of the colon are examined from another angle.

"It does raise the importance of careful and meticulous colonoscopy ... not just rushing through it," Choti said.

Newstex ID: KRTB-0034-23924791