Colonoscopies miss more cancers on right side than left, study suggests

Colonoscopies may not be as effective at reducing deaths from certain polyps as thought, a study by researchers in Ontario suggests. Screening may help reduce deaths from colon cancer by 60 to 70 per cent, not 90 per cent.

Colonoscopies may not be as effective at reducing cancer deaths from polyps on the right side of the colon as those on the left, an Ontario study suggests.

In Monday's online issue of the Annals of Internal Medicine, Dr. Nancy Baxter, a colorectal surgeon at Toronto's St. Michael's Hospital, and her colleagues compared health records of more than 10,000 people in Ontario who died of colorectal cancer by 2003 and a control group who did not die of the disease.

In a complete colonoscopy, a doctor inserts a long, flexible tube into a patient's rectum to scan the entire colon for potentially cancerous growths.

"It prevents about two-thirds of cancer deaths on the left side of the colon, but seems to be much less and perhaps not effective at catching cancers developing on the right side of the colon," Baxter said.

"Colonoscopy is an effective intervention," Dr. David Ransohoff of the University of North Carolina in Chapel Hill wrote in a journal editorial accompanying the study.

But the evidence on the effectiveness of colonoscopy is indirect and does not support current claims of 90 per cent effectiveness and a 60- to 70-per-cent reduction in risk seems more reasonable, Ransohoff added.

"Until we have better data, we can be grateful and optimistic to have a useful intervention to offer our patients, but we should be realistic and cautious when talking with them about the magnitude of both benefits and risks," he concluded.

The 60- to 70-per-cent reduction in mortality due to colon cancer should not be considered disappointing, Ransohoff said, given that breast cancer screening has a 25-per-cent cancer mortality reduction at best and prostate cancer screening has no proven reduction.

Biological differences?

Colonoscopy also appeared to be less effective for right-side colorectal cancer in similar studies in the U.S. and Germany, Baxter and her co-authors noted.

As for why colonoscopy may not work as well at preventing deaths from right-side colon cancers, the study's authors offered three possible explanations:

  1. Some colonscopies may not cover the entire right colon.
  2. Bowel preparation may be worse in the right colon.
  3. Right and left tumours may differ biologically. For example, right-sided tumours may grow faster, or may be less likely to have a fleshy stalk and are occasionally flat, which makes them harder to identify and remove.

The researchers were unable to identify if the colonscopies were given for screening purposes in people showing no symptoms or to diagnose cancer. The study's authors tried to reduce possible bias by excluding people whose cancers were diagnosed within six months of the colonoscopy. 

Since the study used administrative rather than medical records, there is no way to judge whether bowel preparations were poor in the study, Ransohoff said.

An earlier Canadian colonoscopy study found that internists and surgeons did about 70 per cent of colonoscopies, and that these doctors missed colorectal cancers more often than when gastroenterologists did the procedure.

A randomized control trial of screening colonoscopy is in the planning stages and may provide more evidence on how well the procedure helps to detect right side tumours.

Colonoscopy is generally safe, but the procedure has a 0.2-per-cent rate of serious complications, according to the editorial.

Colorectal cancer is the second leading cause of cancer-related deaths in Canada, accounting for 10 per cent of cancer deaths.

The study included 10, 292 patients aged 52 to 90, 719 or seven per cent who had a colonoscopy. The researchers also looked at records from 5,031 controls, 9.8 per cent of whom had a colonoscopy. The patients were diagnosed between 1996 and 2001.

The study was funded by the Canadian Institutes of Health Research and American Society of Clinical Oncology.