Wednesday, April 14, 2004
BY JONATHAN MAZE
Of The Post and Courier Staff

Charleston, SC
As invasive and uncomfortable as it may be, a traditional colonoscopy -- rather than reliance on computer-generated imaging -- is still a patient's best bet for early detection of deadly colon cancer, according to a new study led by researchers at the Medical University of South Carolina.

Many hoped that a "virtual colonoscopy" would become the better weapon against cancer and encourage more people to get screened.

The exam uses a pen-sized probe to take a computerized image of the abdomen that doctors can use to find tumors. The traditional exam, by contrast, detects and removes tumors using a long, flexible scope the size of a garden hose inserted through the rectum.

The virtual colonoscopy's results varied widely among participants in the study, published in this week's Journal of the American Medical Association, and its authors concluded that the procedure isn't yet ready for widespread use.

The procedure detected large suspicious growths, those of more than 10 millimeters, in 55 percent of the patients who had them, and 39 percent of smaller growths of more than 6 millimeters. The conventional exam found growths, regardless of the size, practically each and every time.

The results stand in direct contrast with other studies, including one that gained widespread publicity in December, that said the virtual colonoscopy was more effective than the traditional treatment. That study helped spur imaging centers around the country to begin offering the exams.

"Ours is just like a bucket of cold water to say, 'Let's step back a little bit before everybody opens a colonic scanning unit and people open their checkbook,'" said Dr. Peter Cotton, director of MUSC's Digestive Disease Center and the study's lead author.

The MUSC study looked at what would happen if virtual colonoscopies were put in ordinary practice situations at multiple medical centers, rather than performed at a few centers by experts under controlled conditions.

The study looked at more than 600 procedures performed at nine different medical centers, eight of them in the United States and one in Great Britain.

Participating radiologists had to have performed at least 10 virtual colonoscopies. But only one center, Wake Forest University, had radiologists with more extensive experience.

That center had the best results by far, detecting 82 percent of patients with one or more smaller growths. Combined, the other eight centers detected only 24 percent of the polyps.

Cotton and others were disappointed in the findings.

Colon cancer is the second-leading cause of cancer death in the country, killing 57,000 people a year. The disease can be cured if caught early, and experts widely recommend that all people over age 50 get tested once every 10 years.

Because of that recommendation, and the publicity about colon cancer in general, the number of colonoscopies more than doubled in South Carolina between 1999 and 2002, according to state figures. With more than 45,000 of the exams performed in 2002, it is by far the most common outpatient surgery performed in the state.

The result has led to an explosion in the number of outpatient facilities devoted to the procedure.

Despite these figures, experts say that only 30 percent of South Carolinians over 50 get any kind of test for colon cancer. Nationally, according to the American Cancer Society, only half of people over 50 get a test for colon cancer.

"We were disappointed," Cotton said. "We're all in favor of developing less invasive methods for doing things that need to be done. We were really surprised that the results were what they were."

A big reason that people stay away from colonoscopies is the invasive nature of the traditional procedure, which, unlike the virtual colonoscopy, requires that a patient be sedated.

Dr. Pamela Morris, chief medical officer at prevecare on Daniel Island, agreed that the traditional procedure is frightening for many patients.

Prevecare, which focuses on preventive medicine, is one of the local facilities that performs virtual colonoscopies. Morris said that the December study proved that the virtual colonoscopy is an accurate screen for the disease. She said the procedure's effectiveness depends on the type of software used, the technology the doctors use and what they do to prepare the patient.

Still, others say that the variation in the results requires that doctors use caution.

Dr. David Ransohoff, an epidemiologist at the University of North Carolina, said in an editorial accompanying the Medical Association article that the biggest problem for the virtual colonoscopy is that it may be implemented too widely without regard to technological problems that affect the test's sensitivity.

"The differences between what virtual colonoscopy can do and what it will do if applied in ordinary practice circumstances are so great that physicians must be cautious," Ransohoff said. "There are many important steps yet to be taken in learning how to implement the technology appropriately."

Still, experts aren't writing off the virtual colonoscopy. Cotton said that with time, improvements in technology and software, as well as training of radiologists, the procedure can be a more effective exam when used by doctors.

"We're certainly hopeful that it can improve," he said. "I'm not out to destroy the procedure. I'm very keen on the idea of less-invasive procedures that could get more patients screened. But it needs to improve, and very well may improve. CT scanners are getting better, software is getting better and radiologists probably need some extra training to do this."

Jonathan Maze covers health care and nonprofits. Reach him at 937-5719 or jmaze@postandcourier.com.