Colon cancer is the second leading cause of cancer-related deaths in the Western world. There has been no consensus as to the optimal method for widespread colon cancer screening. While many family physicians have incorporated flexible sigmoidoscopy into their practices for routine screening of asymptomatic patients, a recent report by Lieberman and associates suggests that this modality is inadequate; specifically, that many clinically significant lesions are in the proximal colon and require colonoscopy for detection. The challenge of performing screening colonoscopy on all eligible patients is formidable. Alternative methods of assessing the bowel wall need to be investigated. New diagnostic imaging techniques are currently under investigation as screening tools. These include the use of two-dimensional computerized tomographic imaging. The disadvantage to this modality is the relatively high radiation exposure. A promising technique for virtual colonoscopy is three-dimensional (3D) magnetic resonance imaging. Pappalardo and associates present a summary of a small study on this modality in comparison with conventional colonoscopy.

Over a seven-month period, 70 patients who were referred for diagnostic conventional colonoscopy agreed to undergo magnetic resonance colonography (MRC). Indications for conventional colonoscopy included rectal bleeding (46 patients, 65.7 percent), positive fecal occult blood test (35 patients, 50 percent) and altered bowel habits (24 patients, 34.2 percent). Nine patients (12.8 percent) had a history of previous colon polyps. All patients underwent the same bowel preparation. MRC was followed within 60 minutes by conventional colonoscopy. For the MRC, patients were given intravenous scopolamine to optimize colonic distention and alleviate spasm. They were placed in a supine position, and their colon was filled with 1,500 to 2,000 mL of water containing a solution of gadopentetate dimeglumine. When the contrast material reached the cecum, the 3D colon imaging data were acquired. Patients were required to hold their breath for 30 seconds. Each MRC examination lasted 20 to 25 minutes, excluding the time needed for administering scopolamine, placing the rectal tube and instilling the gadolinium enema. Postprocessing took 60 to 90 minutes, and image interpretation took 40 to 45 minutes.

MRC identified eight patients as negative for lesions. Conventional colonoscopy identified five of these eight MRC–negative patients as negative and three as having lesions. Overall, MRC detected 125 colonic lesions in 54 patients, including one false-positive lesion. Patients tended to find MRC easier to tolerate than conventional colonoscopy. The authors concluded that MRC could be useful in screening programs of patients at normal risk for colon cancer.

editor's note: The debate over the best method of screening the population for colon cancer is ongoing. The Lieberman report is concerning. An accompanying editorial by Podolsky suggested that we “go the distance” in screening the entire colon. Even if physicians, patients and health plans were motivated to ensure this type of primary screening, it is unlikely that we have the ability to do so. MRC may offer a method of performing screening on patients followed by conventional colonoscopy in those found to have lesions. We need larger studies and well-done cost-effectiveness analyses on this new tool before a complete assessment can be made.—j.n.