Report from Stanford MDCT: Trial establishes equivalence between 64-slice CT colonography and optical colonoscopy for colorectal cancer screening

June 15, 2007

A large single-center prospective trial has established that CT colonography and optical colonoscopy are both about 95% sensitive as a screening test for adenomas 6 mm and larger in the colon.

A large single-center prospective trial has established that CT colonography and optical colonoscopy are both about 95% sensitive as a screening test for adenomas 6 mm and larger in the colon.

CT section chief Dr. Christoph Becker and colleagues at the University of Munich in Germany recruited 320 asymptomatic adults over the age of 50 for same-day colon cancer screening. They used a 64-slice Siemens Sensation 64 CT scanner, then optical colonoscopy with digital endoscopy. The prospective study also tested the diagnostic power of the fecal occult blood test and M2 pyruvate kinase.

Dr. Anno Graser, director of the University of Munich's CT colongraphy research group, presented unpublished results at the Stanford Multidetector-Row CT Symposium Thursday in San Francisco.

Becker and Graser, both experienced readers, evaluated the CT studies using primarily 3D reconstructions with 2D scans for correlation and problem solving. Images were acquired with 0.6-mm collimation and reconstructed with 0.75-mm overlapping slices.

All patients adhered to full bowel preparation. Bowel distention was performed using carbon dioxide with patients in prone and supine positions. They were exposed to a dose of 140 mA at 120 kV. Dose modulation reduced the effective dose about 35%, Graser said.

CT imaging and endoscopic colonscopy both performed well. Overall, 367 polyps were detected. The sensitivity of CT colonography was 93.4% for large adenomas (>10 mm), 94.7% for medium adenomas (6 to 8 mm), and 82.8% for small adenomas (<6 mm). The sensitivities of optical colonoscopy for the three categories were 96.4%, 97.4%, and 93%, respectively. CT missed two flat adenomas, while endoscopic screening overlooked one large, two intermediate, and eight small adenomas.

The specificity of CT colonography was about 95% for medium and large polyps, but this was still lower than optical screening. When performed by a skilled physician, this technique rarely misses medium and large polyps in a clean colon that has been properly prepared, Graser said.

The laboratory tests were washouts for diagnosing colon cancer. The overall sensitivity of the fecal occult blood test for large, medium, and small adenomas was about 10%, compared with 17% for the M2 pyruvate kinase test, Graser said.

Graser was encouraged by the CT results, especially in light of influential studies published in the Journal of the American Medical Association in 2004 and The Lancet in 2005 that concluded that sensitivity of CT colonography was only slightly higher than 50%.

"This underlines that CT colonoscopy is ready for prime time. The reality of CTC is that its sensitivity to colon cancer is over 90%, and we think this can be reproduced," he said.

For more information from the Diagnostic Imaging archives:

Multi-energy CT differentiates polyps from fecal matter

Low-dose CT, 3D tools benefit colon screening

Report from RSNA: Alternative bowel preps prove useful for CT colonography