Comparison of polyp distance on CT colonography between supine and prone scans using an automated path-distance...

To investigate the proximity of polyp distance on CT colonography (CTC) between supine and prone scans using an automated path-distance measurement tool and to correlate the path distance of polyps with that of colonoscopy.

Of 196 polyps in 85 patients, 98 polyps (25 <6 mm, 42 6-9.9 mm, 31 > or =10 mm) detected on supine and prone positions in 72 patients, were included. The location of each polyp, which was expressed as the distance from the tip of the rectum, was measured using an automated path-distance measurement tool and was compared between the two positions. The effect of colonic collapse on the proximity of polyp distance between the two scans was analyzed. The automated path distance of 50 polyps in the rectosigmoid colon was correlated with that of colonoscopy.

Mean difference of the automated path distances of polyps between the two positions was 2.6 cm and was not significantly different between the two positions (P > 0.05). Correlation coefficient (gamma) between the two positions was 0.9977. The difference of the distance of polyps between the groups with or without colonic collapse was not significant (P > 0.05). Automated path distance of 50 polyps in the sigmoid colon or rectum was generally well correlated with that on colonoscopy (gamma = 0.8005, P < 0.0001) and the mean difference was 5.1 cm. The mean difference increased as the polyp distance from the point of reference became further located, and reached significance when the distance was further than 30 cm from the tip of the rectum (P = 0.002).

Automated path distance of polyps matches closely between the two positions and is not influenced by the presence of collapsed segments. Polyps located 30 cm or further from the tip of the rectum on CTC do not match closely with that on colonoscopy.