Controversy exists as to whether 2D or 3D image displays are more suitable for initial polyp detection at CT colonography (CTC). The purpose of this study is to compare the time and distance that colorectal polyps are visualized on 2D axial and 3D endoluminal displays.
METHOD AND MATERIALS
The study set consisted of 20 proven colorectal polyps (17 adenomas, 3 hyperplastic polyps; size range, 6-25 mm, mean size, 11.8 mm) on CTC studies from 14 patients. The time and distance of 3D polyp visualization were measured for both retrograde and antergrade flight using the automated centerline for endoluminal navigation. The default (fastest) speed of 3D fly through (1.54 cm/sec) was utilized. A conservative 2D axial reading rate of 0.77 cm/sec (50% the 3D rate) was assumed for 2D time of polyp visualization. Unlike 3D, the 2D time and distance of polyp visualization were both directly related to the number of axial slices on which the polyp is seen.
Mean time and distance of polyp visualization for the 2D axial display were 1.4 ± 0.5 sec and 1.1 ± 0.4 cm, respectively. Mean 3D endoluminal time and distance of polyp visualization for retrograde flight, antegrade flight, and total (retrograde + antegrade) flight were 3.8 ± 2.2 sec and 5.8 ± 3.5 cm, 3.0 ± 2.4 sec and 4.6 ± 3.7 cm, and 6.8 ± 3.7 sec and 10.4 ± 5.8 cm, respectively. The differences between mean 2D and 3D values were significant (p < 0.001).
The opportunity for polyp detection, including both time and distance of polyp visualization, is significantly greater for the 3D endolumuinal display compared with the 2D axial display. This may help explain the increased sensitivity reported for primary 3D polyp detection in low-prevalence cohorts compared with a primary 2D approach.