Patricia M. Carrascosa MD, Presenter: Nothing to Disclose

Carlos Capuñay MD, Abstract Co-Author: Nothing to Disclose

Elba Martin Lopez, Abstract Co-Author: Nothing to Disclose

Adriana Bou-Khair, Abstract Co-Author: Nothing to Disclose

Ines Ninomiya, Abstract Co-Author: Nothing to Disclose

Norma Castagnino, Abstract Co-Author: Nothing to Disclose

Marina Ulla MD, Abstract Co-Author: Nothing to Disclose

Jorge Manuel Carrascosa MD, Abstract Co-Author: Nothing to Disclose

et al, Abstract Co-Author: Nothing to Disclose

et al, Abstract Co-Author: Nothing to Disclose

To determine the usefulness of an unfolded haustra software called “filet view” (FV) for the evaluation of elevated lesions in comparison with the virtual colonoscopy conventional analysis (VC) and conventional colonoscopy (CC) in children.

Sixty one patients (mean age 5 years old), scheduled for CC due to signs of rectal bleeding were evaluated with CT and CC the same day. The day before, they followed a standard colonic cleansing. CCs were carried out with an E-P X30S EVE (Fujimon) colonoscope. CT analysis: it was first performed a virtual colonoscopy reprocess in the conventional modality, reading images in 2D and confirming findings in 3D and voyager. A second reprocess was done with the unfolded haustra software, that shows the colon opened. Findings were sub-divided in normal and pathologic. The presence, location and size of the lesions were assessed in eigth colonic segments. Findings of VC with the FV were compared separately with CC. Elevated lesions found in any segment were considered as true-positive findings. Segments without lesions were considered as true-negative findings. The 95% confidence intervals for the proportions were calculated by the exact binominal method. The average reading time and the standard deviation was calculated for each method.

A total of 38 patients presented negative findings. Rectal bleeding was due to chronic constipation (n= 20), internal hemorroids (n= 5) and external hemorroids (n=8). In the 23 pathologic patients CC detected 30 elevated lesions. Sensitivity (S) and specificity (Sp) were calculated per patient and per polyp. Per patient: for VC, S was 86%, Sp: 98%; for FV, S was 91% and Sp 99%. Per polyp: for VC, S was 81%, Sp: 88%; for FV, S was 82% and Sp 90%. The average reading time in the VC was 18 minutes (+/- 3) whereas for the FV was 4 minutes (+/-1).

Although there were no significant differences between VC and FV results, virtual colon dissection with open folds is a helpful tool for the evaluation of VC in children, reducing significantly the post-processing time.

The unfolded haustra software facilitate the virtual colonoscopy analysis in children, making the exam analysis less time consuming.

Carrascosa, P, Capuñay, C, Martin Lopez, E, Bou-Khair, A, Ninomiya, I, Castagnino, N, Ulla, M, Carrascosa, J, et al, , et al, , Role of Unfolded Haustra Software for the Detection of Elevated Colonic Lesions in Children.  Radiological Society of North America 2006 Scientific Assembly and Annual Meeting, November 26 - December 1, 2006 ,Chicago IL. http://archive.rsna.org/2006/4433992.html