P112. iSCAN-High definition colonoscopy correlates with white light endoscopy and histology assessment in mucosal healing for ulcerative colitis
M. Iacucci1, M. Fort Gasia1, X. Gui2, R. Panaccione1, G. Kaplan1, J. Love1, S. Ghosh1, 1University of Calgary, Gastroenterology, Calgary, Canada, 2University of Calgary, Pathology, Calgary, Canada
White light (WL) colonoscopy is used to assess the endoscopic appearance in ulcerative colitis (UC). High definition iSCAN offers the potential to better characterize the mucosa in patients with UC and may provide information about both inflammation and mucosal healing. Our aim was to assess high definition iSCAN in patients with UC and compare findings to WL endoscopy and histopathology.
45 patients (21 female, median age = 41, range = 19–90 years) with UC were recruited and assessed by both high definition-iSCAN colonoscopy (Pentax, Tokyo) and WL colonoscopy. Mayo endoscopy subscores were assigned to patients according to findings on white light endoscopy. Targeted biopsies were taken and assigned a histological grade using the Harpaz score (0 = normal or inactive chronic colitis, 1 = mildly active chronic colitis, 2 = moderately active chronic colitis, 3 = severe chronic active colitis). The pathologist was blinded to the endoscopy findings. Mucosal pattern on iSCAN was graded as 1 = normal, 2 = mosaic pattern, 3 = roundish rosette, 4 = tubular-gyrus, 5 = nodular roundish rosette. The vascular pattern was graded as 1 = normal, 2 = drop-out vessels, 3 = spiral isolated vessels, 4 = crowded tortuous vessels, 5 = honeycomb irregular crowded vessels. iSCAN appearance was correlated with endoscopic Mayo subscore (0–3) and histologic Harpaz score by using Spearman correlation coefficient.
There was a high degree of correlation between the I-SCAN scores and histological grading and the Mayo endoscopy subscore. I-SCAN mucosal pattern was significantly correlated with Harpaz score (rs = 0.759; p < 0.00001). Similarly, iSCAN vascular pattern was significantly correlated with Harpaz score (rs = 0.520; p = 0.0002). The overall iSCAN score was also significantly correlated with Harpaz score (rs = 0.6702; p < 0.00001). Finally, endoscopic Mayo subscore was also significantly correlated with overall iSCAN score (rs = 0.802; p < 0.00001). However in patients with a Mayo subscore of 0 (9 patients), seven patients had abnormal vascular pattern and 6 patients had abnormal mucosal pattern and 3 patients had abnormal histology according to Harpaz score.
A high degree of correlation was demonstrated between iSCAN mucosal and vascular pattern scores and the Mayo subscore and Harpaz grading. However, patients with endoscopic Mayo subscore 0 demonstrated high proportion of abnormal vascular pattern and mucosal pattern on iSCAN colonoscopy. The significance of this needs to be further explored.