P146 Detection and characterization of colonic dysplastic lesions in IBD surveillance colonoscopy - a randomised comparison of high definition alone with high definition dye spraying and electronic virtual chromoendoscopy using iSCAN
M. Iacucci*1, M. Fort Gasia1, S. Urbanski2, P. Minoo2, G. Kaplan1, R. Panaccione1, S. Ghosh1
1University of Calgary, Department of Gastroenterology, Inflammatory Bowel Disease Clinic, Alberta, Canada, 2University of Calgary, Department of Pathology, Calgary, Canada
The standard of practise for IBD surveillance colonoscopy is now considered dye spraying chromoendoscopy. However, high definition endoscopy has improved in its resolution significantly. Therefore, it is important to determine the best technique for detection of dyspastic lesions (DL) in IBD surveillance colonoscopy. We aimed to determine the frequency of DL during surveillance colonoscopy in IBD and to define the endoscopic features of these lesions by using three different techniques: high definition white light colonoscopy (HD), dye spraying (0.2% indigo carmine) chromoendoscopy (DSC) and electronic virtual chromoendoscopy using i-SCAN (EVC).
A randomized study (NCT02098798) was conducted to determine the detection rates of DL with HD, DSC or EVC in 95 patients (46 female, median age 52 years, range 22-77 years) with long standing colitis (8 years from diagnosis, including both UC and CD). Patients with inactive disease were enrolled in 1:1:1 ratio into three arms of the study. Colonoscopy was performed using a Pentax EPKi processor and high-resolution video colonoscope (EC-3490Fi; Pentax Tokyo). Endoscopic colonic lesions were classified by size, Kudo pit pattern and Paris classification. Lesions of dysplasia-associated lesion or masses and adenoma-like masses (DALMs/ALMs), sessile serrated adenomas (SSAs), adenoma-like polyps (ALPs), hyperplastic polyps (HPs), and inflammatory polyps (IPs) were identified.
Patients were randomized into three groups, HD (n=32, 33.7%), VEC (n=33, 34.7%) and DSC (n=30, 31.6%). 47 DL were found in total. Thirty (63.8%) were detected in the HD group (8 SSAs, 20 ALPs and 2 DALMs), 6 (12.8%) in the DCE group (6 ALPs) and 11 (23.4%) in the EVD group (7 SSAs, 1 ALP and 3 DALMs). The endoscopic characteristics of SSA were <5mm in size (66.6%), non-polypoid (53.3%) and Kudo pit pattern IIO (93.3%). Similarly, ALPs were <5mm in size (77.7%), polypoid (55.5%) and Kudo pit pattern III (77.7%). Finally, DALMs were <5mm (60%), non-polypoid (60%) and Kudo pit pattern IIIL (40%) and IV (40%). Among the three groups, HD had a sensitivity of 86.67%, specificity of 89.29%, PPV 89.66% and NPV 86.21% in detecting DL. DSC had a sensitivity of 66.67%, specificity of 88.57%, PPV 50%, NPV 93.94% and EVC had a sensitivity of 100%, specificity of 83.3%, PPV 64.71% and NPV 100%.
DL are frequently detected in long-standing IBD. Our Results indicate that DSC is not more accurate than either HD or VEC in detecting DL. In fact, the majority of dysplastic lesions were detected in the HD group, suggesting that advances in high definition technology may favour this technique as the surveillance method of choice for IBD in future.