46 A meta-analysis has demonstrated that chromoendoscopy has medium to high sensitivity (83.3%, 95% confidence interval [CI]: 35.9–99.6), specificity

(91.3%, 95% CI: 43.8–100), and high diagnostic accuracy (odds ratio 17.544, 95% CI: 1.245–247.14) for dysplastic lesions47 and is superior to white light colonoscopy for the proportion of lesions detected by biopsies (44%, 95% CI: 28.6–59.1) as well as for flat dysplasia (27%, 95% CI: 11.2–41.9) in patients with UC.26 Kiesslich and colleagues20 reported 165 patients with long-standing UC who were randomized to conventional colonoscopy or colonoscopy with chromoendoscopy using 0.1% methylene blue. More targeted biopsies were possible, and significant intraepithelial neoplasia was detected in the chromoendoscopy PI3K inhibitor group (32 vs 10; P = .003). Rutter and colleagues 23 reported the importance of indigo carmine dye spraying for the detection of dysplasia in UC. They emphasized that no dysplasia was detected in 2904 nontargeted biopsies. In comparison,

chromoendoscopy with targeted biopsy led to fewer biopsies and detected 9 dysplastic lesions, 7 of which were only visible after indigo carmine application. They concluded that the indigo carmine dye spraying of the whole colon is feasible, Ku-0059436 in vitro and dysplasia detection may be more effective than taking large numbers of random biopsies. Hurlstone and colleagues 31 also emphasized that indigo carmine–assisted high-magnification chromoendoscopy and improved the detection of intraepithelial neoplasia in the endoscopic screening of patients with UC. However, pancolonic chromoendoscopy has potential limitations: dye on the mucosa is not always

equally spread; dye pooling can lead to difficult observation; more time is needed; and some biopsies may be false negative. In the authors’ institution, they routinely perform high-magnification colonoscopy with indigo carmine chromoendoscopy after they suspect the presence of NP-CRN in patients with cIBD. Morphologically, NP-CRN in IBD appear to be slightly elevated, completely flat, or slightly depressed as compared with the surrounding mucosa. In order to detect them, the authors look for the presence of a slightly elevated lesion, focal friability, obscure vascular pattern, discoloration (uneven redness or a patch or redness), villous mucosa not (velvety appearance), and irregular nodularity. The finding of any of these signs typically alerts the authors to become suspicious of the possible presence of NP-CRN and leads them to wash out the mucus or debris from the surface on the target lesion and apply the dye for magnifying colonoscopy.15 After dye spraying but before the authors perform a biopsy or resection, they will typically evaluate the border of the lesion. The authors look for the presence of dye pooling within the lesion, which would suggest the diagnosis of a depressed lesions.