P226 Relevant factors and significant endoscopic findings for detecting colitis-associated neoplasms using pancolonic narrow band imaging surveillance colonoscopy in patients with ulcerative colitis: A sub-analysis of prospective randomised trial

• Published in: Journal of Crohn's and colitis 2018-01, Vol.12 (supplement_1), p.S214-S214
• Main Authors: Watanabe, K, Kawano, R, Nishishita, M, Shimamoto, F, Fukuchi, T, Esaki, M, Okamoto, Y, Maehata, Y, Oka, S, Nishiyama, S, Fujii, S, Hirai, F, Matsui, T, Kakimoto, K, Okada, T, Inoue, T, Hida, N, Nozaki, R, Sakurai, T, Kashida, H, Takeuchi, K, Ohmiya, N, Saruta, M, Saito, S, Saito, Y, Nakamura, S, Tanaka, S, Suzuki, Y, Ajioka, Y, Tajiri, H
• Format: Article
• Language: English
• ISSN: 1873-9946; 1876-4479
• DOI: 10.1093/ecco-jcc/jjx180.353

Abstract Background We recently reported the results of a prospective multicenter randomised controlled trial comparing pancolonic observation using a newly developed narrow band imaging (NBI) (CF-HQ290I) and panchromoendoscopy (PCE) for surveillance colonoscopy in patients with ulcerative colitis (UC). The results revealed that the pancolonic NBI observation was not inferior to PCE for the detection of both neoplastic lesions (13.4% vs 9.0%; p = 0.33) and colitis-associated dysplasia or cancer (CC/D) (6.3% vs. 4.9%; p = 0.86). Moreover, the total examination time of NBI (15.0 min, 12.4–20.1 min) was significantly shorter than that of PCE (19.8 min, 16.6–24.5 min) (p < 0.01). This sub-analysis aimed to evaluate the relevant factors or NBI findings of the background mucosa of CC/D in the preceding Navigator Study (UMIN000013527). Methods In total, 263 patients were randomised to the PCE group (n = 130) and the NBI group (n = 133). Surveillance colonoscopy in both groups was performed mainly with targeted biopsy. The pathological diagnoses were made by two expert pathologists who performed including immunohistochemical staining (p53, Ki-67) while being blinded to clinical information. Results A total of 21 sporadic adenoma, 2 traditional serrated adenoma, 2 sessile serrated adenoma/polyp, 7 low-grade dysplasia, and 7 high-grade dysplasia or cancer cases were detected. There was no statistical difference in the contents of detected lesions between the two groups (p = 0.55). The median number of biopsy samples was not different between the two groups (1.80/case in NBI, 1.85/case in PCE; p = 0.88). Multivariate analysis demonstrated that age (odds ratio [OR] 1.05, 95% confidence interval [CI] 1.02–1.08, p < 0.01) and past history of CC/D (OR 4.26, 95% CI 1.29–14.05, p = 0.02) were the relevant factors for the detection of neoplastic lesions, and past history of CC/D (OR 5.70, 95% CI 1.33–24.50, p = 0.02) was the relevant factor for the detection of CC/D. CC/D was significantly more frequently detected than sporadic adenoma in cases in which the background mucosa had an obscure vascular pattern or scar rather than a fair view of the vascular pattern or out of range for inflammation (OR 25.0.95% CI 3.80–164.4, p < 0.01). Conclusions In the Navigator Study, a past history of endoscopic resection or detection of dysplasia on random biopsy was the relevant factor for the detection of CC/D. The NBI findings of the background mucosa were useful in differentiating CC/D