Abstract
Purpose: Colorectal cancer (CRC) risk is elevated in pts with long-term ulcerative colitis (UC). Endoscopic surveillance with biopsies is performed to identify dysplasia, a precursor lesion of CRC. Use of dye-based chromoendoscopy (CE) in UC surveillance is supported by European GI societies and the CCFA, however, recently published guidelines by ACG and AGA recommend random and targeted biopsies of raised lesions and do not yet suggest the routine use of CE. In the US, few physicians use CE in colonoscopic evaluation of IBD pts. Our study aim is to compare the diagnostic yield of CE vs. white light colonoscopy (WLE) for dysplasia during surveillance colonoscopy in the published literature. Methods: Databases examined in this meta-analysis include MEDLINE, EMBASE, Cochrane Collaboration, [email protected], Cochrane, CAB, and OVID Healthstar. Full-length, English, prospective studies were included for analysis. The pooled estimate of odds ratio (OR) for the association between CE/WLE and dysplasia (IN) detection was calculated with random effects models and the heterogeneity in ORs across studies was examined. Results: We identified 2525 articles; 2494 were rejected, 31 were included in full review and 6 met inclusions criteria. There were 1584 pts with UC undergoing surveillance colonoscopy including 878 pts with CE exam and 868 with WLE exam. CE had a significant benefit in detecting dysplasia. The difference in the number of pts with CE detected dysplasia (IN) was statistically significant [pooled OR=2.79 (95% CI 1.91-4.08), but not with significant heterogeneity (p=0.5)]. CE had a significant benefit in detecting IN in flat mucosa [pooled OR=2.47(0.67-8.67) with significant heterogeneity (p=0.036)]. CE had no benefit in detecting IN in raised mucosa (“polypoid” mucosa): [pooled OR=0.74 (0.33-1.65) with no significant heterogeneity (p=0.3)].CE had a significant benefit in detecting IN in targeted biopsies [pooled OR=2.15 (1.38-3.33) without significant heterogeneity (p=0.67)]. Conclusion: This study represents the first meta-analysis of the role of CE in detecting dysplasia during surveillance colonoscopy in pts with long term UC. This confirms an overall significant benefit of the CE for the detection of dysplasia in flat mucosa and in targeted biopsies but it does not demonstrate a benefit in detecting dysplasia in visible, raised (“polypoid”) mucosal lesions.Figure
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call