Improving detection of colorectal dysplasia in inflammatory bowel disease surveillance

Patients with ulcerative colitis or Crohn’s colitis have an increased risk of colorectal cancer compared with the general population. Most colorectal cancer cases in inflammatory bowel disease (IBD) are believed to arise from dysplasia, and, thus, surveillance colonoscopy is recommended to detect dysplasia. High-quality surveillance colonoscopy in this patient population is important but can be challenging because of the subtlety of dysplastic lesions. Although in earlier endoscopic eras most dysplasia was “invisible” and discovered on random biopsy samples, today most dysplasia detected at surveillance colonoscopy is visible. Improvements in endoscopic technologies, such as high-definition video endoscopy, and use of image enhancement methods, such as chromoendoscopy, seem likely to explain the increased proportion of visible dysplastic lesions identified at surveillance colonoscopy.

Varying recommendations regarding the detection and management of colorectal dysplasia in IBD patients have resulted in significant practice disparity. Recently, an international multidisciplinary group, Surveillance for Colorectal Endoscopic Neoplasia Detection and Management in Inflammatory Bowel Disease Patients: International Consensus (SCENIC), which represented a wide spectrum of stakeholders regarding IBD surveillance, aimed to optimize strategies for the detection and management of dysplasia in IBD patients. SCENIC followed the prescribed processes for guideline development, including systematic reviews, full synthesis of evidence and deliberations by panelists, and incorporation of the Grading of Recommendations, Assessment, Development and Evaluations methodology. The key summary recommendations from SCENIC for surveillance colonoscopy included (1) the use of high-definition endoscopy and performance of chromoendoscopy for the detection of dysplasia, (2) the endoscopic removal of visible dysplasia, and (3) referral to an endoscopist with expertise in IBD surveillance using chromoendoscopy with high-definition colonoscopy for patients with endoscopically invisible dysplasia confirmed by a GI pathologist.

In this issue of Gastrointestinal Endoscopy, Deepak et al retrospectively measure the diagnostic yield of chromoendoscopy for dysplasia in a high-risk referral IBD group: 95 patients with dysplasia identified on white-light surveillance colonoscopy. The authors report the miss rate of colorectal dysplasia with white-light colonoscopy by assessing the findings at chromoendoscopy examinations performed a median of 6 months (range, 2.6-12.1 months) after the white-light examinations. Fifty of 95 patients had visible dysplastic lesions identified on the first follow-up chromoendoscopy examination.

The article by Deepak et al again illustrates that most dysplastic lesions in IBD are visible with modern endoscopic technologies. A systematic review of studies in the video-endoscopic era revealed that among IBD patients with dysplasia, the dysplasia is visible in approximately 80% of patients examined with standard-definition colonoscopy and in approximately 90% of patients examined with high-definition white-light colonoscopy or chromoendoscopy.

The findings of Deepak et al also highlight the value of referring patients with endoscopically invisible dysplasia (confirmed by a GI pathologist) to an endoscopist with experience in IBD surveillance using chromoendoscopy with high-definition colonoscopy, as suggested by the SCENIC Consensus Statement. Among the 40 patients with dysplasia identified only on random biopsy sampling at an initial white-light colonoscopy, 12 (30%) had dysplastic lesions identified at subsequent chromoendoscopy. Whether chromoendoscopy increases the detection of dysplasia even when using high-definition colonoscopy has been a point of contention. The relatively high incidence of missed visible dysplasia in this study might be taken to suggest that high-definition white-light examination without chromoendoscopy is not sufficient because 90% of the index examinations performed at the Mayo Clinic used high-definition colonoscopy. However, information on the index examinations for the 45 index colonoscopies performed at other institutions is not provided, limiting our ability to draw conclusions on this topic. The SCENIC Consensus Statement suggested chromoendoscopy even when using high-definition white-light colonoscopy based on a significant benefit in a single prospective trial. Since then, a prospective, randomized, controlled trial also showed significantly higher detection of dysplastic lesions with high-definition plus chromoendoscopy compared with high-definition white-light colonoscopy alone: .26 versus .12 lesions per patient, respectively.

Chromoendoscopy involves the application of dye to the colon mucosa, thereby providing contrast enhancement to improve visualization of epithelial surface detail and accentuate the changes in the surface topography. To date, the chromoendoscopy studies in IBD patients have used either indigo carmine or methylene blue. Currently, indigo carmine supply has been limited in the United States because of manufacturing issues. Endoscopists have been therefore required to use methylene blue, but costs of this dye are now higher. Given the lack of physician or facility reimbursement for chromoendoscopy, the expense may become prohibitive for many centers. Interestingly, the study by Deepak et al is the first to report longitudinal data on the use of an alternative and relatively inexpensive chromoendoscopy agent, .2% FD&C Blue #2 Solution. Further high-quality studies of alternative chromoendoscopy dyes, such as this food coloring agent, will be important to determine if high-quality examination can be achieved at lower cost.

Deepak et al report high rates of incomplete endoscopic resection that could contribute to recurrent dysplasia. Among patients with endoscopically treated dysplastic lesions at index colonoscopy, almost half of the 67 lesions had incomplete resection. Cold forceps or cold snare was predominantly used. These results emphasize the urgency of focusing efforts on improving the endoscopic management of dysplastic lesions. When dysplasia is detected, it should be characterized as “endoscopically resectable” or “nonendoscopically resectable.” The term “endoscopically resectable” indicates that (1) distinct margins of the lesion could be identified, (2) the lesion appears to be completely removed on visual inspection after endoscopic resection, (3) histologic examination of the resected specimen is consistent with complete removal, and (4) biopsy specimens taken from mucosa immediately adjacent to the resection site are free of dysplasia on histologic examination. If the lesion is assessed to be endoscopically resectable, then using the proper technique to ensure complete endoscopic removal is critical. Endoscopic mucosal resection is the preferred method. Hybrid endoscopic mucosal resection and endoscopic submucosal dissection are other potential techniques, often determined by the lesion characteristics as well as the endoscopist’s skill.

Widespread improvement in the endoscopic recognition of dysplastic colorectal lesions could have important implications for the surveillance and management of dysplasia. Targeted sampling and resection of visible lesions may allow a shift in practice away from the random biopsy technique, where less than .1% of the colonic mucosal surface area is blindly sampled, and away from colectomy for the management of dysplasia.