Augmented Endoscopy for Surveillance of Colonic Inflammatory Bowel Disease: Systematic Review With Network Meta-analysis

Introduction Considering the high risk of dysplasia and cancer in inflammatory bowel disease [IBD], surveillance is advocated. However, international guidelines do not reach a uniform recommendation on the way to perform surveillance. We performed a systematic review with a meta-analysis to assess the best endoscopic surveillance strategy in colonic IBD. Methods The systematic review was performed in PubMed/MEDLINE, EMBASE, SCOPUS, and Cochrane databases to identify studies comparing white light endoscopy [WLE] and augmented endoscopy [AE] in the detection of dysplasia/neoplasia in colonic IBD. A sub-analysis between dye-spray chromoendoscopy [DCE], narrow-band imaging [NBI], I-SCAN, full-spectrum endoscopy [FUSE], and auto-fluorescence imaging [AFI] was also performed. Furthermore, a meta-regression and a network meta-analysis were also performed. Results A total of 27 studies [6167 IBD patients with 2024 dysplastic lesions] met the inclusion criteria. There was no publication bias. AE showed a higher likelihood of detecting dysplasia than WLE (19.3% vs 8.5%, odds ratio [OR] = 2.036), with an incremental yield [IY] of 10.8%. DCE [OR = 2.605] and AFI [OR = 3.055] had higher likelihood of detecting dysplasia than WLE; otherwise, I-SCAN [OR = 1.096], NBI [OR = 0.650], and FUSE [OR = 1.118] were not superior to WLE. Dysplasia was found in 1256/7267 targeted biopsies [17.3%] and in 363/110 040 random biopsies [0.33%] [OR = 66.559, IY = 16.9%]. Meta-regression found no variable impacting on the efficacy of AE techniques. Network meta-analysis identified a significant superiority of DCE to WLE in detecting dysplasia [OR 2.12], but no other single technique was found to be superior to all others in dysplasia detection. Conclusions DCE was associated with higher likelihood of discovering dysplastic lesions than WLE. Chromoendoscopy is the best supported endoscopic technique for IBD surveillance.