The impact of Endocuff addition to Real-Time Computer-Aided Detection of Colorectal Neoplasia in a randomised tandem colonoscopy trial: ENOSIS

Aims Adenoma miss rate during colonoscopy is directly linked to risk of post-colonoscopy colorectal cancer. Articifial intelligence (AI) and Endocuff Vision when used indepedently are proven to increase adenoma detection rate (ADR). The aim of the study is to assess whether combining AI with Endocuff Vision, compared to using AI alone, increases the ADR [1] [2] [3] [4] [5].

Methods This is a single-centre randomized, tandem colonoscopy trial performed by the same endoscopists on the same day, first with AI and Endocuff Vision followed by AI alone and vice versa. All procedures were performed by 2 experienced endoscopists. Use of antispasmosdics and position changes during insertion and withdrwal were recorded. Patients with a Boston Bowel Preparation (BPP) score<6 were excluded from the study.

Results A total of 85 patients were enroled (M/F: 51/34, mean age 63±8). AI alone was used in the initial colonoscopy in 39 patients and the combination of AI and Endocuff in 45. The indication for colonoscopy was colorectal cancer screening in 78 (91.8%), and polyp surveillance in 7 (8.2%). There was no difference in average bowel preparation (BPP=8.6±1.0) (p=0.594), age (p=0.467), gender (p=0.113), or indication (p=0.356) between the two groups.The polyp detection rate did not differ with Endocuff (25/39, 64.1%) or without it (25/43, 58.1%, p=0.373), although the ADR was numerically higher without Endocuff (19/39, 48.7% vs 14/43, 32.6%, p=0.107), as was the number of polyps detected per colonoscopy (1.7±2.5 vs 1.2±1.4, p=0.272).The polyp detection rate in the left or right colon did not differ between the two groups (with Endocuff 16.3% and 44.2% vs without Endocuff 28.2% and 51.3%, p=0.150 and p=0.307, respectively).Withdrawal time did not differ between the two groups (with Endocuff 7.2±2.5 vs 7.2±1.7, p=0.887), but the number of patients who reported moderate/severe discomfort during the examination tended to be statistically higher with Endocuff use (4/43, 9.3% vs 0/39, p=0.071).During tandem colonoscopy, an additional 0.4±0.8 polyps per examination were found (Polyp miss rate=0.08±0.15), and 0.3±0.7 adenomas (adenoma miss rate=0.10±0.25). However the addition of Endocuff in tandem colonoscopy among patients initially examined with only AI did not show an increased Adenoma miss rate compared to those with initial AI+Endocuff-assisted endoscopy followed by tandem AI-only (0.11±0.24 vs. 0.92±0.27, p=0.727), nor did it affect the polyp miss rate (0.10±0.15 vs. 0.07±0.14, p=0.415). No adverse events were reported during the study.

Conclusions Our data show that the addition of Endocuff in AI-assisted colonoscopy does not seem to increase its effectiveness in terms of overall polyp and adenoma detection.

Publication History

Article published online:
27 March 2025

• References

• 1 Pohl H., Robertson D.J.. “Colorectal cancers detected after colonoscopy frequently result from missed lesions.,”. Clin. Gastroenterol. Hepatol. 2010; 8 no. 10 pp.: 858-64
• 2 Kaminski M.F.. et al. “Quality indicators for colonoscopy and the risk of interval cancer.,”. N. Engl. J. Med. 2010; 362 no. 19 pp: 1795-803
• 3 Zhao S.. et al. “Magnitude, Risk Factors, and Factors Associated With Adenoma Miss Rate of Tandem Colonoscopy: A Systematic Review and Meta-analysis.,”. Gastroenterology 2019; 156 no. 6 pp: 1661-1674.e11
• 4 Hassan C, Spadaccini M, Iannone A. et al. Performance of artificial intelligence in colonoscopy for adenoma and polyp detection: a systematic review and meta-analysis. Gastrointest Endosc 2021; 93 (01): 77-85.e6
• 5 Bevan R, Ngu WS, Saunders BP. et al. The ADENOMA Study. Accuracy of Detection using Endocuff Vision™ Optimization of Mucosal Abnormalities: study protocol for randomized controlled trial. Endosc Int Open 2016; 4 (02): E205-E212