Computer-assisted detection versus conventional colonoscopy for proximal colonic lesions: a multicenter, randomized,...

Area of Science:

• Gastroenterology
• Medical Technology
• Endoscopy

Background:

• Computer-assisted detection (CADe) is a technology aimed at improving adenoma detection during colonoscopy.
• The effectiveness of CADe in reducing missed colonic lesions remains uncertain.

Purpose of the Study:

• To evaluate the miss rates of proximal colonic lesions using CADe versus conventional colonoscopy.
• To determine if CADe can decrease the number of missed adenomas and polyps in the proximal colon.

Main Methods:

• A prospective, multicenter, randomized, tandem-colonoscopy study was conducted across 3 Asian centers.
• Patients were randomized to undergo either CADe or conventional white-light colonoscopy for the initial examination.
• A tandem colonoscopy was performed in all patients to assess the miss rate, defined by lesions found in the second examination.

Main Results:

• No significant difference was observed in the miss rates for proximal adenomas (20.0% vs 14.0%) or polyps (26.7% vs 19.6%) between CADe and conventional colonoscopy.
• The CADe group demonstrated significantly higher detection rates for proximal polyps (58.0% vs 46.7%) and adenomas (44.7% vs 34.6%).
• CADe showed enhanced detection in specific subgroups, including patients with suboptimal bowel preparation or shorter withdrawal times.

Conclusions:

• Computer-assisted detection (CADe) significantly increases the detection of proximal adenomas and polyps during colonoscopy.
• Despite improved detection rates, CADe did not significantly reduce the miss rate of proximal colonic lesions in this study.
• CADe may be particularly beneficial for endoscopists with lower adenoma detection rates.