No Effect of Computer-Aided Diagnosis on Colonoscopic Adenoma Detection in a Large Pragmatic Multicenter Randomized...

Area of Science:

• Gastroenterology
• Medical Technology
• Clinical Trials

Background:

• Computer-assisted detection (CADe) has shown promise in improving colonoscopic adenoma detection rates (ADR).
• However, recent studies have yielded inconsistent results regarding CADe's effectiveness.

Purpose of the Study:

• To evaluate the impact of CADe on ADR in patients aged 50 and older undergoing colonoscopy.
• To compare secondary outcomes such as adenoma detection rates across different subgroups and adverse events.

Main Methods:

• A randomized controlled trial involving 1627 patients across 12 German institutions.
• Patients were randomized to either a CADe-assisted or a standard colonoscopy group.
• Primary outcome was the adenoma detection rate (ADR); secondary outcomes included adenomas per colonoscopy and detection rates of specific adenoma types.

Main Results:

• No significant difference in the observed or adjusted ADR between the CADe and control groups (40.0% vs 37.5% observed; 40.6% vs 38.3% adjusted).
• CADe did not improve detection rates for adenomas based on size, morphology, location, or histology.
• Examiner, patient age, and sex significantly influenced ADR, while CADe led to a higher detection of hyperplastic polyps.

Conclusions:

• Computer-assisted detection (CADe) did not increase the adenoma detection rate (ADR) in patients over 50 with mixed colonoscopy indications.
• Further research is recommended to determine the selective use of CADe in clinical practice, considering the endoscopist's baseline ADR.