Safety and Efficacy of the ENDO-AID Computer-Aided Detection (CADe) system in detection of colorectal neoplasia during real-time colonoscopy: A European Prospective, Multicenter, Randomized Controlled Trial (EuroCADe)

Aims This study aimed to compare the effectiveness and safety of colonoscopy assisted by the ENDO-AID (Olympus, Japan) CADe (Computer-Aided Detection) system against standard colonoscopy in a real-world clinical setting.

Methods A total of 845 patients with indication for colonoscopy to screen for polyps were recruited for this prospective, multicenter, singe-blind, randomized controlled trial across 6 European hospitals. Participants were randomly assigned to either standard colonoscopy without CADe (CADe-) or CADe-augmented colonoscopy (CADe+), both using latest generation high-definition EZ1500 colonoscopes (CF-EZ1500, Olympus) and EVIS X1 processors (CV-1500, Olympus). Randomization was stratified by the colonoscopist and the indication for the procedure (screening vs surveillance). The majority of colonoscopists involved in this study were experts, with 93.6% in the CADe+group and 95.1% in the CADe- group.

Results The overall adenoma detection rate did not significantly differ between the CADe+group (52.8%) and the CADe- group (47.3%) (OR 1.2, 95%CI 0.9-1.6, p=0.19). Similarly, no significant differences were observed when stratifying the adenoma detection rate for polyp size [diminutive (p=0.12); small (p=0.98; large (p=0.19)] and morphology [pedunculated (p=0.58); sessile (p=0.20); flat (p=0.55)] between both groups. However, this study did show a difference in polyp detection rates, with CADe+achieving 79.6% (95%CI 75.4-83.4) compared to 67.2% (95% CI 62.4-71.8) in the CADe- group. No difference was found in the withdrawal time, with CADe+averaging 12.5 minutes (95%CI 11.9-13.1) and CADe- averaging 12.1 minutes (95%CI 11.6-12.7). Regarding safety endpoints, the positive predictive value per patient was not significantly different between the two groups: CADe+(0.57, 95% CI 0.53-0.62) and CADe- (0.64, 95% CI 0.59-0.68). No relevant CADe-related serious adverse event (SAE) were reported.

Conclusions This study demonstrated that the use of the CADe system did not significantly improve the overall adenoma detection rate in expert endoscopists using the latest scopes and processors. However, the use of CADe was associated with a higher polyp detection rate. These findings suggest that the use of the Olympus Endoscopy CADe system is on par with the identification of polyps by highly experienced endoscopist, potentially improving screening outcomes for non-expert endoscopists. Importantly, the procedural times and safety profiles were comparable, indicating that the integration of CADe does not compromise the efficiency or safety of colonoscopy.

 

MissingLinkText

Conflicts of Interest

EuroCADe study was sponsored by Olympus

Publication History

Article published online:
27 March 2025