Download PDF
Endoscopy 2023; 55(12): 1152
DOI: 10.1055/a-2147-0532
Letter to the editor

Reply to Spadaccini M. et al.

Ahmir Ahmad 
1   Wolfson Unit for Endoscopy, St Mark’s Hospital, London, United Kingdom
,
Paul Bassett
2   Statsconsultancy Ltd, Amersham, United Kingdom
,
Brian P. Saunders
1   Wolfson Unit for Endoscopy, St Mark’s Hospital, London, United Kingdom
› Author Affiliations
› Further Information
Also available at
    Permissions and Reprints

See also:

Colonoscopy setting and impact on statistical significance in Artificial Intelligence trialsEndoscopy 2023; 55(12): 1151-1151
DOI: 10.1055/a-2131-3315

Thank you for the opportunity to respond to the letter from Spadaccini et al. [1]; we are very grateful for their comments.

We chose a fecal immunochemical test-based colonoscopy screening setting to explore how effective computer-aided polyp detection (CADe) would be in this polyp-enriched group, as it makes up a significant percentage of our colonoscopy practice. The higher baseline adenoma detection rate (ADR) of endoscopists in our study was probably due to both the high technical skill level of the bowel cancer screening-accredited endoscopists and the use of the Endocuff Vision device (Olympus, Tokyo, Japan), which was used in 71.7 % of CADe colonoscopies and 69.2 % of standard colonoscopies. This probably reduced the potential for the CADe system to show a more significant polyp detection difference owing to a “ceiling effect.”

Our reason for choosing the polyp detection rate (PDR) as the primary outcome measure was that CADe systems are designed to detect polyps of all shapes, sizes, and morphology and not just adenomas [2]. PDR is a simple measure of colonoscopy performance that does not require histology, and the British Society of Gastroenterology quality standards advise it “may be used where it has been demonstrated to accurately reflect ADR for that unit/clinician” [3].

Undoubtedly, larger studies will be required within a bowel cancer screening setting to definitively identify the place of CADe in this setting. However, our preliminary study strongly suggests that with highly experienced endoscopists using mucosal exposure devices, the clinically significant benefits of CADe are limited.



Publication History

Article published online:
28 November 2023

© 2023. Thieme. All rights reserved.

Georg Thieme Verlag KG
Rüdigerstraße 14, 70469 Stuttgart, Germany