Capsule endoscopy: Is the software TOP 100 a reliable tool in suspected small bowel bleeding?

Abstract

Small bowel capsule endoscopy (SBCE) is the reference standard tool for diagnosing small bowel bleeding (SBB). The "TOP 100", which performs an automatic selection of the 100 images that mostly likely contain abnormalities, emerged as a new functionality of the RAPID Reader® software in 2017. Background Small bowel capsule endoscopy (SBCE) is the reference standard tool for diagnosing small bowel bleeding (SBB). The “TOP 100”, which performs an automatic selection of the 100 images that mostly likely contain abnormalities, emerged as a new functionality of the RAPID Reader® software in 2017.

Aim To compare the concordance of findings between the standard reading (SR) and the use of TOP 100 in suspected SBB.

Methods Retrospective study, including consecutive patients submitted to SBCE for suspected SBB. Two experienced readers performed SR and reported the most important findings. Another experienced reader, who was blinded to the SR results, reviewed all the SBCE videos using TOP 100 and reported the most important findings. The relevant findings were defined as the presence of high bleeding potential lesions (P2).

Results 97 patients were included. The TOP 100 detected 81/97(83.5%) of the P2 lesions, in particular 64/67(95.5%) of the angioectasias and 17/30(56.7%) of the ulcers. The TOP 100 identified all sites of active bleeding (n = 9).

Conclusion The TOP 100 identified all sites of active bleeding, as well as the vast majority of significant lesions (83.5%); in particular, it detected over 95% of the angioectasias. Although SR remains the reference standard in the SBCE review, these findings demonstrate that TOP 100 allows for a quick preview reading constituting an important asset in the identification of lesions that may require priority full review and intervention planning