Endoscopy 2020; 52(S 01): S52
DOI: 10.1055/s-0040-1704159
ESGE Days 2020 oral presentations
Thursday, April 23, 2020 14:30 – 16:00 Polyp forensics: Colon advanced Wicklow Meeting Room 3 Imaging 2
© Georg Thieme Verlag KG Stuttgart · New York

THE INCREMENTAL BENEFIT OF DYE-BASED CHROMOENDOSCOPY AS COMPARED TO HIGH-DEFINITION WHITE LIGHT AND VIRTUAL CHROMOENDOSCOPY FOR LESION ASSESSMENT AND PREDICTION OF SUBMUCOSAL INVASIVE CANCER

M Sidhu
1   Westmead Hospital, Westmead, Australia
,
DJ Tate
1   Westmead Hospital, Westmead, Australia
,
MJ Bourke
1   Westmead Hospital, Westmead, Australia
› Author Affiliations
Further Information

Publication History

Publication Date:
23 April 2020 (online)

Also available at
 

Aims Analysis of the surface pit pattern (SPP), of large (>/=20mm) laterally spreading colonic lesions (LSL) can help predict the risk of submucosal invasion (SMI) has relied upon the use of dye-based chromoendoscopy (DBC) for a reliable assessment of SPP. However, the utility of virtual chromoendoscopy (VC), combined with high definition white light (HDWL) for lesion assessment remains unknown.

We sought to assess the incremental benefit of DBC in addition to HDWL plus VC for the assessment of SPP and prediction of SMI in colonic LSL referred for EMR.

Methods A prospective observational study of consecutive lesions referred for EMR at a single tertiary referral centre was performed (NCT03506321). Prior to resection lesion assessment of surface pit pattern was initially performed using HDWL + VC [Narrow band imaging (NBI), Olympus, Tokyo)] by two trained independent blinded observers. The results were recorded by a third independent observer. Thereafter, DBC (0.2% indigo carmine) was performed and a repeat assessment performed by the same blinded observers.

Results Over 22 months to September 2019, 205 consecutive LSL in 205 patients (50.7% - male) were enrolled. The overall rate of SMI was 9.2% (19/205).

Presence of a demarcated area on the lesion surface for all lesions had a NPV of 95.1%, 95%CI(90.5-97.6) for predicting SMI at histology with no additional benefit from the addition of DBC (NPV [94.6%, 95%CI(90.0-97.3)].

Rate of inter-observer agreement was high between the two blinded observers, independent of whether DBC was used; k-0.98 (SE - 0.03) with HDWL plus VC and k-0.96 (SE - 0.03) with HDWL, VC and DBC.

Conclusions The absence of a demarcated area (where a regular neoplastic pit-pattern becomes disordered) within LSL is strongly predictive for the absence of submucosal invasion histologically, can be determined without the need for dye-based chromoendoscopy and has a high rate of interobserver agreement amongst experts.