Dye-based Chromoendoscopy Added Nothing to Prediction of Cancer in Colorectal Nonpedunculated Lesions
Douglas K. Rex, MD, MASGE, reviewing Sidhu M, et al. Gastrointest Endosc 2021 Dec 4.
In a prospective study, 2 experts independently assessed the surfaces of 400 large, laterally spreading colorectal lesions (median size, 35 mm) for predictors of submucosal invasive cancer (SMIC). This predictor was the presence of a demarcated area of vascular disruption.
Overall sensitivity and specificity of SMIC with high-definition white-light and virtual chromoendoscopy were 73.1% and 92.3%, respectively. As has been previously shown, in flat versus nodular lesions, sensitivity was 90.9% versus 60%, and negative predictive values were 99.5% versus 91.7%, respectively. The experts identified demarcated areas (with the addition of high-definition white-light plus virtual chromoendoscopy) in 51 out of 400 lesions (12.3%). Dye-based chromoendoscopy added nothing to the accuracy of predicting SMIC in either the flat or nodular lesions. The kappa was 0.96 for interobserver agreement between the 2 experts in identifying demarcated areas. There were only 3 cases of discordance between the 2 observers.
Note to readers: At the time we reviewed this paper, its publisher noted that it was not in final form and that subsequent changes might be made.
Sidhu M, Shahidi N, Vosko S, van Hattem WA, Tate DJ, Bourke MJ. The incremental benefit of dye-based chromoendoscopy to predict the risk of submucosal invasive cancer in large nonpedunculated colorectal polyps. Gastrointest Endosc 2021 Dec 4. (Epub ahead of print) (https://doi.org/10.1016/j.gie.2021.11.032)