Chromoendoscopy Using Blue Laser Imaging in the Prediction of Submucosal Invasion In Colorectal Neoplastic Lesions

Aims Endoscopic diagnosis of submucosal invasion allows an optimized treatment strategy for colorectal polyps. Virtual chromoendoscopy, namely with Narrow Band Imaging, can help to predict the presence and depth of submucosal invasion.Data regarding Blue Laser Imaging (BLI) for predicting deep submucosal invasive carcinoma (dSMC) are still missing. The present research aimed to evaluate the diagnostic performance of endoscopic classifications when using BLI.

Methods Consecutive nonpedunculated colorectal lesions sized ≥ 20mm were prospectively evaluated by 1-2 expert(s) gastroenterologist(s) using high-definition colonoscopy with BLI, without magnification. Paris, Lateral Spreading Tumors (LST), Kudo pit pattern and Japan NBI Expert Team (JNET) classifications were applied and correlated with histopathology after endoscopic and/or surgical resection.

Results Ninety-six lesions were included (86 patients, median age 69-years-old [IQR:64-76]), median size 30mm (IQR:20-45), consisting of: 6 (6.3 %) sessile serrated polyps; 54 (56.2 %) low-grade dysplasia; 27 (28.1 %) high-grade dysplasia (HGD)/intramucosal carcinoma (imC)/superficial submucosal carcinoma (sSMC); and 9 (9.4 %) dSMC. Eighty-eight lesions were resected by EMR (19 en bloc). Three patients underwent hybrid endoscopic submucosal dissection and 9 colorectal surgery (5 ab initio; 4 after non-curative endoscopic resection).The accuracy, sensitivity, specificity, positive predictive value and negative predictive value of JNET-Type 3 for dSMC was 95.8 %, 55.6 %, 100 %, 100 % and 95.6 %, respectively, with an area under the ROC curve (AUROC) of 0.78 (95 %CI 0.57-0.99); of JNET-Type 2b for HGD/imC/sSMC was 67.1 %, 26.9 %, 92.8 %, 58.3 % and 77.4 %, respectively; of Kudo V for dSMC was 88.5 %, 100 %, 87.4 %, 45 % and 100 %, respectively, with an AUROC of 0.94 (95 %CI 0.89-0.99)).

Conclusions JNET classification using BLI was as effective as original JNET for dSMC diagnosis. Kudo pit pattern estimated dSMC risk with high accuracy, even without magnification. Advanced imaging with BLI allows to establish the indication for endoscopic resection with high accuracy, being less effective in decision-making on which resection technique to use (en bloc versus piecemeal).

Citation: Estorninho J, Gravito-Soares E, Amaro P et al. OP179 CHROMOENDOSCOPY USING BLUE LASER IMAGING IN THE PREDICTION OF SUBMUCOSAL INVASION IN COLORECTAL NEOPLASTIC LESIONS. Endoscopy 2021; 53: S73.

Publication History

Article published online:
19 March 2021