Evaluation of blue laser endoscopy for detecting colorectal non-pedunculated adenoma

Ying Zhu1, Wei Hu1, Fang Wang1, Yan Zhou3, Guili Xia1, Jianguo Xu2, Wei Gong1

Arab J Gastroenterol. 2021 Mar 15;S1687-1979(20)30135-0. doi: 10.1016/j.ajg.2020.12.002. Online ahead of print.

Background and study aims: Non-pedunculated lesions are easily missed on endoscopy, and histopathological examination shows that some of these lesions are adenomas. Adenoma is a precursor of colorectal cancer, a common tumor of the digestive tract. This study was conducted to compare the detection efficacy of non-pedunculated lesions in the same patient under different modes of blue laser endoscopy and to determine whether the surface pattern of the sample was consistent with its histopathological results.

Patients and methods: A total of 91 patients with non-pedunculated lesions diagnosed at our hospital between April 2018 and March 2019 were included in this study. White light imaging (WLI), linked color imaging (LCI), and blue laser imaging (BLI) modes were used to record the location, number, and Hiroshima classification of the surface patterns of the non-pedunculated lesions. The lesions were removed by different endoscopic excision methods for histopathological examination; the histopathological results were compared with the surface patterns.

Results: A total of 105, 198, and 223 lesions were detected using the WLI, BLI, and LCI modes, respectively. The Wilcoxon signed rank test revealed a significant difference in the number of lesions detected using each observation mode (p < 0.01). The non-pedunculated lesions were primarily located in the rectum and transverse colon, followed by the sigmoid, descending, and ascending colon. The efficacy of LCI and BLI modes was better than that of WLI mode for detecting the non-pedunculated lesions measuring < 5 mm in size (p < 0.05). The surface pattern was not detected by the WLI mode. The surface patterns detected using the LCI and BLI modes were primarily types A and B. Histopathological results of the non-pedunculated lesions included inflammatory polyp, hyperplastic polyp, tubular adenoma, and adenoma. Surface patterns could not be detected using the WLI mode. The McNemar’s test revealed a significant difference between the WLI mode findings and the histopathological results (p < 0.01). No significant difference was observed between the histopathological results and the surface patterns detected using the LCI mode (kappa = 0.57); the agreement was poor. There was also no significant difference between the histopathological results and the surface patterns detected using the BLI mode (kappa test, p < 0.01; kappa = 0.88); hence, there was good agreement between the surface patterns detected using the BLI mode and the histopathological results.

Conclusion: The detection rate of colorectal non-pedunculated lesions may be improved using blue laser endoscopy. Non-pedunculated colorectal adenomas could be identified more accurately using the BLI mode, which might improve the adenoma detection rate, thus indicating that BLI is a feasible option in the practical settings.

1 Department of Gastroenterology, Shenzhen Hospital of Southern Medical University, Shenzhen, Guangdong 518100, China
2 Department of Liver Disease Center, Shenzhen Hospital of Southern Medical University, Shenzhen, Guangdong 518100, China
3 Information Management Section, Bethune International Peace Hospital, Shijiazhuang City, Hebei Province 050000, China