Minimal Influence of Expertise on the Evaluation of Colorectal Neoplastic Lesions by Confocal Laser Endomicroscopy

Xue Min Yuan,Zhen Li,Rui Ji,Chang Qing Li,Xiu Li Zuo,Tao Yu,Xiao Meng Gu,Jun Liu,Peng Wang,Ming Li,Yan Qing Li
DOI: https://doi.org/10.1111/jgh.12357
2014-01-01
Journal of Gastroenterology and Hepatology
Abstract:Background and AimsThree different confocal laser endomicroscopy (CLE) diagnostic systems including Maiz, Sanduleanu, and Qilu had been developed to differentiate between benign and neoplastic colorectal lesions. This study was designed to compare and evaluate these three diagnostic systems by different levels of expertise.MethodsThirty-nine patients with 50 colorectal polyps, including 23 hyperplastic polyps and 27 adenomas diagnosed by histopathology, were recruited. Four confocal images (two superficial images and two deeper images) and one conventional white-light endoscopic image were selected from each of the 50 lesions in this study by an experienced endomicroscopist. Selected images were evaluated by three experienced CLE investigators and three non-experienced ones. The accuracy and interobserver agreement between experienced and non-experienced investigators for prediction of colorectal neoplasia were determined.ResultsThe diagnostic accuracy was 84%, 81%, and 87%, respectively, using the Maiz, Sanduleanu, and Qilu diagnostic system, while the sensitivity was 85%, 79%, and 85%, the specificity was 83%, 84%, and 89%, respectively. There is no significant difference on diagnostic accuracy between experienced and non-experienced investigators. In addition, there is a short learning curve for non-experienced CLE investigators identified in this study.ConclusionsThe three diagnostic systems for the prediction of colorectal hyperplastic polyp or adenoma have a high accuracy, sensitivity, and specificity. The diagnostic accuracy was not significantly influenced by the expertise in CLE.
What problem does this paper attempt to address?