Confocal Laser Endomicroscopy for in Vivo Diagnosis of Gastric Intraepithelial Neoplasia: A Feasibility Study
Zhen Li,Tao Yu,Xiu-Li Zuo,Xiao-Meng Gu,Cheng-Jun Zhou,Rui Ji,Chang-Qing Li,Peng Wang,Ting-Guo Zhang,Khek-Yu Ho,Yan-Qing Li
DOI: https://doi.org/10.1016/j.gie.2010.08.031
IF: 10.396
2010-01-01
Gastrointestinal Endoscopy
Abstract:Background: Confocal laser endomicroscopy (CLE) is a novel endoscopic modality that allows subsurface analysis of the gastric mucosa during ongoing endoscopy. Several studies have reported that this technique is of value in the diagnosis of premalignant lesions in the GI tract, but as yet no investigations have reported its application in the analysis of gastric intraepithelial neoplasia (GIN).Objective: To assess the feasibility of CLE for the identification and grading of GIN.Design: Prospective double-blind feasibility study.Setting: Qilu Hospital, Shandong University, Jinan, China.Patients: CLE images of 33 patients were first evaluated to establish the diagnostic criteria for gastric lesions. Eligible patients were then prospectively investigated by CLE using the newly established criteria.Interventions: All endoscopically suspicious lesions were examined by CLE, and CLE diagnoses were compared with corresponding histopathologic results.Main Outcome Measurements: Sensitivity, specificity, and positive and negative likelihood ratios of CLE diagnosis of biopsy-proven intraepithelial neoplasia by per-lesion analysis.Results: The sensitivity, specificity, and positive and negative likelihood ratios of CLE diagnosis of GIN were 77.8%, 81.8%, 4.28, and 0.27, respectively. The mean kappa value for interobserver agreement for the diagnosis of GIN was 0.70 among endoscopists and 0.71 between endoscopist and GI pathologist. Intraepithelial neoplasia score >= 5 differentiated high-grade from low-grade intraepithelial neoplasia with a sensitivity of 66.7% and a specificity of 88.0%.Limitations: Nonrandomized single-center study, limited number of patients.Conclusions: CLE is an acceptable and potentially useful technology for the identification and grading of GIN in vivo. The diagnostic accuracy needs to be improved. (Gastrointest Endosc 2010;72:1146-53)