Value of Magnifying Endoscopy With Narrow-Band Imaging and Confocal Laser Endomicroscopy in Detecting Gastric Cancerous Lesions

Although the respective potentials of magnifying endoscopy with narrow-band imaging (ME-NBI) and confocal laser endomicroscopy (CLE) in predicting gastric cancer has been well documented, there is a lack of studies in comparing the value and diagnostic strategy of these 2 modalities. Our primary aim is to investigate whether CLE is superior to ME-NBI for differentiation between gastric cancerous and noncancerous lesions. A secondary aim is to propose an applicable clinical strategy. We conducted a diagnostic accuracy study involving patients with suspected gastric superficial cancerous lesions. White light endoscopy, ME-NBI, and CLE were performed diagnostic accuracy, sensitivity, specificity, positive predictive value, and negative predictive value between ME-NBI and CLE were assessed, as well as agreements between ME-NBI/CLE and histopathology. This study involved 86 gastric lesions in 82 consecutive patients who underwent white light endoscopy, ME-NBI, and CLE before biopsy. The accuracy, sensitivity, and specificity for ME-NBI were 93.75%, 91.67%, and 95.45%, compared with 91.86%, 90%, and 93.48%, respectively, for CLE, for discrimination cancerous/noncancerous lesion (all P > 0.05). For undifferentiated/differentiated adenocarcinoma, CLE had a numerically but not statistically significantly higher accuracy than ME-NBI (81.25% vs 73.33%, P = 0.46). Agreements between ME-NBI/CLE and histopathology were near perfect (ME-NBI, κ = 0.87; CLE, κ = 0.84). CLE is not superior to ME-NBI for discriminating gastric cancerous from noncancerous lesions. Endoscopist could make an optimal choice according to the specific indication and advantages of ME-NBI and CLE in daily practices.