Hiroto Suzuki1, Takeshi Yamamura2, Masanao Nakamura1, Chen-Ming Hsu3,4, Ming-Yao Su3,4, Tsung-Hsing Chen3,4, Cheng-Tang Chiu3,4, Yoshiki Hirooka5, Hidemi Goto1. 1. Department of Gastroenterology and Hepatology, Nagoya University Graduate School of Medicine, Nagoya, Japan. 2. Department of Endoscopy, Nagoya University Hospital, Nagoya, Japan, tyamamu@med.nagoya-u.ac.jp. 3. Department of Gastroenterology and Hepatology, Linkou Chang Gung Memorial Hospital, Taoyuan, Taiwan. 4. Department of Gastroenterology and Hepatology, Chang Gung University College of Medicine, Taoyuan, Taiwan. 5. Department of Endoscopy, Nagoya University Hospital, Nagoya, Japan.
Abstract
BACKGROUND: The Japan narrow-band imaging Expert Team (JNET) classification of colorectal polyps based on magnifying endoscopy is used in Japan, but not worldwide. The objective of this study was to clarify differences of diagnostic accuracy between JNET users in Japan and non-JNET users in other countries. METHODS: A total of 185 colorectal tumors were assessed. Six endoscopists (3 each from Japan and Taiwan) participated in the study. The Japanese endoscopists normally used the JNET classification and the Taiwanese endoscopists normally used the narrow-band imaging International Colorectal Endoscopic classification for diagnosis of colorectal tumors. After receiving a lecture on the JNET classification, they all observed one blue laser imaging magnified image per lesion and performed diagnosis based on the JNET classification. RESULTS: Diagnostic ability was equivalent for Type 1, Type 2A, and Type 2B. The sensitivity, specificity, positive predictive value (PPV), and negative predictive value of Type 3 for deep submucosal invasive carcinoma was, respectively, 44.4, 98.3, 57.1, and 97.2% in Group J and 70.0, 94.7, 40.4, and 98.4% in Group T. The PPV for diagnosis of Type 3 with a high confidence was significantly higher in Group J than in Group T (81.8% [55.4-94.6] vs. 44.4% [33.6-50.9], p < 0.05). CONCLUSIONS: The PPV for Type 3 differed between the 2 groups, suggesting the need to become familiar with differentiation between Type 2B and Type 3.
BACKGROUND: The Japan narrow-band imaging Expert Team (JNET) classification of colorectal polyps based on magnifying endoscopy is used in Japan, but not worldwide. The objective of this study was to clarify differences of diagnostic accuracy between JNET users in Japan and non-JNET users in other countries. METHODS: A total of 185 colorectal tumors were assessed. Six endoscopists (3 each from Japan and Taiwan) participated in the study. The Japanese endoscopists normally used the JNET classification and the Taiwanese endoscopists normally used the narrow-band imaging International Colorectal Endoscopic classification for diagnosis of colorectal tumors. After receiving a lecture on the JNET classification, they all observed one blue laser imaging magnified image per lesion and performed diagnosis based on the JNET classification. RESULTS: Diagnostic ability was equivalent for Type 1, Type 2A, and Type 2B. The sensitivity, specificity, positive predictive value (PPV), and negative predictive value of Type 3 for deep submucosal invasive carcinoma was, respectively, 44.4, 98.3, 57.1, and 97.2% in Group J and 70.0, 94.7, 40.4, and 98.4% in Group T. The PPV for diagnosis of Type 3 with a high confidence was significantly higher in Group J than in Group T (81.8% [55.4-94.6] vs. 44.4% [33.6-50.9], p < 0.05). CONCLUSIONS: The PPV for Type 3 differed between the 2 groups, suggesting the need to become familiar with differentiation between Type 2B and Type 3.