AIM: Gastritis is an important pathological state that causes gastric atrophy and cancer. The Sydney System is a well-used classification for histological evaluation for gastritis. However, there is no concordance with endoscopic findings. In the present study, we tried to establish endoscopic criteria and diagnosis for the inflammation activity of gastric mucosa. METHODS: A prospective multicenter study was conducted and 24 facilities participated. Two hundred and seventy patients received endoscopic examinations and 15 endoscopic features were evaluated. Biopsy specimens were taken from five points, and evaluated by a single pathologist for mononuclear cell infiltration and polymorphonuclear cell infiltration. Sensitivity, specificity, positive predictive value, negative predictive value, area under curve of receiver operating characteristics (AUC/ROC) of each endoscopic finding to histological gastritis were calculated. RESULTS: There was no single endoscopic finding that was highly specific for mononuclear cell infiltration and polymorphonuclear cell infiltration. In the corpus, the combination of swelling of areae gastrica by the indigo carmine contrast method (IC method) and lack of a regular arrangement of collecting venules (RAC) in angle for mononuclear cell infiltration (0.887), and the combination of swelling of areae gastrica by the IC method and diffuse redness for polymorphonuclear cell infiltration (0.851) showed the highest AUC/ROC. In the antrum, the combination of diffuse redness and visibility of a vascular pattern for mononuclear cell infiltration (0.780), and the combination of visibility of vascular pattern and swelling of areae gastrica by the IC method for polymorphonuclear cell infiltration (0.795) showed the highest AUC/ROC. CONCLUSION: Combination of endoscopic findings can improve diagnostic accuracy, and sensitivity of examination for inflammation.
AIM: Gastritis is an important pathological state that causes gastric atrophy and cancer. The Sydney System is a well-used classification for histological evaluation for gastritis. However, there is no concordance with endoscopic findings. In the present study, we tried to establish endoscopic criteria and diagnosis for the inflammation activity of gastric mucosa. METHODS: A prospective multicenter study was conducted and 24 facilities participated. Two hundred and seventy patients received endoscopic examinations and 15 endoscopic features were evaluated. Biopsy specimens were taken from five points, and evaluated by a single pathologist for mononuclear cell infiltration and polymorphonuclear cell infiltration. Sensitivity, specificity, positive predictive value, negative predictive value, area under curve of receiver operating characteristics (AUC/ROC) of each endoscopic finding to histological gastritis were calculated. RESULTS: There was no single endoscopic finding that was highly specific for mononuclear cell infiltration and polymorphonuclear cell infiltration. In the corpus, the combination of swelling of areae gastrica by the indigo carmine contrast method (IC method) and lack of a regular arrangement of collecting venules (RAC) in angle for mononuclear cell infiltration (0.887), and the combination of swelling of areae gastrica by the IC method and diffuse redness for polymorphonuclear cell infiltration (0.851) showed the highest AUC/ROC. In the antrum, the combination of diffuse redness and visibility of a vascular pattern for mononuclear cell infiltration (0.780), and the combination of visibility of vascular pattern and swelling of areae gastrica by the IC method for polymorphonuclear cell infiltration (0.795) showed the highest AUC/ROC. CONCLUSION: Combination of endoscopic findings can improve diagnostic accuracy, and sensitivity of examination for inflammation.