Background and Aims
To establish a modified model of the Kyoto classification score and verify its accuracy for predicting Helicobacter pylori (Hp) infection during endoscopy.
Patients who underwent gastroscopy from June 2020 to March 2021 were included in this study. Atrophy, intestinal metaplasia, hypertrophy of gastric fold, nodularity, diffuse redness, sticky mucus, spotty redness, xanthoma, map-like redness, fundic gland polyp, and regular arrangement of collecting venules (RAC) were recorded according to the Kyoto classification of gastritis. The Hp infection status of the participants was determined by 13C breath test, anti-Hp antibody, and histopathological hematoxylin-eosin staining. The modified Kyoto classification scoring model was established based on univariate analysis and logistic regression analysis. The modified scoring model was used to judge the status of Hp infection in patients undergoing gastroscopy from July to September 2021 and evaluate the accuracy of the prediction.
A total of 667 participants comprised the derivation data set, including 326 cases of Hp infection and 341 cases of Hp non-infection. Atrophy, hypertrophy of gastric fold, nodularity, diffuse redness, sticky mucus, and spotty redness were associated with Hp current infection. Thus, a new scoring model termed the modified Kyoto classification scoring model was constructed, which included atrophy, hypertrophy of gastric fold, nodularity, diffuse redness, sticky mucus, spotty redness, fundic gland polyp, and RAC as indicators. A total of 808 subjects, including 251 Hp-positive patients, comprised the validation data set to test the model.
The modified Kyoto classification scoring model improves the accuracy of endoscopic determination of Hp current infection and has clinical application potential in Chinese population.