NGHIÊN CỨU GIÁ TRỊ CỦA PHÂN LOẠI JNET TRONG TIÊN ĐOÁN MÔ BỆNH HỌC POLYP ĐẠI TRỰC TRÀNG

Abstract

Introduction: Colorectal polyps are closely related to colorectal cancer. The JNET classification with magnified narrow-banding imaging (NBI) aids in predicting the histology of colorectal polyps, thus allowing for the appropriate treatment. However, the diagnostic yield of JNET classification on NBI mode with dual focus (DF) magnification in predicting the histology of colorectal polyps is still under-researched in Vietnam. Objective: To determine the diagnostic stratification ability of JNET classification on NBI-DF mode in predicting the histology of colorectal polyps. Methods: A cross-sectional descriptive study was conducted on 666 patients with 1087 colorectal adenomatous polyps from October 2021 to February 2023 at the University Medical Center. Data were analyzed using SPSS 25.0 software. The EVIS EXERA III CV-190 processing system and CF-HQ190I endoscope were used to evaluate the polyps according to JNET classification. Results: The sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of the JNET classification types for predicting the histology of colorectal adenomatous polyps were as follows: type 1, 86.5%, 95.7%, 88.3%, 95.0%, and 93.2%; type 2A, 91.9%, 81.4%, 90%, 84%, and 87.7%; type 2B, 54.7%, 96.6%, 54.7%, 96.6%, and 93.7%; type 3, 66.7%, 99.9%, 93.3%, 99.4%, and 99.4%. The sensitivity for differentiating neoplastic lesions from benign non-neoplasia lesions was 97.8%, the specificity for distinguishing malignant neoplasia from benign neoplasia was 95.9%, and the specificity in the differentiation deep submucosal cancer from other neoplasia was 99,8%. Conclusion: JNET classification based on NBI-DF has a high value in predicting the histology of colorectal polyps. Thus, JNET might contribute to appropriate treatment choices and avoid unnecessary surgery. This classification should be utilized in the Vietnamese setting.