日本 NBI 专家小组分类法（双焦距放大）对结直肠肿瘤的诊断效能。
Diagnostic efficacy of the Japan NBI Expert Team classification with dual-focus magnification for colorectal tumors.
机构信息
Department of Gastroenterology and Hepatology, Tokyo Medical University Hospital, 6-7-1 Nishishinjuku, Shinjuku-ku, Tokyo, 160-0023, Japan.
Endoscopy Center, Tokyo Medical University Hospital, Tokyo, Japan.
出版信息
Surg Endosc. 2022 Jul;36(7):5032-5040. doi: 10.1007/s00464-021-08863-7. Epub 2021 Nov 29.
BACKGROUND AND AIMS
The Japan NBI Expert Team (JNET) classification is the first unified classification criteria for colorectal tumors using magnifying narrow-band imaging (NBI) in Japan. However, the diagnostic stratification ability of the JNET classification with dual-focus magnifying NBI (DF-JNET) has remained obscure. The aim of this study was to validate the diagnostic stratification ability of DF-JNET for colorectal tumors in two Japanese referral centers.
METHODS
A multicenter retrospective image evaluation study was conducted by three experienced endoscopists, including an original JNET member who was also involved in establishing the diagnostic criteria. A total of two images, namely, one representative non-magnified white light image and one representative DF-NBI image for each of the 557 consecutive lesions were used in the evaluation study. The diagnostic value of DF-JNET was calculated based on the evaluation data.
RESULTS
The sensitivity, specificity, positive and negative predictive values, and accuracy of DF-JNET Type 1 for differentiating between non-neoplastic and neoplastic lesions were 78.1%, 98.6%, 89.1%, 96.8%, and 95.9%, respectively; of Type 2A lesions for differentiating low-grade dysplasia from others were 98.0%, 76.5%, 94.9%, 89.7%, and 94.1%, respectively; of Type 2B lesions for differentiating high-grade dysplasia and shallow submucosal invasive carcinoma from others were 43.5%, 99.1%, 66.7%, 97.6%, and 96.8%, respectively; and of Type 3 lesions for differentiating deep submucosal invasive carcinoma from others were 83.3%, 99.5%, 62.5%, 99.8%, and 99.3%, respectively.
CONCLUSIONS
All DF-JNET types had an over 90% diagnostic accuracy for the histological prediction of colorectal tumors. DF-JNET might contribute to appropriate treatment choices, such as endoscopic resection or surgery, not only in Japan but also in Western countries in which the use of optical zoom endoscopy is limited.
背景与目的
日本 NBI 专家团队（JNET）分类是日本首次使用放大窄带成像（NBI）对结直肠肿瘤进行的统一分类标准。然而，双焦点放大 NBI（DF-JNET）的 JNET 分类的诊断分层能力仍不清楚。本研究的目的是验证 JNET 分类在两个日本转诊中心对结直肠肿瘤的诊断分层能力。
方法
通过三位经验丰富的内镜医生进行多中心回顾性图像评估研究，其中包括一位参与建立诊断标准的原始 JNET 成员。评估研究共使用了 557 例连续病变的 2 张图像，即每例病变的 1 张代表性非放大白光图像和 1 张代表性 DF-NBI 图像。根据评估数据计算了 DF-JNET 的诊断价值。
结果
DF-JNET Type 1 区分非肿瘤性和肿瘤性病变的敏感性、特异性、阳性和阴性预测值以及准确性分别为 78.1%、98.6%、89.1%、96.8%和 95.9%；Type 2A 病变区分低级别异型增生与其他病变的敏感性、特异性、阳性和阴性预测值以及准确性分别为 98.0%、76.5%、94.9%、89.7%和 94.1%；Type 2B 病变区分高级别异型增生和浅黏膜下浸润癌与其他病变的敏感性、特异性、阳性和阴性预测值以及准确性分别为 43.5%、99.1%、66.7%、97.6%和 96.8%；Type 3 病变区分深黏膜下浸润癌与其他病变的敏感性、特异性、阳性和阴性预测值以及准确性分别为 83.3%、99.5%、62.5%、99.8%和 99.3%。
结论
所有 DF-JNET 类型对结直肠肿瘤的组织学预测均具有 90%以上的诊断准确性。DF-JNET 可能有助于做出适当的治疗选择，例如内镜切除或手术，不仅在日本，而且在光学变焦内镜应用受限的西方国家也是如此。
Zotero 插件