作者：林贵德 1,2，马东升2，周伟新3，杨辉1

单位：1.广州医科大学附属第二医院 消化内科，广东 广州 510000；2.深圳市龙岗区第二人民医院 内镜科，广东 深圳 518000；3.深圳市龙岗区第二人民医院 病理科，广东 深圳 518000

Authors: Lin Guide1,2，Ma Dongsheng2，Zhou Weixin3，Yang Hui1

Unit: 1.Department of Gastroenterology, the Second Affiliated Hospital of Guangzhou Medical University, Guangzhou 510000，Guangdong，China；2.Department of Endoscopy, the Second People's Hospital of Longgang District, Shenzhen 518000，Guangdong，China；3.Department of Pathology, the Second People's Hospital of Longgang District, Shenzhen 518000，Guangdong，China

摘要：

目的 评估及比较内镜下蓝激光成像(blue laser imaging,BLI)基础上的国际结直肠内镜窄带成像技术(narrow band imaging international colorectal endoscopic,NICE)分型及日本窄带成像技术专家组(Japan Narrow-Band Imaging Expert Team, JNET)分型诊断结直肠肿瘤的临床价值。方法 回顾性分析内镜下BLI观察的130处结直肠息肉,应用NICE分型及JNET分型进行分类,对比分析内镜切除标本的病理结果。结果 病理诊断非肿瘤性息肉37处,肿瘤性息肉93处。NICE分型诊断肿瘤性息肉的总体敏感度、特异度、阳性预测值、阴性预测值和准确率分别为79.6%、86.5%、93.7%、62.7%和81.5%,JNET分型诊断肿瘤性息肉的总体敏感度、特异度、阳性预测值、阴性预测值和准确率分别为89.2%、97.3%、98.8%、78.3%和91.5%。放大镜观察JNET分型诊断肿瘤性病准确率明显优于非放大镜观察的NICE分型,差异有统计学意义(X2=5.58,P<0.05)。在直径臆5 mm的92处息肉中,NICE分型诊断肿瘤性息肉的总体敏感度、特异度和准确率分别为67.8%、87.9%和75.0%,JNET分型诊断肿瘤性息肉的总体敏感度、特异度和准确率分别为84.7%、97.0%和89.1%,两者准确率差异有统计学意义(X2=6.24,P=0.012)。而>5 mm的38处息肉先后依据NICE分型和JNET分型诊断肿瘤性息肉的敏感度、特异度和准确率差异均无统计学意义(均P>0.05)。结论 蓝激光内镜下JNET分型诊断肿瘤性病变明显优于NICE分型,其主要优势在于诊断微小息肉。

关键词：结直肠息肉；蓝色激光成像；放大内镜；NICE分型；JNET分型

Abstract：

Objective To evaluate and compare the clinical value of blue laser imaging (BLI) endoscopy in the diagnosis of colorectal neoplasm by NICE classification and JNET classification. Methods 130 colorectal polyps observed by BLI were analyzed retrospectively. NICE Classification and JNET classification were used to classify the polyps, and the pathological results of endoscopic resection specimens were compared and analyzed. Results Among 130 polyps, 51 were classified as type 1, 79 as type 2 according to NICE classification; 46 as type 1, 84 as type 2 according to JNET classification, 37 were non -tumorous polyps and 93 were tumorous polyps. The overall sensitivity, specificity, positive predictive value, negative predictive value and accuracy of NICE Classification for colorectal neoplasm were 79.6%, 86.5%, 93.7%, 62.7% and 81.5% respectively. The overall sensitivity, specificity, positive predictive value, negative predictive value and accuracy of JNET Classification for colorectal neoplasm were 89.2%, 97.3%, 98.8%, 78.3% and 91.5% respectively. The difference in accuracy between NICE Classification and JNET Classification was statistically significant (X2=5.58, P<0.05). the overall sensitivity, specificity and accuracy of 92 polyps with diameter≤5 mm were 67.8% , 87.9% , and 75.0% , respectively, according to NICE classification, and 84.7% , 97.0% , and 89.1%, respectively, according to JNET classification, the difference in accuracy between the two classification was statistically significant (X2 =6.24, P=0.012).There was no significant difference in sensitivity, specificity and accuracy between NICE classification and JNET classification in the diagnosis of 38 polyps with diameter>5 mm(P>0.05). Conclusions JNET classification with blue laser imaging endoscopy is better than that of NICE classification in the diagnosis of colorectal neoplasm, and its main advantage lies in diminutive polyps.

Key Words: Colorectal polyps；Blue laser imaging；Magnifying endoscopy；NICE classification；JNET classification