非放大内镜应用窄带成像国际结直肠内镜分型判定结直肠息肉样病变性质的临床价值

Clinical value of application by using narrow-band-imaging international colorectal endoscopic classification based on non-magnifying endoscopy in distinguishing the nature of colorectal polypoid lesions

目的评价非放大内镜应用窄带成像国际结直肠内镜(NICE)分型判定结直肠息肉样病变性质的效能及其临床实用价值。方法选取2017年3月至2019年12月在上海市浦东新区公利医院消化内镜中心接受肠镜检查的患者286例,共计547处结直肠息肉样病变纳入分析。以病理诊断结果为标准,分别计算NICE分型判定结直肠息肉样病变性质的灵敏度、特异度、准确率、阳性预测值和阴性预测值,并与病理诊断结果及Sano分型结果进行对比分析及一致性检验。同时对经验组和普通组两组医师的诊断结果进行一致性评价。结果NICE分型判定结直肠息肉样病变病理分型的灵敏度、特异度、准确率、阳性预测值、阴性预测值分别为97.30%、87.88%、96.16%、98.32%、81.69%,判定≤0.5 cm病变病理分型的灵敏度、特异度、准确率、阳性预测值、阴性预测值分别为92.31%、87.23%、88.33%、66.67%、97.62%。NICE分型判定不同直径结直肠息肉样病变性质的灵敏度、特异度及准确率等比较差异均无统计学意义(均P>0.05)。NICE分型结果与病理诊断结果及Sano分型结果一致性均良好(Kappa=0.755、0.824,均P<0.01)。两组医师在诊断灵敏度、特异度、准确率等方面比较差异均无统计学意义(均P>0.05),组间观察一致性良好(Kappa=0.895,P<0.01)。结论非放大内镜应用NICE分型具有良好的判定结直肠息肉样病变性质的效能。建议仅将结直肠肿瘤性息肉纳入内镜下切除或外科手术治疗,而对于结直肠微小息肉(≤0.5 cm)可以执行“切除-丢弃”和“不切除”的这一内镜处置策略。

Objective To evaluate the application of narrow-band-imaging international colorectal endoscopic(NICE)classification in distinguishing the natures of colorectal polypoid lesions which based on non-magnifying endoscopy.Methods A total of 286 patients with 547 colorectal polypoid lesions who underwent colonoscopy in the Digestive Endoscopy Center of Shanghai Pudong New Area Gongli Hospital from March 2017 to December 2019 were included in the analysis.According to the pathological results of colorectal polypoid lesions,the sensitivity,specificity,accuracy,positive predictive value and negative predictive value of NICE classification to determine the natures of colorectal polypoid lesions were calculated respectively.The results of NICE classification were compared with those of pathology and Sano classification.The consistent test between NICE classification and pathology,and between NICE classification and Sano classification was performed respectively.Meanwhile,the consistency of diagnosis by experienced physicians(working experience≥5 years)and ordinary physicians(working experience<5 years)was also tested.Results The sensitivity,specificity,accuracy,positive predictive value and negative predictive value of NICE classification to determine the pathological types of colorectal polypoid lesions were 97.30%,87.88%,96.16%,98.32%and 81.69%.While for those with diameter≤0.5 cm,the sensitivity,specificity,accuracy,positive predictive value and negative predictive value of NICE classification were 92.31%,87.23%,88.33%,66.67%,97.62%.There were no significant differences in the diagnostic sensitivity,specificity and accuracy of NICE classification for distinguishing the natures of colorectal polypoid lesions with different diameters(all P>0.05).Results of NICE classification were consistent with the results of pathologi-cal diagnosis and Sano classification(Kappa=0.755 and 0.824,all P<0.01).There were no significant differences in the diagnostic sensitivity,specificity and accuracy between two physician groups(all P>0.05),and the observation consistency of two groups was excellent(Kappa=0.895,P<0.01).Conclusion NICE classification based on non-magnifying endoscopy has an excellent effect for distinguishing the natures of colorectal polypoid lesions.It is suggested that endoscopic resection or surgical operation should be performed only for colorectal tumorous lesions,while for small colorectal polypoid lesions(≤0.5 cm),the"resect and discard"and"do-not-resect"endoscopic management strategies can be adopted.