醋酸靛胭脂混合三明治染色法联合智能分光比色技术结肠镜下Pit pattern分型对结直肠病变的诊断价值

Diagnostic value of acid indigocarmine mixture sandwich staining combined with flexile spectral imaging color enhancement colonoscopy Pit pattern classification for colorectal lesions

摘要:

目的 探讨醋酸靛胭脂混合（AIM）三明治染色法联合智能分光比色技术（FICE）结肠镜下Pit pattern分型对结直肠病变的诊断价值。方法 选择2022年6月－2023年10月该院收治的100例结直肠病变患者作为研究对象，共222处病灶；分别采用普通内镜、FICE和AIM三明治染色+FICE进行检查，并记录Pit pattern分型的检出情况、病理学类型；计算不同模式下Pit pattern分型诊断的敏感度、特异度、阳性预测值（PPV）、阴性预测值（NPV）和准确度，采用Kappa检验评估不同模式下Pit pattern分型诊断与病理学检查的一致性，采用受试者操作特征曲线（ROC curve）评估诊断效能。结果 与普通内镜（74.32%）相比，FICE（92.34%）和AIM三明治染色+FICE（97.30%）Pit pattern分型检出与病理结果符合率更高，且AIM三明治染色+FICE高于FICE，差异均有统计学意义（P < 0.05）；与普通内镜相比，FICE和AIM三明治染色+FICE诊断结直肠肿瘤性病变的准确度更高，且AIM三明治染色+FICE高于FICE，差异均有统计学意义（P < 0.05）；与普通内镜相比，FICE和AIM三明治染色+FICE诊断早期结直肠癌的准确度更高，差异均有统计学意义（P < 0.05）；普通内镜、FICE和AIM三明治染色+FICE预测结直肠肿瘤性病变的曲线下面积（AUC）分别为0.815（95%CI：0.711～0.859）、0.881（95%CI：0.752～0.904）和0.933（95%CI：0.793～0.961）；普通内镜、FICE和AIM三明治染色+FICE预测早期结直肠癌的AUC分别为0.850（95%CI：0.720～0.866）、0.938（95%CI：0.764～0.951）和0.947（95%CI：0.803～0.972）；AIM三明治染色 + FICE预测结直肠肿瘤性病变和早期结直肠癌的Youden指数最大，分别为0.955和0.968。结论 AIM三明治染色 + FICE下Pit pattern分型诊断结直肠肿瘤性病变和早期结直肠癌的准确度较高，可有效提高内镜的诊治质量。

Abstract:

Objective To explore the diagnostic value of acetic acid indigocarmine mixture (AIM) sandwich staining combined with flexile spectral imaging color enhancement (FICE) technology for colonoscopic Pit pattern classification for colorectal lesions.Methods 100 patients with colorectal lesions from June 2022 to October 2023 were selected as the research subjects, and 222 lesions; Patients were examined using conventional endoscopy, FICE, AIM sandwich staining + FICE, and the detection status and pathological type of Pit pattern classification were recorded. Calculate the sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy of Pit pattern classification diagnosis under different modes. Kappa was used to evaluate the consistency between Pit pattern classification and pathological examination under different modes. Evaluate diagnostic effectiveness using receiver operator characteristic curve (ROC curve).Results Compared with ordinary endoscopy (74.32%), FICE technology (92.34%) and AIM staining + FICE (97.30%) detected more lesions that matched the pathological results through Pit pattern classification, and AIM sandwich staining + FICE detected more than FICE, the differences were statistically significant (P < 0.05). Compared with ordinary endoscopy, the diagnostic accuracy of FICE and AIM staining + FICE for colorectal carcinoma lesions were higher, and AIM sandwich staining+FICE higher than FICE, the differences were statistically significant (P < 0.05). Compared with ordinary endoscopy, the diagnostic accuracy of FICE and AIM sandwich staining + FICE for early colorectal cancer were higher, the differences were statistically significant (P < 0.05). The area under the curve (AUC) predicted by ordinary endoscopy, FICE and AIM sandwich staining + FICE for colorectal carcinoma lesions were 0.815 (95%CI: 0.711 ~ 0.859), 0.881 (95%CI: 0.752 ~ 0.904), and 0.933 (95%CI: 0.793 ~ 0.961), respectively; The AUC predicted by ordinary endoscopy, FICE and AIM sandwich staining+FICE for early colorectal cancer were 0.850 (95%CI: 0.720 ~ 0.866), 0.938 (95%CI: 0.764 ~ 0.951), and 0.947 (95%CI: 0.803 ~ 0.972), respectively. For predicting colorectal carcinoma lesions and early colorectal cancer, the Youden index of AIM staining + FICE technology was the highest, with values of 0.955 and 0.968, respectively.Conclusion Under AIM sandwich staining + FICE, Pit pattern classification can detect more carcinoma lesions and early colorectal cancer, with high diagnostic accuracy and effective improvement of endoscopic diagnosis and treatment quality.