Department of Gastroenterology and Hepatology, Nagoya University Graduate School of Medicine, Nagoya, Japan.
Department of Endoscopy, Nagoya University Hospital, Nagoya, Japan.
Pancreatology. 2021 Mar;21(2):451-458. doi: 10.1016/j.pan.2020.12.012. Epub 2020 Dec 31.
/Objectives: Endoscopic ultrasound elastography (EUS-EG) is useful for diagnosis of small solid pancreatic lesions (SPLs), particularly in excluding pancreatic cancer (PC), but its dependence on main pancreatic duct dilatation (MPDD) has not been examined. We aimed to investigate EUS-EG for diagnosis of small SPLs with and without MPDD.
Patients with pathologically diagnosed SPLs of ≤20 mm were included and retrospectively analyzed. Using the blue:green ratio, an EUS-EG image was classified as blue-dominant, equivalent, or green-dominant. Using multiple EUS-EG images per patient, a lesion with a greater number of blue-dominant than green-dominant images was classified as stiff, and the others as soft. EUS-EG images in random order were judged by three raters. Considering stiff SPLs as PC, diagnostic performance of EUS-EG was examined for SPLs with and without MPDD.
Of 126 cases analyzed, 65 (52%) were diagnosed as PC, and 63 (50%) had MPDD. A total of 1077 EUS-EG images were examined (kappa coefficient = 0.783). Lesions were classified as stiff in 91 cases and soft in 35 (kappa coefficient = 0.932). The ratio of stiff to soft lesions was significantly higher in PC than in non-PC (62:3 vs. 29:32, P < 0.001). The sensitivity, specificity, and negative predictive value of a stiff lesion with vs. without MPDD for diagnosis of PC were 94%, 23%, and 50% vs. 100%, 60%, and 100%, respectively.
Using the EUS-EG stiffness classification for small SPLs, PC can be excluded with high confidence and concordance for a soft lesion without MPDD.
目的:内镜超声弹性成像(EUS-EG)有助于诊断小的胰腺实体病变(SPL),特别是在排除胰腺癌(PC)方面,但尚未研究其对主胰管扩张(MPDD)的依赖性。我们旨在研究 EUS-EG 对有或无 MPDD 的小 SPL 的诊断价值。
纳入经病理诊断为直径≤20mm 的 SPL 患者,并进行回顾性分析。使用蓝绿比值,EUS-EG 图像被分为蓝色占优、均等或绿色占优。对于每个患者的多个 EUS-EG 图像,如果蓝色占优的图像多于绿色占优的图像,则将病变归类为硬,其余归类为软。EUS-EG 图像以随机顺序由 3 位判读者进行判断。将硬 SPL 视为 PC,根据有无 MPDD 对 EUS-EG 对 SPL 的诊断性能进行了检查。
在分析的 126 例中,65 例(52%)被诊断为 PC,63 例(50%)有 MPDD。共检查了 1077 个 EUS-EG 图像(kappa 系数=0.783)。91 例病变被归类为硬,35 例为软(kappa 系数=0.932)。PC 中硬病变与软病变的比例明显高于非 PC(62:3 比 29:32,P<0.001)。有 MPDD 和无 MPDD 的硬病变诊断 PC 的敏感性、特异性和阴性预测值分别为 94%、23%和 50%比 100%、60%和 100%。
对于小 SPL,使用 EUS-EG 硬度分类,在无 MPDD 的情况下,软病变可高度置信且一致地排除 PC。