Abstract：
OBJECTIVE: There were limited data on the diagnostic efficacy of liquid-based cytology (LBC) for pancreatic tissues acquired by percutaneous ultrasound-guided fine-needle aspiration (US-FNA). This study aimed to evaluate the diagnostic value of LBC acquired via percutaneous US-FNA for pancreatic tumors compared with LBC combined with smear cytology (SC).
METHODS: A retrospective database search (January 2014 and February 2022) was performed for patients who underwent percutaneous US-FNA with both LBC and SC. Clinical and pathological data were collected from 298 patients; eventually, 251 cases met the inclusion criteria. Diagnostic accuracy, sensitivity (SEN), specificity (SPE), positive predictive value (PPV) and negative predictive value (NPV) were compared. Rapid on-site evaluation (ROSE) was not available in all cases.
RESULTS: Based on the pancreaticobiliary cytology guidelines published by the Papanicolaou Society of Cytopathology, 224 (89.2%), 13 (5.2%) and 14 (5.6%) cases were diagnosed as malignant, pre-malignant and benign lesions, respectively. The diagnostic accuracy of the LBC + SC (88.5%) was better than that of LBC (87.3%) but without statistical significance (P = 0.125). The SEN, SPE, PPV and NPV were 87.5%, 85.2%, 98.0% and 45.1%, respectively, in the LBC group and 88.8%, 85.2%, 98.0% and 47.9%, respectively, in the LBC + SC group. According to univariate and multivariate analyses, there were no factors have significant association with the diagnostic sensitivity of LBC.
CONCLUSIONS: LBC obtained via percutaneous US-FNA provides good diagnostic value for pancreatic lesions and there was no significant difference between the diagnostic accuracy of LBC and LBC + SC when ROSE was unavailable.
METHODS: A retrospective database search (January 2014 and February 2022) was performed for patients who underwent percutaneous US-FNA with both LBC and SC. Clinical and pathological data were collected from 298 patients; eventually, 251 cases met the inclusion criteria. Diagnostic accuracy, sensitivity (SEN), specificity (SPE), positive predictive value (PPV) and negative predictive value (NPV) were compared. Rapid on-site evaluation (ROSE) was not available in all cases.
RESULTS: Based on the pancreaticobiliary cytology guidelines published by the Papanicolaou Society of Cytopathology, 224 (89.2%), 13 (5.2%) and 14 (5.6%) cases were diagnosed as malignant, pre-malignant and benign lesions, respectively. The diagnostic accuracy of the LBC + SC (88.5%) was better than that of LBC (87.3%) but without statistical significance (P = 0.125). The SEN, SPE, PPV and NPV were 87.5%, 85.2%, 98.0% and 45.1%, respectively, in the LBC group and 88.8%, 85.2%, 98.0% and 47.9%, respectively, in the LBC + SC group. According to univariate and multivariate analyses, there were no factors have significant association with the diagnostic sensitivity of LBC.
CONCLUSIONS: LBC obtained via percutaneous US-FNA provides good diagnostic value for pancreatic lesions and there was no significant difference between the diagnostic accuracy of LBC and LBC + SC when ROSE was unavailable.
摘要：
目的：关于液基细胞学（LBC）对经皮超声引导细针穿刺（US-FNA）获得的胰腺组织的诊断效能的数据有限。本研究旨在评估通过经皮US-FNA获得的LBC与LBC联合涂片细胞学（SC）对胰腺肿瘤的诊断价值。
方法：回顾性数据库搜索（2014年1月和2022年2月），对同时接受LBC和SC的经皮US-FNA患者进行。收集298例患者的临床和病理资料;最终,251例符合纳入标准。诊断准确性,灵敏度（SEN），特异性（SPE），比较阳性预测值(PPV)和阴性预测值(NPV)。并非在所有情况下都可以进行快速现场评估（ROSE）。
结果：根据Papanicolaou细胞病理学学会发布的胰胆管细胞学指南，224（89.2%），13例（5.2％）和14例（5.6％）被诊断为恶性，癌前和良性病变,分别。LBC+SC(88.5%)诊断准确率优于LBC(87.3%),但无统计学意义(P=0.125)。SEN,SPE,PPV和NPV分别为87.5%，85.2%,98.0%和45.1%,分别,在LBC组和88.8%，85.2%,98.0%和47.9%，分别,LBC+SC组。根据单变量和多变量分析，没有因素与LBC的诊断敏感性有显著关联.
结论：通过经皮US-FNA获得的LBC对胰腺病变具有良好的诊断价值，当ROSE不可用时，LBC和LBC+SC的诊断准确性没有显着差异。
方法：回顾性数据库搜索（2014年1月和2022年2月），对同时接受LBC和SC的经皮US-FNA患者进行。收集298例患者的临床和病理资料;最终,251例符合纳入标准。诊断准确性,灵敏度（SEN），特异性（SPE），比较阳性预测值(PPV)和阴性预测值(NPV)。并非在所有情况下都可以进行快速现场评估（ROSE）。
结果：根据Papanicolaou细胞病理学学会发布的胰胆管细胞学指南，224（89.2%），13例（5.2％）和14例（5.6％）被诊断为恶性，癌前和良性病变,分别。LBC+SC(88.5%)诊断准确率优于LBC(87.3%),但无统计学意义(P=0.125)。SEN,SPE,PPV和NPV分别为87.5%，85.2%,98.0%和45.1%,分别,在LBC组和88.8%，85.2%,98.0%和47.9%，分别,LBC+SC组。根据单变量和多变量分析，没有因素与LBC的诊断敏感性有显著关联.
结论：通过经皮US-FNA获得的LBC对胰腺病变具有良好的诊断价值，当ROSE不可用时，LBC和LBC+SC的诊断准确性没有显着差异。
