探讨细针穿刺细胞学(FNAC)检查对乳腺肿块良恶性筛查及其对乳腺癌分型诊断的准确性。

根据纳入、排除标准，选取2015年2月至2017年2月清华大学第一附属医院诊治的279例同时行FNAC和组织病理学检查的乳腺肿块患者进行回顾性研究。将患者的FNAC和病理组织学检查结果进行对照分析。采用一致性检验Kappa值评价FNAC与病理组织学检查在乳腺肿块筛查及分型中的一致性，并用受试者工作特征(ROC)曲线下面积评价FNAC的诊断准确性。

在279例患者中，FNAC诊断乳腺病变的准确率为90.3%(252/279)。FNAC诊断乳腺病变的敏感度为85.0%(125/147)，特异度为96.2%(127/132)，假阳性率为3.8%(5/132)，假阴性率为15.0%(22/147)，阳性似然比为22.4,阴性似然比为0.2。FNAC与组织病理学检查方法的一致性检验显示，Kappa值为0.81(P<0.001)。FNAC的诊断性能评价指标ROC曲线下面积为0.95(P<0.001)。在本组患者中，有125例患者FNAC与组织病理学均确诊为乳腺恶性肿瘤。而在125例乳腺癌患者中，2种方法进行分型诊断的一致性检验显示，Kappa值为0.68(P<0.001)。

FNAC可以作为一种很好的乳腺肿块良恶性筛查方法，在诊断为乳腺癌的患者中也有较好的术前分型价值，可以尝试进行临床分型，为临床的治疗提供依据。

To investigate the accuracy of fine needle aspiration cytology (FNAC) in breast mass screening and breast cancer classification.

Based on the inclusion and exclusion criteria, 279 patients with breast mass undergoing FNAC and histopathological examination in the Department of Pathology, the First Affiliated Hospital of Tsinghua University from February 2015 to February 2017 were enrolled for a retrospective study. The results in FNAC and histopathological examination were compared . The Kappa value was used to evaluate the consistency of FNAC with histopathological examination in breast mass screening and classification. The area under the receiver operating characteristic (ROC) curve was used to evaluate the diagnostic accuracy of FNAC.

In 279 patients, the accuracy of FNAC was 90.3%(127/132) in breast lesions. For breast lesions, the diagnostic sensitivity of FNAC was 85.0% (125/147), specificity 96.2% (127/132), false positive rate 3.8% (5/132), false negative rate 15.0% (22/147), positive likelihood ratio 22.4, negative likelihood ratio 0.2. The consistency test of FNAC and histopathology showed that the Kappa value was 0.81 (P<0.001). The area under ROC curve was 0.95 (P<0.001) for FNAC diagnosis. Totally 125 patients were diagnosed as malignant breast tumor by FNAC and histopathology. In those 125 breast cancer patients, the consistency test of the two methods for classification showed that the Kappa value was 0.68 (P<0.001).

FNAC can be used as a good screening method for breast mass and it can provide guidance for preoperative classification of breast cancer and clinical treatment.