【摘要】 目的探讨传统超声、超色多普勒和弹性成像技术在预测乳腺癌腋窝淋巴结转移的价值。方法选择病理证实的腋窝区56枚良性淋巴结和26枚乳腺癌转移性淋巴结进行研究,术前均经传统超声、超声多普勒和弹性成像检查。观察并记录淋巴结短轴、长轴、短长轴比、皮质厚度、血流特点和弹性评分,并且对两组数据进行。绘制上述指标的ROC曲线,观察曲线下面积、最佳阈值、敏感性和特异性;另外对淋巴结皮质厚度和硬度进行相关性分析。结果转移组淋巴结的短轴、短长轴比和皮质厚度均明显大于非转移组(P=0.00)。与非转移组淋巴结相比,转移组淋巴结主要表现为Ⅲ型和Ⅳ型的血流特征。转移组淋巴结弹性成像分数主要为3～4分,而非转移组淋巴结多为1～2分。淋巴结短轴、皮质厚度、短长轴比、血流特征及弹性评分诊断敏感性分别为84.6%、84.6%、84.6%、76.9%和84.6%,而特异性分别为73.9%、97.8%,84.8%、95.7%和97.8%;传统超声(皮质厚度>2.25 mm)和弹性评分(≥3分)具有最高的诊断敏感性和特异性。淋巴结皮质厚度和硬度具有显著性相关(r=0.899,P<0.00)。结论皮质厚度和超声弹性成像技术是诊断腋窝转移性淋巴结与非转移性淋巴结可行性指标。更多还原

【Abstract】 Objective To discuss the value of traditional ultrasound, color doppler and ultrasound elastography for diagnosing axillary lymph node(LN) metastasis in breast cancer. Methods Axillary LNs in healthy probands(n=56) and metastatic LNs in breast cancer patients(n=26) were examined by suing traditional ultrasound, color doppler and ultrasound elastography. Short radius, long axis, S/L, cortical thickness, blood flow characteristics and elastic characteristics of all LNs were detected, and then these indicators were compared between the two groups. Receiver operating characteristic(ROC)curve of all above indicators were drawed, and AUC, threshold value, sensitivity(SE) and specificity(SP) were calculated. Moreover, the relationship between cortical thickness and the hardness was studied. Results The short radius, S/L, and cortex thickness of metastatic LNs were all significantly larger than that of non-metastatic LNs(P=0.000). The blood types in metastatic LNs were almost III and IV types. UE displayed that the scores of most metastatic LNs was 3-4, but the scores of most non-metastatic LNs was 1-2(P<0.01). The SE of short radius, cortical thickness, S/L, doppler ultrasound and sonoelastography was revealed to be 84.6%,84.6%, 84.6%, 76.9% and 84.6%, and the SP was 73.9%, 97.8%, 84.8%, 95.7% and 97.8%, respectively. The highest SE and SP was achieved by the B-mode(cortex >2.25 mm) and elastographic features(3 score). The cortex thickness had a strong correlation with hardness of LN(r=0.899, P<0.00). Conclusion Cortex thickness and UE of LNs are both feasible method to identify the metastatic LNs from non-etastatic LNs.更多还原