自动乳腺容积成像联合声脉冲辐射力成像在可疑乳腺肿块鉴别诊断中的应用价值

Application value of automated breast volume scanning combined with acoustic radiation force impulse in differential diagnosis of suspicious breast mass

目的:探讨自动乳腺容积成像(automated breast volume scanning,ABVS)和声脉冲辐射力成像(acoustic radiation force impulse,ARFI)两种超声技术联合对常规超声诊断为可疑乳腺肿块的良恶性鉴别诊断能力。方法:对118例乳腺肿块常规超声评估分类为乳腺影像报告和数据系统(Breast Imaging Reporting And Data System,BI-RADS)4类及以上的患者,应用ABVS分析肿块有无冠状面图像特征(汇聚征与微钙化),应用ARFI检测声触诊组织成像(virtual touch tissue imaging,VTI)中肿块在弹性图与二维图的面积比和声触诊组织量化成像(virtual touch tissue imaging quantification,VTIQ)中肿块剪切波速度的平均值(V_(mean)),对乳腺肿块良恶性的图像按照冠状面有无汇聚征、有无微钙化分类,有任意一种即记为阳性;VTI肿块面积比值大于1.65记为阳性;V_(mean)大于4.14 m/s为阳性。计算并比较ABVS、ARFI单独以及两者联合诊断BI-RADS 4类乳腺肿块的灵敏度、特异度、准确度、阳性预测值、阴性预测值。结果:乳腺肿块在ARFI的检查中,恶性肿块的VTI面积比(1.76±0.61)高于良性肿块VTI面积比(1.14±0.48),恶性肿块的V_(mean)[(6.38±1.76)m/s]高于良性肿块[(4.17±1.43)m/s],差异均有统计学意义(P<0.01)。在BI-RADS 4类乳腺肿块中,ABVS检查中汇聚征或微钙化对恶性乳腺肿块检出的灵敏度、特异度、准确度、阳性预测值及阴性预测值分别为92%、78.3%、84.5%、78.0%及92.2%;VTI诊断乳腺恶性肿块的灵敏度、特异度、准确度、阳性预测值及阴性预测值分别为70.0%、81.7%、76.4%、76.1%及76.6%;VTIQ诊断恶性乳腺肿块的灵敏度、特异度、准确度、阳性预测值、阴性预测值分别为96.0%、58.3%、75.5%、65.8%及94.6%。ABVS与ARFI两者联合诊断的灵敏度、特异度、准确度、阳性预测值、阴性预测值分别为100.0%、50.0%、72.7%、62.5%及100.0%。结论:ABVS与ARFI可补充常规二维超声无法获得的乳腺肿块信息,丰富肿块诊断的依据,扩展疾病鉴别诊断的思路,两者结合可达到优势互补的效果,有助于提高诊断的准确度。

Objective:To investigate the differential diagnosis ability of automated breast volume scanning(ABVS)and acoustic radiation force impulse(ARFI)in the diagnosis of suspicious breast mass by conventional ultrasound.Methods:A total of 118 patients with breast masses classified as Breast Imaging Reporting And Data System(BI-RADS)class 4 or above by conventional ultrasound evaluation were analyzed using ABVS and ARFI.ABVS was to determine whether the masses had coronal imaging features(convergence sign and microcalcification),then ARFI was to examine the area ratio of masses in elastic and two-dimensional images in virtual touch tissue imaging(VTI),as well as the average shear wave velocity of masses(V_(mean))in virtual touch tissue imaging quantification(VTIQ).The benign and malignant images of breast masses were classified according to whether there was convergence sign or microcalcification in the coronal plane,and any of the above or more were recorded as positive.The area ratio of tumor in VTI greater than 1.65 was considered positive.The V_(mean) greater than 4.14 m/s was positive.The sensitivity,specificity,accuracy, positive predictive value and negative predictive value of ABVS and ARFI alone or combined in the diagnosis of BI-RADS4 breast masses were calculated and compared. Results: In the examination of breast mass in ARFI, mass area ratio of malignant tumorin VTI (1.76±0.61) was higher than that of benign tumor (1.14±0.48), and V_(mean) of malignant tumor [(6.38±1.76) m/s] was higher than that ofbenign tumor [(4.17±1.43) m/s], the differences were statistically significant (P<0.01). There were significant differences in VTIand VTIQ between the two groups (P<0.01). In BI-RADS 4 breast masses, the sensitivity, specificity, accuracy, positive predictivevalue, and negative predictive value were 92%, 78.3%, 84.5%, 78.0%, and 92.2%, respectively for the detection of malignant breastmass in ABVS. The sensitivity, specificity, accuracy, positive predictive value, and negative predictive value of VTI in diagnosis ofmalignant breast masses were 70.0%, 81.7%, 76.4%, 76.1%, and 76.6%, respectively. The sensitivity, specificity, accuracy, positivepredictive value, and negative predictive value of VTIQ in diagnosis of malignant breast masses were 96.0%, 58.3%, 75.5%,65.8%, and 94.6%, respectively. The sensitivity, specificity, accuracy, positive predictive value, and negative predictive value of thecombined diagnosis of ABVS and ARFI were 100.0%, 50.0%, 72.7%, 62.5%, and 100.0%, respectively. Conclusion: ABVS andARFI can supplement the information of breast mass that cannot be obtained by conventional two-dimensional ultrasound, enrich thebasis of diagnosis of breast mass, and expand the idea of differential diagnosis of disease. The combination of the two can achieve theeffect of complementary advantages, and help to improve the accuracy of diagnosis.