摘要目的:探讨自动乳腺容积超声(ABUS)联合手持超声(HHUS)在1 463例乳腺良恶性病变中鉴别诊断BI-RADS-US 3~5类的临床实用价值。方法:选取BI-RADS 3类随访2年以上和BI-RADS 4类、5类经手术或粗针穿刺活检证实的1 463例乳腺病灶,术前均行HHUS和ABUS检查,获取二维影像和三维容积数据及BI-RADS分类诊断结果,以病理为金标准,分析ABUS联合HHUS评估乳腺良恶性病灶BI-RADS 3~5类的诊断效能。结果:1 463例乳腺病灶中,良性935(63.9%)例、恶性528(36.1%)例。①依照第5版BI-RADS-US分类指南,以病理为金标准,两种检查方法联合诊断与病理对照的一致性(Kappa值=0.85)高于单独HHUS(Kappa值=0.81),具有统计学差异(P<0.05)。比较两种方法评估BI-RADS 3~5类恶性风险范围:3类恶性概率下降,两者联合0(0) vs. HHUS 0.3%(2/685);4A类恶性概率下降,两者联合3.4%(5/146) vs. HHUS 13.6%(22/162);4B类恶性概率下降,两者联合57.6%(95/165) vs. HHUS 60.1%(98/163);4C类恶性概率上升,两者联合84.2%(139/165) vs. HHUS 81.4%(131/161);5类恶性概率上升,两者联合99.3%(298/291) vs. HHUS 94.2%(275/292)。②ABUS联合HHUS与单独HHUS评估乳腺良恶性病灶的诊断效能分别为:敏感性(99.05% vs. 95.45%)、特异性(89.52% vs. 88.02%)、准确性(92.96% vs. 90.70%)、阳性预测值(99.41% vs. 97.17%)、阴性预测值(84.22% vs. 81.82%)、漏诊率(0.95% vs. 4.55%)与误诊率(10.48% vs. 11.98%)及ROC曲线下面积(0.95 vs. 0.91),具有统计学差异(P<0.001)。结论:AUBS联合HHUS较单独HHUS对BI-RADS 3~5类,尤为4类病灶具有更准确地估测恶性风险的能力,有助于提高BI-RADS分类的诊断效能,降低漏误诊率,缩小诊断偏差,利于尽早检出不典型异常病灶,为鉴别乳腺良恶性结节、临床穿刺方案选择及术后评价提供更有意义的影像佐证。
Abstract:Objective: To explore the clinical value of automated breast ultrasound system(ABUS) combined with handheld ultrasonography(HHUS) in the differential diagnosis of BI-RADS-US 3~5 among 1 463 cases of benign and malignant breast lesions. Methods: 1 463 breast lesions with BI-RADS 3 followed-up for more than 2 years and BI-RADS 4 and 5 confirmed by surgery or needle biopsy were selected. HHUS and ABUS were performed before operation to obtain 2D images, 3D volume data and the BI-RADS classification. Corresponding to the pathology as the gold standard, ABUS combined with HHUS was used to evaluate the diagnostic efficacy of BI-RADS 3~5 in benign and malignant breast lesions. Results: Among 1 463 cases of breast lesions, 935(63.9%) were benign and 528(36.1%) were malignant. ①According to the ACR version 5 BI-RADS-US classification guidelines, using pathology as the gold standard, the combined diagnostic evaluation of BI-RADS classification and pathological results compared with results by two examination methods(Kappa value=0.85) was higher than that of HHUS(Kappa value=0.81), with statistical significance(P<0.05). Comparing the two methods to assess the risk range of BI-RADS 3~5 categories of malignancy: the probability of 3 type malignancy decreased while the combination of both 0(0) vs. HHUS 0.3%(2/685); 4A type of malignant probability decreased while the two combined 3.4%(5/146) vs. HHUS 13.6%(22/162); 4B malignant probability decreased while the two combined 57.6%(95/165) vs. HHUS 60.1%(98/163); the probability of 4C type of malignancy increased while the combination of the two 84.2%(139/165) vs. HHUS 81.4%(131/161); the probability of 5 type of malignancy increased while the two combined 99.3%(298/291) vs. HHUS 94.2%(275/292). ②The diagnostic efficacy of ABUS combined with HHUS or HHUS alone in the evaluation of benign and malignant breast lesions were: sensitivity(99.05% vs. 95.45%), specificity(89.52% vs. 88.02%), accuracy(92.96% vs. 90.70%). Positive predictive value(99.41% vs. 97.17%), negative predictive value(84.22% vs. 81.82%), false negative rate(0.95% vs. 4.55%) and false positive rate(10.48% vs. 11.98%), area under ROC curve(0.95 vs. 0.91, P<0.001), which is statistically different(P<0.001). Conclusion: Compared with HHUS alone, AUBS combined with HHUS has the ability to accurately estimate the risk of malignancy in BI-RADS 3~5 lesions, especially BI-RADS 4 lesions, which helps to improve the diagnostic efficiency of BI-RADS classification, reduce misdiagnosis, reduce diagnostic errors, and facilitate the early detection of atypical abnormal lesions, and provide more meaningful imaging evidence for distinguishing breast benign and malignant nodules, clinical puncture options, and postoperative evaluation.
赵 妮,高喜璨,舒 瑞,吴建峰,宋宏萍. ABUS联合HHUS对BI-RADS 3~5类乳腺病灶
恶性风险评估的应用价值[J]. 中国临床医学影像杂志, 2019, 30(7): 457-462.
ZHAO Ni, GAO Xi-can, SHU Rui, WU Jian-feng, SONG Hong-ping. Application of ABUS combined with HHUS in malignant risk assessment of BI-RADS type 3~5 breast lesions. JOURNAL OF CHINA MEDICAL IMAGING, 2019, 30(7): 457-462.
[1]杨祖凤. 乳腺超声自动容积成像技术与普通超声对乳腺肿物的诊断价值比较[J]. 影像研究与医学应用,2018,2(2):18-20.
[2]Ackerman S. ACR BI-RADS for breast ultrasound[J]. Ultrasound Med Biol, 2015, 41(4): S95.
[3]朱庆莉,姜玉新. 乳腺影像报告与数据系统指南(第5版)超声内容更新介绍[J]. 中华医学超声杂志:电子版,2016,13(1):5-7.
[4]阳练,赵玉珍,杜晶晶,等. 乳腺自动容积成像汇聚征与常规超声边界不规整对乳腺癌的诊断价值[J]. 中国超声医学杂志,2017,33(7):587-590.
[5]Chen L, Chen Y, Diao XH, et al. Comparative study of automated breast 3-D ultrasound and handheld B-mode ultrasound for differentiation of benign and malignant breast masses[J]. Ultrasound Med Biol, 2013, 39(10): 1735-1742.
[6]闵贤,庞芸,王燕,等. 超声自动乳腺全容积扫查对乳腺肿块诊断额外临床价值的研究[J]. 中国超声医学杂志,2017,33(5):412-415.
[7]Golatta M, Franz D, Harcos A, et al. Interobserver reliability of automated breast volume scanner(ABVS) interpretation and agreement of ABVS findings with handheld breast ultrasound(HHUS), mammography and pathology results[J]. Eur J Radiol, 2013, 82(8): 332-336.
[8]Zhan J, Diao XH, Pang Y, et al. Is there an extraclinical value of automated breast volume scanner compared with hand-held ultrasound? A pilot study[J]. Medicine, 2017, 96(37): e7765.
[9]谭艳娟,包凌云,黄安茜,等. 乳腺恶性肿瘤冠状面汇聚征与临床病理学相关因素分析[J]. 中国超声医学杂志,2015,31(7):587-589.
[10]Wang Z, Guo Q, Huang B, et al. The accuracy of automated breast volume scanning for the measurement of breast tumor size[J]. Chin J Med Ultrasound, 2015, 12(6): 484-487.
[11]包凌云,谭艳娟,朱罗茜,等. 自动乳腺全容积成像在乳腺导管原位癌诊断中的应用价值[J]. 中华医学超声杂志:电子版,2015,12(1):61-67.
[12]包凌云,朱罗茜,孔凡雷,等. 自动乳腺全容积成像和常规超声对乳腺微钙化诊断的对比研究[J]. 中华超声影像学杂志,2012,22(3):220-223.
[13]Huang A, Zhu L, Tan Y, et al. Evaluation of automated breast volume scanner for breast conservation surgery in ductal carcinoma in situ[J]. Oncol Lett, 2016, 12(4): 2481-2484.