超声弹性成像对乳腺乳头状病变的诊断能提供什么帮助
纳入本院89个乳头状病变(良性74个、恶性15个)及198个非乳头状良性肿瘤作对照,评估应变弹性成像(SE)在鉴别乳头状病变良恶性及与非乳头状良性病变鉴别中的价值。鉴别乳头状病变良恶性时,SE未能提高BI-RADS诊断效能;但BI-RADS联合SE鉴别乳头状病变与非乳头状良性病变时,特异度和准确率显著高于单用BI-RADS且敏感度未下降。
要点
- 鉴别乳头状病变良恶性时,BI-RADS联合SE的AUC为0.509,与单用BI-RADS的0.568比较无统计学差异(p>0.05)。
- 鉴别乳头状病变与非乳头状良性病变时,BI-RADS联合SE的AUC为0.835,高于单用BI-RADS的0.775和SE的0.648/0.661(p<0.001)。
- BI-RADS联合SE鉴别乳头状病变与非乳头状良性病变的特异度和准确率显著高于单用BI-RADS,且敏感度未下降(p<0.05)。
核心结论
- SE不能提高BI-RADS鉴别乳头状病变良恶性的诊断效能;但BI-RADS联合SE可在不降低敏感度的情况下提高BI-RADS鉴别乳头状病变与非乳头状良性病变的特异度。
研究设计与样本
- 单中心研究,纳入本院89个乳头状病变(良性74个、恶性15个)及198个非乳头状良性肿瘤作对照(纤维腺瘤126个、纤维囊性乳腺病72例),所有患者均行常规超声和SE检查。
关键影像表现/征象
- 既往研究中SE假阳性诊断最常见的组织学类型为乳头状瘤。
诊断标准与关键数值
- 鉴别乳头状病变良恶性:BI-RADS的AUC为0.568,SE评分、应变比及BI-RADS联合SE的AUC分别为0.517、0.584和0.509(p>0.05)。
- 鉴别乳头状病变与非乳头状良性病变:BI-RADS联合SE的AUC为0.835,高于BI-RADS的0.775、SE评分的0.648和应变比的0.661(p<0.001)。
- 鉴别乳头状病变与非乳头状良性病变:BI-RADS联合SE的特异度和准确率显著高于单用BI-RADS,敏感度未下降(p<0.05)。
临床应用与阅片要点
- 作者指出,SE不能改善BI-RADS鉴别乳头状病变良恶性的诊断效能。
- 作者指出,BI-RADS联合SE可提高BI-RADS鉴别乳腺乳头状病变与非乳头状良性病变的特异度,且不降低敏感度。
局限性
- 原文未报告局限性。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
英文原文(摘要)· 点击展开
On the basis of results of our previous studies and the findings of other scholars, the most common histologic type of false-positive diagnosis with strain elastography (SE) was papilloma. The objectives of our study were to evaluate whether SE could contribute to conventional ultrasound differentiation between benign and malignant papillary lesions and between papillary lesions and other common benign breast lesions. Data on 89 papillary lesions at our hospital, including 74 benign and 15 malignant papillary lesions, were included in our study. In addition, 198 non-papillary benign tumors were selected as the control group, including 126 fibroadenomas and 72 cases of fibrocystic mastopathy. All patients gave written informed consent. All patients with breast lesions underwent conventional ultrasound and SE examination. Breast Imaging Recording and Data System (BI-RADS) category and SE score were compared with respect to sensitivity, specificity and accuracy in differentiating between benign and malignant papillary lesions. We then explored the possibility of using BI-RADS combined with SE to differentiate papillary lesions from non-papillary benign tumors. For differentiating between benign and malignant papillary lesions, the area under the receiver operating characteristic curve (AUC) of BI-RADS was 0.568, whereas the AUC values of SE score, strain ratio and BI-RADS combined with SE were 0.517, 0.584 and 0.509, respectively (p > 0.05). For differentiating between papillary lesions and non-papillary benign lesions, the AUC of BI-RADS combined with SE was 0.835, which was higher than the values for BI-RADS (0.775) and SE (SE score: 0.648, strain ratio: 0.661) (p < 0.001). The specificity and accuracy of BI-RADS combined with SE were significantly higher than those for BI-RADS alone without a decrease in sensitivity (p < 0.05). SE could not improve the diagnostic efficiency of BI-RADS in differentiating between benign and malignant papillary lesions. However, BI-RADS combined with SE could improve the specificity of BI-RADS without decrease in sensitivity for differentiating breast papillary lesions from non-papillary benign lesions.
引用
- Li LJ, Yao JY, Zhou XC, et al. What Help Could Ultrasound Elastography Give to the Diagnosis of Breast Papillary Lesions? Ultrasound Med Biol. 2017. PMID: 28256344. doi:10.1016/j.ultrasmedbio.2017.01.004
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。