文献速递乳腺US

致密乳腺筛查超声检出的乳晕后导管内病变:恶性率及预测性影像特征

Retroareolar Intraductal Lesions Detected by Screening Ultrasound in Dense Breasts: Malignancy Rate and Predictive Imaging Features.
PubMed 文献 · Korean J Radiol · 2026-01-01

这项多中心回顾性研究纳入544名致密乳腺无症状女性,共550个乳晕后导管内病变,恶性率为3.8%。病变超出导管形态(调整OR 11.4)和彩色多普勒血流信号(调整OR 10.0)是恶性独立预测因子。超出导管形态特异度高达99.1%,血流信号敏感度95.2%。

要点

  • 乳晕后导管内病变恶性率低(3.8%),但需警惕超出导管形态和彩色多普勒血流信号。
  • 超出导管形态特异性高(99.1%),血流信号敏感性高(95.2%),两者联合判读可提高诊断准确性。
  • 超声特征阅片者间一致性为中等至优秀,应统一观察标准。

核心结论

  • 乳晕后导管内病变在致密乳腺筛查超声中恶性率仅为3.8%,总体预后良好。
  • 超出导管形态(调整OR 11.4)和彩色多普勒血流信号(调整OR 10.0)是独立预测恶性的影像特征。
  • 超出导管形态具有高特异度(99.1%)但敏感度低(14.3%),血流信号敏感度高(95.2%)但特异度低(41.2%)。

研究设计与样本

  • 多中心回顾性研究,纳入2012年1月至2021年12月筛查超声检出的乳晕后导管内病变的致密乳腺无症状女性。
  • 共550个病变,来自544名女性(平均年龄49.8±9.3岁)。
  • 三名放射科医生评估超声和钼靶特征,进行逻辑回归分析确定恶性相关因素。
  • 进行阅片者研究评估超声特征的一致性。
  • 参考标准为病理结果,恶性定义为手术病理证实为恶性。

关键影像表现/征象

  • 病变形态:导管内低回声或等回声病变,位于乳晕后方区域。
  • 超出导管形态:病变延伸超出导管壁,可呈不规则或分叶状。
  • 彩色多普勒超声显示血流信号:病变内部可见彩色血流。
  • 超声特征阅片者间一致性从中等至优秀(Fleiss' kappa 95%CI下限0.39-0.89)。

诊断标准与关键数值

  • 恶性率:3.8%(21/550)。
  • 超出导管形态预测恶性:敏感度14.3%(3/21),特异度99.1%(524/529)。
  • 彩色多普勒血流信号预测恶性:敏感度95.2%(20/21),特异度41.2%(218/529)。
  • 多变量分析:超出导管形态调整OR=11.4(95%CI: 2.4-53.3,P=0.002);血流调整OR=10.0(95%CI: 1.3-76.7,P=0.027)。

临床应用与阅片要点

  • 对乳晕后导管内病变,应仔细评估是否有超出导管形态和彩色多普勒血流信号,以提高恶性病变检出。
  • 血流信号敏感性高但特异度低,不宜单独作为活检依据,以免不必要的活检。
  • 超出导管形态特异度极高,一旦出现应高度怀疑恶性并建议活检。
  • 恶性率低,但需结合临床和其他影像特征综合判断,避免漏诊。
  • 注意阅片者间一致性中等,建议统一超声特征定义和评估标准。

局限性

  • 回顾性设计,可能存在选择偏倚。
  • 仅纳入致密乳腺女性,结果可能不适用于非致密乳腺人群。
  • 病变均来自筛查超声,未包括其他乳腺密度或症状性患者。
  • 阅片者研究仅评估了超声特征,未评估钼靶特征的一致性。

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

OBJECTIVE: The clinical significance of intraductal lesions detected as primary findings on screening breast ultrasound (US) remains unclear. This study aimed to determine the malignancy rate of retroareolar intraductal lesions detected by supplemental screening breast US in asymptomatic women with mammographically dense breasts and to identify imaging features associated with malignancy.

MATERIALS AND METHODS: This multicenter retrospective study included asymptomatic women with dense breasts who had intraductal lesions located in the retroareolar area detected on screening breast US between January 2012 and December 2021. Sonographic and mammographic features were evaluated by three radiologists. Logistic regression analysis was conducted to identify imaging and clinicopathologic factors associated with malignancy. The diagnostic performance of imaging features for predicting malignant retroareolar intraductal lesions was assessed. A reader study was performed to evaluate interreader agreement for sonographic features.

RESULTS: Among 550 intraductal lesions in 544 women (mean age ± standard deviation, 49.8 ± 9.3 years), 529 (96.2%) were benign, and 21 (3.8%) were malignant. Multivariable analysis demonstrated that a beyond-the-duct appearance (adjusted odds ratio [OR], 11.4; 95% confidence interval [CI], 2.4-53.3; P = 0.002) and the presence of vascularity on color Doppler US (adjusted OR, 10.0; 95% CI, 1.3-76.7; P = 0.027) were independently associated with malignancy. The beyond-the-duct appearance on US demonstrated a sensitivity of 14.3% (3 of 21) and a high specificity of 99.1% (524 of 529). The presence of vascularity on color Doppler US demonstrated a sensitivity of 95.2% (20 of 21) and a low specificity of 41.2% (218 of 529). Interreader agreement for sonographic features ranged from marginal to excellent (lower bounds of the 95% CI for Fleiss' kappa, 0.39-0.89).

CONCLUSION: Retroareolar intraductal lesions detected on screening US in women with dense breasts exhibited a low malignancy rate of 3.8%. A beyond-the-duct appearance and the presence of vascularity on color Doppler US were independently associated with malignancy in retroareolar intraductal lesions detected on screening breast US.

来源:Korean J Radiol,查看原文。版权归原作者所有。

引用

  1. Ha, Choi, Yoen, et al. Retroareolar Intraductal Lesions Detected by Screening Ultrasound in Dense Breasts: Malignancy Rate and Predictive Imaging Features. Korean J Radiol. 2026. PMID: 42543755. doi:10.3348/kjr.2026.0215

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。