文献速递乳腺技术与安全US

乳腺病灶CEUS报告与数据系统:多中心验证

A New Contrast-Enhanced Ultrasound (CEUS) Reporting and Data System for Breast Lesions
RSS 订阅 · Ultrasound in Medicine and Biology · 2026-09-25

本研究回顾性纳入3个中心1465例患者(1624个乳腺病灶),基于常规超声与CEUS特征开发并验证CEUS BI-RADS。内部验证中CEUS BI-RADS的AUC为0.95,显著高于常规US BI-RADS的0.80(p<0.001),且不必要的活检率从40%降至29%(p=0.037)。

要点

  • CEUS BI-RADS在内部验证中AUC达0.95,显著优于常规US BI-RADS的0.80。
  • CEUS BI-RADS可将不必要活检率从40%降至29%,利于减少患者焦虑与资源消耗。
  • CEUS定性特征结合常规US征象进行评分,可作为乳腺病灶恶性风险分层的实用工具。

核心结论

  • 基于常规超声与CEUS特征构建的新型CEUS BI-RADS具有良好的诊断效能,优于现有BI-RADS。
  • CEUS BI-RADS在保持较高敏感度和特异度的同时,显著降低了乳腺病灶不必要的活检率。

研究设计与样本

  • 回顾性多中心研究,共纳入1465例患者的1624个乳腺病灶,均行常规US、CEUS并获得最终诊断。
  • 中心1病例随机分为训练集(70%)和内部验证集(30%),中心2与中心3构成外部验证集。
  • 采用单变量和多变量logistic回归筛选定性超声特征,通过回归系数取整赋分构建CEUS BI-RADS。

关键影像表现/征象

  • 分析常规US与CEUS的乳腺病灶定性特征,以logistic回归筛选与恶性相关的影像征象。

诊断标准与关键数值

  • 内部验证集中CEUS BI-RADS的AUC为0.95(95% CI:0.92–0.97),常规US BI-RADS为0.80(95% CI:0.76–0.85),p<0.001。
  • CEUS BI-RADS的敏感度为90%(122/135),常规US BI-RADS为76%(103/135),p<0.05。
  • CEUS BI-RADS的特异度为80%(205/255),常规US BI-RADS为73%(187/255),p<0.05。
  • CEUS BI-RADS的阳性预测值为71%(122/172),阴性预测值为94%(205/218),均优于常规US BI-RADS。
  • CEUS BI-RADS将不必要活检率从40%(68/171)降至29%(50/172),p=0.037。

临床应用与阅片要点

  • 日常阅片中可将常规US与CEUS定性特征结合评分,利用CEUS BI-RADS进行恶性风险分层。
  • 对于常规US BI-RADS判定为需活检的病例,可参考CEUS BI-RADS再评估以降低不必要穿刺。
  • 报告时可标注CEUS BI-RADS类别及对应恶性风险,便于临床决策和患者沟通。

局限性

  • 本研究为回顾性多中心设计,原文采用的定性特征评分系统在各中心间的一致性需要进一步验证。
英文原文(摘要)· 点击展开

OBJECTIVE: To develop and validate a breast imaging reporting and data system (BI-RADS) based on conventional ultrasonography (US) and contrast-enhanced US (CEUS) to stratify the malignancy risk of breast lesions and reduce unnecessary punctures.

MATERIALS AND METHODS: This retrospective multicenter study evaluated 1465 patients with 1624 breast lesions who underwent conventional US, CEUS and received a final diagnosis from three centers. Patients from Center 1 were randomized into a training set (70%) and internal validation set (30%). Patients from Center 2 and Center 3 formed external validation sets. Qualitative US features of the breast lesions were analyzed with univariable and multivariable logistic regression. By rounding the regression coefficients of the multivariable logistic analysis to obtain scores, we developed the CEUS BI-RADS. The performance of CEUS BI-RADS in terms of discrimination, calibration and clinical utility was evaluated by receiver operating characteristic curves, calibration curves and decision curve analysis, respectively.

RESULTS: In the internal validation set, the area under the receiver operating characteristic curve value of CEUS BI-RADS was significantly higher than that of conventional US BI-RADS (0.95 [95% confidence interval (CI): 0.92, 0.97] vs. 0.80 [95% CI: 0.76, 0.85], p < 0.001). Moreover, CEUS BI-RADS yielded a considerably better sensitivity (90% [122/135] vs. 76% [103/135]), specificity (80% [205/255] vs. 73% [187/255]), positive predictive value (71% [122/172] vs. 60% [103/171]) and negative predictive value (94% [205/218] vs. 85% [187/219]) than that of conventional US BI-RADS (p < 0.05 for all). Significantly also, the unnecessary biopsy rate declined from 40% (68/171) to 29% (50/172) (p = 0.037).

CONCLUSION: A new, practical BI-RADS based on the features of conventional US and CEUS was developed, which had better diagnostic value than existing BI-RADS.

CLINICAL RELEVANCE STATEMENT: The CEUS BI-RADS, which combined the features of conventional US and CEUS, significantly reduced the rate of unnecessary biopsy of breast lesions while maintaining relatively high sensitivity and specificity, showing its potential in alleviating patients' anxiety, reducing the burden on medical resources and improving diagnostic efficiency.

来源:Ultrasound in Medicine and Biology,查看原文。版权归原作者所有。

引用

  1. Ultrasound in Medicine and Biology:A New Contrast-Enhanced Ultrasound (CEUS) Reporting and Data System for Breast Lesions
  2. Qiu, Zhang, Liu, et al. A New Contrast-Enhanced Ultrasound (CEUS) Reporting and Data System for Breast Lesions. Ultrasound Med Biol. 2026. PMID: 42791173. doi:10.1016/j.ultrasmedbio.2026.08.017

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。