文献速递甲状腺与浅表头颈技术与安全US

超分辨率超声成像鉴别木乃伊化甲状腺结节与乳头状癌

Super-resolution Ultrasound Imaging for Differentiating Mummified Thyroid Nodules from Papillary Thyroid Carcinomas.
PubMed 文献 · Acad Radiol · 2026-08-26

本研究纳入63个木乃伊化甲状腺结节和66个甲状腺乳头状癌,均行常规超声、CEUS及超分辨率超声成像。SRUS定量分析微血管参数,结果显示乳头状癌多项微血管参数高于木乃伊化结节,而MFV及Std Vel较低。联合Roi1%、Std Vel和血管计数的诊断模型AUC为0.919。

要点

  • SRUS可显示甲状腺结节内微血管构筑,补充常规超声及CEUS信息。
  • Roi1%<59.3、Std Vel>3.97、血管计数<16.5提示木乃伊化结节可能。
  • 联合三项SRUS参数的模型鉴别木乃伊化结节与乳头状癌AUC达0.919。

核心结论

  • 定量SRUS可有效鉴别木乃伊化甲状腺结节与甲状腺乳头状癌。
  • 基于微血管特征的多因素诊断模型对识别木乃伊化结节显示出良好诊断效能。

研究设计与样本

  • 研究类型为回顾性对照研究,共纳入63个木乃伊化甲状腺结节和66个甲状腺乳头状癌。
  • 所有结节均接受常规超声、CEUS及SRUS检查,评估大小、形态、边缘、回声、钙化、双黑白晕征及CEUS增强模式。
  • 参考标准未在摘要中说明,样本来自单中心或中心数量未在摘要中说明。

关键影像表现/征象

  • 甲状腺乳头状癌在SRUS多项微血管参数上高于木乃伊化结节,反映更丰富的微血管构筑。
  • 木乃伊化结节表现为Roi1%<59.3、Std Vel>3.97、血管计数<16.5的微血管特征组合。

诊断标准与关键数值

  • 多因素logistic回归显示Roi1%<59.3、Std Vel>3.97、血管计数<16.5为木乃伊化结节的独立预测因子。
  • 联合三项参数的诊断模型鉴别木乃伊化结节与乳头状癌的AUC为0.919(Z=0.871,0.966,p<0.001)。
  • 甲状腺乳头状癌中MFV及MFV标准差(Std Vel)低于木乃伊化结节(p<0.05)。

临床应用与阅片要点

  • 对常规超声及CEUS难以定性的甲状腺结节,可应用SRUS定量微血管参数辅助鉴别。
  • 阅片时可关注Roi1%、Std Vel及血管计数三项指标,符合阈值组合时倾向木乃伊化结节。
  • SRUS提供的微血管定量信息有助于减少不必要的细针穿刺或手术。

局限性

  • 摘要未报告研究设计为前瞻性或回顾性、样本来源中心数及参考标准细节。
  • 样本量相对有限,仅纳入63个木乃伊化结节和66个乳头状癌。

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

RATIONALE AND OBJECTIVES: Super-resolution ultrasound imaging (SRUS) overcomes the acoustic diffraction limit, enabling improved visualization of the microvascular architecture. This study aimed to evaluate its diagnostic performance in distinguishing mummified thyroid nodules (MTNs) from papillary thyroid carcinomas (PTCs), with the goal of enhancing diagnostic accuracy for MTNs and reducing unnecessary fine-needle aspirations or surgeries.

MATERIALS AND METHODS: A total of 63 MTNs and 66 PTCs were enrolled as the MTNs and PTCs groups, respectively. All nodules were examined by conventional ultrasound (US) and contrast-enhanced ultrasound (CEUS) to assess features including size, morphology, margins, echogenicity, calcification, double black-and-white halo sign, and CEUS enhancement patterns. SRUS was performed to quantify microvascular parameters within the nodule area, encompassing microvessel density, density fit, microvessel flow velocity (MFV), vascular tortuosity and complexity, and perfusion indices. Logistic regression analysis was employed to identify independent predictors for differentiating MTNs from PTCs.

RESULTS: Quantitative SRUS analysis revealed that PTCs exhibited higher values in multiple parameters compared to MTNs. In contrast, parameters like MFV, standard deviation of MFV (Std Vel) were lower in PTCs (all p < 0.05). Logistic regression identified Roi1% <59.3, Std Vel >3.97, and vessel count <16.5 as independent predictors for MTNs. The combined diagnostic model integrating these three parameters demonstrated excellent performance, with an area under the curve (AUC) of 0.919 (Z = 0.871, 0.966, p < 0.001).

CONCLUSION: Quantitative SRUS enables effective differentiation between MTNs and PTCs. The multifactorial diagnostic model based on distinct microvascular characteristics demonstrates promising diagnostic performance for identifying MTNs.

来源:Acad Radiol,查看原文。版权归原作者所有。

引用

  1. Luo, Chen, Niu, et al. Super-resolution Ultrasound Imaging for Differentiating Mummified Thyroid Nodules from Papillary Thyroid Carcinomas. Acad Radiol. 2026. PMID: 42648927. doi:10.1016/j.acra.2026.08.032

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。