一种基于ACR TI-RADS(2017)评分与分级的实用CEUS TI-RADS用于甲状腺结节
本研究回顾性纳入3家医院2020年10月至2024年6月接受常规超声、CEUS及FNA的1275例患者共1521个甲状腺结节,在不改变ACR TI-RADS(2017)常规超声初始评分的前提下,筛选出动脉期强度、消退速度、环状增强和增强方向4个独立恶性危险因素构建CEUS TI-RADS。CEUS TI-RADS在训练集、内部验证集和外部验证集的AUC分别为0.94、0.94、0.92,均高于ACR TI-RADS的0.92、0.91、0.80(P均<0.001),训练集不必要活检率CEUS TI-RADS为43%(81/190)低于ACR TI-RADS的59%(156/264)。
要点
- 动脉期强度、消退速度、环状增强、增强方向为恶性独立危险因素(P均<0.001)。
- CEUS TI-RADS在训练集、内部及外部验证集AUC均高于ACR TI-RADS。
- 训练集不必要活检率CEUS TI-RADS 43%(81/190)低于ACR TI-RADS 59%(156/264)。
核心结论
- 基于ACR TI-RADS(2017)且不改变常规超声初始评分体系的CEUS TI-RADS简单有效,可用于甲状腺结节良恶性鉴别,较ACR TI-RADS降低不必要FNA率。
研究设计与样本
- 回顾性研究,纳入3家医院2020年10月至2024年6月接受常规超声、CEUS和FNA的1275例患者共1521个甲状腺结节,采用单变量和多变量logistic回归筛选CEUS特征,以FNA为参考标准构建并验证CEUS TI-RADS。
关键影像表现/征象
- 动脉期强度、消退速度、环状增强和增强方向为恶性结节的独立危险因素(P均<0.001)。
- CEUS TI-RADS在训练集、内部验证集和外部验证集的AUC分别为0.94、0.94、0.92,ACR TI-RADS分别为0.92、0.91、0.80(P均<0.001)。
- 训练集不必要活检率CEUS TI-RADS为43%(81/190个结节),ACR TI-RADS为59%(156/264个结节)。
诊断标准与关键数值
- CEUS TI-RADS在ACR TI-RADS(2017)常规超声初始评分基础上纳入动脉期强度、消退速度、环状增强和增强方向4个CEUS特征进行评分与分级。
- CEUS TI-RADS在训练集、内部验证集和外部验证集的AUC分别为0.94、0.94、0.92,ACR TI-RADS分别为0.92、0.91、0.80(P均<0.001)。
- 训练集不必要活检率CEUS TI-RADS为43%(81/190个结节),ACR TI-RADS为59%(156/264个结节)。
临床应用与阅片要点
- 可在不改变ACR TI-RADS(2017)常规超声初始评分的前提下,参考动脉期强度、消退速度、环状增强和增强方向4个CEUS特征辅助鉴别甲状腺结节良恶性。
- CEUS TI-RADS较ACR TI-RADS可降低不必要FNA率,训练集为43%(81/190)对59%(156/264)。
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
OBJECTIVE: To construct a contrast-enhanced ultrasound thyroid imaging reporting and data system (CEUS TI-RADS) based on ACR TI-RADS (2017) without changing the initial scores of conventional ultrasound features and reduce unnecessary fine-needle aspiration (FNA) of thyroid nodules.
METHODS: This retrospective study involved patients with thyroid nodules who underwent conventional ultrasound, CEUS, and FNA from October 2020 to June 2024 in three hospitals. Univariable and multivariable logistic regression analysis was employed to filter out useful CEUS features. A CEUS TI-RADS was constructed to differentiate thyroid nodules between benign and malignant and indicate the necessity of FNA.
RESULTS: A total of 1275 patients with 1521 thyroid nodules were recruited in this study. Arterial phase intensity, wash-out speed, ring enhancement, and enhancement direction (P-value all<0.001) were independent risk factors for malignant nodules in the training cohort. CEUS TI-RADS outperformed in diagnostic efficacy compared to ACR TI-RADS (AUC[CEUS] =0.94, 0.94, 0.92 vs AUC[ACR]=0.92, 0.91, 0.80 in training, internal validation and external validation cohort respectively, P-value all<0.001). CEUS TI-RADS could achieve a lower unnecessary biopsy rate (UBR) of 43% (81 of 190 nodules) in the training cohort, compared with 59% (156 of 264 nodules) of UBR in ACR TI-RADS. Similar results were confirmed concerning the validation cohorts.
CONCLUSION: The CEUS TI-RADS based on ACR TI-RADS (2017) without changing the initial scoring system of conventional ultrasound, was simple and effective in the differential diagnosis of thyroid nodules, which could reduce the unnecessary FNA rate of thyroid nodules than ACR TI-RADS.
引用
- Xu S, Liang S, Su H, et al. A Practical CEUS TI-RADS Refer to Score and Grading of ACR TI-RADS (2017) for Thyroid Nodules. Ultrasound Med Biol. 2025. PMID: 40414781. doi:10.1016/j.ultrasmedbio.2025.05.005
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。