Stoian Dana, Ivan Viviana, Sporea Ioan, Florian Varcus, Mozos Ioana, Navolan Dan, Nemescu Dragos
2nd Department of Internal Medicine, "Victor Babes" University of Medicine, Timisoara, Romania.
Dr. D Center for Ultrasound in Endocrinology, Timisoara, Romania.
Ther Clin Risk Manag. 2020 Jan 21;16:21-30. doi: 10.2147/TCRM.S224060. eCollection 2020.
Current major guidelines recommend risk stratification of the thyroid nodules, after each diagnostic evaluation, in order to focus attention on potentially risky nodules. The main aim of our study was to evaluate the performance of combined advanced ultrasound techniques in this process, compared with conventional stratification models, in order to reduce unnecessary fine-needle biopsies, respectively, surgery.
We evaluated 261 cases (261 nodules) using conventional ultrasound (2B), real-time Doppler evaluation (4D) respectively, real-time elastography, using a linear multifrequency probe and a linear volumetric probe (Hitachi Prerius Machine, Hitachi Inc, Japan). All the nodules were classified using a risk stratification model comprising seven conventional US characteristics, two 4 D characteristics and a color map RTE aspect. The results were compared with the pathology results, considered the golden standard diagnosis.
The prevalence of malignant nodules was 21.83% (57 cases). Conventional risk classification generated: 106 low-risk cases, 113 intermediate-risk and 42 high-risk cases. Our proposed risk classification changes the conventional risk classification with a risk upgrade in 27 cases and with a risk downgrade in 69 cases. The diagnostic quality of the combined risk stratification model was better, considering a low-risk category predictive for benignancy and a high category predictive for malignancy: Sensitivity: 80.88% versus 49.01%, respectively, Specificity: 91.22% versus 54.38. The diagnostic power differences were observed regardless of the nodule size.
Advanced ultrasound techniques did add diagnostic value in the presurgical risk assessment of the thyroid nodules.
目前的主要指南建议在每次诊断评估后对甲状腺结节进行风险分层，以便将注意力集中在潜在风险结节上。我们研究的主要目的是评估联合先进超声技术在此过程中的表现，与传统分层模型相比，以分别减少不必要的细针穿刺活检和手术。
我们分别使用传统超声（2B）、实时多普勒评估（4D）以及使用线性多频探头和线性容积探头（日本日立公司的Hitachi Prerius机器）的实时弹性成像对261例病例（261个结节）进行了评估。所有结节均使用包含七个传统超声特征、两个4D特征和一个彩色图弹性成像特征的风险分层模型进行分类。将结果与被视为金标准诊断的病理结果进行比较。
恶性结节的患病率为21.83%（57例）。传统风险分类产生：106例低风险病例、113例中风险病例和42例高风险病例。我们提出的风险分类改变了传统风险分类，其中27例风险升级，69例风险降级。考虑到低风险类别预测为良性、高风险类别预测为恶性，联合风险分层模型的诊断质量更好：敏感性分别为80.88%和49.01%，特异性分别为91.22%和54.38%。无论结节大小如何，均观察到诊断能力差异。
先进超声技术在甲状腺结节术前风险评估中确实增加了诊断价值。
