Application of Sonoelastography in Differential Diagnosis of Benign and Malignant Thyroid Nodules

The current value of quantitative shear wave sonoelastography in parotid gland tumors

Background

The preoperative differentiation between salivary gland tumor entities using computed tomography, magnetic resonance imaging (MRI) and ultrasound (US) is still limited. Biopsies are often regarded as indispensable for properly characterizing these various lesions. The aim of this study was to analyze the value of acoustic radiation force impulse (ARFI) sonoelastography as an US differentiation tool when examining parotid gland (PG) lesions.

Methods

We included 104 patients with PG masses in this study, employing two different US devices using quantitative ARFI-sonoelastography (Siemens Acuson-S3000, n=59; Siemens Acuson-Sequoia, n=45). The ability of sonoelastographic measurements to differentiate between different neoplasms was compared and analyzed for both US machines.

Results

Quantitative shear wave sonoelastography is limited in its ability to reliably differentiate between tumor entities of the PG as a stand-alone parameter. Measurement results were unsystematically distributed and not transferable between the two US devices. A significant differentiation of benign and malignant lesions was not possible with either US machine (S3000: P=0.770, Sequoia: P=0.382). A differentiation between pleomorphic adenomas (PA) and Warthin tumors was only possible with the Acuson S3000 system (P=0.001, Spearman-Rho =0.492, sensitivity 73.9%, specificity 65.0%).

Conclusions

A reliable identification and differentiation of PG tumors as well as clinical treatment decisions cannot be made with the sole use of ARFI-sonoelastography. The results emphasize the device-dependence and high error-proneness of this US technique when examining lesions of the PG.

Strain Elastography as a Valuable Diagnosis Tool in Intermediate Cytology (Bethesda III) Thyroid Nodules

Fine needle aspiration (FNA) is considered the gold standard in the diagnostic of thyroid nodules. Using the recommended BETHESDA reporting system, up to 20% of results are classified as intermediate cytology. As there is no consensus whether ultrasound evaluation, lobectomy or surgery is the best treatment option, intermediate cytology results are considered a grey zone of the FNA. The main aim of our study was to evaluate the performance of combined advanced ultrasound techniques in the process of diagnosis and evaluation of the intermediate cytology cases after FNA. We evaluated 54 consecutive cases with intermediate cytology on FNA, using conventional B-mode ultrasound (2B), and strain elastography, using a linear multifrequency 6-13 MHz linear probe (Hitachi Prerius Machine, Hitachi Inc, Japan). All nodules were classified with our Thyroid Imaging Report and Data System (TI-RADS) proposed model, considering: vertical appearance, with antero-posterior diameter bigger than the transvers diameter, the so called taller than wide shape, irregular borders, intranodular inhomogeneity, marked hypoecogenicity, micro calcifications, the presence of suspect lymph nodes, and increased stiffness as suspicious for malignancy. The classification outcomes were compared with the pathology results, considered the gold standard diagnosis. The prevalence of cancer was 28.8%, with 13/45 cases having a clear diagnostic of cancer. Six cases were diagnosed with borderline follicular neoplasia, a category with unclear evolution, also considered as malignant in the analysis of the imaging results. In total, 16/19 cancer cases had increased stiffness on elastography. The cancer prevalence increased with TI-RADS category, being 25% in TI-RADS 4b category and 92.8% in TI-RADS 5 category. The AUROC (Area Under Receiver Operating Curve) of elastography alone, in differentiation of malignant thyroid nodules was 74.9%; the combination of elastographic and conventional ultrasound characteristics generated an even better AUROC, of 84.5%. The combined conventional ultrasound and elastography identified thyroid cancer in cases with intermediate cytology with a sensitivity of 89.5% with a specificity of 50%. High risk thyroid nodules, identified by combined high risk conventional ultrasound characteristics and increased stiffness, on strain elastography, are highly predictive for malignancy, in the intermediate cytology cases.

Combination of Sonoelastography and TIRADS for the Diagnostic Assessment of Thyroid Nodules

To evaluate the diagnostic performance of elastography alone and combined with Thyroid Imaging Reporting And Data System (TIRADS) for the assessment of non-autonomous thyroid nodules. We included 244 thyroid nodules and analyzed the visual elasticity scores, strain value (SV) and TIRADS classification. Histologic examination revealed 38 malignant (16%) and 206 benign nodules. The SV was lower in malignant nodules than in benign with an optimal cutoff ≤0.225. The visual elasticity scores showed a better diagnostic performance than the SV measurement. The risk for malignancy increased with higher TIRADS category. The sensitivity, specificity, positive predictive value and negative predictive value of TIRADS were superior to sonoelastography. The combination of TIRADS ≥4C and SV ≤0.225 showed the highest odds ratio to predict malignancy. Kwak-TIRADS classification is superior to elastography for the differentiation of benign and malignant thyroid nodules. Our data demonstrate that a high TIRADS class alone is predictive for thyroid carcinoma and the clinical relevance of sonoelastography is negligible.