Diagnostic Value of Shear Wave Sonoelastography in Differentiation of Benign From Malignant Thyroid Nodules Publisher Pubmed



Aghaghazvini L¹ ; Maheronnaghsh R¹ ; Soltani A² ; Rouzrokh P² ; Chavoshi M¹ Show Affiliations
Source: European Journal of Radiology Published:2020

Abstract

Purpose: To study the efficacy of shear wave elastography (SWE), using both qualitative and quantitative methods, alone and in conjunction with other B-mode features. Method: 117 patients with 123 nodules were studied both by conventional ultrasonography and SWE. Size, echogenicity, margins, internal calcification (micro- or macro-calcification), composition, shape and color Doppler were assessed for each nodule. The elasticity was assessed both qualitatively and quantitatively. Velocity in the ROI (Region of Interest) was calculated in the stiffest portions for 3 times, and maximum and mean velocity were obtained. ROC curve was analyzed to calculate the best cut-off value of the SWV (Shear Wave Velocity). Univariate logistic regression was used to examine the maximum and mean SWV as discrete variables and the results were compared to key variables of conventional US (Ultrasound) features. Result: 123 nodules in 117 patients were evaluated. Poor margins, hypoechogenicity, micro-calcification, color Doppler grades III and IV, color map grades IV and V, maximum and minimum velocity had significant correlation with malignancy. The highest Nagelkerke R2 belonged to maximum and mean velocity (R2 = 41.2 and 39 respectively) which propose them as the strongest predictors of malignancy. The best cut-off point for differentiation of benign from malignant nodules was 3.63 m/s for maximum velocity (sensitivity of 90 %, specificity of 78.2 %,) and 3.44 m/s for mean velocity (sensitivity of 90 %, specificity of 76.4 %). Conclusion: The Real-time SWE is a promising test for the preoperative malignancy risk stratification of patients and maximum velocity has the strongest predictive value for both conventional and elastography variables. © 2020 Elsevier B.V.