• Yong-Mei Sun, Hai Dong, Zong-Yan Du, Zong-Li Yang, Cheng Zhao, Jing Chong, and Ping Li.
• Department of Abdominal Ultrasound, the Affiliated Hospital of Qingdao University, No.16 Jiang Su Rd, Qingdao 266000, Shandong, China.
• Clinics (Sao Paulo). 2020 Jan 1; 75: e1691.

ObjectiveImaging diagnosis of cervical lymphadenopathy has conventionally used ultrasonography. Shear wave elastography (SWE) is a recent ultrasound technological advancement that has shown promise in the important medical problem of differentiating between benign and malignant cervical lymph nodes based on quantitative measurements of elasticity modulus. However, widely varying elasticity modulus metrics and regions-of-interest (ROIs) were used in existing studies, leading to inconsistent findings and results that are hard to compare with each other.MethodsUsing a large dataset of 264 cervical lymph nodes from 200 patients, we designed a study comparing three elasticity modulus metrics (Emax, Emean, and standard deviation-SD) with three different ROIs to evaluate the effect of such selections. The metric values were compared between the benign and malignant node groups. The different ROI and metric selections were also compared through receiver operating characteristics curve analysis.ResultsFor all ROIs, all metric values were significantly different between the two groups, indicting their diagnostic potential. This was confirmed by the ≥0.80 area under the curve (AUC) values achieved with these metrics. Different ROIs had no effect on Emax, whereas all ROIs achieved high performance at 0.88 AUC. For Emean, the smallest ROI focusing on the area of the highest elasticity achieved the best diagnostic performance. In contrast, the larger ROIs achieved higher performances for SD.ConclusionsThis study illustrated the effect of elasticity modulus and ROI selection on the diagnostic performance of SWE on cervical lymphadenopathy. These new findings help guide relevant future studies and clinical applications of this important quantitative imaging modality.

13 keywords...

hide…