Value of shear wave elastography during second-look breast ultrasonography for suspicious lesions on magnetic resonance imaging

• Published in: Journal of medical ultrasonics (2001) 2022-10, Vol.49 (4), p.719-730
• Main Authors: Gil, Bo Mi, Jung, Na Young, Kim, Sung Hun, Kang, Bong Joo, Lee, Jeongmin, Chung, Myung Hee
• Format: Article
• Language: English
• Subjects: Biopsy; Breast; Breast cancer; Color; Comparative analysis; Data systems; Diagnostic systems; Elasticity; Imaging; Lesions; Magnetic resonance imaging; Mammography; Medical imaging; Medicine; Medicine & Public Health; Original Article–Breast -Thyroid; Parameter sensitivity; Pathology; Patients; Radiology; Sensitivity; Ultrasonic imaging; Ultrasound; Ultrasound imaging
• ISSN: 1346-4523; 1613-2254
• DOI: 10.1007/s10396-022-01253-z

Purpose This study aimed to determine whether the addition of shear wave elastography (SWE) helps to improve diagnostic performance of second-look ultrasonography (SLUS) for suspicious lesions on magnetic resonance imaging (MRI). Methods We retrospectively reviewed 76 breast lesions in 62 patients who underwent SLUS and SWE for suspicious lesions on MRI from August 2017 to November 2019. The six-point color scale ( E col ) and four-type color pattern ( E pattern ) were used for qualitative evaluation, and the mean ( E mean ) and maximum elasticity ( E max ) and standard deviation of elasticity ( E SD ) were recorded as quantitative parameters. Clinical and imaging findings between benign and malignant lesions were compared, and the diagnostic performance was assessed using receiver-operating characteristic (ROC) analysis. Results The biopsies revealed 52 benign and 24 malignant lesions. Of all SWE parameters, only E SD was significantly higher in malignant lesions than in benign lesions ( p  = 0.012). The sensitivity of B-mode US was 100%, but the specificity was low (19.2%). Using SWE parameters to classify lesions improved specificity at the expense of sensitivity. When lesions assigned to Breast Imaging Reporting and Data System (BI-RADS) category 3 or 4a were reclassified considering each SWE parameter, the area under the ROC curve (AUC) and sensitivity increased. The AUC of the US BI-RADS category adjusted by E SD was higher than that of B-mode US BI-RADS (0.770 vs. 0.746). Conclusion SWE parameters, and especially E SD , may play a complimentary role in improving the specificity of SLUS. However, the decision to omit biopsies for suspicious lesions with soft features should be made with caution.