Return To Abstract Listing

Breast Imaging

E2440. Shear Wave Elastography: Value in Targeted Ultrasound After Breast MRI

Au F1,  Ghai S1,  Lu F2,  Crystal P.1 1. University Health Network/Mount Sinai Hospital/Women’s College Hospital, University of Toronto, Toronto, Canada; 2. Sunnybrook Health Sciences Center, University of Toronto, Toronto, Canada

Address correspondence to F. Au (frederick.au@uhn.ca)

Objective: To determine the value of shear wave elastography (SWE) in lesion detection and identification of benign lesions for potential of avoiding biopsy, when added to targeted ultrasound (US) after breast MRI.

Materials and Methods: From July 2015 to June 2016, 24 patients who underwent targeted US assessment of MRI detected breast lesions with BI-RADS category 0, 4, or 5 were enrolled in this prospective study. MRI was performed mostly for high-risk screening and in patients with biopsy-proven breast cancer to evaluate extent of disease. Lesions demonstrated on grayscale US were correlated to MRI findings. BI-RADS category of the US correlates was recorded. SWE was performed to categorize the US lesions as soft (dark blue color on a 6-point color scale, maximum elasticity [E max] 30 kPa). If a lesion was not seen with certainty on US, SWE was performed to determine if it could be demonstrated. Biopsy was performed with US or MR guidance. Pathology findings of the biopsy or surgical excision were correlated to MRI, US, and SWE findings. The value of SWE in lesion detection and identification of benign lesions for potential of avoiding biopsy was determined.

Results: The mean age of the patients was 51 years (range, 31–74 years). There were 31 lesions detected on MRI, including 18 masses (mean size, 9 mm; range, 6–15 mm) and 13 nonmass enhancement (NME) lesions (8 focal, four linear, and one segmental) (mean size, 21 mm; range, 10–47 mm). There were 11 (36%) cancers, two (6%) high-risk lesions, 16 (52%) benign lesions and two (6%) lesions not visualized at MR-guided biopsy, with 6-month MRI follow-up arranged. US correlation was demonstrated for 22 (71%) of lesions detected by MRI, including 14 masses (mean size, 9 mm) and eight NME lesions (four focal, three linear, and one segmental) (mean size, 23 mm), with 10 cancers and 12 benign lesions. There were five lesions detected by US in BI-RADS category 3, three in category 4a, 10 in category 4b, and four in category 4c. All category 3 and 4a lesions were benign; all were soft except one category 3 lesion. Eight category 4b lesions were stiff; seven of these were malignant. Two category 4b lesions were soft and benign. Three category 4c lesions were stiff and malignant. One category 4c lesion was soft and benign. SWE helped in detection of one category 4b lesion largely obscured by recent postsurgical changes on grayscale US (turquoise rim, E max 90.1 kPa); biopsy revealed invasive cancer, which was 1 of 10 or 10% of cancers detected by US.

Conclusion: SWE was a valuable component of targeted US for assessment of breast lesions detected by MRI as it helped in lesion detection. Also, because all cancers demonstrated moderately or highly suspicious morphology on US and were stiff, whereas all soft BI-RADS category 3 and 4a US lesions were benign, there is potential of avoiding biopsy of these soft lesions, reducing rate of benign biopsies.