Abstract
OBJECTIVES: To determine if quantitative and qualitative shear wave elastography have roles in evaluating musculoskeletal masses. METHODS: 105 consecutive patients, prospectively referred for biopsy within a specialist sarcoma centre, underwent B-mode, quantitative (m/s) and qualitative (colour map) shear wave elastography. Reference was histology from subsequent biopsy or excision where possible. Statistical modelling was performed to test elastography data and/or B-mode imaging in predicting malignancy. RESULTS: Of 105 masses, 39 were malignant and 6 had no histology but benign characteristics at 12 months. Radiologist agreement for B-mode and elastography was moderate to excellent Kw 0.52-0.64; PABAKw 0.85-0.90). B-Mode imaging had 78.8% specificity, 76.9% sensitivity for malignancy. Quantitatively, adjusting for age, B-mode and lesion volume there was no statistically significant association between longitudinal velocity and malignancy (OR [95% CI] 0.40[0.10, 1.60], p=0.193), but some evidence that higher transverse velocity was associated with decreased odds of malignancy (0.28[0.06, 1.28], p=0.101). Qualitatively malignant masses tended to be towards the blue spectrum (lower velocities); 39.5% (17/43) of predominantly blue masses were malignant, compared to 14.3% (1/7) of red lesions. CONCLUSIONS: Quantitatively and qualitatively there is no statistically significant association between shear wave velocity and malignancy. There is no clear additional role to B-mode imaging currently. KEY POINTS: • Correlation between shear wave velocity and soft tissue malignancy was statistically insignificant • B-mode ultrasound is 76.9 % sensitive and 78.8 % specific • Statistical models show elastography does not significantly add to lesion assessment.