Tharakeswara Kumar Bathala MD, Presenter: Nothing to Disclose

Gaiane M. Rauch MD, PhD, Abstract Co-Author: Nothing to Disclose

Melanie Bass, Abstract Co-Author: Nothing to Disclose

Deborah Borst, Abstract Co-Author: Nothing to Disclose

Brian Hobbs PhD, Abstract Co-Author: Nothing to Disclose

Deepak G. Bedi MBBCh, Abstract Co-Author: Consultant, Koninklijke Philips Electronics NV

To evaluate diagnostic performance of conventional ultrasonography (US) combined with US Strain Elastography (USE) for differentiation between benign and malignant subcutaneous soft tissue mass lesions, with the pathology as reference standard.

After Institutional IRB approval, we identified 74 patients with a subcutaneous soft tissue mass who had US and USE from January 2009 to May 2012. Three radiologists retrospectively reviewed US and USE images in consensus. Gray scale US imaging features were classified as benign, malignant and indeterminate. USE images were assessed according to tissue elasticity based on color scale and classified as soft, intermediate and hard. Pathological diagnosis obtained either by percutaneous biopsy or surgical excision was used as reference standard. The statistical analysis included evaluation of sensitivity and specificity for US and USE separately, as well as a composite evaluation of US + USE; Bowker’s test was used for evaluation of matched US and USE outcomes for symmetry.

Out of 74 lesions, US 37 were classified as benign, 8 indeterminate, 29 malignant. USE found 35 benign, 14 indeterminate, 25 malignant lesions. US+USE classified 40 as benign, 4 indeterminate and 30 malignant. The estimated sensitivity and 95% CI for US, USE and US+USE was 100% (82-100%), 95% (75-100%), and 100% (82-100%) respectively. The specificity for US, USE and US+USE was 71% (57-82%), 67% (53-79%), and 77% (63-87%) respectively. Significant evidence for the lack of agreement among the matched US and Elastography results was not found (p=0.51). The data suggest US alone is highly sensitive for detecting and characterization of subcutaneous soft tissue lesions. Only 8 patients resulted in an indeterminate US, all of which had benign lesions on pathology. Among these 8, USE was determinate for only 4, of which 3 were correctly classified as benign and 1 was incorrectly classified as malignant. Thus, estimated specificity is improved for the US+USE.

In the presence of an indeterminate result on conventional US, USE may improve specificity for diagnosing subcutaneous soft tissue lesions. Combination of US and USE could provide a better diagnostic performance than conventional US alone.

Addition of USE evaluation to conventional gray scale US imaging improves imaging-based diagnostic information for soft tissue nodule work up.

Bathala, T, Rauch, G, Bass, M, Borst, D, Hobbs, B, Bedi, D, Diagnostic Performance of Conventional Ultrasonography Combined with US Strain Elastography for Differentiation between Benign and Malignant Subcutaneous Soft Tissue Mass Lesions.  Radiological Society of North America 2013 Scientific Assembly and Annual Meeting, December 1 - December 6, 2013 ,Chicago IL. http://archive.rsna.org/2013/13020161.html