ECR 2010 / C-0418
Sensitivity and specificity of breast sonoelasticity imaging utilizing color elastography and strain ratio measurements
Keywords:
Breast, Ultrasound
Authors:
T. R. Kumm, J. R. Grajo, A. Chau, T. Tucker, M. Szabunio; Tampa, FL/US
DOI:
10.1594/ecr2010/C-0418
Conclusion
- Strain ratio assessment provides additional information about a lesion’s strain characteristics and serves as an objective complement to visual elastography scoring.
- Combining SR assessment with ES improved sensitivity, specificity, PPV, NPV and accuracy compared to visual scoring alone. In this series, 16 lesions with false negative ES score showed SR values in the true positive range for SR.
- Study limitations include a small sample size, operator experience, and potential selection bias as our site is a cancer referral center. Our sensitivity, specificity, NPV/PPV and accuracy were less than expected as compared to other reported trials. Our positive biopsy rate of 27% approximates the expected yield in a screening population.
- A total of 14 malignant lesions (18%) showed an ES and SR in the test negative range. The majority of these were small invasive ductal cancers (Fig. 1) with a mean size of 7.3 mm (range 3 - 13 mm). All false negatives had a ES or 2 and all but one case (5 mm lymphoma) showed a SR > 2.5.
- Narrowing the criteria of test negative to ES of 0 or 1 and a SR of > 2.5 yields a specificity of 0.99 and NPV of 0.99 in this sample. With these parameters, no invasive or in situ cancer would have fallen into the false negative category. Our ultimate clinical goal with elastography would be to reduce the biopsy rate in patients with low-risk breast lesions. Our data showa prospective trial with "tightened" elastography criteria could be done safely in patients with small lesions.