RESUMO
PURPOSE: The purpose of this study was to evaluate the diagnostic performance of sonoelastography by using real-time strain ratio and qualitative 5-stage elasticity score in breast lesions for which ultrasonographic evaluation suggested malignancy (BI-RADS 4 and 5 lesions). MATERIALS AND METHODS: From January 2012 to October 2012, 168 solid breast lesions were investigated using sonoelastography. The strain ratios and the elasticity scores were calculated. Final diagnosis was made by histopathological analysis. Areas under the curve and cut-off points were used to assess diagnostic performance of sonoelastography. The sensitivity, and specificity of these two imaging tests were compared using McNemar test. RESULTS: The strain ratios of malignant lesions (mean value=9.3) and benign lesions (mean value=3.75) were significantly different (P<0.00001). Using a cut-off value of 4.79, strain ratio had 78.8% sensitivity, 78.3% specificity, 86.7% positive predictive value (PPV) and 67% negative predictive value (NPV). Using a 5-stage elasticity score system, a 83.3% sensitivity, 74.6% specificity, 85.7% PPV and 71.0% NPV were obtained. CONCLUSION: Our study indicates that sonoelastography has good clinical value to discriminate between benign and malignant breast lesions. There are no differences in terms of diagnostic performance in differentiating malignant from benign lesions when strain ratio or color scoring are used alone. However the diagnostic performance is increased when strain ratio and color scoring evaluations are used in combination.
