The contribution of elastography to differential diagnosis of breast lesions that have been determined by radiological methods and identified histopatologically

Improvement of new methods has been performed to increase the diagnostic specifity and accuracy of ultrasonography (US) without affecting its sensitivity in the evaluation of breast lesions. For this purpose the elastography imaging technique which shows the hardness of tissues against mechanical stress has been introduced newly. The aim of this study was to compare the findings of elastography with histologic findings in the characterization of breast masses and to evaluate its contribution to mammography (MG), magnetic resonance imaging (MRI), and US. After taking the confirmation of Gaziantep University School of Medicine Ethic Committee, the study includes 133 breast lesions (72 benign, 61 malign) in 106 patients having histopatologic diagnose which has determined by radiological methods, between the age of 17- 87. Conventional US, MG, and MRI evaluation was performed according to BI-RADS classification. Elastography scoring was performed with two different devices according to tsukuba scoring. Measurement of Shear Wave Velocity and strain rate was performed. The mean age of patients was 44,5 (min:17, max:87), and mean diameter of all lesions was 20mm (min:2mm, max:90mm). The most common histology of the benign lesions was fibroadenoma (63%) and the most common malign lesions were invazive ductal carsinomas (88%). Most common location on breast was left upper outer quadrant (34,6%). The area under the ROC curve values were 0,894 for conventional ultrasonography, and the sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and accuracy were 96%, 54%, 63%, 95%, and 72%, respectively, when a best cutoff point of between BI- RADS 3 and 4 was used. The area under the ROC curve values were 0,850 for device 1 and 0, 947 for device 2 according to tsukuba score, and the sensitivity, specificity, false negative ratio, false positive ratio, accuracy, PPV, NPV were 86%, 77%, 13%, 23%, and 81%, 75%, 87% for device 1 and 100%, 84%, 0%, %21%, 84%, 64%, 100% for device 2 respectively, when a best cutoff point of between score 3,5 and 4,5 was used. The mean strain ratio of the 51 lesions which were evaluated by device 2 for the benign lesions (n=37) was 1,55±0,15 (min:0,20; max:3,8), and the malign lesions (n=14) 2,44±0,4 (min:0,83; max:5,8). Malignant lesions, as measured by the strain rate was statistically significantly higher in benign lesions (p=0,014). The area under the ROC curve values were 0,681 for strain ratio which evaluatde by device 2, and the sensitivity, specificity, false positive ratio, false negative ratio, positive predictive value (PPV), negative predictive value (NPV), accuracy were 71%, 65%, 35%, 28%, and 43%, 85%, 67% respectively, when a best cutoff point 1,65 was used. 9 lesions of the 14 malign lesion?s Vs values which assessed at the device 2 was X.XX and the other 5 lesions mean Vs value was 1,4±0,3 (min:0,5; max: 2,1). The 3 lesions of 37 benign lesion?s Vs values was X.XX and the other 34 benign lesion?s mean Vs value was 2,73±0,16 (min:0,70; max:6,1). The diagnostic performance of tsukuba scores that has been obtained by two devices are similar with each other and conventional ultrasonography, and could be used for differentiation between benign and malignant focal breast lesions as a highly specific, noninvasive complementary method. The diagnostic performance of strain ratio, which is mentioned as a quantitative value, has been appeared lower than the accuracy and sensitivity of conventional B mode ultrasonography. Therefore, it might be said that the utilization of strain ratio with conventional B mode is much more accurate and reliable. ARFI method needs more improvement of technolohy, and additional studies with high patient numbers. Vs values might be useful in terms of improving subjective qualitative evaluation on elastograpy. Key words: Elastography, Strain ratio, Shear Wave Velocity, Tsukuba Score