Clinical Application, Histopathological And Biomechanical Basis Of Static And Dynamic Ultrasound Elastography Of Breast Lesions

Part Ⅰ Role of static and dynamic ultrasound elastography in the evaluation of breast lesionsChapter I:Breast lesions evaluated by static ultrasound elastographyObjective:To evaluate the diagnostic performance of static ultrasound elastography for the differentiation of breast lesions, compared with conventional ultrasound (US).Materials and Methods:A total of223breast lesions (78malignant,145benign) in207women were enrolled and evaluated by using static ultrasound elastography in this prospective study. For each lesion, the elastic score was recorded according to a five-point scoring system, the strain ratio was calculated, and the mean value of the color coded strain histogram of the lesion was collected. All lesions were subjected to core needle biopsy or surgical excision. The difference among the three types of static ultrasound elastographic features in both malignant and benign groups was evaluated. A final Breast Imaging Reporting and Data System (BI-RADS) assessments based on conventional ultrasound (US), and combined conventional US and the three types of static ultrasound elastographic features were performed. The sensitivity, specificity, and the areas under the receiver operating characteristic (ROC) curve (Az value) in determining malignancy were calculated for conventional US, the elastic score, the strain ratio, the mean value of strain histogram, and combined conventional US and the three types of static ultrasound elastographic features.Results:All three types of static ultrasound elastographic features were significantly different between malignant and benign groups. The Az values of the three types of features were0.831,0.842, and0.852, respectively, and there were no significant difference among them. The Az value (0.941) of the combined conventional US and three types of static ultrasound elastographic features was significantly higher than that of the conventional US (0.891). Founded on this combination, benign lesions recommended for interventional procedure would drop from57.9%(based on conventional US) to26.9%, and all malignant lesions could be correctly selected for biopsy or aspiration.Conclusions:Static ultrasound elastography had potential to differentiate breast lesions and the combined conventional US and static ultrasound elastographic features should be recommended in the diagnosis of breast diseases. Chapter Ⅱ:Breast lesions evaluated by shear wave elastography Objective:To analyze the diagnostic performance of shear wave elastography (SWE) in differentiating between benign and malignant breast lesions, specially emphasizing on the value of the "stiff rim" sign, compared with conventional ultrasound (US).Materials and Methods:A total of193consecutive patients with212breast lesions (59malignant,153benign) were enrolled in this study. The "stiff rim" sign, a qualitative SWE feature, was evaluated at the display setting of lower than180kPa and180kPa. The quantitative SWE features were assessed. A final Breast Imaging Reporting and Data System (BI-RADS) assessments based on conventional US, SWE, and combined conventional US and SWE were performed. The sensitivity, specificity, and the areas under the receiver operating characteristic (ROC) curve (Az value) in determining malignancy were calculated for conventional US, SWE features, and combined conventional US and SWE.Results:The Az value for the "stiff rim" sign at the display setting of lower than180kPa (0.904) was significantly higher than that for all other qualitative and quantitative SWE features, and was comparable with that for conventional US (0.894). The combination of the "stiff rim" sign at the display setting of lower than180kPa and conventional US obtained the highest Az value (0.975) than that based on only conventional US and any single SWE feature alone, and yielded higher sensitivity (98.3%) and similar specificity (90.9%) than conventional US. Founded on this combination, benign lesions recommended for interventional procedure would drop from56.9%(based on conventional US) to16.3%, and all malignant lesions could be correctly selected for biopsy or aspiration.Conclusions:Adding SWE features, specially the "stiff rim" sign at the display setting of lower than180kPa, to conventional US has the potential to improve the differentiation of breast lesions. Chapter Ⅲ:Breast lesions evaluated by acoustic radiation force impulse imaging (ARFI)Objective:To analyze the value of acoustic radiation force impulse imaging (ARFI) with Virtual Touch tissue quantification (VTQ) in the characterization of breast lesions, compared with conventional ultrasound (US).Materials and Methods:A total of175breast lesions (67malignant,108benign) in173women were enrolled in this prospective study. By using VTQ technique, shear wave velocity (SWV) was measured within the breast lesion (internal SWV, SWVi), in the boundary zone (boundary SWV, SWVb), in the normal-appearing glandular tissue (glandular SWV, SWVg), and in the subcutaneous fatty tissue (fatty SWV, SWVf). All lesions were subjected to core needle biopsy or surgical excision. The difference among the four types of SWV in both malignant and benign groups was evaluated. A final Breast Imaging Reporting and Data System (BI-RADS) assessments based on conventional ultrasound (US), and combined conventional US and SWVi, SWVb were performed. The sensitivity, specificity, and the areas under the receiver operating characteristic (ROC) curve (Az value) in determining malignancy were calculated for conventional US, SWVi, SWVb, and combined conventional US and SWVi, SWVb.Results:All four types of SWV were significantly higher in malignant group than in benign group. The sensitivity, specificity and Az for differentiating malignant lesions were55.2%,95.4%and0.706for SWVi,85.1%,55.6%and0.747for SWVb,68.2%and69.6%for SWVg, and83.6%,84.3%and0.907for combined conventional US and SWVi, SWVb. The Az value of the combined conventional US and SWVi, SWVb was significantly higher than that of the conventional US. Founded on this combination, benign lesions recommended for interventional procedure would drop from72.2%(based on conventional US) to49.1%.Conclusions:VTQ had potential to characterize breast lesions and reflect the architectural distortion in glandular tissue and subcutaneous fatty tissue caused by malignant breast tumor. Part Ⅱ Histopathological and biomechanical basis of static and dynamic ultrasound elastography of breast lesionsChapter Ⅰ:Histopathological basis of static and dynamic ultrasound elastography of breast lesionsObjective:To assess the relationship between static and dynamic breast ultrasound elastographic features and the content and arrangement of collagen fibers of breast lesions.Materials and Methods:A total of74breast lesions (33malignant,41benign) in70women were enrolled and evaluated by using both static ultrasound elastography and acoustic radiation force impulse imaging (ARFI) with Virtual Touch tissue quantification (VTQ) in this prospective study. For each lesion, the mean value of the color coded strain histogram within the lesion was recorded in static ultrasound elastography, and shear wave velocity (SWV) was measured within the breast lesion (internal SWV, SWVi), in the boundary zone (boundary SWV, SWVb). The SWVi and SWVb were then converted into internal elastic modulus (Ei) and boundary elastic modulus (Eb). All lesions were subjected to surgical excision. Masson’s Trichrome Staining was applied to compare the content and arrangement of collagen fibers in both benign and malignant lesions.Results:There were significant differences in static and dynamic ultrasound elastographic values, and significant differences in the content and arrangement of collagen fibers between malignant and benign groups. Except for the mean value of the strain histogram within the benign lesions, there were no correlations between ultrasound elastographic values and collagen contents in both malignant and benign groups.Conclusions:The static and dynamic ultrasound elastographic features of breast lesions might not determined by the content of collagen fibers, but might be affected by the arrangement of collagen fibers. Chapter Ⅱ:Biomechanical basis of static and dynamic ultrasound elastography of breast lesionsObjective:To assess the relationship between static and dynamic ultrasound elastographic features and the elastic moduli measured under compression in vitro.Materials and Methods:A total of93breast lesions (38malignant,55benign) in89women were enrolled and evaluated by using both static ultrasound elastography and acoustic radiation force impulse imaging (ARFI) with Virtual Touch tissue quantification (VTQ) in this prospective study. For each lesion, the mean value of the color coded strain histogram within the lesion was recorded in static ultrasound elastography, and shear wave velocity (SWV) was measured within the breast lesion (internal SWV, SWVi), in the boundary zone (boundary SWV, SWVb). The SWVi and SWVb were then converted into internal elastic modulus (Ei) and boundary elastic modulus (Eb). All lesions were subjected to surgical excision and then subjected to elastic moduli measurement.Results:There were significant differences in static and dynamic ultrasound elastographic values, and the elastic moduli in malignant group were significantly higher than that in benign group. Moderate significant negative correlation between the mean value of the color coded strain histogram and the elastic moduli measured in vitro was found, low correlations between the dynamic ultrasound elastographic values and the elastic moduli measured in vitro were also revealed.Conclusions:The static and dynamic ultrasound elastographic features of breast lesions were related to the elastic moduli measured in vitro.