- Email: [email protected]
The Role of Sonoelastography in Breast Lesions
The Role of Sonoelastography in Breast Lesions Richard G. Barr, MD, PhD, FACR, FSRU, FAIUM There is a large body of published material that supports the use of elastography, both strain and shear wave, for characterization of breast lesions. To a lesser extent, elastography can be used in the detection of breast abnormalities. This article reviews the principles of elastography regarding breast imaging, reviews the techniques to perform both strain and shear wave elastography, and reviews the literature and discusses how elastography can be used to improve the characterization of breast lesions to allow for decrease in the number of short-term follow-up examinations and benign biopsies. Semin Ultrasound CT MRI ]:]]]-]]] C 2017 Elsevier Inc. All rights reserved.
Introduction
Basic Principals
T
There are 2 types of ultrasound elastography, strain elastography (SE) and shear wave elastography (SWE). These techniques use different methods to determine tissue stiffness and are complementary techniques. Each has its advantages and disadvantages. The choice of which (or both) to use is dependent on the availability as well as which organ and disease being evaluated.2
he use of palpation for the detection and characterization of disease states has been employed for thousands of years.1 The ancient Egyptians were known to use palpation to detect pathology. Many disease states cause a change in the stiffness of tissue, particularly in most malignancies. Elastography is the imaging equivalent to clinical palpation. Unlike manual palpation, ultrasound elastography can semiquantitate or quantitate the degree of stiffness of a mass or a tissue. It can also assess stiffness of deep tissues that may not be accessible for clinical palpation. Ultrasound elastography has been employed in multiple tissues with varying levels of accuracy in detection and characterization of disease states. Organs where ultrasound elastography may improve assessment of disease states include breast, thyroid, liver, prostate, and tendons.2-7 For focal lesions, the stiffness of malignant breast lesions is much greater than benign lesions with very little overlap allowing for high sensitivity and specificity in characterization of breast lesions as benign or malignant.5,8,9
Department of Radiology, Northeastern Ohio Medical University and Southwoods Imaging, Youngstown, OH. Address reprint requests to Richard G. Barr MD, PhD, FACR, FSRU, FAIUM, Department of Radiology, Northeastern Ohio Medical University and Southwoods Imaging, 7623 Market Street, Youngstown, OH. E-mail: [email protected]
http://dx.doi.org/10.1053/j.sult.2017.05.010 0887-2171/& 2017 Elsevier Inc. All rights reserved.
Precompression When the breast is compressed with the transducer the stiffness of the breast increases.8 This increase in stiffness is identified on both SE and SWE and must be avoided for accurate stiffness values. In general, when performing B-mode ultrasound of the breast some compression is needed to avoid refraction artifacts from Cooper's ligaments. Compressing the breast also helps in bringing deep lesions closer to the optimal focal zone. However, these advantages in B-mode imaging will lead to inaccurate elastography results. The addition of compression while scanning is known as precompression. One method to avoid precompression is to identify an object in the image, and then lift the transducer until that object is as deep as possible in the image and transducer contact is still adequate.10 Another method is to apply a large amount of coupling gel and have some coupling gel between the transducer and the skin while imaging.
Strain Elastography SE evaluates how a tissue changes when a force is applied to it.11 Soft tissues deform more than stiff tissues. By comparing 1
R.G. Barr
2
A
B
Figure 1 Strain elastography evaluated the frame-to-frame changes in a tissue when a stress is applied. In this simple model of an almond within a bowel of gelatin, when a force is applied with the spoon, the gelatin changes shape while the almond does not change shape. The system, therefore, displays the gelatin as soft and the almond as stiff. The display dynamic range used would be almond black and gelatin that changed the most as white.
the amount of deformation of the various tissues in the field of view (FOV) the tissue types can be mapped based on their relative stiffness (Fig. 1). As the exact force applied is not known, a quantitative measurement of a given tissue stiffness cannot be calculated. The image only displaces the relative stiffness of the tissues in the FOV. The force used to cause the deformation can be from the transducer, patient motion (respiration or cardiac motion), or from an Acoustic Radiation Force Impulse (ARFI) ultrasound pulse. A semiquantitative method of strain is the strain ratio. The strain ratio is the stiffness of the tissue or mass of interest divided by the stiffness of a reference tissue. For breast elastography the reference tissue for strain ratio is fat.8 Technique Each ultrasound system has a “sweet spot” of the degree of compress or release cycle to obtain optimal elastogram. That stress can be from movement of the transducer, from patient motion (respiration or cardiac motion or both), or from an ARFI pulse. If the compress or release is too great, only noise will be obtained on the elastogram, and if not enough compress or release is applied, no elastogram is obtained. Most systems have a numerical scale or diagram that displaces how close the compress or release is to optimal. The technique for each vendor is slightly different and can be learned with practice. It is important to maintain the same imaging plane while performing SE as the technique requires identifying changes in the same location when the force is applied and released. As SE is a relative technique, only comparing the stiffness of 1 lesion to another tissue is possible. Therefore, many tissue types should be included in the FOV. For breast, the optimal SE imaging is performed when the FOV includes the lesion, fat, glandular tissue, and the pectoralis muscle.5 This allows for a relatively stable display dynamic range.
Shear Wave Elastography SWE is a quantitative technique, that is, a value of the stiffness is obtained. The stiffness value can be displayed as the speed of the shear wave in meters/second (m/s) or making some
assumptions as Young's modulus in kilo pascals (kPa). SWE can be performed in a single small region of interest (ROI) (point SWE, p-SWE) or over a larger FOV in which the pixels are color-coded (2D-SWE). The 2D-SWE can be performed as a single shot or in real-time. The use of p-SWE in breast is very limited because there is marked variability of the stiffness with a malignant lesion, and the area of maximum stiffness cannot be identified accurately. Therefore, the remainder of this article will discuss only 2D-SWE. 2D-SWE is performed by applying a strong pulse designed for momentum transfer. This pulse is named ARFI. This pulse generates shear waves that are perpendicular to the ARFI pulse. An analogy is dropping a stone in water; the stone is the ARFI pulse and the ripples are the shear waves (Fig. 2). Technique When an area of interest is located, precompression is removed, and the 2D-SWE is turned on. The transducer and patient should remain still during the data acquisition. The image obtained is color-coded with stiff lesions red and soft
Figure 2 Shear wave elastography applies a push pulse (thick red arrow) that generates shear waves. The shear wave speed varies with the stiffness of the tissue it traverses, fast in stiffer tissue and slow in softer tissue. A region of interest (ROI) is placed at the site of interest. B-mode detection pulses are used to monitor the shear wave progression through tissue and calculate the shear wave speed.
Role of sonoelastography in breast lesions
3
A
B
Figure 3 (A) The E/B ratio of a lesion in a 31-year-old patient presenting with a palpable mass. The E/B ratio is 1.22 suggestive of a malignant lesion (ratio 41). Biopsy of the lesion was an invasive ductal carcinoma. (B) The E/B ratio of a palpable lesion in a 91-year-old patient is 0.73 suggestive of a benign lesion (ratio o1.0). On biopsy, the lesion was a benign fibroadenoma.
lesion blue (note in SE this may not be the color scale used). An ROI can then be placed usually at the site of the greatest stiffness in the lesion or in the rim surrounding the lesion.
Clinical Applications Breast elastography can be used to characterize lesions as benign or malignant, identify multiple lesions that may appear as 1 lesion on B-mode, help characterize complex lesions, determine if a lesion is a fat lobule, determine if a lesion is a benign cystic lesion, and help facilitate where to biopsy a lesion.9
Strain Elastography There is a unique characteristic of elastography in breast evaluation. Malignant lesions appear larger on elastography than on B-mode, whereas benign lesions appear smaller.9 There are 3 methods that can be used to characterize using SE. These are the distance ratio of the lesion size on elastography divided by the size of the lesion on B-mode (E/B ratio), a 5-point color scale, and the strain ratio.8 E/B Ratio For reasons not fully understood, malignant breast lesions appear larger than on B-mode, whereas benign lesions appear
smaller. Using the ratio of the lesion, length on elastography divided by the length on B-mode can be used to characterize breast lesions as benign or malignant with high sensitivity and specificity (Fig. 3).
5-Point Color Scale A 5-point color scale can be used to characterize breast lesions. A score of 1 corresponds to a lesion that is soft; a score of 2, a lesion with both soft and hard components; a score of 3, the lesion is stiff and smaller on the elastogram than on B-mode; a score of 4, the lesion is stiff and the same size as B-mode; and a score of 5, the lesion is stiff and larger than B-mode. A score of 1-3 suggests a benign lesion, whereas a score of 4-5 suggests a malignant lesion (Fig. 4). Care must be taken because some use a color scale where blue is stiff and red is soft and others where red is stiff and blue is soft.
Strain Ratio Most systems allow the positioning of an ROI in the lesion and an ROI in a reference tissue and provide the ratio of the stiffness between the 2 tissues. This is the strain ratio. The stiffness of fat is fairly constant within a breast as well as between individuals and is used as reference for breast strain ratio, and in breast it is often referred to as the lesion-to-fat ratio (LFR) (Fig. 5).
R.G. Barr
4
A
B
Figure 4 (A) The 5-point color scale is presented here as an illustration. A score of 1 is given to a lesion that is entirely soft, a score of 2 when the lesion has both soft and hard components, a score of 3 when the lesion is stiff and smaller than the lesion on B-mode, a score of 4 when the lesion is stiff and the same size as B-mode, and a score of 5 when the lesion is stiff and larger than on B-mode. (B) A color-coded SE image where blue is stiff and red is soft. Note that the lesion is stiff and large on the elastogram suggestive of a malignant lesion. The lesion was a biopsy-proven invasive ductal cancer in a 63-yearold patient.
Cyst Artifacts SEy has artifacts in cystic lesions.8 With some ultrasound systems, a “Bull's Eye” appearance is identified in benign simple and complicated cysts (Fig. 6). If there are solid (nonmobile) components in the cyst these will be evident as stiff areas within the Bull's Eye. This artifact has been shown to have a 100% sensitivity and a 100% specificity and can lead to a substantial decrease in the number of benign biopsies (improve positive biopsy rate).12 In other ultrasound systems,
a Blue/Green/Red (BGR) artifact is identified; however, this artifact has not been fully evaluated as to the accuracy of the artifact in characterization of benign cystic lesions.8
Clinical Results There is a unique finding in breast elastography; malignant lesions are larger on elastography than B-mode while benign lesions are smaller. Several papers have used the E/B ratio to
Figure 5 SE image from a 64-year-old patient presenting with a palpable mass. On B-mode image the lesion is a BI-RADS category 4C lesion. The SE image is interpreted using the strain ratio technique. The ROI A is placed in the lesion and the ROI B is placed in fat. The system then calculated that the strain ratio (lesion-to-fat ratio, LFR) is 8.83 suggestive of a malignant lesion. Using the 5-point color scale, the lesion would be a score of 4 also suggestive of a malignancy. Biopsy of the lesion was an invasive ductal carcinoma.
Role of sonoelastography in breast lesions
5
Figure 6 Images from a 40-year-old patient with a palpable mass and an unremarkable mammogram and B-mode ultrasound. On the elastogram, a Bull's eye artifact is identified. The Bull's eye artifact is characterized by a central white spot (red arrow) in a black lesion (white arrows) with a white distal spot (blue arrow). The lesion is an isoechoic complicated cyst (white arrows left image). The lesion was aspirated with a 25-gauge needle with the palpable mass and lesion on B-mode disappearing.
characterize breast masses. These studies have found a sensitivity of 99%-100% and a specificity of 87%-99%.13-15 The E/B ratio has been shown to correlate with tumor grade, with low-grade tumors such as ductal carcinoma in situ and mucinous cancers having a ratio closer to 1, while more aggressive tumors can have a ratio of up to 3.5.16 Using the 5-point color scale, where a score of 1, 2, or 3 is classified as benign and a score of 4 or 5 is classified as malignant, multiple studies have been performed. These studies report a sensitivity of 87%-93% and specificity of 86%-90%.17-19 Results have been more variable with using the strain ratio to characterize breast lesions. The cutoff ratio varies between vendors owing to differences in the algorithm used to determine strain. The application of precompression can also affect the ratio as fat tends to increase in stiffness faster than other tissues with compression.10 Reported cut-offs vary from 2.46-4.8. When using this technique, a literature cutoff value from the same equipment should be used. The studies using the strain ratio have sensitivities of 88%-94% and specificities of 83%-87%.20-24
Interpretation Based on the literature, the maximum stiffness value (Emax) of the lesion or the approximately 3 mm surrounding tissue or both should be used to characterize the lesion. Breast cancers are very heterogeneous and have disorganized structure. In some breast cancers shear waves may not be generated or propagated. In these breast cancers, the lesion will not color-code accurately, as the shear wave speed cannot be calculated. Often there is a rim of higher stiffness around the lesion that can be used to characterize the lesion. In a small
A
B
Shear Wave Elastography Techniques As opposed to SE, SWE provides a quantitative value of stiffness expressed either in m/s or kPa (Fig. 7). For breast SWE, the transducer is placed on the breast with light pressure. No movement of the transducer is required; in fact that is contraindicated. Precompression is a major concern as it is in SE, and the discussion and methods to standardize light compression in SE apply to SWE. Both p-SWE and 2D-SWE have been used to evaluate breast lesions. Breast masses, especially malignancies, tend to be very heterogeneous in stiffness. Therefore, the 2D-SWE technique is preferred, as the larger FOV can depict the differences in stiffness and the area of highest stiffness identified.
Figure 7 (A) SWE image of a 57-year-old patient presenting for an ultrasound guided biopsy for a hypoechoic BI-RADS category 4 lesion demonstrates that the lesion has a stiffness of 2.72 m/s suggestive of a benign lesion. Core biopsy of the lesion was a benign fibroadenoma. (B) SWE of a 67-year-old patient presenting with a BI-RADS category 4C lesion demonstrates that the lesion has a maximum stiffness of 9.77 m/s suggestive of a malignant lesion. Note the heterogeneity of the stiffness in the mass.
R.G. Barr
6
A
B
Figure 8 (A) In the SE image of a 31-year-old with a BI-RADS category 3 hypoechoic lesion it is difficult to identify the lesion on the elastogram. The reason is that the lesion has a stiffness similar to the surrounding breast tissue. The lesion was a fibroadenoma on biopsy. (B) the same lesion has a stiffness value of 2.42 m/s suggestive of a benign lesion on SWE. Note that the SWE confirms that the lesion and surrounding breast tissue have similar stiffness values. This example and that in Fig. 9 demonstrate the advantage of performing both SE and SWE to characterize breast lesions.
number of breast cancers, the lesion may color-code soft and not have a rim of high stiffness, and these can be identified as a false negative by using a quality measure which is evaluated if adequate shear waves were generated to accurately calculate the stiffness value.25 Purely simple cysts do not allow for shear wave speed calculation and are color-coded black. In cysts with some internal echoes or higher viscosity, shear wave velocities can be calculated and these lesions will be color-coded as soft (blue). They will not be distinguishable from some fibroadenomas.8 The cutoff values to distinguish benign from malignant lesions are reported as 4.5 m/s (60 kPa) to 5.2 m/s (80 kPa) in the literature. Many investigators and Breast Imaging and Reporting Data System (BI-RADS) version 5 have recommended that elastography be used to downgrade BI-RADS category 3 lesions to BI-RADS category 2 if the elastography is suggestive of a benign lesion, and upgrading the lesion to a BI-RADS category 4A lesion if the elastography is suggestive of a malignant lesion. If the elastography is suggestive of a benign lesion in a BI-RADS category 4A lesion, the lesion can be downgraded to a BI-RADS category 3 lesion. The interobserver
reliability for maximum and mean stiffness values was highly reproducible.
Review of the literature Multiple studies using 2D-SWE have found a sensitivity of 89%-97% and a specificity of 81%-85%. In a large multicenter study26 using a cut-off of 5.2 m/s (80 kPa) the addition of SWE increased the characterization of lesion over BI-RADS alone, with a sensitivity and specificity of 93.1% and 59.4% in BI-RADS and 92.1% and 7.4% with the addition of SWE.
Combining SE and SWE The difficulties with SE evaluation of breast lesions are usually in benign lesions as the stiffness value of benign lesions are similar to glandular tissue. Therefore, they are often difficult to visualize and the E/B ratio cannot be determined accurately (Fig. 8). In SWE difficulties usually occur in breast malignancies in that shear waves are not generated or are inadequate
Role of sonoelastography in breast lesions
A
B
C
Figure 9 (A) Example of an invasive ductal cancer (biopsy-proven) in a 60-year-old where the central portion of the cancer (red arrow) is not color-coded and the surrounding tissue has stiffness values suggestive of a benign lesion. (B) In the corresponding quality map, the area of the cancer is yellow confirming that the shear wave calculation was inaccurate and should not be used. In general, if the quality map is of poor quality (yellow or red), the lesion is either a simple cyst or a malignancy. (C) The corresponding SE image has an E/B ratio of 1.42 suggestive of a malignant lesion. This case demonstrates the utility of both SE and SWE in characterizing breast lesions.
7
8 for accurate stiffness assessment (Fig. 9). Combining the 2 techniques may lead to improved accuracy. When both SE and SWE are the same, probably benign or probably malignant, there is increased confidence in the result. If the SE and SWE have different results, additional evaluation is needed.
Guidelines Guidelines recommending the use of elastography for characterizing breast lesions have been published by the European Federation Societies for Ultrasound in Medicine (EFSUMB)2 and the World Federation of Ultrasound in Medicine and Biology (WFUMB).5 Both guidelines recommend the addition of elastography to conventional ultrasound to improve the characterizations of breast lesions as benign or malignant. The WFUMB guidelines provide a detailed description of how to perform breast elastography, how to interpret the results, and how to correlate findings obtained on SE and SWE. WFUMB also has guidelines and recommendations for basic science of elastography.11
Conclusions The addition of elastography to a standard breast ultrasound (screening or diagnostic) can be performed in a few minutes. Both SE and SWE are highly sensitive and specific for characterizing breast lesions. The addition of elastography to a standard breast ultrasound will decrease the number of benign biopsies.
Disclosures RGB has equipment grants from Siemens Ultrasound, Philips Ultrasound, SuperSonic Imagine, Hitachi-Aloka Ultrasound, and B and K Ultrasound; Speakers Bureau for Philips Ultrasound, Hitachi-Aloka, and Bracco Diagnostics; Advisory boards for Bracco Diagnostics and Lantheus Medical; and Royalties from Thieme Publishers.
References 1. Tanter M, et al: Quantitative assessment of breast lesion viscoelasticity: initial clinical results using supersonic shear imaging. Ultrasound Med Biol 34(9):1373-1386, 2008 2. Cosgrove D, et al: EFSUMB guidelines and recommendations on the clinical use of ultrasound elastography. Part 2: clinical applications. Ultraschall Med 34(3):238-253, 2013 3. Cosgrove D, et al: WFUMB guidelines and recommendations on the clinical use of ultrasound elastography: Part 4. Thyroid. Ultrasound Med Biol 43(1):4-26, 2017
R.G. Barr 4. Barr RG, et al: WFUMB guidelines and recommendations on the clinical use of ultrasound elastography: Part 5. Prostate. Ultrasound Med Biol 43 (1):27-48, 2017 5. Barr RG, et al: WFUMB guidelines and recommendations for clinical use of ultrasound elastography: Part 2: breast. Ultrasound Med Biol 41 (5):1148-1160, 2015 6. Ferraioli G, et al: WFUMB guidelines and recommendations for clinical use of ultrasound elastography: Part 3: liver. Ultrasound Med Biol 41 (5):1161-1179, 2015 7. Barr RG, et al: Elastography assessment of liver fibrosis: Society of Radiologists in Ultrasound Consensus Conference Statement. Ultrasound Q 32(2):94-107, 2016 8. Barr RG: Sonographic breast elastography: a primer. J Ultrasound Med 31 (5):773-783, 2012 9. Barr RG: Breast Elastography. New York, NY: Thieme Publishers, 2014 10. Barr RG, Zhang Z: Effects of precompression on elasticity imaging of the breast: development of a clinically useful semiquantitative method of precompression assessment. J Ultrasound Med 31(6):895-902, 2012 11. Shiina T, et al: WFUMB guidelines and recommendations for clinical use of ultrasound elastography: Part 1: basic principles and terminology. Ultrasound Med Biol 41(5):1126-1147, 2015 12. Barr RG, Lackey AE: The utility of the "bull's-eye" artifact on breast elasticity imaging in reducing breast lesion biopsy rate. Ultrasound Q 27 (3):151-155, 2011 13. Barr RG: Real-time ultrasound elasticity of the breast: initial clinical results. Ultrasound Q 26(2):61-66, 2010 14. Barr RG, et al: Evaluation of breast lesions using sonographic elasticity imaging: a multicenter trial. J Ultrasound Med 31(2):281-287, 2012 15. Destounis S, et al: Clinical experience with elasticity imaging in a community-based breast center. J Ultrasound Med 32(2):297-302, 2013 16. Grajo JR, Barr RG: Strain elastography in the prediction of breast cancer tumor grade. J Ultrasound Med 33:129-134, 2014 17. Chang JM, et al: Breast mass evaluation: factors influencing the quality of US elastography. Radiology 259(1):59-64, 2011 18. Itoh A, et al: Breast disease: clinical application of US elastography for diagnosis. Radiology 239(2):341-350, 2006 19. Raza S, et al: Using real-time tissue elastography for breast lesion evaluation: our initial experience. J Ultrasound Med 29(4):551-563, 2010 20. Zhi H, et al: Semi-quantitating stiffness of breast solid lesions in ultrasonic elastography. Acad Radiol 15(11):1347-1353, 2008 21. Farrokh A, Wojcinski S, Degenhardt F: Diagnostic value of strain ratio measurement in the differentiation of malignant and benign breast lesions. Ultraschall Med 32(4):400-405, 2011 22. Thomas A, et al: Significant differentiation of focal breast lesions: calculation of strain ratio in breast sonoelastography. Acad Radiol 17 (5):558-563, 2010 23. Alhabshi SM, et al: Semi-quantitative and qualitative assessment of breast ultrasound elastography in differentiating between malignant and benign lesions. Ultrasound Med Biol 39(4):568-578, 2013 24. Stachs A, et al: Differentiating between malignant and benign breast masses: factors limiting sonoelastographic strain ratio. Ultraschall Med 34 (2):131-136, 2013 25. Barr RG, Zhang Z: Shear wave elastography of the breast: value of a quality measure and comparison to strain elastography 275(1):45-53, 2015 26. Berg WA, et al: Shear-wave elastography improves the specificity of breast US: the BE1 multinational study of 939 masses. Radiology 262 (2):435-449, 2012
Introduction
Basic Principals
T
There are 2 types of ultrasound elastography, strain elastography (SE) and shear wave elastography (SWE). These techniques use different methods to determine tissue stiffness and are complementary techniques. Each has its advantages and disadvantages. The choice of which (or both) to use is dependent on the availability as well as which organ and disease being evaluated.2
he use of palpation for the detection and characterization of disease states has been employed for thousands of years.1 The ancient Egyptians were known to use palpation to detect pathology. Many disease states cause a change in the stiffness of tissue, particularly in most malignancies. Elastography is the imaging equivalent to clinical palpation. Unlike manual palpation, ultrasound elastography can semiquantitate or quantitate the degree of stiffness of a mass or a tissue. It can also assess stiffness of deep tissues that may not be accessible for clinical palpation. Ultrasound elastography has been employed in multiple tissues with varying levels of accuracy in detection and characterization of disease states. Organs where ultrasound elastography may improve assessment of disease states include breast, thyroid, liver, prostate, and tendons.2-7 For focal lesions, the stiffness of malignant breast lesions is much greater than benign lesions with very little overlap allowing for high sensitivity and specificity in characterization of breast lesions as benign or malignant.5,8,9
Department of Radiology, Northeastern Ohio Medical University and Southwoods Imaging, Youngstown, OH. Address reprint requests to Richard G. Barr MD, PhD, FACR, FSRU, FAIUM, Department of Radiology, Northeastern Ohio Medical University and Southwoods Imaging, 7623 Market Street, Youngstown, OH. E-mail: [email protected]
http://dx.doi.org/10.1053/j.sult.2017.05.010 0887-2171/& 2017 Elsevier Inc. All rights reserved.
Precompression When the breast is compressed with the transducer the stiffness of the breast increases.8 This increase in stiffness is identified on both SE and SWE and must be avoided for accurate stiffness values. In general, when performing B-mode ultrasound of the breast some compression is needed to avoid refraction artifacts from Cooper's ligaments. Compressing the breast also helps in bringing deep lesions closer to the optimal focal zone. However, these advantages in B-mode imaging will lead to inaccurate elastography results. The addition of compression while scanning is known as precompression. One method to avoid precompression is to identify an object in the image, and then lift the transducer until that object is as deep as possible in the image and transducer contact is still adequate.10 Another method is to apply a large amount of coupling gel and have some coupling gel between the transducer and the skin while imaging.
Strain Elastography SE evaluates how a tissue changes when a force is applied to it.11 Soft tissues deform more than stiff tissues. By comparing 1
R.G. Barr
2
A
B
Figure 1 Strain elastography evaluated the frame-to-frame changes in a tissue when a stress is applied. In this simple model of an almond within a bowel of gelatin, when a force is applied with the spoon, the gelatin changes shape while the almond does not change shape. The system, therefore, displays the gelatin as soft and the almond as stiff. The display dynamic range used would be almond black and gelatin that changed the most as white.
the amount of deformation of the various tissues in the field of view (FOV) the tissue types can be mapped based on their relative stiffness (Fig. 1). As the exact force applied is not known, a quantitative measurement of a given tissue stiffness cannot be calculated. The image only displaces the relative stiffness of the tissues in the FOV. The force used to cause the deformation can be from the transducer, patient motion (respiration or cardiac motion), or from an Acoustic Radiation Force Impulse (ARFI) ultrasound pulse. A semiquantitative method of strain is the strain ratio. The strain ratio is the stiffness of the tissue or mass of interest divided by the stiffness of a reference tissue. For breast elastography the reference tissue for strain ratio is fat.8 Technique Each ultrasound system has a “sweet spot” of the degree of compress or release cycle to obtain optimal elastogram. That stress can be from movement of the transducer, from patient motion (respiration or cardiac motion or both), or from an ARFI pulse. If the compress or release is too great, only noise will be obtained on the elastogram, and if not enough compress or release is applied, no elastogram is obtained. Most systems have a numerical scale or diagram that displaces how close the compress or release is to optimal. The technique for each vendor is slightly different and can be learned with practice. It is important to maintain the same imaging plane while performing SE as the technique requires identifying changes in the same location when the force is applied and released. As SE is a relative technique, only comparing the stiffness of 1 lesion to another tissue is possible. Therefore, many tissue types should be included in the FOV. For breast, the optimal SE imaging is performed when the FOV includes the lesion, fat, glandular tissue, and the pectoralis muscle.5 This allows for a relatively stable display dynamic range.
Shear Wave Elastography SWE is a quantitative technique, that is, a value of the stiffness is obtained. The stiffness value can be displayed as the speed of the shear wave in meters/second (m/s) or making some
assumptions as Young's modulus in kilo pascals (kPa). SWE can be performed in a single small region of interest (ROI) (point SWE, p-SWE) or over a larger FOV in which the pixels are color-coded (2D-SWE). The 2D-SWE can be performed as a single shot or in real-time. The use of p-SWE in breast is very limited because there is marked variability of the stiffness with a malignant lesion, and the area of maximum stiffness cannot be identified accurately. Therefore, the remainder of this article will discuss only 2D-SWE. 2D-SWE is performed by applying a strong pulse designed for momentum transfer. This pulse is named ARFI. This pulse generates shear waves that are perpendicular to the ARFI pulse. An analogy is dropping a stone in water; the stone is the ARFI pulse and the ripples are the shear waves (Fig. 2). Technique When an area of interest is located, precompression is removed, and the 2D-SWE is turned on. The transducer and patient should remain still during the data acquisition. The image obtained is color-coded with stiff lesions red and soft
Figure 2 Shear wave elastography applies a push pulse (thick red arrow) that generates shear waves. The shear wave speed varies with the stiffness of the tissue it traverses, fast in stiffer tissue and slow in softer tissue. A region of interest (ROI) is placed at the site of interest. B-mode detection pulses are used to monitor the shear wave progression through tissue and calculate the shear wave speed.
Role of sonoelastography in breast lesions
3
A
B
Figure 3 (A) The E/B ratio of a lesion in a 31-year-old patient presenting with a palpable mass. The E/B ratio is 1.22 suggestive of a malignant lesion (ratio 41). Biopsy of the lesion was an invasive ductal carcinoma. (B) The E/B ratio of a palpable lesion in a 91-year-old patient is 0.73 suggestive of a benign lesion (ratio o1.0). On biopsy, the lesion was a benign fibroadenoma.
lesion blue (note in SE this may not be the color scale used). An ROI can then be placed usually at the site of the greatest stiffness in the lesion or in the rim surrounding the lesion.
Clinical Applications Breast elastography can be used to characterize lesions as benign or malignant, identify multiple lesions that may appear as 1 lesion on B-mode, help characterize complex lesions, determine if a lesion is a fat lobule, determine if a lesion is a benign cystic lesion, and help facilitate where to biopsy a lesion.9
Strain Elastography There is a unique characteristic of elastography in breast evaluation. Malignant lesions appear larger on elastography than on B-mode, whereas benign lesions appear smaller.9 There are 3 methods that can be used to characterize using SE. These are the distance ratio of the lesion size on elastography divided by the size of the lesion on B-mode (E/B ratio), a 5-point color scale, and the strain ratio.8 E/B Ratio For reasons not fully understood, malignant breast lesions appear larger than on B-mode, whereas benign lesions appear
smaller. Using the ratio of the lesion, length on elastography divided by the length on B-mode can be used to characterize breast lesions as benign or malignant with high sensitivity and specificity (Fig. 3).
5-Point Color Scale A 5-point color scale can be used to characterize breast lesions. A score of 1 corresponds to a lesion that is soft; a score of 2, a lesion with both soft and hard components; a score of 3, the lesion is stiff and smaller on the elastogram than on B-mode; a score of 4, the lesion is stiff and the same size as B-mode; and a score of 5, the lesion is stiff and larger than B-mode. A score of 1-3 suggests a benign lesion, whereas a score of 4-5 suggests a malignant lesion (Fig. 4). Care must be taken because some use a color scale where blue is stiff and red is soft and others where red is stiff and blue is soft.
Strain Ratio Most systems allow the positioning of an ROI in the lesion and an ROI in a reference tissue and provide the ratio of the stiffness between the 2 tissues. This is the strain ratio. The stiffness of fat is fairly constant within a breast as well as between individuals and is used as reference for breast strain ratio, and in breast it is often referred to as the lesion-to-fat ratio (LFR) (Fig. 5).
R.G. Barr
4
A
B
Figure 4 (A) The 5-point color scale is presented here as an illustration. A score of 1 is given to a lesion that is entirely soft, a score of 2 when the lesion has both soft and hard components, a score of 3 when the lesion is stiff and smaller than the lesion on B-mode, a score of 4 when the lesion is stiff and the same size as B-mode, and a score of 5 when the lesion is stiff and larger than on B-mode. (B) A color-coded SE image where blue is stiff and red is soft. Note that the lesion is stiff and large on the elastogram suggestive of a malignant lesion. The lesion was a biopsy-proven invasive ductal cancer in a 63-yearold patient.
Cyst Artifacts SEy has artifacts in cystic lesions.8 With some ultrasound systems, a “Bull's Eye” appearance is identified in benign simple and complicated cysts (Fig. 6). If there are solid (nonmobile) components in the cyst these will be evident as stiff areas within the Bull's Eye. This artifact has been shown to have a 100% sensitivity and a 100% specificity and can lead to a substantial decrease in the number of benign biopsies (improve positive biopsy rate).12 In other ultrasound systems,
a Blue/Green/Red (BGR) artifact is identified; however, this artifact has not been fully evaluated as to the accuracy of the artifact in characterization of benign cystic lesions.8
Clinical Results There is a unique finding in breast elastography; malignant lesions are larger on elastography than B-mode while benign lesions are smaller. Several papers have used the E/B ratio to
Figure 5 SE image from a 64-year-old patient presenting with a palpable mass. On B-mode image the lesion is a BI-RADS category 4C lesion. The SE image is interpreted using the strain ratio technique. The ROI A is placed in the lesion and the ROI B is placed in fat. The system then calculated that the strain ratio (lesion-to-fat ratio, LFR) is 8.83 suggestive of a malignant lesion. Using the 5-point color scale, the lesion would be a score of 4 also suggestive of a malignancy. Biopsy of the lesion was an invasive ductal carcinoma.
Role of sonoelastography in breast lesions
5
Figure 6 Images from a 40-year-old patient with a palpable mass and an unremarkable mammogram and B-mode ultrasound. On the elastogram, a Bull's eye artifact is identified. The Bull's eye artifact is characterized by a central white spot (red arrow) in a black lesion (white arrows) with a white distal spot (blue arrow). The lesion is an isoechoic complicated cyst (white arrows left image). The lesion was aspirated with a 25-gauge needle with the palpable mass and lesion on B-mode disappearing.
characterize breast masses. These studies have found a sensitivity of 99%-100% and a specificity of 87%-99%.13-15 The E/B ratio has been shown to correlate with tumor grade, with low-grade tumors such as ductal carcinoma in situ and mucinous cancers having a ratio closer to 1, while more aggressive tumors can have a ratio of up to 3.5.16 Using the 5-point color scale, where a score of 1, 2, or 3 is classified as benign and a score of 4 or 5 is classified as malignant, multiple studies have been performed. These studies report a sensitivity of 87%-93% and specificity of 86%-90%.17-19 Results have been more variable with using the strain ratio to characterize breast lesions. The cutoff ratio varies between vendors owing to differences in the algorithm used to determine strain. The application of precompression can also affect the ratio as fat tends to increase in stiffness faster than other tissues with compression.10 Reported cut-offs vary from 2.46-4.8. When using this technique, a literature cutoff value from the same equipment should be used. The studies using the strain ratio have sensitivities of 88%-94% and specificities of 83%-87%.20-24
Interpretation Based on the literature, the maximum stiffness value (Emax) of the lesion or the approximately 3 mm surrounding tissue or both should be used to characterize the lesion. Breast cancers are very heterogeneous and have disorganized structure. In some breast cancers shear waves may not be generated or propagated. In these breast cancers, the lesion will not color-code accurately, as the shear wave speed cannot be calculated. Often there is a rim of higher stiffness around the lesion that can be used to characterize the lesion. In a small
A
B
Shear Wave Elastography Techniques As opposed to SE, SWE provides a quantitative value of stiffness expressed either in m/s or kPa (Fig. 7). For breast SWE, the transducer is placed on the breast with light pressure. No movement of the transducer is required; in fact that is contraindicated. Precompression is a major concern as it is in SE, and the discussion and methods to standardize light compression in SE apply to SWE. Both p-SWE and 2D-SWE have been used to evaluate breast lesions. Breast masses, especially malignancies, tend to be very heterogeneous in stiffness. Therefore, the 2D-SWE technique is preferred, as the larger FOV can depict the differences in stiffness and the area of highest stiffness identified.
Figure 7 (A) SWE image of a 57-year-old patient presenting for an ultrasound guided biopsy for a hypoechoic BI-RADS category 4 lesion demonstrates that the lesion has a stiffness of 2.72 m/s suggestive of a benign lesion. Core biopsy of the lesion was a benign fibroadenoma. (B) SWE of a 67-year-old patient presenting with a BI-RADS category 4C lesion demonstrates that the lesion has a maximum stiffness of 9.77 m/s suggestive of a malignant lesion. Note the heterogeneity of the stiffness in the mass.
R.G. Barr
6
A
B
Figure 8 (A) In the SE image of a 31-year-old with a BI-RADS category 3 hypoechoic lesion it is difficult to identify the lesion on the elastogram. The reason is that the lesion has a stiffness similar to the surrounding breast tissue. The lesion was a fibroadenoma on biopsy. (B) the same lesion has a stiffness value of 2.42 m/s suggestive of a benign lesion on SWE. Note that the SWE confirms that the lesion and surrounding breast tissue have similar stiffness values. This example and that in Fig. 9 demonstrate the advantage of performing both SE and SWE to characterize breast lesions.
number of breast cancers, the lesion may color-code soft and not have a rim of high stiffness, and these can be identified as a false negative by using a quality measure which is evaluated if adequate shear waves were generated to accurately calculate the stiffness value.25 Purely simple cysts do not allow for shear wave speed calculation and are color-coded black. In cysts with some internal echoes or higher viscosity, shear wave velocities can be calculated and these lesions will be color-coded as soft (blue). They will not be distinguishable from some fibroadenomas.8 The cutoff values to distinguish benign from malignant lesions are reported as 4.5 m/s (60 kPa) to 5.2 m/s (80 kPa) in the literature. Many investigators and Breast Imaging and Reporting Data System (BI-RADS) version 5 have recommended that elastography be used to downgrade BI-RADS category 3 lesions to BI-RADS category 2 if the elastography is suggestive of a benign lesion, and upgrading the lesion to a BI-RADS category 4A lesion if the elastography is suggestive of a malignant lesion. If the elastography is suggestive of a benign lesion in a BI-RADS category 4A lesion, the lesion can be downgraded to a BI-RADS category 3 lesion. The interobserver
reliability for maximum and mean stiffness values was highly reproducible.
Review of the literature Multiple studies using 2D-SWE have found a sensitivity of 89%-97% and a specificity of 81%-85%. In a large multicenter study26 using a cut-off of 5.2 m/s (80 kPa) the addition of SWE increased the characterization of lesion over BI-RADS alone, with a sensitivity and specificity of 93.1% and 59.4% in BI-RADS and 92.1% and 7.4% with the addition of SWE.
Combining SE and SWE The difficulties with SE evaluation of breast lesions are usually in benign lesions as the stiffness value of benign lesions are similar to glandular tissue. Therefore, they are often difficult to visualize and the E/B ratio cannot be determined accurately (Fig. 8). In SWE difficulties usually occur in breast malignancies in that shear waves are not generated or are inadequate
Role of sonoelastography in breast lesions
A
B
C
Figure 9 (A) Example of an invasive ductal cancer (biopsy-proven) in a 60-year-old where the central portion of the cancer (red arrow) is not color-coded and the surrounding tissue has stiffness values suggestive of a benign lesion. (B) In the corresponding quality map, the area of the cancer is yellow confirming that the shear wave calculation was inaccurate and should not be used. In general, if the quality map is of poor quality (yellow or red), the lesion is either a simple cyst or a malignancy. (C) The corresponding SE image has an E/B ratio of 1.42 suggestive of a malignant lesion. This case demonstrates the utility of both SE and SWE in characterizing breast lesions.
7
8 for accurate stiffness assessment (Fig. 9). Combining the 2 techniques may lead to improved accuracy. When both SE and SWE are the same, probably benign or probably malignant, there is increased confidence in the result. If the SE and SWE have different results, additional evaluation is needed.
Guidelines Guidelines recommending the use of elastography for characterizing breast lesions have been published by the European Federation Societies for Ultrasound in Medicine (EFSUMB)2 and the World Federation of Ultrasound in Medicine and Biology (WFUMB).5 Both guidelines recommend the addition of elastography to conventional ultrasound to improve the characterizations of breast lesions as benign or malignant. The WFUMB guidelines provide a detailed description of how to perform breast elastography, how to interpret the results, and how to correlate findings obtained on SE and SWE. WFUMB also has guidelines and recommendations for basic science of elastography.11
Conclusions The addition of elastography to a standard breast ultrasound (screening or diagnostic) can be performed in a few minutes. Both SE and SWE are highly sensitive and specific for characterizing breast lesions. The addition of elastography to a standard breast ultrasound will decrease the number of benign biopsies.
Disclosures RGB has equipment grants from Siemens Ultrasound, Philips Ultrasound, SuperSonic Imagine, Hitachi-Aloka Ultrasound, and B and K Ultrasound; Speakers Bureau for Philips Ultrasound, Hitachi-Aloka, and Bracco Diagnostics; Advisory boards for Bracco Diagnostics and Lantheus Medical; and Royalties from Thieme Publishers.
References 1. Tanter M, et al: Quantitative assessment of breast lesion viscoelasticity: initial clinical results using supersonic shear imaging. Ultrasound Med Biol 34(9):1373-1386, 2008 2. Cosgrove D, et al: EFSUMB guidelines and recommendations on the clinical use of ultrasound elastography. Part 2: clinical applications. Ultraschall Med 34(3):238-253, 2013 3. Cosgrove D, et al: WFUMB guidelines and recommendations on the clinical use of ultrasound elastography: Part 4. Thyroid. Ultrasound Med Biol 43(1):4-26, 2017
R.G. Barr 4. Barr RG, et al: WFUMB guidelines and recommendations on the clinical use of ultrasound elastography: Part 5. Prostate. Ultrasound Med Biol 43 (1):27-48, 2017 5. Barr RG, et al: WFUMB guidelines and recommendations for clinical use of ultrasound elastography: Part 2: breast. Ultrasound Med Biol 41 (5):1148-1160, 2015 6. Ferraioli G, et al: WFUMB guidelines and recommendations for clinical use of ultrasound elastography: Part 3: liver. Ultrasound Med Biol 41 (5):1161-1179, 2015 7. Barr RG, et al: Elastography assessment of liver fibrosis: Society of Radiologists in Ultrasound Consensus Conference Statement. Ultrasound Q 32(2):94-107, 2016 8. Barr RG: Sonographic breast elastography: a primer. J Ultrasound Med 31 (5):773-783, 2012 9. Barr RG: Breast Elastography. New York, NY: Thieme Publishers, 2014 10. Barr RG, Zhang Z: Effects of precompression on elasticity imaging of the breast: development of a clinically useful semiquantitative method of precompression assessment. J Ultrasound Med 31(6):895-902, 2012 11. Shiina T, et al: WFUMB guidelines and recommendations for clinical use of ultrasound elastography: Part 1: basic principles and terminology. Ultrasound Med Biol 41(5):1126-1147, 2015 12. Barr RG, Lackey AE: The utility of the "bull's-eye" artifact on breast elasticity imaging in reducing breast lesion biopsy rate. Ultrasound Q 27 (3):151-155, 2011 13. Barr RG: Real-time ultrasound elasticity of the breast: initial clinical results. Ultrasound Q 26(2):61-66, 2010 14. Barr RG, et al: Evaluation of breast lesions using sonographic elasticity imaging: a multicenter trial. J Ultrasound Med 31(2):281-287, 2012 15. Destounis S, et al: Clinical experience with elasticity imaging in a community-based breast center. J Ultrasound Med 32(2):297-302, 2013 16. Grajo JR, Barr RG: Strain elastography in the prediction of breast cancer tumor grade. J Ultrasound Med 33:129-134, 2014 17. Chang JM, et al: Breast mass evaluation: factors influencing the quality of US elastography. Radiology 259(1):59-64, 2011 18. Itoh A, et al: Breast disease: clinical application of US elastography for diagnosis. Radiology 239(2):341-350, 2006 19. Raza S, et al: Using real-time tissue elastography for breast lesion evaluation: our initial experience. J Ultrasound Med 29(4):551-563, 2010 20. Zhi H, et al: Semi-quantitating stiffness of breast solid lesions in ultrasonic elastography. Acad Radiol 15(11):1347-1353, 2008 21. Farrokh A, Wojcinski S, Degenhardt F: Diagnostic value of strain ratio measurement in the differentiation of malignant and benign breast lesions. Ultraschall Med 32(4):400-405, 2011 22. Thomas A, et al: Significant differentiation of focal breast lesions: calculation of strain ratio in breast sonoelastography. Acad Radiol 17 (5):558-563, 2010 23. Alhabshi SM, et al: Semi-quantitative and qualitative assessment of breast ultrasound elastography in differentiating between malignant and benign lesions. Ultrasound Med Biol 39(4):568-578, 2013 24. Stachs A, et al: Differentiating between malignant and benign breast masses: factors limiting sonoelastographic strain ratio. Ultraschall Med 34 (2):131-136, 2013 25. Barr RG, Zhang Z: Shear wave elastography of the breast: value of a quality measure and comparison to strain elastography 275(1):45-53, 2015 26. Berg WA, et al: Shear-wave elastography improves the specificity of breast US: the BE1 multinational study of 939 masses. Radiology 262 (2):435-449, 2012