Jay R. Parikh, MD

Over the past 2 decades, one of the exciting advances in medicine and imaging research has been the marked expansion of the capabilities of breast ultrasound in the evaluation of breast disease. Breast ultrasound has become a fundamental component of a state-of-the-art, comprehensive breast imaging center.

A key to understanding ultrasound is knowledge of the nature of the ultrasound transducer. A transducer is, fundamentally, a device that converts one form of energy to another. Modern ultrasound transducers are handheld units that convert electric signals into ultrasonic energy that is then transmitted into the tissues. Typically, a piezoelectric crystal near the face of the transducer generates high-frequency sound when voltage is applied. The sonic beams used in diagnostic breast ultrasound typically have frequencies of more than 7 million cycles per second (7 MHz). Following interaction of the sound waves with the tissues, the transducer receives and reconverts ultrasound energy back into an electrical signal, which is used to create the image.

EQUIPMENT AND TECHNICAL ISSUES

Bruce Porter, MD

Breast ultrasound does not expose the patient to ionizing radiation. Hence, ultrasound is considered by the 2000-2001 American College of Radiology (ACR) Standard for the Performance of Breast Ultrasound Examination¹ and the National Comprehensive Cancer Care Network guidelines² to be the appropriate modality for the initial evaluation of a palpable breast mass in a pregnant woman and for the evaluation of a palpable breast mass in a woman under age 30.

Breast ultrasound has some drawbacks, which include its relatively higher cost, compared with mammography; operator-skill dependence; difficulty in providing reproducible results between different facilities; and the time required to carry out the study. Perhaps the biggest shortcoming of ultrasound is its higher false-negative rate, when compared with mammography, for general screening, especially for the malignant microcalcifications that are typically better seen mammographically.

Figure 1. High-frequency radial ultrasound image with spatial compounding of the right breast, 2:00 location at the site of a 38-year-old female's self-detected and physician-detected thickening. Microcalcifications are identified sonographically as multiple hyperechoic foci (arrows). These are distinguished from speckle artifact by the reproducibility of these calcifications from various angles during real-time scanning. These sonographically detected microcalcifications correlated with mammographic microcalcifications and palpable thickening.

The use of state-of-the-art, high-resolution breast ultrasound equipment is important. A dedicated breast ultrasound unit is preferable. High-frequency linear array transducers are required because linear transducers have a wider near field and can more easily guide intervention procedures. The 2000-2001 ACR Standard for the Performance of Breast Ultrasound Examination suggests transducer frequencies of 7 MHz or higher. If broadband systems are used, the ACR standard states that a center frequency of 6 MHz or higher is needed. Current transducer frequencies are typically 10 MHz or higher. Some new ultrasound machines contain image-enhancing hardware such as compound imaging and tissue harmonics, which can be of appreciable help in specific circumstances. If possible, color-Doppler capability should also be available, as explained below.

A transducer of the correct frequency should be used. The frequency must be appropriate to the size and depth of the area of abnormality. The ACR standard states that the frequency should be high enough to permit differentiation of fluid versus solid breast masses; the standards recognize that this may not always be possible.

Figure 2. Ultrasound postfiring image demonstrates the 12-gauge core biopsy needle passing through the malignant calcifications and an adjacent hypoechoic area of concern. This image was obtained immediately after firing of the biopsy needle. Note the trajectory of the needle, parallel to the pectoralis major muscle. This minimizes the risk of puncturing the chest wall and creating a pneumothorax.

The power, time-gain-compensation (TGC) curve, and focal zone settings must be optimized. Preferably, the preset values will be used as a starting point, with individualized adjustments made as necessary. As a guide, the power is kept as low as possible, allowing the beam just to penetrate the chest wall. The TGC should be set to allow even penetration of the entire field of breast tissue. The ACR standard also states that gain settings should be adjusted to allow simple cysts to be distinguished from solid masses. Power and gain should not be so high as to create artifactual echoes within a simple cyst (causing it to appear as a solid lesion), but should not be so low as to miss real internal echoes in a solid mass. The focal zone is set at the lesion's depth. Multiple focal zones are often needed.

SCANNING TECHNIQUE

Placing the patient in a supine position minimizes the depth of tissue penetration needed for imaging by the ultrasound beam. Raising the ipsilateral hand behind the head flattens the breast and minimizes the tissue depth. For lateral lesions, the ACR standard suggests supine-oblique positioning for scanning. Turning the patient away from the side to be examined flattens the lateral tissue against the chest wall. For medial lesions, the supine position is preferred.

Figure 3. Specimen mammography using magnification technique demonstrates multiple microcalcifications within the core-needle biopsy samples. The presence of these microcalcifications in the specimens confirms successful biopsy of the microcalcifications that were seen using both mammography and ultrasound. Histologically, the calcifications were found to be related to ductal carcinoma in situ (DCIS) of high nuclear grade with necrosis. No invasion was identified. At mastectomy, a 4-cm area of high-nuclear-grade DCIS was identified in the upper inner to central portion of the right breast. No invasion was identified.

Various scanning methods have been proposed. At Swedish Medical Center, Seattle, and First Hill Diagnostic Imaging, Seattle, scanning in the radial and antiradial planes is preferred. Systematic radial and antiradial scanning ensures sonographic evaluation of the area of concern. Palpation and imaging can be done simultaneously with the hand-walk technique. While the transducer is moved with one hand, the index, middle, and ring fingers of the contralateral hand are placed at the leading end of the transducer.

Skin and superficial breast tissue lesions will be better visualized with higher frequency transducers or the use of a stand-off pad, if necessary. Scanning of the retroareolar region is often limited by shadowing from the nipple. Angling the transducer behind and beneath the nipple helps in visualization of this difficult area.

PERSONNEL ISSUES

The 2000-2001 ACR Standard for the Performance of Breast Ultrasound Examination states that diagnostic ultrasound examinations "should be supervised and interpreted by trained and qualified physicians . . . Physicians who perform and/or interpret diagnostic ultrasound examinations should be licensed medical practitioners who have a thorough understanding for the indications for ultrasound. . . and they should be capable of correlating the results of other procedures with sonographic findings."¹

Figure 4. High-resolution radial ultrasound image of a septated, but otherwise smooth-walled, anechoic subareolar cyst. Due to the septation, as well as a history of prior breast cancer and patient concern, fine needle aspiration was requested. Both cavities were completely evacuated, and the cytology was negative for malignant cells, as expected.

The 2000-2001 ACR standard also recognizes diagnostic ultrasound by trained and qualified diagnostic medical sonographers, stating, "The qualification can be demonstrated by certification or eligibility for same by a nationally recognized certifying body."¹

Breast ultrasound is unique, compared with other ultrasound examinations, because of the necessity of correlating the ultrasound findings with mammograms and physical examinations. In our experience, most sonographers have little training in mammographic triangulation and correlation, and have not received formal training in breast physical examination. In the absence of such personnel at a breast imaging center, we would advocate the active supervision of breast ultrasound by, a trained physician. This enables practitioners to provide optimal patient care and reduces medicolegal risk for the facility.

The interpreting physician (usually a radiologist) should be able to understand triangulation principles for mammographic abnormalities and to correlate breast ultrasound with mammograms. The physician should also be capable of, and comfortable with, breast physical examinations. With a self-referred patient, the interpreting physician may be responsible for the physical examination, especially if the patient has not had a recent physical examination elsewhere. The physician should be knowledgeable about breast ultrasound anatomy, pathology, and imaging artifacts. Ideally, the physician should be comfortable scanning the patient's breast himself or herself and making a comprehensive assessment, which can then be communicated to the patient.

BREAST ULTRASOUND INDICATIONS

Figure 5. High-frequency radial ultrasound image of the left breast, 12:00 location, at a 40-year-old female's self-detected lump demonstrating a markedly hypoechoic mass with microlobulations and posterior shadowing. These features are malignant characteristics, according to the criteria of Stavros.5 This is a prefiring image captured during ultrasound-guided core breast biopsy, and it shows the benefit of sonographic guidance. The trajectory of the needle is kept parallel to the chest wall to avoid the complication of pneumothorax. Core biopsy pathology demonstrated an infiltrating ductal carcinoma.

With progressive advances in technology and clinical research, the indications and role for breast ultrasound have increased dramatically over the past few years. The indications for diagnostic breast ultrasound, according to the 2000-2001 ACR standard, include the identification and characterization of palpable and nonpalpable abnormalities and the further evaluation of clinical and mammographic findings.

Breast ultrasound is an essential component of the imaging evaluation of mammographic abnormalities. When there is a mammographically detected mass, developing focal asymmetric density, or architectural distortion, breast ultrasound is an essential component of the algorithm of care. If a woman has mammographically dense tissue and suspicious microcalcifications on her mammogram, ultrasound can be an invaluable tool to locate masses associated with suspicious calcifications and to guide core-needle biopsy of these areas.

Self-detected or clinically detected breast masses, focal thickening, or focal breast pain in the screening age group should first be evaluated mammographically. A negative mammogram, however, is an incomplete evaluation of an area of dominant clinical concern. It is important to recognize, and inform patients, that the false-negative rate of mammography in the detection of breast cancer has been reported to be approximately 10% in the Breast Cancer Detection Project.^3,4 The sensitivity of mammography is especially lowered in women with mammographically dense tissue. Therefore, in the setting of a dominant clinical concern and negative mammogram, the next step in evaluation should be a targeted breast ultrasound.

Figure 6. Radial and antiradial ultrasound of the right breast, 12:00 location, at the site of a 46-year-old female's self-detected lump. The patient previously had bilateral augmentation with saline implants. Mammographic evaluation showed normal, dense breast tissue. Radial and antiradial scanning confirmed the presence of a well-defined, ellipsoid, 1.2-cm, solid, uniform, isoechoic mass with a thin, echogenic pseudocapsule. The features were consistent with a benign lesion. Surgical excision revealed a fibroadenoma. Incidentally noted during scanning were normal implant folds.

The role of ultrasound in the characterization of breast masses has expanded over the past 2 decades. Initially, sonography was largely limited to distinguishing cystic from solid masses in the breast. This concept is outdated, but remains prevalent in the imaging community.

The diagnosis of a simple cyst is made with 96% to 100% certainty by breast ultrasound.^5,6 A simple cyst is defined by breast ultrasound as an anechoic mass with smooth margins and uniform through transmission. If the cyst has internal echoes, wall irregularity, mural nodularity or septation, shadowing, nonuniform transmission, and/or any other feature not associated with a simple cyst, it is by definition a complex cyst, and aspiration and/or biopsy should be contemplated. Ultrasound is a rapid, easy, and efficient method to guide intervention in these settings.

Ultrasound is also useful in the characterization of solid masses. Various criteria have been proposed to help characterize benign versus malignant solid masses. Articles by Stavros et al,⁵ Skaane et al,⁷and others permit a systematic analysis of malignant and benign features of solid breast masses.

MALIGNANT SOLID BREAST MASSES

Figure 7. Transverse ultrasound image with spatial compounding in a 48-year-old female with suspected silicone implant rupture demonstrates an intensely echogenic 1.8-cm collection within a 2.5-cm lymph node in the right axilla (see calipers on image). A similar second intensely echogenic 1.7-cm collection is seen in the adjacent tissue from a second axillary lymph node. In both lymph nodes, the anterior margin of this snowstorm or echogenic noise is well delineated, but the posterior structures are obscured by the intensely echogenic noise. These two lymph nodes corresponded to two high-density lymph nodes seen mammographically. This appearance is characteristic of silicone lymphadenopathy.

Spiculated margins, as demonstrated by sonography, are the ultrasound findings with highest positive predictive value. The spiculation is thought to correlate with mammographic spiculation. It can be postulated that the spiculations represent tumor tentacles or desmoplastic reaction. Sonographic spiculation consists of straight lines that radiate perpendicularly from the surface of the mass. Typically, there are alternating hypoechoic and hyperechoic lines.

Angular margins, as defined by Stavros et al, are seen as an angular configuration of the junction between the relatively hypoechoic or isoechoic central part of the solid mass and the surrounding tissue. These angles can range from acute to obtuse. These have also been referred to as jagged or irregular margins. These should not be confused with lobulations, which are more smooth and rounded. In the study conducted by Stavros et al, angular margins had the greatest sensitivity and overall accuracy as a predictor of malignancy.

Microlobulations are the presence of many 1-mm to 2-mm lobulations on the surface of the solid nodule. They are similar to the mammographic equivalent findings. The risk of malignancy increases as the number of microlobulations increases. Microlobulations are often best visualized with antiradial scanning of the periphery of the mass. These microlobulations are likely to represent different patterns of tumor involvement at the margins, including small fingers of invasive cancer, cancerization of lobules, and intraductal extensions of tumor.

Figure 8. A 51-year-old female with infiltrating ductal carcinoma in the left breast previously documented by core biopsy. Lumpectomy was unsuccessful at an outside facility without preoperative needle localization. Postoperative high-frequency longitudinal and transverse ultrasound images with color Doppler demonstrate the feeding blood vessel coursing to the malignancy, which is separated from the postoperative cavity. Internal tumor vessels are also seen. Surgical excision after ultrasound-guided needle localization successfully removed a 7-mm infiltrating ductal carcinoma.

A ductal extension is a radially oriented projection arising from the malignancy along the axis oriented toward the nipple. This projection may be either within or around the milk duct. Occasionally, there can be a ductal bridge of tumor seen extending between two or more multifocal malignancies.

In contradistinction, Stavros et al defines the branch pattern as multiple extensions arising from the mass that are extending away from the nipple. Again, the projections may be either within or around the milk duct. The branch pattern represents advancement of tumor away from the nipple, whereas the ductal extension pattern represents advancement of tumor toward the nipple.

A solid breast mass that is taller than it is wide is suspicious for malignancy. If any part of the mass is longer in the anteroposterior dimension than in either the sagittal or transverse dimensions, it is reason to suspect malignancy. This orientation can be considered to represent malignancies having a predilection for growth toward the nipple. The normal tissue planes of the breast are horizontally oriented in patients who are  scanned in the supine position. The breast malignancy can be conceptualized as having the aggressive ability to overcome tissue planes and barriers and, therefore, to have a vertical orientation.

A solid lesion that is markedly hypoechoic is suspicious for malignancy. These masses are intensely black relative to the surrounding isoechoic fat. Malignancies can also be isoechoic and hyperechoic. Subtle, small hypoechoic and isoechoic malignancies can be detected sonographically with careful scrutiny and scanning. This is especially important in the detection of multifocal disease.

Shadowing posterior to a solid breast mass is another sonographic sign suspicious for malignancy. The desmoplastic response around the tumor is thought to attenuate the sonic beam more than the adjacent normal tissue. This should be considered present even if it is mild, or only found posterior to a portion of the mass. This should not be confused with edge or refractive shadowing, where shadowing occurs at the curved edge of a smooth, benign mass. This is related to the interface of the edge of the mass with the surrounding tissue, and may be found in benign and, occasionally, malignant breast masses. Shadowing is more commonly seen in low-grade to intermediate-grade tumors than in high-grade aggressive tumors. The lower-grade tumors grow slowly enough that the host can mount the desmoplastic reaction. Higher-grade tumors are more uniformly cellular, have associated lymphocytic infiltrates, and exhibit tumor necrosis, all of which may lead to increased through transmission, as opposed to shadowing; they are often more homogeneously hypoechoic, as well.

Punctate calcifications seen within a solid mass are more likely to be associated with a breast malignancy. Ultrasound is far less sensitive than mammography in the detection of breast microcalcifications. Calcifications seen using ultrasound are typically bright, punctate foci that do not create shadows because of their small size. Since normal breast glandular tissue includes a mixture of hyperechoic and heterogeneous fibrous tissue, benign calcifications are typically difficult to detect. Malignancies, however, are either intensely or mildly homogeneously hypoechoic solid masses, providing a background that enhances the ability of the imager to detect calcifications, especially with advanced techniques such as spatial compounding. Hence, while calcifications are not frequently seen, their detection in a hypoechoic mass is suspicious for malignancy by ultrasound criteria.

BENIGN SOLID BREAST MASSES

Marked and uniform hyperechogenicity, according to Stavros et al, probably represents normal fibrous change or focal fibrous change. In their study, this was the benign characteristic with the highest negative predictive value. If, however, there are areas of isoechogenicity or hypoechogenicity within this tissue that are larger than normal ducts or terminal ductal-lobular units, and are not entrapped fat lobules, the tissue should be considered indeterminate and biopsy should be recommended. In addition, the margins must be well defined. It is possible for a small malignant mass with a 4-mm to 6-mm central nidus to have a thick,  ill-defined hyperechoic halo.

Fibroadenomas tend to grow along the tissue planes of the breast. In a patient scanned in the supine position, the normal tissue planes of the breast are horizontally oriented (parallel to the pectoralis muscle and chest wall). Thus, fibroadenomas are usually horizontally oriented, and wider than they are tall. The flattened oval shape of a fibroadenoma during real-time scanning may also be a reflection of the greater compressibility of benign lesions with normal probe pressure.

A mild undulation in contour can be seen in solid benign masses such as fibroadenomas. It should be noted that the maximum number of lobulations allowed for benign solid masses is three. This should be distinguished during real-time scanning from microlobulations, which are smaller, sharper, and more numerous (and are common for malignant solid masses).

By definition, a thin, echogenic capsule is well circumscribed on both its inner and outer surfaces. It is usually a pseudocapsule of the compressed adjacent tissue. This implies that the mass is pushing against, as opposed to infiltrating, adjacent breast tissue. During real-time scanning, the capsule is best visualized at the orthogonal interface with the ultrasound beam, at both the anterior and posterior margins. The lateral edges of the capsule are least well seen because they are parallel to the ultrasound beam, unless a form of spatial compounding is used. The entire mass must be scanned with a systematic sweep to evaluate the mass and the capsule fully. The capsule is best demonstrated using high-frequency, broad-bandwidth probes that use shorter pulse lengths.

The criteria by Stavros et al for benign solid masses must be strictly applied. If any malignant characteristics are identified using ultrasound, a mass is considered suspicious for malignancy and is excluded from the benign classification. To be declared a benign mass, the mass must have no malignant features and must fulfill the conditions for one of three combinations of benign characteristics:

• intense and uniform hyperechogenicity;
• ellipsoid shape plus a thin, echogenic capsule; or
• two or three gentle lobulations plus a thin, echogenic capsule

If a solid breast mass does not have any of the malignant sonographic characteristics, but does not strictly meet one of three eligible combinations for a benign mass, it is considered indeterminate according to the Stavros criteria, and tissue sampling should be considered.

INTERVENTION AND IMPLANTS

Ultrasound has become the imaging modality of choice for the guidance of interventional breast procedures. Ultrasound-guided breast procedures are typically done with the patient in the comfortable supine position. A linear transducer permits long-axis visualization of a needle introduced into the sonographic abnormality, confirming accurate needle positioning in real time. In experienced hands, ultrasound-guided breast procedures, including cyst aspiration, core-needle biopsy, and preoperative needle localization, can be done rapidly (in 15 minutes or less per lesion). This efficient use of physician time translates into economic savings. If a rare vasovagal complication such as loss of consciousness arises, the patient is already supine for conservative management measures.

More than a million women in the United States have received silicone breast-augmentation implants.⁸ In 1992, a US Food and Drug Administration moratorium banned the routine use of silicone implants in the United States, except in controlled clinical trials.⁹ Many of the theorized systemic complications of silicone have since been challenged and refuted by multiple studies. This debate extends beyond the scope of this article; nevertheless, silicone gel freed into the breast may incite a localized inflammatory response and granuloma formation, with an ensuing mass effect. Plastic surgeons, therefore, frequently advocate removal of ruptured implants.

Silicone-implant ruptures may be divided into intracapsular and extracapsular categories. A silicone implant is composed of silicone gel contained within a silicone-polymer membrane. When a silicone implant is placed in a breast, the host treats it as a foreign body and contains it within a fibrous capsule. With disruption of the membrane, and when silicone gel is confined within the capsule, this is referred to as intracapsular leakage. Leakage of silicone outside the fibrous capsule into the surrounding breast tissues and beyond is called extracapsular rupture. Mammography is accurate in the evaluation of extracapsular rupture, but is limited in detection of intracapsular rupture. Breast ultrasound is more sensitive for the evaluation of extracapsular and intracapsular rupture than mammography.

Mammographic positioning of breast tissue can be limited and obscured by high-radiodensity augmentation implants. Positioning and breast compression may be particularly limited in patients with fibrous encapsulation (hardening of the fibrous capsule around the implants). Palpable masses and areas of focal pain may not always be visualized mammographically. We advocate the use of breast ultrasound in these clinical circumstances to evaluate the area of clinical concern, as well as the surrounding tissues and regional nodes.

EVALUTING PALPABLE MASSES

In women less than 30 years old, most masses are cysts that can be definitively evaluated by sonography, which can then guide aspiration or needle biopsy, if necessary. Mammography is not used as the initial modality of choice because of the low incidence of cancer in this age group, the prevalence and limitations of dense breast tissue seen in this age group, and the risk of unnecessary breast exposure to radiation. If ultrasound is nondiagnostic or inconclusive, mammography may be a valuable aid.

Women who are lactating often have dense breast tissue on mammograms because of engorgement of the milk ducts. This lowers the sensitivity of mammography. In addition, many breast masses are related to cysts, galactoceles, or abscesses in this setting. These can be visualized by sonography, which can then guide aspiration as needed.

Since ultrasound has no ionizing radiation, it is preferred as the initial imaging modality for evaluation of a lump in a pregnant patient. Many lumps in pregnant patients are cysts, which are readily visualized and diagnosed using ultrasound. If needed, ultrasound can guide cyst aspiration. If sonography demonstrates a malignant-appearing mass at the site of palpable concern, mammography with abdominal shielding or MRI is indicated. If sonography is normal or nonspecific at the area of concern, and there is persistent clinical suspicion, mammography with abdominal shielding may be obtained. The radiation exposure of the fetus is low with current film-screen mammography techniques.

According to the 2000-2001 ACR standard,¹ ultrasound is not currently indicated for screening studies for occult masses or calcifications. Nonetheless, some published studies^10-14 have shown that screening ultrasound can detect mammographically occult masses, particularly in women with mammographically dense breast tissue. These studies vary individually with type of equipment, scan time, and the operator. The published true-positive biopsy rate for screening breast ultrasound in these studies has been consistently lower than that of most published studies for mammography. Thus, while initial results are promising for screening ultrasound, further research is needed to help standardize the scanning technique, scan time, required training, and equipment standards, as well as to define an acceptable false-positive biopsy rate. Prospective studies are also needed to demonstrate reductions in breast-cancer mortality attributable to screening ultrasound.

COLOR DOPPLER BREAST ULTRASOUND

The Doppler effect refers to a change in the perceived frequency of sound emitted by a moving source. In many continuous-wave Doppler ultrasound systems, there are two piezoelectric crystals in the transducer.¹⁵ One crystal transmits an outgoing known sonic frequency, while the second crystal receives returning echoes and records the frequency. The Doppler shift can be conceptualized as the algebraic subtraction of the initially transmitted frequency from the returning frequency.

Color Doppler has been developed to produce imaging of blood flow throughout a chosen field. Color Doppler presents flow information in most contemporary ultrasound systems by superimposing a color image on the gray-scale real-time image. Thus, the operator has the opportunity to  assess anatomy and blood flow in the breast simultaneously. Detected flows of varying velocities are assigned specific colors, usually shades of red and blue, for motion to and from the transducers.

The role of color Doppler in breast ultrasound has not specifically been incorporated into the 2000-2001 ACR Standard for the Performance of Breast Ultrasound Examination,¹ yet in skilled hands it is a powerful tool to help evaluate breast abnormalities.

While not highly specific, color Doppler has different patterns for benign and malignant solid masses. Benign solid masses typically have more peripheral and circumferential flow patterns. Malignant masses typically have feeding vessels and, most important, prominent internal vascularity. The large, feeding blood vessels frequently seen are postulated to arise from tumor-associated angiogenesis.

High-resolution color Doppler ultrasound frequently demonstrates these feeding vessels surrounding a malignancy. The plane of intervention, under ultrasound guidance, can be chosen by the interventional radiologist to avoid these feeding vessels. This logically minimizes the risk of hematoma development from the procedure, particularly in large-gauge core-needle breast biopsy.

In approximately 2% of surgical excisions after preoperative needle localization of nonpalpable mammographic lesions, the surgical specimen will fail to contain the lesion.¹⁶ This may be secondary to inaccurate positioning or migration of the needle localization wire, surgical error, or loss of the lesion during pathological processing. In the immediate postoperative period, imaging to detect residual malignancy can be difficult. Mammography is compromised because the patient's postoperative breast tenderness limits compression. In addition, postoperative hematoma can limit visualization of the area. MRI is expensive, is not always available, and may require  a wait of several months to distinguish postoperative scarring from residual malignancy reliably. Ultrasound can be readily done with minimal discomfort to the patient, but can be technically challenging to interpret because of postoperative changes. Color Doppler is an occasionally valuable tool in the immediate postoperative period. Feeding blood vessels and the internal vascularity of malignancies can be seen with color Doppler, and this helps to identify residual malignancy. This then allows preoperative localization and excision of the area of concern.

CONCLUSION

Breast ultrasound is a critical component of a current, comprehensive breast-imaging center. Understanding the fundamental principles, advantages, and disadvantages of breast sonography is essential for the implementation of breast ultrasound in clinical practice. Equipment, technical, and personnel issues must be meticulously addressed when integrating breast ultrasound into a breast imaging center. The indications for breast ultrasound have increased over the past 2 decades and will probably continue to expand, given recent acceleration in clinical ultrasound research and advances in high-resolution ultrasound technology.

NOTE: For further reading recommendations, see the online version of this article at www.imagingeconomics.com.

Jay R. Parikh, MD, is medical director, Interventional Breast Imaging, Swedish Breast Care Centers/WDIC, Swedish Medical Center, Seattle. Bruce Porter, MD, is medical director, First Hill Diagnostic Imaging, Seattle.

FOR FURTHER READING

Angell M. Shattuck lecture-evaluating the health risks of breast implants: the interplay of medical science, the law, and public opinion. N Engl J Med. 1996;334:1513-1518.

Baker JA, Kornguth PJ, Soo MS, Walsh R, Mengoni P. Sonography of solid breast lesions: observer variability of lesion description and assessment. AJR Am J Roentgenol. 1999;172:1621-1625.

Bassett LW, Jackson VP, Jahan R, et al. Diagnosis of Diseases of the Breast. Philadelphia: WB Saunders; 1997.

Beekman WH, van Straalen WR, Hage JJ, Taets van Amerongen AH, Mulder JW.. Imaging signs and radiologists' jargon of ruptured implants. Plast Reconstr Surg. 1998;102:1281-1289.

Brown SL, Silverman BG, Berg WA. Rupture of silicone-gel breast implants: causes, sequelae, and diagnosis. Lancet. 1997;350:1531-1537.

Dennis MA, Parker SH, Klaus AJ, Stavros AT, Kaske TI, Clark SB. Breast biopsy avoidance: the value of normal mammograms and normal sonograms in the setting of a palpable lump. Radiology. 2001;219:186-191.

Feu J, Tresserra F, Fabregas R, et al. Metastatic breast carcinoma in axillary lymph nodes: in vitro US detection. Radiology. 1997;205:831-835.

Ford D, Easton DF, Bishop DT, Narod SA, Goldgar DE. Risk of cancer in BRCA1-mutation carriers. Lancet. 1994;343:692-695.

Garcia CJ, Espinoza A, Dinamarca V, et al. Breast US in children and adolescents. Radiographics. 2000;20:1605-1612.

Hoskins KF, Stopfer JE, Calsone KA. Assessment and counseling for women with a family history of breast cancer. JAMA. 1995;273:577-585.

Kaplan SS. Clinical utility of bilateral whole-breast US in the evaluation of women with dense breast tissue. Radiology. 2001;221:641-649.

Moon WK, Im JG, Koh YH, Noh DY, Park IA. US of mammographically detected clustered microcalcifications. Radiology. 2000;217:849-854.

Parikh JR, Cardenosa G, Coll D, Chadha D, Chilcote WA. Breast ultrasound. In: Taveras JM, Ferrucci JT, eds. Radiology. Philadelphia: JB Lippincott; 1999:92.

Rockhill B, Colditz GA. Making sense of breas

Jay R. Parikh, MD, is medical director, Interventional Breast Imaging, Swedish Breast Care Centers/WDIC, Swedish Medical Center, Seattle.

Bruce Porter, MD, is medical director, First Hill Diagnostic Imaging, Seattle.

References:

1. ACR Standard for the Performance of Breast Ultrasound Examination, 2000-2001 ACR Standards. Reston, Va: American College of Radiology; 389-392.
2. National Comprehensive Cancer Network Practice Guidelines: screening for and evaluation of suspicious breast lesions. NCCN Proceedings. Oncology. 1998;12(11A):89-138.
3. Seidman H, Gelb SK, Silverberg E, LaVerda N, Lubera JA. Survival experience in the Breast Cancer Detection Demonstration Project. CA Cancer J Clin. 1987;37:258-290.
4. Soo MS, Rosen EL, Baker JA, Vo TT, Boyd BA. Negative predictive value of sonography with mammography in patients with palpable breast lesions. AJR Am J Roentgenol. 2001;177:1167-1170.
5. . Stavros AT, Thickman D, Rapp CL, Dennis MA, Parker SH, Sisney GA. Solid breast nodules: use of sonography to distinguish between benign and malignant lesions. Radiology. 1995;196:123-134.
6. Sickles EA. Detection and diagnosis of breast cancer with mammography. Perspect Radiol. 1988;1:36-65
7. Skaane P, Engedal K. Analysis of sonographic features in the differentiation of fibroadenoma and invasive ductal carcinoma. AJR Am J Roentgenol. 1998;170:109-114.
8. Gerszten K, Gerszten PC. Silicone breast implants: an oncologic perspective. Oncology. 1998;12:1427-1433.
9. Kessler DA: The basis of the FDA's decision on breast implants. N Engl J Med. 1992;326:1713-1715.
10. Buchberger W, DeKoekkoek-Doll P, Springer P, Obrist P, Dunser M. Incidental findings on sonography of the breast: clinical significance and diagnostic workup. AJR Am J Roentgenol. 1999;173:921-927.
11. Kolb TM, Lichy J, Newhouse JH. Occult cancer in women with dense breast: detection with screening USâ€"diagnostic yield and tumor characteristics. Radiology. 1998;207:191-199.
12. Berg WA, Gilbreath PL. Multicentric and multifocal cancer: whole-breast US in preoperative evaluation. Radiology. 2000; 214:59-66.
13. Gordon PB, Goldenberg SL. Malignant breast masses detected only by ultrasound. Cancer. 1995;76:626-630.
14. Weinstein BJ, Phillips G, Kumari-Subaiya S, et al. Ultrasound detection of non-palpable, non-mammographically visible breast cancer [abstract]. J Ultrasound Med. 1996;15(suppl):117.
15. Curry TS, Dowdey JE, Murry RC. Christensen's Physics of Diagnostic Radiology. 4th ed. Philadelphia: Lea & Febiger; 1990:323-371.
16. Jackman RJ, Marzoni FA. Needle-localized breast biopsy: why do we fail? Radiology. 1997;204:677-684.