Figure 1. Simple cyst. ultrasound showed a circumscribed, thin-walled, anechoic
structure with increased through transmission of sound.Radiologic Clinics of North America
Volume
39 • Number 3 • May 2001
Äëÿ ó÷àñòíèêîâ áåñåäû î ñêðèíèíãå ðàêà ìîëî÷íîé æåëåçû.
Ultrasound (US) is an indispensable tool in the breast imaging armamentarium. It is easily the most valuable and readily used adjunct to mammography. Wild and Neal[78] first described US imaging of the breast in 1951. The evolution of this powerful technique continues, as do controversies over its appropriate use.[31] No longer used only to differentiate masses as cystic versus solid, US today contributes significantly to the characterization of solid masses through feature analysis. It is also used extensively in image-guided interventional procedures. Research continues in the application of Doppler in the evaluation of breast masses and in the possible use of US as a screening tool in some patient populations. This article reviews these and other indications for breast US.
The importance of proper technique in this highly operator-dependent modality has been emphasized.[4] [8] [12] [23] Early experience with water-bath and whole-breast scanners has evolved to the use of hand-held devices. Transducers should be in the range of 7.5 MHz and higher. Linear array transducers are preferred because of their wider near field. Offset pads are sometimes required to image superficial lesions. The time gain compensation curve compensates for the attenuation of sound waves in tissue. For example, the rapid attenuation in very dense tissue requires a steep time gain compensation curve. Power and focal zone must be optimized. The depth of focus is usually set at 3 cm or less. All settings should be adjusted for each patient to ensure that fat looks gray, not dark.
Proper positioning can minimize the depth of tissue that must be penetrated during the US examination. The patient should be placed supine with elevation of the ipsilateral hemithorax by a support (pillow or wedge). The ipsilateral arm should be raised above the head to decrease breast thickness, particularly in the upper outer quadrant. Specialized techniques may be required to image the nipple-areolar complex because it often causes posterior acoustic shadowing. This is related to abrupt changes in the surface contour in this area and to the fibrous elements within nipple. It is often helpful to place the transducer slightly off center to the nipple and then use angulation to image the retroareolar tissues.
If a question exists as to whether an US-detected lesion corresponds to a mammographic abnormality, a radiopaque marker may be placed during the US examination and a follow-up mammographic view obtained. Because of different positioning between the US and mammogram, attention should be paid to proper placement of the metallic marker.[37]
Investigations into the use of Doppler US examinations of breast lesions have produced mixed results.[5] [9] [13] [21] [22] [37] [51] [59] [79] Doppler adds significantly to cost and continues to have questioned efficacy. Because Doppler technique relies on increased vascularity (which can be seen in both benign and malignant lesions), significant overlap of positive findings is not unexpected.
Dock[21] studied 75 solid breast neoplasms using duplex US to establish the optimal threshold value of systolic peak flow velocity to differentiate benign from malignant tumors. This series showed that only positive results were potentially reliable. Negative findings were unhelpful.
Studies of color Doppler have produced similar results. As color Doppler equipment has become more and more sensitive, vascular signals can been found in normal, benign, and malignant tissue.[13] [79] Wilkens et al[79] found that color Doppler US yielded no additional information beyond that of gray-scale US in discriminating between benign and malignant solid breast nodules. This study suggested several potential causes of false-positive results (inflammatory lesions, recent fine-needle aspiration [FNA], and some vascular fibroadenomas) and false-negative results (poorly vascularized tumors, insensitive Doppler parameters, and incidental cancers) on color Doppler US.
Reports of the evaluation of power Doppler imaging have yielded similar results. Raza and Baum[59] specifically investigated power Doppler imaging of tumor vascularity by focusing on the morphology and pattern of the distribution of vessels within a mass rather than performing quantitative analysis. By using penetrating vessels to indicate malignancy, the sensitivity of power Doppler imaging was 68% and specificity was 95% in a series of 86 solid breast masses (25 cancers, 38 fibroadenomas, and 23 other benign lesions). Birdwell et al[5] analyzed the extent of flow on power Doppler imaging as estimated as a percentage of the lesion on multiple longitudinal and transverse static US images in 69 biopsy proved solid masses. Interestingly, of the avascular lesions, nine were malignant and eight were benign. Significant overlap was seen in the vascularity of the remaining 52 lesions with both malignant and benign lesions showing a similar extent of vascularity. This finding suggests that power Doppler imaging is of limited value in the diagnostic evaluation of solid breast masses.
Ultrasound has the ability to image many normal anatomic structures including the skin, fibroglandular parenchyma, fat lobules, the nipple-areolar complex, milk ducts, blood vessels, and intramammary lymph nodes. The skin, which is normally 1 to 3 mm thick, appears as two parallel echogenic lines. The normal parenchymal and glandular tissue are echogenic or have mixed echogenicity. Fat lobules in the breast have a hypoechoic appearance as opposed to fat seen elsewhere in the body that is usually hyperechoic. Fat lobules typically appear ovoid or flattened because they are easily compressed. Careful examination in both transverse and sagittal planes often suggests a somewhat tubular shape following fascial planes in the breast. Although in transverse section a fat lobule can appear as a hypoechoic mass containing a central echogenic nidus and surrounded by a thin connective tissue border, the sagittal plane of the same lobule shows smooth tapering toward the ends of the long axis. Milk ducts appear as branching, hypoechoic, tubular structures that stream toward the nipple as they increase in size near the lactiferous sinuses. Pectoral muscles can be seen deep to retromammary fat and are sharply demarcated by their echogenic fascia. Costochondral cartilages appear as hypoechoic oval masses deep to the muscle and calcified ribs show prominent shadowing. Beneath the thoracic wall, a linear echogenic interface can be seen that includes chest wall, pleura, and lung border. Anatomic structures that can be seen in axilla include blood vessels, lymph nodes, and the proximal portions of the pectoral muscles.
Benign breast cysts are common mass lesions occurring in the female breast constituting 25% of all palpable or mammographically detected lesions.[36] Although distinguishing a cyst from a solid lesion is not possible mammographically, US is extremely useful for this differentiation. On mammography, a cyst is usually a low- to medium-density circumscribed round mass. The typical ultrasonographic appearance of a cyst is a circumscribed, thin-walled, anechoic structure with increased through transmission of sound (Fig. 1) . US is 96% to 100% accurate in the diagnosis of cyst if strict criteria are applied.[36] [43] [64] Such criteria include round or oval lesions, lack of internal echoes, sharp margins, and posterior acoustic enhancement. Cysts may also show deformability and refractive lateral wall shadowing. It is important to note that US does not show all lesions. A mammographic lesion that cannot be seen on US should be presumed solid and its management must be based on the mammographic findings. Any decision to obtain a biopsy or follow a palpable mass that is inapparent on US and mammography must be based on the clinical assessment.
Figure 1. Simple cyst. ultrasound showed a circumscribed, thin-walled, anechoic
structure with increased through transmission of sound.
Some breast cysts may appear complicated because of proteinaceous material, blood, cellular debris, infection, or cholesterol crystals. This appearance occasionally can make it difficult to differentiate a complex or debris-filled cyst from a solid lesion. Image-guided aspiration of nonpalpable possible breast cysts may obviate the need for surgical biopsy or short-term mammographic follow-up in this group of patients.[55] A review by Venta et al[76] of 4562 breast USs performed during an 18-month period showed 308 complex cysts in 252 women. The malignancy rate of the lesions classified as complex cysts in this study was 0.3% (1 per 308). The single malignancy detected was a papilloma with a 3-mm focus of ductal carcinoma-in-situ (DCIS). Smith et al[68] reported a series of 660 breast cysts that were evaluated by aspiration cytology. Ninety-five percent of the breast cyst fluid samples showed either no malignant cells or were considered nondiagnostic (because of insufficient cellular material for analysis). The cytologic examination of cyst fluid in the remaining 33 (5%) patient series showed atypia. None of these 33 lesions were found to represent malignancy on surgical excision (N = 9) or clinical follow-up (N = 24). For most accurate diagnosis, imaging before instrumentation is recommended. Imaging cysts after an aspiration attempt can be misleading because changes related to the procedure (hematoma, trauma, or edema) can appear suspicious for malignancy.
Although US was initially limited to differentiating palpable and mammographic lesions as either cystic or solid, recent investigations argue for broadening the role of US to include characterization of some solid nodules. A landmark study by Stavros et al[70] attempted to validate a set of US features that have a very low risk of carcinoma. If such a group of lesions could be reliably identified, they would be similar to the "probably benign" mammography category used by Sickles.[63] This probably benign category showed a frequency of carcinoma less than 2%. Stavros et al[70] prospectively classified 750 solid nodules based on US features. Malignant features included spiculation, angular margins, marked hypoechogenicity, shadowing, calcifications, duct extension, branch pattern, and microlobulation. As described by Fornage et al,[24] nodules that are "taller than wide" are more likely to be malignant. This finding is particularly worrisome because it implies growth across normal tissue planes (oriented horizontally in a supine patient) (Fig. 2) . Benign lesions tend to remain within tissue planes; they are usually wider than they are tall. Features that were associated with a very low risk of carcinoma were (1) an intensely hyperechoic mass, (2) an ellipsoid shape with a thin echogenic capsule that was considered to be a pseudocapsule of compressed adjacent normal breast tissue (Fig. 3) , (3) two or three gentle well-circumscribed lobulations plus a thin echogenic capsule (Fig. 4) , and (4) absence of any malignant findings. Indeterminate features included maximum diameter, isoechogenicity, mild hypoechogenicity, normal sound transmission, enhanced transmission, heterogeneous texture, and homogeneous texture. Posterior enhancement was an unreliable feature (Fig. 5) . No malignant tumors were found to be hyperechoic, a finding that has been supported by others.[24] [53] If even a single malignant feature is present, the lesion cannot be classified as benign. In this series reported by Stavros et al[70] only two of 424 prospectively classified masses with benign features were malignant (0.5%). Seventy-eight percent of masses prospectively classified with malignant features were malignant, whereas 12.3% of masses classified as indeterminate were malignant.
Figure 2. Invasive ductal carcinoma. ultrasound showed a microlobulated, hypoechoic mass that is
taller than wide. There is prominent posterior acoustic
shadowing.
Figure 3. Fibroadenoma showing an echogenic pseudocapsule
(arrow).
Figure 4. Fibroadenoma showing gentle well-circumscribed
lobulation (arrow).
Figure 5. Invasive ductal carcinoma that enhanced through
transmission.
Although this classification system has been validated,[65] others have found that there can be significant practice and interpreter variability in differentiating benign from malignant solid breast masses.[58] [66] Radiologists can differ substantially in the interpretation of breast US particularly when mammographically obvious cancers are excluded from the group being analyzed. There continues to be a lack of uniformity among observers' use of descriptive terms for solid breast masses that results in inconsistent diagnoses.[3] Combined readings (mammography and US) offer the highest diagnostic accuracy. If a probably benign lesion is recommended for follow-up rather than tissue sampling, a complete mammographic work-up (including spot compression views with or without magnifications) should be performed to make sure there are no suspicious mammographic findings. State-of-the-art, optimized US images should be obtained. Usually short-interval (6 months) follow-up is conducted for 2 full years.
ultrasound is the primary study for patients younger than 30 years who present with a palpable breast mass. This limits the amount of ionizing radiation in this population with a low prevalence of breast cancer. If US shows a simple cyst, no further imaging is indicated. Should a solid mass be detected, mammographic images may be helpful in further evaluating the lesion. The presence of suspicious calcifications or other suspicious mammographic findings argues for biopsy. Benign mammographic findings (coarse calcifications suggestive of fibroadenoma or areas of fat within the lesion suggestive of a hamartoma) may obviate tissue sampling. Newly discovered palpable solid masses are considered to be growing in size and are typically recommended to undergo biopsy.
ultrasound is used extensively for imaging guidance for interventional procedures. * As in cyst aspirations, needle-wire localizations can be performed with direct US visualization.[10] After the surgical excision, ultrasound should be used to image the biopsy specimen to ensure that the targeted lesion has been removed.[25]
Fine-needle aspiration biopsy has been studied as a method of diagnosing nonpalpable lesions.[7] [27] [41] A series of 1885 US-guided FNA of nonpalpable breast lesions was reported in 1999.[7] A total of 480 lesions were shown to be benign cysts. Of the remaining 1405 lesions, 41% were diagnosed as benign and 36% were diagnosed as malignant after FNA. The remaining 23% of these 1405 lesions were deemed atypical, suspicious, or nondiagnostic. Follow-up showed a malignancy rate of 3.7% in lesions diagnosed as benign, 99% in lesions diagnosed as malignant, 53% in lesions diagnosed as atypical, 76% in lesions diagnosed as suspicious, and 34% in lesions reported as nondiagnostic. The widespread use of this technique has been limited by several factors. Insufficient specimens (particularly in benign lesions), lack of adequate training in aspiration techniques, and a paucity of dedicated breast cytopathologists pose important problems for FNA biopsy.[27]
Ultrasound-guided 14-gauge core breast biopsy was first described in a series of 181 patients by Parker et al.[57] In this report, 49 of these 181 lesions were eventually excised (including 34 malignancies) and showed complete agreement between pathology from the surgical excision and pathology from the large-core needle biopsy. No cancers were found during follow-up (12 to 36 months) of the unexcised benign lesions. This technique (Fig. 6) is typically easier for patients and faster than large-core needle biopsy using stereotactic guidance. Chare et al[11] described a series of 125 lesions sampled by 16-gauge US large-core needle biopsy. A definite decision regarding patient management could be made in only 88 (70%) lesions in this series. The remaining 37 patients underwent surgical excision and 12 additional malignancies were found. The lack of adequate tissue sampling in this series may have been related to the smaller (16-gauge) needle used. Liberman et al[48] pointed out that cost savings for US-guided large-core needle biopsy should exceed those with stereotactic biopsy. The equipment is less expensive and multifunctional US machines are readily available in most radiology departments. Core biopsy has a number of advantages over FNA in the evaluation of breast masses. These include lower inadequacy and equivocal rates and the ability to make a definitive diagnosis of a fibroadenoma. In addition, a core result of normal breast tissue alerts the clinician to the possibility of sampling error and the need for repeat biopsy.
Figure 6. ultrasound (US)
guided biopsy. A, Prefire image from 14-gauge US-guided large core needle
biopsy of a 6-mm solid mass shows needle placement under direct US
visualization. B, Postfire image shows the needle traversing the lesion.
Pathology showed invasive lobular carcinoma.
DeBiose et al[16] reported that US can be helpful in planning radiation treatment. They found that US seems to be a more accurate means of identifying the full extent of the lumpectomy cavity when compared with clinical estimates (physical examination, operative report, mammographic information, and location of the surgical scar).
Breast infections usually are diagnosed and treated clinically. Should an abscess be suspected, however, US is helpful in identifying a drainable fluid collection. Often pain prohibits these patients from undergoing mammography. The US appearance of an abscess is usually a heterogeneous hypoechoic mass with a thick echogenic wall (Fig. 7) . Surrounding breast parenchyma can often be hyperechoic secondary to inflammation and edema.
Figure 7. Breast abscess in a lactating woman. ultrasound shows an irregular hypoechoic fluid collection
surrounded by echogenic edematous breast tissue.
Perhaps one of the most controversial uses of breast US is to screen asymptomatic patients. Reports in the literature so far have produced conflicting results. Maestro et al[49] reported a series of 350 bilateral US examinations in 350 asymptomatic women with dense breasts. US examinations were read as normal in 53% of the patients and abnormal in 47%. The abnormal category included 117 cysts, 44 solid nodules, and four mixed echogenicity nodules. All but two solid nodules were benign (both of which were invasive carcinomas that in retrospect were palpable). In the benign category, US falsely diagnosed four fibroadenomas as cancers. The largest series reported so far by Kolb et al[45] described screening US performed in 3626 women with normal mammographic and physical examinations. The study group was limited to women with BIRADS density grades of 2 to 4 (any breast density other than a breast that is almost entirely fat). These authors reported an average scanning time of 3 minutes and 59 seconds per examination. Among these 3626 women, 215 solid masses were detected by the screening US procedure. FNA or surgical biopsy was eventually used to sample 123 of these 215 solid masses. Eleven additional cancers were discovered. The authors caution that "before widespread implementation of screening US, the detection of these additional cancers must be weighed against the additional cost and false-positive rate of screening US, and a demonstration of mortality rate reduction is necessary."[45] A consensus statement by the European Group for Breast Cancer Screening[73] reads "The use of US in population screening of asymptomatic women is associated with unacceptably high rates of both false-positive and false-negative outcomes. At present there is little evidence to support the use of US in population breast cancer screening at any age."[73] Additional work with large patient populations is needed before widespread use of screening US in asymptomatic patients can be justified.
Several studies have examined the ability of US to detect suspicious microcalcifications.[2] [46] [81] It seems that the calcifications that occur within breast cancers are more likely to be seen in part because most malignant solid nodules provide a hypoechoic background, which enhances the US demonstration of the bright punctate calcification echoes. In contrast, benign calcifications within normal breast tissue (comprising hyperechoic and heterogeneous fibrous tissue) are less reliably detected. It is unlikely that US will be able to perform better than mammography in the detection of microcalcifications in DCIS lesions that do not have an invasive component. Currently, there are no in vivo studies in the literature directly comparing the ability of mammography and US to detect microcalcifications in the size range of clinical interest.
Although less well proved, US can be helpful in evaluating a potential mass seen only in one mammographic projection. This should not be used as an alternative to high-quality mammographic positioning. It should also be noted that the absence of an ultrasonographic abnormality does not rule out a suspicious lesion.
US has been used extensively in the evaluation of breast implants for intracapsular and extracapsular rupture.[17] [29] [32] [56] [75] A variety of linear echoes can be seen in most silicone breast implants on gray-scale US. Unfortunately, the presence or absence of linear echoes is not useful in predicting implant rupture because radial folds of the lumen (a normal finding) can appear similar to a collapsed internal capsule. Complete absence of internal echoes, although highly predictive of an intact implant, is infrequently seen. DeBruhl et al[17] described a stepladder sign, which refers to multiple, discontinuous, parallel linear or curvilinear echoes traversing the lumen of the implant. This finding corresponds to the linguine sign[29] on MR imaging that can be seen when the internal capsule is no longer intact. A false-positive stepladder sign can be caused by envelope folds or by infolding sometimes seen in multiple lumen implants. Such folds or septations should be in continuity with the implant envelope. Extracapsular rupture of free silicone has a typical snowstorm appearance.[32] The term snowstorm describes the intense echogenic pattern of scattered and reverberating echoes in the soft tissues surrounding a ruptured breast implant with a well-defined anterior margin but loss of detail posterior to the echogenic area (Fig. 8) . Although this sign can differentiate true extracapsular rupture from herniation of an intact implant through the fibrous capsule, the ultrasonographer must be aware of any prior surgery. Free extracapsular silicone left behind during previous implant surgery may not reflect rupture of the current implant. Unfortunately, sensitivity and specificity of US features of implant rupture are not sufficient to recommend it as the definitive study. A series of 236 implants with surgical correlation were reported in 1996 by Venta et al.[75] US had a sensitivity of 50% and specificity of 55% in the detection of breast implant rupture. A meta-analysis by Goodman et al[28] in 1998 that included studies with a total of 1099 breast implants over 7 years showed only minimally better results (sensitivity 59% and specificity 77%). MR imaging is emerging as the gold standard for the evaluation of implants.[29] This technique offers a non-operator-dependent global view of the entire implant and surrounding tissues in multiple planes. Additionally, specialized fat and silicone saturation techniques can be used. MR imaging can easily detect extracapsular rupture. MR imaging is also more accurate in the detection of intracapsular rupture than mammography or US, but is also more costly.
Figure 8. Extracapsular rupture. ultrasound shows the typical snowstorm appearance of free
silicone with breast tissue (arrow).
The classic appearance of invasive ductal carcinoma on breast US is a hypoechoic mass with irregular margins (Fig. 9) . Other malignant features include spiculation (Fig. 10) , angular margins, calcifications, duct extension, branch pattern, and microlobulation (Fig. 11) . As mentioned previously, carcinomas tend to be "taller than wide" because of their invasive tendency to cross fascial planes, whereas benign lesions tend to displace adjacent normal breast tissue. Posterior acoustic shadowing and maximum diameter should both be considered as indeterminate findings.[70]
Figure 9. Invasive ductal carcinoma. ultrasound shows a microlobulated hypoechoic
mass.
Figure 10. Invasive ductal carcinoma. ultrasound shows a hypoechoic mass with prominent
spiculations (arrows).
Figure 11. Invasive ductal carcinoma. Screening mammography detected a nonpalpable area of increased
density. ultrasound showed a microlobulated 2-cm
solid mass. Excisional biopsy showed invasive ductal
carcinoma.
The second most commonly occurring histologic subtype, invasive lobular carcinoma (ILC), has a similar appearance on US. Butler et al[10] reviewed US images from 81 biopsy proved invasive lobular carcinomas. In 61% of this series, the invasive lobular carcinoma was a heterogeneous hypoechoic mass with angular or ill-defined margins and posterior acoustic shadowing. Fifteen percent, however, showed only focal shadowing without a discrete mass (Fig. 12) ; 12% appeared as a lobulated, well-circumscribed mass; and 12% were not seen by US.
Figure 12. Screening mammography detected a new 5-mm upper outer quadrant mass.
ultrasound shows prominent posterior acoustic
shadowing. Surgical excision after ultrasound-guided wire localization showed
invasive lobular carcinoma.
Tubular carcinomas are usually seen on mammography as irregularly shaped masses with spiculated margins. The US appearance of a hypoechoic mass with ill-defined margins is identical to invasive ductal carcinoma.[62]
Medullary carcinoma of the breast is an uncommon tumor that may mimic a benign mass at both mammography and US. These lesions have been reported as well-defined masses with an heterogeneous, hypoechoic texture that do not attenuate sound, but rather show posterior acoustic enhancement.[54]
Ductal carcinoma-in-situ is a common lesion that typically presents as microcalcifications that are detected mammographically. In approximately 10% of cases, however,[71] this lesion presents as a noncalcified soft tissue mass. US of this subset of DCIS lesions has a variable appearance that ranges from solid masses that are circumscribed, round or oval, to heterogeneous masses with irregular contours.[20]
Intracystic carcinoma is a rare lesion. Czernobilsky[15] found the incidence to be 0.5% in 2500 breast cancers. It is likely that the cancers reported in other large series included not only intracystic carcinoma, but also instances when a solid tumor invaded an adjacent benign cyst and solid tumors that had undergone central cystic necrosis.[1] [14] [44] [51] US can be helpful in identifying this rare lesion (Fig. 13) , which can appear as a mural mass within a cystic space.[59]
Figure 13. Intracystic carcinoma. ultrasound shows cystic lesion containing a large mural
mass.
Hematogenous metastases are usually seen as single or multiple well-circumscribed hypoechoic masses.[80] Similar to the appearance elsewhere in the body, the ultrasound appearance of a lymphomatous mass is typically hypoechoic. Paralleling their mammographic appearance, these lesions can vary from nodules with well-circumscribed borders to masses with ill-defined margins.[19] [80]
A fibroadenoma is the most common benign solid mass of the breast and accounts for up to 40% to 50% of all breast biopsies.[30] The classic appearance is an oval, wider than tall hypoechoic mass with an echogenic pseudocapsule (see Fig. 3) . Other variations (Fig. 14) from the classic appearance have been reported.[42] In the study by Jackson et al,[42] only 15% of the fibroadenomas had the classic US appearance of a smooth round or oval mass with homogeneous internal echoes. Eighteen percent were not visible on the US. Most (66%) of lesions in this study had at least one or more atypical signs including border irregularity, lobulations, inhomogeneous internal echo texture, or posterior shadowing. Eleven percent of lesions were misdiagnosed as fibroadenoma, and included five benign lesions and four carcinomas. This study argues for the histologic evaluation of all solid masses. As mentioned previously, the posterior acoustic appearance is an indeterminate finding.[70]
Figure 14. Atypical
fibroadenoma. ultrasound shows a hypoechoic solid mass containing
multiple foci of calcification. Excisional biopsy showed fibroadenoma with
calcifications.
Phyllodes can appear similar to fibroadenoma on US.[47] Liberman et al[47] reported the US findings in 30 phyllodes. Both benign and malignant subtypes typically appeared as hypoechoic masses. Aside from tumor size (greater than 3 cm) there were no US features that indicated a benign or malignant histologic nature. Cystic areas were detected in both benign and malignant phyllodes tumors.
Hamartomas also have a variable appearance on US (Fig. 15) . Black et al[6] described a series of nine patients with mammographically detected hamartomas. Two lesions could not be identified by US. Six were well-circumscribed masses of heterogeneous echogenicity with no posterior acoustic shadowing. One lesion was a well-defined hypoechoic mass with no posterior shadowing. The mammographic appearance as a lesion containing fat should preclude biopsy.
Figure 15. Nonspecific appearance of a
hamartoma. ultrasound shows a 3-cm heterogeneous mass. mammography showed a circumscribed mass containing fat
consistent with a hamartoma.
Fibrous nodules have a variable appearance on mammography.[35] In the study by Harvey et al,[35] 40% manifested as circumscribed masses, 33% as masses with indistinct margins, and 27% had imaging findings suggestive of malignancy. Venta et al[74] reported similar findings on US for 36 fibrous nodules with 58% of nodules showing circumscribed margins, 28% showing lobulated margins, and 14% with ill-defined margins.
Fat necrosis can present as a discrete mass.[69] The report by Soo et al [69] described the US appearance of 31 fat necrosis lesions. Fifteen masses were solid, 11 were complex nodules, two were anechoic with posterior acoustic enhancement, two were anechoic with posterior acoustic shadowing, and in one case no mass was visible. The US appearance of fat necrosis can evolve, often becoming more solid over time. During follow-up no mass increased in size, and four decreased in size. Harvey et al[34] reported the US appearance of 26 oil cysts. In this series, only 8% appeared similar to simple apocrine cysts. Twelve percent showed an intracystic mass. Most had smooth walls (88%); were hypoechoic (65%); and showed neither posterior acoustic enhancement nor shadowing (50%). Recognizing the typical mammographic appearance of these lesions as a radiolucent mass surrounded by a thin membrane can avoid unnecessary biopsy.
Lumpectomy scars can have a US appearance similar to invasive carcinomas (irregular hypoechoic mass with posterior acoustic shadowing). Close correlation with the mammographic findings (including presurgical and postsurgical imaging) is needed for complete evaluation.
Epidermal inclusion cysts (sometimes referred to as sebaceous cysts) are dermal lesions that can present as palpable breast lumps. The most common mammographic finding is an isodense to high-density mass with a circumscribed margin. US typically shows a superficial circumscribed hypoechoic mass with through transmission. Recognizing extension of the mass into the dermis can help make the correct diagnosis.[18]
Lactating adenomas can present as painless new breast masses in pregnant or postpartum patients. The differential diagnosis includes galactocele, fibroadenoma, and cancer. US features of lactating adenomas are generally benign (wider than tall, well-defined margins, homogenous echotexture, and posterior acoustic enhancement).[80] As reported by Sumkin et al,[72] however, four lesions had US features that resembled malignancy, with irregular margins, heterogeneous echotexture, and posterior acoustic shadowing. Galactoceles typically appear similar to complex cysts as oval lesions with multiple internal echoes.
The US appearance of hematomas reflects their natural evolution.[77] They are usually spherical or ovoid and generally decrease over time. They can have irregular walls and contain a variable amount of internal echoes initially, although they usually become anechoic (Fig. 16) . The US appearance is not specific and can be similar to an abscess.
Figure 16.
Hematoma. ultrasound of palpable mass at the site of prior lumpectomy
shows a complex cystic lesion consistent with postoperative seroma/hematoma.
This lesion is unchanged at 5-year follow-up.
ultrasound is second in importance only to mammography in the evaluation of breast abnormalities. The ability of US to diagnose benign simple cysts accurately has spared innumerable women from needless biopsies and aspirations. When a suspicious solid abnormality is detected by ultrasound, it readily provides convenient, inexpensive, accurate imaging guidance for interventional procedures. Future studies of US as a screening tool must weigh potential benefits against the potential risks because of increased biopsies performed for false-positive results.
The author thanks Sharare Soleymani for her assistance in the preparation of this manuscript.
Назад к Русскому Медицинскому Серверу
К медицинскому сайту Участкового