Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Benign findings in the breast are common. Most findings recalled from screening and even the majority of lesions that undergo biopsy are ultimately found to be benign. Findings that require additional imaging assessment and biopsy are stressful for patients and add costs to the health care system. These cases contribute to what is termed false positives—recalls or biopsies yielding clinically insignificant information—and are a commonly used metric to argue against mammographic screening. Of course, it is only retrospective analysis that defines these as false positives. Better prospective imaging to distinguish between benign and malignant lesions could reduce false-positive work-ups and patient anxiety.
Even after a thorough diagnostic evaluation including spot compression and/or magnification views, the complete assessment of a lesion with two-dimensional (2D) mammography can still remain limited. Overlapping tissue and obscured margins may limit the accuracy of the assessment. Although ultrasound is almost always used in conjunction with mammography for evaluation of masses or densities, it is the combined information from both modalities that ultimately leads to a recommendation to dismiss, follow, or biopsy a lesion. If any findings are concerning, a recommendation for tissue sampling will commonly be made. With improved visualization of lesion shape and margins provided by tomosynthesis, along with a better assessment of multiplicity of lesions and distribution of findings, one can potentially designate more lesions as benign versus those that are suspicious enough to require additional attention, reducing unnecessary imaging follow-ups or biopsies. Tomosynthesis provides the opportunity to reduce false positives and increase the positive predictive value of biopsy.
However, tomosynthesis can also present a challenge in that many more lesions previously undetected on 2D mammography may become more apparent. A benign mass potentially present for many years may be obvious on tomosynthesis, though not detected on prior 2D mammograms. Knowing how to manage a multitude of additional findings can be difficult at first. Close scrutiny of prior mammograms and breast ultrasound images may indeed suggest stability. Even if the margins were not discernable on prior mammograms, it may be possible that the tissue pattern suggests stability and therefore the newly seen finding on tomosynthesis can be safely followed. Certainly, increasing recalls for such additional benign findings is not desired. Analogous to ditzels or UBOs (unidentified bright objects) on magnetic resonance imaging (MRI) or small benign-appearing oval masses on screening ultrasound, many such benign-appearing lesions ultimately have to be disregarded. Learning over time to dismiss these benign findings seen only on tomosynthesis takes confidence and experience to achieve.
Many findings can be assessed thoroughly enough with the routine craniocaudal (CC) and mediolateral oblique (MLO) screening tomosynthesis views such that they do not require recall, including skin lesions or dermal calcifications, looped vessels, or other clearly benign findings on screening tomosynthesis exams. Such cases are discussed in Chapter 5 . This chapter concentrates on additional benign findings, some of which may still require recall for more thorough assessment, and discusses their imaging characterization and management.
Like in all aspects of breast imaging, when a finding is encountered on tomosynthesis, Breast Imaging Reporting and Data System (BI-RADS) descriptors help to guide interpretation toward a benign, probably benign, or suspicious assessment. Oval shape and circumscribed margins are typical characteristics of benign masses. In many cases, screening mammography with tomosynthesis provides sufficient information on shape and margins that can be used in place of common diagnostic views.
Benign masses often have circumscribed margins. Although a small percentage of malignant masses are characterized as circumscribed on 2D imaging, tomosynthesis may depict subtle irregular or spiculated margins that were previously unrecognized. The complete assessment of margins of a mass requires evaluation of all tomosynthesis slices in which the mass is visualized. A circumscribed margin on a single 1-mm thick tomosynthesis image does not necessarily mean that the mass is circumscribed in its entirety and may lead to an inaccurate assessment. There will still be a small percentage of cancers that remain circumscribed-appearing, even on tomosynthesis. Use of all information, such as interval growth, dominant lesion, age of patient, and ultrasound appearance, needs to be taken into account before determining management. Nevertheless, the vast majority of lesions with circumscribed margins will be benign, and tomosynthesis will provide a better differentiation of benign and malignant masses compared with 2D imaging alone.
Multiplicity of bilateral similar findings is indicative of benign processes on 2D mammography and screening breast ultrasound. Multiple circumscribed masses can be regarded as likely representing cysts, fibroadenomas, or other benign etiology, and recall is unnecessary. Although tomosynthesis can initially present a challenge when additional masses are detected that were not seen on 2D imaging, it can also help in some cases to demonstrate multiplicity and bilaterality of findings that suggest benign etiology. Of course, in the case of multiple bilateral findings, any lesion that is dominant or different in terms of shape, margin, or density should be considered for recall unless it has been demonstrated to be stable or previously shown to represent a benign finding, such as a cyst.
The majority of mammographic asymmetries represent overlapping parenchymal tissue. After implementing tomosynthesis, asymmetries account for a smaller percentage of recalled abnormalities at screening mammography. Most asymmetries seen on 2D mammography can be dismissed after reviewing tomosynthesis images. Tomosynthesis most frequently shows that an asymmetry represents overlapping parenchymal tissue and may also reveal a benign cause of the asymmetry, such as a looped vessel or skin lesion. There are some asymmetries, particularly focal asymmetries, that may remain questionable on routine tomosynthesis images, especially if the asymmetry is new or more prominent than prior exams. Sometimes the 2D component is sufficiently concerning that even if tomosynthesis views do not demonstrate a definitive finding, a recall is still generated. Particularly in areas of dense tissue, even tomosynthesis cannot always definitively resolve whether or not a focal mass exists.
Cases of one-view asymmetries are far less common with tomosynthesis than with 2D mammography. Appropriate use of scrolling through the tomosynthesis images can usually determine if a true lesion exists or if the finding is more likely superimposed tissue. If real, one should be able to determine where in the breast it is located. Such findings can usually be located upon review of the corresponding area on the orthogonal tomosynthesis view and can help to further determine if the finding simply represents tissue, especially if the area is seen to be stable, versus a focal asymmetry or mass. Areas of questioned asymmetries in denser areas will be more difficult to assess and frequently require ultrasound for further assessment.
Diagnostic work-up for a patient recalled from screening may include spot compression views. Importantly, spot compression views can be performed as combination images (with both 2D and tomosynthesis image acquisition), as tomosynthesis-only images, or as conventional 2D image only. When performing spot compression views with tomosynthesis, the 2D component is helpful to assess if the appropriate location was “spotted” and if the asymmetry persists. These combined spot compression views are especially useful when first learning tomosynthesis interpretation because the familiar 2D images are obtained along with the tomosynthesis views. If available, synthesized mammography can be used to reduce the total radiation exposure. A word of warning: malignant lesions can efface. This has been observed with 2D mammography but can also happen with tomosynthesis. Spot compression views are helpful when they show a definitive lesion, allowing for better feature and location characterization; however, uncertainty can remain even with a “negative” spot compression view. Ultrasound may be used in the majority of diagnostic evaluations, even when a potentially suspicious finding appears to efface on spot compression views.
Ultrasound should be performed in most cases of a suspected asymmetry. If the tissue is not dense and the general location of the asymmetry is defined, this can be performed as a targeted exam. If the tissue is dense and/or the finding cannot be as precisely located, more extensive or even whole breast scanning may be necessary. If ultrasound is negative, the finding can usually be confidently assessed as benign and the patient can return to routine screening. The decision to designate a finding BI-RADS 3 (probably benign, for which short-term imaging follow-up is performed) is dependent on the level of confidence after fully assessing the area with tomosynthesis mammography and ultrasound imaging ( Fig. 8.1 ). In the authors’ practice, use of tomosynthesis in diagnostic mammography has resulted in the percentage of asymmetries designated BI-RADS 3, probably benign lesions, to be substantially reduced.
Benign masses are commonly encountered on tomosynthesis in both the screening and diagnostic settings. Although tomosynthesis has been shown to reduce overall screening recall rates, some studies have shown a shift in the type of finding recalled. Although relatively more masses are recalled, this is offset by a decrease in the number of asymmetries recalled with tomosynthesis exams. By far the most common benign masses encountered are cysts, fibroadenomas, and lymph nodes.
In many instances, these findings may be confidently dismissed as BI-RADS 2 findings on screening exams based on clearly benign features, multiplicity, and/or stability. Some findings may still require recall for complete assessment. Of the recalled findings, many may be first evaluated with ultrasound because standard tomosynthesis views provide adequate information about lesion location and morphology compared with 2D mammography alone ( Fig. 8.2 ).
Close inspection of prior 2D imaging may reveal that “new” masses seen on tomosynthesis were actually present previously.
Cysts are commonly encountered in mammography in both the screening and diagnostic settings. Cysts are typically round or oval and have sharply defined margins. A solitary cyst may be encountered, but more commonly, cysts are multiple and the diagnosis is more easily established. Tomosynthesis accentuates the visibility of cysts. This is especially true in women with dense tissue ( Fig. 8.3 ). If available, close review of prior images may demonstrate the characteristic waxing and waning pattern of these benign entities and prevent recall.
Cysts are a very common finding and may be seen at any age but peak in frequency during the perimenopausal years. Exogenous hormonal therapy can perpetuate fibrocystic changes in some women that might otherwise have subsided with menopause. Nevertheless, a new circumscribed mass in an older woman has to be viewed with suspicion. Careful assessment of the margins must always be performed, and ultrasound can be used for diagnostic evaluation. Beware—some malignant masses, such as medullary or mucinous cancers, metastases, and high-grade invasive cancers, can mimic cysts on mammography; however, ultrasound will generally show a corresponding solid mass.
If a solitary, growing, dominant, or otherwise concerning mass is recalled, ultrasound is necessary to complement tomosynthesis and to further characterize the cystic or solid nature of the mass ( Fig. 8.4 ). In cases of benign simple cysts, ultrasound will reveal an anechoic, oval, or round structure with sharply defined, thin walls, with posterior acoustic enhancement. Complicated cysts are common, particularly in the setting of multiple bilateral cysts. They appear similar to simple cysts except, rather than being anechoic, they have internal echoes. Clustered small cysts are also a common occurrence. Occasionally, calcifications seen on the 2D image will be seen to be within a cyst on tomosynthesis. When the calcifications demonstrate the typical layering “teacup” appearance within the cyst, further imaging evaluation is unnecessary.
Fibroadenomas are benign fibroepithelial tumors of the breast. They are the most common benign solid tumors of the breast in women of all ages and can be solitary or multiple and bilateral. Fibroadenomas may be indistinguishable from cysts on tomosynthesis, most frequently demonstrating an oval shape and circumscribed margins ( Fig. 8.5 ). Some fibroadenomas may have been considered to be part of the patient’s parenchymal pattern on 2D mammography. When “newly” discovered on tomosynthesis, some of these fibroadenomas can be shown to be stable after careful review of patients’ prior mammograms and ultrasound images. Detailed characterization of such common benign masses afforded by tomosynthesis imaging may preclude additional diagnostic work-up, ultrasound, and short interval follow-up in many patients ( Fig. 8.6 ).
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here