LEARNING OBJECTIVES
1. Imaging findings and evolution of biopsy-related changes
2. Imaging findings of trauma-related changes in the breast
3. Imaging findings and evolution of changes expected following lumpectomy and radiation therapy
4. Medical complications related to breast hypertrophy
5. Imaging findings associated to reduction mammoplasty
6. Imaging women with implants
7. Imaging implant-related complications
EXCISIONAL BIOPSY CHANGES
Accurate pre-operative wire localization methods facilitate minimal volume biopsies and an increasing number of minimally invasive imaging-guided needle biopsies have reduced the need for excisional biopsies. Consequently, the likelihood of an excisional biopsy creating a permanent change that affects subsequent mammographic interpretations has decreased significantly. When post-operative changes occur, they often resolve in the first several months after the procedure. Unless a mammogram is done in the first 6 months following a biopsy, annual mammograms are normal in over 50% of these women (1–3) and, in those women in whom a change is seen, it is often recognizable as related to a biopsy particularly when correlated with history to the site of the biopsy.
The technologists take a careful history, inspect the breast, and document the site of any scars on the patient’s history sheet. This allows the interpreting radiologist to correlate any findings on the mammogram with prior biopsy sites. Although we do not routinely mark biopsy scars at the time screening studies are done, thin wires or metallic BBs can be used (2). Given the expense associated with these markers, in conjunction with the high number of women who have had at least one breast biopsy, we reserve the use of markers for the few patients a year in whom there is a question about an area of mammographic concern correlating with a biopsy site. Most importantly, when reviewing images, we want to minimize distractions that may interfere with our ability to identify early-stage breast cancers; metallic makers (e.g., for nipples) and wires are a distraction that, in most patients, provide no useful diagnostic information (Fig. 11.1).
Changes that may be seen on a mammogram correlating with excisional biopsy sites include skin thickening and retraction, distortion (Fig. 11.2) that is often seen best in one of the standard views and subtle or not apparent in the other view, and a round, oval, or irregular mass possibly fat-containing with spiculated or indistinct margins (Fig. 11.3) (4). Depending on the amount of tissue excised, generalized observations may include asymmetry in the size of the breasts with the operated breast being smaller and parenchymal asymmetry such that less tissue is present at the surgery site compared with remaining tissue at the corresponding site in the contralateral breast (Fig. 11.4). Except for the asymmetric breast size and resulting parenchymal asymmetry seen in some patients, focal changes usually resolve completely on serial mammograms, remain stable after the first year or slowly evolve with time (Fig. 11.5). Increasing amounts of fat centrally in a mass, dispersion of the density, oil cyst formation, and the development of dystrophic calcifications are changes that can be seen on follow up (1–4). Occasionally, retained fragments of the localization wire (Fig. 11.6A), needle tips, or foreign bodies (Fig. 11.6B) may be seen at prior surgical sites.
When biopsy changes are noted mammographically and correlated to a biopsy site, evaluation with ultrasound or magnetic resonance imaging (MRI) is not usually indicated. Ultrasound and MRI findings, however, will vary significantly depending on the timing of these studies to the biopsy. Acutely, biopsy changes on ultrasound may be characterized by increases in the echogenicity of the tissue, loss of normal soft tissue planes, fluid collections (Fig. 11.7A) that may be complex and skin thickening. Distortion and shadowing (5) that may be seen best in one plane and become less apparent as the transducer is rotated 90-degrees are common findings at long-standing biopsy sites.
On MRI, an irregular mass with mural or internal areas of high T1 and T2 signal and minimal or no enhancement is a common finding for acute hematomas (Fig. 11.7B–D; also see Fig. 5.9); if enhancement is present, it is often thin and peripheral in location. Subacute and chronic post-operative fluid collections may demonstrate a heterogeneously high T2 signal with intermediate to low T1 signal and internal non-enhancing nodules. Minimal enhancement may be noted involving the wall of the collection. Following re-absorption of post-operative fluid collections, increased non-enhancing soft tissue with distortion, an associated fat signal and skin changes (retraction, focal thickening) may be apparent.
Post-operative changes are characterized by progressive resolution or stability needs to be established on annual studies. This is particularly important in patients with a high-risk lesion (e.g., atypical ductal hyperplasia, lobular neoplasia, and multiple peripheral papillomas) diagnosed on excisional biopsy, or those with a personal or significant family history of breast cancer. In these patients, a mammogram 6 months after the biopsy is helpful in establishing the presence of any biopsy change (e.g., a new baseline for the patient). Changes seen 6 months after a biopsy are unlikely to represent recurrent or interval cancer. On subsequent studies, however, we expect the changes to stabilize or, more commonly, evolve, becoming less prominent with time. Increases in distortion or density at a biopsy site after the initial 6-month post-operative study warrant a biopsy recommendation and a review of the original pathology (Fig. 11.8). As yearly studies accrue on a patient, comparison is made to the earliest post-operative study available. Subtle progressive changes may not be readily apparent from year to year, but can be quite obvious, if comparison is made to the earliest available post-operative study (Fig. 11.9).
VACUUM-ASSISTED IMAGING-GUIDED BIOPSY
Vacuum-assisted, imaging-guided breast biopsies with an 11G (less commonly 14G) needle can result in complete removal of small lesions. In these patients, a radiopaque marker (often a titanium clip) is deployed to mark the site of the lesion. In patients having biopsies for MRI detected, mammographically occult lesions, the clip marks the site of the MRI detected abnormality for future reference. If a high-risk lesion or malignancy is diagnosed, the radiopaque marker is localized pre-operatively so that the high-risk lesion can be more completely characterized or wide excision of a malignancy can be accomplished. The clip usually remains at, or close to the biopsy site; however, migration of the clip can occur (6,7). If the patient does not undergo surgery, the radiopaque marker is seen on follow-up studies.
If a clip is deployed during an imaging-guided biopsy, orthogonal images are done to document the accuracy of clip placement immediately after the biopsy procedure; we also comment if the original lesion is removed in its entirety. In addition to the radiopaque marker, an air locule or locules (Fig. 11.10) increased density in a tubular configuration (Fig. 11.11) in one of the images and more mass-like on the orthogonal view (e.g., denoting the needle track), and increased soft tissue stranding or a mass may be seen at the biopsy site (Fig. 11.12). If present, imaging-guided biopsy changes usually resolve within the first several days following the biopsy and do not produce long-term sequelae. If an MRI is done following an imaging-guided biopsy, fluid collections will usually have a high T2 signal, and the T1 signal will vary depending on the timing of the MRI with respect to the biopsy (in the acute setting they may have a high T1 signal that decreases with time); these are sometimes noted along the distribution of the needle track (Fig. 11.13).
TRAUMA
Following trauma to the breast, patients may present acutely with an ecchymosis at the site of the trauma. Irregular areas of increased density, a mass reflecting the presence of a hematoma (Fig. 11.14A; also see Fig. 7.55), parenchymal asymmetry, or focal prominence of the trabecular markings (Fig. 11.15; also see Fig. 9.7) may be seen mammographically. On ultrasound, a hyperechoic or complex cystic and solid mass (Fig. 11.14B; also see Fig. 7.55B) may be seen alternatively; mass-like areas of hyperechogenicity with internal hypo- to anechoic areas are common sonographically. Clinically, the ecchymosis resolves. Mammographic and sonographic findings also usually resolve; however, some patients develop a hard palpable mass on physical exam that is often a fat-containing mass mammographically (Figs. 11.14D, 11.16A, and 11.17A). Alternatively, an oil cyst and dystrophic calcifications may develop at the site of trauma (Fig. 11.16B). Many patients do not present acutely following the trauma but rather months after the incident. Consequently, if a fat-containing mass is seen mammographically with hyperechogenicity and internal hypoechogenicity on ultrasound (Fig. 11.17B), specifically ask the patient about trauma. Also, before assuming a mammographic finding is related to trauma, consider the location of the lesion; trauma is typically going to involve the upper quadrants of the breasts, and it is unlikely to involve the lower central aspects of the breasts posteriorly. In some patients on anti-coagulants, hematomas may develop with the patient having no recollection of significant trauma (Fig. 11.17).
Following a seat belt injury, patients may present with a band-like area of increased density corresponding to the course of the seat belt. The findings are localized to the upper inner or central quadrants of the left breast or the lower inner quadrant of the right breast when the patient is the driver. Findings in the upper inner quadrant of the right breast (Fig. 11.18) or lower inner quadrant of the left breast may be identified when the patient is the front seat passenger (8).
CONSERVATIVE BREAST CANCER TREATMENT
The primary aim of breast conserving treatment is adequate local control of breast cancer. Wide surgical margins are desired in minimizing local recurrences. The cosmetic results, however, are an important secondary consideration. If a substantial amount of tissue needs to be removed to obtain clear margins in a patient with a small breast, cosmesis may not be acceptable. Radiation therapy is used to control any residual occult breast cancer. It is begun 2 to 5 weeks after the lumpectomy. Treatment is given 5 days a week for a total of 5 weeks and delivers 45 to 50 Gy to the affected breast. A boost to the lumpectomy site may be given using an electron beam or iridium implant, increasing the total dose delivered to 60 to 66 Gy (2). Although not all patients with breast cancer are good candidates, high doses of radiation therapy delivered over a 5-day period to the site of the tumor (brachytherapy) are being used with early results comparable to those of whole breast radiation. The other alternative being used with increasing frequency is neoadjuvant therapy as the initial treatment. This approach has been traditionally reserved for patients with inflammatory carcinoma; however, it is now being extended to include two other groups of patients: (i) those with known metastatic disease to the axilla in whom systemic disease is a concern and (ii) those with larger primary lesions in whom conservative therapy is not a good option at the time of presentation but for whom it can become an option if the tumor shrinks (see Chapter 5 for additional discussion regarding imaging in these patients).
Mammographic changes related to whole breast radiation therapy involve the breast diffusely and are primarily related to edema. These include skin and trabecular thickening resulting in increases in parenchymal density and reduced breast compressibility. The changes usually resolve within the first 2 years following treatment (Fig. 11.19; also see Fig. 9.1). Technical factors for adequate exposure of the remaining breast tissue may need to be adjusted accordingly. In the acute setting, skin thickening, dilated lymphatics, and increased tissue echogenicity may be seen on ultrasound following radiation therapy. Residual skin thickening is seen in approximately 20% of women 2 years after radiation therapy (2). Rarely, patients develop fairly extensive fat necrosis that involves the breast diffusely (e.g., not localized to the lumpectomy site) with progressive fibrosis, contracture, and patient discomfort (Fig. 11.20). It is unclear if this process is related solely to the radiation or if it reflects a combination of radiation therapy and surgical effect. Long-term complications of radiation therapy are now also being seen with increasing frequency in part a function of an increase in the number of follow-up years in women who have had whole breast radiation therapy for breast cancer. Patients may present with pulmonary fibrosis, pleural effusions, rib fractures (osteonecrosis), and secondary malignancies often sarcomatous in nature involving the skin (e.g., cutaneous angiosarcomas), breast parenchyma (Fig. 11.21), or soft tissues of the chest and upper abdominal wall (Fig. 11.22) included in, or surrounding, the radiation field (9,10).
The findings seen mammographically following lumpectomy are variable, usually evolve with time and are localized to the surgical site (Fig. 11.19). These include irregular increases in density, distortion, a mass with spiculated margins (fat necrosis), and a mass at the lumpectomy site reflecting a post-operative fluid collection as well as localized skin thickening and retraction. On MRI, non-enhancing distortion and a non-enhancing mass with spiculated margins are the most common findings noted at prior lumpectomy sites; associated skin thickening and retraction are also seen and, in some patients, depending on the location of the original tumor, tenting of the pectoral muscle.
Fat necrosis developing at the lumpectomy site is variably sized and characterized by an irregular mass with indistinct, or more commonly spiculated margins, or distortion alone (Fig. 11.19C, D). As the acute inflammatory changes associated with fat necrosis resolve, the mass or area of distortion decreases in size and density, and oil cysts or dystrophic calcifications (Fig. 11.23) can be seen developing at the lumpectomy site (2,4,11–17). On ultrasound, fat necrosis often results in disruption of normal soft tissue planes as well as distortion and intense shadowing at the surgical site sometimes associated with an irregular mass (5).
Fluid collections are seen in as many as 50% of patients within the first 4 weeks after the lumpectomy (2). An oval or round (Figs. 11.19A, B and 11.24A) mass with circumscribed to indistinct or spiculated margins (Fig. 11.24C) is identified at the lumpectomy site. These are variable in density; however, some are relatively low in density particularly when considering their size. They may have lucent locules or fluid–fluid levels. An internal halo may be present partially outlining the inner margin of the mass (Fig. 11.25). On ultrasound, complex cystic and solid masses with septations (Fig. 11.26A; also see Fig. 4.29B), thickened walls, or echogenic nodules (Fig. 11.26B; also see Fig. 4.29A) are imaged corresponding to the masses seen mammographically. Less commonly a complex cystic and solid mass that is predominantly solid with cystic spaces (Fig. 11.24B and 11.26C) or a nearly anechoic mass is seen. Most fluid collections resolve within the first 2 years after the lumpectomy; however, some may persist for years. If the patient is asymptomatic, aspiration is not indicated and should be avoided because, in many patients, the fluid re-accumulates rapidly after percutaneous drainage, the re-accumulated fluid collection may be larger than the starting point and chronic draining sinuses can develop (Fig. 11.27; also see Fig. 4.30). Draining sinuses can be hard to manage and affect the patient’s quality of life significantly. The signal characteristics of post-operative fluid collections on MRI are variable depending on the age of the collection. Acutely, they may demonstrate high T1 and T2 signal with minimal rim enhancement. As the collection ages, it decreases in size, the margins become better defined and the T1 signal usually decreases. Fluid–fluid levels may be apparent (Fig. 11.28).
Currently, there is no consensus on an appropriate follow-up protocol for patients following lumpectomy. If the patient is asymptomatic following conservative breast cancer therapy, the American College of Radiology practice guidelines on screening and diagnostic mammography (18) leaves the decision on how to schedule (e.g., as screening or diagnostic studies) these patients at the discretion of the imaging facility. Some facilities recommend 6-month follow ups of the treated breast for 3, 5, or 7 years and annual imaging of the contralateral, presumably normal breast (2). In our practice, we obtain a pre-radiation mammogram on patients who presented with an extensive area of calcifications. Rarely, we identify patients with residual calcifications at the lumpectomy site (14) who may benefit from re-excision before radiation therapy is started. Following completion of radiation therapy, we schedule patients for diagnostic studies annually for 7 years after which we return them to screening. We are not routinely obtaining 6-month follow ups of the treated breast following conservative therapy since there is nothing to suggest that recurrences grow any faster than primary lesions. In addition to routine views (CC and MLO), we obtain a spot compression magnification view of the lumpectomy site in tangent to the x-ray beam.
When evaluating these patients, it is helpful to have information on those features of the initial tumor that may influence the likelihood of recurrence including tumor size and grade, proximity of tumor to margins, presence of an extensive intraductal component, lymphovascular space involvement, and lymph node status. Additionally, it is helpful to know if the patient had radiation, chemotherapy, and if she is being treated with tamoxifen. Ideally, the patient’s imaging evaluation at the time of diagnosis is available for review since recurrences often resemble the appearance of the primary. The likelihood of recurrence is low in the first 2 years following treatment; it may be that some of the lesions identified within this time period reflect residual (inadequately treated) disease and not necessarily a recurrence. In the first 7 years following treatment, recurrences are likely to arise at or close to the lumpectomy site (Figs. 11.29 and 11.30); after this, recurrences or second primaries develop with an equal frequency anywhere in the breast (Fig. 11.30) (19–21). Fine pleomorphic calcifications with linear forms or demonstrating linear distribution, developing at prior lumpectomy sites should be biopsied (Fig. 11.29). Developing masses (Fig. 11.31) or increases in the size and density of architectural distortion at the lumpectomy site may also indicate recurrence. Rarely, recurrences will present with diffuse changes (Fig. 11.32; also see Fig. 9.17). In addition to meticulously evaluating the treated breast, carefully evaluate the contralateral side, since patients with a personal history of breast cancer are at increased risk for the development of breast cancer in the contralateral side (Fig. 11.33; also see Figs. 2.47 and 8.18). Aggressively pursue developing changes with spot compression or spot compresson magnification views, correlative physical examination and ultrasound.